DRC-2021-005548 - 0901a06880e6d465

Radioactive Material License Application / Federal Cell Facility

Page Q-1 Appendix Q April 9, 2021
Revision 0

APPENDIX Q

DEPLETED URANIUM PERFORMANCE ASSESSMENT

(Neptune, 2015 and 2021c)

 NAC-0024_R4

Final Report for the Clive DU PA
Model

Clive DU PA Model v1.4
November 24, 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Final Report for the Clive DU PA Model
24 November 2015 ii

Final Report for the Clive DU PA Model
24 November 2015 iii
1. Title: Final Report for the Clive DU PA Model
2. Filename: Clive DU PA Model Final Report v1.4.docx
3. Description: This report describes details of the Clive DU PA Model v1.4, and references report Appendices
(white papers) for further details not included in the main body of this report.

Name Date
4. Originator K. Catlett, G. Occhiogrosso, R. Lee 24 November 2015
5. Reviewers D. Levitt, P. Black, M. Sully 25 November 2015
56. Remarks

This report describes the Clive DU PA Model v1.4.

6/5/2014: Added revisions due to ES comments.

5 Aug 2014: Modifications completed in response to interrogatories. – R Perona, J Tauxe

28 Oct 2015: This document is incomplete for v1.3. It is superseded by v1.4. Changes were made to engineering
drawings and associated fallout, including inventory to make v1.4. U solubility in deep time was reverted to v1.2 in
v1.4.

28 Oct 2015: Updated for v1.4 with new disposal unit description and updated model results based on latest
engineering drawings. – G. Occhiogrosso
8-11 Nov 2015: Review, edit, modify text throughout. K. Catlett
12-24 Nov 2015: Review, edit, add SA results. K. Catlett and G. Occhiogrosso

Final Report for the Clive DU PA Model
24 November 2015 iv

This page intentionally left blank, aside from this statement.
Final Report for the Clive DU PA Model
24 November 2015 v
CONTENTS
Executive Summary ........................................................................................................................ 1

1.0
 Background .......................................................................................................................... 11

1.1
 Depleted Uranium .......................................................................................................... 11

1.2
 The Clive Waste Disposal Facility ................................................................................. 12

1.3
 Regulatory Context ........................................................................................................ 12

1.4
 Performance Assessment ............................................................................................... 16

1.5
 Technical Evolution of PA and PA Modeling ............................................................... 18

1.6
 Report Structure ............................................................................................................. 19

2.0
 Introduction .......................................................................................................................... 20

2.1
 General Approach .......................................................................................................... 20

2.2
 General Facility Description .......................................................................................... 23

3.0
 Features, Events and Processes ............................................................................................ 26

4.0
 Conceptual Site Model ......................................................................................................... 26

4.1.1
 Disposal Site Location .............................................................................................. 27

4.1.2
 Disposal Site Description ......................................................................................... 27

5.0
 Model Structure ................................................................................................................... 43

5.1
 Summary of Important Assumptions ............................................................................. 43

5.1.1
 Points of Compliance ............................................................................................... 43

5.1.2
 Time Periods of Concern .......................................................................................... 43

5.1.3
 Closure Cover Design Options ................................................................................. 44

5.1.4
 Waste Concentration Averaging ............................................................................... 44

5.1.5
 Environmental Media Concentration Averaging ...................................................... 44

5.1.6
 Members of the Public ............................................................................................. 44

5.1.7
 Inadvertent Human Intrusion .................................................................................... 44

5.1.8
 Deep Time Evaluation .............................................................................................. 45

5.2
 Distribution Averaging ................................................................................................... 45

5.3
 Model Evaluation through Uncertainty and Sensitivity Analysis .................................. 46

5.4
 Clive DU PA Model Structure ....................................................................................... 47

5.4.1
 Materials ................................................................................................................... 48

5.4.2
 Processes .................................................................................................................. 48

5.4.3
 Inventory .................................................................................................................. 48

5.4.4
 Disposal .................................................................................................................... 49

5.4.5
 Exposure and Dose ................................................................................................... 50

5.4.6
 Groundwater Protection Level Calculations ............................................................ 51

5.4.7
 Deep Time ................................................................................................................ 51

5.4.8
 Supplemental Containers .......................................................................................... 51

6.0
 Results of Analysis .............................................................................................................. 52

6.1
 Groundwater Concentrations ......................................................................................... 54

6.1.1
 Summary of Results for Groundwater ...................................................................... 54

6.1.2
 Sensitivity Analysis for Groundwater ...................................................................... 58

6.2
 Receptor Doses ............................................................................................................... 60

6.2.1
 Summary of Results for Doses ................................................................................. 60

6.2.2
 Sensitivity Analysis for Doses ................................................................................. 61

6.3
 Receptor Uranium Hazard Indices ................................................................................. 63

6.3.1
 Summary of Results for Uranium Hazard ................................................................ 63

Final Report for the Clive DU PA Model
24 November 2015 vi
6.3.2
 Sensitivity Analysis for Uranium Hazard Index ...................................................... 63

6.4
 ALARA .......................................................................................................................... 64

6.5
 Deep Time Results ......................................................................................................... 66

6.5.1
 Sedimentation and Lake Timing Results .................................................................. 68

6.5.2
 Lake Sediment Concentrations ................................................................................. 68

6.5.3
 Lake Water Concentrations ...................................................................................... 75

6.5.4
 Radon flux results after the first lake ....................................................................... 79

6.5.5
 Rancher radon results after the first lake .................................................................. 81

7.0
 Summary .............................................................................................................................. 82

7.1
 Interpretation of Results ................................................................................................. 82

7.2
 Comparison to Performance Objectives ......................................................................... 84

8.0
 Conclusions .......................................................................................................................... 85

9.0
 References ............................................................................................................................ 87

List of Appendices ........................................................................................................................ 90

Final Report for the Clive DU PA Model
24 November 2015 vii
Figures
Figure 1. Location of the Clive site operated by EnergySolutions (base image from Google
Earth). .......................................................................................................................... 13

Figure 2. Disposal and Treatment Facilities operated by EnergySolutions. ................................. 25

Figure 3. Top level of the Clive DU PA Model v1.4. ................................................................... 49

Figure 4. Control Panel for the Modeling of the Clive Disposal Facility. .................................... 50

Figure 5. Time history of 99Tc well concentrations; 1000 realizations shown. ............................ 56

Figure 6. Time history of 99Tc well concentrations: statistical summary of the 1000
realizations shown in Figure 5. ................................................................................... 57

Figure 7. Partial dependence plot for peak 99Tc groundwater concentration in 500 years. .......... 59

Figure 8. Partial dependence plots for the mean ranch worker dose, assuming waste below
grade. ........................................................................................................................... 62

Figure 9. Evolution of sediment thickness in deep time. .............................................................. 69

Figure 10. Time of appearance of first intermediate lake to reach the Clive elevation. ............... 70

Figure 11. Time history of concentrations of uranium-238 in sediments ..................................... 72

Figure 12. Time history of concentrations of thorium-230 in sediments ..................................... 73

Figure 13. Time history of concentrations of radium-226 in sediments ....................................... 74

Figure 14. Time history of concentrations of uranium-238 in lake water, 1000 realizations
shown. ......................................................................................................................... 76

Figure 15. Time history of concentrations of uranium-238 in lake water .................................... 77

Figure 16. Time history of concentrations of thorium-230 in lake water ..................................... 78

Figure 17. Time history of concentrations of radium-226 in lake water ...................................... 79

Figure 18. 222Rn ground surface flux in deep time. ...................................................................... 81

Final Report for the Clive DU PA Model
24 November 2015 viii
Tables
Table ES-1. Peak TEDE: statistical summary ................................................................................ 7

Table ES-2. Peak groundwater activity concentrations within 500 yr, compared to GWPLs ........ 7

Table ES-3. Cumulative population TEDE: statistical summary ................................................... 8

Table ES-4. Statistical summary of lake water concentrations at peak lake occurrence, 90 ky ..... 9

Table ES-5. Statistical summary of sediment concentrations at peak lake occurrence, 90 ky ....... 9

Table ES-6. Summary of the results of the Clive DU PA Model ................................................... 9

Table 1. Exposure Pathways Summary ........................................................................................ 38

Table 2. Summary statistics for peak mean groundwater activity concentrations within 500 yr,
compared to GWPLs ................................................................................................... 54

Table 3. Sensitivities of select peak groundwater concentrations within 500 years. .................... 58

Table 4. Peak of the mean TEDE: statistical summary within 10,000 yr. .................................... 60

Table 5. Sensitivities of peak TEDE within 10,000 yr ................................................................. 61

Table 6. Peak of the mean uranium hazard index within 10,000 years. ....................................... 63

Table 7. Sensitivities of uranium hazard index within 10,000 yr ................................................. 64

Table 8. Cumulative population TEDE: statistical summary ....................................................... 64

Table 9. Cumulative receptor population: statistical summary .................................................... 65

Table 10. Statistical summary of the flat rate ALARA costs ....................................................... 65

Table 11. Statistical summary of deep time sediment concentrations at model year 90,000.
Based on 1000 realizations. ........................................................................................ 71

Table 12. Statistical summary of deep time lake concentrations at model year 90,000. Based
on 1000 realizations. ................................................................................................... 75

Table 13. Statistical summary of radon-222 flux concentrations after the first lake recedes. ...... 80

Table 14. Statistical summary of doses to ranchers after the first lake recedes. ........................... 82

Table 15. Summary statistics for peak mean groundwater activity concentration of 99Tc
within 500 yr ............................................................................................................... 84

Table 16. Peak mean TEDE for ranch worker: statistical summary ............................................. 84

Table 17. Summary of results of the Clive DU PA Model ........................................................... 85

Final Report for the Clive DU PA Model
24 November 2015 1
Executive Summary
Neptune and Company, Inc., (Neptune) under contract to EnergySolutions, LLC
(EnergySolutions), has developed a computer model (the Clive DU PA Model, or the Model) to
support decision making related to the proposed disposal of depleted uranium (DU) wastes at the
low-level radioactive waste (LLW) disposal facility at Clive, Utah, operated by EnergySolutions.
The Model provides a platform on which to conduct analyses relevant to performance assessment
(PA), as required by the State of Utah in Utah Administrative Code (UAC) R313-25, License
Requirements for Land Disposal of Radioactive Waste (Utah 2015). Specifically, a PA is
required in UAC R313-25-9, Technical Analyses. The model may also serve to inform decisions
made by the Site operator to gain maximum utility of the resource that is the Clive Facility.
Depleted uranium is the remains of the uranium enrichment process, of which the fissionable
uranium isotope 235U is the product. The leftover uranium, depleted in 235U, is predominantly
238U, but may include small amounts of other U isotopes. In general, DU will contain very small
amounts of decay products in the uranium, thorium, actinium, and neptunium series of decay
chains. Some specific DU waste, resulting from introduction of uranium recycled from used
nuclear reactor fuel (reactor returns) into the separations process, contains varying amounts of
contaminants, in the form of fission and activation products. Since some of the DU evaluated in
this PA includes reactor returns, it is here termed “DU waste”. The national inventory of DU is
on the order of 700 Gg (700,000 Mg, or metric tons) in mass as uranium hexafluoride (DUF6),
and the bulk of it exists in its original storage cylinders, awaiting conversion to oxide form for
disposal. This conversion is being performed at the Portsmouth, Ohio, and Paducah, Kentucky
gaseous diffusion plant (GDP) sites, using new purpose-built “deconversion” plants to produce
triuranium octoxide( U3O8). A much smaller mass of DU waste was generated by the Savannah
River Site (SRS) in the form of uranium trioxide (UO3), a powder stored in several thousand
200-L (55-gal) drums. While the composition of the SRS DU is reasonably well known, the
content of the GDP DU is not well documented. For the purposes of this assessment, it was
necessary to assume that some uncertain fraction of the GDP DU waste was contaminated to the
same extent as the SRS DU waste. DU waste from both sources is considered in the Clive DU PA
Model.
The Model is written using the GoldSim probabilistic systems analysis software, which is well-
suited for the purpose. In order to provide decision makers with a broad perspective of the
behavior and capabilities of the Facility, the model considers uncertainty in input parameter
values. This probabilistic assessment methodology is encouraged by the Nuclear Regulatory
Commission (NRC) and the Department of Energy (DOE) in constructing PAs and the models
that support them. The Model can be run in deterministic mode, where a single set of median
model inputs is used, but running in probabilistic Monte Carlo mode provides greater insight into
the model behavior, and especially into model sensitivity to the distribution of input parameter
values. In Monte Carlo mode, a large number of realizations are executed with values drawn at
random from the input parameter distributions using Latin Hypercube Sampling to ensure equal
probability across the range of the input distributions. The distributions of results, therefore,
reflect the uncertainty in these values. To the extent that the model reflects the uncertain state of
knowledge at a site, the model provides insight about how the site works, and what should be
expected if different actions are taken or different wastes are disposed. In this way, the model
aids in decision making, even in the face of uncertainty.
Final Report for the Clive DU PA Model
24 November 2015 2
The Clive Facility is located at the eastern edge of the Great Salt Desert, west of the Cedar
Mountains, and approximately 100 km (60 mi) west of Salt Lake City, Utah. Clive is a remote
and environmentally inhospitable area for human habitation. Human activity at Clive has
historically been very limited, due largely to the lack of potable water, or even water suitable for
irrigation. The site is located on flat ground, with the bottom of the waste disposal cells shallowly
excavated into local lacustrine silts, sands, and clays. A single waste disposal cell, or
embankment, is considered in this model: the Federal cell housing DU. This cell is modeled with
an engineered cover, as per design documents. The top of the cell is above grade, and the cover
has layers of an evapotranspiration (ET) cover system of earthen origin. In time, this cover is
expected to become vegetated with native plants, and occupied to a limited extent by animals
including insects and mammals. As plant communities become established, they are likely to
keep the cover system fairly dry through transpiration.
Water balance modeling of the cover indicates that some water penetrates the cover system, and
this infiltration has the potential to leach radionuclides from the waste and transport them down
through the cell liner and unsaturated zone to the aquifer. In the saturated zone (aquifer),
contaminants are transported laterally to a hypothetical monitoring well located about 27 m
(90 ft) from the edge of the interior of the cell. Since the side slopes of the cell are modeled to not
contain DU waste, the effective distance to the well from the DU waste itself is about 73 m
(240 ft). This environmental transport pathway is relevant for long-lived and readily-leached
radionuclides such as 99Tc. Contributions to groundwater radionuclide concentrations from the
proposed DU waste are calculated for comparison to groundwater protection limits (GWPLs)
during the next 500 years, as stipulated in the water discharge permit (UWQB 2009).
In addition to water advective transport, radionuclides are transported via diffusion in both water
and air phases within the cover system, which can provide upward transport pathways. Gaseous
radionuclides, such as 222Rn, partition between air and water. Soluble constituents partition
between water and solid porous media. Coupled with all these process are the activities of biota,
with plants transporting contaminants to their above-ground surface tissues via their roots, and
burrowing animals (ants and small mammals) moving bulk materials upward and downward
through burrow excavation and collapse. Biota do not play a major role in contaminant transport
contributing to human doses or uranium hazard according to model results. The model does not
consider the effects of enhanced radon diffusion from a compromised radon barrier, but the
model does include an expanded assessment of the performance of the radon barriers with respect
to infiltration.
Once radionuclides reach the ground surface at the top of the engineered cover via the
aforementioned processes, they are subject to suspension into the atmosphere and dispersion to
the surrounding landscape. Atmospheric transport of gases (222Rn) and contaminants sorbed to
suspended particles is modeled using a standard modeling platform approved by the U.S.
Environmental Protection Agency (EPA), called AERMOD. The results of this model are
abstracted into the Clive DU PA Model, and contributions of airborne radionuclides to dose and
uranium toxicity hazard are evaluated.

Final Report for the Clive DU PA Model
24 November 2015 3
The impact of sheet and gully erosion in the Model is evaluated by the application of results of
landscape evolution models of hillslope erosion loss and channel development conducted for a
borrow pit at the site. The model domain for the borrow pit includes the borrow pit floor, a 10-ft
high pit face at a 1:1 slope and several hundred meters of ground surface upslope from the pit
face at a slope of 0.003 (0.3 percent). The soil characteristics are consistent with the Unit 4 silty
clay, though the landscape evolution model did not consider the presence of vegetation or rock
cover. While composed of similar soil, the surface layer of the top slope of the ET cover
proposed for the Federal DU Cell has a slope of about 2 percent, a gravel composition of 15
percent, and will be re-vegetated with a mix of native and non-native species. While the cover on
the top slope part of the embankment has a greater slope than that of the undisturbed area upslope
from the borrow pit face, the top slope characteristics included vegetation and gravel admix that
would act to slow erosion and channel formation.
A subset of the borrow pit model domain was selected to represent the cover. Gully depths
estimated by the erosion model were extrapolated to 10,000 years and a statistical model was
developed that generated values of the percentage of the cover where gullies ended within a given
depth interval. This model provided an estimate of the volume of embankment cover material
removed by gullies. The depositional area of the gully fan is assumed to be the same as the area
of waste exposed in the gullies, using projections onto the horizontal plane. If these embankment
materials include DU waste components, then this leads to some contribution to doses and
uranium hazards. No associated effects, such as biotic processes, effects on radon dispersion, or
local changes in infiltration are considered within the gullies.
Given the remote and inhospitable environment of Clive, it is not reasonable to assume that the
traditional residential receptors considered in other PAs will be present here. Traditionally, and
based on DOE (DOE M 435.1) and NRC guidance (10 CFR 61), members of the public are
evaluated outside the fence line or boundary of the disposal facility, and inadvertent intruders are
assumed to access the disposal facility and the disposed waste directly, in activities such as well
drilling or house construction. For disposal facilities in the arid west, these types of strictly
defined default scenarios do not adequately describe likely human activities. Their inclusion in a
PA for a site in the arid west, such as Clive, will usually result in unrealistic underestimation of
the performance of a disposal system, which does not lend itself to effective decision making for
the Nation’s needs to dispose of radioactive waste.
At Clive, there is no potable water resource to drill for, and historical evidence suggests there is
little likelihood that anyone would construct a residence on or near the site. There are present day
activities in the vicinity, however, that might result in receptor exposures if these activities are
projected into the future when the facility is closed and after institutional control is lost. Large
ranches operate in the area, so ranch hands work in the vicinity. Pronghorn antelope are found in
the region, and hunters will follow them. Both of these activities are facilitated by the use of off-
highway vehicles (OHVs). OHV enthusiasts also ride recreationally for sport in areas near the
facility.
In addition to these receptors, there are specific points of exposure within the vicinity of the Clive
Facility where individuals might be exposed. About 12 km (8 miles) to the west, OHV
enthusiasts use the Knolls Recreation Area. Interstate-80 and a railroad are located to the north,
with an associated rest area on the highway. Closer to the Clive Facility, the Utah Test and
Final Report for the Clive DU PA Model
24 November 2015 4
Training Range access road is used on occasion. The Model hence evaluates dose and uranium
hazard to these site-specific receptors.
The State of Utah follows federal guidance by categorizing receptors in a PA in UAC Rule
R313-25-9 and 10 CFR 61.41 according to the labels “member of the public” (MOP) and
“inadvertent human intruder” (IHI). NRC offers two definitions of inadvertent intruders in
10 CFR 61:
§ 61.2 Definitions. Inadvertent intruder means a person who might occupy the
disposal site after closure and engage in normal activities, such as agriculture,
dwelling construction, or other pursuits in which the person might be
unknowingly exposed to radiation from the waste.
§ 61.42 Protection of individuals from inadvertent intrusion. Design,
operation, and closure of the land disposal facility must ensure protection of any
individual inadvertently intruding into the disposal site and occupying the site or
contacting the waste at any time after active institutional controls over the
disposal site are removed.
NRC offers one reference to an MOP in the context of the general population:
§ 61.41 Protection of the general population from releases of radioactivity.
Concentrations of radioactive material which may be released to the general
environment in ground water, surface water, air, soil, plants, or animals must not
result in an annual dose exceeding an equivalent of 25 millirems [0.25 mSv] to
the whole body, 75 millirems [0.75 mSv] to the thyroid, and 25 millirems
[0.25 mSv] to any other organ of any member of the public. Reasonable effort
should be made to maintain releases of radioactivity in effluents to the general
environment as low as is reasonably achievable.
DOE definitions in DOE M 435.1 (the Manual accompanying DOE Order 435.1) are much more
specific. However, the applicable federal agency that regulates disposal of low-level radioactive
waste at the Clive Facility is NRC. For the Clive Facility and the Model, based on the NRC
definitions, the ranch hand, hunter and OHV enthusiast are expected to engage in activities both
on and off the site. These receptors fit the NRC definition of inadvertent intrusion because they
are assumed to occupy the site, albeit for limited periods of time, and also because the use of
OHVs on the cover may precipitate the creation of gullies. The receptors that are located at
specific offsite locations, instead, fit the NRC definition of MOP. The Model presents predicted
doses to the receptors identified above, under the conditions and assumptions that provide the
basis for the Model. These doses are presented as the results of the Model. A comparison of
doses to both MOP and IHI performance objectives is also presented.
The Model addresses radiation doses to human receptors who might come in contact with
radionuclides released from the disposal facility into the environment subsequent to facility
closure. In accordance with UAC Rule R313-25-9, doses are calculated within a 10,000-year
compliance period. The doses are compared to a performance criterion of 25 mrem in a year for a
MOP, and 500 mrem in a year for an inadvertent intruder. The dose assessment component of the
PA model, like the transport modeling components described above, supports probabilistic Monte
Final Report for the Clive DU PA Model
24 November 2015 5
Carlo analysis. Spatiotemporal scaling is a critical component of the Model development. For
example, the Model differentiates the impact of short-term variability in exposure parameters
(values applicable over a few years or decades, such as individual physiological and behavioral
parameters) from the longer-term variability of transport parameters (values applied over the full
10,000-year performance period, such as hydraulic and geochemical parameters). This distinction
facilitates assessment of uncertainties that relate to physical processes from uncertainties relating
to inter-individual differences in potential future receptors.
In addition to radiation dose, uranium is also associated with non-radiological toxicity. The
potential chemical toxicity of uranium disposed at the Clive Facility is evaluated in the Model.
Potential receptor chronic daily intake of uranium is compared to toxicological criteria developed
by EPA that pertain to a threshold of adverse effect associated with kidney toxicity.
These doses and the supporting contaminant transport modeling that provides the dose model
with radionuclide concentrations in exposure media are evaluated for 10,000 yr, in accordance
with UAC R313-25-9(5)(a). After that time, the modeling focus turns to long-term, or “deep
time” scenarios. Peak activity of the waste occurs when the progeny of the principal parent, 238U
(with a half-life that is approximately the age of the earth— over 4 billion years), reach secular
equilibrium. This occurs at roughly 2.1 My from the time of isotopic separation, and the model
evaluates the potential future of the site in this context. At 2.1 My the activity of the last modeled
member of the chain, 210Pb, is equal to that of 238U, within less than one half of one percent.
While the calculation could be carried out further in time to achieve a greater degree of accuracy,
there is no benefit in doing so for decision-making purposes. This time frame borders on
geologic, and needs to take into account the likely possibility of future deep lakes in the
Bonneville Basin. The return of such lakes is understood to be inevitable, and the Clive Facility,
as constructed, will not survive in its current configuration. Many lakes, of intermediate and deep
size, are expected to occur in the 2.1-My time frame, following the climate cycle periodicity of
about 100,000 yr, based on current scientific understanding of paleoclimatology. In these
timeframes, it is also important to consider processes such as eolian (i.e., wind-borne) deposition,
which can be seen in geologic formations in the Clive area. Deposition builds up the ground
surface over time, such that the ground surface when a lake returns is 2 – 3 m higher than the
current ground surface.
As each lake returns, estimates are made of the radionuclide concentrations in the lake and in the
sediments surrounding and subsuming the site. Because the exact behavior of lake intrusion and
site destruction is speculative, the model makes several conservative assumptions. Upward
movement of radionuclides, via diffusion and biota, is assumed to occur until the first lake
returns. At that point in time, the radionuclides that are above ground are assumed to comingle
with sediments, dispersed over an uncertain area approximately the size of an intermediate lake.
In the presence of a lake, the radionuclides migrate into the water, in accordance with their
aqueous solubility. For U3O8, which is considered to be the only form of uranium oxide
remaining by the time the first lake arrives (since UO3 moves out of the waste first and what is
left will become more like U3O8 or UO2 in the presence of a wetter climate), the solubility of U is
very low. As each lake recedes, radionuclides are co-deposited with the sediment, only to be
dissolved into the water again with the next lake. This is a very conservative approach, especially
for the lake concentrations, since in reality each blanket of sediment could entrap constituents,
and the concentrations in water and sediment over time should decrease consequently. The
Final Report for the Clive DU PA Model
24 November 2015 6
analysis, therefore, focuses on the arrival of the first lake, which will be the most destructive in
terms of sudden release of radionuclides, and would provide the least amount of sediment to
encapsulate them. Subsequent lakes would see progressively less radionuclide activity as the site
is slowly buried under ever-deeper lacustrine deposits through the eons.
The utility of such a calculation, aside from responding to the UAC, is to inform decisions
regarding the placement of wastes in the embankment. With downward pathways influencing
groundwater concentrations, and upward pathways influencing dose and uranium hazard, a
balance must be achieved in the placement of different kinds of waste. In version 1.0 of the Clive
DU PA Model (Neptune 2011), three different options for configuration of the DU waste within
the Class A South embankment (subsequently renamed the Federal DU Cell) were evaluated.
These options included a “3-m model”, named because the top of the DU waste was 3 m below
the embankment cover, and also 5-m and 10-m models. No DU waste is included under the side
slopes for this PA. In addition to these disposal options, two scenarios related to embankment
erosion were evaluated in the Clive DU PA Model v1.0. The first essentially assumed a stable
embankment for 10 ky, with infilling of the cap and continual airborne deposition replacing fine
sediments that are resuspended themselves and subsequently dispersed offsite. The second
scenario was one in which gullies were formed that, depending on the DU waste disposal
configuration, might intersect and expose the DU waste to the environment.
In version 1.4 of the Model, which supports the results described herein, the erosion modeling as
described above and all modeling was conducted under the assumption that gullies will occur on
the embankment. Additionally, the only DU waste configuration presently evaluated is for
disposal of these wastes in layers of the embankment below the current grade of surrounding soil.
Dose results for each type of potential receptor are presented in Table ES-1.
There is a question of which statistic from the many Model realizations is most appropriate for
comparison to performance criteria. The statistics in Table ES-1 represent summaries of the
mean, median, and 95th percentiles of the dose at 10,000 yr for the 10,000 realizations. The peak
mean dose is sometimes of interest for comparison with performance objectives, and in this
model, the peak mean dose occurs at or near 10 ky. In effect, 10 ky is the worst case year in terms
of dose. Under these circumstances, the 95th percentile is analogous to the 95% upper confidence
interval of the mean at 10 ky that is commonly used to represent reasonable maximum exposure
in CERCLA risk assessments.
Compliance with the performance objectives for the inadvertent intruder dose of 500 mrem in a
year and for the MOP of 25 mrem in a year is clearly established for all three types of potential
future receptors. This indicates that for the disposal configuration where DU wastes are placed
below grade, doses are expected to remain well below applicable dose thresholds even if gullies
are assumed to occur on the embankment.
Results are also available for the offsite (MOP) receptors. None of the 95th percentile dose
estimates for these receptors exceeds 1 mrem in a year, and all of the peak mean dose estimates
are at or below 0.1 mrem in a year.

Final Report for the Clive DU PA Model
24 November 2015 7
Table ES-1. Peak TEDE: statistical summary
peak TEDE (mrem in a yr) within 10,000 yr
receptor mean median
(50th %ile) 95th %ile
ranch worker 6.2E-2 5.1E-2 1.5E-1
hunter 4.5E-3 3.8E-3 9.9E-3
OHV enthusiast 8.4E-3 7.5E-3 1.8E-2
Results are based on 10,000 realizations of the Model.
TEDE: Total effective dose equivalent
For those radionuclides for which GWPLs exist, as specified in the facility’s permit (UWQB
2009), results are shown in Table ES-2. For all such radionuclides compliance with the GWPLs
is clearly demonstrated. The mean values for 99Tc and 129I are much greater than the median,
indicating that the distributions of these concentrations have a very strong degree of skewness.
Table ES-2. Peak groundwater activity concentrations within 500 yr, compared to GWPLs
peak activity concentration within 500 yr (pCi/L)
radionuclide GWPL1
(pCi/L) mean median
(50th %ile) 95th %ile
90Sr 42 0 0 0
99Tc 3790 26 4.3E-2 150
129I 21 1.7E-2 4.3E-11 1.1E-1
230Th 83 2.2E-28 0 0
232Th 92 1.4E-34 0 0
237Np 7 1.5E-19 0 3.7E-27
233U 26 5.6E-24 0 3.9E-28
234U 26 2.1E-23 0 2.2E-28
235U 27 1.6E-24 0 2.0E-29
236U 27 2.7E-24 0 3.3E-29
238U 26 1.5E-22 0 1.8E-27
1GWPLs are from UWQB (2009) Table 1A.
Results are based on 10,000 realizations of the Model.
Sensitivity analyses on the Model results indicate that receptor doses are dominated by radon
inhalation, whereas the downward migration pathway is dominated by groundwater
concentrations of 99Tc. A trade-off is indicated in terms of DU waste placement. The lower the
DU waste is placed, particularly the 99Tc-contaminated DU waste, the greater the groundwater
concentrations of 99Tc, but the lower the doses due to increases in the diffusion path length to the
ground surface. Conversely the higher the DU waste is placed in the embankment, the lower the 99Tc groundwater concentrations, and the greater the dose to receptors. Placement of DU waste
below surface grade in the Federal DU cell satisfies both dose and groundwater performance
objectives. Sensitivity analyses on the groundwater concentration of 99Tc indicate that these
Final Report for the Clive DU PA Model
24 November 2015 8
results are primarily sensitive to the α parameter of van Genuchten equation and secondarily to
the molecular diffusion coefficient.
In addition to the dose assessment for hypothetical individuals described above, the structure of
the model allows the cumulative population dose to be tracked. For the objective of keeping
doses as low as reasonably achievable (ALARA), estimated dose to the entire population of ranch
workers, hunters, and OHV enthusiasts over the 10,000-yr simulation was evaluated. These
cumulative population doses are shown in Table ES-3.
The population doses presented in Table ES-3 may be evaluated relative to doses received from
natural background radiation and by considering the person-rem costs suggested in recent NRC
(2015) guidance. The NRC has suggested value of a statistical life (VSL)-based cost of $5,100
per person rem. Using such a cost, the total ALARA cost over 10 ky (for example, $61,200 using
the mean estimate of total population dose, or $6 per yr.) is very small compared to the cost of
waste operations and disposal. Average annual individual background dose related to natural
background radiation in the United States is approximately 3.1 mSv (310 mrem; NCRP, 2009),
which for the total cumulative receptor population of about 3,200,000 individuals in 10,000 years
is approximately 992,000 rem—a level that is many orders of magnitude greater than the
population doses shown in Table ES-3. ALARA is intended to support evaluation of options to
reduce doses in a cost-effective manner. Given the results of this ALARA analysis, it is not clear
that further reduction in dose is necessary.
Table ES-3. Cumulative population TEDE: statistical summary
population TEDE (person-rem) within 10,000 yr
receptor type mean median
(50th %ile) 95th %ile
total population 12 11 26
ranch worker 2.8 2.5 5.7
hunter 1.5 1.3 3.0
OHV enthusiast 8.3 7.4 17
Results are based on 10,000 realizations of the Model.
TEDE: Total effective dose equivalent
The final set of analyses conducted with the Model are the deep-time analyses. As described
above, the deep-time model is very conservative in many ways with respect to dispersal of the
DU waste material. Deep lakes that obliterate the Federal DU Cell are assumed to return
periodically. Simplified processes are used to keep the deep time model from becoming overly
complicated for the amount of uncertainty in both parameters and processes.
Concentrations of 238U in lake water and sediment at the time of peak lake occurrence (90,000
years) are presented in Tables ES-4 and ES-5. These results simply show the concentrations that
might occur in response to obliteration of the site by wave action during return of a lake to the
elevation of Clive and subsequent dispersal of the waste in a relatively confined system. The
concentrations presented would continue to decrease with each lake and climate cycle as more
sediment is deposited with each lake event, and each lake event allows radionuclides to be
dispersed ever further afield.
Final Report for the Clive DU PA Model
24 November 2015 9
Table ES-4. Statistical summary of lake water concentrations at peak lake occurrence,
90 ky
Lake concentrations (pCi/L) at 90,000 yr
radionuclide mean median
(50th %ile) 95th %ile
uranium-238 2.1E-5 0.018 0.11
radium-226 0.15 0.54 2.4
thorium-230 0.15 0.55 2.4
Results are based on 1,000 simulations of the Model
Table ES-5. Statistical summary of sediment concentrations at peak lake occurrence, 90 ky
Sediment concentrations (pCi/g) at 90,000 yr
radionuclide mean median
(50th %ile) 95th %ile
uranium-238 1.8E-3 2.0E-2 9.5E-2
radium-226 1.2E-3 5.0E-3 2.2E-2
thorium-230 1.2E-3 5.0E-3 2.3E-2
Results are based on 1,000 simulations of the Model
The deep-time model disperses the above-ground radionuclides that have migrated upward from
the DU waste prior to the occurrence of the first returning lake. The current disposal scenario has
the entire DU waste disposed below grade. The model assumes that no material below grade is
dispersed. Based on these results, it is reasonable to expect that the deep-time concentrations
could be close to or possibly less than background concentrations for uranium in soil of about
1 pCi/g (Myrick, et al., 1981, Table 30) and approximately 2 pCi/L for background uranium
concentrations in the Great Salt Lake (CRWQCB, 1990, Table 5). In addition, the return of the
first lake is considered likely to be several tens of thousands of years, or even a few hundreds of
thousands of years, into the future, at which point eolian deposition will result in sedimentation
deposits around the site of several meters. This deposition will both stabilize the site and make it
even less likely that any below-grade material will be dispersed.
The quantitative results for all Model analyses are summarized in Table ES-6. Doses to all
receptors are always less than the 500-mrem (IHI) and 25-mrem (MOP) annual performance
criteria. Groundwater concentrations are always less than the GWPLs. Even in the case of 99Tc,
the peak median, mean and 95% groundwater concentrations are well below the GWPL of
3,790 pCi/L.

Table ES-6. Summary of the results of the Clive DU PA Model
performance objective meets performance
objective?
Dose to MOP below regulatory threshold of 25 mrem in a year Yes
Dose to IHI below regulatory threshold of 500 mrem in a year Yes
Final Report for the Clive DU PA Model
24 November 2015 10
Groundwater maximum concentration of 99Tc in 500 years < 3790
pCi/L Yes
ALARA average total population cost equivalent over 10,000 years: $61,200

The results overall suggest clearly that the below-grade disposal configuration can be used to
dispose of the quantities of DU waste included in the Model in a manner adequately protective of
human health and the environment.

Final Report for the Clive DU PA Model
24 November 2015 11
1.0 Background
One of the responsibilities of the Nuclear Regulatory Commission (NRC) is to ensure the safe
disposal of commercially generated low-level radioactive waste. Non-defense-related depleted
uranium (DU) waste falls under the jurisdiction of NRC, and requires a disposal option that is
protective of human health and the environment. NRC currently regulates the disposal of DU
waste as a low-level radioactive waste, in cooperation with “Agreement States”. The
EnergySolutions low-level radioactive waste disposal facility at Clive, Utah is a candidate for
disposal of DU waste, and Utah is an Agreement State that has regulatory authority to determine
if such disposal can occur in compliance with Utah and NRC regulatory requirements.
Adequate protection of human health and the environment is evaluated by conducting a
Performance Assessment (PA). A PA is used to model potential transport of radionuclides from
the disposed inventory to the accessible environment, and to estimate radiation dose to potential
human receptors. The estimated doses are compared to performance objectives, which are
specified as dose limits. If the estimated doses are less than the performance objectives, then
adequate protection of human health has been demonstrated.
The purpose of this report is to present the results of the Clive DU PA Model v1.4 (the Model), a
computer model developed to inform PA for disposal of specific DU waste materials at the Clive
Facility. This report provides a summary of the approach taken and the results that can be
obtained from the Model, and is accompanied by supporting documentation that includes details
of the Model development and quality assurance program.
1.1 Depleted Uranium
In order to produce suitable fuel for nuclear reactors and/or weapons, uranium has to be enriched
in the fissionable 235U isotope. Uranium enrichment in the US began during the Manhattan
Project in World War II. Enrichment for civilian and military uses continued after the war under
the U.S. Atomic Energy Commission, and its successor agencies, including the DOE.
The uranium fuel cycle begins by extracting and milling natural uranium ore to produce "yellow
cake," which is a varying mixture of uranium oxides. Low-grade natural ores contain about 0.05
to 0.3% by weight of uranium oxide while high-grade natural ores can contain up to 70% by
weight of uranium oxide. Uranium found in natural ores contains two principal isotopes –
uranium-238 (99.3% 238U) and uranium-235 (0.7% 235U). The uranium is enriched in 235U before
being made into nuclear fuel, which generates a product consisting of 3% to 5% 235U for use as
nuclear fuel and a by-product of DU (between 0.1% and 0.5 235U). The DU has some commercial
applications including counterweights and military applications as artillery. However, the
commercial demand for depleted uranium is currently much less than the amounts generated for
nuclear fuel. Use of 238U as fuel for breeder reactors has not been seriously considered in this
country. The U.S. Department of Energy (DOE) has about 700 Gg (700,000 Mg or metric tons)
of DUF6 in storage, containing roughly 464 Gg of uranium. Hence, the need to find disposal
options for DU waste.
Final Report for the Clive DU PA Model
24 November 2015 12
1.2 The Clive Waste Disposal Facility
EnergySolutions operates a low-level radioactive waste disposal facility west of the Cedar
Mountains in Clive, Utah, as shown in Figure 1. Clive is located along Interstate-80,
approximately 5 km (3 mi) south of the highway, in Tooele County. The facility is approximately
80 km (50 mi) east of Wendover, Utah and approximately 100 km (60 mi) west of Salt Lake City,
Utah. The facility sits at an elevation of approximately 1302 m (4275 ft) above mean sea level
(amsl) and is accessed by both road and rail transportation.
Currently, the Clive Facility receives low-level radioactive waste shipped via truck and rail. The
Clive disposal facility is licensed to accept Class A low-level radioactive waste. Under current
NRC regulations, DU waste is considered Class A waste, in which case the Clive site is an option
for disposal. However, NRC is currently considering options for updating 10 CFR 61, and the
State of Utah has updated their regulations (UAC-R313-25-9 [Utah 2015]), which force the
requirement of a PA for disposal of DU. Pending the findings of the Clive DU PA, DU waste will
be disposed in an above-ground engineered disposal embankment that is clay-lined with clay
barriers and an ET cover. The disposal embankment is designed to perform for a minimum of
500 years based on requirements of 10 CFR 61.7, and hence provides a possible solution for the
long-term disposal of DU.
Clive is a remote and environmentally inhospitable area. Human activity at Clive has,
historically, been very limited. The regulations (10 CFR 61 and Utah regulations R313-25-9)
indicate the need to evaluate performance with respect to members of the public and inadvertent
human intruders. However, the difference between these two categories of human receptors is
somewhat blurred because of the types of human activities that are reasonable to consider in the
general area of the disposal facility. These two categories of receptors are described further below
in the context of the regulatory context of the Clive DU PA.
1.3 Regulatory Context
EnergySolutions is permitted by the State of Utah to receive Class A Low Level under Utah
Administrative Code (UAC) R313 25, License Requirements for Land Disposal of Radioactive
Waste. The wastes that are received must be classified in accordance with the UAC
R313 15 1009, Classification and Characteristics of Low-Level Radioactive Waste. The
classification requirements in UAC R313-15-1009 reflect those outlined in NRC’s 10 CFR 61
Section 55, but include additional references to radium 226 (226Ra). Further, groundwater
protection levels (GWPLs) must be adhered to, as outlined in the site’s Ground Water Quality
Discharge Permit (UWQB, 2010).
Title 10 CFR 61 (Code of Federal Regulations, 2007) is the Federal regulation for the disposal of
certain radioactive wastes, including land disposal at privately-operated facilities such as that
managed and operated by EnergySolutions at Clive, Utah. It contains procedural requirements,
performance objectives, and technical requirements for near-surface disposal, including disposal
in engineered facilities with protective earthen covers, which may be built fully or partially
above-grade. Near-surface disposal is defined as disposal in or within the upper 30 m (100 ft) of
the earth’s surface (10 CFR 61.2).
Final Report for the Clive DU PA Model
24 November 2015 13

Figure 1. Location of the Clive site operated by EnergySolutions (base image from Google
Earth).

Performance objectives are evaluated by preparing a PA model. DU presents an interesting case
because the uranium is nearly all 238U, meaning that secular equilibrium is not attained for more
than 2 My, and during that time, activity associated with the DU continues to increase. At the
time of the development of the regulation, DU waste as such did not, and was not expected to,
exist in significant quantities. The nature of the radiological hazards associated with DU presents
challenges to the estimation of long-term effects from its disposal. Recognition of this special
behavior of DU has prompted the NRC to revisit the regulation. Until that process is complete,
however, 10 CFR 61 stands as the controlling regulation.
The key endpoints of a PA are estimated future potential doses to members of the public (MOP).
The performance objectives specified in Subpart C of 10 CFR 61 are in the following section:
§ 61.41 Protection of the general population from releases of radioactivity.
Concentrations of radioactive material which may be released to the general
environment in ground water, surface water, air, soil, plants, or animals must not
result in an annual dose exceeding an equivalent of 25 millirems [0.25 mSv] to
the whole body, 75 millirems [0.75 mSv] to the thyroid, and 25 millirems
Final Report for the Clive DU PA Model
24 November 2015 14
[0.25 mSv] to any other organ of any member of the public. Reasonable effort
should be made to maintain releases of radioactivity in effluents to the general
environment as low as is reasonably achievable.
The location of a member of the public (MOP) is not defined clearly in the NRC statute. Under
DOE Order 435.1 the MOP is defined as someone who does not access the disposal facility, but
is located outside of the fence line or boundary of the facility. However, NRC does not similarly
define an MOP, unless the disposal facility is not considered part of the natural environment.
Otherwise, an MOP is not restricted other than through the activities in which the MOP might
engage.
In addition to addressing MOP, 10 CFR 61 requires additional assurance of protecting individuals
from the consequences of inadvertent intrusion. An inadvertent intruder is someone who is
exposed to waste without intent, and without realizing that exposure might occur (after loss of
institutional control). This is distinct from the intentional intruder, who might be interested in
deliberately disturbing the site, or extracting materials from it, or who might be driven by
curiosity or scientific interest. Intentional intruders are not evaluated in a PA.
§ 61.42 Protection of individuals from inadvertent intrusion. Design,
operation, and closure of the land disposal facility must ensure protection of any
individual inadvertently intruding into the disposal site and occupying the site or
contacting the waste at any time after active institutional controls over the
disposal site are removed.
The distinction between MOP and an inadvertent intruder is clear in DOE Order 435.1, but is not
as clear in NRC 10 CFR 61. Under DOE Orders, a MOP does not engage in activities within the
boundaries of the disposal facility, and an inadvertent intruder inadvertently accesses the waste
material directly. Consequently, the locations of MOP and intruder are different under DOE
Orders. However, the NRC indicates that an inadvertent intruder is defined as follows:
§ 61.2 Definitions. Inadvertent intruder means a person who might occupy the
disposal site after closure and engage in normal activities, such as agriculture,
dwelling construction, or other pursuits in which the person might be
unknowingly exposed to radiation from the waste.
Because of the remoteness of the Clive Facility and, hence, the types of activities in which
humans might engage, the distinction is made for this PA that ranchers, hunters and OHV
enthusiasts are inadvertent intruders because they “engage in normal activities, such as
agriculture, dwelling construction, or other pursuits in which the person might be unknowingly
exposed to radiation from the waste”. This facility is regulated under NRC, in which case the
definitions in 10 CFR 61 are most relevant. However, it is noted that the ranchers, hunters and
OHV enthusiasts do not intrude into the waste to create a direct exposure. Other receptors
evaluated in the PA Model who are located offsite are regarded as MOPs. The results of this
Model are calculated without regard for MOP and IHI categorization. The Model simply
evaluates dose to each receptor, providing the information necessary for comparison with
performance objectives.
Final Report for the Clive DU PA Model
24 November 2015 15
No dose limit is specified in 10 CFR 61 for the inadvertent intruder. However, since Part 61 has
been issued, the standard used by NRC and others for LLW disposal licensing has been an annual
dose of 500 mrem. The 500 mrem-in-a-year standard is also used in the DOE waste
determinations implementing the Part 61 performance objectives (NUREG-1854), and as part of
the license termination rule dose standard for intruders (10 CFR 20.1403).
The scope of a PA may be limited to the evaluation of MOP and inadvertent intrusion, and also to
the issue of site stability. The performance standard for stability requires the facility to be sited,
designed, and closed to achieve long-term stability to eliminate to the extent practicable the need
for ongoing active maintenance of the site following closure. The intent was to provide
reasonable assurance that long-term stability of the disposed waste and the disposal site will be
achieved. To help achieve stability, the NRC suggested to the extent practicable that disposed
waste should maintain gross physical properties and identity over 300 years, under the conditions
of disposal, with a further suggestion that the disposal facility should be evaluated for at least a
500-year time frame. About the same time as Part 61 was promulgated, the NRC also put in place
requirements for design of uranium mill tailings piles such as the Vitro site which is collocated
with the Clive Facility. The NRC specified that the design shall provide reasonable assurance of
control of radiological hazards to be effective for 1,000 years to the extent reasonably achievable,
and, in any case, for at least 200 years.
This raises the issue of appropriate compliance periods for a waste form that does not reach peak
radioactivity for more than 2 My. Section 2(a) of R313-25-9(5)(a) states:
For purposes of this performance assessment, the compliance period shall be a
minimum of 10,000 years. Additional simulations shall be performed for the
period where peak dose occurs and the results shall be analyzed qualitatively.
The intent of this Model, therefore, is to evaluate impacts to receptors for a period of 10,000
years, and long-term performance of the disposal system beyond that time. The regulation does
not address time frame for site stability. Given the long period of time before DU reaches secular
equilibrium, it is difficult to determine when peak dose might occur. Consequently, the Clive DU
PA Model has been implemented quantitatively for 10 ky, and has run additional simulations for
2.1 My, the time at which DU reaches peak activity. The results of the PA Model will be used to
inform decisions about the suitability of the Clive facility for disposal of DU waste, the amount
of DU waste that can be disposed safely, and different options for the engineered design and the
placement of the waste within the disposal system. These decisions will be made in light of the
doses to the receptors identified for the Model, groundwater concentrations of 99Tc and other
radionuclides, and the long-term effects on site stability and dispersal of DU waste in returning
lakes and lake sediment.
Site stability might also be considered to be a qualitative criterion for evaluating the concept of
maintaining receptor impacts to be “as low as reasonably achievable” (ALARA). However, the
10 CFR 20.1003 defines ALARA in the context of dose to populations. In addition, 10 CFR
61.42 states that "reasonable effort should be made to maintain releases of radioactivity in
effluents to the general environment as low as is reasonably achievable". The ALARA process is
described in more detail in the white paper Decision Analysis Methodology for Assessing ALARA
Collective Radiation Doses and Risks (Appendix 12). ALARA is evaluated in terms of
Final Report for the Clive DU PA Model
24 November 2015 16
population doses for the design options that are considered. This allows design options to be
compared, and, ultimately, to be optimized. . NRC (2015) suggests a value of $5,100 for the cost
per person rem, with a possible range of $3000 to $7500 (NRC 2015).
In addition to the radiological criteria, the State of Utah imposes limits on groundwater
contamination, as stated in the Ground Water Quality Discharge Permit (UWQB, 2010). Part
I.C.1 of the Permit specifies that GWPLs in Table 1A of the Permit shall be used for the Class A
LLW Cell. Table 1A in the Permit specifies general mass and radioactivity concentrations for
several constituents of interest to DU waste disposal. These GWPLs are derived from Ground
Water Quality Standards listed in UAC R317-6-2 Ground Water Quality Standards. Exceptions
to values in that table are provided for specific constituents in specific wells, tabulated in Table
1B of the Permit. This includes values for mass concentration of total uranium, radium, and gross
alpha and beta radioactivity concentrations for specific wells where background values were
found to be in exceedance of the Table 1A limits.
According to the Permit, groundwater at Clive is classified as Class IV, saline ground water,
according to UAC R317-6-3 Ground Water Classes, and is highly unlikely to serve as a future
water source. The underlying groundwater in the vicinity of the Clive site is of naturally poor
quality because of its high salinity and, as a consequence, is not suitable for most human uses,
and is not potable for humans. However, the Clive DU PA Model calculates estimates of
groundwater concentrations at a virtual well near the Federal DU cell for comparison with these
GWPLs. Part I.D.1 of the Permit specifies that the performance standard for radionuclides is 500
years.
1.4 Performance Assessment
Within the regulatory framework described above, a PA addresses doses to potential human
receptors within a time frame of compliance. The Clive DU PA Model also addresses
performance of the system for approximately 2.1 My—until secular equilibrium of 238U and its
decay products is reached. The PA process starts with the regulatory context but is itself a
decision support process. Decisions may be made based on the results of the PA modeling that is
performed. In the context of decision analysis, this requires steps that include:
1. State a problem,
2. Identify objectives (and measures of those objectives – i.e., attributes or criteria),
3. Identify decision alternatives or options,
4. Gather relevant information, decompose and model the problem (structure, uncertainty,
preferences),
5. Choose the “best” alternative (the option that maximizes the overall benefit),
6. Conduct uncertainty analysis, sensitivity analysis and value of information analysis to
determine if the decision should be made, or if more data/information should be collected
to reduce uncertainty and, hence, increase confidence in the decision, and
7. Go back (iterate) if more data/information are collected.
The problem addressed here is one of potential disposal of DU waste at the Clive Facility. The
objectives are to minimize risk to human health and the environment. Risk is measured in terms
of dose and uranium toxicity hazard to the human receptors that are identified for analysis. The
Final Report for the Clive DU PA Model
24 November 2015 17
decision options that are evaluated relate to different waste configuration options for DU waste
disposal. Given that context, the next step of the PA process is to gather information, and build a
PA model. There are several steps involved, each one building on the previous step. The
modeling process starts with evaluating features, events and processes (FEPs) that might be
important for evaluating performance, and using the FEPs analysis to build a conceptual site
model (CSM). These steps are described in full in the FEP Analysis for Disposal of Depleted
Uranium at the Clive Facility (Appendix 1), and the Conceptual Site Model for Disposal of
Depleted Uranium at the Clive Facility (Appendix 2).
Development of the CSM sets the stage for subsequent model structuring, which is the first step
needed to build the numerical model of the system. All relevant FEPs are captured in the model
structure, from waste inventory, mechanisms for transport through the engineered system,
migration through the natural environment to the accessible environment, to identification of
human receptors, exposure pathways and dose assessment. The model structure leads to
specification of the model. Probability distributions are specified for each input parameter. The
type of information available for each input parameter is highly variable, hence requiring varied
approaches for specification. Different methods that are used are described in the white paper
Development of Probability Distributions (Appendix 14).
Model structuring and specification completes the numerical model. The model is computed
using the GoldSim systems analysis software (GTG, 2010). GoldSim is probabilistic simulation
software that includes a graphical user-interface that is convenient for developing PA models.
GoldSim is inherently a systems-level software framework. The focus of a GoldSim model is on
the decision making process, which includes managing uncertainty and coupling all processes.
This PA model is intended to reflect the current state of knowledge with respect to the proposed
DU disposal, and to support environmental decision making in light of inherent uncertainties.
The development of the model is iterative, where the iterations depend on model evaluation,
which is performed at various levels. During model construction the model is evaluated
iteratively as new components are added. Once a complete model is assembled then the model is
subjected to uncertainty and sensitivity analysis. The goals of the uncertainty analysis are to
evaluate results against the performance objectives and to understand the values of the results
with respect to the model formation. The sensitivity analysis is used to identify components of
the model that are most influential on the output. This leads to model iteration as suggested in
Step 7 above.
Building a model to inform PA is a large undertaking. There are many intricacies that must be
accommodated starting with development of FEPs, moving through the CSM, mathematical
abstraction of environmental processes, numerical model structuring, development of probability
distributions for the input parameters, and model evaluation. This complex process is described
briefly in this document, and is described in more detail in the supporting documents (see
Appendices). In addition to complete documentation, the GoldSim model itself is fully contained,
with internal documentation of every aspect of the model structure. The extensive documentation
is provided for two reasons: The first is simply that it provides access to all information used in
the Model. This is done in the spirit of openness, transparency and, hence, defensibility. The
second is in the context of the quality assurance program that requires tracking of all information
Final Report for the Clive DU PA Model
24 November 2015 18
from its source through to the final model. The QA program implemented for this Model is
described in full in the Quality Assurance Project Plan (Appendix 17).
1.5 Technical Evolution of PA and PA Modeling
Since PA modeling began in the late 1970s through early 1990s at many of the radioactive waste
disposal facilities around the U.S., many different approaches to modeling have been used. These
approaches span the range from deterministic process-level modeling to probabilistic systems-
level modeling. Early PA models tended towards deterministic modeling for several reasons: 1)
PA modeling was initially performed with a focus on groundwater modeling, which was, and still
is, often performed using deterministic process-level models, 2) there were computational or
technological difficulties with taking a probabilistic approach, and 3) PA regulations and
guidance were established mostly with deterministic performance objectives, which was
interpreted as a reason for performing deterministic modeling. In particular, PA for low-level
radioactive waste (LLW) disposal facilities followed deterministic performance objectives.
However, the regulations for the Waste Isolation Pilot Plant and the Yucca Mountain Project
(YMP) (Title 40, Code of Federal Regulations (CFR), Part 191, “Environmental Radiation
Protection Standards for Management and Disposal of Spent Nuclear Fuel, High-Level and
Transuranic Radioactive Wastes,” and Title 40, CFR Part 197, “Public Health and Environmental
Radiation Protection Standards for Yucca Mountain, Nevada”) provide an exception to the
deterministic objectives, and consequently, PA models for these radioactive waste disposal
facilities have been developed probabilistically.
Technological advances in the last decade have also allowed more PA modeling to move towards
a probabilistic approach. Finally, PA modeling is multi-disciplinary, and as more technical
disciplines have been brought into PA modeling, there has been increased recognition of the
potential benefits of probabilistic systems-level modeling.
Systems-level models are usually computationally simpler than process-level models. However,
the systems-level PA model might still have large numbers of parameters, which reveals the
complexity of dealing with PA modeling even at a systems-level scale. The large number of
parameters is a consequence of the many constituents of concern that are usually included in PA
models, and the need to characterize transport properties for each of these constituents
(e.g., partitioning coefficients, solubility, plant uptake factors). However, it is unlikely that more
than a few of these parameters are important predictors for a given PA endpoint (e.g., dose to a
member of the public, groundwater protection levels). Along these lines, another advantage of
systems-level modeling performed in a probabilistic environment is the ability to identify
parameters that are most important or sensitive for a given endpoint. Because system-level
models may be probabilistic, global sensitivity analysis methods can be used to identify the most
sensitive parameters (see the white paper entitled Sensitivity Analysis in Appendix 15).
The advantages of system-level models are that they are capable of 1) coupling of different
processes without the need for the application of ad hoc boundary conditions, 2) using an
appropriate spatial and temporal scaling relative to the decisions that need to be made, 3) having
the ability to characterize and manage uncertainty through probabilistic modeling, and 4) being
used to perform global sensitivity analysis. Use of the global sensitivity analysis can potentially
lead to refinement and enhancements of the underlying models or the identification and collection
Final Report for the Clive DU PA Model
24 November 2015 19
of new data (e.g., research studies or monitoring) as necessary to reduce uncertainty of certain
parameters or variables. Use of a system-level model can also provide the ability to rapidly and
efficiently explore alternative conceptualizations of the system, which allows a greater ability to
address scenario and conceptual model uncertainties.
System-level models are often supported by process-level models. Each component of a system-
level model requires model building, which can include abstraction from a process-level model.
The purpose of the abstraction is to be able to capture the essence of the process-level model in
the probabilistic system-level model, so that its relative importance or sensitivity can be
evaluated. As a consequence of the development of system-level modeling frameworks such as
GoldSim, PA models are often developed following this approach, with global sensitivity
analysis driving iteration until the model results indicate a clear response and decision path.
1.6 Report Structure
The remainder of this report provides a more complete introduction to the PA modeling process
applied to the Clive DU waste disposal option, briefly describes the FEPs process, and follows
with a brief description of the CSM. The CSM description is aimed more at identifying
components of the model that might be significant in the model results. Model building always
leads to insights into the important components of a model, and that is conveyed in terms of
important aspects of the CSM.
The model structure is described prior to presentation of results, which are the main focus of this
report. Results are presented for the 10-ky quantitative model and for the deep-time model. For
the 10-ky model, the important results from a regulatory perspective include doses to the
receptors that have been identified as critical. Groundwater concentrations are evaluated for the
next 500 yrs. For the deep-time model, which models the performance of disposal of DU at Clive
for the next 2.1 My, results are presented in terms of lake water concentrations assuming the
return of a deep pluvial lake in the Bonneville Basin, and sediment concentrations that remain
after the pluvial lake recedes.
A summary is provided that includes further interpretation of results and comparison with
performance objectives. More complete documentation of the details of the model development
is contained in the Appendices, and also in the GoldSim model itself. This compendium of
documents provides a thorough treatise of the Clive DU PA Model v1.4.

Final Report for the Clive DU PA Model
24 November 2015 20
2.0 Introduction
The safe storage and disposal of DU waste is essential for mitigating releases of radioactive
materials and reducing exposures to humans and the environment. Currently, a radioactive waste
facility located in Clive, Utah and operated by EnergySolutions is proposed to receive and store
DU waste that has been declared surplus from radiological facilities across the nation. The Clive
Facility has been tasked with evaluating disposal of the DU waste in an economically feasible
manner that protects humans from future radiological releases.
To assess whether the Clive Facility location and containment technologies are suitable for
protection of human health, specific performance objectives for land disposal of radioactive
waste set forth in Title 10 Code of Federal Regulations Part 61 (10 CFR 61) Subpart C, and
promulgated by the Nuclear Regulatory Commission (NRC), must be met. In order to support the
required radiological PA, a model is needed to evaluate doses to human receptors that would
result from the disposal of DU and its associated radioactive contaminants.
This section provides an introduction to the general approach taken to developing version 1.4 of
the Clive DU PA Model. The focus is on methods that have been undertaken at each step along
the path, from description of the problem and the disposal facility under consideration, FEPs
identification, CSM development, approaches to numerical modeling and evaluation of results.
2.1 General Approach
Performance Assessment models are complex probabilistic systems-level models that evaluate
the long-term effects to human health and the environment of disposal of radioactive waste. The
approach includes the following steps:
1. Identification of disposal options – in this case use of the Federal DU cell at the Clive
Facility in Utah for disposal of DU waste, and specifics of the disposal configuration. This
includes consideration of the regulatory environment in which the PA model is to be
evaluated.
2. Identification of important FEPs that should be considered in the evaluation of the Clive
disposal facility. This includes identification of human receptors who might be engaged in
activities near or on the disposal facility.
3. Development of a CSM that captures the relevant FEPs. This includes evaluation of the
FEPs for the likelihood of occurrence and their consequence. If, for a given FEP the
likelihood of occurrence or consequence is considered too small, then the FEP is not
included in the CSM.
4. Development of a numerical or computational model for the PA. This translates the CSM
into numerical code for processing. This includes model structure and model specification.
The Clive DU PA Model is developed fully probabilistically, with coupling of all
processes included in the model.

Final Report for the Clive DU PA Model
24 November 2015 21
5. Model evaluation, including:
a. uncertainty analysis, which compares the probabilistic output to the performance
objectives,
b. sensitivity analysis, which is used to identify the important parameters or components
of the model in terms of prediction of the model output. This leads to model refinement
or data collection if the uncertainties in the decisions that need to be made are
considered to be too large.
6. Reporting of the PA model and its results, including:
a. doses to potential human receptors,
b. population doses evaluated in the context of ALARA,
c. groundwater concentrations at a specified location, and
d. deep time concentrations in lake water and lake sediment.
7. Quality Assurance.
A PA is a type of systematic (risk) analysis that addresses what can happen, how likely it is to
happen, what the resulting impacts are, and how these impacts compare to regulatory standards.
The essential elements of a performance assessment are
• a description of the site and engineered system,
• an understanding of events and processes likely to affect long-term facility performance,
• a description of processes controlling the movement of contaminants from waste sources
to the general environment,
• a computation of metrics reflecting system performance including concentrations, doses,
and other human health risk metrics to members of the general population, and
• an evaluation of uncertainties in the modeling results that support the assessment.
The role of PA in a regulatory context is often restricted to the narrow use of evaluating
compliance. In the present case, the Clive DU PA Model v1.4 can be used to evaluate
compliance—and inform a PA document that presents the argument that demonstrates
compliance—with 10 CFR 61 Subpart C and the corresponding provisions of the Utah
Administrative Code. In addition to that role, however, and because of the long-term nature of the
analysis, the intent of the Model is not to estimate actual long-term human health impacts or risks
from a closed facility. We believe that it is technically inappropriate to view the model results in
terms of actual long-term human health effects. The purpose of the Model is to provide a robust
analysis that can examine and identify the key elements and components of the site, the
engineered system, and the environmental setting that could contribute to potential long-term
impacts. Because of the time-scales of the analysis and the associated uncertainty in knowledge
of characteristics of the site, the waste inventory, the engineered system and its potential to
degrade over time, and changing environmental conditions, a critical part of the PA process is
also the consideration of uncertainty and evaluation of model and parameter sensitivity in
interpretation of PA modeling results.
Final Report for the Clive DU PA Model
24 November 2015 22
A probabilistic model includes a mathematical analysis of stochastic events or processes and their
consequences. Probabilistic analysis acknowledges that events and processes are inherently
uncertain, and hence involves characterization of uncertainty around expectation. Model output
hence is expressed with the same characteristics of expectation and uncertainty, which lends itself
to a global or probabilistic sensitivity analysis. Sensitivity analysis for probabilistic models is
used to identify the parameters (variables) that are the most important predictors of the output for
a given endpoint (e.g., dose to a resident, concentrations in groundwater). The important
predictors are those that explain most of the variability in the output variable of interest. Usually,
for a given endpoint of interest, this is no more than a handful of input or explanatory variables.
Because PA models are usually complex, dynamic, non-linear systems, these global sensitivity
analysis methods involve complex non-linear regression models that capture the impact of each
input variable across its specified range (range of its probability distribution).
Performance Assessment concerns modeling radioactive waste disposal facilities into the long-
term future. As such, PA models must address both the spatial and temporal magnitude of PA. It
is critical in a PA model to addresses the scale of the decisions that need to be made. Modeling is
performed at the spatial and temporal scale that is needed to support PA decisions related to
closure. In effect, system-level models might be fairly coarse, but this has advantages for
evaluating how the system evolves over time. For example, all processes involved are fully
coupled in the same model, probabilistic modeling can be performed to both characterize and
manage uncertainty, and statistics and decision analysis can be incorporated into the modeling
framework.
Results from a systems-level model are aimed at the decision objectives at the spatial and
temporal scales of interest. These results are presented as probability distributions for the
endpoints of interest (peak doses, concentrations, etc.), and comparisons are made with
performance objectives where appropriate (dose, groundwater concentrations).
Given the PA model construction with respect to the spatio-temporal scales of the model, there
are two levels of response. The first is for each hypothetical individual included in the model.
Dose results are available for each receptor in every year of the model, up to 10 ky. Each dose
result at this level represents individual doses resulting from the concentrations in various
exposure media predicted by the model at that time. The dose parameters, however, are specific
to the individual. This approach to modeling dose was taken for a few reasons: 1) There are not
many receptors at Clive, in which case, from a computational perspective it was feasible to
consider each individual receptor, and 2) this approach allows population dose to be estimated
directly from the individual doses.
Although individual peak doses are available in the model, the output of interest is the mean
dose. Traditionally this has been estimated as the mean dose to a hypothetical average individual.
With this model, the mean dose is estimated directly from the individual doses. Mean doses are
evaluated in each year of the model, but traditionally for PA, interest lies primarily in the worst
case year, in which case the peak mean dose across time is the metric of interest.
The effect is that average (mean) doses are available at multiple scales. Traditional comparison
with performance objectives is performed with the peak mean dose, meaning the highest mean
dose in a year across the 10-ky performance period. In this model, for which radioactivity is
Final Report for the Clive DU PA Model
24 November 2015 23
increasing with time for the DU waste, the greatest dose almost always occurs at 10 ky, and 500
years for groundwater concentrations. So peak mean dose results at 10 ky are presented. Note
that there will be 10,000 estimates of dose for each receptor if 10,000 realizations are run. This is
usually enough simulations to stabilize an estimate of the mean. The dose assessment model is
described in detail in the white paper entitled Dose Assessment (Appendix 11).
If the distribution of the peak of the means is treated as if each simulation result is independent,
then the 95th percentile of the distribution is somewhat analogous to the notion of a 95% upper
confidence interval that is commonly used under CERCLA. Comparisons may be made with the
PA performance objectives using the median, mean and 95th percentile of the output distribution
for each endpoint of interest.
For the ALARA analysis, the model is set up so that the population dose can be estimated for
each receptor class in each year of the model. The 10,000 realizations provide 10,000 estimates
of population dose in each year of the model. The population dose distribution can also be
processed to include the cost to human health and society by assigning a dollar value to
person-rem. This process is described in detail in the Decision Analysis white paper (Appendix
12).
Once the results are obtained and compared to the performance objectives, a global sensitivity
analysis is performed to identify the parameters that are the most influential in predicting each
endpoint of interest. Often this is only a handful of parameters for each endpoint. The results of
the sensitivity analysis can be used to determine if it might be useful to collect more data or
otherwise refine the model before making final decisions. This is ostensibly a decision analysis
task, which can be performed using the sensitivity analysis results as a basis for determining the
benefit of collecting new data. The potential benefits would be seen in reduction in uncertainty in
the model results. The sensitivity analysis methods used for this model are described in the white
paper entitled Sensitivity Analysis Methods (Appendix 15).
This holistic approach to PA modeling is aimed at providing insights into disposal system
performance. Although the model predicts or estimates doses to human receptors, among other
endpoints, the more important aspect of this type of modeling is to gain an understanding of how
the system might evolve over the time frames of interest, and to use this understanding to support
decision making including ability to safely dispose of waste and optimization of waste placement
within the disposal system.. No matter what doses are predicted, it is important to understand
why those modeled doses are observed, and hence, what are the important features of the disposal
system with regards to protection of human health and the environment.
2.2 General Facility Description
The EnergySolutions low-level radioactive waste disposal facility is west of the Cedar Mountains
in Clive, Utah, as shown in Figure 2. Clive is located along Interstate-80, approximately 5 km
(3 mi) south of the highway, in Tooele County. The facility is approximately 80 km (50 mi) east
of Wendover, Utah and approximately 100 km (60 mi) west of Salt Lake City, Utah. The facility
sits at an elevation of approximately 1302 m (4275 ft) above mean sea level (amsl). The Clive
Facility is adjacent to the above-ground disposal cell used for uranium mill tailings that were
Final Report for the Clive DU PA Model
24 November 2015 24
removed from the former Vitro Chemical company site in South Salt Lake City between 1984
and 1988 (Baird et al., 1990).
Currently, the Clive Facility receives waste shipped via truck and rail. Pending the findings of the
PA, DU waste will be stored in a permanent above-ground engineered disposal embankment that
is clay-lined with composite clay barriers and an ET cover. The disposal embankment is designed
to perform for a minimum of 500 years based on requirements of 10 CFR 61.7. The
EnergySolutions Clive Facility is divided into three main areas (Figure 2):
• the Bulk Waste Facility, including the Mixed Waste, Low Activity Radioactive Waste
(LARW), 11e.(2), and Class A LLW areas,
• the Containerized Waste Facility (CWF), located within the Class A LLW area, and
• the Treatment Facility (TF), located in the southeast corner of the Mixed Waste area.
The DU waste under consideration is proposed for disposal in the Federal DU cell. The terms
“cell” and “embankment” are here used interchangeably. That is, this Clive DU PA Model
considers only to the long-term performance of DU disposed in this waste cell. The Federal Cell
housing DU is next to the 11e.(2) cell, which is dedicated to the disposal of uranium processing
by-product waste and not considered in this analysis (Figure 2).
The general aspect of the Federal DU cell is that of a hipped cap, with relatively steeper sloping
sides nearer the edges. The upper part of the embankment, known as the top slope, has a
moderate slope, while the side slope is markedly steeper (20% as opposed to 2.4%). For this PA
Model, no waste is placed under the side slopes, in which case modeling focuses on waste placed
under the top slope. The embankment is also constructed such that a portion of it lies below-
grade. Details of the design of the embankment are contained in the white paper entitled
Embankment Modeling (Appendix 3).
DU waste from the Savannah River Site (SRS) and the gaseous diffusion plants (GDP) at
Portsmouth, Ohio and Paducah, Kentucky has been proposed for disposal at the Clive facility.
There are three categories of DU waste that are considered:
1. Depleted uranium oxide (UO3) waste from the Savannah River Site (SRS) proposed for
disposal at the Clive facility,
2. DU from the GDPs, which exists in two principal populations:
a) DU contaminated with fission and activation products from reactor returns introduced
to the diffusion cascades, and
b) DU consisting of only “clean” uranium, with no such contamination.

Final Report for the Clive DU PA Model
24 November 2015 25

Figure 2. Disposal and Treatment Facilities operated by EnergySolutions.

Final Report for the Clive DU PA Model
24 November 2015 26
The DU oxides that are to be produced at these sites “deconversion” plants will be primarily
U3O8. The contamination problem arises from the past practice of introducing irradiated nuclear
materials (reactor returns) into the isotopic separations process. Irradiated nuclear fuel underwent
a chemical separation process to remove the plutonium for use in nuclear weapons. Uranium,
then thought to be a rare substance, was also separated out, but contained some residual
contamination from activation and fission products. This uranium was again converted to UF6 for
re enrichment, and was introduced to the gaseous diffusion cascades, contaminating them and the
storage cylinders as well. Decay products (226Ra), activation products (241Am, 237Np, 238Pu, 239Pu,
240Pu, 241Pu, 242Pu), and fission products (90Sr, 99Tc, 129I, 137Cs) potentially contaminate the DU
waste. The proposed inventory that is evaluated in the Model is described fully in the white paper
entitled Waste Inventory (Appendix 4).
3.0 Features, Events and Processes
The conceptual site model (CSM) describes the physical, chemical, and biological characteristics
of the Clive facility. The CSM, therefore, encompasses everything from the inventory of disposed
wastes, the migration of radionuclides contained in the waste through the engineered and natural
systems, and the exposure and radiation doses to hypothetical future humans. These site
characteristics are used to define variables for the quantitative PA model that is used to provide
insights and understanding of the future potential human radiation doses from the disposal of DU
waste.
The content of the CSM informs the Model with respect to regional and site-specific features,
events and processes, such as climate, groundwater, and human receptor scenarios. The CSM
accounts for and defines relevant features, events, and processes (FEPs) at the site, materials and
their properties, interrelationships, and boundaries. These constitute the basis of the Model, on
which, or through which, radionuclides are transported to locations where receptors might be
exposed.
A key activity in developing a PA for a radiological waste repository is the comprehensive
identification of relevant external factors that should be included in quantitative analyses. These
factors, termed “features, events, and processes” (FEPs), form the basis for scenarios that are
evaluated to assess site performance.
The universe of FEPs that were screened and identified as relevant for the Clive Facility PA are
documented in the white paper entitled FEP Analysis for Disposal of Depleted Uranium at the
Clive Facility (Appendix 1) and further elaborated in the CSM document (Conceptual Site Model
for Disposal of Depleted Uranium at the Clive Facility – Appendix 2).
4.0 Conceptual Site Model
The important components of the conceptual site model are described in the following sections.
Details are contained in the white paper entitled Conceptual Site Model for Disposal of Depleted
Uranium at the Clive Facility (Appendix 2).
Final Report for the Clive DU PA Model
24 November 2015 27
4.1.1 Disposal Site Location
EnergySolutions operates a low-level radioactive waste disposal facility west of the Cedar
Mountains in Clive, Utah, as shown in Figure 1. Clive is located along Interstate-80,
approximately 5 km (3 mi) south of the highway, in Tooele County. The facility is approximately
80 km (50 mi) east of Wendover, Utah and approximately 100 km (60 mi) west of Salt Lake City,
Utah. The facility sits at an elevation of approximately 1,302 m (4,275 ft) above mean sea level
(amsl) and is accessed by both highway and rail transportation. The Clive Facility is adjacent to
the above-ground disposal cell used for uranium mill tailings that were removed from the former
Vitro Chemical company site in South Salt Lake City between 1984 and 1988 (Baird et al.,
1990).
4.1.2 Disposal Site Description
Currently, the Clive Facility receives waste shipped via truck and rail. DU waste is proposed for
disposal in a permanent above-ground engineered disposal embankment that is clay-lined with
clay barriers and an ET cover. The disposal embankment is designed to perform for a minimum
of 500 years based on requirements of 10 CFR 61.7, which provides a long-term disposal
solution with minimal need for active maintenance after site closure. More detail relating to the
properties of the disposal embankment is provided in Section 0.
The EnergySolutions Clive Facility is divided into three main areas (Figure 2): the Bulk Waste
Facility, including the Mixed Waste, Low Activity Radioactive Waste (LARW), 11e.(2), and
Class A LLW areas, the Containerized Waste Facility (CWF), located within the Class A LLW
area, and the Treatment Facility (TF), located in the southeast corner of the Mixed Waste area.
This analysis considers only the Federal DU cell.
4.1.2.1 Embankment
Depleted uranium waste is proposed for disposal in the Federal DU cell. The Federal DU Cell is
about 541 × 402 m (1,775 × 1,318 ft), with an area of approximately 22 ha (54 acres), and an
estimated total waste volume of about 2.0 million m3 (71 million ft3). A drainage ditch surrounds
the disposal cell. The cell is constructed on top of a compacted clay liner covered by a protective
cover. Waste will be placed above the liner and will be covered with a layered engineered cover
constructed of natural materials. The top slopes will be finished at a 2.4% grade while the side
slopes will be no steeper than 5:1 (20% grade).
The design of Federal DU cell cover has been engineered to discourage erosion, reduce the
effects of infiltration, and to protect workers and the public from radionuclide exposure. The cell
cover consists of layers including two clay radon barriers, a frost protection layer, an evaporative
zone layer, and a surface layer. The clay radon barriers are designed to minimize infiltration of
precipitation and runoff and reduce the migration of radon from the waste cell. The detailed
properties of each cell layer may be found in the white paper on Unsaturated Zone Modeling
(Appendix 5).
Final Report for the Clive DU PA Model
24 November 2015 28
4.1.2.2 Waste Inventory
The waste inventory is limited to the disposal of DU wastes of two general waste types:
1) depleted uranium trioxide (DUO3) waste from the Savannah River Site (SRS) and 2)
anticipated DU waste as U3O8 from gaseous diffusion plants (GDPs) at Portsmouth, Ohio and
Paducah, Kentucky. The quantity and characteristics of DU waste from other sources that has that
already been disposed of at the Clive Facility was not included. A full list of radionuclides has
been established for the PA modeling effort. The radionuclide species list was based upon
process knowledge, radionuclides analyzed for (though not necessarily detected) in the DU waste
material, and decay products with half-lives over five years. The species list consists of the
following radionuclides:
fission products:
Sr-90, Tc-99, I-129, Cs-137
progeny of uranium isotopes:
Pb-210, Rn-222, Ra-226, -228, Ac-227, Th-228, -229, -230, -232, Pa-231
uranium isotopes:
U-232, -233, -234, -235, -236, -238
transuranic radionuclides:
Np-237, Pu-239, -239, -240, -241, -242, Am-241
The waste inventory is discussed in more detail in the Waste Inventory white paper (Appendix 4)
and in the Conceptual Site Model white paper (Appendix 1).
4.1.2.3 Climate
The following sections briefly describe the aspects of the regional climate that influence the
performance of the site and engineered features. Further details are provided in the Conceptual
Site Model white paper (Appendix 1), and in the Unsaturated Zone Modeling white paper
(Appendix 5). In general the climate is dry, with evapotranspiration potential that exceeds
precipitation on an annual basis. This leads to low infiltration rates, and subsequent relatively
slow movement of radionuclides to groundwater. Also, the embankment is largely above grade,
and the dry, sometimes windy, environment could lead to drying out of the embankment beyond
what is considered in typical unsaturated zone models.
4.1.2.3.1 Temperature
Regional climate is regulated by the surrounding mountain ranges, which restrict movement of
weather systems in the vicinity of the Clive facility. The most influential feature affecting
regional climate is the presence of the Great Salt Lake, which can moderate downwind
temperatures since it never freezes (NRC, 1993). The climatic conditions at the Clive Facility are
characterized by hot and dry summers, cool springs and falls, and moderately cold winters (NRC,
1993). Frequent invasions of cold air are restricted by the mountain ranges in the area. Data from
the Clive Facility from 1992 through 2009 indicate that monthly temperatures range from
about -2°C (29°F) in December to 26°C (78°F) in July (Whetstone, 2006).
Final Report for the Clive DU PA Model
24 November 2015 29
4.1.2.3.2 Precipitation
The Clive Facility is characterized as being an arid to semi-arid environment where evaporation
greatly exceeds annual precipitation (Adrian Brown, 1997). Data collected at the Clive Facility
from 1992 through 2004 indicate that average annual rainfall is on the order of 22 cm (8.6 in) per
year (Whetstone, 2006). Precipitation generally reaches a maximum in the spring (1992-2004
monthly average of 3.2 cm [1.25 in] in April), when storms from the Pacific Ocean are strong
enough to move over the mountains (NRC, 1993; Whetstone, 2006). Precipitation is generally
lighter during the summer and fall months (1992-2004 monthly average of 0.8 cm [0.32 in] in
August) with snowfall occurring during the winter months (Whetstone, 2006; NRC, 1993; Baird
et al., 1990).
4.1.2.3.3 Evaporation
Because of warm temperatures and low relative humidity, the Clive Facility is located in an area
of high evaporation rates. NRC (1993) indicates that average annual pond evaporation rate at the
Clive Facility is 150 cm/yr (59 in/yr), with the highest evaporation rates between the months of
May and October. Previous modeling studies indicate that the Dugway climatological station
nearby is comparable to the Clive site with respect to evaporation and have reported pan-
evaporation estimates of 183 cm/yr (72 in/yr), which is considerably greater than average annual
rainfall (Adrian Brown, 1997). Because of the high evaporation rate, the amount of groundwater
recharge due to precipitation is likely very small.
4.1.2.4 Unsaturated Zone
The engineered features of the landfill, including cap, waste, and liner, are all in the unsaturated
zone (UZ), at least within the 10,000-yr duration of the quantitative model. The part of the UZ
that extends from the bottom of the cell liner to the water table consists of naturally-occurring
lake sediments from the ancestral Lake Bonneville.
Diffusion in the water phase may also play a role in the transport of waterborne contaminants in
the UZ, since the advective flux is expected to be small. The concentration gradients in the UZ
are also expected to be predominantly vertical, so diffusion will also occur in the vertical
direction, oriented with the column of cells.
Diffusion in the air phase within the UZ below the facility will not be modeled, since the only
diffusive species would be radon, which is of greater concern at the ground surface. Upward
radon diffusion to the ground surface will be dominated by radon parents in the waste zone, and
is modeled within the engineered cap. Unsaturated zone processes, material properties, and
parameters represented in the PA model are described in detail in the Unsaturated Zone Modeling
White Paper. The primary concerns for the PA are movement through the unsaturated zone of
mobile radionuclides, such as 90Sr, 99Tc, and 129I to groundwater and the upward diffusive
movement of radon.
4.1.2.4.1 Infiltration
Recharge is an important process in controlling the release of contaminants to the groundwater
pathway. Site characteristics influencing movement of water from precipitation through the
Final Report for the Clive DU PA Model
24 November 2015 30
vadose zone to the water table at the Clive site include climate, soil characteristics, and native
vegetation. Engineered barriers are used at the Clive site to control the flow of water into the
waste. A hydrologic model of the waste disposal system must realistically represent precipitation,
the source of water to the system, runoff, evaporation, transpiration, and changes in storage to
estimate the flow through the system. Under natural conditions plants remove water from the
upper soil zone through root uptake and transpiration reducing the water available for seepage
deeper into the profile. The same processes occur in an engineered cover layer that has been
revegetated. Seepage through a cover system can occur when soils become wet enough to
increase their conductivity to water. Cover surface layers with adequate storage capacity can hold
the water in the near surface until it can move back into the atmosphere through evaporation
reducing the seepage of water to the waste.
Steady-state water infiltration rates and water contents for the cover layers required as input for
the Clive DU PA GoldSim model were calculated from a regression model developed from
infiltration modeling using the HYDRUS-1D software package. This section describes the
development of HYDRUS-1D models for the Clive DU PA model and the abstraction of the
HYDRUS-1D results into the probabilistic framework employed by GoldSim. The HYDRUS-1D
model (Šimůnek et al., 2009) was selected for simulating the performance of the ET cover
proposed for the DU waste cell because of its ability to simulate processes known to have a
significant role in water flow in landfill covers in arid regions.
The one-dimensional version of the software rather than two-dimensional version was selected
for simulating flow in the Federal DU cell ET cover since previous numerical modeling of flow
in the similar ET cover design for the Class A West cover demonstrated that subsurface lateral
flow was not significant (EnergySolutions, 2012). To test the importance of 2-D flow effects in
the ET cover design 2-D transient flow simulations were conducted for representative sections of
the cover. The approach taken was to model a section of the side slope in two-dimensions.
Representative hydraulic properties were assigned to the ET cover layers and the models were
run with daily atmospheric boundary conditions for 100 years. Root water uptake was modeled
assuming the roots extended to the bottom of the evaporative zone layer and a rooting density
that decreased with depth. The results of these 2-D simulations demonstrated that water flow in
the cover system for both designs is predominantly vertical with no significant horizontal
component. These results demonstrate that 1-D models can be used to provide a defensible
analysis of cover performance for the ET cover design due to the lack of lateral flow.
Model development requires construction of a computational grid based on the geometry of the
model domain. Hydraulic properties for each layer required for the model were available from
previous studies at the site or were estimated from site-specific measurements such as particle
size distributions. Some of the hydraulic properties were variable in this modeling as described
below. HYDRUS requires daily values of precipitation, potential evaporation, and potential
transpiration to represent the time-variable boundary conditions on the upper surface of the cover.
Representative boundary conditions were developed from records of nearby meteorological
observations. Parameters for describing root water uptake were available from the literature.
The process of abstracting a detailed flow model into a probabilistic model that could be
implemented in GoldSim required the development of distributions for hydraulic property
parameters for the cover materials that influence water balance. Included in the distributions used
Final Report for the Clive DU PA Model
24 November 2015 31
was a distribution for the saturated hydraulic conductivity (Ks) of the radon barriers for the
modeling. This distribution included values from a range of in-service (“naturalized”) clay barrier
Ks values described by Benson et al. (2011, Section 6.4, p. 6-12). Multiple HYDRUS-1D
simulations with varying hydraulic property inputs were conducted to provide values of
infiltration flux into the waste zone, and water content within each ET cover layer as a function
of hydraulic property parameter values. From these simulation results a statistical model was
developed that related values of hydraulic properties from the statistical distributions to values of
infiltration flux and cover layer water content. This statistical model was then implemented in
Clive DU PA model to provide for each realization a steady-state infiltration flux and layer water
contents that included the uncertainty in these parameters. The ET cover and unsaturated zone
infiltration modeling approaches and results are described in detail in the Unsaturated Zone
Modeling white paper (Appendix 5).
4.1.2.5 Geochemical
The conceptual model for the transport of radionuclides at the Clive Facility allows sufficient
meteoric water infiltration into the waste zone to allow dissolution of uranium and daughters,
fission products and potential transuranic contaminants (along with native soluble minerals). At
first, leaching is likely to be solubility-limited with respect to uranium, and the leachate will
migrate away from the source with the uranium concentration at the solubility limit. The other
radionuclides are unlikely to be at a solubility limit. Depending upon the amount of water
available, these radionuclides will either re-precipitate, once the thermodynamic conditions for
saturation are reached, or remain in solution and be transported to the saturated zone. This water
is expected to be oxidizing, with circum-neutral to slightly alkaline pH (similar to the upper
unconfined aquifer), and an atmospheric partial pressure of carbon dioxide. However, the amount
of total dissolved solids (TDS) is expected to be initially lower than the upper aquifer.
The composition of this aqueous phase will change as it reaches the saturated zone, with some
increase in dissolved solids and potentially lower dissolved oxygen and carbon dioxide. The
saturated zone for this PA model includes only the shallow, unconfined aquifer. Transport of
radionuclides is expected to be restricted to this aquifer and not migrate to the lower aquifer due
to a natural upward gradient at the facility. The chemical composition of the saturated zone is
characterized as somewhat alkaline pH likely due to the presence of carbonates, mainly oxidizing
though transient reduced conditions may exist, with high levels of dissolved ions of mainly
sodium and chlorine.
The transport of dissolved radionuclides can also be limited by sorption onto the solid phase of
associated minerals and soils within each of the zones considered in this PA model. The transport
of uranium is limited by both solubility and the sorption of radionuclides in groundwater.
Sorption consists of several physicochemical processes including ion exchange, adsorption, and
chemisorption. Sorption is represented in the PA model as a partitioning coefficient (Kd) value.
Distributions of radionuclide-specific partitioning coefficients and solubilities were developed for
the PA model considering the geochemical conditions in the cell, the unsaturated zone, and the
shallow aquifer at the Clive facility. The development of these distributions is described in detail
in the Geochemical Modeling white paper (Appendix 6). The primary concerns for the model
Final Report for the Clive DU PA Model
24 November 2015 32
include the geochemical properties of 99Tc as they affect movement to groundwater and of
uranium in its different chemical forms for the 10-ky and deep-time models.
4.1.2.6 Saturated Zone
Contaminants moving vertically in the UZ below the cell enter the saturated zone (SZ) beneath
the disposal facility. The rate of recharge is the same as the Darcy flux (the rate of volume flow
of water per unit area) through the overlying UZ, and is expected to be small enough that vertical
transport within the SZ would be small. Most SZ waterborne contaminant transport will be in the
horizontal direction, following the local pressure gradients, which are reflected in water table
elevations in the shallow aquifer. A point of compliance in the groundwater has been established
at 27 m (90 ft) from the edge of the embankment interior, so saturated transport is modeled to
that point. Note that in the case of the proposed DU waste disposal, only the top slope section of
the embankment would contain DU waste, so the effective distance from the DU waste to the
well is lengthened by the width of the side slope section, to about 73 m (240 ft).
Saturated zone groundwater transport generally involves the processes of advection-dispersion
and diffusion. Mean pore water velocity in the saturated zone is assumed to be determined by the
Darcy flux and the porosity of the sediment. A range of values will allow the sensitivity analysis
(SA) to determine if this is a sensitive parameter in the determination of concentrations at the
compliance well and resultant potential doses. Modeling of fate and transport for the saturated
zone pathway will include advection, linear sorption, mechanical dispersion, and molecular
diffusion. Saturated zone processes and parameters represented in the PA model are described in
detail in the Saturated Zone Modeling white paper (Appendix 7). The primary concern for the
model is the breakthrough of 99Tc at the monitoring well.
4.1.2.7 Air Modeling
Gaseous and particle-bound contaminants that have migrated to the surface soil layer are
potentially subject to dispersion in the atmosphere. The effect of mechanical disturbance on
human exposure to soil particulates is evaluated in the PA based on the effect of off-highway
vehicle (OHV) use. However, although this mechanism may be consequential for human
exposure, it is not likely to be a significant contributor to the overall rate of fine particulates
emissions from the embankment over time. Eolian (wind-related) disturbance is the primary
cause of particulates emissions from the embankment and is the process modeled in the PA to
estimate particulate emissions.
In addition to particulate emissions of contaminated surface soil due to eolian erosion, emissions
of gas-phase radionuclides diffusing across the surface of the embankment into the atmosphere
are considered in the PA model. Note that this effect is counter-balanced by replacement with
eolian material that moves onto the cap. Diffusion modeling of radionuclide gases in the
embankment, and estimation of flux into the atmosphere, is described in the Radon white paper
(Appendix 18). For both particulate-bound and gaseous radionuclides, atmospheric dispersion
modeling employing local meteorological data is conducted to calculate breathing-zone air
concentrations above the embankment and at specific locations in the area where off-site
receptors may be exposed (see Dose Assessment white paper – Appendix 11).
Final Report for the Clive DU PA Model
24 November 2015 33
Atmospheric dispersion may result in significant bulk transport of fine particles modeling off of
the embankment. Atmospheric dispersion modeling is also used to calculate the deposition flux
of resuspended embankment particles in the areas adjacent to the embankment where ranchers
and recreational receptors may be exposed. As particulates from the embankment are deposited
on surrounding land, this surrounding area may become a secondary source of radionuclide
exposure.
Atmospheric dispersion modeling was conducted outside of the GoldSim modeling environment,
into which the model was abstracted. An atmospheric dispersion model is a mathematical model
that employs meteorological and terrain elevation data, in conjunction with information on the
release of contamination from a source, to calculate breathing-zone air concentrations at locations
above or downwind of the release. Some models may also be used to calculate surface deposition
rates of contamination at locations downwind of the release.
Both particle resuspension and atmospheric dispersion are first modeled outside of the GoldSim
PA model, and the results are then incorporated into GoldSim. The particulate emission model
used is a relatively simple model that has been adopted by EPA to estimate an annual-average
emission rate of respirable particulates (approximately 10 µm and less, i.e., PM10) from the
ground surface. The air dispersion model used is AERMOD, which is EPA’s recommended
regulatory air modeling system for steady-state releases and suitable for calculating annual-
average contaminant breathing zone air concentrations at various distances and in various
directions from a source release. These models are described in detail in the Atmospheric
Transport Modeling white paper (Appendix 8). Given the massive dilution that occurs for
windblown sediments, it is unlikely that this pathway will result in offsite accumulation of large
amounts of transported radionuclides. Accumulation onsite is more likely.
4.1.2.8 Biologically Induced Transport
Biological organisms play an important role in soil mixing processes, and therefore are
potentially important mediators of transport of buried wastes from deeper layers to shallower
layers or the soil surface. Three broad categories are evaluated for their potential effect on the
redistribution of radionuclides at the Clive facility: plants, ants, and burrowing mammals. The
impact of these flora and fauna will be limited largely to the top several meters, as their potential
influence as contaminant transport mechanisms is greater in the cover layers than in the
underlying waste, although contaminant concentrations are lower in the cover layers. Details for
all three categories can be found in the Biological Modeling white paper (Appendix 9).
4.1.2.8.1 Plants
Biotic fate and transport models have been developed to evaluate the redistribution of soils, and
contaminants within the soil, by native flora and fauna. The Clive Facility is located in the
eastern side of the Great Salt Lake Desert, with flora and fauna characteristic of Great Basin
alkali flat and Great Basin desert shrub communities.
Plant-induced transport of contaminants is assumed to proceed by absorption of contaminants
into the plants roots, followed by redistribution throughout all the tissues of the plant, both above
ground and below ground. Upon senescence, the above-ground plant parts are incorporated into
surface soils, and the roots are incorporated into soils at their respective depths.
Final Report for the Clive DU PA Model
24 November 2015 34
Functional factors that contribute to the plant section of the biotic transport model include
identifying dominant plant species, grouping plant species into categories that are significantly
similar in form and function with respect to the transport processes, estimating net annual
primary productivity (NAPP), a measure of combined above-ground and below-ground biomass
generation), determining relative abundance of plants or plant groups, evaluating root/shoot mass
ratios, and representing the density of plant roots as a function of depth below the ground surface.
Field surveys of the Clive site and surrounding areas were conducted by SWCA Environmental
Consultants in September and December 2010 to identify plant species present in different
vegetative associations around the Clive Site (SWCA, 2011). Five different vegetative
associations were surveyed, with three associations representing the alkali flat/desert flat type
soils found in the vicinity of Clive, and two associations representative of desert scrub/shrub-
steppe habitat characteristic of slopes and slightly higher elevations with less-saline soil
chemistry. A one hectare (100 m × 100 m) plot was established in each vegetative association,
and each plot was surveyed for dominant plant species present, and the percent cover and density
of each species. In addition, a small number of black greasewood, shadscale, halogeton, and
Mojave seablite plants were excavated to obtain root profile measurements and above-ground
plant dimensions. Plots 3 through 5 represent current vegetation at the Clive site, while Plots 1
and 2 are representative of less-saline soils that may develop on top of the waste cell cover.
A total of 41 plant species were identified on the five survey plots. Eighteen species each
comprised at least 1% of the total cover on at least one plot. These 18 species were considered the
most important for the purpose of modeling plant mediated transport of radiochemical
contaminants at Clive. Species were grouped into five functional plant groups: grasses, forbs,
greasewood, other shrubs, and trees. Greasewood is separated from other shrubs because of its
status as a phreatophyte that can extend taproots in excess of five meters to reach groundwater.
Annual and perennial grasses were grouped due to similar maximum rooting depths. Despite the
ability of Greasewood to extend taproots, it will only do so if there is a water source to mine.
There is no evidence in the Clive data that greasewood in the area of Clive extends to the water
table. Also, the radon barrier acts as an impediment to deep rooting. Consequently, plant
pathways for radionuclide transport are likely to have a limited effect in the current model.
4.1.2.8.2 Ants
Ants fill a broad ecological niche in arid ecosystems as predators, scavengers, trophobionts and
granivores. However, it is their role as burrowers that is of main concern for the purposes of this
model. Ants burrow for a variety of reasons but mostly for the procurement of shelter, the rearing
of young and the storage of foodstuffs. How and where ant nests are constructed plays a role in
quantifying the amount and rate of subsurface soil transport to the ground surface at the Clive
site. Factors relating to the physical construction of the nests, including the size, shape, and depth
of the nest, are key to quantifying excavation volumes. Factors limiting the abundance and
distribution of ant nests such as the abundance and distribution of plant species, and intra-specific
or inter-specific competitors, also can affect excavated soil volumes. Important parameters
related to ant burrowing activities include nest area, nest depth, rate of new nest additions,
excavation volume, excavation rates, colony density, and colony lifespan.
Final Report for the Clive DU PA Model
24 November 2015 35
Modeling soil and contaminant transport by ant species assumes that ants move materials from
lower cells to those cells above while excavating chambers and tunnels within a nest. These
chambers and tunnels are assumed to collapse over time and return soil from upper cells back to
lower cells.
Surveys for ants at Clive were limited to surface surveys of ant colonies, including identification
of ant species, measurements (length, width, and height) of ant mounds, and determination of ant
nest densities in each vegetative association (SWCA, 2011). No excavations of ant nests were
performed at Clive to support this initial PA model, although excavations could be conducted to
support future model iterations if ant nest depth and volume are found to be sensitive parameters.
Total nest depth and nest volume were extrapolated from mound surface dimensions based on
correlations from data observed at the Nevada National Security Site (NNSS) (Neptune 2006) for
the dominant ant species at Clive. Only two species of ants were identified during the surveys,
with the western harvester ant, Pogonomyrmex occidentalis, accounting for 62 of the 64 nests
identified. The second ant species, a member of the genus Lasius, was only encountered twice,
both times in the mixed grassland plot. Harvester ants also tend to create the largest and deepest
burrows. Consequently, the characteristics of the harvester ants were included in the model. For
details of biological models, refer to the Biological Modeling white paper. Although the effect of
burrowing ants is modeled, it is not expected to have a large influence on model results because
ant nests will not penetrate to the waste layer, which is about 5m or more below ground surface
for the disposal configurations considered. This is based on site-specific investigations indicating
most ant burrowing will occur in the upper layers of the cover and be minimal below a depth of
42 inches (SWCA 2013, p.28).
4.1.2.8.3 Burrowing Mammals
Burrowing mammals can have a profound impact on the distribution of soil and its contents near
the soil surface. The degree to which mammals influence soil structure is dependent on the
behavioral habits of individual species. While some species account for a large volume of soil
displacement, others are less influential. Functional factors such as burrowing depth, burrow
depth distributions, percent burrow by depth, tunnel cross-section dimension, tunnel lengths, soil
displacement by weight, soil displacement by volume and animal density per hectare play a
critical role in determining the final soil constituent mass by depth within the soil.
Modeling soil and contaminant transport by mammal species within the Clive PA model assumes
animals move materials from lower cells to those cells above while excavating burrows. Burrows
are assumed to collapse over time and return soil from upper cells back to lower cells. Thus, the
balance of materials is preserved through time.
Each Clive plot was surveyed for small mammal burrows during September and October 2010
(SWCA 2011). Burrows were identified by animal category. Within the survey area four
categories of mammal burrows were identified: ground squirrels, kangaroo rats, mouse/rats/voles,
and one badger. Due to the small number of badger and ground squirrel burrows, the decision
was made to treat all burrowing mammals as a single unit for modeling purposes. Small mammal
trapping was conducted on the five Clive plots during the new moon in October 2010 to identify
the principal small mammal fauna present in each vegetative association. Each 1-ha plot was
Final Report for the Clive DU PA Model
24 November 2015 36
subdivided into 25 20-m × 20-m subplots. At the center of the each subplot, two Sherman® live
traps were placed, for a total of 50 traps per plot.
Deer mice (Peromyscus maniculatus) were the most abundant small mammal captured during
trapping, and were the only mammal captured in the plots located on the Clive Facility (Plots 3,
4, and 5). Plots 3, 4, and 5 were characterized by very low mammal densities, as evidenced by
both the trapping results and the burrow surveys. With such a small population in plots 3, 4, and
5, the decision was made to average these plots. While the surface layer materials for the cap of
the Clive embankment may be conducive to the development of mammal burrows, the burrows
are sufficiently shallow that it is unlikely that they will have a significant impact on radionuclide
transport, and hence on doses to human receptors.
4.1.2.9 Erosion
The Federal DU cell is subject to erosion by the forces of wind and water. The conceptual model
assumes that wind blows material off-site (see Section 4.1.2.7), even while it replaces material
that is removed from the cap. Water removes cap material through sheet erosion and the
formation of channels (gullies).
Once an initiating event has occurred, wherein a “nick” is formed in the surface of the cover (by
natural or anthropogenic events), gully formation follows from water flowing in narrow channels,
particularly during heavy rainfall events. Gully erosion typically results in a gully that has an
approximate “V” cross section which widens (lateral growth) and deepens (vertical growth)
through time until the gully stabilizes. The formation of gullies is a concern on uranium mill
tailings sites and other long-term above-ground radioactive waste sites (NRC 2010). Gully
erosion has the potential to move substantial quantities of both cover materials and waste, should
the waste material be buried close to the surface. Gully outwash forms depositional fans on the
slopes of the embankment. Gullies might form initially on the embankment through disturbance
attributed to animal burrowing, or by human induced mechanisms such as cattle paths or OHV
tracks.
Two approaches have been used in the Clive DU PA model to evaluate the influence of erosion
on embankment performance. The first is a screening gully model that was applied in version 1.0
of the Clive DU PA model. The current approach used in the Clive DU PA model to evaluate the
influence of erosion on embankment performance is to apply results from a landscape evolution
model of a borrow pit area at the Clive Site as an analogue for embankment cover erosion.
Assumptions for this approach include:
• The geometry of the borrow pit wall and upslope area are sufficiently similar to that of the
embankment top slope and side slope.
• The borrow pit materials (unit 4) are sufficiently similar to the layers of the embankment
(unit 4 with gravel, unit 4, and radon barrier clays).
• Surface elevation changes at 10,000 years can be extrapolated from SIBERIA model
results from 100 yr, 500 yr and 1000 yr.
• The results at 10,000 years approximate steady state of gullies. This steady state situation
is implemented from time zero in this model.
Final Report for the Clive DU PA Model
24 November 2015 37
• The area of waste that is deposited on the fan is the same as the area of waste exposed in
the gullies, using projections onto the horizontal plane.
• The excavation of ET Cover cells was not considered in the calculations below for
contaminants in the excavated mass from the gully because it was assumed that
significantly more contaminant mass was in the waste than in the cap and that the material
extracted from the waste layers would be on the top of the fan.

Implementation of this approach in GoldSim is described in more detail in the Erosion Modeling
white paper (Appendix 10).
4.1.2.10 Dose Assessment
The dose assessment in the Model addresses potential radiation dose to any receptor who may
come in contact with radioactivity released from the disposal facility into the general
environment (10 CFR 61.41).The objective of a dose assessment in a radiological PA is to
provide estimates of potential doses to humans over time from radioactive releases from a
disposal facility after closure, as described in Section 3.3.7 of NRC (2000 – NUREG 1573). As
described below, the critical groups in the Model are defined as ranchers and recreationalists.
The radiation dose limit for protection of the general population is 25 mrem/yr, as a total
effective dose equivalent (TEDE). Dose limits for radiological PAs are defined in UAC Rule
R313-25-20 and10 CFR 61.41as an equivalent of 0.25 mSv (25 mrem) to the whole body,
0.75 mSv (75 mrem) to the thyroid, and 0.25 mSv (25 mrem) to any other organ of any member
of the public. However, the radiation dosimetry underlying these dose metrics is based on a
methodology published by the International Commission on Radiation Protection (ICRP) in
1959. More recent dose assessment methodology has been published as ICRP Publication 30
(ICRP, 1979) and ICRP Publication 56 (ICRP, 1989), employing the TEDE approach. As stated
in Section 3.3.7.1.2 of NRC (2000), “As a matter of policy, the Commission considers 0.25
mSv/year (25 mrem/year) TEDE as the appropriate dose limit to compare with the range of
potential doses represented by the older limits…”
The period of performance for a radiological PA defined in UAC Rule R313-25-9 requires
evaluation for a minimum compliance period of 10 ky, with additional simulations for a
qualitative analysis for the period where peak hypothetical dose occurs. The scope of this Model
includes modeling of the disposal system performance to the time of peak hypothetical
radiological dose (or peak radioactivity, as a proxy), and to quantify dose within the time frame
of 10 ky.
4.1.2.10.1 Receptors and Exposure Scenarios
Receptors in a PA are categorized in UAC Rule R313-25-20 and -21 and10 CFR 61.41 according
to the labels “member of the public” (MOP) and “inadvertent human intruder” (IHI). The
regulatory basis for, and interpretation of these categories of receptors is provided in Section 1.3.
The MOP is essentially a receptor who is exposed outside the boundaries of the facility. Refer to
Section 5.1.7 where the definition of IHI as specifically applied in the PA is described:
Final Report for the Clive DU PA Model
24 November 2015 38
“Inadvertent intrusion is often used in terms of direct but inadvertent access to the
waste (e.g. through well drilling or basement construction), for which the initiator
is exposed. However, such direct activities are unlikely at this site. The types of
activities here do not result in direct exposure to the waste by the initiator, but
potentially to future receptors.”
Ranching Scenario. The land surrounding the Clive Facility is currently utilized for cattle and
sheep grazing. Ranchers typically use off-highway vehicles (OHVs, including four-wheel drive
trucks) for transport. Activities are expected to include herding, maintenance of fencing and other
infrastructure, and assistance in calving and weaning. Ranchers may be exposed to contamination
via the pathways outlined in Table 1.
Recreational Scenario. Recreational uses on the land surrounding the Clive Facility may involve
OHV use, hunting, target shooting of inanimate objects, rock-hounding, wild-horse viewing, and
limited camping. As soil develops on the rip-rap surface of the cap and plant succession
proceeds, the disposal unit may become more attractive for different types of recreational
activities. It is assumed in the Clive DU PA Model that recreational OHV riders (“Sport”
OHVers; i.e., OHV users who use their vehicles for recreation alone) and hunters using OHVs
(“Hunters”), both of whom may also camp at the site, represent the most highly-exposed
recreational receptors. Recreationalists may be exposed to contamination via the pathways
outlined in Table 1.
Table 1. Exposure Pathways Summary
exposure pathway ranching recreation
Inhalation (wind derived dust) × ×
Inhalation (mechanically-generated dust) × ×
Inhalation (gas phase radionuclides) × ×
Ingestion of surface soils (inadvertent) × ×
Ingestion of game meat

× (Hunter)
Ingestion of beef ×
External irradiation – soil × ×
External irradiation – immersion in air × ×

The ranching and recreation scenarios are characterized by potential exposure related to activities
both on the disposal site and in the adjoining area. Specific off-site points of potential exposure
also exist for other receptors based upon present-day conditions and infrastructure. Unlike
ranching and recreational receptors who might be exposed by a variety of pathways on or
adjacent to the site, these off-site receptors would likely only be exposed to wind-dispersed
contamination, for which inhalation exposures are likely to predominate. Five specific off-site
locations and receptors are evaluated in the Clive PA, including:
Final Report for the Clive DU PA Model
24 November 2015 39
• Travelers on Interstate-80, which passes 4 km to the north of the site;
• Travelers on the main east-west rail line, which passes 2 km to the north of the site;
• Workers at the Utah Test and Training Range (UTTR, a military facility) to the south of
the Clive facility, who may occasionally drive on an access road immediately to the west
of the Clive Facility fence line;
• The resident caretaker at the east-bound Interstate-80 rest facility (the Grassy Mountain
Rest Area at Aragonite) approximately 12 km to the northeast of the site, and,
• OHV riders at the Knolls OHV area (BLM land that is specifically managed for OHV
recreation) 12 km to the west of the site.
4.1.2.11 ALARA
CFR (Section 61.42) defines a second decision rule that pertains to populations as well as
individuals. The regulation states "reasonable effort should be made to maintain releases of
radioactivity in effluents to the general environment as low as is reasonably achievable" (or
ALARA). The ALARA concept can be applied to either individuals or populations. In the context
of the Clive DU PA Model, ALARA is applied to collective doses germane to the receptor
populations described in Section 4.1.2.10.
The ALARA process is also described in DOE regulations and associated guidance documents
such as 10 CFR Part 834 and DOE 5400.5 ALARA (10 CFR 834; DOE 1993, 1997), and in other
NRC documents (NRC, 1995, 2000, 2004, 2015). The definitions in each case are very similar;
indicating that exposures should be controlled so that releases of radioactive material to the
environment are as low as is reasonable taking into account social, technical, economic, practical,
and public policy considerations.
The probabilistic Clive DU PA Model is designed to estimate individual annual doses to
hypothetical individuals in future populations that may be exposed to radionuclide releases from
the Clive Facility. The model is also able to aggregate individual doses into estimates of
collective and cumulative population dose on an annual basis as well as over the 10-ky period of
performance. Given this model structure, an opportunity exists with the Clive DU PA Model to
evaluate ALARA in the context of population dose.
The overall implication of the various Agency regulations and guidance documents regarding
ALARA is that many factors should be taken into account when considering the potential
benefits of different options for disposal of radioactive waste. Previous guidance from NRC
(2004) suggests several different options for addressing consequences over thousands of years, as
is necessary for the DU PA. The options essentially correspond to different discount rates. NRC
recommends using 3 percent and 7 percent discount rates, where the former approximates the real
rate of return on long-term government debt, and the latter approximates the marginal pretax real
rate of return on average investment in the private sector. NRC relied on OMB (2003) for its
central arguments, noting that OMB also recognizes that special circumstances might arise when
considering long time frames, for which ethical and technical arguments might support the use of
lower discount rates. Consequently, NRC suggests also performing analysis with a zero percent
Final Report for the Clive DU PA Model
24 November 2015 40
discount rate, and sensitivity analysis across a range of possible discount rates. The most recent
(NRC, 2015) guidance does not mention discount rates, so none are applied here (which
implicitly assumes a zero rate).
In order to implement ALARA in a logical system, and so that economic factors are taken into
consideration, a decision analysis is implied. Decision analysis is the appropriate mechanism for
evaluating and optimizing disposal, closure and long term monitoring and maintenance of a
radioactive waste disposal system. Decision options for disposal at Clive include engineering
options and waste placement. More generally, if decision analysis is applied, then a much wider
range of options can be factored into the decision model, such as transportation of waste, risk to
workers, and effect on the environment.
The decision analysis in this context is essentially a benefit-cost analysis, within which different
options for the placement of waste are evaluated. For each option, the Model predicts doses to the
array of receptors, and the consequences of those doses are assessed as part of an overall cost
model, which also includes the costs of disposal of waste for each option. The goal is to find the
best option, which is the option that provides the greatest overall benefit. The consequences of
risk can be measured through a simplification that is available in ALARA guidance, including
NRC 2015, which provides the basis for, and history of, assigning a dollar value to person-rem as
a measure of radiation dose.
In assigning a value to the person-rem cost to society of radiation dose, the agencies have
simplified the basis for a full decision analysis. This is reasonable for a first pass at a decision
analysis associated with the proposed disposal at Clive. Hence, the value of $5,100 is applied to
the population dose. Application of the ALARA process to the Clive DU PA Model is described
more completely in the Decision Analysis white paper (Appendix 12).
4.1.2.12 Groundwater Concentrations
Apart from individual and population dose evaluations, evaluation of the PA also requires
comparison of groundwater concentrations with groundwater protection levels, or GWPLs. That
is, the State of Utah imposes limits on groundwater contamination, as stated in the Ground Water
Quality Discharge Permit (UWQB, 2010). Part I.C.1 of the Permit specifies that GWPLs in Table
1A of the Permit shall be used for the Class A LLW Cell. Table 1A in the Permit specifies
general mass and radioactivity concentrations for several constituents of interest to DU waste
disposal. This includes values for mass concentration of total uranium, radium, and gross alpha
and beta radioactivity concentrations for specific wells where background values were found to
be in exceedence of the Table 1A limits. Part I.D.1 of the Permit specifies that the performance
standard for radionuclides is 500 years. Relevant GWPLs for Clive are:
• Strontium-90 42 pCi/L,
• Technetium-99 3,790 pCi/L,
• Iodine-129 21 pCi/L,
• Thorium-230 83 pCi/L,
• Thorium-232 92 pCi/L,
• Neptunium-237 7 pCi/L,
• Uranium-233 26 pCi/L,
Final Report for the Clive DU PA Model
24 November 2015 41
• Uranium-234 26 pCi/L,
• Uranium-235 27 pCi/L,
• Uranium-236 27 pCi/L, and
• Uranium-238 26 pCi/L.
The main concern for the PA model is the potential for transport of 99Tc, a contaminant in the DU
waste, to the point of compliance.
Note that according to the Permit, groundwater at Clive is classified as Class IV, saline ground
water, according to UAC R317-6-3 Ground Water Classes, and is highly unlikely to serve as a
future water source. The underlying groundwater in the vicinity of the Clive site is of naturally
poor quality because of its high salinity and, as a consequence, is not suitable for most human
uses, and is not potable for humans. However, the Clive DU waste PA will calculate estimates of
groundwater concentrations at the location of a virtual well near Federal DU cell for comparison
with the GWPLs.
4.1.2.13 Deep Time Assessment
The approach to deep time modeling is briefly described in the Conceptual Site Model for
Disposal of Depleted Uranium at the Clive Facility white paper (Appendix 2). A more in-depth
discussion of the deep time modeling methodology is described in Deep Time Assessment for the
Clive PA white paper (Appendix 13). The focus of the deep time evaluation is to assess the
potential impact of glacial epoch pluvial lake events on the Federal DU cell from 10 ky through
2.1 My post-closure. (note that this model is termed the “deep-time” model.) A pluvial lake is a
consequence of periods of extensive glaciation, and results from low evaporation, increased cloud
cover, increased albedo, and increased precipitation in landlocked areas. Given that long-term
climatic cycles of 100 ky are considered very likely in the next 2.1 My (Hayes et al., 1976;
Shackleton, 2000), it is assumed that deep lakes will return to the Bonneville Basin in the future.
In addition to deep lakes, intermediate sized lakes are also assumed to occur, periodically during
a 100-ky glacial cycle. Events that might occur in deep time other than the occurrence of
intermediate lakes and the cyclic return of deep lakes (e.g., meteor strikes and a large eruption at
Yellowstone) are not considered further in this model because their likelihood is relatively small,
and their consequences are likely to be much greater and far reaching for human civilization.
For the deep time scenarios, the PA model provides a qualitative assessment of the future
consequences of present-day disposal of DU waste to the environment. While no exposure or
dose assessment is attempted, tracking of radioactive species concentrations provides insight into
waste disposal and embankment construction design and performance. Long-term historical
information on the area surrounding the Clive site is sparse, providing only a broad depiction of
historical behavior of lake cycles in the Bonneville Basin. Thus, the model utilized for projecting
into the long-term future is largely conceptual or stylized, providing a similarly broad depiction
of future behavior
There are two components of the model used to represent the deep time scenarios. The first is
modeling lake formation and dynamics in the Bonneville Basin. The second is modeling the fate
of the Federal DU cell and disposed DU waste.
Final Report for the Clive DU PA Model
24 November 2015 42
For the first component, the deep time evaluation focuses on potential releases of radioactivity
following a series of pluvial lake events caused by glacial cycles assumed to occur
(approximately) every 100 ky. The 100-ky glacial periodicity is based on historical ice core and
the benthic marine isotope data for the past 800 ky. These cycles are also consistent with
information regarding orbital forcing, and the periodicity suggested by the Milankovitch cycles.
These 100 ky glacial cycles form the basis for modeling the return and recurrence of lake events
in the Bonneville Basin. The lake formation model is applied to each 100 ky cycle similarly. One
deep lake is assumed to occur every in each 100 ky cycle, and several intermediate lakes are
allowed to form during the transgressive and regressive phases of the deep lake. Note that the
current 100-ky cycle is not modeled differently than future glacial cycles, despite evidence that
the current inter-glacial period might last for another 50 ky (Berger and Loutre, 2002). In the
model, therefore, an intermediate lake can return sooner than might be expected in the current
100-ky cycle. The precise timing of the return of a lake at or greater than the elevation of Clive is
not as important as the event itself.
For the second component, it is assumed that destruction of the Federal DU cell and fate of the
DU waste will result from the effects of wave action from an intermediate or deep lake. In effect,
it is assumed that a lake is large enough that obliteration of the embankment will occur. In this
obliteration scenario, all of the embankment material above grade is dispersed across a large
localized area through wave action, although this includes all the DU waste, even if some DU
waste was disposed below grade. Inclusion of the below grade waste is conservative, since it
allows more DU waste to migrate into returning lakes and future sediment. The waste material is
mixed with sediment and then enters the lake system via dissolution. A simplifying, conservative
assumption is to limit dissolution to a column above the waste dispersal area. This assumption is
conservative because lake water will probably mix more extensively, creating greater dilution. As
a result, these assumptions lead to greater concentrations of waste than is probably reasonable.
The conservatism is included in this model because of the lack of data that exists to quantify the
processes.
The deep-time model assumes that the form of DU available for deep-time transport is U3O8,
which is far less soluble than UO3. Fate and transport modeling performed using the PA Model
indicates that the relative soluble UO3 will have migrated transported to groundwater within 50
ky. Consequently, the deep time model focuses on U3O8 as the form of DU available for deep-
time transport. While the lake is present, some waste in the water column will bind with
carbonate ions and precipitate out into oolitic sediments, while the remaining waste will fall out
with the sediment as the lake eventually recedes. The model assumes the waste is fully mixed
with the accumulated sediments, a conservative assumption, since some waste is likely to be
buried rather than mixed with future lake sediments. The extent of mixing of previous sediment
with new sediment is not well understood; hence an assumption that the sediments completely
mix is expedient, and probably leads to conservative results. All of the waste that has dissolved
into the lake re-enters the lake sediment once the lake recedes. Overall sediment concentrations
decrease over time because the amount of waste does not change other than through decay and
ingrowth, whereas more sediment is added over time.
Thus the deep-time model should be regarded as conceptual and heuristic. The intent is to present
a picture of what the long-term future might hold for the DU waste disposal embankment, rather
Final Report for the Clive DU PA Model
24 November 2015 43
than to provide a quantitative, temporally-specific, prediction of future conditions, or an
assessment of exposure or dose to human receptors. The type of glacial climate change
envisioned in the deep-time model will probably have wide-reaching consequences for the planet
and human society that are far beyond the scope of a PA for disposal of radioactive waste.
5.0 Model Structure
5.1 Summary of Important Assumptions
The results of the Clive DU PA Model depend critically on the model structure, the model
specification (input probability distributions, for example) and the assumptions that underlie the
model. That is, the results are fully dependent, or conditional, on the Model. The most important
assumptions are identified in this section.
5.1.1 Points of Compliance
Points of compliance in a PA are usually defined in terms of the location in the accessible
environment at which human health is evaluated in the dose assessment, and the location at
which groundwater concentrations are used for comparison to GWPLs. For this model, the
primary receptors (ranchers, recreators) are assumed to spend time on the site, and off the site in
the general vicinity. Other receptors are defined at points in space (See Section 4.1.2.10.1). Note
that the ALARA analysis addresses the same points of compliance.
Groundwater concentrations are evaluated at a virtual well located 27 m (90 ft) from the interior
of the waste embankment. In the case of the proposed DU waste disposal, only the top slope
section of the embankment would contain DU waste, so the effective distance from the DU waste
to the well is lengthened by the width of the side slope section, to about 73 m (240 ft).
For the deep-time model, there are no receptors that are considered, and doses are not calculated.
Instead, concentration of radionuclides are estimated in lake water and in lake sediment in the
general vicinity of the Federal DU cell.
5.1.2 Time Periods of Concern
There are four time periods that have import in this PA. The PA model is run fully quantitatively
for dose endpoints for 10 ky. Peak mean dose is estimated and used for comparison with
performance objectives for this time frame. The ALARA analysis is also performed for this
period of time.
An institutional control period of 100 y is assumed, during which time doses are not calculated,
because access to the site is assumed to be not possible.
Groundwater concentrations are compared to GWPLs for the first 500 years of the model, since
this is the compliance period that is applied to the GWPLs under Utah Code.
The deep-time model is run for 2.1 My because the DU does not achieve secular equilibrium
until about that time. That is, the model is run to peak activity of the DU, rather than to peak
dose, which is undefined that far into the future.
Final Report for the Clive DU PA Model
24 November 2015 44
5.1.3 Closure Cover Design Options
The engineered system in the PA model allows for evaluation of many different disposal
configurations. DU waste is assumed to not be disposed under the side slopes. There are 27 waste
layers in the model, each about 0.45 m thick, starting with Layer 1 directly under the cap. The
layers are numbered one through 27, with the 27th layer at the bottom of the waste cell. Layers 22
through 27 are below grade. Only one type of waste can be placed in a specific layer. Although
the model is setup to allow for many different waste disposal configurations only one is
considered in this version of the Clive DU PA model:
GDP contaminated waste in Layer 22 –GDP uncontaminated waste in Layers 23-26 –
SRS waste in Layer 27. Note that fill material is assumed for the 9 m between the cap and
Layer 21. This model places all waste below grade.
The initial model v1.0 had three configurations that spanned a fairly wide range of options, from
disposal near the cap, to disposal primarily below grade. In the current model, DU waste is only
being considered to be disposed below grade.
5.1.4 Waste Concentration Averaging
Within each waste layer the contents of the waste are assumed to include the waste material and
the fill material needed to occupy the layer volume. Since each layer represents a mixing cell, the
concentration of the radionuclides is averaged throughout the layer. That is, each drum or
cylinder is not modeled separately. This is typical of PA models, and is reasonable provided
transport from the actual configuration does not differ greatly from transport from the modeled
configuration.
5.1.5 Environmental Media Concentration Averaging
Similarly to the waste layers, concentrations in the environmental media are averaged throughout
the cell that represents the medium. For example, the concentration of uranium in deep-time lake
sediment is the average concentration throughout the sediment layer that is defined by its model
cell.
5.1.6 Members of the Public
MOP is defined in terms of the receptors who perform activities in the vicinity of the Clive
facility. This includes receptors at specific locations offsite as described in Section 0.
5.1.7 Inadvertent Human Intrusion
Following NRC 10 CFR 61, inadvertent intrusion is defined in terms of receptors who might
perform some activities onsite. This includes ranchers, hunters and OHV enthusiasts. Inadvertent
intrusion is often used in terms of direct but inadvertent access to the waste (e.g. through well
drilling or basement construction), for which the initiator is exposed. However, such direct
activities are unlikely at this site. The types of activities here do not result in direct exposure to
the waste by the initiator, but potentially to future receptors. However, the receptors identified
here are engaged in onsite activities, and are hence indirectly exposed to the DU waste.
Final Report for the Clive DU PA Model
24 November 2015 45
5.1.8 Deep Time Evaluation
The deep-time evaluation depends on the return of a lake in the Bonneville Basin that is large
enough to obliterate the Federal DU cell. Such a lake is assumed to occur more than once in each
100-ky glacial cycle. Once the Federal DU cell is obliterated, the material is assumed to disperse
within the vicinity of Clive. The dispersed radionuclides then migrate into lake water through
diffusion. All radionuclides that leave the sediment return to the sediment as the lake recedes,
either physically or chemically. The wastes are assumed to mix with lake sediment in each lake
cycle.
The outputs of interest are concentrations of radionuclides in lake water and in lake sediment, as
well as radon flux after the first lake recedes.
Note that in version 1.0 of the model, all DU waste is assumed dispersed with the arrival of the
first intermediate lake. However, the DU waste in versions 1.2 and 1.4 of the model is disposed
below grade in which case dispersion of the waste itself does not occur. In these updated model
versions only the radionuclides that have moved above the ground surface are dispersed.
5.2 Distribution Averaging
Most parameters in the Clive DU PA Model correspond to physical quantities that represent an
average of some type. Some parameters represent averages over time, as they represent typical
behavior that will be used throughout the 10-ky performance period, such as annual precipitation.
Other parameters represent averages over space. For example, properties of vegetation represent
an average vegetation effect across a model area, while soil properties represent an average
across a volume of material represented by a model cell. When data are available that represent
small amounts of time relative to the 10,000 years, or small areas/volumes relative to the model
cells, then it is the mean of the data distribution that needs to be modeled.
To capture the temporal domain of the model, time steps in this type of systems-level dynamic
probabilistic model are usually on the order of several to many years. Consequently, the average
effects over long time frames, assuming no catastrophic changes in the system, are far more
important than the effects on the scale of days, hours, minutes or seconds. Spatial and temporal
scaling of available data, which are usually collected at points in time and space, is critical for the
success of systems-level models. Scaling in this context is essentially an averaging process both
spatially and temporally. Simple averaging works well if the effect on the response of a variable
or parameter is linear. Otherwise, some care needs to be taken in the spatio-temporal averaging
process. In addition, these types of models are characterized by differential equations and
multiplicative terms. Averaging is a linear construct that does not translate directly in non-linear
systems. Again, care needs to be taken to capture the appropriate systems-level effect when
dealing with differential equations and multiplicative terms.
Another important statistical issues that is often overlooked in PA is correlation between inputs.
Many parameters in the Clive DU PA Model are related to one another. One parameter may be
physically constrained by the value of another parameter, or they may simply tend to vary
together. When joint data are available, a simple approach is to simply calculate the sample
correlation of the parameters in the data and apply the same correlation to the parameters in the
model to induce a joint distribution. A simple correlation structure may not fully capture the
Final Report for the Clive DU PA Model
24 November 2015 46
relationship between two parameters but often provides a reasonable first approximation. Where
a correlation structure is used in the Clive DU PA Model, the correlation algorithms implemented
in GoldSim for Gaussian copula are used (Iman and Conover 1982, Embrechts et al. 2001).
Where data and expertise are available, it is generally preferable to construct joint distributions
for the parameters by constructing a marginal distribution for one parameter and conditional
distributions for the remaining parameters. By fitting a distinct conditional distribution of the
second parameter for each possible value of the first parameter, a more realistic relationship
might be constructed than can be achieved through simple correlation
The statistical methods used for appropriate spatio-temporal scaling and correlation effects are
described in the Development of Probability Distributions white paper (Appendix 14).
5.3 Model Evaluation through Uncertainty and Sensitivity Analysis
The Clive DU PA is built as a probabilistic systems-level model. Systems-level modeling is
geared towards decision objectives, and is a style of bottom-up modeling for which model
refinement and iteration is performed in response to model evaluation. Model evaluation is
performed throughout model development, but in the final stages it involves uncertainty analysis
and sensitivity analysis. Quantitative assessment of the importance of inputs is necessary when
the level of uncertainty in the system response exceeds the acceptable threshold specified in the
decision making framework. One of the goals of sensitivity analysis is to identify which variables
have distributions that exert the greatest influence on the response.
Uncertainty is captured directly for probabilistic system-level models. The input probability
distributions are used to capture the range of possible parameter values. For probabilistic models,
sensitivity analysis is performed simultaneously for all input parameters. This approach is termed
global sensitivity analysis. It is a very powerful tool at the disposal of probabilistic modeling for
identifying parameters that are important predictors of the model output, and it is not constrained
by the user’s preconceptions of what may be important. In addition to global sensitivity analysis,
probabilistic models can be evaluated numerically in an uncertainty analysis and for value of
information. Uncertainty analysis in this context involves comparison of the output distribution to
performance metrics. A determination can then be made based on the comparison of the
compliance of the disposal system. Value of information analysis can be performed to identify
parameters for which uncertainty reduction in the output of interest might best be achieved, if it is
necessary to reduce uncertainty. This approach can also be used in the context of ALARA
contamination goals, to determine if further uncertainty reduction can reasonably be performed.
Sensitivity analysis is a very important tool for understanding the model. For those parameters
that are deemed as important, and if the uncertainty analysis indicates, then there are options for
further model refinement. These options include further data collection, and refinement of the
model. Uncertainty and sensitivity analysis are applied to each endpoint (model output)
separately. Consequently, it is reasonable to expect that some of the endpoints are sensitive to
different inputs. For example, output doses might be sensitive to parameters that are related to
radon production and transport, whereas the groundwater concentrations might be sensitive to 99Tc inventory or Kd. Consequently, each endpoint might have different needs regarding further
data collection or model refinement.
Final Report for the Clive DU PA Model
24 November 2015 47
Sensitivity analysis can be used to help identify those inputs for which uncertainty reduction
through further information collection will have the most impact on reducing uncertainty in the
model response. However, sensitivity analysis of high dimensional probabilistic models can be
computationally challenging. These challenges can be met through machine learning methods
applied to probabilistic simulation results. Further details are provided in the Sensitivity Analysis
Methods white paper (Appendix 15).
Another aspect of uncertainty when running probabilistic simulations is simulation stability. The
final statistics of interest might relate to the mean output, or a percentile of the output, and
therefore may require a large number of simulations for stability of the estimate of the statistic.
The question is, how large? The number of simulations needed can be determined by running a
different number of simulations for each endpoint and statistic of interest. Otherwise, simulation
uncertainty could interfere with the uncertainty and sensitivity analysis.
5.4 Clive DU PA Model Structure
The Clive DU PA Model is written using the GoldSim systems modeling software. Like other
such models, its structure is hierarchical, with nested “Containers” providing the means to
organize the model into different conceptual parts (see

Final Report for the Clive DU PA Model
24 November 2015 48
Figure 3). This model uses Containers to basic modeling constructs such as Materials, and
contaminant transport Processes that are global (model-wide) in scope. Other containers are
devoted to distinct topics, such as Inventory definitions, Disposal calculations, Exposure and
Dose calculations, comparisons to GWPLs, and the development of Deep Time Scenarios.
Supplemental containers define dashboards used for running the model and displaying results,
collected Results from calculations around the model, Simulation Settings for model controls,
and Documentation. The role of each of these is discussed below. For instructions on how to use
the model, consult the Clive DU PA Model User Guide.
The purpose of this model is to simulate, to a degree sufficient for decision making, the fate and
transport of radionuclides proposed for disposal in the Clive Facility, and to assess their potential
effects on future individuals and populations. This is done in the realm of environmental
transport modeling coupled with the modeling of health physics and toxicity to humans.
5.4.1 Materials
Any physical model of an environmental system must contain some sort of materials as a basis
for representing the physical environment. Water, air, waste, soils, and other porous media are
defined in this container, and are referenced throughout the model. The arrangement of these
materials in space, and their interconnectivity, is intended to represent a large block of the
environment, including the Clive Facility, or in this case the Federal DU Cell within that facility,
and its surroundings. The spatial definition of the environment is in the Disposal container.
5.4.2 Processes
Contaminant transport in the environment is driven by several processes in this model, including
advection in water, diffusion in water, diffusion in air, uptake and redistribution by plants, and
disturbance by burrowing animals. The parameters defining these processes are global in model
scope, and so are defined at this high level. The actual implementation of these processes in
moving radionuclides in the environment, is done mostly in the Disposal container.
Radioactive decay and ingrowth, chemical solubility in water, soil/water partitioning, and
air/water partitioning are also fundamental processes that determine fate and transport of
radionuclides, though these are defined in the Materials container, since they are directly related
to materials.
5.4.3 Inventory
The mass of radionuclides introduced as waste into the model is called the inventory. Inside this
container, the total mass of various types of DU waste is defined, as are the concentrations of the
radionuclides in each type of waste. These inventories can be selected individually or in
combination by the user by using the Control Panel dashboard (see Figure 4), and is then
introduced to the modeling cells that represent the waste layers, in the Disposal container.
Final Report for the Clive DU PA Model
24 November 2015 49

Figure 3. Top level of the Clive DU PA Model v1.4.

5.4.4 Disposal
For the first 10,000 yr following disposal, calculations are performed for the fate and transport of
radionuclides from the inventory into and throughout the modeled environment, in the Disposal
container. Here the physical location of modeling cells is defined, each with materials
representing what would be found at that location. For example, modeling cells represent the
cover container Unit 4, Unit 2 (shallow aquifer), clays, and other porous media, as well as water
and air. Cells representing the aquifer contain Unit 2 sediments and water, but no air, since this
regions is saturated with water by definition. Waste cells contain waste and backfill as porous
media, air and water, and are provided a mass of radionuclides from the inventory. As the model
progresses through time, these radionuclides migrate into other part of the physical system, and
eventually are found in environmental media (air, water, soils) that receptors will encounter. The
Disposal container performs essentially all the contaminant transport calculations to necessary to
estimate future concentrations of radionuclides in these exposure media.
Final Report for the Clive DU PA Model
24 November 2015 50

Figure 4. Control Panel for the Modeling of the Clive Disposal Facility.

5.4.5 Exposure and Dose
The exposure and dose calculations, which also include estimates of uranium toxicity hazard, are
performed in this Exposure_Dose container. Receptors are hypothetical future humans who have
behaviors similar to those of people around the site today: There are ranch workers, hunters, and
OHV enthusiasts, all of whom are expected to have direct access to the site after institutional
control is lost. There are also receptors who travel in the area, using highways, railroads, and
access roads. These receptors are represented with a range of attributes and behaviors, from age
to time spent on an OHV, and each encounters exposure media. As they breathe dust-laden air
and walk on contaminated soils, for example, their exposures result in doses from radionuclides
and toxic effects from uranium as a heavy metal. All of these calculations are performed in this
container, and provide results that can be compared to performance objectives such as peak dose
limits.
Final Report for the Clive DU PA Model
24 November 2015 51
5.4.6 Groundwater Protection Level Calculations
In addition to the performance objectives provided by the State of Utah and the NRC for dose
limits, there are GWPLs to be considered. In the Disposal container, the model provides
radionuclide concentrations at a hypothetical monitoring well located about 27 m (90 ft) from the
interior of the waste embankment. In the case of the proposed DU waste disposal, only the top
slope section of the embankment would contain DU waste, so the effective distance from the DU
waste to the well is lengthened by the width of the side slope section, to about 73 m (240 ft). For
those radionuclides that have GWPLs defined, the maximum well concentrations within 500 yr
are compared to the GWPL values. These comparison calculations are performed in the GWPLs
container.
5.4.7 Deep Time
All the calculations described above are aimed at producing results for comparisons to
performance objectives that pertain to the first 10,000 yr after disposal. Following that, and out to
the time of peak activity, is considered deep time. Peak activity of the DU waste, which is
predominantly 238U, is the time at which the decay products of the parent reach secular
equilibrium with the parent. In this case, the peak activity is at about 2.1 million years. For the
purposes of the model, then deep time is that duration from 10,000 y to 2.1 My.
Given the distinct time frame, the deep time calculations are independent of much of the rest of
the model, except that the radionuclide mass in the embankment, as calculated in the Disposal
container, is used as a source of radionuclides for dispersal in future lakes. The
DeepTimeScenarios container produces estimates of radionuclide concentrations in the water of
future lakes, in the sediments that they deposit, as well as radon flux calculations and rancher
scenario dose estimates after the first lake recedes.
5.4.8 Supplemental Containers
The Dashboards container is simply a location in the model for storing Dashboard elements,
which are dialog-box-like controls for operating the model and for conveniently viewing results.
The model can be executed and browsed without using any dashboards, though their convenience
makes them quite useful.
The Simulation Settings container hosts a small number of elements that are used simply to
control the simulation. Logical switches and values controlled by the dashboards are kept here,
and the container will probably be of little interest to the average user.
The dashboards provide access to several results of general interest, most of which are collected
in the Results container. In addition to those referenced by the dashboards, there are many other
results that provide a more detailed look into the model. Also inside this container are the results
needed for performing sensitivity analyses, such as those discussed later in this report.
Documentation contains records pertinent to model development, such as the Change Log,
illustrations about particular model processes, and a large collection of references supporting the
model. The sub-container Documentation\References holds nearly 1 GB of reference materials in
PDF format.
Final Report for the Clive DU PA Model
24 November 2015 52
6.0 Results of Analysis
The Clive DU PA Model was run in order to evaluate the performance of the disposal system and
to understand the sensitivity of input parameters on those results. Endpoints of interest include:
• groundwater concentrations of radionuclides for which GWPLs are specified,
• dose and uranium toxicity hazard to various receptors,
• 222Rn flux in the deep time analysis, and
• lake water and sediment concentrations of 238U in the deep time analysis.
Statistical results (e.g. mean, median, 95th percentile) are based on probabilistic simulations of
10,000 realizations. Sensitivity analysis has been performed on each of the model endpoints of
interest.
The DU waste is disposed below the grade of the surface soil surrounding the embankment, about
11 m (36 ft) below the surface of the embankment. The disposal volume above the DU waste is
assumed to be backfilled with clean material for the purposes of this DU analysis. The waste is
arranged as follows: The bottom waste layer contains SRS DU, the four waste layers above that
contain Clean GDP DU, and the top waste layer contains Contaminated GDP DU. Details
regarding these wastes can be found in the Waste Inventory white paper.
Each waste layer is roughly 0.45 m (18 in) in thickness. In general, the effect of the layer is that
the higher the waste is emplaced in the volume, the greater influence it has on doses, which are
derived from surface soils. The lower the waste, the greater its influence on groundwater
concentrations. For this reason, the contaminated GDP DU wastes are placed above the clean
GDP DU wastes, in order to position the 99Tc that is present in contaminated wastes as far from
the groundwater as possible. Details on this modeling can be found in the Embankment Modeling
white paper. This arrangement allows exploration of the Clive DU PA Model and hence the
performance of the system.
Groundwater protection levels are defined in the Clive Facility’s groundwater discharge permit
(UWQB 2009). Radionuclides with GWPLs and for which concentrations are evaluated include 90Sr, 99Tc, 129I, 230Th, 232Th, 237Np, 233U, 234U, 235U, 236U, and 238U (see Section 4.1.2.12). The
Clive DU PA Model estimates contributions to groundwater concentrations from the DU wastes
for 500 yr, assuming transport to a hypothetical monitoring well. Details on the groundwater
transport calculations are provided in the Unsaturated Zone Modeling and Saturated Zone
Modeling white papers (Appendices 5 and 7).
Possible human receptors are of the following basic types, and details are available in the Dose
Assessment white paper (Appendix 11):
• Ranch workers (mostly ranch hands), hunters, and OHV enthusiasts are expected to be
present on and near the embankment after the institutional control period.
Final Report for the Clive DU PA Model
24 November 2015 53
• Other receptors have doses evaluated at specific locations, including the nearby highway
(Interstate-80), the Knolls OHV Recreations Area (Knolls), the nearby rail road
(Railroad), the Grassy Mountain Rest Area on I-80 (Rest Area), and the Utah Test and
Training Range access road (UTTR).
• All receptors are considered in population dose calculations.
Erosion and the formation of gullies in the embankment cap are modeled using SIBERIA, a
landscape evolution model, abstracted into this version of the Clive DU PA Model. It is
considered more realistic than the screening exercise applied in v1.0 but has limitations because a
Borrow Pit at Clive was used for the modeling rather than the Federal DU Cell and assumptions
had to be made to apply the SIBERIA results to the Federal DU Cell. As well, the formation of
gullies is not integrated with contaminant transport within the column in that erosion is not
considered for infiltration, diffusion, biotic transport or other processes in the embankment
column. The model may be run with or without inclusion of gully formation, so that their effect
on modeled doses may be explored. In the following presentation of results, gully calculations are
included. Details on the erosion calculations are provided in the Erosion Modeling white paper
(Appendix 10).
Deep time is considered to be that time after 10,000 yr, the period of performance for assessing
dose as specified in the Utah regulation. Endpoints related to the deep time assessment include
lake sediment concentrations of 238U, 230Th and 226Ra, and concentrations of 238U, 230Th and 226Ra
in lakewater, when lakes are present, as well as 222Rn flux and corresponding rancher dose after
the first lake recedes. Lake and sediment concentrations are presented in graphical format to
illustrate effects with lakes coming and going with time. Summary statistics of these results are
presented for 90,000 years, which is the timestep at which the greatest percentage of lakes is
present in the 10,000 realizations that were run. Details on the deep time calculation methods are
provided in the Deep Time Assessment white paper (Appendix 13).
Results for all these endpoints are summarized in tables below. The means and 95th percentiles
are used for comparison with performance objectives. Graphs of time histories and sensitivity
analysis results are also shown, although in cases where results are qualitatively similar, only a
single representative graph is presented. The results presented below in tables and sensitivity
analysis results and figures are primarily from the Clive DU PA Model v1.4 run for 10,000
realizations with seed 2. Where noted, the results presented in other figures and tables with 1000
realizations are from the Clive DU PA Model v1.4 run for 1000 realizations with seed 1. In both
cases, Latin Hypercube Sampling is enabled using mid-points of strata, and Repeat Sampling
Sequences is enabled. GoldSim solution precision is set to High.
Some output distributions are positively skewed, with a long tail. The long tails are probably due
to a combination of factors that include skewed input distributions that reasonably reflect
uncertainty in upper values of a parameter, multiplicative effects in the model, and missing
correlations between some input parameters. This can lead to implausible combinations of input
values. Consequently, results that are far into the tail of the output distributions might be
unreliable.
Final Report for the Clive DU PA Model
24 November 2015 54
6.1 Groundwater Concentrations
Peak groundwater activity concentrations within 500 yr resulting from the proposed waste
disposal are calculated for all radionuclides at a hypothetical monitoring well placed about 27 m
(90 ft) from the toe of the waste in the DU Federal Cell. In the case of the proposed DU waste
disposal, only the top slope section of the embankment would contain DU waste, so the effective
distance from the DU waste to the well is lengthened by the width of the side slope section, to
about 73 m (240 ft).
6.1.1 Summary of Results for Groundwater
For those radionuclides for which GWPLs exist, as specified in the facility’s permit (UWQB
2009), results are shown in Table 2. It should be noted that these statistics summarize the
concentrations at 500-yrs. In general, concentrations increase with time, in which case the
statistics presented are of the concentrations at 500 yrs. Because all modeled estimates are of
mean concentrations, the statistics represent the mean, median and 95th percentile of the (peak of
the) mean concentration. As such, the 95th percentile is analogous to a 95% upper confidence
limit on the mean. The large difference between the mean and median concentrations, when
values are reported for each, indicates that these output distributions are markedly positively
skewed.
Table 2. Summary statistics for peak mean groundwater activity concentrations within
500 yr, compared to GWPLs
activity concentration at 500 yr (pCi/L)
radionuclide GWPL1
(pCi/L) mean median
(50th %ile) 95th %ile
90Sr 42 0 0 0
99Tc 3790 26 4.3E-2 150
129I 21 1.7E-2 4.3E-11 1.1E-1
230Th 83 2.2E-28 0 0
232Th 92 1.4E-34 0 0
237Np 7 1.5E-19 0 3.7E-27
233U 26 5.6E-24 0 3.9E-28
234U 26 2.1E-23 0 2.2E-28
235U 27 1.6E-24 0 2.0E-29
236U 27 2.7E-24 0 3.3E-29
238U 26 1.5E-22 0 1.8E-27
1GWPLs are from UWQB (2009) Table 1A.
Results are based on 10,000 realizations, seed 2

Final Report for the Clive DU PA Model
24 November 2015 55
Since the DU waste is emplaced below grade in the embankment, modeled monitoring well
concentrations are greater than they would be if the DU were emplaced at a higher elevation
within the embankment for two reasons: 1) The waste is closer to the groundwater, and so has a
shorter travel distance, bringing the peak closer in time, and 2) the waste is more concentrated if
it is arranged into a smaller volume, thereby decreasing the duration of breakthrough at the well,
while increasing its amplitude.
For most radionuclides in Table 2 the groundwater concentrations are negligible compared to the
GWPLs. The exceptions are 99Tc and 129I, although the 95th percentile values for these
radionuclides are still more than an order of magnitude below the respective GWPLs. The
distributions of these concentrations are highly skewed, largely because of the skew in some if
the input distributions. For example, the distributions for Kd for 99Tc and 129I are expressed as
left-truncated normal distributions, which is a skewed distribution.
In the case of 129I, this radionuclide was not detected in any samples collected from the SRS
drums (see the Waste Inventory white paper – Appendix 4). Not only was 129I not detected, but it
was not identified in any sample. However, because 129I may be present at concentrations below
the detection limits these limits were used directly for creating the input distribution for inventory
of 129I. This probably greatly overestimates the inventory of 129I in the DU waste.
The 99Tc inventory concentration distribution is derived from three datasets that suggest very
different potential waste concentrations, with particular uncertainty in the concentration of 99Tc in
the GDP waste. Consequently, the input distribution covers more than one order of magnitude of
possible 99Tc concentrations. With more data or better information, it is reasonable to expect that
this uncertainty could be reduced.
Technetium-99 is selected to represent a time history of monitoring well concentrations, as
shown in Figure 5. Figure 5 shows each of the 1,000 realizations, and Figure 6 shows a statistical
summary of those realizations. Note that the model results are based on 10,000 realizations, on
which the summary statistics in Table 2 are based. Subsequent time histories will show only the
statistical summaries. Of particular interest is that the concentrations of 99Tc are considerably less
than the GWPL, and concentrations of 99Tc increase over time up to the 500-yr compliance
period.
Final Report for the Clive DU PA Model
24 November 2015 56

Figure 5. Time history of 99Tc well concentrations; 1000 realizations shown.
Final Report for the Clive DU PA Model
24 November 2015 57

Figure 6. Time history of 99Tc well concentrations: statistical summary of the 1000
realizations shown in Figure 5.

Final Report for the Clive DU PA Model
24 November 2015 58
6.1.2 Sensitivity Analysis for Groundwater
A sensitivity analysis of the 99Tc and 129I groundwater concentrations was performed in order to
determine which modeling parameters are most significant in predicting its value. As seen in
Figure 7 and tablulated in Table 3, the most sensitive parameter for both 99Tc and 129I
groundwater concentrations was van Genuchten’s α. This parameter is included in the regression
equations for water content and infiltration, which affect both advection and diffusion of
radionuclides to groundwater. The next most sensitive parameter for both radionuclide
concentrations was molecular diffusivity in water, which controls diffusion. The soil-water
partition coefficient (Kd) was the 3rd most sensitive parameter for both radionuclides. Kd controls
sorption to the solid phase, with higher Kd resulting in increased sorption which retards migration
of the radionuclides.
Overall, the controlling mechanism is infiltration. The infiltration rate is sufficiently small that
waste inventory and Kd are not the primary drivers. Note also that the sensitivity indices are not
very large. It is typical in a model is well structured and specified, that a few input variables have
sensitivity indices that are quite large collectively. Otherwise, the model might not be well
formed, or the signal is very low. The latter appears to be the case here. Consider that the 50th
percentile is essentially zero, in which case the ability of an input variable to differentiate signal
responses is a challenge.
Sensitivity analyses for other results are presented in Appendix 19.
Table 3. Sensitivity Indices of select peak groundwater concentrations within 500 years.
Radionuclide SI rank input parameter sensitivity
index (SI)
waste emplaced below grade
99Tc 1 Unit 4 ET Layers log of van Genuchten’s α 32
2 Molecular Diffusivity in Water 25
3 Kd for Tc 1 14
4 Activity Concentration of Tc-99 in SRS DU
Waste
11
5 Unit 4 ET Layers log of van Genuchten’s n 4
129I 1 Unit 4 ET Layers log of van Genuchten’s α 36
2 Molecular Diffusivity in Water 30
3 Kd for I 1 18
4 Unit 4 ET Layers log of van Genuchten’s n 6
1 For iodine for technetium, the same Kd value was used for all materials.

Final Report for the Clive DU PA Model
24 November 2015 59

Figure 7. Partial dependence plot for peak 99Tc groundwater concentration in 500 years.

Final Report for the Clive DU PA Model
24 November 2015 60
6.2 Receptor Doses
Radiation doses to receptors are calculated as the total effective dose equivalent (TEDE), and are
compared to the performance objective of a peak dose of 0.25 mSv (25 mrem) in a year, achieved
within 10,000 yr (Utah 2010). Comparison with the inadvertent intrusion standard of 5 mSv
(500 mrem) in a year is also considered in relation to human-induced gully erosion.
6.2.1 Summary of Results for Doses
The dose results are summarized in Table 4, which shows the statistics for peak TEDE for all
receptors for DU waste emplaced below surface grade. These results include consideration of
dose related to gully erosion. The greatest doses occur at or near 10,000 years. The peak mean
dose results are presented below at 10,000 years, along with median and 95th percentile values of
the doses at 10,000 years.

Table 4. Peak of the mean TEDE: statistical summary within 10,000 yr.
TEDE (mrem in a yr) at 10,000 yr
receptor mean median
(50th %ile) 95th %ile
ranch worker 6.2E-2 5.1E-2 1.5E-1
hunter 4.5E-3 3.8E-3 9.9E-3
OHV enthusiast 8.4E-3 7.5E-3 1.8E-2
I-80 receptor 1.6E-6 1.3E-6 4.2E-6
Knolls receptor 4.6E-6 1.9E-6 1.8E-5
rail road receptor 2.5E-6 2.0E-6 6.6E-6
rest area receptor 4.1E-5 3.4E-5 9.7E-5
UTTR access road receptor 9.1E-4 7.4E-4 2.2E-3
Results are based on 10,000 realizations, seed 2.

Note that the doses to the offsite receptors are very small relative to the doses for receptors that
may be exposed on the embankment. Consequently, the attributes of the dose results for offsite
receptors are not explored in this discussion. Of greater interest are the doses to the ranchers,
hunters and OHVers. These three classes of receptors were modeled with the intent of capturing
dose to each hypothetical individual in the relevant populations (see the Dose Assessment white
paper – Appendix 11). The data presented hence represent summary statistics for the peak of the
mean dose to a diverse set of hypothetical individuals within each group of receptors. The peak
of the mean doses occurs at 10,000 years in the Clive DU PA Model, because dose increases with
time for DU. Consequently, the 95th percentile is analogous to a 95% upper confidence limit of
the mean dose at 10,000 years that might be used under CERCLA, for example.
The greatest doses are to ranch workers, which are greater than the doses to hunters and OHV
enthusiasts by about an order of magnitude or more. In all cases the summary statistics present
values that are far below the IHI performance objective of 5 mSv (500 mrem) in a year. Although
Final Report for the Clive DU PA Model
24 November 2015 61
the model results include consideration of dose that may occur subsequent to gully formation
initiated by inadvertent intrusion, these values are also far less than the MOP performance
objective of 0.25 mSv (25 mrem) in a year.
An evaluation of pathway-specific doses for the three onsite receptors indicates that effectively
100% of the dose is associated with the inhalation exposure pathway. There is practically zero
dose related to external radiation from soil or inadvertent soil ingestion, which is because
virtually no radionuclides have been transported to surface soil on the cap through the overlying
11 m (36 ft) of embankment within 10,000 years. Even in gullies, soil concentrations of 210Pb,
deposited as the progeny of 222Rn subsequent to air-phase diffusion, only reaches concentrations
of approximately 0.03 pCi/g. Because surface soil particulate radionuclide concentrations and
associated dose pathways are so low, the inhalation pathway doses are necessarily related to
inhalation of gas-phase radionuclides such as 222Rn.

6.2.2 Sensitivity Analysis for Doses
Sensitivity analysis was performed on the results for the mean TEDE at 10 ky to ranch workers,
hunters, and to OHV enthusiasts. Sensitive parameters are summarized in Table 5, and the partial
dependence plot for the ranch worker is shown in Figure 8.
Table 5. Sensitivities of peak TEDE within 10,000 yr
receptor SI rank input parameter sensitivity
index (SI)
waste emplaced below grade
ranch worker 1 Radon Escape/Production Ratio for Waste 44
2 Kd for Ra in sand 8
3 Molecular
 Diffusivity
 in
 Water
  4
hunter 1 Radon Escape/Production Ratio for Waste 61
2 Kd for Ra in sand 10
3 Molecular Diffusivity in Water 6
4 Resuspension Flux 5
OHV enthusiast 1 Radon Escape/Production Ratio for Waste 69
2 Kd for Ra in sand 12
3 Molecular Diffusivity in Water 7

Final Report for the Clive DU PA Model
24 November 2015 62

Figure 8. Partial dependence plots for the mean ranch worker dose, assuming waste below
grade.
As shown in Table 5, the most sensitive input parameter for all receptors is the radon E/P ratio,
which defines the fraction of 222Rn that escapes into the mobile environment when formed by
radioactive decay from its parent, 226Ra. Radon that does not escape but remains within the
matrix of the radium-containing waste material stays in place and decays to polonium and then to 210Pb. Note that the higher the E/P ratio, the higher the dose.
The next most sensitive input is the soil/water partition coefficient (Kd) for radium in sand, the
parent radionuclide of radon. Radon gas inhalation is an important dose pathway for all receptors
at the ground surface. Increased radium partitioning to the solid phase tends hinders migration of
dissolved radium, which reduces surface radon flux and thus doses, as can be seen in the partial
dependence plot in Figure 8. Molecular diffusivity in water is the next most important input;
increased diffusion of radionuclides through the unsaturated zone tends to increase doses.
The sensitivity analysis confirms that radon is the greatest dose driver in the model. The sensitive
parameters for radiation dose are associated with the release and transport of radon. Diffusivity
and Kd affect transport to the ground surface, while higher values of the radon E/P ratio are
Final Report for the Clive DU PA Model
24 November 2015 63
associated with higher radon doses. As described in the Dose Assessment white paper (Appendix
11), radon dose is not often calculated in a PA. Instead, radon flux at the surface of a disposal
system is commonly calculated and compared to a radon-specific flux criterion. This example
perhaps indicates the importance of considering the impact of radon in a dose calculation. If dose
due to radon inhalation was not included in the results, the rancher doses shown in Table 4 would
be orders of magnitude lower than those shown.
6.3 Receptor Uranium Hazard Indices
Uranium hazard indices (HIs) within 10,000 yr are calculated for each receptor scenario as the
sum of hazard quotients (HQs) for the ingestion exposure pathways defined in Table 1. A HQ is
the ratio of the average daily dose (i.e., chemical intake) of a chemical to the corresponding
reference dose for that chemical, where a reference dose is an estimate of daily exposure likely to
be without appreciable risk of adverse health effects. The uranium HI values are compared to
EPA’s standard HI threshold of 1.0, a level that indicates that the average daily dose is below the
dose associated with health effects.
6.3.1 Summary of Results for Uranium Hazard
The uranium HI results are summarized in Table 6, which shows the statistics for the peak of the
mean uranium HI for all receptors.
The HIs for uranium are extremely small relative to threshold of 1.0, indicating essentially no
possibility of observing health effects from uranium toxicity. Similar to the dose results presented
above, this indicates that disposal of DU waste below grade, at the bottom of the embankment, is
protective of human health and the environment. These values are in compliance with the
regulatory standards. Peak mean uranium HI results, across time, occur essentially at 10,000
years since concentrations at the ground surface increase with time within 10,000 years.
Table 6. Peak of the mean uranium hazard index within 10,000 years.
uranium hazard index at 10,000 yr
receptor mean median
(50th %ile) 95th %ile
ranch worker 3.0E-8 4.4E-16 8.5E-9
hunter 1.6E-9 3.4E-17 7.9E-10
OHV enthusiast 2.2E-9 4.8E-17 1.1E-9

6.3.2 Sensitivity Analysis for Uranium Hazard Index
Sensitivity analysis was performed on the results for the mean uranium hazard index to ranch
workers, hunters, and to OHV enthusiasts, summarized in Table 7. Sensitivities of uranium
hazard index within 10,000 yr. The most sensitive input parameter for the ranch worker uranium
HI is the beef transfer factor for Tc. Transfer factors define the amount of an element taken up
into muscle tissue of animal per unit of intake by the animal. As such, this is an important
parameter when determining dose via the ingestion pathway. For the hunter and OHV receptors,
Final Report for the Clive DU PA Model
24 November 2015 64
all inputs exhibited relatively low sensitivity indices, suggesting that the absolute uranium HI are
so low that the model signal it too small to find differentiating factors that explain the results.
Table 7. Sensitivities of uranium hazard index within 10,000 yr
receptor SI rank input parameter sensitivity
index (SI)
waste emplaced below grade
ranch worker 1 Beef Transfer Factor for Tc 44
2 Unit 4 ET Layers Bulk Density 7
hunter 1 Unit 3 Porosity 9
2 Contaminated Fraction of GDP DU 9
3 Mammal Burrow Excavation Rate 7
4 Tree Root/Shoot Ratio 7
OHV enthusiast 1 Contaminated Fraction of GDP DU 13
2 Mammal Burrow Excavation Rate 12
3 Tree Root/Shoot Ratio 6

6.4 ALARA
The focus of the assessment for establishing doses as low as reasonably achievable (ALARA) is
an evaluation of potential doses to the entire population of hypothetical individuals. This
calculation addresses the cumulative dose to all ranch workers, hunters, and OHV enthusiasts,
summed across all individuals and all years of the 10,000-yr simulation. These cumulative
population doses, expressed as the TEDE, are shown in Table 8.
Table 8. Cumulative population TEDE: statistical summary
population TEDE (person-rem) within 10,000 yr
receptor type mean median
(50th %ile) 95th %ile
total population 12 11 26
ranch worker 2.8 2.5 5.7
hunter 1.5 1.3 3.0
OHV enthusiast 8.3 7.4 17

These population doses represent the sum of the doses to all hypothetical individuals in each year
over the 10,000-yr simulation. Table 9 below shows statistics of the average number of
cumulative individuals at 10,000 years for the total population as well as the different receptor
types.
Final Report for the Clive DU PA Model
24 November 2015 65
Table 9. Cumulative receptor population: statistical summary
population at 10,000 yr
receptor type mean median
(50th %ile) 95th %ile
total population 3.2E6 3.2E6 3.3E6
ranch worker 1.0E5 1.0E5 1.1E5
hunter 7.6E5 7.6E5 7.9E5
OHV enthusiast 2.3E6 2.3E6 2.4E6
One measure for evaluating the population dose levels shown in Table 8 is by comparing these
doses with radiation doses related to natural sources. Average annual individual background
doses related to ubiquitous natural background radiation in the United States is approximately
3.1 mSv (310 mrem) (NCRP, 2009). For the total population of about 3 million individuals,
natural background radiation dose is therefore approximately 930,000 rem, a level that is many
orders of magnitude higher than the population doses shown in Table 8.
A second measure for these population doses can be obtained by considering the person-rem
costs suggested in NRC (see the Decision Analysis white paper – Appendix 12). Prior to 1995,
NRC suggested a flat $1,000 per person-rem cost. Subsequent to 1995, NRC suggested a value of
$2,000 with a discounting factor of 7% for the first 100 years, and 3% thereafter. NRC also
suggested that a range of $1,000 to $6,000 might be reasonable, with a best estimate of $2,000.
NRC noted that the intent of raising the person-rem costs from $1,000 to $2,000 was to
accommodate discounting in an economic analysis. Note that the intent of the NRC approach is
to capture the societal effects of added dose to the public. However, more recently, NRC (2015)
suggests a value of $5,100 per person-rem. Further discussion is provided in Appendix 12. If a
flat rate of $5,100 is applied to the population dose estimates provided above in Table 8, then the
costs associated with these scenarios are provided in Table 10.
Table 10. Statistical summary of the flat rate ALARA costs
population ALARA costs over 10,000 yr
simulation scenario mean median
(50th %ile) 95th %ile
total population $61,200 $56,100 $132,600
ranch worker $14,280 $12,750 $29,070
hunter $7,650 $6,630 $15,300
OHV enthusiast $42,330 $37,740 $86,700
Note that discounting could also be applied, but this would simply result in lower costs. This
analysis shows that the ALARA costs involved are small (for the total population, about $13 per
year, or considerably less than $1 per day) and that the estimated population dose is a fraction of
natural background radiation dose. The reasons for this are that there are few receptors in the
model that are involved in ranching, hunting or OHV activities at the site at any particular time,
the concentrations are low, and, hence, the individual doses are also low.
Final Report for the Clive DU PA Model
24 November 2015 66
6.5 Deep Time Results
The deep time model addresses in a heuristic fashion the fate of the Federal DU Cell from 10 ky
to 2.1 My, the time at which DU reaches secular equilibrium. The model addresses the needs
identified in the Section 5(a) of R313-25-9 of the UAC to perform additional simulations for the
period where peak dose occurs, for which the results are to be analyzed qualitatively. Even
though the deep-time model runs to 2.1 My and there is huge uncertainty in predicting human
society and evolution that far into the future, rancher doses are calculated to provide a context for
radon fluxes which are calculated when no lake is present. The output of the deep time model is
also presented in terms of concentrations of radionuclides in relevant environmental media.
The deep-time model considers the return of lakes in the Bonneville Basin that reach or exceed
the elevation of Clive. Two classes of lakes are considered. The first is a deep lake similar to
Lake Bonneville that inundates the Clive facility. It is deep and adds to materials that are
currently on Bonneville Basin floor. This type of lake is assumed to occur once every 100 ky in
line with the 100-ky climate cycles that have occurred for the past 1 My or so. The second type of
lake is shallower and is termed an intermediate lake. It is also assumed to inundate the Clive
facility and adds sediment materials but is not a deep lake like Lake Bonneville. It is more similar
to the Gilbert Lake that occurred at the end of the last ice age. This type of lake is assumed to
occur several times in each climate cycle in response to colder, wetter conditions.
Return of a lake at or above the elevation of Clive is assumed to result in the destruction of the
Federal DU Cell. The above-grade embankment material and radionuclides are assumed to be
dispersed through wave action. The dispersal area forms the basis for the lake volume in which
radionuclides are dissolved and ultimately settle back to the basin floor through precipitation or
through evaporation as the lake recedes. The lake cycle involves movement of the radionuclides,
subject to continuing decay and ingrowth, from the sediment into lake water and back to
sediment as the lake forms and recedes. The dispersed radionuclides are assumed to be fully
mixed with the accumulated sediment. Sediment accumulates on average at the rate of about
17 m per 100-ky climate cycle. The current Unit 3 layer of sediment at Clive, which is derived
from Lake Bonneville, is assumed to be a confining layer.
The lake cycle effects on transport processes are complex. Sediment core records show
significant mixing of sediment, but also can be used to identify significant lake events in the past
several hundred thousand years. The extent of sediment mixing is not well understood. The
mechanisms for dispersal of a relatively soft pile of material in the middle of a desert flat are not
well understood. The extent of mixing of dissolved materials in a deep lake is also not well
understood. The Model, consequently, is simplified to the point of acknowledging lake return,
destruction of the Federal DU Cell, and cycling of radionuclides between periodic lakes and basin
sediments.
In particular, the model overly simplifies the lake cycle processes and the effect of those
processes on the transport of radionuclides. It limits the dispersal of radionuclides through time.
Destruction of the Federal DU Cell is assumed to occur with a lake that at least reaches the
elevation of Clive. This means that even a very shallow lake is assumed to destroy the
embankment. With the sediment acting as one large mixing cell, lake water diffusion can occur
across the entire depth of the sediment, no matter how deep. The simplified model ignores
Final Report for the Clive DU PA Model
24 November 2015 67
increased precipitation and cooler conditions as the time of lake return approaches, which would
move radionuclides downwards in the sediment. With these simplifying assumptions, some
(perhaps unreasonably) high lake water and sediment concentrations are predicted by the Model.
The area of dispersal of the Federal DU Cell is captured with a simple distribution that reflects
the area of an intermediate lake. This fixes a dispersal area. Dissolution into the lake is assumed
to occur and to be mixed in the entire lake. The same dispersal area is used for both intermediate
and deep lakes, limiting both the volume of water within which dissolved materials might mix
and the area in which precipitates and evaporates can return.
Although the embankment material is dispersed within a specified dispersal area, isolation of any
part of the sediment profile is assumed not to occur. That is, the sediment is assumed to
completely mix with previous sediment for every lake event. Lake sedimentation does not allow
burial or isolation of previously formed sediment layers. Since different lakes can be identified in
sediment cores, this again limits the dispersal of the radionuclides.
The model, therefore, represents a closed system that cycles radionuclides from lake water to
sediment and back again. Decreased concentrations in sediment are obtained because of the
increased sediment load, but the mass of radionuclides available to diffuse into each lake is not
different in time, except from decay and ingrowth.
Deep Time Model results such as radon flux are considered in the context of gauging system
performance and may provide limited insight into the behavior of the disposal system in deep
time. Based on potential future radon fluxes, a rancher dose was calculated in deep time to
provide a context for the radon flux results, consistent with the rancher scenario from the first
10,000 years of the model.
Conceptually, deep time will result in a combination of repeated isolation of sediment layers and
more dispersal than modeled. This will cause mixing over ever increasing areas and volumes,
rather than mixing within a closed system. Consequently, concentrations of radionuclides will
decrease with each lake cycle and with each climate cycle. However, the constraints of the model
do not allow lake water concentrations to decrease with each cycle, and sediment concentrations
decrease only because of the additional mass of sediment within which the DU waste is mixed.
In light of the simplifications in the model, the results for the deep time scenario are presented
primarily within the first 100-ky cycle, in which the first intermediate or deep lake will return and
the Federal DU Cell will be obliterated. Consideration of model assumptions should be used
when interpreting results beyond the first 100-ky cycle. Summary statistics lake water
concentrations are presented at 90,000 years, which is the timestep at which the greatest
percentage of lakes is present. The focus of the deep-time results is, consequently, the effects of
dispersal on concentrations of 238U and its progeny in lake water and sediments within the first
100-ky climate cycle, as well as 222Rn flux and rancher dose after the first lake recedes. Progeny
of 238U presented include 230Th and 226Ra. Unless otherwise noted, deep time results are
presented for 1000 realizations in order to capture the temporal changes in these results most
clearly.
Final Report for the Clive DU PA Model
24 November 2015 68
6.5.1 Sedimentation and Lake Timing Results
Thickness of the sediment above the DU waste is shown in Figure 9. The next lake to reach the
elevation of Clive is assumed to occur no sooner than 50 ky into the future, so only eolian
deposition, at a constant (uncertain) rate, contributes to accumulation of sediments in the vicinity
of Clive. Note that the embankment exists until the advent of the first lake, so the eolian
deposition thickness up to 50 ky is the only sediment accumulation in the vicinity of Clive. When
a lake reaches the Clive elevation, eolian deposition is augmented by the deposition of lake-
derived sediments. Because the number and timing of such lakes and the depth of deposited
sediment are uncertain, the variability in sediment thickness after 50 ky is considerably greater
than in the initial 50-ky modeling period. The change in the slope of the sediment thickness curve
at approximately 75 ky reflects the deposition of sediment from deep lakes that often appear at
this time within the 100-ky climate cycle.
The increasing depth of material covering the disposed DU waste over time will result in
attenuation of radon flux. However, this rate of attenuation will be partly offset by the slowly
increasing activity of the radioactive progeny of 238U. Modeling results indicate that sediment
accumulation overwhelms the influence of progeny ingrowth. This is revealed by inspection of
the results of individual model realizations, where radon flux is always highest at the model time
when the first intermediate lake recedes and then decreases over time to the end of the modeling
period. Hence, the time of peak radon flux is equivalent to the time when the first lake to reach
the elevation of Clive (and destroy the embankment by wave action) has just receded from the
location of the below-grade disposed DU waste.
The time when the first intermediate lake returns after 50 ky is modeled as a Poisson process and
varies with each model realization. Approximately 95% of intermediate lakes occur within the
first 90 ky of the simulation. Deep lake start times are modeled as a log-normal distribution
which typically occur before 100 ky but sometime occur after that point in time. As shown in
Figure 10, the likelihood that the first lake to reach Clive has appeared increases with time from
50 ky such that there is approximately a 80% probability that a lake will have appeared by
approximately 80 ky. At that time the advent of an intermediate lake is overtaken by the
probability that a deep lake will begin within this 100-ky climate cycle.
6.5.2 Lake Sediment Concentrations
Results are presented similarly in Table 11 for concentrations of 238U and its progeny in sediment
derived from successive lakes. These results are statistical summaries of lake concentrations at
90 ky. The peak occurrence of a lake across 10,000 realizations, the time at which a lake is most
likely to be present, is at 90 ky. By that point in time, 230Th and 226Ra have ingrown sufficiently
to be present concentrations greater than those of 238U.

Final Report for the Clive DU PA Model
24 November 2015 69
Figure 9. Evolution of sediment thickness in deep time.

As described in the Deep Time White Paper, sediment thickness increase with time at a rate of
about 12m per 100 ky climate cycle.
Final Report for the Clive DU PA Model
24 November 2015 70

Figure 10. Time of appearance of first intermediate lake to reach the Clive elevation.

In the model the first lake is expected to appear in the bottom half of the current climate cycle.
However, more recent work on climate change suggests instead that the first lake might not reach
the elevation of Clive for several 100 ky (see the Deep Time White Paper).

Final Report for the Clive DU PA Model
24 November 2015 71
Table 11. Statistical summary of deep time sediment concentrations at model year 90,000.
Based on 1000 realizations.
25th Percentile Median Mean 95th Percentile
U-238 sediment
concentration
(pCi/g)
1.7E-4 1.8E-3 2.0E-2 9.5E-2
Ra-226 sediment
concentration
(pCi/g)
6.9E-5 1.2E-3 5.0E-3 2.2E-2
Th-230 sediment
concentration
(pCi/g)
7.0E-5 1.2E-3 5.0E-3 2.3E-2

Time history plots of radionuclide concentrations in future lake sediments for 238U and its
progeny 230throium and 226radium are presented in Figure 11, Figure 12, and Figure 13,
respectively, over 2.1 My. These plots show a large increase in concentrations as a consequence
of the first lake event, with subsequent decreases as the sediment load increases.
Final Report for the Clive DU PA Model
24 November 2015 72

Figure 11. Time history of concentrations of uranium-238 in sediments
Final Report for the Clive DU PA Model
24 November 2015 73

Figure 12. Time history of concentrations of thorium-230 in sediments
Final Report for the Clive DU PA Model
24 November 2015 74

Figure 13. Time history of concentrations of radium-226 in sediments

Final Report for the Clive DU PA Model
24 November 2015 75
6.5.3 Lake Water Concentrations
A summary of lake water concentrations of 238U and some of its progeny are presented in Table
12. These results are statistical summaries of lake concentrations at 90 ky, the time at which a
lake is most likely to be present. By that point in time, 230Th and 226Ra have ingrown sufficiently
to be present computable concentrations.
Table 12. Statistical summary of deep time lake concentrations at model year 90,000. Based
on 1000 realizations.
25th Percentile Median Mean 95th Percentile
U-238 lake
concentration
(pCi/L)
1.4E-7 2.1E-5 1.8E-2 1.1E-1
Ra-226 lake
concentration
(pCi/L)
8.5E-3 1.5E-1 5.4E-1 2.4
Th-230 lake
concentration
(pCi/L)
8.7E-3 1.5E-1 5.5E-1 2.4

To illustrate the dynamic nature of lake returns and related lake concentrations a time history plot
of 238U lake concentrations for all 1000 realization is presented in Figure 14. Intermediate lakes
appear as single peaks, whereas deep lakes increase in concentration over their approximately
20-ky cycle.
Time history plots of lake water concentration statistics for 238U and its progeny, 230Th and 226Ra,
are presented in Figure 15, Figure 16, and Figure 17, respectively, across 2.1 My. These are
presented on a log scale to capture the full concentration range. The jagged nature of the plots is
due to the fact that lake water concentrations are zero when there is no lake present, and
intermediate lakes only occur on average 3 times per 100 ky. Peak lake water concentrations
tend to occur near the end of the period of the deep lake, which provides time for the
radionuclides to dissolve into the lake.

Final Report for the Clive DU PA Model
24 November 2015 76

Figure 14. Time history of concentrations of uranium-238 in lake water, 1000 realizations
shown.

Final Report for the Clive DU PA Model
24 November 2015 77

Figure 15. Time history of concentrations of uranium-238 in lake water
Final Report for the Clive DU PA Model
24 November 2015 78

Figure 16. Time history of concentrations of thorium-230 in lake water
Final Report for the Clive DU PA Model
24 November 2015 79

Figure 17. Time history of concentrations of radium-226 in lake water

6.5.4 Radon flux results after the first lake
A statistical summary of radon flux results after the first lake recedes are presented in Table 13.
The time when the first intermediate lake returns after 50 ky is modeled as a Poisson process and
varies with each model realization. Therefore, the time of peak radon flux also varies with each
realization. Mean and median values are below the 10,000-yr timeframe regulatory limit of 20
pCi/m2s. An interpretation of the significance of these concentrations is presented in the
calculations of dose to a ranch worker in Section 6.5.5, below.
Final Report for the Clive DU PA Model
24 November 2015 80
Table 13. Statistical summary of radon-222 flux concentrations after the first lake recedes.
Radon-222 flux after first lake recedes (pCi/m2-s)
simulation scenario mean median
(50th %ile) 95th %ile
radon-222 18 4.0 80
Results are based on 10,000 realizations, seed 2

Radon flux over time is shown in Figure 18. Although radon flux will be highest at times closest
to 50 ky, in most realizations a lake will not have occurred until closer to 60 ky. The change in
the slope of radon flux curve before 100 ky in Figure 18 reflects the deposition of sediment from
a deep lake that appears by this time within the 100-ky climate cycle.
The peak of the mean radon flux shown in Figure 18 is approximately 13 pCi/m2-s. The peak
occurs in the Model at about 65,000 yr. This value is lower than the mean radon flux after the
first lake recedes (above) since that value occurs at various points in time and the mean flux in
Figure 18 is calculated for each point in time.
Although the median and mean sediment thickness track closely (Figure 9), the mean radon
ground surface flux is much larger than the median. This strongly skewed result for radon flux is
a consequence of the nonlinearities inherent in the NRC radon ground surface flux calculation.
These are equations (9) through (12) in NRC (1989), here reproduced without detailed
explanation:

The definitions of variables are available in the NRC Regulatory Guide (1989), but it is clear that
these equations will produce a highly nonlinear result, Jc, which is the ground surface flux of
radon. So, even though all the inputs to the calculation are essentially normal distributions, the
complexity of dividing one by another and involving powers (e.g. ex) and hyperbolic tangent,
produces a nonlinear result.

Final Report for the Clive DU PA Model
24 November 2015 81

Figure 18. 222Rn ground surface flux in deep time.

6.5.5 Rancher radon results after the first lake
Doses to a rancher receptor are calculated to provide a context for the radon flux calculations,
using radon flux after the first lake recedes. The rancher exposure scenario provided the greatest
dose to a receptor in the Model from 10,000 years, so it was used here for comparison. Rancher
dose is less than 1 mrem/yr even at the 95th percentile of the results.
Final Report for the Clive DU PA Model
24 November 2015 82
Table 14. Statistical summary of doses to ranchers after the first lake recedes.
Rancher dose after first lake recedes (mrem/yr)
simulation scenario mean median
(50th %ile) 95th %ile
rancher dose 0.14 0.03 0.62
Results are based on 10,000 realizations, seed 2

One of the objectives of a PA, as defined in the UAC R313-25-9 is site stability. The
performance standard for stability requires the facility must be sited, designed, and closed to
achieve long-term stability to eliminate to the extent practicable the need for ongoing active
maintenance of the site following closure. If the intent is to minimize the need for ongoing active
maintenance, as stated, then obliteration of the Federal DU Cell in deep time achieves this goal,
since concentrations are low and the need to maintain the site disappears completely. In addition,
continued deposition through eolian processes in inter-glacial periods, or through lake deposition
otherwise, will continue to affect the site, either by providing additional cover, or through
continued mixing with newly formed sediment layers.
7.0 Summary
This report has laid out the approach taken to developing the PA model for DU waste disposal
options at the Clive facility, and has presented results of the updated Model (Clive DU PA Model
v1.4) with accompanying sensitivity analyses. The purpose of this section is to summarize the
results, provide additional interpretation of the results, and to compare the results more directly to
performance objectives in a compliance evaluation.
7.1 Interpretation of Results
Important results of the quantitative PA Model can be summarized, given the compliance time
frames of interest, in terms of doses to receptors, groundwater concentrations of soluble
radionuclides, and disposal system evolution in deep time. The DU waste disposal configuration
evaluated in the Model assumed burial below the grade of the area surrounding the embankment.
The Model was run assuming that gullies are static in the simulation period, forming immediately
in the first year.
Doses to all receptors are driven primarily by inhalation exposure to radon, with higher doses
expected when the waste is emplaced closer to the embankment surface. However, groundwater
concentrations of 99Tc would be expected to increase as the waste is emplaced lower in the
disposal facility. These concentrations are driven van Genuchten’s α and molecular diffusivity.
These results highlight the trade-off between disposal configurations that place DU waste higher
or lower in the disposal facility. Transport mechanisms move waste either up into the accessible
environment or down towards groundwater. The modeling results indicate that the groundwater
performance objectives can still be satisfied when DU waste is placed below grade, which also
minimizes dose to receptors on the ground surface.
For the configuration used in the Model v1.4, erosion was included in the model, but it was not
expected to make a difference in the dose results since the waste was buried much deeper (11 m)
Final Report for the Clive DU PA Model
24 November 2015 83
than the gully maximum depth (2 m). The impact of gullies has not been fully developed in terms
of their effect on biotic activity, radon transport, or infiltration.
The ALARA analysis results indicate that population doses are small compared to natural
background radiation dose, and costs are low compared to disposal and other costs. The
population doses are small because the population itself is small, and the doses to any
hypothetical individual in the population are also small. Taking this ALARA approach to site
performance would suggest that this is a good site for disposal of DU waste. There is room for
improvement in this simple ALARA decision analysis. For example, other factors could be
included in the analysis such as transportation and worker safety factors, and the cost per person
rem could be reevaluated. However, the small population, because of the remoteness of the
facility and the low individual doses suggest that the disposal system meets ALARA-based
performance objectives.
The deep-time model should be regarded as heuristic or highly stylized. Nevertheless, it models
the basic concepts of the return of lakes in the Bonneville Basin at or above the elevation of the
Clive facility. A sufficiently deep lake destroys the DU disposal facility, redistributes
radionuclides that have moved above ground into the lake sediment, and repeats the cycles of
radionuclides moving into lake water and settling back into sediment. Sedimentation rates are
about 12 m per 100 ky, and the DU waste is assumed to mix with the sediment across time. There
are several components of this heuristic model that could be regarded as conservative in the sense
of over-predicting concentration in both lake water and lake sediment. For example,
1. In version 1.4 of the model all DU waste is disposed below grade. With this waste
disposal configuration, none of the waste is dispersed directly. Waste material that would
be dispersed under this scenario only includes radionuclides that have transported into the
above grade volume of the disposal system. Note also that eolian deposition occurs until
the first lake returns, in which case the site will be more stable than at present and the
below grade waste will be further below grade. Dispersal of the waste on occurs for the
small fraction of waste that has migrated into the above ground component of the disposal
system. The model does not account for increased wetter and cooler conditions that occur
before the first lake returns and would move radionuclides downward from the
embankment.
2. In the model a lake is assumed to destroy the site when it reaches the Clive elevation,
which can cause mixing of waste in a very shallow lake, a lake that perhaps does not have
sufficient power to destroy the facility. Research into the power needed for a lake to
destroy the facility might indicate the minimum elevation needed for such an event.
3. Sediment mixing is assumed to occur with every lake cycle, even though some lake
cycles might result only in burial with new sediments. The resulting concentrations reflect
concentrations associated with the first lake event, consistent with the timing of the
maximum lake water and lake sediment concentrations.
Peak lake water concentrations of 238U at 90,000 yr average about 2.1E-E pCi/L, even given the
conservatism in the model, with a 95th percentile of about 0.11 pCi/L. The peak of the mean
concentrations of 238U in sediment average about 0.02 pCi/g, with a 95th percentile of about
0.1 pCi/g. Given the simplified and biased model structure, these lake water and sediment
concentrations are substantial overestimates.
Final Report for the Clive DU PA Model
24 November 2015 84
7.2 Comparison to Performance Objectives
Comparisons to performance objectives are presented for doses to ranch workers, since dose to
other receptors are smaller, and groundwater concentration for 99Tc, the radionuclide with
concentrations closest to the GWPL. The evaluations are for waste disposed of below grade and
include erosion. Quantitative performance objectives do not exist for the ALARA analysis or for
the deep-time concentrations endpoints.
The concentrations reported by the PA model represent estimates of the concentration in each
year, or the peak concentration within the 500-yr period of groundwater compliance. The peaks
of those concentrations are collected. Because the groundwater concentration of 99Tc increases
with time, the peak concentrations occur at 500 yr. Since the timing of these peaks in different
realizations is the same, the peak of the mean concentrations is identical to the mean of the peak
concentrations. The 10,000 model realizations provide 10,000 estimates of the peak
concentrations. Summary statistics for the distribution of the mean of the peak 99Tc
concentrations are presented in Table 15. Summary statistics for peak mean groundwater activity
concentration of 99Tc within 500 yr. The mean, median, and 95th percentile values are below the
GWPL.
Table 15. Summary statistics for peak mean groundwater activity concentration of 99Tc
within 500 yr
activity concentration at 500 yr (pCi/L)
radionuclide GWPL
(pCi/L) mean median
(50th %ile) 95th %ile
99Tc 3790 26 0.043 150

The results of the analyses depend critically on the model structure, specification and underlying
assumptions. For example, the release of 99Tc to the environment in the early modeling period
would be restricted if waste containerization were taken into account; and 99Tc inventory
concentrations might be overestimated. The model could be optimized for compliance with
GWPLs and dose performance objectives so that both are fully met.
The dose results for ranch workers are presented in Table 16. Peak mean TEDE for ranch worker:
statistical summary. The statistics represent summaries of the peak mean doses achieved within
10,000 yr. The 95th percentile is analogous to the 95% upper confidence interval of the mean that
is commonly used to represent reasonable maximum exposure conditions in CERCLA risk
assessments. Both the mean and the 95th percentile are much lower than the MOP performance
objective of 25 mrem/yr.
Table 16. Peak mean TEDE for ranch worker: statistical summary
TEDE (mrem in a yr) at 10,000 yr
receptor mean median
(50th %ile) 95th %ile
ranch worker 0.062 0051 0.15

Final Report for the Clive DU PA Model
24 November 2015 85
8.0 Conclusions
Model results are dependent on the model structure, model specification and assumptions upon
which they are based. All conclusions depend on the model structure, specification and
assumptions. Changes in any aspect of the model could cause different results. Within this
context the Clive DU PA Model v1.4 demonstrates that the below-surface-grade configuration
option for the subject DU waste is adequately protective of human health and the environment as
projected for the next 10,000 years. Protectiveness is assessed under Utah Administrative Code
R313-25-9 Section 5(a) by consideration in this PA Model of:
• dose to site-specific receptors,
• concentrations in groundwater (to 500 years),
• ALARA, and
• consideration of deep-time scenarios.
The model was run with the waste buried below grade, beneath extra fill material. It was also run
with gully formation assumed to occur near the beginning of the simulation period. Simplified
summary results for these scenarios are presented in Table 17.
Table 17. Summary of results of the Clive DU PA Model
performance objective
meets
performance
objective?

Dose to MOP below regulatory threshold of 25 mrem in a year Yes
Dose to IHI below regulatory threshold of 500 mrem in a year Yes
Groundwater maximum concentration of 99Tc in 500 years < 3790 pCi/L 2 Yes
ALARA average total population cost equivalent over 10,000 years $61,200
2Groundwater concentrations of all other radionuclides are significantly less than their respective GWPLs.

The configuration evaluated for the Clive DU PA Model v1.4, including erosion, demonstrates
that the disposal facility can adequately protect human health and the environment when
disposing of the subject DU waste:
• all disposal options evaluated exhibit doses that are less than the inadvertent intrusion
performance objective,
• there are clearly disposal configurations for which the predicted doses are less than the
MOP performance objective, and
Final Report for the Clive DU PA Model
24 November 2015 86
• there are disposal options for which groundwater concentrations do not exceed GWPLs.
In addition, the ALARA analysis indicates that ALARA costs from population doses that might
be realized for the duration of the 10 ky model are small. On a per year basis, the ALARA costs
are less than $13 per year at the 95th percentile of total population dose.
The Federal DU cell was assumed to be destroyed by the return of a deep lake. The deep-time
model indicates that concentrations in media such as lake water and sediment will continue to
decrease with each lake and climate cycle and that destruction of the site will lead to dispersal of
radionuclides in the Bonneville Basin.

Final Report for the Clive DU PA Model
24 November 2015 87
9.0 References
Adrian Brown (Adrian Brown Consultants), 1997. Volume I, LARW Infiltration Modeling Input
Parameters and Results, Report 3101B.970515.
Baird R.D., Bollenbacher, M.K., Murphy, E.S., et al. 1990. Evaluation of the Potential Public
Health Impacts Associated with Radioactive Waste Disposal at a Site Near Clive, Utah.
Rogers and Associates Engineering Corporation, Salt Lake City UT.
Benson, C.H., W.H. Albright, D.O. Fratta, J.M. Tinjum, E. Kucukkirca, S. H. Lee, J. Scalia, P. D.
Schlicht, and X. Wang. 2011. Engineered Covers for Waste Containment: Changes in
Engineering Properties & Implications for Long-Term Performance Assessment,
NUREG/CR-7028, Office of Research, U.S. Nuclear Regulatory Commission, Washington,
DC.
Berger, A. and M. F. Loutre, 2002. “An exceptionally long interglacial ahead?” Science, 297:
1287- 1288.
California Regional Water Quality Control Board (CRWQCB). 1990. Water and Sediment
Quality Survey of Selected Inland Saline Lakes. CRWQCB Central Valley Region.
October 1990.
http://www.waterboards.ca.gov/rwqcb5/water_issues/swamp/historic_reports_and_faq_she
ets/bckgrnd_saline_lakes/survey_select_inlandsalinelakes_90.pdf
DOE (US Department of Energy). 1997. Applying the ALARA Process for Radiation Protection
of the Public and Environmental Compliance with 10 CFR Part 834 and DOE 5400.5
ALARA Program Requirements, Volume 1 Discussion, DOE-STD-ALARA1draft. United
States Department of Energy, Washington DC. April 1997.
Embrechts, P., Lindskog, F., and McNeil, A. (2001). Modelling Dependence with Copulas and
Applications to Risk Management, Department of Mathematics, Swiss Federal Institute of
Technology, Zurich.
EnergySolutions, 2012, Utah Low-Level Radioactive Material License (RML UT2300249)
Updated Site-Specific Performance Assessment, October 8, 2012, EnergySolutions, LLC,
Salt Lake City, UTHays, J.D., J. Imbrie, and N.J. Shackleton, 1976, Variations in the
Earth’s orbit; Pacemaker of the Ice Ages, Science, Vol. 194, No. 4270, pp. 1121-1132. (see
p. 1126. )Iman, R.L., and Conover, W.J. (1982). “A Distribution-Free Approach to
Inducing Rank Correlation Among Input Variables,” Communications in Statistics:
Simulation and Computation,11 (3): 311-334.
ICRP. Radiation protection recommendations as applied to the disposal of long-lived solid
radioactive waste: ICRP Publication 81. Annals of the ICRP 28:13-22, 1998.
Myrick, T.E., B.A. Berven, and F.F. Haywood, 1981, State background-radiation levels: Results
of measurements taken during 1975-1979, Oak Ridge National Laboratory report
ORNL/TM-7343, Oak Ridge TN. http://www.osti.gov/scitech/servlets/purl/5801538
Neptune. 2006. Ant Parameter Specifications for the Area 5 and Area 3 RWMS Models. Neptune
and Company, Inc.
Final Report for the Clive DU PA Model
24 November 2015 88
NCRP (National Council on Radiation Protection). 2009. Report 160. Ionizing Radiation
Exposure of the Population of the United States. National Council on Radiation Protection
and Measurements, Washington, D.C. Online at
http://radiology.rsna.org/content/253/2/293.full.pdf.
Neptune 2011. Final Report for the Clive DU PA Model, version 1.0. Submitted to
EnergySolutions, June 1, 2011.
NRC (Nuclear Regulatory Commission), 1989. Calculation of Radon Flux Attenuation by
Earthen Uranium Mill Tailings Covers, U.S. Nuclear Regulatory Commission, Office of
Nuclear Regulatory Research, Task WM 503-4, June 1989.
NRC (U.S. Nuclear Regulatory Commission). 1993. Final Environmental Impact Statement to
Construct and Operate a Facility to Receive, Store, and Dispose of 11e.(2) Byproduct
Material Near Clive, Utah, NUREG-1476, US Nuclear Regulatory Commission,
Washington, DC.
NRC. 1995. Reassessment of NRC’s Dollar Per Person-Rem Conversion Factor Policy,
NUREG-1530, US Nuclear Regulatory Commission, Washington, DC. December 1995.
NRC, 2000. A Performance Assessment Methodology for Low-Level Radioactive Waste Disposal
Facilities. NUREG-1573. Division of Waste Management, Office of Material Safety and
Safeguards, U.S. Nuclear Regulatory Commission, Washington D.C., October 2000
NRC, 2004. Regulatory Analysis Guidelines of the U.S. Nuclear Regulatory Commission,
NUREG/BR-0058, Office of Nuclear Regulatory Research, Revision 4, September 2004.
NRC, 2010. Workshop on Engineered Barrier Performance Related to Low-Level Radioactive
Waste, Decommissioning, and Uranium Mill Tailings Facilities. Nuclear Regulatory
Commission. August 3 – 5, 2010.
NRC. 2015. Reassessment of NRC’s Dollar per Person-Rem Conversion Factor Policy. Draft
report for comment. NUREG-1530, Rev. 1. U.S. Nuclear Regulatory Commission,
Washington, D.C.
Office of Management and Budget, 2003. Regulatory Analysis, Circular No. A-4, September 17,
2003.
Shackleton, N.J., 2000, The 100,000-year Ice-Age cycle identified and found to lag temperature,
carbon dioxide, and orbital eccentricity, Science, Vol. 289(5486), pp. 1897-1902.
Šimůnek, J., M. Šejna, H. Saito, M. Sakai, and M. Th. van Genuchten, 2009, The HYDRUS-1D
Software Package for Simulating the One-Dimensional Movement of Water, Heat, and
Multiple Solutes in Variably-Saturated Media, Department of Environmental Sciences,
University of California Riverside, Riverside, CA.
SWCA 2011. Field Sampling of Biotic Turbation of Soils at the Clive Site, Tooele County, Utah.
SWCA Environmental Consultants Inc. Prepared for EnergySolutions, Salt Lake City, UT.
SWCA, 2013, EnergySolutions. Updated Performance Assessment –SWCA’s Response to First
Round DRC Interrogatories, SWCA Environmental Consultants, Salt Lake City, Utah,
September 2013.
Final Report for the Clive DU PA Model
24 November 2015 89
Utah 2015. License Requirements for Land Disposal of Radioactive Waste. Utah Administrative
Code Rule R313-25. As in effect on September 1, 2015.
UWQB (State of Utah, Division of Water Quality, Utah Water Quality Board), 2009. Ground
Water Quality Discharge Permit No. 450005, 23 Dec 2009.
Whetstone, 2006. EnergySolutions Class A Combined (CAC) Disposal Cell Infiltration and
Transport Modeling Report, Salt Lake City Utah, May 2006.

Final Report for the Clive DU PA Model
24 November 2015 90
List of Appendices
Appendix 1 FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
Appendix 2 Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
Appendix 3 Embankment Modeling
Appendix 4 Waste Inventory
Appendix 5 Unsaturated Zone Modeling
Appendix 6 Geochemical Modeling
Appendix 7 Saturated Zone Modeling
Appendix 8 Air Modeling
Appendix 9 Biological Modeling
Appendix 10 Erosion Modeling
Appendix 11 Dose Assessment
Appendix 12 Decision Analysis (ALARA)
Appendix 13 Deep Time Assessment
Appendix 14 Development of Probability Distributions
Appendix 15 Sensitivity Analysis Methods
Appendix 16 Model Parameters
Appendix 17 Quality Assurance Project Plan
Appendix 18 Radon
Appendix 19 Sensitivity Analysis Results
Appendix 20 Comparison of Results across Models
Appendix 21 Technical Responses to April 2015 Draft SER

 NAC-0020_R2

FEP Analysis for Disposal of
Depleted Uranium at the
Clive Facility
5 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 ii
1. Title: FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
2. Filename: Clive DU PA FEP Analysis v1.4.docx
3. Description: This documents the development and analysis of features, events, and processes
for disposal of depleted uranium at the Clive, Utah Facility.
Name Date
4. Originator Jenifer Linville 28 May 2011
5. Reviewer John Tauxe 28 May 2014
6. Remarks
5 Nov 2015: Updated from v1.2 to v1.4. - D.Levitt.

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 iii

This page intentionally left blank, aside from this statement.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 iv
CONTENTS
TABLES .......................................................................................................................................... v

1.0
 Introduction ............................................................................................................................ 1

2.0
 Identification of Features, Events, and Processes .................................................................. 1

2.1
 Compilation of FEPs ......................................................................................................... 2

2.2
 Normalization and Consolidation of FEPs ....................................................................... 2

3.0
 Classifying Features, Events, and Processes .......................................................................... 3

4.0
 Screening of FEPs .................................................................................................................. 4

4.1
 Regulatory Considerations, Guidance, and Supporting Information ................................ 4

4.1.1
 Nuclear Regulatory Commission: 10 CFR 61 ............................................................ 5

4.1.2
 Utah Administrative Code R313: Radiation Control .................................................. 5

4.1.3
 Additional Guidance ................................................................................................... 6

4.2
 Scope of Assessment and Physical Reasonableness ......................................................... 7

5.0
 Screening Results ................................................................................................................... 7

6.0
 Use of FEPs for Conceptual Model and Scenario Development ........................................... 9

7.0
 References ............................................................................................................................ 13

Appendix: FEP Listings ................................................................................................................ 15

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 v
TABLES
Table 1. List of Initial FEPs by Reference .................................................................................... 16

Table 2. List of consolidated FEPs evaluated for inclusion in the conceptual site model and
scenarios ...................................................................................................................... 49

Table 3. List of FEPs dismissed from further consideration. ........................................................ 57

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 1
1.0 Introduction
The safe storage and disposal of depleted uranium (DU) waste is essential for mitigating releases
of radioactive materials and reducing exposures to humans and the environment. Currently, a
radioactive waste facility located in Clive, Utah (the “Clive facility”) operated by
EnergySolutions is proposed to receive and store DU waste that has been declared surplus from
radiological facilities across the nation. The Clive facility has been tasked with disposing of the
DU waste in an economically feasible manner that protects humans from radiological releases.
To assess whether that the proposed Clive facility DU disposal location and containment
technologies are suitable for protection of human health, specific performance objectives for land
disposal of radioactive waste set forth in Title 10 Code of Federal Regulations Part 61
(10 CFR 61) Subpart C, promulgated by the U.S. Nuclear Regulatory Commission (NRC), must
be met. In order to support the required radiological performance assessment (PA), a detailed
computer model is being developed to evaluate the potential detrimental effects on human health
that would result from the disposal of DU and its associated radioactive contaminants.
A key activity in developing a PA for a radiological waste repository is the comprehensive
identification of relevant external factors that should be included in quantitative analyses. These
factors, termed “features, events, and processes” (FEPs), form the basis for scenarios that are
evaluated to assess site performance.
Although it is not a governing regulation for the disposal of LLW and DU at Clive, Title 40 CFR
Part 191, promulgated by the U.S. Environmental Protection Agency (EPA), provides a useful
and general definition for the scope of a PA analysis of a radiological disposal facility. The PA
1) identifies the processes and events that might affect the disposal system, 2) examines the
effects of these processes and events on the performance of the disposal system, and 3) estimates
the cumulative releases of radionuclides considering the associated uncertainties caused by all
significant processes and events (40 CFR 191). The identification of FEPs is essential to the
development of the conceptual site model (CSM) and model scenario development process (see
Conceptual Site Model white paper).
This report serves to document and examine the universe of FEPs that may apply to the disposal
of depleted uranium (DU) waste at the Clive Facility. FEPs that are screened and identified as
relevant for the Clive facility PA are identified in this white paper and are further elaborated in
the CSM white paper.
This document is considered to be a living document that is synchronized with current
conceptual models, analysis, and modeling of the PA. As concepts and modeling evolve, so too
will this document.
2.0 Identification of Features, Events, and Processes
The identification of FEPs for use in the Clive DU PA Model was an iterative process that began
with compiling an exhaustive list of candidate FEPs that could affect the long-term performance
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 2
of the radiological waste repository. As an initial step, all potentially relevant FEPs from a
variety of reference sources were collected. The initial list from external sources was modified as
additional FEPs were identified that are specific to the Clive facility.
This exhaustive initial compilation of FEPs led to significant redundancy across the original
sources. Redundancy was addressed by the modification of the candidate list of FEPs through
normalization (removal of redundant FEPs) and assignment of FEPs categories (grouping of
common FEPs). This section describes the FEP identification process, including implementation
of the normalization, categorization and screening processes.
2.1 Compilation of FEPs
The initial list of FEPs pertaining to the efficacy of disposal of radioactive wastes in general was
compiled from several scenario development documents published for other nuclear waste
disposal facilities, including those for Yucca Mountain Project, the Waste Isolation Pilot Plant,
and several foreign radioactive waste projects. The primary literature source for FEP analysis is
Guzowski and Newman (1993). They compiled over 700 potentially disruptive FEPs from a
review of scenario documentation from other waste repositories around the world.
The facilities considered in Guzowski and Newman have substantially different geological,
environmental and regulatory settings from those of the Clive facility. Consequently, the
collection of FEPs in Guzowski and Newman provides a substantial list that should be
considered for any PA, but they are also missing FEPs that pertain more particularly to the waste
disposal facility at Clive. Site-specific understanding of the environmental and engineered
attributes of the Clive facility, and the potentially affected region and population, was used to
augment the initial compilation of FEPs.
Additional FEPs were also identified from the Nuclear Energy Agency database (NEA, 2000). In
this initial compilation step, nearly 1,000 FEPs were identified from the literature and site-
specific considerations. Initial FEPs compiled from all sources are listed in Table 1 in the
Appendix.
2.2 Normalization and Consolidation of FEPs
Subsequent to the initial compilation of FEPs, steps were taken to reduce redundancy. Initially,
FEPs were sorted alphabetically and duplicates were deleted. Recorded FEP values that were
different only in vernacular/diction (e.g., “climate change” versus “change in climate”) were
normalized to capture a single primary FEP value for a series of identical or closely-related
concepts. To address duplication of FEPs where similar terminology was stated dissimilarly,
initial FEPs were grouped by keyword content (e.g., “climate” “waste” “groundwater”) and
evaluated for possible normalization or consolidation. Where possible, FEPs were normalized to
a standard terminology.
Similar but not identical FEPs were maintained, to be evaluated as part of the consolidation step.
At this point, each FEP was considered for its similarity to other FEPs, so that they could be
grouped into fewer classes, making the list more manageable. For example, all geochemical
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 3
processes were grouped together. These would be easier to address as a group for inclusion in the
CSM. Likewise, all coastal processes could be considered for exclusion as a group. For each
FEP, the rationale behind its grouping was noted. No FEPs were excluded at this step, but nearly
all were consolidated with others. This consolidation process reduced the total number to 135
unique FEP groupings.
3.0 Classifying Features, Events, and Processes
Following the normalization and consolidation steps, the 135 unique FEP groups were carried
forward to the classification step and were considered for inclusion in the conceptual model
scenarios. The classification is principally an organizational tool for the FEP analysis, although
the categories identified also relate to components of the CSM. The 135 unique FEP groups
were classified into the following 18 categories:
• Celestial
• Celestial
• Climate change
• Containerization
• Contaminant Migration
• Engineered Features
• Exposure
• Hydrology
• Geochemical
• Geological
• Human Processes
• Hydrogeological
• Marine
• Meteorology
• Model Settings
• Other Natural Processes
• Source Release
• Tectonic/Seismic/Volcanic
• Waste

These categories are relevant to the development of scenarios and are integral to the CSM for the
Clive Facility. Occasionally, a FEP could have been classified into more than one category.
However, the overall goal of the FEP analysis is to identify those processes that should be
carried forward into the CSM, and subsequently into the modeling. Provided each FEP is
identified in one of the categories, it was carried forward to the CSM. Ultimately, each FEP was
given due consideration, and the implementation of relevant FEPs in the final modeling was
rather independent of the classification.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 4
4.0 Screening of FEPs
The long list of FEPs was screened in consideration of regulatory concern and professional
judgment based on physical reasonableness, probability of occurrence, severity of consequence,
and assessment scope.
The most basic screening criterion is regulatory concern. Regulatory requirements for
performance of EnergySolutions’ Clive facility are published in 10 CFR 61 and Utah
Administrative Code R313. While the mention of something that can be construed as a feature,
event, or process in the text of a regulation triggers its consideration in this FEP analysis, it does
not mean that the FEP must become part of the PA analysis or modeling.
A subjective element of the FEP screening process is consideration of assessment scope and
physical reasonableness. Physical reasonableness is a professional judgment based on logical
arguments using available data and information to support a conclusion of whether or not
conditions can exist within the period of regulatory concern that will result in the occurrence of a
particular event or process that affects disposal system performance. In addition to meeting
screening criteria, some FEPs were retained as model parameters specifically because they
pertain to scenario development itself (e.g., exposure terms).
The inclusion or dismissal of FEPs and associated rationale is documented in support of
constructing the conceptual model and scenarios. The product of this screening procedure is the
identification of those FEPs that, either alone or in conjunction with others, could affect the
performance of the disposal system.
4.1 Regulatory Considerations, Guidance, and Supporting
Information
This section discusses the regulatory language, guidance, and other supporting information to be
considered in developing scenarios and conceptual models for the Clive DU PA Model. Specific
considerations of NRC’s land disposal performance requirements (10 CFR 61 Subpart C) are
required for the scenario development and are important to document as part of the FEP
compilation and screening activity. In addition, observations and recommendations previously
published by radioactive waste disposal facility working groups and technical advisers are also
considered, although most of these are focused on geologic disposal of radioactive wastes.
Specific provisions of regulations for the operation and closure of a land-disposal LLW facility
were specifically considered if they were mentioned in a regulatory document.
Based on these provisions, 55 of 135 FEPs were identified as relevant for evaluation in the
conceptual model or exposure scenarios. The remaining FEPs were dismissed from further
consideration for various reasons. Some, like a direct impact from a large meteorite, are simply
beyond the scope of the analysis. Tsunami and other marine phenomena do not apply at the Clive
facility. Several FEPs from the original sources were dismissed because they apply only to
geologic repositories, or to specific types of containment like copper canisters for used nuclear
fuel.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 5
4.1.1 Nuclear Regulatory Commission: 10 CFR 61
This regulation contains Federal procedural requirements and performance objectives applicable
to land disposal of radioactive waste. Specific considerations of 10 CFR 61 include attributes of
facility siting, facility engineering (including post-closure stability and control), site monitoring,
record-keeping, protection of health and safety, and a minimum time frame for which an
assessment must be conducted to ensure long-term stability of the disposal site. The types of
objectives mentioned in 10 CFR 61 include:
• long-term effectiveness based on physical siting of the disposal unit (including site
geology and hydrology),
• protection of the general population (in terms of radiological dose),
• protection of inadvertent intruders (dose),
• protection of individuals during operations (dose),
• isolation and segregation of wastes,
• limitation of releases of radionuclides via pathways in air, water, surface water, plant
uptake, or exhumation by burrowing animals,
• long-term stability of the disposal site,
• evaluation of engineering failures, including erosion, mass wasting, slope failure,
settlement of wastes and backfill, infiltration through covers, and surface drainage,
• site monitoring requirements,
• identification of natural resources whose exploitation could result in inadvertent
exposure, and
• efficacy of institutional controls.
4.1.2 Utah Administrative Code R313: Radiation Control
The Utah Administrative Code (UAC) Rules 313-15 (Standards for Protection Against
Radiation) and 313-25 (License Requirements for Land Disposal of Radioactive Waste) mirror
the provisions for land disposal of radioactive waste provided in 10 CFR 61. Notable
performance objectives of near-surface disposal sites established of UAC Rule R313-25 include:
• protection of the general population,
• protection of inadvertent intruders,
• consideration of releases of radionuclides through pathways via air, water, surface water,
plant update, and exhumation of burrowing animals,
• protection of individuals during operations,
• long-term stability of the disposal site,
• prevention of erosion, mass wasting, slope failure, settlement of wastes and backfill,
infiltration through covers, and surface drainage,
• site monitoring requirements, and
• identification of natural resources whose exploitation could result in inadvertent
exposure.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 6
The majority of the FEPs identified as relevant under 10 CFR 61 are also applicable under UAC
Rule R313-25 and are retained for analysis.
4.1.3 Additional Guidance
The NRC’s PA working group has identified additional considerations in NRC’s Performance
Assessment Methodology (NRC 2000). The working group identifies two specific areas of
interest in conducting a PA: pathway analysis and dose assessment.
Pathway analysis involves the mechanisms of radionuclide transfer through the biosphere to
humans. These mechanisms, or transport and exposure pathways, must be identified and
modeled. Pathway analysis should result in the determination of the total intake of radionuclides
by the average member of the critical group. The critical group is defined as the “...group of
individuals reasonably expected to receive the greatest dose from radioactive releases from the
disposal facility over time, given the circumstances under which the analysis would be carried
out” (NRC 2000).
Various considerations should be taken into account when analyzing the transport of
radionuclides through the biosphere (to humans). These considerations should include
• modeling the movement of radionuclides through the environment and the food chain,
adequately reflecting complex symbiotic systems and relationships,
• considering mechanisms of (biotic and) human uptake of radionuclides, and
• identifying usage, production, and consumption parameters, for various food products
and related systems, that may vary widely, depending on regional climate conditions,
local or ethnic diet, and habits.

The dose assessment requires that the dosimetry of the exposed individual be modeled. The
objective of dose modeling in a LLW PA is to provide estimates of potential doses to humans, in
terms of the average member of the critical group, from radioactive releases from a LLW
disposal facility, after closure.
A “current conditions” philosophy is initially applied to determine which pathways are to be
evaluated. That is to say that current regional land use and other local conditions in place at the
time of the analysis will strongly influence pathways that are considered to be significant. The
conceptual model and scenarios must consider each of the general pathways discussed in
10 CFR 61.13. Additional pathways for consideration are published in NUREG/CR-5453
(Shipers, 1989) and NUREG-1200 (NRC, 1994). NUREG-1200 discusses example potential
“scenarios by which radioactivity may be released from the disposal facility and cause the
potential for radiological impacts on individuals.” Shipers (1989) identifies exposure pathways,
and scenarios regarding transport mechanisms that could contribute to the release of radioactive
materials from the disposal facility leading to human exposure, in the context of near-surface
LLW disposal.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 7
4.2 Scope of Assessment and Physical Reasonableness
The final phase of FEP screening is the application of professional judgment in terms of the
scope of the PA and the physical reasonableness of evaluating those FEPs in the CSM and
scenarios. Performance objectives include protection of the general population from releases of
radioactivity (10 CFR 61.41), protection of individuals from inadvertent intrusion (§61.42), and
stability of the site after closure (§61.44). Assumptions of the scope of the PA include:
• Performance assessment reflects post-closure conditions. Because PA considers the site
only after closure, consideration of the protection of individuals during operations
(§61.43) is not within the scope of the evaluation and FEPs related to operations are not
considered relevant to the CSM or scenarios.
• Land-use assumptions relative to human exposures post-closure are based on current
conditions and likely future conditions. Therefore urban settlement, residential use,
farming, and aquaculture and FEPs pertaining to these incongruous uses are not included
in the CSM or scenarios because of the high concentrations of salt in the soil and
groundwater of this site. However, hunting, ranching, and recreational use are considered
viable scenarios.
• Intentional human intruders are not protected.

5.0 Screening Results
Using the identification and screening processes described in Sections 1 through 3, FEPs
were consolidated from an exhaustive list of over 900 to 135 FEPs or FEP
categories. Of this consolidation, 90 FEPs are retained for further consideration
and 45 FEPs were dismissed from inclusion in the PA model. All FEPs
considered and retained for inclusion in the CSM and scenarios are reported in

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 8
Table 2 in the Appendix. FEPs that were considered and dismissed from evaluation in the CSM
and scenarios are listed in Table 3, along with a brief rationale for their exclusion.
In summary, FEPs retained for consideration in the PA, CSM, and scenarios pertain to regulatory
aspects of post-closure protection of human health and long-term stability of the disposal facility
for the duration and spatial scope of the assessment period. FEPs that were dismissed from
consideration in the PA include those that do not fall within the scope of the PA, were
characterized as extremely unlikely to occur or having a low magnitude of consequence of
affecting the performance of the repository, or were dismissed based on site-specific
considerations.

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 9
6.0 Use of FEPs for Conceptual Model and Scenario
Development
The CSM provides detailed descriptions of the physical environment, the engineered disposal
facility, the sources and chemical forms of disposed wastes, potentially affected media, potential
release pathways and exposure routes, and potential receptors. The CSM considers broad
categories of FEPs that are relevant to these attributes, but individual FEPs may or may not be
addressed in the CSM based on the scope of the assessment and the scenarios developed. This
section identifies the FEPs that are considered for inclusion in the CSM and are addressed in the
development of scenarios for the PA model. These are grouped into several categories, and listed
in tabulated form in Appendix B. Those FEPs that were dismissed from consideration in the
modeling are listed in Appendix C. Some FEPs may overlap or repeat between categories.
Meteorology
Frost weathering and other meteorological events (e.g., precipitation, atmospheric dispersion,
resuspension) are considered in the conceptual model. Weathering may occur from frost cycles.
Resuspension of particulates from surface soils allows redistribution by atmospheric dispersion,
which is a meteorological phenomenon. Dust devils are also possible at the site and a tornado
occurred in Salt Lake City in 1999, which was the first tornado in Utah in over 100 years.
Climate change
Features, events, and processes of climate change considered in the conceptual model include
effects on hydrology (including lake effects), hydrogeology, biota, and human behaviors. Lake
effects include appearance/disappearance of large lakes and associated phenomena
(sedimentation, wave action, erosion/inundation). Wave action, including seiches, is included in
the CSM.
Hydrology
Hydrology is addressed in the conceptual model since it influences many processes in
contaminant transport. Examples of FEPs considered for the conceptual model include
groundwater transport, inundation, and water table changes.
Hydrogeological
Several hydrogeological FEPs were identified for consideration in the conceptual model.
Groundwater transport, in both the unsaturated and saturated zones, is potentially a significant
transport pathway. For some model endpoints, such as groundwater concentrations that are
compared to groundwater protection levels (GWPLs), it is the only pathway of concern.
Groundwater flow and transport processes include advection-dispersion, diffusion, fluid
migration, waterborne contaminant transport, changes in the flow system, recharge, water table
movements, and brine interactions. Inundation of the site may occur due to changes in lakes or
reservoirs, which is included in lake effects of climate change.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 10
Geochemical
Geochemical effects include chemical sorption and partitioning between phases, aqueous
solubility, precipitation, chemical stability, complexation, changes in water chemistry (redox
potential, pH, Eh), fluid interactions, speciation, interactions with clays and other host materials,
and leaching of radionuclides from the waste form. These processes are addressed in the model.
Other Natural Processes
The broad category of other natural processes considered for the conceptual model include
ecological changes and pedogenesis (soil formation). Ecological changes are associated with
catastrophic events (e.g., inundation), evolution, or climate change. Pedogenesis is expected on
the cap, giving rise to vegetation growth or habitation by wildlife.
Denudation (cap erosion) may be sufficient to expose waste. Erosion of the repository resulting
from pluvial, fluvial or aeolian processes can result from extreme precipitation, changes in
surface water channels, and weathering. Sediment transport is an inherent aspect of erosion.
Sedimentation/deposition onto the repository would also affect disposal at the site.
Note that seismic activity is unlikely to impact the Clive facility. Faults are not present within the
vicinity of Clive, although effects of isostatic rebound are still possible in the Lake Bonneville
area.
Engineered Features
Engineered features are intended to promote containment and inhibit migration of contaminants.
Conditions potentially affecting site performance include failure of general engineered features,
repository design, repository seals, material properties, and subsidence of the repository.
Containerization
Two key components of containerization were identified as FEPs: containment degradation and
corrosion. Canister degradation, including fractures, fissures, and corrosion (pitting, rusting)
could result in containment failure. These processes are evaluated in the conceptual model
(Conceptual Site Model white paper, Section 8.1).
Waste
Attributes of waste that could influence the performance of the Clive facility include the
inventory of radionuclides, physical and chemical waste forms, container performance, matrix
performance, leaching, radon emanation, and other waste release mechanisms.
Source Release
Source release can result from many mechanisms, including containment failure, leaching, radon
emanation, plant uptake, and translocation by burrowing animals. FEPs that fit in the category of
source release include gas generation, radioactive decay and in-growth, and radon emanation.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 11
Contaminant Migration
Contaminant migration for the CSM includes the mechanisms and processes by which
radionuclides may come to be located outside of the containment unit. The following
contaminant migration processes were identified for consideration in the conceptual model:
resuspension, atmospheric dispersion, biotically-induced transport, contaminant transport,
diffusion, dilution, advection-dispersion, dissolution, dust devils, tornados, infiltration, and
preferential pathways.
Animal ingestion is part of the human exposure model, both as ingestion of fodder and feed by
livestock, and ingestion of livestock by humans. Transport by atmospheric dispersion is modeled
and is associated with limited resuspension, dust devils, and tornados. Modeling of biotic (plant-
and animal-mediated) processes leading to contaminant transport, and the evolution of these
processes in response to climate change and other influences, including bioturbation, burrowing,
root development, and contaminant uptake and translocation are considered. Contaminant
transport includes transport media (water, air, soil), transport processes (advection-dispersion,
diffusion, plant uptake, soil translocation), and partitioning between phases. Diffusion occurs in
gas and water phases. Dilution occurs when mixing with less concentrated water. Hydrodynamic
dispersion is associated with water advection. Dissolution in water is limited by aqueous
solubility. Transport in the gas phase includes gas generation in the waste, partitioning between
air and water phases, diffusion in air and water, and radioactive decay and ingrowth. Infiltration
of water through the cap, into wastes, and potentially to the groundwater is another contaminant
migration concern. Preferential pathways for contaminant transport are also addressed.
Human Processes
The FEPs identified as human processes encompass human behaviors and activities, resource
use, and unintentional intrusion into the repository. Human process FEPs identified for
assessment are related to the human exposure model and include anthropogenic climate change,
human behavior, human-induced processes related to engineered features at the site, human-
induced transport, inadvertent human intrusion, institutional control, land use, post-closure
subsurface activities, waste recovery, water resource management, and weapons training such as
that occurring at nearby bombing ranges.
Exposure
Exposure is an integral part of the conceptual model, and may result from reduced site
performance. Exposure-relevant FEPs identified for evaluation include those related to
dosimetry, exposure media, human exposure, ingestion pathways, and inhalation pathways.
Dosimetry as a science is not a FEP per se but physiological dose response is accounted for in
the PA model.
Transport pathways (e.g. food chains) that lead to foodstuff contamination, and human exposures
due to inhalation of gaseous radionuclides and particulates are included. Exposure media include
are foodstuffs, drinking water, and environmental media. Exposure pathways (ingestion,
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 12
inhalation, etc.) and physiological effects from radionuclides and toxic contaminants (e.g.
uranium) are also assessed.
Model Settings
Model settings that were identified during the FEP compilation process include model
parameterization, period of performance, regulatory requirements, and spatial domain. While
these are not FEPs in and of themselves, they are important considerations in the performance
assessment model and are included with the FEPs for completeness.

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 13
7.0 References
Andersson, J., T. Carlsson, T., F. Kautsky, E. Soderman, and S. Wingefors, 1989. The Joint
SKI/SKB Scenario Development Project. SKB-TR89-35, SvenskKarnbranslehantering Ab,
Stockholm, Sweden.
Burkholder, H.C., 1980. “Waste Isolation Performance Assessment—A Status Report”, in
Scientific Basis for Nuclear Waste Management, Ed. C.J.M. Northrup, Jr., Plenum Press,
New York, NY, Vol. 2, p. 689-702.
Code of Federal Regulations, Title 10, Part 61 (10 CFR 61), Licensing Requirements for Land
Disposal of Radioactive Waste, Government Printing Office, 2007.
Code of Federal Regulations, Title 40, Part 191 (40 CFR 191), Environmental Radiation
Protection Standards for Management and Disposal of Spent Nuclear Fuel, High-Level
and Transuranic Radioactive Waste, Government Printing Office, 1993.
Guzowski, R.V., 1990. Preliminary Identification of Scenarios That May Affect the Escape and
Transport of Radionuclides From the Waste Isolation Pilot Plant, Southeastern New
Mexico, SAND89-7149, Sandia National Laboratories, Albuquerque, NM.
Guzowski, R.V., and G. Newman, 1993, Preliminary Identification of Potentially Disruptive
Scenarios at the Greater Confinement Disposal Facility, Area 5 of the Nevada Test Site,
SAND93-7100, Sandia National Laboratories, Albuquerque, NM.
Hertzler, C.L., and C.L. Atwood, 1989. Preliminary Development and Screening of Release
Scenarios for Greater Confinement Disposal of Transuranic Waste at the Nevada Test Site,
EGG-SARE-8767, EG&G Idaho, Inc., Idaho Falls, ID.
Hunter, R.L., 1983. Preliminary Scenarios for the Release of Radioactive Waste From a
Hypothetical Repository in Basalt of the Columbia Plateau, SAND83-1342 (NUREG/CR-
3353), Sandia National Laboratories, Albuquerque, NM.
Hunter, R.L., 1989. Events and Processes for Constructing Scenarios for the Release of
Transuranic Waste From the Waste Isolation Pilot Plant, Southeastern New Mexico,
SAND89-2546, Sandia National Laboratories, Albuquerque, NM.
Koplik, C.M., M.F. Kaplan, and B. Ross, 1982. “The Safety of Repositories for Highly
Radioactive Wastes,” Reviews of Modern Physics, Vol. 54, no. 1, p. 269-310.
Merrett, GJ., and P.A. Gillespie, 1983. Nuclear Fuel Waste Disposal: Long-Term Stability
Analysis, AECL-6820, Atomic Energy of Canada Limited, Pinawa, Manitoba.
NEA (Nuclear Energy Agency), 1992, Systematic Approach to Scenario Development. A report
of the NEA Working Group on the Identification and Selection of Scenarios for
Performance Assessment of Radioactive Waste Disposal, Nuclear Energy Agency, Paris,
France.
NEA, 2000. Features, Events, and Processes (FEPs) for Geologic Disposal of Radioactive
Waste. An International Database. Nuclear Energy Agency, Organization for Economic
Cooperation and Development.
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 14
NRC (U.S. Nuclear Regulatory Commission), 1994. Standard Review Plan for the Review of a
License Application for a Low-Level Radioactive Waste Disposal Facility, NUREG-1200,
U.S. Nuclear Regulatory Commission, Washington, D.C.
NRC, 2000. A Performance Assessment Methodology for Low-Level Radioactive Waste Disposal
Facilities, NUREG-1573, U.S. Nuclear Regulatory Commission, Washington, D.C.
Shipers, L.R., 1989, Background Information for the Development of a Low-Level Waste
Performance Assessment Methodology, Identification of Potential Exposure Pathways,
NUREG/CR-5453, Vol. 1 , U.S. Nuclear Regulatory Commission, December 1989.

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 15
Appendix: FEP Listings
This appendix lists the features, events, and processes (FEPs) identified for evaluation in
the Conceptual Site Model and Performance Assessment Scenario development.
Table 1 contains all initial FEP values, listed and numbered by reference
document.

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 16
Table 2 lists those FEPs retained for analysis, and Table 3 includes all those FEPs that were
dismissed from further consideration.
Table 1. List of Initial FEPs by Reference
Table 1 (continued)
FEP ID Initial FEP Reference1
1 meteorite Andersson et al., 1989
2 change in sea level Andersson et al., 1989
3 desert and unsaturation Andersson et al., 1989
4 no ice age Andersson et al., 1989
5 glaciation Andersson et al., 1989
6 permafrost Andersson et al., 1989
7 creeping of copper Andersson et al., 1989
8 common cause canister defects - Quality control Andersson et al., 1989
9 cracking along welds Andersson et al., 1989
10 degradation of hole- and shaft seals Andersson et al., 1989
11 electro-chemical cracking Andersson et al., 1989
12 internal pressure Andersson et al., 1989
13 radiation effects on canister Andersson et al., 1989
14 random canister defects - Quality control Andersson et al., 1989
15 reactions with cement pore water Andersson et al., 1989
16 role of chlorides in copper corrosion Andersson et al., 1989
17 thermal cracking Andersson et al., 1989
18 corrosive agents, sulphides, oxygen etc Andersson et al., 1989
19 pitting Andersson et al., 1989
20 stress corrosion cracking Andersson et al., 1989
21 accumulation in peat Andersson et al., 1989
22 colloid generation and transport Andersson et al., 1989
23 colloid generation - source Andersson et al., 1989
24 colloids, complexing agents Andersson et al., 1989
25 accumulation in sediments Andersson et al., 1989
26 loss of ductility Andersson et al., 1989
27 matrix diffusion Andersson et al., 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 17
Table 1 (continued)
FEP ID Initial FEP Reference1
28 saturation of sorption sites Andersson et al., 1989
29 solubility and precipitation Andersson et al., 1989
30 sorption Andersson et al., 1989
31 extreme channel flow of oxidants and nuclides Andersson et al., 1989
32 radiation effects on bentonite Andersson et al., 1989
33 solubility within fuel matrix Andersson et al., 1989
34 thermal buoyancy Andersson et al., 1989
35 thermochemical changes Andersson et al., 1989
36 diffusion - surface diffusion Andersson et al., 1989
37 dilution Andersson et al., 1989
38 dispersion Andersson et al., 1989
39 dissolution chemistry Andersson et al., 1989
40 dissolution of fracture fillings/precipitations Andersson et al., 1989
41 methane intrusion Andersson et al., 1989
42 accumulation of gases under permafrost Andersson et al., 1989
43 gas transport Andersson et al., 1989
44 gas transport in bentonite Andersson et al., 1989
45 flow through buffer/backfill Andersson et al., 1989
46 preferential pathways in the buffer/backfill Andersson et al., 1989
47 poorly designed repository Andersson et al., 1989
48 backfill effects on copper corrosion Andersson et al., 1989
49 backfill material deficiencies Andersson et al., 1989
50 changed hydrostatic pressure on canister Andersson et al., 1989
51 degradation of the bentonite by chemical reactions Andersson et al., 1989
52 erosion of buffer/backfill Andersson et al., 1989
53 excavation/backfilling effects on nearby rock Andersson et al., 1989
54 external stress Andersson et al., 1989
55 hydraulic conductivity change - excavation/backfilling
effect
Andersson et al., 1989
56 hydrostatic pressure on canister Andersson et al., 1989
57 movement of canister in buffer/backfill Andersson et al., 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 18
Table 1 (continued)
FEP ID Initial FEP Reference1
58 thermal effects on the buffer material Andersson et al., 1989
59 voids in the lead filling Andersson et al., 1989
60 swelling of bentonite into tunnels and cracks Andersson et al., 1989
61 swelling of corrosion products Andersson et al., 1989
62 uneven swelling of bentonite Andersson et al., 1989
63 mechanical effects - excavation/backfilling effects Andersson et al., 1989
64 mechanical failure of buffer/backfill Andersson et al., 1989
65 mechanical failure of repository Andersson et al., 1989
66 sudden energy release Andersson et al., 1989
67 coagulation of bentonite Andersson et al., 1989
68 chemical toxicity of wastes Andersson et al., 1989
69 complexing agents Andersson et al., 1989
70 far field hydrochemistry - acids, oxidants. nitrate Andersson et al., 1989
71 change of ground-water chemistry in nearby rock Andersson et al., 1989
72 chemical effects of rock reinforcement Andersson et al., 1989
73 coupled effects (electrophoresis) Andersson et al., 1989
74 effects of bentonite on ground-water chemistry Andersson et al., 1989
75 isotopic dilution Andersson et al., 1989
76 near field buffer chemistry Andersson et al., 1989
77 oxidizing conditions Andersson et al., 1989
78 Pb-I reactions Andersson et al., 1989
79 pH-deviations Andersson et al., 1989
80 recrystallization Andersson et al., 1989
81 redox front Andersson et al., 1989
82 redox potential Andersson et al., 1989
83 diagenesis Andersson et al., 1989
84 accidents during operation Andersson et al., 1989
85 human-induced climate change Andersson et al., 1989
86 non-sealed repository Andersson et al., 1989
87 unsealed boreholes and/or shafts Andersson et al., 1989
88 explosions Andersson et al., 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 19
Table 1 (continued)
FEP ID Initial FEP Reference1
89 geothermal energy production Andersson et al., 1989
90 enhanced rock fracturing Andersson et al., 1989
91 thermo-hydro-mechanical effects Andersson et al., 1989
92 altered surface water chemistry by humans Andersson et al., 1989
93 city on the site Andersson et al., 1989
94 underground dwellings Andersson et al., 1989
95 loss of records Andersson et al., 1989
96 archeological intrusion Andersson et al., 1989
97 postclosure monitoring Andersson et al., 1989
98 underground test of nuclear devices Andersson et al., 1989
99 unsuccessful attempt of site improvement Andersson et al., 1989
100 poorly constructed repository Andersson et al., 1989
101 future boreholes and undetected past boreholes Andersson et al., 1989
102 other future uses of crystalline rock Andersson et al., 1989
103 reuse of boreholes Andersson et al., 1989
104 chemical sabotage Andersson et al., 1989
105 nuclear war Andersson et al., 1989
106 waste retrieval, mining Andersson et al., 1989
107 human-induced actions on ground-water recharge Andersson et al., 1989
108 human-induced changes in surface hydrology Andersson et al., 1989
109 water producing well Andersson et al., 1989
110 weathering of flow paths Andersson et al., 1989
111 erosion on surface/sediments Andersson et al., 1989
112 geothermally induced flow Andersson et al., 1989
113 sedimentation of bentonite Andersson et al., 1989
114 changes of ground-water flow Andersson et al., 1989
115 enhanced ground-water flow Andersson et al., 1989
116 groundwater recharge/discharge Andersson et al., 1989
117 resaturation Andersson et al., 1989
118 saline or fresh ground-water intrusion Andersson et al., 1989
119 river meandering Andersson et al., 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 20
Table 1 (continued)
FEP ID Initial FEP Reference1
120 microbes Andersson et al., 1989
121 repository induced Pb/Cu electrochemical reactions Andersson et al., 1989
122 Gas generation Andersson et al., 1989
123 gas generation: He production Andersson et al., 1989
124 radiolysis Andersson et al., 1989
125 radiolysis Andersson et al., 1989
126 recoil of alpha-decay Andersson et al., 1989
127 reconcentration Andersson et al., 1989
128 chemical reactions (copper corrosion) Andersson et al., 1989
129 I, Cs-migration to fuel surface Andersson et al., 1989
130 interactions with corrosion products and waste Andersson et al., 1989
131 internal corrosion due to waste Andersson et al., 1989
132 natural telluric electrochemical reactions Andersson et al., 1989
133 perturbed buffer material chemistry Andersson et al., 1989
134 radioactive decay; heat Andersson et al., 1989
135 release of radionuclides from failed canister Andersson et al., 1989
136 role of the eventual channeling within the canister Andersson et al., 1989
137 soret effect Andersson et al., 1989
138 earthquakes Andersson et al., 1989
139 faulting Andersson et al., 1989
140 intruding dikes Andersson et al., 1989
141 changes of the magnetic field Andersson et al., 1989
142 stress changes of conductivity Andersson et al., 1989
143 creeping of rock mass Andersson et al., 1989
144 intrusion into accumulation zone in the biosphere Andersson et al., 1989
145 uplift and subsidence Andersson et al., 1989
146 effect of plate movements Andersson et al., 1989
147 tectonic activity - large scale Andersson et al., 1989
148 undetected discontinuities Andersson et al., 1989
149 undetected fracture zones Andersson et al., 1989
150 volcanism Andersson et al., 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 21
Table 1 (continued)
FEP ID Initial FEP Reference1
151 criticality Andersson et al., 1989
152 H2/02 explosions Andersson et al., 1989
153 co-storage of other waste Andersson et al., 1989
154 damaged or deviating fuel Andersson et al., 1989
155 decontamination materials left Andersson et al., 1989
156 near storage of other waste Andersson et al., 1989
157 stray materials left Andersson et al., 1989
158 Meteorites Burkholder, 1980
159 climate modification Burkholder, 1980
160 Glaciation Burkholder, 1980
161 corrosion Burkholder, 1980
162 Transport Agent Introduction Burkholder, 1980
163 fluid migration Burkholder, 1980
164 dissolutioning Burkholder, 1980
165 biochemical gas generation Burkholder, 1980
166 decay product gas generation Burkholder, 1980
167 differential elastic response Burkholder, 1980
168 dewatering Burkholder, 1980
169 canister movement Burkholder, 1980
170 fluid pressure changes Burkholder, 1980
171 material property changes Burkholder, 1980
172 non-elastic response Burkholder, 1980
173 shaft seal failure Burkholder, 1980
174 geochemical alterations Burkholder, 1980
175 diagenesis Burkholder, 1980
176 gas or brine pockets Burkholder, 1980
177 reservoirs Burkholder, 1980
178 undiscovered boreholes Burkholder, 1980
179 Undetected Past Intrusion Burkholder, 1980
180 Intentional Intrusion Burkholder, 1980
181 archeological exhumation Burkholder, 1980
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 22
Table 1 (continued)
FEP ID Initial FEP Reference1
182 irrigation Burkholder, 1980
183 establishment of new population center Burkholder, 1980
184 improper waste emplacement Burkholder, 1980
185 resource mining (mineral hydrocarbon, geothermal, salt) Burkholder, 1980
186 mine shafts Burkholder, 1980
187 sabotage Burkholder, 1980
188 war Burkholder, 1980
189 waste recovery Burkholder, 1980
190 intentional artificial ground-water recharge or withdrawal Burkholder, 1980
191 weapons testing Burkholder, 1980
192 Denudation and Stream Erosion Burkholder, 1980
193 sedimentation Burkholder, 1980
194 flooding Burkholder, 1980
195 radiolysis Burkholder, 1980
196 waste package - geology interactions Burkholder, 1980
197 breccia pipes Burkholder, 1980
198 diapirism Burkholder, 1980
199 far-field faulting Burkholder, 1980
200 near-field faulting Burkholder, 1980
201 faults, shear zones Burkholder, 1980
202 static fracturing Burkholder, 1980
203 impact fracturing Burkholder, 1980
204 surficial fissuring Burkholder, 1980
205 local fracturing Burkholder, 1980
206 Igneous emplacement Burkholder, 1980
207 intrusive magmatic activity Burkholder, 1980
208 hydraulic fracturing Burkholder, 1980
209 isostasy Burkholder, 1980
210 lava tubes Burkholder, 1980
211 Orogenic Diastrophism Burkholder, 1980
212 Epeirogenic Displacement Burkholder, 1980
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 23
Table 1 (continued)
FEP ID Initial FEP Reference1
213 undetected features Burkholder, 1980
214 extrusive magmatic activity Burkholder, 1980
215 criticality Burkholder, 1980
216 chemical liquid waste disposal Burkholder, 1980
217 storage of hydrocarbons or compressed air Burkholder, 1980
218 non-nuclear waste disposal Burkholder, 1980
219 Celestial bodies Guzowski, 1990
220 meteorite impact Guzowski, 1990
221 sea-level variations Guzowski, 1990
222 pluvial periods Guzowski, 1990
223 glaciation Guzowski, 1990
224 seiches Guzowski, 1990
225 formation of dissolution cavities Guzowski, 1990
226 excavation induced stress/fracturing in host rock Guzowski, 1990
227 subsidence and caving Guzowski, 1990
228 thermally induced stress/fracturing in host rock Guzowski, 1990
229 shaft and borehole seal degradation Guzowski, 1990
230 explosions Guzowski, 1990
231 Inadvertent Future Intrusions Guzowski, 1990
232 injection wells Guzowski, 1990
233 irrigation Guzowski, 1990
234 drilling Guzowski, 1990
235 mining Guzowski, 1990
236 damming of streams or rivers Guzowski, 1990
237 withdrawal wells Guzowski, 1990
238 mass wasting Guzowski, 1990
239 erosion/ sedimentation Guzowski, 1990
240 flooding Guzowski, 1990
241 hydrologic stresses Guzowski, 1990
242 hurricanes Guzowski, 1990
243 tsunamis Guzowski, 1990
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 24
Table 1 (continued)
FEP ID Initial FEP Reference1
244 diapirism Guzowski, 1990
245 faulting Guzowski, 1990
246 formation of interconnected fracture systems Guzowski, 1990
247 regional subsidence or uplift (also applies to subsurface) Guzowski, 1990
248 seismic activity Guzowski, 1990
249 magmatic activity Guzowski, 1990
250 volcanic activity Guzowski, 1990
251 meteorite impact Hertzler and Atwood, 1989
252 climatic change Hertzler and Atwood, 1989
253 sea level change Hertzler and Atwood, 1989
254 dam and reservoir formation from natural causes Hertzler and Atwood, 1989
255 glacial activity Hertzler and Atwood, 1989
256 radial dispersion Hertzler and Atwood, 1989
257 fluid interactions Hertzler and Atwood, 1989
258 dissolution Hertzler and Atwood, 1989
259 decay product gas generation Hertzler and Atwood, 1989
260 infiltration and evapotranspiration Hertzler and Atwood, 1989
261 thermal changes in burial zone caused by heat generation Hertzler and Atwood, 1989
262 mechanical effects Hertzler and Atwood, 1989
263 shaft/borehole seal failure Hertzler and Atwood, 1989
264 geochemical changes from natural causes Hertzler and Atwood, 1989
265 diagenesis Hertzler and Atwood, 1989
266 landslide Hertzler and Atwood, 1989
267 local subsidence/caving Hertzler and Atwood, 1989
268 climate control Hertzler and Atwood, 1989
269 fire and explosion Hertzler and Atwood, 1989
270 fire and explosion of waste after burial Hertzler and Atwood, 1989
271 geochemical changes from manmade causes Hertzler and Atwood, 1989
272 earthquake from man-made causes Hertzler and Atwood, 1989
273 human surface activities Hertzler and Atwood, 1989
274 hydrology change from man-made causes Hertzler and Atwood, 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 25
Table 1 (continued)
FEP ID Initial FEP Reference1
275 unanticipated intrusion Hertzler and Atwood, 1989
276 undetected past intrusion Hertzler and Atwood, 1989
277 undetected features or processes Hertzler and Atwood, 1989
278 intentional intrusion Hertzler and Atwood, 1989
279 improper waste emplacement Hertzler and Atwood, 1989
280 mining inadvertent intruder Hertzler and Atwood, 1989
281 dam and reservoir, man-made Hertzler and Atwood, 1989
282 well-drilling inadvertent intruder Hertzler and Atwood, 1989
283 weapons testing Hertzler and Atwood, 1989
284 land erosion Hertzler and Atwood, 1989
285 sedimentation/ aggradation Hertzler and Atwood, 1989
286 lateral ground-water flow in the unsaturated zone Hertzler and Atwood, 1989
287 hydrology change from natural causes Hertzler and Atwood, 1989
288 hurricane Hertzler and Atwood, 1989
289 tornado Hertzler and Atwood, 1989
290 brush fire Hertzler and Atwood, 1989
291 chemical effects Hertzler and Atwood, 1989
292 diapirism Hertzler and Atwood, 1989
293 earthquake from natural causes Hertzler and Atwood, 1989
294 faulting Hertzler and Atwood, 1989
295 igneous activity Hertzler and Atwood, 1989
296 regional subsidence or uplift Hertzler and Atwood, 1989
297 criticality Hertzler and Atwood, 1989
298 chemical liquid waste disposal Hertzler and Atwood, 1989
299 unanticipated waste composition Hertzler and Atwood, 1989
300 permafrost affects repository Hunter, 1983
301 fluids do not recirculate in response to thermal gradients Hunter, 1983
302 fluids leave along new fault Hunter, 1983
303 fluids recirculate in response to thermal gradients Hunter, 1983
304 fluids recirculate in response to thermal gradients Hunter, 1983
305 normal flow increases Hunter, 1983
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 26
Table 1 (continued)
FEP ID Initial FEP Reference1
306 diffusive mixing occurs Hunter, 1983
307 flux through repository is altered Hunter, 1983
308 head is above outfall Hunter, 1983
309 head is below outfall Hunter, 1983
310 subsidence fractures end above repository Hunter, 1983
311 subsidence fractures reach repository Hunter, 1983
312 fluids carry waste to rivers or tributaries Hunter, 1983
313 fluids carry waste to wells or springs Hunter, 1983
314 ground-water flow paths are shortened Hunter, 1983
315 water from a confined aquifer enters repository Hunter, 1983
316 water from the unconfined aquifer enters repository Hunter, 1983
317 location of river channel changes Hunter, 1983
318 location of river channel changes and flow through
repository is altered
Hunter, 1983
319 flow channels close and reopen later Hunter, 1983
320 meteorite impact Hunter, 1989
321 climatic change Hunter, 1989
322 glaciation Hunter, 1989
323 leaching Hunter, 1989
324 diffusion out of the repository Hunter, 1989
325 dissolution Hunter, 1989
326 dissolution other than leaching Hunter, 1989
327 thermal effects Hunter, 1989
328 seal performance Hunter, 1989
329 subsidence Hunter, 1989
330 exhumation Hunter, 1989
331 drilling into repository Hunter, 1989
332 effects of mining for resources Hunter, 1989
333 sabotage Hunter, 1989
334 warfare Hunter, 1989
335 sedimentation Hunter, 1989
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 27
Table 1 (continued)
FEP ID Initial FEP Reference1
336 ground-water flow Hunter, 1989
337 migration of brine aquifer Hunter, 1989
338 migration of intracrystalline brine inclusions Hunter, 1989
339 effects of brine pocket Hunter, 1989
340 gas generation waste effect Hunter, 1989
341 radiolysis waste effect Hunter, 1989
342 waste/rock interaction Hunter, 1989
343 breccia-pipe formation Hunter, 1989
344 induced diapirism Hunter, 1989
345 faulting Hunter, 1989
346 Igneous intrusion Hunter, 1989
347 nuclear criticality Hunter, 1989
348 meteorite impact IAEA 1983
349 climatic change IAEA 1983
350 sea level change IAEA 1983
351 glacial erosion IAEA 1983
352 geochemical change IAEA 1983
353 corrosion IAEA 1983
354 transport agent introduction IAEA 1983
355 fluid interactions IAEA 1983
356 fluid migration IAEA 1983
357 decay-product gas generation IAEA 1983
358 faulty design IAEA 1983
359 exploration bore-hole seal failure IAEA 1983
360 thermal effects IAEA 1983
361 canister movement IAEA 1983
362 fluid pressure, density, viscosity changes IAEA 1983
363 differential elastic response IAEA 1983
364 material property changes IAEA 1983
365 mechanical effects IAEA 1983
366 non-elastic response IAEA 1983
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 28
Table 1 (continued)
FEP ID Initial FEP Reference1
367 shaft seal failure IAEA 1983
368 geochemical change IAEA 1983
369 diagenesis IAEA 1983
370 gas or brine pockets IAEA 1983
371 climate control IAEA 1983
372 reservoirs IAEA 1983
373 inadvertent future intrusion IAEA 1983
374 undetected past intrusion IAEA 1983
375 undiscovered boreholes IAEA 1983
376 Intentional intrusion IAEA 1983
377 archeological exhumation IAEA 1983
378 irrigation IAEA 1983
379 faulty operation IAEA 1983
380 faulty waste emplacement IAEA 1983
381 resource mining (mineral, water, hydrocarbon,
geothermal, salt, etc)
IAEA 1983
382 exploratory drilling IAEA 1983
383 mine shafts IAEA 1983
384 sabotage IAEA 1983
385 war IAEA 1983
386 waste recovery IAEA 1983
387 intentional artificial ground-water recharge or withdrawal IAEA 1983
388 denudation IAEA 1983
389 stream erosion IAEA 1983
390 sedimentation IAEA 1983
391 flooding IAEA 1983
392 ground-water flow IAEA 1983
393 brine pockets IAEA 1983
394 large-scale alterations of hydrology IAEA 1983
395 hydrology change IAEA 1983
396 gas generation IAEA 1983
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 29
Table 1 (continued)
FEP ID Initial FEP Reference1
397 radiolysis IAEA 1983
398 waste package-rock interactions IAEA 1983
399 breccia pipes IAEA 1983
400 diapirism IAEA 1983
401 faulting/seismicity IAEA 1983
402 faults, shear zones IAEA 1983
403 local fracturing IAEA 1983
404 intrusive IAEA 1983
405 intrusive dikes IAEA 1983
406 Isostatic IAEA 1983
407 lava tubes IAEA 1983
408 orogenic IAEA 1983
409 uplift/subsidence IAEA 1983
410 epeirogenic IAEA 1983
411 magmatic activity IAEA 1983
412 extrusive IAEA 1983
413 nuclear criticality IAEA 1983
414 chemical liquid waste disposal IAEA 1983
415 meteorites Koplik et al., 1982
416 climate modification Koplik et al., 1982
417 climatic fluctuations Koplik et al., 1982
418 glaciation Koplik et al., 1982
419 corrosion Koplik et al., 1982
420 biosphere alteration Koplik et al., 1982
421 local fluid migration Koplik et al., 1982
422 dissolutioning Koplik et al., 1982
423 decay product gas generation Koplik et al., 1982
424 Improper design of operation Koplik et al., 1982
425 Thermal effects Koplik et al., 1982
426 canister movement Koplik et al., 1982
427 change in local state of stress Koplik et al., 1982
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 30
Table 1 (continued)
FEP ID Initial FEP Reference1
428 readjustment of rock along joints Koplik et al., 1982
429 fluid pressure changes Koplik et al., 1982
430 canister migration Koplik et al., 1982
431 convection Koplik et al., 1982
432 differential elastic response Koplik et al., 1982
433 material property changes Koplik et al., 1982
434 Mechanical effects Koplik et al., 1982
435 nonelastic response Koplik et al., 1982
436 stored energy Koplik et al., 1982
437 shaft seal failure Koplik et al., 1982
438 seal - rock interactions Koplik et al., 1982
439 subsidence of canister Koplik et al., 1982
440 geochemical alterations Koplik et al., 1982
441 diagenesis Koplik et al., 1982
442 gas or brine pockets Koplik et al., 1982
443 reservoirs Koplik et al., 1982
444 Inadvertent future intrusion Koplik et al., 1982
445 Undetected past intrusion Koplik et al., 1982
446 undiscovered boreholes Koplik et al., 1982
447 Intentional intrusion Koplik et al., 1982
448 archeological exhumation Koplik et al., 1982
449 irrigation Koplik et al., 1982
450 establishment of population center Koplik et al., 1982
451 improper waste emplacement Koplik et al., 1982
452 resource mining (salt, mineral, hydrocarbon, geothermal) Koplik et al., 1982
453 mine shafts Koplik et al., 1982
454 sabotage Koplik et al., 1982
455 war Koplik et al., 1982
456 waste recovery Koplik et al., 1982
457 Perturbation of ground-water system Koplik et al., 1982
458 intentional artificial ground-water recharge or withdrawal Koplik et al., 1982
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 31
Table 1 (continued)
FEP ID Initial FEP Reference1
459 weapons testing Koplik et al., 1982
460 Denudation and stream erosion Koplik et al., 1982
461 Sedimentation Koplik et al., 1982
462 Flooding Koplik et al., 1982
463 Modification of hydrologic regime Koplik et al., 1982
464 gas generation Koplik et al., 1982
465 Radiation effects Koplik et al., 1982
466 radiolysis Koplik et al., 1982
467 Chemical effects Koplik et al., 1982
468 waste package - geology interactions Koplik et al., 1982
469 breccia pipes Koplik et al., 1982
470 diapirism Koplik et al., 1982
471 far-field faulting Koplik et al., 1982
472 near-field faulting Koplik et al., 1982
473 faults, shear zones Koplik et al., 1982
474 Static fracturing Koplik et al., 1982
475 impact fracturing Koplik et al., 1982
476 surficial fissuring Koplik et al., 1982
477 local fracturing Koplik et al., 1982
478 Igneous emplacement Koplik et al., 1982
479 intrusive magmatic activity Koplik et al., 1982
480 hydraulic fracturing Koplik et al., 1982
481 isostasy Koplik et al., 1982
482 lava tubes Koplik et al., 1982
483 Orogenic diastrophism Koplik et al., 1982
484 Epeirogenic displacement Koplik et al., 1982
485 Magmatic activity Koplik et al., 1982
486 extrusive magmatic activity Koplik et al., 1982
487 criticality Koplik et al., 1982
488 storage of hydrocarbons, compressed air, or hot water Koplik et al., 1982
489 non-nuclear waste disposal Koplik et al., 1982
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 32
Table 1 (continued)
FEP ID Initial FEP Reference1
490 chemical liquid waste disposal Koplik et al., 1982
491 Meteorite impact Merrett and Gillespie, 1983
492 determination of meteorite impact frequencies Merrett and Gillespie, 1983
493 probability of meteorite damage Merrett and Gillespie, 1983
494 Glaciation Merrett and Gillespie, 1983
495 glacial erosion Merrett and Gillespie, 1983
496 fracture mechanics analysis Merrett and Gillespie, 1983
497 vault-related events Merrett and Gillespie, 1983
498 presence of a heat source Merrett and Gillespie, 1983
499 excavation Merrett and Gillespie, 1983
500 use of explosive devices Merrett and Gillespie, 1983
501 drilling and mining Merrett and Gillespie, 1983
502 Denudation and fluvial erosion Merrett and Gillespie, 1983
503 denudation Merrett and Gillespie, 1983
504 fluvial erosion Merrett and Gillespie, 1983
505 alteration of hydrological conditions Merrett and Gillespie, 1983
506 new fault formation Merrett and Gillespie, 1983
507 rapid fault growth Merrett and Gillespie, 1983
508 slow fault growth Merrett and Gillespie, 1983
509 stress analysis Merrett and Gillespie, 1983
510 glacially induced faulting Merrett and Gillespie, 1983
511 subsidence and rebound Merrett and Gillespie, 1983
512 Seismic activity Merrett and Gillespie, 1983
513 jointed rock motion Merrett and Gillespie, 1983
514 Volcanic activity Merrett and Gillespie, 1983
515 hot-spot volcanic activity Merrett and Gillespie, 1983
516 rift system volcanic activity Merrett and Gillespie, 1983
517 Presence of a radioactive source Merrett and Gillespie, 1983
518 Meteorite impact NEA OECD, 2000
519 Climate change, Global NEA OECD, 2000
520 Climate change, regional and local NEA OECD, 2000
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 33
Table 1 (continued)
FEP ID Initial FEP Reference1
521 Ecological response to climate changes NEA OECD, 2000
522 Hydrological/hydrogeological response to climate
changes
NEA OECD, 2000
523 Sea Level change NEA OECD, 2000
524 Warm climate effects (tropical and desert) NEA OECD, 2000
525 Glacial and ice sheet effects, local NEA OECD, 2000
526 Periglacial effects NEA OECD, 2000
527 Container materials and characteristics NEA OECD, 2000
528 Atmospheric transport of contaminants NEA OECD, 2000
529 Vegetation NEA OECD, 2000
530 Animal populations NEA OECD, 2000
531 Biological/biochemical processes and conditions (in
geosphere)
NEA OECD, 2000
532 Biological/biochemical processes and conditions (in waste
and EBS)
NEA OECD, 2000
533 Species evolution NEA OECD, 2000
534 Animal, plant and microbe mediated transport of
contaminants
NEA OECD, 2000
535 Colloids. contaminant interactions and transport with NEA OECD, 2000
536 Contaminant transport path characteristics (in geosphere) NEA OECD, 2000
537 Chemical/complexing agents, effects on contaminant
speciation/transport
NEA OECD, 2000
538 Solid-mediated transport of contaminants NEA OECD, 2000
539 Sorption/desorption processes, contaminant NEA OECD, 2000
540 Speciation and solubility, contaminant NEA OECD, 2000
541 Dissolution, precipitation, and crystallization, contaminant NEA OECD, 2000
542 Noble gases NEA OECD, 2000
543 Volatiles and potential for volatility NEA OECD, 2000
544 Gas-mediated transport of contaminants NEA OECD, 2000
545 Geological resources NEA OECD, 2000
546 Geological units, other NEA OECD, 2000
547 Host rock NEA OECD, 2000
548 Repository assumptions NEA OECD, 2000
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 34
Table 1 (continued)
FEP ID Initial FEP Reference1
549 Thermal processes and conditions (in geosphere) NEA OECD, 2000
550 Excavation disturbed zone, host rock NEA OECD, 2000
551 Buffer/backfill materials and characteristics NEA OECD, 2000
552 Other engineered features materials and characteristics NEA OECD, 2000
553 Thermal processes and conditions (in wastes and EBS) NEA OECD, 2000
554 Emplacement of wastes and backfilling NEA OECD, 2000
555 Repository design NEA OECD, 2000
556 Mechanical processes and conditions (in geosphere) NEA OECD, 2000
557 Mechanical processes and conditions (in wastes and
EBS)
NEA OECD, 2000
558 Seals. cavern/tunnel/shaft NEA OECD, 2000
559 Closure and repository sealing NEA OECD, 2000
560 Dose response assumptions NEA OECD, 2000
561 Dosimetry NEA OECD, 2000
562 Drinking water, foodstuffs and drugs, contaminant
concentrations in
NEA OECD, 2000
563 Environmental media, contaminant concentrations in NEA OECD, 2000
564 Impacts or concern NEA OECD, 2000
565 Human characteristics (physiology, metabolism) NEA OECD, 2000
566 Chemical/organic toxin stability NEA OECD, 2000
567 Exposure modes NEA OECD, 2000
568 Non-food products, contaminant concentrations in NEA OECD, 2000
569 Nonradiological toxicity/effects NEA OECD, 2000
570 Radiological toxicity/effects NEA OECD, 2000
571 Radon and radon daughter exposure NEA OECD, 2000
572 Diet and fluid Intake NEA OECD, 2000
573 Food and water processing and preparation NEA OECD, 2000
574 Food chains, uptake of contaminants in NEA OECD, 2000
575 Chemical/geochemical processes and conditions (in
geosphere)
NEA OECD, 2000
576 Chemical/geochemical processes and conditions (In
wastes and
NEA OECD, 2000
577 Organics and potential for organic forms NEA OECD, 2000
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 35
Table 1 (continued)
FEP ID Initial FEP Reference1
578 Diagenesis NEA OECD, 2000
579 Gas sources and effects (in geosphere) NEA OECD, 2000
580 Human influences on climate NEA OECD, 2000
581 Social and Institutional developments NEA OECD, 2000
582 Excavation/construction NEA OECD, 2000
583 Explosions and crashes NEA OECD, 2000
584 Future human action assumptions NEA OECD, 2000
585 Future human behavior (target group) assumptions NEA OECD, 2000
586 Habits (non-diet related behavior) NEA OECD, 2000
587 Leisure and other uses of environment NEA OECD, 2000
588 Human response to climate changes NEA OECD, 2000
589 Surface environment, human activities NEA OECD, 2000
590 Technological developments NEA OECD, 2000
591 Adults, children, Infants and other variations NEA OECD, 2000
592 Human-action-mediated transport of contaminants NEA OECD, 2000
593 Community characteristics NEA OECD, 2000
594 Dwellings NEA OECD, 2000
595 Motivation and knowledge issues (inadvertent/deliberate
human actions)
NEA OECD, 2000
596 Administrative control , repository site NEA OECD, 2000
597 Records and markers, repository NEA OECD, 2000
598 Unintrusive site investigation NEA OECD, 2000
599 Site Investigation NEA OECD, 2000
600 Rural and agricultural land and water use (incl. fisheries) NEA OECD, 2000
601 Urban and Industrial land and water use NEA OECD, 2000
602 Wild and natural land and water use NEA OECD, 2000
603 Monitoring of repository NEA OECD, 2000
604 Remedial actions NEA OECD, 2000
605 Schedule and planning NEA OECD, 2000
606 Quality control NEA OECD, 2000
607 Retrievability NEA OECD, 2000
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 36
Table 1 (continued)
FEP ID Initial FEP Reference1
608 Drilling activities (human intrusion) NEA OECD, 2000
609 Mining and other underground activities (human intrusion) NEA OECD, 2000
610 Accidents and unplanned events NEA OECD, 2000
611 Water management (wells, reservoirs. dams) NEA OECD, 2000
612 Coastal features NEA OECD, 2000
613 Topography and morphology NEA OECD, 2000
614 Erosion and deposition NEA OECD, 2000
615 Erosion and sedimentation NEA OECD, 2000
616 Hydraulic/hydrogeological processes and conditions (in
geosphere)
NEA OECD, 2000
617 Hydraulic/hydrogeological processes and conditions (in
wastes and EBS)
NEA OECD, 2000
618 Hydrological/hydrogeological response to geological
changes
NEA OECD, 2000
619 Hydrothermal activity NEA OECD, 2000
620 Marine features NEA OECD, 2000
621 Soil and sediment NEA OECD, 2000
622 Aquifers and water-bearing features, near surface NEA OECD, 2000
623 Water-mediated transport of contaminants NEA OECD, 2000
624 Hydrological regime and water balance (near-surface) NEA OECD, 2000
625 Lakes, rivers, streams and springs NEA OECD, 2000
626 Atmosphere NEA OECD, 2000
627 Meteorology NEA OECD, 2000
628 Model and data Issues NEA OECD, 2000
629 Timescale of concern NEA OECD, 2000
630 Regulatory requirements and exclusions NEA OECD, 2000
631 Spatial domain or concern NEA OECD, 2000
632 Ecological/biological microbial systems NEA OECD, 2000
633 Microbial/biological/plant-mediated processes, NEA OECD, 2000
634 Gas sources and effects (in wastes and EBS) NEA OECD, 2000
635 Radioactive decay and in-growth NEA OECD, 2000
636 Radiation effects (In wastes and EBS) NEA OECD, 2000
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 37
Table 1 (continued)
FEP ID Initial FEP Reference1
637 Inorganic solids/solutes NEA OECD, 2000
638 Salt diapirism and dissolution NEA OECD, 2000
639 Discontinuities, large scale (in geosphere) NEA OECD, 2000
640 Metamorphism NEA OECD, 2000
641 Deformation, elastic, plastic or brittle NEA OECD, 2000
642 Seismicity NEA OECD, 2000
643 Undetected features (In geosphere) NEA OECD, 2000
644 Tectonic movements and orogeny NEA OECD, 2000
645 Volcanic and magmatic activity NEA OECD, 2000
646 Nuclear criticality NEA OECD, 2000
647 Inventory, radionuclide and other material NEA OECD, 2000
648 Waste form materials and characteristics NEA OECD, 2000
649 Waste allocation NEA OECD, 2000
650 meteorite impact NEA, 1992
651 no ice age NEA, 1992
652 sea-level rise/fall NEA, 1992
653 ecological response to climatic change NEA, 1992
654 glaciation (erosion/deposition, glacial loading,
hydrogeological change)
NEA, 1992
655 periglacial effects (permafrost, high seasonality) NEA, 1992
656 river flow and lake level changes NEA, 1992
657 fracturing NEA, 1992
658 embrittlement and cracking NEA, 1992
659 metallic corrosion (pitting/uniform, internal and external
agents, gas generation e.g. H2)
NEA, 1992
660 animal uptake NEA, 1992
661 plant uptake NEA, 1992
662 uptake by animal, plant, root NEA, 1992
663 uptake by deep rooting species NEA, 1992
664 soil and sediment bioturbation NEA, 1992
665 plant and animal evolution NEA, 1992
666 colloid formation, dissolution and transport NEA, 1992
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 38
Table 1 (continued)
FEP ID Initial FEP Reference1
667 accumulation in soils and organic debris NEA, 1992
668 advection and dispersion NEA, 1992
669 matrix diffusion NEA, 1992
670 multiphase flow and gas driven flow NEA, 1992
671 solubility limit NEA, 1992
672 sorption (linear/non-linear, reversible/irreversible) NEA, 1992
673 non-radioactive solute plume in geosphere (effect on
redox, ph and sorption)
NEA, 1992
674 diffusion NEA, 1992
675 mass, isotopic and species dilution NEA, 1992
676 dissolution, precipitation, and crystallization NEA, 1992
677 natural gas intrusion NEA, 1992
678 gas flow NEA, 1992
679 gas mediated transport NEA, 1992
680 inadequate backfill or compaction voidage NEA, 1992
681 dewatering of host rock NEA, 1992
682 common cause failures NEA, 1992
683 investigation borehole seal failure and degradation NEA, 1992
684 stress field changes, settling, subsidence or caving NEA, 1992
685 thermal effects (concrete hydration) NEA, 1992
686 Thermal (nuclear and chemical) NEA, 1992
687 canister or container movement NEA, 1992
688 changes in in-situ stress field NEA, 1992
689 subsidence / collapse NEA, 1992
690 differential elastic response NEA, 1992
691 material defects (e.g. early canister failure) NEA, 1992
692 material property changes NEA, 1992
693 Mechanical NEA, 1992
694 non-elastic response NEA, 1992
695 Design and construction NEA, 1992
696 design modification NEA, 1992
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 39
Table 1 (continued)
FEP ID Initial FEP Reference1
697 shaft or access tunnel seal failure and degradation NEA, 1992
698 altered soil or surface water chemistry NEA, 1992
699 chemical transformations NEA, 1992
700 chemical gradients (electrochemical effects and osmosis) NEA, 1992
701 complexing agents NEA, 1992
702 diagenesis NEA, 1992
703 land slide NEA, 1992
704 accidents during operation NEA, 1992
705 agricultural and fisheries practice changes NEA, 1992
706 anthropogenic climate changes (greenhouse effect) NEA, 1992
707 abandonment of unsealed repository NEA, 1992
708 poor closure NEA, 1992
709 tunneling NEA, 1992
710 underground construction NEA, 1992
711 geothermal energy production NEA, 1992
712 repository flooding during operation NEA, 1992
713 co-disposal of reactive wastes (deliberate) NEA, 1992
714 undetected past intrusions (boreholes, mining) NEA, 1992
715 injection of liquid wastes NEA, 1992
716 loss of records NEA, 1992
717 archeological investigation NEA, 1992
718 irrigation NEA, 1992
719 demographic change, urban development NEA, 1992
720 land use changes NEA, 1992
721 post-closure monitoring NEA, 1992
722 underground nuclear testing NEA, 1992
723 effects of phased operation NEA, 1992
724 Operation and closure NEA, 1992
725 poor quality construction NEA, 1992
726 radioactive waste disposal error NEA, 1992
727 Post-closure surface activities NEA, 1992
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 40
Table 1 (continued)
FEP ID Initial FEP Reference1
728 exploitation drilling NEA, 1992
729 exploratory drilling NEA, 1992
730 resource mining NEA, 1992
731 quarrying, near surface extraction NEA, 1992
732 sabotage NEA, 1992
733 malicious intrusion (sabotage, act of war) NEA, 1992
734 recovery of repository materials NEA, 1992
735 recovery of repository materials NEA, 1992
736 ground-water abstraction NEA, 1992
737 dams and reservoirs, built/drained NEA, 1992
738 coastal erosion and estuarine development NEA, 1992
739 denudation (eolian and fluvial) NEA, 1992
740 chemical denudation and weathering NEA, 1992
741 freshwater sediment transport and deposition NEA, 1992
742 fracture mineralization and weathering NEA, 1992
743 rock heterogeneity (permeability, mineralogy), affecting
water and
NEA, 1992
744 river, stream, channel erosion (downcutting) NEA, 1992
745 marine sediment transport and deposition NEA, 1992
746 extremes of precipitation, snow melt and associated
flooding
NEA, 1992
747 effects at saline-freshwater interface NEA, 1992
748 ground-water conditions (saturated/unsaturated) NEA, 1992
749 ground-water discharge (to surface water, springs, soils,
wells, and marine)
NEA, 1992
750 ground-water flow (Darcy, non-Darcy, intergranular
fracture,
NEA, 1992
751 recharge to ground water NEA, 1992
752 saline or freshwater intrusion NEA, 1992
753 natural thermal effects NEA, 1992
754 induced hydrological changes (fluid pressure, density
convection, viscosity)
NEA, 1992
755 site flooding NEA, 1992
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 41
Table 1 (continued)
FEP ID Initial FEP Reference1
756 rivers rechanneled NEA, 1992
757 river meander NEA, 1992
758 frost weathering NEA, 1992
759 solar insolation NEA, 1992
760 coastal surge, storms, and hurricanes NEA, 1992
761 precipitation, temperature, soil, water balance NEA, 1992
762 ecological change (ex. forest fire cycles) NEA, 1992
763 microbial interactions NEA, 1992
764 microbiological (effects on corrosion/degradation,
solubility/complexation, gas generation, ex. CH.C02)
NEA, 1992
765 pedogenesis NEA, 1992
766 gas effects (pressurization, disruption, explosion, fire) NEA, 1992
767 radioactive decay and ingrowth (chain decay) NEA, 1992
768 radiolysis NEA, 1992
769 Radiological NEA, 1992
770 heterogeneity of waste forms (chemical, physical) NEA, 1992
771 cellulosic degradation NEA, 1992
772 interactions of host materials and ground water with
repository material (ex. concrete carbonation, sulphate
attack)
NEA, 1992
773 interactions of waste and repository materials with host
materials (electrochemical corrosive agents)
NEA, 1992
774 introduced complexing agents and cellulosics NEA, 1992
775 induced chemical changes (solubility sorption, species
equilibrium, mineralization)
NEA, 1992
776 diapirism NEA, 1992
777 fault activation NEA, 1992
778 fault generation NEA, 1992
779 host rock fracture aperture changes NEA, 1992
780 metamorphic activity NEA, 1992
781 changes in the earth's magnetic field NEA, 1992
782 uplift and subsidence (orogenic, isostatic) NEA, 1992
783 seismicity NEA, 1992
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 42
Table 1 (continued)
FEP ID Initial FEP Reference1
784 plate movement/tectonic change NEA, 1992
785 undetected features (faults, fracture networks, shear
zones, brecciation, gas pockets)
NEA, 1992
786 magmatic activity (intrusive, extrusive) NEA, 1992
787 nuclear criticality NEA, 1992
788 inadvertent inclusion of undesirable materials NEA, 1992
789 Recurrance of Lake Bonneville Neptune
790 Wave action Neptune
791 Animal burrowing Neptune
792 Dust devils Neptune
793 Barrier stability during inundation Neptune
794 inhalation pathways Neptune
795 human induced hydraulic fracturing Neptune
796 natural hydraulic fracturing (hydrogeological) Neptune
797 Sedimentation Neptune
798 Inundation Neptune
799 radon emanation Neptune
800 natural hydraulic fracturing (tectonic/seismic/volcanic) Neptune
801 Off-Site Residents: impacts on the site by people who
might use the area but don’t live on the site itself.
Neptune
802 On-Site Residents: water well with desalinization;
construction-related activities like basements, footings,
and utilities; enhanced infiltration from septic; altered
plant/animal communities; effect of grading on infiltration;
effect of buildings and pavement on evapotranspiration.
Neptune
803 Agricultural activities Neptune
804 Explosions and Crashes related to plane crashes, bombs Neptune
805 Accidental Intrusion, facility properties intact: mineral, oil
and gas, geothermal or other resource exploration; water
well with desalinization; construction-related activities
Neptune
806 Accidental Intrusion, facility properties altered due to prior
volcanic or seismic event
Neptune
807 FEPs related to post-closure inhabitation of the area Neptune
808 Deliberate Intrusion (purposeful waste retrieval;
archeology; terrorism, etc)
Neptune
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 43
Table 1 (continued)
FEP ID Initial FEP Reference1
809 FEPs related to post-closure intrusion by nonresidents
who come looking for something, or to some kind of major
accident like a plane crash either before or after closure
Neptune
810 meteorite Prij et al. 1991
811 climatic variability Prij et al. 1991
812 minor climatic changes Prij et al. 1991
813 sea-level changes Prij et al. 1991
814 ecological response to climate Prij et al. 1991
815 glaciation Prij et al. 1991
816 periglacial effects Prij et al. 1991
817 canister defects Prij et al. 1991
818 common cause (canister) failures Prij et al. 1991
819 fracturing Prij et al. 1991
820 embrittlement, cracking Prij et al. 1991
821 metallic corrosion Prij et al. 1991
822 bioturbation of soil sediment Prij et al. 1991
823 radiocolloid formation Prij et al. 1991
824 accumulation in soils, organic debris Prij et al. 1991
825 transport of radionuclides Prij et al. 1991
826 advection and dispersion Prij et al. 1991
827 matrix diffusion Prij et al. 1991
828 multiphase flow Prij et al. 1991
829 leaching of nuclides Prij et al. 1991
830 non-radioactive solute in geosphere Prij et al. 1991
831 diffusion Prij et al. 1991
832 dilution of mass Prij et al. 1991
833 dissolution/precipitation reactions Prij et al. 1991
834 natural gas intrusion Prij et al. 1991
835 gas mediated transport Prij et al. 1991
836 inadequate backfill compaction, voidage Prij et al. 1991
837 convergence of openings Prij et al. 1991
838 dewatering of host rock Prij et al. 1991
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 44
Table 1 (continued)
FEP ID Initial FEP Reference1
839 stress field changes Prij et al. 1991
840 thermal effects Prij et al. 1991
841 Thermal Prij et al. 1991
842 degradation of buffer/backfill Prij et al. 1991
843 canister or container movement Prij et al. 1991
844 changes in in-situ stress field Prij et al. 1991
845 readjustment of host rock along joints Prij et al. 1991
846 heat production Prij et al. 1991
847 fracture aperture changes Prij et al. 1991
848 canister migration Prij et al. 1991
849 dehydration of salt minerals Prij et al. 1991
850 differential elastic response Prij et al. 1991
851 material defects Prij et al. 1991
852 swelling of backfill (clay) Prij et al. 1991
853 swelling of corrosion products Prij et al. 1991
854 material property changes Prij et al. 1991
855 Mechanical Prij et al. 1991
856 non-elastic response Prij et al. 1991
857 release of stored energy Prij et al. 1991
858 Design and construction Prij et al. 1991
859 design modification Prij et al. 1991
860 seal failure Prij et al. 1991
861 subsidence, collapse Prij et al. 1991
862 alteration of soil, surface water chemistry Prij et al. 1991
863 Geochemical Prij et al. 1991
864 chemical transformations Prij et al. 1991
865 ionic strength Prij et al. 1991
866 speciation equilibrium reactions Prij et al. 1991
867 texture Prij et al. 1991
868 acidity Prij et al. 1991
869 adsorption and desorption reactions Prij et al. 1991
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 45
Table 1 (continued)
FEP ID Initial FEP Reference1
870 chemical equilibrium reactions Prij et al. 1991
871 counter, competitive, and potential determining ions Prij et al. 1991
872 physico-chemical characteristics influencing chemical
equilibria
Prij et al. 1991
873 redox conditions Prij et al. 1991
874 geochemical alterations Prij et al. 1991
875 diagenesis Prij et al. 1991
876 land slide Prij et al. 1991
877 accidents during operation Prij et al. 1991
878 agricultural developments and changes Prij et al. 1991
879 anthropogenic climate changes (greenhouse effect) Prij et al. 1991
880 abandonment of unsealed repository Prij et al. 1991
881 poor closure Prij et al. 1991
882 tunneling Prij et al. 1991
883 underground construction Prij et al. 1991
884 fisheries developments and changes Prij et al. 1991
885 geothermal energy production Prij et al. 1991
886 co-disposal of reactive wastes (deliberate) Prij et al. 1991
887 Human Induced Phenomena Prij et al. 1991
888 undetected past intrusions Prij et al. 1991
889 injection of fluids Prij et al. 1991
890 loss of records Prij et al. 1991
891 archaeological investigation Prij et al. 1991
892 irrigation Prij et al. 1991
893 changes in land use Prij et al. 1991
894 demographic developments and changes Prij et al. 1991
895 urban developments and changes Prij et al. 1991
896 post-closure monitoring Prij et al. 1991
897 underground nuclear testing Prij et al. 1991
898 Operation and closure Prij et al. 1991
899 phased operation effects Prij et al. 1991
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 46
Table 1 (continued)
FEP ID Initial FEP Reference1
900 attempt of site Improvement Prij et al. 1991
901 poor quality construction Prij et al. 1991
902 improper waste emplacement Prij et al. 1991
903 radioactive waste disposal error Prij et al. 1991
904 Post-closure sub-surface activities Prij et al. 1991
905 Post-closure subsurface activities (intrusion) Prij et al. 1991
906 Post-closure surface activities Prij et al. 1991
907 exploitation drilling Prij et al. 1991
908 exploratory drilling Prij et al. 1991
909 resource mining Prij et al. 1991
910 quarrying, surface mining Prij et al. 1991
911 sabotage Prij et al. 1991
912 malicious intrusion, sabotage/war Prij et al. 1991
913 ground-water abstraction/recharge Prij et al. 1991
914 construction of dams/reservoirs Prij et al. 1991
915 drainage of dams reservoirs Prij et al. 1991
916 coastal erosion development of estuaries Prij et al. 1991
917 denudation, erosion Prij et al. 1991
918 channel erosion Prij et al. 1991
919 chemical denudation Prij et al. 1991
920 channeling and preferential pathways Prij et al. 1991
921 effects on suberosion Prij et al. 1991
922 sediment transport Prij et al. 1991
923 solifluction Prij et al. 1991
924 rock heterogeneity Prij et al. 1991
925 subrosion Prij et al. 1991
926 flooding of repository during operation Prij et al. 1991
927 extreme precipitation Prij et al. 1991
928 flooding of site Prij et al. 1991
929 changes in ground-water system Prij et al. 1991
930 ground-water conditions Prij et al. 1991
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 47
Table 1 (continued)
FEP ID Initial FEP Reference1
931 ground-water discharge Prij et al. 1991
932 ground-water flow Prij et al. 1991
933 ground-water recharge Prij et al. 1991
934 saline-freshwater interface Prij et al. 1991
935 brine migration Prij et al. 1991
936 natural thermal effects Prij et al. 1991
937 induced hydrological changes Prij et al. 1991
938 changes in river regime, lake levels Prij et al. 1991
939 intrusion of saline/fresh water Prij et al. 1991
940 rechanneling of rivers Prij et al. 1991
941 meandering of river Prij et al. 1991
942 water table changes Prij et al. 1991
943 frost weathering Prij et al. 1991
944 solar insolation Prij et al. 1991
945 coastal surge, storms Prij et al. 1991
946 precipitation, temperature, soil, water balance Prij et al. 1991
947 temperature Prij et al. 1991
948 ecological response to sudden change (forest fires) Prij et al. 1991
949 evolution Prij et al. 1991
950 microbial interactions Prij et al. 1991
951 microbiological effects Prij et al. 1991
952 pedogenesis Prij et al. 1991
953 gas generation, explosions Prij et al. 1991
954 gas generation effects Prij et al. 1991
955 radioactive decay/ingrowth Prij et al. 1991
956 Radiological Prij et al. 1991
957 radiolysis Prij et al. 1991
958 heterogeneity of waste forms; chemical or physical Prij et al. 1991
959 cellulosic degradation Prij et al. 1991
960 electrochemical reactions Prij et al. 1991
961 introduced complexing agents, cellulosics Prij et al. 1991
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 48
Table 1 (continued)
FEP ID Initial FEP Reference1
962 material interactions Prij et al. 1991
963 redox potential, pH Prij et al. 1991
964 induced chemical changes Prij et al. 1991
965 diapirism, halokinesis Prij et al. 1991
966 fault activation Prij et al. 1991
967 fault generation Prij et al. 1991
968 fracturing Prij et al. 1991
969 metamorphic activity Prij et al. 1991
970 changes in magnetic field Prij et al. 1991
971 creep of rock Prij et al. 1991
972 uplift and subsidence Prij et al. 1991
973 seismicity Prij et al. 1991
974 undetected geological features Prij et al. 1991
975 plate tectonics Prij et al. 1991
976 undetected features Prij et al. 1991
977 magmatic activity Prij et al. 1991
978 nuclear criticality Prij et al. 1991
979 inadvertent inclusion of undesirable materials Prij et al. 1991
980 radon emanation Neptune
981 resuspension Neptune
1 References for Andersson et al. (1989), Burkholder (1980), Guzowski (1990), Hertzler and Atwood (1989),
Hunter (1983), Hunter, (1989), IAEA (1983), Koplik et al. (1982), Merrett and Gillespie, NEA (1992) and Prij et al.
(1991) were found in Guzowski and Newman (1993).

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 49
Table 2. List of consolidated FEPs evaluated for inclusion in the conceptual site model and
scenarios
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
Climate change climate change Climate change can have a large influence on
site performance. Climate change includes
natural and anthropogenic changes and its
effects on hydrology (including lake effects),
hydrogeology, glaciation, biota, and human
behaviors.
2, 3, 4, 159, 221,
222, 252, 253,
254, 321, 349,
350, 416, 417,
519, 520, 521,
522, 523, 524,
651, 652, 653,
811, 812, 813,
814
lake effects A large lake could have detrimental effects on
the repository. Lake effects include appearance/
disappearance of large lakes and associated
phenomena (sedimentation, wave action,
erosion/inundation, isostasy). This is covered
within climate change scenarios.
Regulations suggest consideration.
656, 789
wave action Wave action, including seiches, could influence
site performance and is included in long-term
scenarios. See lake effects and
erosion/inundation.
224, 790
Containerization containment
degradation
A number of processes can contribute to
degradation of waste containment. These are
accounted for in release of the source term. It is
expected that no credit will be given to
containment.
Regulations suggest consideration.
7, 8, 9, 10, 11,
12, 13, 14, 15,
16, 17, 352, 496,
527, 657, 658,
817, 818, 819,
820
corrosion Corrosion is one of the processes that would
contribute to degradation of waste containment.
Regulations suggest consideration.
18, 19, 20, 161,
353, 419, 659,
821
Contaminant
Migration
biotically-
induced
transport
Plant uptake and burrow excavation are
potential contaminant transport (CT) pathways.
Modeling includes biotic (plant- and animal-
mediated) processes leading to contaminant
transport, and the evolution of these processes
in response to climate change and other
influences, including bioturbation, burrowing,
root development, and contaminant uptake and
translocation.
Regulations suggest consideration.
21, 420, 529,
530, 531, 532,
533, 534, 661,
662, 663, 664,
665, 791, 822
colloid transport Colloid formation could be a CT pathway. This
process will be considered in the geochemistry
conceptual model.
22, 23, 24, 535,
666, 823
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 50
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
contaminant
transport
CT is a large class of processes that govern the
migration of contaminants in the environment,
including transport media (water, air, soil)
processes (advection-dispersion, diffusion, plant
uptake, soil translocation) and partitioning
between phases; much overlap with
atmospheric, groundwater, surface water, and
biotically-induced transport.
Regulations suggest consideration.
25, 26, 27, 28,
29, 30, 31, 32,
33, 34, 35, 162,
163, 257, 301,
302, 303, 304,
305, 323, 354,
355, 356, 421,
536, 537, 538,
539, 540, 667,
668, 669, 670,
671, 672, 673,
824, 825, 826,
827, 828, 829,
830
diffusion Diffusion is a basic CT process that could affect
performance. Diffusion occurs in gas and water
phases.
36, 306, 324,
674, 831
dilution Dilution is a basic CT process that could affect
performance. Dilution occurs when mixing with
less concentrated water.
37, 675, 832
dispersion Dispersion is a basic CT process that could
affect performance. Hydrodynamic dispersion is
associated with water advection.
38
dissolution Dissolution will govern leaching of the waste
form into water, limited by aqueous solubility.
39, 40, 164, 225,
258, 325, 326,
422, 541, 676,
833
dust devils Dust devils are common on the flats, and could
disperse contaminants. These are included in
atmospheric dispersion.
792
gas transport Radon produced in the waste is likely to be
transported via gaseous diffusion. Transport in
the gas phase includes gas generation in the
waste, partitioning between air and water
phases, diffusion in air and water, and
radioactive decay and ingrowth.
42, 43, 44, 165,
166, 259, 357,
423, 542, 543,
544, 678, 679,
835
infiltration Infiltration through the cap materials, the waste,
and unsaturated zone could be an important CT
mechanism. This includes infiltration of meteoric
water (precipitation minus abstractions) through
the cap, into wastes, and potentially to the
groundwater.
45, 260, 307
local geology This feature will control some aspects of CT and
is included implicitly in other processes.
Regulations suggest consideration.
545, 546, 547
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 51
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
preferential
pathways
Preferential pathways could contribute to CT.
Their presence is accounted for in the definition
of advective and diffusive processes.
Regulations suggest consideration.
46
Engineered
Features
compaction
error
Inadequate compaction could result in
subsidence. This overlaps with subsidence and
closure failure.
680, 836
engineered
features
Many engineered features are intended to
improve performance. This large collection of
features is intended to promote containment and
inhibit migration of contaminants.
Regulations suggest consideration.
48, 49, 50, 51,
52, 53, 54, 55,
56, 57, 58, 59,
167, 168, 169,
170, 226, 227,
228, 261, 308,
309, 327, 359,
360, 361, 362,
363, 425, 426,
427, 428, 429,
430, 431, 432,
497, 498, 548,
549, 550, 551,
552, 553, 554,
555, 681, 682,
683, 684, 685,
686, 687, 688,
689,690
material
properties
Material properties are an essential feature of
any model, and include density, porosity,
hydraulic conductivity, permeability, texture,
tortuosity, etc. of waste, backfill, cap materials,
and naturally occurring materials.
60, 61, 62, 171,
364, 433, 692,
852, 853, 854
repository
design
Respository design clearly influences its
performance. This is accounted for implicitly in
the modeling of the repository.
Regulations suggest consideration.
695, 696, 858,
859
source release Source release is an essential part of the model,
and can result from many mechanisms,
including containment failure, leaching, radon
emanation, plant uptake, and translocation by
burrowing animals.
128, 129, 130,
131, 132, 133,
134, 135, 136,
137, 196, 291,
342, 398, 467,
468, 637, 770,
771, 772, 773,
774, 775, 958,
959, 960, 961,
962, 963, 964
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 52
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
subsidence of
repository
Subsidence can compromise performance,
leading to failure of the cap, and enhanced
infiltration.
Regulations suggest consideration.
310, 311, 329,
439, 861
waste Waste form and inventory are essential parts of
the model. Inventory and source release
includes initial inventory of radionuclides and its
physical and chemical form, container
performance, matrix performance, leaching, and
other release mechanisms.
517, 647, 648,
649
Exposure animal
ingestion
Human ingestion of livestock and game exposed
to contaminants is an exposure pathway, and is
implemented as part of the human exposure
model, as ingestion of fodder and feed by
livestock, and ingestion of livestock by humans,
and similar pathways for game.
Regulations suggest consideration.
660
dosimetry Dosimetry hints at human dose response, which
is an integral part of PA. Physiological dose
response will be estimated in the PA model.
Dosimetry as a science is not a FEP, per se.
Regulations suggest consideration.
560, 561
exposure
media
Exposure media are a fundamental part of
exposure pathways, and include foodstuffs,
drinking water, other environmental media.
These are included in the human exposure
model.
Regulations suggest consideration.
562, 563
human
behavior
Behavior is part of human exposure pathway.
Future human behaviors include activities and
their frequency and duration, distinct from food
and water ingestion.
Regulations suggest consideration.
584, 585, 586,
587, 588
human
exposure
Human exposure, in terms of dose and toxicity,
is considered in the model, and includes
exposure pathways (ingestion, inhalation, etc.)
and physiological effects from radionuclides and
toxic contaminants.
Regulations suggest consideration.
68, 564, 565,
566, 567, 568,
569, 570, 571,
801, 802
ingestion
pathways
Ingestion of food, water, and soils are modeled
human exposure pathways. These include
human exposures due to ingestion of water and
foodstuffs, and transport pathways (e.g. food
chains) that lead to foodstuffs.
Regulations suggest consideration.
572, 573, 574
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 53
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
inhalation
pathways
Inhalation of gases and fine particles are
modeled human exposure pathways.
Regulations suggest consideration.
794
Geochemical geochemical
effects
Geochemical processes control CT in waste
sources, water, and geologic media. These
include chemical sorption and partitioning
between phases, aqueous solubility,
precipitation, chemical stability, complexing,
changes in water chemistry (redox potential,
pH, Eh), fluid interactions, halokinesis,
diagenesis, speciation, cellulosic degradation
effects, interactions with clays and other host
materials, effects of corrosion products, effects
of cementitious materials, and leaching.
69, 70, 71, 72,
73, 74, 75, 76,
77, 78, 79, 80,
81, 82, 174, 264,
368, 440, 575,
576, 577, 698,
699, 700, 701,
862, 863, 864,
865, 866, 867,
868, 869, 870,
871, 872, 873,
874
Human Processes anthropogenic
climate change
This is addressed as part of climate change in
general.
85, 580, 706,
879
community
development
Development of communities and other human
habitation overlaps with land use and habitation,
and is considered in the human exposure
assessment, albeit unlikely. See inhabitation,
land use.
Regulations suggest consideration.
581
excavation Excavation includes construction of basements
and other construction, and is included as part of
the human intrusion scenarios.
330, 499, 582,
709, 710, 882,
883
explosions Human-caused explosions include bombs, plane
crashes, and conventional weapons training.
230, 500, 583,
804
human-induced
processes
Human-induced processes are limited to
repository design, inadvertent human intrusion,
or human-induced climate change. Engineered
features include repository design and new
technological developments. Intentional intrusion
is not considered. Anthropogenic climate change
is considered under climate change.
90, 91, 92, 177,
271, 272, 372,
443, 589, 590,
712, 713, 886
human-induced
transport
Human activities that could contribute to release
are considered. Humans can induce
contaminant transport through a variety of
activities. See inadvertent human intrusion.
273, 274, 591,
592, 795, 887
inadvertent
human
intrusion
Inadvertent human intrusion into the waste is
considered in the development of exposure
pathways.
Regulations suggest consideration.
178, 179, 231,
275, 276, 277,
373, 374, 375,
444, 445, 446,
714, 805, 806,
888
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 54
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
inhabitation Inhabitation on or near the site, including the
establishment of surface or underground
dwellings, communities, or cities, is extremely
unlikely. See community development, land use.
Regulations suggest consideration.
93, 94, 593, 594,
807
institutional
control
Institutional control affects human exposures,
and includes records of site knowledge,
markers, barriers, and security, and the loss
thereof.
Regulations suggest consideration.
95, 595, 596,
597, 716, 890
land use Land use in general could affect exposure
scenarios. Land use changes are related to
demographics, including development of
agricultural, industrial, urban, or wild land uses.
Regulations suggest consideration.
183, 450, 600,
601, 602, 719,
720, 893, 894,
895
post-closure
subsurface
activities
Subsurface human activities are covered to the
extent that they are inadvertent. This could
include intrusion, construction, investigation,
drilling, or mining.
Regulations suggest consideration.
727, 904, 905,
906
Hydrogeological denudation Denudation could expose wastes, and is
combined with erosion and inundation.
Regulations suggest consideration.
192, 388, 460,
502, 503, 739,
917
erosion Erosion of the repository resulting from pluvial,
fluvial, or aeolian processes can result from
extreme precipitation, changes in surface water
channels, and weathering.
Regulations suggest consideration.
110, 238, 284,
389, 504, 613,
740, 918, 919,
920, 921
erosional
transport
Erosional (sediment) transport could be a CT
mechanism. Sediments may move during
erosion; includes solifluction.
Regulations suggest consideration.
111, 239, 614,
615, 741, 742,
922, 923
hydrogeological
effects
Hydrogeological and groundwater hydraulics
changes may occur in response to geological
changes, including hydrothermal activity. This is
generally covered under groundwater transport.
Regulations suggest consideration.
112, 616, 617,
618, 619, 743,
744, 796, 924
sedimentation Sedimentation would occur on a lake bottom,
and could affect performance. This includes
sedimentation/aggradation onto the repository.
113, 193, 285,
335, 390, 461,
621, 797
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 55
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
Hydrology groundwater
transport
Groundwater transport includes waterborne
contaminant transport (CT) in the unsaturated
and saturated zones, and is a principal CT
mechanism. Groundwater flow and transport
mechanisms include advection-dispersion,
diffusion, fluid migration, waterborne
contaminant transport, changes in the flow
system, recharge and discharge, water table
movements, and brine interactions.
114, 115, 116,
117, 118, 286,
312, 313, 314,
315, 316, 336,
337, 338, 339,
392, 393, 622,
623, 747, 748,
749, 750, 751,
752, 929, 930,
931, 932, 933,
934, 935, 942
hydrological
effects
Hydrological processes are considered under
the topics of surface water and groundwater.
Regulations suggest consideration.
463, 505, 624,
753, 754, 936,
937
inundation Inundation by a large lake or reservoir is likely to
affect the site in the long term. (See also: wave
action, and lake effects).
Regulations suggest consideration.
755, 798, 938,
939
Meteorology frost
weathering
Weathering from frost cycles is included in cap
degradation modeling.
758, 943
meteorology Meteorology is considered indirectly;
meteorology as a science is not a FEP, per se,
but contributes to other processes, such as
precipitation and atmospheric dispersion, which
are covered elsewhere.
Regulations suggest consideration.
626, 627, 761,
946, 947
resuspension Resuspension will affect site performance,
allowing particulates from surface soils to be
redistributed by atmospheric dispersion.
981
atmospheric
dispersion
Atmospheric dispersion is a potential CT
pathway and is modeled. See also: dust devils.
Regulations suggest consideration.
256, 528
tornado Tornados are possible in the area. 289
Model Settings model
parameteri-
zation
Parameterization is a fundamental part of
modeling, though is not a FEP, per se.
628
period of
performance
Definition of a period of performance is a
fundamental part of PA modeling, though is not
a FEP, per se.
629
regulatory
requirements
Regulatory requirements drive much of the
modeling in PA, though is not a FEP, per se.
630
spatial domain Definition of a spatial domain is a fundamental
part of modeling, though is not a FEP, per se.
631
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 56
Table 2 (continued)
Neptune
Subgroup
Normalized
FEP
(accepted)
Discussion Representative
FEP IDs1
Other Natural
Processes
ecological
changes
Changes in the types and abundance of plants
and animals could affect performance. Changes
in the ecology can be associated with
catastrophic events (e.g. fire, inundation),
evolution, or climate change.
762, 948, 949
gas generation Uranium wastes are expected to produce radon
which will affect site performance in terms of
doses. See also gas transport.
122, 123, 340,
396, 464, 634,
766, 953, 954
pedogenesis Soils are likely to develop on the cap and may
affect performance.
765, 952
radioactive
decay and in-
growth
Radioactive decay and ingrowth processes are
essential to the model.
635, 767, 799,
955
radon
emanation
Radon emanation directly affects the mass of
radon released into the environment, and hence
site performance.
980
reconcentration Possible reconcentration of radiological
materials during transport is accounted for in the
CT modeling.
127
Tectonic/ Seismic/
Volcanic
geophysical
effects
Geophysical changes to the engineered features
of the site are accounted for in degradation.
Geophysical effects include pressure, stress,
density, viscosity, deformation, magnetics,
creep, and elasticity.

141, 142, 143,
509, 641, 781,
970, 971
1 The Representative FEP IDs correspond to the FEP IDs given in Table 1.

FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 57
Table 3. List of FEPs dismissed from further consideration.
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
Celestial meteorite
impact
The occurrence and consequences of a direct
hit by a meteorite are out of the scope of this
model.
1, 158, 219, 220,
251, 320, 348,
415, 491, 492,
493, 518, 650,
810
Climate change glacial effects Glacial effects include presence of continental
glaciers and resulting isostatic effects, glacial
erosion, and periglacial effects. Glaciers in the
basin are not modeled. Return of a large lake is
expected should a glacial epoch return and is
covered within climate change scenarios.
5, 160, 223, 255,
322, 351, 418,
494, 495, 525,
526, 654, 655,
815, 816
permafrost The effects of permafrost are bounded by those
of cap degradation, which considers more
damaging freeze/thaw cycles. See frost
weathering.
6, 300
Contaminant
Migration
gas intrusion No mechanism for intrusion of naturally-
produced gases into the repository has been
identified.
41, 677, 834
Engineered
Features
convergence of
openings
This FEP applies to mined repositories only. 837
design error Errors in design could compromise performance
but are not included in the modeling. Design
error is distinct from construction or operational
error.
47, 358, 424
material defects Material defects are covered by degradation,
and include material defects in source
containment, closure cap, and other engineered
materials.
691, 851
mechanical
effects
Mechanical effects are covered implicitly by
degradation, and include changes in mechanical
properties and conditions, including failure.
63, 64, 65, 172,
262, 365, 366,
434, 435, 556,
557, 693, 694,
855, 856
release of
stored energy
No significant energy is stored within the wastes. 66, 436, 857
repository seals Regulations suggest consideration, but, the
sealing of the repository shafts, boreholes, and
construction and failure of such is applicable
only to mined repositories.
67, 173, 229,
263, 328, 367,
437, 438, 558,
559, 697, 860
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 58
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
Exposure agriculture Agriculture includes establishment, evolution,
and abandonment of agriculture and aquaculture
at and near the site. Regulations suggest
consideration, however, none of these are
expected to occur because of the high salinity of
soils and groundwater at the site.
705, 803, 878
Geological diagenesis Diagenesis in local lacustrine sediments could
include the formation of interstitial evaporites,
but is not expected to change site performance.
83, 175, 265,
369, 441, 578,
702, 875
gas or brine
pockets
No gas or brine pockets have been identified
below the site.
176, 370, 442,
579
landslide Regulations suggest consideration, but
landslides are not expected to occur in the flat
lacustrine basin. Mass wasting of the site itself is
covered under erosion.
266, 703, 876
local
subsidence
Geological subsidence in the area is unlikely to
seriously affect performance, and is not
expected in the basin of lacustrine sediments.
267
Human
Processes
accidents
during
operations
Regulations suggest consideration, but
operational performance is not within the scope
of the PA model.
84, 704, 877
climate control No climate control at the facility is assumed.
Climate control is a feature of certain mined
repositories.
268, 371
closure failure Regulations suggest consideration; however,
poor closure includes abandonment or other
failure to close the facility as planned, and is not
modeled.
86, 87, 707, 708,
880, 881
fire The waste is not combustible or explosive. Fires
in the waste itself or following explosions are
distinct from wildfire.
269, 270
fisheries Regulations suggest consideration, but
development of fisheries is not credible at the
site.

884
geothermal
energy
production
No geothermal resources are identified at the
site.
89, 711, 885
injection wells Given the regional history, the construction of
injection wells nearby for disposal of liquid
wastes is possible. The effect of drilling such
wells in the vicinity would be negligible,
however.
232, 715, 889
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 59
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
intentional
intrusion
Intentional intruders are not protected and are
not modeled as receptors. Intentional intrusion
includes exhumation of waste, sabotage,
terrorism, or archeological research.
96, 180, 181,
278, 376, 377,
447, 448, 717,
808, 891
investigation Site investigation is considered intentional, and
receptors are not covered.
598, 599, 809
irrigation Regulations suggest consideration, and
irrigation could affect site performance, but will
not occur since there is no suitable water
source.
182, 233, 378,
449, 718, 892
monitoring Monitoring of the site is required, but persons
performing the activity are not protected since it
is intentional and informed. Monitoring activities
will not affect the performance of the site.
97, 603, 721,
896
nuclear testing Regulations suggest consideration; however,
testing of nuclear devices underground, at the
ground surface, or in the atmosphere is
considered intentional disruption of the site and
is not covered.
98, 722, 897
operational
effects
Operations could affect performance, and
include normal site operation, closure, and later
attempts at site improvement. Regulations
suggest consideration; however, operations are
not part of the PA.
99, 604, 605,
723, 724, 898,
899, 900
operational
error
Covered under operational effects. Operational
errors include poor quality site construction,
waste emplacement, and site closure.
Regulations suggest consideration, however,
operations are not part of the PA.
100, 184, 279,
379, 380, 451,
725, 726, 901,
902, 903
quality control Quality control is important to site operations,
but is not a FEP that lends itself to modeling.
606
resource
extraction
Regulations suggest consideration. Resource
extraction is a type of intentional intrusion,
including drilling, mining, or quarrying into the
repository, or in such a way as to affect
performance, in search of resources such as
petroleum, natural gas, salt, rock, or geothermal
resources. See intentional intrusion.
101, 102, 103,
185, 186, 234,
235, 280, 331,
332, 381, 382,
383, 452, 453,
501, 608, 609,
728, 729, 730,
731, 907, 908,
909, 910
sabotage Sabotage is by its nature intentional. See
intentional intrusion.
104, 187, 333,
384, 454, 732,
733, 911, 912
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 60
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
unplanned
events
This category is too vague to be considered
explicitly; unplanned events are generally
subsumed by other FEPs.
610
war Intrusion or disruption as part of an act of war
would be intentional. See intentional intrusion.
105, 188, 334,
385, 455
waste recovery Regulations suggest consideration, but waste
recovery, retrieval, or mining are considered
intentional acts. See intentional intrusion.
106, 189, 386,
456, 607.734,
735
water resource
management
Water resource activities include construction of
dams, reservoirs, and wells, and could affect the
site as water is extracted or retained.
Regulations suggest consideration; however,
this is not specifically modeled, as it is bounded
by the large lake scenario.
107, 108, 109,
190, 236, 237,
281, 282, 387,
457, 458, 611,
736, 737, 913,
914, 915
weapons
testing
Any nuclear and conventional weapons testing
would be done with cognizance of the site, and
is intentional. See also explosions and
intentional intrusion.
191, 283, 459
Hydrogeological subrosion No subsurface erosion has been reported in the
vicinity.
925
Hydrology flooding Regulations suggest consideration; however,
temporary flooding of the site due to extreme
precipitation is not plausible due to site
topography in the midst of the flats. This is
distinct from inundation by the return of a large
lake, which is included.
194, 240, 391,
462, 746, 926,
927, 928
surface water
transport
Surface water transport includes formation and
changes in rivers, lakes, and streams, and
transport of dissolved and suspended solids,
and sediments. Such effects are not anticipated
at the facility. This is distinct from inundation by
the return of a large lake, which is included.
119, 241, 287,
317, 318, 319,
394, 395, 625,
756, 757, 940,
941
Marine coastal
processes
Coastal processes will not apply at the site,
since no sea or ocean is expected in relevant
time frames. However, see wave action.
612, 738, 760,
916, 945
hurricanes No hurricanes occur in the area. 242, 288
insolation Insolation (the amount of sunshine on the site)
has no direct effect on site performance. See
ecological changes.
759, 944
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 61
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
marine effects Marine processes will not apply at the site, since
no sea or ocean is expected in relevant time
frames. Marine processes include sea-level
change. See also coastal processes and
tsunami.
620, 745
tsunami No tsunami will occur at the site. See coastal
processes and marine effects.
243
Natural
Processes
microbial
effects
Microbial action is not expected to affect
performance. Microbial processes include
corrosion, changes in chemistry, and dissolution
of glasses, but biotically-induced transport is
limited to macrobiological processes.
120, 632, 633,
763, 764, 950,
951
radiological
effects
Regulations suggest consideration. Radiological
processes such as radiolysis are a concern for
waste containment in some geological
repositories, but are not modeled here, since
waste containment is not given credit.
124, 125, 126,
195, 341, 397,
465, 466, 636,
768, 769, 956,
957
wildfire Occasional wildfire (brush fire, forest fire, either
local or widespread) is not likely to affect site
performance in the long run, since this is a
natural part of plant community dynamics.
290
Source Release electrochemical
effects
Electrochemical effects are not a relevant
process at the site. Electrochemical reactions
are a concern for the SKB repository.
121
explosions Explosive gases are not present in the
repository.
88
Tectonic/ Seismic/
Volcanic
breccia pipes Regulations suggest consideration, and the
formation of breccia pipes or mud volcanoes
could affect performance, but is considered
highly unlikely.
197, 343, 399,
469
diapirism Salt deposits in the strata below the site will not
result in the formation of diapirs.
198, 244, 292,
344, 400, 470,
638, 776, 965
discontinuities No major geological discontinuities are
envisioned at the site.
639
earthquake Earthquakes, either from natural or man-made
causes, would not change the performance of
this shallow unconsolidated site.
138, 293
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 62
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
faulting Faulting is unlikely to significantly affect
performance of this shallow unconsolidated site
and is not explicitly modeled. Geologic faulting
includes all type of faults, shear zones,
diastrophism, existing and future. See also see
fracturing.
139, 199, 200,
201, 245, 294,
345, 401, 402,
471, 472, 473,
506, 507, 508,
777, 778, 966,
967
fracturing Tectonic fracturing will not affect unconsolidated
site performance.
202, 203, 204,
205, 246, 403,
474, 475, 476,
477, 779, 968
geological
intrusion
Magmatic and intrusive igneous activity has not
been identified in the vicinity of the site.
Geological intrusion includes dikes, intrusive and
magmatic activity, and metamorphism due to
such activity. This is distinct from breccia pipes
(mud volcanoes) and human intrusion.
140, 206, 207,
295, 346, 404,
405, 478, 479,
640, 780, 969
hydraulic
fracturing
Hydraulic fracturing is performed in solid rock,
and has no applicaton at the site. Hydraulic
fracturing ("hydrofracking") is induced by
humans to enhance resource recovery or liquid
waste disposal by injection.
208, 480
intrusion into
accumulation
zone in the
biosphere
No accumulation zone in the biosphere has
been identified at the site.
144
isostatic effects Isostatic changes could influence lake levels,
which are accounted for elsewhere. Isostasy
includes that caused by tectonics, large bodies
of water, and by continental glaciers.
209, 406, 481,
510, 511
lava tubes No lava tubes exist at the site or are expected in
the future.
210, 407, 482
orogeny No significant orogeny is expected in relevant
time frames. Orogeny (mountain-building)
caused by tectonic movements or regional uplift.
211, 247, 296,
408, 483
regional
subsidence
Regional subsidence could influence lake levels,
which are accounted for elsewhere.
145, 409, 782,
972
seismic effects Regulations suggest consideration, but effects of
seismic activity (see also earthquakes) would be
insignificant for shallow land burial.
248, 512, 513,
642, 783, 973
FEP Analysis for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 63
Table 3 (continued)
Neptune
Subgroup
Normalized
FEP
(dismissed)
Discussion Representative
FEP IDs1
tectonic effects Tectonic effects could influence lake levels,
which are accounted for elsewhere.
146, 147, 148,
149, 212, 213,
410, 484, 643,
644, 784, 785,
974, 975, 976
volcanism No significant volcanism is expected in relevant
time frames.
150, 214, 249,
250, 411, 412,
485, 486, 514,
515, 516, 645,
786, 800, 977
Waste nuclear
criticality
Nuclear criticality, while a concern for
repositories of used nuclear fuel, is not a
concern at this LLW site.
151, 152, 215,
297, 347, 413,
487, 646, 787,
978
other waste The current analysis is constrained to examine
depleted uranium wastes only, including
associated "contaminant" waste. This rather
vague reference to "other waste" will be
addressed as the scope of wastes under
consideration expands.
153, 154, 155,
156, 157, 216,
217, 218, 298,
299, 414, 488,
489, 490, 788,
979
1 The Representative FEP IDs correspond to the FEP IDs given in Table 1.

 NAC-0018_R4

Conceptual Site Model for
Disposal of Depleted Uranium
at the Clive Facility

Clive DU PA Model v1.4
5 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 ii
1. Title: Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
2. Filename: Clive DU PA CSM v1.4.docx
3. Description: This document describes the site conditions, chemical and radiological
characteristics of the wastes, contaminant transport pathways, and potential exposure routes at
the Clive facility that are used to structure the quantitative Clive DU PA Model.
Name Date
4. Originator John Tauxe 21 May, 2014
5. Reviewer Dan Levitt, Mike Sully and Bruce Crowe 22 May, 2014
6. Remarks
10/20/2015 MS: Saved as v1.4

21 Oct 2015: Modified figures to be consistent with Clive DU PA Model v1.4, and to clarify use
of the “Federal Cell.” Corrected internal figure references that were out of sequence. – J Tauxe

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 iii

This page is intentionally blank, aside from this statement.

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 iv
CONTENTS
1.0
 Introduction ............................................................................................................................ 1

2.0
 Scope of the Conceptual Site Model ...................................................................................... 1

3.0
 Site Description ...................................................................................................................... 5

3.1
 Land Management ............................................................................................................ 6

3.2
 Climate .............................................................................................................................. 8

3.2.1
 Temperature ................................................................................................................ 8

3.2.2
 Clive facility Precipitation .......................................................................................... 8

3.2.3
 Evaporation ................................................................................................................. 8

3.3
 Geology ............................................................................................................................. 8

3.3.1
 Site Geology ................................................................................................................ 8

3.3.2
 Site Seismotectonics ................................................................................................. 10

3.3.3
 Eolian Deposition ...................................................................................................... 13

3.4
 Hydrology ....................................................................................................................... 14

3.4.1
 Surface Water ............................................................................................................ 14

3.4.2
 Groundwater .............................................................................................................. 15

3.5
 Ecology ........................................................................................................................... 16

3.5.1
 Local Vegetation ....................................................................................................... 16

3.5.2
 Local Wildlife ........................................................................................................... 17

3.6
 Engineered Features ........................................................................................................ 18

3.6.1
 Federal Cell Disposal Cell Design ............................................................................ 18

3.6.2
 Degradation of Engineered Features ......................................................................... 18

4.0
 Regulatory Context ............................................................................................................... 18

4.1
 Nuclear Regulatory Commission Regulations ................................................................ 19

4.1.1
 Section 61.55: Waste Classification .......................................................................... 19

4.1.2
 Section 61.41: Protection of the Public ..................................................................... 20

4.1.3
 Section 61.42: ALARA and Collective Dose ........................................................... 21

4.1.4
 Section 61.42: Protection of the Inadvertent Intruder ............................................... 22

4.1.5
 Proposed Rule-Making Regarding 10 CFR 61 ......................................................... 22

4.2
 State of Utah Regulations ............................................................................................... 22

4.2.1
 Section R313-25: Licensing Requirements ............................................................... 23

4.2.2
 Section R313-15-1009: Waste Classification ........................................................... 23

4.2.3
 Groundwater Protection Limits ................................................................................. 24

5.0
 Summary of Features, Events, and Processes ...................................................................... 25

6.0
 Waste Forms ......................................................................................................................... 28

6.1
 Savannah River Site Uranium Trioxide .......................................................................... 29

6.2
 Depleted Uranium Oxide from the Gaseous Diffusion Plants ........................................ 30

6.3
 Depleted Uranium Already Disposed at the Clive Facility ............................................ 31

6.4
 Modeled Radionuclides .................................................................................................. 31

6.5
 Chemical Characteristics of DU Wastes ......................................................................... 31

7.0
 Modeling of the Natural Environment ................................................................................. 32

7.1
 Current Conditions .......................................................................................................... 32

7.1.1
 Groundwater Flow and Transport ............................................................................. 32

7.1.2
 Surface Water ............................................................................................................ 36

7.1.3
 Air and Atmosphere .................................................................................................. 37

7.1.4
 Biota 39

7.1.5
 Native Animals ......................................................................................................... 41

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 v
7.2
 Deep Time Conditions .................................................................................................... 43

7.2.1
 Background on Long-term Controls on Site Conditions .......................................... 44

7.2.2
 Long-Term Scenarios ................................................................................................ 49

8.0
 Modeling of Engineered Features ........................................................................................ 51

8.1
 Waste Form and Containment ........................................................................................ 51

8.2
 Liners .............................................................................................................................. 51

8.3
 Cover ............................................................................................................................... 52

9.0
 Radionuclide Transport ........................................................................................................ 53

9.1
 Modeled Radionuclides .................................................................................................. 54

9.1.1
 Reported Inventory ................................................................................................... 54

9.1.2
 Radioactive Decay and In-growth ............................................................................. 54

9.1.3
 Short-lived Radionuclides ......................................................................................... 54

9.1.4
 Radionuclides with Small Branching Fractions ........................................................ 56

9.2
 Source Release ................................................................................................................ 57

9.2.1
 Containment Degradation ......................................................................................... 57

9.2.2
 Matrix Release .......................................................................................................... 57

9.2.3
 Radon Emanation ...................................................................................................... 57

9.3
 Waterborne Radionuclide Transport ............................................................................... 58

9.4
 Airborne transport ........................................................................................................... 59

9.4.1
 Diffusion Through Porous Media ............................................................................. 59

9.4.2
 Atmospheric Dispersion ............................................................................................ 60

9.5
 Biotically Induced Transport .......................................................................................... 60

9.5.1
 Transport via Plants .................................................................................................. 60

9.5.2
 Burrowing Animals ................................................................................................... 61

10.0
 Modeling Dose and Risk to Humans .................................................................................... 61

10.1
 Period of Performance .................................................................................................... 62

10.2
 Site Characteristics and Assumptions ............................................................................. 63

10.3
 Receptor Scenarios ......................................................................................................... 63

10.3.1
 Ranching Scenario .................................................................................................... 63

10.3.2
 Recreational Scenario ............................................................................................... 64

10.3.3
 Remote Off-Site Receptors ....................................................................................... 65

10.4
 Transport Pathways ......................................................................................................... 65

10.5
 Exposure Pathways ......................................................................................................... 66

10.6
 Risk Assessment Endpoints ............................................................................................ 66

11.0
 Summary ............................................................................................................................... 68

12.0
 References ............................................................................................................................ 70

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 vi
FIGURES
Figure 1. Conceptual diagram of the performance assessment process. ......................................... 2

Figure 2. Location of the Clive site operated by EnergySolutions. ................................................. 6

Figure 3. Disposal and treatment facilities operated by EnergySolutions, with Federal Cell
identified. ....................................................................................................................... 7

Figure 4. Eolian silt in trench located at Clive Pit 29 overlying Lake Bonneville sedimentary
deposits (Neptune 2015). ............................................................................................. 13

Figure 5. An example of upper soil-modified eolian silt in Pit 29. Basal contact of the silt is
approximately located at the middle of the pick handle. Lake Bonneville marl is at
the bottom of the pick handle. ..................................................................................... 14

Figure 6. Waste classification Tables 1 and 2 from 10 CFR 61.55. .............................................. 20

Figure 7. Waste classification Table I from R313-15-1009. ......................................................... 24

Figure 8. Section and Plan views of the Federal Cell, with top slope shown in blue and side
slope in green. The brown dotted line in the West-East Cross section represents
below-grade (below the line) and above-grade (above the line) regions of the
embankment. ............................................................................................................... 33

Figure 9. Evapotranspiration (ET) cover system. .......................................................................... 35

Figure 10. Hydrostratigraphic profile showing ET cover, waste zone, and hydrostratigraphy
below the Federal Cell. ................................................................................................ 36

Figure 11. Conceptual model for plant induced contaminant transport ........................................ 40

Figure 12. Whittaker Biome Diagram ........................................................................................... 48

Figure 13. Scenarios for the long-term fate of the Clive facility ................................................... 50

Figure 14. Principal decay chains for the four actinide series. Radionuclides in black are
included in the fate and transport model, and those in green are considered only in
the dose model. ............................................................................................................ 55

Figure 15. Detailed decay chains for actinides. Radionuclides in black are included in the fate
and transport model, those in green are considered only in the dose model, and
those in gray are not modeled. ..................................................................................... 56

Figure 16. Conceptual model for transport and exposure pathways at the Clive facility .............. 67

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 vii
TABLES
Table 1. Known lake cycles in the Bonneville Basin .................................................................... 46

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 viii
Acronyms and Abbrev.
Ac actinium
Am americium
amsl above mean sea level
bgs below ground surface
BLM Bureau of Land Management
Bq becquerel (1 disintegration per second)
CAW Class A West (embankment)
CEDE committed effective dose equivalent
CFR U.S. Code of Federal Regulations
Ci curie (37 GBq)
CSF cancer slope factor
CSM conceptual site model
CWF Containerized Waste Facility
DCF dose conversion factor
DOE U.S. Department of Energy
DU depleted uranium
DUF6 depleted uranium hexafluoride
EIS Environmental Impact Statement
EPA U.S. Environmental Protection Agency
ETTP East Tennessee Technology Park
FEIS Final Environmental Impact Statement
FEP features, events, and processes
FR Federal Register
ft foot/feet
g gram
GDP gaseous diffusion plant
GWPL groundwater protection limit(s)
GTCC greater than Class C waste
ha hectare
IAEA International Atomic Energy Agency
ICRP International Commission on Radiation Protection
IHI inadvertent human intruder
ka thousand years ago
Kd soil/water partition coefficient
kg kilogram
KH Henry’s Law constant (air/water partition coefficient)
km kilometer
ky thousand years
L liter
LARW low-activity radioactive waste
LLW low-level radioactive waste
MCL maximum contaminant level(s)
m meter
Ma million years ago
mg milligram
Mg megagram (one metric ton)
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 ix
MLLW mixed [hazardous and] low-level radioactive waste
MOP member of the public
MPa megapascal
mrem millirem
mSv millisievert
My million years
NRC U.S. Nuclear Regulatory Commission
NNSS Nevada National Security Site
NUREG an NRC publication
OHV off-highway vehicle
Pa protactinium
PA performance assessment
PAWG Performance Assessment Working Group (DOE)
pCi picocurie
Po polonium
ppm part per million
Pu plutonium
QA quality assurance
Ra radium
RfD reference dose
Rn radon
SRS Savannah River Site
Sv Sievert
Tc technetium
TDS total dissolved solids
TEDE total effective dose equivalent
TF Treatment Facility
Th thorium
U uranium
UAC Utah Administrative Code
UNF used nuclear fuel
UWQB Utah Water Quality Board
yr year

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 1
1.0 Introduction
The safe storage and disposal of depleted uranium (DU) waste is essential for mitigating releases
of radioactive materials and reducing exposures to humans and the environment. Currently, a
radioactive waste facility located in Clive, Utah (the Clive facility) operated by EnergySolutions
is proposed to receive and store DU waste that has been declared surplus by the U.S. Department
of Energy (DOE). The Clive facility has been tasked with disposing of the DU waste in an
economically feasible manner that protects humans from future radiological releases.
To assess whether the proposed Clive facility location and containment technologies are suitable
for protection of human health, specific performance objectives for land disposal of radioactive
waste set forth in Title 10 Code of Federal Regulations Part 61 (10 CFR 61) Subpart C, and
promulgated by the Nuclear Regulatory Commission (NRC), must be met. In order to support the
required radiological performance assessment (PA), a detailed computer model is developed in
order to evaluate the doses to human receptors that would result from the disposal of DU and its
associated radioactive contaminants (collectively termed “DU waste”), and conversely to
determine how much DU waste can be safely disposed at the Clive facility.
This conceptual site model (CSM) document describes the site conditions, chemical and
radiological characteristics of the wastes, contaminant transport pathways, and potential
exposure routes at the Clive facility that are used to structure the quantitative Clive DU PA
Model. The Model is probabilistic, taking into account uncertainties inherent to model variables
and site-specific conditions. The GoldSim systems analysis software (GTG, 2010) is used to
construct the probabilistic PA model. This PA model is intended to reflect the current state of
knowledge with respect to the proposed DU disposal, and to support environmental decision
making in light of inherent uncertainties.
This CSM report, and the associated features, events and processes (FEPs) report, are regarded
as “living documents.” That is, as further information is gathered during the course of model
development, the CSM might evolve and, consequently, be updated. Changes to the CSM will be
tracked so that the evolution is well documented.
2.0 Scope of the Conceptual Site Model
The overall scope of this analysis is to evaluate the long-term siting and performance integrity of
the Federal Cell (a discrete section of what was formerly known as the Class A South
Embankment, which included other wastes as well; and interchangeably termed the Federal DU
Cell in other documents because of the focus of this model on disposal of DU) at the Clive
facility for the proposed disposal of DU waste. The need for a PA is driven by Federal and State
of Utah regulations, which require an evaluation of the potential human radiation doses and
consequences of disposal of radioactive waste. The regulations contain procedural requirements,
performance objectives, and technical requirements for near-surface disposal, including disposal
in engineered facilities with protective earthen covers, which may be built fully or partially
above-grade, such as the radioactive waste disposal cells at the Clive facility. The overall PA
process is illustrated in Figure 1.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 2
This CSM describes the physical, chemical, and biological characteristics of the Clive facility.
The CSM, therefore, encompasses everything from the inventory of disposed wastes, the
migration of radionuclides contained in the waste through the engineered and natural systems,
and the exposure and radiation doses to hypothetical future humans. These site characteristics are
used to define variables for the quantitative Clive DU PA Model that are used to provide insights
and understanding of the future potential human radiation doses from the disposal of DU waste.
The content of the CSM informs the Model with respect to regional and site-specific FEPs, such
as climate, groundwater, and human receptor scenarios. The CSM accounts for and defines
relevant FEPs at the site, materials and their properties, interrelationships, and boundaries. These
constitute the basis of the Clive DU PA Model, on which, or through which, radionuclides are
transported to locations where receptors might be exposed.
The quantitative probabilistic Clive DU PA Model will be used to evaluate the migration of
radionuclides contained in the DU wastes, and the subsequent human doses resulting from
potential exposure to radionuclides, based on projecting current societal conditions up to 10,000
years into the future. However, because the radioactivity from the DU wastes (including
progeny) will increase for more than 2 million years, and will persist for at least a billion years,
further modeling of potential long-term future scenarios will be performed beyond the
10,000-year compliance period. The longer term model will address mechanisms by which
radionuclides might be dispersed in the environment, suggesting concentrations of radionuclides
in various media.
However, the long term future model will not directly address human doses, because it is not
clear what human exposure scenarios might be reasonable given events in the long term future
that might dramatically alter human society and civilization. Therefore, the focus of the longer-
Figure 1. Conceptual diagram of the performance assessment process.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 3
term modeling will be scenarios developed to represent potential features, events and processes
that affect contaminant fate and transport over these much longer periods.
The quantitative model is used to evaluate potential human radiation doses from exposure to
radionuclides contained in the DU wastes that may result from migration through the engineered
and natural systems to the potentially exposed population. Note that regulations specify
estimation of dose, rather than risk, though there are risks implied in the regulatory dose limits
(see Section 4). Risk-based decision-making is best supported with probabilistic modeling, and
has been used to assess compliance and inform decision making at many challenging radioactive
waste sites under various regulatory requirements. The U.S. Environmental Protection Agency
(EPA) has published probabilistic risk assessment guidance for human exposure to chemicals
(EPA, 2001) and promotes the use of probabilistic methods for performance assessments of
radioactive disposal facilities in its Environmental Radiation Protection Standards (40 CFR
191). The DOE has implemented probabilistic PAs at the Waste Isolation Pilot Plant, at the
Yucca Mountain Project, and for low-level radioactive waste (LLW) disposal facilities at the
Nevada National Security Site (NNSS, formerly the Nevada Test Site), Los Alamos National
Laboratory, and the Savannah River Site. The NRC has adopted this approach as well, as
documented in its Performance Assessment Methodology for LLW Disposal Facilities (NRC,
2000). Further, the National Research Council has argued in favor of the risk-based approach in
its recent book, Risk and Decisions (National Research Council, 2005). More generally, various
agencies and professional organizations (e.g., EPA’s Council for Regulatory Environmental
Modeling, Society for Risk Analysis) have consistently moved in the direction of supporting
risk-based decisions with probabilistic analysis so that the potential risks are modeled more
realistically (as opposed to conservatively) and uncertainty is numerically characterized.
Thus, the quantitative PA model is probabilistic, with uncertainties associated with the complex
evolution from waste disposal to human exposure and dose captured through input parameter
probability distributions. Attention is paid to developing model input parameter distributions that
reflect both the uncertain state of knowledge and the appropriate spatio-temporal scaling. The
focus of the uncertainty analysis in the Clive DU PA Model will be parameter uncertainty. The
Model is also developed with the capability of running the model under various FEP scenarios to
allow for an assessment of scenario uncertainty. This is important for the longer-term scenarios
in particular.
As noted above, the probabilistic approach models future conditions by projecting current
conditions as reasonably as possible while including uncertainty in the parameters or
assumptions of the model. This is differentiated from “conservative” (i.e., biased toward safety)
modeling that is sometimes performed, typically using point values for parameters (implying a
great deal of confidence; i.e., no uncertainty). This type of conservative modeling is often termed
“deterministic” modeling, and has often been used to support compliance decisions. However,
supposed conservatism in parameter estimates (or distributions) is often difficult to judge in fully
coupled models in which all transport processes are contained in the same overall PA model.
More importantly perhaps, conservative dose results from PA models do not support the full
capability of a disposal facility. Conservative, deterministic models may have utility at a
“screening” level, but, they do not provide the full range of information that is necessary for
important decisions such as compliance or rule-making (Bogen 1994, Cullen 1994).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 4
Of further concern is the type of modeling environment that is needed to support the types of
decisions that are made on the basis of PA models. The GoldSim modeling environment is
focused on development of “systems-level” models. These models are intended to characterize
the effects and consequences of system level dynamics. In this case, the system consists of the
waste disposal facility and the interaction of the facility with the environment (e.g., weather,
water, biota, etc.) in the 10,000-yr duration for which quantitative modeling will be performed
with human dose as the endpoint of interest, as well as the longer duration for which media
concentrations resulting from potential future scenarios involving, for example, climate change,
re-occurrence of large lakes, will be evaluated. That is, the domain of the model is large both
spatially and temporally. However, decisions need to be made in the face of uncertainty
regarding the applicability of the Clive facility for disposal of DU, and more generally, for the
design of the disposal facility.
Systems-level models are aimed precisely at supporting decision making in this type of context.
More detailed “process-level” models, which might model at a much more refined spatial scale
(and perhaps temporal scale), can provide useful input to the systems-level model, but they do
not as readily support decision-making at the more holistic scale of the systems-level response.
For example, a systems-level model will evaluate the movement of radionuclides from the waste
zone, through the unsaturated zone, to the saturated zone, by considering the average effects
across those system components, as opposed to the effects at a more refined scale such as every
cubic meter, which is more common for process-level modeling. Process-level models are often
geared towards capturing variability at small spatial scales, whereas systems-level models are
aimed at capturing uncertainty in the system as a whole. PA modeling is concerned with the
latter, including demonstration of compliance followed by a decision analysis in the spirit of
achieving ALARA (as low as reasonably achievable; see Section 4.0) releases and doses to
optimize disposal and closure (e.g., engineered barriers, institutional controls).
To capture the temporal domain of the model, time steps in this type of systems-level dynamic
probabilistic model are usually on the order of several to many years. Consequently, the average
effects over long time frames, assuming no catastrophic changes in the system, are far more
important than the effects on the scale of days or seconds. Spatial and temporal scaling of
available data, which are usually collected at points in time and space, is critical for the success
of systems-level models. Scaling in this context is essentially an averaging process both spatially
and temporally. Simple averaging works well if the effect on the response of a variable or
parameter is linear. Otherwise, some care needs to be taken in the spatio-temporal averaging
process. In addition, these types of models are characterized by differential equations and
multiplicative terms. Averaging is a linear construct that does not translate directly in non-linear
systems. Again, care needs to be taken to capture the appropriate systems-level effect when
dealing with differential equations and multiplicative terms.
A further statistical issue of concern is the challenge of capturing dependencies or correlation
structures with this type of dynamic probabilistic system. Inputs for parameters (variables) are
usually provided independently of each other. However, it is very important to capture
correlations between variables in a multiplicative model. Otherwise, system uncertainty is not
adequately constrained. GoldSim provides some limited capability to introduce correlation into a
PA model, but steps will be taken to evaluate the correlation effects of some variables.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 5
Processes that contribute to the fate and transport of these contaminants are also abstracted into
mathematical models. That is, process-level models are sometimes important for providing input
to PA models. Model abstraction is best performed by running process-level models for some
cases or scenarios that correspond to a design over the inputs. The response can be modeled
using a statistical response surface, which can then be carried or abstracted into the PA model.
The systems-level PA model is then fully coupled across processes, meaning that inputs and
outputs from each process affect the prior and posterior processes.
With a probabilistic dynamic PA model, a global sensitivity analysis can be performed to
identify those parameters that are most important for predicting the model results. This type of
sensitivity analysis is performed using statistical methods from data mining allowing all input
parameters to be varied simultaneously. This allows the combined effect of changes in
parameters to be evaluated. The sensitivity analysis tools can then be used to determine whether
more information should be collected to reduce uncertainty.
3.0 Site Description
EnergySolutions operates a low-level radioactive waste disposal facility west of the Cedar
Mountains in Clive, Utah, as shown in Figure 2. Clive is located along Interstate-80,
approximately 5 km (3 mi) south of the highway, in Tooele County. The facility is
approximately 80 km (50 mi) east of Wendover, Utah and approximately 100 km (60 mi) west of
Salt Lake City, Utah. The facility sits at an elevation of approximately 1302 m (4275 ft) above
mean sea level (amsl) and is accessed by both highway and rail transportation. The Clive facility
is adjacent to the above-ground disposal cell used for uranium mill tailings that were removed
from the former Vitro Chemical company site in South Salt Lake City between 1984 and 1988
(Baird et al., 1990).
Currently, the Clive facility receives waste shipped via truck and rail. Pending the findings of the
PA, DU waste will be stored in a permanent above-ground engineered disposal embankment that
is clay-lined with a composite clay and soil cover. The disposal embankment is designed to
perform for a minimum of 500 years based on requirements of 10 CFR 61.7, which provides a
long-term disposal solution with minimal need for active maintenance after site closure. More
detail relating to the properties of the disposal embankment is provided in Section 3.6.1.
The EnergySolutions Clive facility is divided into three main areas (Figure 3 in EnergySolutions,
2008):
• the Bulk Waste Facility, including the Mixed Waste, Low Activity Radioactive Waste
(LARW), 11e.(2), and Class A LLW areas,
• the Containerized Waste Facility (CWF), located within the Class A LLW area, and
• the Treatment Facility (TF), located in the southeast corner of the Mixed Waste area.
The subject of this CSM and associated modeling is DU waste disposed or to be disposed in the
Federal Cell. The terms “cell” and “embankment” are here used interchangeably.

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 6

3.1 Land Management
The Bureau of Land Management (BLM) administers much of the land around the Clive facility.
BLM land is public domain (NRC, 1993). The disposal site is located within a 260-ha (640-acre)
section of land that was originally selected for the disposal of the Vitro Chemical Company
uranium tailings (see “Vitro” in Figure 3). This section of land occupies approximately 40 ha
(100 acres), while the remaining 220 ha (540 acres) is owned and operated by EnergySolutions.
The Tooele County Commission zoned the Clive site as a “Hazardous Industrial District,” which
falls within the West Desert Hazardous Industry Area, an area that prohibits future residential
housing in the near vicinity of the Clive site (NRC, 1993).
NRC (1993) and the BLM (BLM staff, personal communication, 2010) indicates that the area
surrounding the Clive facility is used for cattle and sheep grazing purposes and recreation. While
the site is zoned for hazardous waste disposal by Tooele County, the lack of potable water at this
site makes the surrounding area an unlikely location for any residential, commercial, or industrial
developments (Baird et al., 1990).

Figure 2. Location of the Clive site operated by EnergySolutions.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 7

Figure 3. Disposal and treatment facilities operated by EnergySolutions, with Federal Cell
identified.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 8
3.2 Climate
3.2.1 Temperature
Regional climate is regulated by the surrounding mountain ranges, which restrict movement of
weather systems in the vicinity of the Clive facility. The most influential feature affecting
regional climate is the presence of the Great Salt Lake, which can moderate downwind
temperatures since it never freezes (NRC, 1993). The climatic conditions at the Clive facility are
characterized by hot and dry summers, cool springs and falls, and moderately cold winters
(NRC, 1993). Frequent invasions of cold air are restricted by the mountain ranges in the area.
Data from the Clive facility from 1992 to 2009 indicate that monthly temperatures range from
about -2.4°C (27.7°F) in December to 26.4°C (79.5°F) in July (MSI, 2010) where monthly
average temperatures are assumed to be calculated as the monthly average of hourly air
temperatures for that month based on comparison with hourly data collected for 2009 and
reported in MSI (2010).
3.2.2 Clive facility Precipitation
Clive facility Data collected at the Clive facility from 1992 through 2004 indicate that average
annual rainfall is on the order of 22 cm (8.6 in) per year (Whetstone, 2006). Precipitation
generally reaches a maximum in the spring (1992-2004 monthly average of 3.2 cm [1.25 in] in
April), when storms from the Pacific Ocean are strong enough to move over the mountains
(NRC, 1993; Whetstone, 2006). Precipitation is generally lighter during the summer and fall
months (1992-2004 monthly average of 0.8 cm [0.32 in] in August) with snowfall occurring
during the winter months (Whetstone, 2006; NRC, 1993; Baird et al., 1990).
3.2.3 Evaporation
Because of warm temperatures and low relative humidity, the Clive facility is located in an area
of high evaporation rates. NRC (1993) indicates that average annual pond evaporation rate at the
Clive facility is 150 cm/yr (59 in/yr), with the highest evaporation rates between the months of
May and October. Previous modeling studies indicate that the Dugway climatological station
nearby is comparable to the Clive site with respect to evaporation and have reported pan-
evaporation estimates of 183 cm/yr (72 in/yr), which is considerably greater than average annual
rainfall (Adrian Brown, 1997a). While the data range for the site is more limited, annual pan
evaporation measured at the site greatly exceeds annual precipitation (MSI 2010). Average
annual pan evaporation is 132 cm (52 in) (MSI 2010, p. 4-7) while average annual precipitation
is 22 cm (8.5 in) (MSI 2010, p. 4-8).
3.3 Geology
3.3.1 Site Geology
The Clive facility rests on lacustrine deposits from the ancestral Lake Bonneville, which was a
pluvial lake that existed during the late Pleistocene. The geology is characterized by north-south
trending mountain ranges surrounded by sediment filled basins. The site is bounded by the Cedar
Mountains to the east and the Great Salt Lake Desert to the west. Surficial drainage is generally
in a westward direction away from the nearest mountain range.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 9
NRC (1993) indicates that based on subsurface borehole logs, lacustrine deposits extend to at
least 75 m (250 ft) underneath the site, however these estimates are limited to the depths of
boreholes drilled from previous hydrogeologic investigations (e.g., Envirocare [2004]). Oviatt et
al. (1999) examined the upper 110 m (361 ft) of the Burmester core, a sediment core that was
collected to a depth of 307 m (1007 ft) in the 1970s to characterize major pluvial lake cycles in
the Bonneville Basin. Brodeur (2006) also indicates that sediments can be up to a thousand
meters thick in some regions of the basin and greater than 200 m (700 ft) thick in the basin at the
Clive site.
The sediments underlying the Clive site are described as four separate hydrostratigraphic units
based on grain size and sediment characteristics. These units are described in NRC (1993),
Adrian Brown (1997a), and Envirocare (2004) and are introduced from the ground surface down:
• Unit 4 (surface) is composed primarily of silt and clay between 1.8 and 5 m (6 and 16.5
ft) thick, with an average thickness of 3 m (10 ft). Minor amounts of sand within the silt
and clay can be found along with some evaporite mineral content. This layer has a low
permeability and a high capacity to store moisture.
• Unit 3 lies beneath Unit 4 and is composed of a silty sand between 2.1 and 7.6 m (7 and
25 ft thick, with an average thickness of 3 m (10 ft). The water table of a shallow,
unconfined aquifer occurs near the bottom of this Unit on the western side of the site.
This shallow aquifer is saline.
• Unit 2 lies beneath Unit 3 and is composed of clay with occasional lenses or interbeds of
silty sand. This unit is between 0.76 and 7.6 m (2.5 and 25 ft) thick and is saturated with
saline groundwater.
• Unit 1 underlies Unit 2 and is saturated beneath the facility, containing a locally confined
aquifer. Unit 1 extends from approximately 14 m (45 ft) bgs and contains the deep
aquifer. The deeper aquifer is reported to be made up of lacustrine deposits consisting of
deposits of silty sand with some silty clay layers. One or possibly more silty clay layers
overlie the aquifer (Bingham Environmental 1994).
The aquifer system in the vicinity of the Clive facility is described by Bingham Environmental
(1991, 1994) and Envirocare (2000, 2004) as consisting of unconsolidated basin-fill and alluvial
fan aquifers. Characterization of the aquifer system is based on subsurface stratigraphy
observations from borehole logs and from potentiometric measurements. The aquifer system is
described as being composed of two aquifers; a shallow, unconfined aquifer and a deep confined
aquifer. The shallow unconfined aquifer extends from the water table to a depth of
approximately 13 to 14 m (40 to 45 ft) bgs. The water table in the shallow aquifer is reported to
be located in Unit 3 on the west side of the site and in Unit 2 on the east side.
The deep confined aquifer is encountered at approximately 14 m (45 ft) bgs and extends through
the valley fill (Bingham 1994). The boring log from a water supply well drilled in adjoining
Section 29 indicated continuous sediments to a depth of 190 m (620 ft) bgs (DWR 2014, water
right number 16-816 and associated well log 11293). The deepest portion of the basin in the
Clive area is believed to be north of Clive in Ripple Valley where the basin fill was estimated to
be 900 m (3,000 ft) thick (Baer and Benson, as cited in Black et al., 1999).
Deeper saturated zones in Unit 1 below approximately 14 m (45 ft) bgs are reported to show
higher potentiometric levels than the shallow unconfined aquifer. Differences in potentiometric
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 10
levels are attributed to the presence of the Unit 2 clays. These observations are interpreted as
indicating that the shallow unconfined aquifer below the site does not extend into Unit 1 but is
contained within Units 2 and 3 (Bingham Environmental, 1994).
Vertical gradients between shallow and deeper screened intervals in the monitor well clusters
were calculated by Bingham Environmental (1994). An upward vertical gradient was observed
ranging in magnitude from 0.02 to 0.04 based on the distance between the screen centers.
Hydraulic conductivities measured from bailing tests are reported to average 2.6 × 10-3 cm/s
(7.45 ft/day) by Envirocare (2004). Bailing tests in boreholes provide a saturated hydraulic
conductivity more representative of the horizontal hydraulic conductivity than the vertical. Based
on 3 measurements of vertical hydraulic conductivity on silty clay cores made by Bingham
Environmental (1991), Envirocare (2004) and Bingham Environmental (1994), Envirocare
(2004) use a value of 1 × 10-6 cm/s for the vertical hydraulic conductivity. This corresponds to an
anisotropy ratio Kv/Kh of 1:2600. Average linear vertical groundwater velocity ranged from 1.5
to 3.0 cm/yr (0.05 to 0.10 ft/yr) based on these vertical gradients, a porosity of 0.4 and a vertical
hydraulic conductivity of × 10-6 cm/s (Bingham, 1994).
Horizontal groundwater velocities were calculated by Bingham Environmental (1994) for 17
monitoring wells having measurements of hydraulic conductivity and estimated gradients.
Hydraulic conductivities ranged from 2.9 × 10-5 cm/s to 9.5 × 10-4 cm/s and horizontal hydraulic
gradients ranged from 2 × 10-4 to 1 × 10-3. Average linear horizontal groundwater velocity
ranged from less than 0.6 to 64 cm/yr (0.02 to 2.1 ft/yr) based on a porosity of 0.3. The ratio of
linear horizontal velocities to linear vertical velocities ranged from 0.4 to 21.
The influence of downward hydraulic gradients on shallow groundwater flow is discussed in
Envirocare (2004) for two cases. In the first, flow was affected by localized recharge from a
surface water retention pond in the southwest corner of the facility in the spring of 1999 and in
the second, a ground water mound formed between March 1993 and spring 1997 below a borrow
pit excavated near the 11e.(2) cells that occasionally filled with rain water. The mound decreased
and was negligible by the time of the report in 2004.
3.3.2 Site Seismotectonics
The Clive site does not have any known active faults in its vicinity. NRC (1993) indicates that
the nearest faulting is located 29 km (18 miles) to the north, having occurred between 1 million
to 25 million years ago (1 to 25 Ma). Although the site is not located near any active faults,
isostatic rebound is suspected to be the cause of any recent seismic activity in the Lake
Bonneville area.
NRC (1993) cites two seismic investigations that were conducted for the Vitro tailings disposal
facility and a proposed site for a supercollider that was to encompass a 24-km (15-mile) elliptical
ring around the Clive site. Based on these studies, NRC (1993) indicated that nearby structures
and seismogenic areas that could pose a hazard include the fault zones within a 72-km (45-mile)
radius of the site. These include the eastern flank of the Cedar Mountains, western flank of the
Lakeside Mountains, Northwest Puddle Valley, eastern flank of the Newfoundland mountains,
and the western flank of the Stansbury Mountains. However, NRC (1993) concluded that no
active fault zones lie beneath the Clive site, and there is no macroseismic evidence of a capable
fault in the vicinity of the site.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 11
The lack of Quaternary and/or capable faults in the vicinity of the Clive site is not sufficient
evidence to dismiss seismic activity as a potential issue of concern. While the absence of surface
faults in the site is consistent with a low probability of surface-fault rupture, ground shaking
associated with background earthquakes require assessments (i.e. moderate-size earthquakes (M
5.5 – 6.5) that do not cause surface rupture, see Wong et al., 2013).
Seismic hazard assessments have been evaluated previously for the Clive site including
assessments of active or potentially active faults in the region and background earthquakes. The
peak ground accelerations for both seismic sources is 0.24 g. The peak ground accelerations for
the Clive site are within the range of estimated ground accelerations for two DOE regulated and
approved low-level waste disposal sites (Area G, Los Alamos, New Mexico (LANL, 2008), and
Area 5, NNSS, Shott et al. 2008). Performance assessments for these sites conclude that the
impacts of ground shaking on waste disposal systems are minor and are overshadowed by the
longer-term effects of subsidence.
The negligible effects of the peak ground accelerations on the long-term stability of Clive’s
embankments has previously been demonstrated and found acceptable by the Division. No new
information on seismic hazards has been identified that would change or require revisions of the
previous work.
The following sections summarize the results of seismic hazard assessments for the Clive site:
“The seismic hazard assessment is based on an assessment of the peak ground acceleration
(PGA) associated with the Maximum Credible Earthquake (MCE) for known active or
potentially active faults in the site region, and the PGA obtained from a probabilistic seismic
hazard analysis (PSHA) to assess the seismic hazard for earthquakes that may occur on
unknown faults in the area surrounding the project site (i.e., background seismicity). For
fault sources, the PGA is calculated at the 84th percentile level and is based on the maximum
rupture length and rupture area for each fault. The return period for ground motions resulting
from a background earthquake is identified as 5000 years (equal to a one percent probability
of exceedance [sic] in 50 years). The approach to select a MCE PGA from the larger of the
values associated with the deterministic MCE for faults or the PSHA result for background
earthquakes at a 5000 year return period is consistent with the discussions among AMEC,
ES, Utah DEQ and their peer reviewer, URS Corporation, and is consistent with the
recommendations of the Utah Seismic Safety Commission (2003) and as required by the
Utah Division of Water Rights (Dam Safety Section) for assessment of dams.
The deterministic assessment follows the approach described in our October 25, 2011 letter,
and is updated in the following paragraphs. Potential fault sources are shown on Figure B-1.1
and are listed in Table B-1.1 of Appendix B, including an assessment of the fault parameters,
source to site distance, and PGA. Specific fault parameters and other information in Table B-
1.1 include fault name, slip type, maximum magnitude, location of site on hanging wall or
footwall, fault dip, rake, maximum rupture length (fault length), downdip rupture width,
distance measures required for ground motion attenuation relationships, and PGA for median
and 84th percentile levels. We use a suite of four Next Generation Attenuation (NGA)
relationships . . . all of which are applicable for the site conditions and types of sources in
Utah and the Intermountain Region. Additional parameters for attenuation relationships
include site shear wave velocity, VS30, taken as 305 m/s as described in the October 25
Letter, and depth to top of bedrock (Z1.0 and Z2.5), taken as default values calculated from
the site VS30 as recommended by the authors of the NGA relationships (also as described in
the October 25 Letter).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 12
The maximum magnitude for each fault is based on rupture of the full length of the fault, and
where available is taken as the maximum value published by the Utah Working Group on
Earthquake Probabilities (WGUEP, 2011), except for the Stansbury fault as noted below. For
faults not assessed in the previous studies, including the Skull Valley fault, the maximum
magnitude was assessed using the same methodology as the WGUEP study, based on
maximum rupture length, rupture width, and the empirical relationships of Wells and
Coppersmith (1994). For short faults where the calculated maximum magnitude is less than
MW 6.5, a maximum magnitude of 6.5 is adopted because this is judged to be a reasonable
minimum value of magnitude for earthquakes that rupture to the ground surface.
For the Stansbury fault, the maximum magnitude is assessed as MW 7.3 based on
consideration of the maximum rupture length, fault width, and maximum fault displacement
identified in previous investigations. . . The value of MW 7.5 listed in the October 25 Letter
and by the WGUEP is judged to be too conservative because it is higher than the maximum
value obtained from empirical relationships, considering all combinations of rupture length,
rupture width, and maximum fault displacement cited in those previous investigations. We
note that it may be reasonable to consider an extreme value with a very low weighting (e.g.,
less than 10 percent) in a probabilistic analysis, but that it is not reasonable practice to adopt
an extreme value for the MCE for a deterministic analysis.
The maximum of the 84th percentile PGA values calculated for the Mmax events on the fault
sources is equal to 0.24 g, as obtained for the Stansbury and the Skull Valley faults (Table
B-1.1). For the PSHA, we used the current version (Ver. 7.62) of commercial program
EZ-FRISK to calculate the PGA for the background earthquake. The program developer,
Risk Engineering, has prepared input fault and background seismicity files for Utah for use
in calculating seismic hazard; these files are based on the same fault source parameters and
independent seismicity catalog used by the U.S. Geological Survey (USGS) to prepare the
2008 National Seismic Hazard Maps.
The seismicity catalog is an independent (de-clustered) catalog based on moment magnitude
(MW) that covers the Western United States; the seismicity in the vicinity of the project site
is shown on Figure B-1.1. The recurrence rates for the background seismicity are based on
the same recurrence models and maximum magnitudes used by USGS, which is a spatially
smoothed gridded approach, with a maximum magnitude of 7.0 for Utah (Peterson et al.,
2008). As for the deterministic analysis, we use the same suite of four NGA relationships
and the site VS30 of 305 m/s. The PGA is taken as the weighted average of the mean values
for the four NGA relationships at a return period of 5000 years (equal to 0.24 g, Table
B-1.1).
The largest PGA from the deterministic assessment of fault-specific sources and the
probabilistic assessment of the background earthquake is 0.24 g. The maximum magnitude
varies from 7.0 to 7.3 for the sources that result in the maximum PGA; we identify the
largest value, MW 7.3, as appropriate for use in the seismic stability analyses for this
project.” (EnergySolutions, 2012, pg. 2-3).
In review of this information and its implications on the Class A West Embankment (CAW)
design, the Division concluded, “Based on the information summarized above, the Division
concludes that the Licensee’s proposed design basis conditions and justification for the design
criteria for waste placement and backfill for the CAW Embankment are acceptable.” (DRC,
2012, pg. 33).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 13
3.3.3 Eolian Deposition
Recent field studies (Neptune 2015) provide evidence for a site-specific conceptual model of a
Holocene history of weak development of soil profiles (limited pedogenesis) in a setting
influenced by low rates of deposition of eolian silt. The Site is within a region of significant
eolian activity evidenced by locally thick accumulation of gypsum dunes west and southwest of
the site and a laterally continuous layer of suspension fallout silts preserved beneath the modern
surface throughout the Clive site. Clive quarry exposures examined in a field study (Neptune
2015) showed sections of eolian silts immediately below a modern vegetated surface (Figure 4).
The bottom of the eolian silt formed a gradational but definable contact with the lake muds and
marl below. The upper vegetated surface at the top of the eolian section was distinct and noted as
being partially indurated. In addition, buried soils were found in the eolian and lake sediments
below the Lake Bonneville lacustrine sequence.
The eolian deposits in the upper part of the stratigraphic section shown in Figure 4 represent a
10,000-year-old record of deposition and soil formation (Neptune 2015). Primary soil features
developed over this time interval include an indurated Av-zone, and slight reddening of the silt
profile with local platy structure from formation of clays (Figure 5). These observations are
consistent with slow processes of pedogenesis in a high elevation semi-arid setting and
continuing suppression and burial of developing soils by a relatively low rate of deposition of
eolian silt. There is no evidence of soil structure development extensive enough to influence soil
hydraulic properties.
Figure 4. Eolian silt in trench located at Clive Pit 29 overlying Lake Bonneville
sedimentary deposits (Neptune 2015).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 14
Observations of Holocene eolian silt throughout the Clive site support a conceptual model of
long-term eolian deposition on a stable surface that promotes and preserves concurrent eolian
deposits which are only slightly modified by slow processes of soil formation. The past
Holocene depositional conditions at the Clive site are promoted by a combination of extensive
wet playa sources of eolian source material to the west and southwest of the Clive site and the
extremely low gradient paleo-Lake Bonneville surface surrounding the site with sparse surface
vegetation and limited surface erosion. These conditions will persist at the Clive site as long as
the lake levels remain below the site elevation. Rates of eolian deposition would be expected to
increase as future lakes approach the site with increased formation of dunes (deposition of eolian
sands). Recurring lakes during ice ages (climate cycles) will rework and mix the eolian deposits
with aggrading clastic lake sediments. The expectation is that eolian deposits will drape and
slightly stabilize closure covers until future lakes return to the Clive site.
3.4 Hydrology
3.4.1 Surface Water
The Clive site is located within a hydrologically closed basin west of the Cedar Mountains. As
there is no outlet from the basin, any water that would flow by the site would pond several miles
to the west in a playa (NRC, 1993).
No surface water bodies are present on the Clive site and any stream flows from higher
elevations usually evaporate and/or infiltrate before reaching flatter land (NRC, 1993). Indicators
Figure 5. An example of upper soil-modified eolian silt in Pit 29. Basal contact of the silt is
approximately located at the middle of the pick handle. Lake Bonneville marl is at the
bottom of the pick handle.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 15
of channelized flow are not present on the Clive site (Baird et al., 1990). The nearest stream
channel ends about 3.2 km (2 mi) east of the site, and the nearest water body that is utilized is
approximately 45 km (28 mi) to the east. The only significant water body in the region is Great
Salt Lake. NRC (1993) indicates that no historical (chronic) flooding has occurred in the vicinity
of the site. Given the 1300-m elevation of the Clive facility, it is not subject to flooding from the
Great Salt Lake, which is not expected to exceed 1285 m (4217 ft) amsl (NRC, 1993).
3.4.2 Groundwater
The NRC recognizes “groundwater” to include all subsurface water, in both unsaturated and
saturated zones. This convention is used in the following descriptions.
3.4.2.1 Groundwater Flow Regime
Groundwater at the Clive site is found within a low-permeability saline aquifer starting near the
bottom of the Unit 3 stratigraphic unit, and saturating the Unit 2 stratigraphic unit. The depth to
groundwater is between approximately 6 and 9 m (20 and 30 ft) bgs at an approximate elevation
of 1295 m (4250 ft) amsl (Brodeur, 2006).
The regional (saturated) groundwater system flows primarily to the east-northeast toward the
Great Salt Lake (Envirocare 2004) and the local shallow groundwater follows a slight horizontal
gradient to the north-northeast (Brodeur, 2006).
Recharge to the aquifer in the vicinity of Clive is thought to be composed of three components: a
small amount due to vertical infiltration from the surface, some small amount of lateral flow
from recharge areas to the east of the site, and the majority of recharge believed to be from
upward vertical leakage from the deeper confined aquifer (Bingham Environmental (1994).
Average annual groundwater recharge from the surface in the southern Great Salt Lake Desert in
the precipitation zone typical of Clive was estimated by Gates and Krauer (1981, Table 2). An
estimated 0.37 hm3/yr (300 acre-feet per year) were recharged to lacustrine deposits and other
unconsolidated sediments over an area of 19,000 ha (47,100 acres). This is a recharge rate of
approximately 2 mm/yr (0.08 in/yr). Groundwater recharge from lateral flow occurs due to
infiltration at bedrock and alluvial fan deposits away from the Site which moves laterally through
the unconfined and confined aquifers (Bingham Environmental, 1994). This is evidenced by the
increasing salinity of the groundwater due to dissolution of evaporate minerals as water moves
from the recharge area to the aquifers below the Facility (Bingham Environmental, 1994). The
majority of recharge to the shallow aquifer is believed by Bingham Environmental (1994) to be
due to vertical leakage upward from the deep confined aquifer due to the presence of upward
hydraulic gradients.
Deeper saturated zones in Unit 1 below approximately 14 m (45 ft) bgs are reported to show
higher potentiometric levels than the shallow unconfined aquifer. Differences in potentiometric
levels are attributed to the presence of the Unit 2 clays (Bingham Environmental, 1994). Vertical
gradients between shallow and deeper screened intervals in the monitor well clusters were
calculated by Bingham Environmental (1994). An upward vertical gradient was observed
ranging in magnitude from 0.02 to 0.04 based on the distance between the screen centers. For a
vertical hydraulic conductivity of 1 × 10-6 cm/s (Bingham Environmental 1994) this corresponds
to a recharge range from 6 to 13 mm/yr (0.25 to 0.5 in/yr).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 16
Estimates of vertical recharge from the surface take into account natural processes such as snow
accumulation and melting, concentration of water in topographic depressions, drainages,
fractures, holes, or burrows and increased surface permeability due to frost heave or plant roots.
When features such as topographic depressions, drainages, or fractures result in enhanced
infiltration, the vertical infiltration below the localized recharge points flows laterally at the
water table toward the lower elevations of the water table (Freeze and Cherry, 1979). The effect
of animal burrowing on subsurface moisture content was investigated in a field experiment at the
Hanford Site by Landeen (1994). Over the course of five testing periods, three during the
summer and two during the winter soil moisture measurements showed no influence of
burrowing activities on long-term water storage.
Degradation models for changes in cover properties over time leading to increased vertical flow
were discussed in the Benson et al. (2011) report published by the NRC. While this is a useful
report, the topic of cover performance is a complex topic with a wide range of research and
programmatic applications (for example, ongoing work in the NRC, DOE, CERCLA/RCRA and
international communities). Any modifications in data and model assumptions used for cover
properties and cover performance should be based on information from multiple referenced
sources. More importantly, the long-term performance and changes in cover performance over
time are strongly dependent on the type of closure cover (for example, engineered, ET cover)
and the climate setting for the cover application.
3.4.2.2 Groundwater Quality
The underlying groundwater in the vicinity of the Clive site is of naturally poor quality because
of its high salinity and its high content of total dissolved solids (TDS), as a consequence, is not
suitable for most human uses (NRC, 1993). Brodeur (2006) reports that groundwater beneath the
Clive site had a TDS content of 40,500 mg/L (40.5‰). The majority of the cations and anions
are sodium and chloride, respectively. This is not potable for humans or livestock, nor is it
suitable for irrigation. For comparison purposes, seawater typically has a salinity of about 35‰,
making the Clive groundwater only slightly higher than average seawater.
3.5 Ecology
NRC (1993) and Envirocare (2000) characterized the Clive facility as a homogeneous, semi-
desert low shrubland, primarily composed of shadscale (Atriplex confertifolia). The shrubland is
part of the Northern Great Basin Desert Shrub Biome and has been described as a saltbrush-
greasewood shrub complex. The development of modeling of biotic processes is detailed in the
Biological Modeling white paper.
3.5.1 Local Vegetation
The vegetation communities that occur on and near Clive were documented during 2010 and
2012 field studies (SWCA 2011, 2012). Inter-Mountain Basins Mixed Salt Desert Scrub (Lowry
2007) is the dominant vegetation cover type on analogs to the Clive site. The target vegetation
community on the ET cover consists of approximately 15% cover of small stature native shrub
species (Atriplex confertifolia, Atriplex canescens, Bassia americana, Picrothamnus desertorum,
and Suaeda torreyana), with additional cover provided by sparse native forbs and grasses (p.35,
SWCA 2013).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 17
Several plant communities identified include shadscale-gray molly (Kochia americana var.
vestita), shadscale-gray molly-black greasewood (Sarcobatus vermiculatus), and black
greasewood-gardner saltbrush (Atriplex nuttallii). Shrubs are widely spaced, totaling between
1.5% and 20% ground cover, depending upon vegetation association. The shadscale-gray molly
community covers most of the South Clive site, with black greasewood becoming prominent
only on the eastern quarter of the site. SWCA (2011) found very little transition between the
shadscale-gray molly and black greasewood vegetation associations, and that shadscale and gray
molly totaled less than 0.5% cover in the greasewood association, suggesting that the shadscale-
gray molly-black greasewood community identified by Envirocare (2000) is perhaps better
classified as a pure greasewood community. Envirocare reported that the black greasewood-
gardner saltbush community only occurs in the far northeast corner of the Clive site. Seepweed
(Suaeda torreyana), perfoliate pepperweed (Lepidium perfoliatum), and halogeton (Halogeton
glomeratus) are the most common understory plants. Sage (Artemisia spp.) and rabbitbrush
(Chrysothamnus spp.) which are characteristic of much of the Great Basin shrubland, do not
occur on the valley floors around Clive due to their low salt tolerance, but may occur on bajadas
and well-drained slopes. No threatened or endangered plant species are known to occur in the
near vicinity of the Clive site (NRC, 1993).
3.5.2 Local Wildlife
The Clive site consists of two main habitat types, shadscale flats and greasewood.
Comprehensive faunal surveys have not been conducted around the Clive site, but NRC (1993)
indicates that species diversity is low. Species typical of these shrubland habitats include black-
tailed jackrabbit (Lepus californicus), Townsend’s ground-squirrel (Spermophilus townsendii),
Ord’s kangaroo rat (Dipodomys ordii), deer mouse (Peromyscus maniculatus), horned lark
(Eremophila alpestris), and the desert horned lizard (Phrynosoma platyrhinos). Jackrabbits, deer
mice, and grasshopper mice (Onychomys leucogaster) were the only mammals trapped during
surveys conducted for the 1993 Environmental Impact Statement (EIS) (NRC 1993). Additional
trapping conducted in October 2010 collected only deer mice at the Clive site, and deer mice,
grasshopper mice, Ord’s kangaroo rat, and chisel-toothed kangaroo rat in neighboring areas with
steeper slopes and greater density of grasses (SWCA 2011). Pronghorn antelope can also be
found near the facility, but the area is considered to be poor habitat (NRC, 1993). The bald eagle
and the peregrine falcon are two federally-listed species that could occur in the project area.
However, NRC (1993) indicates that the U.S. Fish and Wildlife Service concurs with the
conclusion that the project site would not affect either species due to the distance to the nearest
nesting site.
A variety of invertebrates is expected to occur at the Clive site. Invertebrates, particularly ants,
play a key role in maintenance of desert shrub communities. Harvester ants of the genus
Pogonomyrmex create large, easily recognizable nests, and play an important role in the
development of desert soils and the dispersal of plant seeds. Surveys conducted in 2010 found
that the Western harvester ant (Pogonomyrmex occidentalis) was by far the dominant ant species
at the site, independent of vegetative association (SWCA 2011).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 18
3.6 Engineered Features
3.6.1 Federal Cell Disposal Cell Design
Depleted uranium waste is proposed for disposal in the Federal Cell. The waste footprint in the
Federal Cell is about 541 × 402 m (1,775 × 1,318 ft), with an area of approximately 22 ha
(54 acres), and an estimated total waste volume of about 2.1 million m3 (2.7 million yd3). A
drainage ditch surrounds the disposal cell. The cell is constructed on top of a compacted clay
liner covered by a protective cover. Waste will be placed above the liner and will be covered
with a layered engineered cover constructed of natural materials. The top slopes will be finished
at a grade of 2.4% while the side slopes will be no steeper than 5:1 (20% grade).
The design of the Federal Cell cover has been engineered to prevent the effects of erosion,
reduce the effects of infiltration, and to protect workers and the public from radionuclide
exposure. The Cell cover is a layered composite of a clay radon barrier, frost protection material,
an evaporative layer composed of Unit 4 material, and a surface layer composed of Unit 4
material with 15% gravel on the top slopes and 50% gravel on the side slopes. The Surface Layer
of silty clay provides storage for water accumulating from precipitation events, enhances losses
due to evaporation, and provides a rooting zone for plants that will further decrease the water
available for downward movement. The purpose of the Evaporative Zone Layer is to provide
additional storage for precipitation and additional depth for plant rooting zone to maximize ET.
The detailed properties of each cell layer may be found in the Unsaturated Zone Modeling white
paper accompanying the Clive DU PA Model.
3.6.2 Degradation of Engineered Features
Whereas the engineered liner and cover are expected to be constructed as designed, and to
perform well over the coming decades, they will likely degrade with time. Sheet erosion by wind
and water is expected to be minor, and is likely to be counteracted by eolian deposition of loess
(wind-blown sediment) filling the interstices of the gravel. It is possible, however, that the
surface layer may be degraded by processes such as unusual weather events (e.g., tornadoes),
animal and plant activity, or human activities after the loss of institutional control. These events
may result in damage to the cover, though the damage is likely to be localized. Details are
provided in the Erosion Modeling white paper accompanying the Clive DU PA Model.
4.0 Regulatory Context
EnergySolutions is permitted by the State of Utah to receive Class A low-level and mixed
low-level radioactive waste (LLW and MLLW) under Utah Administrative Code (UAC)
R313-25, License Requirements for Land Disposal of Radioactive Waste (Utah, 2015a). The
wastes that are received must be classified in accordance with the UAC R313-15-1009,
Classification and Characteristics of Low-Level Radioactive Waste (Utah, 2015b). The
classification requirements in UAC R313-15-1009 reflect those outlined in NRC’s 10 CFR 61
Section 55, but include additional references to radium-226 (226Ra). Further, groundwater
protection levels (GWPLs) must be adhered to, as outlined in the site’s Ground Water Quality
Discharge Permit (UWQB, 2010). The regulatory context within the Federal and State
regulations is discussed in the following sections.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 19
4.1 Nuclear Regulatory Commission Regulations
Title 10 CFR 61 (Code of Federal Regulations, 2007) is the Federal regulation for the disposal of
certain radioactive wastes, including land disposal at privately-operated facilities such as that
managed and operated by EnergySolutions at Clive, Utah. It contains procedural requirements,
performance objectives, and technical requirements for near-surface disposal, including disposal
in engineered facilities with protective earthen covers, which may be built fully or partially
above-grade. Near-surface disposal is defined as disposal in or within the upper 30 meters of the
earth’s surface (10 CFR 61.2).
The promulgation of 10 CFR 61 required a Final Environmental Impact Statement (FEIS) which
was issued in 1982 (NRC, 1982). The FEIS focused on the waste streams typically disposed by
NRC licensees at the time, and did not take into account facilities that generated high
concentrations and large quantities of DU, which was not then considered to be waste. As a
result, the NRC did not establish a concentration limit for uranium isotopes in the waste
classification tables presented in 10 CFR 61.55.
4.1.1 Section 61.55: Waste Classification
Section 61.55 defines three classes of radioactive waste for near surface disposal—Class A,
Class B, Class C—and discusses the fourth, commonly called “greater than Class C” (GTCC)
waste, which, “in the absence of specific requirements in this part […] must be disposed of in a
geologic repository […] unless proposals for disposal of such waste in a disposal site licensed
pursuant to this part are approved by the Commission” (§61.55[2][iv]). The Class A, B, and C
wastes are defined based on concentrations of specific long-lived radionuclides (defined in Table
1 of §61.55), or, in the absence of long-lived ones, on specific short-lived radionuclides (defined
in Table 2 of §61.55). These tables are reproduced in Figure 6 for convenience.
Wastes containing radionuclides listed on both tables are classified using a combination
approach as specified in §61.55(5):
§61.55(5) Classification determined by both long- and short-lived radionuclides. If
radioactive waste contains a mixture of radionuclides, some of which are listed in Table 1,
and some of which are listed in Table 2, classification shall be determined as follows:
(i) If the concentration of a nuclide listed in Table 1 does not exceed 0.1 times the value
listed in Table 1, the class shall be that determined by the concentration of nuclides listed in
Table 2.
(ii) If the concentration of a nuclide listed in Table 1 exceeds 0.1 times the value listed in
Table 1 but does not exceed the value in Table 1, the waste shall be Class C, provided the
concentration of nuclides listed in Table 2 does not exceed the value shown in Column 3 of
Table 2.
The scope of the Clive DU PA Model includes the disposal of DU, which by default falls
into the category of Class A waste:
§61.55(6) Classification of wastes with radionuclides other than those listed in Tables 1 and
2. If radioactive waste does not contain any nuclides listed in either Table 1 or 2, it is Class
A.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 20
Nevertheless, DU presents an interesting case, as the uranium it contains is fundamentally
different from the Class A wastes that NRC had in mind when it devised the classifications.
Uranium does not appear in Table 1 of 10 CFR 61.55 (Figure 6) because, at the time of the
development of the regulation, uranium waste did not, and was not expected to, exist in
significant quantities. The nature of the radiological hazards associated with DU presents
challenges to the estimation of long-term effects from its disposal. As DU evolves toward secular
equilibrium with its progeny, a process that will take over 2 million years, it becomes a greater
radiological hazard due to the in-growth of its decay products. Recognition of this special
behavior of DU has prompted the NRC to revisit the regulation in a rule-making. This is
discussed in Section 4.1.5. Until that rule-making is complete, however, 10 CFR 61 stands as the
controlling regulation.
4.1.2 Section 61.41: Protection of the Public
The key endpoints of a PA are estimated future potential doses to members of the public (MOP)
and the general population. The performance objectives specified in Subpart C of 10 CFR 61 are
in the following section:
§ 61.41 Protection of the general population from releases of radioactivity.
Concentrations of radioactive material which may be released to the general environment in
ground water, surface water, air, soil, plants, or animals must not result in an annual dose
exceeding an equivalent of 25 millirems [0.25 mSv] to the whole body, 75 millirems [0.75
mSv] to the thyroid, and 25 millirems [0.25 mSv] to any other organ of any member of the
public. Reasonable effort should be made to maintain releases of radioactivity in effluents to
the general environment as low as is reasonably achievable.

Figure 6. Waste classification Tables 1 and 2 from 10 CFR 61.55.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 21
However, the approach to dose assessment suggested by §61.41 is now dated, and NRC
recommends the current International Commission on Radiological Protection 30 (ICRP 1984)
methodology in their Performance Assessment Methodology, NUREG-1573 (NRC 2000):
3.3.7.1.2 Internal Dosimetry
The NRC performance objective set forth in Section 61.41, is based on the ICRP 2 dose 3-79
methodology (ICRP, 1979), but current health physics practices follow the dose
methodology used in Part 20, which is currently based on ICRP 30 methodology
(ICRP,1979). The license application will contain many other assessments of potential
exposures (e.g., worker exposure, accident exposures, and operational releases) that will
need to use ICRP 30 dose methodology. For internal consistency in the application, it is
recommended that the performance assessment be consistent with the methodology approved
by the NRC in Part 20 for comparison with the performance objective. Therefore, PAWG
[the performance assessment working group] believes that calculation of a TEDE [total
effective dose equivalent] for the LLW performance assessment—a summation of the annual
external dose and the CEDE [committed effective dose equivalent]—is acceptable for
comparison with the performance objective. As a matter of policy, the Commission considers
0.25 mSv/year (25 mrem/year) TEDE as the appropriate dose limit to compare with the
range of potential doses represented by the older limits that had whole-body dose limits of
0.25 mSv/year (25 mrem/year) (NRC, 1999, 64 FR 8644; see Footnote 1). Applicants do not
need to consider organ doses individually because the low value of the TEDE should ensure
that no organ dose will exceed 0.50 mSv/year (50 mrem/year).
The estimation of dose to receptors in the Clive DU PA Model therefore uses the ICRP 30 TEDE
approach. There are a number of implicit assumptions in using dose as a performance metric, in
that it is being used as a proxy for risk. Risk involves a biological effect. The biological effect of
greatest interest at the doses evaluated here is cancer. The risk of cancer to an exposed individual
depends upon a large number of assumptions, the most influential being 1) that the major source
of data for radiological risk assessment; i.e., the Hiroshima/Nagasaki atomic bomb survivors, is
relevant for the doses evaluated, and 2) that risks can be extrapolated from large doses to small
doses in a linear fashion, with no threshold of effect (i.e., no dose is without some risk of
cancer). Both of these assumptions are controversial, yet provide the basis for most radiation
regulation. The implications of these assumptions are discussed in the Dose Assessment white
paper.
4.1.3 Section 61.42: ALARA and Collective Dose
A second potential decision rule pertains to populations. There is no clear decision rule as far as
collective (cumulative population) doses are concerned. However, the regulations state that
"reasonable effort should be made to maintain releases of radioactivity in effluents to the general
environment as low as is reasonably achievable" (ALARA). There are, however, other
competing objectives, and the resource implications are large to achieving ALARA on a
collective level. Additionally, the words "reasonably" and "achievable" are not precise. The two
words perhaps imply some degree of consideration of trade-offs, but no clear definition is
published. Assuming that there are trade-offs, then this implies that an analysis that explicitly
evaluates the trade-offs, and how different disposal options, designs, or sites may differentially
satisfy the objectives and resource constraints (e.g., a decision or economic analysis) should be
performed. Yet, at present, this has yet to be conducted in the context of the PA process, and
there are no current specific regulations. However, the ICRP (1984) provides guidance regarding
potential approaches.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 22
4.1.4 Section 61.42: Protection of the Inadvertent Intruder
In addition to protecting any MOP, 10 CFR 61 requires additional assurance of protecting
individuals from the consequences of inadvertent intrusion. An inadvertent intruder is someone
who is exposed to waste without meaning to, and without realizing it is there (after loss of
institutional control). This is distinct from the intentional intruder, who might be interested in
deliberately disturbing the site, or extracting materials from it, or who might be driven by
curiosity or scientific interest.
§ 61.42 Protection of individuals from inadvertent intrusion.
Design, operation, and closure of the land disposal facility must ensure protection of any
individual inadvertently intruding into the disposal site and occupying the site or contacting
the waste at any time after active institutional controls over the disposal site are removed.
Because the definition of inadvertent intruders encompasses exposure of individuals who engage
in normal activities without knowing that they are receiving radiation exposure, there is no
practical distinction made here between a MOP and inadvertent intruders with regard to
exposure/dose assessment.
4.1.5 Proposed Rule-Making Regarding 10 CFR 61
In 2005, the NRC proposed to consider whether or not large quantities of DU, such as that
produced from uranium enrichment facilities, warrant an amendment of the waste classification
tables currently defined in 10 CFR 61 (NRC, 2005). In 2008, NRC staff responded to the
October 2005 order that evaluated a generic case to determine if Part 61 standards could be met
for near-surface disposal of DU (NRC, 2008). The results of this evaluation indicated that it may
be possible, given certain conditions, to meet the standards for near-surface disposal of DU.
Furthermore the NRC staff prepared several regulatory options. NRC staff also recommended
that no classification change be made for DU, retaining its status as Class A waste, but that
additional language be included requiring a site-specific PA prior to the acceptance of DU for
disposal. In March 2009, the NRC agreed with the course of action recommended by the NRC
staff in SECY-08-0147 and decided to keep DU classified as a Class A waste (NRC, 2009a).
They also decided to initiate rule- making that would propose enhanced PA requirements for
those facilities that plan to dispose of large quantities of DU (NRC, 2009b).
Most of the proposed changes to 10 CFR 61 involve the concept that no matter what
classification DU is given, any disposal of the material should involve an analysis that will
inform decision makers about the doses associated with such a disposal to individuals who might
be exposed at some time after site closure. This position is substantially in concordance with that
put forth by the National Research Council (2005), and with the approach that will be used in the
Clive DU PA Model.
4.2 State of Utah Regulations
Utah is an NRC agreement state, meaning that it is granted authority to enforce NRC regulation,
or regulations of its own drafting that are compatible with the NRC regulation, 10 CFR 61. The
State of Utah has done so, in two Rules of the Utah Administrative Code (UAC): UAC Rule
R313-25 License Requirements for Land Disposal of Radioactive Waste, and Rule R313-15
Standards for Protection Against Radiation (Utah, 2015). Each of these is discussed below.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 23
4.2.1 Section R313-25: Licensing Requirements
Section R313-25-9 Technical Analyses. Parts (4)(a) and (b) of this Section are patterned closely
after 10 CFR 61.41 and 42:
(4) The licensee or applicant shall also include in the specific technical information the
following analyses needed to demonstrate that the performance objectives of Rule R313-25
will be met: (a) Analyses demonstrating that the general population will be protected from
releases of radioactivity shall consider the pathways of air, soil, ground water, surface water,
plant uptake, and exhumation by burrowing animals. The analyses shall clearly identify and
differentiate between the roles performed by the natural disposal site characteristics and
design features in isolating and segregating the wastes. The analyses shall clearly
demonstrate a reasonable assurance that the exposures to humans from the release of
radioactivity will not exceed the limits set forth in Section R313-25-20.
(b) Analyses of the protection of inadvertent intruders shall demonstrate a reasonable
assurance that the waste classification and segregation requirements will be met and that
adequate barriers to inadvertent intrusion will be provided. Analyses of the protection of
inadvertent intruders shall demonstrate a reasonable assurance that the waste classification
and segregation requirements will be met and that adequate barriers to inadvertent intrusion
will be provided.
In addition, a new section for R313-25-9 has recently been adopted, and is reproduced here:
(5)(a) Notwithstanding Subsection R313-25-9(1), any facility that proposes to land dispose
of significant quantities of concentrated depleted uranium (more than one metric ton in total
accumulation) after June 1, 2010, shall submit for the Director's review and approval a
performance assessment that demonstrates that the performance standards specified in 10
CFR Part 61 and corresponding provisions of Utah rules will be met for the total quantities
of concentrated depleted uranium and other wastes, including wastes already disposed of and
the quantities of concentrated depleted uranium the facility now proposes to dispose. Any
such performance assessment shall be revised as needed to reflect ongoing guidance and
rulemaking from NRC. For purposes of this performance assessment, the compliance period
shall be a minimum of 10,000 years. Additional simulations shall be performed for the
period where peak dose occurs and the results shall be analyzed qualitatively.
4.2.2 Section R313-15-1009: Waste Classification
Rule R313-15 contains section R313-15-1009 Classification and Characteristics of Low- Level
Radioactive Waste. The definitions in this section are essentially identical to those in 10 CFR
61.55, with one exception: Utah adds 226Ra to the list of long-lived radionuclides in the
regulation’s Table I (see Figure 7), with a concentration limit of 100 nCi/g (Utah, 2010). 226Ra is
a decay product of uranium-238 (238U), the principal component of DU, it is of direct interest to
the disposal of DU waste. The EnergySolutions Clive facility is licensed by the State of Utah for
disposal of Class A waste. The DU wastes under consideration for disposal in the present PA,
however, contain isotopes of uranium, potentially including some radionuclides listed in the
tables shown in Figure 6 in addition to the 226Ra added by Utah (Figure 7). In particular, the DU
from certain sources contains some amount of technetium-99 (99Tc). Therefore, the
determination of classification is driven not by the presence of uranium, but by the presence of
radionuclides in the tables, as discussed in the quotation from §61.55(5) above.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 24

4.2.3 Groundwater Protection Limits
In addition to these radiological criteria, the State of Utah imposes limits on groundwater
contamination, as stated in the Ground Water Quality Discharge Permit (UWQB, 2010). Part
I.C.1 of the Permit specifies that GWPLs in Table 1A of the Permit shall be used for the Class A
LLW Cell. Table 1A in the Permit specifies general mass and radioactivity concentrations for
several constituents of interest to DU waste disposal. These GWPLs are derived from Ground
Water Quality Standards listed in UAC R317-6-2 Ground Water Quality Standards. Exceptions
to values in that table are provided for specific constituents in specific wells, tabulated in Table
1B of the Permit. This includes values for mass concentration of total uranium, radium, and gross
alpha and beta radioactivity concentrations for specific wells where background values were
found to be in exceedence of the Table 1A limits. Note that according to the Permit, groundwater
at Clive is classified as Class IV, saline ground water, according to UAC R317-6-3 Ground
Water Classes, and is highly unlikely to serve as a future water source. As noted in Section 0, the
underlying groundwater in the vicinity of the Clive site is of naturally poor quality because of its
high salinity and, as a consequence, is not suitable for most human uses, and is not potable for
humans.
The Clive DU PA Model calculates estimates of groundwater concentrations at a virtual well
near the Federal Cell for comparison with these GWPLs.
Figure 7. Waste classification Table I from R313-15-1009.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 25
5.0 Summary of Features, Events, and Processes
A requirement for the PA scenario development process is the preliminary identification of
possible future states of the disposal system as it is subjected to external changes and factors
(e.g., climate, weathering, demographic changes) over time. The identification of features,
events, and processes (FEPs) is a key activity in developing scenarios for the Clive DU PA
Model . The identification, compilation, and screening of FEPs form the basis for scenarios and
quantitative analyses used to evaluate site performance.
The list of FEPs pertaining to the efficacy of disposal and storage of DU waste at the Clive
Facility was compiled from several PA-related FEPs documents published for other radiological
waste disposal facilities (e.g., NEA, 1992; NEA, 2000; Guzowski, 1990; Guzowski and
Newman, 1993). In addition to existing PA literature sources for FEPs, site-specific
understanding of the environmental and engineered attributes of the Clive facility, geographical
region, and population were also addressed in the compilation of FEPs for this assessment.
All FEPs identified in the literature and developed internally were compiled into an exhaustive
initial list. This list was iteratively reviewed to reduce duplication among sources and to more
broadly (or more precisely) group related FEPs for incorporation in the CSM. For each group of
related FEPs, the rationale for its inclusion in or dismissal from the model was documented.
This section of the CSM identifies the FEPs and conditions pertaining to the conceptual model
that are retained for use in developing the Clive DU PA Model. Details related to the
identification and screening processes are discussed in the accompanying FEP Analysis
document. Features, events, and processes were grouped into several categories based on
groupings listed in the original source documents, and include some overlap and redundancy.
Nevertheless, the groupings are not significant with respect to the CSM. What is important is that
the FEPs are considered in the appropriate parts of the model. Only those FEPs retained for
further consideration are discussed here. Once identified, these FEPs are qualitatively evaluated
for inclusion in the CSM based on considerations of their likelihood and consequence.
Meteorology
Frost weathering and other meteorological events (e.g., precipitation, atmospheric dispersion,
resuspension) are included in the CSM. Weathering may occur from frost cycles. Resuspension
of particulates from surface soils allows them to be redistributed by atmospheric dispersion,
which is a meteorological phenomenon. Dust devils are also possible at the site and a tornado
occurred in Salt Lake City in 1999, which was the first tornado in Utah in over 100 years.
Climate change
Features, events, and processes of climate change considered in the conceptual model include
effects on hydrology (including lake effects), hydrogeology, biota, and human behaviors. Lake
effects include appearance/disappearance of large lakes and associated phenomena
(sedimentation, wave action, erosion/inundation). Wave action, including seiches, is included in
the CSM.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 26
Hydrology
Several hydrogeological FEPs were identified for consideration in the conceptual model.
Groundwater transport, in both the unsaturated and saturated zones, is potentially a significant
transport pathway. For some model endpoints, such as groundwater concentrations that are
compared to GWPLs, it is the only pathway of concern.
Groundwater flow and transport processes include advection-dispersion, diffusion, changes in
the flow system, recharge, and brine interactions. Inundation of the site may occur due to
changes in lakes or reservoirs, which is included in lake effects of climate change.
Geochemical
Geochemical effects include chemical sorption and partitioning between phases, aqueous
solubility, precipitation, chemical stability, complexation, changes in water chemistry (redox
potential, pH, Eh), speciation, and leaching of radionuclides from the waste form. These
processes are addressed in the model.
Other Natural Processes
The broad category of other natural processes considered for the conceptual model include
ecological changes and pedogenesis (soil formation). Ecological changes are associated with
catastrophic events (e.g., inundation), evolution, or climate change. Pedogenesis is expected on
the cover, giving rise to vegetation growth or habitation by wildlife.
Denudation (cover erosion) may be sufficient to expose waste. Erosion of the repository resulting
from pluvial, fluvial or eolian processes can result from extreme precipitation, changes in surface
water channels, and weathering. Sediment transport is an inherent aspect of erosion.
Sedimentation/deposition onto the cell may also affect cell performance.
Note that seismic activity is unlikely to impact the Clive facility. Faults are not present within the
vicinity of Clive, although effects of isostatic rebound are still possible in the Lake Bonneville
area.
Engineered Features
Engineered features are intended to promote containment and inhibit migration of contaminants.
Conditions potentially affecting site performance include failure of engineered features, cell
design, material properties, and subsidence of the cell.
Containerization
Two key components of containerization were identified as FEPs: containment degradation and
corrosion. Canister degradation, including fractures, fissures, and corrosion (pitting, rusting)
could result in containment failure. These processes are evaluated in the conceptual model (See
Section 8.1).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 27
Waste
Attributes of waste that could influence the performance of the Clive facility include the
inventory of radionuclides, physical and chemical waste forms, container performance, matrix
performance, leaching, radon emanation, and other waste release mechanisms.
Source Release
Source release can result from many mechanisms, including containment failure, leaching, radon
emanation, plant uptake, and translocation by burrowing animals. FEPs that fit in the category of
source release include gas generation, radioactive decay and in-growth, and radon emanation.
Contaminant Migration
Contaminant migration for the CSM includes the mechanisms and processes by which
radionuclides may come to be located outside of the containment unit. The following
contaminant migration processes were identified for consideration in the CSM: resuspension,
atmospheric dispersion, biotically-induced transport, contaminant transport, diffusion, dilution,
advection-dispersion, dissolution, dust devils, tornadoes, infiltration, and preferential pathways.
Human exposure pathways could include animal ingestion, both as ingestion of fodder and feed
by livestock, and ingestion of livestock by humans. Transport by atmospheric dispersion could
be associated with limited resuspension, dust devils, and tornadoes. Modeling of biotic (plant-
and animal-mediated) processes leading to contaminant transport, and the evolution of these
processes in response to climate change and other influences, including bioturbation, burrowing,
root development, and contaminant uptake and translocation are considered.
Contaminant transport includes transport media (water, air, soil), transport processes (advection-
dispersion, diffusion, plant uptake, soil translocation), and partitioning between phases.
Diffusion occurs in gas and water phases. Dilution occurs when mixing with less concentrated
water. Hydrodynamic dispersion is associated with water advection. Dissolution in water is
limited by aqueous solubility. Transport in the gas phase includes gas generation in the waste,
partitioning between air and water phases, diffusion in air and water, and radioactive decay and
ingrowth. Infiltration of water through the cover, into wastes, and potentially to the groundwater
is another contaminant migration concern. Preferential pathways for contaminant transport are
also addressed.
Human Processes
The FEPs identified as human processes encompass human behaviors and activities, resource
use, and unintentional intrusion into the repository. Human process FEPs identified for
assessment are related to the human exposure model and include anthropogenic climate change,
human behavior, human-induced processes related to engineered features at the site, human-
induced transport, inadvertent human intrusion, institutional control, land use, post-closure
subsurface activities, waste recovery, water resource management, and military activities.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 28
Exposure
Exposure is an integral part of the conceptual model, and may result from reduced site
performance. Exposure-relevant FEPs identified for evaluation include those related to
dosimetry, exposure media, human exposure, ingestion pathways, and inhalation pathways.
Dosimetry as a science is not a FEP per se but physiological dose response is accounted for in
the PA model.
Transport pathways (e.g. food chains) that lead to foodstuff contamination, and human exposures
due to inhalation of gaseous radionuclides and particulates are included. Exposure media include
soil/dust and food. Exposure pathways (ingestion, inhalation, etc.) and physiological effects from
radionuclides and toxic contaminants (e.g. uranium) are also assessed.
6.0 Waste Forms
The scope of this CSM is limited to the disposal of DU wastes of two general waste types: 1)
depleted uranium trioxide (DUO3) waste from the Savannah River Site (SRS) 2) anticipated DU
waste as U3O8 from gaseous diffusion plants (GDPs) at Portsmouth, Ohio and Paducah,
Kentucky. The quantity and characteristics of DU waste from other sources that has that already
been disposed of at the Clive Facility was not included.
The quantity and characteristics of DU waste will constitute source terms in the Clive DU PA
Model. This section provides background on the uranium cycle and origins and nature of DU
waste in particular.
Depleted uranium consists of three isotopes of uranium (238U, 235U, and 234U) and progeny from
radioactive decay. The wastes proposed for disposal contain these isotopes of uranium, but some
also include other “contaminants” in varying amounts (ORNL 2000, EnergySolutions, 2009b).
These associated radionuclides are the result of introduction of used nuclear fuel (UNF) into the
uranium enrichment process. In order to clarify that these wastes contain more than just DU
(uranium isotopes), they are termed “DU waste.” When this term is used, it refers to wastes, such
as those from SRS that contain DU and a small amount of contamination from actinides and
fission products. If uranium hexafluoride derived from irradiated reactor returns is introduced to
the cascade, some of the associated fission products and actinides end up fixed to the walls of the
DU cylinders containing the 238U. These contamination “heels” will remain in the cylinders
through the process of deconversion, since they are again reused for collecting the U3O8 product.
Depleted Uranium Background
The uranium fuel cycle begins by extracting and milling natural uranium ore to produce “yellow
cake,” a mixture of various uranium oxides. Low-grade natural ores contain about 0.05 to 0.3%
by weight of uranium oxide while high-grade natural ores can contain up to 70% by weight of
uranium oxide (NRC, 2010). Naturally occurring uranium contains the isotopes 238U, 235U, and 234U, and radioactive decay products in secular equilibrium with these primordial parents. Each
uranium isotope has the same chemical properties, but differs in terms of radiological properties.
Naturally occurring uranium has a typical isotopic composition of about 99.283% 238U, 0.711% 235U, and 0.006% 234U by mass, although there are varying assays and estimates.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 29
In order to produce fuel for nuclear reactors and weapons, uranium has to be enriched in the
fissionable 235U isotope. Uranium enrichment began in support of the Manhattan Project during
World War II. Enrichment for civilian and military uses continued after the war under the U.S.
Atomic Energy Commission, and its successor agencies, including the DOE.
The official definition of DU given by the NRC is uranium in which the percentage fraction by
weight of 235U is less than 0.711%. (its natural abundance) According to the International Atomic
Energy Agency (IAEA), typical DU percentage concentration by weight of the uranium isotopes
used for military purposes is 99.8% 238U, 0.2% 235U, and 0.001% 234U. Depleted uranium
isotopic ratio values from gaseous diffusion plants, which processed material for both military
and commercial purposes, are reported to be 99.75% 238U, 0.25% 235U, and 0.0005% 234U (Rich
et al. 1988). Because processing of uranium has only been practiced for roughly 60 years, there
has not been sufficient time for noticeable in-growth of the daughter radionuclides in this by-
product. Depleted refined uranium is therefore considerably less radioactive than natural uranium
because it has less 234U, 235U, and progeny, per unit mass.
6.1 Savannah River Site Uranium Trioxide
The SRS produced DU as a byproduct of the nuclear material production programs, where
irradiated nuclear fuels were reprocessed to separate out the fissionable plutonium-239 (239Pu)
(Fussell and McWhorter, 2002). Uranium billets were produced at the DOE Fernald, Ohio site,
fabricated into targets at SRS, then irradiated in one of the SRS production reactors to produce 239Pu. The irradiated targets were processed in F-Canyon, where in acid solution, the fission
products were separated from the plutonium and uranium, which were then separated from each
other. After additional purification, the DU-bearing waste stream was transferred to the FA-Line
Facility where it was processed into uranium trioxide which is now a focus of this PA. This
DUO3 contains small quantities of waste fission products and transuranic elements
(EnergySolutions, 2009b), which will also be included in the Clive DU PA Model.
The DU waste was produced at the SRS from the 1950s to the late 1980s as a by-product in the
manufacture of nuclear materials, as described above. The DUO3 was produced from DUF6
using a classic chemical separation process to separate and recover plutonium and uranium
product.
The DU was purified through multiple processing steps, and then transferred to a final
production plant for conversion to uranium trioxide. Some of this material was sent off-site for
commercial or military use, and the rest was stored on site, and is now slated for disposal.
The chemical separation process was performed in two separate processing cycles. The more
highly radioactive processing, such as dissolution of irradiated target material from the SRS
reactors, and removal of the vast majority of the highly radioactive fission products and
actinides, was performed in the first processing cycle. The final purification of the uranium
product stream to remove the remaining fission product and actinide “contaminants” was
performed in a second processing cycle. A small fraction of these contaminants was carried
forward with the uranium product. This process ceased operations in the late 1980s.
The SRS produced approximately 36,000 200-L (55-gal) steel drums of DUO3 during the
production campaigns (Fussell and McWhorter, 2002). This DUO3, a solid powder at room
temperature and pressure, is considered to be relatively homogeneous, based on known process
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 30
controls and operations. The drums have an average mass of 680 kg (weight of 1,500 lb) apiece
(Fussell and McWhorter, 2002). The condition of the drums varies from good to poor with a high
percentage of the drums having some degree of outer surface corrosion. A significant number of
drums in two facilities (221-21F and 221-22F) have been placed into overpacks as a mitigating
action for corrosion control and to prevent spills. The estimated mass of DU from SRS proposed
for disposal at Clive is 24,500 Mg (megagrams, or metric tons), assuming disposal of all 36,000
drums.
This material was characterized by SRS for uranium isotopes, fission products, and transuranics,
as well as some metals and organic compounds (pesticides, herbicides, semi-volatile and volatile
organic compounds) as recorded in the Waste Profile Record (EnergySolutions, 2009b). No
organic compounds were detected, though low levels (0 to 2 mg/kg) of lead, arsenic, cadmium,
chromium, selenium, silver, zinc and copper were found. These low levels of metal make up less
than 5 parts per million (ppm) mass of the DU waste. Based on the physical properties
description in the Waste Profile Record, the DU is stoichiometrically 83.22% uranium (100%
UO3) with over 99% 238U. Beals et al. (2002) provide additional information on trace
radionuclides in the SRS DU waste.
6.2 Depleted Uranium Oxide from the Gaseous Diffusion Plants
Three large GDPs were constructed to produce enriched uranium. The first diffusion cascades
were built in Oak Ridge, Tennessee, at what was the K-25 Site, but is now known as the East
Tennessee Technology Park (ETTP). Two others of similar design were constructed in Paducah,
Kentucky (PGDP), and Portsmouth, Ohio (PORTS) (DOE 2004a and 2004b). The cascades at
the K-25 Site ceased operations in 1985, the Portsmouth plant ceased in 2001, the Paducah GDP
continues to operate. The two more recent GDPs are host to a large inventory of stored DUF6,
including the ETTP material that was moved to Portsmouth.
The DOE is currently managing approximately 60,000 cylinders at both PGDP and PORTS
(DOE 2004a, 2004b). For many years, interest has been expressed in converting the DUF6 in
these cylinders to an oxide form to support their long-term disposal. In May, 1995 an
independent DOE oversight board recommended a study to determine a suitable chemical form
for long-term storage of DU. Two Environmental Impact Statements (EIS) were prepared as part
of the plan, one for Paducah, DOE/EIS-0359, (DOE 2004a) and one for Portsmouth, EIS-0360,
DOE 2004b). These EISs describe the background and alternatives for DUF6 conversion. With
the completions of the EISs, “deconversion” plants were built at both the PORTS and PGDP
locations. In 2002, DOE awarded a contract to design, construct, and operate two DUF6
deconversion facilities at these locations. As of this writing, both plants have been built and have
begun test processing DUF6 into oxide form.
Of the DUF6 cylinders that will be reused for disposal of the DU oxide, a fraction are
contaminated with fission and activation products from introduction of reactor returns into the
diffusion cascades. The contamination is similar in nature to that found in the SRS DU, and is
modeled as such until more information is gained from the generation of DU oxide at
Portsmouth and Paducah. Since the contaminated cylinders are a low priority for conversion, this
information is unlikely to be available for several years.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 31
6.3 Depleted Uranium Already Disposed at the Clive Facility
The DU PA Model does not account for DU that is already disposed at the Clive site, some of
which is from the same SRS DU population (Fussell and McWhorter, 2002).
6.4 Modeled Radionuclides
A full list of radionuclides has been established for the CSM and the contaminant transport
modeling effort:
fission products: 90Sr-, 99Tc-, 129I-, 137Cs-

progeny of uranium and transuranics: 210Pb, 222Rn, 226,228Ra, 227Ac, 228,229,230,232Th 231Pa

uranium isotopes: 232,233,234,235,236,238U

transuranic radionuclides:
237Np-237, 239,240,241,242Pu, 241Am

This radionuclide species list is based upon process knowledge, radionuclides analyzed for
(though not necessarily detected) in the DU waste material, and decay products with half-lives
over five years. A diagram showing each decay species is shown in the Radionuclide Transport
section (Section 9.0). The decay chains are informative as they provide an understanding of how
each species derived from a parent radionuclide. Many more short-lived progeny are accounted
for in dose assessment calculations. Note that in several instances where the inventory has been
set to zero, these species may be daughters of a known parent with inventory of a potential future
inventory species.
6.5 Chemical Characteristics of DU Wastes
Both forms of uranium oxide have some limited solubility in water, thus hydrologic transport is
expected to occur to some extent. The solubilities of the two waste forms are dependent upon the
geochemistry and their own inherent solubility. Other specific waste forms will be modeled as
information becomes available if needed. This transport will start with release from the
containment (e.g., drums, cylinders), followed by leaching of the radionuclides from the DU
waste which is primarily a function of solubility. The solubility of the radionuclide species,
including uranium, will depend upon two main geochemical processes: dissolution/precipitation
and adsorption/desorption. These processes are largely controlled by the redox condition, pH,
carbonate chemistry, and ionic strength of the local environments. The parameters used to model
the transport of the uranium oxides and associated radionuclides are described in Section 9.0.
Retarded transport will be modeled using a solid/water partition (or distribution) coefficient (Kd)
for each radionuclide species. The values (represented as statistical distributions) used for each
radionuclide will depend upon the expected geochemical conditions within the various wastes
and natural media.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 32
The release of radon-222 (222Rn) from its 226Ra parent in the 238U/234U decay chain is also
described in Section 9.0. The transport of radon in the saturated zone and in the unsaturated zone
from the waste to the ground surface is included in the Clive DU PA Model. Radon transport is
controlled by the emanation factor, diffusion, advection, and partitioning parameters that will be
incorporated into the transport modeling.
7.0 Modeling of the Natural Environment
The natural environment consists of those materials that surround the engineered facility, and
make up its environs. This includes the lacustrine sediments of the Great Salt Lake Desert
underlying the site, the groundwater within those sediments, the air above, and the biota living
on and near the ground surface. Each of these environments is introduced below, along with their
conceptual models for the PA.
7.1 Current Conditions
The basic conceptual model of the present day site is that the facility is located on a desert flat,
with a biotic community established on the ground surface, and with unsaturated and saturated
zones of groundwater below. This scenario is assumed to apply for the 10,000-yr duration of the
quantitative model for this base case.
In general, natural processes in the environs will tend to make the site and its engineered features
more like the natural environment. Wind and water will modify the cover, and biota will
populate it. Throughout this evolving and mixing system, radionuclides that have been disposed
within the facility will tend to migrate out to the natural system. A fundamental function of the
Clive DU PA Model is to estimate the rate and extent of that migration.
7.1.1 Groundwater Flow and Transport
Groundwater is considered in two parts: unsaturated zone (UZ) and the saturated zone (SZ). The
UZ, often called the vadose zone, extends from the ground surface down to the water table, and
is characterized by having both water and air in the porous spaces in the sediment. The SZ lies
below the water table, and extends deep into the earth’s crust. For the purposes of modeling,
however, contaminants are assumed to penetrate only so far into the saturated sediments, which
include natural horizontal barriers confining the vertical flow, as discussed in Section 3.3.1.
7.1.1.1 The Unsaturated Zone
The engineered features of the landfill, including cover, waste, and liner, are all in the UZ, at
least within the 10,000-yr duration of the quantitative model. Engineered barriers are used at the
Clive site to control the flow of water into the waste. A stylized drawing of the Federal Cell and
its relationship to the 11e.(2) cell is shown in Figure 8.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 33
The general aspect of the Federal Cell is that of a hipped cover, with relatively steeper sloping
sides nearer the edges. The upper part of the embankment, known as the top slope, has a
moderate slope, while the side slope is markedly steeper (20% as opposed to 2.4%). These two
distinct areas, shown in different colors in Figure 8, are modeled separately in the Clive DU PA
Model. Each is represented in the Model as a separate one-dimensional column, with a total area
equivalent to the Federal Cell footprint. In the current Clive DU PA Model, there is no waste
Figure 8. Section and Plan views of the Federal Cell, with top slope shown in blue and side
slope in green. The brown dotted line in the West-East Cross section represents
below-grade (below the line) and above-grade (above the line) regions of the
embankment.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 34
located below the side slope portion of the model. The embankment is also constructed such that
a portion of it lies below-grade. A detailed description of embankment dimensions and a
discussion of representation of the Federal Cell in the Model are provided in the Embankment
Modeling white paper accompanying the Clive DU PA Model.
Disposal involves placing waste on a prepared clay liner that is approximately 2.5 m (8 ft) below
the ground surface. For the Federal Cell design, the depth of the waste below the top slope is a
maximum of 14.5 m (47.5 ft). A cover system is constructed above the waste. The objective of
the cover system is to limit contact of water with the waste. The cover is sloped to promote
runoff and designed to limit water flow by increasing evapotranspiration (ET). The arrangement
of the layers used for the ET cover design is shown in Figure 9. Beginning at the top of the cover
the layers above the waste used for the ET cover design are:
• Surface Layer: This layer is composed of native vegetated Unit 4 material with 15
percent gravel mixture on the top slope and 50 percent gravel mixture for the side slope.
This layer is 15.2 cm (6 in) thick. The functions of this layer are to control runoff,
minimize erosion, and maximize water loss from ET. This layer of silty clay provides
storage for water accumulating from precipitation events, enhances losses due to
evaporation, and provides a rooting zone for plants that will further decrease the water
available for downward movement.
• Evaporative Zone Layer: This layer is composed of Unit 4 material. The thickness of this
layer is 30.5 cm (12 in). The purpose of this layer to provide additional storage for
precipitation and additional depth for plant rooting zone to maximize ET.
• Frost Protection Layer: This material ranges in size from 40 cm (16 in) diameter to clay
size particles. This layer is 45.7 cm (18 in) thick. The purpose of this layer is to protect
layers below from freeze/thaw cycles, wetting/drying cycles, and inhibit plant, animal, or
human intrusion.
• Upper Radon Barrier: This layer consists of 30.5 cm (12 in) of compacted clay with a low
hydraulic conductivity. This layer has the lowest conductivity of any layer in the cover
system. This is a barrier layer that reduces the downward movement of water to the waste
and the upward movement of gas out of the disposal cell.
• Lower Radon Barrier: This layer consists of 30.5 cm (12 in) of compacted clay with a
low hydraulic conductivity. This is a barrier layer placed directly above the waste that
reduces the downward movement of water.

The part of the UZ that extends from the bottom of the landfill liner to the water table consists of
naturally-occurring lake sediments from the ancestral Lake Bonneville. The texture class, and
average thickness for the hydrostratigraphic units underlying the Clive site are shown in Figure
10. The characteristics of the units are described in Section 3.3.1.
The natural UZ below the facility will be modeled as a column of discrete elements, called Cell
Pathway elements in the GoldSim modeling framework. Each of these is connected in series to
model the one-dimensional advective flow path to the water table. Diffusion in the water phase
may also play a role in the transport of waterborne contaminants in the UZ, since the advective
flux is expected to be small. The concentration gradients in the UZ are also expected to be
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 35
predominantly vertical, so diffusion will also occur in the vertical direction, oriented with the
column of cells.
Diffusion in the air phase within the UZ below the facility will not be modeled, since the only
diffusive species would be radon, which is of greater concern at the ground surface. Upward
radon diffusion to the ground surface will be dominated by radon parents in the waste zone, and
is modeled within the engineered cover.
7.1.1.2 The Saturated Zone
Contaminant transport in the water phase in the SZ is fed by contaminants entering the water
table beneath the disposal facility as recharge. The rate of recharge is the same as the Darcy flux
Figure 9. Evapotranspiration (ET) cover system.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 36
(the rate of volume flow of water per unit area) through the overlying UZ, and is expected to be
small enough that vertical transport within the SZ would be small. Most SZ waterborne
contaminant transport will be in the horizontal direction, following the local pressure gradients
which are reflected in water table elevations in an unconfined aquifer such as this. A point of
compliance in the groundwater has been established to be 27 m (90 ft) from the toe of the waste
embankment, so transport is modeled to that point.
Saturated zone groundwater transport generally involves the processes of advection-dispersion
and diffusion. Mean pore water velocity in the saturated zone is assumed to be determined by the
Darcy flux and the porosity of the sediment. A range of values will allow the sensitivity analysis
(SA) to determine if this is a sensitive parameter in the determination of concentrations at the
compliance well and resultant potential doses. Modeling of fate and transport for the saturated
zone pathway will include advection, linear sorption, mechanical dispersion, and molecular
diffusion.
The modeling of the SZ is similar to the modeling of the UZ, except that the “column” of
GoldSim Cell Pathway elements is arranged horizontally. This will be modeled as a row of cells
between the region below the disposal unit and the compliance well. These cells are saturated
with water that flows along the row, in order to represent the aquifer.
7.1.2 Surface Water
The Clive facility is sited in an area of extremely low topographic relief, and surface water
features such as stream channels are rare. The ancestral lake bed is quite flat, so there is little in
Figure 10. Hydrostratigraphic profile showing ET cover, waste zone, and
hydrostratigraphy below the Federal Cell.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 37
the way of land surface gradients which might drive surface water flow. Most if not all meteoric
water that lands on the ground is assumed to be returned to the atmosphere by
evapotranspiration, and essentially none is abstracted by runoff.
The embankment cells on the waste disposal site have significant relief, and surface water runoff
should be expected from these structures. The runoff and associated sediment transport will be
local, and is likely to remain in the vicinity of the site. The principal effect of surface water flow
is expected to be contribution to the formation of gullies, as discussed in Section 10.4.
7.1.3 Air and Atmosphere
Contaminant transport in the air phase takes on two distinct forms: diffusion in the interstitial air
in porous media below ground, and dispersion by the atmosphere above. Diffusion in interstitial
air of porous media is a means by which contaminants reach the atmosphere at the ground
surface. Dispersion of contaminants in the atmosphere can occur through direct diffusion of
gaseous contaminants into ambient air, and through resuspension and movement of wind borne
contaminated soil particles.
Airborne transport is a secondary contaminant transport mechanism at the Clive Facility. As
containment features such as the cover become contaminated from the result of natural processes
(e.g. radon diffusion, burrow excavation, plant senescence), radionuclides will migrate to surface
soils, serving as a source for atmospheric transport. As these contaminants accumulate on the
ground surface, either in a gaseous form (e.g. radon) or attached to solid particles, they undergo
resuspension or volatilization into the atmosphere, leading to airborne transport. Airborne
contaminants will be carried into ambient air by the wind and either inhaled directly by receptor
populations or deposited onto exposure media such as vegetation or soils in the vicinity.
7.1.3.1 Diffusion Through Air in Porous Media
Contaminants released from the waste (or generated by decay of parents in any location) may be
transported via the air pathway by migration of gaseous species through soil pore space. Over
time, cracks, fissures, animal burrows, and plant roots can also provide preferential pathways
that reduce the effectiveness of the engineered barrier. These effects are difficult to quantify and
are not modeled for diffusion in air in the Clive DU PA Model. Efforts at quantification could be
included as part of future cover modeling. Factors that influence the diffusion of contaminants
through porous media include the volatility of the chemical species, its molecular weight,
physical properties of the soil matrix (e.g., porosity, grain size distribution, and moisture content,
which determine phasic tortuosity – that is, tortuosity in either the air or water phase), and
temperature gradients. Diffusion in porous media and along preferential pathways is also driven
by concentration gradients and mediated by effective diffusion coefficients through the tortuous
diffusion path.
Diffusion rates are determined from the defined values for effective diffusivities, diffusive areas,
diffusive lengths, and the calculated concentration gradients between adjacent cells, which varies
as time progresses. Diffusion can take place in both air and water. In coordination with diffusion
is radioactive decay and ingrowth, advection of water, partitioning of contaminants between
water and air and between water and soils, and biotic processes. All these differential equations
and transfer functions are solved at each time step by the PA model.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 38
An important consideration related to the disposal of DU is the production of radon. Since 222Rn
is a descendent of 238U and 234U, through 230Th and 226Ra, it will be generated wherever 226Ra
occurs. As the radium, or any parent in the chain, migrates into the cover, either by diffusion in
the water phase or translocation by biotic processes (see Section 7.1.4), it provides a source for 222Rn in more locations beyond the disposed waste. Furthermore, not all of the radon that is
produced enters the environment for transport. Some of it is retained within the solid material
that held its parent, and decays to polonium-218 (218Po) without moving. This phenomenon is
called radon emanation, and is discussed in the radionuclide transport section (Section 9.0).
Radon that does enter the environment partitions between air and water. Soil moisture therefore
retards the migration of radon as it migrates through the soil, making it less available to diffusion
in air under wetter soil conditions.
7.1.3.2 Atmospheric Dispersion
Atmospheric dispersion of airborne gaseous and particulate contaminants found in surface soils
is expected. To the extent that contaminated subsurface soils are exposed or exhumed and plant
litter is deposited on the surface, they become surface soils and as such will also be subject to
atmospheric dispersion.
Atmospheric dispersion of contaminants is regulated by several factors. Contaminant chemistry,
contaminant mobility, soil texture, effects of vegetation on the atmospheric boundary layer,
topography, and meteorological conditions (predominant wind direction and speed, precipitation,
temperature, and humidity) may influence dispersion of airborne contaminants as well as soil
erosion and contaminant resuspension rates.
The Clive facility is sited in an exposed area, with little around it to protect from the winds.
Wind dispersion is a likely mechanism of airborne transport. Contaminants deposited over or
adsorbed onto soil may migrate from this area source as airborne particulates. Depending on the
particle- size distribution and associated settling rates, these particulates may be deposited
downwind or remain suspended, resulting in contamination of surface soils and/or exposure of
regional receptors through inhalation, immersion, or external irradiation pathways.
Ancestral lake sediments prevalent at the Clive facility are fine-grained, and are susceptible to
resuspension and entrainment in the wind, and to subsequent atmospheric dispersion. This
resuspension of naturally-occurring sediments, however, is moderated by local plant growth,
which creates a boundary layer of lower-velocity air at the ground surface, and by the formation
of desert crust, making the cemented particles of sediment in effect much larger.
The embankments on the site have significant relief in relation to the surrounding environment.
Eventually, enough wind-driven (eolian) sediment may be deposited that the disposal site will
approach the surrounding natural lake bed in appearance and behavior. Although these eolian
deposits will consist of uncontaminated material at first, they may become contaminated by the
process of radon diffusion upward from the waste (with radon progeny left behind in the soils)
and through the biotic processes discussed in the following section. Once radon gas and
resuspended particles have entered the atmosphere directly above the cells, they can be dispersed
over a wide area by the wind. Given these possible transport pathways, atmospheric dispersion of
gases (e.g. radon and other volatile constituents) and of fine particles of sediment must be taken
into consideration in the model.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 39
Entrainment of contaminants into the atmosphere will contribute to the air inhalation exposure
pathways for receptors that are present on the site itself. As particulates eroding from the
embankment are deposited on surrounding land, this surrounding area may become a secondary
source of radionuclide exposure. Atmospheric dispersion calculations in the Clive DU PA Model
will support estimation of gas and particulate air concentrations above the embankment, and off-
site particulate deposition rates that can be used to estimate radionuclide soil concentrations in
the area surrounding the embankment.
7.1.4 Biota
Biota of primary importance for movement of buried waste and subsurface soils are burrowing
animals (both vertebrates and invertebrates, which provide constant mixing of the soil column)
and plants, which can move buried wastes through root-uptake and translocation of contaminants
to various parts of the plant.
7.1.4.1 Native Plants
Plants represent an important potential pathway for waste transport by way of rooting and
conditioning of soil aggregates and particulates, nutrient exchange with soil surfaces, transport of
nutrients from soil through plant tissues, deposition of organic materials and non-nutritive waste
products at or near the soil surface, and physical mixing of soils through the addition of organic
materials to soil due to root collapse and surface deposition. In particular, nutrient exchanges
between the subsurface and surface also create the potential for the exchange of non-nutritive
chemicals, such as with anthropogenic wastes.
Plant induced transport of contaminants is assumed to occur primarily through absorption of
contaminants into the roots, after which the contaminants are redistributed throughout all the
tissues of the plant, both aboveground and belowground. Upon senescence, the above-ground
plant parts are incorporated into surface soils, and the roots are incorporated into soils at their
respective depths. This process is illustrated in Figure 11, which shows the conceptual model for
plant uptake, redistribution, and senescence. Note that relatively clean surface soils become more
contaminated over time as subsurface contaminants are translocated to aboveground portions of
the plant, and ultimately to the surface soil as the plant senesces.
The degree to which plants can move contaminants from the subsurface, and the rate at which
that transport can occur are dependent upon a number of factors such as plant rooting depth, total
above ground plant biomass, total below ground plant biomass, relative abundance of plants, and
density of plants roots by depth.
Plant rooting depths are influenced by a number of physical and physiological factors, but the
ultimate limiting factor is the availability of water. Roots of desert plants generally do not exceed
the depth to which water from precipitation infiltrates on a consistent basis. The maximum
rooting depth of any desert plant is physically limited to the maximum depth from which the
plant can obtain water. Of the plants that dominate the Clive site, black greasewood (Sarcobatus
vermiculatus) is likely the most deeply rooted. Black greasewood is phreatophytic, meaning that
it can utilize shallow groundwater, or derive supplementary water from the overlying capillary
fringe and deplete soil water potential to values less than 4.0 megapascals (MPa). However, in
areas where precipitation does not infiltrate to groundwater, black greasewood will not form
taproots and will maintain a more shallowly rooted growth form. Excavations of several
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 40
greasewood plants at the Clive site by SWCA (2011) found roots that did not exceed one meter
in depth. Several investigators have documented the types and metrics of plant species in
bajadas, desert valleys, and saline mounds (Robinson 1958, Meinzer 1927, Groenveld 1990,
Blank et al. 1998, Hansen and Ostler 2003, Rundel and Nobel 1991, and Holmgren and Brewster
1972).
The plant species currently inhabiting the Clive site are generally halophytic, meaning that they
are adapted to saline environments. Dominant plant species in the saline environments around
Clive include the halophytic shrubs black greasewood, shadscale, and the non-native forb
halogeton. Soil chemistry of the alkali flat environment is a limiting factor that regulates the
local plant community assemblages. It could be anticipated that the soil chemistry of constructed
mounds such as the disposal cells may change over time as precipitation leaches salts from the
mound soils, which are elevated above the surrounding terrain and decoupled from the saline
groundwater. This change in soil chemistry could allow for the establishment of less salt-tolerant
species, such as sage (Artemesia spp) and rabbit brush (Chrysothamnus spp.), which are common
in less saline cool desert habitats. Current closure plans include a revegetated surface layer
composed of Unit 4 material with 15% gravel on the top slope and 50% gravel on the side slope.
Figure 11. Conceptual model for plant induced contaminant transport
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 41
This layer is underlain by an evaporative zone layer composed of Unit 4 material. The soils and
plant species in these layers will be similar to surrounding undisturbed areas.
7.1.5 Native Animals
Only limited biotic surveys of the Clive site have been conducted, so site-specific information
about the utilization of the site by specific animal species is likewise limited. However, based on
the limited Clive studies and more comprehensive studies at other sites, burrowing animals,
including invertebrates and mammals, are of importance when evaluating the mixing of soils and
the potential for transporting buried wastes from the subsurface to the surface.
Ants
Ants fill a broad ecological niche as predators, scavengers, trophobionts and granivores, but it is
their role as burrowers that is of main concern for evaluating transport of buried materials from
the subsurface to the surface. Ants burrow for a variety of reasons but mostly for the
procurement of shelter, the rearing of young and the storage of foodstuffs.
In arid areas of the Great Basin and southwestern U.S., harvester ants of the genera
Pogonomyrmex and Messor are widespread, form large colonies, and often construct elaborate
nests. A preliminary survey of the Clive site and surrounding areas in October 2010 found that
the Western harvester ant (Pogonomrymex occidentalis) is by far the most common ant at the
site, with nest densities ranging from two nests per hectare in mixed sage/juniper community, to
33 nests per hectare in areas with abundant grasses (SWCA 2011). Only a single other ant
species (Lasius sp.) was identified at the Clive site during the preliminary surveys, and it
occurred only in the mixed grass vegetative association.
Several investigations have focused on ants as a taxonomic group of importance for the potential
to move buried waste at locations such as the Idaho National Laboratory (INL), and the Hanford
Site in southeastern Washington (Blom 1990, Fitzner et al. 1979, Gano et al., 1985). These
studies indicate that large colonies of Pogonomyrmex spp. may nest to depths of 3 to 4 meters
(10 to 13 ft) and may colonize areas with great densities of nests (over 100 per hectare), thus
potentially excavating large volumes of contaminated soil to the ground surface.
How and where ant nests are constructed plays a role in quantifying the amount and rate of soil
movement and the mixing of the soil column. Factors relating to the physical construction of the
nests including the size, shape, and depth of the nest are necessary in order to quantify
excavation volumes. Factors limiting the abundance and distribution of ant nests such as the
abundance and distribution of plant species, and intra- and inter-species competition also can
affect excavated soil volumes. Therefore, the amount and rate of soil movement is based on a
variety of factors, including nest area, nest depth, rate of new nest additions, colony density and
colony lifespan.
Due to its dominance at the Clive site, the initial model will be parameterized using available
data for Pogonomyrmex occidentalis. The geometry and structure of ant nests appears to be more
of a species-specific trait that does not exhibit significant flexibility in variable environments
(MacKay 1981). The mound’s height, width, distribution of particles, color, and exposure
significantly impact the colony for predatory defense and environmental regulation, but for any
given species, these mound traits are the same from place to place (MacKay 1981). Therefore,
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 42
there is defensibility for using data collected elsewhere for the same species in order to
parameterize the potential for ant-mediated transport in the Clive model. Site specific data
collected by SWCA (2011) on mound surface dimensions will be used to predict overall nest
volume and depth, and habitat-specific information of ant nest density will be used to help
predict the overall rate of soil movement on a per hectare basis for each habitat type. Additional
site specific data may be needed dependent on the outcome of the initial model.
A number of authors contend that it is reasonable to expect that over the 15- to 30-year life of
some Pogonomyrmex colonies, the entire soil column of the nest is turned over at least once
(Mandel and Sorenson 1981). For important and long-lived Pogonomyrmex ants in the desert
southwestern U.S., Lavigne (1969) and MacKay (1981) have investigated nest structure rather
extensively, and conclude that the net effect of soil movement within an ant colony’s lifetime is a
general homogenization of soils throughout the nest profile. In general, it is likely that this
homogenization occurs more rapidly in the top third of the nest, as this is where most of the
colony’s burrowing takes place, but over the life of the nest, burrowing at the greatest depths of
the nest can be extensive (Lavigne 1969, MacKay 1981). It is expected that ants will colonize the
cover, however, ants will not directly transport the larger particles from layers with gravel.
Therefore, mixing of the gravel particles downward will be minimal, though transport of soil and
clay particles from lower layers of the cover upward is expected.
Mammals
Burrowing mammals such as gophers, pocket gophers, moles, voles, squirrels, mice, rats,
kangaroo rats, and their predators have a profound influence on soil mixing. Burrowing
mammals rework the entire near-surface of soil over most of the North American continent on a
persistent basis, but at varying rates (Nevo 1999). Each of these mammalian species contributes
to soil turnover to a varying degree, depending upon their burrowing habits, geographic location,
and prevailing climate and soil conditions (Laundré and Reynolds 1993). Mammalian biotic
transport of soils also includes the deposition of fecal material in soils, the intermixing of
vegetation, and the significant aeration of upper layers. All of these actions dramatically affect
soil fertility, permeability by air and water, and increase soils’ susceptibility to invasion by
microorganisms (e.g., bacteria, fungi, nematodes, microarthropods).
Some mammals such as pocket gophers (Thomomys spp.), ground squirrels (Spermophilus spp.,
Sciuridae spp., and others), and kangaroo rats are considered obligately fossorial, i.e., they spend
most of their time underground, including foraging underground. Other organisms, however, will
utilize burrows only for shelter (temporary or permanent) and reproduction. These include hares
(Lepus spp.), rabbits (Sylvilagus spp.), sagebrush voles (Lagurus curtatus), pocket mice
(Perognathus spp.), kangaroo mice (Microdipodops spp.), foxes (Vulpes spp., Urocyon
cinereoargenteus), and coyotes (Canis latrans).
Biotic transport of soils by mammals at waste burial sites includes the potential direct movement
of waste from the subsurface to the surface, as well as secondary transport, such as food chain
transfer, transport by way of fecal deposition, and carcass degradation (Arthur and Markham,
1982; Smallwood et al., 1998). Intrusion into buried wastes and active physical transport occur
when animals penetrate protective barriers and cause vertical or horizontal redistribution of
waste material (Hakonson et al., 1982; Arthur and Markham, 1982). As animals excavate
burrows they either relocate buried material to the surface, or relocate soils from depth into
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 43
below-ground chambers lateral to the point of entry, as is common with pocket gophers or other
obligately fossorial mammals (Smallwood et al., 1998).
Because mammal burrows facilitate natural ventilation and aeration of the soils, burrowing
activity may also enhance the potential for contaminant release in gaseous form by allowing
increased communication between the atmosphere and buried waste. Mammal burrows also may
provide preferential pathways for water infiltration, as some studies have shown that recharge
quantities and depth of recharge were positively correlated with burrow density, and also found
that ground squirrels can increase precipitation infiltration into the soils by as much as 34% as a
consequence of burrowing activity (Laundré, 1993). Other studies, however have shown little
effect of animal burrowing on water balance (Section 3.4.2.1). The effect of animal burrowing on
subsurface moisture content was investigated in a field experiment at the Hanford Site by
Landeen (1994). Over the course of five testing periods, three during the summer and two during
the winter soil moisture measurements showed no influence of burrowing activities on long-term
water storage.
Preliminary investigations of mammals at Clive have focused on surveying the different habitat
associations for mammal burrows, quantification of the amount of soil excavated by burrowing
mammals, and trapping to determine dominant small mammal species in each vegetative
association. Results suggest that burrowing mammals are relatively scarce on the alkali flat
habitats (greasewood, shadscale), becoming more abundant in the less saline soils associated
with mixed grass and juniper-sage habitats. Deer mice were the most abundant mammals trapped
in all habitat types, with lesser numbers of kangaroo rats (two species), and grasshopper mice
also found in the traps.
7.2 Deep Time Conditions
The deep time frame over which the analysis is concerned is defined by the period of time
beyond 10,000 years until radioactivity from the DU parents and its progeny is at its peak. This
occurs when the progeny, identified in Section 9.1.2, are in secular equilibrium with the parent.
For decay of a refined 238U parent (the longest-lived uranium isotope), progeny reach secular
equilibrium at about 2.1 million years (My). With its exceedingly long half-life of over 4 billion
years, the parent 238U decays only by about one half-life before the end of the solar system, and
the peak achieved at 2.1 My wanes only slightly in that time. The analysis devoted to deep time
scenarios is sufficiently representative of this entire duration when considered out to only 2.1 My
in the future, as changes in radioactivity are minor after that time.
The model developed to evaluate the deep time performance of the Clive facility focuses on
concentrations in various media, and does not attempt to translate these concentrations into
human dose metrics. This approach is used because of the overwhelming uncertainty associated
with evaluating human receptor scenarios that far into the future. This uncertainty is associated
both with projecting human behavior and environmental conditions.
A scenario is considered that involves the return of large lakes in the Bonneville Basin over the
next few million years, since secular equilibrium is reached at about 2.1 My. Following that, the
radioactivity of the DU will persist effectively forever.
Understanding the phases associated with the change from current climatic conditions to future
climatic conditions can help construct a qualitative picture of how the Clive facility will respond
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 44
to those changes. The following section provides a brief overview of how major environmental
changes in the past are directly coupled to major shifts in climatic regimes. This section also
provides context with respect to how these past changes may occur in the future and their
implications on the stability of the Clive facility.
7.2.1 Background on Long-term Controls on Site Conditions
7.2.1.1 Climate processes
Large-scale climatic fluctuations over the last 2.58 My (the beginning of the Quaternary Period)
have been studied extensively in order to understand the mechanism underlying those changes
(Hays et al., 1976, Berger, 1988, Paillard, 2001, Berger and Loutre, 2002). These large-scale
fluctuations in climate have resulted in glacial and interglacial cycles which have waxed and
waned throughout the Quaternary Period. The causes of the onset of the Northern Hemisphere
glaciation about 3 million years ago (3 Ma) remain uncertain, but several studies suggest that the
closing of the Isthmus of Panama caused a marked reorganization of ocean circulation patterns
that resulted in continental glaciation (Haug and Tiedemann, 1998, Driscoll and Haug, 1998).
Changes in the periodicity of glacial cycles have been linked to variations in Earth’s orbit around
the Sun. These variations were described by Milankovitch and are based on changes that occur
due to:
• the eccentricity of Earth’s orbit – about every 100,000 years (100 ky),
• the obliquity of Earth’s axis– about 41 ky, and,
• the precession of the equinoxes (or solstices) – about 21 ky.
For the first two million years of the Pleistocene (the first major Epoch of the Quaternary
Period), Northern Hemispheric glacial cycles occurred about every 41 ky, while the last million
years have indicated larger glacial cycles occurring about once every 100 ky, with strong
cyclicity in solar radiation every ~23 ky (Berger and Loutre, 2002; Paillard, 2006). The results of
Hays et al. (1976), who analyzed changes in the isotopic δ18O composition of deep-sea sediment
cores, suggest that major climatic changes have followed both the variations in obliquity and
precession through their impact on planetary insolation. Variations in δ18O reflect changes in
oceanic isotopic composition caused by the waxing and waning of Northern Hemispheric ice
sheets, and are thus used as a proxy for the climatic record. However, the shift from shorter to
longer cycles is one of the greatest uncertainties associated with utilizing the Milankovitch
orbital theory to explain the onset of glacial cycles alone (Paillard, 2006).
Various studies have highlighted the importance of atmospheric carbon dioxide (CO2) variations
in the dynamics of glaciations across the Northern Hemisphere in addition to the insolation due
to orbital forcing (Clark et al., 2009; Paillard, 2006). Direct measurement of past CO2 trapped in
the Vostok and EPICA Dome C ice cores from Antarctica show that atmospheric CO2
concentrations decreased during glacial periods due to greater storage in the deep ocean, thereby
causing cooler temperatures from a reduction of the atmosphere’s greenhouse effect (EPICA,
2004). Warmer temperatures resulting from elevated concentrations of CO2 that are released
from the ocean on the other hand contribute to further warming and could support hypotheses of
rapid wasting at the end of glacial events (Hays et al., 1976). Berger and Loutre (2002)
conducted simulations forced with insolation and CO2 variations over the next 100 ky and report
that the current interglacial period could last another 50 ky with the next glacial maximum
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 45
occurring about 100 ky from now. They also report, however, that future increases in
atmospheric CO2 from anthropogenic activity along with small insolation variations could result
in a transition between the Quaternary and the next geologic period due to the potential wasting
of the Greenland and west Antarctic Ice Sheets.
There is a strong likelihood that there will be major climatic shifts within the next million years,
and strong evidence that the 100 ky cycle has impacted the Bonneville basin in the form of large
lake recurrence (Oviatt, 1997; Asmerom et al., 2010). Thus, due to the destructive potential of a
lake to the waste embankment, the deep time scenarios of most interest are the return of large
lakes in the Bonneville Basin.
7.2.1.2 Large Lake Cycle Events
The Clive facility is located in the Bonneville Basin where Lake Bonneville, the largest of the
late Pleistocene pluvial lakes, last existed between 30-10 ka. Pluvial lakes are lakes that show
evidence of expansion due to pluvial episodes (wetter climatic phases) as well as contraction due
to what is assumed to reflect interpluvial episodes (warmer, dryer climatic phases). Various FEPs
fall within the lake cycle scenario which include wave action, sedimentation, and site inundation.
At its maximum (between ~15-16 ka BP), Lake Bonneville is estimated to have covered an area
of 51,300 km2 (~19,800 sq mi) and was over 370 m (1200 ft) deep (Lowe and Walker, 1997).
Following the Bonneville flood at ~18 to 18.5 ka (Miller et al 2013), during which the lake level
dropped by ~114 m (~375 ft) as it spilled over and eroded a spill point, the lake level continued
to decline leaving behind modern-day Great Salt Lake. Geomorphological evidence is present
that shows the variability in the levels of the last major lake cycle as indicated by the exposed
shoreline features in areas of the Bonneville basin.
Oviatt et al. (1999) examined sediments from the Burmester core and suggested that a total of
four deep-lake cycles occurred during the past 780 ky. They found that the four lake cycles
correlated with marine oxygen isotope stages 2 (Bonneville lake cycle: ~24-12 ka), 6 (Little
Valley lake cycle: ~186-128 ka), 12 (Pokes Point lake cycle: ~478-423 ka), and 16 (Lava Creek
lake cycle: ~659-620 ka), which suggests that large lake formation in the Bonneville basin
occurred only during the most extensive Northern Hemisphere glaciations. In addition to these
large lake cycles, a smaller cycle known as the Cutler Dam cycle occurred between ~80-40 ka
(Link et al., 1999). Each major lake cycle and its corresponding estimated maximum shoreline
elevations are listed in Table 1. As a point of reference, the Clive facility is located at an
elevation of 1302 m (4275 ft) amsl, and the airport at Salt Lake City, SLC, is at 1288 m (4227
ft).
During the large pluvial lake events, large amounts of calcium carbonate were precipitated as
tufas, marls, shells (of mollusks), and ostracodes (Hart et al., 2004). Brimhall and Merritt (1981)
reviewed previous studies that analyzed sediment cores of Utah Lake, a freshwater remnant of
Lake Bonneville that formed ~10 ka. It is suggested that up to 8.5 m (28 ft) of sediment has
accumulated since the beginning of Utah Lake, implying an average sedimentation rate of
~0.00085 m/y (nearly 1 mm/y) over 10 ky. Within the Bonneville basin as a whole it is
suggested that the major lake cycles resulted in substantial accumulations of sediment based on
the depth of the cores analyzed (e.g., 110-meter core that corresponds to the past 780 ky, or four
major lake cycles for an average sedimentation rate of 0.00014 m/yr including non-lake phases;
Oviatt et al., 1999).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 46
Table 1. Known lake cycles in the Bonneville Basin
Lake Cycle Approximate
Age*
Maximum Elevation Lake level control
Great Salt Lake
(current level)
present 1284 m (4212 ft) in 1873 climate; human intervention
Gilbert 11–10 ka 1295 m (4250 ft) climate
Provo 14.5–13.5 ka 1445 m (4740 ft) threshold at Zenda near Red
Rock Pass, Idaho
Bonneville ~28–12 ka
(14C)
1552 m (5090 ft) threshold at Zenda near Red
Rock Pass, Idaho
Stansbury 23–20 ka 1372 m (4500 ft) climate
Cutler Dam ~80–40 ka < 1380 m (< 4525 ft)
Little Valley ~128–186 ka 1490 m (4887 ft)
Pokes Point 417–478 ka 1428 m (4684 ft)
Lava Creek ~620–659 ka 1420 m (4658 ft)
*Approximate ages derived from Currey, et al. (1984) Link et al. (1999) and Oviatt et al. (1999).
Elevations are not corrected for isostatic variations

There is a lack of peer-reviewed literature that considers the direct effects of future climate
change on major lake formation in the Bonneville basin. However, if the current geologic era
continues, the probability of another major lake cycle occurring in the Bonneville basin within
the next 100 ky in conjunction with variation in Earth's orbital characteristics is high, considering
the correspondence between past global temperature fluctuations and past known lake events.
Assuming that past conditions will apply in the future, variations in orbital characteristics are
very likely lead to another major ice age and thus alter long-term climatic patterns in the
Bonneville region making it suitable for lake formation. Each 100 ky glacial cycle is different,
depending on orbital forcing, but it is clear from the historical record that the current period is
inter-glacial, and colder conditions are likely in the future. Unless the current geologic period
ends in response to anthropogenic forcing effects on atmospheric CO2 concentrations (Berger
and Loutre, 2002), it is expected that the Clive facility will be subjected to lake formation in the
future. Return of a large lake is considered unlikely without climatic change.
7.2.1.3 Isostatic Rebound
Isostasy refers to the gravitational equilibrium between Earth’s lithosphere (the rocky outer
crust) and asthenosphere (the semiliquid layer below the crust) such that the lithosphere “floats”
at an elevation that depends on its local thickness and density. When large amounts of sediment,
water, (in the case of Lake Bonneville) or ice occur over a particular region over time, the weight
of the new mass may cause the crust below to sink. Hetzel and Hampel (2005) examined the
effects of the removal of Lake Bonneville on isostatic rebound of the lithosphere. They found
that the removal of Lake Bonneville triggered an increase in fault slip rates in the Wasatch region
resulting in clustering of earthquakes during the early Holocene. Former islands present during
the Lake Bonneville cycle also indicate that isostatic rebound occurred after the regression of the
lake. This is evidenced by the paleo-shorelines on the islands which are located tens of meters
above the paleo-shorelines along the lake periphery (Hetzel and Hampel, 2005).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 47
Although it is difficult to predict potential impacts from future seismic events, it is expected that
if isostatic rebound effects were to occur, the effects of future seismic events would be mitigated
by the site’s burial by lacustrine sediment.
7.2.1.4 Volcanism
The principal effects of volcanism on the Clive site are indirect. Hart et al. (1997) suggest that
lava flows near Grace, Idaho during the Pleistocene diverted the upper Bear River between the
Snake River drainage to the Bonneville Basin through the formation of lava dams. Link et al.
(1999) report that the permanent addition of the Bear River discharge to Lake Bonneville likely
occurred around 50 ka (±10 ka), and in conjunction with cooler and wetter conditions during this
time, it is thought to be responsible for the lake reaching its highest level (i.e., the Bonneville
shoreline). Although the lava dams resulted in the alteration of the path of the Bear River, at
certain times during the Pleistocene the upper Bear River was diverted into the Snake River
which deprived the Bonneville basin of significant discharge. Future changes in the regional
hydrology in response to any future lava flows or regional volcanic activity could result in
similar implications for future pluvial lake events (i.e., increase or decrease in discharge to the
basin).
7.2.1.5 Ecological Changes
Changes in biotic assemblages have been shown to occur in the past (Davis and Moutoux, 1998)
and will likely occur in the future in response to shifts in climatic regimes. Temperature and
precipitation have a profound effect on plant community assemblages, as does soil chemistry.
Areas where salt pans remain in place will remain largely unvegetated regardless of changes in
temperature and precipitation. Valley areas around the margins of salt pans will remain restricted
to halophytic plants until salinity levels drop. Because Clive is somewhat centrally located within
the Great Basin cold desert biome, vegetation assemblage changes associated with climate
change will occur more slowly than in areas closer to biome transition zones. As the climate
changes, vegetation changes will occur on steppes and slopes, but soil chemistry will remain the
constraining factor on the valley floors.
Pollen studies from sediment cores in the Great Salt Lake show that the vegetation of the
Bonneville Basin and surrounding area has been desert for approximately the last 5 My (Davis
and Moutoux, 1998). The pollen studies indicate that Sarcobatus, Artemisia, and various
Chenopodaceae (the family that includes the various saltbush species) have dominated during
interglacial periods, with montane conifers (Picea, Abies, and Pseudotsuga) increasing during
glacial periods. For the purposes of this CSM, it is assumed that climatic shifts could occur
resulting in any one of four different conditions: cooler-wetter, cooler-drier, warmer-wetter,
warmer-drier. The direction of the climatic shift will affect both the vegetative and faunal
assemblages occupying the site. Figure 12 illustrates a general biome diagram based on
temperature and precipitation, as well as the approximate location of the Clive site within this
temperature-precipitation gradient.
Cooler, wetter conditions will likely result in transition first to Artemisia sage communities, then
to Pinyon-Juniper woodland characterized by the presence of Juniperus osteosperma and Pinus
monophylla, and finally to montane spruce/fir woodlands as seen during past glacial periods.
These woodlands are not likely to ever occupy the valley floor unless profound changes in soil
chemistry occur. All of these changes occur over geologic time, and prediction of the occurrence
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 48
of specific species represents a great uncertainty. Cooler, drier conditions will likely maintain
similar plant communities as are currently present, unless temperatures get cold enough to
support taiga/tundra conditions.
Warmer, drier conditions will result in plant assemblages similar to those that occur in the
Mojave desert, where valley floors are dominated by creosote bush (Larrea tridentata), white
bursage (Ambrosia dumosa), and pale desert-thorn (Lycium pallidum). Warmer, wetter
conditions could lead to establishment of grasslands, and eventually temperate forest, as existed
more than 10 Ma when the pollen record shows that elm (Ulmus), hickory (Carya), yew (Taxus),
and hemlock (Tsuga) were common in the area (Davis and Moutoux, 1998). Again,
establishment of these vegetative complexes on the valley floor would require a major shift in
soil structure and chemistry.
7.2.1.6 Human Intervention
Various scenarios can be constructed that look at each of these impacts on the Clive facility in
the ultra long-term future. One major difference between the past 3 My and the present is the
existence of well-developed human civilization, technology, and greater ability to adapt to
changing conditions. If in the future another ice age were to occur similar to those that have
occurred during the Pleistocene, disposal cell design could help mitigate the effects of future
events that could jeopardize the stability of the engineered facility at Clive.
Figure 12. Whittaker Biome Diagram
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 49
In the event of another major lake cycle, human intervention is likely to be employed in
surrounding areas (e.g., Salt Lake City) and could result in modifying engineered features like
those that were installed to alleviate the effects of flooding in the early 1980s, when a pumping
system was built to divert flood waters into the west desert (see www.water.utah.gov/
Construction/GSL/GSLpage.htm). In fact, the Utah Division of Water Resources proposed
various options to handle flooding events of Great Salt Lake due to natural variations in
precipitation (see www.water.utah.gov/Construction/GSL/GSLflood.htm). Some of the options
that were proposed included the exportation of flood flows from the Great Salt Lake drainage
basin to the Bear River and Sevier River drainages, consumption of water via evapotranspiration
through the development of new agricultural lands, and creating a dike around the lake to protect
major facilities and resources.
While it is difficult to predict the level of human intervention in response to these events, it
should be taken into consideration for all future scenarios considered for the performance
assessment of Clive facility.
7.2.2 Long-Term Scenarios
The primary scenario of concern in the deep time scenario is the return of a lake to the
Bonneville Basin that reaches the elevation of the Clive facility. There is historical evidence of
large lakes covering the Clive site with more than 100 meters of water, so large lakes will be
modeled as recurring in the future. There is weaker historical record of intermediate-sized lakes,
lakes that are relatively shallow at the Clive elevation. The lack of historical record for
intermediate lakes is not necessarily surprising, since the combined effects of wave erosion and
lake sedimentation during transgressive and regressive lake cycles are likely to bury and or
obscure evidence of intermediate lakes. However, there is evidence of two relatively recent
intermediate lakes – Cutler Dam and Gilbert, as well as stratigraphy in sediment cores that
suggest many lakes rising and falling at the Clive elevation (Oviatt, 1997), which might be
associated with either intermediate lakes or fluctuations in large lake transgression and
regression. The expected consequence of the formation of a lake in the Bonneville Basin is the
destruction of the waste embankment due to wave energy, resulting in physical dispersal of the
site material. Waste entrained in the sediment can partially dissolve into the lake, and
contaminant complexes will precipitate from the lake water back into the sediment. This process
is depicted in the conceptual model shown in Figure 13.
The deep time model is thus constructed to represent the following components:
• Continuation of natural processes in the waste embankment. After 10,000 years, natural
processes such as eolian erosion and/or deposition of silts and sand, groundwater
transport, and biotic uptake will continue to be modeled as long as the embankment is
intact.
• Returns of large and intermediate lakes to the Clive site. Large lakes will be treated as
occurring regularly with the 100,000-year orbital cycle, while intermediate lakes will
occur according to a random process between large lake cycles, with greater probability
of occurrence further in time from the end of the inter-glacial period (i.e., as the
temperature decreases and precipitation increases).
• Site destruction. When the first lake returns at or above the elevation of Clive, the waste
embankment will be treated as destroyed. The result is dispersal of above-grade waste
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 50
into the sediments near the site, along with dissolution into the lake water. Once the
waste embankment is destroyed, the evolution of the waste embankment is no longer
modeled.
• Sedimentation and mixing. The presence of a lake implies sedimentation at the site. As
the waste is dispersed, it will be mixed with the embankment materials and sediment.
Waste material that dissolves into the water column will be assumed to precipitate out of
the water column back into the sediment at the site as the lake recedes. Subsequent lakes
are likely to at least partially bury the waste beneath subsequent sediment. However,
since the deep time model is intended to be qualitative, a conservative choice is made to
model all sediments containing waste as mixing with sediments of subsequent lakes.
• Activity levels. The results tracked in the deep time model are the radioactive
concentrations in lake water and in sediment.
Figure 13. Scenarios for the long-term fate of the Clive facility
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 51
8.0 Modeling of Engineered Features
The engineered features of the disposal facility are the waste form itself (including containment),
and the liner and cover, which surround the wastes. Other than these, the natural environment is
relied upon to moderate the migration of contaminants. These engineered features are expected
to degrade with time, gradually assuming a form more like the natural surroundings. The model
will attempt to capture the performance of the engineered features, including the essential
processes contributing to their degradation, as described in this section.
8.1 Waste Form and Containment
The waste forms are discussed in detail in Section 6.0, but a brief discussion is included here for
completeness as an engineered feature. The waste form, for the purposes of this discussion,
includes the matrix that contains radionuclides, and any drums, boxes, or other materials that
contain that matrix. Generally, wastes are not designed with their long-term resistance to
degradation in mind, but rather for the convenience of the generator and shipper. Also, waste
form and containment on waste profiles or shipping manifests are sufficient for disposal
purposes, but not necessarily for PA purposes.
Low-level radioactive waste matrices are in general quite heterogeneous, including bulk soils,
debris from decontamination and decommissioning activities, protective equipment, tools,
laboratory wastes, chemical residues, resins and filters, and such, but in the case of DU waste,
the form is unusually uniform. Leachability and solubility can be modeled for well-documented
DU oxide waste forms. Details on the chemical characteristics of DU waste are given in Section
6.5.
Steel barrels and boxes, “burrito-wrap” fabrics, cardboard, or even bulk uncontainerized
materials are common in LLW. Most of these offer little in the way of long-term containment,
especially after compaction to reduce void spaces, which often crushes or otherwise
compromises containment. Container integrity is not typically given credit in LLW PA models.
In the case of DU, the containers, which consist of steel 200-L (55-gal) drums or the various
specialized designs of steel UF6 cylinders, are not expected to provide much in the way of long-
term containment. Pitting, rusting, and other forms of corrosion have already been documented
for the cylinders, and a number of steel drums have had to be repackaged. This degradation has
taken place in the last few decades, so it would be unreasonable to assume that containers would
remain intact for any appreciable length of time in the environment of the embankment cell. The
model, therefore, will not take credit for containment (refer to Section 6 of the FEPs Analysis
white paper accompanying the Clive DU PA Model). All wastes are assumed to have the
characteristics of local Unit 3 sandy soil.
8.2 Liners
The Clive facility’s embankment cells are constructed similarly to those designed for landfills
under the Resource Conservation and Recovery Act (RCRA), using a variety of natural and
engineered materials. Liners are constructed on the floor of the facility, and the waste is placed
on top of them. Caps are constructed over the waste, and are designed to shed water.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 52
Previous PA modeling at the Clive site, which addressed a performance period of hundreds of
years, included modeling of the installed performance of the cover and liner, degradation of the
cover, and bio-intrusion scenarios (Whetstone, 2000). Liner degradation allows for increased
contaminant transport from the waste layers to the UZ below the facility, and subsequently to the
SZ through recharge. The performance of the liner is not expected to degrade significantly. The
principal role of the liner in the contaminant transport model is to regulate flow from the waste to
the underlying UZ, so all that matters, in the end, is the rate at which water may penetrate it, plus
any chemical retardation involved as it flows through.
8.3 Cover
Engineered covers can be subject to degradation processes such as biointrusion, freeze-thaw, and
erosion. These processes are discussed in the following paragraphs.
Current closure plans include a revegetated Surface Layer composed of Unit 4 material with 15%
gravel on the top slope and 50% gravel on the side slope. This layer is underlain by an
Evaporative Zone Layer composed of Unit 4 material. The soils and plant species in these layers
will be similar to surrounding undisturbed areas. The cover will differ from the surrounding
areas in slope. Potential changes in cover performance due to time dependent evolution of the
cover layers after closure are driven by the following processes:
• Site-specific field studies (SWCA, 2013) indicate that although the plants and animals in
the vicinity of the Clive site are found at low densities and are small in size, the local
animals and plants described in Sections 7.1.4 and 7.1.5 are expected to penetrate the
upper soil layers of the ET cover. These studies concluded that the amount of soil
disturbance would be insignificant in comparison with the total soil volume of the cover.
Quantitative estimates of soil displacement are contained in SWCA (2013).

• The frost protection layer consists of bank run materials with sizes ranging from cobbles
to clays. This material contains large- and medium-sized cobble that cannot be moved by
small animals, pore sizes small enough to prevent passage by small animals, and a fine
soil component that fills the pores of the coarse component providing a further deterrent
to burrowing (SWCA, 2013).

• Observations made during a biological survey at the Clive facility (SWCA, 2011)
indicate that plant roots often form on top of clay layers that are a meter or more below
the top surface, such as the upper radon barrier. Some of these roots may penetrate the
radon barriers, based on observations of plant roots in clay layers in boring logs, although
the recent biological survey did not dig through clay layers to confirm this. It is possible
that ants may also penetrate the clay layers by following root holes or possible cracks in
the clay layers. On balance, the biological survey evidence suggests that bioturbation and
homogenization of the radon barriers will probably occur very slowly relative to the
10,000-year time frame for the PA.

• Sheet erosion is a uniform process over the area of the cover, and depends largely on its
slope. In the central area of the embankment, where slopes are gradual, sheet erosion
would be slower than on the steeper side slopes of the cell. As soil moves downslope,
however, it is expected that the volume would be replenished by deposition of clean loess
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 53
from the surrounding environs. In the end, the soil volumes do not change, though there
would be a slow movement of soils downslope, along with the contaminants they could
potentially contain. Sheet erosion is not included in this model since the top slope of the
cover is gradual (about 2%), and since the overall effect of sheet erosion is likely to be
considerably less than the effect of gully erosion. The revegetation plan proposed by
EnergySolutions (SWCA, 2013) includes steps to promote the regrowth of the biological
soil crusts found on undisturbed areas in the vicinity of the site. An established biological
crust will provide long-term reduction of sediment transport by sheet erosion.

• Gully erosion has the potential to move substantial quantities of both cover materials
and waste. Once a “nick” is started somewhere on the surface of the cover, by an
animal burrow or off-highway vehicle (OHV) track, for example, the feedback
processes inherent in gully formation will cause erosion upward to the top of the
slope, and downward to the surrounding grade. SWCA (2013) notes that there is
minimal evidence of soil erosion at the Clive site or in the vicinity. SWCA cites
observations of small berms of soil created by eolian accumulation of soils that show
no evidence of water erosion over long periods of time.

• Freeze/thaw cycles will also tend to degrade performance of the cover. This process is
anticipated in the design, however, which includes a frost protection layer to
accommodate it.

• Subsidence of the wastes could also contribute to decreased performance of the cover
(Smith et al., 1997). Differential subsidence would be expected to cause vertical shearing
of the cover layers, creating enhanced transport pathways, and the formation of
depressions which could capture water, increasing local infiltration. However, it is
expected that any depression would fill in rather quickly by windblown sediments.
Subsidence is not expected to be an important process atIn the Clive facility, since the
waste is aggressively compacted in order to prevent this occurrence (EnergySolutions,
2009c).

At the Clive site these processes are expected to be slowed significantly by the effects of eolian
deposition. As discussed in Section 3.3.3 above, examination of eolian deposits in the upper part
of the stratigraphic section at the Clive site show slow processes of pedogenesis and continuing
suppression and burial of developing soils by a relatively low rate of deposition of eolian silt.
These conditions will persist at the Clive site as long as the lake levels remain below the site
elevation. The expectation is that eolian deposits will drape and slightly stabilize closure covers
until future lakes return to the Clive site.

9.0 Radionuclide Transport
This section describes the aspects of modeling that involve radionuclides. The modeling of the
natural environment, including groundwater flow, atmospheric dispersion, and other processes
that are not specific to radionuclides, is discussed in Section 7.0. Following the determination of
the list of radionuclide species under consideration, this section discusses the mechanisms
governing their fate and transport in the environment.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 54
9.1 Modeled Radionuclides
Unlike general LLW, DU waste contains only a select number of radionuclides. These are mostly
uranium isotopes (by mass), the most common of which is 238U. The non-uranium radionuclides
are either fission products or actinides.
9.1.1 Reported Inventory
Based on laboratory analysis of the contents of DU waste (including all radionuclides in the
containers), the species in the disposed inventory include (Beals, et al. 2002, EnergySolutions
2009b, Johnson 2010):
uranium isotopes 233U, 234U, 235U, 236U, 238U
other actinides (and radium) 226Ra, 241Am, 237Np, 238Pu, 239Pu,
240Pu, 241Pu fission products 90Sr, 99Tc, 129I, 137Cs
9.1.2 Radioactive Decay and In-growth
Radioactive decay and in-growth are fundamental physical processes. There are several types of
radiological transformations, including alpha, beta, gamma, electron capture, spontaneous
fission, etc. While these processes are not specifically detailed in this subsection, they are
accounted for in terms of their dose effects on humans, and their change in elemental (chemical)
nature. As they experience decay and in-growth, the radionuclides in the reported inventory will
change and these progeny must also be included in the modeling.
Simplified decay chains for the actinides are shown in Figure 14. Decay and in-growth continue
until a stable nuclide is reached. In the case of the actinides, the stable nuclide is always bismuth
or lead.
9.1.3 Short-lived Radionuclides
Not all of the members of a decay chain are modeled in the fate and transport calculations. Given
the long duration of the analysis, and the short half-life of many of the radionuclides, it is
impractical to model their transport, as they could not travel any appreciable distance before
decaying to the next nuclide of the decay chain. Attempting to include short-lived radionuclides
in the fate and transport model adds unnecessary complexity to the model. Therefore,
radionuclides with half-lives less than five years are excluded from the fate and transport
analysis, with one exception: 222Rn. Radon is a special case, since as a noble gas it has unique
transport characteristics, even though it has a half-life of under four days. It diffuses in both air
and water, partitioning between the two, and can migrate significant distances.
It must be noted that while the short-lived radionuclides are not included in the fate and transport
calculations, they are included in the dose assessment. It is often short-lived nuclides that
contribute most to dose.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 55
Figure 14. Principal decay chains for the four actinide series. Radionuclides in black are
included in the fate and transport model, and those in green are considered only in
the dose model.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 56
9.1.4 Radionuclides with Small Branching Fractions
Similar to the short-lived radionuclides, there are radionuclides that have exceedingly small
branching fractions, in addition to being short-lived. These are included in neither the fate and
transport calculations, nor the dose calculations, as their omission is invariably inconsequential
and promotes computational efficiency. In addition, most of these small branching fraction
radionuclides have no dose conversion factors available.
The detailed sections of the actinide decay chains that contain these radionuclides, showing all
the short-lived and small-branching-fraction radionuclides, are provided in Figure 15.
List of Radionuclides Species for Fate and Transport
The complete list of radionuclides accounted for in the fate and transport model follows,
Figure 15. Detailed decay chains for actinides. Radionuclides in black are included in the
fate and transport model, those in green are considered only in the dose model, and
those in gray are not modeled.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 57
organized into decay chains:
241Pu → 241Am → 237Np → 233U → 229Th
242Pu → 238U → 234U → 230Th → 226Ra → 222Rn → 210Pb
238Pu → 234U → (joins the above chain)
239Pu → 235U → 231Pa → 227Ac
236U → 232Th → 228Ra → 228Th
232U → 228Th → (joins the above chain)
Several radionuclides are not part of the actinide series:
137Cs → 137mBa
129I
90Sr → 90Y
99Tc
The decay of the last species listed in the chain is also included in the fate and transport
modeling.
9.2 Source Release
The disposed DU waste is assumed to be uncontainerized, since standard operations at the site
include significant compaction of disposed waste.
9.2.1 Containment Degradation
As discussed in Section 8.1, no credit will be given to the ability of steel containers to inhibit
release of wastes.
9.2.2 Matrix Release
In the absence of detailed information regarding the chemical and physical form of the uranium
oxides, release of radionuclides from the waste matrix will be assumed to be instantaneous. That
is, release into infiltrating water that migrates through the waste will be controlled only by the
geochemical constraints of the waste/water partition coefficient (Kd) and solubility (see Section
9.3). If information can be provided for a basis of a measured release from the waste matrix, that
can also be incorporated into the model.
9.2.3 Radon Emanation
A special consideration for DU is the production and release of radon, especially 222Rn. As 222Rn
is produced by alpha decay from 226Ra, the recoil from the ejection of the alpha particle may be
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 58
of sufficient energy to expel the 222Rn atom from the waste matrix. If it is not so energetic, the
radon atom will stay in the matrix, and will in a matter of days decay to 218Po and then to other
progeny, and will not be available for environmental transport as radon.
The fraction of decaying radium atoms that result in a radon atom being expelled into a transport
medium (water or air) is called the radon emanation factor or the escape/production ratio (E/P)
ratio, and has a value between 0 and 1. If the E/P ratio for a given waste form is 0, no radon ever
escapes the matrix; if it is 1, all radon escapes. A dense solid matrix such as metal, crystal, or
glass could have a low E/P ratio, and a fine powder or surface contamination would have a
relatively high value.
9.3 Waterborne Radionuclide Transport
Water enters the modeled system as infiltration from meteoric waters (precipitation) at the
embankment cell surface, and as groundwater below the ground surface. The approach to
modeling different groundwater zones is discussed in Section 7.1.1. This section focuses on the
transport of radionuclides within that water system. For many contaminants waterborne transport
is influenced by geochemical processes.
While the radiogeochemistry of contaminant transport is in reality exceedingly complex, it is
typically simplified for the purposes of PA. A full geochemical model considers the mineralogy
of neighboring geological materials and the full geochemical makeup of water, on a highly
refined scale. It considers the speciation and complexation of ions, which is especially involved
for those cations with multiple valence states, such as uranium and plutonium. It considers the
formation and transport of colloids, and the fine-scale adsorption of chemical species onto
sediment particles and fracture coatings. For the PA modeling, the geochemistry of contaminant
transport in groundwater is approached at the macro scale, and a few key concepts are assumed
to account for all the small-scale variation. A simple equilibrium sorption model using soil/water
partition coefficients or Kds is used to model the partitioning process. While simplified, the Kd
approach is conservatively representative of the solid-water partitioning process and is in
common usage in PA models. The Kd model assumes that a given constituent dissolved in the
water (e.g. uranium) has some propensity to sorb to the solid phase of a porous medium, while
maintaining some presence dissolved in the aqueous phase as well. The definition of the
solid/water distribution coefficient, with units of mL/g (or sometimes m3/kg) is:
𝐾!=𝑚𝑎𝑠𝑠  𝑜𝑓  𝑐𝑜𝑛𝑠𝑡𝑖𝑡𝑢𝑒𝑛𝑡  𝑠𝑜𝑟𝑏𝑒𝑑  𝑜𝑛  𝑎  𝑢𝑛𝑖𝑡  𝑚𝑎𝑠𝑠  𝑜𝑓  𝑠𝑜𝑙𝑖𝑑  (𝑔/𝑔)
𝑚𝑎𝑠𝑠  𝑜𝑓  𝑐𝑜𝑛𝑠𝑡𝑖𝑡𝑢𝑒𝑛𝑡  𝑤𝑖𝑡ℎ𝑖𝑛  𝑎  𝑢𝑛𝑖𝑡  𝑣𝑜𝑙𝑢𝑚𝑒  𝑜𝑓  𝑤𝑎𝑡𝑒𝑟  (𝑔/𝑚𝐿) (1)
The sorption is assumed to be instantaneously reversible and independent of concentration. That
is, no dynamics are accounted for, and the ratio is always simply linear—a constituent’s
concentration in water is always the same ratio with respect to its sorbed concentration onto the
solid, and it takes no time for the change between solid or liquid phases to occur. This is the
linear isotherm assumption, and is commonly employed.
Aqueous solubility, however, places limits on the amount of a constituent that can be dissolved
in the water phase. Each chemical species (in this case, each chemical element, including all
isotopes) has a limit as to how much of that chemical can exist in the water phase. Solubility is
expressed in moles per unit volume of water (typically mol/L), where one mole is Avogadro’s
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 59
number of atoms (or molecules). If, then, the solubility of uranium were 1 mol/L, one liter of
water could hold one mole of uranium, which could be a mix of 235U, 236U, 238U, or other
isotopes. Any attempt to add uranium to the water will result in the precipitation of uranium.
The Kd model expressed in Equation 1 is applied only when the solubility limit for a given
constituent is not in effect. This is a particularly important point to keep in mind when modeling
the leaching of a concentrated waste form, such as uranium oxides. At first, the leaching is likely
to be solubility-limited with respect to uranium, and the leachate will migrate away with uranium
at the solubility limit. Eventually, as enough uranium is removed from the source, the leachate
concentration will be limited only by Kd, and will be less and less concentrated until the source is
depleted. This occurs for all other elements as well, though the synergistic effect of various
similar chemicals (e.g. other heavy metals like plutonium and lead) is not modeled.
Note that partitioning and solubility are independent of isotopic variation, as the radiological
aspect of contaminants does not enter into their chemistry. That is, isotopes all behave
identically, chemically speaking. 234U, 235U and 238U are isotopes, and therefore compete together
for sorption sites, or for aqueous solubility. A model that considers 235U and 238U in separate
simulations cannot couple these effects, and may produce inaccurate results, especially in the
presence of solubility limitations. GoldSim recognizes the concept of isotopes, and accounts for
their interrelated chemical behavior.
9.4 Airborne transport
As discussed in the section on modeling the natural environment (Section 7.1.3), the two distinct
types of airborne transport include diffusion in the air-filled pore spaces of porous media, and
dispersion above the ground surface by wind. Radiological aspects of these processes are
discussed below.
9.4.1 Diffusion Through Porous Media
Diffusion within porous media, in either air or water, is driven by concentration gradients.
Diffusion is mediated by diffusion coefficients, and it follows tortuous paths through the specific
medium. Partitioning between air and water phases also occurs, which adds to the number of
simultaneous equations to be solved.
The principal radionuclides of interest in the modeling of DU waste are the isotopes of radon,
since radon, a noble gas, is the only radionuclide to be found in a gaseous form. The parents and
progeny of radon isotopes are of interest as well. Radon has several isotopes that occur in the
various actinide decay series, including 217Rn, -218, -219, -220, and -222 (see Figure 14 and
Figure 15). Radon isotopes with half-lives ranging from milliseconds to just under 1 minute
quickly undergo decay to polonium and therefore can travel no appreciable distance. Radon-222,
however, has a half-life of just under 4 days, and is able to migrate for some distance by
diffusion in interstitial air before it, too, decays to polonium. When regulations such as DOE’s
Radioactive Waste Management Order 435.1 address radon ground surface flux as a
performance objective, 222Rn is the isotope of concern.
Since 222Rn is a direct descendent of 238U and 234U, and hence 230Th and 226Ra, it will be
generated anywhere in the environment that 226Ra occurs. As the radium migrates into the
embankment cell cover, either by diffusion in the water phase or translocation by biotic
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 60
processes (see Section 9.5), it provides a source for 222Rn in more locations than just the disposed
waste. This form of translocation and transport is accounted for in the modeling.
A phenomenon unique to the production and release of radon is the E/P ratio, introduced in
Section 9.2.3 with respect to release from the waste form. If, however, the 226Ra parent is present
in other locations, such as cover materials or surface soils, radon will be in water or adsorbed
onto solids, rather than bound in some crystalline matrix. The E/P ratio in the environment is
assumed to be 1, and thereby all of the decay of 226Ra outside the waste form results in 222Rn that
is available for transport.
Radon partitions between air and water, per its Henry’s Law constant (KH). For this reason, wet
soils are much better at attenuating radon migration than dry soils. To mitigate the diffusion of
radon through the engineered cover, the layering within the cover design includes a substantial
layer of clay. Clay has a low permeability to air and to water, and also can maintain a high
moisture content, which retards the migration of radon as it partitions into soil (Ota et al., 2007).
The effectiveness of this clay radon barrier, however, depends on its resistance to degradation by
erosion and biotic processes. Cracks, fissures, animal burrows, and plant roots can all provide
fast diffusion pathways that reduce the effectiveness of the radon barrier.
Diffusion in the porous medium air phase, as well as the water phase, is implemented in the
Clive DU PA Model through diffusive flux links between all GoldSim Cell Pathway elements in
a column, from the atmosphere to the water table.
9.4.2 Atmospheric Dispersion
The basic modeling of atmospheric dispersion is covered in Section 7.1.3.2. The only effect of
radon and radionuclides attached to particles that is related to radioactive processes is that during
transport, as in other transport pathways, radionuclides undergo radioactive decay and in-growth.
For the purposes of this model, however, the assumption is made that atmospheric transport is
sufficiently fast relative to rates of decay that no decay need be accounted for during the
transport.
9.5 Biotically Induced Transport
Plants and fossorial (burrowing) animals have the potential to move radioactive material in
addition to the more commonly implemented waterborne and airborne transport pathways. The
full conceptual model of biota at the site is discussed in Section 7.1.4, and the relevance to
radionuclide transport is discussed here.
9.5.1 Transport via Plants
Plants obtain many nutrients and minerals from the soil, through root uptake. Some chemical
species are preferred over others, and this preference differs between plant species, as does the
effectiveness of uptake. This selective uptake is coupled with radioactive decay and in-growth.
Plants are conceived to selectively absorb chemical species from the soils, with roots exposed to
different soil layers and thus different suites of chemicals at various depths. The absorbed
radionuclides, then, are distributed evenly within the plant tissues, both above-ground and
below-ground.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 61
When the plant dies, the below-ground parts return radionuclides to whatever soil layer they are
in, and the above-ground plant parts all return their constituents to the top layer of soil.
9.5.2 Burrowing Animals
Burrowing animals include various mammals, reptiles, and insect species. They move bulk soil
from the depths where they construct burrows directly to the ground surface. Bulk soil includes
soil and any interstitial water and air, and all radionuclides contained in the volume that the
animals remove.
After a burrow is abandoned, it eventually collapses, moving bulk soils back down from the
surface, in accordance with the volume excavated. This preserves the mass balance of soil in the
soil column. The overall effect of this burrowing activity is a consistent churning of the soil
layers (bioturbation). This effect may be deep, with ant nests having been observed to penetrate
over 4 meters (~13 ft) below the ground surface at another western radioactive waste disposal
site (see Section 7.1.5).
10.0 Modeling Dose and Risk to Humans
Evaluation of radiation dose (with implied risk) to potential human receptors is a requirement of
the PA. The individual dose assessment addresses potential radiation dose to any member of the
public who may come in contact with radioactivity released from the disposal facility into the
general environment (10 CFR 61.41). Radiation dose limits for protection of the general
population are defined in 10 CFR 61.41. Design, operation, and closure of the land disposal
facility must also ensure protection of any individual inadvertently intruding into the disposal
site and occupying the site or contacting the waste at any time after loss of active institutional
control of the site (10 CFR 61.42). Because the definition of inadvertent intruders encompasses
exposure of individuals who engage in normal activities without knowing that they are receiving
radiation exposure (10 CFR 61.2), there is no practical distinction made in the dose assessment
between any MOP and inadvertent intruders with regard to modeling radiation dose for
protection of the general population.
Protection of inadvertent intruders from the consequences of disturbing disposed waste can
involve two principal controls: 1) institutional control over the site after operations by the site
owner to ensure that no such occupation or improper use of the site occurs, or 2) designating
which waste could present an unacceptable risk to an intruder, and disposing of this waste in a
manner that provides some form of intruder barrier that is intended to prevent contact with the
waste (10 CFR 61.7(3)).
The objective of modeling annual radiation dose to an individual in a radiological PA is to
provide estimates of potential doses to humans, in terms of an “average” member of the critical
group, from radioactive releases from a disposal facility after closure, as described in Section
3.3.7 of NUREG-1573, A Performance Assessment Methodology for Low-Level Radioactive
Waste Disposal Facilities (NRC, 2000). As described below, the critical groups in this PA are
defined as Ranchers, Sport OHVers, and Hunters. An “average” member of such a group may be
considered as either a statistical construct, or more subjectively as simply a hypothetical
individual whose behavioral and physiological attributes do not place them on either the lower of
higher extreme of the range of possible individual doses.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 62
NUREG-1573 describes two aspects of dose modeling: First, the mechanisms of radionuclide
transfer through the biosphere, to humans, needs to be identified and modeled. This is termed the
pathway analysis. Second, the dosimetry of the exposed individual must be modeled. This is
termed the individual dose assessment.
Pathway analysis, as defined in NUREG-1573, results in the determination of the total intake of
radionuclides by the average member of the critical group. The critical group is defined as the
group of individuals reasonably expected to receive the greatest dose from radioactive releases
from the disposal facility over time, given the circumstances under which the analysis would be
carried out. Modeling of radionuclide transport by plants and animals, and of human activities, is
captured within the scope of this pathway analysis. The dosimetry component of the dose
modeling refers to estimation of the effective dose equivalent from internal radiation dose
following radionuclide intake, and from external radiation dose.
In order to estimate collective doses for the purpose of determining whether disposal options
satisfy ALARA, a population needs to be assessed. A population is comprised of multiple
individuals, so individual doses need to be added over some period of time to estimate the
collective dose. The ‘answer’, at the end of the performance period (10,000 years post-closure, in
this case) might then be the individual annual doses added up over a period of 10,000 years.
Although there is no collective dose performance metric that currently exists, this analysis may
be useful in the context of comparing how one site or disposal option might perform compared to
another.
10.1 Period of Performance
No specific time frame is defined in 10 CFR 61 for the dose assessment. In the context of
inadvertent human intrusion, Section 61.42 states,
“Design, operation, and closure of the land disposal facility must ensure
protection of any individual inadvertently intruding into the disposal site
and occupying the site or contacting the waste at any time after active
institutional controls over the disposal site are removed.” (emphasis added.)

UAC Rule R313-25-9 is more specific, requiring a PA for DU to have a minimum compliance
period of 10,000 years, with additional simulations for a qualitative analysis for the period where
peak hypothetical dose occurs. The estimation of doses at such long time frames is uncertain, but
if total radioactivity is used for a proxy, accounting for decay and ingrowth from the disposed
DU, then a peak value would occur once the progeny of U-238 have reached secular equilibrium
in about 2.1 million years.
The scope of this PA is to model the disposal system performance to the time of peak
hypothetical radiological dose (or peak radioactivity, as a proxy), but to quantify dose only
within the regulatory time frame of 10,000 yr. This approach is consistent with UAC
R313-25-9(5)(a).
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 63
10.2 Site Characteristics and Assumptions
Key land use characteristics and assumptions for the Clive facility that pertain to the
development of receptor scenarios and dose modeling are summarized in the Site Description
(Section 3.0).
As addressed in the FEP Analysis white paper, the distinction between deliberate and inadvertent
intrusion for this PA is based on the motive underlying the activity. Intrusive activities not
related to a deliberate attempt to excavate materials underlying the protective cover will be
considered inadvertent. The performance objectives of 10 CFR 61.43 specifically address
protection of individuals from the consequences of inadvertent intrusion after active institutional
controls are removed. Because deliberate intrusion at the site is omitted from the performance
objectives, whereas inadvertent intrusion is specifically mentioned, modeling of dose resulting
from deliberate intrusion into the disposal site is not included in this PA. Therefore, radiation
doses due to intrusion based on motives such as archeology, sabotage, or waste retrieval for
constructive or malicious reasons, are not evaluated.
10.3 Receptor Scenarios
Potential activities of interest for this model are based on the predominant present day uses of the
general area as identified in the FEP analysis: ranching and recreation. Other scenarios that are
often considered for PAs, including agriculture and homesteading, are not applicable for the
Clive site for reasons described below. There are other populations that might be exposed at
locations remote from the disposal embankment, such as drivers along Interstate-80, a resident
caretaker at the Aragonite rest area off I-80, rail workers and riders, and workers at the Utah Test
and Training Range. Although these receptors are likely exposed for short amounts of time
and/or at lower concentrations compared to ranchers and recreationists, these off-site receptors
will also be evaluated in the PA model.
From a regulatory perspective, two categories of receptors require consideration. These are often
labeled “member of the public” (MOP) and “inadvertent human intruder” (IHI). Both categories
are described in related guidance: the MOP essentially as a receptor who resides at the boundary
of the facility, and the IHI as someone who directly contacts the waste (e.g., by well drilling, or
basement construction). There is no historical evidence of non-transient human activities in the
near vicinity of Clive, however, other than current activities and a temporary maintenance camp
at the nearby railroad over 50 years ago. Furthermore, while the area in which the site is located
is zoned for hazardous waste disposal by Tooele County, the lack of potable water makes the
surrounding area an unlikely location for other residential, commercial, or industrial
developments (Baird et al., 1990). Consequently, an IHI or MOP receptor as described in
regulatory guidance is extremely unlikely. Therefore, consideration will be given to ranching and
recreational scenarios to describe plausible human activities under current conditions. The
potential for these human activities to result in inadvertent human intrusion will also be
considered.
10.3.1 Ranching Scenario
The land surrounding the disposal facility is used for cattle and sheep grazing (NRC, 1993;
BLM, 2010). Leases are administered by the BLM, and are generally up to 6 months in length,
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 64
from autumn to spring. The ranching exposure scenario includes exposure to radionuclides that
have entered the available environment due to natural processes described in the transport model.
Receptors may be directly exposed while working upon or in the vicinity of the disposal unit.
Evaluation of potential radiation dose in this scenario is partially dependent upon assumptions
regarding the nature of plant community succession on the disposal unit over time. Because
ecological succession on the disposal unit over time could potentially result in grazing habitat
upon the disposal unit, a variety of potential future plant community assemblages are evaluated
in the PA model.
Inputs for developing exposure parameter values under the ranching scenario include
information on the characteristic activities of ranch hands and restrictions related to BLM leases
for ranching. Activities are expected to include herding, maintenance of fencing and other
infrastructure, and assistance in calving and weaning. The primary exposure pathways for the
ranching scenario include incidental ingestion of soil, inhalation, external irradiation, and
ingestion of beef from cattle grazing in contaminated areas. Exposure to respirable particulates
may occur from natural wind disturbance of surface soil as well as mechanical disturbance due to
rancher use of OHVs for transportation within the impacted area.
10.3.2 Recreational Scenario
The recreational exposure scenario encompasses receptors such as hunters and recreational OHV
riders on, or in the vicinity of, the disposal unit. Based upon discussions with the BLM and
reasonable judgment regarding anticipated land use, all recreational activities are likely to
involve some OHV use and may encompass sport OHV riding, hunting, target shooting of
inanimate objects, rock-hounding, wild-horse viewing, looking for ghost towns, and limited
camping. The recreational scenario evaluated in the PA model includes two distinct receptor
groups:
1. “Sport OHVers” who use their vehicles primarily for recreation and who may visit
the area as either a day trip or by camping overnight; and,

2. “Hunters” who, in addition to purely recreational visits, also visit the area for the
purpose of hunting game and who may also visit the area as either a day trip or by
camping overnight.

The desirability of recreational activities on or around the Clive facility is, like suitability for
ranching, dependent on assumptions regarding ecological succession at the Clive facility over
time. With the possible exception of OHV use and use of the cover as a vantage point for
hunting, recreational uses of Clive facility in an as-closed state of re-vegetated soil surfaces is
likely to be minimal. As soil develops on the cover and plant succession proceeds, the Clive
facility may become more attractive for activities such as camping and therefore support higher
exposure intensity.
The primary exposure pathways for the Sport OHV scenario modeled in the PA (described in
more detail below) include incidental ingestion of soil, inhalation, and external irradiation. The
Hunter scenario includes these same pathways and adds ingestion of game meat from animals
grazing in contaminated areas. Exposure to respirable particulates is evaluated for both natural
wind disturbance of surface soil as well as mechanical disturbance due to Sport OHV and Hunter
use of OHVs for transportation within the impacted area.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 65
10.3.3 Remote Off-Site Receptors
The ranching and recreation scenarios are characterized by potential exposure related to activities
both on the disposal site and in the adjoining area. Specific off-site points of potential exposure
also exist for other receptors based upon present-day conditions and infrastructure. Unlike
ranching and recreational receptors who may be exposed by a variety of pathways, these off-site
receptors are likely to be exposed solely to wind-dispersed contamination, for which inhalation
exposures are likely to predominate. The remote locations and receptors for which inhalation
exposures are evaluated in the PA model include:
• Travelers on Interstate-80, which passes 4 km to the north of the site;
• Travelers on the main east-west rail line, which passes 2 km to the north of the site;
• Workers at the Utah Test and Training Range (a military facility) to the south of the Clive
facility, who may occasionally drive on an access road immediately to the west of the
Clive facility fenceline;
• The resident caretaker at the east-bound Interstate-80 rest facility (Aragonite [Grassy
Mountain]) approximately 12 km to the northeast of the site, and,
• Recreational OHVers at the Knolls OHV area (BLM land that is specifically managed for
OHV recreation) 12 km to the west of the site.
10.4 Transport Pathways
Various considerations should be taken into account when analyzing the transport of
radionuclides through the biosphere to humans. Pathway identification is discussed in various
literature sources, such as Volume 1 of NUREG/CR-5453 (NRC, 1989) and NUREG-1200
(NRC, 1994), and NUREG-1573 (NRC, 2000). Components of the disposal system that can
affect transport include aspects of the source term and engineered barriers. Principal transport
media at many low-level waste disposal sites include groundwater, surface water, and air (NRC,
2000).
Pathways that will be evaluated for the protection of exposed individuals from releases of
radioactivity include those related to air (gas diffusion, air dispersion, and eolian erosion and
deposition of soil), soil (contaminant migration via upward flux from subsurface soil, deposition
of wind-borne material), groundwater (groundwater flow, geochemical effects, radon
emanation), surface water (water erosion leading to gullies, infiltration), plants (uptake of
contaminants in the waste, engineered cover, or soil), and animals (exhumation by burrowing).
Exposure media subsequently affected by transport processes include air, surface soil, plants,
game, and livestock. Figure 16 depicts the conceptual model for contaminant transport at the
Clive facility.
The transport processes figure depicts those processes relating contaminant release mechanisms
to environmental media that are the subject of the dose assessment. Many of these transport
pathways may not be complete or may not contribute sufficiently to exposures to warrant explicit
modeling.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 66
10.5 Exposure Pathways
Exposure pathways describe the activities and exposure routes between the environmental media
described in Section 7.0 and human receptors in the ranching and recreation exposure scenarios.
The primary exposure routes related to radionuclides in environmental media include ingestion,
inhalation, and external irradiation.
The ingestion exposure route may pertain to inadvertent ingestion of contaminated soil at either
on-site or off-site locations for the ranching and recreation scenarios. In addition to incidental
ingestion of soil, ingestion of meat containing radionuclides taken up from contaminated soil by
grazing animals is possible. Ingestion of meat from livestock grazing on or around the Clive
facility is characterized in the Ranching scenario. Ingestion of hunted meat from pronghorn
grazing in the region of the Clive facility is characterized for the Hunter receptor in the
recreational scenario.
The inhalation exposure route consists of the inhalation of either gas-phase radiological
contaminants or of respirable particulates originating from contaminated soil. The inhalation
exposure route is evaluated for both the ranching and recreational scenarios. Concentrations of
respirable particulates in air is assessed as a function of both wind erosion and mechanical
disturbance from the use of OHVs for all potential receptors.
External irradiation refers to the external exposure to a radiological source such as contaminated
surface soil (a two-dimensional source) or air (a three-dimensional source). External irradiation
from contaminated soil may occur when a receptor travels across the ground surface during
either ranching or recreational activities. Atmospheric immersion occurs when a receptor is
exposed to external irradiation via bodily immersion in contaminated air. Atmospheric
immersion is tied to the gaseous diffusion and air dispersion transport pathways, and is a viable
exposure route for both the ranching and recreational scenarios.
10.6 Risk Assessment Endpoints
Title 10 CFR 61.41 specifies assessment endpoints related to radiation dose. The specific metrics
described in §61.41 are organ-specific doses, and restrict the annual dose to an equivalent of 0.25
mSv (25 mrem) to the whole body, 0.75 mSv (75 mrem) to the thyroid, and 0.25 mSv (25 mrem)
to any other organ of any member of the public. However, as described below, the dose
assessment for the PA will employ a total effective dose equivalent (TEDE) for comparison with
the 0.25-mSv/yr threshold.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 67

Figure 16. Conceptual model for transport and exposure pathways at the Clive facility
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 68
As discussed in Section 3.3.7.1.2 of NUREG-1573 (NRC, 2000), the radiation dosimetry
underlying these dose metrics was based on a methodology published by the International
Commission on Radiation Protection (ICRP) in 1959. More recent dose assessment methodology
has been published as ICRP Publication 30 (ICRP, 1979) and ICRP Publication 56 (ICRP, 1989),
employing the TEDE approach. The TEDE uses weighting factors related to the radiosensitivity
of each target organ to arrive at an effective dose equivalent across all organs. The text of
Section 3.3.7.1.2 of NUREG-1573 (NRC, 2000) states
“As a matter of policy, the Commission considers 0.25 mSv/year (25
mrem/year) TEDE as the appropriate dose limit to compare with the range of
potential doses represented by the older limits... Applicants do not need to
consider organ doses individually because the low value of TEDE should
ensure that no organ dose will exceed 0.50 mSv/year (50 mrem/year).”

Radiation dose conversion factors (DCFs) applicable for calculating the TEDE are published by
DOE, EPA, and the ICRP. Section 3.3.7.3 of NUREG-1573 specifies DCFs published by EPA in
Federal Guidance Reports 11 (EPA, 1988) and 12 (EPA, 1993). EPA subsequently made use of
age-specific DCFs published in ICRP Publication 72 (ICRP, 1996) to compute radionuclide
cancer slope factors in Federal Guidance Report 13 (EPA, 1999). DCFs published in Federal
Guidance Report 13 are employed in this PA where possible.
DU waste can also be associated with toxicological risks that are independent of radioactive
properties. Unlike carcinogenic agents, EPA typically views toxicants with non-cancer effects as
having thresholds; i.e., levels below which effects would be unlikely. Reference doses (RfDs)
essentially amount to such thresholds, usually with several layers of ‘safety’ factors added. The
basic modeling process for evaluating uranium toxicity is very similar to that conducted for
radionuclides, except that kidney toxicity (as opposed to radiation dose) of DU is evaluated, and
the toxicity of DU does not change over time (as radioactive decay is not important in this
context).
11.0 Summary
This CSM describes the dynamic systems model that will be implemented for the Clive DU PA.
The CSM describes the regulatory environment that constrains the PA, and the technical
components that transport radionuclides associated with the DU waste to the accessible
environment. Transport starts with characterization of the waste, and continues with release of
radionuclides from the waste, migration through the engineered barriers system that initially
confines the waste, fate and transport through the local environment to the accessible
environment where human receptors might be exposed, including radioactive decay and
ingrowth through time and space. The dynamic systems model will be implemented using the
GoldSim systems modeling platform, which facilitates fully-coupled dynamic systems modeling
and is ideally suited to performing radiological performance assessments. The modeling will be
performed in a probabilistic manner so that uncertainties are fully captured and global sensitivity
analysis can be performed in order to identify the critical parameters. Consideration will be given
to spatio-temporal scaling and correlation in the modeling, so that input probability distributions
are properly specified. For some inputs to the model (e.g., radon diffusion, water content in the
unsaturated zone, erosion of the cover) process-level models may be developed and then
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 69
abstracted into the GoldSim systems-level model so that these model components are fully
integrated into the overall model.
The modeling effort will be split into two overlapping but distinct time frames of primary
interest. The regulatory compliance period for the first time frame is 10,000 years, requiring a
quantitative model that predicts radioactive dose to potential receptors. For this model, current
conditions of society and the environment will be projected into the future. Potential receptors of
interest for this model are based on present day use of the general area, as discussed in Section
10.3, including ranching, hunting, and recreation.
The second modeling time-frame will consider much longer term consequences of disposal of
DU waste at Clive, since peak radioactivity of the DU waste occurs beyond 2 million years into
the future. This model will overlap the short term assessment in that it will share many of the
same modeling components, such as waste inventory, source release, and fate and transport
through the local environment. However, this model will consider changes in the general
environment that might affect major changes in the environmental conditions of Clive. For
example, climate change is inevitable within this time frame, so its consequences will be
considered. Earth is in a glacial epoch, consisting of long glacial periods interspersed with
shorter interglacial periods. For example, the current interglacial period is one in which the
population of the human race has expanded to unprecedented levels. This current interglacial
period may continue for tens of thousands of years or longer because of the effects of
anthropogenic production of CO2 (Tzedakis et al 2012; Masson-Demotte et al. 2013). However,
based on the geological record, return of a glacial period is eventually inevitable driven by the
Milankovitch cycles. Based on geological evidence, the return of a glacial period will probably
result in the re-formation of a large lake covering most of northwestern Utah, so lake recurrence
is included in this model. Human exposure scenarios, however, will not be evaluated that far into
the future, because receptor scenarios cannot be defensibly developed and the consequences of
radioactive dose cannot be reasonably understood that far into the future. Many changes in
climate will have occurred within the next 2.1 My, the period over which it takes DU to reach
secular equilibrium. During such a long time frame there is likely to be massive disruption in
human society and changes in human evolution. Consequently, instead of attempting to model
dose to hypothetical human receptors that far into the future, the spatial distribution and
concentrations of radionuclides that might migrate from the disposal cells to the environment
will be modeled. The processes by which the radionuclides might move around, include the
formation of large lakes and the return to lower lake levels once the lake subsides again.
Consideration will also be given to the potential effects of wave action at the Clive facility as the
lake forms.
This two-tiered approach is consistent with the requirements of the Utah regulations to perform
fully quantitative modeling for 10,000 years, and qualitative modeling until peak activity.
Consequently, these two models will be used together to support the required regulatory analysis
of DU waste disposal at the Clive facility.

Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 70
12.0 References
Code of Federal Regulations, Title 10, Part 61 (10 CFR 61), Licensing Requirements for Land
Disposal of Radioactive Waste, Government Printing Office, 2007.
Code of Federal Regulations, Title 40, Part 191 (40 CFR 191), Environmental Radiation
Protection Standards for Management and Disposal of Spent Nuclear Fuel, High-Level
and Transuranic Radioactive Waste, Government Printing Office, 1993.
Adrian Brown (Adrian Brown Consultants), 1997a. Volume I, LARW Infiltration Modeling Input
Parameters and Results, Report 3101B.970515, 15 May 1997.
Arthur, W.J. and O.D. Markham, 1983. “Small Mammal Soil Burrowing as a Radionuclide
Transport Vector at a Radioactive Waste Disposal Area in Southeastern Idaho,” J. Environ.
Qual. 12: 117–122.
Asmerom, Y., Polyak, V. J., and S. J. Burns, 2010. “Variable winter moisture in the
southwestern United States linked to rapid glacial climate shifts,” Nature Geoscience 3:
114-117.
Baird, R. D., Bollenbacher, M. K., Murphy, E. S., Shuman, R., and R. B. Klein, 1990.
Evaluation of the Potential Public Health Impacts Associated with Radioactive Waste
Disposal at a Site Near Clive, Utah. Rogers and Associates Engineering Corporation, Salt
Lake City Utah.
Beals, D., S.P. LaMont, J.R. Cadieux, C.R. Shick, Jr., and G. Hall, 2002. Determination of Trace
Radionuclides in SRS Depleted Uranium, WSRC-TR-2002-00536, Westinghouse
Savannah River Company, Aiken, SC, 19 November 2002.
Benson, C.H., W.H. Albright, D.O. Fratta, J.M. Tinjum, E. Kucukkirca, S. H. Lee, J. Scalia, P.
D. Schlicht, and X. Wang. 2011. Engineered Covers for Waste Containment: Changes in
Engineering Properties & Implications for Long-Term Performance Assessment,
NUREG/CR-7028, Office of Research, U.S. Nuclear Regulatory Commission, Washington,
DC.
Berger, A., 1988. “Milankovitch Theory and Climate,” Reviews of Geophysics, 26(4): 624-657.
Berger, A. and M. F. Loutre, 2002. “An exceptionally long interglacial ahead?” Science,
297:1287-1288.
Bingham Environmental. 1991. Hydrogeologic report Envirocare Waste Disposal Facility South
Clive, Utah. October 9, 1991. Prepared for Envirocare of Utah. Salt Lake City, UT.
Bingham Environmental. 1994. Hydrogeologic report Mixed Waste Disposal Area Envirocare
Waste Disposal Facility South Clive, Utah. November 18, 1994. Prepared for Envirocare of
Utah. Salt Lake City, UT.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 71
Black, B.D., B.J. Solomon, and K.M. Harty, 1999, Geology and geologic hazards of Tooele
Valley and the West Desert Hazardous Industry Area, Tooele Count, Utah, Special Study
96, Utah Geological Survey, Salt Lake City, Utah.
Blank, R.R., Young, J.A., Trent, J.D., and D.E. Palmquist, 1998. “Natural History of a Saline
Mound,” Great Basin Naturalist, 58(3): 217-230.
BLM (U.S. Bureau of Land Management). 2010. Rangeland Administration System. U.S.
Blom, P.E., 1990. Potential Impacts of Radioactive Waste Disposal Situations by the Harvester
Ant, Pogonomyrmex salinus Olsen (Hymenoptera: Formicidae). A thesis for degree of
Masters of Science. Regents of the University of Idaho, Moscow, ID. 198 pp.
Bogen, K.T., "A Note on Compounded Conservatism", Risk Analysis, 14 (4) , pp. 379 – 381,
August 1994
Brimhall W. H. and L. B. Merritt, 1981. “The Geology of Utah Lake – Implications for Resource
management,” Great Basin Naturalist Memoirs, 5: 24-42.
Brodeur, J. R., 2006. Mixed Low-Level Radioactive and Hazardous Waste Disposal Facilities,
Energy Sciences and Engineering, Kennewick, Washington.
Clark, P. U., Dyke, A. S., Shakun, J. D., Carlson, A. E., Clark, J., Wohlfarth, B., Mitrovica, J. X.,
Hostetler, S. W., and A. M. McCabe, 2009. “The Last Glacial Maximum,” Science, 325:
710-714.
Cullen, A. C., "Measures of Compounding Conservatism in Probabilistic Risk Assessment", Risk
Analysis, 14 (4) , pp. 389 – 393, August 1994
Currey, D.R., G. Atwood, and D.R. Mabey, Map 73 Major Levels of Great Salt Lake and Lake
Bonneville, Utah Geological and Mineral Survey, Salt Lake City, UT, May 1984
Davis, O.K., and T.E. Moutoux, 1998. “Tertiary and Quaternary Vegetation History of the Great
Salt lake, Utah, USA.” J. Paleolimnology, 19: 417-427.
DOE, 2004a. Final Environmental Impact Statement for Construction and Operation of a
Depleted Uranium Hexafluoride Conversion Facility at the Paducah, Kentucky, Site,
DOE/EIS-0359, U.S. DOE Environmental Management, June 2004.
DOE, 2004b. Final Environmental Impact Statement for Construction and Operation of a
Depleted Uranium Hexafluoride Conversion Facility at the Portsmouth, Ohio, Site,
DOE/EIS-0360, U.S. DOE Environmental Management, June 2004.
Driscoll, N. W. and G. H. Haug, 1998. “A Short Circuit in Thermohaline Circulation: A Cause
for Northern Hemisphere Glaciation,” Science, 282: 436-438.
EnergySolutions, 2008. Bulk Waste Disposal and Treatment Facilities Waste Acceptance
Criteria, Rev. 7, Clive, Utah.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 72
EnergySolutions, 2009a. Energy Solutions Clive Facility, EnergySolutions License Amendment
Request: Class A South/11e.(2) Embankment, Revision 1, 9 Jun 2009, Clive, Utah.
EnergySolutions, 2009b. Radioactive Waste Profile Record, EC-0230, Rev. 7, plus attachments
(Form 9021-33), Clive, Utah.
EnergySolutions, 2009c, LLRW and 11e.(2) Construction Quality Assurance/Quality Control
Manual, rev 24e, EnergySolutions, Salt Lake City, Utah, 16 September 2009
Envirocare, 2000. Assessment of Vegetative Impacts on LLRW, Salt Lake City, Utah, November
2000.
Envirocare 2004. Revised Hydrogeologic Report for the Envirocare Waste Disposal Facility
Clive, Utah, Version 2.0, Salt Lake City Utah, August 2004.
Envirocare, 2005. Cover Test Cell Data Report Addendum: Justification to Change EZD from
18-inches to 24-inches. Attachment to letter dated October 13, 2005 from Daniel B. Shrum,
Director of Safety and Compliance, Envirocare of Utah, LLC, to Dane L. Finefrock,
Executive Secretary, Division of Radiation Control.
EPA (U.S. Environmental Protection Agency), 1988. Limiting Values of Radionuclide Intake and
Air Concentration and Dose Conversion Factors for Inhalation, Submersion, and
Ingestion, EPA 520/1-88-020, Federal Guidance Report (FGR) 11.
EPA, 1993. External Exposure to Radionuclides in Air, Water, and Soil, EPA 402-R-93-81, FGR
12.
EPA, 1999. Cancer Risk Coefficients for Environmental Exposure to Radionuclides: Updates
and Supplements, EPA 402-R-99-001, FGR 13.
EPA, 2001. Risk Assessment Guidance for SuperFund: Volume III, Part A, Process for
Conducting Probabilistic Performance Assessment. EPA 540-R-02-002, December 2001.
EPICA (European Project for Ice Coring in Antarctica). 2004. Eight glacial cycles from and
Antarctic ice core. Nature. 429, 623-628.
Fitzner, R.E., K.A. Gano, W.H. Rickard, and L.E. Rogers, 1979. Characterization of the
Hanford 300 Area Burial Grounds: Task IV - Biological Transport, PNL-2774, Pacific
Northwest Laboratory, Richland, Washington. Pp. 24–27.
Freeze, R.A. and J.A. Cherry. 1979. Groundwater. Prentice-Hall, Inc. Englewood Cliffs, NJ.
604p.
Fussell, G.M, and D. L. McWhorter, 2002. Project Plan for the Disposition of the SRS Depleted,
Natural, and Low-Enriched Uranium Materials. WSRC-RP-2002-00459, Washington
Savannah River Site, November 21, 2002.
Gano, K.A., D.W. Carlile and L.E. Rogers, 1985. A Harvester Ant Bioassay for Assessing
Hazardous Chemical Waste Sites, Pacific Northwest Laboratory, Battelle Memorial
Institute, Richland, Washington. 19 pp., appendices.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 73
Gates, J.S. and S.A. Kruer, Hydrologic reconnaissance of the Southern Great Salt Lake Desert
and summary of the hydrology of West-Central Utah, State of Utah Department of Natural
Resources, Technical Publication No. 71, Salt Lake City, UT.
Groenveld, D.P., 1990. Shrub Rooting and Water Acquisition in Threatened Shallow
Groundwater Habitats in the Owens Valley, California. Symposium on Cheatgrass
Invasion, Shrub Die-off, and Other Aspects of Shrub Biology and Management. pp. 221-
236, in: U.S. Dept. Agr. General Technical Report INT-267. Intermountain Research
Station, Ogden, Utah.
GTG (GoldSim Technology Group), 2010. GoldSim: Monte Carlo Simulation Software for
Decision and Risk Analysis, http://www.goldsim.com
Guzowski, R.V., 1990. Preliminary Identification of Scenarios That May Affect the Escape and
Transport of Radionuclides From the Waste Isolation Pilot Plant, Southeastern New
Mexico, SAND89-7149, Sandia National Laboratories, Albuquerque, NM.
Guzowski, R.V., and G. Newman, 1993, Preliminary Identification of Potentially Disruptive
Scenarios at the Greater Confinement Disposal Facility, Area 5 of the Nevada Test Site,
SAND93-7100, Sandia National Laboratories, Albuquerque, NM.
Hakonson, T.E., J.L. Martinez, and G.C. White, 1982. “Disturbance of a Low-level Waste Burial
Site Cover by Pocket Gophers,” Health Physics 42: 868–871.
Hansen, D.J., and W.K. Ostler, 2003. Rooting Characteristics of Vegetation Near Areas 3 and 5
Radioactive Waste Management Sites at the Nevada Test Site, DOE/NV/11718-595, U.S.
Department of Energy National Nuclear Security Administration Nevada Site Office, 125
pp.
Hart, W. S., Quade, J., Madsen, D. B., Kaufmann, D. S., and C. G. Oviatt, 2004. “The 87Sr/86Sr
Ratios of Lacustrine Carbonates and Lake-level History of the Bonneville Paleolake
System,” GSA Bulletin, 116: 1107-1119.
Haug, G. H. and R. Tiedemann, 1998. “Effect of the Formation of the Isthmus of Panama on
Atlantic Ocean Thermohaline Circulation,” Nature, 393: 673-676.
Hays, J. D., Imbire, J., and N. J. Shackleton, 1976. “Varitations in the Earth's Orbit: Pacemaker
of the Ice Ages,” Science, 194: 1121-1132.
Hetzel, R. and A. Hampel, 2005. “Slip rate variations on normal faults during glacial-interglacial
changes in surface loads,” Nature, 435: 81-84.
Holmgren. R.C. and S.F. Brewster, 1972. Distribution of Organic Matter Reserve in a Desert
Shrub Community, Paper INT-130, U.S. Dept. Agr. Forest Serv. Research, Intermountain
Forest and Range Experiment Station, Ogden, Utah.
ICRP (International Commission on Radiation Protection), 1979. ICRP Publication 30 Limits for
Intakes of Radionuclides by Workers, Ann. ICRP 2(3-4), 1979.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 74
ICRP, 1984. ICRP Publication 30. Cost-Benefit Analysis in the Optimization of Radiation
Protection. Ann ICRP. 1983;10(2-3):1-75.
ICRP, 1989. ICRP Publication 56 Age-Dependent Doses from Intakes of Radionuclides: Part 1,
Ann. ICRP 20(2), 1989.
ICRP, 1996. ICRP Publication 72 Age-Dependent Doses to Members of the Public from Intake of
Radionuclides, Part 5: Compilation of Ingestion and Inhalation Dose Coefficients, Ann.
ICRP 26(1), 1996. P 072 errata in SG 03 JAICRP 32 (1-2).
Johnson R. 2010. State of Utah, DEQ. Memo – April 6, 2010 Subj. Savannah River Deplete
Uranium Sampling
Landeen, D.S., 1994, The influence of small mammal burrowing activity on water storage at the
Hanford Site, WHC-EP0730 UC-702, Westinghouse Hanford Company, Richland, WA.
Laundré, J.W., 1993. “Effects of Small Mammal Burrows on Water Infiltration in a Cool Desert
Environment,” Oecologia, 94: 43–48.
Laundré, J.W. and T.D. Reynolds, 1993. “Effects of Soil Structure on Burrowing Characteristics
of Five Small Mammal Species,” Great Basin Naturalist 53(4): 358–366.
Lavigne, R.J. ,1969. “Bionomics and Nest Structure of Pogonomyrmex occidentalis
(Hymenoptera: Formicidae),” Ann. Ent. Soc. Am. 62: 1166–1175.
Link, P. K., Kaufman, D. S., and G. D. Thackray, 1999. Field guide to Pleistocene lakes
Thatcher and Bonneville and the Bonneville Flood, southeastern Idaho, in Hughes, S. S.
and G. D. Thackray (eds.), Guidebook to the Geology of Eastern Idaho, Idaho Museum of
Natural History, pp. 251-266.
Los Alamos National Laboratory (LANL), 2008, Performance assessment and composite
analysis for Los Alamos National Laboratory Technical Area G, Revision 4. LA-UR-08-
06764. Los Alamos National Laboratory, Los Alamos, NM.
Lowe, J. J. and M. J. C. Walker, 1997.Reconstructing Quaternary Environments, 2nd Edition.
Prentice Hall, London, 446 pp.
Lowry, J.H, Jr., R.D. Ramsey, K.A. Thomas, D. Schrupp, W. Kepner, T. Sajwaj, J. Kirby, E.
Waller, S. Schrader, S. Falzarano, L. Langs Stoner, G. Manis, C. Wallace, K. Schulz, P.
Comer, K. Pohs, W. Rieth, C. Velasquez, B. Wolk, K.G., Boykin, L. O’Brien, J. Prior-
Magee, D. Bradford, and B. Thompson. 2007. Land cover classification and mapping
(Chapter 2). In Southwest Regional Gap Analysis Final Report, edited by J. S. Prior-
Magee, et al. Moscow, Idaho: U.S. Geological Survey, Gap Analysis Program.
MacKay, W.P., 1981. “A Comparison of the Nest Phenologies of Three Species of
Pogonomyrmex Harvester Ants (Hymenoptera: Fomicidae),” Psyche 88: 25–74.
Mandel, R.D. and C.J. Sorenson, 1981. “The Role of the Western Harvester Ant (Pogonomyrmex
occidentalis) in Soil Formation,” Soil Sci. Soc. Am. J. 46: 785–788.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 75
Masson-Delmotte, V., M. Schulz, A. Abe-Ouchi, J. Beer, A. Ganopolski, J.F. Gonzalez Rouco, E.
Jansen, K. Lambeck, J. Luterbacher, T. Naish, T. Osborn, B. Otto-Bliesner, T. Quinn, R..
Ramesh, M. Rojas, X. Shao, and A. Timmermann, 2013, Information from Paleoclimate
Arechive. In: Climate Change 2013: The Physical Science Basis. Contribution of
Working group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J.
Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA.
Meinzer, O.E., 1927. Plants As Indicators of Groundwater, United States Department of Interior
Geological Survey Water Supply Paper 577. Washington, D.C.
Miller, D.M., C.G. Oviatt, and J.P. Mcgeehin, 2013. Stratigraphy and chronology of Provo
shoreline deposits and lake-level implications, Late Pleistocene Lake Bonneville, eastern
Great Basin, U.S.A., Boreas, Vol. 42, pp. 342–361, 2013.
MSI (Meteorological Solutions, Inc.) 2010. January 2009 Through December 2009 and January
1993 Through December 2009 Summary Report of Meteorological Data Collected At
EnergySolutions’ Clive, Utah Facility. Report for EnergySolutions by Meteorological
Solutions, Inc. February 2010.
National Research Council, 2005. Risk and Decisions, National Academies Press, Washington,
DC, 2005.
NEA (Nuclear Energy Agency), 1992, Systematic Approach to Scenario Development. A Report
of the NEA Working Group on the Identification and Selection of Scenarios for
Performance Assessment of Radioactive Waste Disposal, Nuclear Energy Agency, Paris,
France.
NEA, 2000. Features, Events, and Processes (FEPs) for Geologic Disposal of Radioactive
Waste. An International Database. Nuclear Energy Agency, Organization for Economic
Cooperation and Development.
Neptune, 2015. Neptune Field Studies, December, 2014, Eolian Depositional History Clive
Disposal Site, NAC-0044_R0, Neptune and Company Inc., Los Alamos NM, March
2015.
Nevo, E., 1999. Mosaic Evolution of Subterranean Mammals: Regression, Progression and
Global Convergence. Oxford University Press, Oxford England. 413 pp.
NRC (U.S. Nuclear Regulatory Commission), 1982. Final Environmental Impact Statement on
10 CFR Part 61 Licensing Requirements for Land Disposal of Radioactive Waste,
NUREG-0945, Nuclear Regulatory Commission, Washington, DC, November 1982.
NRC, 1989a. Methodology, Identification of Potential Exposure Pathways, NUREG/CR-5453,
Nuclear Regulatory Commission, Washington, DC.
NRC, 1993. Final Environmental Impact Statement to Construct and Operate a Facility to
Receive, Store, and Dispose of 11e.(2) Byproduct Material Near Clive, Utah, NUREG-
1476, Nuclear Regulatory Commission, Washington, DC.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 76
NRC, 1994. Standard Review Plan for the review of a license application for a Low-Level
Radioactive Waste Disposal Facility, NUREG-1200, Rev. 3, Nuclear Regulatory
Commission, April 1994.
NRC, 2000. A Performance Assessment Methodology for Low-Level Radioactive Waste Disposal
Facilities, NUREG-1573, Nuclear Regulatory Commission, Washington, DC, October
2000.
NRC, 2005. In the Matter of Louisiana Energy Services (National Enrichment Facility), CLI-05-
10, Docket 70-3103-ML, Nuclear Regulatory Commission, Washington, DC, October 19,
2005.
NRC, 2008. Response to Commission Order CLI-05-20 Regarding Depleted Uranium, SECY-
08-0147, Nuclear Regulatory Commission, Washington, DC.
NRC, 2009a. Memorandum, Staff requirements – SECY-08-0147 – Response to commission
order CLI-05-20 regarding depleted uranium, Nuclear Regulatory Commission,
Washington, DC, March 18, 2009.
NRC, 2009b. NRC News, No. 09-052, Nuclear Regulatory Commission, Washington, DC, March
18, 2009.
NRC, 2010. “Stages of the Nuclear Fuel Cycle,” U.S. Nuclear Regulatory Commission website,
http://www.nrc.gov/materials/fuel-cycle-fac/stages-fuel-cycle.html
ORNL. 2000. Strategy for Characterizing Transuranics and Technetium Contamination in
Depleted UF6 Cylinders, ORNL/TM-2000/242, Oak Ridge National Laboratory, October
2000.
Ota, M., Iida, T., Yamazawa, H., Nagara, S., Ishimori, Y., Sato, K., and T. Tokizawa, 2007.
“Suppression of Radon Exhalation from Soil by Covering with Clay-mixed Soil,” Journal
of Nuclear Science and Technology, 44(5): 791-800.
Oviatt, C. G., 1997. Lake Bonneville fluctuations and global climate change. Geology, 25(2):
155-158.
Oviatt, C. G., Thompson, R. S., Kaufman, D. S., Bright, J., and R. M. Forester, (1999).
“Reinterpretation of the Burmester Core,” Bonneville Basin, Utah, Quaternary Research,
52, 180-184.
Paillard, D., 2001. Glacial cycles: toward a new paradigm, Reviews of Geophysics, 39(3): 325-
346.
Paillard, D., 2006. “What drives the Ice Age cycle?” Science, 313: 455-456.
Rich, B.L., S.L. Hinnefeld, C.R. Lagerquist, W.G. Mansfield, L.H. Munson, E.R. Wagner, and
E.J. Vallario, 1988. Health Physics Manual of Good Practices for Uranium Facilities,
EGG-2530, Idaho National Engineering Laboratory, Idaho Falls, ID, June 1988.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 77
Robinson, T.W., 1958. Phreatophytes. United States Department of Interior Geological Survey
Water Supply Paper 577, Washington, D.C.
Rundel, P.W., and P.S. Nobel, 1991. “Structure and Function in Desert Root Systems,” In: D.
Atkinson (Ed.), Plant Root Growth: An Ecological Perspective, Special Publication #10 of
the British Ecological Society. Blackwell Scientific Pub. London, England. Pp. 349-378.
Shott, G., V. Yucel and L. Desotell, 2008, Special Analysis of Transuranic Waste in Trench
T04C at the Area 5 Radioactive Waste Management Site, Nevada Test Site, Nye County,
Nevada, Revision 1.0, DOE/NV/25946–470, National Security Technologies, LLC
Smallwood, K.S., M.L. Morrison, and J. Beyea, 1998. Animal Burrowing Attributes Affecting
Hazardous Waste Management. Environmental Management, 22(6): 831–847.
Smith, E.D., R.J. Luxmoore, and G.W. Suter, II, 1997. Natural Physical and Biological
Processes Compromise the Long-Term Performance of Compacted Soil Caps, Barrier
Technologies for Environmental Management: Summary of a Workshop. National
Academy of Sciences, 1997.
SWCA, 2011, Field Sampling of Biotic Turbation of Soils at the Clive Site, Tooele County, Utah,
SWCA Environmental Consultants, Salt Lake City, Utah, January 2011.
SWCA, 2012, Vegetated Cover System for the EnergySolutions Clive Site: Literature Review,
Evaluation,of Existing Data, and Field Studies Summary Report, SWCA Environmental
Consultants, Salt lake City, Utah, August 2012.
SWCA, 2013, EnergySolutions Updated Performance Assessment –SWCA’s Response to First
Round DRC Interrogatories, SWCA Environmental Consultants, Salt Lake City, Utah,
September 2013.
Tzedakis, P.C., J.E.T. Channel, D.A. Hodell, H.F. Kleiven, and L.C. Skinner, 2012b),
Determining the natural length of the current interglacial, Nature Geoscience, 5, 138-141.
Utah, State of, 2015, Utah Administrative Code Rule R313-25. License Requirements for Land
Disposal of Radioactive Waste - General Provisions. As in effect on September 1, 2015.
(http://www.rules.utah.gov/publicat/code/r313/r313-025.htm, accessed 5 Nov 2015).
Utah, State of, 2015, Utah Administrative Code Rule R313-15. Standards for Protection Against
Radiation. As in effect on September 1, 2015.
(http://www.rules.utah.gov/publicat/code/r313/r313-015.htm, accessed 5 Nov 2015).
UWQB (State of Utah, Division of Water Quality, Utah Water Quality Board), 2010. Ground
Water Quality Discharge Permit No. 450005, 23 Dec 2010.
Whetstone (Whetstone Associates, Inc.), 2000. Revised Envirocare of Utah Western LARW Cell
Infiltration and Transport Modeling, Lakewood, Colorado, 19 July 2000.
Whetstone, 2006. EnergySolutions Class A Combined (CAC) Disposal Cell Infiltration and
Transport Modeling Report, Salt Lake City Utah, May 2006.
Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
5 November 2015 78
Wong, I., W. Arabasz, B. Carey, C. DuRoss, W. Lund, J. Pechmann, and B. Welliver, 2013,
Opinion, Seismological Research Letters, Vol. 84, 165-169.

 NAC-0019_R4

Embankment Modeling for the
Clive DU PA

Clive DU PA Model v1.4
21 October 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Embankment Modeling for the Clive DU PA
21 October 2015 ii
1. Title: Embankment Modeling for the Clive DU PA
2. Filename: Embankment Modeling v1.4.docx
3. Description: This white paper addresses specific details relating to the dimensional
components of the Federal Cell, located at the Clive facility. This paper includes a
description of the parameters and calculations used to estimate the various dimensional
components of the Federal Cell.
Name Date
4. Originator Dan Levitt May 23, 2014
5. Reviewer Mike Sully May 26, 2014
6. Remarks

23 May 2014: DL. Revision to R1 of this white paper is not yet complete. Some references to
“Class A South” remain in this document and cannot be updated until model parameters are
also updated).

2 Jul 2014: R2; Modified Figure 9 in response to Interrogatory 161. – J Tauxe

15 Jul 2015: Amir Mohktari. Updated “Class A South” terminology to “Federal DU.”

15 October 2015: Paul Black, Katie Catlett, final review and finalization of v1.3.

19 October 2015: Final review and finalization of v1.3. – J Tauxe

20 October 2015: Revisions to figures and terminology for v1.4. – J Tauxe

21 October 2015: Additional revisions related to new figures for v1.4. – G. Occhiogrosso;
reviewed v1.4. – K. Catlett

Embankment Modeling for the Clive DU PA
21 October 2015 iii

This page is intentionally blank, aside from this statement.
Embankment Modeling for the Clive DU PA
21 October 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ........................................................................................................................................ vi

1.0
 Summary of Parameter Values .............................................................................................. 1

2.0
 Introduction ............................................................................................................................ 2

3.0
 Physical Dimensions .............................................................................................................. 2

3.1
 Federal Cell Dimensions .................................................................................................. 2

3.1.1
 Federal Cell Interior Waste ........................................................................................ 5

3.1.2
 Federal Cell Cover and Liner Dimensions ................................................................. 9

4.0
 Original Grade Elevation ....................................................................................................... 9

4.1
 Federal Cell Original Grade ........................................................................................... 12

5.0
 Model Implementation using GoldSim ................................................................................ 13

5.1
 Representation of the Federal Cell ................................................................................. 13

5.1.1
 Federal Cell Dimensions .......................................................................................... 13

5.1.2
 Federal Cell Columns ............................................................................................... 13

6.0
 References ............................................................................................................................ 16

Embankment Modeling for the Clive DU PA
21 October 2015 v
FIGURES
Figure 1. The Clive Facility, with the location of the Federal Cell outlined in green. This
orthophotograph is roughly 1 mile across, and north is up. .......................................... 3

Figure 2. Section and Plan views of the Federal Cell, with top slope shown in blue and side
slope in green. The brown dotted line in the West-East Cross section represents
below-grade (below the line) and above-grade (above the line) regions of the
embankment. ................................................................................................................. 4

Figure 3. Dimensions of the Federal Cell that are used in the Clive DU PA Model. Not to
scale. .............................................................................................................................. 6

Figure 4. Federal Cell and 11e.(2) Cell engineering drawing 14004 V1A. (EnergySolutions
2014c) ........................................................................................................................... 7

Figure 5. Federal Cell and 11e.(2) Cell engineering drawing 14004 V3A (west-east cross
section) (EnergySolutions 2014d). ................................................................................ 8

Figure 6. Federal Cell engineering drawing 14004 V7: cap dimensions. (EnergySolutions
2014a) ......................................................................................................................... 10

Figure 7. Federal Cell and 11e.(2) Cell engineering drawing 14004 L1A (west-east cross
section) (EnergySolutions 2014b). .............................................................................. 11

Figure 8. Section 32 within the Aragonite quadrangle, as it appeared in 1973, before
construction of the Clive Facility. Note elevation contours at 4270 and 4280 ft
amsl. ARAGONITE NW is the next quadrangle to the west. ....................................... 12

Figure 9. Geometrical deconstruction of the Federal Cell waste volumes. .................................. 14

Figure 10. Waste layering definitions within the two columns of the Federal Cell. .................... 15

Embankment Modeling for the Clive DU PA
21 October 2015 vi
TABLES
Table 1. Summary of embankment engineering parameters .......................................................... 1

Table 2. Cover layer thicknesses for the Federal Cell .................................................................... 9

Embankment Modeling for the Clive DU PA
21 October 2015 1
1.0 Summary of Parameter Values
The parameters that define the characteristics of the Federal Cell at the Clive facility are
summarized in Table 1. Of principal interest to the model are the interior dimensions of the
volume occupied by waste, and the thicknesses of the various layers in the engineered cover.

Table 1. Summary of embankment engineering parameters
Parameter Value Units Reference / Comment
average original grade elevation 4272 ft amsl* USGS (1973) see §4.1
height of top of the waste at the ridgeline 47.5 ft amsl EnergySolutions (2014c)
see §3.1.1
height of top of the waste at the break in
slope
35.0 ft amsl EnergySolutions (2014c)
see §3.1.1
average elevation of the bottom of the
waste
4264 ft amsl EnergySolutions (2014d)
see §3.1.1
height of the clay liner 2 ft EnergySolutions (2014a)
see §3.1.2
length overall 1317.8 ft EnergySolutions (2014c)
see §3.1.1
width overall 1775.0 ft EnergySolutions (2014c)
see §3.1.1
length to break 175.0 ft EnergySolutions (2014c)
see §3.1.1
width to break 175.0 ft EnergySolutions (2014c)
see §3.1.1
break to ridge length (west) 521 ft EnergySolutions (2014c)
see §3.1.1
break to ridge length (east) 447 ft EnergySolutions (2014c)
see §3.1.1
break to ridge width 521 ft EnergySolutions (2014c)
see §3.1.1
ET Cover Layer Thicknesses
surface 0.5 ft EnergySolutions (2014a)
Federal Cell Drawing
14004 V7
evaporative zone 1.0 ft ibid.
frost protection 1.5 ft ibid.
upper radon barrier 1.0 ft ibid.
lower radon barrier 1.0 ft ibid.
*above mean sea level
Embankment Modeling for the Clive DU PA
21 October 2015 2
2.0 Introduction
The safe storage and disposal of depleted uranium (DU) waste is essential for mitigating releases
of radioactive materials and reducing exposures to humans and the environment. Currently, a
radioactive waste facility located in Clive, Utah (the “Clive facility”) operated by the company
EnergySolutions, Inc., is being considered to receive and dispose DU waste that has been
declared surplus from radiological facilities across the nation. The Clive facility has been tasked
with disposing of the DU waste in a manner that protects humans and the environment from
future radiological releases.
To assess whether the proposed Clive facility location and containment technologies are suitable
for protection of human health, specific performance objectives for land disposal of radioactive
waste set forth in Utah Administrative Code (UAC) Rule R313-25 License Requirements for
Land Disposal of Radioactive Waste— General Provisions (Utah 2015) must be met—
specifically R313-25-9 Technical Analyses. In order to support the required radiological
performance assessment (PA), a probabilistic computer model has been developed to evaluate the
doses to human receptors and the concentrations in groundwater that would result from the
disposal of radioactive waste, and conversely to determine how much waste can be safely
disposed at the Clive facility. The GoldSim systems analysis software (GTG, 2015) was used to
construct the probabilistic PA model.
The site conditions, chemical and radiological characteristics of the wastes, contaminant transport
pathways, and potential human receptors and exposure routes at the Clive facility that are used to
structure the quantitative PA model are described in the conceptual site model documented in the
white paper titled Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility.
The purpose of this white paper, Embankment Modeling for the Clive DU PA, is to address
specific details relating to the dimensional components of the Federal Cell, located at the Clive
facility. This paper is organized to give a brief overview of where the Federal Cell section is
located at the Clive facility followed by a description of the parameters and calculations used to
estimate the various dimensional components of the Federal Cell.
This probabilistic PA takes into account uncertainty in many input parameters, but the
dimensions of the Federal Cell are not considered to be uncertain. Given that the disposal cell is
carefully designed and constructed, any uncertainty in its dimensions is considered insignificant.
Stochastic representation of parameters is reserved for those values about which there is
uncertainty.
3.0 Physical Dimensions
The Clive DU PA Model considers only a single embankment. For the purposes of this PA, only
the Federal Cell is considered for disposal (Figure 1).
3.1 Federal Cell Dimensions
The Federal Cell, or embankment, location at the Clive facility is identified in Figure 1. A
stylized drawing of the Federal Cell is shown in Figure 2.
Embankment Modeling for the Clive DU PA
21 October 2015 3
The general aspect of the Federal Cell is that of a hipped cap, with relatively steeper sloping sides
nearer the edges. The upper part, known as the top slope, has a moderate slope, while the side
slope is markedly steeper (20% as opposed to 2.4%). These two distinct areas, shown in different
colors in the Plan View diagram of Figure 2, are modeled separately in the Clive DU PA Model.
Each area is modeled as a separate one-dimensional column, with an area equivalent to the
corresponding embankment footprint. The embankment is also constructed such that a portion of
it lies below-grade (Figure 2).

Figure 1. The Clive Facility, with the location of the Federal Cell outlined in green. This
orthophotograph is roughly 1 mile across, and north is up.
Embankment Modeling for the Clive DU PA
21 October 2015 4

Figure 2. Section and Plan views of the Federal Cell, with top slope shown in blue and side
slope in green. The brown dotted line in the West-East Cross section represents
below-grade (below the line) and above-grade (above the line) regions of the
embankment.

Embankment Modeling for the Clive DU PA
21 October 2015 5
3.1.1 Federal Cell Interior Waste
The Clive DU PA Model requires information about embankment dimensions to be able to
determine the footprint areas and the volumes of waste within each area. From this, an average
thickness of the waste is determined, since the 1-D column represents a single thickness over its
entire area. All dimensions provided in this white paper are with respect to the waste itself, and
do not include the liner or cover materials, with a few exceptions as noted. The dimensions of
interest that are used in the Clive DU PA Model are shown in Figure 3.
The values of the dimensions shown in Figure 3 are derived from various engineering drawings
as noted below.
As shown in the engineering drawings, the exact dimensions of the Federal Cell are somewhat
irregular, with a gently sloping bottom and ridge line. The shape of the cell has been somewhat
idealized to facilitate calculations, and it is assumed to have a horizontal floor and ridge line.
Height of the top of the waste at the ridge line: Elevations for the top of the waste are shown in
drawing 14004 V1A (Figure 4), which has the note “1. All elevations shown are for top of
waste...” At the ridge, the top of the waste is at height of about 47.5 ft.
Height of the top of the waste at the slope break: Also shown in 14004 V1A (Figure 4), the
height of the top of the wastes at the slope break (shoulder) is given as about 35 ft.
Elevation of the bottom of the waste: This is estimated from the values provided in the drawing
14004 V3A (Figure 5). The top of the liner protective cover is at an elevation of 4263 ft at the
west end and 4266 ft at the east end, which includes the area under the neighboring 11e.(2) Cell.
The elevation of the bottom of the waste at the midpoint of the Federal Cell is estimated by linear
interpolation to be at 4264 ft.
Length overall: Defined in Figure 3, the overall length (east to west) of the waste footprint in the
Federal Cell is shown in drawing 14004 V1A (Figure 4) to be 1317.8 ft.
Width overall: Defined in Figure 3, the overall width (north to south) of the waste footprint in
the Federal Cell is shown in drawing 14004 V1A (Figure 4) to be 1775.0 ft.
Length to break: Defined in Figure 3, the length from edge of the unit to the break in slope is
shown in drawing 14004 V1A (Figure 4) to be 175.0 ft. This value is the same on the east and
west sides of the disposal unit.
Width to break: Defined in Figure 3, the width from the edge of the unit to the break in slope is
shown in drawing 14004 V1A (Figure 4) to be 175.0 ft. This value is the same on the north and
south sides of the disposal unit.
Break to ridge length (west): Defined in Figure 3, on the west side of the disposal unit, the
length from the break in slope to the ridge is shown in drawing 14004 V1A (Figure 4) to be
521 ft.
Break to ridge length (east): Defined in Figure 3, on the east side of the disposal unit, the length
from the break in slope to the ridge is shown in drawing 14004 V1A (Figure 4) to be 447 ft.
Embankment Modeling for the Clive DU PA
21 October 2015 6

Figure 3. Dimensions of the Federal Cell that are used in the Clive DU PA Model. Not to
scale.
Embankment Modeling for the Clive DU PA
21 October 2015 7

Figure 4. Federal Cell and 11e.(2) Cell engineering drawing 14004 V1A. (EnergySolutions
2014c)
Embankment Modeling for the Clive DU PA
21 October 2015 8

Figure 5. Federal Cell and 11e.(2) Cell engineering drawing 14004 V3A (west-east cross
section) (EnergySolutions 2014d).
Embankment Modeling for the Clive DU PA
21 October 2015 9
Break to ridge width: Defined in Figure 3, the width from the break in slope to the ridge is
shown in drawing 14004 V1A (Figure 4) to be 521 ft. This value is the same on north and south
sides of the disposal unit.
Ridge length: Defined in Figure 3, the length along the ridge (north-south) is derived from
drawing 14004 V1A (Figure 4). It is equal to the “overall width” less twice the distance from the
edge of the unit to the ridge in the north-south direction. Based on the quantities above, the ridge
length is calculated in the GoldSim model as: 1775.0 ft – ( 2 × ( 521 ft + 175 ft ) ) = 383 ft.
3.1.2 Federal Cell Cover and Liner Dimensions
The engineered cover designs for the top slope and side slope sections of the Federal Cell are
shown in drawing 14004 V7 (Figure 6). The values chosen from the sections labeled “ET Cover
Top Slopes” and “ET Cover Side Slopes” are summarized in Table 2. The properties of the
various layers within the engineered cover and liner are discussed in detail in the Unsaturated
Zone Modeling white paper.
Table 2. Cover layer thicknesses for the Federal Cell
layer thickness (ft)
top slope side slope
surface 0.5 0.5
evaporative zone 1.0 1.0
frost protection 1.5 1.5
upper radon barrier 1.0 1.0
lower radon barrier 1.0 1.0
The waste layers of the embankment are underlain by a clay liner, as shown in Figure 7. The
thickness of the clay liner is defined in the engineering drawing 14004 L1A (EnergySolutions,
2014b) as 2 ft.
Elevation of the bottom of the clay liner: This is calculated simply as the average elevation of
the bottom of the waste minus the thickness of the liner. The elevation of the bottom of the clay
liner is then 4264 ft – 2 ft = 4262 ft and is calculated as such in the GoldSim model. Note that for
model simplification, the liner protective cover is assumed to be a part of the unsaturated zone, in
essence below the clay liner instead of above it.
Note that this is also the elevation of the top of the unsaturated zone.
4.0 Original Grade Elevation
The original grade is of interest for determining the vertical location of wastes inside the
embankment. Above-ground waste or other material can be considered erodible, and, conversely,
below-ground waste to be inherently not erodible. It is therefore of interest to determine the
disposal volume that lies below grade since placing waste below grade greatly reduces the
potential for erosion during lake cycles. Again, only the Federal Cell is considered at this time.
Embankment Modeling for the Clive DU PA
21 October 2015 10

Figure 6. Federal Cell engineering drawing 14004 V7: cap dimensions. (EnergySolutions
2014a)

Embankment Modeling for the Clive DU PA
21 October 2015 11

Figure 7. Federal Cell and 11e.(2) Cell engineering drawing 14004 L1A (west-east cross
section) (EnergySolutions 2014b).
Embankment Modeling for the Clive DU PA
21 October 2015 12
4.1 Federal Cell Original Grade
The elevation of the original grade is interpreted from the elevations indicated on a 1:24,000
scale quadrangle map for Aragonite, UT (USGS, 1973). The relevant section of this map as it
applies to the Federal Cell is shown in Figure 8. This 1-square mile section, Section 32, is the site
of the Clive Facility (Figure 8). The southwest corner of Section 32 is at elevation 4270 ft amsl
(above mean sea level) while the ground surface (original grade) slopes gently and fairly
uniformly up to the northeastern corner, crossing the 4280-ft amsl contour.
The Federal Cell occupies the southwestern corner of Section 32 (refer to Figure 1), and its center
is approximately at an elevation of 4272 ft amsl. This is the value used for original grade of the
Federal Cell.

Figure 8. Section 32 within the Aragonite quadrangle, as it appeared in 1973, before
construction of the Clive Facility. Note elevation contours at 4270 and 4280 ft amsl.
ARAGONITE NW is the next quadrangle to the west.
Embankment Modeling for the Clive DU PA
21 October 2015 13
5.0 Model Implementation using GoldSim
5.1 Representation of the Federal Cell
The representation of the Federal Cell in the Clive DU PA Model is essentially one-dimensional
(1-D), and is therefore necessarily simplified. The top slope and side slope sections of the
embankment are modeled as independent 1-D columns, as discussed below. The volumes of
waste and the layers of engineered cap and liner are preserved. Since the cap and liner are
laterally continuous and do not vary in dimension within a column, the thicknesses in the model
correspond directly to thicknesses in the real world. The waste layers, however, are of a shape
that changes in the horizontal, and must be rearranged to produce a shape that is a rectangular
prism of equal volume to the actual waste volume.
5.1.1 Federal Cell Dimensions
The dimensions developed in Section 3.1 are documented in the Clive DU PA GoldSim model
(GoldSim model) in the container \Disposal\FederalCell\Federal_Cell_Dimensions. The
calculation of the waste volumes within the side slope and within the top slope, as identified in
Figure 2, is performed within the Model by assembling pieces that have volumes that are easily
calculated using basic geometry, as shown in Figure 9. Once the waste volumes for top and side
slope are known, the average waste thicknesses are calculated. These are used as the waste
thicknesses in the columns within the Model, as described in the following section.
5.1.2 Federal Cell Columns
The top slope and side slope columns are modeled in parallel, since they have different waste and
cap layer thicknesses. That is, each column has only vertical flow of water. The vertical flow
feeds into the unsaturated zone and thence to groundwater at the bottom. The top slope column
has a much thicker waste layer than the side slope, and this is reflected in the overall thickness of
the two columns. In order to capture the flexibility available in locating waste during disposal
operations, the user can select which waste types go where in the top slope column, using the
Waste Layering Definition dashboard. No DU wastes are to be disposed in the side slope column
in this model. An example of this selection in the GoldSim model is shown in Figure 10. The
waste configuration in Figure 10 is consistent with the most recent engineering drawings,
locating the waste in the bottom 7 - 7.8 ft of the embankment (EnergySolutions 2014b).

Embankment Modeling for the Clive DU PA
21 October 2015 14

Figure 9. Geometrical deconstruction of the Federal Cell waste volumes.
Embankment Modeling for the Clive DU PA
21 October 2015 15

Figure 10. Waste layering definitions within the two columns of the Federal Cell.

Embankment Modeling for the Clive DU PA
21 October 2015 16
6.0 References
EnergySolutions, 2009. EnergySolutions License Amendment Request: Class A South/11e.(2)
Embankment, Revision 1, 9 June 2009 (file: Class A South-11e.(2) Eng Drawings.pdf).
EnergySolutions, 2014a. Engineering Drawings (file: Federal Cell drawing 14004.pdf). [Note:
Three drawings in this drawing set (14004 V1, 14004 V2, and 14004 L1), are superseded
for v1.4; see the three references below.]
EnergySolutions, 2014b. Engineering Drawing 14004 L1A, “Conceptual DU Disposal Plan”,
dated 11/13/2014 with revision on 12/1/2014. (file: FederalCell DUplan 14004 L1A.pdf).
EnergySolutions, 2014c. Engineering Drawing 14004 V1A, “Cell Layout”, dated 11/06/2014
with revision on 12/1/2014. (file: FederalCell plan 14004 V1A.pdf).
EnergySolutions, 2014d. Engineering Drawing 14004 V3A, “Cell Cross Section – East/West”,
dated 11/06/2014 with revision on 12/1/2014. (file: FederalCell section 14004 V3A.pdf).
GTG (GoldSim Technology Group), 2015. GoldSim: Monte Carlo Simulation Software for
Decision and Risk Analysis, http://www.goldsim.com
USGS (United States Geological Survey), 1973. 1:24,000 topographic quadrangle map for
Aragonite, UT, revised 1973 (file: UT_Aragonite_1973_geo.pdf).
Utah, State of, 2015, Utah Administrative Code Rule R313-25. License Requirements for Land
Disposal of Radioactive Waste - General Provisions. As in effect on September 1, 2015.
(http://www.rules.utah.gov/publicat/code/r313/r313-025.htm, accessed 5 Nov 2015).

 NAC-0023_R4

Radioactive Waste Inventory for
the Clive DU PA

Clive DU PA Model v1.4
12 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 ii
1. Title: Radioactive Waste Inventory for the Clive DU PA
2. Filename: Waste Inventory v1.4.docx
3. Description: Description of the waste inventory input distributions for the Clive DU PA Model
v1.4
Name Date
4. Originator Paul Black 31 May 2014
5. Reviewer Mike Sully 31 May 2014
6. Remarks

2015 Oct 28. Added information on new waste disposal dimensions and corresponding calculations for
number of GDP cylinders to be disposed of at Clive. Kate Catlett

2015 Oct 29. Added section numbers to Table 1 and text to calculations for GDP cylinders. Gregg
Occhiogrosso

2015 Oct 30. Review & edit. Kate Catlett

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 iii

This page is intentionally blank, aside from this statement.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

1.0
 Waste Inventory Parameters Summary .................................................................................. 1

2.0
 Uranium Oxide Inventory ....................................................................................................... 3

2.1
 Depleted Uranium ............................................................................................................. 3

2.2
 Savannah River Site Depleted Uranium ........................................................................... 5

2.2.1
 Mass of SRS Depleted Uranium Proposed for Disposal ............................................. 5

2.2.2
 Composition of SRS Depleted Uranium ..................................................................... 5

2.3
 Depleted Uranium Oxide from the Gaseous Diffusion Plants .......................................... 6

2.3.1
 Mass of GDP Depleted Uranium ................................................................................ 7

2.3.2
 Composition of GDP Depleted Uranium .................................................................... 8

3.0
 Input Parameter Distribution Development ............................................................................ 8

3.1
 Parameters for Depleted Uranium from the Savannah River Site .................................... 8

3.1.1
 Mass of SRS Depleted Uranium ................................................................................. 8

3.1.2
 Composition of SRS Depleted Uranium ..................................................................... 9

3.2
 Analysis of Uranium Composition in SRS Depleted Uranium ...................................... 11

3.2.1
 Exploratory Comparison of Uranium Data ............................................................... 12

3.2.2
 Partitioning 233+234U and 235+236U .............................................................................. 14

3.2.3
 SRS Depleted Uranium Activity Concentration ....................................................... 15

3.3
 Analysis of Technetium Concentrations in SRS DU ...................................................... 18

3.4
 Concentrations of Other Radionuclides in the SRS Depleted Uranium ......................... 22

3.5
 Parameters for Depleted Uranium Oxide from the GDPs .............................................. 22

3.5.1
 Mass of GDP DU ...................................................................................................... 23

3.5.2
 Number of GDP DU cylinders disposed ................................................................... 23

3.5.3
 Composition of GDP DU .......................................................................................... 24

3.5.3.1
 Clean GDP DU ........................................................................................ 24

3.5.3.2
 Contaminated GDP DU .......................................................................... 24

3.5.3.3
 Fraction of Contaminated GDP DU ........................................................ 25

4.0
 References ............................................................................................................................ 29

Appendix ....................................................................................................................................... 31

Appendix References ..................................................................................................................... 38

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 v
FIGURES
Figure 1. Comparison of activity percent for the SRS DU uranium isotopes ............................... 14

Figure 2. Distribution of mean activity concentration values from bootstrap resampling. ........... 17

Figure 3. Tc-99 Activity Concentration. Sample sizes: SRS-2002 = 33; ES-2010 = 11; Utah-
2010 = 173. .................................................................................................................. 19

Figure 4. Distribution of Tc-99 mean values. Red lines indicate mean values of Utah-2010,
ES-2010 and SRS-2002 results. The dashed lines indicate the 5th and 95th
percentiles of the mean values of the resampled data. ................................................ 21

Figure 5. Additional radionuclide data (SRS-2002). Sample size = 33. ....................................... 23

Figure 6. Probability density function for the proportion of contaminated cylinders. .................. 28

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 vi
TABLES
Table 1. Summary input parameter values and distributions .......................................................... 1

Table 2. Summary of mean and standard deviations for SRS DUO3 concentrations, assuming
a normal distribution ...................................................................................................... 2

Table 3. Radionuclide constituents of contaminated depleted uranium .......................................... 4

Table 4: Summary of available uranium and technetium data for the SRS DU .............................. 9

Table 5: Summary of probability distributions of mean activity concentrations (pCi/g of DU
waste) for uranium and technetium ............................................................................. 10

Table 6: Summary of probability distributions for mean activity concentrations (pCi/g of DU
waste) for other radioisotopes. (Source: SRS-2002.) .................................................. 11

Table 7: Summary statistics for the uranium activity% data ......................................................... 14

Table 8: Partitioning Ratios for Uranium Isotopes ........................................................................ 15

Table 9: Summary statistics for Technetium data (concentration in pCi/g of DU waste) ............. 18

Table 10: Categorization of Paducah Cylinders Using Cylinder History Cards (reproduced
from Table 1 in Henson, 2006) .................................................................................... 27

Table 11: Inputs for the Simulation of the Fraction of Contaminated GDP Cylinders ................. 27

Table 12. Uranium isotopic abundances by mass spectrometry, atomic percent, including
replicates (data summarized in Table 16, Beals, et al. 2002) ...................................... 31

Table 13. Uranium isotopic abundances by alpha spectrometry (as percent of total uranium
activity) (Table 17, Beals, et al. 2002) and Technetium concentrations in the SRS-
2002 data (Beals, et al. 2002) ...................................................................................... 32

Table 14. January 2010 EnergySolutions Data Analyzed by GEL (GEL 2010a and 2010b) ........ 33

Table 15. April 2010 EnergySolutions Data Analyzed by GEL (GEL 2010c) ............................. 34

Table 16. Technetium-99 concentrations collected by State of Utah, (Johnson, 2010) ................ 35

Table 17. Concentration data for other radioisotopes, SRS-2002. (Beals, et al. 2002) ................. 37

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 1
1.0 Waste Inventory Parameters Summary
This section is a brief summary of parameters and distributions employed in the waste inventory
component of the Clive Depleted Uranium (DU) Performance Assessment (PA) Model that is the
subject of this white paper.
For distributions, the following notation is used:
• Beta( µ, σ, min, max ) represents a generalized beta distribution with mean µ, standard
deviation σ, minimum min, and maximum max.
A summary of values and distributions for waste inventory modeling inputs is provided in
Table 1.
Table 1. Summary input parameter values and distributions
parameter value or distribution units comments
Number of SRS DU drums 5,408 — see Section 2.2.1
Mass of a 208-L (55-gal) drum 20 kg see Section 2.2.1
Total mass of SRS DUO3
(including drums) proposed for
disposal at Clive
3,577 Mg see Section 3.1.1
Number of DUF6 cylinders from
Paducah GDP 36,191 — see Section 3.5.1
Number of DUF6 cylinders from
Portsmouth GDP 16,109 — see Section 3.5.1
Number of DUF6 cylinders from
K-25 GDP 4,822 — see Section 3.5.1
Mass of DUF6 from Paducah
GDP 436,400 Mg see Section 3.5.1
Mass of DUF6 from Portsmouth
GDP 195,800 Mg see Section 3.5.1
Mass of DUF6 from K-25 GDP 54,300 Mg see Section 3.5.1
Diameter of cylinders 4 ft see Section 2.3.1
Length of cylinders 12 ft see Section 2.3.1
Fraction of GDP DU that is
contaminated Beta( 0.0392, 0.0025, 0, 1 ) — see Section 3.5.3.3
Number of DUF6 cylinders
disposed of in the Federal Cell 48628 see Section 3.5.2

Mean and standard deviation values for uranium isotopes and other fission products in the DU
trioxide (UO3) from the Savannah River Site (SRS) are developed in Section 3. These
concentrations are summarized in Table 2. Note that the standard deviations are those used in the
GoldSim PA model. They are intended to be estimates of the standard deviation of the mean
concentration, hence addressing the spatio-temporal scale of the input distribution.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 2
Table 2. Summary of mean and standard deviations for SRS DUO3 concentrations,
assuming a normal distribution
SRS DUO3 concentration
radionuclide mean
(pCi/g of DU waste)
standard deviation
(pCi/g of DU waste)
90Sr 4.70E+1 1.28E+1
99Tc 2.38E+4 1.10E+4
129I 1.86E+1 1.59E+0
137Cs 1.21E+1 7.10E-1
210Pb 0 0
222Rn 0 0
226Ra 3.17E+2 1.91E+1
228Ra 0 0
227Ac 0 0
228Th 0 0
229Th 0 0
230Th 0 0
232Th 0 0
231Pa 0 0
232U 0 0
233U 5.29E+3 4.78E+2
234U 3.31E+4 2.17E+3
235U 2.97E+3 7.50E+2
236U 4.91E+3 1.17E+3
238U 2.72E+5 6.64E+3
237Np 5.68E+0 1.17E+0
238Pu 2.10E-1 4.00E-2
239Pu 1.28E+0 2.00E-1
240Pu 3.40E-1 5.00E-2
241Pu 4.04E+0 7.40E-1
242Pu 0 0
241Am 1.42E+1 9.10E-1

The DU inventories from the gaseous diffusion plants (GDPs) are based upon estimates from the
DOE (DOE 2004a and 2004b) for mass of DUF6 and U3O8 produced. The inventories for the
other actinides and fission products is highly uncertain, but is informed to some extent by studies
performed by Oak Ridge National Laboratory (ORNL 2000a, 2000b, 2000c, 2000d), and reports
written by Bechtel Jacobs Company, LLC (BJC, 2000a, 2000b, 2000c). However, these studies
and reports do not provide specific information on concentrations that can be used directly to
develop input probability distributions. Until adequate information concerning DU inventory is
received from the GDPs, which may not happen until the DU oxide product has been produced
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 3
and sampled, the actinides and fission products are assumed to be in relative concentrations in
the DUF6 waste equal to those in the SRS DUO3 waste, as shown in Table 2. This is only a rough
approximation and will need to be revised as data from the GDP waste are made available.
Note that the amounts of transuranic materials within these actinides and fission products are
significantly less than the 10 nCi/g limit on transuranic nuclides required by the U.S. Nuclear
Regulatory Commission (NRC) in 10 CFR 61.55 (CFR 2014), and as limited by the Northwest
Interstate Compact on Low-level Radioactive Waste Management
(http://www.ecy.wa.gov/nwic/resolution_3.pdf). For example, the mean concentrations of 237Np
and 241Am are 0.0057 and 0.014 nCi/g, respectively (see Table 2). The highest concentration of
any of the Pu isotopes is 0.040 nCi/g for 241Pu (see Table 2).
2.0 Uranium Oxide Inventory
This document describes three categories of depleted uranium waste form at the Clive, Utah
disposal facility:
1. Depleted uranium oxide (UO3) waste from the Savannah River Site (SRS) proposed for
disposal at the Clive facility,
2. DU from the GDPs at Portsmouth, Ohio and Paducah, Kentucky, which exists in two
principal populations:
a) DU contaminated with fission and activation products from reactor returns introduced to the
diffusion cascades, and
b) DU consisting of only “clean” uranium, with no such contamination.
The DU oxides that are to be produced at these sites’ “deconversion” plants will be primarily
U3O8. The remainder of this section provides background on the uranium cycle and origins and
nature of DU waste in particular.
2.1 Depleted Uranium
In order to produce suitable fuel for nuclear reactors and/or weapons, uranium has to be enriched
in the fissionable 235U isotope. Uranium enrichment in the US began during the Manhattan
Project in World War II. Enrichment for civilian and military uses continued after the war under
the U.S. Atomic Energy Commission, and its successor agencies, including the DOE.
The uranium fuel cycle begins by extracting and milling natural uranium ore to produce "yellow
cake," a varying mixture of uranium oxides. Low-grade natural ores contain about 0.05 to 0.3%
by weight of uranium oxide while high-grade natural ores can contain up to 70% by weight
uranium oxide (NRC, 2010). Naturally occurring uranium contains three isotopes, 238U, 235U, and
234U. Each isotope has the same chemical properties, but they differ in radiological properties.
Naturally occurring U has an isotopic composition of about 99.2739±.0007% 238U,
0.7204±.0007% 235U, and 0.0057±.0002% 234U (Rich et al., 1988).
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 4
The milled ore is refined to remove the decay products (226Ra, 230Th, etc.) that have built up in
the material naturally to the degree of secular equilibrium, leaving more or less pure uranium
oxide. This uranium, still at natural isotopic abundances, is enriched to obtain the 235U, with vast
quantities of 238U as a by-product. Although a variety of technologies exist for enrichment, the
most prevalent enrichment process at the time was by gaseous diffusion, which requires that the
uranium be converted to a gaseous form: uranium hexafluoride (UF6). This gas is introduced to a
diffusion cascade, which separates the isotopes, generating enriched uranium as a product, and
depleted uranium hexafluoride (DUF6) as a waste stream. Depleted uranium isotopic ratio values
from gaseous diffusion plants are roughly 99.75% 238U, 0.25% 235U, and 0.0005% 234U (Rich, et
al., 1988), but the 235U assay found in the cylinders today varies with fluctuating enrichment
goals, operational conditions, and where in the cascade process the DU was removed. Because
processing of uranium has been practiced for only about 60 years, there has not been sufficient
time for appreciable in-growth of decay products in this by-product. Depleted uranium is
therefore considerably less radioactive than natural uranium because it has less 234U and other
decay products per unit mass. The bulk of this material is still stored in the original cylinders in
which it was first collected at the GDPs.
Uncontaminated (clean) depleted uranium consists principally of three isotopes of uranium (238U, 235U, and 234U) and a small amount of progeny from radioactive decay of these isotopes. Trace
amounts of other uranium isotopes (232U, 233U, and 236U) may also exist. The bulk of the DU at
the GDPs is clean uranium, but a significant amount of contaminated DU also exists, both at the
GDPs and in all the DU waste from the SRS.
The contamination problem arises from the past practice of introducing irradiated nuclear
materials (reactor returns) into the isotopic separations process. Irradiated nuclear fuel underwent
a chemical separation process to remove the plutonium for use in nuclear weapons. Uranium,
then thought to be a rare substance, was also separated out, but contained some residual
contamination from activation and fission products. This uranium was again converted to UF6 for
re-enrichment, and was introduced to the gaseous diffusion cascades, contaminating them and
the storage cylinders as well. Based on laboratory analysis of the contents of contaminated DU
waste (including all radionuclides in the containers), the species in the disposed inventory
include those in Table 3 (Beals, et al. 2002, EnergySolutions 2009b, and ORNL 2000c).
Table 3. Radionuclide constituents of contaminated depleted uranium
category radionuclides
uranium isotopes 232U, 233U, 234U, 235U, 236U, 238U
decay products 226Ra
activation products 241Am, 237Np, 238Pu, 239Pu, 240Pu, 241Pu, 242Pu
fission products 90Sr, 99Tc, 129I, 137Cs

In order to clarify that the contaminated DU wastes contain more than just uranium or DU, they
are termed “DU waste”. When this term is used, it refers to wastes that contain DU and a perhaps
small but potentially significant amount of contamination from actinides and fission products.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 5
2.2 Savannah River Site Depleted Uranium
Depleted uranium was generated at the SRS as a byproduct of the nuclear material production
programs (Fussell and McWhorter, 2002). Depleted uranium billets were produced at the DOE
Fernald, Ohio, site, fabricated into targets at SRS, then irradiated in one of the SRS production
reactors. The irradiated targets were transported to F-Canyon where the targets were dissolved.
After dissolution, the fission products were separated from the plutonium and uranium which
were then separated from each other. After additional purification, the uranium stream was
transferred to the FA-Line Facility where it was processed into uranium trioxide (UO3) for
storage in about 36,000 drums (Fussell and McWhorter, 2002). Since the chemical separations
process is imperfect, the DUO3 contains trace quantities of fission products and transuranic
elements (Beals et al, 2002, EnergySolutions, 2009b) as discussed above.
2.2.1 Mass of SRS Depleted Uranium Proposed for Disposal
The SRS DUO3 is a solid powder at room temperature and pressures. This DU oxide is stored in
208-L (55-gal) steel drums, with plastic liners. Steel drums have a tare mass of about 20 kg each.
The drums are approximately 2/3 full with an average mass of about 1500 lbm (750 kg) apiece
(Fussell and McWhorter, 2002). This DUO3 is considered to be relatively homogeneous, based
on known process controls and operations. The condition of the drums varies from good to poor
with a high percentage of the drums having some degree of outer surface corrosion.
In December 2009, SRS made a shipment of drums to the Clive, Utah facility. This shipment
contained 52 rail-cars (referred to as gondolas in the manifests), each holding 104 drums, for a
total of 5,408 drums. This shipment of DU waste is considered in this PA.
2.2.2 Composition of SRS Depleted Uranium
There are three main sources of data for establishing the concentration of uranium isotopes,
fission products, and transuranics in the SRS DU. In 2002 SRS sampled and analyzed their DU
oxide in preparation for shipment to Utah (Beals, et al., 2002). A total of 33 drums were
sampled; this is approximately 1% of 3300 drums that were available for sampling. The samples
were analyzed at the Savannah River Technology Center (SRTC) and by a Utah certified
laboratory (BWXT Services, Inc) for uranium, fission, and transuranic radionuclides. The
analytical results from SRTC are presented in Beals et al, 2002, and in an EnergySolutions
Radioactive Waste Profile Record, referred to here as the 2002 Waste Profile Record
(EnergySolutions, 2009b).
The 2002 Waste Profile Record (EnergySolutions, 2009b) provides activity concentration data
for isotopes of uranium and for potential contaminants such as 99Tc. The latter are used to
characterize the contaminant radionuclides for the PA (see Section 3). The data for uranium
isotopes are in the form of both activity concentration by alpha spectrometry, and atomic percent
by mass spectrometry. 233U was not detected by mass spectrometry. The alpha spectrometry, also
used to characterize the samples, cannot differentiate between 233U and 234U (or 235U and 236U)
thereby requiring the mass spectrometry analysis. Note, the 235U and 236U results are also based
on mass spectrometry analysis.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 6
The 33 samples were characterized for uranium isotopes, fission products, transuranics, and
some metals and organic compounds (pesticides, herbicides, semi-volatile and volatile organic
compounds) as recorded in the Waste Profile Record (EnergySolutions, 2009b). No organic
compounds were detected but low levels (mg/kg) of lead, arsenic, cadmium, chromium,
selenium, silver, zinc and copper were found. These low levels of metal make up less than 5 ppm
of the DU, and are not considered in this PA because they are not radioactive, and they are not in
excess of minimum regulated concentrations for hazardous waste (i.e., the DU waste is not
classified as “mixed waste”).
Data for other characteristics of the DU waste are also available from the 52 Waste Manifests
(EnergySolutions, 2009d). The shipment consisted of 52 gondola railroad cars, each car
containing 104 drums. The 2009 Waste Manifests from that shipment provide the volume (total
1,133.2 m3) and weight (total of 7,886,738 pounds, corresponding to a mass of 3,577 Mg). This
weight was calculated from information provided on the Uniform Low-Level Radioactive Waste
Manifest – Forms 540 and 541. On these forms, the material description (Form 540, box 11) is
listed as “RQ, UN 3221, Radioactive material, low specific activity (LSA-II), 7, Fissile
Excepted.” In the Radiological Description (Form 541, box 15) uranium component is described
as “U-(dep).”
The mass of the empty drums is assumed to be approximately 108 Mg, so the total waste mass is
3577 Mg of drummed waste - 108 Mg drum mass = 3469 Mg of DU waste which is a mix of
uranium isotopes and contaminants, and where the uranium is assumed to be in the form of
DUO3.
Based on the physical properties description in the Waste Profile Record (EnergySolutions,
2009b), the DU is stoichiometrically 83.22% uranium, indicating that the DU is essentially 100%
UO3. The isotopic mass percent of 238U is over 99%.
Since the arrival in Clive of the 52 gondolas of SRS DU waste, EnergySolutions has performed
two separate sampling and analysis events. In January of 2010, EnergySolutions collected 11
samples that were analyzed for uranium isotopes (Table 14, in the Appendix). In April 2010
EnergySolutions collected 15 samples that were analyzed for uranium isotopes and 99Tc (Table
15, in the Appendix). In August of 2010 the State of Utah analyzed 173 samples that
EnergySolutions collected from the drums (Johnson, 2010). These samples were analyzed for 99Tc only. The data are described in greater detail in Section 3, in which input distributions for
the GoldSim PA model are developed.
2.3 Depleted Uranium Oxide from the Gaseous Diffusion Plants
Three large GDPs were constructed to produce enriched uranium. The first diffusion cascades
were built in Oak Ridge, Tennessee, at what was the K-25 Site, but is now known as the East
Tennessee Technology Park (ETTP). Two others of similar design were constructed in Paducah,
Kentucky (PGDP), and Portsmouth, Ohio (PORTS) (DOE 2004a and 2004b). The cascades at
the K-25 Site ceased operations in 1985, the Portsmouth plant ceased in 2001, the Paducah GDP
continues to operate. The two more recent GDPs are host to a large inventory of stored DUF6,
including the ETTP material that was moved to Portsmouth.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 7
The DOE is currently managing approximately 60,000 cylinders at both PGDP and PORTS
(DOE 2004a, 2004b). For many years, interest has been expressed in converting the DUF6 in
these cylinders to an oxide form to support their long-term disposal. In May, 1995 an
independent DOE oversight board recommended a study to determine a suitable chemical form
for long-term storage of DU. Also, in 1994 the DOE began work on a Programmatic
Environmental Impact Statement for Alternative Strategies for the Long-Term Management and
Use of Depleted Uranium Hexafluoride (DOE 1999a). Later, DOE issued the Final Plan for the
Conversion of Depleted Uranium Hexafluoride as Required by Public Law 105-204 (DOE
1999b). As a result of these efforts the DOE developed a Conversion Plan that describes the
steps that would allow DOE to convert the DUF6 inventory to a more stable chemical form. Two
Environmental Impact Statements (EIS) were prepared as part of the plan, one for Paducah,
DOE/EIS-0359, (DOE 2004a) and one for Portsmouth, EIS-0360 (DOE 2004b). These EISs
describe the background and alternatives for DUF6 conversion. With the completions of the
EISs, “deconversion” plants were built at both the PORTS and PGDP locations. In 2002, DOE
awarded a contract to Uranium Disposition Service, LLC (UDS) to design, construct, and operate
two DUF6 deconversion facilities at these locations. As of this writing, both plants have been
built by UDS and have begun test processing DUF6 into oxide form.
The UDS dry conversion is a continuous process in which DUF6 is vaporized and converted to a
mixture of uranium oxides (primarily DU3O8 but with some UO2) by reaction with steam and
hydrogen in a fluidized-bed conversion unit. The hydrogen is generated using anhydrous
ammonia (NH3). Nitrogen is also used as an inert purging gas and is released to the atmosphere
through the building stack as part of the clean off-gas stream. The DU3O8 powder is collected
and packaged in the former DUF6 cylinders for disposition. The process equipment is arranged in
parallel lines. Each line consists of two autoclaves, two conversion units, a HF recovery system,
and process off-gas scrubbers (DOE 2004a).
2.3.1 Mass of GDP Depleted Uranium
According to the EISs the PGDP facility has been designed to convert approximately 18,000 Mg
(one Mg is one metric tonne, or about 2,200 lbm) of DUF6 per year, which will require
approximately 25 years for full conversion of the PGDP inventory. At Portsmouth, 13,500 Mg of
DUF6 per year (approximately 1,000 cylinders per year) is expected to be converted.
Several different cylinder types are in use. Most cylinders are expected to range from 11 to 12
Mg full. The cylinders with a 12-Mg capacity are 12 ft (3.7 m) long by 4 ft (1.2 m) in diameter;
most have a steel wall that is 5/16 in (0.79 cm) thick. Similar but slightly smaller cylinders with a
capacity of 9 Mg are also in use. Most of the cylinders were manufactured in accordance with an
American National Standards Institute standard (ANSI N14.1, Uranium Hexafluoride Packaging
for Transport) as specified in 49 CFR 173.420, the Federal regulations governing transport of
DUF6.
To develop an estimate for the mass of DU oxide from the two GDPs, the mass of DUF6 was
converted to mass of uranium and thence to mass of U3O8. This simple stoichiometric
conversion, based on moles of uranium, fluorine, and oxygen, is performed within the Clive DU
PA Model. Details are provided in Section 3.5.1.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 8
2.3.2 Composition of GDP Depleted Uranium
The depleted uranium oxides from Portsmouth and Paducah that are proposed for disposal have
yet to be manufactured. Until their production is complete, with associated testing of
composition, estimates of composition must be relied upon to construct distributions and make
decisions. At the most coarse level, there are two distinct populations of GDP DU composition:
1) DU derived from "clean" (a.k.a. "green") uranium, which contains no contamination, and 2)
contaminated DU, which contains varying amounts of fission and activation products, as well as
transuranics, resulting from the introduction of reactor returns into the gaseous diffusion cascade.
The clean DU is characterized by its abundance of uranium isotopes, and includes those
radionuclides as well as their decay products. Isotopic abundance analyses were focused on
determining the amount of U-235 in the DU, since this isotope was the "product" of the entire
enrichment enterprise, and little attention was given to the exact abundance of other uranium
isotopes, all of which were considered waste products.
Little information is available at this time regarding the exact nature and extent of the
contamination within the contaminated DU population. The uranium isotopic abundance
estimates are the same as for the clean DU. Estimates of the contamination by reactor return
radionuclides, however, must rely on the SRS DU as a proxy until better GDP-specific
information becomes available. For the purposes of this PA, then, the contaminated fraction of
the GDP DU is assumed to have the same contaminant composition as the SRS DU.
3.0 Input Parameter Distribution Development
The probabilistic Clive DU PA Model relies on stochastic parameters in order to evaluate
uncertainty and sensitivity. The statistical development of input parameter distributions is
provided here.
3.1 Parameters for Depleted Uranium from the Savannah River Site
Parameters of interest for the PA include the mass of DU waste, and the concentrations of each
radio-isotope contained in the DU waste. The contents of the SRS drums were described in
Section 2.1. The purpose of this section is to describe the characterization of the mass of DU, and
the concentrations of the radioisotopes. The mass of DU is considered fixed for the purpose of
this PA, and is presented without uncertainty. The concentrations are presented in terms of the
best estimate of the mean concentration, and the uncertainty of the mean concentration for each
radio-isotope.
3.1.1 Mass of SRS Depleted Uranium
The single source of information regarding the mass of total depleted uranium shipped from SRS
to Clive are shipping manifests (EnergySolutions, 2009d). Key pieces of information on these
forms include the following
• Total mass in kg and corresponding weight in US tons
• Total volume in cubic meters and in cubic feet
• Net waste volume in cubic meters and in cubic feet
• Net mass in kg and corresponding net weight in US tons
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 9
Reviewing these manifests suggest that each gondola rail car was weighed empty (tared) and
fully loaded, and the tare weight was subtracted to arrive at the “Net Waste Weight” reported on
the manifests. Since this is a measured amount, it will be considered a fixed value and a
distribution will not be assigned. There is no reason to believe that the mass of the drums was
deducted from this net weight. Such drums do not have a standardized tare weight, but for the
purposes of calculation it is assumed that each drum has a mass of 20 kg. This is considered a
representative weight for a 55-gallon drum.
The net weights from the manifests were summarized by W. Johns in a spreadsheet (“100105
9021-33 Iso With Calcs.xls”) sent to Neptune. These values have been summed to create a total
mass data value for total mass of the depleted uranium shipped from SRS to Clive, Utah. Masses
of DU plus drums for the individual 52 rail cars range from 50.37 Mg to 75.56 Mg. The total
amount shipped is 3,577 Mg.
3.1.2 Composition of SRS Depleted Uranium
Three data sources are available for the development of probability distributions for the
concentrations of radio-isotopes in the SRS DU waste: The SRS-2002 dataset consists of activity
concentration data and uranium isotopic abundance as atomic percent from Beals, et al. (2002).
The ES-2010 dataset has uranium activity concentration and total uranium mass concentrations
from two EnergySolutions sampling and analysis events: GEL (2010a and 2010b), and GEL
(2010c). Finally, the Utah-2010 analysis obtained activity concentrations of 99Tc from
EnergySolutions sampling and State of Utah requested analysis (Johnson 2010). These datasets
are briefly described in Table 4 and the individual values are presented in Appendix A.
Note that the 33 samples included in the SRS-2002 data also include concentrations of the other
contaminants presented in Table 3 (decay, activation and fission products), which are used to
developed input probability distributions for the concentrations of these radionuclides.
The spatio-temporal scale of interest for the Clive DU PA Model includes a large volume of DU
waste and fill material in the Class A South embankment, a 10 ky quantitative analysis followed
by a 2.1 My qualitative analysis. This, and the dynamic nature of the PA modeling environment
in which time steps of many years are used, affects the approach to characterizing probability
distributions of the inventory. Conceptually, the PA model incorporates compartments or cells
that are fully mixed at each time step.
The physical samples used in this statistical analysis represent very small volumes of waste, but
the mean concentrations are representative of the entire inventory. This approach is reasonable so
long as there is not a strong non-linear effect due to spatial variation within the waste cell. For
this model the waste is fully mixed within a waste layer. The appropriate spatio-temporal scaling
suggests that characterization of the mean activity concentration of each radionuclide is needed.
This is the basic approach that is taken in each case, however, because the data sources are
different for some of the radionuclides, different approaches are needed for estimation of the
probability distributions (Table 5 and Table 6):

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 10
Table 4: Summary of available uranium and technetium data for the SRS DU
Source Date Number Constituents Units1
SRS-2002: Table 16 of
Beals et al, (2002)2
2002 6 (2 replicates
per sample)
233U, 234U, 235U, 236U, 238U
Isotopic abundances
(atomic % U)
SRS-2002: Table 17 and
Table 4 of Beals et al,
(2002)
2002 33 233+234U, 235+236U,
238U, 99Tc
Activity % U
ES-2010 (GEL, 2010 a,b) January
2010
15 Total U, 233+234U, 235+236U, 238U
µg/g for Total U;
pCi/g for others
ES-2010 (GEL 2010 c) April
2010
11 Total U, 99Tc, 233+234U, 235+236U,
238U
µg/g for Total U;
pCi/g for others
Utah-2010 (Johnson,
2010)3
August
2010
173 (plus 30
duplicates)
99Tc pCi/g
1 Concentration units for the data are expressed in terms of activity per gram of DU waste.
2 Although these data are referenced to Beals et al (2002), the data used actually come from a Waste Profile Record
file that is labeled Waste Profile Record SRS DU 9021-33_r0.pdf. It is an EnergySolutions radioactive waste
profile record that is signed by a DOE representative. The DOE signature is dated November, 2009. It is clear in
this Waste Profile Record that the original 33 samples were used to characterize most radionuclides, and that
basically the same samples were used for the atom% data. However, Beals et al includes 7 samples with no
replicates, whereas the waste profile record includes only 6 of those 7 samples with replicates for 12 samples in all.
It is not clear why Sample #8 is missing from the atom% table (listed as Attachment 2 in the Waste Profile
Record), or why there are replicate results presented for each of the six samples that are included. This discrepancy
does not make a large difference to the input distribution development, but the 12 sample results were selected
instead of the 7 results in Beals et al because 12 results are assumed to provide more information.
3 Note that splits of these samples were also submitted for analysis by EnergySolutions.

Table 5: Summary of probability distributions of mean activity concentrations (pCi/g of
DU waste) for uranium and technetium
Radioisotope Mean Standard Error Source
99Tc 23,800 11,000 SRS-2002, ES-2010 (Jan), Utah-2010
233U* 5,290 478 ES 2010 (Jan/Apr)
234U* 33,100 2,170 ES 2010 (Jan.Apr)
235U* 2,970 750 ES 2010 (Jan.Apr)
236U* 4,910 1,170 ES 2010 (Jan.Apr)
238U 272,000 6,640 ES 2010 (Jan.Apr)
* Isotopes are partitioned using SRS-2002 atomic percentage data.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 11
Table 6: Summary of probability distributions for mean activity concentrations (pCi/g of
DU waste) for other radioisotopes. (Source: SRS-2002.)
Radioisotope N Mean Std. Error
241Am 33 14.2 0.91
137Cs 33 12.1 0.71
129I 33 18.6 1.59
237Np 33 5.68 1.17
238Pu 31* 0.21 0.04
239Pu 31* 1.28 0.20
240Pu 31* 0.34 0.05
241Pu 31* 4.04 0.74
226Ra 33 316.8 19.1
90Sr 33 47.0 12.8
* Note that results for plutonium isotopes were not reported for 2 samples in the SRS-2002 data.
• The probability distribution of mean activity concentration for uranium isotopes is
estimated from the ES-2010 data. Because activity from combinations of isotopes 233+234U and 235+236U is reported in ES-2010, the atomic percent data from SRS-2002 is
used to partition these isotopes.
• There are three sources of 99Tc data: SRS-2002, ES-2010, and Utah-2010. These datasets
are used to estimate mean 99Tc activity concentrations. Note that the duplicate
measurements in Utah-2010 were not used because there are many samples (173) without
the duplicates, and the duplicates were found to be dependent on their original samples
(separating out those dependencies statistically is complicated and unnecessary given the
large number of samples available).
• The SRS-2002 data provide the only data available for the other radionucludes
(americium, cesium, radon, iodine and plutonium). Consequently, these data are used to
estimate distributions of mean activity concentrations for these radionuclides. The
parameter estimates for the probability distributions of the mean activity concentrations
for these radionuclides are presented in Table 5.
Therefore, the approach for distribution development is to establish the uncertainty distribution
of the mean activity concentration for each radionuclide. Each individual data set available is
reasonably well-behaved statistically, not exhibiting large skew or multi-modality. There are also
enough data that the Central Limit Theorem can be applied, implying a normal distribution for
the distribution of the mean. The normal distributions are characterized with the mean
concentration and the standard error (i.e., the standard deviation of the mean).
While available site knowledge and historical information suggest that the SRS waste is from
similar processes and is similar in composition, the sampling events were treated as if they were
sampling different populations. The results from different sampling events for 99Tc and U form
clusters, the lack of information suggesting other reasons for these clusters indicate potentially
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 12
different sampling and analysis methods between sampling events. Consequently, for 99Tc and
uranium isotopes, bootstrap re-sampling of the samples and the sampling events is used to
address possible differences between sampling events. For the remaining radionuclides, the SRS
data are used directly to estimate the parameters. The final distributions are presented in Table 5
and Table 6. Details of the development of these distributions are in the following sections.
3.2 Analysis of Uranium Composition in SRS Depleted Uranium
Direct comparison between uranium concentrations represented in the SRS-2002 data and in the
ES-2010 data is complicated by several factors. The ES-2010 data represent activity
concentrations for uranium, where the SRS-2002 data represent isotopic abundance as activity
percent (%) of uranium, rather than activity concentration. These different expressions of
uranium activity cannot be reconciled without recourse to the total proportion of uranium in each
sample—information that is not available. Further, the pedigree of the SRS-2002 data is not
clear. Information is available in Beals et al. (2002) about the analytical methods performed in
the laboratory, but the actual laboratory reports for the SRS-2002 data are not available. In
contrast, the pedigree of the ES-2010 data is well known, and the laboratory reports are available
to support the reported uranium activity concentrations. Consequently, only the ES-2010 data are
used to generate distributions of the mean uranium activity concentration for each uranium
isotope. However, an exploratory comparison is made between the SRS-2002 and the ES-2010
activity data to understand the differences between the SRS and ES uranium data. Development
of input probability distributions is presented after the exploratory comparison.
For the PA model, separation is also needed for the uranium isotopes in the pairs 233+234U and 235+236U. The ES-2010 laboratory analysis and subsequent uranium data do not distinguish
between these pairs of isotopes, but report 233+234U and 235+236U activity concentrations
combined. However, the SRS-2002 study also includes some uranium isotopic abundance data
presented as atomic percent (%) for all uranium isotopes. These SRS-2002 atomic% data are
used to partition the 233+234U and 235+236U activity concentration data obtained from ES-2010.
3.2.1 Exploratory Comparison of Uranium Data
In SRS-2002, activity% for all uranium isotopes was measured at SRS using alpha spectrometry.
In ES-2010, activity concentrations (pCi/g) were measured for 233+234U, 235+236U, and 238U. As
noted above, only the ES-2010 data will be used to develop input distributions for uranium
concentrations for the PA model. However, a comparison of the ES-2010 and SRS-2002 data is
presented to better understand the limitations of the SRS-2002 data, and to support the
contention that the ES-2010 data are more appropriate for use in developing input distributions
for uranium activity concentrations for the PA model.
A major consideration in the decision to focus on the ES-2010 for development of input
distributions for the PA model is the lack of supporting documentation for the SRS-2002 data
and the difficulty of converting from data presented in activity% to activity concentration. The
ES-2010 and SRS-2002 data are compared by first translating one of the datasets to the units of
the other dataset. The approach taken is to convert the ES-2010 data to activity%. This is a
relatively simple step that facilitates comparison of the SRS-2002 and ES-2010 datasets.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 13
Activity% can be calculated directly from activity concentrations (Equation 1).
100×∑j
iic
c=A (1)
where
Ai = activity% of uranium component i,
ci = activity concentration for uranium component of interest i, and
cj = activity concentration for all enumerated uranium components j, which indexes
233+234U, 235+236U, and 238U.
The results of this conversion are presented graphically in Figure 1. This figure shows pairs of
scatter plots for the different uranium components. These plots show clear difference between
the datasets. For example, there is a cluster of points from the SRS-2002 dataset (circles). As
originally ordered and labeled in Beals et al. (2002), the first 21 samples form the close cluster of
points while the last 12 points form the more dispersed cluster of points. Without any further
information, this is suggestive of either sampling or laboratory differences or biases within the
SRS-2002 data. Sample IDs could be surrogates for sample location, perhaps representing
samples from barrels of similar wastes, which would be an example of a potential sampling bias
if the entire waste stream is not relatively homogeneous. Alternatively, the samples could have
been analyzed in separate batches on different days—with different ambient background
concentrations being subtracted from each batch—which would be an example of laboratory
bias. No information has been found to explain these differences, but this provides further
evidence for why these data are not included in the development of the probability distributions
for uranium isotopic inventory for the Clive DU PA.
Data from the two 2010 ES sampling events form clusters that are different but with some
overlap. The data from the ES-2010-January sampling event have greater standard deviation than
those from the April sampling event. The 235+236U data tend to be slightly greater for the January
sampling event, whereas the 238U data tend to be slightly greater for the April sampling event.
The greatest overall difference is between the first cluster (21 samples) from SRS-2002 and the
rest of the data. This cluster has markedly lower 233+234U activity% values than the remainder of
the data, and, consequently, markedly greater 238U activity% values.
The summary statistics for each dataset in Figure 1 are presented in Table 7. They further
demonstrate the differences between the datasets. The questionable pedigree and difference
between the two clusters in the SRS-2002 data are sufficient to justify not using these data for
distribution development for the PA. The differences, particular in standard deviation, between
the two ES datasets suggest that these two datasets should not be combined when estimating
input probability distributions for the uranium activity concentrations for the PA model.
The next stage in this exploratory analysis of the SRS-2002 and ES-2010 Uranium data is to
convert the SRS-2002 data from activity% to activity concentrations. This is done to see if the
same basic results are obtained, considering different inputs are needed for this conversion.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 14
Table 7: Summary statistics for the uranium activity% data
Radioisotope SRS-2002 (33 samples) ES-2010-January (11) ES-2010-April (15)
Mean Std. Dev. Mean Std. Dev. Mean Std. Dev.
233+234U 8.0% 1.8% 11.6% 1.1% 12.4% 2.0%
235+236U 2.0% 0.3% 1.7% 0.4% 3.2% 1.2%
238U 90.0% 2.0% 86.7% 1.1% 84.4% 2.3%

Figure 1. Comparison of activity percent for the SRS DU uranium isotopes

3.2.2 Partitioning 233+234U and 235+236U
The Clive DU PA Model requires probability distributions of activity concentration for each
uranium isotope. Because of the methods used to measure radioactivity, most samples collected
in 2002 and in 2010 do not distinguish between 233U and 234U or between 235U and 236U, but
rather report combined quantities. To separate the isotopes, some data on the relative
contributions of each isotope in each pair is needed.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 15
From the SRS-2002 data, 6 samples were analyzed using mass spectrometry. These 6 samples
are from the original 33 samples that were analyzed for activity% of uranium. The mass
spectrometry method identified all uranium isotopic abundances and the results are expressed as
atomic% (see Table 12 in the Appendix). The dataset provides two values for each sample.
These values are treated as duplicates and the values are averaged for use in subsequent analyses.
All abundance values for 233U are reported as 0.0000%, because it was not identified in any
sample. However, to allow for the possibility of a trace quantity of 233U in the SRS DU, for both
SRS-2002 and ES-2010 datasets, 233U atomic percentage values are assumed to be 0.00005%, a
value that was chosen because any value smaller than that would be recorded as 0.0000% to four
decimal places. This essentially treats the values as non-detects, and allows for very small values
that would have been rounded to zero. This is conservative with respect to the possible
abundance of 233U.
To partition activity% and activity concentrations for 235+236U and 233+234U, uranium abundances
expressed as atomic% are multiplied by their respective specific activities, and renormalized to
calculate activity%. Ratios are presented in Table 8. The atomic% data do not sum to exactly
100%, hence the renormalization causes small differences in the 233U activity% values.
Table 8: Partitioning Ratios for Uranium Isotopes
Radionuclide Ratios
Sample 233U 234U 235U 236U 234U/233U 236U/235U
3 1.29% 7.03% 0.73% 1.12% 5.45 1.54
9 1.29% 6.73% 0.73% 1.11% 5.20 1.53
17 1.29% 7.13% 0.73% 1.14% 5.54 1.56
20 1.28% 7.33% 0.74% 1.16% 5.70 1.58
25 1.22% 11.50% 0.83% 1.56% 9.43 1.89
30 1.25% 9.80% 0.78% 1.46% 7.87 1.86

Both sets of ratios show similar patterns, clearly demonstrating that the last two samples are
different than the first four samples. This also matches the differences observed in the activity%
data reported in the 33 samples, for which the first 21 samples are clearly different than the last
12 samples (see Figure 1). However, all six samples are used to separate these isotopes for the
PA model, the effect of which is to increase the variance of the ratios, which introduces more
uncertainty in the PA model. In general, the differences this causes in uranium activity
concentrations are fairly small relative to the likely effect on the PA model results, however, this
will be tested in the model evaluation and sensitivity analysis. If the uranium isotopic
distributions prove to be sensitive in the PA model, then it might be necessary to collect data that
are aimed more specifically at the needs of the PA.
3.2.3 SRS Depleted Uranium Activity Concentration
As illustrated in Figure 1, there are differences between concentrations measured by ES in the
January and April, 2010, data. (Note, as described in Section 3.2.1, the SRS-2002 uranium data
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 16
are not included in the development of input distributions for uranium activity concentrations for
the PA model.) The focus is on the ES-201 datasets. The data from these two ES-2010 dataset
are not considered independent or exchangeable, in which case they cannot be directly
combined. Consequently, in order to estimate the population mean and the standard deviation of
the mean, a bootstrap method is used giving equal weight to both ES-2010 sampling events.
To simulate the two sampling events, all combinations of the ES-2010 January and April
sampling events were used. The samples are bootstrapped within each sampling event, the mean
value is calculated for each study, and the study means are averaged to obtain an overall mean
value. The bootstrap method is applied as follows:
1. The two sampling events are selected with replacement. Since there are only 4 possible
combinations of sampling events (select the January event twice, select the April event
twice, select the January event followed by the April event, and select the April event
followed by the January event – this is analogous to the results that could be obtained by
tossing a coin twice), all combinations are used and weighted equally.
2. For each sampling event selected, the data are sampled with replacement and a mean
calculated. An overall mean is calculated as an average of the two means.
3. This simulation is repeated 10,000 times for each of the 4 sampling event combinations,
to construct a distribution of means. The simulations were selected at random. This large
number of simulations provided adequate convergence of the distribution of the mean.
The effect of this approach is that the effective sample size is related more to the two sampling
events than to the 26 samples. This leads to a comparatively wide distribution. If instead, all 26
samples had been treated as independent, then the standard deviation would be considerably
smaller. The conceptual difference between the two possible approaches is that treating the data
as independent assigns the information content, or uncertainty, to each sample, whereas, the
approach used assigns the information content to the sampling events. That is, the sampling
events themselves are considered more important for characterizing the distribution of the
uranium isotopes than the individual sample results.
10,000 bootstrap samples are used, to create the distributions of mean values for each uranium
component shown in Figure 2. The distributions for the uranium components are presented to
show how the distributions relate to the two ES-2010 datasets. The red lines on the plots show
how the April data exhibit greater activity concentrations for all three uranium components. The
plots also show how the distributions bound the means of the two datasets for all three uranium
components. If an approach had been taken that treated all 26 data points as independent, then
the distributions of the means would probably have fallen between the two means.
The distributions of the uranium components 233+234U and 235+236U are partitioned using a
randomly assigned ratio from one of the 6 ratios presented in Table 8. That is, each of the 10,000
simulated means is partitioned, so that there are 10,000 realizations of the distributions of the
individual uranium isotopes. The resulting distributions of the mean uranium isotope activity
concentrations were fit using a normal distribution. The resulting distributions are presented in
Table 5.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 17
The activity concentrations of uranium are dominated by 238U at an average of 272,000 pCi/g.
This is to be expected, although the mean activity concentration of 234U is also large compared to
the other isotopes. These distributions could be narrowed (i.e., reduced uncertainty) by collecting
new data under an experimental design that is aimed at the needs of the PA. This includes
activity concentrations over a wide range of drums, locations in drums, and laboratory analysis
that provides activity concentrations for every uranium isotope. The ES-2010 datasets provide
reasonable data, but the two datasets present different mean uranium activity concentrations, in
which case there would be benefit from a more complete study of uranium in the SRS DU waste.
If, given these relatively broad distributions, the uranium isotopes are not sensitive to any PA
model endpoint, then the need to refine these distributions will be less.
Mean concentration from each input data set are denoted by vertical red lines. To compare with
original ES data, mean concentrations of 233+234U, 235+236U and 238U components are shown (red
lines) for both the ES-2010 January and April datasets.

Figure 2. Distribution of mean activity concentration values from bootstrap resampling.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 18
3.3 Analysis of Technetium Concentrations in SRS DU
Technetium-99 is the most important of the contaminants contained in the SRS DU waste,
because of its potential for relatively fast transport to groundwater. Other mobile radionuclides
were reported as not detected in the SRS-2002 samples.
Three sources of data exist for 99Tc from the following sampling events: SRS-2002 (33 samples),
ES-2010 (11 samples), and Utah-2010 (173 samples – without duplicates). Figure 3 shows that
the samples from these three sampling events have different mean concentrations and different
standard deviations. The original SRS-2002 data show the greatest concentrations.
EnergySolutions attempted to verify these concentrations in January 2011. However, the ES-
2010 99Tc showed lower concentrations. Given the uncertainty and importance of understanding
the 99Tc concentrations, the State of Utah commissioned a study involving sampling and analysis
of 99Tc for 173 samples (Johnson, 2010). However, these exhibited lower concentrations again.
The boxplots shown in Figures 3 and 5 are standard typical boxplots (Tukey, 1977) used to
illustrate and summarize the distribution of groups of data. The top, middle and bottom lines
indicate the 75th, 50th (median) and 25th percentile of the data. The vertical lines “whiskers”
extend to the largest or smallest point within 1.5 times the interquartile range (75th – 25th
percentiles) of the 25th and 75th percentiles. Results falling outside the whiskers are considered to
be outliers. This indicates that there is a reasonable chance they are from a different distribution.
With several groups of data, boxplots can be used to informally compare the central values
(median), spread or variances (width of the boxes) or distributions (symmetry).
Table 9: Summary statistics for Technetium data (concentration in pCi/g of DU waste)
Data Source
Statistic SRS-2002 ES-2010 (January) Utah-2010
Number of Samples 33 11 173
Mean 49,370 17,800 4,340
Standard Deviation 29,260 5,910 3,550

The pattern of 99Tc concentrations in the SRS-2002 data is similar to the pattern seen in the
uranium data. That is, the concentrations are considerably greater in the last 12 samples
(particularly in the last 9 samples) than in the first 21 samples, by sample ID (see Table 13). This
could be reason to exclude the SRS-2002 99Tc data from the distribution development. The data
do not seem to come from one population, possibly because of sampling or laboratory
differences or biases, and the pedigree of the data is lacking because there are no laboratory
reports available for the data. However, these data have been included because they show greater
concentrations than the two datasets from 2010, which causes the developed distribution of 99Tc
concentrations to extend out to cover the SRS-2002 data. The effect of the inclusion of these data
has been tested during model evaluation and is reported as part of the sensitivity analysis. If, as
might be expected, the 99Tc concentrations are a sensitive part of the model, then it might
warrant reconsideration of the available data.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 19
Figure 3. Tc-99 Activity Concentration. Sample sizes: SRS-2002 = 33; ES-2010 = 11; Utah-
2010 = 173.
Of further concern is the difference between the ES-2010 data and the Utah-2010 data. These
data were collected less than a year apart, and several of the samples from the Utah-2010 data
were from the same drums used for the ES-2010 samples. The only clear difference between the
two datasets is that different analytical laboratories were used in each case. The ES-2010
samples were analyzed by GEL Laboratories. The Utah-2010 samples were analyzed at a
different laboratory. It is possible that the differences are analytical
As a consequence of the differences in 99Tc concentrations between the different sampling
events, the approach taken to development of an input distribution of mean 99Tc concentrations is
similar to the one used for uranium. That is, it is considered more important to model the
information content in the sampling events rather than each individual sample. This approach
reduces the effect of the Utah-2010 data, which would otherwise dominate estimation of the
input distribution.
A simple approach to distribution development is to treat each measurement across all three
sampling events as independent and identically distributed and calculate the mean and standard
error using all the data. However, this approach weights the data based on the number of
samples, giving the Utah-2010 data the most influence. Further, to the extent that the data within
each study are not independent, the standard error would be artificially small. The individual data
points might not be independent because analyses were often performed on samples from the
same drum. To address these issues, a bootstrap method was developed and used to estimate the
distribution of the mean 99Tc value that treats the three datasets as independent, rather than each
data point across sampling events.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 20
Note that the Utah-2010 dataset contains 18 laboratory and 12 field duplicate measurements.
These data were examined and found to be correlated with the associated primary samples. Since
these measurements cannot be considered independent and a relatively large number of samples
(173) were analyzed, the duplicates are not included in this distribution analysis.
The three datasets are treated independently in the bootstrap approach, which leads to a wide
distribution that covers the range of all three datasets combined. The more simple approach of
treating each data point as independent across the three sampling events would result in a very
narrow distribution, because of the large number of data points, and the center of the distribution
would be lower because the Utah-2010 dataset would dominate given the large sample size.
The bootstrap method is applied as follows:
1. The three studies are selected with replacement from the three available sources of 99Tc
data (SRS-2002, ES-2010 and Utah-2010). Since there are only 27 possible combinations
of sampling events, all combinations were used and weighted equally.
2. For each study, the data are sampled with replacement and a study mean calculated. An
overall mean is calculated as an average of the three study means.
3. This simulation is repeated 10,000 times for each of the 27 study combinations, to
construct a distribution of the estimated mean concentrations for 99Tc.
The density plot describes the distribution of the overall mean (Figure 4). Because of smoothing
in the plotting algorithm, the distribution appears to include negative values, however, the
smallest value from the simulations is 3,800 pCi/g.
This distribution is reasonably described by a normal distribution, which is used in the PA model
(see Table 5). The mean of the distribution is 23,800 pCi/g, and the standard deviation is 11,000
pCi/g. In the PA model, the distribution is truncated at zero, so that negative mean concentrations
are not possible. Since this is a distribution of the mean concentration, this distribution indicates
that the mean concentration of 99Tc could be as low as zero, or greater than 60,000 pCi/g (see
Figure 4). This is a large range, and reflects the uncertainty in the three data sources because of
their differences. Different decisions regarding combination of the available data would almost
certainly lead to a narrower distribution of the mean concentration, given the large number of
data points available. For example, if the Utah-2010 data were used alone, then the 173 data
points would lead to a mean of about 4,340 pCi/g and a standard error of about 270 pCi/g, which
is the distribution that would then be used in the PA. That is, most of the distribution of the mean
concentration would fall between 3,800 pCi/g and 4,880 pCi/g. This is very different than the
distribution that is currently proposed for use in the PA.
Note in Figure 4 that the mean concentrations for the three data sources are also presented. These
show clearly that the distribution of the mean 99Tc concentration spans the means of the available
datasets. As noted above, if the mean 99Tc concentrations proves to be sensitive for any given
endpoint of the PA model (dose, groundwater concentrations, or deep time concentrations), then
the development of this input distribution should be revisited, including a re-examination of how
the three data sources have been combined.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 21

Figure 4. Distribution of Tc-99 mean values. Red lines indicate mean values of Utah-2010,
ES-2010 and SRS-2002 results. The dashed lines indicate the 5th and 95th percentiles of the
mean values of the resampled data.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 22
3.4 Concentrations of Other Radionuclides in the SRS Depleted
Uranium
As noted in Section 2.1, there are other potential contaminants in the SRS DU, including decay,
activation and fission products (see Table 3). Given the only source of data for these
radionuclides in SRS-2002, the concentrations are very low, and are unlikely to significantly
contribute to the PA, however, input distributions for the mean concentrations of each of these
radionuclides are developed and included in the PA to confirm that this is the case.
The measurement of other radionuclides is reported only in the SRS-2002 dataset. These include 241Am, 226Ra, 137Cs, 90Sr, 237Np, 238Pu, 239Pu, 240Pu, 241Pu and 129I. Distributions of these values
are shown in Figure 5. With the exception of the plutonium isotopes, all measurements were
below the detection limit. Non-detects were set to their detection limits for this analysis. This is a
conservative approach, which over-estimates the activity concentrations of these radionuclides.
However, the impact of these radionuclides on the PA is expected to be very small, in which case
use of the detection limits probably has insignificant effect on the concentrations and doses
output by the PA model1. The final distributions are presented in Table 6. The distributions are
assumed to be normal, and they are truncated at zero in the PA model.
3.5 Parameters for Depleted Uranium Oxide from the GDPs
The exact nature of the DU oxides that will be generated by the deconversion plants at
Portsmouth and Paducah will not be known until their production, so this PA relies on the best
information available to develop estimates. What is known is that the oxides will be primarily
U3O8, and that they will be shipped and disposed in used DUF6 cylinders, some of which will
contain residual contamination from reactor returns.

1 Note that iodine-129 was not detected in any of the 33 samples from SRS-2002. However, upon further research,
the lower limits of detection (LLDs) are likely to over-estimate the iodine-129 inventory by about five orders of
magnitude. Very small quantities of 129I might be expected given the presence of 99Tc, given that they are both
fission products. Using the ratio of 99Tc to 129I could provide a better path to a more reasonable estimate of 129I
concentrations. EPRI (2005) provides some information on acceptable knowledge, however, the EPRI reference
does not contain sufficient information and acknowledges that there are very few actual 129I measurement included
in the data. However, Cox (2014 – personal communication from Billy Cox, EPRI, to Paul Black, Neptune and
Company, Inc., April 2014) indicated that there is process knowledge that may be brought to bear: The equilibrium
burnup ratio for 99Tc to 129I is about 200:1. That is, in spent fuel, the activity of 99Tc is about 200 times the activity
of 129I. The first step of fuel reprocessing is to dissolve the fuel in nitric acid, in order to facilitate the wet chemistry
extraction of U, Pu, or other desirable constituents. In this process of dissolution in nitric acid, about 99% of the
iodine is volatilized, and none of the technetium is volatilized. This alters the ratio of 99Tc to 129I by another factor of
about 100. Once the acid has been neutralized in preparation for other processes, including whatever processes were
used to bring the contaminated reactor return uranium to its current form as UO3 powder, this ratio of 100×200:1, or
about 20,000:1, is maintained. If we take advantage of this process knowledge, then, the activity concentration of 129I can be estimated as 0.00005 times the activity concentration of 99Tc. There are a number of reports written by
DOE and contractors regarding the fate of reactor return uranium (DOE, 2000a, 200b, BJC, 200a-c) on this issue.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 23

Figure 5. Additional radionuclide data (SRS-2002). Sample size = 33.

3.5.1 Mass of GDP DU
The total mass of anticipated GDP DU oxide is estimated from the reported mass of DUF6
currently residing in the cylinder yards and a mass conversion from DUF6 to DU3O8.
Although the exact number of cylinders at each facility varies from day to day, the Depleted
Uranium Management Information Network reports the numbers as 36,191 at Paducah, 16,109
from the Portsmouth GDP, and 4,822 from the K-25 GDP, now moved to Portsmouth (DOE,
2010). However, there are discrepancies in the available information regarding the numbers of
cylinders. Consequently, these numbers are used only for rough estimates of the volume needed
for disposal.
Estimates of the total mass of DUF6 from each of the GDPs is also provided at the Depleted UF6
Management Information Network web site (DOE, 2010). These estimates are 436,400 Mg for
Paducah, 195,800 Mg for Portsmouth, and 54,300 Mg for the K-25 GDP, now stored at
Portsmouth. These estimates are used in the PA model. No uncertainty is assigned to them. They
are a condition of the PA model until more information is made available. Uncertainty is,
instead, included in the concentration estimates, which serves as a reasonable measure in this PA
model for inventory uncertainty.
3.5.2 Number of GDP DU cylinders disposed
The number of GDP DU cylinders disposed at the Clive facility is constrained by the available
below-grade volume, as all DU wastes will be disposed below grade. The number of GDP DU
cylinders which could be disposed at the Clive site was estimated from engineering
specifications, packing dimensions and SRS DU drum volume.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 24
Based on the dimensions of the waste footprint and depth below-grade, the number of 12-ft long,
4-ft diameter cylinders that could be disposed was estimated to be 48,906 cylinders could be
disposed (EnergySolutions, 2015). A detailed description of embankment dimensions and a
discussion of representation of the Federal Cell in the GoldSim model are provided in the
Embankment Modeling for the Clive DU PA Model white paper. Given that each 48Yd cylinder
has a volume of about 4041 L (Argonne National Laboratory, n.d.) this translates to a total waste
disposal volume of 197,629 m3.
A portion of this total waste volume is allocated to the SRS DU wastes described in Section 2.2.
The total volume of the 5408 SRS 55-gallon drums is approximately 1,125 m3. Assuming similar
packing efficiency of SRS drums and GDP cylinders (i.e., the ratio of cylinder/drum volume to
total volume), we subtract this volume from the total GDP cylinder volume to give 196,504 m3
available for GDP DU cylinders. This is equivalent to about 48,628 cylinders.
3.5.3 Composition of GDP DU
As of this writing, only a single cylinder of oxide has been produced from the deconversion
plants, and only one sample from that cylinder has been analyzed. The DUF6 processed for this
sample was of low 235U assay, and contained no TRU or fission product contaminants, and is
therefore not representative of the entire populations of GDP DU oxides. The GDP DU is
considered to have two distinct compositions: Clean DU is pure uranium, derived from natural
sources, and Contaminated DU includes at least some TRU and fission products from reactor
returns. Each of these is discussed below, and the fraction of the total that is contaminated is
estimated for use in the PA model.
3.5.3.1 Clean GDP DU
The constituents comprising the clean DU are naturally-occurring isotopes of uranium,
significantly depleted in everything but 238U, and whatever decay products may have developed
in the short time since their purification and separation. No quantitative information is available
about the relative abundance of the uranium isotopes that characterizes the entire waste stream.
Given the lack of definitive information about the relative abundances of the uranium isotopes, it
is assumed that Clean DU from the GDPs shares the same uranium composition as the DU from
SRS. The same isotopic abundances and contaminant concentrations developed for the SRS DU
in Section 3.2.3 are therefore applied to the uranium fraction of GDP DU cylinders.
3.5.3.2 Contaminated GDP DU
No quantitative information is available about the contamination of the GDP DU Cylinders,
other than limited research determining that some are contaminated and some are not. Given the
lack of definitive information about the degree of contamination, it is assumed that contaminated
DU from the GDPs shares the same composition as the DU from SRS. The same isotopic
abundances and contaminant concentrations developed for the SRS DU in Section 3.2.3 are
therefore applied to the contaminated fraction of GDP DU cylinders. There are no other data that
are available at this time. The processes under which the DU waste is generated is similar in both
case, with material being processed in a diffusion cascade. In both cases the cascades were
contaminated, and this is the source of the contaminants in the DU. Without further information
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 25
on the contamination concentration levels, use of the SRS DU contaminant concentrations is the
only information available, even though it is surrogate information.
3.5.3.3 Fraction of Contaminated GDP DU
Assuming that each GDP cylinder is either “clean” or “contaminated”, an estimate is needed for
the number of each type, so that the total amount of contaminant radionuclides in the GDP
inventory can be estimated. At the time of this writing, the best available information about this
comes from a study by Henson (2006): DUF6-G-G-STU-003 (Draft for UDS review). This
document reviews information about the Paducah population of cylinders as recorded on cylinder
history cards, which were used until 1988, and all contaminated cylinders are represented in this
population. Table 1 (reproduced here as Table 10) in Henson (2006) categorizes the cylinders as
follows:
• "Category 1 – 13,240 cylinders: Cleared" cylinders, which are not contaminated,
• "Category 2 – 1,335 cylinders: TRU and/or Tc" cylinders, which are confirmed to have
some degree of contamination,
• "Category 3 – 971 cylinders: >1% U235" cylinders, which do not contain DU and so are
not considered in this PA, and
• “Category 4 – 22,382 cylinders: To Be Determined" cylinders which have unknown
status regarding contamination. 9,407 of these cylinders have history cards and 12,975 do
not.
Note that these values are in numbers of cylinders, rather than mass of DU, so an assumption is
made for the purposes of estimating the fraction of waste that is contaminated that each cylinder
contains the same mass of DU. Note also that the total number of cylinders here is not the same
as the number of cylinders suggested in Section 3.5.1. This reflects both uncertainty in the total
number of cylinders, and the change in number through time as cylinders are reprocessed or
transferred.
The Paducah data can be summarized as follows for the purposes of building a distribution for
the fraction of cylinders that are contaminated:
• 13,240 are known to not be contaminated
• 1,335 are known to be contaminated
• Of the unknowns 9,407 have history cards, and, hence, can be considered part of the
same population of reconciled cylinders. These are assumed to be pre-1988 cylinders.
• Of the unknowns, 12,975 do not have history cards. These are post-1988 cylinders.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 26
The cylinder history card system at Paducah was discontinued May 31, 1988 (Henson, 2006).
Paducah cylinders post-1988 are considered much more likely to be clean of contaminants.
Consequently, unknown cylinders are modeled differently for pre-1988 and post-1988.
The cylinders at Portsmouth also need to be considered. The Depleted Uranium Management
Information Network reports the numbers as 16,109 from the Portsmouth GDP, and 4,822 from
the K-25 GDP, now moved to Portsmouth (DOE, 2010). These cylinders are also considered
unlikely to be contaminated (personal communication, Tammy Stapleton, May 2011).
This completes the summary of the population of cylinders that are considered for disposal at the
Clive facility. The available information is used to construct an estimate of the total fraction of
the cylinders that are contaminated. In effect the proportion contaminated at Paducah for the
cylinders that have known status is used as an estimate of the fraction of all cylinders with
history cards that are contaminated. These are presumed to be all of the pre-1988 cylinders. For
the post-1988 cylinders at Paducah, which have no history cards, and the Portsmouth cylinders, a
much smaller fraction of the cylinders is assumed to be contaminated.
Consequently, the fraction of Pre-1988 cylinders at Paducah that is assumed to be contaminated
is about 9% [1,335 / (1,335 + 13,240)]. The Portsmouth cylinders might also have a small
fraction that are contaminated. Using expert opinion, this is estimated at less than 1%, with a best
guess at no more than 10 cylinders contaminated (personal communication, Tammy Stapleton,
May 2011). These values were interpreted as expert judgment of the 95th and 50th percentiles of
the distribution, respectively. A beta distribution was fit to these values, following the procedures
outlined in the Fitting Probability Distributions white paper.
The total number of contaminated cylinders was then simulated by adding the number of
confirmed contaminated cylinders with simulated numbers for the unknown cylinders. Table 11
shows the inputs that were used for the simulations. A distribution was constructed based on the
simulation output for the overall proportion of cylinders that are contaminated. This
Beta( 0.0392, 0.0025 ) probability density function is shown in Figure 6.
In terms of the number of contaminated cylinders, this distribution has 1st, 50th, and 99th
percentiles of 1,946, 2,266, and 2,619, respectively. This is a fairly narrow distribution given the
lack of information available. It is narrow because nearly 15,000 of the Paducah cylinders have
been characterized, an assumption is made that all other pre-1988 cylinders will be show a
similar ratio, and the remaining cylinders are expected to be clean of contamination. As more
information is gathered when the depleted uranium is prepared for disposal, then input
distributions used to characterize the GDP waste should be revisited. Information that will be
needed will include total amount of DU, chemical speciation of DU, and activity concentrations
of the DU and contaminants.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 27
Table 10: Categorization of Paducah Cylinders Using Cylinder History Cards (reproduced
from Table 1 in Henson, 2006)
Category 1:
Cleared
Category 2:
TRU and/or Tc
Category 3:
>1% 235U
Category 4:
To Be
Determined
Filled once with
natural normal or
depleted material.
(9,728)
Never filled with 1% or
greater assay, but have a
history of containing
recycled feed material.
These cylinders may have
“hidden heels” containing
both transuranics (TRU)
and Tc.
(1,334)
Filled at some time with
material >1% assay, and
also used to contain
recycled material. These
cylinders may have
“hidden heels” containing
both transuranics (TRU)
and Tc.
(584)
No Paducah
history card.
(12,975)
Filled more than
once, but only with
natural normal or
depleted material.
(2,681)
No history of recycled
feed service, but used to
hold Paducah product (at
<1% enrichment). These
cylinders may also have
“hidden heels” which
could contain Tc.
(1)
No history of recycled
feed service, but used to
hold Paducah product (at
>1% enrichment). These
cylinders may also have
“hidden heels” which
could contain Tc.
(387)
History card
does not provide
enough
information.
(9,407)
Washed and
subsequently filled
with only natural
normal or depleted
material.
(832)
Filled at some time with
>1% assay, but have
never contained recycled
uranium or Paducah
product.
(n/a for Phase II)

TOTAL = 13,240 TOTAL = 1,335 TOTAL = 971 TOTAL = 22,382

Table 11: Inputs for the Simulation of the Fraction of Contaminated GDP Cylinders
Cylinder
Type
Paducah
Category 2
Paducah
Category 1
Paducah
Category 4
Pre-1988
Paducah
Category 4
Post-1988
Portsmouth
(not from
Oak Ridge)
Portsmouth
(from Oak
Ridge)
Number 1,335 13,240 9,407 12,975 16,109 4,822
Simulated
Binomial
Proportion
NA
(confirmed
value)
NA
(confirmed
value)
Beta( 0.092,
0.0024) Beta( 0.0020, 0.0042 )
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 28

Figure 6. Probability density function for the proportion of contaminated cylinders.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 29
4.0 References
Argonne National Laboratory. (n.d.) UF6 Cylinder Data Summary. Retrieved from
http://web.ead.anl.gov/uranium/guide/prodhand/sld035.cfm
Beals D.M., LaMont S.P., Cadieux J.R., et al. 2002. Determination of Trace Radionuclides in
SRS Depleted Uranium (DU). WSRC-TR-2002-00536, Westinghouse Savannah River
Company, Savannah River Site, Aiken, SC.
BJC (Bechtel Jacobs Company LLC), 2000a, Recycled Uranium Mass Balance Project Oak
Ridge Gaseous Diffusion Plant Site Report, BJC/OR-584, June 2000.
BJC, 2000b, Recycled Uranium Mass Balance Project Paducah Gaseous Diffusion Plant Site
Report, BJC/PGDP-167, 14 Jun 2000.
BJC, 2000c, Recycled Uranium Mass Balance Project Portsmouth, Ohio Site Report,
BJC/PORTS-139/R1, 19 Jun 2000.
CFR (Code of Federal Regulations), 2014. 10 CFR 61, Licensing Requirements for Land
Disposal of Radioactive Waste, United States Code of Federal Regulations, United States
Government Printing Office, Washington DC, January 2014.
DOE (U.S. Department of Energy) 1999a. Programmatic Environmental Impact Statement for
Alternative Strategies for the Long-Term Management and Use of Depleted Uranium
Hexafluoride (DUF6 PEIS) (DOE/EIS-0269).
DOE 1999b. Final Plan for the Conversion of Depleted Uranium Hexafluoride as Required by
Public Law 105-204.
DOE, 2004a. Final Environmental Impact Statement for Construction and Operation of a
Depleted Uranium Hexafluoride Conversion Facility at the Paducah, Kentucky, Site,
DOE/EIS-0359, U.S. DOE Environmental Management, June 2004.
DOE, 2004b. Final Environmental Impact Statement for Construction and Operation of a
Depleted Uranium Hexafluoride Conversion Facility at the Portsmouth, Ohio, Site,
DOE/EIS-0360, U.S. DOE Environmental Management, June 2004.
DOE. 2010. Depleted UF6 Management Information Network. URL:
http://web.ead.anl.gov/uranium/mgmtuses/storage/index.cfm
EnergySolutions. 2009b. Radioactive Waste Profile Record, EC 0230, Rev. 7, plus attachments
(Form 9021 33), EnergySolutions Inc. Clive UT.
EnergySolutions. 2009d. Uniform Low-level Radioactive Waste Manifest Shipping Papers,
(Form 540), EnergySolutions Inc. Clive UT.
EnergySolutions, 2015. Engineering Drawing 14004 SK1, “Conceptual DU Disposal Plan”,
dated 10/23/2015. (file: FederalCell DUplan 14004-SK1.pdf).
Electric Power Research Institute (EPRI). 1985. Radionuclide Correlations in Low-Level
Radwaste, EPRI NP-4037, June 5, 1985.
Fussell, G.M, and D. L. McWhorter, 2002. Project Plan for the Disposition of the SRS Depleted,
Natural, and Low-Enriched Uranium Materials. WSRC-RP-2002-00459, Washington
Savannah River Site, November 21, 2002.
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 30
GEL 2010a. GEL Work Order 243721. Laboratory report dated January 12, 2010.
GEL 2010b. GEL Work Order 244495. Laboratory report dated January 19, 2010.
GEL 2010c. GEL Work Order 249710. Laboratory report dated April 8, 2010.
Henson (Henson Technical Projects, LLC), 2006, Contents Categorization of Paducah DUF6
Cylinders Using Cylinder History Cards – Phase II, DUF6-G-G-STU-003, Draft for
UDS Review, Uranium Disposition Services, LLC, Lexington, KY, 30 September 2006
(file: DUF6-G-G-STU-003 Henson 2006.pdf)
Johnson R. 2010. State of Utah, DEQ. Memo – April 6, 2010 Subj. Savannah River Depleted
Uranium Sampling
NRC (U.S. Nuclear Regulatory Commission). 2010. Stages of the Nuclear Fuel Cycle, URL:
http://www.nrc.gov/materials/fuel-cycle-fac/stages-fuel-cycle.html
ORNL (Oak Ridge National Laboratory). 2000a. Depleted Uranium Storage and Disposal Trade
Study: Summary Report, ORNL/TM 2000/10, Oak Ridge National Laboratory, Oak
Ridge TN, February, 2000
ORNL. 2000b. Assessment of Preferred Depleted Uranium Disposal Forms, ORNL/TM
2000/161, Oak Ridge National Laboratory, Oak Ridge TN, June 2000
ORNL. 2000c. Strategy for Characterizing Transuranics and Technetium Contamination in
Depleted UF6 Cylinders, ORNL/TM-2000/242, Oak Ridge National Laboratory, October
2000.
ORNL. 2000d. Evaluation of the Acceptability of Potential Depleted Uranium Hexafluoride
Conversion Products at the Envirocare Disposal Site, ORNL/TM-2000/355, Oak Ridge
National Laboratory, October 2000.
Rich, B.L., S.L. Hinnefeld, C.R. Lagerquist, W.G. Mansfield, L.H. Munson, E.R. Wagner, and
E.J. Vallario, 1988. Health Physics Manual of Good Practices for Uranium Facilities,
EGG-2530, Idaho National Engineering Laboratory, Idaho Falls, ID, June 1988.
Stapleton, Tammy, Uranium Disposition Services, LLC, personal communication via telephone
to John Tauxe, Neptune and Company, Inc., 3 May 2011.
Tukey, John (1977). Exploratory Data Analysis. Addison-Wesley.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 31
Appendix
Table 12. Uranium isotopic abundances by mass spectrometry, atomic percent, including
replicates (data summarized in Table 16, Beals, et al. 2002)
Sample Replicate 234U 235U 236U 238U
3 a 0.0004% 0.1270% 0.0065% 99.87%
3 b 0.0004% 0.1260% 0.0065% 99.87%
9 a 0.0004% 0.1260% 0.0064% 99.87%
9 b 0.0004% 0.1250% 0.0064% 99.87%
17 a 0.0004% 0.1260% 0.0066% 99.87%
17 b 0.0004% 0.1260% 0.0066% 99.87%
20 a 0.0005% 0.1270% 0.0068% 99.87%
20 b 0.0004% 0.1290% 0.0067% 99.86%
25 a 0.0008% 0.1510% 0.0096% 99.84%
25 b 0.0007% 0.1510% 0.0095% 99.84%
30 a 0.0006% 0.1410% 0.0088% 99.85%
30 b 0.0006% 0.1400% 0.0086% 99.85%

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 32
Table 13. Uranium isotopic abundances by alpha spectrometry (as percent of total uranium
activity) (Table 17, Beals, et al. 2002) and Technetium concentrations in the
SRS-2002 data (Beals, et al. 2002)
Sample 238U 235+ 236U 234U 99Tc (nCi/g)
1 91.7 1.72 6.57 44.2
2 91.0 1.74 7.28 57.5
3 91.3 2.04 6.63 21.2
4 91.3 1.86 6.82 33.3
5 91.6 1.73 6.67 15.7
6 91.2 1.76 7.07 19.1
7 91.2 1.85 6.91 18.5
8 91.6 1.71 6.67 24.5
9 91.3 1.98 6.72 90.2
10 91.8 1.7 6.55 79.7
11 91.6 1.7 6.75 89.8
12 91.8 2.04 6.18 79.7
13 91.3 1.95 6.74 37.5
14 91.2 1.7 7.09 75.3
15 91.6 1.74 6.63 34.2
16 91.4 1.86 6.7 74.2
17 91.2 2.07 6.7 41.4
18 91.4 1.86 6.71 64.7
19 91.7 1.97 6.32 16.1
20 90.8 2.25 6.92 14.9
21 91.6 1.73 6.69 27.2
22 87.5 2.11 10.42 8.1
23 88.4 2.11 9.46 15.7
24 85.9 2.51 11.55 9
25 86.9 2.41 10.71 93.8
26 86.7 2.36 10.9 92.7
27 87.3 2.27 10.41 32.5
28 88.0 2.26 9.72 55.3
29 87.3 2.84 9.91 53.8
30 88.5 2.27 9.2 88.5
31 85.9 2.77 11.32 93.7
32 88.6 2.8 8.61 54.3
33 88.2 1.83 9.99 73
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 33
Mean 90 2.05 7.99 49.37
Std.Dev 2.03 0.34 1.77 29.26

Table 14. January 2010 EnergySolutions Data Analyzed by GEL (GEL 2010a and 2010b)
Sample ID bulk density
(g/cm3)
99Tc
(pCi/g DU
waste)
total uranium
(µg/g DU
waste)
233+234U
(pCi/g DU
waste)
235+236U
(pCi/g DU
waste)
238 U
(pCi/g DU
waste)
243721001 3.31 2.28E+4 7.93E+5 4.84E+4 1.11E+4 2.65E+5
243721002 3.45 9.78E+3 8.54E+5 4.50E+4 7.21E+3 2.86E+5
243721003 2.84 1.78E+4 8.06E+5 3.83E+4 1.89E+4 2.68E+5
243721004 3.15 9.04E+3 8.27E+5 3.26E+4 4.92E+3* 2.77E+5
243721005 2.50 1.44E+4 8.48E+5 4.25E+4 7.27E+3* 2.85E+5
243721006 3.21 2.08E+4 8.80E+5 3.04E+4 1.28E+4 2.94E+5
243721007 4.00 2.25E+4 9.90E+5 6.44E+4 1.28E+4 3.31E+5
243721008 2.36 1.14E+4 6.50E+5 3.37E+4 1.19E+4 2.17E+5
244495001 3.46 2.60E+4 8.44E+5 3.57E+4 6.72E+3 2.83E+5
244495002 3.66 2.35E+4 8.00E+5 3.65E+4 1.17E+4 2.67E+5
244495003 4.00 1.81E+4 8.76E+5 4.70E+4 9.94E+3 2.93E+5
Mean 3.27 1.78E+4 8.33E+5 4.13E+4 1.15E+4 2.79E+5
Std.Dev 0.54 5.91E+3 8.15E+3 9.73E+3 3.57E+3 2.74E+3
* - reported as non-detects – detection limits used for statistical analysis.

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 34
Table 15. April 2010 EnergySolutions Data Analyzed by GEL (GEL 2010c)
Sample ID bulk density
(g/cm3)
99Tc (pCi/g
DU waste)
total uranium
(µg/g DU
waste)
233+234U
(pCi/g DU
waste)
235+236U
(pCi/g DU
waste)
238 U
(pCi/g DU
waste)
249710001 - - 7.95E+5 3.42E+4 6.34E+3 2.66E+5
249710002 - - 8.31E+5 3.65E+4 6.31E+3 2.78E+5
249710003 - - 8.15E+5 3.35E+4 5.12E+3 2.73E+5
249710004 - - 8.74E+5 3.84E+4 5.17E+3 2.93E+5
249710005 - - 8.28E+5 3.66E+4 4.30E+3 2.78E+5
249710006 - - 8.74E+5 4.17E+4 4.31E+3 2.93E+5
249710007 - - 7.07E+5 2.94E+4 4.86E+3 2.37E+5
249710008 - - 6.46E+5 3.78E+4 4.43E+3 2.17E+5
249710009 - - 7.42E+5 3.66E+4 6.80E+3 2.48E+5
249710010 - - 7.97E+5 3.86E+4 4.95E+3 2.67E+5
249710011 - - 8.29E+5 3.51E+4 4.36E+3 2.78E+5
249710012 - - 7.58E+5 2.98E+4 7.60E+3 2.54E+5
249710013 - - 7.45E+5 3.16E+4 4.89E+3 2.50E+5
249710014 - - 7.71E+5 3.09E+4 4.14E+3 2.58E+5
249710015 - - 8.97E+5 4.02E+4 5.75E+3 3.01E+5
Mean 7.88E+5 3.54E+4 5.34E+3 2.64E+5
Std.Dev. 6.61E+4 3.60E+3 1.07E+3 2.21E+4

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 35
Table 16. Technetium-99 concentrations collected by State of Utah, (Johnson, 2010)
Sample ID pCi/g DU
waste Sample ID pCi/g DU
waste Sample ID pCi/g DU
waste
1337 6.30E+3 3800 1.24E+4 0249 3.28E+3
1348 1.27E+4 3824 5.59E+3 0370 4.77E+3
1423 2.13E+3 3849 4.13E+3 0434 2.80E+3
1428 3.45E+3 3857 2.56E+3 0461 4.09E+3
1429 7.05E+3 3870 1.55E+4 0488 3.09E+3
1467 2.66E+3 3951 1.79E+3 0499 8.22E+2
1584 3.50E+3 4052 2.07E+3 0555 1.12E+3
1622 7.99E+3 4104 2.44E+3 0562 2.08E+3
1697 3.09E+3 4138 4.23E+3 0565 7.19E+3
1712 5.21E+3 4162 3.51E+3 0571 4.11E+3
1739 5.62E+3 4172 6.85E+3 0626 1.78E+3
1794 2.74E+3 4185 2.64E+3 0629 4.41E+3
1808 2.54E+3 4207 2.01E+3 0662 2.74E+2
1834 1.53E+4 4244 1.56E+3 0670 1.95E+3
1835 7.12E+3 4275 1.22E+3 0697 1.63E+3
1853 2.49E+3 4303 8.86E+2 0739 2.37E+3
1876 1.47E+3 4322 1.01E+3 0756 3.56E+3
1918 2.90E+3 4362 3.06E+3 0800 1.57E+3
1946 2.08E+3 4376 6.66E+3 0809 5.73E+2
2061 1.84E+4 4384 2.32E+3 0813 2.22E+3
2077 1.83E+3 4385 9.72E+3 0852 4.45E+3
2098 1.10E+4 4393 3.58E+3 0853 2.31E+3
2102 7.65E+2 4414 3.78E+3 0854 2.83E+3
2140 7.86E+3 4415 8.86E+3 0879 4.52E+3
2250 6.71E+3 4425 5.87E+3 0884 4.76E+3
2256 7.19E+3 4431 1.29E+4 0893 2.02E+3
2343 1.30E+3 4486 5.83E+3 0910 2.24E+2
2424 6.27E+2 4487 2.63E+3 0911 8.23E+2
2449 4.86E+3 4504 8.48E+3 0927 6.38E+2
2481 1.32E+3 4535 5.25E+3 0928 7.42E+2
2497 1.62E+4 4606 1.72E+3 1000 5.85E+3
2517 8.06E+2 4611 3.47E+3 1021 1.24E+3
2528 1.66E+3 4687 1.51E+3 1030 1.63E+3
2550 3.02E+3 4760 3.04E+3 1117 6.56E+3
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 36
Sample ID pCi/g DU
waste Sample ID pCi/g DU
waste Sample ID pCi/g DU
waste
2614 1.49E+3 4790 2.28E+3 1140 1.76E+3
2674 1.89E+3 4817 2.25E+3 1147 1.29E+3
2675 2.92E+3 4822 2.62E+3 1216 1.44E+3
2823 4.89E+3 4851 1.32E+4 1505 2.26E+3
2827 1.61E+4 4866 1.45E+4 1511 3.96E+3
2878 2.86E+3 4940 4.41E+3 1646 6.19E+3
3035 7.59E+3 4955 3.68E+3 1678 1.00E+4
3059 5.01E+3 4962 5.89E+3 2393 4.08E+3
3067 1.77E+3 5023 1.89E+3 2657 5.52E+3
3080 4.36E+3 5054 2.36E+3 2693 1.97E+3
3085 1.53E+3 5061 1.68E+3 3127 3.31E+3
3089 2.37E+3 5084 6.22E+3 3160 6.34E+3
3197 2.28E+3 5191 1.13E+4 3288 7.08E+3
3234 4.62E+3 5224 5.75E+3 3336 5.12E+3
3303 5.61E+3 5277 1.58E+3 3337 5.37E+3
3347 5.53E+3 5322 7.33E+2 3446 3.25E+3
3543 1.67E+3 0023 3.89E+3 3471 2.86E+3
3668 3.12E+3 0057 1.15E+3 3546 4.73E+3
3685 3.03E+3 0157 1.28E+3 4016 1.09E+4
3695 7.46E+3 0162 6.53E+3 4098 8.93E+3
3717 4.56E+3 0168 3.42E+3 4200 1.68E+3
3726 4.94E+3 0180 2.80E+3 4514 3.28E+3
3728 1.28E+4 0210 3.72E+3 4581 1.81E+3
3760 2.38E+3 0214 2.24E+3

Number of samples = 173
Average 99Tc concentration = 4,340 pCi/g
Standard Deviation = 3,550 pCi/g

Radioactive Waste Inventory for the Clive DU PA
12 November 2015 37
Table 17. Concentration data for other radioisotopes, SRS-2002. (Beals, et al. 2002)
Sample
241Am
*(<)
pCi/g
DU
waste
226Ra
*(<)
pCi/g
DU
waste
137Cs
*(<)
pCi/g
DU
waste
90Sr
*(<)
pCi/g
DU
waste
237Np
pCi/g
DU
waste
238Pu
pCi/g
DU
waste
239Pu
pCi/g
DU
waste
240Pu
pCi/g
DU
waste
241Pu
pCi/g
DU
waste
129I *(<)
pCi/g
DU
waste
1 6 120 6 8.6 0.44 0.114 0.53 0.14 2.80 13
2 24 500 19 5.9 2.34 0.099 0.69 0.15 nd 7
3 21 450 17 3.4 0.33 0.065 0.48 0.12 1.00 7
4 17 330 14 6.7 4.61 0.129 0.84 0.17 nd 4
5 25 600 20 7.2 12.8 0.086 0.95 0.23 2.50 12
6 20 390 15 14 8.89 0.163 0.40 0.10 nd 10
7 16 314 13 8 14.3 0.090 0.34 0.10 1.60 9
8 16 310 12 7.7 3.85 1.420 0.91 0.48 10.00 4
9 10 240 9 50.7 6.52 0.350 3.43 1.14 11.00 8
10 21 470 19 32.7 2.43 0.244 0.48 0.18 3.80 6
11 16 370 14 23.4 13.6 0.240 3.10 0.68 13.00 20
12 11 250 10 29.3 11.9 0.090 1.15 0.29 2.70 14
13 11 260 10 46.6 8.55 0.230 5.09 1.14 17.00 18
14 13 340 12 31.2 1.3 0.123 2.46 0.55 7.50 20
15 17 360 13 40 6.38 0.127 0.36 0.09 0.90 16
16 12 300 11 68.2 33.5 0.099 0.66 nd nd 16
17 11 230 10 28.4 6.08 0.125 1.63 0.50 4.00 17
18 11 230 8 38.3 2.86 0.081 0.75 0.20 nd 19
19 10 210 7 51 10.2 0.043 3.74 0.86 11.00 26
20 6 170 5 45.6 11.3 0.088 1.07 0.27 nd 32
21 14 300 13 27.1 1.92 0.094 0.50 0.12 1.10 33
22 9 250 8 28.6 0.77 0.149 0.81 0.22 3.40 27
23 18 380 15 45.7 1.67 0.186 1.81 0.52 5.30 24
24 16 340 13 26.9 0.69 0.242 1.30 0.36 nd 27
25 13 280 11 45.7 1.18 0.178 0.88 0.24 2.60 26
26 9 250 9 100.5 0.65 0.560 0.79 0.22 2.70 7
27 10 280 10 59.1 0.94 0.181 0.79 0.22 2.80 30
28 25 550 21 28 1.61 0.154 0.74 0.21 3.40 34
29 16 410 14 57.9 11.1 0.420 0.79 0.18 nd 24
30 10 190 10 32.9 0.87 0.123 0.85 0.22 4.00 27
31 16 350 15 78.9 1.04 0.250 1.02 nd nd 26
32 9 190 7 438.2 1.32 0.155 1.09 0.32 2.50 22
33 9 240 9 35.8 1.58 0.153 0.82 0.24 1.70 28
Radioactive Waste Inventory for the Clive DU PA
12 November 2015 38
Appendix References
Beals D.M., LaMont S.P., Cadieux J.R., et al. 2002. Determination of Trace Radionuclides in
SRS Depleted Uranium (DU). WSRC-TR-2002-00536, Westinghouse Savannah River
Company, Savannah River Site, Aiken, SC.
GEL 2010a. GEL Work Order 243721. Laboratory report dated January 12, 2010.
GEL 2010b. GEL Work Order 244495. Laboratory report dated January 19, 2010.
GEL 2010c. GEL Work Order 249710. Laboratory report dated April 8, 2010.
Johnson R. 2010. State of Utah, DEQ. Memo – April 6, 2010 Subj. Savannah River Depleted
Uranium Sampling

 NAC-0015_R4

Unsaturated Zone Modeling for
the Clive PA

Clive DU PA Model v1.4
23 October 2015

Prepared for EnergySolutions by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Unsaturated Zone Modeling for the Clive PA
23 October 2015 iii
1. Title: Unsaturated Zone Modeling for the Clive PA
2. Filename: Unsaturated Zone Modeling v1.4.docx
3. Description: This white paper provides documentation of the development of parameter values
and distributions used for modeling liquid phase transport in the unsaturated zone for the
Clive DU PA Model.
Name Date
4. Originator Michael Sully 5 May 2014
5. Reviewer Dan Levitt 21 May 2014
6. Remarks
5/8/2014: DL. Added new section 14.0 that discusses nine H1D sensitivity runs that evaluate
effects of Rn barrier Ksat and rooting depth on infiltration.
5/9/2014: MS. The element names in GoldSim for α and n were changed. Element names were
revised in white paper.
5/13/2014: MS. Revised distribution for van Genuchten alpha and n from using standard
deviations to standard errors.
5/13/2014: DL. Added text justifying use of 1D model.
5/17/2014: MS: Expanded discussion of 2D vs 1D.
6/8/2014: MS: Added reference to method described in Appendix for estimating water content
for waste, clay liner, and unsaturated zone.
10/6/2015: GO/DL: Updated regression coefficients in Section 12 based on latest results of 50
HYDRUS runs, and made corresponding text updates. Deleted section 14 (from
v1.2) as it refers to a sensitivity analysis for the 20 reps described in v1.2.
10/19/2015: MS/DL: Revised porosity distributions for surface, ET, and frost protection layers.
More edits to parameter names to be more consistent with the GoldSim model v1.4.
10/21/2015: MS/DL: Delete Federal DU cell drawing. Version change to v1.4.
10/23/2015: MS/DL/KC: Edits for consistency with model v1.4.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 iv
CONTENTS
1.0
 Summary of Parameter Values and Distributions .................................................................. 8

2.0
 Introduction .......................................................................................................................... 12

3.0
 Disposal Cell Design ............................................................................................................ 12

4.0
 Unsaturated Zone and Shallow Aquifer ............................................................................... 14

5.0
 Climate ................................................................................................................................. 17

6.0
 Vegetation ............................................................................................................................. 19

7.0
 Properties of Unit 3 and Radon Barriers .............................................................................. 20

7.1
 Laboratory Measurements ........................................................................................... 20

7.2
 Grain Size Distributions for the Cores ........................................................................ 20

7.3
 Soil Material Properties ............................................................................................... 24

7.4
 Soil Moisture Content ................................................................................................. 26

7.4.1
 Unit 3 Brooks-Corey Parameters ..................................................................... 30

7.4.2
 Unit 4 Brooks-Corey Parameters ..................................................................... 30

8.0
 Properties of Upper Cover Layers ........................................................................................ 30

9.0
 Properties of Waste ............................................................................................................... 31

10.0
 Properties of the Clay Liner ................................................................................................. 31

11.0
 Properties of the Unsaturated Zone below the Clay Liner ................................................... 31

12.0
 Modeling of Net Infiltration and Water Content for the Clive DU PA Model .................... 32

12.1
 Description of HYDRUS ............................................................................................ 32

12.2
 Conceptual Model ....................................................................................................... 34

12.3
 Climate and Vegetation Parameters ............................................................................ 34

12.4
 Model Geometry ......................................................................................................... 39

12.5
 Material Properties ...................................................................................................... 39

12.6
 Boundary Conditions ................................................................................................... 44

12.7
 Initial Conditions ......................................................................................................... 44

12.8
 Cases Simulated .......................................................................................................... 44

12.9
 Model Results .............................................................................................................. 44

13.0
 Implementation in GoldSim ................................................................................................. 45

14.0
 Contaminant Fate and Transport in Porous Media ............................................................... 46

14.1
 Porous Medium Water Transport ................................................................................ 46

14.1.1
 Advection of Water ......................................................................................... 46

14.1.2
 Diffusion in Water ........................................................................................... 46

14.1.3
 Water Phase Tortuosity ................................................................................... 47

14.2
 Porous Medium Air Transport .................................................................................... 48

14.2.1
 Advection of Air .............................................................................................. 48

Unsaturated Zone Modeling for the Clive PA
23 October 2015 v
14.2.2
 Diffusion in Air ............................................................................................... 48

14.2.3
 Air-Phase Tortuosity ....................................................................................... 49

15.0
 References ............................................................................................................................ 52

Appendix A ................................................................................................................................... 56

Appendix B .................................................................................................................................... 58

1.
 Purpose ................................................................................................................................... 58

2.
 Method .................................................................................................................................... 58

3.
 Darcy Equation Solution by the Runge-Kutta Method .......................................................... 60

4.
 Verification of the Runge-Kutta Method ............................................................................... 61

5.
 Implementation in the DU PA Model .................................................................................... 66

6.
 Numerical Testing of the Top Slope Model in GoldSim ....................................................... 67

8.
 References .............................................................................................................................. 75

Unsaturated Zone Modeling for the Clive PA
23 October 2015 vi
FIGURES
Figure 1. Evapotranspiration (ET) cover profile showing materials, observation nodes, and
root distribution used in the HYDRUS-1D models. .................................................... 14

Figure 2. Stratigraphic profile showing ET cover, waste zone, and stratigraphy below the
Federal DU Cell. .......................................................................................................... 15

Figure 3. Monthly mean precipitation for the Clive Site and monthly mean pan evaporation
for the NOAA BYU station at Provo, Utah. ................................................................ 17

Figure 4. Monthly mean temperatures for the Clive Site and the NOAA BYU station at
Provo, Utah. ................................................................................................................. 18

Figure 5. Comparison of water retention data (wetting cycle) for four core samples ................... 23

Figure 6. 100-year daily precipitation record generated from monthly average values of daily
measurements at the site based on 17 years of observations. ...................................... 37

Figure 7. 100-year daily Tmax and Tmin record generated from a 30-year record available
from the Dugway, Utah NOAA station. ...................................................................... 37

Figure 8. 100-year daily potential evaporation generated using the Hargreaves method. ............ 37

Figure 9. Root density with depth at the Clive Site for Shadscale and Black Greasewood
[SWCA 2011]. ............................................................................................................. 38

Figure 10. Water stress response function for root water uptake model. ...................................... 38

Figure 11. Comparison of air-phase tortuosity models by Penman (equation (44)), Millington
and Quirk (MQ1, equation (45)), Millington and Quirk as modified by Jin and
Jury (1996) (MQ2, equation (46)), and Lahvis et al. (1999) (equation (47)). ............. 50

Figure 12. Comparison of effective to bulk diffusivity ratios with air phase porosity for air
phase tortuosity models. .............................................................................................. 51

Unsaturated Zone Modeling for the Clive PA
23 October 2015 vii
TABLES
Table 1. Summary of Parameter Values and Distributions ............................................................. 8

Table 2. Assignment of solid/water partition coefficients Kd values. ........................................... 12

Table 3. Texture class, thickness range, and average thickness for the hydrostratigraphic units
underlying the Clive site. ............................................................................................. 15

Table 4. Grain size distributions for cores from Unit 4, a silty clay. ............................................ 21

Table 5. Grain size distributions for cores from Unit 3, a silty sand. ............................................ 22

Table 6. Theoretical porosities based on particle packing geometry. ........................................... 24

Table 7. Bulk density, porosity, and calculated particle density data from water retention
experiments. ................................................................................................................. 25

Table 8. Hydraulic properties of topslope cover used for HYDRUS modeling. ........................... 41

Table 9. Parameter sets of van Genuchten α and n, and Ks used for HYDRUS modeling. .......... 42

Table 10. Coefficients calculated from multiple linear regression models. .................................. 45

Table 11. Atmosphere volume parameters for creating a surface boundary condition in the
porous medium air diffusion model. ........................................................................... 49

Unsaturated Zone Modeling for the Clive PA
23 October 2015 8
1.0 Summary of Parameter Values and Distributions
A summary of material properties and parameter values used in the Clive DU PA Model is
provided in Table 1. For distributions, the following notation is used:
• N( µ, σ, [min, max] ) represents a normal distribution with mean µ and standard deviation
σ, and optional truncation at the specified minimum and maximum,
• LN( GM, GSD, [min, max] ) represents a log-normal distribution with geometric mean
GM and geometric standard deviation GSD, and optional min and max,
• U( min, max ) represents a uniform distribution with lower bound min and upper bound
max,
• Beta( µ, σ, min, max ) represents a generalized beta distribution with mean µ, standard
deviation σ, minimum min, and maximum max,
• Gamma( µ, σ ) represents a gamma distribution with mean µ and standard deviation σ,
and
• TRI( min, m, max ) represents a triangular distribution with lower bound min, mode m,
and upper bound max.

Note that a number of these distributions are truncated at a minimum value of 0 or a value of
Small, an arbitrarily small number just greater than 0 defined in the GoldSim model, and a
maximum of Large, an arbitrarily large value defined in the GoldSim model. The truncation at
the low end is a matter of physical limits (e.g. precipitation cannot be negative), and in
GoldSim’s distribution definitions, if truncations are made, they must be made at both ends, so
the very large value is chosen for the upper end.
Table 1. Summary of Parameter Values and Distributions
Parameter Distribution
[Comments]
Units Internal Reference
Infiltration and Water Content
VG_logAlpha N( µ=-1.79, σ=0.121,
(min=-Large, max=0 )
log10(1/cm) Section 12.5
VG_logN N( µ=0.121, σ=0.019,
(min=Small, max=Large );
— Section 12.5
RnBarrierKsat_Natdist LN( 3.37, 3.23); [right
shift of 0.00432]
cm/day Section 12.5
WaterContentResidual — Section 12.5, Table 8
SurfaceSoil
EvapLayer
FrostLayer
UpperRnBarrier
LowerRnBarrier
0.11
0.11
0.065
0.1
0.1
—
—
—
—
—

Section 12.5, Table 8
Section 12.5, Table 8
Section 12.5, Table 8
Section 12.5, Table 8
Section 12.5, Table 8

Unsaturated Zone Modeling for the Clive PA
23 October 2015 9
Cover Layers Infiltration and Water Content Regression Parameters
Response Variable β0
Infiltration flux (through all layers) -­‐0.32921
 — Section 12.9
Water content in Surface Layer 0.48155
 — Section 12.9
Water content in Evaporative zone layer 0.57947
 — Section 12.9
Water content in Frost Protection layer 0.04282
 — Section 12.9
Water in Upper Radon Barrier 0.14737
 — Section 12.9
Water in Lower Radon Barrier 0.14740
 — Section 12.9
Response Variable β1
Infiltration flux (through all layers) N/A — Section 12.9
Water content in Surface Layer 0.00000 — Section 12.9
Water content in Evaporative zone layer 0.00000 — Section 12.9
Water content in Frost Protection layer 0.00000 — Section 12.9
Water in Upper Radon Barrier -­‐0.00076 — Section 12.9
Water in Lower Radon Barrier -­‐0.00076 — Section 12.9
Response Variable β2
Infiltration flux (through all layers) 5.56826 — Section 12.9
Water content in Surface Layer 0.54920 — Section 12.9
Water content in Evaporative zone layer 0.73997 — Section 12.9
Water content in Frost Protection layer 0.43297 — Section 12.9
Water in Upper Radon Barrier 1.70702 — Section 12.9
Water in Lower Radon Barrier 1.70648 — Section 12.9
Response Variable β3
Infiltration flux (through all layers) 0.19538 — Section 12.9
Water content in Surface Layer -­‐0.20020 — Section 12.9
Water content in Evaporative zone layer -­‐0.24790 — Section 12.9
Water content in Frost Protection layer 0.01617 — Section 12.9
Water in Upper Radon Barrier 0.06353 — Section 12.9
Water in Lower Radon Barrier 0.06351 — Section 12.9

Fate and Transport
Water tortuosity water content exponent N( µ=7/3, σ=0.01) — Section 15.1.3
Unsaturated Zone Modeling for the Clive PA
23 October 2015 10
Water tortuosity porosity exponent N( µ=2.0, σ=0.01 — Section 15.1.3
Thickness of the atmosphere layer N( µ=2.0, σ=0.5,
min=Small, max=Large )
M Section 15.2.2, Table
12
Wind speed N( µ=3.14, σ=0.5,
min=Small, max=Large )
m/s Section 15.2.2, Table
12
Atmospheric diffusion length N( µ=0.1, σ=0.02,
min=Small, max=Large )
m Section 15.2.2, Table
12
Thickness of the Unsat zone (below the
embankment clay liner)
N(12.9, 0.25, min=Small,
max=Large )
ft Section 11
Unit 3
Porosity_Unit3 N( 0.393, 6.11e-3,
min=Small,
max=1-Small )
— Section 7.3
BulkDensity_Unit3 N( ParticleDensity_Unit3
× ( 1 – Porosity_Unit3 ),
0.1, min=Small,
max=Large )
g/cm3 Section 7.3
ParticleDensity_Unit3 2.65 g/cm3 Section 7.3
D_Unit3 N( 2.73, 5.21e-3, min=0,
max=3 )
— Section 7.4.1
Hb_Unit3 N( 8.85, 0.929,
min=Small, max=Large );
[-0.85 correlation with
D_Unit3]
cm Section 7.4.1
MCres_Unit3 N( 6.78e-3, 2.05e-3,
min=Small, max=Large )
— Section 7.4.1
MCsat_Unit3 equal to Porosity_Unit3 — Section 7.4.1
Ksat_Unit3 N( 5.14e-5, 5.95e-6,
min=Small, max=Large );
[-0.98 correlation with
D_Unit3]
cm/s Section 7.4.1
Unit 4
Porosity_Unit4Compacted N(0.428, 9.08e-3,
min = small, max = 1-
small);
— Section 7.4.2
BulkDensity_Unit4Compacted N( ParticleDensity_Unit4
× (1 – Porosity_Unit4 ),
0.1, min=Small,
max=Large );
g/cm3 Section 7.4
ParticleDensity_Unit4 2.65 g/cm3 Section 7.4
D_Unit4Compacted N( 2.81, 9.93e-5, min=0,
max=3 )
— Section 7.4.2
Hb_Unit4Compacted N( 104., 1.72, min=Small, cm Section 7.4.2
Unsaturated Zone Modeling for the Clive PA
23 October 2015 11
max=Large ); [correlated
to D_Unit4 as -0.66]
MCres_Unit4Compacted N( 0.108, 8.95e-4,
min=Small, max=Large );
[truncated just above 0]
— Section 7.4.2
MCsat_Unit4Compacted equal to Porosity_Unit4 — Section 7.4.2
Radon Barrier Clay
Porosity_UpperRnBarrierClay assigned value for Unit 4
BulkDensity_UpperRnBarrierClay assigned value for Unit 4
Porosity_LowerRnBarrierClay assigned value for Unit 4
BulkDensity_LowerRnBarrierClay assigned value for Unit 4
UpperRnBarrierKsat_AsBuilt 5.00E-8 cm/s Section 3
LowerRnBarrierKsat_AsBuilt 1.00E-6 cm/s Section 3
Unit 4 ET Layers
Porosity_Unit4_ETLayers N( 0.481, 0.015) — Section 8.0
BulkDensity_Unit4_ETLayers N( ParticleDensity_Unit4
× (1 –
Porosity_Unit4_ETLayers
), 0.1, min=Small,
max=Large )
g/cm3 Section 8.0
Frost Protection Layer
Porosity_SiltSandGravel N(0.41, 0.0026) — Section 8.0
BulkDensity_SiltSandGravel N( ParticleDensity_Unit4
× (1 –
Porosity_SiltSandGravel),
0.1, min=Small,
max=Large )
g/cm3 Section 8.0
Generic, UO3, and U3O8 Waste
Porosity_Generic_Waste assigned value for Unit 3 — Section 7.3
BulkDensity__Generic_Waste assigned value for Unit 3 g/cm3 Section 7.3
D_Generic_Waste assigned value for Unit 3 — Section 7.4.1
Hb_Generic_Waste assigned value for Unit 3 cm Section 7.4.1
MCres_Generic_Waste assigned value for Unit 3 — Section 7.4.1
MCsat_Generic_Waste assigned value for Unit 3 — Section 7.4.1
Ksat_Generic_Waste assigned value for Unit 3 cm/s Section 7.4.1
Liner Clay
Porosity_LinerClay assigned value for Unit 4 — Section 7.4.2
BulkDensity__LinerClay assigned value for Unit 4 g/cm3 Section 7.4
D_LinerClay assigned value for Unit 4 — Section 7.4.2
Hb_LinerClay assigned value for Unit 4 cm Section 7.4.2
Unsaturated Zone Modeling for the Clive PA
23 October 2015 12
MCres_LinerClay assigned value for Unit 4 — Section 7.4.2
MCsat_LinerClay assigned value for Unit 4 — Section 7.4.2
Ksat_LinerClay LN( 1e-6, 1.2 )

cm/s Section 10.0

Porous medium solid/water partition coefficients for various radionuclides in these materials are
assigned one of three representative and generic collections of Kd values for the materials sand,
silt and clay. These assignments are listed in Table 2. Distributions for the values themselves are
documented in the Geochemical Modeling white paper.

Table 2. Assignment of solid/water partition coefficients Kd values.
material Kd material
Unit 2 (includes saturated zone medium) clay
Unit 3 (includes unsaturated zone medium and all wastes) sand
Unit 4 (includes surface layer, evaporative zone, clay liner, and upper and lower
radon barrier clays)
silt

2.0 Introduction
This white paper provides documentation of the development of parameter values and
distributions used for modeling gas and liquid phase transport in the unsaturated zone for the
Clive DU PA Model. Data sources are identified and the rationale applied in developing
distributions is described. The intent of this white paper is to describe the characteristics and
processes in the disposal cell, waste, and the underlying unsaturated zone above the shallow
aquifer. Estimates of net infiltration through the evapotranspiration (ET) cover system layers and
material water content required by the GoldSim model (the DU PA Model) were made using the
HYDRUS-1D software package (Šimůnek et al., 2009) and are described in this white paper.
Saturated zone characteristics and processes are described in the Saturated Zone Modeling white
paper.
3.0 Disposal Cell Design
The general aspect of the Federal Cell (interchangeably termed the Federal DU Cell in this
document because of the focus of this model on disposal of DU) is that of a hipped cap, with
relatively steeper sloping sides nearer the edges. The upper part of the embankment, known as
the top slope, has a moderate slope, while the side slope is markedly steeper (20% as opposed to
2.4%). These two distinct areas, are modeled separately in the Clive DU PA Model. Each is built
in GoldSim to be modeled as a separate one-dimensional column, with an area equivalent to the
Federal DU Cell footprint. In the current Clive DU PA Model, the sideslope portion of the model
is inactive. The embankment is also constructed such that a portion of it lies below-grade. The
Unsaturated Zone Modeling for the Clive PA
23 October 2015 13
overall length of the embankment is 1317.8 ft and the overall width is 1775.0 ft. A detailed
description of embankment dimensions and a discussion of representation of the Federal DU Cell
in the GoldSim model are provided in the Embankment Modeling for the Clive DU PA Model
white paper.
Disposal involves placing waste on a prepared clay liner that is approximately 8 ft below the
ground surface. For the Federal DU Cell design, the depth of the waste below the top slope is a
maximum of 47.5 ft (14.5 m). A cover system is constructed above the waste. The objective of
the cover system is to limit contact of water with the waste. The cover is sloped to promote
runoff and designed to limit water flow by increasing evapotranspiration (ET). The arrangement
of the layers used for the ET cover design is shown in Figure 1. Beginning at the top of the
cover, the layers above the waste used for the ET cover design are:
• Surface layer: This layer is composed of native vegetated Unit 4 material with 15 percent
gravel mixture on the top slope and 50 percent gravel mixture for the side slope. This
layer is 6 inches thick. The functions of this layer are to control runoff, minimize erosion,
and maximize water loss from ET. This layer of silty clay provides storage for water
accumulating from precipitation events, enhances losses due to evaporation, and provides
a rooting zone for plants that will further decrease the water available for downward
movement.
• Evaporative Zone layer: This layer is composed of Unit 4 material. The thickness of this
layer is 12 inches. The purpose of this layer is to provide additional storage for
precipitation and additional depth for plant rooting zone to maximize ET.
• Frost Protection Layer: This material ranges in size from 16 inches to clay size particles.
This layer is 18 inches thick. The purpose of this layer is to protect layers below from
freeze/thaw cycles, wetting/drying cycles, and to inhibit plant, animal, or human
intrusion.
• Upper Radon Barrier: This layer consists of 12 inches of compacted clay with a low
hydraulic conductivity. This layer has the lowest conductivity of any layer in the cover
system. This is a barrier layer that reduces the downward movement of water to the waste
and the upward movement of gas out of the disposal cell. The as-built saturated hydraulic
conductivity (Ksat) of this layer is 5.00E-08 cm/s (Whetstone Associates, Inc.
[Whetstone], 2011, Table 15).
• Lower Radon Barrier: This layer consists of 12 inches of compacted clay with a low
hydraulic conductivity. This is a barrier layer placed directly above the waste that reduces
the downward movement of water. The as-built Ksat of this layer is 1.00E-06 cm/s
(Whetstone 2011, Table 15).

Unsaturated Zone Modeling for the Clive PA
23 October 2015 14

Figure 1. Evapotranspiration (ET) cover profile showing materials, observation nodes, and
root distribution used in the HYDRUS-1D models.

4.0 Unsaturated Zone and Shallow Aquifer
The following description of the Clive site hydrology is taken from the review prepared by
Envirocare (2004). The site is described as being located on lacustrine (lake bed) deposits
associated with the former Lake Bonneville. The sediments underlying the facility are principally
interbedded silt, sand, and clay. Sediments at the site are described by Bingham Environmental
(1991, 1994) and Envirocare (2000, 2004) as being classified into four hydrostratigraphic units
(HSU). Predominant sediment textural class, layer thickness range, and average layer thickness
for each unit are listed in Table 3. A diagram of the unsaturated zone is shown in Figure 2.
Unit 4: This unit begins at the ground surface and extends to between 6 ft and 16.5 ft below the
ground surface (bgs). The average thickness of this unit is 10 ft. This unit is composed of finer
grained low permeability silty clay and clay silt.
Unit 3: Unit 3 underlies Unit 4 and ranges from 7 ft to 25 ft in thickness. The average thickness
of this unit is 15 ft. Unit 3 is described as consisting of silty sand with occasional lenses of silty
to sandy clay.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 15
Unit 2: Unit 2 underlies Unit 3 and ranges from 2.5 ft to 25 ft in thickness. The average
thickness of this unit is 15 ft. Unit 2 is described as being composed of clay with occasional silty
sand interbeds. A structure map was prepared by Envirocare (2004, Figure 5) with contours
representing the elevations of the top of the unit. This map shows that the top surface of Unit 2
slopes downward gradually from east to west in the vicinity of the Class A South cell.
Unit 1: Unit 1underlies Unit 2 and is saturated beneath the facility, containing a locally confined
aquifer. Unit 1 extends from approximately 45 ft bgs and contains the deep aquifer. The deeper
aquifer is reported to be made up of lacustrine deposits consisting of deposits of silty sand with
some silty clay layers. One or possibly more silty clay layers overlie the aquifer (Bingham
Environmental 1994).
Table 3. Texture class, thickness range, and average thickness for the hydrostratigraphic
units underlying the Clive site.
Unit Sediment Texture Class Thickness
Range (ft)
Average
Thickness (ft)
4 silt and clay 6 – 16.5 10
3 silty sand with interbedded silt and clay layers 7 – 25 15
2 clay with occasional silty sand interbeds 2.5 – 25 15
1 silty sand with interbedded clay and silt layers >620 >620

Figure 2. Stratigraphic profile showing ET cover, waste zone, and stratigraphy below the
Federal DU Cell.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 16
The aquifer system in the vicinity of the Clive Facility is described by Bingham Environmental
(1991, 1994) and Envirocare (2000, 2004) as consisting of unconsolidated basin-fill and alluvial
fan aquifers. Characterization of the aquifer system is based on subsurface stratigraphy
observations from borehole logs and from potentiometric measurements. The aquifer system is
described as being composed of two aquifers: a shallow, unconfined aquifer and a deep confined
aquifer. The shallow unconfined aquifer extends from the water table to a depth of
approximately 40 ft to 45 ft bgs. The water table in the shallow aquifer is reported to be located
in Unit 3 on the west side of the site and in Unit 2 on the east side.
The deep confined aquifer is encountered at approximately 45 ft bgs and extends through the
valley fill (Bingham 1994). The boring log from a water supply well drilled in adjoining Section
29 indicated continuous sediments to a depth of 620 ft bgs (DWR 2014, water right number 16-
816 and associated well log 11293). The deepest portion of the basin in the Clive area is believed
to be north of Clive in Ripple Valley where the basin fill was estimated to be 3,000 ft thick (Baer
and Benson (as cited in Black et al., 1999)).
Deeper saturated zones in Unit 1 below approximately 45 ft bgs are reported to show higher
potentiometric levels than the shallow unconfined aquifer. Differences in potentiometric levels
are attributed to the presence of the Unit 2 clays. These observations are interpreted as indicating
that the shallow unconfined aquifer below the site does not extend into Unit 1 but is contained
within Units 2 and 3 (Bingham Environmental, 1994). The aquifer systems are described in more
detail in the Saturated Zone Modeling white paper.
Recharge to the shallow aquifer in the vicinity of Clive is thought to be composed of three
components: a small amount due to vertical infiltration from the surface; some small amount of
lateral flow from recharge areas to the east of the site; and the majority of recharge believed to be
from upward vertical leakage from the deeper confined aquifer (Bingham Environmental, 1994).
Average annual groundwater recharge from the surface in the southern Great Salt Lake Desert in
the precipitation zone typical of Clive was estimated by Gates and Kruer (1981). An estimated
300 acre-feet per year were recharged to lacustrine deposits and other unconsolidated sediments
over an area of 47,100 acres. This is a recharge rate of approximately 0.08 in/yr.
Groundwater recharge from lateral flow occurs due to infiltration at bedrock and alluvial fan
deposits away from the Site, which moves laterally through the unconfined and confined aquifers
(Bingham Environmental, 1994). This is evidenced by the increasing salinity of the groundwater
due to dissolution of evaporate minerals as water moves from the recharge area to the aquifers
below the Facility (Bingham Environmental, 1994). The majority of recharge to the shallow
aquifer is believed by Bingham Environmental (1994) to be due to vertical leakage upward from
the deep confined aquifer due to the presence of upward hydraulic gradients.
Deeper saturated zones in Unit 1 below approximately 45 ft bgs are reported to show higher
potentiometric levels than the shallow unconfined aquifer. Differences in potentiometric levels
are attributed to the presence of the Unit 2 clays (Bingham Environmental, 1994). Vertical
gradients between shallow and deeper screened intervals in the monitor well clusters were
calculated by Bingham Environmental (1994). An upward vertical gradient was observed
ranging in magnitude from 0.02 to 0.04 based on the distance between the screen centers. For a
Unsaturated Zone Modeling for the Clive PA
23 October 2015 17
vertical hydraulic conductivity of 1 x 10-6 cm/s (Bingham Environmental, 1994), this
corresponds to a recharge range from 0.25 in/yr to 0.5 in/yr.
5.0 Climate
Precipitation measurements taken at the site over the 17-year period 1992 to 2009 show a mean
annual value of 8.53 inches (21.7 cm) (Whetstone 2011). The distribution of precipitation
throughout the year is shown in Figure 3. Precipitation exceeds the annual average from January
through June and again in October and is below average for the remaining months. The nearest
National Oceanographic and Atmospheric Administration (NOAA) station with a long-term
record is located in Dugway, Utah, approximately 40 miles to the south. The mean annual
precipitation for the same 17-year period measured at the Dugway station is 8.24 inches (20.9
cm). A comparison of the Dugway precipitation data for the 17-year period 1992 to 2009 with
the long-term average for Dugway was made by Whetstone (2011). This comparison indicated
that annual average precipitation during this 17-year period has been greater than the long-term
average at Dugway by 8 percent. Whetstone (2011) concluded that simulations of cover
performance using precipitation data from this 17-year period might be overestimating this
component of the site water balance.

Figure 3. Monthly mean precipitation for the Clive Site and monthly mean pan evaporation
for the NOAA BYU station at Provo, Utah.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 18
The HYDRUS-1D modeling performed is based on the 17-year record for consistency with the
modeling results reported in Whetstone (2011). However, an additional 2 years of monthly
precipitation data are available from Meteorological Solutions (2012). The 19-year average
precipitation is 8.62 inches (21.9 cm). This difference is driven primarily by the 4.28 inches of
rainfall in May 2011. The small change in the overall average suggests that the modeling results
presented for this analysis would not change significantly if the 19-year precipitation record had
been used instead of the 17-year record.
The close correspondence between mean monthly temperatures measured at the Clive site and
the Dugway NOAA station was demonstrated by Whetstone (2011). Average monthly
temperatures measured at the Clive site over the 17-year period 1992–2009 ranged from 27.7 oF
in December to 79.5 oF in July.
Mean monthly values of pan evaporation measured at the BYU NOAA station in Provo, Utah,
over the period 1980 to 2005 are shown in Figure 3. Mean annual pan evaporation over this time
period is 49.94 inches. This station is located 83 miles to the southeast of the Clive facility. Data
from this station are used because pan evaporation data are not available for the Dugway station.
Although the Clive site is warmer than Provo during the summer months as shown in Figure 4,
the data provide insight into the water balance at the site.

Figure 4. Monthly mean temperatures for the Clive Site and the NOAA BYU station at
Provo, Utah.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 19
Assuming pan evaporation is approximately equal to potential evapotranspiration (PET), the
ratio of annual average precipitation to PET is 0.17. Although PET greatly exceeds precipitation
on an annual basis, monthly means in Figure 3 show precipitation exceeds PET from November
through February. This indicates the potential for recharge during these months under natural
conditions at the site. This is only a coarse measure, however, that neglects other factors. Actual
recharge is estimated through modeling of net infiltration.
6.0 Vegetation
Actual transpiration is dependent on the characteristics of the plant communities at the site.
Vegetation cover at the site is less than 20 percent, with soils supporting a range of native and
invasive shrubs. Excavations at the site have shown plant rooting depths extending to
approximately 31 inches (80 cm) below the ground surface, with root density decreasing with
depth (SWCA 2011).
Vegetation surveys of three field plots on or adjacent to the Clive Site were conducted by SWCA
(2011). The three low desert vegetation associations were characterized as: black greasewood,
Plot 3; halogeton-disturbed, Plot 4; and shadscale-gray-molly, Plot 5. The dominant shrub in Plot
3 was black greasewood with a percent cover of 4.5% and the dominant forb was halogeton with
a percent cover of 0.7%. In Plot 4 the dominant shrub was shadscale saltbush with a percent
cover of 2.3% and the dominant forb was halogeton with a percent cover of 3.3%. In Plot 5 the
dominant shrub was shadscale saltbush with a percent cover of 12.5% and the dominant forb was
halogeton with percent cover of 0.9%.
Black greasewood, shadscale saltbush, and halogeton are all classified as facultative halophytes
(Anderson, 2004; Simonin, 2001; and Pavek, 1992). Facultative halophytes are known to benefit
from high salt concentrations in their growth media (Shabala, 2013). Halophytes are able to
adjust to saline environments through various physiological adaptations such as
compartmentalization of ions in cell vacuoles, succulence, and the elimination of salt through
salt-secreting glands and bladders (Shabala, 2013). Optimal growth for halophytes has been
demonstrated by Shabala (2013) to occur in media with a concentration of approximately 50 mM
NaCl for monocots, and between 100 and 200 mM for dicots. For the optimum range for dicots
of 100 to 200 millimoles per liter (mM), the corresponding range of electrical conductivity for a
NaCl solution is 9.7 to 18.3 mmho/cm (CRC, 1985).
Depending on the extent of the area defined on and adjacent to the Clive Site, approximately 80
to 90 percent of the soils are mapped as the Skumpah silt loam on 0 to 2 percent slopes (NRCS,
2013). This Unit is characterized as having maximum salinity ranging from 8.0 to 16.0
mmhos/cm. The top end of this range of maximum salinity does not exceed the maximum of the
range of salinity considered optimum for halophyte growth of 18.3 mmho/cm. Given the
similarity in ranges of salinity in the surface soils at the Clive Site and for optimum halophyte
growth, the influence of the osmotic head reduction in the root-water uptake water stress
response function is considered negligible and was, consequently, not included in the model.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 20
7.0 Properties of Unit 3 and Radon Barriers
7.1 Laboratory Measurements
As shown in Figure 2 above, Unit 3 underlies the clay liner and extends into the shallow
unconfined aquifer. The upper and lower radon barriers in the cover system are constructed using
Unit 4 material. This section describes the development of material property distributions for
Unit 3 and for the engineered radon barrier layers constructed from Unit 4 material. Although the
properties developed in this section are used for the radon barriers they will be referred to in this
section as Unit 4.
The hydraulic properties for Units 3 and 4 are based on laboratory measurements by the
Colorado State University (CSU) Porous Media Laboratory for the moisture retention and
hydraulic conductivity of core samples from Units 3 and 4 at the Clive Site (Bingham
Environmental, 1991). Measurements of water retention as a function of matric pressure (called
suction head in this report) are available for the drying and wetting cycles. These measurements
were performed on four cores: GW19A B1 and GW17A B2 from Unit 4 (a silty clay), and
GW18 B4 and GW17A B5 from Unit 3 (a silty sand). Measurements of hydraulic conductivity as
a function of moisture content are available for three cores: GW19A B1, GW18 B4, and GW17A
B5. The focus in this work (and in previous work) is on the wetting cycle data because
infiltration after rain, which is a major driver for downward flow and transport, is driven by a
rewetting front that passes through the engineered cover, waste, and clay layers. Appendix A
documents the hydraulic data for Units 3 and 4, based on data reported in Bingham
Environmental (1991, pp. B 19 through B 31).
7.2 Grain Size Distributions for the Cores
Tables 3 and 4 summarize the grain size distributions according to the Unified Soil Classification
System (Bingham Environmental, 1991) for cores from Units 4 and 3, respectively. Table 4 is
sorted by increasing percent of clay plus silt content. Table 5 is sorted by increasing percent of
sand content. The four cores that were tested by CSU have the following properties:
• GW17A B2 has 55.6% clay, the highest measured clay content with a trace of sand in
Table 4 for Unit 4,
• GW19A B1 has 56.2% silt, the highest measured silt content with a trace of sand in Table
4 for Unit 4,
• GW18 B4 has 45.5% sand, the lowest measured sand content in Table 5 for Unit 3, and
• GW17A B5 has 83.3% sand, the highest measured sand content in Table 5 for Unit 3.
The core samples that were selected for testing span the extremes of the clay, silt, and sand
contents for Units 3 and 4. The core samples that were tested are in a bold font in Tables 3 and 4.
The water retention data are consistent with these material distributions, as shown in Figure 5. In
particular, the core that has the greatest clay content retains a greater moisture content than the
cores that are high in silt or sand at a given suction head, and the core that has the greatest sand
content demonstrates the abrupt changes in moisture content that are typical of a sandy material.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 21
Table 4. Grain size distributions for cores from Unit 4, a silty clay.
Well/Sample
No.
Depth
(ft)
Description %
Gravel
%
Sand
%
Silt
%
Clay
%
Clay +
Silt
Reference
I-3-50 (SE) 1.5 Silty Clay 0 39.3 60.7 Bingham 1994,
page 23
I-4-50 (SE) 10.5 Silty Clay 0 19.6 80.4 Bingham 1994,
page 32
I-3-50 (SE) 10.5 Silty Clay 0 16.6 83.4 Bingham 1994,
page 24
I-1-50 (NW) 7.5 Silty Clay 0 11.7 88.3 Bingham 1994,
page 13
GW-16/S-1 3 - 5 Brown Silty Clay
w/Trace Fine Sand
0.1 11.2 50.3 38.4 88.7 Bingham 1991,
page B-13
GW-19A/S-1 5-7 Brown Silty Clay
w/Trace Fine Sand
0 2.8 56.2 41.0 97.2 Bingham 1991,
page B-17
GW-17A/L-2 7-9.5 Brown Silty Clay
w/Trace Fine Sand
0 2.1 42.3 55.6 97.9 Bingham 1991,
page B-15
GW-18/B-1 5-6.5 Brown Silty Clay
w/Trace Fine Sand
0 2.0 49.9 48.1 98.0 Bingham 1991,
page B-16
I-4-50 (SE) 7.5 Silty Clay 0 1.2 98.8 Bingham 1994,
page 31
Cores in bold font were tested by CSU.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 22
Table 5. Grain size distributions for cores from Unit 3, a silty sand.
Well/Sample
No.
Depth
(ft)
Description %
Gravel
%
Sand
%
Silt
%
Clay
%
Clay +
Silt
Reference
GW-18/S-4 20-22 Brown Silty Fine
Sand w/Some Clay
0 45.5 38.7 15.8 54.5 Bingham 1991,
page B-16
I-1-50 (NW) 18.0 Silty Sand 0 48.2 51.8 Bingham 1994,
page 15
DH-48/B-2 17-19 Tan Silty Sand 0 55.5 44.5 Bingham 1994,
page B-11
GW-16/B-4 19.5-
21
Tan Silty Fine Sand 0 59.4 40.6 Bingham 1991,
page B-14
I-3-50 (SE) 19.5 Silty Sand 0 62.3 37.7 Bingham 1994,
page 26
GW-41/B-6 10-12 Tan Silty Sand 0 65.3 34.7 Bingham 1994,
page B-10
GW-41/B-9 16-18 Tan Silty Sand 0 66.3 33.7 Bingham 1994,
page B-10
I-1-50 (NW) 10.5 Silty Sand 0 66.6 33.4 Bingham 1994,
page 14
GW-19B/B-4 17-19 Tan Silty Fine Sand 0 66.7 33.3 Bingham 1991,
page B-18
GW-55/B-8 14-16 Tan Silty Sand 1.1 69.5 29.4 Bingham 1994,
page B-11
DH-33/L-7 16.5 Tan Silty Sand 0.1 72.9 27 Bingham 1994,
page B-9
GW-16/B-3 14.5-
16
Tan Silty Fine Sand 0.2 74.7 25.1 Bingham 1991,
page B-13
I-3-50 (SE) 15 Silty Sand 0 75.8 24.2 Bingham 1994,
page 25
I-4-50 (SE) 21 Silty Sand 0 76.4 23.6 Bingham 1994,
page 33
GW-16/B-2 9.5-11 Tan Silty Fine Sand 1.6 79.8 18.6 Bingham 1991,
page B-13
Unsaturated Zone Modeling for the Clive PA
23 October 2015 23
Well/Sample
No.
Depth
(ft)
Description %
Gravel
%
Sand
%
Silt
%
Clay
%
Clay +
Silt
Reference
GW-19A/S-3 15-16 Brown Silty Fine
Sand
0 82.0 18 Bingham 1991,
page B-17
GW-17A/L-5 19.5-
22
Brown Silty Fine
Sand w/Trace Clay
0 83.8 8.4 7.8 16.2 Bingham 1991,
page B-15
GW-19B/L-5 22-
24.5
Tan Silty Fine Sand 0 83.8 16.2 Bingham 1991,
page B-18
Cores in bold font were tested by CSU.

Figure 5. Comparison of water retention data (wetting cycle) for four core samples
Unsaturated Zone Modeling for the Clive PA
23 October 2015 24
7.3 Soil Material Properties
Particle density ρs is defined as the ratio of the mass of the solid to the volume of the solid:
ρs = Msolid / Vsolid.
Particle density depends on the chemical composition and crystalline structure of the mineral
particles. Particle density is not influenced by particle size, packing arrangement, or pore space.
Dry bulk density ρb is defined as the ratio of the mass of dried alluvium to its total volume,
ρb = Msolid / Vtotal.
For a dried sample, Vtotal = Vsolid + Vgas.
Porosity, ϕ, (often also denoted as n) is the relative pore volume of the medium,
(Vliquid + Vgas )/ (Vsolid + Vliquid + Vgas).
For a dry sample, porosity is Vgas / (Vsolid + Vgas). Total porosity can be determined from dry bulk
density and particle density by ϕ = 1 – ρb / ρs. Therefore, relating these equations,
ϕ=1– ρb /ρs= (ρs - ρb )/ρs = [Msolid /Vsolid –Msolid /(Vsolid + Vgas)]/( Msolid/Vsolid =Vgas /( Vsolid+Vgas ).
The structure of coarse dry alluvium is generally single grained. The actual packing arrangement
depends on grain size distribution, grain shape, and the processes under which the alluvium was
deposited. The grain size distribution can consist of a single grain size (monodisperse) or
multiple grain sizes (polydisperse). The packing arrangements of spherical grains of uniform size
can be represented by models for regular packing that allow the calculation of the spacing of
layers, the volume of a unit cell, and thus the bulk density. Although monodisperse systems are
idealizations of natural porous materials such as alluvium, calculated relationships between
particle density and bulk density gives some insight into potential particle density—bulk density
correlation. The unit cell volume, bulk density, and porosity are given in Table 6 below for five
models of regular packing of uniform spheres.
Table 6. Theoretical porosities based on particle packing geometry.
Model Unit Cell Volume
(R is grain radius)
Bulk Density Porosity
simple cubic 8R3 πρs/6 47.64
cubic tetrahedral 4√3 R3 πρs/3√3 39.54
tetragonal sphenoidal 6R3 2 πρs/9 30.19
pyramidal 4√2R3 πρs/3√2 25.95
tetrahedral 4√2R3 πρs/3√2 25.95
Unsaturated Zone Modeling for the Clive PA
23 October 2015 25
These calculations show that the bulk density of a volume of monodisperse spheres of constant
particle density depends on the packing arrangement. Thus, correlation between particle density
and bulk density would only be expected for a sample characterized by a single packing
arrangement.
Polydisperse systems are more complex with grains of smaller radii filling in the pore spaces
between larger grains. The increase in bulk density due to infilling by smaller particles depends
on the grain size distribution. Natural materials are more likely to be characterized by a range of
particle sizes leading to many diverse packing arrangements. The large range of possible packing
arrangements in coarse alluvium makes a physically based correlation between particle density
and bulk density unlikely.
Given the conclusion that particle density and bulk density are not physically dependent and
given the need to restrict the sampling of material properties and moisture content parameters to
physically meaningful and consistent values, the following approach was taken:
1. Separate up-scaled distributions for Units 3 and 4 for saturated water content and residual
water content are estimated from borehole water retention curve and hydraulic
conductivity data. This estimation approach is detailed in subsequent sections.
2. Porosity is assumed to be equal to the saturated water content.
3. Based on particle density data presented in Table 7 and best professional judgment, a
constant value of 2.65 g/cm3 was chosen for particle density for both Units 3 and 4, and
the frost protection layer.
4. Based on bulk density data presented in Table 7 and best professional judgment, an up-
scaled distribution for bulk density was specified as a normal distribution with a mean of
(1- porosity) times particle density and a standard deviation of 0.1. This was applied to
both Units 3 and 4, and the frost protection layer.
This approach allows the uncertainty in water content and bulk density to be modeled while
maintaining a physically coherent probabilistic unsaturated zone model.
Table 7. Bulk density, porosity, and calculated particle density data from water retention
experiments.
Borehole Unit Bulk Density
(g/cm)
Porosity Calculated Particle Density
(g/cm3)
GW18-B4 3 1.567 0.409 2.65
GW17A-B5 3 1.673 0.32 2.46
GW19A-B1 4 1.397 0.473 2.65
GW17A-B2 4 1.326 0.505 2.68
from CSU Porous Media Laboratory
Unsaturated Zone Modeling for the Clive PA
23 October 2015 26
7.4 Soil Moisture Content
The flow of water in porous media occurs in response to a gradient in the total potential energy
of water. The total potential can be composed of a number of components but this analysis will
be restricted to gravitational and matric potentials. Water potential components are often
expressed in units of energy per unit weight rather than units of energy per unit mass. When the
quantity of water is expressed as a weight, the units of potential are defined in terms of head. The
gravitational potential refers to the energy of water with respect to reference elevation and is
written here as Z. Although not a formal definition, the matric potential relates to the energy of
the tension imposed on the pore water by the soil matrix. Matric potential is a negative value and
is written here as ψ. The total potential is then H = ψ + Z.
Steady-state fluid flow in an unsaturated medium is defined by the Buckingham-Darcy equation
(Jury and Horton, 2004, p. 95). In the following discussion this equation will be referred to
simply as the Darcy equation. The one-dimensional form of Darcy’s equation for unsaturated
flow is given by Fayer (2000, Eqns. 4.2 and 4.5):
𝑞=−𝐾!(𝜓)∂𝐻
∂𝑧 (1)
where
q is the flux of liquid per unit area,
KL is the unsaturated conductivity as a function of the matric head ψ,
H is the matric plus gravitational potentials [cm], and
z is the depth below ground surface [cm].

It is convenient to define two sign conventions for the total potential (Fayer 2000, page 4.2):
(1) the z-coordinate is zero at the soil surface and positive downward. With this convention, the
gravitational head in the soil, which is defined as the elevation of a point with respect to the soil
surface, is negative and defined as -z; and (2) the suction head, h, is the negative of the matric
potential or matric head, ψ. With this convention, the suction head, h, is always greater than zero
for an unsaturated soil. It follows that
𝐻=𝜓+𝑍=−(ℎ+𝑧) (2)
and the flux is then given by
𝑞=𝐾!ℎ∂ℎ
∂𝑧+1 (3)
The unsaturated conductivity, KL, is formulated based on the Brooks-Corey (1964) representation
for moisture content as a function of suction head
𝛩=ℎ
ℎ!
!!
for                            ℎ>ℎ!
                                   =1              for                                0 ≤ℎ≤ℎ!
(4)
Unsaturated Zone Modeling for the Clive PA
23 October 2015 27
where
Θ is the effective saturation,
h is the suction head (cm),
hb is the bubbling pressure head (cm) at which moisture first drains from the
material, and
l is a constant that is fit to data.
Alternatively, expressed in terms of the fractal dimension, D
𝛩=ℎ
ℎ!
!!!
for                    ℎ>ℎ!
                                       =1                    for                      0 ≤ℎ≤ℎ!
(5)
The suction head is positive for an unsaturated material and 0 at saturation. Θ, the effective
saturation, is defined as
𝛩=𝜃−𝜃!
𝜃!−𝜃!
(6)
where
θ is the moisture content,
θr is the residual moisture content, and
θs is the saturated moisture content.

Combining Equations
𝜃=𝜃!+𝜃!−𝜃!
ℎ
ℎ!
!!
(7)
This equation can then be fit to core data.
Alternatively, expressing in terms of D and assuming
𝜃=𝜃!+(𝜃!−𝜃!)ℎ
ℎ!
!!!
(8)
Using the Mualem theory for predicting hydraulic conductivity (Mualem 1976), the unsaturated
hydraulic conductivity is defined as
𝐾!=𝐾!𝛩!!!! (9)
Substituting Equation 6 into Equation 9 gives:
𝐾!=𝐾!
𝜃−𝜃!
𝜃!−𝜃!
!!!
! (10)
Unsaturated Zone Modeling for the Clive PA
23 October 2015 28
Setup (e.g. Unit 3):
1. from 4 measurements estimate mean and standard error for porosity (φ ) and θr, use these
as priors for θs and θr (assumes θs = φ ).
2. for each borehole core there are 2 separate measurements:
1. moisture content, θ ; and suction head, h
2. moisture content, θ ; and hydraulic conductivity KL
3. estimate hb, D, θs, θr , and Ks as described below.
Here is the Brooks-Corey θ ~ f (h) equation:
𝜃=𝜃!+(𝜃!−𝜃!)ℎ
ℎ!
(!!!)
(11)
Here is KL ~ f ( θ )
𝐾!=𝐾!
𝜃−𝜃!
𝜃!−𝜃!
!(!!!/(!!!))
(12)
where the data are
θ the water content,
h is the suction head (cm),
KL is hydraulic conductivity (cm/sec),

and the parameters to be fit are

hb is the air entry pressure head (cm),
D is the soil fractal dimension,
θs is the saturated water content,
θr is the residual water content,
τ is the Mualem empirical parameter = 2,
KS is saturated hydraulic conductivity (cm/sec).

Typically these relationships are fit using non-linear least squares. However for these boreholes
the least squares optimization had trouble converging and the uncertainty in parameter estimates
was difficult to estimate. To allow combining of information across the available borehole
moisture content and hydraulic conductivity datasets and to provide an estimate of the
uncertainty in these parameter estimates, a Bayesian Markov Chain Monte Carlo (MCMC)
simulation approach was taken that allows the parameters to be constrained via prior
distributions and generates parameter posterior distributions. This also allows the two sets of
information from a borehole to be combined as well as allowing for combining information
across boreholes for a unit (borehole data are presented in Appendix A).
Unsaturated Zone Modeling for the Clive PA
23 October 2015 29
In a Bayesian approach sources of information on model parameters can be combined through a
prior distribution or through a data likelihood. The priors integrate expert judgment and scientific
knowledge while the likelihood integrates information available in observed data. In effect, the
priors can be used to constrain the results parameter distribution to physically meaningful values.
The priors listed below (Equations 13–19) are all uniform distributions. As such they are
relatively non-informative, which allow the data to determine the distribution and also constrain
the parameter values to a physically meaningful range.
𝑝(𝜃!)=𝑈[0.3,0.55] (13)
𝑝(𝜃!)=𝑈[0.001,0.2] (14)
𝑝(ℎ!)=𝑈[1,500] (15)
𝑝(𝐷)=𝑈[1,2.999] (16)
𝑝(𝜎)=𝑈[0.001,1000] (17)
𝑝(𝐾!)=𝑈[10e −10,10e −3] (18)
𝑝(𝜎!!)=𝑈[1e −9,1e −4] (19)
The likelihood based on the moisture content matrix pressure data:
𝑝(𝜃!,ℎ!,𝐷,𝜎|𝜃!"#$!!"#!,𝜃!"#$!!"#!,ℎ!"#$!!"#!,ℎ!"#$!!"#!)=
𝑁!"#$!!"#!𝜃!+(𝜃!−𝜃!)ℎ!"#$!!"#!
ℎ!
(!!!)
,𝜎
𝑁!"#$!!"#!𝜃!+(𝜃!−𝜃!)ℎ!"#$!!"#!
ℎ!
(!!!)
,𝜎
(20)

The likelihood based on the moisture content hydraulic conductivity data:

𝑝(𝜃!,𝜃!,𝐷,𝐾!,𝜎!!|𝜃!"#$!!"#!,𝜃!"#$!!"#!,𝐾!!"#$!!"!!,𝐾!!"#$!!"#!)=
𝑁!"#$!!"#!𝐾!
(𝜃−𝜃!)
(𝜃!−𝜃!)
!(!!!/(!!!))
,𝜎!!
𝑁!"#$!!"#!𝐾!
(𝜃−𝜃!)
(𝜃!−𝜃!)
!(!!!/(!!!))
,𝜎!!
(21)
Markov Chain Monte Carlo (MCMC) simulation of the joint distribution defined by equations
13-21 was used to generate samples from the marginal parameter distributions for the moisture
content and hydraulic conductivity models. Results for Units 3 and 4 are presented in the
following sections.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 30
7.4.1 Unit 3 Brooks-Corey Parameters
The MCMC sampling using likelihoods incorporating the two Unit 3 borehole cores resulted in
the following marginal parameter distributions:
𝑝(ℎ!)=𝑁[𝑚𝑒𝑎𝑛=8.85,𝑠𝑑=0.929] (22)
𝑝(𝐷)=𝑁[𝑚𝑒𝑎𝑛=2.73,𝑠𝑑=5.21e −3] (23)
𝑝(𝐾!)=𝑁[𝑚𝑒𝑎𝑛=5.14e −05,𝑠𝑑=5.95e −6] (24)
𝑝(𝜃!)=𝑁[𝑚𝑒𝑎𝑛=0.393,𝑠𝑑=6.11e −03] (25)
𝑝(𝜃!)=𝑁[𝑠ℎ𝑎𝑝𝑒=6.78e −3,𝑠𝑐𝑎𝑙𝑒=2.05e −3] (26)
Significant correlations from these simulations were found between D and hb (-0.85) and
between Ks and D (-0.98).
7.4.2 Unit 4 Brooks-Corey Parameters
The MCMC sampling using likelihoods incorporating the two Unit 4 borehole cores resulted in
the following marginal parameter distributions:
𝑝(ℎ!)=𝑁[𝑚𝑒𝑎𝑛=104.,𝑠𝑑=1.72] (27)
𝑝(𝐷)=𝑁[𝑚𝑒𝑎𝑛=2.81,𝑠𝑑=9.93e −5] (28)
𝑝(𝐾!)=𝑁[𝑚𝑒𝑎𝑛=5.16e −05,𝑠𝑑=5.97e −7] (29)
𝑝(𝜃!)=𝑁[𝑚𝑒𝑎𝑛=0.428,𝑠𝑑=9.08e −3] (30)
𝑝(𝜃!)=𝑁[𝑠ℎ𝑎𝑝𝑒=0.108,𝑠𝑐𝑎𝑙𝑒=8.95e −4] (31)
Significant correlations from these simulations were found between D and hb (-0.66) and
between Ks and D (-0.37).
8.0 Properties of Upper Cover Layers
Upper cover layers include the surface, evaporative zone, and frost protection layers. The surface
and evaporative zone layers are constructed from Unit 4 material. These layers will be
revegetated so the objective in construction will be to make their properties similar to that of
undisturbed Unit 4 silty clay. As a result, the porosity of these layers will be greater than the
porosity of the clay liner and radon barriers that will be more highly compacted. Uncertainty in
the porosity and bulk density for surface and evaporative zone layers was estimated using a
distribution (mean and standard error) for the saturated water content taken from the Rosetta
database of hydraulic parameters for the textural class of silty clay (Schaap 2002). This
distribution was normal with a mean of 0.481 and a standard deviation of 0.015.
The frost protection layer is modeled as a combination of sand, silt, and gravel. Uncertainty in
the porosity and bulk density for this layer was estimated using a distribution (mean and standard
Unsaturated Zone Modeling for the Clive PA
23 October 2015 31
error) for the saturated water content taken from the Carsel and Parrish (1988) database of
hydraulic parameters for the textural class of sandy loam. A sandy loam was chosen because it
represented a coarse-grained material with some silt and clay. This distribution was normal with
a mean of 0.41 and a standard deviation of 0.0026.
9.0 Properties of Waste
Test data are not available for the unsaturated porous media properties of the wastes. However,
the DU waste is expected to be in a powdered form or possibly compressed into small
“briquettes” for safety during transportation to the Clive facility. In this condition, the DU waste
will behave like a mixture of fine sand to fine gravel. Since there is so little information on
which to base material properties for the waste, it is assigned the properties of Unit 3.
Three types of waste materials are considered in the DU PA: Generic LLW, the UO3 waste from
the SRS, and the U3O8 wastes from the gaseous diffusion plants (GDPs) at Portsmouth, OH, and
Paducah, KY. The generic LLW is used only as an inert filler in the model, with no inventory,
and is assumed to simply have the properties of local silty sandy soil: Unit 3.
The uranium oxide wastes, both UO3 and U3O8, will be disposed in an indeterminate mix of
materials, including containers (55 gallon drums and DU cylinders of various types) and possibly
concrete, grout, bulk LLW, and local soils as backfill. This complex mix of heterogeneous
materials is not modeled at this point, and the assumption is made instead that the overall
material properties are again simply that of local silty sandy soil: Unit 3.
So, in summary, all waste materials in the Clive DU PA Model are assumed to have the same
physical properties as Unit 3 soils.
10.0 Properties of the Clay Liner
The Liner is constructed of compacted local clay, Unit 4 material. Porosity and bulk density
values for the clay liner are assumed to be the same as the Radon Barrier Clays, as these clays
are all compacted, unlike the surface and ET layer Unit 4 material. Brooks-Corey parameters
were assigned to be the same as Unit 4, as described in Section 7.4.2. The distribution for
saturated hydraulic conductivity was developed using the design value from Table 8 of
Whetstone (2007) for the clay liner of 1 × 10-6 cm/s as the geometric mean of a lognormal
distribution. A geometric standard deviation of 1.2 was chosen to provide an approximate order
of magnitude variation above and below the geometric.
11.0 Properties of the Unsaturated Zone below the Clay Liner
The Federal DU Cell is constructed by excavating through Unit 4, and into the top of Unit 3. The
entire unsaturated zone below the embankment, from the bottom of the clay liner to the top of the
saturated zone, is modeled as Unit 3 material, sharing all the properties and characteristics of
Unit 3 as outlined in this white paper. The saturated zone is modeled as Unit 2 (see the Saturated
Zone Modeling white paper). In the GoldSim PA Model, this zone below the embankment is
called the “Unsat zone” and does not include overlying waste and cover materials. It is part of
both the top slope and side slope columns.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 32
The thickness of the Unsat zone below the Federal DU Cell is determined by the difference in
average elevations of the bottom of the clay liner and the water table. The clay liner is uniformly
about 60 cm (2 ft) thick by design, though the bottom of the waste cell has a gentle slope to it as
documented in the Embankment Modeling white paper.
A distribution for the thickness of the unsaturated zone was established based on measurements
for groundwater wells, engineering drawings for the Federal DU Cell (see the Embankment
Modeling white paper), and consideration of the accuracy of the elevation measurements. The
four wells are selected from a map of wells (Figure 7 in Bingham Environmental, 1991): GW
19A, GW 25, GW 27, and GW-60, since the location of these four wells bound the Class A
waste cell. Each groundwater well is in the vicinity of one of the four corners of the Federal DU
Cell, so their measurements are treated as approximations to the water table elevation at the four
corners. These water table elevations are also used to establish the distributions for the thickness
of the saturated zone, and are documented in the Saturated Zone Modeling white paper.
12.0 Modeling of Net Infiltration and Water Content for the
Clive DU PA Model
Steady-state water infiltration rates and water contents for the cover layers required as input for
the Clive DU PA GoldSim model were calculated from a regression model developed from
infiltration modeling using the HYDRUS-1D software package. This section describes the
abstraction of the HYDRUS-1D results into the probabilistic framework employed by GoldSim.
12.1 Description of HYDRUS
HYDRUS-1D was selected for simulating the performance of the ET cover proposed for the DU
waste cell. The HYDRUS-1D platform was selected for this project because of its ability to
simulate processes known to have a significant role in water flow in landfill covers in arid
regions. HYDRUS includes the capabilities to simulate:
• water flow in variably saturated porous media,
• material hydraulic property functions,
• atmospheric surface boundary conditions including precipitation and evapotranspiration,
• root water uptake, and
• free-drainage boundary conditions.

The flow component of unsaturated flow and transport software packages with atmospheric
boundary conditions such as HYDRUS solve modified forms of the Richards equation for
variably saturated water flow. The flow equation incorporates a sink term to account for water
uptake by plant roots. HYDRUS can be applied to one-, two-, and three-dimensional problems.
The HYDRUS software includes grid generators for structured and unstructured finite element
meshes. Programs such as HYDRUS require detailed data to represent the atmospheric boundary
conditions and plant responses that are the dominant influences on flow in the cover in arid and
semi-arid conditions. These programs use the infiltration capacity of the soil at any time as
calculated in the model to partition precipitation into infiltration and overland flow. HYDRUS
has been used for many applications for unsaturated zone modeling and has received numerous
Unsaturated Zone Modeling for the Clive PA
23 October 2015 33
favorable reviews such as Scanlon’s (2004) review of HYDRUS-1D, Diodato’s (2000) review of
HYDRUS-2D, and McCray’s (2007) review of the most recent program, HYDRUS (2D/3D).
HYDRUS-1D was selected for simulating flow in the Federal DU Cell ET cover since previous
numerical modeling of flow in the similar ET cover design for the Class A West cover
demonstrated that subsurface lateral flow was not significant (EnergySolutions, 2012). To test
the importance of 2-D flow effects in the ET cover design, 2-D transient flow simulations were
conducted for representative sections of the cover. The approach taken was to model a section of
the side slope in two dimensions. Representative hydraulic properties were assigned to the ET
cover layers and the models were run with daily atmospheric boundary conditions for 100 years.
Root water uptake was modeled assuming the roots extended to the bottom of the evaporative
zone layer and that rooting density decreased with depth.
The results of these 2-D simulations demonstrated that water flow in the cover system for both
designs is predominantly vertical with no significant horizontal component. These results
demonstrate that 1-D models can be used to provide a defensible analysis of cover performance
for the ET cover design due to the lack of lateral flow.
HYDRUS-1D models were developed for the evapotranspiration cover designs for the DU waste
cell (Figure 1). Model development requires construction of a computational grid based on the
geometry of the model domain. Hydraulic properties for each layer required for the model are
available from previous studies at the site or can be estimated from site-specific measurements
such as particle size distributions. HYDRUS requires daily values of precipitation, potential
evaporation, and potential transpiration to represent the time-variable boundary conditions on the
upper surface of the cover. Representative boundary conditions were developed from records of
nearby meteorological observations. Parameters for describing root water uptake were available
from the literature.
HYDRUS implements the soil-hydraulic functions of van Genuchten (1980), who used the
statistical pore-size distribution model of Mualem (1976) to obtain a predictive equation for the
unsaturated hydraulic conductivity function in terms of soil water retention parameters. The
expressions of van Genuchten (1980) are given by
𝜃ℎ=
𝜃!+𝜃!−𝜃!
1 +𝛼ℎ!!ℎ<0
𝜃!ℎ≥0
(32)
𝐾(ℎ)=𝐾!𝑆!![1 −1 −𝑆!
!!
!
]! (33)
where
𝑚=1 −1/𝑛, 𝑛>1 (34)
The above equations contain five independent parameters: θr, θs, α, n, and Ks. The pore-
connectivity parameter “l” (lower-case L) in the hydraulic conductivity function was estimated
Unsaturated Zone Modeling for the Clive PA
23 October 2015 34
(Mualem, 1976) to be about 0.5 as an average for many soils. The value for l is commonly taken
to be 0.5, and this value was used for all simulations for all soil types. The effective saturation,
Se, is identical to Θ in Equation 6.
12.2 Conceptual Model
Recharge is an important process in controlling the release of contaminants to the groundwater
pathway. Site characteristics influencing movement of water from precipitation through the
vadose zone to the water table at the Clive Site include climate, soil characteristics, and native
vegetation. Engineered barriers are used at the Clive Site to control the flow of water into the
waste. A hydrologic model of the waste disposal system must realistically represent
precipitation, the source of water to the system, runoff, evaporation, transpiration, and changes in
storage to estimate the flow through the system. Under natural conditions plants remove water
from the upper soil zone through root uptake and transpiration, reducing the water available for
seepage deeper into the profile. The same processes occur in an engineered cover layer that has
been revegetated. Seepage through a cover system can occur when soils become wet enough to
increase their conductivity to water. Cover surface layers with adequate storage capacity can
hold the water in the near surface until it can move back into the atmosphere through
evaporation, reducing the seepage of water to the waste. These processes would be expected to
show temporal variability at the Clive Site on the time scale of minutes to hours in the near
surface and days to years deeper in the disposal cell. Processes that tend to change cover
properties such as plant and animal activity and climate influences (e.g. frost heave, erosion) are
expected to be slowed by the effects of aeolian deposition.
12.3 Climate and Vegetation Parameters
Infiltration of precipitation, surface runoff, and evaporation under time-varying climate
conditions are modeled by HYDRUS. The data required includes daily values of precipitation,
potential evaporation, and potential transpiration to represent the time-variable boundary
conditions on the upper surface of the cover. The location of nearby meteorological stations and
the time period of available records were discussed in Section 5. The long-term evaluation period
for this analysis makes it necessary to generate a representative climate record with a longer term
than the existing data.
The WGEN model (Richardson and Wright 1984) was used to generate a 100-year synthetic
precipitation record for the site. The WGEN model is a component of the HELP model
(Schroeder et al. 1994a, 1994b). A 100-year precipitation record was generated using the
monthly average values from measurements at the site based on 17 years of observations. This
100-year record is shown in Figure 6. The annual mean was 8.42 inches (21.38 cm/yr) with a
maximum daily precipitation of 1.09 inches (2.77 cm).
Daily potential evapotranspiration (PET) was calculated with values of daily maximum (Tmax),
minimum (Tmin), and mean (Tmean) temperatures and extraterrestrial radiation using the
Hargreaves method (Neitsch et al. 2005). This approach is used extensively and is documented in
the HYDRUS manuals (Šimůnek et al. 2009). Using the Hargreaves method, PET is calculated
as
Unsaturated Zone Modeling for the Clive PA
23 October 2015 35
 λ𝐸!=0.0023 ∗𝐻!∗𝑇!"#−𝑇!"#!/!∗(𝑇!"#$+17.8) (35)
where
λ latent heat of vaporization [MJ kg-1],
E0 potential evapotranspiration [mm d-1],
H0 extraterrestrial radiation [MJ m-2 d-1],
Tmax maximum air temperature for the day [°C],
Tmin minimum air temperature for the day [°C],
Tmean mean temperature for the day [°C].

Monthly mean values for Tmax and Tmin based on a 30-year record are available from the Dugway,
Utah, NOAA station (WRCC 2012). Monthly average temperatures were used from this long-
term record in HELP to provide daily 100-year records for Tmax and Tmin. Tmax ranged from 14.7
to 110.7°F with a mean of 66.4 oF. Tmin ranged from -9.1 to 75.3°F with a mean of 36.5°F. Tmean
ranged from 2.8 to 93°F with a mean of 51.4°F. Daily maximum and minimum air temperatures
for a 100-year record are shown in
Figure 7. Daily PET values for a 100-year record were then calculated from these temperature
data using the Hargreaves method described above. The daily 100-year PET record is shown in
Figure 8.
The HYDRUS atmospheric boundary condition requires that potential soil evaporation and
potential transpiration be specified separately. Potential evaporation (Ep) and potential
transpiration (Tp) can be calculated from PET using the Beer-Lambert law (Varado et al. 2006;
Wang et al. 2009). This calculation requires an estimate of the vegetation leaf area index (LAI).
The leaf area index is the one-sided active leaf area per unit ground surface area. Using the Beer-
Lambert law
𝑇!=PET ∗1 −exp −𝑎!"∗𝐿𝐴𝐼
𝐸!=𝑃𝐸𝑇∗exp −𝑎!"∗𝐿𝐴𝐼 (36)
where the abl coefficient accounts for radiation intercepted by vegetation and is given the default
value of 0.5 (Varado et al. 2006).
A single LAI value of 0.082 was used for all the HYDRUS-1D simulations. This value was
provided by Goodman (1973) for the total yield (all spp.) for a mixed vegetation plot for the
month of April. The Goodman (1973) study was located in the Curlew Valley, UT, portion of the
glacial Lake Bonneville, located approximately 75 miles north of the Clive Site.
Root water uptake depends on the estimation of daily potential transpiration (described above),
the depth of the rooting zone, the variation of root density with depth, and the parameters used to
describe the water stress function. Measurements of rooting depth and root distribution were
Unsaturated Zone Modeling for the Clive PA
23 October 2015 36
made in two excavations by SWCA (2011). Rooting depths and density for the two most
prevalent species are shown in Figure 9.
Root distribution was modeled as extending into the frost protection layer with a maximum depth
of 31 inches (80 cm). Root density was modeled as decreasing linearly with depth.
The van Genuchten S-shaped model (van Genuchten, 1987) was used to model root water
uptake. In this model the actual root water uptake is given by the potential transpiration
multiplied by a water stress response function. For soil water pressures above the wilting point
the water stress response function is given by
𝛼(ℎ,ℎ!)=1
1 +ℎ+ℎ!ℎ!"
! (37)
where h is the soil pressure head, hφ is the osmotic head, and h50 and p are parameters. Given the
discussion in Section 6 on osmotic potential, the osmotic stress is assumed to be negligible for
these simulations, so hφ is zero. The parameter h50 corresponds to the pressure head at which
water uptake is reduced by 50 percent. A value of -200 cm was used for these simulations. A
HYDRUS default value of 3 was used for the exponent p. The water stress response function
with these parameters is shown in Figure 10.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 37

Figure 6. 100-year daily precipitation record generated from monthly average values of
daily measurements at the site based on 17 years of observations.

Figure 7. 100-year daily Tmax and Tmin record generated from a 30-year record available
from the Dugway, Utah NOAA station.

Figure 8. 100-year daily potential evaporation generated using the Hargreaves method.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 38

Figure 9. Root density with depth at the Clive Site for Shadscale and Black Greasewood
[SWCA 2011].

Figure 10. Water stress response function for root water uptake model.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 39
12.4 Model Geometry
The HYDRUS-1D models were constructed using the maximum number of nodes (1001), with
nodes evenly spaced down a 152-cm deep profile such that each node had a 0.152-cm spacing.
The top slope of the waste cover was simulated, with a slope set to 2.4% (1.4 degrees).
The HYDRUS-1D model geometry for all simulations is shown in Figure 1, which shows the
thickness of each material layer in the ET cover. Observation nodes were placed in the center of
each layer, with an additional node at the bottom boundary.
12.5 Material Properties
The hydraulic properties for each of the layers within the ET cover for the HYDRUS-1D
modeling are summarized in Table 8. The source of each hydraulic property for each layer is
provided in this table. Bingham (1991, p. B-20) is the source of hydraulic properties measured on
core samples collected at the Clive Site. Whetstone (2011, Table 17) is the source of the design
specifications for the Ks of the two radon barriers. For the frost protection layer, hydraulic
properties for a sandy loam were used and taken from the HYDRUS-1D pull-down menu, which
includes properties from the database of Carsel and Parrish (1988). Table 8 also identifies several
properties as “Variable.” These properties were associated with an infiltration and water content
model based on statistical distributions of hydraulic properties developed to provide net
infiltration and volumetric water content to the GoldSim DU PA Model.
The nine cores sampled from Unit 4 at the site and listed in Table 4 are all described as a silty
clay texture. However, hydraulic properties were available for only two of the nine cores (see
Appendix A). To provide a better estimate of the uncertainty of the hydraulic properties of Unit 4
that compose the surface and evaporative zone layers of the ET cover, the α and n values were
taken from the distributions (mean and standard deviation) for each parameter from the Rosetta
database of hydraulic parameters for the textural class of silty clay (Schaap 2002). The standard
deviations were converted to standard errors by dividing by √n where n is the number of
samples, 28 in this case. The distributions for α and n are summarized here:
A: log (base-10) mean = -1.79, log (base-10) standard error = 0.121 (38)
N: log (base-10) mean = 0.121, log (base-10) standard error = 0.019 (39)
where α = 10A and n = 10N. The units of α are 1/cm and n is dimensionless.
Normal distributions of A and N were sampled 50 times, and then transposed from log space by
calculating 10A, and 10N for the 50 sampled values. In addition, N was truncated such that it
could not be less than 0.0 (required in Equation 32).
An expanded assessment of the performance of the radon barriers was made possible by
developing a distribution for the saturated hydraulic conductivity (Ks) of the radon barriers to use
for the modeling. The Ks values for the radon barriers were sampled from a distribution
developed from a minimum value of 4.32×10-3 cm/day corresponding to the design specification
for the upper radon barrier (Whetstone 2007, Table 8), and 1st, 50th, and 99th percentile values of
0.65 cm/day, 3.8 cm/day, and 52 cm/day, respectively, which are from a range of in-service
Unsaturated Zone Modeling for the Clive PA
23 October 2015 40
(“naturalized”) clay barrier Ks values described by Benson et al. (2011, Section 6.4, p. 6-12). A
shifted lognormal distribution was fit to the 1st, 50th, and 99th percentiles, and the minimum value
of 4.32E-3 cm/day was used as a shift. The resulting distribution is:
𝐾𝑠  ~  𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙𝑔𝑒𝑜𝑚.𝑚𝑒𝑎𝑛:3.37  𝑐𝑚/𝑑𝑎𝑦,𝑔𝑒𝑜𝑚.𝑠𝑑:  3.23  𝑐𝑚/𝑑𝑎𝑦, with a right shift of
0.00432 cm/day
For all HYDRUS simulations, the same Ks value was applied to both the upper and lower radon
barriers.
Correlations between α and n were investigated by analyzing the combinations of α and n for the
12 textural classes in Rosetta (Schaap, 2002), and no correlations were evident. There were also
no statistically significant correlations between Ks and α or n.
The developed 50 sets of uncertain parameters for α, n, and Ks were then used as hydraulic
property inputs to 50, 1000-year simulations using HYDRUS-1D.
The 50 HYDRUS-1D simulations were conducted to evaluate the uncertainty in infiltration flux
into the waste zone, and water content within each ET cover layer as a function of hydraulic
property uncertainty. While it is preferable to sample distributions of uncertain hydraulic
parameters for all waste layers, a modified approach was used where van Genuchten (1980) α
and n parameters for the surface and evaporative zone layers, and the Ks of the radon barriers
were randomly sampled from distributions for each, to generate 50 parameter sets of α, n, and Ks.
These 50 parameters sets are shown in Table 9.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 41
Table 8. Hydraulic properties of topslope cover used for HYDRUS modeling.
Layer Parameter Value Units Source Notes
Surface θr 0.111 [-] Rosetta database for
Silty clay

θs 0.4089 [-] Rosetta database for
Silty clay
Adjusted for 15% gravel
α Variable 1/cm Rosetta database See Table 8.
n Variable [-] Rosetta database See Table 8.
Ks 4.46 cm/day Table 1, Unit 4
surface and ET
layers

Evaporative
Zone
θr 0.111 [-] Rosetta database for
Silty clay

θs 0.481 [-] Rosetta database for
Silty clay

α Variable 1/cm Rosetta database See Table 8.
n Variable [-] Rosetta database See Table 8.
Ks 4.46 cm/day Table 1, Unit 4
surface and ET
layers

Frost
Protection
θr 0.065 [-] Carsel and Parrish
(1988)
Šimůnek and Šejna (2011),
Table 7, Sandy Loam
θs 0.41 [-] " "
α 0.075 1/cm " "
n 1.89 [-] " "
Ks 106.1 cm/day " "
Upper Radon
Barrier
θr 0.1 [-] Whetstone (2011)
Table 15, p. 25
Compacted Unit 4 borrow
soils
θs 0.432 [-] " "
α 0.003 1/cm " "
n 1.172 [-] " "
Ks Variable cm/day Whetstone (2011)
Table 15, p. 25;
Benson et al., (2011)
See Table 8.
Lower Radon
Barrier
θr 0.1 [-] Whetstone (2011)
Table 15, p. 25
Compacted Unit 4 borrow
soils
θs 0.432 [-] " "
α 0.003 1/cm " "
n 1.172 [-] " "
Ks Variable cm/day Whetstone (2011)
Table 15, p. 25;
Benson et al., (2011)
See Table 8.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 42
Table 9. Parameter sets of van Genuchten α and n, and Ks used for HYDRUS modeling.
Replicate α (1/cm) n Ks (cm/d)
1 0.013091 1.359766 3.285794
2 0.014317 1.371086 12.497148
3 0.010969 1.357776 3.736272
4 0.018089 1.342287 5.162964
5 0.019954 1.316356 2.325706
6 0.010797 1.279182 4.168751
7 0.016004 1.396199 2.595876
8 0.012816 1.308572 0.838501
9 0.014744 1.372326 2.055096
10 0.014791 1.360367 5.052781
11 0.020639 1.276159 3.234858
12 0.019501 1.327968 2.194697
13 0.015766 1.334194 1.307280
14 0.019048 1.373538 1.719640
15 0.018539 1.338996 1.635838
16 0.017045 1.267606 1.749758
17 0.019983 1.413655 5.126214
18 0.012494 1.326223 10.753272
19 0.019503 1.356646 1.845171
20 0.028186 1.378016 3.643845
21
 0.010929
 1.244500
 6.738214

22
 0.020973
 1.282170
 6.943533

23
 0.017971
 1.372107
 1.099495

24
 0.016549
 1.467656
 3.648668

25
 0.012120
 1.330512
 6.338780

26
 0.011984
 1.382991
 0.792890

27
 0.012782
 1.382761
 7.005276

28
 0.017094
 1.275082
 4.768674

Unsaturated Zone Modeling for the Clive PA
23 October 2015 43
Replicate α (1/cm) n Ks (cm/d)
29
 0.013032
 1.382671
 9.861743

30
 0.024165
 1.349583
 7.758327

31
 0.016054
 1.386282
 1.478986

32
 0.024889
 1.310637
 2.501489

33
 0.017247
 1.320670
 2.459523

34
 0.014338
 1.265236
 66.503659

35
 0.016633
 1.286526
 31.683457

36
 0.014343
 1.383885
 1.005712

37
 0.022207
 1.236303
 3.733521

38
 0.012511
 1.317326
 4.565641

39
 0.018395
 1.333180
 6.167757

40
 0.013735
 1.294514
 2.206236

41
 0.015243
 1.229113
 4.106400

42
 0.018063
 1.282922
 3.299065

43
 0.017010
 1.326811
 32.484809

44
 0.020072
 1.323515
 31.128008

45
 0.015950
 1.357247
 2.326748

46
 0.018944
 1.252554
 2.976567

47
 0.015677
 1.301147
 1.241111

48
 0.024293
 1.287802
 4.617869

49
 0.018819
 1.264178
 0.737824

50
 0.017781
 1.263628
 2.880623

Unsaturated Zone Modeling for the Clive PA
23 October 2015 44
12.6 Boundary Conditions
The atmospheric boundary condition in HYDRUS provides the top boundary of the model with
daily values of precipitation, potential evaporation, and potential transpiration at the soil-air
interface. A free drainage boundary condition is applied at the bottom of the model as a unit
gradient boundary condition where the water flux across the boundary is equal to the flux due to
gravity at the water content of the material. HYDRUS calculates and reports surface runoff,
evaporation, and infiltration fluxes for the atmospheric boundary and fluxes for the free drainage
boundary.
12.7 Initial Conditions
An initial pressure head condition of -200 cm was applied to the entire model domain. This
pressure head corresponds to a slightly unsaturated condition for the fine-grained materials. The
model is deliberately run for a long period of time (1,000 years) in order to reach a near-steady
state net infiltration rate that is not influenced by the initial conditions.
12.8 Cases Simulated
As discussed above, 50 HYDRUS-1D simulations were conducted to evaluate the uncertainty in
infiltration flux into the waste zone, and water content within each ET cover layer as a function
of hydraulic property uncertainty. The fifty simulations with varying van Genuchten α and n and
Ks values are shown in Table 9.
Simulations were run for 1,000 years. The mean of the fluxes into the top of the waste layer and
the mean water contents for the surface layer, evaporative zone, frost protection layer, upper and
lower radon barriers over years 900 to 1000 were calculated.
12.9 Model Results
The 50 HYDRUS-1D simulations resulted in a distribution of average annual infiltration into the
waste zone, and average volumetric water contents for each ET cover layer. Infiltration flux into
the waste zone ranged from 0.0067 to 0.18 mm/yr, with an average of 0.024 mm/yr, and a log
mean of 0.018 mm/yr for the 50 replicates.
Multiple linear regression models were fit to the HYDRUS infiltration results, and water
contents for each ET cover layer. The general form of the regression was:
𝑌=β!+β!∗𝐾!+β!∗α +β!∗𝑛 (40)
Net infiltration is in units of mm/yr and volumetric water content is dimensionless. For the net
infiltration flux regressions, Ks was dropped as a predictor due to poor fit of the models. The
regressions were fit using the ‘lm()’ function in the software package R, which uses least squares
optimization for estimating parameters. All values of β coefficients are summarized in Table 10.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 45
Table 10. Coefficients calculated from multiple linear regression models.
Coefficient βo β1 β2 β3

SurfaceWC
 0.48155
 0.00000
 0.54920
 -­‐0.20020

EvapWC
 0.57947
 0.00000
 0.73997
 -­‐0.24790

FrostWC
 0.04282
 0.00000
 0.43297
 0.01617

Rn1WC
 0.14737
 -­‐0.00076
 1.70702
 0.06353

Rn2WC
 0.14740
 -­‐0.00076
 1.70648
 0.06351

Flux
 (mm/yr) -­‐0.32921
 N/A
 5.56826
 0.19538

13.0 Implementation in GoldSim
Average annual infiltration flux into the waste zone, and the volumetric water content of each ET
cover layer was calculated using Equations 41 and 42, developed from HYDRUS-1D simulation
results. GoldSim calculates values using Equations 41 and 42 for each ET cover layer.
The resulting equations for solving infiltration and water content in GoldSim become:
𝐼𝑛𝑓𝑖𝑙=β!+β!∗α +β!∗𝑛 (41)
𝑊𝐶=𝛽!,!+𝛽!,!∗𝐾!+𝛽!,!∗𝛼+𝛽!,!∗𝑛  (42)
where Infil is net infiltration in mm/yr, WC is average volumetric water content, and β values are
linear regression coefficients with the subscript i corresponding to Surface, Evaporative zone,
Frost protection, Upper radon barrier, and Lower radon barrier layers. The necessary
distributions in GoldSim are VG_logAlpha, VG_logN, and RnBarrierKsat_Natdist.
α and n are calculated from values drawn from distributions using:
𝛼=10VG_logAlpha,𝑤ℎ𝑒𝑟𝑒  VG_logAlpha  ~  𝑁𝑜𝑟𝑚𝑎𝑙𝑚𝑒𝑎𝑛:  −1.79,𝑠𝑒:  0.121 and (43)
𝑛=10VGlogN,𝑤ℎ𝑒𝑟𝑒  VGlogN~  𝑁𝑜𝑟𝑚𝑎𝑙𝑚𝑒𝑎𝑛:  0.121,𝑠𝑒:  0.019 .  (44)
Ks is sampled using:
RnBarrierKsat_Natdist =
𝐾!,~𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙𝑔𝑒𝑜𝑚.𝑚𝑒𝑎𝑛:3.37  𝑐𝑚/𝑑𝑎𝑦,𝑔𝑒𝑜𝑚.𝑠𝑑:  3.23  𝑐𝑚/𝑑𝑎𝑦, with right shift
of 0.00432 cm/day.  
(45)
Volumetric water contents for the waste, clay liner, and native Unit 3 soil below the Federal DU
Cell at the EnergySolutions Clive Facility are calculated using a numerical method. The
development and testing of this method implemented in the GoldSim DU PA Model are
described in the Appendix B.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 46
14.0 Contaminant Fate and Transport in Porous Media
Once all the hydraulic properties and states have been developed, as in the previous sections, we
can turn to transport mechanisms within the various porous media. Contaminant transport takes
place in fluid phases—in the present case, this is limited to air and water. Fluids move through
the pores by advection in response to fluid pressure gradients, carrying dissolved contaminants
with them. Fluids are also a medium for diffusive transport, in which contaminants move simply
in response to concentration gradients, and do not require movement of the fluid. Both these
processes occur simultaneously, along with all the other mechanisms identified in the model for
contaminant transport (radioactive decay and ingrowth, geochemical partitioning, biotically
induced transport, erosion, etc.). This section discusses advective and diffusive contaminant
transport mechanisms in fluids.
14.1 Porous Medium Water Transport
Water is a transport pathway considered at Clive, and the conceptual model includes the
advection of solutes in water moving down from the waste to the shallow aquifer as well as
diffusion of solutes in pore water.
14.1.1 Advection of Water
The flow of water is discussed at length in the previous sections of this document. Contaminant
transport in this flowing water is essentially passive, with solutes moving along with the fluid,
though of course concentrations are affected by other simultaneous processes.
14.1.2 Diffusion in Water
The Clive DU PA Model employs a modified version of GoldSim’s native diffusive flux links to
calculate diffusive fluxes in porous media. The modifications are necessary to account for
unsaturated media, since GoldSim assumes that porous media are saturated in its basic
implementation of diffusive flux calculations. The standard GoldSim diffusive flux mathematics
are covered in Appendix B of the GoldSim User’s Guide (GTG, 2011), and the modifications
that have been developed by Neptune are discussed in detail in the Neptune document entitled
Modeling Diffusion in GoldSim, but are also covered briefly here. The modifications required to
model diffusion in unsaturated media take two phenomena into consideration: 1) The diffusive
area is reduced by the saturation (with respect to air or water, whichever medium is of interest)
and 2) the diffusive length is increased to account for tortuosity in the respective medium.
If a porous medium contains only a single fluid phase, the diffusive area between two cells
containing that medium is simply the total area times the porosity, since the pores are occupied
by the fluid, and the diffusion takes place only in the fluid. In the case of two fluids, such as air
and water in unsaturated media, the diffusive area is further reduced, since the area of the fluid of
interest across the plane of diffusion is less. If we are interested in diffusion in the water phase,
for example, the area of water that intersects the plane is equal to the total area times the water
content, which equals the total area times the porosity times the saturation with respect to water.
If we are interested in diffusion in the air phase, we use the same construct, substituting air for
Unsaturated Zone Modeling for the Clive PA
23 October 2015 47
water. Because the diffusive area is always less, the diffusion in an unsaturated medium will
always be less than that in a fully saturated medium.
Diffusion in unsaturated media is also attenuated because of increased tortuosity. In any porous
medium, a diffusing solute must travel through pores, following a tortuous path that is always
longer than if it were traveling in a straight line. The ratio of the straight line distance to this
tortuous path is called the tortuosity. If the porous medium is unsaturated, this path becomes
even longer, since the three-dimensional shape of the fluid of interest gets even more tortuous.
This increases the diffusive length, which is used in calculating the concentration gradient. The
gradient in concentration of a solute is what drives diffusion.
14.1.3 Water Phase Tortuosity
Tortuosity is a term used to describe the resistive and retarding influence of pore structure for a
variety of transport processes (Clennell, 1997). Definitions of tortuosity are not consistent in the
literature and depend on the discipline and the particular transport process of interest. The
tortuosity τ for molecular diffusion in porous media can be written as the ratio of effective
diffusivity Deff to bulk diffusivity Dbulk, often seen in two forms:
𝜏!=𝐷!""
𝐷!"#$
(46)
or alternatively, if the measured porosity n is explicit (Clennell, 1997), as
𝜏!=𝐷!""
𝑛  𝐷!"#$
(47)
In this definition, consistent with the assumptions of GoldSim’s internal calculations, the value
of tortuosity varies between 0 and 1, with lower values indicating a longer path for porous
medium solute transport via diffusion.
For unsaturated systems, n is replaced in equation (47) by water content θw for water phase
diffusion, or by the volumetric air content θa for gaseous phase diffusion. The form shown in
equation (46) is found in Freeze and Cherry (1979) and Marsily (1986), while that in equation
(47) is used by Hillel (1980) and Koorevaar et al. (1983).
For consistency with GoldSim the second form is used. The equations for diffusive transport in
GoldSim explicitly specify the effective porosity (or in the case of unsaturated flow, water
content or air filled porosity) as in equation (47). For more information on the diffusive mass
flux equations in GoldSim, see Appendix B of the GoldSim User's Guide (GTG, 2011). In the
following sections, the equations from the literature have been converted where necessary to be
consistent with equation (47) so that they can be directly applied to GoldSim models.
Two options were considered for modeling liquid phase tortuosity in the models. The Millington-
Quirk model is commonly used to estimate tortuosity in non-fractured porous media (Millington
and Quirk, 1961). (See Jury and Horton, 2004, eq. 7.14, modified by division by water content
for consistency with GoldSim.) The water phase tortuosity τw is calculated as
Unsaturated Zone Modeling for the Clive PA
23 October 2015 48
𝜏!=𝐷!""
𝜃!  𝐷!"#$
=𝜃!!!
𝑛! (48)
Water phase tortuosity will be implemented in the Clive DU PA Model using the form shown in
equation (47). The exponents will be treated as distributions in order to allow the sensitivity
analysis to determine if the model is sensitive to the values of the exponents. The water content
exponent is described by a normal distribution with a mean of 7/3 and a standard deviation of
0.01 and the porosity exponent is described by a normal distribution with a mean of 2 and a
standard deviation of 0.01.
14.2 Porous Medium Air Transport
14.2.1 Advection of Air
Air-phase advection is not included in the Clive DU PA Model. It is assumed that the advective
flux of gases is negligible compared to the diffusive gas flux.
14.2.2 Diffusion in Air
Air-phase diffusion is included in the model, and this is the principal process by which gases are
moved. The “built-in” diffusion calculations in GoldSim are used to estimate diffusion in the air
phase. These gaseous diffusive fluxes are modified to handle the unsaturated porous media
(described above in Section 14.1.2), but also include a calibration to counteract numerical
dispersion for radon (discussed in the Radon Transport white paper), which at this time is the
only radionuclide that is considered to be present in the gaseous phase.
Diffusion in the air phase is modeled throughout the top slope column, bounded at the bottom by
the saturated zone, and at the top by the atmosphere. The bottom boundary condition is one of no
diffusion, since there is no air in the saturated zone to diffuse into, by definition. The boundary
condition at the top is effectively a zero-concentration sink, since the volume of air in the
atmosphere flowing over the embankment is sufficiently large that concentrations are kept much
lower than in the pore air of the cover and wastes below. In order to model this, the air directly
above the embankment is represented by an Atmosphere Cell Pathway element in GoldSim. The
volume of air is defined by a thickness times the area of each respective modeled column, and
this air volume is flushed out by the wind. The diffusive flux from the uppermost cover cell in
the column to the Atmosphere cell is defined by the diffusive area, as discussed above, and the
diffusive length, discussed in the following section. Since the atmosphere is not a porous
medium, a diffusive length unrelated to its thickness is adopted. Since the wind will maintain
low concentrations in the atmosphere, amounting to a zero-concentration boundary condition, the
choice of the parameters defining the Atmosphere is not expected to have much influence on the
diffusive flux from the embankment cover. Uncertainties have been included for these values, as
shown in Table 11, in order to evaluate the model’s sensitivity.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 49
Table 11. Atmosphere volume parameters for creating a surface boundary condition in the
porous medium air diffusion model.
Parameter Distribution Units
Thickness of the atmosphere
layer
N( µ=2.0, σ=0.5, min=Small, max=Large ) m
Wind speed N( µ=3.14, σ=0.5, min=Small, max=Large ) m/s
Atmospheric diffusion length N( µ=0.1, σ=0.02, min=Small, max=Large ) m

14.2.3 Air-Phase Tortuosity
A number of tortuosity models have been proposed for air phase diffusion in porous media.
Using the form for tortuosity shown in (42) above, models reviewed by Jin and Jury (1996)
include the Penman model (Penman, 1940) and two models attributed to Millington and Quirk.
In the Penman model, air phase tortuosity τa is a constant:
𝜏!=0.66. (49)
In the more commonly used Millington-Quirk model (MQ1), which is analogous to equation
(48), tortuosity is expressed as
𝜏!=𝜃!!!
𝑛! (50)
And, in an alternative Millington-Quirk model (MQ2) evaluated by Jin and Jury (1996),
tortuosity is expressed as
𝜏!=𝜃!
𝑛!! (51)
Note that as θa approaches n (e.g. as the porous medium becomes drier), τa approaches n1/3 for
both formulations (50) and (51).
An air-phase tortuosity model was developed by Lahvis et al. (1999) by calibrating a transport
model to steady-state gas concentration data obtained from seven column experiments using silt
and fine sand sediments. In this model, air phase tortuosity is dependent only on the volumetric
water content:
𝜏!=0.765 −2.02𝜃! (52)
Comparison of these models for alluvium with an effective porosity of 0.37 and tortuosity as
defined in equation (47) is shown in Figure 11. Due to the similarity of the Lahvis et al. (1999)
model to the MQ2 model over a wide range of volumetric water content, it will not be considered
further.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 50
The Penman and the two Millington-Quirk models were compared by Jin and Jury (1996) with
measured Deff /Dbulk ratios from six studies that included a total of approximately 50
measurements on predominantly agricultural soils. While this ratio corresponds to the definition
of tortuosity given in equation (47), it is useful in comparing the predictions of the various
models. Over the range of air phase porosity investigated (0.05 to 0.5), the Penman model tended
to overestimate tortuosity, while the MQ1 model in equation (50) underestimated tortuosity. Of
the three models, the MQ2 model given by (51) provided the best fit to the measured tortuosities.
A comparison of the Penman and Millington-Quirk models for a material with an effective
porosity of 0.37 is shown in Figures 11 and 12. Note that in both these figures, the points are
merely points of calculation, and do not represent data. The values produced by the Penman and
Millington-Quirk models converge for dry and wet conditions but diverge at intermediate values
of air porosity. Given its median behavior as seen in Figures 11 and 12, the alternative
Millington-Quirk model (MQ2, equation (51)) is used in the Clive DU PA Model.

Figure 11. Comparison of air-phase tortuosity models by Penman (equation (44)),
Millington and Quirk (MQ1, equation (45)), Millington and Quirk as modified by Jin
and Jury (1996) (MQ2, equation (46)), and Lahvis et al. (1999) (equation (47)).
Unsaturated Zone Modeling for the Clive PA
23 October 2015 51

Figure 12. Comparison of effective to bulk diffusivity ratios with air phase porosity for air
phase tortuosity models.
Tortuosity is implemented in the GoldSim model as a multiplier to the diffusive length, which is
defined for each Cell Pathway element using the common method of setting it equal to 1/2 the
cell length that is parallel to flow. In this case, that is the vertical dimension.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 52
15.0 References
Anderson, Michelle D. 2004. Sarcobatus vermiculatus. In: Fire Effects Information System,
[Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research
Station, Fire Sciences Laboratory (Producer). Available from:
http://www.fs.fed.us/database/feis/ [2013, November 7].
Benson, C.H., W.H. Albright, D.O. Fratta, J.M. Tinjum, E. Kucukkirca, S. H. Lee, J. Scalia, P.
D. Schlicht, and X. Wang. 2011. Engineered Covers for Waste Containment: Changes in
Engineering Properties & Implications for Long-Term Performance Assessment,
NUREG/CR-7028, Office of Research, U.S. Nuclear Regulatory Commission,
Washington, DC.
Bingham Environmental. 1991. Hydrogeologic report Envirocare Waste Disposal Facility South
Clive, Utah. October 9, 1991. Prepared for Envirocare of Utah. Salt Lake City, UT.
Bingham Environmental. 1994. Hydrogeologic report Mixed Waste Disposal Area Envirocare
Waste Disposal Facility South Clive, Utah. November 18, 1994. Prepared for Envirocare
of Utah. Salt Lake City, UT.
Black, B.D., B.J. Solomon, and K.M. Harty, 1999. Geology and geologic hazards of Tooele
Valley and the West Desert Hazardous Industry Area, Tooele Count, Utah, Special Study
96, Utah Geological Survey, Salt Lake City, Utah.
Brooks, R.J. and A. T. Corey, 1964. Hydraulic properties of porous media, Hydrology Paper 3,
Colo. State Univ., Fort Collins.
Carsel, R. F., and R. S. Parrish. 1988. Developing joint probability distributions of soil water
retention characteristics. Water Resources Research, 24(5):755-770, 1988.
Clennell, M.B. 1997. Tortuosity: a guide through the maze. in Developments in Petrophysics,
Lovell, M.A. and P.K. Harvey (eds). Geological Society Special Publication No. 122, pp.
299-344.
CRC. 1985. CRC Handbook of Chemistry and Physics, 66th Edition. R.C. Weast, M.J. Astle,
and W.H. Beyer (eds.). Osmotic parameters and electrical conductivities of aqueous
solutions. p. D-269. CRC Press Inc. Boca Raton, FL.
Diodato, D.M., 2000. Review: HYDRUS-2D, Ground Water 38:10–11.
EnergySolutions, 2012, Utah Low-Level Radioactive Material License (RML UT2300249)
Updated Site-Specific Performance Assessment, October 8, 2012, EnergySolutions, LLC,
Salt Lake City, UT
Envirocare of Utah, Inc. 2000. Revised hydrogeologic report for the Envirocare Waste Disposal
Facility Clive, Utah. Version 1.0. Envirocare of Utah, Inc. Salt Lake City, UT.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 53
Envirocare of Utah, Inc. 2004. Revised hydrogeologic report for the Envirocare Waste Disposal
Facility Clive, Utah. Version 2.0. Envirocare of Utah, Inc. Salt Lake City, UT.
Fayer, M.J., 2000. UNSAT-H Version 3.0: Unsaturated Soil Water and Heat Flow Model,
Theory, User Manual, and Examples. PNNL-13249. Pacific Northwest National
Laboratory, Richland, Washington. June, 2000.
Freeze, R.A and J.A. Cherry. 1979. Groundwater. Prentice-Hall, Inc., Englewood Cliffs, NJ.
Gates, J.S. and S.A. Kruer, 1981. Hydrologic reconnaissance of the Southern Great Salt Lake
Desert and summary of the hydrology of West-Central Utah, State of Utah Department of
Natural Resources, Technical Publication No. 71, Salt Lake City, UT.
GTG (GoldSim Technology Group), 2011. User's Guide: GoldSim Contaminant Transport
Module, GoldSim Technology Group, Issaquah, WA, December 2010.
Goodman, P.J. 1973. Physiological and Ecotypic Adaptations of Plants to Salt Desert Conditions
in Utah. Journal of Ecology, Vol. 61, No. 2, pp. 473-494.
Hillel, D. 1980. Fundamentals of Soil Physics. Academic Press Inc. San Francisco, CA.
Jin, Y., and W.A. Jury. 1996. Characterizing the Dependence of Gas Diffusion Coefficient on
Soil Properties, Soil Science Society of America Journal, Vol. 60, pp. 66-71.
Jury, W.A. and R. Horton. 2004. Soil Physics. 6th ed. John Wiley and Sons Inc. New Jersey.
Koorevaar, P., G. Menelik, and C. Dirksen. 1983. Elements of Soil Physics. Elsevier. New York,
NY.
Lahvis, M.A., A.L. Baehr, and R.J. Baker. 1999. Quantification of Aerobic Biodegradation and
Volatilization Rates of Gasoline Hydrocarbons Near the Water Table Under Natural
Attenuation Conditions. Water Resour. Res. v. 27, 753-765.
Marsily, G. de. 1986. Quantitative Hydrogeology. Academic Press Inc. San Diego, CA.
McCray, J., 2007. HYDRUS: Software review, Southwest Hydrol. 6(1):41.
Meteorological Solutions, 2012. January 2011 Through December 2011 and January 1993
through December 2011 Summary Report of Meteorological Data Collected at
EnergySolutions’ Clive, Utah Facility, Project No. 011110111, February 2012.
Millington, R.J., and J.P. Quirk. 1961. Permeability of porous solids. Trans. Faraday Society
(57) pp. 1200-1207
Mualem, Y. 1976. A New Model for Predicting the Hydraulic Conductivity of Unsaturated
Porous Media. Water Resources Research, Vol. 12, No. 3, pp. 513-522. June, 1976.
Neitsch, S.L., J.G. Arnold, J.R. Kinry, and J.R. Williams, 2005. Soil and Water Assessment Tool
Theoretical Documentation: Version 2005, Grassland, Soil and Water Research
Unsaturated Zone Modeling for the Clive PA
23 October 2015 54
Laboratory Agricultural Research Service and Blackland Research Center, Texas
Agricultural Experiment Station, Temple TX.
http://swatmodel.tamu.edu/media/1292/swat2005theory.pdf Accessed 7/24/2012.
NRCS. 2013. U.S. Department of Agriculture, Natural Resources Conservation Service. Web
Soil Survey. Available from:
http://websoilsurvey.sc.egov.usda.gov/App/HomePage.htm
[Accessed 2013, November 7].
Pavek, Diane S. 1992. Halogeton glomeratus. In: Fire Effects Information System, [Online].
U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available from:
http://www.fs.fed.us/database/feis/
Penman, H.L. 1940. Gas and vapor movements in the soil. I. The diffusion of vapors through
porous solids. Journal of Agricultural Science (30) pp. 437-462.
Richardson, C.W. and D.A. Wright, 1984, WGEN: A Model for Generating Daily Weather
Variables. U. S. Dept. of Agriculture, Agricultural Research Service, ARS-8.
Scanlon, B. 2004. Review of Hydrus-1D, Southwest Hydrology, 3(4):37.
Schaap, M. 2002. Rosetta: A computer program for estimating soil hydraulic parameters with
hierarchical pedotransfer functions. Available from:
http://ag.arizona.edu/research/rosetta/download/rosetta.pdf
Schroeder, P.R., N.M. Aziz, C.M. Lloyd, and P.A. Zappi. 1994a. The Hydrological Evaluation of
Landfill Performance (HELP) Model: User’s Guide for Version 3. EPA/600/R-94/168a.
U.S. Environmental Protection Agency Office of Research and Development.
Washington, D.C.
Schroeder, P.R., T.S. Dozier, P.A. Zappi, B.M. McEnroe, J.W. Sjostrom, and R.L. Peyton.
1994b. The Hydrological Evaluation of Landfill Performance (HELP) Model:
Engineering Documentation for Version 3. EPA/600/R-94/168b. U.S. Environmental
Protection Agency Office of Research and Development. Washington, D.C.
Shabala, S. 2013. Learning from halophytes: physiological basis and strategies to improve
abiotic stress tolerance in crops. Ann. Bot. doi: 10.1093/aob/mct205. Available from:
http://aob.oxfordjournals.org/content/early/2013/09/30/aob.mct205.full [2013, November
7].
Simonin, Kevin A. 2001. Atriplex confertifolia. In: Fire Effects Information System, [Online].
U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available from:
http://www.fs.fed.us/database/feis/ [2013, November 7].
Šimůnek, J., M. Šejna, H. Saito, M. Sakai, and M. Th. van Genuchten, 2009, The HYDRUS-1D
Software Package for Simulating the One-Dimensional Movement of Water, Heat, and
Unsaturated Zone Modeling for the Clive PA
23 October 2015 55
Multiple Solutes in Variably-Saturated Media, Department of Environmental Sciences,
University of California Riverside, Riverside, CA.
Šimůnek, J. and M. Šejna. 2011. Software Package for Simulating the Two- and Three-
Dimensional Movement of Water, Heat and Multiple Solutes in Variably-Saturated
Media: User Manual Version 2, PC-Progress, Prague, Czech Republic.
SWCA. 2011. Field Sampling of Biotic Turbation of Soils at the Clive Site, Tooele County, Utah,
January, 2011, SWCA Environmental Consultants, Salt Lake City, UT.
Utah Division of Water Rights, (DWR), water rights and well log database at
http://waterrights.utah.gov/wrinfo/query.asp. Accessed March 18, 2014.
van Genuchten, M.Th. 1980. A closed-form equation for predicting the hydraulic conductivity of
unsaturated soils. Soil Science Society of America Journal 44 (5): 892–898.
van Genuchten, M. Th. 1987. A numerical model for water and solute movement in and below
the root zone. Research Report No 121, U.S. Salinity laboratory, USDA, ARS, Riverside,
California, 1987.
Varado, N., I Braud, and P.J. Ross, 2005. Development and assessment of an efficient vadose
zone module solving the 1D Richards’ equation including root extraction by plants,
Journal of Hydrology, 323, 258-275.
Wang, T., V. A. Zlotnik, J. Šimunek, and M. G. Schaap, 2009. Using pedotransfer functions in
vadose zone models for estimating groundwater recharge in semiarid regions, Water
Resour. Res., 45, W04412, doi:10.1029/2008WR006903.
Western Regional Climate Center, 2012. Dugway, Utah, 30 year daily temperature and
precipitation summary. http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?ut2257.
Whetstone Associates, Inc., 2011. EnergySolutions Class A West Disposal Cell Infiltration and
Transport Modeling Report, dated November 28, 2011. Document Number
4104K111128.
Whetstone Associates, Inc., 2007. EnergySolutions Class A South Cell Infiltration and Transport
Modeling. December 7, 2007.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 56
Appendix A
Soil Moisture Data for Units 3 and 4
The data for soil moisture characteristics in Unit 3, a silty sand, and in Unit 4, a silty clay, are
reproduced in the following tables, and are based on testing performed by Colorado State
University (Bingham Environmental 1991, Appendix B, pages B 20 and B 26). Cores GW18 B4
and GW17A B5 are from Unit 3, and cores GW19A B1 and GW17A B2 are from Unit 4. Bulk
density is defined in the units of g/cm3. Conductivity data have units of cm/s.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 57

Unsaturated Zone Modeling for the Clive PA
23 October 2015 58
Appendix B
Runge-Kutta Method for Calculating Water Content
1. Purpose
This Appendix describes the development and testing of a numerical method implemented in the
GoldSim DU PA model for estimating the volumetric water content of the waste, clay liner and
native Unit 3 soil below the Federal DU cell at the EnergySolutions Clive Facility.
2. Method
The flow of water in porous media occurs in response to a gradient in the total potential energy
of water. The total potential can be composed of a number of components but this analysis will
be restricted to gravitational and matric potentials. Water potential components are often
expressed in units of energy per unit weight rather than units of energy per unit mass. When the
quantity of water is expressed as a weight, the units of potential are defined in terms of head.
The gravitational potential refers to the energy of water with respect to reference elevation and is
written here as Z. Although not a formal definition, the matric potential relates to the energy of
the tension imposed on the pore water by the soil matrix. Matric potential is a negative value
and is written here as ψ. The total potential is then H = ψ + Z.
Steady-state fluid flow in an unsaturated medium is defined by the Buckingham-Darcy equation
(Jury and Horton, 2004, p.95). In the following discussion this equation will be referred to
simply as the Darcy equation. The one dimensional form of Darcy’s equation for unsaturated
flow is given by Fayer (2000, Eqns. 4.2 and 4.5):
𝑞=−𝐾!(𝜓)∂𝐻
∂𝑧 (1)
where
q is the flux of liquid per unit area,
KL is the unsaturated conductivity as a function of the matric potential ψ,
H is the matric plus gravitational potentials [cm], and
z is the depth below ground surface [cm].

It is convenient to define two sign conventions for the total potential (Fayer 2000, page 4.2): (1)
the z-coordinate is zero at the soil surface and positive downward. With this convention, the
gravitational head in the soil, which is defined as the elevation of a point with respect to the soil
surface, and negative and defined as -z; and (2) the suction head, h, is the negative of the matric
potential or matric head, ψ. With this convention, the suction head, h, is always greater than zero
for an unsaturated soil. It follows that:
𝐻=𝜓+𝑍=−(ℎ+𝑧) (2)

Unsaturated Zone Modeling for the Clive PA
23 October 2015 59
and the flux is then given by
q =K!(h)∂h
∂z +1 (3)
The unsaturated conductivity, KL, is formulated based on the Brooks-Corey representation for
moisture content as a function of suction head
Θ =h
h!
!!
for h >h!,
                   =1              for 0 ≤h ≤h!
(4)
where
Θ is the effective saturation,
h is the suction head (cm),
hb is the bubbling pressure head (cm) at which moisture first drains from the
material, and
l is a constant that is fit to data.

Alternatively, expressed in terms of the fractal dimension, D
Θ =        h!!!
h!
for h >h!,
=1    for 0 ≤h ≤h!
(5)
The suction head is positive for an unsaturated material and 0 at saturation. θs, the effective
saturation, is defined as
Θ =θ −θ!
θ!−θ!
, (6)
where
Ɵ is the moisture content,
Ɵr is the residual moisture content, and
Ɵs is the saturated moisture content.

Combining Equations
θ =θ!+θ!−θ!∗h
h!
!!
(7)
This equation can then be fit to core data. Alternatively, expressing in terms of D and assuming
θ =θ!+(θ!−θ!)h
h!
(!!!)
(8)
Unsaturated Zone Modeling for the Clive PA
23 October 2015 60
Using the Mualem theory for predicting hydraulic conductivity (Mualem 1976), the unsaturated
hydraulic conductivity is defined as:
K!=K!Θ!!!
!. (9)
Substituting Equation 6 into Equation 9 gives:
K!=K!
θ −θ!
θ!−θ!
!!!
!. (10)
The computational method implemented in the Clive DU PA Model solves Equation 3 for steady
state flow at constant infiltration flux, q. (At steady state, the vertical infiltration flux must be
constant in all layers of the cell below the radon barriers, which includes the waste, the clay liner,
and the unsaturated zone.) No iterations are required with the selected solution technique. The
approach in the Clive DU PA Model differs from the solution technique in the UNSAT-H code,
which solves the transient (unsteady) equation for one-dimensional unsaturated flow and iterates
to a steady state solution with constant infiltration rate.
3. Darcy Equation Solution by the Runge-Kutta Method
Equation 3 is a nonlinear, first order differential equation for the suction head that can be solved
by numerical approximation. The Runge-Kutta method is attractive for this application because
it allows variable spacing (i.e., variable Δz) between nodes, because it is highly stable, and
because it does not require iteration to converge to a solution. Equation 3 can be rewritten as a
first order differential equation in the form h′ = f(h) :
𝜕ℎ
𝜕𝑧=𝑞
𝐾!(ℎ)−1 (11)
A second order Runge-Kutta solution for this first order differential equation is given by
Abramowitz and Stegun (1970, Section 25.5.6):
ℎ!!!=ℎ!+𝑘!+𝑘!
2 +𝑂(ℎ!), (12)
with
𝑘!=𝛥𝑧𝑞
𝐾!(ℎ!)−1 (13)

𝑘!=𝛥𝑧𝑞
𝐾!(ℎ!+𝑘!)−1 (14)
and
𝛥𝑧=𝑧!!!−𝑧!. (15)
Unsaturated Zone Modeling for the Clive PA
23 October 2015 61
Equations 12 through 15 define a procedure for calculating hn+1 from the known values of hn, Δz,
and the (constant) infiltration flux, q. These equations constitute a predictor-corrector
calculation, where k1 is the predictor and k2 is the corrector. No iteration is involved in this
solution because Equations 13, 14, and 12 can be solved sequentially for each node of the grid,
beginning with the lowest node at the top of the water table with h = 0 (because the suction head
is zero for a saturated soil) and KL = Ks, and integrating upward through the various unsaturated
soil layers. Stable solutions do require a finer discretization than the layers that are defined for
the 1-D columns used in the Clive PA model.
The value of Δz does not have to be constant over the domain of integration, and has been
adjusted to provide reasonable accuracy where the head gradient is greatest. In practice, these
regions occur at the capillary fringe just above the water table and at the interface between the
clay liner and waste. The value of Δz has to be small enough that the predictor step (Equation 13)
does not generate a value of k1 that is so large and negative that (hn + k1) becomes negative.
Suction head is always positive, and KL(hn + k1) in Equation 14 cannot be evaluated for negative
values of (hn + k1). In practice, an initial node spacing of 2 cm provides a stable solution in
Unit 3, directly above the water table, for the infiltration fluxes of interest. However, an initial
node spacing of 0.1 mm was required to provide a stable solution in the waste, directly above the
clay liner, at high infiltration rates. This fine spacing is required because the head gradient at the
interface between the waste and clay liner is quite large. A node spacing of 25 cm provides a
stable solution in the main body of the waste and in Unit 3 where the head gradients are smaller.
A constant node spacing of 15 cm provides adequate resolution in the clay liner and in the upper
and lower radon barrier. Solutions at these variable grid spacings are mapped to the Clive DU PA
Model’s regular grid that is used to represent wastes and other layers, in the top slope and side
slope columns.
4. Verification of the Runge-Kutta Method
The UNSAT-H modeling program (Fayer 2000) has been used to analyze infiltration through the
Federal DU cell at the EnergySolutions facility (Whetstone 2007). A model built with UNSAT-H
predicted moisture content and suction head from the radon barriers in the cover downward
through the waste, clay liner, and Unit 3 silty sand to the top of the aquifer (Whetstone 2007,
Section 4 and Table 17). The results from the UNSAT-H calculation for the top and side slope
models have been used to verify the steady state unsaturated flow solutions with the Runge-Kutta
method outlined in Section 3.
The UNSAT-H calculations are based on a van Genuchten representation for soil moisture
content and for soil hydraulic conductivity. For verification purposes, the Runge-Kutta solution
was programmed into a spreadsheet using the identical van Genuchten models as UNSAT-H.
The Runge-Kutta verification used the same total thicknesses for the radon barriers, waste, clay
liner, and Unit 3 sand as the UNSAT-H model, but the spacing of individual nodes (i.e., the
values of Δz) is different. Table 1 summarizes the thicknesses of the major components.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 62

Table 12. Layer thicknesses and coordinates for top slope validation calculations.
Layer Thickness z-Coordinate
Upper Radon Barrier 1 ft (30.48 cm) 0 to 30.48 cm
Lower Radon Barrier 1 ft (30.45 cm) 30.48 cm to 60.96 cm
Waste 45 ft (1371.6 cm) 60.96 cm to 1432.56 cm
Clay Liner 2 ft (60.96 cm) 1432.56 cm to 1493.52 cm
Unit 3 Silty Sand 10.8 ft (329.2 cm) 1493.52 cm to 1822.7 cm

Figure 1(a) compares the calculated values for moisture content from the UNSAT-H model
(Whetstone 2007, Table 17) and from the Runge-Kutta solution for the top slope model with an
infiltration rate of 0.276 cm/yr. Both solutions encompass the radon barriers, the waste, the clay
liner beneath the waste, and Unit 3 from the bottom of the clay liner to the top of the water table.
The results are essentially identical, providing validation for the Runge-Kutta method. Figure
1(b) provides a more detailed comparison of moisture content near the bottom and top of the clay
liner, again demonstrating the close agreement between the UNSAT-H model and the Runge-
Kutta method.
A similar comparison was also performed for the side slope model with an infiltration rate of
0.595 cm/yr. The side slope model is similar to the top slope model, except the average waste
thickness is 5.64 m (18.5 ft) rather than 13.7 m (45 ft). Figures 2(a) and 2(b) again demonstrate
the close agreement between the UNSAT-H model and the Runge-Kutta method.
The calculated values for suction head from the UNSAT-H model and from the Runge Kutta
method were also compared for the top and side slope models. The suction head profiles in the
radon barriers, waste, clay liner and Unit 3 are shown in Figure 3 for the top and side slope
models. A qualitative comparison between the Runge-Kutta solution and the UNSAT-H results
was performed because the UNSAT-H data for suction head were not tabulated, only presented
graphically (Whetstone 2007, Figures 8 and 9). The comparison of suction heads from both
methods again demonstrates that the Runge-Kutta solution is in excellent agreement with the
results from the UNSAT-H model.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 63

a) Comparison of moisture content in Unit 3, clay liner, waste, and radon barriers

b) Comparison of moisture content in and adjacent to the clay liner
Figure 13. Comparison of the Runge-Kutta and UNSAT-H solutions for top slope model.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 64

(a) Comparison of moisture content in Unit 3, clay liner, waste, and radon barrier

(b) Comparison of moisture content in and adjacent to the clay liner
Figure 14. Comparison of the Runge-Kutta and UNSAT-H solutions for side slope model.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 65

(a) Top Slope Model

b) Side Slope Model
Figure 15. Suction head profiles in Unit 3, clay liner, waste, and radon barriers for the top
slope and side slope models.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 66
The results in Figures 1 and 2 highlight three important features of the response of the Federal
DU cell to infiltration. First, the clay liner has a moisture content of about 0.42 (see Figures 1(b)
and 2(b)) in the top and side slope models. This value is just below θs, which is 0.432 for the van
Genuchten model. The radon barriers have slightly higher moisture contents, approximately
0.425 to 0.43 (see left-hand side of Figures 1(a) and 2(a)), again just below the saturated
moisture content of 0.432. These results confirm that the clay liner and radon barriers remain
very close to saturation for either model (top or side slope) and for two different infiltration rates
(0.276 cm/yr or 0.595 cm/yr) in the Federal DU cell. Second, the waste drains to a relatively low
moisture content, on the order of 0.06 for either slope model and infiltration rate. This behavior
is consistent with the low moisture retention of a sandy material. Finally, suction head shows
greater differences than moisture content for the top and side slope models. The suction head is
more directly dependent on flow rate (see Equation 11) than moisture content, and the factor of
two difference in the flow rates for the top and side slope models is the probable cause of the
differences in Figure 3(a) and 3(b).
5. Implementation in the DU PA Model
The Runge-Kutta method has been incorporated into the Clive PA model for infiltration through
the radon barriers, waste, clay liner and Unit 3 of the Federal DU cell at the EnergySolutions
facility. The PA model of the Federal DU cell has a number of differences with the verification
calculations discussed in the previous section. The major differences are as follows:
1. The moisture retention and hydraulic conductivity of the radon barriers and clay liner are
defined by a Brooks-Corey/Mualem model that is based on the test data from Colorado
State University (Bingham Environmental 1991, Appendix B, pages B-20 and B-26) for
Unit 4 cores GW17A B2 and GW19A B1.
2. The moisture retention and hydraulic conductivity of the Unit 3 silty sand between the
clay liner and water table are defined by a Brooks-Corey/Mualem model that is based on
the test data from Colorado State University (Bingham Environmental 1991, Appendix B,
pages B-20 and B-26) for Unit 3 cores GW18 B4 and GW17A B5.
Integration of the Darcy equation from node n, with a known value of the suction head, hn, and a
known value of Δzn = zi+1 – zn, to node n+1 is based on the following sequential steps:
1. Calculate the moisture content, θn, corresponding to the suction head, hn. The calculation
of θn, is based on Equations 4 and 6.
2. Calculate the conductivity, K(hn), based on the effective saturation, Θn, at θn. Equations 6
and 9 define the formulas.
3. Calculate k1 = Δzn(q/K(hn) – 1) (see Equations 13 and 15).
4. Calculate the trial value of the suction head, hn + k1.
5. Calculate the trial value of the moisture content, θ (hn + k1) using Equations 4 and 6.
6. Calculate the trial value of the conductivity, K(hn + k1), based on the effective saturation
at θ (hn + k1). Equations 6 and 9 in Section 2 define the formulas.
7. Calculate k2 = Δzn(q/K(hn + k1) – 1) (see Equations 14).
8. Calculate hn +1 = hn + (k1 + k2)/2 (see Equation 12).
Unsaturated Zone Modeling for the Clive PA
23 October 2015 67
Numerical testing demonstrated that the trial value of the suction head, hn + k1, can become
negative, leading to an undefined value for K(hn + k1). Negative values of K(hn + k1)occurred at
the interface between the waste and clay liner when the infiltration rate increased from 0.3 to 0.5
cm/yr for the as-designed cover to approximately 5 cm/yr. The numerical problem appears in the
waste, adjacent to its interface with the clay liner, because the gradient of suction head is greatest
at this location (for example, see Figure 3(a) at a depth of about 1,400 cm).
The verification testing in Section 3 used the following spacing for nodes in the waste, adjacent
to the clay liner: (1) 2 cm node spacing for the first five nodes in the waste, (2) 5 cm node
spacing for the next 4 nodes in the waste, and (3) 25 cm node spacing for all other nodes in the
waste. The GoldSim implementation of this solution uses a geometric spacing between the first
12 nodes in the waste, beginning with an initial spacing of 0.1 mm, which increases by a ratio of
approximately 1.93 for each subsequent node. The spacing between the 11th and 12th nodes is
0.135 m and the total width of the 12 nodes with geometric zoning is 0.281 m. All subsequent
nodes in the waste have a constant spacing of 0.281 m in the GoldSim implementation.
Numerical testing demonstrated that the geometric zoning produces stable solutions for the top
slope and side slope models with the Runge-Kutta method up to flow rates of 5 cm/year.
6. Numerical Testing of the Top Slope Model in GoldSim
Validation of a top slope infiltration model for the Federal DU cell was performed in GoldSim,
using the same Runge-Kutta method and the same descriptions of soil properties, providing a
direct comparison of results and a means of identifying errors in programming. Deterministic
calculations were performed with Brooks-Corey/Mualem models for the individual cores (Unit 4
core GW17A B2 or GW19A B1, and Unit 3 core GW17A B5 or GW18 B4) to compare
unsaturated flow conditions calculated using GoldSim. Stochastic calculations were performed
with GoldSim for 20 realizations using randomly sampled values for the Brooks-Corey/Mualem
input parameters for Units 3 and 4. The GoldSim results for Realization 18 were identical to a
calculation for Realization 18 to 5 or 6 significant digits. This testing also provided useful
insights into the range of conditions in the Federal DU cell during unsaturated flow.
Figures 4 and 5 compare the profiles for moisture content and suction head, respectively, in the
radon barriers, waste, clay liner, and Unit 3 for the four deterministic calculations that use Unit 3
(silty sand) properties for GW18 B4 or GW17A B5 and use Unit 4 (silty clay) properties for
GW17A B2 or GW19A B1. All calculations have an infiltration rate of 0.276 cm/yr (0.109
in/yr). These results confirm previous observations: (1) The moisture contents of the clay liner
and radon barriers remain close to saturation, and (2) the waste retains a low moisture content of
0.06. In addition, the suction heads in the radon barriers are identical because the hydraulic
conductivity is identical for either core (because conductivity was only measured for one of the
two cores).

Unsaturated Zone Modeling for the Clive PA
23 October 2015 68

Figure 16. Profiles of moisture content in Unit 3, clay liner, waste, and radon barriers for
the top slope model with 0.276 cm/yr infiltration.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 69

Figure 17. Profiles of suction head in Unit 3, clay liner, waste, and radon barriers for the
top slope model with 0.276 cm/yr infiltration.
Figures 6 and 7 compare the profiles for moisture content and suction head, respectively, in the
radon barriers, waste, clay liner, and Unit 3 for deterministic calculations that use soil properties
for GW17A-B5 (Unit 3) and GW17A-B2 (Unit 4) at three different infiltration rates: 0.168
cm/year, 0.276 cm/yr, and 5.0 cm/yr. In general, Figures 8 and 9 demonstrate that moisture
content is more sensitive to infiltration rate than to the differences between soil properties for the
various cores. The major difference in Figure 6 is the degree of drainage in the waste, with the
high infiltration rate increasing the retained moisture from 0.055 at 0.168 cm/yr to 0.084 at 5.0
Unsaturated Zone Modeling for the Clive PA
23 October 2015 70
cm/yr infiltration. The moisture content in the waste also shows a small oscillation between
0.082 to 0.086 at the 5.0 cm/yr infiltration rate. This could have be eliminated by having finer
spacing between the nodes in the waste, but the accuracy of the current solution is considered
more than adequate. Similar calculations were also performed for soil properties with GW17A-
B5 for Unit 3 and GW19A-B1 for Unit 4. The results are very similar to those shown in Figures
6 and 7 and are not repeated here.

Figure 18. Profiles of moisture content in Unit 3, clay liner, waste, and radon barriers for
the top slope model with different infiltration rates.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 71

Figure 19. Profiles of suction head in Unit 3, clay liner, waste, and radon barriers for the
top slope model with different infiltration rates.

Figures 8 through 12 compare the time dependent moisture content at the mid-points of Unit 3,
of the clay liner, of the waste, of the lower radon barrier, and of the upper radon barrier,
respectively, for a GoldSim calculation with 20 realizations and randomly sampled soil
properties for Units 3 and 4. The duration of each realization is 3,000 years and the lower filter
layer is assumed to become degraded at 2,640 years after closure for test purposes.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 72
The results in Figures 8 through 12 confirm the observations from the previous calculations: (1)
the moisture contents in the clay liner, lower radon barrier, and upper radon barrier remain close
to saturation (note the expanded vertical scale for Figures 11 and 12), and (2) the waste drains to
low moisture content, 0.03 to 0.08, for these 20 realizations, and (3) the moisture content in Unit
3 also has a limited range of 0.13 to 0.20 for the infiltration rates generated by the cover
infiltration model.

Figure 20. Time dependent moisture content from 20 realizations at the mid-height of Unit
3 with sampled soil properties for Units 3 and Unit 4.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 73

Figure 21. Time dependent moisture content from 20 realizations at the mid-height of the
clay liner with sampled soil properties for Units 3 and Unit 4.

Unsaturated Zone Modeling for the Clive PA
23 October 2015 74

Figure 22. Time dependent moisture content from 20 realizations at the mid-height of the
waste with sampled soil properties for Units 3 and Unit 4.

Figure 23. Time dependent moisture content from 20 realizations at the mid-height of the
lower radon barrier with sampled soil properties for Units 3 and Unit 4.
Unsaturated Zone Modeling for the Clive PA
23 October 2015 75

Figure 24. Time dependent moisture content from 20 realizations at the mid-height of the
upper radon barrier with sampled soil properties for Units 3 and Unit 4.
8. References
Abramowitz, Milton, and Irene A. Stegun, 1970. Handbook of Mathematical Functions with
Formulas, Graphs, and Mathematical Tables. National Bureau of Standards, Applied
Mathematics Series 55, ninth printing. November, 1970.

Bingham Environmental, 1991. Hydrogeologic Report Envirocare Waste Disposal Facility South
Clive, Utah. Final version October 9, 1991.

Fayer, M.J., 2000. UNSAT-H Version 3.0: Unsaturated Soil Water and Heat Flow Model,
Theory, User Manual, and Examples. PNNL-13249. Pacific Northwest National
Laboratory, Richland, Washington. June, 2000.

Jury, W.A. and R. Horton. 2004. Soil Physics. 6th ed. John Wiley and Sons Inc. New Jersey.

Mualem, Yechezkel, 1976. A New Model for Predicting the Hydraulic Conductivity of
Unsaturated Porous Media. Water Resources Research, Vol. 12, No. 3, pp. 513-522.
June, 1976.

Whetstone Associates, Inc., 2007. EnergySolutions Class A South Cell Infiltration and Transport
Modeling. December 7, 2007.
 NAC-0025_R3

Geochemical Modeling for
the Clive DU PA

Clive DU PA Model v1.4
5 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Geochemical Modeling for the Clive DU PA
5 November 2015 ii
1. Title: Geochemical Modeling for the Clive DU PA
2. Filename: Geochemical Modeling v1.4.docx
3. Description: This white paper provides documentation of the development of parameter values
and distributions used for modeling geochemical processes, such as solubility and adsorption,
in the transport of radionuclides for the Clive DU PA Model.

Name Date
4. Originator Katie Catlett 17 April 2014
5. Reviewer Dan Levitt 5 November 2015
6. Remarks

5 Nov 2015: Updated v1.2 to v1.4. – D. Levitt.

Geochemical Modeling for the Clive DU PA
5 November 2015 iii
CONTENTS
FIGURES ....................................................................................................................................... iv

TABLES .......................................................................................................................................... v

1.0
 Summary of Solubility, Partitioning (Kd), and Diffusion Parameters ................................... 1

2.0
 Geochemical Conditions ........................................................................................................ 3

2.1
 Hydrostratigraphic Units ................................................................................................... 6

2.2
 Shallow Unconfined Aquifer ............................................................................................ 7

3.0
 Method for Estimating Distributions for Solubility and Partitioning Parameters .................. 9

4.0
 Solid/Water Partition Coefficients (Kd) ................................................................................ 11

4.1
 Partitioning by Element .................................................................................................. 13

4.1.1
 Actinium .................................................................................................................... 13

4.1.2
 Americium ................................................................................................................ 13

4.1.3
 Cesium ...................................................................................................................... 13

4.1.4
 Iodine ........................................................................................................................ 14

4.1.5
 Lead ......................................................................................................................... 14

4.1.6
 Neptunium ................................................................................................................. 15

4.1.7
 Plutonium .................................................................................................................. 15

4.1.8
 Protactinium .............................................................................................................. 16

4.1.9
 Radium ...................................................................................................................... 16

4.1.10
 Strontium ................................................................................................................... 16

4.1.11
 Technetium ................................................................................................................ 17

4.1.12
 Thorium ..................................................................................................................... 17

4.1.13
 Uranium .................................................................................................................... 17

5.0
 Element and Species Solubility ............................................................................................ 18

5.1
 Solubility by Element ..................................................................................................... 21

5.1.1
 Actinium .................................................................................................................... 21

5.1.2
 Americium ................................................................................................................ 21

5.1.3
 Cesium ...................................................................................................................... 21

5.1.4
 Iodine ........................................................................................................................ 21

5.1.5
 Lead ......................................................................................................................... 21

5.1.6
 Neptunium ................................................................................................................. 22

5.1.7
 Plutonium .................................................................................................................. 22

5.1.8
 Protactinium .............................................................................................................. 22

5.1.9
 Radium ...................................................................................................................... 22

5.1.10
 Radon ........................................................................................................................ 22

5.1.11
 Strontium ................................................................................................................... 23

5.1.12
 Technetium ................................................................................................................ 23

5.1.13
 Thorium ..................................................................................................................... 23

5.1.14
 Uranium .................................................................................................................... 23

5.1.14.1
 Uranium Forms and Geochemical Model Parameters ............................ 23

5.1.14.2
 Uranium Solubilities based on Schoepite ............................................... 26

5.1.14.3
 Uranium Solubilities based on U3O8 ....................................................... 28

6.0
 Ionic and Molecular Diffusion Coefficients ......................................................................... 29

7.0
 References ............................................................................................................................ 31

Geochemical Modeling for the Clive DU PA
5 November 2015 iv
FIGURES
Figure 1: Example of probability distribution function for log-uniform distribution. Value for
Kd of Ac in silt with a range of values from 15.7 to 1,910 mL/g. ............................... 10

Figure 2: Distribution of Kd values for I in sand. Values less than 0 are set equal to zero. ........ 11

Geochemical Modeling for the Clive DU PA
5 November 2015 v
TABLES
Table 1: Distribution Parameters for Partitioning Coefficients (Kd) for materials (mL/g).
Unless noted otherwise, distributions are described by the log uniform
distribution. .................................................................................................................... 1

Table 2: Log Uniform Parameters for Solubilities .......................................................................... 2

Table 3. Ranges of Values Used to Develop Distribution Ranges for Kd Values. .......................... 2

Table 4. Solubility Ranges Used to Develop Solubility Distributions. For most of these
elements a log-uniform distribution was chosen, so the central value was not used. ... 3

Table 5. Distribution Parameters for Ionic and Molecular Diffusion Coefficients ......................... 3

Table 6: Soil and Mineralogy within the Four Hydrostratigraphic Units ....................................... 7

Table 7: Geochemical parameter ranges from Groundwater Wells at Clive, Utah ......................... 8

Table 8: Ion Concentrations from GW Wells Surrounding the Waste Cell. Negative and
positive percent charge balance contributions are given on a molar basis. ................... 8

Table 9: Model Results for High TDS System analogous to the Upper Aquifer. Uranium
solubility limit based on Schoepite. * .......................................................................... 27

Table 10: Total uranium, low TDS (ionic strength 0.127 M). Uranium solubility limit based
on schoepite. ................................................................................................................ 27

Table 11: Major dissolved uranium (VI) species included in geochemical models. ..................... 28

Table 12: Total Uranium, low TDS (ionic strength 0.127 M). Uranium solubility limit based
on the mineral U3O8. * ................................................................................................. 28

Table 13. Diffusion coefficients for selected cations and anions. ................................................. 30

Geochemical Modeling for the Clive DU PA
5 November 2015 1
1.0 Summary of Solubility, Partitioning (Kd), and Diffusion
Parameters
This section is a brief summary of parameters and distributions used for modeling geochemical
processes for the Clive Depleted Uranium (DU) Performance Assessment (PA) Model. For
distributions, the following notation is used:
• N( µ, σ, [min, max] ) represents a normal distribution with mean µ and standard deviation
σ, and optional truncation at the specified minimum and maximum, and
• LN( GM, GSD, [min, max] ) represents a log-normal distribution with geometric mean.

Water partitioning coefficients for the sand, clay and silt fractions used in the GoldSim transport
model are summarized in Table 1 (discussed in Section 4.0). The minimum and maximum
values used in the log uniform distribution of the aqueous solubility ranges used in the GoldSim
transport model are summarized in Table 2 (discussed in Section 5.0). Table 3 details the ranges
used to develop the Kd distributions listed in Table 1. Note that the Kd distributions were chosen
with the assumption of high carbonate concentrations at the site, as evidenced by the low range
of Kd values for U (e.g., 0.34 ml/g to 6.8 ml/g for sand and a maximum of 66 ml/g for clay
versus maximum U Kd values of 630,000 ml/g [EPA 1999b] and 1600 ml/g [Sheppard and
Thibault 1990]). Table 4 illustrates the min, max and central values used to determine solubility
distributions. Minimum and maximum values for a uniform distribution for ionic and molecular
diffusion coefficients are summarized in Table 5 (see Section 6.0).

Table 1: Distribution Parameters for Partitioning Coefficients (Kd) for materials (mL/g).
Unless noted otherwise, distributions are described by the log uniform distribution.
Chemical
Element
Sand Silt Clay
Min Max Min Max Min Max
Ac 1.68E+1 5.35E+2 1.57E+1 1.91E+3 8.36E+1 2.99E+3
Am 4.32E+1 8.11E+2 8.80E+1 1.14E+3 8.80E+1 1.14E+3
Cs 2.70E+0 2.22E+1 4.23E+0 1.18E+2 6.69E+0 2.39E+2
I N(4.28e-1, 6.05e-1) N(4.28e-1, 6.05e-1) N(4.28e-1, 6.05e-1)
Np 3.92E-1 5.10E+1 8.05E-1 6.21E+001 4.32E+0 8.11E+1
Pa 8.32E+0 3.31E+2 1.84E+2 9.78E+2 1.80E+2 1.56E+3
Pb 2.70E+0 2.22E+1 4.23E+0 1.18E+2 6.69E+0 2.39E+2
Pu 6.69E+1 2.39E+3 8.05E+1 6.21E+3 9.14E+2 5.47E+3
Ra 3.87E-1 6.46E+1 7.97E-1 7.53E+1 1.42E+0 1.41E+3
Rn 0.00E+0 0.00E+0 0.00E+0 0.00E+0 0.00E+0 0.00E+0
Sr 2.70E+0 2.22E+1 4.23E+0 1.18E+2 6.69E+0 2.39E+2
Tc N(1.02e-1, 1.45e-1) N(1.02e-1, 1.45e-1) N(1.02e-1, 1.45e-1)
Th 1.92E+1 4.16E+1 3.44E+1 6.97E+2 8.47E+1 2.36E+3
U 3.44E-1 6.77E+0 8.80E-1 1.14E+1 9.05E+0 6.63E+001
Geochemical Modeling for the Clive DU PA
5 November 2015 2
Table 2: Log Uniform Parameters for Solubilities
Chemical Element Min (mol/L) Max (mol/L)
Ac 6.81E-9 1.47E-5
Am 6.81E-10 1.47E-6
Cs 6.81E-3 1.47E+1
I 5.99E-5 1.67E+0
Np 6.81E-6 1.47E-2
Pa 6.81E-9 1.47E-5
Pb 6.81E-9 1.47E-5
Pu 5.27E-11 1.90E-5
Ra 5.99E-10 1.67E-5
Rn 7.74E-4 1.29E-1
Sr 6.81E-7 1.47E-3
Tc 7.74E-5 1.29E-2
Th 7.74E-9 1.29E-6
U* 3.58E-6 2.79E-3
U3O8 1.0E-16 6.5E-10
UO3 3.58E-6 2.79E-3
* See GoldSim model note Section .
Table 3. Ranges of Values Used to Develop Distribution Ranges for Kd Values.

Salt Water
Ranges: (units are L/kg) soil/water partition coefficients (Kds)
Element Sand Silt Clay
Ac 20 to 450 20 to 1500 100 to 2500
Am
50 to 700: 100
central
100 to 1000: 200
central 100 to 1000: 200 central
Cs 3 to 20 5 to 100 8 to 200
I 0 0 0
Np 0.5 to 40 1 to 50 5 to 70
Pa 10-275 200 to 900 200 to 1400
Pb 3 to 20 5 to 100 8 to 200
Pu 80 to 2000 100 to 5000 1000 to 5000
Ra 0.5 to 50 1 to 60 2-1000
Rn 0 0 0
Sr 3 to 20 5 to 100 8 to 200
Tc 0 0 0
Th 20 to 40 40 to 600 100 to 2000
U 0.4 to 6 1 to 10 10 to 60
Geochemical Modeling for the Clive DU PA
5 November 2015 3
Table 4. Solubility Ranges Used to Develop Solubility Distributions. For most of these
elements a log-uniform distribution was chosen, so the central value was not used.
Element Solubility (M)
Range, Min
Solubility (M)
Range, Max
Solubility (M)
Central Value
Ac 1.00E-08 1.00E-05 1.00E-06
Am 1.00E-09 1.00E-06 5.00E-07
Cs 1.00E-02 1.00E+01 1.00E+00
I 1.00E-04 1.00E+00 1.00E-01
Np 1.00E-05 1.00E-02 1.40E-04
Pa 1.00E-08 1.00E-05 1.00E-07
Pb 1.00E-08 1.00E-05 1.00E-06
Pu 1.00E-10 1.00E-05 5.10E-07
Ra 1.00E-09 1.00E-05 1.00E-06
Rn 1.00E-03 1.00E-01 1.35E-02
Sr 1.00E-06 1.00E-03 1.00E-04
Tc 1.00E-04 1.00E-02 1.00E-03
Th 1.00E-08 1.00E-06 5.00E-07
U 5.00E-06 2.00E-03 5.00E-04

Table 5. Distribution Parameters for Ionic and Molecular Diffusion Coefficients
Parameter Ion/Molecule Units Distribution
Dm All cm2/s U( 3 × 10-6, 2 × 10-5 )

2.0 Geochemical Conditions
The Clive Disposal Facility is located on the eastern side of the Great Salt Lake Desert. The
geochemistry of the Clive, Utah location is dominated by weathering and erosion of the local
basin and mountains and by recharge via meteorological precipitation. The area consists of a
large basin surrounded by mountains formed of Paleozoic limestones, dolomites, shales,
quartzites, and sandstones. Isolated areas of the Great Salt Lake desert region are underlain with
tertiary extrusive igneous basaltic flows and pyroclasts. The valley sediments consist of alluvial
fans, evaporites, and unconsolidated and semi-consolidated valley fill (Bingham Environmental
1991, Schaefer et al., 2003). Within the valley, where the Clive facility is located, the valley fill
is formed by quaternary-age lacustrine lake deposits associated with the former Lake Bonneville.
The surface deposits are mainly low-permeability silty clays with sand and gravel outcrops and
lenses in the subsurface. Bedrock appears to be at least 75 m (250 ft) below ground surface (bgs)
and potentially much lower. The regional groundwater flow is to the east-northeast towards the
Great Salt Lake.
There are four zones within the PA model domain that are included in the radionuclide transport
model, moving downward beginning with the DU waste cell there is a clay liner beneath this
Geochemical Modeling for the Clive DU PA
5 November 2015 4
waste—part of the engineered closure system. Beneath the clay liner is the unsaturated zone
which extends to the upper aquifer in the saturated zone. Each zone has unique properties that
will influence the dissolved transport of the radionuclides modeled. The DU in the waste cell,
like the unsaturated zone below it, is expected to be largely devoid of a significant water phase
during the period of this PA model. The DU waste will be initially contained in cylinders or
drums within the embankment. For the Conceptual Site Model (CSM) and associated
geochemical modeling, there is no assumption that the waste cell will have any type of grout or
concrete added. However, it is likely that fill will be placed between the waste containers before
the cell is closed. It is expected that within the 10,000-year time period the containers will fail to
a significant extent such that the DUoxide will be mixed with the degraded steel containers and
surrounding fill material. No credit is given for containment by the steel drums or cylinders, nor
is any credit taken for adsorption of radionuclides onto the steel drums. Water will occur as
inclusions in the waste and fill pores. Transport through this zone, either downward or upward,
via a dissolved phase, is modeled using the solubility conditions and partitioning (Kd) values
described below.
The conceptual model for the transport of radionuclides at the Clive Facility allows sufficient
meteoric water infiltration into the waste zone such that dissolution of uranium and daughters,
fission products and potential transuranic contaminants (along with native soluble minerals) will
occur. Depending upon the amount of water available, these radionuclides will either
re-precipitate, once the thermodynamic conditions for saturation are reached or remain in
solution and be transported to the saturated zone. This water is expected to be oxidizing, with
circumneutral to slightly alkaline pH (similar to the upper unconfined aquifer), and an
atmospheric partial pressure of carbon dioxide. However, the amount of total dissolved solids
(TDS) is expected to be initially lower than the upper aquifer. The composition of this aqueous
phase will change as it reaches the unsaturated zone, with some increase in dissolved solids and
potentially lower dissolved oxygen and carbon dioxide. This is a fairly simplistic representation
geochemically, yet the use of stochastics for the material properties, element solubilities, and
sorption parameters provides for variability in this model.
The saturated zone for this PA model includes only the shallow, unconfined aquifer. The water
table in the shallow aquifer is reported to be located in Unit 3 on the west side of the site (under
the Federal DU cell) and in Unit 2 on the east side (Bingham 1994). The influence of off-normal
conditions on shallow groundwater flow is discussed in Envirocare (2004) for two cases. In the
first, flow was affected by localized recharge from a surface water retention pond in the
southwest corner of the facility near well GW-19A in the spring of 1999. The potential for pond
overflow and localized groundwater mounding was eliminated by rerouting surface water
drainage to the pond. In the second, a groundwater mound formed between March 1993 and
spring 1997 below a borrow pit excavated near the 11e.(2) cells that occasionally filled with rain
water. The mound decreased and was negligible by the time of the report in 2004. The latter of
these conditions was captured by the hydraulic gradient data set used to develop the distribution
for the Clive DU PA model. The influence of these conditions on the hydraulic gradient appear
to be transient and of small magnitude.
Transport of radionuclides is expected to be restricted to this aquifer and not migrate to the deep
aquifer due to a natural upward gradient at the facility. The Unsaturated Zone Modeling and
Saturated Zone Modeling white papers discuss transport in more detail. The chemical
Geochemical Modeling for the Clive DU PA
5 November 2015 5
composition of the saturated zone was established by using site-specific groundwater quality
measurements. This groundwater is characterized as somewhat alkaline pH likely due to the
presence of carbonates, oxidizing, with high levels of dissolved ions of mainly sodium and
chlorine. The presence of carbonates can have a significant influence on uranium solubility.
The aqueous chemistry for the unsaturated zone is expected to be relatively oxidizing. However,
reducing conditions can exist in some areas of the saturated zone as evidenced by low Eh values
and zero dissolved oxygen in some wells at the Clive Facility.
The radionuclides of interest for this PA model include uranium and its daughter products with
relatively long half-lives, along with fission products and potential contaminant transuranic
elements (ORNL 2000, Beals, et al. 2002). The inventory and speciation of the radionuclides in
the waste layer will determine the source term. The total inventory and uranium oxide waste
forms are described in a separate white paper (Waste Inventory).
The three major types of chemical reactions that affect water composition include dissolution
and precipitation, ion-exchange and sorption, occurring as gas-phase and aqueous reactions.
Precipitation and dissolution are the major reactions between the solid and aqueous phases.
When the dissolved concentration of a radionuclide exceeds the solubility limit for any possible
mineral form, the solid phase will theoretically precipitate and control the maximum
concentration. Precipitation and dissolution are governed by thermodynamic and kinetic
considerations that include water temperature, redox conditions, concentration (activity) of
dissolved constituents, pH, and partial pressure of gases including carbon dioxide. The rate of
dissolution (kinetics) is not considered in the PA model. Due to the ratio of dissolution rate to the
time frame of interest for contaminant transport (10,000 years), it is assumed that any dissolution
is instantaneous within this time frame. Ionic strength is also a critical parameter, especially in
waters with high dissolved solids, as activity is influenced by this parameter. The
thermodynamic activity of a dissolved species is the product of its actual concentration and
activity coefficient. For dilute systems, the activity is close to unity but will deviate substantially
at high ionic concentrations. Under equilibrium conditions, the composition of the aqueous phase
within each zone will react with the surrounding solid phases to establish the chemistry that will
define the radionuclide solubilities discussed in Section 5.0. Sorption at the solid-solution
interface is also important in transport modeling and is discussed in Section 4.0.
By definition, isotopes behave identically from a chemical standpoint. As such, both solubility
and sorption parameters are treated as equal for each isotope of a single element. For example,
uranium-234, -235 and -238, is isotopes, are given equal solubility and sorption constraints,
competing for sorption sites and for aqueous solubility.
Colloid-mediated transport of actinides is possible within nuclear waste; however, this process is
complex and controversially discussed (Geckeis and Rabung 2008). Kim (1991) has reported
that the transport of polyvalent actinides can be enhanced when sorbed to colloids (e.g.,
nanoparticles), whereas experiments at the Girmsel Test Site (GTS) in Switzerland have shown
that clay colloids promote retardation and retention of radionuclides (Möri et al. 2003; Geckeis et
al. 2004). The colloidal transport/retardation process is known to be controlled by variables such
as the radionuclide-colloid interaction mechanism, colloid dissolution, agglomeration, filtration
or colloid attachment to surfaces (Geckeis and Rabung 2008). Colloid-mediated transport of
Geochemical Modeling for the Clive DU PA
5 November 2015 6
radionuclides is considered to be more significant in areas where the rock is fractured and
porous, allowing for access to groundwater pathways (Geckeis and Rabung 2008). This transport
is less significant in rock formations that are nonporous (Voegelin and Kretzschmar, 2002),
retarding migration to water sources. In addition, colloid retention is favored at high ionic
strength, low pH and in impermeable rock (Ryan and Elimelech 1996; Degueldre, et al. 2000;
CRWMS 2000). The high ionic strength conditions in the saturated zone at Clive are not
considered favorable for colloid transport. Thus, colloid-mediated transport has not been
incorporated in the PA model.
Additionally, since the site conditions are not considered favorable for colloid transport, the
effect of colloids on adsorption is that they could provide another surface to which adsorption
occurs. The effect of colloids on Kd distributions is highly uncertain as it depends on the
availability of colloid surfaces in the waste layer and the strength of sorption to colloids as
compared to surrounding minerals. If there were a high concentration of colloids in the waste and
if radionuclide-colloid Kds were greater than radionuclide-mineral Kds, then the Kds derived
from minerals alone would be low, allowing for greater transport of radionuclides in the system
than what would be expected in reality. In the current model Kds are derived from mineral
sorption coefficients. Since the presence and amount of colloids in the Clive DU waste is
unknown and the effects of colloids on Kds speculative, colloids are not considered in the
development of Kd distributions at this time.
2.1 Hydrostratigraphic Units
Sediments at the Clive site are divided into four hydrostratigraphic units within the unsaturated
and saturated zones (Table 6). Unit 4 is the uppermost unit, with Unit 1 beginning
approximately 12 to 14 m (40 to 45 feet) bgs (Envirocare 2004).
The Unit 4 soils have cation exchange capacity (CEC) values in the range of 10 to 20 meq/100 g
(USDA, 2009). These values were used qualitatively in the derivation of sorption parameters
described below.
The waste zone will contain two forms of DU oxide: UO3 produced from the Savannah River
Site (SRS) and other DOE facilities, and what is predominantly U3O8 from the gaseous diffusion
plants (GDPs). In both cases the waste will be initially stored in steel cylinders and drums that
are assumed to be backfilled with Unit 3 soils.
Geochemical conditions and water movement have not been extensively studied in the
unsaturated zone at the Clive Facility. As described above, the upper level pore water within the
unsaturated zones is expected to contain lower TDS than is found within the saturated zone,
though these levels could increase with depth in the unsaturated zone. The relative anion and
cation constituents of this pore water are likely very similar to those in the saturated zone. This
is expected as the ions in the saturated zone appear to be largely due to the presence of evaporites
and alluvium from the valley and former Lake Bonneville. Dissolved oxygen and carbon
dioxide are expected to be largely in equilibrium with atmospheric conditions, at least in the
upper profile including the DU waste zone. For derivation of the solubility and sorption
parameters a pH range of 6.5-8.5, pCO2 range of slightly above atmospheric to slightly below
atmospheric, and geochemical make up similar to the saturated zone but lower TDS was used.
Geochemical Modeling for the Clive DU PA
5 November 2015 7
Table 6: Soil and Mineralogy within the Four Hydrostratigraphic Units
Unit Number Soil and Mineral Type Unit Description
4 Fine-grained silty clay, clay silt.
Carbonates, quartz, feldspars,
clay minerals (kaolinite, smectite,
and illite/mica) trace gypsum.
From 6 to16.5 ft thick with an average
thickness of 10 ft. Unsaturated.
3 Silty sand, occasional silty to
sandy clay lenses.
10 to 25 ft thick with an average of thickness
of 15 ft. Largely unsaturated, with lower
portion saturated in western part of site. The
unconfined water-bearing zone in Unit 3 and
the upper part of Unit 2 has been
designated as the shallow aquifer.
2 Silty clay. 2.5 to 25 ft thick with an average of
thickness of 15 ft. Unit 2 is saturated below
the Clive Facility.
1 Silty sand with occasional silty
clay. Confined aquifer.
Begins at a depth of approximately 45 ft
bgs. The thickness of Unit 1 is unknown.
2.2 Shallow Unconfined Aquifer
The geochemistry of the shallow, unconfined aquifer consists of very high levels of dissolved
solutes as outlined above. The groundwater table occurs near the bottom of Unit 3, with the
shallow aquifer mainly within Unit 2. For the purposes of the PA model, the water table is
assumed to be coincident with this stratigraphic interface. This unconfined aquifer contains very
high dissolved solids, with TDS values ranging from 20 to 70 parts per thousand and specific
gravity from 1.02 to 1.06 g/mL (Envirocare 2004, and recent site specific groundwater data
acquired by EnergySolutions). The shallow aquifer consists of a brine with sodium and chloride
comprising approximately 90 percent of the ions (see Table 7 and Table 8). This brine is likely a
result of the dissolution from the Lake Bonneville evaporite sediments. Prior geochemical
modeling (Bingham Environmental, 1991) indicates the aquifer is supersaturated with calcite and
dolomite. Geochemical modeling for this PA also indicates these minerals to be at saturated
conditions. The deep confined aquifer, in Unit 1, also has high values of TDS of up to 20 parts
per thousand, but the average is well below the average of the shallow aquifer. The higher
salinity of the shallow aquifer is thought to be due to concentration of salts through
evapotranspiration (ET) and/or localized dissolution of evaporite deposits in the unsaturated
zone.
The Clive Facility has a large number of monitoring wells with completion zones in the shallow
aquifer and monitoring data are currently collected from these wells on at least an annual basis.
Prior to geochemical modeling performed for this PA, geochemical data from seven of these
monitoring wells were summarized and are provided in Table 8 and Table 9 below. These wells
are in close proximity to the DU waste cell. All wells are completed within the upper unconfined
aquifer, and are located surrounding the cell in all four horizontal directions. Data ranges and
averages were taken from quarterly, and in some cases monthly, monitoring reports. At least two
years of data were used, and in most cases data goes back to at least the year 2000.
Geochemical Modeling for the Clive DU PA
5 November 2015 8
Table 7: Geochemical parameter ranges from Groundwater Wells at Clive, Utah
Well ID pH TDS
(mg/L)
Eh
(mV)
DO
(mg/L)
Bicarbonate
(mg/L)
Temp
(°C)
GW-16R 6.65 to 7.63 26,000 to 46,400 -21 to 489 0.2 to 3 300 to 350 11.40 to 13.60
GW-25 6.62 to 7.62 40,000 to 55,000 -34 to 500 0 to 6.7 160 to 330 10.90 to 15.50
GW-19A 7.16 to 7.25 69,000 to 75,000 61 to 212 1.6 to 2.51 120 to 140 13.40 to 14.29
GW-57 6.64 to 7.69 35,000 to 52,700 -43.70 to 480 0.19 to 5.37 102 to 140 10.80 to 15.10
GW-100 6.95 to 7.63 31,000 to 42,000 30.8 to 209 0.4 to 4.0 120 to 140 12.13 to 14.00
GW-110 7.24 to 7.57 29,000 to 38,000 -18 to 168 0.14 to 7.52 160 to 204 12.60 to 13.59
GW-125 7.09 to 7.52 28,000 to 40,000 48 to 233 0.76 to 4.58 160 to 180 12.60 to 13.84
Max Range 6.62 to 7.69 26,000 to 75,000 -43.70 to 500 0 to 7.52 102 to 350 10.8 to 15.5

Table 8: Ion Concentrations from GW Wells Surrounding the Waste Cell. Negative and
positive percent charge balance contributions are given on a molar basis.
GW Well
Br–
(mg/L)
F–
(mg/L)
Cl–
(mg/L)
NO3–
(mg/L)
SO42–
(mg/L)
Ca2+
(mg/L)
Mg2+
(mg/L)
K+
(mg/L)
Na+
(mg/L)
GW-16R 22 3.8 22,914 1.4 1,769 354 486 476 14,263
GW-25 23 8.8 25,783 1.1 4,420 527 853 565 16,465
GW-19A 0 0.0 37,800 0.0 0 1,028 1,580 616 23,800
GW-57 18 8.5 23,110 1.9 4,652 707 844 530 14,398
GW-100 26 1.8 20,254 1.1 2,911 496 683 457 12,993
GW-110 17 1.5 17,989 2.1 2,226 322 469 432 11,400
GW-125 16 0.9 20,813 2,494 427 637 488 12,813
Average
(mg/L) 20 4.2 24,094 1.5 3,079 552 793 509 15,162
Average
(mol/L) 2.2E-04 1.9E-04 6.8E-01 1.8E-05 2.7E-02 1.4E-02 3.3E-02 1.3E-02 6.6E-01
percent
of charge
balance 0.03% 0.03% 92 % 0.002% 7.5 % 3.6 % 8.5 % 1.7 % 86 %

The groundwater is considered a brine, with TDS values as high as 72,000 mg/L. The redox
conditions are fairly oxidizing with an average Eh of 125 mV. Sodium and chloride are clearly
the dominant ions with slightly alkaline pH. Excellent charge balance is obtained using these
data, indicating all major ions are being accounted for. Note that the dominance of Na and Cl in
the charge balance (86% and 92%) obscures many of the other ion contributions. Groundwater
temperatures range from 11.5 to 14.5 °C. Using the data from the average of all wells shown in
Table 7, the stoichiometric ionic strength is calculated at 0.73 M (mol/L).
Geochemical Modeling for the Clive DU PA
5 November 2015 9
3.0 Method for Estimating Distributions for Solubility and
Partitioning Parameters
The process for developing probability distributions for the geochemical parameters utilized the
following basic scheme:
1. Perform a literature search for parameter values.
2. Based on site characteristics, screen the literature studies to those that could potentially
apply to the Clive site.
3. Weight the remaining literature values based on expert judgment.
4. Develop a distribution based on the weighting.
In nearly every case, once the site specific data and the general literature were screened to retain
studies relevant to the Clive site. Any value within the range of those studies was deemed to be
"equally viable," given the uncertainty associated with various soil and water characteristics for
the site. “Equally viable” indicates that the probability of one order of magnitude range is
equally likely as any other order of magnitude range within the overall viable range. Therefore,
the default probability distribution is a log-uniform distribution. To establish a range for the log-
uniform distribution, the range of values from the relevant literature was considered. To ensure
that the distribution represented the minimum and maximum literature values, these values were
treated as the 5th (Q0.05) and 95th (Q0.95) percentiles of the distribution, respectively, effectively
extending the support of the distribution a small amount beyond the range of literature values.
That is, the geometric mean of the distribution is set to the geometric mean of the quantiles:
𝐺𝑀=exp ln𝑄!.!"+ln𝑄!.!"
2 (1)
To calculate the range of the distribution in log-space, the range of the log-percentiles is
extended from 90% to 100%:
𝑅!=ln𝑄!.!"−ln𝑄!.!"
0.9 (2)
To get the endpoints of the log-uniform distribution, the half-range is subtracted and added to
the geometric mean:
Min =exp ln𝐺𝑀−𝑅!
2        Max =exp ln𝐺𝑀+𝑅!
2 (3)

For example, the literature values for Kd values of Ac in silt ranged from 20 mL/g to 1500 mL/g.
Treating these values as 5th and 95th percentiles gives a geometric mean of GM=173, and a
log-range of Rl=4.80, leading to a log-uniform distribution from 15.7 mL/g to 1910 mL/g. This
distribution is illustrated in Figure 1.
Geochemical Modeling for the Clive DU PA
5 November 2015 10
The exceptions to the log-uniform fit were the Kd values for Tc and I. For these parameters,
values of 0 are possible, yet a log-uniform distribution cannot represent that possibility naturally.
For these parameters, it was decided to fit a distribution that would give an approximate 25%
chance of a 0 value and a median near the Clive-specific data (Adrian Brown Consultants,
Response to UDEQ Kd Interrogatories, 1997). The median values used were 0.11 mL/g for Tc
and 0.46 mL/g for I. For Tc, a maximum literature value of 0.33 was considered as a 95th
percentile. These distributions were fitted using the standard approach of fitting distributions
based on quantiles that is described in the Fitting Probability Distributions white paper. A
normal distribution fit these percentiles well, when values less than 0 were treated as 0. The I
distribution was then scaled to match the shape of the Tc distribution. Figure 2 illustrates how
the distribution for I is represented in GoldSim. Values less than zero are set equal to zero.

Figure 1: Example of probability distribution function for log-uniform distribution. Value
for Kd of Ac in silt with a range of values from 15.7 to 1,910 mL/g.
Geochemical Modeling for the Clive DU PA
5 November 2015 11

Figure 2: Distribution of Kd values for I in sand. Values less than 0 are set equal to zero.
4.0 Solid/Water Partition Coefficients (Kd)
The transport of dissolved radionuclides can be limited by sorption onto the solid phase of
associated minerals and soils within each of the zones considered in this PA model. The
transport of uranium is limited by both solubility and the sorption of radionuclides in
groundwater. Sorption consists of several physicochemical processes including ion exchange,
adsorption, and chemisorption.
Sorption is represented in the PA model as a Kd value. While the geochemistry of contaminant
transport is complex, a representative and standard approach was taken for the purposes of the
PA. Distribution parameters for radionuclide solubilities are derived in Section 5 below. The
current section focuses on the description of sorption and the derivation of parameters for Kd
distributions.
Solid/water partition coefficients, or Kds, are based on a simple equilibrium sorption model, and
are a simplification of the wide range of geochemical processes discussed above. Despite the
simplicity of the Kd models, they are commonly used in performance assessments because of
their ease of implementation in transport codes. Site-specific monitoring tests were used in the
process to derive distributions when this information was available. The Kd model assumes that a
given constituent dissolved in the water (e.g., uranium) has some propensity to sorb to the solid
Geochemical Modeling for the Clive DU PA
5 November 2015 12
phase of a porous medium, while maintaining an aqueous phase. The definition of the solid/water
distribution coefficient, with units of mL/g is:
𝐾!=𝑚𝑎𝑠𝑠𝑜𝑓𝑐𝑜𝑛𝑠𝑡𝑖𝑡𝑢𝑒𝑛𝑡𝑠𝑜𝑟𝑏𝑒𝑑𝑜𝑛𝑎𝑢𝑛𝑖𝑡𝑚𝑎𝑠𝑠𝑜𝑓𝑠𝑜𝑙𝑖𝑑(𝑔/𝑔)
𝑚𝑎𝑠𝑠𝑜𝑓𝑐𝑜𝑛𝑠𝑡𝑖𝑡𝑢𝑒𝑛𝑡𝑤𝑖𝑡ℎ𝑖𝑛𝑎𝑢𝑛𝑖𝑡𝑣𝑜𝑙𝑢𝑚𝑒𝑜𝑓𝑤𝑎𝑡𝑒𝑟(𝑔/𝑚𝐿). (4)
The sorption is assumed to be instantaneously reversible and independent of concentration. That
is, no dynamics are accounted for, and the ratio is always simply linear—a constituent’s
concentration in water is always the same ratio with respect to its sorbed concentration onto the
solid, and sorption is instantaneous. This is the commonly used linear isotherm assumption.
Applying the Kd model outside of the range of concentrations used to obtain the values can lead
to over- or under-estimation of sorption. To account for ranges of geochemical conditions and
the potential deviation from the assumptions underlying the linear sorption model which may
result in variation in Kd values, this PA model includes parameter distributions (stochastics) for
the sorption values. Nominal Kd values were selected using both site-specific monitoring tests
(when available) and the general scientific literature. Data were taken from literature that most
closely matched the geochemical conditions at the site, including TDS range, pH and alkaline
conditions, temperature, and soil properties (CEC, clay types) to the extent possible. Kd values
have been chosen for five individual materials: silt, sand, clay, UO3 waste and U3O8 waste. In all
natrual zones the silt, sand, and clay are mixed to some extent. After including the uranium
oxide material amounts into the GoldSim model, it became apparent that they form such a small
fraction of the profile relative to the other materials that sorption processes within the waste
could be neglected, so these materials are assigned values for Unit 3, represented using the Kds
for sand. It is also recognized that essentially no information on sorption to uranium oxides (as
the waste inventory) is available in the published literature.
The process for selecting Kd values for the elements entailed an extensive literature search to
identify sorption values used in other transport models, and in particular from locations that have
similar solid phase properties and geochemical conditions. The sorption values used by
Whetstone Associates (2009), Bingham Environmental (1995, 1996), Scism (2006), Sheppard
and Thibault (1990), the Yucca Mountain Site Characterization Project (LANL 1997), DOE
(2003), and Envirocare (2000) were evaluated. In addition, the EPA three volume series
Understanding Variation In Partition Coefficient, Kd, Values (EPA, 1999a, 1999b, 2004) was
referenced extensively in this process. The reader is advised to consult that EPA series as it was
used for the derivation of many Kd values and much of that information is not repeated here.
Work by Serne (2007) was also reviewed during this investigation. Serne focused on surface
agricultural soils and Columbia River bank near-surface sediment associated with the Hanford
site. Some of the scenarios investigated by Serne (2007) are “non-groundwater” scenarios which
do not involve direct ingestion of contaminated well water by humans or animals. Serne
specifically states that the values are not to be used in water-borne scenarios except when the
modeling is used to estimate accumulation of contaminants by future surface soils from irrigation
practices, and that they are not appropriate for unsaturated zone transport to the groundwater.
Nevertheless, the results are important from a semi-quantitative perspective. Of note is that the
Hanford soils are slightly acidic (pH 6.2 to 7.8), with organic content of 0.5 to 1.5% organic
carbon somewhat different from the Clive location with organic carbon contents of
Geochemical Modeling for the Clive DU PA
5 November 2015 13
approximately 0.3% to 1%. Serne (2007) also reviews a number of studies that are also
somewhat applicable to the Clive facility and the range of Kd values provided are useful as a first
comparison. As such, a number of values compiled by Serne are provided below. Serne also
refers to work by Last et al. (2004) and Krupka et al. (2004) for systems that include migration to
groundwater, as envisioned in this PA model. Values described by these authors are discussed
below for individual species.
4.1 Partitioning by Element
This section provides a description of the derivation of the partition coefficient for each element
used in the transport model. Data were derived first from site-specific monitoring studies where
this information was available. Second, the data was taken from literature searches, with values
chosen from locations with similar geochemistry and soil/mineral conditions as the Clive facility
as described in Section 4.0 above.
4.1.1 Actinium
Minimal Kd information was found for this element. Values from Sheppard and Thibault (1990)
are as follows: 450 mL/g (sand), 1,500 mL/g (loam, here used to represent silt), 2,400 mL/g
(clay), 5,400 mL/g (organic). In order to derive values for this PA for each of the three materials
(sand, silt, and clay) a range similar to those from Sheppard and Thibault (1990) was
incorporated with adjustments made for each of the three materials.
4.1.2 Americium
Americium will likely occur predominantly as carbonate complex cation Am (III) in the pH
range at the Clive facility, though some speciation as an anion is also possible. This rare earth
element will have a large sorption coefficient. The largest source of Kd values for this element
was found in Serne (2007) with discussion on a number of studies by other researchers. In the
Hanford system, americium adsorbs fairly strongly to soils and sediments. Serne chose a best
value of 500 mL/g with a recommended range of 60 mL/g to 5,000 mL/g for the
non-groundwater scenarios. This range is consistent with studies by others using a matrix within
a groundwater system, with the exception of those done on <1 mm size particles by Tanaka and
Muraoka found in Serne. Krupka et al. (2004) chose a best value of 300 mL/g. Sanchez (in EPA
2004) found no apparent effect of salinity on Kd values and no additional information was
obtained during this research. For the transport model a range from 43 mL/g to 1,140 mL/g was
chosen as shown in Table 1.
4.1.3 Cesium
Cesium sorption is strong in most soils (EPA 1999b). Sorption commonly occurs as the Cs+
cation via cation exchange. In calcareous soils with mica minerals, cesium was essentially
completely absorbed above pH 4.0. However, high salt solution does decrease sorption. At
Idaho National Laboratory (INL) (Hull, 2008), a release of cesium in 1972 has been found to be
essentially immobile. This effect is thought to largely be a function of cation exchange with
clays in the exchange on both the planar and frayed edge sites of clays. The binding on the
frayed edge is considered stronger, resulting in a high Kd.
Geochemical Modeling for the Clive DU PA
5 November 2015 14
Serne (2007) chose a recommended value of 2,000 mL/g for the low ionic strength, circum-
neutral waters in the near surface sediments at Hanford. This was consistent with the value from
Krupka et al. (2004). Serne (2007) recommended a range of 200 to 5,000 mL/g for the non-
water-borne (e.g., unsaturated, agriculture zone) scenarios at that location and a log normal
probability distribution to describe the variation.
Because cesium sorbs by an ion exchange process, sorption can be depressed by high TDS of the
groundwater. Vandergraaf et al. (1993) has performed sorption experiments with Cs examining
the relationship between Cs concentrations and TDS and were able to fit a quadratic equation to
the data. For this PA model, cesium Kd values were selected largely from the look up tables in
EPA (1999c), but were adjusted lower due to the high TDS in the saturated zone. Also note that
the CEC values of 10-20 meq/100 g at the Clive facility are indicative of some but not a
significant presence of clay minerals within the saturated zone. These CEC values apply to the
materials within the saturated zone. The liner especially, and some native materials within the
unsaturated zone do contain clay minerals.
4.1.4 Iodine
Iodine is expected to largely exist as the anion, I– or IO3–, though volatile organic forms are also
possible. Because of the negative charge, sorption will likely not be strong, due to the typical
negative charge of the soils at the Clive site under neutral to alkaline conditions. This is
especially true in the saturated zone where high concentrations of chloride ions will compete for
any available sites to sorb. Sorption appears to increase with increasing organic matter for iodine
(EPA 2004), which may be largely due to microbial processes. Studies of iodine sorption under
oxic conditions on Hanford Site sediments (Kaplan 1998b from EPA 2004) indicated very low
Kd values. Serne (2007) recommended a value of 3 mL/g with a range of 0 to 15 mL/g. Last et
al. (2004) recommended a range of 0 to 2 mL/g. The very low concentrations of organic
compounds found in the sediments at the Clive facility would support the use of a low range for
the Kd. This range is largely derived from Summary of Results, Radionuclide Kd Tests (Bingham
Environmental, Inc. August 3, 1995) where a value of 0.7 mL/g was derived for the Clive facility
using samples from Unit 3 samples. The grain size distributions from these Unit 3 samples
indicated the material was largely sand. Clive site groundwater was used for the sorption
studies. This Kd value was then changed in Adrian Brown Consultants (1997), with a
recommended value of 0.46 mL/g.
4.1.5 Lead
Lead speciation is largely anticipated to be in the form of dissolved PbCO3 at least in the
saturated zone. Lead may be largely in the hydroxide ion form in waters of lower carbonate
concentration, though this is not anticipated to any significant extent. As PbCO3 is expected to
be the dominant form above pH 7, sorption will not be especially significant. Lead has such low
solubility, especially in presence of phosphate and chloride, that solubility often can control
movement. Table 1 in Appendix D of the Bingham Environmental (1991) shows an EPA Arid
Site value of 220 mL/g for lead, with an applicable range of 1 to 10,000 mL/g. Serne (2007)
recommended a value of 400 mL/g for the non-groundwater scenarios at Hanford, within the
range of Sheppard and Thibault (1990). Based on the Pauling ionic radii of Sr2+ and Pb2+ (1.12
and 1.19 Å, respectively), the sorption of lead is expected to be similar to strontium and also to
Geochemical Modeling for the Clive DU PA
5 November 2015 15
be somewhat suppressed by high ionic strength solutions. For this model, the lead Kd values
were chosen in a range from 2 to 200 mL/g with lower values for the sand material. This range
was based upon the expected similarity with strontium and cesium.
4.1.6 Neptunium
Neptunium will most probably exist in the Np(V) form with some as Np(IV), principally as an
uncharged hydroxide, where reducing conditions exist. In the Np(V) form as NpO2+, this species
can sorb to iron oxides and clays but not to a significant extent to common minerals. Np(V) has
a pH dependence, with negligible sorption at values less than pH 5. This ion can also form
carbonate complexes above pH 8.5 or under high carbonate concentration conditions (Serne,
2007).
The transient, mildly reducing conditions that can exist at Clive and the presence of carbonates
may lead to the formation of Np(V) carbonate complexes above pH 7 (EPA 2004). However,
there are a limited number of Kd studies for this element. Heberling et al. (2008) studied Np(V)
adsorption to calcite at four pH values at constant ionic strength. The Kd was found to vary with
both pH and concentration with a value range of 0.0090 ± 0.004 mL/g to 0.0610 ± 0.002 mL/g.
Wooyong et al. (2009) measured Kd values from sediment collected at the Hanford site and
found a range of 0.6 to 4.8 mL/g. The EPA (2004) suggests a minimum Kd of 0.2 mL/g. Serne
(2007) chose a range of 2 to 50 mL/g for the non-groundwater scenario at Hanford with a best
value of 25 mL/g. This range is approximately two times higher than the range recommended by
Last et al. (2004) and Krupka et al. (2004) for groundwater systems.
Vandergraaf et al. (1993) reported values ranging from 0.5 to 68 mL/g. In the presence of Fe(II),
reduction of Np(V) has been observed (Cui and Eriksen 1996, Nakata et al. 2002). Kumata et al.
(1993) observed retention of Np in columns of crushed granite from solutions with low Eh
values finding a relationship between retention and flow rate that suggested that the kinetics of
the redox process were relatively slow. Values from Sheppard and Thibault (1990) are as
follows: sand: 5 mL/g, loam (silt): 25 mL/g, clay: 55 mL/g, and organic: 1200 mL/g. For the
Clive facility, a similar range is recommended, though this range is reduced for the sand matrix.
4.1.7 Plutonium
Plutonium can be found in a number of valence states under the conditions at Clive. The most
likely states are as Pu(V) and Pu(VI) both as cations and complexed with hydroxide and
carbonate, although Pu(IV) may be present in the slightly reducing conditions of the saturated
zone and localized areas of the unsaturated zone due to surface-mediated reduction of Pu(V)
(Keeney-Kennicutt and Norse 1985; Powell et al. 2005; and Sanchez et al. 1985). Plutonium
sorption is known to occur on many common minerals, clays, and oxides. It is noted that
experiments by Linsalata and Cohen (1980) did not find a reduction in Kd with high ionic
strength (salinity increased to 24%). Serne (2007) chose a range of 200 to 5,000 mL/g and a best
value of 600 mL/g for the Hanford non-groundwater scenario. Last et al. (2004) and Krupka et
al. (2004) recommended values of 600 mL/g and 150 mL/g, respectively, with a high range value
of 2000 mL/g. Data by Glover et al. (1976) found in the EPA (EPA 1999b) series may be of
particular relevance to the Clive location since the data demonstrated correlations with carbonate
concentrations and clay content, two factors that are of importance in the Clive DU PA model.
Geochemical Modeling for the Clive DU PA
5 November 2015 16
These data, along with Kd values collected on basalt sediments were used to develop the EPA
look up table. For the Clive location, Kd values corresponding to lower clay content especially
within the saturated zone, and medium to highly soluble carbonate throughout are most
applicable. These Kd values range from 80 mL/g to 520 mL/g. These values appear to be
slightly low compared to other studies, such as Serne (2007), Last et al (2004) and Krupka et al.
(2004) discussed above. For the Clive facility, a slightly higher range was chosen, with the
upper limit increased because of the clay matrix.
4.1.8 Protactinium
Little information was identified that provided sorption values for this element. Serne (2007)
discusses the use of neptunium as an analog for protactinium. In sea water environments, where
particles of very high surface area are encountered, Kd values of greater than 10,000 mL/g have
been measured. Serne recommended a most probable Kd value of 400 mL/g for protactinium,
similar the values used for bismuth and polonium. Serne's recommended range is 150 to 10,000
mL/g. Again, this is for a non-groundwater scenario, different from that at the Clive facility.
Sheppard and Thibault (1990) categorize Kd values by sand, clay, loam, and organic with values
ranging from 550 to 2,700, sand-clay respectively. These values from Sheppard and Thibault
(1990) formed the basis for the distribution used for this PA model. These values were reduced
to account for the high TDS in the saturated zone.
4.1.9 Radium
Based on the compilation by Serne (2007), radium is a fairly strongly sorbing species in low
ionic solutions at circumneutral pH. Radium will co-precipitate with calcium sulfate in high
ionic strength waters and may also do so in barite. Sheppard and Thibault (1990) recommended
the following values: sand: 500 mL/g, loam (silt): 36,000 mL/g clay: 9,100 mL/g and organic:
2,400 mL/g. Serne recommended a best value of 200 mL/g and a range of 5 mL/g to 500 mL/g.
This range was lower than Sheppard and Thibault (1990) based on studies that indicated cation
exchange is a dominant sequestration mechanism and the low CEC of the Hanford soils. Krupka
et al. (2004) recommended a Kd of 14 mL/g with a range of 5 mL/g to 200 mL/g. A range based
on all of the above data was chosen for each material class.
4.1.10 Strontium
Strontium has little tendency to form complexes with inorganic ligands (EPA 1999b).
Reversible cation exchange is expected to be the most important mechanism impacting sorption
in the pH conditions at the Clive facility. This behavior is similar to that of cesium though
sorption is generally not as strong. A point worth noting in this context is that natural Sr in the
groundwater will dilute any radioactive Sr isotopically. The high sulfate concentration in the
groundwater at Clive (4,420 mg/L average for GW-25) may lead to precipitation of SrSO4 or co-
precipitation with CaSO4.
A study at the INL (Hull, 2008) indicated strontium sorption was dependent upon other cations,
primarily Ca2+, Mg2+, and Na+ with Kd decreasing with increasing concentrations of these ions.
The Kd value decreased from 85 mL/g to 4.7 mL/g. This effect was considered a cation-
exchange phenomenon, where the divalent strontium cation competes with calcium. This effect
Geochemical Modeling for the Clive DU PA
5 November 2015 17
is similar to that observed by Patterson and Spoel (1981, as referenced in Hull) at the Chalk
River Nuclear Laboratories. The EPA look up table (EPA 1999) was developed using pH and
CEC values. Using this table, along with the known applicable parameter ranges for Clive of
relatively low abundance of clay minerals within the saturated zone but somewhat higher within
the unsaturated zone, a CEC of 10 to 20 meq/100 g, and pH range of 6.6 to 8.5, the listed Kd
values are within a range from 15 mL/g to approximately 200 mL/g. Note, a higher clay content
will act to increase these values. However, under very high TDS conditions as in the saturated
zone, lower sorption is expected and is reflected in the ranges chosen for this transport model.
4.1.11 Technetium
In oxic conditions technetium will exist as the TcO4– metal oxyanion, which is essentially
non-adsorptive (EPA 2004). EPA did not develop a lookup table for technetium but cite data
indicating Kd ranges from slightly negative to generally less than 1 mL/g. Under chemically
reducing conditions, either in the bulk groundwater or locally on the surface of Fe(II) containing
minerals (biotite, magnetite), or in the presence of microbes, reduction of Tc(VII) to Tc(IV) can
occur and the reduced form of Tc will either sorb strongly or will precipitate (Vandergraaf et al.
1984, Cui and Eriksen 1996). This process will fix technetium to geological material. However,
if redox conditions change, there is the possibility that Tc could be resolubilized and transported
through the geosphere as an anion without retardation. Sheppard and Thibault (1990) indicate
very low Kd values for this species ranging from 0.1 to 1. This low propensity for the TcO4– ion
to sorb, has been noted by many researchers. Wooyong et al. (2009) measured Kd values from
sediment collected at the Hanford site and found a range of 0.08 to 0.4 mL/g. For this model, the
technetium sorption distributions were chosen based on the information above and the derivation
that is provided in Adrian Brown Associates (1997). These data are derived from sorption on to
Unit 3 sand and site-specific groundwater under oxidizing conditions.
4.1.12 Thorium
The solubility of thorium is low (circa 10-9 molar), which has made sorption measurements
difficult. This element occurs only in the +4 oxidation state in natural waters. Thorium can form
many different species including carbonate complexes. The Canadian high level waste program
uses a Kd value of 800 mL/g. Values for thorium were chosen largely based upon the information
in the EPA literature (1999b, 1999c, 2004) though the values were reduced to some extent to
account for the high TDS based upon recommendations from Vandergraaf (personal
communication 2010).
4.1.13 Uranium
The Kd for uranium is important in this PA due to the large mass of this element in the inventory
relative to any other radionuclide. The transport of uranium is expected to be mainly a factor of
the solubility within the waste zone (near source), and potentially within the saturated zone with
time. However, retardation of the uranium via sorption will be important in the clay liner
beneath the waste zone and within both the unsaturated and saturated zones.
Uranium (VI) sorption can be controlled by cation exchange and adsorption processes, especially
in low ionic strength systems (EPA 1999b). As the ionic strength increases, other cations will
Geochemical Modeling for the Clive DU PA
5 November 2015 18
displace the uranyl (UO22+) ion (EPA 1999b). Uranium sorption on iron oxide minerals and
smectite clays is extensive except in the presence of carbonate where this is reduced (EPA
1999b). Aqueous pH values also influence uranium sorption, affecting the speciation as
described above as well as influencing the number of exchange sites on variably charged
surfaces. Dissolved carbonate concentrations and pH appear to be the most important factors
influencing adsorption of U(VI). However, under the slightly alkaline and carbonate dominated
conditions expected in the saturated zone, uranium will likely occur in several forms including a
uranyl-carbonate or oxy-carbonate anion or a non-charged uranyl hydoxide. The speciation
results from the solubility modeling are described in Section 5.
In the range of pH 7 to 9, there were 4 to 5 orders of magnitude variation in Kd values noted in
the data collected by the EPA (1999b, 1999c, 2004). For the pH range of 7 to 8, which is most
likely at the Clive site, the EPA listed a minimum Kd of 0.4 mL/g and a maximum of
630,000 mL/g. The minimum value was based on values calculated for quartz with the
maximum value based on data calculated for ferrihydrite and kaolinite. These very high Kd
values are considered potentially biased by one order of magnitude because of precipitation
occurring as well as adsorption (EPA 1999b). Values from Sheppard and Thibault (1990) are as
follows: sand: 35 mL/g, loam (slit): 15 mL/g clay: 1,600 mL/g and organic: 410 mL/g.
Last et al. (2004) and Krupka et al. (2004) recommend ranges for uranium of 0.2 mL/g to 4 mL/g
and 0.1 mL/g to 80 mL/g respectively. Wooyong et al. (2009) measured Kd values from
sediment collected at the Hanford site and found a range of 0.2 mL/g to 1.5 mL/g. In most cases
these authors found that higher Kd values were associated with the less-than-2-mm particle size
fraction as one would expect based purely on surface area. However in some of their sediments
this relationship was reversed. They attributed this to highly reactive surfaces on gravel at their
location.
Site-specific sorption data for uranium are also available from the Adrian Brown and Associates
(1997) report, performed by Barringer Laboratories. Two data points at a single uranium
concentration (at day 7 and 16) were obtained with tests performed at two higher concentrations
resulting in the precipitation of the uranium. The average Kd value in this study was 6.0 mL/g.
The Kd values chosen for this PA were based on both the site-specific data and literature
information. The U(VI) species in the aqueous environment will not have particularly strong
sorption tendencies. The uranyl ion is mobile in the high ionic-strength solutions and this
mobility is also found with waters containing high carbonates. This indicates uranium sorption
is more likely to be found at the lower ranges of those cited by the EPA and Sheppard and
Thibault (1990).
5.0 Element and Species Solubility
Modeling transport of radionuclides of interest at the Clive Disposal Facility area requires an
understanding of the expected concentration of these species in the dissolved phase starting in
the DU oxide waste zone. Once dissolved from the waste, the radionuclides have the potential
for transport vertically down into the unsaturated zone below and then into the shallow aquifer.
Diffusion both upward and downward in the aqueous phase is also possible. At first, leaching is
likely to be solubility-limited with respect to uranium, and the leachate will migrate away from
Geochemical Modeling for the Clive DU PA
5 November 2015 19
the source with uranium concentration at the solubility limit. The other radionuclides are
unlikely to be at a solubility limit but establishing boundaries is necessary for the modeling. The
concentrations of radionuclides limited by sorption will be less in the dissolved phase farther
from the source.
The importance of solubilities of the individual species in this PA model varies. Uranium is
expected to be solubility limited at the source, but most other elements in the inventory likely
will not be so limited. Therefore, the majority of effort for solubility distribution development
was focused on uranium. Solubilities for the other species were drawn from literature reviews of
studies conducted at locations with similar water chemistry.
At the Clive site, four major physical zones or systems are encountered and can influence the
aqueous movement of radionuclides. These zones include the waste cell, the clay liner beneath
the waste zone, the unsaturated zone, and the saturated zone (shallow aquifer). As described in
the Unsaturated Zone Modeling and Saturated Zone Modeling white papers, the waste zone and
unsaturated zone are considered very similar in terms of the expected geochemistry of the
aqueous phase. Due to the small ratio of DU waste to native materials they are also relatively
similar in mineral composition and both are modeled using physical and chemical properties of
Unit 3 (represented chemically as a sand). The clay liner will mainly influence retardation via
sorption, in addition to decreasing water infiltration. The important saturated zone geochemical
conditions, including aqueous and solid state chemistry, are those that influence the precipitation
and dissolution of the species of interest in this PA. Differences between the saturated and
unsaturated zones are mainly associated with ionic strength and redox conditions with pH
expected to be fairly similar in both saturated and unsaturated zones. The interstitial water in the
waste and unsaturated zones is considered to be highly oxidizing, more so than the aquifer, and
with neutral to slightly alkaline pH. However, the differences in ionic strength and oxidizing
conditions between the unsaturated zone and saturated zone did not have a significant effect on
the calculated solubilities. Data from the saturated zone (Table 7and Table 8) indicate it is
susceptible to localized, transient, anoxic conditions with zero to slightly negative Eh values.
These areas will have a large influence on uranium solubilities since U(IV) is much less soluble
(circa 10-8 M) than U(VI). Other species of interest to this PA model will also have reduced
solubilities in anoxic regions.
Microbial influences on the transport of the radionuclides are not expected to be important.
Little or no organic materials (cellulosics, plastic) are expected in the waste. Therefore, no
microbial influence is included in this model, nor are organic materials such as humic and fulvic
acids expected to be present in any significant amounts. However, radiolytic effects could cause
transient changes in redox conditions or generate carboxylic acids as described below.
Anoxic corrosion of the steels and iron-based alloys used to construct the DU cylinders and
drums can affect the release of actinides. Corrosion would be expected to reduce the oxidation
state of some actinides. The most significant effect would be to decrease the mobility of
uranium, technetium, and plutonium. Uranium transport is again strongly influenced by redox
conditions. However, it is highly uncertain whether anoxic corrosion would take place since this
would require consumption of oxygen. A more conservative approach was taken, where largely
oxidizing conditions are assumed to remain to some extent within the unsaturated zone.
Geochemical Modeling for the Clive DU PA
5 November 2015 20
In many cases the solubility of radionuclide species used in the transport model was based to
some extent on the data provided in the proposed Yucca Mountain Project (LANL 1997) and the
Nevada National Security Site (NNSS, formerly the Nevada Test Site) (Sandia 2001) modeling.
These data provide a starting basis for the central tendency value used in the solubility
distributions for several of the radionuclide species. The Yucca Mountain, NNSS, and the Clive,
Utah locations have many common geochemical conditions such that the solubility for the minor
constituents (those other than uranium) can be modeled similarly. There are noted differences
between the three sites but these references provide a good basis for selecting solubility since
much of the chemistry is similar with respect to redox, carbonate chemistry, low organic matter
content, and pH.
The Yucca Mountain unsaturated zone water is characterized as oxidizing (Eh estimated at 400
to 600 mV) and the partial pressure of carbon dioxide will be variable with depth resulting in a
pH of 7 to 8. The saturated zone water at Yucca Mountain is characterized as having a pH also
in the range of 7 to 8 and oxidizing to reducing conditions depending upon whether the waters
have access to atmospheric oxygen or to reducing agents (Kerrisk, 1987). With the exception of
the very high ionic strength of the shallow aquifer, this is similar to the conditions at the Clive
site. The waste zone at the Clive facility will likely have redox conditions very similar to those
in the unsaturated zone at Yucca Mountain. The high ionic strength brine found in the shallow
aquifer at Clive can increase or decrease the solubilities of some actinides, as shown at WIPP
(DOE 2009). The WIPP site has a higher ionic strength in the pore water (~6 M) than expected at
Clive (~1 M), and WIPP is expected to be a carbonate-free system, unlike Clive. So while the
information from WIPP is not directly transferrable to Clive, the influence of the brine effect on
solubility was incorporated into the decision making for solubility selection and modeling. For
example, the range of solubility values for a particular element might be extended an order of
magnitude higher for Clive than it was for Yucca Mountain (e.g., Section 5.1.7 below). Clive,
Yucca Mountain, NNSS and WIPP have different mineralogy and soil properties that can
influence the ion-exchange, sorption and solubility constraints in this model and the direct
applicability of using data from the literature for Clive. More information is given in the sections
below as to how the influence of the properties of the reference (e.g., high ionic strength) are
included in deriving the solubility distributions.
At the NNSS, data from Frenchman Flat (Sandia 2001) indicate that elemental composition of
minerals and total oxide concentrations of the sediments remain fairly constant with depth. The
alluvium has a composition of approximately 65% SiO2 and 13% Al2O3. Very little clay is
present. Some accumulation of calcium carbonate, in certain horizons are found as coatings on
clasts and with pendants of pebbles and sand beneath, indicating repeated periods of surface
stability in the Quaternary. Water does not move downward under current climate conditions in
the unsaturated zone and this is expected to continue within the next 10,000 years. The
unsaturated zone moisture content is low, roughly 5 to 10% to 40 meters with a pH range of 7 to
9 and a high Eh. The alluvium is dominated by quartz, feldspar, cristobalite, with calcite,
gypsum, and minor amounts of clays and zeolites.
Geochemical Modeling for the Clive DU PA
5 November 2015 21
5.1 Solubility by Element
5.1.1 Actinium
The only stable oxidation form of actinium is the +3 ion (Morss et al., 1977). Actinium forms
hydrolysis complexes with the Ac(OH)3 species and the solubility is reported at 0.74 mg/L (2.6
x 10-6 M).
At Yucca Mountain the actinium solubility used in the Total System Performance Assessment
(TSPA) model ranges from 10-10 M to 10-6 M (LANL 1997). For this PA model, a similar range
was used, though the lower end was raised by a factor of 100.
5.1.2 Americium
Americium exists in the +3 oxidation state in natural waters and forms carbonate complexes at
pH values above 7 (EPA 2004, Serne 2007). The americium solids that would likely control the
solubility include Am (OH)3, AmOHCO3, and Am2(CO3)3. At Yucca Mountain the americium
solubility used in the TSPA model ranged from 10-10 M to 10-6 M (LANL 1997). A similar range
is used in this PA model.
5.1.3 Cesium
Cesium exists in the +1 oxidation state (EPA 1999b) with little tendency to form aqueous
complexes. The dominant form at the site would be as the Cs+ ion. Cesium has a high solubility,
with little tendency to precipitate, therefore a conservatively high solubility was used for this PA
model, with a fairly narrow range.
5.1.4 Iodine
Iodine can form a number of oxidation states, but within the Eh and pH conditions expected at
the Clive facility, iodine is expected to exist in the -1 oxidation form. This is consistent with the
modeling provided by EPA (1999c). In addition to dissolving and sorbing reactions, iodine can
also volatilize to the gas phase either as I2 (molecular iodine) or hydrogen iodide and organic
(e.g. methyl) iodides. Iodine is not likely to form minerals due to the very low concentrations
that would be encountered. Nevertheless, solubility could be controlled via iodine minerals. The
distribution used for iodine in this PA model reflects the high solubility.
5.1.5 Lead
Under the environmental conditions at the site lead will exist in the +2 oxidation state (EPA
1999b). However, lead has very low solubility with values of 10-8 M in natural waters, with the
dissolved species PbCO3 the dominant form above pH 7. Lead species include hydrolysis and
carbonate complexes, with the later more prevalent above pH 7. Lead carbonate (e.g cerussite,
hydrocerussite), sulfate (anglesite) and phosphates (chlorophyromorphite) minerals control lead
solubility under oxidizing conditions (EPA 1999b). At Yucca Mountain the lead solubility used
in the TSPA model ranged from 10-8 M to 10-5 M, with an expected value of 10-6.5 M. (LANL
1997). This same general range is used in all three zones for this PA model.
Geochemical Modeling for the Clive DU PA
5 November 2015 22
5.1.6 Neptunium
Neptunium can exist in several oxidation states, but only +4 and +5 are reasonable for the Clive
site. Np(V) is relatively mobile due to the high solubilities of associated minerals and low
sorption. Np (V) is expected to be present as NpO2 + (EPA 2004). Np (IV) however, forms
solids of low solubility though these are restricted to reducing conditions. Neptunium can form
carbonate complexes but this is generally limited to pH conditions above 8. In carbonate rich
systems with high sodium and potassium, as found at the Clive facility, several sodium (e.g.
2NaNpO2CO3•7H2O) and potassium based mineral forms of Np can control the solubility. Due
to the high solubilities of these minerals Np can be found at levels in the 10-4 M or greater
concentration. At Yucca Mountain the neptunium solubility used in the TSPA model ranged
from 10-8 M to 10-2 M, with an expected value of 10-4 M (LANL 1997).
5.1.7 Plutonium
Plutonium can exist in four different oxidation states: +3, +4, +5, and +6 with Pu(IV), Pu(V),
and Pu(VI) expected under oxidizing conditions, such as those found at the site (EPA 1999b).
Plutonium forms strong hydroxy-carbonate complexes with the tetravalent complex
[Pu(OH)2(CO3)22-] a likely dominant form at the Clive site. Pu(VI) can also form complexes
with chloride ion under oxidizing conditions in high ionic strength solutions (Clark and Tait,
1996). Dissolved plutonium in the natural environmental is typically very low, in the 10-15 M
range (EPA 1999b), though higher levels are possible where a solid phase is present. At Yucca
Mountain the plutonium solubility used in the TSPA model ranged from 10-10 M to 10-6 M, with
an expected value of 10-8 M. (LANL 1997). A similar range is used for this PA model.
5.1.8 Protactinium
Protactinium can exist in two oxidation states in natural waters, +4 and +5. Both forms have a
propensity to form hydrolysis complexes (Morss, et al. 1977). Protactinium will also form
complexes with halides (F–, Cl–, Br–, I–) and sulfate. Very little information is available on the
protactinium species in circumneutral pH range. At Yucca Mountain the protactinium solubility
used in the TSPA model ranged from 10-10 M to 10-5 M (LANL 1997). A similar range is used
for this model.
5.1.9 Radium
Radium only exists in the +2 oxidation state in nature and is generally found uncomplexed as
Ra2+ (EPA 2004). Radium has similar chemical behavior as barium and forms a co-precipitate as
a sulfate [(Ba,Ra) SO4]. This co-precipitate would likely control the solubility at the Clive site if
radium reaches levels for saturation. At Yucca Mountain, the radium solubility used in the
TSPA model ranged from 10-9 M to 10-5 M, with an expected value of 10-7 M. (LANL 1997). For
this PA, a similar range was used with a central tendency value higher by a factor of 10.
5.1.10 Radon
Radon, in the form of 222Rn, is the longest-lived of all radon isotopes with a half-life of 3.8 days
and is considered the most environmentally important isotope. Radon exists as an essentially
inert gas and does not precipitate or sorb to any significant extent, but will partition between
Geochemical Modeling for the Clive DU PA
5 November 2015 23
aqueous and gas phases, according to its Henry’s Law constant, as discussed in the Unsaturated
Zone Modeling white paper. In the unsaturated zone radon will mainly exist in the gas phase, but
is soluble and within all zones the solubility is temperature sensitive. A fairly narrow solubility
range based on values from Langmuir (1997) was used in this PA model since temperature is
likely the largest factor impacting this species solubility.
5.1.11 Strontium
Strontium is expected to exist in the Sr2+ form in the aqueous environments at the Clive site.
Strontium has minimal tendency to form inorganic complexes (EPA 1999b), has a similar ionic
radius to that of calcium, and forms similar minerals including celestite (SrSO4) and strontianite
(SrCO3). In alkaline conditions with sufficient concentration, strontianite, or co-precipitation
with calcite and anhydrite, is expected to control the Sr2+ concentration. At Yucca Mountain the
strontium solubility used in the TSPA model ranged from 10-6 M to 10-3 M. (LANL 1997). A
similar range was used for this PA model.
5.1.12 Technetium
Technetium can exist in multiple oxidation states, but +7 is dominant under oxidizing conditions
(EPA 2004, Langmuir 1997, Wildung et al. 2004). In oxidizing conditions the species is the
oxyanion TcO4– which is highly soluble and is not known to form complexes. Under slightly
reducing conditions technetium exists as an uncharged hydroxide. Under stronger reducing
conditions technetium can form a very insoluble form. For this PA model, a solubility centered
around 10 -3 M was used.
5.1.13 Thorium
Thorium is expected to exist in the 4+ form at the Clive site. Thorium forms hydroxyl
complexes as well as carbonate and inorganic anion complexes. Thorium has very low solubility
(EPA 1999b). Hydrous thorium oxide can be used to develop a maximum solubility. At Yucca
Mountain the thorium solubility used in the TSPA model ranged from 10-10 M to 10-7 M. (LANL
1997). This value is consistent with values in EPA (1999c), which described a range of 10-8.5 M
to 10-9 M for a pH range of 5 to 9. The solubility of hydrous thorium oxide increases 2 to 3
orders of magnitude with increasing ionic strength (EPA 1999b). This behavior is significant for
the shallow aquifer zone of this model resulting in a solubility range that was chosen near the
upper end of that used in the Yucca Mountain repository study.
5.1.14 Uranium
5.1.14.1 Uranium Forms and Geochemical Model Parameters
The DU waste proposed for disposal at the Clive facility is in two main uranium oxide forms:
UO3 and U3O8. Uranium trioxide, UO3, is the waste form received from SRS. The uranium
oxides expected to be produced at the deconversion plants in Portsmouth, OH, and Paducah KY
(the GDP DU) are anticipated to be predominantly U3O8, with small amounts of UO2. The U.S.
DOE has characterized U3O8 as insoluble (ANL 2010, DOE 2001). The exact solid phase that
will control uranium solubility for the Clive Facility is not known and would require extensive
laboratory testing to determine. Based upon the results outlined by the several research groups
Geochemical Modeling for the Clive DU PA
5 November 2015 24
described above, schoepite likely is the major contributor, and this solid was selected to develop
the solubility distribution in this PA for the UO3 form. This is a conservative assumption in that
schoepite is more soluble than uranyl carbonate and much more soluble than U3O8. The
solubility of U3O8 is also incorporated into the GoldSim model as an option for the model user.
The Clive DU PA Model v1.4 defaults to U for solubility for the 10,000-year model. For the
Deep Time model, the solubility of U3O8 is used.
Due to the importance of uranium solubility to this PA, the input distribution was derived from
geochemical modeling. The model Visual MINTEQ (Gustafsson 2011) was utilized. This
geochemical code is based on the EPA MINTEQA2 program, and was used with its default
database. MINTEQ allows for a large number of uranium mineral forms to be examined. The
following were considered the most important for this PA: schoepite, U3O8 (crystalline), U4O9
(crystalline), UO2 (amorphous), B_UO2(OH)2, rutherfordine, and uraninite.
Under oxidizing subsurface conditions U(VI) as the UO22+ uranyl complex, is the predominant
oxidation state and is not easily reduced geochemically. Experiments by Reed et al. (1996)
indicate the uranyl complex can persist for over two years, even under high ionic strength anoxic
conditions. However, with strongly reducing conditions the U(IV) species can form. Based on
the pH and redox conditions at Clive, aqueous uranium is expected to be predominantly in the +6
form [U(VI)] in all three zones (Langmuir 1997, EPA 1999b). However, some of the
groundwater measurements do indicate areas of negative Eh (reducing conditions) where
uranium could exist in either the U(V) or U(IV) form if bioreduction or reduced iron exists.
Under low carbonate levels uranium exists as a polynuclear hydrolysis species. However, the
carbonate chemistry associated with the groundwater and the atmospheric CO2 partial pressure
will promote the formation of carbonate complexes with uranium. These complexes can increase
the overall uranium solubility. The carbonate complex UO2(CO3)34– is accepted as a major
complex at high carbonate concentrations (Clark et al., 1995, Langmuir 1997). The solid, uranyl
carbonate, UO2CO3, can also potentially limit the uranium solubility (DOE 2009). Studies where
this was the dominant species indicates that solubilities decreases with increasing ionic strength.
Divalent metal uranium carbonate complexation [e.g., Ca2UO2(CO3)3(aq)] is also possible
(Bernhard, et al. 2001, Wan et al. 2008).
The scenario within the waste and unsaturated zones is expected to be somewhat analogous to
the experimental system described by Wronkiewics et al. (1992) and also modeled by De Windt
et al. (2003). In the work by Wronkiewics et al. the Unsaturated Test Method was used to study
the dissolution and precipitation of UO2 at 90 degrees Celsius. Note that UO2 is uncommon in
the Clive inventory, but may make up a small amount of the GDP DU. The UO2 was freshly
prepared containing natural uranium isotope abundances and leached with water from well J-13
near Yucca Mountain, Nevada that had been equilibrated with local tuff. During the course of
the test, this leachate was periodically injected into the top of the system, water samples were
collected, and the UO2 was visually inspected. They found an initially low concentration of
uranium in the outflow followed by a slug of uranium that leveled off over the circa 238-week
(4.5-yr) period. Formic and oxalic acids were detected in the leachate but not found in the
starting material. This was attributed this to a potential radiolytic effect. Of particular
importance was the change in uranium oxide phases with time. A number of secondary uranium
oxides were formed as the uranium first dissolved, then later precipitated as a different oxide.
Schoepite (UO3•xH2O), dehydrated schoepite (UO3), uranophane and a number of other uranium
Geochemical Modeling for the Clive DU PA
5 November 2015 25
hydroxide minerals including uranium alkali silicates were formed. In all cases these indicated
uranium was present only in the U(VI) redox state. The solubility limit, based upon the steady-
state uranium release after two years of leaching, was attributed to precipitation of uranyl
silicates on the surface, limiting additional dissolution. De Windt et al. (2003) modeled UO2
oxidative dissolution in a saturated zone under oxidizing conditions. They found that uranium
mobility was controlled by schoepite, the dominant mineral formed. The total uranium
concentration decreased for the first 100 years from a maximum of approximately 400 µM to a
constant 10-5 M after consumption of all of the UO2. This mineral paragenesis is similar to that
observed for oxidized zones in natural uraninite. Modeling by Langmuir (1997) indicates that
uranium solubility based on schoepite ranges from 10-6 M to 10-4 M, depending upon pCO2
levels. This is consistent with the values in the Wronkiewics et al. (1992) study and modeled by
De Windt (2003). Though analogous to what can happen at the Clive facility, it is critical to
realize Wronkiewics et al. and De Windt started with a very soluble form of uranium, UO2, as
compared with the much more stable UO3 and especially U3O8 DU waste at Clive. U3O8 is
considered one of the most thermodynamically and kinetically stable forms of uranium. Also of
note above was the potential for dissolution to be reduced by the formation of less soluble uranyl
silicates on the surface of the starting material.
At the Clive Facility, should formic and oxalic acid be formed radiolytically, they also could
increase the overall uranium solubility (Langmuir, 1997). However, the generation of these
acids is not expected to be significant nor should it have a significant effect on pH. This is
especially true for any water that leaches from the waste zone into the shallow aquifer due to the
buffering capacity of this aquifer. As such, this effect would be transitory at best.
To derive uranium solubility distributions and speciation, the geochemical modeling program
was run in individual batches, where for a given run the parameters pH, temperature (13°C), pe
(Eh), and water density were fixed. Multiple runs were performed by adjusting a single
parameter. The effect of temperature on solubility, with a range of approximately 10°C to 25°C,
is insignificant relative to the uncertainty of measurements and modeling. A single temperature
value of 13°C was used in the geochemical modeling. To account for the thermodynamic
activity in this high TDS saturated zone system, the Bronsted-Guggenheim-Scatchard specific
ion interaction theory (SIT) (Nordstrom and Munoz, 1994) model is utilized. The SIT model is
applicable to high ionic strength solutions up to approximately 4 molal. When lower TDS
parameters were utilized the Davies equation was employed. Most of the thermodynamic
parameters used in Visual MINTEQ for the uranium species are derived from the Nuclear
Energy Agency database (2003) available in the Visual MINTEQ program.
The level of ions other than uranium used in the geochemical modeling was set for two different
conditions. The first condition was set based on the ions found as the average of the data from
well GW-25 at the Clive facility. The groundwater chemistry of this well and several others is
shown in Table 7 and Table 8, representing the conditions in the very high TDS shallow aquifer.
Modeling indicated that pH and bicarbonate/pCO2 had the largest influence on total uranium
concentrations within the Eh range of 800 to -200 mV. When the redox conditions were reduced
to -500 mV, U3O8 precipitated at very low uranium concentrations. These low redox conditions
are not anticipated at Clive except under transient conditions.
Geochemical Modeling for the Clive DU PA
5 November 2015 26
A second set of geochemical modeling runs was performed under much lower TDS conditions.
In this set lower sodium and chlorine levels were chosen to more closely represent the water
phase in the unsaturated zones, at least in the upper waste prior to significant salt dissolution.
The mineral form of uranium anticipated from the two main sites SRS (as schoepite, a form of
UO3,) and the GDPs (as U3O8) were included as infinite solids in separate geochemical models.
Basing solubility limits solely on a solid phase is not without uncertainty (OECD 1997, Chapter
IV). Consequently, solubility distributions are incorporated into the PA to account for this
uncertainty. The pH, Eh, and carbonate conditions were varied and the resulting total uranium
concentration was calculated. Uranium solubility limits were based on the concentration at
which each mineral reached saturation. Differences between the high ionic strength modeling
runs and those with lower sodium and chloride were not significant enough, relative to the
uncertainty of the modeling, to justify separate solubility distributions. Complete reports from
each of the model runs are available.
5.1.14.2 Uranium Solubilities based on Schoepite
The solubility of uranium for the UO3 waste form was derived using the mineral schoepite in the
geochemical code Visual MINTEQ as described above. The parameters were adjusted so that
they would mimic the range of conditions at the site, with schoepite provided as an infinite solid
source. TDS values were elevated, similar to the shallow aquifer. The total uranium solubility
outputs from these geochemical models are shown in Table 10. As shown from these model
results, uranium solubilities range from 10 mg/L to 100 mg/L. Using schoepite [U(VI)] as the
controlling mineral indicates uranium levels could reach as high as 100 mg/L (4.2 × 10-4 M)
under several scenarios. These values are similar to those found experimentally by Wronkiewics
et al. (1992).
Choppin (2000) performed batch solubility studies using a synthetic water and mixed uranium
oxides under both anoxic and oxidizing conditions. In the absence of humic compounds, the
solubility of uranium was 114±2 mg/L and 94±2 mg/L under oxic (Eh of 220 mV) and anoxic
(Eh 138 mV) conditions respectively. The pH of the solution ranged from 8 to 8.2. A 1 ppm
humic acids level had little effect on solubility. The data by Choppin (2000) are also similar to
estimates from this geochemical modeling for schoepite. Tomasko (2001) modeled uranium
transport using UF4 as the disposed form. This is a very soluble form of uranium, much more so
than the uranium oxides that are expected in the Clive facility inventory. The study by Tomasko
considered uranium tetrafluoride disposed in 30- or 50-gallon drums. After exposure to water
the uranium underwent hydrolysis to form schoepite or U3O8. No solubility experiments were
performed by Tomasko though he cited work by others to develop the solubility limits. Under
oxidizing conditions Tomasko envisioned the schoepite being formed from the U3O8. Tomasko
used a range of uranium solubilities from 24 mg/L, the solubility he cited for schoepite, to a
much higher value of to 23,000 mg/L. This higher value was based on the potential formation of
ammonium carbonate uranium complexes. These complexes are not considered significant
species at the Clive location due to the low nitrogen availability of the system.

Geochemical Modeling for the Clive DU PA
5 November 2015 27
Table 9: Model Results for High TDS System analogous to the Upper Aquifer. Uranium
solubility limit based on Schoepite. *
pH Bicarbonate
(mg/L)
Eh
(mV)
Total Uranium
(mg/L)
Total Uranium
(mol/L)
6.5 190 200 28.1 1.18E-4
7.2 10 200 2.3 9.72E-6
7 190 200 75.4 3.17E-4
7.3 190 811 58.5 2.46E-4
7.2 300 200 241 1.01E-3
7.5 500 200 421 1.77E-3
8 300 200 428 1.80E-3
*Data in this table include the following constant parameters: specific gravity of water 1.03, stoichiometric
ionic strength 0.88, and anions and cations levels as shown in Table 5 and 6 above for GW-25.
The geochemical model runs listed in Table 9 were also repeated at a reduced TDS condition as
shown in Table 10. The results were sufficiently similar that the model is considered to be
applicable to both the saturated and unsaturated regions. When the model is adjusted so that
levels of sodium and chloride are reduced significantly (Na = 6,465 mg/L, Cl = 10,783 mg/L) at
pH 7.25, the uranium solubilities with schoepite are not significantly different, as shown in
Table 10. As such, the GoldSim model did not utilize different solubilities for the different
levels of TDS and instead the higher maximum values for the distributions were chosen as a
conservative approach. Though this approach is a bit conservative, as indicated above the range
of TDS that is expected throughout the unsaturated and saturated zones does not have a large
impact on uranium solubility.
Table 10: Total uranium, low TDS (ionic strength 0.127 M). Uranium solubility limit based
on schoepite.
pH Bicarbonate
(mg/L)
Eh
(mV)
Total Uranium
(mg/L)
Total Uranium
(mol/L)
6.5 190 200 152.8 6.42E-4
7.2 10 200 2.28 9.58E-6
7 190 200 162.32 6.82E-4
7.3 190 811 172.55 7.25E-4
7.2 300 200 278.46 1.17E-3
7.5 500 200 485.52 2.04E-3
8 300 200 307.02 1.29E-3

Geochemical Modeling for the Clive DU PA
5 November 2015 28
The major dissolved uranium species considered in the geochemical model simulations included
those shown in Table 11. A mixture of primarily anions, but also uncharged and cations are
included. The uncharged and anionic species make up the major species under the conditions
modeled for schoepite as shown in Table 11.
Table 11: Major dissolved uranium (VI) species included in geochemical models.
Uranium Species
UO2(CO3)3-4
UO2(CO3)2-2
Ca2 UO2(CO3)3 (aq)
CaUO2(CO3)3-2
(UO2)3(CO3 )6-6
UO2(OH)2 (aq)
UO2(OH)3-
(UO2)4(OH)7+
(UO2)3(OH)5+

5.1.14.3 Uranium Solubilities based on U3O8
The solubility of uranium for the U3O8 waste form was derived by directly setting this solid form
as an infinite source in the geochemical code Visual MINTEQ as described above. Results are
provided in Table 12. The differences in solubility between schoepite and U3O8 are pronounced.
U3O8 has significantly lower solubility within the geochemical conditions expected at the Clive
Facility. Only at very anoxic conditions does U3O8 show a solubility approaching that of UO3.
The amount of current experimental data on the solubility of this mineral is limited. As stated
above, the DOE considers U3O8 insoluble.
Table 12: Total Uranium, low TDS (ionic strength 0.127 M). Uranium solubility limit
based on the mineral U3O8. *
pH Bicarbonate
(mg/L)
Eh
(mV)
Total Uranium
(mg/L)
Total Uranium
(mol/L)
6.5 190 200 1.87E-10 7.85E-16
7 190 200 7.14E-11 3.00E-16
8 300 200 2.38E-11 1.00E-16
7.3 190 -10 1.19E-6 4.98E-12
7.3 190 -40 6.00E-6 2.52E-11
7.3 190 -100 1.54E-4 6.45E-10
7.3 190 -300 7.57E+0 3.18E-5
*Data in this table include the following constant parameters: specific gravity of water 1.03, stoichiometric
ionic strength 0.88, and anion and cation levels as shown in Table 7 and Table 8 above for GW-25.
Geochemical Modeling for the Clive DU PA
5 November 2015 29
GoldSim Model Note: The GoldSim model cannot run with solubilities for both UO3 and U3O8
simultaneously within a single 10,000-year simulation because only one solubility is used for
each element. To account for the differences in solubility between the two uranium oxide wastes
in the inventory, the Control Panel for the model provides the ability to use either the default
solubility based on UO3 or to select the solubility based on U3O8. One may also select either or
both waste inventories. Based upon evaluation of the Clive DU PA Model, the UO3 solubility is
governing the uranium concentrations into the groundwater for about 50,000 yr. This indicates
that the inability to have two separate solubilities in the model for the two waste forms of DU is
not affecting the simulation results. For the Deep Time model, the solubility of U3O8 is used for
uranium solubility.
6.0 Ionic and Molecular Diffusion Coefficients
The diffusion coefficient (Dm) is required for calculating the movement of solutes due to
differences in concentration gradient. Movement by diffusion can occur without advective flow
of water. Ionic and molecular diffusion coefficients are derived in theory from the
Stokes-Einstein equation:
𝑫𝒎=𝑹𝑻/𝟔𝝅𝜼𝑩𝒓𝑨 (5)
where
R = universal gas constant,
T = temperature,
ηB = absolute viscosity of the solvent (water), and
rA = radius of the assumed spherical solute.
A variety of empirical equations have been derived based on the Stokes-Einstein equation for
different scenarios. For a dilute solution of a single salt the diffusion coefficient can be derived
from the Nernst-Haskell equation (Reid et al., 1987). This equation includes the valence of the
cation and anions as well as ionic conductances. Specific ionic conductances are required for
each cation and anion species. When two or more chemical species are present at different
concentrations, interdiffusion (counterdiffusion) must be included to satisfy electroneutrality
(Lerman 1979). For a geochemical system as large as that found in radioactive waste disposal
facilities this quickly becomes too complex to model, even if ionic conductivities are available
for each species.
An additional difficulty in deriving ion-specific diffusion coefficients lies in the large number of
potential ions. The number of radioactive waste elements typically modeled may be 30 to 40,
and for each element in this list one can expect multiple forms. For example, U has 4 redox
states, and many soluble species for each of these. Assuming oxic conditions U will be primarily
found as UO2(CO3)34–, UO2(CO3)22–, and UO2CO3, but there are at least 8 additional forms of
U(+6) that may be found. Thus the potential number of ions that would need to be included in
the model would easily be in the hundreds. Obtaining the parameters for each species that would
be required to model the ionic diffusion would be difficult.
Geochemical Modeling for the Clive DU PA
5 November 2015 30
Given these issues with developing ion-specific values of Dm , the approach used in modeling
diffusion in the PA model is to use a range of Dm values. This range can be derived from Table
3.1 in Lerman (1979). For conditions near 25°C, the range of Dm for the elements of interest is
4 × 10–6 to 2 × 10–5 cm2/s. For cooler temperatures, which would be expected in the deeper
subsurface, the values are somewhat lower. The values for 25°C are reproduced in Table 13.
Based on these values, the diffusion coefficient is represented in the Clive DU PA Model as a
uniform distribution with a minimum of 3 × 10–6 cm2/s and a maximum of 2 × 10–5 cm2/s, and is
the same for all elements.

Table 13. Diffusion coefficients for selected cations and anions.
Cation Dm
(10–6 cm2/s)
Anion Dm
(10–6 cm2/s)
K+ 19.6 Cl– 20.3
Cs+ 20.7 I– 20
Sr2+ 7.94 IO3– 10.6
Ba2+ 8.48
Ra2+ 8.89
Co2+ 6.99
Ni2+ 6.79
Cd2+ 7.17
Pb2+ 9.45
UO22+ 4.26
Al3+ 5.59
SOURCE: Table 3.1 Lerman (1979)

Geochemical Modeling for the Clive DU PA
5 November 2015 31
7.0 References
Adrian Brown Consultants, Response to UDEQ Kd Interrogatories, Dated April 22, 1997 Report
3101B.970422.
ANL Characteristics of Uranium. Web site accessed 2010.
http://web.ead.anl.gov/uranium/guide/ucompound/propertiesu/octaoxide.cfm
ANL 2000. Colloid-Associated Radionuclide Concentration Limits: ANL. ANL-ES5-MO-
0000.20 REV 00 leN 01
Beals, D. M., S. P. LaMont, J. R. Cadieux, C. R. Shick, and G. Hall. Determination of Trace
Radionuclides in SRS Depleted Uranium (DU). November 19, 2002. WSRC-TR-2002-
00536 Westinghouse Savannah River Company, SRS, Aiken, SC 29808.
Bernhard G, G Geipel, T Riech, V Brendler, S Amayri, and H Nitsche. 2001. Uranyl (VI)
Carbonate Complex Formation: Validation of the Ca2UO2(CO3)3 (aq) Species.
Radiochimica Acta 89:511-518.
Bingham Environmental, 1991. Hydrogeologic Report Envirocare Waste Disposal FacilitySouth
Clive, Utah. Final version October 9, 1991.
Bingham Environmental. 1994. Hydrogeologic report Mixed Waste Disposal Area Envirocare
Waste Disposal Facility South Clive, Utah. November 18, 1994. Prepared for Envirocare of
Utah. Salt Lake City, UT.Bingham Environmental, Project Memorandum. Summary of
Results, Radionuclide Kd Tests, Envirocare Disposal Landfills, Clive, Utah. August 3,
1995.
Bingham Environmental, Project Memorandum. Summary of Results, Radionuclide Kd Tests,
Envirocare Disposal Landfills, Clive, Utah. January 25, 1996.
Choppin, G. R. Idaho National Engineering and Environmental Laboratory Publication.
INEEL/EXT-01-00762 Rev. 0. November 2000. Actinide Solubility Experiments in
INEEL Perched Simulant Solution.
Clark, D. L. and Tait, C. D. 1996. Monthly Reports Under SNL Contract AP2274, Sandia WIPP
Central File A:WBS 1.1.10.1.1. These data are qualified under LANL QAPjP CST-OSD-
QAP1-001/0. WPO 31106.
Clark, D. L., D. E. Hobart, and M. P. Neu. Actinide Carbonate Complexes and Their Importance
in Actinide Environmental Chemistry. Chemical Reviews, Vol 95: 25. 1995
CRWMS M&O (Civilian Radioactive Waste Management System). 2000. Colloid-Associated
Radionuclide Concentration Limits. ANL-EBS-MD-000020 REV 00 ICN 01.
Cui, D. and Eriksen, T. 1996. Reduction of Tc(VII) and Np(V) in solution by ferrous iron. SKP
TR 96-03.
Geochemical Modeling for the Clive DU PA
5 November 2015 32
De Windt L., Burnol A., Montaranl P., van der Lee J. 2003. Intercomparison of reactive transport
models applied to UO2 oxidative dissolution and uranium migration. Journal of
Contaminant Hydrology 61, 303-312, 2003.
Degueldre, C., I. Triay, J. Kim, P, Vilks, M. Laaksoharju, N. Mieleley. 2000. “Groundwater
colloid properties: a global approach.” Applied Geochemistry, vol.15, p. 1043 – 1052.
DOE 2001. http://web.ead.anl.gov/uranium/pdf/UraniumCharacteristicsFS.pdf
DOE. 2003. Publication 45185. Evaluation of Surface Complexation Models for Radionuclide
Transport at the Nevada Test Site: Data Availability and Parameter Evaluation. D. Decker
and C. Papelis. May, 2003.
DOE. 2009. Waste Isolation Pilot Plant (WIPP), SOTERM-2009. Title 40 CFR Part 191
Subparts B and C Compliance Recertification Application for the Waste Isolation Pilot
Plant. Appendix SOTERM-2009 Actinide Chemistry Source Term.
Envirocare of Utah. Revised Hydrogeological Report for the Envirocare Waste Disposal
Facility, Clive, Utah. Version 2.0, 2004.
Envirocare of Utah. Metals Distribution Coefficient Values Relevant to the Envirocare Site.
Memorandum and Report to the US Nuclear Regulatory Commission. 2000.
EPA. 1999a. Understanding Variation in Partition Coefficient, Kd, Values. Volume I. 402-R-99-
004A. US Environmental Protection Agency, Washington, DC.
EPA. 1999b. Understanding Variation in Partition Coefficient, Kd, Values. Volume II (1999).
402-R-99-004C. US Environmental Protection Agency, Washington, DC.
EPA. 2004. Understanding Variation in Partition Coefficient, Kd, Values. Volume III. 402-R-04-
002C. US Environmental Protection Agency, Washington, DC.
Geckeis, H. and T. Rabung. Actinide Geochemistry: From the Molecular Level to the Real
System. Journal of Contaminant Hydrology 102, 187-195, 2008.
Geckeis, H., Schäfer, Th., Hauser, W., Rabung, Th., Missana, T., Degueldre, C., Möri, A.,
Eikenberg, J., Fierz, Th., Alexander, W.R., 2004. Results of the colloid and radionuclide
retention experiment (CRR) at the Grimsel Test Site (GTS), Switzerland — impact of
reaction kinetics and speciation on radionuclide migration. Radiochim. Acta 92, 765–774.
Glover, P. A., F. J. Miner and W. O. Polzer. 1976. “Plutonium and Americium Behavior in the
Soil/Water Environment. I. Sorption of Plutonium and Americium by Soils.” In
Proceedings of Actinide-Sediment Reactions Working Meeting, Seattle, Washington, pp.
225-254, BNWL-2117, Battelle Pacific Northwest Laboratories, Richland, Washington.
Gustafsson, J.P., 2011. Visual MINTEQ ver 3.0. Based on the USEPA MINTEQA2 software,
(http://www2.lwr.kth.se/English/OurSoftware/vminteq/)
Geochemical Modeling for the Clive DU PA
5 November 2015 33
Heberling, F., B. Brendebach, and D. Bosbach. Neptunium(V) Adsorption to Calcite. Journal of
Contaminant Hydrology 102, 246-252, 2008.
Hull, L. C, and A. L. Schaefer. 2008. Accelerated transport of 90Sr following a release of a
high ionic strength solution in vadose zone sediments. Journal of Contaminant Hydrology
97, 135-157.
Johnson, G.L., and L.M. Toth. 1978. Plutonium(IV) and Thorium(IV) Hydrous Polymer
Chemistry. ORNL/TM-6365. Oak Ridge, TN: Oak Ridge National Laboratory, Chemistry
Division.
Keeney-Kennicutt, W.L., J.W. Morse. 1985. The redox chemistry of Pu(V)O2+ interaction with
common mineral surfaces in dilute solutions and seawater, Geochimica et Cosmochimica
Acta, Vol. 49, pp. 2577–2588.
Kerrisk, J.F. Ground-Water Chemistry at Yucca Mountain, Nevada and Vicinity. LA-10929-MS
February 1987.
Kim, J.I., 1991. Actinide colloid generation in groundwater. Radiochim. Acta 52/53, 71–81.
Krupka K.M., R.J. Serne, and D.I. Kaplan. 2004. Geochemical Data Package for the 2005
Hanford Integrated Disposal Facility Performance Assessment. PNNL-13037, Rev 2,
Pacific Northwest National Laboratory, Richland, Washington.
Kumata, M., Vandergraaf, T.T. And Jujnke, D. G. 1993. The migration behavior of neptunium
under deep geological conditions. Migration '93 Abstracts, p 73.
LANL (Los Alamos National Laboratory), LA-13262-MS. 1997. Summary and Synthesis
Report on Radionuclide Retardation for the Yucca Mountain Site Characterization Project.
Last G.V., E.J. Freeman, K.J. Cantrell, M.J. Fayer, G.W. Gee, W.E. Nichols, B.N. Bjornstad, and
D.G. Horton. 2004. Vadose Zone Hydrogeology Data Package for the 2004 Composite
Analysis. PNNL-14702 Rev. 0, Pacific Northwest National Laboratory, Richland,
Washington.
Lerman, A. 1979. Geochemical Processes in Water and Sediment Environments. Wiley-
Interscience. QES71.L45
Langmuir, D. Aqueous Environmental Chemistry. Prentice Hall 1997.
Linsalata, P. and N. Cohen. Determination of the Distribution Coefficient of Plutonium in
WATER-Secimetn Systems. Health Physics 39:1040-1041. 1980.
Möri, A., A., W.R., Geckeis, H., Hauser, W., Schäfer, Th., Eikenberg, J., Fierz, Th., Degueldre,
C., Missana, T., 2003. The colloid and radionuclide retardation experiment at the Grimsel
Test Site: influence of bentonite colloids on radionuclide migration in a fractured rock.
Coll. Surf. 217, 33.
Geochemical Modeling for the Clive DU PA
5 November 2015 34
Morss, L., N.M Edelstein, J. Fuger, and J.Katz. The Chemistry of the Actinide and Transactinide
Elements. Third Edition, Volume 1. 1977 Springer.
Nakata, K., Nagasaki, S., Tanaka, S., Sakamoto, Y., Tanaka, T. and Ogawa, H. 2002. Sorption
and reduction of neptunium (V) on surface or iorn oxides. Radiochim. Acta 90 pp 665-669.
Nordstrom, D. K, and J. L. Munoz. 1994. Geochemical thermodynamics. 2nd edition.
Cambridge, MA; Blackwell Scientific Publications, Inc.
OECD Publication 1997. Modelling in Aquatic Chemistry. Chapter IV, Raulf Grauer. Scientific
Editors Ingmar Grenthe and Ignasi Puigdomenech
ORNL (Oak Ridge National Laboratory). 2000. J. R. Hightower, et al. Strategy for
Characterizing Transuranics and Technetium Contamination in Depleted UF6 Cylinders.
ORNL/TM-2000-242 Chemical Technology Division, ORNL. 2000.
Patterson, R.J., Spoel, T., 1981. Laboratory measurements of strontium distribution coefficeint
Kd -Sr for sediments from a shallow sand aquifer. Water Resources Research 17 (3), 513-
520.
Powell, B.A., R.A. Fjeld, D.I. Kaplan, J.T. Coates, and S.M. Serkiz. 2005. Pu(V)O2+ adsorption
and reduction by synthetic hematite and goethite, Environ. Sci. Technol., Vol. 39,
pp. 2107–2114.
Reed, D.T., D. R. Wygmans, and M. K. Richman. Actinide Stability/Solubility in Simulated
WIPP Brines. Argonne National Laboratory, Actinide Speciation and Chemistry Group,
Chemical Technology Group. Interim Report 1996.
Reid, R.C., Prausnitz, J.M., and Poling B.E., 1987. The Properties of Gases and Liquids, 4th
Edition. McGraw-Hill, Inc. TP242.R4.
Ryan, J.N., Elimelech, M., 1996. Colloid mobilization and transport in groundwater. Colloids
and Surfaces. 107, 1–56.
Sanchez, A.L., J.W. Murray, and T.H. Sibley. 1985. The adsorption of plutonium-IV and
plutonium-V on goethite, Geochimica et Cosmochimica Acta, Vol. 49, p. 2297, 1985.
Sandia (Sandia National Laboratories) 2001. Compliance Assessment Document for the
Transuranic Wastes in the Greater Confinement Disposal Boreholes at the Nevada Test
Sites. Volume 2: Performance Assessment. Version 2.0.
Schaefer, D. H., S. A. Thiros, and M. R. Rosen. Ground-Water Quality in the Carbonate-Rock
Aquifer of the Great Basin, Nevada and Utah, 2003. U.S. Geological Survey, National
Water-Quality Assessment Program. Scientific Investigations Report 2005-5232.
Scism, C. D. 2006. The Sorption/Desorption Behavior of Uranium in Transport Studies using
Yucca Mountain Alluvium. Los Alamos National Laboratory LA-14271-T. 2006.
Geochemical Modeling for the Clive DU PA
5 November 2015 35
Serne, R. J. 2007. Kd Values for Agricultural and Surface Soils for Use in Hanford Site Farm,
Residential, and River Shore Scenarios. Technical Report for Ground-Water Protection
Project. PNNL-16531. August 2007.
Sheppard, M. and D. H. Thibault. 1990. Default Soil Solid/Liquid Partition Coefficients, KdS,
for Four Major Soil Types: A Compendium. Health Physics Vol 59, No 4, pp 471-482.
October 1990.
Tomasko, D., 2001, Groundwater Calculations for Depleted Uranium Disposed of as Uranium
Tetrafluoride (UF4). ANL/EAD/TM-111, Argonne National Laboratory, Argonne, Ill.
USDA. Nat Resources Conservation Service, Tooele Area Soil Survey Utah, Version 5, Sept 2,
2009.
Vandergraaf, T. T., Ticknor, K.V., and George, I. M. 1984. Reactions between Technetium in
Solution and Iron-Containing Minerals Under Oxic and Anoxic Conditions. ACS
Symposium Series, 246, pp.25-43, Atomic Energy of Canada Limited Report, AECL-7957.
Vandergraaf, T. T., Ticknor, K.V., and Melnyk, T. W. 1993. The selection of a sorption data
base for the geosphere model in the Canadian Nuclear Fuel Waste Management Program.
Journal of Contaminant Hydrology, 13, 327-345.
Voegelin, A., Kretzschmar, R., 2002. Stability and Mobility of Colloids in Opalinus Clay,
Institute of Terrestrial Ecology, ETH Zürich, NTB 02-14, December 2002, 33 pages.
Wan, J., T.K. Tokunaga, Y. Kim, Z. Wang, A. Lanzirotti, E. Saiz, and R.J. Serne, Effect of saline
waste solution infiltration rates on uranium retention and spatial distribution in Hanford
sediments, Environ. Sci. Technol., 42, 1973-1978, 2008.
Whetstone Associates. Technical Memorandum to Energy Solutions from Whetstone
Associates, Oct 30, 2009.
Wildung, R. E., Li S. W., Murray, C. J., Krupka, K. M., Xie, Y., Hess, H. J., and Rogen, E.E.
Technetium reduction in sediments of a shallow aquifer exhibiting dissimilatory iron
reduction potential. FEMS Microbiology Ecology. 49, 151-162, 2004.
Wooyong Um, R. J. Serne, G. V. Last, R. E. Clayton, and E. T. Glossbrenner. The Effect of
Gravel Size Fractions on the Distribution Coefficients of Selected Radionuclides. Journal
of Contaminant Hydrology 107, 82-90, 2009.
Wronkiewicz, D. J, Bates, J. K., Gerding, T. J., Veleckis, E., and Tani B. S. Journal of Nuclear
Materials 190. 107-127. 1992.

NAC-0016_R4

Saturated Zone Modeling for the Clive DU PA
31 October 2015 ii
1. Title: Saturated Zone Modeling for the Clive DU PA
2. Filename: Saturated Zone Modeling v1.4.docx
3. Description: This white paper provides documentation of the development of parameter
values and distributions used for modeling liquid phase transport in the saturated zone for the
Clive DU PA Model v1.4.
Name Date
4. Originator Michael Sully 5 May 2014
5. Reviewer Dan Levitt 20 May 2014
6. Remarks

20 Oct 2015: Revised figure to bring up to date for v1.4. – J Tauxe

Saturated Zone Modeling for the Clive DU PA
31 October 2015 iii

This page is intentionally blank, aside from this statement.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 iv
CONTENTS
FIGURES .........................................................................................................................................v
TABLES ........................................................................................................................................ vi
1.0 Summary of Parameters and Distributions .............................................................................1
2.0 Clive Site Hydrogeology ........................................................................................................2
3.0 Groundwater Flow Parameter Distributions ...........................................................................3
3.1 Saturated Hydraulic Conductivity .....................................................................................3
3.2 Bulk Density and Porosity ................................................................................................4
3.3 Hydraulic Gradient ............................................................................................................4
4.0 Groundwater Transport Parameter Distributions ....................................................................5
4.1 Saturated Zone Dimensions ..............................................................................................6
4.2 Dispersion .......................................................................................................................11
5.0 References .............................................................................................................................14

Saturated Zone Modeling for the Clive DU PA
31 October 2015 v
FIGURES
Figure 1. Location and extent of the saturated zone modeling domain including location of
the DU waste, the point of compliance monitoring well, the buffer zone of the DU
cell, and outer boundaries of property owned and controlled by EnergySolutions. ......6
Figure 2: Schematic representation of unsaturated zone and shallow aquifer transport using
cell pathways; section parallel to groundwater flow direction. .....................................7
Figure 3. Cross-section D-D' modified from Envirocare (2004) showing estimated elevation
of the bottom of the shallow aquifer. ...........................................................................10

Saturated Zone Modeling for the Clive DU PA
31 October 2015 vi
TABLES
Table 1. Summary of saturated zone parameter distributions ..........................................................1
Table 2. Texture class, thickness range, and average thickness for the hydrostratigraphic units
underlying the Clive site. ...............................................................................................3
Table 3. Construction details for selected wells used for estimating the elevation of the
bottom of the shallow aquifer. .......................................................................................8
Table 4. Construction details for selected wells used for water table elevations. ...........................9
Table 5. Water table elevations, aquifer bottom elevations and estimated saturated thickness
of the shallow aquifer.....................................................................................................9

Saturated Zone Modeling for the Clive DU PA
31 October 2015 1
1.0 Summary of Parameters and Distributions
This section is a brief summary of parameters and distributions used for modeling saturated zone
processes for the Clive Depleted Uranium (DU) Performance Assessment (PA) Model. For
distributions, the following notation is used:
• N( μ, σ, [min, max] ) represents a normal distribution with mean μ and standard deviation
σ, and optional truncation at the specified minimum and maximum,
• LN( GM, GSD, [min, max] ) represents a log-normal distribution with geometric mean
GM and geometric standard deviation GSD, and optional min and max,
• U( min, max ) represents a uniform distribution with lower bound min and upper bound
max,
• Beta( μ, σ, min, max ) represents a generalized beta distribution with mean μ, standard
deviation σ, minimum min, and maximum max,
• Gamma( μ, σ ) represents a gamma distribution with mean μ and standard deviation σ,
and
• TRI( min, m, max ) represents a triangular distribution with lower bound min, mode m,
and upper bound max.

Note that some distributions are truncated at a minimum value of 0 or a value of Small, an
arbitrarily small number just greater than 0 defined in the GoldSim model, and a maximum of
Large, an arbitrarily large value defined in the GoldSim model, or sometimes 1 – Small,
depending on physical limits. These truncations are often a matter of physical limits (e.g.
precipitation cannot be negative), and in GoldSim’s distribution definitions, if truncations are
made, they must be made at both ends, so the very large value is chosen for the upper end.
Table 1. Summary of saturated zone parameter distributions
Parameter Distribution Units Comment
Saturated Hydraulic
Conductivity
N( 9.6e-4, 9.67e-5, min=Small,
max=Large )
cm/s See Section
Bulk Density N( 1.57, 0.05, min=Small, max=Large )
[standard deviation is a placeholder]
g/cm3 See Section
Porosity N( 0.29, 0.05, min=Small, max=1-Small )
[standard deviation is a placeholder]
— See Section
Hydraulic Gradient N (6.94 x 10-4, 1.27 x 10 -4 , min=0 ,
max=Large )
— See Section
Aquifer Thickness N ( 16.2, 0.25, min=0, max=Large ) ft See Section

Saturated Zone Modeling for the Clive DU PA
31 October 2015 2
2.0 Clive Site Hydrogeology
The site hydrogeology for the EnergySolutions' Clive facility has been described by Bingham
Environmental (1991, 1994) and Envirocare (2000, 2004). The most recently revised
hydrogeologic report prepared by Envirocare (2004) noted that the interpretations of structure
and stratigraphy presented in their report were consistent with previous presentations described
in Bingham Environmental (1991, 1994) and Envirocare (2000).
The following description of the Clive site hydrology is taken from the review prepared by
Envirocare (2004). The site is described as being located on lacustrine (lake bed) deposits
associated with the former Lake Bonneville. The sediments underlying the facility are principally
interbedded silt, sand, and clay. While the depth of the sediments below the site is not known,
the sediments extend to a depth of at least 620 feet (ft) (DWR 2014, water right number 16-816
and associated well log 11293). This minimum depth is based on a borehole log for a nearby well
that did not encounter bedrock at its total depth of 620 ft.
Sediments at the site are described by Bingham Environmental (1991, 1994) and Envirocare
(2000, 2004) as being classified into four hydrostratigraphic units (HSU). Predominant sediment
textural class, layer thickness range, and average layer thickness for each unit are listed in Table
2.
Unit 4: This unit begins at the ground surface and extends to between 6 ft and 16.5 ft below the
ground surface (bgs). The average thickness of this unit is 10 ft. This unit is composed of
finer grained low permeability silty clay and clay silt.
Unit 3: Unit 3 underlies Unit 4 and ranges from 7 ft to 25 ft in thickness. The average thickness
of this unit is 15 ft. Unit 3 is described as consisting of silty sand with occasional lenses
of silty to sandy clay.
Unit 2: Unit 2 underlies Unit 3 and ranges from 2.5 ft to 25 ft in thickness. The average
thickness of this unit is 15 ft. Unit 2 is described as being composed of clay with
occasional silty sand interbeds. A structure map was prepared by Envirocare (2004,
Figure 5) with contours representing the elevations of the top of the unit. This map
shows that the top surface of Unit 2 slopes downward gradually from east to west in the
vicinity of the Federal Cell (interchangeably termed the Federal DU Cell in this
document because of the focus of this model on disposal of DU).
Unit 1: Unit 1 is the bottom layer of this sequence. This unit is described as silty sand
interbedded with clay and silt layers. The thickness of this layer in the vicinity of the
Clive facility is known to be in excess of 620 ft. (DWR 2014, water right number 16-816
and associated well log 11293).
The aquifer system in the vicinity of the Clive Facility is described by Bingham Environmental
(1991, 1994) and Envirocare (2000, 2004) as consisting of unconsolidated basin-fill and
alluvial-fan aquifers. Characterization of the aquifer system is based on subsurface stratigraphy
observations from borehole logs and from potentiometric measurements.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 3
The aquifer system is described as being composed of two aquifers; a shallow, unconfined
aquifer and a deep confined aquifer. The shallow unconfined aquifer extends from the water
table to a depth of approximately 40 ft to 45 ft bgs. The deep confined aquifer is encountered at
approximately 45 ft bgs and extends through the valley fill (Bingham 1994). The water table in
the shallow aquifer is reported to be located in Unit 3 on the west side of the site and in Unit 2 on
the east side.
Table 2. Texture class, thickness range, and average thickness for the hydrostratigraphic
units underlying the Clive site.
Unit Sediment Texture Class Thickness
Range (ft)
Average
Thickness (ft)
4 silt and clay 6 – 16.5 10
3 silty sand with interbedded silt and clay layers 7 - 25 15
2 clay with occasional silty sand interbeds 2.5 - 25 15
1 silty sand with interbedded clay and silt layers > 620 > 620

Deeper saturated zones in Unit 1 below approximately 45 ft bgs are reported to show higher
potentiometric levels than the shallow unconfined aquifer. Differences in potentiometric levels
are attributed to the presence of the Unit 2 clays. These observations are interpreted as indicating
that the shallow unconfined aquifer below the site does not extend into Unit 1 but is contained
within Units 2 and 3. Unit 1 extends from approximately 45 ft bgs and contains the deep aquifer.
3.0 Groundwater Flow Parameter Distributions
The parameters used to calculate the groundwater flux are the saturated hydraulic conductivity
and the hydraulic gradient. The porosity is needed to calculate the mean groundwater velocity
from the flux.
3.1 Saturated Hydraulic Conductivity
To develop a distribution for saturated hydraulic conductivity (Ks), 253 measurements were
obtained for 122 locations in the vicinity of the cells and ponds. These measurements were
provided to N&C by EnergySolutions in an Excel workbook named “Hydraulic Cond.xls”
prepared by R. Sobocinski.
There are multiple measurements per location. Thus, in order to not over-represent those
locations, a random effects analysis of variance model was fitted, treating location as a random
effect, to produce estimates of the mean Ks and its associated standard error.
The average Ks across locations ranges from 2.23 × 10-6 cm/s to 5.95 × 10-3 cm/s. There is some
right-skew to the average Ks values, which results in a slight overestimate of the standard error in
the random-effects model. However, with 122 locations, the distribution of the mean will be
well-approximated with a normal distribution. The random effects model produces a mean Ks of
9.6 × 10-4 cm/s and standard error of 9.67 × 10-5 cm/s.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 4
3.2 Bulk Density and Porosity
Although no data have been provided, Whetstone (2000) provides some values for material
properties of the shallow aquifer. In Section 7.1.2 of that report, a deterministic value for bulk
density of 1.566 g/cm3 is listed as an input for the Whetstone (2000) model. That value was
adopted as a mean of a normal distribution, and was assigned a placeholder standard deviation of
0.05 g/cm3.
Similarly, section 7.1.3 of Whetstone (2000) offers a porosity for the shallow aquifer of 0.29.
That value was used as the mean of a normal distribution, and a placeholder standard deviation
of 0.05 was assigned.
3.3 Hydraulic Gradient
The statistical distribution for hydraulic gradient developed for the Clive DU PA Model is
specific to horizontal gradients in the shallow aquifer. Vertical gradients were not considered in
the model.

Monthly averages of the site-wide hydraulic gradient from 1999 through 2010 were calculated
by EnergySolutions from water level measurements. These data were used to establish a
distribution for the mean site-wide gradient. The influence of any off-normal conditions
occurring during the time period of the water level measurement data would be included in these
data. The uncertainty related to the mean is typically well-modeled by a normal distribution, due
to the effect of averaging. A difficulty with the gradient data is in establishing an appropriate
standard error for the mean, since there is considerable time correlation in the data. That is, the
values change less from month to month than they do over longer time periods. To account for
this behavior several auto-regressive, moving-average (ARMA) models (Brockwell and Davis
1996) were fit to determine a model that adequately captured the time with an adequate fit for the
time correlation. Amongst these models, a best model was chosen based on the Akaike
information criterion (AIC), and a standard error for the mean was established based on this
model's fit.

A performance assessment is based on estimates of the expected performance of the site. To
achieve a realistic estimate of expected performance, spatio-temporal scaling (upscaling) is
needed for defining parameter distributions in probabilistic models. These upscaled distributions
represent a large area/volume and time frame instead of only points in time and space. Spatio-
temporal scaling is critical for model definition and understanding the impact on uncertainty for
estimating 95th percentiles (for example) of model output distributions. Without proper scaling,
models outputs are compromised.

The influence of off-normal conditions on shallow groundwater flow is discussed in Envirocare
(2004) for two cases: In the first, flow was affected by localized recharge from a surface water
retention pond in the southwest corner of the facility in the spring of 1999 and in the second, a
ground water mound formed between March 1993 and spring 1997 below a borrow pit excavated
near the 11e.(2) Cells (neighboring the Federal Cell) that occasionally filled with rain water. The
mound decreased and was negligible by the time of the report in 2004. The latter of these
conditions was captured by the hydraulic gradient data set used to develop the distribution for the
Saturated Zone Modeling for the Clive DU PA
31 October 2015 5
model. The influence of these conditions on the hydraulic gradient appear to be transient and of
small magnitude. The development of the distribution for hydraulic gradient did not consider
climate change.

The hydraulic gradient (i) is modeled as normal distribution with a mean of 6.9 × 10-4 and a
standard deviation of 1.27 × 10-4. The influence of the range of the gradient given by the
distribution can be evaluated by calculating a range of groundwater velocity derived from the
gradient using Darcy’s law. The saturated hydraulic conductivity (Ks) is modeled as a normal
distribution with a mean of 9.6 × 10-4 cm/s and a standard error of 9.67 × 10-5 cm/s. Porosity (ϕ)
is modeled as a normal distribution with a mean of 0.29 and a standard deviation of 0.05. From
Darcy’s law the groundwater flux (J) is:
𝐽= 𝐾𝑠 𝑖
and the groundwater velocity (v) is:
𝑣=𝐽/𝜙
where ϕ is the porosity.
The range of groundwater velocity is estimated by choosing values from each distribution
corresponding to the mean ± 3 times the standard error and calculating values of v from the
equations above. Maximum and minimum values for groundwater velocity derived from the
hydraulic gradient distribution range from 4.2 times the mean to 1/5th of the mean.
The significance of uncertainty in the value of the hydraulic gradient was evaluated for the Clive
DU PA model through a sensitivity analysis. The sensitivity analysis identifies which variables
have distributions that exert the greatest influence on the response. The response evaluated in the
sensitivity analysis for the PA model was dose. The results showed that hydraulic gradient was
quantitatively determined to not be a sensitive parameter.
4.0 Groundwater Transport Parameter Distributions
Calculations in the PA Model that are needed for estimating transport in the shallow saturated
zone include the cross-sectional area normal to the flow direction (thickness times width),
definitions of the material SatZone_Medium (hydraulic conductivity, porosity, and bulk density
of Unit 2), the Darcy velocity (a function of gradient and hydraulic conductivity) and
radioelement-specific solid/water partition coefficients (Kds). The distributions for bulk density
and porosity have been described previously in Section 3.2 and the hydraulic gradient in Section
3.3. Aquifer dimensions are described in Section 4.1. Since the flow through the saturated zone
is modeled as a horizontal column of discrete GoldSim Cell pathway elements, dispersivity is not
explicitly defined as it would be for an analytical solution such as a plume. This is discussed in
Section 4.2. The distributions for Kds are described in the Geochemical Modeling white paper
accompanying the Clive DU PA Model.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 6
4.1 Saturated Zone Dimensions
The location and extent of the saturated zone modeling domain including the location of the DU
waste, the point of compliance monitoring well, the buffer zone of the DU cell, and outer
boundaries of property owned and controlled by EnergySolutions are shown in Figure 1.

Figure 1. Location and extent of the saturated zone modeling domain including location of
the DU waste, the point of compliance monitoring well, the buffer zone of the DU
cell, and outer boundaries of property owned and controlled by
EnergySolutions.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 7
Both the unsaturated (vadose) and saturated zones are represented in the Clive DU PA Model as
GoldSim Cell pathway elements. A Cell pathway is mathematically equivalent to a continuously-
stirred tank reactor (CSTR), in which the contents are instantaneously and uniformly mixed
throughout the volume. The representation of the saturated zone in the Model consists of a series
of linked cells. The mass and rate of water flowing through the column of cells depends on the
Darcy velocity and the cross-sectional area perpendicular to the flow direction. This area is
simply the (stochastic) thickness of the aquifer times its width, which is dependent on the
geometry of the embankment. The transport of contaminants in water through the vadose zone
and into the saturated zone is modeled as advective mass flux links from the unsaturated zone
vertical column into the various cells underlying the embankment. This contaminated recharge is
distributed along the saturated zone flow pathway, with a fraction entering each saturated zone
cell. The cell pathways and their interconnections are represented schematically in Figure 2.
Note that there are no wastes located under the side slopes in the Clive DU PA Model.
The advective mass flux in a cell pathway is calculated as the concentration of the contaminant
in water multiplied by the rate at which the water is flowing:

Figure 2: Schematic representation of unsaturated zone and shallow aquifer transport
using cell pathways; section parallel to groundwater flow direction.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 8
An assumption of the mixing cell approach is that all contaminant mass that enters the cell is
completely mixed and equilibrated among all media in the cell, consistent with the mathematical
representation of a CSTR. To provide contaminant mass balance, GoldSim requires information
specifying the volume of the cells. For the Clive DU PA Model, the extent of the saturated zone
below the Federal Cell and the distance from the toe of the waste in the Federal Cell to the
compliance point are represented as a horizontal network of linked GoldSim cell pathway
elements (Figure 2). GoldSim requires the specification of the length of each cell in the direction
of flow and the cross-sectional area of the cell. The length of each cell is the transport distance
divided by the number of cells. The choice of the number of cells used is based on standard
modeling practice, with more discussion provided in Section 4.2. The cross sectional area is the
product of the cell width and height. For the Clive DU PA Model, the cell width is set to the
width of the Federal Cell perpendicular to the direction of flow (“length overall” in Figure 3 of
the Embankment Modeling white paper accompanying the Model). The height of the cell
corresponds to the aquifer thickness.
Aquifer thickness in the subsurface below the Federal Cell was estimated considering water table
elevations, mapped stratigraphy, and interpretations described in Envirocare (2000, 2004). Water
table maps provided in Envirocare (2000, 2004) indicate that the flow in the shallow aquifer in
the vicinity of the Federal Cell is generally to the north. This northerly flow direction is
representative of the current conditions reflecting the effects of mounding due to surface water
infiltration. The natural gradient is approximately to the northeast. Given the predominant flow
direction, wells GW-19B, GW-27D, GW-25, and GW-1 were selected as locations providing the
best available borehole logs for estimating the elevation of the bottom of the aquifer. Well
construction details are provided in Table 3.
Table 3. Construction details for selected wells used for estimating the elevation of the
bottom of the shallow aquifer.
Well Number
State Plane Coordinates
(NAD 83) Surface
Elevation
(ft)
Well
Depth
(ft bgs)
Date
Drilled Easting (ft) Northing (ft)
GW-19B 1189865 7420999 4269 102 02/06/91
GW-27D 1190080 7423071 4270 100 12/28/98
GW-25 1191693 7423029 4274 34 12/19/91
GW-1 1191843 7420942 4273 42 03/03/88

Since the shallow aquifer is described as unconfined, the elevation of the top of the aquifer is
determined by the water table elevation. At three of the locations, nearby wells with shallow
screened intervals were used to obtain representative values for the shallow water table elevation.
Well construction details for the wells used for measurement of water level elevations are
provided in Table 4. Well GW-19A is located 8 ft from well GW-19B, well GW-27 is located
45.6 ft from well GW-27D, and well GW-60 is located 37.6 ft from well GW-1. Given the
average hydraulic gradient of 6.94 × 10-4, the maximum error in water table elevation due to
distance between the wells will be 0.03 ft. This error was considered small enough to be
neglected in the estimate of aquifer thickness.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 9
Table 4. Construction details for selected wells used for water table elevations.
Well
Number
State Plane
Coordinates (NAD 83) Screened
Interval
(ft bgs)
Well Depth
(ft bgs) Date Drilled
Easting (ft) Northing (ft)
GW-19A 1189866 7421007 18 – 27.5 31.5 02/07/91
GW-27 1190121 7423091 20 – 29.5 32 12/11/91
GW-25 1191693 7423029 24 – 33.5 34 12/19/91
GW-60 1191832 7420906 22.5 - 27 28 02/02/93

A map of the shallow aquifer showing fresh water equivalent head surface elevation contours
was prepared by Envirocare (2004) using groundwater elevation measurements from February,
2004. These elevations are used for this analysis to provide continuity with past work describing
the shallow aquifer. The fresh water elevations for the four wells were taken from Table 4 of
Envirocare (2004) and are listed below in Table 5.

Table 5. Water table elevations, aquifer bottom elevations and estimated saturated
thickness of the shallow aquifer.
Well Number Water Table
Elevation (ft)*
Bottom Elevation of
Shallow Aquifer (ft)
Saturated
Thickness (ft)
GW-19B 4251 4229 22
GW-27D 4250 4238 12
GW-25 4250 4240 10
GW-1 4251 4231 20
*GW-19B, GW-27D, and GW-1 water table elevations estimated from the elevation in nearby shallow aquifer wells.

The bottom elevations of the shallow aquifer at wells GW-19B and GW-27D were estimated
from hydrologic cross-sections described in Envirocare (2000, 2004). A south to north cross-
section on the west side of the Federal Cell is shown in Error! Reference source not found.. At
well GW-19B the elevation of the bottom of the aquifer is estimated to be where the silty sand
interval grades into a clay interval. The borehole log for this well indicates that this transition
occurs at an elevation of 4,229 ft.
The lower boundary is extended to the top of an extensive clay layer mapped in well GW-27D
shown in Error! Reference source not found.. The borehole log for this well indicates that the
top of the clay layer occurs at an elevation of 4,238 ft.

Saturated Zone Modeling for the Clive DU PA
31 October 2015 10

Figure 3. Cross-section D-D' modified from Envirocare (2004) showing estimated elevation
of the bottom of the shallow aquifer.

Saturated Zone Modeling for the Clive DU PA
31 October 2015 11
Well GW-25 is 40 ft deep and screened in the bottom 10 ft of the well in a unit described as silty
clay. The elevation of the bottom of the well is 4,240 ft amsl. The saturated hydraulic
conductivity measured in this well is reported by Envirocare (2004) as 1.05 × 10-3 cm/s.
Comparing this result with a site-wide mean value of saturated hydraulic conductivity of
9.6 × 10-4 cm/s indicates that this well is completed within the shallow aquifer. The elevation of
the bottom of the aquifer at this well may be deeper than the bottom of the well but is
conservatively taken as 4,240 ft, the elevation of the bottom of the well.
Well GW-1 is 41.5 ft deep and is screened from 20 ft bgs to 40 ft bgs. The driller's log describes
the sediments as a silty sand from 14 ft to 29 ft depth and sandy clay from 29 ft to the bottom of
the borehole at 41.5 ft. Well GW-60 located 37.6 ft from well GW-1 is completed to a depth of
28 ft in sediments described as a silty clay. The interval from 22.5 ft bgs to 27 ft bgs within the
silty clay is screened. Saturated hydraulic conductivity in well GW-60 was determined to be
3.4 × 10-3 cm/s or three times the site-wide average. This relatively high value of saturated
hydraulic conductivity measured in a silty clay indicates the shallow aquifer extends at least as
deep as the bottom of well GW-1. Given this interpretation, the elevation of the bottom of the
aquifer at this borehole is estimated to be 4,231 ft. The estimated elevations of the bottom of the
shallow aquifer and the resulting saturated thicknesses are listed in Table 5.
A distribution for the thickness of the saturated zone was established based on four location
measurements (GW-19B, GW-27D, GW-25, and GW-1), and professional judgment regarding
the accuracy of the measurements. An aquifer thickness for each of the four locations was
calculated as the difference between the recorded elevation of the water table and the elevation
of the bottom of the shallow aquifer. Since the four locations do not quite form a square,
triangulation was used to calculate an average thickness across the region. Only two possible
triangulations exist for these four points, so both were computed, and the average of the two was
used as the mean of the distribution for saturated zone thickness. Professional judgment was that
the measurements are accurate to within 1 foot. Thus, 1 foot was interpreted as a two standard
deviation range, giving a measurement standard deviation of 0.5 ft. Since four measurements are
being averaged (with nearly equal weights), the resulting standard error for the mean is then
0.5 ft divided by the square root of 4. The resulting distribution for the mean thickness of the
saturated zone was thus chosen as a normal distribution with mean equal to 16.2 ft with a
standard deviation of 0.25 ft.
4.2 Dispersion
The process of spreading of a contaminant in groundwater that occurs in addition to movement
by advective flow is represented in mathematical models by the dispersion coefficient. The
dispersion coefficient represents both the mechanical (hydrodynamic) and chemical components
of mixing and is written as:
𝐷𝑙= 𝛼𝑙 𝑣̅ + 𝐷𝑚 (4)
where
Dl = longitudinal dispersion coefficient,
αl = longitudinal dispersivity,
𝑣̅ = mean pore water velocity, and
Dm = molecular diffusion coefficient.
Saturated Zone Modeling for the Clive DU PA
31 October 2015 12
Only longitudinal dispersion is considered for this discussion because of the geometry of the
transport pathway. The width of the disposed waste is the dimension perpendicular to the
groundwater flow direction. This distance is 1,317.8 ft (“length overall” in Figure 3 of the
Embankment Modeling white paper). The distance from the edge of the waste to the compliance
point is 90 ft as required by the groundwater discharge permit. The entire horizontal length of the
saturated zone cells is this 90 ft plus the footprint of the embankment parallel to the direction of
water flow (1775.0 ft, the “width overall” in Figure 3 of the Embankment Modeling white paper),
making a total length of 1865 ft. With this geometry, the width of the source is more than 5 times
the distance from the edge of the source to the point of compliance, making transverse dispersion
insignificant.
In a numerical model such as the Clive DU PA Model, the discretization of the flow path into
cells results in an effective (numerical) longitudinal dispersion (parallel to the flow direction)
due to the full mixing of a CSTR even with no additional dispersivity defined. Because of this
inherent numerical dispersion, no additional dispersion coefficient is included in the saturated
zone transport calculations in the Clive DU PA Model.
Dispersion is discussed in the User’s Guide for the GoldSim Contaminant Transport Module
(GoldSim 2010) in the context of the GoldSim Aquifer pathway element. The Aquifer element is
a collection of linked cell pathway elements, and the saturated zone in the Clive DU PA Model is
also represented as a collection (column) of cell elements, which is somewhat more flexible than
the predefined GoldSim Aquifer element. Longitudinal dispersivity is commonly approximated
as 0.1 times the length of the transport path (GoldSim 2010). For the Clive DU PA Model the
point of compliance is a fixed location 232 ft from the edge of the DU waste, since the length
travelled under the side slope of the embankment, which contains no DU waste (142 ft), is added
to the standard 90 ft. The estimated value of the dispersivity would then be 232 ft /10 = 23 ft. In
order to reduce unwanted numerical dispersion, GoldSim (2010) recommends that the number of
cell elements used in the column be greater than the transport path distance divided by twice the
dispersivity. For the Clive DU PA Model geometry, the number of cells should therefore be
greater than 232 ft / (2×23 ft) = 5. The horizontal column of Cell elements that represents the
saturated zone to the well in the Clive DU PA Model contains 20 cells and there are 2 cells under
the side slope. The number of cells making up the transport path exceeds the minimum
recommended.
The mass balance of water flow is not in question, since it is up to the GoldSim programmer (the
model author) to assure that all flows are properly accounted for. GoldSim performs no solutions
whatsoever to the hydraulics of the model. In the case of the saturated zone, the water flow
through the horizontal column is defined as a constant value all the way through the column.
Since there are no numerical calculations in GoldSim with respect to water flow calculations,
mass balance of water has no mass balance error.
The mass balance of contaminants (radionuclides) is determined internally by the GoldSim
software as part of its proprietary solution algorithms. The internal solver accounts for advective
flows, diffusion in air and water (where applicable), partitioning between air, water, and solid
phases, as well as radioactive decay and ingrowth. The modeler and the user are not privy to the
internal mass balance calculations, but a good indication of how well the model is performing
can be had by experimenting with the settings for solution precision, which are accessible to the
Saturated Zone Modeling for the Clive DU PA
31 October 2015 13
user. Using the GoldSim interface, go to Model | Options dialog, and select the Contaminant
Transport tab. Under the first set of options, General Options, there is a drop-down box where
the user can set the solution precision, in qualitative terms: low, medium, and high. If choosing a
higher solution precision does not result in substantially different results, then the user has an
indication that the mass balance is acceptable, since refining the precision does not improve the
calculation.

Saturated Zone Modeling for the Clive DU PA
31 October 2015 14
5.0 References
Bingham Environmental. 1991. Hydrogeologic report Envirocare Waste Disposal Facility South
Clive, Utah. October 9, 1991. Prepared for Envirocare of Utah. Salt Lake City, UT.
Bingham Environmental. 1994. Hydrogeologic report Mixed Waste Disposal Area Envirocare
Waste Disposal Facility South Clive, Utah. November 18, 1994. Prepared for Envirocare of
Utah. Salt Lake City, UT.
Brockwell, P. J. and Davis, R. A. 1996. Introduction to Time Series and Forecasting. Springer,
New York.
Domenico, P.A. and F.W. Schwartz. 1990. Physical and chemical hydrology. New York: John
Wiley and Sons.
Envirocare of Utah, Inc. 2000. Revised hydrogeologic report for the Envirocare Waste Disposal
Facility Clive, Utah. Version 1.0. Envirocare of Utah, Inc. Salt Lake City, UT.
Envirocare of Utah, Inc. 2004. Revised hydrogeologic report for the Envirocare Waste Disposal
Facility Clive, Utah. Version 2.0. Envirocare of Utah, Inc. Salt Lake City, UT.
Utah Division of Water Rights, (DWR), water rights and well log database at
http://waterrights.utah.gov/wrinfo/query.asp. Accessed March 18, 2014.
Whetstone Associates. 2000, Revised Envirocare of Utah Western LARW Cell Infiltration and
Transport Modeling. July 19, 2000.
Whetstone Associates. 2007. EnergySolutions Class A South Cell Infiltration and Transport
Modeling. December 7, 2007.

 NAC-0021_R2

Atmospheric Transport Modeling
for the Clive DU PA

Clive DU PA Model v1.4
5 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 ii
1. Title: Atmospheric Transport Modeling for the Clive DU PA
2. Filename: Atmospheric Modeling v1.4.docx
3. Description: This white paper provides documentation of the development of parameter values and distributions
used for atmospheric modeling for the Clive DU PA Model.

Name Date
4. Originator Dan Levitt 5 November 2015
5. Reviewer Paul Black 5 November 2015
6. Remarks

Nov 5, 2015: D.Levitt: Updated from v1.2 to v1.4. Updated R313 information.

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 iii

This page is intentionally blank, aside from this statement.

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

1.0
 Summary of PA Model Inputs ................................................................................................ 1

2.0
 Introduction ............................................................................................................................ 1

3.0
 Overview and Framework ...................................................................................................... 2

4.0
 Model Descriptions ................................................................................................................ 4

4.1
 Cowherd Particle Resuspension Model ............................................................................ 4

4.2
 AERMOD ......................................................................................................................... 6

4.3
 CAP-88 ............................................................................................................................. 6

5.0
 Meteorological and Terrain Elevation Data ........................................................................... 7

6.0
 Implementation of Resuspension and Dispersion Models ..................................................... 7

6.1
 Spatial Attributes of Air Dispersion Modeling ................................................................. 9

6.2
 AERMOD Results for Air Concentrations and Off-Site Deposition ................................ 9

6.2.1
 AERMOD Simulated Air Concentrations and Chi/Q Values ................................... 10

6.2.2
 AERMOD Off-Site Particulate Deposition ............................................................... 15

6.3
 Confirmation of AERMOD Results with CAP-88 ......................................................... 16

6.4
 Implementation of Cowherd Unlimited-Reservoir Resuspension Model ....................... 18

7.0
 Electronic Reference ............................................................................................................ 19

8.0
 References ............................................................................................................................ 19

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 v
FIGURES
Figure 1. Wind Rose for Clive, Utah (courtesy of Meteorological Solutions, Inc.) ....................... 8

Figure 2. Off-site air dispersion locations (Note: red line is the rail; green line is UTTR
access road). ................................................................................................................ 10

Figure 3. Off-site air dispersion area (approximate dimensions of largest receptor exposure
area shown as dashed green line). ............................................................................... 13

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 vi
TABLES
Table 1. Summary input parameter values and distributions .......................................................... 1

Table 2. Allocation of particle mass in particle size fraction bins for PM10 emissions. .............. 11

Table 3. Air concentration estimates (ug/m3 of PM10) by location and particle diameter
fraction; 0.25 g/s emission rate. ................................................................................... 12

Table 4. Receptor-specific χ/Q ratios for PM10 particulates. ....................................................... 12

Table 5. Radon air concentrations (0.25 g/s emissions) and χ/Q ratios for each receptor
location. ....................................................................................................................... 14

Table 6. Total deposition of PM10 particulate matter on the disposal embankment. ................... 16

Table 7. Comparison of CAP-88 and AERMOD particle deposition results (g/m2-yr). .............. 17

Table 8. Range of input parameter values for particle resuspension modeling. ............................ 18

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 1
1.0 Summary of PA Model Inputs
A summary of parameter values and distributions employed in the atmospheric modeling
component of the Clive Performance Assessment (PA) model is provided here. Additional
information on the derivation and basis for these inputs is provided in subsequent sections of this
report. With the exception of particulate resuspension flux, the PA model inputs related to
atmospheric modeling are derived from AERMOD air dispersion modeling results. The term
Chi/Q refers to the ratio of breathing-zone air concentration (Chi) to the emission rate (Q) used
in the AERMOD simulations. The term PM10 refers to particulates with a mean aerodynamic
diameter of 10 µm and less, the size fraction employed in regulatory air modeling to represent
respirable particles.
Table 1. Summary input parameter values and distributions
PA Model Parameter Units Value Notes
Chi / Q ratios for PM10 µg/m3 per g/s See Table 4 Based on AERMOD
modeling; see Section 10.
Chi / Q ratios for gases µg/m3 per g/s See
Table 5
Based on AERMOD
modeling; see Section 10.
Embankment PM10
redeposition
g/m2-yr per g/yr See Table 6 Based on AERMOD
modeling; see Section 15.
Resuspension flux of PM10 kg/m2-yr LogUniform (
2.5e-7, 0.3 )
Implementation of Cowherd
et al (1985); see Section 17.
Fraction PM10 deposition
in off-site exposure area
— See Table 6 Based on AERMOD
modeling; see Section 15.
2.0 Introduction
The safe storage and disposal of depleted uranium (DU) waste is essential for mitigating releases
of radioactive materials and reducing exposures to humans and the environment. Currently, a
radioactive waste facility located in Clive, Utah (the “Clive facility”) operated by the company
EnergySolutions Inc. is being considered to receive and store DU waste that has been declared
surplus from radiological facilities across the nation. The Clive facility has been tasked with
disposing of the DU waste in a manner that protects humans from future radiological releases.
To assess whether the proposed Clive facility location and containment technologies are suitable
for protection of human health, specific performance objectives for land disposal of radioactive
waste set forth in Utah Administrative Code (UAC) Rule R313-25 License Requirements for
Land Disposal of Radioactive Waste - General Provisions must be met—specifically R313-25-9
Technical Analyses (Utah 2015). In order to support the required radiological performance
assessment (PA), a probabilistic computer model has been developed to evaluate the doses to
human receptors that would result from the disposal of radioactive waste, and conversely to
determine how much waste can be safely disposed at the Clive facility. The GoldSim systems
analysis software (GTG 2011) was used to construct the probabilistic PA model.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 2
The site conditions, chemical and radiological characteristics of the wastes, contaminant
transport pathways, and potential human receptors and exposure routes at the Clive facility that
are used to structure the quantitative PA model are described in the conceptual site model
documented in Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
(Clive DU PA CSM.pdf). Based on current and reasonably anticipated future land uses, the two
future use exposure scenarios described in the CSM for evaluation in the PA are ranching and
recreation.
The Neptune and Company, Inc. (Neptune) white paper Dose Assessment for the Clive PA (Dose
Assessment.pdf) details the assumptions and computational methods for estimating radiation
doses to future human receptors associated with DU and its decay products. This present white
paper focuses on one aspect of the exposure and radiation dose calculations; atmospheric
modeling to support the calculation of breathing zone air concentrations of radionuclides for
future human receptors. Specifically, this paper addresses the modeling of:
1. Rates of particle resuspension by aeolian (wind derived) processes;
2. Air concentrations of radionuclides above the disposal embankment and at specific
locations of potential off-site exposure; and,
3. Deposition flux of resuspended embankment particles at locations beyond the
embankment.
Particle resuspension related to mechanical disturbances from off-highway vehicle (OHV) use is
also addressed in the PA model and is discussed in the white paper Dose Assessment for the Clive
PA.
3.0 Overview and Framework
Atmospheric dispersion modeling was conducted using computer software outside of the
GoldSim modeling environment, as the GoldSim PA model is a system-level model. An
atmospheric dispersion model is a mathematical model that employs meteorological and terrain
elevation data, in conjunction with information on the release of contamination from a source, to
calculate breathing-zone air concentrations at locations above or downwind of the release. Some
models may also be used to calculate surface deposition rates of contamination at locations
downwind of the release.
Air dispersion models, including the AERMOD (EPA, 2011a) and CAP-88 (EPA, 2011b) models
used in this exercise, commonly assume a Gaussian distribution for estimating vertical and
horizontal dispersion of contamination away from the source. Factors affecting the amount of
dispersion include atmospheric turbulence, the height of the release (e.g., a virtual stack versus
ground level), the buoyancy of the plume, and terrain features. Although they employ different
mathematical models for assessing horizontal and vertical dispersion, both AERMOD and
CAP-88 ultimately calculate annual-average contaminant breathing zone air concentrations at
various distances and in various directions from a source release.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 3
The Clive facility waste disposal embankment will be a large-area emissions source with a gently
sloping surface that will be raised approximately 15 m above the surrounding terrain. There are
two types of future radioactive emissions associated with the embankment:
1. Particulate emissions of contaminated surface soil due to aeolian erosion; and,
2. Emissions of gas-phase radionuclides diffusing across the surface of the embankment
into the atmosphere.
With respect to potential human receptors exposed upon the embankment itself (ranchers and
recreationalists, including hunters, and OHV sport riders—see the Dose Assessment white
paper), the surface of the embankment represents a ground-level (0-m height) emissions source.
For estimating the annual dose to these individuals, the air modeling endpoint of interest is the
annual-average breathing-zone concentration of respirable particles or gaseous radionuclides
above the embankment. For individuals exposed at locations other than the embankment, the
embankment represents a 15-m elevation emissions source, as transport by wind will be
necessary for exposure at these locations. A second air modeling endpoint of interest for these
“off-site” receptors is the same as for the “on-site” receptors; i.e., the annual-average breathing-
zone concentration of respirable particles or gaseous radionuclides released from the
embankment at some specific off-site location.
A third endpoint of interest is the off-site deposition rate of embankment particulates. As
particulates eroding from the embankment are deposited on surrounding land, this surrounding
area may become a secondary source of radionuclide exposure for ranchers and recreationists.
The relative importance of exposure on-site and off-site depends in part on the fraction of total
exposure time a rancher or recreationist spends in each area. However, the importance of on-site
vs. off-site exposure also depends on the rate of aeolian particle erosion from the embankment
and the rate at which contamination from the disposed waste is transported to the the surface of
the embankment by processes such as biotic transport (see Biological Modeling white paper) and
radon diffusion. If transport rates of radioactivity are much higher than the rate at which aeolian
particle erosion removes radioactivity, then embankment surface soil radionuclide concentrations
will steadily increase over time relative to off-site levels. However, if aeolian particle erosion
rates are greater than the transport/accumulation rate of radioactivity in surface soil, then
embankment soil radioactivity will be minimal throughout the modeling period. Because only a
portion of wind-eroded particles remain within the overall receptor exposure area, and because
receptor exposure intensity varies between the embankment and the off-site exposure area, this
can have significant consequences for dose assessment results.
In summary, there are three air modeling endpoints:
1. Annual-average breathing-zone concentration of respirable particles and gaseous
radionuclides above the embankment;
2. Annual-average breathing-zone concentration of respirable particles and gaseous
radionuclides at specific off-site locations; and,
3. Off-site aeolian deposition rate of embankment particulates.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 4
For gas-phase radionuclides, the contaminant transport component of the GoldSim PA model
(see Unsaturated Zone Modeling white paper) provides the diffusive flux (activity per area per
time, as in Bq/m2·s) at the surface of the disposal embankment. A particulate resuspension
model, described below, is employed to calculate the particle flux from the surface of the
disposal embankment. The gas-phase radionuclide and particle fluxes are the site-specific inputs
to the air dispersion model. The third endpoint, the off-site deposition rate of embankment
particulates, is used as an input for modeling radionuclide soil concentrations over time in the
off-site exposure area for ranchers and recreationists.
AERMOD, a United States Environmental Protection Agency (EPA)-recommended regulatory
air modeling system that incorporates state-of-the-art modeling approaches (EPA, 2011a), is used
for the air dispersion modeling to address the three endpoints. As a quality assurance measure, a
second EPA regulatory air dispersion model (CAP-88; EPA, 2011b) is employed to confirm the
AERMOD results (see Section 16).
4.0 Model Descriptions
The following subsections provide a summary of the particle resuspension and air dispersion
models used to support the modeling endpoints described above.
4.1 Cowherd Particle Resuspension Model
Air dispersion models for estimating radionuclide concentrations above, or at some distance
from, a release source require a radionuclide emission rate as an input. In the case of aeolian soil
particulates in ambient air (e.g., dust), an area-averaged particulate resuspension rate is needed.
For screening of potential inhalation risks at contaminated soil sites, EPA recommends a
particulate emission factor (PEF) model to estimate annual average concentrations of respirable
particulates (approximately 10 µm and less; i.e., PM10) in ambient air above contaminated soil
(EPA, 1996; EPA, 2002).
The PEF incorporates PM10 emission models (Cowherd et al, 1985) related to wind erosion
under one of two conditions. The particulate emission model for PM10 used in EPA (1996; 2002)
pertains to a surface with unlimited erosion potential. Cowherd et al (1985) also provide a model
for estimating PM10 particle emissions from surfaces with a limited reservoir of erodible
particles. The decision criterion in choosing between these model types is provided in Figure 3-2
of Cowherd et al. (1985) as, “Is threshold friction velocity > 75 cm/s?” For surfaces not covered
by continuous vegetation, including assumed future states of the disposal embankment (see
Biological Modeling white paper), surfaces with a threshold friction velocity larger than 75 cm/s
tend to be composed of elements too large to be eroded, or of erosion-resistant crusts. An
erosion-resistant crust might be of cryptogamic nature (particles bound by a biological
community consisting of one or more types of cyanobacteria, lichens, mosses, and fungi), or
simply by aggregation of very fine silty-clay particles.
Methods for characterizing threshold friction velocity in Cowherd et al. (1985) rely on site
inspection, which is problematic for this modeling because the future surface characteristics of
the embankment are uncertain. The foreseeable future state of the cap surface likely includes a
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 5
range of particle sizes due to contributions from windblown loess, from decaying plant material,
and from degrading rip rap. A practical constraint on the use of the limited-reservoir model of
soil erosion is that this model is dependent upon the frequency of disturbance of the surface.
When a surface has limited erosion potential, disturbances to expose fresh surface material are
considered necessary to restore erodibility. For the Clive PA model, a range of input parameter
values are used with the unlimited-reservoir model to estimate possible PM10 emission rates
based on the presumption of dynamic steady-state conditions, where PM10 emissions are
presumed to be balanced by deposition of particles from upwind locations.
The equation for particle emissions from a surface with unlimited erosion potential, originally
published as Equation 4-4 in Cowherd et al. (1985), has the form:
E10 = 0.036 × (1 - V) × ([u] / ut-7)3 × F(x) (1)
where:
E10 is the annual-average PM10 emission rate per unit area of contaminated soil
(g/m2·hr);
V is the fraction of vegetative cover (-);
[u] is the mean annual wind speed (m/s);
ut-7 is the threshold value of wind speed at 7 m (m/s); and,
F(x) is a function dependent on the ratio u / ut (-).
and, from Equation 4-3 in Cowherd et al. (1985):
ut-7 = (ut × Fadj / 0.4) × ln(700 cm / z0) (2)
where:
ut is the unadjusted threshold friction velocity (m/s);
Fadj is the threshold friction velocity adjustment factor; and,
z0 is the surface roughness height (cm).
Values of F(x) are estimated based on the function shown graphically in Figure 4-3 of Cowherd
et al. (1985). The value of x is calculated as defined in Equation 4-4 of Cowherd et al. (1985):
x = 0.886 × (ut-7 / [u]) (3)
and the function F(x) is approximated using the following equations:
when x < 1, F(x) = (6 – x3)/π
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 6
when x ≥ 1 and < 2, F(x) = (-1.3 × x) + 2.89
when x ≥ 2, F(x) = [(8 × x3) + (12 × x)] × e-(x^2).
With the exception of the case where x ≥ 2, these equations were fit by Neptune and Company
based on visual approximation to the graphic in Figure 4-3 of Cowherd et al. (1985). For the case
x ≥2, the equation is taken from Appendix B of Cowherd et al (1985).
4.2 AERMOD
AERMOD is EPA's recommended regulatory air modeling system for steady-state emissions. It
is defined by EPA (2011a) as
“A steady-state plume model that incorporates air dispersion based on planetary boundary
layer turbulence structure and scaling concepts, including treatment of both surface and
elevated sources, and both simple and complex terrain.”
AERMOD supports source characterization as an area of user-defined dimensions and elevation
and is thus suitable for modeling the disposal embankment. AERMOD employs two pre-
processors related to handling of meteorological data and terrain data. The AERMET pre-
processor is used to estimate boundary layer parameter values such as mixing height and friction
velocity needed for the air dispersion modeling. AERMET inputs include albedo (a measure of
the reflectivity of the ground surface), surface roughness, and Bowen ratio (a measure of heat
flux to the atmosphere), plus meteorological measurements such as wind speed and direction,
temperature, and cloud cover. The AERMAP pre-processor uses gridded terrain elevation data to
generate receptor grids for the air dispersion modeling.
In the stable boundary layer nearest the earth's surface, AERMOD assumes a Gaussian
concentration distribution on the vertical and horizontal axes. In the convective boundary layer
above, the horizontal distribution is also assumed to be Gaussian, but the vertical distribution is
described using a linear combination of two separate Gaussian functions. In this manner
AERMOD addresses heterogeneity in the planetary boundary layer where wind and associated
mixing is influenced by friction with the earth's surface.
4.3 CAP-88
The Clean Air Assessment Package – 1988 (CAP-88) modeling program (EPA, 2011b) is
recommended for demonstrating regulatory compliance with the requirements of Subpart H of 40
CFR Part 61 (NESHAPS; National Emission Standards for Emissions of Radionuclides Other
Than Radon from Department of Energy Facilities). As described in 10 CFR 40 Part 61.93, 12-
15-1989:
“To determine compliance with the standard, radionuclide emissions shall be
determined and effective dose equivalent values to members of the public
calculated using EPA approved sampling procedures, computer models CAP–88
or AIRDOS-PC, or other procedures for which EPA has granted prior approval.”
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 7
CAP-88 employs a modified Gaussian plume dispersion model to compute ground-level
radionuclide air concentrations for a circular grid around an emission source. Meteorological
data must be processed into STability ARray (STAR) files for CAP-88, which include
assignments of atmospheric turbulence into one of six stability classes labeled A through F.
5.0 Meteorological and Terrain Elevation Data
Raw meteorological data from the the EnergySolutions monitoring station at Clive, Utah were
collected (MSI, 2010). The monitoring station is at 1,306 m above sea level, and is equipped to
measure horizontal wind speed, wind direction, 2-and 10-meter temperature, delta-temperature
for the derivation of atmospheric stability class, solar radiation, precipitation, and evaporation
(MSI, 2010).
Meteorological Solutions Inc. (MSI), processed the raw meteorological data to create AERMET
files (for AERMOD air dispersion modeling) and STAR files (for CAP-88 air dispersion
modeling). STAR files were created by MSI using two different methods. The sigma-theta
method (STAR-ST) assigns an atmospheric stability class based on the standard deviation of the
horizontal wind direction. A second method (STAR-SR) assigns an atmospheric stability class
based on solar radiation and delta-temperature measurements. The processed meteorological
data were then employed by Neptune for the air dispersion modeling. AERMET, STAR-ST, and
STAR-SR input files for the years 2003, 2004, 2006, 2007, and 2009 were made available to
Neptune by MSI. Composite STAR-SR and STAR-ST files integrating meteorological data for
all five years were also created by MSI and provided to Neptune.
A Clive, Utah wind rose from MSI (2010), showing wind speed and direction for the period
January 2009 through December 2009, is duplicated here as Figure 1. As shown in Figure 4.1 of
MSI (2010), the wind rose integrating data for the period 1993 through 2009 is very similar to
that for 2009 shown here. For example, average annual wind speed for both time periods is 7.2
mph and stability class variability for 1993-2009 and just 2009 is less than 5% (MSI, 2010).
Terrain elevation information for each grid cell was derived from the AERMAP interface within
the AERMOD ViewTM (Version 6.7.1) software package (Lakes Environmental, 2010).
AERMAP accesses digital elevation model (DEM) data from webGIS (http://www.webgis.com),
which is then processed for input into AERMOD. For this project, DEM data from the United
States Geological Survey (USGS) for Tooele County, Utah are employed. These data have a
nominal resolution of 90 m and were interpolated to the uniform Cartesian grid (i.e., the
modeling area) using the inverse distance weighting setting, which is the recommended setting in
AERMAP. The nature of the AERMOD ViewTM interface, and the basis of the spatial receptor
grid, are described in Section 8.
6.0 Implementation of Resuspension and Dispersion Models
Neptune implemented AERMOD within the graphical user interface AERMOD ViewTM (Lakes
Environmental, 2010). This software package provides an interface for using base maps to
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 8
define sources and receptors, importing digital elevation data from USGS, and producing
graphical displays of results.

Figure 1. Wind Rose for Clive, Utah (courtesy of Meteorological Solutions, Inc.)

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 9
6.1 Spatial Attributes of Air Dispersion Modeling
As described in Section 1, the intent of the air dispersion modeling was to estimate air
concentrations of radionuclides above the disposal embankment, and for receptors at specific
locations of potential off-site exposure. These receptors and off-site locations, described in the
Dose Assessment white paper, include:
• Travelers on Interstate-80 which passes 4 km to the north of the site;
• Travelers on the main east-west rail line which passes 2 km to the north of the site;
• The resident caretaker present at the east-bound Grassy Mountain (Aragonite) Interstate-
80 rest area 12 km to the northeast of the site;
• Recreational users of the Knolls OHV area (BLM land that is specifically managed for
OHV recreation) 12 km to the west of the site; and,
• Workers at the Utah Test and Training Range (UTTR, a military facility) to the south of
the Clive facility, who may occasionally drive on an access road immediately to the west
of the EnergySolutions fenceline.
These five locations are shown in Figure 2. A uniform Cartesian grid using 1-km2 resolution grid
cells was employed in the AERMOD air dispersion modeling to support calculation of air
concentrations at the first four locations. This grid was constructed of 299 grid cells (23 grid
cells longitudinally by 14 grid cells latitudinally).
To support the estimation of air concentrations above the disposal embankment and particle
deposition onto the embankment, AERMOD was also run with a smaller 0.3-km2 grid size,
which corresponds to the area of the disposal embankment. In this AERMOD simulation, one
grid cell was centered directly above the 0.3-km2 area emissions source representing the
embankment. The results for this grid cell were also applied to the UTTR access road, which is
in close proximity to the disposal embankment.
6.2 AERMOD Results for Air Concentrations and Off-Site Deposition
Two sets of simulations were conducted using AERMOD; one to estimate air concentrations and
total deposition of particulates, and a second to estimate gas concentrations at the specified
receptor locations in Section 8. The air concentration outputs (particulate and gas) from
AERMOD were then used to calculate χ/Q ratios, which are the ratio of breathing-zone air
concentration (χ) to the emission rate (Q) used in the AERMOD simulations (Section 10). These
χ/Q ratios are then employed in the GoldSim model for each receptor location by multiplying
χ/Q by the gas or particle emission rate generated in the model. Particle deposition rates from
AERMOD were used to calculate the fraction of particulates that are redeposited on the
embankment (Section 15). This off-site deposition fraction was used in conjunction with the
particle emission rate generated in the model to calculate the mass of embankment particles
deposited onto the off-site air dispersion area over time.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 10

Figure 2. Off-site air dispersion locations (Note: red line is the rail; green line is UTTR ac-
cess road).

6.2.1 AERMOD Simulated Air Concentrations and Chi/Q Values
As described in Section 8, AERMOD was run using either a 0.3-km2 or a 1.0-km2 resolution
grid, depending on whether the air concentrations above the embankment or at distant off-site
locations were simulated. A consideration in the air dispersion modeling is the elevation of the
area source. For modeling air concentrations in the breathing zone above an area source, it is
necessary to define a zero meter-elevation release height in AERMOD. For modeling air
concentrations at the locations of distant off-site receptors, however, the disposal embankment is
more accurately represented as an area source with a 15-m release height (where 15 m is the
approximate height of the gently sloping top of the embankment).
An assumed PM10 emission rate of 0.25 g/sec was used for all AERMOD simulations. This value
corresponds to an area flux of approximately 0.025 kg/m2-yr, which is near the upper end of
PM10 emission rates derived using Cowherd et al (1985) (see Section 17). The AERMOD results
are used to develop χ/Q ratios, which in principle are independent of the specific emission rate
used in the simulations. The emission rate input to AERMOD was varied over several orders-of-
magnitude, and it was confirmed that the ratio χ/Q is independent of emission rate.

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 11
The relative mass associated with two particle size fractions within the PM10 category can be
distinguished in AERMOD: 0 to 2.5 micron particle diameter, and 2.5 to 10 micron particle
diameter. The actual particle size distribution of future PM10 emissions from the embankment is
unknown. To explore the influence of particle size fraction on on-site and off-site PM10 air
concentrations, the mass of particles in the two categories for a series of eight simulations was
varied as presented in Table 2:
Table 2. Allocation of particle mass in particle size fraction bins for PM10 emissions.
Simulation 0 to 2.5 microns 2.5 to 10 microns
1 0% 100%
2 5% 95%
3 10% 90%
4 20% 80%
5 40% 60%
6 60% 40%
7 80% 20%
8 100% 0%
Note that these fractions represent fine particle fractions only, and assume that less than 10% of the
particle emissions is composed of dust greater than or equal to 10 microns in diameter.

The AERMOD particulate simulations in Table 2 were conducted using meteorological input
data for year 2009, as previously discussed. Additional simulations were conducted using
meteorological data from 2003, 2004, 2006, and 2007. The differences in modeled air
concentrations among the five data sets was minimal. Uncertainty related to meteorological
conditions is overwhelmingly due to extrapolating current conditions (as represented by any of
these five years) to the 10,000-year performance period, which is not possible to quantify at this
time. Therefore, uncertainty related to the slight differences in AERMOD results based upon the
five data sets has not been propagated in the GoldSim PA model.
As described above, two sets of simulations were conducted at different spatial resolutions (0.3-
km2 and 1.0-km2) for the particle size fractions outlined in Table 2. The outputs from these
simulations are summarized in Table 3.
The AERMOD input emission rate and the air concentration outputs from AERMOD were then
used to construct χ/Q ratios for each receptor, as shown in Table 4. The Q term for this ratio is
0.25 g/sec, as described above. The χ term (µg/m3) is from Table 3. These χ/Q ratios were then
directly imported into the GoldSim PA model. For each model realization, one of the eight
simulations is selected and the associated χ/Q ratios are used in the dose calculations.
Differences among the eight sets of χ/Q ratios represent uncertainty in the particle size
distribution of future PM10 emissions from the embankment.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 12
Table 3. Air concentration estimates (ug/m3 of PM10) by location and particle diameter
fraction; 0.25 g/s emission rate.
Simulation
Knolls
OHV
Area
Grassy Mt.
(Aragonite)
Rest Area
I-80 Railroad Embankment UTTR
Access Road
1 0.011 0.0017 0.065 0.11 56 56
2 0.011 0.0017 0.066 0.11 56 56
3 0.011 0.0017 0.066 0.11 56 56
4 0.011 0.0017 0.067 0.11 56 56
5 0.012 0.0018 0.068 0.11 57 57
6 0.013 0.0018 0.069 0.11 58 58
7 0.014 0.0018 0.070 0.11 59 59
8 0.015 0.0019 0.071 0.11 59 59
Concentration estimates for the Embankment and UTTR Access Road receptors are based on simulations
conducted at 0.3-km2 resolution. All other concentrations correspond to simulations conducted at 1.0-km2
resolution. Values for I-80 and Railroad are the largest values for any grid cell containing these features (i.e. at
points close to the Clive facility). Note that the simulation numbers in this table correspond to the particle diameter
fractions in Table 2.

Table 4. Receptor-specific χ/Q ratios for PM10 particulates.
Simulation
Knolls
OHV
Area
Grassy Mt.
(Aragonite)
Rest Area
I-80 Railroad Embankment UTTR
Access Road
1 0.043 0.0069 0.26 0.43 222 222
2 0.044 0.0069 0.26 0.43 223 223
3 0.044 0.0069 0.26 0.43 224 224
4 0.046 0.0070 0.27 0.43 225 225
5 0.049 0.0071 0.27 0.43 228 228
6 0.052 0.0072 0.28 0.44 231 231
7 0.055 0.0073 0.28 0.44 234 234
8 0.058 0.0074 0.28 0.44 238 238
χ/Q ratios for the Embankment and UTTR Access Road receptors are based on simulations conducted at 0.3-km2
resolution. All other off-site receptors correspond to simulations conducted at 1.0-km2 resolution. Values for I-80
and Railroad are the largest values for any grid cell containing these features. Note that the simulation numbers in
this table correspond to the particle diameter fractions in Table 2.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 13
As described in Section 2, air concentrations of gases in the off-site air dispersion area are based
on air dispersion of gas emissions from the cap. The size and basis of the off-site air dispersion
area (see

Figure 3) is discussed in the Dose Assessment white paper, and is that area surrounding the
embankment in which ranchers and recreationists may be exposed to contaminants originating
from the embankment.
Radon-222 is the only gas-phase radionuclide evaluated in the Clive PA model. Breathing zone
concentrations of radon-222 in the off-site air dispersion area are based on releases from the cap,
rather than evolution from any radium-226 deposited with particulates in dispersion area surface
soil, because the former will be by far the more significant source. Radon transport in the
embankment is discussed in the Unsaturated Zone Modeling for the Clive PA white paper.
Radon-222 air concentrations in the off-site air dispersion area have been calculated based on the
smallest potential size of this area (16,000 acres, or approximately 65 km2). The gas
concentration in air for this area was calculated as the arithmetic average of the gas
concentrations in the 65 AERMOD 1-km grid areas with the highest concentrations.
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 14

Figure 3. Off-site air dispersion area (approximate dimensions of largest receptor exposure
area shown as dashed green line).

Radon-222 air concentrations were estimated using the gas deposition module in AERMOD for
the embankment and the 5 other receptor locations described in Section 8. Similar to estimating
air concentrations for PM10 dust, these simulations were conducted with a 0-m elevation source
for a 0.3-km2 grid size (over the embankment and for the adjacent UTTR access road) and a 15-
m elevation source with a 1.0-km2 grid size (all other receptor locations). However, only one
simulation each was conducted for radon gas dispersion because uncertainty related to particle
size fraction is inapplicable to gases.
The input parameters required by AERMOD include diffusivity of the modeled gas in air and
water, cuticular resistance, and Henry's Law constant. For radon diffusivity in air, a value of
0.11 cm2/sec was assumed (Rogers and Nielson, 1991; Nielson and Sandquist, 2011). For radon
diffusivity in water, a value of 100,000 cm2/sec was assumed (Volkovitsky, 2004), while Henry's
Law constant was assumed to be 0.0093 mol/kg-bar (NIST, 2011). The landcover properties were
assigned the default values from AERMOD corresponding to category 8, or “barren land, mostly
desert”. Cuticular resistance, a measure of gas uptake by plants, was set to an arbitrarily low
value of 0.1 sec/cm because this parameter was expected to have little influence for AERMOD
simulations in a desert environment. The low influence of the value of cuticular resistance on
modeled gas concentrations was confirmed by setting the value to 100 sec/cm and observing no
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 15
change in radon air concentrations. As with particulates, radon air concentrations were
simulated using meteorological data for year 2009.
Table 5 presents the output air concentrations for radon for each receptor location and their
associated χ/Q ratios that are input into the GoldSim model.
Table 5. Radon air concentrations (0.25 g/s emissions) and χ/Q ratios for each receptor lo-
cation.
Receptor Location Air Concentration (µg/m3) χ/Q ratio (µg/m3 per g/s)
Embankment (OnSite) 59 234
Knolls OHV Area 0.013 0.053
Grassy Mt. (Aragonite) Rest
Area 0.0022 0.0088
I-80 0.070 0.28
Railroad 0.11 0.44
UTTR Access Road 59 234
Off-Site Exposure Area 0.096 0.38
χ/Q ratios for the Embankment and UTTR Access Road receptors are based on simulations conducted at 0.3-km2
resolution. All other off-site receptors correspond to simulations conducted at 1.0-km2 resolution. Values for I-80
and Railroad are the largest values for any grid cell containing these features.

6.2.2 AERMOD Off-Site Particulate Deposition
In addition to calculating air concentrations of gases and particulates, AERMOD was used to
calculate the fraction of annual mass deposition (g/yr) of resuspended embankment particles
outside the perimeter of the embankment. The total mass of deposited particulates within
AERMOD is a function of the size of the grid area, and is therefore only approximated with a
finite grid area. However, suspended particle re-deposition on the embankment is available as an
output of AERMOD using the 0.3-km2 grid size described in Section 8. The fraction of total
particulate mass deposited outside the embankment area can be calculated by mass balance as:
Depoff-site = 1 – (Depsite / Esite ) (4)
where:
Depoff-site is the fraction of annual PM10 emissions deposited beyond the embankment;
Esite is the annual-average PM10 emission rate per unit area of contaminated soil
(g/m2·yr); and,
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 16
Depsite is the annual deposition rate of resuspended site PM10 within the site perimeter per
unit area of contaminated soil (g/m2·yr).
The majority of PM10 particulates deposited outside the embankment are carried by atmospheric
transport to regions far beyond the vicinity of the embankment. The fraction of all PM10
emissions that is deposited within the combined area of the embankment and the largest potential
size of the off-site dispersion area (64,000 acres, or 260 km2; see the Dose Assessment white
paper) varies depending on PM10 particle size fraction (see Table 2) between approximately 4%
and 11%.
The remaining PM10 mass (89% to 96%) can be expected to be deposited over some very large
region outside the receptor grid at rates no greater than the low values that were calculated with
AERMOD near the receptor grid boundaries. The exact size of this region is influenced by
regional atmospheric conditions and terrain features. At distances beyond approximately 20 to 50
km, AERMOD is unsuitable for air dispersion modeling and a long-range regional model would
be required for quantifying concentrations and deposition rates. The fraction of total particulate
mass deposited within the off-site exposure area is calculated as:
Depoff-site dispersion area = flocal × Depoff-site (5)
where:
flocal is the fraction of annual PM10 deposition occurring within the off-site dispersion
area (see Table 6, Column 4); and ,
Depoff-site is the fraction of annual PM10 emissions deposited beyond the embankment
from Equation 4.
To estimate the total amount of particulate matter deposited on the disposal embankment (Depsite)
for Equation 4, AERMOD simulations were performed using the 0.3-km2 resolution grid for each
of the eight particle size fraction combinations given in Table 2. Table 6 presents the AERMOD
output for total deposition over the disposal embankment. To estimate the amount of redeposited
material, the total mass emitted on an annual basis was calculated based on the AERMOD input
emission rate of 0.25 g/sec. The total annual mass of particulates emitted each year from the
source area is therefore 7,884,000 g. The total mass of particulate matter deposited per square
meter over the embankment (Table 6, Column 2) was then divided by the annual mass emitted to
give an estimate of on-site redeposition of particulate matter (Table 6, Column 3) for each of the
eight simulations. These results were integrated into the GoldSim PA model in a manner
analogous to that described for particle air concentrations in Section 10.
Table 6. Total deposition of PM10 particulate matter on the disposal embankment.
Simulation Total Deposition
(g/m2-yr)
On-site redeposition
(g/m2-yr per g/yr)
Fraction off-site deposition
occurring in off-site exposure area
1 3.3 4.2E-07 0.11
2 3.2 4.1E-07 0.11
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 17
3 3.2 4.0E-07 0.11
4 3.0 3.8E-07 0.099
5 2.6 3.3E-07 0.086
6 2.2 2.8E-07 0.072
7 1.8 2.3E-07 0.057
8 1.4 1.8E-07 0.041

6.3 Confirmation of AERMOD Results with CAP-88
Version 3 of the CAP-88 air dispersion model was used to confirm the results of the AERMOD
simulations. The purpose of this comparison was to perform a quality assurance check on
AERMOD data preparation. As described in Section 7, two types of STAR files for input of
meteorological data to CAP-88 were prepared by MSI. The variability in CAP-88 results using
STAR-ST vs STAR-SR files was about 10-20%, and a number of user input variables (such as
the height of the tropospheric “lid” on mixing) were set at default values. On the AERMOD
side, air concentrations and particle depositions varied by up to a factor of two depending on the
particle size fractions assumed for emissions (see Table 3). Particle size fraction for the emission
rate is not a variable input in the CAP-88 model. These sources of variance are in addition to the
underlying differences in the model frameworks. AERMOD does not employ atmospheric
stability class categories and troposhere "lid" inputs but instead implements planetary boundary
layer methods of estimating atmospheric mixing. Therefore, comparison of AERMOD results
with CAP-88 results is considered on an order-of-magnitude scale, where results within a factor
of 10 or less of each other may be considered nominally equivalent.
Both AERMOD and CAP-88 output air concentrations and ground deposition rates, although
with AERMOD these results are integrated over a receptor grid cell while in CAP-88 they are
associated with specific x,y coordinates. Particle deposition rates were selected as the output for
this comparison. CAP-88 results were obtained for distances of 1 km, 5 km, and 10 km from the
embankment at each of 16 orientations (N, NNW, NW, WNW, etc). Particle deposition results
from AERMOD grid cells overlapping these coordinates were identified. A comparison of these
results for the four cardinal directions is shown in Table 7.
Table 7. Comparison of CAP-88 and AERMOD particle deposition results (g/m2-yr).
Direction Distance (km) CAP-88 deposition AERMOD
deposition
Ratio CAP-88 /
AERMOD
N 1 0.14 0.11 1.2
N 5 0.015 0.016 0.92
N 10 0.0054 0.0047 1.2
W 1 0.13 0.097 1.3
W 5 0.013 0.0037 3.5
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 18
W 10 0.0045 0.00084 5.3
S 1 0.082 0.099 0.83
S 5 0.0081 0.0096 0.85
S 10 0.0029 0.0033 0.89
E 1 0.042 0.21 0.20
E 5 0.0042 0.0045 0.93
E 10 0.0015 0.00056 2.7
Of the 12 comparisons shown in Table 7, CAP-88 and AERMOD particle deposition results were
within a factor of two for all but four results. The largest discrepancies were approximately a
factor of five, for the 10-km distance to the west and the 1-km distance to the east. This
comparison indicates that that there is relatively low variability between the CAP-88 and
AERMOD results considering the differences between these models, and suggests that the
AERMOD results are reliable.

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 19
6.4 Implementation of Cowherd Unlimited-Reservoir Resuspension
Model
A range of input parameter values for the unlimited-reservoir particle resuspension model were
employed to evaluate the possible particle emission rates. Input parameters include fraction of
vegetative cover (V), average annual wind speed (u), surface roughness height (z0), the
unadjusted threshold friction velocity (ut), and the friction velocity adjustment factor. The range
of potential adjustment factors is shown in Figure 3-5 of Cowherd et al (1985). High-end,
middle, and low-end estimates (based on impact to the calculated emission rate (E10) are shown
in Table 8 and discussed in the following paragraphs.
Table 8. Range of input parameter values for particle resuspension modeling.
Parameter units High E10 Middle E10 Low E10
vegetative cover (V) – 0.058 0.172 0.318
average annual wind speed (u) m/s 3.20 3.14 3.10
surface roughness height (z0) cm 5 3.5 2
unadjusted threshold friction velocity (ut) m/s 0.1 0.25 0.7
Friction velocity adjustment factor – 3 4 5
Values for the range of V are based on means for each of the five plant communities evaluated in
test plots near the disposal facility site. The range of u is based on review of five years of Clive
meteorological data. High-end and low-end values are approximate. Values of z0 are based on
Figure 3-6 of Cowherd et al (1985). The value for High E10 is a slightly larger z0 than that of a
wheat field and comparable to "suburban dwellings". This is possibly analogous to widely
spaced shrubs. The z0 of 2 is the lower part of the range for "grassland".
Estimates for ut are the most critical for calculating particle erosion. The range of other
parameters can be estimated, whereas the outcome of soil development on the cap after many
millenia (with respect to particle size distribution, formation of soil crust, amount of projecting
rip rap, etc) is essentially unknown. However, based upon professional judgment, the values
used here are based on examination of Figure 3-4 of Cowherd et al (1985). The value of High
E10 is a factor of 10 below the lowest value for aggregate size distribution (100 µm) shown on the
scale, or 10 µm. This corresponds, by extending the linear function in Figure 3-4, to a ut value of
0.1 m/s. The value of Low E10 corresponds to an aggregate erodible particle size distribution
mode of ~1 mm (1,000 µm). The middle value equates to a 100 µm size. The High E10 value
equates to an aggregate particle diameter smaller than that of silt-size particles (0.05 mm), below
which one may presume a more crusted surface that is not associated with an unlimited-reservoir
erosion model. For the Low E10 value, an aggregate diameter of 1 mm suggests a relatively large
contribution from weathering of rip rap and particle aggregation.

Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 20
The ut adjustment factor estimates were developed based on correlation of expected cap
conditions with photographs in Appendix A of Cowherd et al (1985). Figure A-3, was selected as
the best representation of the likely future cap surface. The associated value of 5 for Figure A-3,
however, is approximately equal to the upper end of the range of adjustment factors shown in
Figure 3-5 of Cowherd et al (1985). Therefore, to capture some range of possible values, factors
of 3, 4, and 5 were used for High E10, Mid E10, and Low E10 calculations, respectively.
Adjustment factors shown in Figure 3-5 span a range between 1 and 7, with the function
steepening rapidly between values of 2 and 7.
The average-annual PM10 emission rates (E10) calculated using Equation 1 are as follows:
• High E10: 0.30 kg/m2-yr;
• Mid E10: 2.5E-07 kg/m2-yr; and,
• Low E10: 1.4E-94 kg/m2-yr.
Because the middle value is effectively zero, these results were represented in the GoldSim PA
model using a log-uniform distribution with boundaries of 2.5E-07 and 0.30 kg/m2-yr.
7.0 Electronic Reference
Atmospheric Modeling Appendix.pdf
This file contains graphical output of air concentrations and particulate deposition related to the
AERMOD simulations described in this white paper.
8.0 References
Cowherd, C., G. E. Muleski, P. J. Englehart, and D. A. Gillette, 1985, Rapid Assessment of
Exposure to Particulate Emissions from Surface Contamination Sites, prepared for
U.S. Environmental Protection Agency, Office of Health and Environmental Assessment,
by Midwest Research Institute, Kansas City, Missouri, EPA/600/8-85/002, February, 1985.
EPA, 1996, Soil Screening Guidance: Technical Background Document, EPA/540/R-95/128,
OSWER Directive 9355.4-17A, Office of Solid Waste and Emergency Response,
U.S. Environmental Protection Agency, Washington, D.C., May 1996.
EPA, 2002, Supplemental Guidance for Developing Soil Screening Levels for Superfund Sites,
OSWER Directive 9355.4-24, U.S. Environmental Protection Agency, Office of Solid
Waste and Emergency Response, Washington, D.C., December 2002.
EPA, 2011a. AERMOD modeling system, model and documentation available on-line at:
http://www.epa.gov/ttn/scram/dispersion_prefrec.htm#aermod
EPA, 2011b. CAP-88 radiation risk assessment software, model and documentation available on-
line at: http://www.epa.gov/rpdweb00/assessment/CAP88/index.html
Atmospheric Transport Modeling for the Clive DU PA
5 November 2015 21
GTG (GoldSim Technology Group), 2011. GoldSim: Monte Carlo Simulation Software for
Decision and Risk Analysis, http://www.goldsim.com
Lakes Environmental, 2010. AERMOD ViewTM, air dispersion modeling package, available on-
line at: http://www.weblakes.com/products/aermod/.
MSI, 2010, January 2009 Through December 2009 and January 1993 Through December 2009
Summary Report of Meteorological Data Collected at EnergySolutions' Clive, Utah
Facility, prepared for EnergySolutions, LLC by Meteorological Solutions Inc, February,
2010.
NESHAPS, National Emission Standards for Emissions of Radionuclides Other Than Radon
from Department of Energy Facilities, 10 CFR 40 Part 61.93, available on-line at:
http://ecfr.gpoaccess.gov/cgi/t/text/text-
idx?c=ecfr&sid=3ae5812c554c6c41807e0fd4dc157bac&rgn=div5&view=text&node=40:8
.0.1.1.1&idno=40
Nielson, K.K., and G.M. Sandquist. 2011. Radon Emanation from Disposal of Depleted Uranium
at Clive, Utah. Report for EnergySolutions by Applied Science Professionals, LLC.
February 2011.
NIST, 2011, NIST Chemistry WebBook, National Institute of Standards and Technology,
available on-line at:
http://webbook.nist.gov/cgi/cbook.cgi?ID=C10043922&Mask=10#Solubility
Rogers, V. C., and K. K. Nielson, 1991. Correlations for predicting air permeabilities and 222Rn
diffusion coefficients of soils, Health Physics 61(2): 225-230.
Utah 2015. License Requirements for Land Disposal of Radioactive Waste. Utah Administrative
Code Rule R313-25. As in effect on September 1, 2015.
Volkovitsky, P., 2004. Radon diffusion and the emanation fraction for NIST polyethylene
capsules containing radium solution. National Institute of Standards and Technology,
Ionizing Radiation Division, available on-line at:
http://www.aarst.org/proceedings/2004/2004_11_Radon_Diffusion_Emanation_Fraction_f
or_NIST_Poly.pdf

 NAC-0022_R2

Biologically Induced Transport
Modeling for the Clive DU PA

Clive DU PA Model v1.4
5 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015
1. Title: Biologically Induced Transport Modeling for the Clive DU PA
2. Filename: Biological Modeling v1.4.docx
3. Description: This documents the methods used in the biologically induced contaminant
transport modeling of the Clive DU PA Model v1.4.

Name Date
4. Originator Dan Levitt 5 November 2015
5. Reviewer Paul Black 5 November 2015
6. Remarks

30 May 2014: Minor edits, including in response to EnergySolutions review. – J Tauxe

5 Nov 2015: Updated from v1.2 to v1.4. – D.Levitt

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015

This page is intentionally blank, aside from this statement.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

1.0
 Summary of Parameters ......................................................................................................... 1

2.0
 Introduction ............................................................................................................................ 5

3.0
 Plant Specifications and Parameters ....................................................................................... 5

3.1
 Plant Conceptual Model .................................................................................................... 5

3.2
 Identification of Plant Functional Groups ........................................................................ 7

3.3
 Estimation of Net Annual Primary Production ................................................................. 8

3.4
 Root/Shoot Ratios ............................................................................................................. 9

3.5
 Maximum Root Depths and Biomass ............................................................................. 11

3.6
 Estimation of Plant Uptake ............................................................................................. 14

4.0
 Ant Specifications and Parameters ....................................................................................... 17

4.1
 Ant Conceptual Model .................................................................................................... 17

4.2
 Clive Field Surveys ......................................................................................................... 17

4.3
 Ant Nest Volume ............................................................................................................ 18

4.4
 Maximum Nest Depth ..................................................................................................... 19

4.5
 Colony Lifespan .............................................................................................................. 19

4.6
 Burrow Density as a Function of Depth ......................................................................... 20

4.7
 Colony Density ............................................................................................................... 20

5.0
 Mammal Specifications and Parameters .............................................................................. 23

5.1
 Mammal Conceptual Model ........................................................................................... 23

5.2
 Clive Site Surveys ........................................................................................................... 24

5.3
 Mound Volume ............................................................................................................... 25

5.4
 Maximum Burrow Depth ................................................................................................ 25

5.5
 Burrow Density as a Function of Depth ......................................................................... 25

6.0
 References ............................................................................................................................ 29

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015
FIGURES
Figure 1. Conceptual model of contaminant uptake and redistribution by plants ........................... 6

Figure 2. Linear regression model to predict ant nest volume based on nest surface area ............ 19

Figure 3 Distribution of ant colony counts for each plot area. ...................................................... 21

Figure 4. Comparison of bootstrapped and a normal distribution for Pogonomyrmex spp. nest
density with depth b parameter .................................................................................... 22

Figure 5. Conceptual diagram of soil movement by burrowing animals ...................................... 24

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015
TABLES
Table 1. Summary of Biotic Transport Parameters ......................................................................... 2

Table 2. Vegetative associations surveyed for embankment cover modeling ................................. 8

Table 3. Species identified at Clive included within each plant group ........................................... 8

Table 4. Measured percent cover of plant groups within each vegetation type (From Tables 1
through 5 in SWCA, 2011) .......................................................................................... 10

Table 5. Great Basin net annual primary productivity .................................................................. 10

Table 6. Root/shoot ratios for plant groups at Clive Site .............................................................. 11

Table 7. Maximum root depths for plant groups at the Clive Site ................................................ 13

Table 8. Proportion root biomass by depth from Clive excavations conducted by SWCA
Environmental Consultants (extrapolated by multiplying average number of roots
per cm in each layer by the total rooting width in each layer, with all layers
summing to 1) .............................................................................................................. 13

Table 9. Fitting parameter b describing root biomass above a given depth for each plant type ... 13

Table 10. Plant/soil concentration ratios ....................................................................................... 15

Table 11. Summary of ant nests in each vegetative association .................................................... 18

Table 12. Summary of Pogonomyrmex nest longevity reported in literature (Adapted from
Neptune 2006, Table 6, p. 32) ..................................................................................... 20

Table 13. Summary of Clive small mammal burrow surveys ....................................................... 25

Table 14. Results of Clive small mammal trapping ...................................................................... 26

Table 15. Soil volume (m3) of excavated mammal burrows ......................................................... 27

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 1
1.0 Summary of Parameters
Following is a brief summary of input parameters used in the biotic transport component of the
Clive Depleted Uranium Performance Assessment Model (Clive DU PA Model) that is the
subject of this white paper.
Table 1 lists the biological transport model parameter distributions for the Clive DU PA Model
that are summarized in this document. For a number of biotic parameters, site specific data were
not available for the Clive site, so the Model makes use of biotic parameters for the same or
similar species developed for the performance assessment of disposal cells at the Nevada
National Security Site (NNSS, formerly the Nevada Test Site), with the assumption that these
species-specific parameters do not vary greatly across North American desert types. The
derivation of these NNSS parameters is detailed in the relevant NNSS documents (Neptune
2005a, 2005b, 2006).
For distributions, the following notation is used:
• N( µ, σ, [min, max] ) represents a normal distribution with mean µ and standard deviation
σ, and optional truncation at the specified minimum and maximum,
• LN( GM, GSD, [min, max] ) represents a lognormal distribution with geometric mean
GM and geometric standard deviation GSD, and optional min and max,
• U( min, max ) represents a uniform distribution with lower bound min and upper bound
max,
• Beta( µ, σ, min, max ) represents a generalized beta distribution with mean µ, standard
deviation σ, minimum min, and maximum max,
• Gamma( µ, σ ) represents a gamma distribution with mean µ and standard deviation σ,
and
• TRI( min, m, max ) represents a triangular distribution with lower bound min, mode m,
and upper bound max.

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 2
Table 1. Summary of Biotic Transport Parameters
Parameter Value Units Reference / Comment
Ant Transport Parameters
Volume of Each Nest
N( µ=0.161,
σ=0.024,
min=0,
max=Large )
m3 SWCA, 2011 (Sec 2.3, Appendix A1)
and Neptune, 2006. See Section 4.3
Lifespan of Each Colony
N( µ=20.2,
σ=3.6, min=0,
max=Large )
yr Neptune, 2006 (Section 6.8, p. 16)
ColonyDensity - area
density of colonies on the
ground
___ ___ SWCA, 2011 (Table 20, p. 23). See
Section 4.7
ColonyDensity_Plot1
Gamma( 33,1,
min=0,
max=Large )
1/ha Ibid.
ColonyDensity_Plot2
Gamma( 2, 1,
min=0,
max=Large )
1/ha Ibid.
ColonyDensity_Plot3
Gamma( 7, 1,
min=0,
max=Large )
1/ha Ibid.
ColonyDensity_Plot4
Gamma( 17, 1,
min=0,
max=Large )
1/ha
SWCA, 2011 (Based on provided
data. Information for this plot in Table
20, p. 23 in the SWCA report is
incorrect.)
ColonyDensity_Plot5
Gamma( 6, 1,
min=0,
max=Large )
1/ha Ibid
MaxDepth - maximum
depth for any colony 212 cm SWCA, 2011 and Neptune, 2006.
See Section 4.4.
b - fitting parameter for
nest shape
N( µ=10,
σ=0.71, min=1,
max=Large )
— Neptune, 2006 (Section 7.3, p. 21)
Mammal Transport Parameters
MoundDensity - area
density of mounds on the
ground
see below for
each plot --- SWCA, 2011 (Section 2.2.2, p. 18 –
22)
_Plot1
Gamma( 235,
1, min=0,
max=Large )
1/ha
_Plot2
Gamma( 239,
1, min=0,
max=Large )
1/ha
_Plot3
Gamma( 1.33,
1, min=0,
max=Large )
1/ha
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 3
_Plot4
Gamma( 1.33,
1, min=0,
max=Large )
1/ha
_Plot5
Gamma ( 1.33,
1, min=0,
max=Large )
1/ha
ExcavationRate -
volumetric rate of a single
burrow excavation
N( µ=0.0006,
σ=0.00015,
min=Small,
max=Large )
m3/yr
Mean of excavated volumes at each
sample location from SWCA, 2011
(Tables 13, 15, 17, 19), corrected for
the number of burrows reported at
each sample location (See Table 14
of this white paper)
MaxDepth - maximum
depth for any burrow 200 cm Neptune 2005b (Table 2)
b - fitting parameter for
burrow shape
N( µ=4.5,
σ=0.84, min=1,
max=Large )
— Fitting parameter for rodent burrows
from Neptune 2005b (Fig. 10, p. 22)
Plant Transport Parameters
BiomassProductionRate U(300,1500) kg/ha yr Approximate Range for Great Basin
from Smith, et al. 1997(Fig 7, p. 37)
PctCover_Plot*_[plant]
Tabulated in
Clive PA Model
Parameters.xls
workbook
— Simulations based on SWCA (2011)
percent cover data. See Section 3.3
Percent cover random
selector
randomly
select between
values 1 to
1000, inclusive
— Modeling construct
Vegetation Association
Picker
Discrete ( 1, 2,
3, 4, 5 ) — Modeling construct
Greasewood Parameters
RootShoot_Ratio U( 0.30, 1.24 ) — Assumed similar to creosote,
Neptune, 2005a (Table 16, p. 38)
MaxDepth 570 cm Robertson, 1983 (p. 311)
b - fitting parameter for
root shape
N( µ=14.6,
σ=0.0807,
min=1,
max=Large )
— Assumed similar to creosote,
Neptune, 2005a (Fig. 9, p. 51)
Grass Parameters
RootShoot_Ratio T( 1, 1.2, 2 ) —
Mode based on Bethlenfalvay and
Dakessian, 1984 (Table 2, p. 314);
bounds based on Neptune, 2005a
MaxDepth 150 cm Based on H. comata from Zlatnik,
1999a (p. 7)
b - fitting parameter for
root shape
N( µ=2.19
σ=0.036,
min=1,
max=Large )
— For perennial grasses, from Neptune
2005a (Fig. 12, p. 55)
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 4
Forb Parameters
RootShoot_Ratio U( 0.40, 1.80 ) — Distribution of “Other Shrubs” used
for conservatism, see Section 3.4
MaxDepth 51 cm Based on Halogeton, from Pavek,
1992 (p. 5)
b – fitting parameter for
root shape
N( µ=23.9 σ
=0.313, min=1,
max=Large )
— Distribution same as “Other Shrubs”,
see Section 3.5
Tree Parameters
RootShoot_Ratio U( 0.55, 0.76 ) — For Juniperus occidentalis from Miller
et al., 2005 (p. 16)
MaxDepth 450 cm For J. occidentalis from Zlatnik,
1999b (p. 6)
b – fitting parameter for
root shape
N( µ=14.6
σ=0.0807,
min=1,
max=Large )
— Distribution for creosote used due to
similar taproot depth, see Section 3.5
Other Shrub Parameters
RootShoot_Ratio U(0.4, 1.8) —
Based on range for Artemisia sp.
from Neptune, 2005a (Table 16, p.
38),
MaxDepth 110 cm Branson et al. 1976 (Fig. 19, p.
1120)
b - fitting parameter for
nest shape
N (µ= 23.9,
σ=0.313,
min=1,
max=Large)
—
Based on fitting parameter for
Atriplex canascens at NNSS, from
Neptune 2005a (Fig 10, p. 52)
Plant/Soil Concentration Ratios
PlantCRs by chemical
element
tabulated in
Clive PA Model
Parameters.xls
workbook
— See Table 10
Plant CR GM for Rn Small — See Table 10
Plant CR GSD for Rn 1 — See Table 10
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 5
2.0 Introduction
Biotic fate and transport models have been developed for the depleted uranium (DU) waste cell
at the Clive repository to evaluate the redistribution of soils, and contaminants within the soil, by
native flora and fauna. The biotic models are part of the larger Clive DU PA Model that has been
built to evaluate the consequences of contaminant migration over time from the DU waste cell.
The purpose of the Model is to provide a decision management system that will support future
disposal, closure and long term monitoring decisions, as well as supporting all regulatory
requirements of PAs and other environmental assessments for these waste disposal systems. The
Clive facility is located in the eastern side of the Great Salt Lake Desert, with flora and fauna
characteristic of Great Basin alkali flat and Great Basin desert shrub communities.
3.0 Plant Specifications and Parameters
The purpose of this chapter is to explain the component of the Clive DU PA Model that
addresses calculation of plant-mediated contaminant mass distributions by depth, and the rate of
contaminant transport from subsurface strata to the ground surface.
3.1 Plant Conceptual Model
Plant-induced transport of contaminants is assumed to proceed by absorption of contaminants
into the plant’s roots, followed by redistribution throughout all the tissues of the plant, both
aboveground and belowground. Upon senescence, the aboveground plant parts are incorporated
into surface soils, and the roots are incorporated into soils at their respective depths (Figure 1).
The calculations of contaminant transport due to plant uptake and redistribution take place in a
series of steps:
1. Calculate the fraction of plant roots in each layer for each plant type.
2. Calculate uptake of contaminants into plant roots in each layer.
3. Sum the contaminant uptake to determine the total uptake by the roots for each
contaminant.
4. Determine the average concentration in the roots, assuming complete redistribution
within the root mass.
5. Assuming that the plant returns all fixed contaminants to adjacent soils upon senescence,
determine how much of each contaminant is returned to each layer. The aboveground
plant parts are mixed in the uppermost layer.
6. Calculate uptake of contaminants into aboveground parts of the plant ("shoots"), based on
the fractions of roots fixing contaminants within each layer and sending it up to the
shoots.
7. Calculate the net flux of contaminants into (or out of) each layer due to steps 1 through 6.
This value is used to adjust contaminant inventories in each layer (each layer is a
GoldSim cell).
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 6

Figure 1. Conceptual model of contaminant uptake and redistribution by plants
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 7
This section describes the functional factors that contribute to the parameterization of the plant
section of the biotic transport model. Such factors include identifying dominant plant species,
grouping plant species into categories that are significantly similar in form and function with
respect to the transport processes, estimating net annual primary productivity (NAPP, a measure
of combined aboveground and belowground biomass generation), determining relative
abundance of plants or plant groups, evaluating root/shoot mass ratios, and representing the
density of plant roots as a function of depth below the ground surface. The data used for each of
the seven steps of the algorithm are presented, outstanding issues with the available data are
identified, and the issues that deserve attention for the next model iteration are described.
In the Clive DU PA Model, the vertical soil horizon is discretized into horizontal layers based on
various functional attributes of the soil-based biotic communities (plants and animals),
requirements related to gas and liquid transport, and the configuration of the disposal cell cover.
The Model is ultimately used to simulate radionuclide transport throughout the soil layers.
Utilizing the information provided in 1 through 6 above, distributions of aboveground and
belowground NAPP for grasses, forbs, shrubs and trees are developed. Radionuclide activity
associated with aboveground biomass is assigned to the uppermost soil/cover layer in the Model.
Radionuclide activity associated with belowground NAPP is apportioned by depth interval
according to root mass distribution. In order to reflect the redistribution of radionuclides, these
calculations require the use of plant uptake factors (plant/soil concentration ratios) to model the
relative uptake of contaminants from soil by plants.
3.2 Identification of Plant Functional Groups
Field surveys of the Clive site and surrounding areas were conducted by SWCA Environmental
Consultants in September and December 2010 to identify plant species present in different
vegetative associations around the Clive Site (SWCA Environmental Consultants, 2011). Five
different vegetative associations were surveyed, with three associations representing the alkali
flat/desert flat type soils found in the vicinity of Clive, and two associations representing the
desert scrub/shrub-steppe habitat characteristic of slopes and slightly higher elevations with less-
saline soil chemistry. A one hectare (100 m × 100 m) plot was established in each vegetative
association, and each plot was surveyed for dominant plant species present, and the percent cover
and density of each species. In addition, a small number of black greasewood, shadscale,
halogeton, and Mojave seablite plants were excavated to obtain root profile measurements and
aboveground plant dimensions. The vegetative associations for each plot are shown in Table 2.
Plots 3 through 5 represent current vegetation at the Clive site, while Plots 1 and 2 are
representative of less-saline soils that may develop on top of the waste cell cover.
A total of 41 plant species were identified on the five survey plots. Eighteen species each
comprised at least 1% of the total cover on at least one plot. These 18 species were considered
the most important for purposes of modeling plant-mediated transport of chemical contaminants
at Clive. Species were grouped into five functional plant groups, as shown in Table 3. The five
functional groups are: grasses, forbs, greasewood, other shrubs, and trees. Greasewood is
separated from other shrubs due to its status as a phreatophyte that can extend taproots in excess
of five meters to reach groundwater. Annual and perennial grasses were grouped due to similar
maximum rooting depths.

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 8
Table 2. Vegetative associations surveyed for embankment cover modeling
Plot Number Plot Name
1 Mixed Grassland
2 Juniper sagebrush
3 Black Greasewood
4 Halogeton-disturbed
5 Shadscale-Gray Molly

Table 3. Species identified at Clive included within each plant group
Plant Group Common Name Species Name
Forbs Halogeton Halogeton glomeratus
Forbs Mojave seablite Suaeda torreyana
Forbs Curveseed butterwort Ranunculus testiculatus
Grasses Needle and thread Hesperostipa comata
Grasses Intermediate wheatgrass Thinopyrum intermedium
Grasses Sandberg bluegrass Poa secunda
Grasses Crested wheatgrass Agropyron cristatum
Grasses Muttongrass Poa fendleriana
Grasses Tall wheatgrass Thinopyrum ponticum
Grasses Slender wheatgrass Elymus trachycaulus
Grasses Western wheatgrass Pascopyrum smithii
Grasses Cheatgrass Bromus tectorum
Greasewood Black greasewood Sarcobatus vermiculatus
Shrubs Big sagebrush Artemisia tridentata
Shrubs Shadscale saltbush Atriplex confertifolia
Shrubs Gray molly Bassia americana
Shrubs Broom snakeweed Gutierrezia sarothrae
Trees Utah juniper Juniperus osteosperma
3.3 Estimation of Net Annual Primary Production
Net annual primary productivity has not been measured at the Clive site or in the adjacent
vegetative associations. NAPP can vary widely on an annual basis and is strongly correlated with
mean annual water availability; in desert ecosystems, it correlates moderately well with annual
precipitation (Smith et al., 1997). Smith et al. (1997, Figure 7, p. 37) show Great Basin NAPP
ranging from approximately 300 to 1500 kg/ha/yr, and report mean NAPP for Great Basin
terrestrial systems of 920 kg/ha/yr. Given the lack of site-specific NAPP data, the variability of
NAPP, and the dependence of NAPP on annual water availability, it is reasonable to assume for
the initial modeling effort that NAPP in the area of Clive has a uniform distribution of 300 to
1500 kg/ha/yr. A total biomass production for the selected plot is drawn from this distribution.
Since these data are not on a per-plant or per-species basis, percent cover of each plant group
will be used to apportion NAPP by vegetation type. This biomass is then apportioned based on
the percent of vegetation from each plant type. Percent cover of each plant species was measured
in 100 separate 1-m2 quadrats located along ten transects in each Plot. Mean percent cover for
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 9
each species was reported by SWCA (2011, Tables 1 through 5) for plant species recorded in
each vegetation association; this information is summarized by plant group in Table 4.
A distribution for percent plant cover was developed using a bootstrap resampling approach to
estimate the sampling distribution of the mean percent plant cover (Efron 1998). The percent
plant cover is to be applied for the full 10 ka performance period, and thus it is the distribution of
the mean percent plant cover that is being modeled, to account for the time averaging. The
bootstrap resampling simulation needs to reflect the same sort of sampling structure as the field
sampling, in order to capture the underlying structure of the data. To simulate this structure, five
transects from two subplots were selected at random from each plot, then 10 quadrats within
those five transects were selected at random. This means that quadrat data originally within a
transect were resampled together, and transect data from within a subplot were resampled
together. Subplot data within a plot were resampled together, and data between plots were not
mixed. As in standard bootstrap resampling, each random selection was done with replacement.
A mean value was then calculated for percent cover of each plant type from the two subplots. To
calculate total percent coverage, percent coverage for each plant type in each simulation was
aggregated. The percent coverage for each plot, for each plant type, and for each simulation was
saved in a table, with the entire process being repeated 1,000 times. Since data was collected on
only two of the four subplots within a plot, there are only four ways in which the two subplots
can be selected. Therefore, in this phase of the bootstrap resampling, all four possibilities are
calculated and assigned equal weight. No standard statistical distribution provided an adequate
fit to the resulting mean percent cover values. Thus, the simulated values were recorded in a
table, and each simulated value is drawn with equal likelihood in the Clive DU PA Model. All
percent cover simulation results are shown in the Clive PA Model Parameters Workbook.
To calculate total biomass by plant type, these percent cover simulations are used with the Total
Biomass distribution to apportion biomass by plant type. For example, if a plot with 20% shrubs,
30% grasses, and 50% bare ground is assumed to produce 1000 kg of biomass, 400 kg is
assumed to be produced by shrubs and 600 kg is assumed to be produced by grasses (Table 5),
since bare ground, which for purposes of this model includes litter and biological crust, is
assumed to produce no biomass.
3.4 Root/Shoot Ratios
Distributions of aboveground and belowground biomass production for plant groups are
developed from the total NAPP based on root/shoot ratio for each plant group. The root/shoot
ratio is the ratio of belowground (root) mass to aboveground (shoot) mass. Estimates of
belowground NAPP are determined by multiplying total NAPP by the root/shoot ratio of the
species of concern. Aboveground NAPP is equivalent to the remaining portion of total NAPP.
Root/shoot ratios for each plant group are shown in Table 6. A triangular distribution was
developed for the grasses root/shoot ratio. Data from Bethlenfalvay and Dakessian (1984, Table
2, p. 314) for Hesperostipa comata suggesting a root/shoot ratio of 1.2 in ungrazed systems was
used for the mode of the distribution. Furthermore, since root/shoot ratios for grasses generally
range from 1:1 to 2:1 (Neptune, 2005a) the endpoints of the distribution were set at a minimum
of one and a maximum of two. For greasewood, the root/shoot ratio is based on information in
Neptune (2005a) for creosote (Larrea tridentata), a warm desert shrub with a similar growth
form to greasewood. The root/shoot ratio for the “Other Shrubs” category is based on the range
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 10
of root/shoot ratios reported for sage (Artemisia spp.) by Neptune (2005a, Table 16, p. 38). Utah
juniper (Juniperus osteosperma) is the only tree found in any of the five survey plots. The
root/shoot ratio for trees is based on western juniper (Juniperus occidentalis), a closely related
species, as reported by Miller et al. (2005, p. 16). No root/shoot information was available for the
primary forbs occupying the site (halogeton and curveseed butterwort). This lack of information
represents a data gap, though biointrusion modeling at NNSS showed that forbs, due to their
more shallow rooting system and smaller contribution to NAPP, contributed very minimally to
the biotic transport of buried wastes. To parameterize this model input, the root/shoot ratio for
other shrubs was used, because this ratio represents a uniform distribution with a wide range and
relatively large upper bound. For modeling of contaminant uptake, this means that the
distribution tends to be conservative, since a large proportion of the plant mass can be
determined to be underground, which results in increased absorption and upward movement of
any contaminants in a given layer where roots occur.

Table 4. Measured percent cover of plant groups within each vegetation type (From Tables
1 through 5 in SWCA, 2011)

Plot 1:
Mixed
Grassland
Plot 2:
Juniper -
Sagebrush
Plot 3:
Greasewood
Plot 4:
Halogeton -
Disturbed
Plot 5:
Shadscale -
Gray Molly
% Tree 0 6.2 0 0 0
%
Greasewood 0 0 4.5 0.2 0.2
% Other
Shrub 2.0 18.9 0.6 5.0 13.1
% Forb 2.2 1.4 0.8 3.9 1
% Grass 26.4 9.8 0 0 0.1
% Bare
Ground 69.4 63.7 94.1 90.9 85.6

Table 5. Great Basin net annual primary productivity
Group Value or Distribution Units References
Total Biomass
(Primary productivity)
U(300, 1500) kg/ha/yr Range for Great Basin from Smith,
et al. 1997. Mean of 920 kg/ha/yr
reported by Le Houerou 1984. Net
primary productivity dependent upon
total moisture availability
Biomass
Greasewood
Apportioned from
above by % cover of
each vegetation type

Biomass Shrubs
Biomass Grasses
Biomass Forbs
Biomass Trees

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 11
Table 6. Root/shoot ratios for plant groups at Clive Site
ES Plant
Type
Value or
Distribution
Units References
Forbs U(0.40, 1.80) — Distribution of “Other Shrubs” used
for conservatism, see text
Grasses Tri(1, 1.2, 2) — Based on H. comata (ungrazed),
Bethlenfalvay and Dakessian, 1984
Greasewood U(0.30, 1.24) — Assumed similar to creosote, from
NTS (Neptune, 2005a)
Other Shrubs U(0.4, 1.8) — Based on range for Artemisia spp.
from Barbour, 1973
Trees U(0.55, 0.76) — For Western Juniper, Miller et al.,
2005

3.5 Maximum Root Depths and Biomass
Maximum root depths for each of the plant groups are based on literature values as shown in
Table 7. Forbs are the most shallowly rooted plant group at Clive, with halogeton roots
extending half a meter or less based on excavations conducted by SWCA (2011, Table 6).
Though roots of some perennial grasses have been shown to extend up to two and a half meters
(Zlatnik, 1999c), maximum rooting depths for the two most abundant grasses identified in the
2011 SWCA surveys of the Clive plots [needle and thread grass (Hesperostipa comata) and
cheatgrass (Bromus tectorum)] extend about 1.5 meters (Zlatnik, 1999a, and Zouhar, 2003).
Greasewood has been reported to extend taproots up to 19 meters to reach groundwater (SWCA
Environmental Consultants, 2000, p. 2), though this extreme situation will only occur when
precipitation can infiltrate to groundwater, as greasewood roots cannot penetrate the very dry soil
that occurs below the zone of infiltration. The vegetative survey of the Clive site found that the
majority of greasewood plants are less than one meter tall, and studies have found that
greasewood of that size tend not to produce taproots (Robertson, 1983). Still, larger plants do
occupy parts of the Clive site, especially where precipitation runoff is concentrated, and these
plants may extend taproots to exploit deeper water. A maximum root depth of 5.7 meters
(Robertson, 1983, p. 311) is used in this model. Maximum root depth for the “Other Shrub”
category is based on rooting depths for shadscale as reported in Branson et al. (1976, Fig. 19, p.
1120). The maximum rooting depth of three shadscale excavated at the Clive site (Table 6 in
SWCA, 2011) was approximately 75 cm.
The proportion of root biomass as a function of depth was determined for greasewood, shadscale
(i.e. other shrubs), and halogeton and mojave seablite (i.e. forbs) based on root profile
excavations conducted by SWCA Environmental Consultants (2011) and is presented in Table 8.
Maximum rooting depth for the only tree species found on any of the five survey plots (Utah
juniper, Juniperus osteosperma) was based on rooting depths of the similar Western juniper
(Juniperus occidentalis), which has been found to extend taproots as deep as 4.5 meters (Zlatnik,
1999b, p. 6). Understanding root biomass by depth is necessary to apportion belowground
biomass production to depth layers or “cells” within the cover component of the Clive DU PA
Model. The first step entails modeling the depth distribution of plant mass for each shrub and
grass species. Once this is accomplished, a model is applied to the aggregate within each layer.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 12
The Clive DU PA Model uses the work done by Neptune
(2005a) at NNSS to fit mathematical functions describing
the root mass by depth for each of the plant groups. Fitting
parameters (b) describing the root biomass as a function of
depth for each of the Clive plant groups are presented in
Table 9. All plant types use the same generic mathematical
function to represent the density of roots with depth, from
which is derived the value for N
if , the fraction of root in
each layer N. Each plant type, however, is assigned specific
distributions of parameter values max
iz and bi to change the
shape of the function in order to fit available root density
data.
The function fi used to represent root densities actually
defines the fraction of all roots above any given depth. At
depth z = 0, the value is obviously 0, and at the maximum
root depth max
izz=the value is 1, meaning that all roots are above that depth (the definition of
maximum root depth). The fraction of roots for plant i above any depth z is
,11
ib
max
i
z
i z
zf ⎟⎟
⎠
⎞
⎜⎜
⎝
⎛−−= (1)
where
z
if = fraction of roots for plant i above any depth z,
max
iz = maximum root depth for plant i, and
bi = fitting parameter for the root density equation, for plant i.

A value of b = 1 indicates a uniform cylindrical “can-shape” distribution of roots from the
surface to maximum rooting depth. Increasing b values result in a narrowing of overall rooting
width with depth, with b = 3 resulting in a “cone-shaped” distribution of roots, and b values
greater than 4 indicating increasingly “funnel-shaped” distributions with depth, as might be
found in plants producing taproots. Neptune’s work at the NNSS did not develop b parameters
for forbs and trees. However, as shown in Table 8, excavations of halogeton, the dominant forb
at the Clive site, show that all root mass is in the top 50 cm of soil. Tilley et al. (2008) report that
halogeton does form a taproot that can extend to approximately 50 cm below the surface.
Therefore, the selected b for forbs at Clive was based on the b for “other shrubs” at the NNSS,
which had deeper maximum rooting depths but similar “shape” of root apportionment with
depth. As discussed previously, the NNSS biointrusion modeling excluded evaluation of forbs
due to their minimal contribution to the biotic transport of buried wastes. Additional excavations
of halogeton to better define distribution of root mass with depth could be performed in the
future if this uncertainty influences modeling results. Neptune’s work at the NNSS also did not
derive b parameters for trees. Therefore, the fitting parameter for juniper roots is based on the b
derived for creosote, which also forms a taproot and has a fairly deep maximum rooting depth
[315 cm (Neptune, 2005a)] as that used here for juniper [450 cm (Zlatnik, 1999b)].
b > 1
b =
1
10
ma
x
i
m
u
m

de
p
t
h
de
p
t
h
fraction above depth
0
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 13
Table 7. Maximum root depths for plant groups at the Clive Site
ES Plant Type Value or
Distribution
Units References
Forbs 51 cm For Halogeton from Pavek, 1992
Grasses 150

cm Based on H. comata (Zlatnik, 1999a)
and B. tectorum (Zouhar, 2003), the two
most abundant grasses at Clive
Greasewood 570 cm Robertson, 1983
Other Shrubs 110 cm Based on shadscale from Branson et
al., 1976
Trees 450 cm Value for Western Juniper from Zlatnik,
1999b
Table 8. Proportion root biomass by depth from Clive excavations conducted by SWCA
Environmental Consultants
Depth
Interval
(cm)
Proportion Rootmass in Layer
Black Greasewood Other Shrubs Forbs
Mean St. Dev. Mean St. Dev. Mean St. Dev.
0–10 0.029 0.025 0.096 0.023 0.217 0.109
10–20 0.405 0.315 0.344 0.227 0.434 0.219
20–30 0.292 0.18 0.306 0.059 0.268 0.213
30–40 0.15 0.065 0.197 0.124 0.07 0.099
40–50 0.078 0.029 0.042 0.019 0.012 0.016
50–60 0.03 0.041 0.003 0.006 0 0
60–70 0.015 0.014 0.002 0.003 0 0
70–80 0.001 0.001 0.003 0.006 0 0
80–90 0 0 0.003 0.006 0 0
90–100 0 0 0.005 0.009 0 0

Table 9. Fitting parameter b describing root biomass above a given depth for each plant
type
ES Plant
Type Value or Distribution References
Forbs N( µ=23.9 σ =0.313, min=1,
max=Large )
Fitting parameter based on “other shrubs” at
NNSS (Neptune, 2005a). See Section 3.5
Grasses N(2.19, 0.036, min=1,
max=Large)
Fitting parameter for perennial grasses
(Neptune, 2005a)
Greasewood N( µ=14.6, σ=0.0807, min=1,
max=Large)
Based on fitting parameter for creosote at
NNSS (Neptune, 2005a)
Other
Shrubs
N(23.9, 0.313, min = 1,
max=Large)
Based on fitting parameter for four-winged
saltbush at NNSS (Neptune, 2005a)
Trees N( µ=14.6 σ=0.0807, min=1,
max=Large )
Based on fitting parameter for creosote at
NNSS (Neptune 2005a). See Section 3.5
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 14
3.6 Estimation of Plant Uptake
Radionuclide concentrations in plant tissues are calculated based on root uptake using plant/soil
concentration ratios (Kp-s), expressed as activity per dry weight plant tissue divided by activity
per dry weight of bulk soil (Bq/g per Bq/g). Element-specific Kp-s values were preferentially
obtained from a recent publication of the International Atomic Energy Agency (IAEA, 2010). A
report by Pacific Northwest National Laboratory (Staven et al., 2003) was used as a secondary
reference when element-specific values were not available in IAEA (2010).
Element-specific values of Kp-s were available in IAEA (2010) for all Clive DU PA radionuclides
of concern with the exception of actinium, iodine, protactinium, and radon. For actinium and
protactinium, americium values were employed as a surrogate as suggested in Staven et al.
(2003). A Kp-s value for iodine was obtained from Stave et al. (2003). A summary of Kp-s values
used in the Clive DU PA is provided in Table 10.
Distributional form for the values of geometric mean and geometric standard deviation reported
in IAEA (2010) was not discussed in this reference. In order to provide a common set of inputs,
values obtained from IAEA (2010) and Staven et al. (2003) were processed to conform to an
assumed lognormal distribution. The value for iodine originally reported as an arithmetic mean
was transformed to a geometric mean equivalent. Kp-s data were reported in IAEA (2010) as a
geometric mean, geometric standard deviation, minimum, and maximum. The geometric
standard deviations are greater than 2 in nearly every case, suggesting high right-skewness in the
data, and the minimum and maximum were consistent with samples from a lognormal
distribution. In order to establish a distribution for the mean, a parametric bootstrap approach
was taken (Efron 1998), simulating bootstrap samples from the lognormal distribution using the
maximum likelihood estimates of the lognormal parameters. A lognormal distribution was then
fit to the resulting bootstrap simulations of the mean, since some right-skewness was still present
in the sampling distribution.
Plant/soil concentration ratios reflect an assumption that there is a linear and unchanging
relationship between soil and plant tissue concentrations. In reality, Kp-s values are liable to
overestimate plant tissue concentrations as soil concentrations increase to levels higher than
those employed in the studies from which the values are derived. This concern may apply in the
Clive DU PA Model to conditions where plant roots are in contact with relatively high uranium
concentrations, such as in disposed DU waste. The Model assumes that plant roots are in contact
with soils in various layers belowground, each of which has its own concentration of
contaminants (“Species” in GoldSim parlance). The roots present in each layer absorb each
Species proportionally to the concentration of that Species in the soil in that layer. These
absorbed Species are distributed uniformly throughout all the plant’s tissues, aboveground and
belowground. The plant is then assumed to die off, and all the Species contained within it are
returned to soils in each layer according to the fraction of roots present in that layer.
Aboveground plant parts are returned to the topmost soil layer. All of these processes take place
in a single time step.

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 15
Table 10. Plant/soil concentration ratios
Element Sample
Size
Geometric
Mean
Geometric
Std. Dev. Notes
Actinium 27 0.0037 1.50 Americium used as a surrogate, based
on Staven et al. (2003)
Americium 27 0.0037 1.50
Cesium 401 0.67 1.13
Iodine 1 0.066 3.87 Geo mean based on Staven et al.
(2003). Geo SD from Sheppard and
Evenden (1997).
Neptunium 16 0.095 1.35
Protactinium 27 0.0037 1.50 Americium used as a surrogate, based
on Staven et al. (2003).
Lead 34 0.29 1.54
Plutonium 22 0.0010 1.35
Radium 42 0.44 1.82
Radon NA arbitrarily
small number
1 Radon gas is inert and has effectively
no potential to establish equilibrium in
plant tissue.
Strontium 172 1.8 1.07
Technetium 18 131 1.39
Thorium 64 0.39 1.47
Uranium 53 0.17 1.49

The concentration of Species j in the plant i with roots in layer N is simply
,,
N
sj
N
ji CCRC⋅= (2)
where
N
jiC, = concentration of Species j in plant i roots in layer N,
CRj = concentration ratio for all plants and Species j (Table 10), and
N
sC = concentration in soil on layer N.
The total mass of Species j extracted by roots of plant i from soils (or wastes) in layer N is
shoot
N
ii
N
jiroot
N
ii
N
ji
N
ji ffMPCffMPCM⋅⋅⋅+⋅⋅⋅=,,,, (3)
where
root
if = mass fraction of plant i that is in the roots (belowground fraction),
N
if = mass fraction of root of plant i that is in layer N (so that the fraction of the
entire plant in layer N is root
if × N
if ),
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 16
shoot
if = mass fraction of plant i that is in the shoots (aboveground fraction),
N
jiM, = mass of Species j extracted by the roots of plant i in layer N, and
MPi = mass of all individuals of plant i over the site (M).

The model assumes that all absorbed Species are distributed uniformly throughout all the plant
tissues, both aboveground parts and roots. The total mass of Species j in plant i is the total mass
extracted by the roots of the plant summed across all N layers:
,,,∑=
N
N
ji
T
ji MM (4)
where
T
jiM, = total mass of Species j extracted by the roots of plant i and redistributed
throughout the plant tissues, and
N
jiM, = mass of Species j extracted by the roots of plant i in layer N.

This total amount of Species mass is divided up into the parts of the plant that occupy each layer,
as well as the aboveground parts, so that we may calculate the mass of contamination N
jiM,
+ that
the plant returns to the various soil layers upon senescence. The total amount of contamination
returned to the soils must equal the amount that was absorbed (not accounting for decay of the
Species) in order to conserve mass of the Species. This total absorbed Species mass is returned to
the soil in proportion to the amount of plant in each layer, with the topmost soil layer also
receiving the aboveground plant parts:

N
iroot
T
ji
N
ji
iroot
T
jiji
shoot
T
jiiroot
T
jiji
ffMM
ffMM
fMffMM
⋅⋅=
⋅⋅=
⋅+⋅⋅=
+
+
+
,,
2
,
2
,
,
1
,
1
,
! (5)

The net mass added to each layer is the redistributed mass from Eq. (5) minus the absorbed mass
from Eq. (3). For plant i, this net mass added is simply
.,,
N
ji
N
ji MM−+ (6)

The Clive DU PA Model contains various plant types. For the sake of simplicity in defining
changes to each cell’s inventory, the Species redistribution for all plants can be combined to
result in a net addition (or subtraction) of mass effected by all plants. To do so, we sum Eq. (6)
over all the plant types:
.and ,,∑∑==++
i
N
ji
N
j
i
N
ji
N
j MMMM (7)

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 17
4.0 Ant Specifications and Parameters
4.1 Ant Conceptual Model
Ants fill a broad ecological niche in arid ecosystems as predators, scavengers, trophobionts and
granivores. However, it is their role as burrowers that is of main concern for the purposes of this
model. Ants burrow for a variety of reasons but mostly for the procurement of shelter, the rearing
of young and the storage of foodstuffs. How and where ant nests are constructed plays a role in
quantifying the amount and rate of subsurface soil transport to the ground surface at the Clive
site.
Factors relating to the physical construction of the nests, including the size, shape, and depth of
the nest, are key to quantifying excavation volumes. Factors limiting the abundance and
distribution of ant nests such as the abundance and distribution of plant species, and intra-
specific or inter-specific competitors, also can affect excavated soil volumes. Parameters related
to ant burrowing activities include nest area, nest depth, rate of new nest additions, excavation
volume, excavation rates, colony density, and colony lifespan. These attributes are described in
this section, along with other considerations involving the impact of ant species and their
inclusion in the Clive DU PA Model.
The calculations of contaminant transport due to ant burrowing involve three steps:
1. Identify which of the ant species overwhelmingly contribute to the rearrangement of soils
near the surface at Clive.
2. Calculate soil and contaminant excavated volume using maximum depth, nest area, nest
volume, colony density, colony life span, and turnover rate for predominant ant species.
3. Calculate burrow density as a function of depth to determine the distribution of
contaminants within the vertical soil profile for each predominant ant species.
4.2 Clive Field Surveys
Surveys for ants at Clive were limited to surface surveys of ant colonies, including identification
of ant species, measurements (length, width, and height) of ant mounds, and determination of ant
nest densities in each vegetative association (SWCA Environmental Consultants, 2011). No
excavations of ant nests were performed at Clive to support the initial Clive DU PA Model,
though excavations could be conducted to support future model iterations if ant nest depth and
volume are found to be sensitive parameters. Only two species of ants were identified during the
surveys, with the western harvester ant, Pogonomyrmex occidentalis, accounting for 62 of the 64
nests identified. The second ant species, a member of the genus Lasius, was only encountered
twice, both times in the mixed grassland plot. A summary of ant nests in each vegetative
association is shown in Table 11.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 18
Table 11. Summary of ant nests in each vegetative association
Vegetative Association Number of
Mounds/Hectare
Average Mound Surface Area
(sq dm)
Plot 1: Mixed Grassland 33 95.03
Plot 2: Juniper-Sagebrush 2 39.77
Plot 3: Greasewood 7 120.18
Plot 4: Halogeton-disturbed 17 84.43
Plot 5: Shadscale-Gray Molly 6 137.73

4.3 Ant Nest Volume
Ant nests were not excavated at the Clive site, so only nest surface area, not nest volume or depth
data, were available. Generally, the surface areas of the Clive sites were smaller than the surface
areas at the sites studied at the NNSS. To obtain estimates of nest volumes, a regression was
made using Pogonomyrmex nest volume surface area data collected at the NNSS (Neptune,
2006) with nest surface area data described in Table 11. The NNSS data and associated
regressions are shown in Figure 2. To be consistent with the data available from NNSS, the areas
calculated are the two-dimensional areas of the mound, not the conical surface area.
To predict nest volume as a function of surface area, the following steps were taken:
1. Using data from NNSS, a linear model was fit to log transformed surface area and
volume data to predict nest volume. Figure 2 shows the fitted model along with the
predicted values based on measured surface area values from the Clive study.
2. To estimate the uncertainty in the predicted volume values, a model-based resampling
method was used. With the statistical model created with the NNSS data, data from Clive
were resampled with replacement. New values were estimated by drawing from a normal
distribution whose mean was the predicted value and whose standard deviation is a
function of both the fitting error and the residual error. This was repeated 10,000 times.
3. The distribution of the mean volume is summarized by the mean and standard deviation
of the resampled values.
Modeling all sample plots together resulted in a volume distribution of N( 0.161 m3, 0.024 m3 ).
Predicted nest volumes were smaller than those observed at NNSS, where the volume
distribution was N( 0.64 m3, 0.091 m3 ).
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 19

Figure 2. Linear regression model to predict ant nest volume based on nest surface area

4.4 Maximum Nest Depth
Again, since ant nests were not excavated, maximum nest depth had to be determined by other
means. As shown in Figure 2, NNSS data support the assumption that larger mound surface area
features correlate with larger nest volumes and deeper maximum depths; therefore, the mound
dimension data collected by SWCA (2011, Table 20, p. 23) was used to predict nest depths. The
upper 95% prediction interval of SWCA-measured surface area was used with the NNSS linear
model predicting depth as a function of surface area. The upper 95% prediction interval was used
in lieu of a maximum value because taking the maximum of simulated values from an
unbounded normal distribution could result in an unrealistically large value. Using this approach,
the predicted maximum nest depth at Clive is 212 cm.
4.5 Colony Lifespan
A critical component in modeling excavation volume is the turnover rate, or the fraction of the
volume of the ant nest that is excavated in any given year. The turnover rate itself is inversely
related to the life span of the colony. Table 12 shows four literature studies that report colony
lifespan for P. occidentalis or Pogonomyrmex spp. These Pogonomyrmex spp. entries are
included because the P. occidentalis study simply suggests colony lifespan is greater than 7
years, indicating that the study did not continue until colony failure. The non-specific studies
include one entry that suggests a range of 15–20 years, one that suggests a range for the Queen
of 17–30 but only 2–17 for the nest, and an entry of 20.2 ± 8.1 (standard deviation) based on 5
observations. The NNSS cover modeling (Neptune, 2006) used the latter entry, including the
●
●
●
●
●
●●
●
●
●
●
●
●
●
0.1 0.5 2.0 10.0
0.
0
5

0
.
1
0

0
.
2
0

0
.
5
0

1
.
0
0
Surface Area m2
Vo
l
u
m
e

m
3
●●
●
●
●●
●
●
●
●●●●
●
●●
●
●
●
●●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●●
●●
●
●
●
●
●
●
●●
●●
●
●
●
●●
●●
●
●
●●
●
NTS Measurements
ES Predicted Vol
●
●
●
●
●
●●
●
●
●
●
●
●
●
0 5 10 15 20
0.
0

0
.
2

0
.
4

0
.
6

0
.
8

1
.
0

1
.
2
Surface Area m2
Vo
l
u
m
e

m
3
●●●●●●
●
●●●●●●●●●
●
●
●●●●
●●●●●●●●
●●
●●●●●●●
●●●
●●●
●
●
●
●
●
●
●●●●
●
●●●●
●●
●
●●
●
●
NTS Measurements
ES Predicted Vol
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 20
information that there were 5 data points. Since the standard deviation was based on 5
observations, the standard deviation of 8.1 was divided by the square of 5 to arrive at a normal
distribution with a mean of 20.2 years and standard deviation of 3.6 years. This same distribution
was used here. To ensure non-negative values as well as allow division by colony life, the
distribution is truncated at 1e-20.
Table 12. Summary of Pogonomyrmex nest longevity reported in literature (Adapted from
Neptune 2006, Table 6, p. 32)
Genera and species
Max nest (n) or
queen (q) longevity
(years)
Number of
observations Authors
Pogonomyrmex 17–30 (q) Hölldobler and Wilson 1990
2–17 (n) Hölldobler and Wilson 1990
20.2 ± 8.1 5 Porter and Jorgensen 1988
Pogonomyrmex
occidentalis (Cresson) >7 (n) Hölldobler and Wilson 1990

4.6 Burrow Density as a Function of Depth
Excavation volume gives an overall picture of how much soil is being transported to the soil
surface. However, it is also important to determine the density of burrowing activities as a
function of depth within the vertical soil profile. The shape of the nest under the surface
expression of the nest gives insight into the quantity of contaminated soils at various depths
being excavated to the surface. The burrow density as a function of depth is described by the
fitting parameter b. Lacking site-specific nest excavations at Clive, the fitting parameter
developed in the NNSS study (Neptune, 2006) for all Pogonomyrmex species is used in the
model. Based on bootstrapping, a normal distribution with a mean of 10 and standard deviation
of 0.71, truncated at 1, was estimated for β (Figure 4) for Pogonomyrmex nests at NNSS
(Neptune, 2006).
4.7 Colony Density
Colony densities in the five Clive plots ranged from two colonies per hectare in the Juniper-Sage
habitat to 33 colonies per hectare in the mixed grassland (SWCA 2011, Table 20, p. 23). For the
initial model, the colony density will use the non-informative prior distribution and the Bayesian
posterior, meaning that for an observed count of X, the posterior distribution for the rate would
be Gamma( X, 1 ) (where the 1 is in the units of data collection, i.e. 1/ha). Expressed another
way, Bayesian statistics combines knowledge about a process generating data (in this case
colony counts) with assumptions about the process. It is reasonable to assume that the colony
counts are non-negative, making the gamma distribution more appropriate than a normal
distribution. A non-informative prior indicates that, other than the fact that counts cannot be
negative, there is no data which might suggest how the colony counts are distributed for each
location. In other circumstances, other data might be used to reduce uncertainty. In this case, the
distributions are conservative and reflect this lack of prior knowledge. Figure 3 illustrates the
shape of the distributions used to describe colony counts for each plot area.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 21
Modeling soil and contaminant transport by ant species within the Clive DU PA Model assumes
that ants move materials from lower cells to those cells above while excavating chambers and
tunnels within a nest. These chambers and tunnels are assumed to collapse over time and return
soil from upper cells back to lower cells. Through this process the balance of materials is
preserved over time. Soil and contaminant movement from one cell to another is calculated as
follows. Within each layer, the fraction of excavated ant nest volume and the fraction of
contaminants contained within that layer are determined. The fraction of contaminants within the
excavated volume is based on the ratio of the excavated volume to total volume of each layer and
is assumed to be distributed homogeneously within the layer. Secondly, the sum of contaminants
from each layer associated with the ant nest is calculated with the assumption that all excavations
from layers below are deposited in the uppermost layer. Finally, downward movement of
contaminants associated with chamber and tunnel collapse from each layer to the layer below is
calculated and the net movement of contaminants into each layer is determined. The amount of
contaminants in each layer is then used to adjust contaminant inventory in each layer for the next
time step.

Figure 3 Distribution of ant colony counts for each plot area.

0 10 20 30 40 50
0.
0

0
.
1

0
.
2

0
.
3
Colony Count (1/ha)
De
n
s
i
t
y

Plot # (mean value)
Plot 1, (33)
Plot 2, (2)
Plot 3, (7)
Plot 4, (17)
Plot 5, (6)
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 22

Figure 4. Comparison of bootstrapped and a normal distribution for Pogonomyrmex spp.
nest density with depth b parameter

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 23
5.0 Mammal Specifications and Parameters
5.1 Mammal Conceptual Model
Burrowing mammals can have a profound impact on the distribution of soil and its contents near
the soil surface. The degree to which mammals influence soil structure is dependent on the
behavioral habits of individual species. While some species account for a large volume of soil
displacement, others are less influential. This section presents the functional factors used to
parameterize the Clive DU PA Model. Factors such as burrowing depth, burrow depth
distributions, percent burrow by depth, tunnel cross-section dimension, tunnel lengths, soil
displacement by weight, soil displacement by volume and animal density per hectare play a
critical role in determining the final soil constituent mass by depth within the soil.
Modeling soil and contaminant transport by mammal species within the Clive DU PA Model
assumes animals move materials from lower cells to those cells above while excavating burrows.
Furthermore, burrows are assumed to collapse over time and return soil from upper cells back to
lower cells (Figure 5). Thus, the balance of materials is preserved through time. Calculating soil
and contaminant movement from one cell to another is straightforward. Within each layer, the
fraction of burrow volume and the fraction of contaminants contained within the burrowed
volume are determined. The fraction of contaminants within the burrowed volume is based on
the ratio of burrow volume to total volume of each layer and is assumed to be distributed
homogeneously within the layer. Secondly, the sum of contaminants from each layer associated
with burrow excavation by all animal types is calculated with the assumption that all excavations
from layers below are deposited in the uppermost layer. Finally, downward movement of
contaminants associated with burrow collapse from each layer to the layer below is calculated
and the net movement of contaminants into each layer is determined. The amount of
contaminants in each layer is then used to adjust contaminant inventory in each layer for the next
time step.
The calculations of contaminant transport due to mammal burrowing involve four steps:
1. Identify which of the mammal species overwhelmingly contribute to the rearrangement
of soils near the surface.
2. Assign these mammal species to categories and determine the excavated volumes.
3. Calculate burrow density as a function of depth for mammal categories.
4. Determine the distribution of the burrow depth fitting parameter b for mammal
categories.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 24

Figure 5. Conceptual diagram of soil movement by burrowing animals

5.2 Clive Site Surveys
Each Clive plot was surveyed for small mammal burrows during September and October 2010
(SWCA 2011). Burrows were identified by animal category, as shown in Table 13. Within the
survey area four categories of mammal burrows were identified: ground squirrels, kangaroo rats,
mice/rats/voles, and one badger. Due to the small number of badger and ground squirrel burrows,
the decision was made to treat all burrowing mammals as a single unit for modeling purposes.
Small mammal trapping was conducted on the five Clive plots during the new moon in October
2010 to identify the principal small mammal fauna present in each vegetative association. Each
1.0-ha plot was subdivided into 25 20–m × 20–m subplots. At the center of each subplot, two
Sherman® live traps were placed, for a total of 50 traps per plot. Results of the small mammal
trapping are presented in Table 14.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 25
Table 13. Summary of Clive small mammal burrow surveys
Badger Ground
Squirrel
Kangaroo
Rat
Mouse/Vole/
Rat
Total
Plot 1: Mixed Grassland 0 2 102 131 235
Plot 2: Juniper-Sage 1 0 222 16 239
Plot 3: Greasewood 0 1 1 1 3
Plot 4: Halogeton-disturbed 0 0 0 0 0
Plot 5: Shadscale-Gray Molly 0 0 0 1 1
Deer mice (Peromyscus maniculatus) were the most abundant small mammal captured during
trapping, and were the only mammal captured in the plots located on the Clive facility (Plots 3,
4, and 5). Plots 3, 4, and 5 were characterized by very low mammal densities, as evidenced by
both the trapping results and the burrow surveys. Consequently, a decision was made to average
these plots. Similar to how the ant mound density data was used to develop distributions for the
model, the resulting mammal burrow population counts were used to develop Gamma
distributions for mound density. For the Clive DU PA Model mound density is defined as
Gamma(X, 1) where X is the number of mammal mound counts for each plot.
5.3 Mound Volume
After burrow surveys were completed, soil volumes were collected in a randomly selected ¼-plot
(0.25 ha) within each plot. The obviously mounded or disturbed soil around a burrow entrance
was collected and its volume measured. This provides an estimate of the volume of soil
excavated from each burrow, with the assumption that the mounded soil represents excavations
for a single year. Results of the mound volume measurements are shown in Table 15. Based on
analysis of the data presented in Table 15, the per-mound volume is defined as a normal
distribution with a mean of 0.0006 m3/yr, and a standard deviation of 0.00015 m3/yr. Total
annual excavated volume is equal to the per mound volume multiplied by the mound density.
5.4 Maximum Burrow Depth
Maximum burrow depth was set at 200 cm based on best professional judgment. This depth is
consistent with that used at NNSS by Neptune (2005b), and represents the likely average vertical
extent of multiple badger excavations (Kennedy et al., 1985).
5.5 Burrow Density as a Function of Depth
The b parameter describes the burrow density as a function of depth, and alters the form and
volume of the excavated burrow. As the value of b increases, the fraction of burrow excavated at
each depth moves from being evenly distributed to a highly skewed distribution with most of the
excavation occurring near the soil surface. Since no belowground measurements were obtained
on mammal burrows at Clive, this version of the Clive DU PA Model uses the b parameter
derived by Neptune (2005b) for rodents at NNSS. The b parameter, defined based on analysis of
NNSS data, resulted in a parameter estimate of 4.5 and a standard error of 0.84. Badger data
were not used in the derivation of the b parameter due to the overall scarcity of badgers in the
survey area, where only one badger burrow was recorded in the five hectares surveyed across all
vegetation types.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 26
Table 14. Results of Clive small mammal trapping
Plot Date Species Count -
Species
Sum - #
Recaptured
Sum - #
Deceased
1 24 7 3

10/5/2010 4 0 0

Peromyscus maniculatus 4 0 0

10/6/2010 4 0 1

Peromyscus maniculatus 4 0 1

10/7/2010 8 3 1

Peromyscus maniculatus 6 3 1

Dipodomys microps 1 0 0

Onychomys leucogaster 1 0 0

10/8/2010 8 4 1

Peromyscus maniculatus 8 4 1
2 43 5 0

10/5/2010 7 0 0

Peromyscus maniculatus 7 0 0

10/6/2010 8 2 0

Peromyscus maniculatus 8 2 0

10/7/2010 14 0 0

Peromyscus maniculatus 10 0 0

Dipodomys microps 3 0 0

Dipodomys ordii 1 0 0

10/8/2010 14 3 0

Peromyscus maniculatus 11 3 0

Dipodomys microps 3 0 0
3 2 1 0

10/6/2010 1 0 0

Peromyscus maniculatus 1 0 0

10/7/2010 1 1 0

Peromyscus maniculatus 1 1 0
4 1 0 0

10/8/2010 1 0 0

Peromyscus maniculatus 1 0 0
5 4 1 0
10/6/2010 1 0 0
Peromyscus maniculatus 1 0 0
10/7/2010 1 0 0
Peromyscus maniculatus 1 0 0
10/8/2010 2 1 0
Peromyscus maniculatus 2 1 0
Total 74 14 3
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 27
Table 15. Soil volume (m3) of excavated mammal burrows

Plot
Burrow
ID
Number of
Burrows
Kangaroo
Rat
Mouse/Vole/Rat Badger Grand
Total
1

0.01203 0.00059

0.01262

1SW104 2 0.0035

0.0035

1SW105 1

0.00001

0.00001

1SW106 2

0.0002

0.0002

1SW107 1

0.00001

0.00001

1SW108 1 0.00005

0.00005

1SW110 1 0.00125

0.00125

1SW111 2 0.0003

0.0003

1SW112 4 0.00056

0.0006

1SW113 1

0.00003

0.00003

1SW114 1

0.00001

0.00001

1SW115 1 0.00025

0.00025

1SW116 1 0.00005

0.00005

1SW117 3 0.0025

0.0025

1SW118 4

0.00008

0.00008

1SW119 1 0.00003

0.00003

1SW120 1 0.00003

0.00003

1SW121 3 0.00009

0.00009

1SW122 2 0.00003

0.00003

1SW123 1 0.00003

0.00003

1SW124 1 0.0002

0.0002

1SW125 1 0.00015

0.00015

1SW126 1 0.0001

0.0001

1SW127 1

0.00001

0.00001

1SW128 4 0.00286

0.00286

1SW129 1 0.00005

0.00005

1SW130 1

0.00004

0.00004

1SW131 2

0.00005

0.00005

1SW132 2

0.00003

0.00003

1SW133 1

0.0001

0.0001

1SW134 1

0.00002

0.00002
2

0.037845 0.00019 0.006 0.044035

2NE002 1 0.00005

0.00005

2NE006 1

0.00001

0.00001

2NE007 1 0.00001

0.00001

2NE009 6 0.00015

0.00015

2NE010 1

0.06000

0.00006

2NE012 1 0.000225

0.000225

2NE015 1

0.006 0.006

2NE019 2 0.00135

0.00135
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 28

Plot
Burrow
ID
Number of
Burrows
Kangaroo
Rat
Mouse/Vole/Rat Badger Grand
Total

2NE020 11 0.00683

0.00683

2NE021 14 0.002975

0.002975

2NE025 1 0.00006

0.00006

2NE026 3 0.000185

0.000185

2NE027 1

0.0001

0.0001

2NE028 1 0.00005

0.00005

2NE029 1 0.0002

0.0002

2NE037 1

0.00001

0.00001

2NE040 1 0.00001

0.00001

2NE041 4 0.00004

0.00004

2NE044 1

0.00001

0.00001

2NE046 3 0.0003

0.0003

2NE048 2 0.0001

0.0001

2NE051 10 0.01501

0.01501

2NE052 3 0.0095

0.0095

2NE104 2 0.0008

0.0008
3

0.001

0.001

3NE003 1

0.001

0.001
5

0.01375

0.01375

5SW001 1

0.01375

0.01375
Grand
Total

124 0.049875 0.01553 0.006 0.071405

Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 29
6.0 References
Bethlenfalvay, G.J., and S. Dakessian. 1985. Grazing effects on mycorrhizal colonization and
floristic composition of the vegetation on a semiarid range in Northern Nevada. J. Range
Management. 37(4): 312–316.
Branson, F.A., Miller, R.F., and I.S. McQueen. 1976. Moisture relationships in twelve northern
desert shrub communities near Grand Junction, Colorado. Ecology. 57: 1104–1124.
Efron B., and Tibshirani R.J. 1998. Introduction to the Bootstrap, CRC Press, Boca Raton, FL.
Hölldobler B. and E.O. Wilson. 1990. The Ants. The Belknap Press of Harvard University Press,
Cambridge, Massachusetts. 732 pp.
IAEA, 2010. Handbook of Values for the Prediction of Radionuclide Transfer in Terrestrial and
Freshwater Environments, Technical Report Series No. 472, International Atomic Energy
Agency, Vienna, 2010.
Kennedy, W.F., L.L. Cadwell, and D.H. McKenzie. 1985. Biotic transport of radionuclide wastes
from a low-level radioactive waste site. Health Physics 49(1): 11–24.
Neptune. 2005a. Plant Parameter Specifications for the Area 5 and Area 3 RWMS Models.
Neptune and Company, Inc.
Neptune. 2005b. Mammal Parameter Specifications for the Area 5 and Area 3 RWMS Models.
Neptune and Company, Inc.
Neptune. 2006. Ant Parameter Specifications for the Area 5 and Area 3 RWMS Models.
Neptune and Company, Inc.
Pavek, Diane S. 1992. Halogeton glomeratus. In: Fire Effects Information System, [Online].
U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [2011,
February 22].
Robertson, J.H., 1983. Greasewood (Sarcobatus vermiculatus (Hook.) Torr.). Phytologia. 54(5):
309–324.
Sheppard, S.C. and W.G. Evenden. 1997. Variation in Transfer Factors for Stochastic Models:
Soil-to-Plant Transfer, Health Physics, 72: 727–33.
Smith, S.D., Monson, R.K., and J.E. Anderson. 1997. Physiological Ecology of North American
Desert Plants. Springer-Verlag, Berlin. 286 pages.
Staven L.H., Napier B.A., Rhoads K., Strenge DL. 2003. A Compendium of Transfer Factors for
Agricultural and Animal Products, Pacific Northwest National Laboratory, Richland
WA.
Biologically Induced Transport Modeling for the Clive DU PA
5 November 2015 30
SWCA Environmental Consultants. 2000. Assessment of Vegetative Impacts on LLRW. Prepared
for Envirocare of Utah, Inc. Salt Lake City, UT. 12 pages.
SWCA Environmental Consultants. 2011. Field Sampling of Biotic Turbation of Soils at the
Clive Site, Tooele County, Utah. Prepared for Energy Solutions, Salt Lake City, UT. 31 pp.
Tilley, D., Ogle, D., and L. St. John. 2008. Halogeton, Halogeton glomeratus (M. Bieb.) C.
Meyer. USDA NRCS Plant Guide. United States Department of Agriculture. Available:
http://plants.usda.gov/plantguide/pdf/pg_hagl.pdf [2011, May 11].
Zlatnik, Elena. 1999a. Hesperostipa comata. In: Fire Effects Information System, [Online]. U.S.
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [2011,
February 22].
Zlatnik, Elena. 1999b. Juniperus osteosperma. In: Fire Effects Information System, [Online].
U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [2011,
February 22].
Zlatnik, Elena. 1999c. Agropyron cristatum. In: Fire Effects Information System, [Online]. U.S.
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [2011,
February 22].
Zouhar, Kris. 2003. Bromus tectorum. In: Fire Effects Information System, [Online]. U.S.
Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire
Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [2011,
February 22].

 NAC-0017_R4

Erosion Modeling for the
Clive DU PA

Clive DU PA Model v1.4
29 October 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Erosion Modeling for the Clive DU PA
29 October 2015 ii
1. Title: Erosion Modeling for the Clive DU PA
2. Filename: Erosion Modeling v1.4.docx
3. Description: This white paper provides documentation of the development of parameter values
and distributions used for modeling erosion for the Clive DU PA Model v1.2.

Name Date
4. Originator Mike Sully 29 October 2015
5. Reviewer Dan Levitt, Kate Catlett 29 October 2015
6. Remarks

17 May 2014: Added documentation of new erosion modeling approach from the SIBERIA
modeling of the borrow pit. – M. Sully
21 May 2014: Major reorganization and accepted track change edits. – Dan Levitt
27 May 2014: Minor edits, clean up for formatting and style, redrafting of many equations. –
John Tauxe
4 July 2014: Revisions for Round 3. – M. Sully
29 Oct 2015: Removed gully screening model per PM direction. Created v1.4 – M. Sully

Erosion Modeling for the Clive DU PA
29 October 2015 iii

This page is intentionally blank, aside from this statement.
Erosion Modeling for the Clive DU PA
29 October 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

1.0
 Erosion Model Input Distribution Summary .......................................................................... 1

2.0
 Introduction ............................................................................................................................ 1

2.1
 Sheet Erosion .................................................................................................................... 1

2.2
 Gully Erosion .................................................................................................................... 2

3.0
 Evapotranspiration Cover Design ........................................................................................... 2

4.0
 Borrow Pit Model Analog ...................................................................................................... 3

4.1
 Simulation of Sheet and Channel Erosion ........................................................................ 3

4.2
 Implementation in the Clive DU PA Model ..................................................................... 5

5.0
 References ............................................................................................................................ 11

Erosion Modeling for the Clive DU PA
29 October 2015 v
FIGURES
Figure 1. Percentile depth of the area with time and fitted functions. ............................................. 6

Figure 2. The 1,000 realizations of fraction of cover area for each elevation change (depth)
interval. .......................................................................................................................... 7

Figure 3. Method for estimating gully volume from SIBERIA elevation change results. .............. 9

Figure 4. Visualization of SIBERIA model simulation of elevation change for bare soil case
for the borrow pit at 1,000 years. ................................................................................ 10

Erosion Modeling for the Clive DU PA
29 October 2015 vi
TABLES
Table 1. Summary of distributions for erosion modeling ................................................................ 1

Erosion Modeling for the Clive DU PA
29 October 2015 1
1.0 Erosion Model Input Distribution Summary
A summary of parameter values and distributions used in the erosion modeling component of the
Clive Depleted Uranium Performance Assessment Model (the Clive DU PA Model) is provided
in Table 1. Additional information on the derivation and basis for these inputs is provided in
subsequent sections of this report.
For distributions, the following notation is used:
• Discrete represents a discrete distribution of a finite number of pre-defined values.
Table 1. Summary of distributions for erosion modeling
GoldSim Model
Parameter
Units Distribution or Value Notes
FractionGully — discrete See Section 4.1
2.0 Introduction
The purpose of this white paper is to address specific details of the erosional processes that may
affect cover performance. This paper is organized to give a brief overview of erosional processes
and present the overall modeling approach, assumptions, and implementation in the Clive DU
PA Model.
Above-ground covers of waste repositories are subject to erosion by the forces of wind and
water. The proposed waste disposal cell for DU at the Clive facility, which has an engineered
above-ground cover, is subject to these erosional processes. Both wind and water erosion are
represented in the Clive DU PA Model. Details of wind erosion modeling and the effects on dose
to potential receptors are addressed in detail in the Atmospheric Transport Modeling white paper,
(Atmospheric Modeling.pdf) and are not addressed further in this white paper. Water erosion via
the return of Lake Bonneville or a small lake is not discussed in this document, but is addressed
in the Deep Time Assessment (Deep Time Assessment.pdf). Other water erosional processes are
described below.
There are two types of water erosion described in the CSM: sheet erosion and gully erosion
(channel formation). The approach used in the Clive DU PA Model to evaluate the influence of
erosion on embankment performance uses results from a landscape evolution model of a borrow
pit area at the Clive site as an analog for embankment cover erosion.
2.1 Sheet Erosion
Sheet erosion is erosion of soil particles by water flowing overland as a “sheet” in a downslope
direction. During rainfall events when rain falls faster than water can infiltrate, runoff can occur,
acting as a mechanism for eroding cover materials. Sheet erosion is a uniform process over the
area of the cover and depends largely on the steepness and shape of the slope, soil texture, and
cover characteristics, as well as rainfall intensity. This is different from erosion that flows in
defined channels (i.e., gully erosion), which is discussed in Section 2.2.
Erosion Modeling for the Clive DU PA
29 October 2015 2
In the top slope of the embankment, where slopes are gradual (about 2% slope), sheet erosion
will be slower than on the steeper side slopes of the cell (about 20% slope) (Embankment
Modeling white paper). As soil moves down slope by sheet erosion, it is likely that this material
would be replenished by deposition of clean eolian silt from the surrounding environs (i.e., a net
balance of zero change). In the end, the total soil volume on the embankment would not change,
though there would be a slow movement of soils down slope, along with the contaminants they
could potentially contain.
2.2 Gully Erosion
Gully erosion is a process that occurs when water flows in narrow channels, particularly during
heavy rainfall events. Gully erosion typically results in a gully that has an approximate “V” cross
section that widens (lateral growth) and deepens (vertical growth) through time until the gully
stabilizes. The formation of gullies is a concern on uranium mill tailings sites and other long-
term above-ground radioactive waste sites (NRC 2010). Gully erosion has the potential to move
substantial quantities of both cover materials and waste, should the waste material be buried
close to the surface. It occurs when surface water runoff becomes channeled and repeatedly
removes soil along drainage lines, creating a depositional fan of the removed materials.

The engineered cover at the Clive facility may be subject to gully erosion via a disturbance
attributed to an animal burrow, large animal tracks, the root of a fallen tree or shrub (tree throw),
or off-highway vehicle (OHV) track. It is assumed that a notch or nick will be created from these
activities at some location on the surface of the cover and the feedback processes inherent in
gully formation will cause erosion downward to the surrounding grade and erosion upward
toward the top slope of the embankment. As water flows across the inner walls of the notch,
erodible solid materials will be transported with it, creating a larger notch (both vertically and
laterally) and thus a greater capacity to remove solid material.
3.0 Evapotranspiration Cover Design
The composition of the embankment cover is an important factor in determining its erodibility.
At the Clive facility, the cover for the portion of the Federal DU Cell is an evapotranspiration
(ET) cover composed of a 6-in. thick Surface Layer of native vegetated Unit 4 material with 15
percent gravel mixture on the top slope and 50 percent gravel mixture for the side slope. The
functions of this layer are to control runoff, minimize erosion, and maximize water loss from ET.
This layer of silty clay provides storage for water accumulating from precipitation events,
enhances losses due to evaporation, and provides a rooting zone for plants that will further
decrease the water available for downward movement. Underlying the surface layer is the
Evaporative Zone Layer. This layer is also composed of Unit 4 material and is 12 in. thick. The
purpose of this layer to provide additional storage for precipitation and additional depth for plant
rooting zone to maximize ET. The Frost Protection Layer is below the Evaporative Zone Layer,
and is 18 in. thick. The purpose of this layer is to protect layers below from freeze/thaw cycles,
wetting/drying cycles, and inhibit plant, animal, or human intrusion.
Erosion Modeling for the Clive DU PA
29 October 2015 3
4.0 Borrow Pit Model Analog
A borrow pit model has been used in the Clive DU PA Model as an analog to evaluate the
influence of erosion on embankment performance. Results from landscape evolution modeling at
the Clive Site are used to project embankment cover erosion at 10,000 years. The following
sections describe the borrow pit model and the implementation of the results into the Clive DU
PA Model.
4.1 Simulation of Sheet and Channel Erosion
Landscape evolution models were developed and applied for a face of a borrow pit at the Clive
Site in order to predict the response of the pit face and upslope land surface to water erosion
processes during runoff events. The models provide a quantitative description of the evolution of
slopes and channels (also called gullies in this white paper) over time. The objective of the
models was to provide a realistic estimate of the rate of progression of hillslope erosion loss and
channel development towards the existing embankments that encase waste. Landscape evolution
models are based on the concept that, while the runoff response of a landform to rainfall depends
on the shape of the landform, the landform shape also adjusts through erosion processes acting
during the runoff event. This concept is applied by considering the interaction of hillslope
erosion processes (sheetflow) with channel growth (gully formation) process in the model
(Willgoose et al. 1991a, 1991b).
The landscape evolution model SIBERIA (Willgoose, 2005) was selected for this analysis.
Landscape evolution models such as SIBERIA capture the interaction between the runoff
response and the elevation changes of the landform surface over long time periods. This
capability makes models such as SIBERIA particularly well-suited for waste site modeling. The
model domain for the borrow pit included the borrow pit floor, a 3-m (10-ft) high pit face at a 1:1
slope and several hundred meters of ground surface upslope from the pit face at a slope of 0.3
percent. The soil was characterized with properties consistent with the Unit 4 silty clay, and had
no vegetation or rock cover.
While composed of similar soil the surface layer of the top slope of the ET cover proposed for
the Federal DU Cell has a slope of 2 percent, a gravel composition of 15 percent, and will be re-
vegetated with a mix of native and non-native species. While the cover top slope has a larger
slope of 2 percent as compared with the slope of 0.3 percent for the undisturbed area upslope
from the borrow pit face, the top slope characteristics include vegetation and gravel admix that
would act to slow erosion and channel formation. Changes in elevation at each node were
obtained at 100 y, 500 y, and 1000 y.
Assumptions for this approach include:
• The geometry of the borrow pit wall and upslope area are sufficiently similar to that of
the embankment top slope and side slope that the borrow pit serves as an analog.
• The borrow pit materials (Unit 4) are sufficiently similar to the layers of the embankment
(Unit 4 with gravel, Unit 4, and radon barrier clays).
Erosion Modeling for the Clive DU PA
29 October 2015 4
• Surface elevation changes at 10,000 y can be extrapolated from SIBERIA model results
from 100 y, 500 y and 1000 y.
• The results at 10,000 y approximate steady state of gullies. This steady state assumption
is implemented from time zero in this model.
• The area of waste that is deposited on the fan is the same as the area of waste exposed in
the gullies, using projections onto the horizontal plane.
• The excavation of ET Cover cells was not considered in the calculations below for
contaminants in the excavated mass from the gully because it was assumed that
significantly more contaminant mass was in the waste than in the cover and that the
material extracted from the waste layers would be on top of the fan.
A subset of the borrow pit model domain was selected to represent the cover. The area extended
from 50 m downslope from the edge of the embankment to 10 m upslope from the borrow pit
face. The model domain was represented by a grid with nodes at equal 0.75-m spacing. Changes
in elevation at each node were obtained at 100 y, 500 y, and 1000 y. Simulations were done for
two rainfall intensities. Since only small differences in elevation change were seen between the
two rainfall intensities, results for both intensities were combined to provide two estimates of
elevation change at each of the three times. The 0.1th, 10th, 20th, and 90th percentiles of the
simulated data were calculated at each of the three times. These percentile plots in most cases
showed a non-linear relationship between the percentile depth of the area and time.
A square-root function was fit to the 0.1th, 10th, 20th, and 90th percentiles using the general form:
𝑓𝑡=  𝐴× 𝑡+error
where
f(t) = percentile depth of the area,
A = amplitude parameter, and
t = time.

The error term was assumed to follow a normal distribution. The nls() function in the ‘stats’
package of the software program R was used to estimate the A parameter and the error term. The
four percentile plots used for the fit are shown in Figure 1.
After the percentile curves were fit using the square root function, parameters were randomly
drawn from the A distributions for three of the curves, and values of the function at 10,000 y
were calculated. For each of the 1,000 iterations, a lognormal distribution was estimated from the
resulting percentiles. The proportion of the lognormal distribution that fell within each specified
depth profile was calculated through simulation. An example iteration is included below for
demonstration purposes:
Erosion Modeling for the Clive DU PA
29 October 2015 5
1. Simulate “A” values for the 10th, 20th, and 90th percentile regression fits. The fits for the
0.1th percentile were not used for stability reasons. These values might be 1.59, 0.670,
and -0.228, respectively.

2. Project the depth value for each of the curves at 10,000 years. For the “A” parameters
above, these would be 159 mm, 67.0 mm, and -22.8 mm, respectively.

3. Fit a lognormal distribution to these projected depth fits. This step involves finding the
best geometric mean, geometric standard deviation, and shift parameter (lower bound) for
the percentiles above. Because the 10th percentile of the data is really the 90th percentile
of “depth,” the percentiles used for fitting are subtracted from 1. So for fitting purposes,
the 10th percentile of the data is the 90th percentile of the fitted distribution (and so on).

4. The parameters of the lognormal distribution with the best fit are: µ = 3.22, σ = 1.56,
θ = -26.2. One thousand values from this lognormal distribution are simulated, and the
proportion that fall within each depth range are calculated and saved to a matrix. As a
point of reference, the theoretical 10th/80th/90th percentiles of this distribution are -22.77,
67.03, and 158.8, respectively. So the lognormal distribution fits very well to these three
percentiles (see Step 2).

5. The matrix of the 1,000 iterations is output to a .csv file and converted to an MS Excel
file.

The 1,000 realizations of fraction of cover area for each elevation change (depth) interval are
shown in Figure 2. The original output file included 0.5-m depth increments from the beginning
of the waste (1.5 m) up to 10 m. It was clear that there are virtually no gullies greater than 3.5 m,
so the depth ranges were cut off there, the proportions were re-normalized, and the 3.5-m to
10-m depth ranges were deleted.
4.2 Implementation in the Clive DU PA Model
In the Clive DU PA Model, the area of the waste exposed by the gullies and the volume of the
waste removed by the gullies are used in the dose calculations. The area of waste exposed by
gullies and the resulting fan of waste from gully excavation of the disposal cell is the exposure
area for gullies. The volume of the waste removed by gullies is used to calculate a concentration
of radionuclides in the waste that was removed. This concentration of waste is assumed to be
spread out over the exposure area of the gullies and fan and is used for dose calculations.

Erosion Modeling for the Clive DU PA
29 October 2015 6

Figure 1. Percentile depth of the area with time and fitted functions.
●
●
●●
0 2000 4000 6000 8000 10000
−50
−40
−30
−20
−10

0
90th Percentile Plot
Years
De
p
t
h

(
m
m
)
Erosion Modeling for the Clive DU PA
29 October 2015 7

Figure 2. The 1,000 realizations of fraction of cover area for each elevation change (depth)
interval.

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 500 1000 1500 2000 2500 3000 3500
Fr
a
c
t
i
o
n

o
f

C
o
v
e
r

A
r
e
a
Gully
 Depth
 (mm)
Realizations
 1-­‐250
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 500 1000 1500 2000 2500 3000 3500
Fr
a
c
t
i
o
n

o
f

C
o
v
e
r

A
r
e
a
Gully
 Depth
 (mm)
Realizations
 251-­‐500
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 500 1000 1500 2000 2500 3000 3500
Fr
a
c
t
i
o
n

o
f

C
o
v
e
r

A
r
e
a
Gully
 Depth
 (mm)
Realizations
 501-­‐750
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 500 1000 1500 2000 2500 3000 3500
Fr
a
c
t
i
o
n

o
f

C
o
v
e
r

A
r
e
a
Gully
 Depth
 (mm)
Realizations
 751-­‐1000
Erosion Modeling for the Clive DU PA
29 October 2015 8
The results of the 1,000 realizations described in the Section above (MS Excel file
SimulatedErosionDepthProportions.xlsx) provide the proportion of the area of
the cover or waste layer that has a gully “end” at the defined cell depths, where a gully “end” is
defined as a cell for which the gully enters in the top of that SIBERIA cell but does not exit out
the bottom of the cell. The gully could exit on the side of the cell, but it would still be considered
to be an “end” of a gully in that cell as can be seen in the example illustrations in Figure 3 below.
The 1,000 realizations of the fraction of cover area as a function of depth for 15 depth increments
are stored in the GoldSim model as a lookup table. Each realization has an assigned index value
ranging from 1 to 1,000.
At the start of each GoldSim simulation, a value is drawn randomly from a discrete distribution
of integers ranging from 1 to 1,000. This number corresponds to the realization number in the
lookup table that will be used for the gully area–depth distribution for that GoldSim simulation.
Since the borrow pit analog showed gullies impacting only the upper 4 layers of waste, the
fraction of gully area from the lookup table for the lower 4 gully depth increments are collected
in a data element in the GoldSim model.
The area of waste exposed by gullies for each waste cell layer is calculated by multiplying the
fraction of gully area by the entire waste area for the topslope. This calculation assumes that the
fraction of borrow pit model grid cells that contain the end of a gully is the same as the fraction
of the topslope area exposed by the gully. This assumption is valid since grid spacing for the
borrow pit model is uniform.
The exposure area of the fan is assumed to be the same as the exposure area of the gullies. This
assumption is supported by output figures from the SIBERIA model such as Figure 4 (from the
file 5YR_Rainfall_1000YR_Bare.png). The fan appears to be similar in size to the area
exposed by the gully. The total exposed waste area is calculated by summing the area of the
gullies and the area of the fan. These areas are used for exposure assessment in the GoldSim
Model.
Next the volume of waste removed by gullies is calculated for each layer. The volume of waste
removed by the gully is estimated as the sum of the volume above every gully bottom or “end”
as shown in Figure 3. The volume of waste removed by the gully in each layer of the waste is the
waste removed by the cell of the gully bottom plus the sum of all of the waste removed above
that bottom cell. It is assumed that the gully is a vertical excavation straight up from the bottom
of the gully to the top. This assumption is conservative but makes the best estimate available
given the level of spatial discretization of the borrow pit modeling.
First the volume of waste removed by the gully from the lowest GoldSim cell representing the
gully is calculated by multiplying the waste cell volume by the fraction of gully area for each
layer. Then a multiplier matrix is created consisting of zeros and ones. This multiplier is used to
account for waste in layers above the layer representing the bottom of the gully. For example, the
multiplier for a gully ending in the first layer would be [1,0,0,0]. The multiplier for a gully
ending in the second layer would be [1,1,0,0], etc. Multiplying the matrix by the volume of waste
removed by the gully from the lowest GoldSim cell described above gives the total amount of
waste by layer removed by gullies.
Erosion Modeling for the Clive DU PA
29 October 2015 9

Figure 3. Method for estimating gully volume from SIBERIA elevation change results.
To clarify the fractions given in the elements FractionWasteCellsGullyEnds and FractionCapCellsGullyEnds, the illustrations
are provided below. The fraction of the cap or waste cells in which a gully "ends" was extracted from the SIBERIA output.
These fractions denote the proportion of the area in a Cap Cell or Waste Cell for which a SIBERIA modeling cell had a gully
enter (from the top) but not exit (from the bottom). The dark gray cells below are the cells that would be counted as having a
gully end. The light gray cells are removed in addition to the dark gray cells for volume of gully calculations. Note that because
we are counting discrete cells, the area and volume estimates are conservative.
Fraction of gully illustration
cell layer 1: 0/3 of the cells are counted for the fraction of gully ends;
1/3 of the cells are removed by gullies for gully volume calculations.
cell layer 2: 0/3 of the cells are counted for the fraction of gully ends;
1/3 of the cells are removed by gullies for gully volume calculations.
cell layer 3: 1/3 of the cells are counted for the fraction of gully ends;
1/3 of the cells are removed by gullies for gully volume calculations.
Example 1.
Example 2.
Gully cross-section with grids denoting cells and cell layers in Siberia. The gullies are roughly drawn in as "V"s. The fraction of
cells in which the gully "ends", is represented by the dark gray cells. The area of waste exposed by the gullies is that fraction
times the surface area of that layer. The volume of waste removed by the gully is the sum of all the cells for which the gully ends
(dark gray cells), plus the sum of the cells directly above, represented by light gray cells.
cell layer 1: 2/3 of the cells are counted for the fraction of gully ends;
3/3 of the cells are removed by gullies for gully volume calculations.
cell layer 2: 0/3 of the cells are counted for the fraction of gully ends;
1/3 of the cells are removed by gullies for gully volume calculations.
cell layer 3: 1/3 of the cells are counted for the fraction of gully ends;
1/3 of the cells are removed by gullies for gully volume calculations.
The gully now has "ends" in the first layer since the gully in those cells does not go through to the layer beneath. Thus the fraction of
cells in which the gully ends is greater than in Example 1. The volume of the waste removed by the gully similarly increases.
This wider gully now has "ends" in the second layer since the gully in those cells does not go through to the layer beneath. Thus the
fraction of cells in which the gully ends is greater than in Example 1 or 2. The volume of the waste removed by the gully increases.
Example 3
cell layer 1: 0/3 of the cells are counted for the fraction of gully ends;
3/3 of the cells are removed by gullies for gully volume calculations.
cell layer 2: 2/3 of the cells are counted for the fraction of gully ends;
3/3 of the cells are removed by gullies for gully volume calculations.
cell layer 3: 1/3 of the cells are counted for the fraction of gully ends;
1/3 of the cells are removed by gullies for gully volume calculations.
Erosion Modeling for the Clive DU PA
29 October 2015 10

Figure 4. Visualization of SIBERIA model simulation of elevation change for bare soil case
for the borrow pit at 1,000 years.
Vertical exaggeration is 18× making the pit face appear nearly vertical.

The volume removed by the gully from each waste layer is multiplied by the concentration of
each radionuclide in that waste layer to get the mass of radionuclides removed by the gully over
time. The activity mass of radionuclides per mass of soil removed is used in the dose calculations
related to gully formation. Note that gully formation in the Clive DU PA Model does not change
over time. The gully areas and volumes are fixed for a realization.

Erosion Modeling for the Clive DU PA
29 October 2015 11
5.0 References
NRC, 2010. Workshop on Engineered Barrier Performance Related to Low-Level Radioactive
Waste, Decommissioning, and Uranium Mill Tailings Facilities. Nuclear Regulatory
Commission. August 3 – 5.
Willgoose, G., et al., 1991a. A Coupled Channel Network Growth and Hillslope Evolution
Model, 1. Theory, Water Resources Research 27 (7) 1671–1684
Willgoose, G., et al., 1991b. A Coupled Channel Network Growth and Hillslope Evolution
Model, 2. Nondimensionalization and Applications, Water Resources Research 27 (7)
1685–1696
Willgoose, G. 2005. User Manual for SIBERIA (Version 8.30). Telluric Research.
http://www.telluricresearch.com/siberia_8.30_manual.pdf

 NAC-0027_R2

Dose Assessment for the
Clive DU PA

Clive DU PA Model v1.4
6 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Dose Assessment for the Clive DU PA
6 November 2015 ii
1. Title: Dose Assessment for the Clive DU PA
2. Filename: Dose Assessment v1.4.docx
3. Description: Documentation of the human exposure and dose assessment for the Clive DU PA
Model v1.4

Name Date
4. Originator M. Sully 29 October 2015
5. Reviewer D. Levitt, K. Catlett 6 November 2015
6. Remarks

6 Nov 2015: Updated v1.2 to v1.4. – D. Levitt.

Dose Assessment for the Clive DU PA
6 November 2015 iii

This page is intentionally blank, aside from this statement.
Dose Assessment for the Clive DU PA
6 November 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

1.0
 Summary of Input Parameter Values ..................................................................................... 1

2.0
 Purpose and Context ............................................................................................................... 9

3.0
 Exposure-Dose Model Implementation ................................................................................ 10

3.1
 Summary of Exposure-Dose Model Scope ..................................................................... 10

3.2
 Exposure Scenarios ......................................................................................................... 11

3.2.1
 Ranching ................................................................................................................... 13

3.2.2
 Recreation ................................................................................................................. 13

3.2.3
 Other Potential Receptors ......................................................................................... 13

3.3
 Assessment Endpoints .................................................................................................... 14

3.3.1
 Individual Dose ......................................................................................................... 14

3.3.2
 As Low As Reasonably Achievable (ALARA) ........................................................ 16

3.3.3
 Collective Dose ......................................................................................................... 17

3.4
 Modeling Doses .............................................................................................................. 19

3.4.1
 Individual Doses ....................................................................................................... 19

3.4.2
 Collective Dose ......................................................................................................... 21

3.4.3
 Dose Conversion Factors .......................................................................................... 22

3.4.4
 Additional Sources of Uncertainty ............................................................................ 27

3.4.5
 Non-Cancer Toxicity Endpoints ............................................................................... 28

4.0
 Equations and Parameters of the Exposure-Dose Container ................................................ 29

4.1
 Organization .................................................................................................................... 29

4.2
 Environmental Concentrations ....................................................................................... 30

4.3
 Exposure Parameters ....................................................................................................... 31

4.4
 DCFs ............................................................................................................................... 33

4.5
 PDCFs ............................................................................................................................. 34

4.5.1
 Inhalation PDCF Equations ...................................................................................... 34

4.5.2
 External PDCF Equations ......................................................................................... 35

4.5.3
 Ingestion PDCF Equations ........................................................................................ 35

4.6
 TEDE .............................................................................................................................. 37

4.6.1
 Inhalation TEDE Equations ...................................................................................... 37

4.6.2
 External Radiation TEDE Equations ........................................................................ 39

4.6.3
 Ingestion TEDE Equations ........................................................................................ 40

5.0
 References ............................................................................................................................ 44

Appendix I: Discussion of Derivations of Selected Parameter Distributions .............................. 49

Dose Assessment for the Clive DU PA
6 November 2015 v
FIGURES
Figure 1. Geometric mean of body weight as a function of age. ................................................... 50

Figure 2. Examples of distributions for body weight. ................................................................... 50

Figure 3. Geometric means for ventilation rate, as a function of age and gender. ........................ 51

Figure 4. Examples of ventilation rate distributions for different activities (20-year-old male). .. 52

Figure 5. Distributions for soil ingestion, representing different tracers. ..................................... 53

Figure 6. Distributions for home-produced meat ingestion rates. ................................................. 54

Figure 7. Example distributions for sedentary plus sleeping time/day and sleeping time/day
(30-year-old female). ................................................................................................... 55

Figure 8. Distributions for light, medium, and heavy activity time/day (30-year-old female). .... 56

Figure 9. Distribution for the total number of individuals at the site during a given year. ........... 57

Figure 10. Distribution for the average day-trip time. ................................................................... 58

Figure 11. Distribution for dust loading (overlaid on a histogram of simulated values). ............. 59

Figure 12. Distribution for Rancher exposure frequency. ............................................................. 60

Figure 13. Distribution for Sport OHVer exposure frequency. ..................................................... 61

Figure 14. Distribution for Hunter exposure frequency. ............................................................... 62

Figure 15. Distribution for rest area caretaker exposure frequency. ............................................. 63

Figure 16. Distributions for meat loss (preparation and post-cooking). ........................................ 64

Figure 17. Distribution for the average cattle range acreage. ........................................................ 65

Figure 18. Distribution for alpha particle REF. ............................................................................. 66

Figure 19. Distribution for electron and photon REFs. ................................................................. 67

Dose Assessment for the Clive DU PA
6 November 2015 vi
TABLES
Table 1. Exposure dose input parameters summary ........................................................................ 1

Table 2. Exposure pathways summary .......................................................................................... 12

Table 3. Beef transfer factors (Bq/kg per Bq/d) ............................................................................ 43

Dose Assessment for the Clive DU PA
6 November 2015 1
1.0 Summary of Input Parameter Values
Following is a brief summary of input values used parameters employed in the “exposure-dose”
(ED) component of the Clive Depleted Uranium (DU) Performance Assessment (PA) model that
is the subject of this white paper. See Appendix I in this document, the companion spreadsheet
Dose Assessment Appendix II, and the Model Parameters white paper (Appendix 16) for further
justifications of selected values, and the text for further explanation.
For distributions, the following notation is used:
• N( µ, σ, [min, max] ) represents a normal distribution with mean µ and standard deviation
σ, and optional truncation at the specified minimum and maximum,
• LN( GM, GSD, [min, max] ) represents a log-normal distribution with geometric mean
GM and geometric standard deviation GSD, and optional min and max,
• U( min, max ) represents a uniform distribution with lower bound min and upper bound
max,
• Beta( µ, σ, min, max ) represents a generalized beta distribution with mean µ, standard
deviation σ, minimum min, and maximum max,
• Gamma( µ, σ ) represents a gamma distribution with mean µ and standard deviation σ,
and
• TRI( min, m, max ) represents a triangular distribution with lower bound min, mode m,
and upper bound max.
Table 1. Exposure dose input parameters summary
Parameter Units Value Dependencies Source Table Notes
“Inner Loop” human exposure and dose factors; sampled multiple times within a realization
Dose conversion
factors (DCFs)
Sv/Bq;
Sv-m3 / Bq-s
Distributions for some DCFs
are derived based upon
Kocher et al, 2005 REFs (see
below). See also Dose
Assessment Appendix II.xls
EPA, 1999; and
others

Radiation
effectiveness factors
(REFs)
Unitless Alpha:
LN( 1.81e+01, 2.37+00)

Photon < 30 keV:
LN( 2.45, 1.55 )

Photon 30-250 keV:
LN( 1.96, 1.48)

Electron:
LN( 2.41, 1.44)
Kocher et al.,
2005
14, 15; p.
26
Particle- and energy-
specific values.
Based upon
lognormal fits to
percentiles presented
in Kocher et al., 2005
Uranium oral
reference dose
mg/kg-day Discrete(
0.5, 0.0006;
0.5, 0.003)
EPA, 2011;
EPA, 2000
Equal probability
assigned to Office of
Water and Superfund
criteria.
Age yr N( 25.7, 20.3 ), truncated at
16 and 60
USFS, 2005 2, p. 8
Gender Male: 60.8%
Female: 39.2%
USFS, 2005 2, p. 8
Dose Assessment for the Clive DU PA
6 November 2015 2
Parameter Units Value Dependencies Source Table Notes
Body weight kg Male:
LN( exp( 4.08+1.64e-2*Age-
1.69e-4*Age2 ), 1.24 )

Female:
LN( exp( 3.94+1.51e-2*Age-
1.51e-4*Age2 ), 1.28 )
Age, Gender EPA, 2009a 8-4, p. 8-
12;, 8-5, p.
8-13

Ventilation rate:
sleeping
m3/min-kg Male, age 16-20:
LN( 6.91e-5, 1.24 )

Male, age 21-60:
LN( exp( -9.91+4.93e-3*Age
), 1.26 )

Female, age 16-20:
LN( 6.71e-5, 1.29 )

Female, age 21-60:
LN( exp( -9.93+3.57e-3*Age
), 1.30 )
Age, Gender,
units in terms of
Body Weight
EPA, 2009a,
EPA, 2009b
6-13, p. 6-
33;, 6-14,
p. 6-35

Ventilation rate:
sedentary activity
m3/min-kg Male, age 16-20:
LN( 7.58e-5, 1.20 )

Male, age 21-60:
LN( exp( -9.82+5.14e-3*Age
), 1.19 )

Female, age 16-20:
LN( 7.37e-5, 1.23 )

Female, age 21-60:
LN( exp( -9.86+3.89e-3*Age
), 1.24 )
Age, Gender,
units in terms of
Body Weight
EPA, 2009a,
EPA, 2009b
6-13, p. 6-
33;, 6-14,
p. 6-35

Ventilation rate: light
activity
m3/min-kg Male, age 16-20:
LN( 1.77e-4, 1.18 )

Male, age 21-60:
LN( exp( -8.82+2.01e-3*Age
), 1.17 )

Female, age 16-20:
LN( 1.72e-4, 1.18 )

Female, age 21-60:
LN( exp( -8.88+2.55e-3*Age
), 1.20 )
Age, Gender,
units in terms of
Body Weight
EPA, 2009a,
EPA, 2009b
6-13, p. 6-
33;, 6-14,
p. 6-35

Ventilation rate:
moderate activity
m3/min-kg Male, age 16-20:
LN( 3.80e-4, 1.21 )

Male, age 21-60:
LN( exp( -8.02+1.93e-3*Age
), 1.25 )

Female, age 16-20:
LN( 3.56e-4, 1.21 )

Female, age 21-60:
LN( exp( -8.10+1.40e-3*Age
), 1.25 )
Age, Gender,
units in terms of
Body Weight
EPA, 2009a,
EPA, 2009b
6-13, p. 6-
34; 6-14, p.
6-36

Ventilation rate: high
activity
m3/min-kg Male, age 16-20:
LN( 6.92e-4, 1.25 )

Male, age 21-60:
LN( exp( -7.38+5.56e-4*Age
Age, Gender,
units in terms of
Body Weight
EPA, 2009a,
EPA, 2009b
6-13, p. 6-
34; 6-14, p.
6-36

Dose Assessment for the Clive DU PA
6 November 2015 3
Parameter Units Value Dependencies Source Table Notes
), 1.27 )

Female, age 16-20:
LN( 6.76e-4, 1.27 )

Female, age 21-60:
LN( exp( -7.37-4.88e-4*Age
), 1.30 )
Adult incidental soil
ingestion rate
mg/d Silicon:
LN( 12.2, 3.29 ),
truncated at 0 and 197

Aluminum:
LN( 32.7, 3.81 ),
truncated 0 and 814

Titanium:
LN( 296, 2.76 ),
truncated at 0 and 2900
Selection of
tracer element
performed
outside of the
“inner loop”
EPA, 2009a;
Davis et al,
2006.
5-11, p. 5-
37
Only study with
applicable adult data.
Truncation maxima
based upon maxima
reported in Davis et
al, 2006, as
pathological soil
ingestion is not of
interest here.
Ingestion rate: “home-
produced” beef
g/kg-d Age 16-39:
Gamma( 2.12, 1.77 )

Age 40-60:
Gamma( 1.89, 1.39 )
Age, units in
terms of Body
Weight
EPA, 2009a 13-33, p.
13-40

Ingestion rate: “home-
produced” game
g/kg-d Age 16-39:
Gamma( 0.84, 0.68 )

Age 40-60:
Gamma( 0.99, 0.83 )
Age, units in
terms of Body
Weight
EPA, 2009a 13-41, p.
13-48

Daily exposure time;
sedentary+sleeping
hr/day Males:
LN( exp( 2.79-1.55e-
2*Age+2.09e-4*Age2 ), 1.09
)

Females:
LN( exp( 2.84-1.71e-
2*Age+2.10e-4*Age2 ), 1.08
)
Truncated at 24 hr/day
Age, Gender EPA, 2009a,
EPA, 2009b
6-15, p. 6-
37
Sedentary duration
alone constructed by
subtracting sleeping
time.
Daily exposure time;
sleeping
hr/day Males:
LN( exp( 2.31-1.01e-
2*Age+1.05e-4*Age2 ), 1.06
)

Females:
LN( exp( 2.35-9.94e-
3*Age+9.94e-5*Age2 ), 1.06
)
Truncated at
Sedentary+Sleeping time
Age, Gender,
Sedentary+Sleepi
ng time
EPA, 2009a,
EPA, 2009b
6-15, p. 6-
37
Sleep duration is
excluded for daily-
use receptors.
Daily exposure time;
light activity
hr/day (un-
normalized)
Males:
LN( exp( 2.38-3.44e-
2*Age+4.05e-4*Age2 ), 1.49
)

Females:
LN( exp( 2.09-1.37e-
2*Age+1.69e-4*Age2 ), 1.34
)
Age, Gender EPA, 2009a,
EPA, 2009b
6-15, p. 6-
37
Light, moderate, and
high activities are
normalized to equal:
24 hr/day –
(sedentary + sleeping
time).
Daily exposure time;
moderate activity
hr/day (un-
normalized)
Males:
LN( exp( 1.86e-1+6.74e-
2*Age-8.16e-4*Age2 ), 1.88
)
Age, Gender EPA, 2009a,
EPA, 2009b
6-15, p. 6-
38
Light, moderate, and
high activities are
normalized to equal:
24 hr/day –
Dose Assessment for the Clive DU PA
6 November 2015 4
Parameter Units Value Dependencies Source Table Notes

Females:
LN( exp( 2.21e-1+6.49e-
2*Age-7.85e-4*Age2 ), 1.65
)
(sedentary + seeping
time).
Daily exposure time;
high activity
hr/day (un-
normalized)
Males:
LN( exp( -1.12-2.19e-
2*Age+3.14e-4*Age2 ), 3.04
)

Females:
LN( exp( -1.97+4.04e-
3*Age+6.27e-5*Age2 ), 2.84
)
Age, Gender EPA, 2009a,
EPA, 2009b
6-15, p. 6-
38
Light, moderate, and
high activities are
normalized to equal:
24 hr/day –
(sedentary + sleeping
time).
Total number of
individuals in vicinity
of site
# TRI(100, 350, 500) BLM, personal
communication
, 2010
Assumes area up to
approximately 100 sq
mi around site. This
value, minus the
number of ranchers
(see text), defines the
number of Sport
OHVers and Hunters
Number of Ranchers
in vicinity of site
# U(1, 20) BLM, personal
communication
, 2010

Number of Hunters in
vicinity of site
# Binomial( N, 0.25 ), where N
is the number of non-rancher
individuals in vicinity of site

Total number of
individuals,
number of
ranchers
USFS, 2005 22, p. 32 "Big game" hunters,
all OHV users.
Rounded to two
significant figures.
Number of Sport
OHVers in vicinity of
site
# Number(Recreationalists) -
Number(Hunter)
Total number of
individuals,
number of
ranchers and
hunters
Number of
Recreationists
defined as all
individuals minus
Ranchers.
Ranchers; day trip
time in exposure area
hr/d U(4, 12) Professional
judgment.
Sport OHVers; day
trip time in exposure
area
hr/d Beta(6.3, 2.11, 1, 20) Burr et al, 2008 21, p. 18 Utah data. Minimum
, maximum, and
standard deviation
based upon
professional
judgment. Rounded
to two significant
figures.
Hunter/Rancher;
fraction of day trip
time spent OHVing
fraction U(0.1, 0.75) Professional
judgment. OHV use
related to higher dust
concentrations in air.
All receptors; camp
trip time spent
OHVing
hr/d U(2.0, 8.0) Professional
judgment. All
overnight users
assumed to have
similar OHV use.
OHV use related to
higher dust
concentrations in air.
Exposure time;
overnight trip
hr/d 24 Professional
judgment; overnight
trip assigned a 24 hr
duration.
Dose Assessment for the Clive DU PA
6 November 2015 5
Parameter Units Value Dependencies Source Table Notes
All receptors; fraction
of camp trip exposure
time on disposal cell
fraction U(0.25, 0.75) Professional
judgment.
Corresponds to 6 to
18 hr/day. Campers
are assumed to set up
camp on the disposal
cell.
Hunter; fraction of
hunting day trip
exposure time on
disposal cell
fraction U(0.02, 0.17)

Professional
judgment.
Corresponds to 0.5 to
4 hr/day.
Rancher and Sport
OHVer; fraction of
day trip exposure time
on disposal cell
fraction Disposal cell area / Exposure
area
Assumes that
Ranchers and Sport
OHVers visiting the
area for a day trip
cover the exposure
area randomly over
the course of a year.
Rancher; exposure
frequency
d/yr Beta( 135, 34.9, 0, 180 ) BLM, personal
communication
, 2010; BLM,
2010
All leases are 6 mo.,
from November 1 to
April 30, but can be
reduced depending
upon grazing
conditions. It is
assumed that
Ranchers only work
5 days per week (i.e.
130 days per year).
distribution based
upon professional
judgment.
Sport OHVer;
exposure frequency
d/yr LN(11.3, 3.45, 1, 200) USFS, 2005 19, p. 27 Western region, "all
groups". Minimum
and and maximum
based upon
professional
judgment.
Hunter; exposure
frequency
d/yr LN(4.66, 3.45, 1, 100) USFWS, 2006 pg. 10 Utah data.
Recreationists who
are not Hunters are
defined as Sport
OHVers: # Sport
OHVers = #
Recreationists in
total - # Hunters.
Mean calculated
based upon number
of hunters and days
of hunting.
Minimum,
maximum, and
standard deviation
based upon
professional
judgment.
Ranchers; fraction of
exposure frequency
related to overnight
trips
fraction U(0.5, 0.67) BLM, personal
communication
, 2010
Corresponds to 15 –
20 day/month
overnight.
Remaining days in
ranching EF assumed
to be day trips.
Hunters; fraction of
exposure frequency
fraction U(0, 1.0) Professional
judgment.
Dose Assessment for the Clive DU PA
6 November 2015 6
Parameter Units Value Dependencies Source Table Notes
related to overnight
trips
Sport OHVers;
fraction of exposure
frequency related to
overnight trips
fraction U(0, 1.0) Professional
judgment.
Off-Site Receptor Distributions (“Inner Loop”)
Exposure frequency
rest area caretaker
d/yr TRI(327,350,365) Professional
judgment. Minimum
represents 28 days of
vacation, 10
holidays, mode is
EPA default (EPA,
1989), high is
maximum.
Exposure time rest
area caretaker
hrs/day 24 Professional
judgment (residential
receptor).
Exposure frequency I-
80 and west-side
access road traveller
d/yr U(250, 365) Professional
judgment (minimum
reflects average
number of work days
per year).
Exposure time
travelers on I-80 and
train
min/d U(2.3, 7.2) Professional
judgment. Minimum
represents 80 mph/3
miles 1-way;
maximum 50 mph/3
miles 2-way. 3 miles
represents 'densest'
part of off-site
dispersion plume.
Exposure time cars on
west-side access road
(Utah Test and
Training Range
access)
min/d U(2.4,4.0) Professional
judgment. Minimum
represents 50 mph/1
mile 2-way upper 30
mph/1 mile 2-way. 1
mile represents size
of ES property.
Knolls area Sport
OHVer; exposure
frequency
d/yr LN(11.3, 3.45, 1, 200) USFS, 2005 19, p. 27 Western region, "all
groups". Minimum
and maximum based
upon professional
judgment.
Knolls area Sport
OHVers; exposure
time
hr/d Beta(6.3, 2.11, 1, 20) Burr et al, 2008 21, p.18 Utah data. Minimum,
maximum, and
standard deviation
based upon
professional
judgment. Rounded
to two significant
figures.
“Outer Loop” human exposure factors; sampled once each model realization
Dose Assessment for the Clive DU PA
6 November 2015 7
Parameter Units Value Dependencies Source Table Notes
Receptor area
(exposure area)
acres U(16000,64000) BLM, personal
communication
, 2010; BLM,
2010
Professional
judgment. High-end
reflects area between
I-80 and UTTR,
bounded by salt flats
and Cedar Mt
foothills. Low-end
reflects Aragonite
and E. Grassy range
leases. This defines
the exposure area for
ranching and
recreational
receptors.
Meat preparation loss fraction N(0.27,0.07, 0.01, 1) EPA, 1997b 13-5 Converted from
fractions. Fraction of
meat (which is based
upon beef, uncooked
weight) lost in
preparation.
Minimum and
maximum based
upon professional
judgment.
Meat post-cooking
loss
fraction N(0.24, 0.09, 0.01, 1) EPA, 1997b 13-5 Converted from
fractions. Fraction of
meat (which is based
upon beef, uncooked
weight) lost in
preparation.
Minimum and
maximum based
upon professional
judgment.
OHV dust loading multiplier for
ambient dust
concentration
LN(98.1, 1.65) EPA, 2008 2 Activity based; i.e.
OHVs generate
increased dust.
Exposure frequency;
food
d/yr 365 EPA, 1997b Food intake rates are
annual averages.
Soil ingestion tracer
element
Discrete(0.333) Professional
judgment; equal
probability assigned
to distributions based
upon aluminum,
silicon, and titanium.
Cattle and game radionuclide uptake exposure factors (“Outer Loop”)
Cattle range area, per
operation
acres See 'outer loop'
parameter definition
for Receptor area
(exposure area).
Pronghorn range area acres U(995, 9192) Huffman, 2004 Foraging distances
for summer and
winter were equally
weighted and
assigned as diameters
of a circular home
range, from 0.1-0.8
km in the spring and
summer to 3.2-9.7
km in the fall and
winter.
Dose Assessment for the Clive DU PA
6 November 2015 8
Parameter Units Value Dependencies Source Table Notes
Cattle beef transfer
factor
Bq/kg per
Bq/d
(element-specific; see Table
3)
IAEA, 2010;
and others
Also applied to
pronghorn.
Cattle water ingestion
rate
kg/day U(33, 53) MSUE, 2011 Range of average
daily water intake for
“finishing cattle” of
weights 600 – 1200
lb is 8.6 to 14
gallons.
Cattle forage
ingestion rate
kg/day U(8.85, 14.75) EPA 2005 B-3-10, p.
B-138
Recommended value
is 11.8 kg/day; range
of +/- 25% is
professional
judgment. Value is
dry weight.
Cattle soil ingestion
rate
kg/day U(0.05, 0.95) EPA 2005 B-3-10, p.
B-139
Recommended value
is 0.5 kg/day; range
of +/- 100% is
professional
judgment.
Cattle time fraction in
exposure area
fraction Discrete(1.0) Professional
judgment. Time
grazing around the
site is presumed to be
sufficient to reach
the equilibrium
represented by
transfer factors.
Pronghorn water
ingestion rate
kg/day U(0.1, 1) UDWR, 2009 p. 4 Professional
judgment. Pronghorn
may drink no water
at all when fresh
browse is available
and up to 0.79
gal/day (3.0 L)
during dry periods.
Maximum set at 1
L/day.
Pronghorn body
weight
kg U(38, 41) Huffman, 2004
Pronghorn forage
ingestion rate
kg/day 0.577 x Body Weight
Factor0.727 x 0.001
EPA, 1993b Equation
3-9, p. 3-6
Allometric scaling
based upon body
weight for
mammalian
herbivore. Units
converted to kg/d.
Pronghorn soil
ingestion rate
kg/day U(0.005, 0.095) Professional
judgment. Set equal
to 10% of soil
ingestion distribution
for cattle based upon
body mass.
Plant ingestion screening calculations exposure factors (“Outer Loop”)
Dry-wet plant weight
conversion factor
fraction U (0.05, 0.30) EPA, 2009a 9-33, p. 9-
59
Professional
judgment. Based
upon approximate
range of moisture
contents for edible
parts of fruits and
vegetables.

Dose Assessment for the Clive DU PA
6 November 2015 9
2.0 Purpose and Context
A radioactive waste disposal facility located in Clive, Utah (the “Clive facility”) and operated by
EnergySolutions is proposed to receive and store DU and associated contaminants (called "DU
waste" here). To assess whether the proposed Clive facility location and containment
technologies are suitable for protection of human health, specific performance objectives for land
disposal of radioactive waste set forth in Utah Administrative Code (UAC) Rule R313-25-9
(Utah, 2015) must be met. In order to support the required radiological PA, a detailed computer
model has been developed to evaluate the potential future radiation doses to human receptors that
may result from the disposal of DU waste, and conversely to determine how much DU waste can
be safely disposed at the Clive facility.
The site conditions, chemical and radiological characteristics of the wastes, contaminant
transport pathways, and potential human receptors and exposure routes at the Clive facility that
are used to structure the quantitative PA model are described in the conceptual site model (CSM)
documented in the Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility
white paper (Appendix 2). The PA model has been developed as a probabilistic model taking
into account site-specific conditions and uncertainties inherent to model variables (termed
“parameters” here). The GoldSim systems analysis software (GTG, 2010) was used to construct
the probabilistic PA model. This software supports probabilistic analysis of the release and
transport of radionuclides from disposal systems. The PA model is intended to reflect the current
state of knowledge with respect to the proposed DU disposal, and to support environmental
decision making in light of inherent uncertainties.
The dynamic aspects of the PA model may be grouped into two domains. The contaminant
transport (CT) component of the PA model encompasses the release of contaminants from
disposed wastes and subsequent migration through the environment. The output of the CT
component (documented in other white papers) is a time series of contaminant concentrations in
different environmental media. These concentrations serve as inputs to the exposure-dose (ED)
component of the PA model that is the subject of this white paper. Because the ED component of
the PA model is organized within a single “container” in GoldSim, the terms ED model and ED
container are used interchangeably.
Assumptions and mathematical equations describing contaminant intake, including external
exposure to ionizing radiation, for each exposure scenario are provided here. Equations for
estimating radionuclide dose, and non-carcinogenic toxicity associated with uranium, are also
provided. The implementation of methods for evaluating uncertainty in the ED calculations are
also described. The bases of the deterministic values and/or statistical distributions for each of
the ED parameters are discussed in the text below, the attached Appendix I, the spreadsheet Dose
Assessment Appendix II, and the Model Parameters white paper (Appendix 16).

Dose Assessment for the Clive DU PA
6 November 2015 10
3.0 Exposure-Dose Model Implementation
3.1 Summary of Exposure-Dose Model Scope
The ED container addresses potential radiation exposure, dose and non-carcinogenic toxicity to
human receptors who may come in contact with contaminants released from the disposal facility
into the environment subsequent to facility closure. Radiation dose limits for protection of the
general population are defined in UAC Rule R313-25-9 (Utah, 2015), and in 10 CFR 61.41
(CFR, 2007). These dose limits implicitly assume a level of health risk (discussed further below).
The regulations specify that design, operation, and closure of the land disposal facility must also
ensure protection of individuals inadvertently intruding into the disposal site and occupying the
site or contacting the waste at any time after loss of active institutional control (e.g., fences,
guards, etc.) of the site. Because the definition of inadvertent human intruders (IHI) encompasses
exposure of individuals who engage in normal activities without knowing that they are receiving
radiation exposure, there is no practical distinction made between a member of the public (MOP)
and IHI with regard to receptors and dose calculations.
The UAC Rule R313-25-9 (Utah, 2015) requires a PA for DU to have a minimum compliance
period of 10,000 years, with additional simulations for a “qualitative analysis” (i.e., one in which
only contaminant migration, and not doses, are modeled) for the period where peak hypothetical
dose occurs. The estimation of doses in such long time horizons would be speculative at best, but
if total radioactivity is used for a proxy (accounting for radiological decay and ingrowth from the
disposed DU), then a peak value would occur once the progeny of U-238 have reached secular
equilibrium in about 2.5 million years. With respect to radiation dose and non-carcinogenic
uranium toxicity, the ED container quantifies dose only within the regulatory time frame of
10,000 yr. This approach is consistent with the requirements of UAC R313-25-9 (Utah, 2015).
No specific time frame is defined in 10 CFR 61 (CFR, 2007) for the exposure/dose assessment.
Key land use characteristics of the Clive facility that pertain to the development of receptor
scenarios and dose modeling are summarized in the CSM (Appendix 2) and in the Features,
Events, and Processes (FEPs) Analysis for Disposal of Depleted Uranium at the Clive Facility
white paper (Appendix 1). Current human use of the area surrounding the Clive facility is very
limited. Note that a residential scenario is not evaluated here, as there is no evidence that humans
have permanently resided at the immediate Clive facility environs in recent history (see CSM).
The closest current dwelling is approximately 12 km to the northeast of the site (a caretaker at
the Aragonite/Grassy Mountain rest stop on east-bound Interstate-80).
Rancher and recreationist scenarios for the area surrounding the Clive facility are conditioned
only on a continuation of present-day land use, whereas the conditions related to other scenarios
would be much more speculative. It is not possible to project changes in human biology, society,
technology, or behavior over a 10,000 year time frame; thus, current land use characteristics are
projected throughout this period of performance, as recommended in NRC (2000). Uncertainty
associated with this assumption is not quantified at this time. However, general justifications for
this assumption in addition to NRC guidance can be made. The Clive facility environs are
currently not amenable to permanent habitation due to the lack of potable groundwater and other
factors. Dramatic changes in climate, such as large increases in average annual temperature or
decreases in precipitation, would make the site even less hospitable. Changes in the opposite
Dose Assessment for the Clive DU PA
6 November 2015 11
direction; i.e., large decreases in average annual temperature or increases in precipitation, have
historically only been associated with ice ages and thus again would result in the site becoming
less hospitable than it is today (see the CSM). Therefore, the assumption that future land use and
receptors will be similar to today's is likely conservative (i.e., protective).
It is possible that the Clive facility disposal cap could become more amenable to plant cover and
perhaps increased human use than the surrounding areas post-closure due to the presence of the
rip rap cover (e.g., in terms of accumulation of aeolian or wind-borne soil and dust and lower
evaporation rates from soil below the rip rap). Nearby areas hosting vegetation (e.g., the alluvial
fan of the Cedar Mountains east of the Clive facility, rocky outcrops west of the site) thus
potentially offer analogous sites that will be considered for characterizing potential future plant
communities on the disposal cap.
3.2 Exposure Scenarios
Based upon current and reasonably anticipated future land uses as summarized above, and as
described in the FEP analysis (Appendix 1), two future use exposure scenarios were identified
for inclusion in the ED model: ranching and recreation. After institutional controls are no longer
maintained, exposures to contamination in the ranching and recreation scenarios could occur
both on the Clive facility site as well as nearby off-site locations.
Modeling of ranching and recreation scenarios is discussed here. Exposure scenarios are defined
according to various human activities, which may result in a complete exposure pathway existing
between the contaminant source and receptors. Exposure pathways describe the media, activities
and exposure routes by which contamination becomes available to human receptors in the
exposure scenarios. Every complete exposure pathway contains the following elements (EPA,
1989):
• Known or potential sources and/or releases of contamination;
• Contaminant transport pathways;
• Potential exposure media;
• A point of potential receptor contact with the impacted medium; and,
• An exposure route (such as ingestion or inhalation).
The primary exposure routes for the ranching and recreation scenarios include ingestion,
inhalation, and external irradiation. A summary of potentially complete exposure pathways for
each scenario is provided in Table 2. Figure 10 in the CSM (Appendix 2) depicts the transport
mechanisms by which contaminants in the disposed waste may reach the exposure media
discussed in this section.

Dose Assessment for the Clive DU PA
6 November 2015 12
Table 2. Exposure pathways summary
Exposure Pathway Ranching Recreation
Inhalation (wind derived dust) × ×
Inhalation (mechanically-generated dust) × ×
Inhalation (gas phase radionuclides) × ×
Ingestion of surface soils (inadvertent) × ×
Ingestion of game meat ×
Ingestion of beef ×
Ingestion of wild plant material ×* ×*
Ingestion of seasonal surface water ×* ×*
External irradiation – soil × ×
External irradiation – immersion in air × ×
*Not included in the ranching or recreation scenarios; see text.
Note that a single individual could potentially engage in both ranching and recreation in the same
area, but these scenarios are modeled separately because they are expected to be distinct.
Groundwater ingestion is not directly evaluated in the ED model, although groundwater concen-
trations are compared to State of Utah Ground Water Protection Levels (GWPLs). As described
in the CSM (Appendix 2), the aquifers underlying the area are more saline than seawater, and
would not be potable without extensive desalinization. This situation is unlikely to change under
any foreseeable conditions that would allow human habitation in the vicinity of the facility.
It is possible that humans may be exposed by ingestion of native plants. Several plants identified
in Clive area vegetation plots were historically used as traditional food or medicine. These
include shadscale saltbrush (Atriplex confertifolia), black greasewood (Sarcobatus vermiculatus),
and rockcress (Arabis sp.), among others. However, present-day use of these plants by potential
receptors in the area is unknown. In the absence of such information for plant uses and quantities
thereof, a screening-level calculation will be performed to determine what quantity of plant
material from the disposal cap would need to be consumed to exceed the radiation dose
performance objective.
A second possible exposure pathway not directly assessed in the ranching and recreation
scenarios is human ingestion of intermittent (seasonal) surface water from puddles that may form
in the air dispersion area. This surface water is likely to be salty, due to the saline nature of soils
adjacent to the Clive facility, and direct human exposure is considered to be unlikely. Although
present-day use of surface water by potential receptors in the area is unknown, a screening-level
calculation will be performed to determine what volume of water would need to be consumed to
exceed the radiation dose performance objective.
Dose Assessment for the Clive DU PA
6 November 2015 13
3.2.1 Ranching
The land surrounding the Clive facility is currently utilized for cattle and sheep grazing (BLM,
2010). Livestock apparently utilize the area more during winter periods when snow is present
and when puddles exist during wet periods (NRC, 1993). The Bureau of Land Management
(BLM) currently issues leases for 6 months of the year (November 1 to April 30; BLM, 2010,
personal communication: Salt Lake Field Office). The personnel who spend time with the herds
in the field are called "Ranchers" here (although this may include a variety of job classifications).
Activities are expected to include herding, maintenance of fencing and other infrastructure, and
assistance in calving and weaning. Ranchers may be exposed to contamination via the routes
outlined in Table 1. It is assumed that any future ranching-related structures that might be
constructed will be rough-built, with sufficient air flow that indoor radon accumulation is not an
issue.
Ranchers typically use off-highway vehicles (OHVs; including four-wheel drive trucks) for
transport. Beef consumption (from cattle exposed to contamination released from the site), is
evaluated for the Ranchers, assuming that they may consume some of their own product. Beef,
rather than lamb or mutton, is used as a food in the ED ranching scenario because regulatory
bodies such as EPA (2005) and others have published information related to modeling of tissue
concentrations for cattle.
3.2.2 Recreation
The recreational exposure scenario could potentially encompass a variety of activities.
Information is limited regarding current use, as the BLM, the manager of much of the
surrounding land, does not specifically track recreational usage in the area. However, based upon
discussions with the BLM and reasonable judgment regarding anticipated land use, recreation
may involve OHV use, hunting, target shooting of inanimate objects, rock-hounding, wild-horse
viewing, and limited camping.
The desirability of recreational activities on or around the disposal units, similar to suitability for
ranching, is partially dependent upon assumptions regarding ecological succession on the
disposal unit over time. With the possible exceptions of OHV use and as a vantage for hunting
(e.g., for pronghorn), recreational use of the disposal unit in an as-closed state is likely to be
minimal. As plant succession proceeds the disposal unit may become more attractive for
different types of recreational activities. However, for the purpose of exposure assessment, it is
assumed that sport OHV riders ("Sport OHVers; i.e., OHV users who use their vehicles for
recreation alone) and hunters using OHVs ("Hunters"), both of whom may also camp at the site,
would represent the most highly-exposed receptors (due to exposure to mechanically-generated
dust, game meat ingestion, etc.), and other types of recreationists would have lower exposures.
3.2.3 Other Potential Receptors
The ranching and recreation scenarios are characterized by potential exposure related to activities
both on the disposal site and in the adjoining area. Specific off-site points of potential exposure
also exist for other receptors based upon present-day conditions and infrastructure. These
locations and receptors include:
Dose Assessment for the Clive DU PA
6 November 2015 14
• Travelers on Interstate-80, which passes 4 km to the north of the site;
• Travelers on the main east-west rail line, which passes 2 km to the north of the site;
• Workers at the Utah Test and Training Range (a military facility) to the south of the
Clive facility, who may occasionally drive on a gravel road immediately to the west
of the Clive facility fenceline;
• The resident caretaker at the east-bound Interstate-80 rest facility (the Grassy
Mountain Rest Area at Aragonite) approximately 12 km northeast of the site, and,
• Sport OHV enthusiasts at the Knolls OHV area (BLM land that is specifically
managed for OHV recreation) 12 km to the west of the site.
Exposure to individuals at these off-site locations is expected to be minimal due to either the
large distance from the site (Interstate-80 rest area and Knolls OHV area) or because the
exposure time for any individual will be very brief (travelers on road, rail, and highway). Unlike
ranching and recreational receptors who may be exposed by a variety of pathways on or adjacent
to the site, these off-site receptors would likely only be exposed to wind-dispersed
contamination, for which inhalation exposures are likely to predominate. These receptors will be
evaluated to determine whether exposures at these off-site locations may be important.
3.3 Assessment Endpoints
The biological effect of greatest interest to regulatory agencies for environmental exposure to
radionuclides is cancer. Ionizing radiation is a clear cause of cancer and other health effects at
high doses. However, the risk of cancer to an individual exposed to radiation at environmental
levels is highly uncertain and depends upon a large number of assumptions, the most influential
being: 1) That the major source of data for radiological risk assessment; i.e., the high doses
experienced by the Hiroshima/Nagasaki atomic bomb victims in World War II, is relevant for the
much lower doses in the range of regulatory dose limits; and, 2) that risks can be extrapolated
from large doses to small doses in a linear fashion, with no threshold of effect (i.e., the
hypothesis that no dose is without some risk of cancer) (Brenner et al., 2003). Both of these
assumptions are controversial (Scott, 2008), but they provide substantive bases for NRC and
DOE radiation regulation and guidance at this time. Uncertainty associated with these
assumptions is not evaluated in the PA model at this time.
3.3.1 Individual Dose
There are two performance goals that may be applicable in the PA. The first is the individual
dose limit. Title 10 CFR 61.41 (CFR, 2007) specifies assessment endpoints for a radiological PA
that are related to annual radiation dose. The specific metrics described in §61.41 are organ-
specific doses, and restrict the annual dose to an equivalent of 0.25 mSv (25 mrem) to the whole
body, 0.75 mSv (75 mrem) to the thyroid, and 0.25 mSv (25 mrem) to any other organ. As
described below, the ED model will employ a total effective dose equivalent (TEDE) for
comparison with the 0.25 mSv/yr threshold. This dose level will be considered as a deterministic
performance goal, with no uncertainty.
Dose Assessment for the Clive DU PA
6 November 2015 15
As discussed in Section 3.3.7.1.2 of NUREG-1573 (NRC, 2000), the radiation dosimetry
underlying the §61.41 dose metrics was based upon a methodology published by the
International Commission on Radiation Protection (ICRP) in 1959. Subsequent to Title
10 CFR 61.41, more recent dose assessment methodology has been published by the ICRP
(ICRP, 1979; 1991; 1995) that employs the TEDE approach. The TEDE uses weighting factors
related to the radiosensitivity of each target organ to arrive at an effective dose equivalent across
all organs. The text of Section 3.3.7.1.2 of NUREG-1573 (NRC, 2000) states:
As a matter of policy, the Commission considers 0.25 mSv/year (25 mrem/year) TEDE as
the appropriate dose limit to compare with the range of potential doses represented by
the older limits... Applicants do not need to consider organ doses individually because
the low value of TEDE should ensure that no organ dose will exceed 0.50 mSv/year (50
mrem/year).
The regulations state that this dose limit is applicable to any member of the public, yet NRC PA
guidance (NRC, 2000) suggests a practical approach of applying the dose limit to an average
member of a "critical group" (i.e., a group of public receptors who might be reasonably expected
to live near or experience exposure to the facility site). The ED model has been developed to
support estimates of both average individual dose and various percentiles of the distribution of
the mean individual dose for Ranchers, Sport OHVers, and Hunters at any model year of a
simulation.
Thus, in terms of PA performance objectives, the modeling question relates to estimating the
probability that the total radiation dose attributable to future releases from the site to any or an
average member of a critical group (defined here as a Rancher, Sport OHVer, or Hunter) will
exceed 25 mrem TEDE in any particular year, during the performance period of the site. As
institutional controls in place while the site is operating are designed to prevent public access,
there will be no public exposure during this time period. The period of time of interest, therefore,
in the ED portion of the PA model is from the time of loss of institutional control to 10,000 years
post-closure, although physical transport processes are evaluated beginning at model year zero.
The US Environmental Protection Agency (EPA) has estimated that 15 mrem/year is equivalent
to a 3-in-10,000 excess risk of cancer (EPA, 1997a), and has defined that level as:
...consistent with levels generally considered protective in other governmental
actions, particularly regulations and guidance developed by EPA in other
radiation control programs.
A 1- in-1-million excess risk level is typically viewed as a de minimus level; i.e. one that is
below a level of concern (CFR, 1994). If the estimated EPA risk equivalence for 15 mrem/year is
extrapolated to 1- in-1-million, this results in a 0.05 mrem/year de minimus dose. This is
potentially important both when evaluating the dose to any receptor and when collective dose is
assessed (discussed below).

Dose Assessment for the Clive DU PA
6 November 2015 16
3.3.2 As Low As Reasonably Achievable (ALARA)
A second decision rule pertains to the ALARA concept. Ionizing radiation protection limits have
been utilized since the 1920s (Hendee and Edwards, 1987). These limits have changed over time
as more information regarding the negative biological effects of radiation has become available
(especially after World War II). Concurrently, therapeutic and diagnostic (i.e., beneficial) uses of
radiation have increased dramatically. Radiation in high doses kills cells, which can be harmful
or beneficial to the receptor of the doses (e.g., in the latter case, targeted radiation is used to kill
cancer cells). The effects of low doses of radiation are more uncertain. There is ample evidence
that ionizing radiation can damage DNA and enhance cell proliferation in doses below those that
kill cells, and thus can potentially cause cancer. However, it is uncertain at dose this becomes a
concern.
For many years, there has been a presumption in radiation protection, based upon statistical
analysis of animal and human data, that ionizing radiation has a linear dose-response curve at
low doses and that there is no threshold of effect; i.e. any dose of radiation can result in an
increased probability of cancer (this is termed the linear no-threshold, or LNT, hypothesis). This
is not supported by all experimental and clinical observation (Scott, 2008) and multiple highly-
efficient molecular and cellular defense and repair mechanisms for radiation damage exist.
Regardless, this LNT hypothesis is the basis for most regulatory standards today, and indeed for
the ALARA concept.
ALARA (or the older but similar concept "as low as practicable"; ALAP) essentially assumes no
carcinogenic threshold of radiation carcinogenesis. If this assumption is taken at face value,
ALARA seems to be a reasonable objective. If not, then a threshold of effect would be a more
tractable and achievable objective. ALARA could perhaps be applied even in the case of a
threshold or 'target' concentration; the threshold would simply be a limit on the amount of risk
reduction that should be achieved by a particular management alternative. Proper evaluation of
uncertainty associated with the LNT hypothesis would be a large task in itself, but the influence
of a LNT assumption can still in principle be evaluated using sensitivity analysis.
A different sort of threshold exists with regard to natural background levels of radiation. The
doses that the public receives from all environmental sources (e.g., local geology, extraterrestrial,
etc.) can be quite variable. For example, population X who live at high altitude in a location with
geologically high levels of uranium may have a much higher level of annual exposure than
population Y who live at sea level with low levels of uranium in soil (e.g., see
http://www.epa.gov/radon/zonemap.html). If population sizes were equivalent, one could then
consider that a larger incremental dose might be acceptable for population Y compared to
population X.
Uranium and many other metals are also associated with non-radiological toxicity; e.g. kidney or
liver damage. In such cases, toxicology has developed concepts such as the reference dose and
benchmark dose, to account for the clear thresholds of effect that are associated with non-
carcinogenic toxicity (Filipsson, 2003). Similar to the discussion above, in these cases the
threshold can be viewed as a target, below which risks are not of substantial concern.
Dose Assessment for the Clive DU PA
6 November 2015 17
The modern ALARA concept, as germane to radiation protection on both individual and
population levels, was described by the ICRP in 1977 (ICRP, 1977):
Most decisions about human activities are based on an implicit form of balancing of
costs and benefits leading to the conclusion that the conduct of a chosen practice is
'worthwhile.' Less generally, it is also recognized that the conduct of the chosen practice
should be adjusted to maximize the benefit to the individual or to society. In radiation
protection, it is becoming possible to formalize these broad decision-making procedures.
The ICRP (1977) basically recommended a system of radiation protection that included the
following principles:
• No practice shall be adopted unless its introduction produces a positive net benefit;
• All exposures shall be kept as low as reasonably achievable, economic and social
factors being taken into account; and,
• The dose equivalent to individuals shall not exceed the limits recommended for the
appropriate circumstances by the Commission.
• These three components are identified by the ICRP by the abbreviated terms:
• The justification of the practice;
• The optimization of radiation protection; and,
• The limits of individual dose equivalent.
For present purposes, as regulatory agencies have adopted and applied clear dose limits for
individuals, evaluation of ALARA here will be restricted to population doses, termed collective
dose. This is appropriate in the context of design and siting of radioactive waste facilities; as it is
likely, if any substantial future risks occur, that health concerns will be at a population level.
Further, we assume that facility workers will be protected under existing health and safety
regulations and guidance, and will not be evaluated here.
ICRP 101b (2006) describes updates to previous ICRP publications addressing ALARA. Section
3.3.3 discusses calculation of collective dose in the context of this publication.
3.3.3 Collective Dose
In order to estimate collective dose, a population needs to be assessed. If cumulative doses are to
be estimated over some period of time, then the doses are added over that time period. The
collective dose at the end of the performance period (10,000 years post-closure, in this case) is
then the individual annual doses added up over a period of 10,000 years (minus the period of
time when institutional controls are in place).

Dose Assessment for the Clive DU PA
6 November 2015 18
For a hypothetical example, say a total population of 50 people is potentially exposed to the site
for every year during the performance period (note that all radioactive waste repositories that
have been recently evaluated in the US are in fairly remote areas, so a large urban population
would be inappropriate). Say institutional controls are in place for 100 years. Then, the
cumulative population dose will be the sum of 50 individual doses in mrem/year, multiplied by
9,900 years. Say that every person in the population is exposed just below the individual dose
limit (say, 24 mrem/year TEDE). Thus, the cumulative population dose will be
50×24×9900=11,880,000 mrem, or 11,880 person-rem. This number has no meaning by itself, as
there is no standard or basis for declaring this is 'unacceptable' or not, or whether it is
"reasonable" or "achievable" (according to ALARA). It is only useful in the context of
comparing how one site or disposal option might perform compared to another. This is best
determined in the context of a decision or economic analysis, which is discussed in the Decision
Analysis (ALARA) white paper (Appendix 12).
In lieu of guidance that defines what an 'acceptable' population dose might be, a means must be
applied so that all populations (e.g., the entire United States) are not assessed, as this would be
burdensome and meaningless. For instance, it is known that a large population will indeed be
exposed to the site if current conditions continue; i.e., the population of drivers on Interstate-80.
However, as previously mentioned, each of these drivers would be exposed for very short
periods of time. Furthermore, the exposure levels would be a small fraction of those experienced
by the Ranching and Recreation receptors described in Section 3.2.
In order to gauge the importance of quantifying dose for this population, and indeed any remote
population that might be exposed for brief periods and/or to very low concentrations, a de
minimus risk approach will be considered. As explained previously, according to the EPA a 0.05
mrem/year dose corresponds to approximately a 1-in-1-million excess cancer risk. Individual
doses for receptors other than Ranchers, Sport OHVers, or Hunters will be evaluated relative to
this individual dose threshold to determine whether doses to remote receptors should be
considered when computing collective dose. Cumulative population dose will not include
contributions from remote receptors if individual doses for these receptors are far below 0.05
mrem/year.
Note that NRC was required under Section 10 of the Low-Level Waste Policy Amendments Act
of 1985 to “establish standards for determining when radionuclides in waste streams were in
sufficiently low concentrations or quantities as to be below regulatory concern, thereby
potentially exempting them from NRC Low-Level Waste regulation” (NRC, 2007; NUREG-
1853, Section 3.5). The de minimus risk level discussed above is in no way related to
establishing concentrations or quantities “below regulatory concern” in disposed waste. Rather,
this level is employed to support a methodology for meaningful evaluation of collective radiation
dose in relation to the ALARA assessment endpoint of the Performance Assessment.

Dose Assessment for the Clive DU PA
6 November 2015 19
3.4 Modeling Doses
3.4.1 Individual Doses
Studies of the health of existing populations (i.e., epidemiological studies) have struggled with
how to infer individual risk from population statistics. For example, a study of cigarette smokers
and lung cancer may show a clear statistical relationship between the exposure and disease, with
a high degree of confidence; yet, for instance, it does not tell me what my additional risk of
cancer will be if I smoke one cigarette. It is indeed impossible to directly estimate health risk for
individuals for the majority of exogenous exposures (there are exceptions in the case of some
genetic abnormalities; if the abnormality is known to exist in an individual, then the risk of
disease in that individual associated with that abnormality is known with almost perfect
confidence). Risk for individuals must generally be inferred from populations. In addition to
various designs of epidemiological studies, insurance companies, for example, use life tables
stratified on gender, age, disease history, etc. to estimate premiums.
In the present case, the issue is estimation of individual radiation doses. As mentioned above,
risk is implicit in radiation dose, with many inherent assumptions. Additionally, the PA is
projecting into the future, to individuals who do not exist yet. As information as to how humans
may or may not change biologically in the space of a 10,000-year performance period does not
exist, it is only reasonable to assume that humans will remain essentially the same.
One approach to estimating individual risk, based upon how the EPA has historically conducted
exposure assessment (EPA, 1989), is to define a 'simulated' individual based upon their exposure
characteristics. The simulated individual is therefore the product of a number of physiological
and behavioral parameters. Historically, this has been done deterministically; i.e., single values
are used for the exposure and physiological parameters, and a single simulated individual results.
With more recent applications of probabilistic methods, this process has been expanded to
address variance in the exposure parameter values.
For the Clive facility, following are some major sources of variance related to radiation dose that
are directly germane to the ED model at any particular point during the assessment time horizon:
1. The number of receptors, if any, in the vicinity of the disposal site at any point in time;
2. The physiological characteristics of the receptors;
3. The nature and intensity of exposure by various potential exposure routes (ingestion,
inhalation, external radiation) based upon behavioral characteristics of the receptors;
4. The concentrations of radionuclides in potential exposure media; and,
5. The annual radiation dose associated with the exposure.
Within some of these five categories there may be multiple exposure parameters employed in the
modeling and hence numerous sources of variance. In particular, radionuclide concentrations in
exposure media include all the variance from the contaminant transport modeling conducted in
the PA that are propagated to the ED assessment.
Dose Assessment for the Clive DU PA
6 November 2015 20
As discussed above, the PA guidance (NRC, 2000) suggests that the annual dose to an "average
member of a critical group" should be estimated. Specifically:
The average member of the critical group is that individual who is assumed to represent
the most likely exposure situation, based on cautious but reasonable exposure
assumptions and parameter values. It is generally not practicable, when analyzing future
potential doses, to calculate individual doses for each member of a critical group and
then re-calculate the average dose to these same members. In general, it is more
meaningful to designate a single hypothetical individual, representative of that critical
group, who has habits and characteristics equal to the mean value of the various
parameter ranges that define the critical group. In this fashion, the dose to the "average
member" of the critical group approximates the average dose obtained if each member of
the critical group were separately modeled and the results averaged.
Thus, the guidance appears to request definition of:
• A critical group;
• An average member of the critical group; and,
• The annual dose to this member.
The critical groups, in the case of the present PA, are defined as Ranchers, Sport OHVers, and
Hunters. An "average member" of these groups is a theoretical or statistical construct, as such a
person does not and never will exist. Thus, we can interpret the guidance as referring to the
statistical average dose (i.e., arithmetic mean) of a population of individuals' doses. In order to
estimate the average simulated individual's dose at a particular time step, doses to a population of
simulated individuals need to be estimated (note that hardware and software capabilities have
increased dramatically since the NRC's guidance, so it is indeed now possible to calculate doses
at an individual level).
In the context of human health risk assessment, variance in parameter values is traditionally split
into the categories of variability and uncertainty (EPA, 2001). The term variability refers to
natural, irreducible variance in the range of values a parameter may take (say, body weights in a
population), and uncertainty refers to incomplete, imprecise and/or inaccurate knowledge
associated with parameter values (Bogen et al., 2009). These particular definitions are not
universally accepted however, and in practice may have more or less utility as a basis for the
methodology used to assess overall variance in model output.
Returning to the issue of doses to a population of simulated individuals, and to the five major
sources of variance for these dose estimates, the first 3 sources of variance apply to population
variability. In particular, in any year the physiological and behavioral characteristics of the
exposed individuals govern the degree of variance related to sources #2 and #3. The variance
related to parameters contributing to exposure concentrations and to radiation dose coefficients
do not vary over time and do not vary for different hypothetical individuals. For example,
models of carcinogenesis for low-dose radiation are highly uncertain, but this uncertainty does
not appreciably differ among individuals nor does it vary from one model year to another.
Similarly, we assume essentially static environmental conditions over the 10,000-year
performance period for any given model realization; a soil-water distribution coefficient that
applies at model year 2,000 also applies at model year 3,000.
Dose Assessment for the Clive DU PA
6 November 2015 21
There are multiple methods that may be employed to model two different types of variance, but a
typical method is termed 2-dimensional (2D) or nested-loop Monte Carlo simulation (Bogen et
al., 2009). In the ED model, the exposure parameters are grouped into long-term model
uncertainty and population variability categories. The physiological and behavioral parameters
related to sources #2 and #3, as well as the number of individuals exposed in any year (source
#1), are evaluated annually in the “inner loop” of the 2D Monte Carlo simulation. The remainder
of the model parameters, including all aspects of the Contaminant Transport modeling and the
radiation dose conversion factors (DCFs) are defined in the “outer loop” of the 2D Monte Carlo
simulation. This categorization is further discussed below.
3.4.2 Collective Dose
As described above, an issue of ALARA interest is the collective dose over the performance
period. To reiterate, this estimate is of little value in itself as there are no performance objectives
for this endpoint; rather, it should ideally be viewed in the context of decision analysis.
Estimating population dose is simple. It is the sum of individual annual doses over the period of
time from loss of institutional control to the 10,000 year mark. Contributions from off-site
receptors who are anticipated to have very low annual dose rates will only be included in the
collective dose sum if individual doses are approaching a 0.05 mrem/yr threshold (equivalent to
approximately a 1-in-1-million excess cancer risk).
The calculation of collective dose is consistent with recommendations of the ICRP (2006). For
example, the PA’s methodology specifically addresses the following characteristics of the popu-
lation (ICRP, 2006; Table 3.1):
• Gender
• Age
• Habits
• Characteristics of the exposure
• Distribution of exposures in time and space
• Number of individuals
• Minimum individual dose
• Maximum individual dose
• Mean individual dose
• Statistical deviations
• Collective dose associated with ranges of individual doses.
Dose Assessment for the Clive DU PA
6 November 2015 22
3.4.3 Dose Conversion Factors
For both individual doses and population doses, exposures or intakes are converted to TEDEs via
DCFs, or dose equivalents per unit intake. DCFs have been published by EPA and ICRP. Section
3.3.7.3 of NUREG-1573 specifies DCFs published by EPA in Federal Guidance Reports (FGR)
11 (EPA, 1988) and 12 (EPA, 1993a). EPA subsequently made use of age-specific DCFs
published in ICRP Publication 72 (ICRP, 1995) to estimate radionuclide cancer risk coefficients
in FGR 13 (EPA, 1999). The DCFs published in EPA (1999) are used in the dose assessment and
are available online (http://ordose.ornl.gov/downloads.html). The radionuclide-specific DCFs
used in the dose assessment are also provided in the spreadsheet Dose Assessment Appendix II.
DCFs are derived using models and data that represent the physics and biology of the interaction
of the human body with radiation or radioactive material. Briefly, internal DCFs (typically in
units of Sv/Bq) are used to convert from an exposure or intake to an internal dose delivered to
target organs. DCFs are radionuclide, receptor-age, and exposure-route dependent (external,
inhalation, or ingestion). In addition, separate inhalation dose coefficients are published for
different lung absorption rate classes. For external exposure the dose coefficient depends upon
whether the receptor is immersed in a plume of radioactive contaminants (such as air) or is
standing on the surface of contaminated ground (surface water sources are not evaluated here).
A number of groups have investigated uncertainty in radiation dose that is delivered to internal
target organs (i.e., effective dose, via use of DCFs). For example, the US National Committee on
Radiation Protection and Measurements (NCRP) has published a general methodological guide
for uncertainty analysis in dose and risk assessments (NCRP 1996), a guide for evaluating the
reliability of the biokinetic and dosimetric models used to assess individual doses (NCRP 1998),
and assessments of uncertainties associated with internal (NCRP 2009) and external (NCRP
2007) dosimetry. Additionally, the United Kingdom’s Health Protection Agency’s (HPA’s)
Centre for Radiation has conducted uncertainty analyses of internal and external dosimetry
(Puncher and Harrison 2012, 2013).
Major sources of uncertainty associated with effective dose estimation include the following
(Puncher and Harrison 2012):
• Biokinetic models and their parameter values that are used to predict the dynamic
distribution of radioactivity within the body
• The geometric relationship of source and target tissues, their dimensions and masses.
These influence the amount of energy deposited in tissues
• The relative effectiveness of different radiation types in causing cancer and
differences between tissues in their sensitivity to radiation induced cancer
Estimation of disease dose-response and risk (i.e., risk assessment) and associated uncertainties
involves ‘translating’ effective dose into estimation of additional disease (typically cancer)
probability. The Biological Effects of Ionizing Radiation (BEIR) VII report (National Research
Council 2006) contains extensive information on the state of knowledge regarding radiation
dose-response, including a limited uncertainty analysis. Both NCRP (2012) and EPA (EPA
2007) have investigated some sources of uncertainty in risk assessment.
Dose Assessment for the Clive DU PA
6 November 2015 23
With regard to evaluating radiation risk, major sources of uncertainty include the following
(NCRP 2012):
• Issues associated with epidemiological and animal study design and application,
including low statistical power and precision
• Inadequate or simplistic modeling of radiation risk (especially at low doses), or
assumption of one generic model (typically the the linear no-threshold hypothesis, or
LNT, model)
• Extrapolation or generalization of risk estimates to different populations
As an example, EPA (2007) estimated uncertainties for radionuclides that have published risk
coefficients in EPA’s Federal Guidance Report (FGR) No. 13 (EPA 1999). They addressed the
following sources of uncertainty:
• Biokinetic models describing the biological behavior of ingested or inhaled
radionuclides
• Specific energies that relate emissions from source organs to energy deposition in
target organs
• Risk model coefficients representing the risk of cancer per unit absorbed dose to
sensitive tissues from radiation at high dose and high dose rates
• Tissue-specific dose and dose rate effectiveness factors (DDREF); and tissue-specific
high-dose relative biological effectiveness (RBE)
Uncertainties associated with alternative dose-response statistical models (i.e., aside from the
LNT model) were not addressed by EPA (2007). EPA (2007) employed a combination of
modeling and expert opinion in the analysis, and concluded that “the assessed uncertainty in the
radiation risk [as opposed to dose] model was found to be the main determinant of the
uncertainty category for most risk coefficients, but conclusions concerning the relative
contributions of risk and dose models to the total uncertainty in a risk coefficient may depend
strongly on the method of assessing uncertainties in the risk model”.
All groups that have attempted to analyze uncertainties associated with radiation effective dose
and risk have acknowledged that this is a difficult undertaking, and there is no generic “one-size-
fits-all” solution. Each type of radiation and target organ dose-response has unique
characteristics. Therefore, the most straightforward way to evaluate uncertainties in dose and risk
may be to employ the FGR 13 central values and ‘uncertainty categories’ published by EPA
(1999, 2007). These are represented as a ratio of the 95th to the 5th quantiles. As an example, if
an uncertainty factor is 100, then a risk coefficient could vary from the published FGR 13 value
by a factor as great as 10 (the square root of 100). Most radionuclides fall within categories A or
B.

Dose Assessment for the Clive DU PA
6 November 2015 24
Unlike any other sources reviewed, ratios are available for a large (>800) number of
radionuclides. The exact ratio values (as opposed to the letter categories) are available for all
radionuclides with risk coefficients in FGR 13 (EPA 1999). Assuming a distributional shape
such as lognormal, distributions can then be developed.
If uncertainties associated with effective dose only are evaluated (which is the approach taken at
this time in the DU PA), the scope of existing and published work is much more limited. In order
to be useful for probabilistic modeling, the uncertainties associated with DCFs must be
represented as statistical distributions. A search of the published literature indicates that
uncertainty distributions for DCFs per se have only been developed in a few instances; largely
focused on a few radionuclides (e.g., I-131, tritium) that have been to focus of worker protection
assessments, legal cases, and related dose reconstruction scenarios (e.g., Hamby, 1999; Harvey et
al., 2006). Puncher and Harrison (2012, 2013) evaluated uncertainties for 9 radionuclides via
ingestion and inhalation. For the purpose of the PA, uncertainty distributions for a large number
of DCFs would ideally be available. No such 'global' source was identified in the literature.
However, there has been published work that has focused on components of DCFs that are
generalizable to different classes of radionuclides. The most relevant work that was identified is
the work of Kocher et al. (2005), in the context of "probability of causation" in cases of worker
exposure to radiation. This work has been incorporated into the National Institute for
Occupational Safety and Health's "Interactive RadioEpidemiological Program" (IREP;
http://www.cdc.gov/niosh/ocas/ocasirep.html), which is employed to determine the probability
that a cancer was caused by workers' exposure to radiation during nuclear weapons production.
Similar work has been applied in the context of probabilistic dose reconstruction (Linkov et al.,
2001.)
Kocher et al. (2005) estimate:
…so-called radiation effectiveness factors (REFs) [note: not to be confused with 'radon
emanation factors'] that are intended to represent the biological effectiveness of different
types of ionizing radiation for the purpose of estimating cancer risks and probability of
causation of radiogenic cancers in identified individuals. An REF is a dimensionless
factor used to modify an estimate of average absorbed dose from a given radiation type
in an organ or tissue of concern in an identified individual to obtain a biologically
significant dose on which the risk of induction of cancer in that organ or tissue is
assumed to depend.
Kocher et al. (2005) specify that they are ultimately interested in risks, not doses; but, the
estimates of uncertainty associated with REFs are relevant to the current application. They state
that their REFs are essentially analogous to radiation weighting factors (wR). The wR is is an
additive function of a dimensionless “quality factor” Q, that is dependent upon radiation type;
and a dimensionless N, which is dependent upon the tissues irradiated, the time and volume
relevant to irradiation, and biological characteristics of the receptor. Consistent and thorough
documentation of these terms appear to be lacking in published reports. Regardless, in most
cases, these terms have been superseded by another term; Relative Biological Effectiveness
(RBE). The radiation dose unit employed in this PA, the sievert (Sv), can vary considerably
based upon the RBE.
Dose Assessment for the Clive DU PA
6 November 2015 25
Kocher et al. (2005) state that their
…new term “radiation effectiveness factor” (REF) is used in this work to distinguish a
quantity that represents biological effectiveness for purposes of estimating cancer risks
and probability of causation in identified individuals from similar quantities, including
relative biological effectiveness (RBE), which strictly applies only to results of specific
radiobiological studies under controlled conditions.
For the purpose of establishing initial uncertainty distributions for DCFs for incorporation into
the PA, these philosophical and semantic issues will take a subservient position. We will
therefore assume that for the carcinogenic effects of radiation, that the REF is equivalent to the
RBE, which is in turn equivalent to wR. This is not strictly the case, but the intent here is to
estimate uncertainty in biologically-relevant radiation dose, not exact numerical quantities. REFs
account for the fact that some types of radioactive decay result in more biological damage than
others. The "reference" type of radiation is typically Co-60 high-dose/dose-rate gamma decay,
as this is the type of radiation germane to the atomic-bomb survivor data and similar sources of
epidemiological data on cancer resulting from radiation exposure. The REF (or wR) for such
radiation is set at 1.0. However, larger particles such as alpha particles and neutrons can cause
more biological damage, thus the REFs for these types of ionizing radiation are larger, and
function as multipliers to the DCFs.
In this PA model, radiation-type specific REFs per Kocher et al. (2005) will be used as
modifying distributions to the DCF point estimates presented in FGR 13 (note that DCFs are not
presented in the written report of FGR 13, but are available via an online database:
http://ordose.ornl.gov/downloads.html). Kocher et al. (2005) developed probability distributions
for REFs, based upon a combination of exhaustive literature review, statistical analysis,
modeling, and subjective judgment. Tables 14 and 15 in that reference provide summaries.
These REF distributions can be essentially viewed as modifiers to published DCFs, in lieu of the
published deterministic wR's used in radiation protection (ICRP, 1991). For example, the
published deterministic wR for alpha particles is 20. The Kocher et al. (2005) REF for alpha
particles can be represented by a lognormal distribution with a median of 18, and a 95%
confidence interval from 3.4 to 100. Thus, for an alpha-emitting radionuclide, the published DCF
would be divided by 20, then multiplied by the distribution provided. As the REFs are radiation-
type specific, they are generally applicable to the predominant radiation characteristics of the
particular radionuclide of concern.
In the present model, there are no species that decay by neutron emission. The REFs employed
represent alpha, beta (electron), and photon (gamma, X-ray) decay. For each radionuclide, the
dominant radiation type and its energy are defined based upon information from ICRP (using the
program RadSum32, available from http://ordose.ornl.gov/downloads.html). For some
radionuclides, the energy of electron or photon emissions is essentially equivalent to the
reference radiation (high-energy gamma), resulting in an REF of 1.0 with no uncertainty. For
others, an REF distribution is defined based upon the information in Kocher et al. (2005) and this
REF is used as a multiplier to the DCF. Please note that radon is evaluated differently from other
radionuclides (see Section 4.4); thus the REF distribution development process outlined below
does not apply.
Dose Assessment for the Clive DU PA
6 November 2015 26
Following is a summary of the specific process by which REF distributions are generated and
applied in the PA model, along with assumptions (please see Kocher et al. (2005) for
assumptions made in that work). Radionuclide-specific deterministic DCFs, and the inputs
necessary to calculate stochastic DCFs, are provided in the spreadsheet Dose Assessment
Appendix II.
1. The 27 radionuclide Species in the PA model were expanded to 63 radionuclides to
account for short-lived progeny (Species radionuclides have a half-life of approximately 2 years
or longer). The decay chains for identifying progeny were taken from the Nuclear Wallet Cards
(Tuli, 2005).
2. DCFs were taken from the the EPA FGR 13 database (available from
http://ordose.ornl.gov/downloads.html). DCFs are available for particulate and vapor-phase
inhalation, ingestion, and external exposure (including "submersion", "ground plane", and "soil
volume" values). In all cases, DCFs for adults are selected (as the receptors of interest are
adults), and “effective dose” DCFs (a weighted composite of all organs) are employed.
Inhalation DCFs related to the default inhalation absorption class from Table 2.1 of FGR 13 were
used. If no default class was specified, the “medium” (Class M) inhalation DCF was usually
selected because it is commonly between the DCF values for slow and fast absorption classes,
and is therefore considered to be the least biased point estimate. For external exposure to
contaminated soils, the “soil volume” external DCFs are used in this PA consistent with the
physical models of contaminant transport over time.
3. A dominant form of radiological decay was assigned for internal DCFs and external DCFs
for each of the 63 radionuclides using information from the RadSum32 code. For internal DCFs,
the dominant decay mode was identified as the highest contributor to total emitted energy of any
radiation type (gamma + x-ray; electron (the maximum of beta, internal conversion electrons, or
auger electrons); and, alpha). In all cases, this protocol resulted in alpha emissions being selected
as the dominant decay mode when alpha decay occurs. For external DCFs, the dominant decay
mode was identified as the energy of gamma + x-ray. If there are no photon emissions for a
radionuclide, dominant decay for external irradiation was identified as the highest energy among
beta, internal conversion electrons, and auger electrons. Because alpha particles cannot penetrate
the stratum corneum to the biologically active lower strata of the skin, alpha particles are not
evaluated for the purpose of assigning REF distributions to external DCFs.
4. For radionuclides where the dominant decay mode is electron or photon, the average
particle energy of that decay mode (in million electron volts, or MeV) is identified from the
RadSum32 code.
5. REF distributions are defined for four categories of decay mode and energy, based upon
percentiles in Tables 14 and 15 in Kocher et al. (2005). For radionuclides where the dominant
decay mode is photon or electron emission with a mean energy higher than the particular
threshold, an REF of 1.0 is assigned, as the REF for these emissions are essentially equivalent to
the reference radiation (Co-60 gamma). The REF distribution categories include:

Dose Assessment for the Clive DU PA
6 November 2015 27
• alpha (any energy)
• electron (<0.015 MeV)
• photon (>0.03 and <=0.25 MeV)
• photon (<=0.03 MeV)
6. With regard to the alpha REF, please note that Kocher et al. (2005) assumed that 100
represented the 97.5th percentile of the distribution. This is likely conservative, as the highest
value ever estimated from experimental studies is 100, and this only applies to particular forms
of inhaled plutonium (Kocher et al., 2005).
7. The DCFs for each of the 63 radionuclides are divided by the ICRP weighting factor (wR)
in order to apply the REF distributions. For alpha emitters, the wR value is 20, and for electrons
and photons it is 1.0. Stochastic DCFs are then calculated as the product of the DCF and the
appropriate REF.
8. DCFs for the 27 radionuclide Species defined in the PA model are assembled using the
decay chains and branching fractions from the Nuclear Wallet Cards (Tuli, 2005). These are
equivalent to the “plus daughters” (+D) DCFs for primary radionuclides provided in radiological
dose software such as the RESRAD computer code (http://web.ead.anl.gov/resrad/home2/).
9. The stochastic +D DCFs may then be employed in the PA model for radiation dose
calculations. Alternatively, a model user may select the option of using the deterministic FGR 13
DCFs in a simulation. This is permitted even when the PA model is run in stochastic mode for all
other model parameters.
As previously discussed, this method only addresses one component of uncertainty associated
with DCFs, and thus must be viewed as a pilot effort. DCF distributions are available for some
radionuclides, and could be incorporated into future modeling. Use of EPA (2007) risk coeffi-
cients in addition to or in lieu of dose estimations would be a logical next step in expanding the
scope of the uncertainty analysis for the health effects of radionuclides.
3.4.4 Additional Sources of Uncertainty
In addition to variance in the definition of model parameter values, there are other important
sources of uncertainty and/or bias to potentially consider. For example, if radiation dose-
response model uncertainty (particularly at low doses) were to be considered, it is possible that
the uncertainties associated with radiation risk would swamp those associated with the remainder
of the PA model, as it is by no means clear that ionizing radiation has no threshold of
carcinogenic effect.
here is uncertainty associated with the mathematical models defining contaminant transport in
the environment over time. These models are designed to represent the system as best they can
(although sometimes with known protective biases) but they like all models are simply
approximations of reality. Other aspects of the PA model have similar issues associated with
model uncertainty.
Dose Assessment for the Clive DU PA
6 November 2015 28
Most importantly, the overall uncertainty associated with what the natural world and human
society will be like in 1,000 or 10,000 years from today is likely much greater than the
uncertainty associated with the model form, yet this 'future world' uncertainty is not quantifiable
or readily bounded. Such sources of uncertainty must be discussed qualitatively rather than being
quantitatively modeled.
3.4.5 Non-Cancer Toxicity Endpoints
DU waste (and potentially other compounds) associated with the Clive facility can be associated
with toxicological risks that are independent of radioactive properties. EPA has evaluated
available dose response information for many chemicals and has published this information in
the form of toxicity values and accompanying information. Potential health effects related to
intake of chemicals is assessed by means of slope factors for suspected carcinogens, and
reference doses (RfDs) for noncarcinogenic effects of chemicals. Unlike carcinogenic agents,
EPA typically views toxicants with non-cancer effects as having thresholds; i.e., levels below
which effects would be unlikely. RfDs essentially amount to such thresholds, usually with
several layers of 'safety' factors added.
A limited evaluation of the effect of science policy uncertainty in the value of the uranium oral
RfD on chemical hazard results is included in this assessment. The modeling process is very
similar to that conducted for radionuclides, other than kidney toxicity (as opposed to radiation
dose) of DU will be evaluated, and the toxicity of DU will not change over time (as radioactive
decay is not important in this context).
Oral toxicity criteria for uranium are published by EPA in relation to the Superfund program
(EPA, 2011) and by EPA's Office of Water in relation to drinking water standards (EPA, 2000).
There is a five-fold difference between these criteria, and both will be employed in the
assessment of uranium toxicity to determine the sensitivity of uranium health effect results to
differences in these recommended toxicity criteria for uranium.
A discrete distribution is used to represent the uranium oral RfD based on current EPA science
policy associated with EPA’s Superfund Program and Office of Water. A uranium oral RfD of
0.0006 mg/kg-day is associated with the derivation of the final uranium drinking water
maximum contaminant level (MCL) as defined on page 76713 of Federal Register, Volume 65,
No. 236, December 7, 2000 (Section I.D.2d). A uranium oral RfD of 0.003 mg/kg-day for
soluble salts of uranium is published in the Integrated Risk Information System (IRIS)
supporting the Superfund Program. A 50/50 probability is assigned to these oral RfDs to
determine in the Sensitivity Analysis whether selecting one or the other of these published values
is a significant contributor to uncertainty in the uranium Hazard Index in any exposure scenario.
Dose Assessment for the Clive DU PA
6 November 2015 29
4.0 Equations and Parameters of the Exposure-Dose Con-
tainer
4.1 Organization
The implementation of the exposure and dose calculations, and associated results, are organized
within different subcontainers in the ED container. A description of the main subcontainers and
their contents are described below:
• Environmental Concentrations: Concentrations of species in various environmental
media developed in the Contaminant Transport (CT) component of the PA model are
tracked here. These elements are the link between the CT and ED components of the
PA model, and take the form of GoldSim vectors defined by the array Species.
Environmental concentrations are subsequently defined as two-dimensional matrices
with the addition of arrays for different receptor groups in order to track doses for
multiple individuals to tally a population dose.
• Behavioral Parameters: Input parameter values related to human activities and
behaviors for the Rancher, Sport OHVer, and Hunter exposure scenarios. With few
exceptions, these parameters are defined within an 'inner-loop' container that has a
separate internal timestep so that they can be sampled on an annual basis regardless
of the timestep length of the CT model.
• DCFs: Dose conversion factors for radionuclides are grouped in a subcontainer
outside the inner-loop container..
• Dose Calculations: A series of subcontainers are defined within the inner-loop
container for calculation of TEDE related to inhalation, ingestion, and external
radiation exposures for the Rancher, Sport OHVer, and Hunter exposure scenarios. A
container for off-site receptor doses is also provided. Screening-level dose
calculations for ingestion of edible plant materials gathered on the waste disposal
cell, and ingestion of standing surface water, are grouped in a subcontainer outside
the inner-loop container.
• Uranium Hazard: A subcontainer within the inner-loop container holding calculations
for systemic toxicity (hazard) related to the nonradiological effects of uranium.
In terms of parameter definitions, GoldSim uses a variety of methods, including deterministic
values, scalars, time series data, and “stochastics”, which are user-defined statistical
distributions. Parameter distributions employed in the PA model reflect a mixture of site- and
receptor-specific data, information modeled in 'upstream' portions of the PA model, literature
information, and subjective judgment; as appropriate.
Dose Assessment for the Clive DU PA
6 November 2015 30
4.2 Environmental Concentrations
The principal link between the CT component and the ED component of the PA model are
concentrations of contaminants in different environmental media. Major environmental media
evaluated in the ED container include:
• Soil. There are several soil concentration terms that are used in the ED container. The
contaminant transport portion of the PA model employs a homogenized waste source
term and simulates transport over time to produce estimates of soil concentrations for
the embankment top slope and the embankment side slopes. The principal soil term in
the ED container is the area-weighted average concentration in the top layer of both
the top slope and side slope of the disposal cap. This is the disposal cap soil
concentration. Contaminant concentrations in these soils, plus possible contribution
from lower soil layers and even the disposed waste itself, are used to calculate soil
exposure concentrations for the embankment. Embankment soil concentrations are
defined as the area-averaged soil concentrations of the disposal cap and of one or
more gullies and fans that may develop in the future. Finally, particle resuspension
and deposition models are used to calculate area-averaged soil concentrations off-site
air dispersion area based upon the embankment soil concentrations.
Area-averaged soil concentrations for the embankment and the off-site air dispersion
area are employed because there is no basis for specifying greater or lesser individual
exposure intensity as a function of location within these regions. Individuals are
presumed to be exposed at random in these areas, and an area-averaged exposure
concentration reflects this presumed behavior.
The human exposure area surrounding the Clive site is where the Ranchers, Sport
OHVers, and Hunters identified as likely receptor populations conduct their activities.
The maximum size of this area is the approximate area between I-80 and the Utah
Test and Training Range (UTTR) in an east-west orientation, and the Cedar Mountain
foothills and salt/mud flats in a north-south orientation. The minimum size of this
area is the approximate minimum size of the four current grazing leases in the vicinity
of the Clive facility. Because the maximum area is roughly equivalent to the largest
of the four current grazing leases, the human exposure area and the size of the area
over which cattle may graze are equivalent.
• Air. Air concentrations of gaseous and particulate contaminants in the atmosphere are
calculated using the AERMOD atmospheric dispersion model for breathing-zone air
above the embankment and above the off-site dispersion area. Off-site air
concentrations are also calculated at the specific exposure locations described in
Section 3.2.3. These calculations are documented in the Atmospheric Transport
Modeling white paper (Appendix 8). To evaluate the impacts of dust generated during
off-highway vehicle (OHV) use, an adjustment factor for particulate air
concentrations is used based upon dust generation data collected by EPA Region 9
for OHV users wearing personal air monitors in a recreational area in California
(EPA, 2008).
Dose Assessment for the Clive DU PA
6 November 2015 31
• Game. Contaminant concentrations in the meat of game animals that incorporate the
embankment and nearby areas as part of their home range. Based upon
communications with BLM, pronghorn are modeled as the most likely game species
of interest to future Hunters. Contaminant concentrations in game tissue are modeled
as a function of ingestion of browse plants, standing surface water, and soil
inadvertently ingested while browsing.
• Beef. Contaminant concentrations in beef from cattle that incorporate the
embankment and nearby areas as part of their range. Similar to game tissue
concentrations, beef concentrations are related to plants, surface water, and soil. The
number of cattle grazing in impacted areas is assumed to be sufficient to provide
ranchers with beef commensurate with the specified intake rates.
• Plants. Wet weight contaminant concentrations in plant tissues. These concentrations
are used as an interim step in the calculation of tissue concentrations in cattle and
game and are calculated assuming equilibrium with soil defined by element-specific
plant-soil concentration ratios. They are also used for screening-level calculations to
determine if potential direct human exposures by plant ingestion may be of concern.
• Surface Water. Contaminant concentrations in standing surface water in the air
dispersion area. Water concentrations are calculated assuming equilibrium with soil,
as defined by element-specific soil-water partition coefficients. These water
concentrations are used as an interim step in the calculation of tissue concentrations
in cattle and game. They are also used for screening-level calculations to determine if
potential direct human exposures by surface water ingestion may be of concern.
Groundwater is not an exposure medium per se, because the aquifer below the Clive facility is
too saline to be used as a drinking water source, and so is classified by the State of Utah as Class
IV (nonpotable) in the ground water quality discharge permit for the Clive facility. However, the
permit also states that concentrations of contaminants in groundwater will nevertheless be
compared to State of Utah GWPLs.
4.3 Exposure Parameters
The basis of the deterministic values and/or statistical distributions for each of the ED equation
parameters is discussed in the Model Parameters white paper (Appendix 16), the attached
Appendix I, and the spreadsheet Dose Assessment Appendix II. A major source of exposure
parameter values is the 2009 update to the EPA Exposure Factors Handbook (EPA, 2009a).
Although this reference exists as an external review draft, it is much more current and extensive
than the 1997 version, and much more distributional information is included. For physiological
variables in particular, the primary studies that EPA employed as the basis of recommendations
in EPA (2009a) were also reviewed.
Three non-residential human receptor scenarios (Rancher, Sport OHV recreationist, and Hunter
recreationist) are defined, each with its own set of exposure parameter values but with similar
computational exposure models. Exposure parameters that pertain to inter-individual population
variability have been assigned to the “inner loop” of the 2D Monte Carlo simulation. These
Dose Assessment for the Clive DU PA
6 November 2015 32
parameters pertain to physiological characteristics, the fraction of time an individual spends on
or near the site, and the number of receptors present at the site. These categorizations of inner or
outer loop are noted in Section 1 and discussed in Section 3.4.1.
Exposure parameters related to inter-individually varying population characteristics, and to the
number of receptors within the exposure area, are defined within an “inner-loop” sub-container
in the ED model. This sub-container has an annual time step so that the stochastic parameters
relating to the number of individuals appearing in the exposure area, and the inter-individual
characteristics of these individuals, are sampled annually. This sub-container is the "inner loop"
of the 2-dimensional Monte Carlo simulation.
The remainder of the exposure parameters, which include the exposure concentrations in
environmental media, the DCFs, and a few other parameters, are defined by uncertainty
distributions that apply to each individual in the population over the entire 10,000-yr
performance period. These parameters, and all components of the contaminant transport model
that produce estimates of exposure concentrations over time, are in the "outer loop" of the 2-
dimensional Monte Carlo simulation. The uncertainty distributions for stochastic parameters in
the outer loop outside this sub-container are sampled only once at the beginning of each model
realization.
In the 2-dimensional model, it is assumed that uncertainties are independent for each member of
the ranching and recreational scenario populations. The fraction of time that each individual
spends on the disposal cell or in the adjacent off-site area is variable. Because the processes that
lead to concentration terms in these two areas are different, they have different uncertainty
characteristics. This results in independence in the uncertainties of the individual annual dose
results.
The inhalation rate distributions activities are specified according to exertion level as heavy,
moderate, light, sedentary, and sleeping. For each exertion level, EPA (2009a) provides
information for breathing (ventilation) rate and associated fraction of daily time spent at that
level. In the absence of scenario-specific information, the fraction of daily time spent at each
exertion level for the general population described in EPA (2009a) has been applied to ranching
and recreation receptors. Stochastic distributions for the inhalation rates, and also for meat
ingestion rates, are tied to the age and (for inhalation rate) gender of an individual receptor, and
are specified as a linear function of their body weight as described in EPA (2009a; 2009b). An
adult between the ages of 16 and 60 is defined for the ranching and recreation receptor groups.
The behavioral exposure parameters defined in the inner-loop sub-container relate primarily to
the fraction of daily and yearly time spent by receptors in the exposure area generally, and within
the exposure area the fractional time spent on the embankment versus other locations. Based
upon discussion with BLM, Ranchers are assumed to work within a ranching lease during the
day and may also camp overnight. Both Sport OHVers riders and Hunters may visit the area for
either a day trip or an overnight trip.
Dose Assessment for the Clive DU PA
6 November 2015 33
4.4 DCFs
The TEDE is not an effect per se, but rather a measure of radiation dose absorbed by a tissue.
The DCFs used in the ED model account for the biological effectiveness of the radiation (e.g.,
alpha particles, photons) in causing cellular damage in different tissues, as well as the sensitivity
of different tissues to the effects of ionizing radiation. For external dose, this “effective dose” is
calculated. For internal dose, the committed effective dose is calculated, which accounts for
continued dose over time from radionuclides retained in the body. Distribution development for
one source of uncertainties inherent in DCFs (i.e., associated with REFs) is described in Section
3.4.3.
Section 3.3.7 of NUREG-1573 (NRC, 2000) discusses modeling of radiation dose, including
internal and external dosimetry. NRC (2000) notes that the performance objectives set forth in
Section 61.41 of Title 10 CFR 61.41 (CFR, 2007) are based upon ICRP 2 dose assessment
methods, which pre-date the development of TEDE methodology. NRC recommends the use of
current ICRP dosimetry employing TEDE methods in lieu of calculation of individual organ
doses.
The internal and external DCFs used in the ED model were obtained from the electronic database
accompanying FGR 13 (EPA, 1999), available online at http://ordose.ornl.gov/downloads.html
and also provided in the spreadsheet Dose Assessment Appendix II. The DCFs for all species, as
well as the individual short-lived progeny of these parent nuclides, were developed using
appropriate decay chains and branching fractions as described in the CSM and documented in the
electronic attachment.
The DCF for radon-222 and progeny was derived from recommendations provided in an ICRP
draft report for consultation (ICRP, 2009). A range of 3 - 6 mSv-m3/mJ-hr is given for the radon-
222 DCF, calculated using ICRP's Human Respiratory Tract Model. The main sources of
uncertainty related to this range are the activity size distribution of aerosols for radon progeny,
and the breathing rates (ICRP 2009; Appendix B, paragraph B 6).
In paragraph B 11 of Appendix B to ICRP (2009), the inhalation rate for a "standard worker"
associated with the upper-end DCF estimate of 6 mSv-m3/mJ-hr is given as 1.2 m3/hr. ICRP
states,
For typical aerosol conditions in home and mines the effective dose is about 3.7 mSv-
m3/mJ-hr. . . However, assuming the same aerosol conditions as for a home but with a
breathing rate for a standard worker (1.2 m3/hr) the effective dose increases from 3.7 to 6
mSv-m3/mJ-hr.
This indicates that approximately 75% of the range of 3 - 6 mSv-m3/mJ-hr given for the Rn-222
DCF may be related to inhalation rate. Based upon this observation, a breathing rate normalized
radon-222 DCF was calculated for use in the ED model. The units for alpha energy (mJ) were
converted to an equivalent activity (Bq) for radon-222 according to units definitions in the
glossary of ICRP (2009).

Dose Assessment for the Clive DU PA
6 November 2015 34
A radon-222 DCF of 2.8 × 10-8 Sv/Bq was calculated as:
Radon-222 DCF = (0.006 Sv-m3/mJ-hr × 5.56 × 10-6 mJ/Bq) / 1.2 m3/hr
Note that the REFs discussed earlier are not applicable to radon, as the DCF was estimated in a
different fashion than the other species.
4.5 PDCFs
A PDCF is an equation that combines Exposure Parameter values and DCFs, as described in
Section 3.3.7.2 of NRC (2000). PDCFs are combined with estimates of radionuclide
concentrations in exposure media to calculate a TEDE. PDCF equations for each exposure route
are described in subsections below.
4.5.1 Inhalation PDCF Equations
PDCF for inhalation of particulates and gases
PDCF_Inh (Sv·m3/Bq·yr) = DCF_Inh × InhalationRate × EF × ET (1)
where
DCF_Inh is the inhalation DCF (Sv/Bq)
InhalationRate is the activity-weighted inhalation rate (m3/hr)
EF is the yearly exposure frequency (d/yr), and
ET is the total daily exposure time (hr/d).
and
InhalationRate (m3/hr) = ∑i (Inhal_acti × ET_fraci ) (2)
where
Inhal_acti is the inhalation rate for activity level i (m3/hr), and
ET_fraci is the fraction of daily exposure time for activity level i (-)
Activity levels (i) for which population-average breathing rates and daily exposure times are
defined include sleeping, sedentary activity, light activity, medium activity, and heavy activity.
Breathing rates are body weight adjusted. Population distributions of both breathing rates and
daily exposure times at different activity levels are defined as functions of age and gender, as
described in EPA (2009a).

Dose Assessment for the Clive DU PA
6 November 2015 35
4.5.2 External PDCF Equations
PDCF for external radiation from soil
PDCF_Ext_Soil (Sv·g/Bq·yr) = DCF_Ext × EF × ET × ρb × CF1 (3)
where
DCF_Ext is the external DCF for a 3-dimensional soil source (Sv·m3/Bq·s)
EF is the yearly exposure frequency (d/yr)
ET is the total daily exposure time (hr/d)
ρb is the bulk soil density (g/m3), and
CF1 is a unit conversion factor (3600 s/hr)
PDCF for external radiation from immersion in air
PDCF_Imm (Sv·m3/Bq·yr) = DCF_Imm × EF × ET × CF1 (4)
where
DCF_Imm is the external DCF for air immersion (Sv·m3/Bq·s)
EF is the yearly exposure frequency (d/yr)
ET is the total daily exposure time (hr/d), and
CF1 is a unit conversion factor (3600 s/hr)
4.5.3 Ingestion PDCF Equations
PDCF for inadvertent ingestion of soil
PDCF_Ing_Soil (Sv·g/Bq·yr) = DCF_Ing × SoilIngRate × EF × CF2 (5)
where
DCF_Ing is the ingestion DCF (Sv/Bq)
SoilIngRate is the daily soil ingestion rate (mg/day)
EF is the yearly exposure frequency (d/yr), and
CF2 is a unit conversion factor (0.001 g/mg).

Dose Assessment for the Clive DU PA
6 November 2015 36
PDCF for ingestion of game meat or beef
PDCF_Ing_Meat (Sv·g/Bq·yr) = DCF_Ing × MeatConsumpRate × (1 - Prep_loss) × (1 –
PostCook_loss) × EF_food (6)
where
DCF_Ing is the ingestion DCF (Sv/Bq)
MeatConsumpRate is the daily consumption rate of beef or game meat (g/kg body
weight/d)
Prep_loss is the fractional preparation and cooking loss of consumed meat related to
dripping and volatile losses during cooking (-)
PostCook_loss is the fractional post-cooking loss of consumed meat related to trimming,
bones, scraps, etc (-)
EF_food is the intrinsic exposure frequency assumed in the time-averaged ingestion rate
data (d/yr)
PDCF for plant ingestion (screening calculation)
PDCF_Ing_Plant (Sv·g/Bq·yr) = DCF_Ing × PlantIngRate (7)
where
DCF_Ing is the ingestion DCF (Sv/Bq, and
PlantConsumpRate is the yearly consumption rate of wild plants (g/yr)
PDCF for water ingestion (screening calculation)
PDCF_Ing_Water (Sv·g/Bq·yr) = DCF_Ing × WaterIngRate × WatDens (8)
where
DCF_Ing is the ingestion DCF (Sv/Bq,
WaterConsumpRate is the yearly consumption rate of standing water (L/yr), and
WatDens is the density of water (g/L)

Dose Assessment for the Clive DU PA
6 November 2015 37
4.6 TEDE
The calculation of dose, represented here by TEDE, is the product of a PDCF and the exposure
concentration. Separate soil concentrations are developed in the contaminant transport model for
the disposal cap and the off-site area impacted by deposition of wind-dispersed particles.
Particulate air concentrations, which are related to resuspension of soil, and concentrations of
gas-phase radionuclides in air, are also calculated separately for these three exposure areas.
Other exposure concentrations used in the dose model include radionuclide concentrations in
animal tissue, as well as plant tissue and standing surface water in screening calculations.
All TEDE calculations reference the PA model element describing the time after site closure
when institutional controls fail and a receptor can gain access to the site. If this time has not been
reached in the model realization, ranching and recreation doses are assigned a zero value.
Note that potential embankment gullies are modeled in a preliminary manner in the PA model to
evaluate possible consequences given the current waste disposal configuration. Gully formation
can be 'switched' on or off by the model user.
4.6.1 Inhalation TEDE Equations
Gas and particulate inhalation TEDE results (mSv/yr) are vectors dimensioned by Species in the
PA model related to the inhalation PDCFs. Concentrations of respirable particles and gas-phase
radionuclides in air are calculated by methods described in the Atmospheric Transport Modeling
white paper (Appendix 8). The inhalation TEDE equation for particulate inhalation is:
TEDE_Inh (mSv/yr) = PDCF_Inh × Cair (9)
where
PDCF_Inh is the inhalation PDCF (Sv·m3/Bq·yr), and
Cair is the spatially-averaged air concentration (Bq/m3)
Exposure concentrations on the embankment are calculated in an area-weighted manner. This
calculation presumes that exposures across the embankment occur in a random manner. Air
concentrations above the embankment are calculated as:
Cembnk (Bq/m3) = {([Ccap × Acap + Cgullies × Agullies] / [Acap + Agullies]) (10)
where
Cembnk is the air concentration above the embankment (Bq/m3)
Ccap is the air concentration above the disposal cap (Bq/m3)
Acap is the area of the embankment cap (m2)
Cgullies is the air concentration above the gullies and associated fans (Bq/m3)
Agullies is the surface area of the gullies and associated fans (m2)
Dose Assessment for the Clive DU PA
6 November 2015 38
The terms Cgullies and Agullies are calculated using a model for possible erosive effects of
precipitation subsequent to gully initiation due to OHV activity, grazing animals, or other
processes.
With respect to ranching and recreation exposure, there are two concentrations terms to address:
a concentration term for the embankment (per Equation 10) and a concentration term for the off-
site air dispersion area. A weighted exposure concentration for particulates in ambient air is
calculated for these two concentration terms as follows:
Cair-dust (Bq/m3) = { OHV_timefrac × OHV_dust × (Cembnk × ET_fracembnk) + (Ca-disp × [1
– ET_fracembnk] } + { [ 1 - OHV_timefrac] × (Cembnk × ET_fracembnk) + (Ca-disp × [1 –
ET_fracembnk] }
(11)
where
OHV_timefrac is the fraction of exposure time spent OHVing (-)
OHV_dust is the off-highway vehicle dust factor, used to account for the contribution of
mechanical dust creation (-)
Cembnk is the air concentration above the embankment (Bq/m3)
ET_fracembnk is the fraction of total daily exposure time spent on the embankment (-)
Ca-disp is the air concentration above the air dispersion area (Bq/m3)
For particulates, Cembnk and Ca-disp are calculated using a particle erosion model, which calculates
the amount of dust released from the ground surface, and the AERMOD air dispersion model
(see the Atmospheric Transport Modeling white paper, Appendix 8). Particle erosion is assessed
as a function of both wind and mechanical disturbance from the use of OHVs, but the
mechanical dust creation factor is applied as a multiplier to the baseline (wind-derived) dust
concentration. The AERMOD air dispersion model is used to estimate particulate deposition in
the offsite air dispersion area as well as breathing zone concentrations of respirable particles
above contaminated soil.
For radon and other gas-phase radionuclides, Cembnk and Ca-disp are calculated using AERMOD
(see the Atmospheric Transport Modeling white paper, Appendix 8) based upon the embankment
surface flux computed in the PA model. The air dispersion area is not a definite region with
respect to particle definition, because it's size is defined by the size of the receptor exposure area,
which varies as described in Section 4.2. Based upon AERMOD calculations, a protective
estimate of respirable particle deposition beyond the embankment is assigned to this area. Radon
air concentrations in the off-site air dispersion area are calculated as the average across the entire
area.
Because mechanical dust generation by OHVs is not an issue for calculating air concentrations of
radon and other gas-phase radionuclides, Equation 11 reduces to:
Cair-gas (Bq/m3) = (Cembnk × ET_fracembnk) + (Ca-disp × [1 – ET_fracembnk]) (12)
Dose Assessment for the Clive DU PA
6 November 2015 39
The current version of the PA model does not fully integrate gully formation into the physical
model of the embankment. Therefore, Radon air concentrations in the gully are modeled from
estimated radium-226 surface soil concentrations on the gully ‘floor’. The contribution of radon
from disposed waste below this surface soil layer is presently accounted for. Also, the influence
of gully walls on radon air concentrations within the gully has not been modeled. For these
reasons, gully radon exposures may be underestimated.
4.6.2 External Radiation TEDE Equations
Soil and air immersion external dose results (mSv/yr) are vectors dimensioned by Species in the
PA model related to the external PDCFs.
The air immersion external dose equation is:
TEDE_Imm (mSv/yr) = PDCF_Imm × Cair (13)
where
PDCF_Imm is the immersion PDCF (Sv·m3/Bq·yr), and
Cair is the spatially-averaged air concentration (Bq/m3)
The derivation of Cair for air immersion is identical to that described in Equations 10, 11 and 12.
The soil external dose equation is:
TEDE_Ext_Soil (mSv/yr) = PDCF_Ext_Soil × Csoil (14)
where
PDCF_Ext_Soil is the soil ingestion PDCF (Sv·g/Bq·yr), and
Csoil is the spatially-averaged soil concentration (Bq/g)
Similar to Equation 8, soil concentrations on the embankment are calculated as:
Cembnk (Bq/g) = {([Ccap × Acap + Cgullies × Agullies] / [Acap + Agullies]) (15)
where
Cembnk is the embankment soil concentration (Bq/g)
Ccap is the disposal cap soil concentration (Bq/g)
Acap is the area of the disposal cap (m2)
Cgullies is the soil concentration of the gullies and associated fans (Bq/g)
Agullies is the surface area of the gullies and associated fans (m2)
Dose Assessment for the Clive DU PA
6 November 2015 40
Analogous to Equation 12, a weighted exposure concentration for embankment and air
dispersion area soil is calculated as follows:
Csoil (Bq/g) = (Cembnk × ET_fracembnk) + (Ca-disp × [1 – ET_fracembnk])

(16)
where
ET_fracembnk is the fraction of total daily exposure time spent on the embankment (-)
Ca-disp is the soil concentration for the air dispersion area (Bq/g)
4.6.3 Ingestion TEDE Equations
Inadvertent soil ingestion (i.e., via soil on hands, food, etc.) and meat ingestion dose results
(mSv/yr) are vectors dimensioned by Species in the PA model related to the ingestion PDCFs.
The soil inadvertent ingestion dose equation is:
TEDE_Ing_Soil (mSv/yr) = PDCF_Ing_Soil × Csoil (17)
where
PDCF_Ext_Soil is the soil ingestion PDCF (Sv·g/Bq·yr), and
Csoil is the spatially-averaged soil concentration (Bq/g)
(Csoil is calculated according to Equation 16.)
The meat ingestion dose equation is:
TEDE_Ing_Meat (mSv/yr) = PDCF_Ing_Meat × Cmeat (18)
where
PDCF_Ext_Soil is the soil ingestion PDCF (Sv·g/Bq·yr), and
Cmeat is the concentration in beef or game meat (Bq/g)
The calculation of Cmeat is based upon grazing models for beef cattle and pronghorn and uses as
inputs the soil concentrations Cembnk and Ca-disp. Both beef cattle and pronghorn may be exposed
to soil contamination by direct soil ingestion while grazing, by ingestion of browse plants
growing in contaminated soil, and by ingestion of standing water on contaminated soil.
Radionuclide concentrations in beef and game tissue are calculated based upon three animal
exposure pathways: direct ingestion of soil while browsing, ingestion of plants growing in
contaminated soils, and drinking standing surface water. Cattle and pronghorn are assumed to
graze randomly across the entire range area. Hence, exposure to radionuclides in the
embankment and air dispersion areas is based upon the relative size of these areas.
Dose Assessment for the Clive DU PA
6 November 2015 41
For soil, exposure concentrations for cattle are calculated as:
Csoil-cattle (Bq/g) = ([Cembnk × Aembnk] + [Ca-disp × Aa-disp]) / Arange-cattle (19)
where
Cembnk is the embankment soil concentration (Bq/g)
Aembnk is the area of the embankment (m2)
Ca-disp is the soil concentration for the air dispersion area (Bq/g)
Aa-disp is the surface area of the air dispersion area (m2), and
Arange-cattle is the size of the cattle range area (m2)
Soil radionuclide exposure concentrations for pronghorn are calculated in an identical manner,
substituting the size of the pronghorn grazing area.
Equation 19 is also used to calculate exposure concentrations in browse plants for cattle and
pronghorn. However, plant concentrations on the disposal cap are based upon uptake of
contamination across the entire root depth profile of the plants. Different types of plants
(differentiated by root depth distributions, biomass, and leaf litter production) are employed in
the Contaminant Transport component of the PA model to evaluate transport of radionuclides on
the disposal cap. Plant concentrations on the disposal cap are calculated in the contaminant
transport portion of the PA model as the weighted average (based upon leaf litter production) of
all plants. Soil concentrations in the air dispersion area, and in the gullies and fans, are only
calculated for a single surface soil layer. 100% of plant roots are assumed to be situated in this
layer.
For standing surface water, exposure concentrations for cattle and pronghorn are calculated for
puddles in the air dispersion area. Puddle water concentrations are based upon bulk soil
concentrations using element-specific soil water partition coefficients.
Using the exposure concentrations described above, radionuclide concentrations in beef are
calculated as:
Cbeef (Bq/g) = TF_beef × (Cplant-cattle × cattle_forage) + (Csoil-cattle × cattle_soil) + (Cwater-
cattle × cattle_water) (20)
where
TF_beef is the amount of an element taken up into muscle tissue as a function of the
daily intake rate of that element by the animal. (Bq/g per Bq/d)
Cplant-cattle is the area-weighted plant concentration on the cap, gullies and fans, and air-
dispersion area (Bq/g dry wt)
Dose Assessment for the Clive DU PA
6 November 2015 42
cattle_forage is the dry-weight forage intake rate for browsing cattle (g/day dry wt)
Csoil-cattle is the weighted soil concentration on the embankment and air-dispersion areas
(Bq/g)
cattle_soil is the soil ingestion rate for browsing cattle (g/day)
Cwater-cattle is the water concentration for the puddles in the air-dispersion areas (Bq/g)
cattle_water is the water ingestion rate for browsing cattle (g/day)
Concentrations in pronghorn tissue (Cgame) are calculated in a manner analogous to Equation 20,
substituting weighted exposure concentrations and intake rates for pronghorn.
Transfer factors (TFs) determine the amount of an element taken up into muscle tissue as a
function of the daily intake rate of that element by the animal. The units are expressed as Bq/kg
per Bq/d (d/kg). Element-specific beef transfer factors were preferentially obtained from a recent
publication of the International Atomic Energy Agency (IAEA, 2010). A report by Pacific
Northwest National Laboratory (Staven et al., 2003) was used as a secondary reference. For
many elements, these values are reported as a geometric mean and geometric standard deviation.
For a subset of elements with only a single reference, an arithmetic mean is provided with no
measure of variance. In these cases (actinium, americium, neptunium, protactinium, radium, and
technetium), an estimate of variance was produced by taking the average geometric standard
deviation for the all other elements excepting plutonium, which was considered an outlier. A
summary of the beef TFs with accompanying notes is provided in Table 3.
Distributional form for the values of geometric mean and geometric standard deviation reported
in IAEA (2010) was not discussed in this reference. Also, for sample sizes of less than 3, IAEA
(2010) values were originally reported as the arithmetic mean and standard deviation. In order to
provide a common set of inputs, values obtained from IAEA (2010) and Staven et al. (2003)
were processed to conform to an assumed lognormal distribution. Values originally reported as
arithmetic mean and standard deviation were transformed to geometric equivalents.
Beef TF data were reported in IAEA (2010) as a geometric mean, geometric standard deviation,
minimum, and maximum. The geometric standard deviations are greater than 2 in nearly every
case, suggested high right-skewness in the data, and the minimum and maximum were consistent
with samples from a lognormal distribution. In order to establish a distribution for the mean, a
parametric bootstrap approach was taken [Efron 1998], simulating bootstrap samples from the
lognormal distribution using the maximum likelihood estimates of the lognormal parameters. A
lognormal distribution was then fit to the resulting bootstrap simulations of the mean, since some
right-skewness was still present in the sampling distribution.

Dose Assessment for the Clive DU PA
6 November 2015 43
Table 3. Beef transfer factors (Bq/kg per Bq/d)
Element Sample
size
Geometric
Mean
Geometric
Std. Dev. Notes
Actinium 1 0.0004 generic*
Mean based upon Staven et al. (2003; table 2-6, p.
2.7); no value in IAEA (2010). Geometric
standard deviation based upon 6 surrogate
elements.
Americium 1 0.0005 generic* Geometric standard deviation based upon 6
surrogate elements. (IAEA, 2010; table 30, p. 93)
Cesium 58 0.032 1.15 Based upon values provided in IAEA (2010; table
30, p. 93)).
Iodine 5 0.0107 1.85 Based upon values provided in IAEA (2010; table
30, p. 93).
Neptunium 1 0.001 generic*
Mean based upon Staven et al. (2003; table 2.6, p.
2.7); no value in IAEA 2010. Geometric standard
deviation based upon surrogate elements.
Protactinium 1 0.0005 generic*
Americium (IAEA 2010 value) used as a
surrogate based upon Staven et al. (2003; table
2.6, p. 2.7).
Lead 5 0.000952 1.59 Based upon values provided in IAEA (2010; table
30, p. 93).
Plutonium 5 0.0000128 7.42 Based upon values provided in IAEA (2010; table
30, p. 93).
Radium 1 0.0017 generic* Geometric standard deviation based upon
surrogate elements. (IAEA 2010; table 30, p. 93)
Radon -- arbitrarily
small value 1 Radon gas is inert and has effectively no potential
to establish an equilibrium in animal tissue.
Strontium 35 0.00223 1.26 Based upon values provided in IAEA (2010; table
30, p. 93).
Technetium 1 0.0001 generic*
Mean based upon Staven et al. (2003; table 2.6, p.
2.7); no value in IAEA 2010. Geometric standard
deviation based upon surrogate elements.
Thorium 6 0.000355 1.68 Based upon values provided in IAEA (2010; table
30, p. 93).
Uranium 3 0.000421 1.32 Based upon values provided in IAEA (2010; table
30, p. 93).
* A generic GSD for these elements is 1.475

Dose Assessment for the Clive DU PA
6 November 2015 44
5.0 References
BLM (United States Bureau of Land Management) 2010. Rangeland Administration System.
United States Bureau of Land Management, www.blm.gov/ras.
Bogen K.T., Cullen A.C., Frey H.C., Price P. 2009. Probabilistic exposure analysis for chemical
risk characterization. Toxicological Sciences 109(1): 4-17.
Brenner D.J., Doll R., Goodhead D.T., et al. 2003. Cancer risks attributable to low doses of
ionizing radiation: Assessing what we really know. Proc. Natl Acad Sci 100(24): 13761-
6.
Burr S.W., Smith J.W., Reiter D., et al. 2008. Recreational Off-Highway Vehicle Use on Public
Lands in Utah. Utah State University, Institute for Outdoor Recreation and Tourism,
Logan UT.
CFR (Code of Federal Regulations). 1994. National Oil and Hazardous Substances Pollution
Contingency Plan, Code of Federal Regulations 40 CFR 300, US Government Printing
Office, Washington, DC.
CFR. 2007. Licensing Requirements for Land Disposal of Radioactive Waste, Code of Federal
Regulations Title 10, Part 61 (10 CFR 61), US Government Printing Office, Washington,
DC.
Davis S., Mirick D.K. 2006. Soil ingestion in children and adults in the same family. Journal of
Exposure Science and Environmental Epidemiology, 16: 63-75.
Efron B.,Tibshirani R.J. 1998. Introduction to the Bootstrap, CRC Press, Boca Raton, FL.
EPA (United States Environmental Protection Agency). 1988. Limiting Values of Radionuclide
Intake and Air Concentration and Dose Conversion Factors for Inhalation, Submersion,
and Ingestion, EPA 520/1-88-020, Federal Guidance Report (FGR) 11, US
Environmental Protection Agency, Washington, DC.
EPA 1999. Cancer Risk Coefficients for Environmental Exposure to Radionuclides. Federal
Guidance Report No. 13, United States Environmental Protection Agency and Oak Ridge
National Laboratory, EPA 402-R-99-001, Oak Ridge TN.
EPA 2007 (Pawel et al.) Uncertainties in Cancer Risk Coefficients for Environmental Exposure
to Radionuclides: An Uncertainty Analysis for Risk Coefficients Reported in Federal
Guidance Report No. 13, United States Environmental Protection Agency and Oak Ridge
National Laboratory, ORNL/TM-2006/583, Oak Ridge TN.
EPA, 2011a. Integrated Risk Information System (IRIS), Office of Research and Development
and National Center for Environmental Assessment, Electronic Database. Available
online at: http://www.epa.gov/iris/.
EPA. 1989. Risk Assessment Guidance for Superfund, Volume I: Human Health Evaluation
Manual (Part A), Interim Final, Office of Emergency and Remedial Response, US
Environmental Protection Agency, Washington, DC.
Dose Assessment for the Clive DU PA
6 November 2015 45
EPA. 1993a. External Exposure to Radionuclides in Air, Water, and Soil, EPA 402-R-93-81,
Federal Guidance Report (FGR) 12, US Environmental Protection Agency, Washington,
DC.
EPA. 1993b. Wildlife Exposure Factors Handbook. Vol. I. EPA/600/R-93/187, Office of
Research and Development, US Environmental Protection Agency, Washington, DC.
EPA. 1997a, Establishment of Cleanup Levels for CERCLA Sites with Radioactive
Contamination, Memorandum from Stephen D. Luftig, Director, Office of Emergency
and Remedial Response, and Larry Weinstock, Acting Director, Office of Radiation and
Indoor Air, to addressees, OSWER No. 9200.4 18, US Environmental Protection Agency,
Washington, DC.
EPA. 1997b. Exposure Factors Handbook. Office of Research and Development. US
Environmental Protection Agency, Washington, DC.
EPA. 1999. Cancer Risk Coefficients for Environmental Exposure to Radionuclides: Updates
and Supplements, EPA 402-R-99-001, FGR 13, US Environmental Protection Agency,
Washington, DC.
EPA. 2000. National Primary Drinking Water Regulations; Radionuclides; Final Rule, Federal
Register: December 7, 2000 (Volume 65, Number 236)
EPA. 2001, Risk Assessment Guidance for Superfund (RAGS), Volume III: Process for
Conducting Probabilistic Risk Assessment (Part A), EPA 540-R-02-002, OSWER
9285.7-45, Office of Emergency and Remedial Response, US Environmental Protection
Agency, Washington, DC.
EPA. 2005, Human Health Risk Assessment Protocol for Hazardous Waste Combustion
Facilities, Final, EPA/530-R-05-006, Office of Solid Waste and Emergency Response,
US Environmental Protection Agency, Washington, DC.
EPA. 2008, Clear Creek Management Area Asbestos Exposure and Human Health Risk
Assessment, US Environmental Protection Agency, Region 9, San Francisco, CA.
EPA. 2009a. Exposure Factors Handbook: 2009 Update. External Review Draft, July 2009.
Office of Research and Development, National Center for Environmental Assessment,
US Environmental Protection Agency, Washington, DC.
EPA. 2009b. Metabolically Derived Human Ventilation Rates. May 2009. Office of Research
and Development, National Center for Environmental Assessment, US Environmental
Protection Agency, Washington, DC.
Filipsson A.F., Sand S., Nilsson J., et al. 2003. The benchmark dose method- review of available
models, and recommendations for application in health risk assessment. Crit Rev Toxicol
33:505-542.
GTG (GoldSim Technology Group), 2010. GoldSim: Monte Carlo Simulation Software for
Decision and Risk Analysis, http://www.goldsim.com
Hamby D.M. 1999. Uncertainty of the tritium dose conversion factor. Health Phys 77(3): 291-7.
Dose Assessment for the Clive DU PA
6 November 2015 46
Harvey R.P., Hamby D.M., Palmer T.S., 2006. Uncertainty of the thyroid dose conversion factor
for inhalation intakes of I-131 and its parametric uncertainty. Radiat Prot Dosimetry
118(3):296-306.
Hendee W.R., Edwards F.M. 1986. ALARA and an integrated approach to radiation protection.
Seminars in Nuclear Medicine 16:142-150.
Huffman B. 2004. Antilocapra americana. Pronghorn. Ungulate Fact Sheet.
http://www.ultimateungulate.com/Artiodactyla/Antilocapra_americana.html.
IAEA (International Atomic Energy Agency). 2010. Handbook of Values for the Prediction of
Radionuclide Transfer in Terrestrial and Freshwater Environments, International Atomic
Energy Agency Technical Report Series No. 472, Vienna.
ICRP, 1977. Radiation Protection. International Commission of Radiological Protection
Publication No. 26. Permagon, NY.
ICRP. 1979. Limits for Intakes of Radionuclides by Workers, International Commission on
Radiological Protection Publication No. 30, Ann. ICRP 2(3-4).
ICRP. 1983. Cost-Benefit Analysis in the Optimization of Radiation Protection. International
Commission of Radiological Protection Publication No. 37. Permagon, NY.
ICRP. 1989. Age-Dependent Doses from Intakes of Radionuclides: Part 1, International
Commission on Radiological Protection Publication No. 56, Ann. ICRP 20(2).
ICRP. 1991. Recommendations of the International Commission on Radiological Protection I;
International Commission on Radiological Protection Publication No. 60. Ann. ICRP 21
(1–3)
ICRP. 1995. Age-Dependent Doses to Members of the Public from Intake of Radionuclides, Part
5: Compilation of Ingestion and Inhalation Dose Coefficients, International Commission
on Radiological Protection Publication No. 72, Ann. ICRP 26(1). P 072 errata in SG 03
JAICRP 32 (1-2).
ICRP, 2006. The Optimisation of Radiological Protection – Broadening the Process,
International Commission on Radiological Protection Publication 101b, Ann. ICRP 36
(3)6.
ICRP. 2009. ICRP Statement on Radon and Lung Cancer Risk from Radon and Progeny, Draft
Report for Consultation,
http://www.icrp.org/docs/ICRP_Statement_on_Radon_AND_Lung_cancer_risk_from_ra
don_and_progeny%28for_consultation%29.pdf.
Kocher D.C., Apostoaei A.I., Hoffman F.O. 2005. Radiation effectiveness factors for use in
calculating probability of causation of radiogenic cancers. Health Phys 89(1): 3-32.
Linkov I., Burmistrov D. 2001. Reconstruction of doses from radionuclide inhalation for nuclear
power plant worker using air concentration measurements and associated uncertainties.
Health Phys 81(1): 70-75.
MSUE (Michigan State University Extension). 2011. Agricultural Water Use Reporting.
Michigan State University Extension. Available at
web1.msue.msu.edu/waterqual/WQWEB/Beef.doc.
Dose Assessment for the Clive DU PA
6 November 2015 47
National Research Council 2006. Health Risks from Exposure to Low Levels of Ionizing Radia-
tion, BEIR VII Phase II, National Academies Press, Washington DC.
NCRP (National Council on Radiation Protection and Measurements). 1996. A Guide for
Uncertainty Analysis in Dose and Risk Assessments Related to Environmental
Contamination. National Council on Radiation Protection and Measurements
Commentary No. 14, Bethesda, MD.
NCRP 1996, A Guide for Uncertainty Analysis in Dose and Risk Assessments Related to Envi-
ronmental Contamination, NCRP Commentary No. 14, National Council on Radiation
Protection and Measurements, Bethesda MD.
NCRP 1998, Evaluating the Reliability of Biokinetic and Dosimetric Models and Parameters
used to Assess Individual Doses for Risk Assessment Purposes, NCRP Commentary
No. 15, National Council on Radiation Protection and Measurements, Bethesda MD.
NCRP 2007, Uncertainties in the Measurement and Dosimetry of External Radiation, NCRP
Report No. 158, National Council on Radiation Protection and Measurements, Bethesda
MD.
NCRP 2009, Uncertainties in Internal Radiation Dose Assessment, NCRP Report No. 164, Na-
tional Council on Radiation Protection and Measurements, Bethesda MD.
NCRP 2012, Uncertainties in the Estimation of Radiation Risks and Probability of Disease Cau-
sation, NCRP Report No. 171, National Council on Radiation Protection and Measure-
ments, Bethesda MD.
NRC (United States Nuclear Regulatory Commission). 1993. Final Environmental Impact
Statement to Construct and Operate a Facility to Receive, Store, and Dispose of 11e.(2)
Byproduct Material Near Clive, Utah, NUREG 1476, US Nuclear Regulatory
Commission, Washington, DC.
NRC. 2000. A Performance Assessment Methodology for Low-Level Radioactive Waste Disposal
Facilities, NUREG 1573, US Nuclear Regulatory Commission, Washington, DC.
Puncher, M. and J.D. Harrison 2012, Assessing the Reliability of Dose Coefficients for Ingestion
and Inhalation of Radionuclides by Members of the Public, Health Protection Agency,
Center for Radiation, Chemical and Environmental Hazards, Chilton, Didcot, Oxford-
shire OX11 0RQ, HPA-CRCE-048, ISBN 978-0-85951-741-6.
Puncher, M. and J.D. Harrison 2013, Assessing the Reliability of Dose Coefficients for Inhaled
and Ingested Radionuclides. Journal of Radiological Protection 32:223-241.
Scott B.R. 2008. It's time for a new low-dose-radiation risk assessment paradigm- one that
acknowledges hormesis. Dose-Response 6:333-351.
Staven L.H., Napier B.A., Rhoads K., Strenge DL. 2003. A Compendium of Transfer Factors for
Agricultural and Animal Products, Pacific Northwest National Laboratory, Richland
WA.
Tuli J. K., 2005. Nuclear Wallet Cards, 7th Edition, Brookhaven National Laboratory, Upton,
NY.
Dose Assessment for the Clive DU PA
6 November 2015 48
UDWR (Utah Division of Wildlife Resources) 2009. Utah Pronghorn Statewide Management
Plan., Utah Department of Natural Resources, Salt Lake City UT.
USFS (United States Forest Service), 2005. Off-Highway Vehicle Recreation in the United
States, Regions, and States: A National Report from the National Survey on Recreation
and the Environment. June, 2005. US Forest Service, University of Georgia; Athens GA.
USFWS (United States Fish and Wildlife Service) 2006. National Survey of Fishing, Hunting,
and Wildlife-Associated Recreation: Utah. US Fish and Wildlife Service, US
Department of Commerce, and US Census Bureau; Washington, DC.
Utah, State of, 2015. Utah Administrative Code Rule R313-25. License Requirements for Land
Disposal of Radioactive Waste - General Provisions. As in effect on September 1, 2015.
(http://www.rules.utah.gov/publicat/code/r313/r313-025.htm, accessed 5 Nov 2015).

Dose Assessment for the Clive DU PA
6 November 2015 49
Appendix I: Discussion of Derivations of
Selected Parameter Distributions
Distribution development utilized data where available, and exercised professional judgment
where it was not available. For the parameter distributions discussed below, unless specified
otherwise, the approach followed the Probability Distribution Development white paper
(Appendix 14).
Age: Based upon the observed age quantile breakdown reported in USFS (2005) for recreational
receptors, ignoring the age groups outside of the defined adult age range 16-60. For simplicity,
and because age data specific to ranchers in the vicinity of Clive were unavailable, the same age
distribution was also used for rancher receptors. The age range corresponds to bins used to
aggregate ventilation rate data by EPA (2009b).
Gender: Based upon the observed percentage in USFS (2005).
Body Weight: EPA (2009a) reports body weights as quantiles, broken down by various age and
gender categories. Mean body weight changes gradually with age, and is significantly different
between genders. A lognormal distribution was fit for each gender separately, with the log of the
geometric mean was fit as a constant, a linear function of age, and a quadratic function of age,
using the quantile likelihood fitting described in the Probability Distribution Development white
paper (Appendix 14). The quadratic model produced the best fit, capturing the mean decrease in
the population for the oldest age group:
𝜇=𝛽!+𝛽!Age +𝛽!Age! (21)
where 𝑒!is the geometric mean.

Dose Assessment for the Clive DU PA
6 November 2015 50

Figure 1. Geometric mean of body weight as a function of age.

Figure 2. Examples of distributions for body weight.
Dose Assessment for the Clive DU PA
6 November 2015 51
Ventilation Rate: EPA (2009a) reports inhalation rates as quantiles, broken down by various
activity, age, and gender categories. The data are reported as both weight-adjusted and non-
weight-adjusted inhalation rates. In order to incorporate correlation in inhalation rates between
activity categories, the weight-adjusted data are utilized. That is, a weight-adjusted inhalation
rate will be simulated for each activity level, and then the single simulated body weight for the
individual is multiplied by the weight-adjusted inhalation rates to obtain the inhalation rates:
𝑉!,!=𝑉!,!
(!")⋅𝐵𝑊 (22)
where 𝑉!,!is inhalation rate for activity level i in m3/min, 𝑉!,!"is body-weight adjusted
inhalation rate for activity level i in m3/kg-min, and 𝐵𝑊is body weight in kg. This approach to
constructing inhalation rate is similar to the approach taken in EPA (2009b). Inhalation rate is
significantly different between genders, and mean ventilation rate changes gradually with age. A
lognormal distribution was fit for each gender separately. The log of the geometric mean was fit
as a constant, a linear function of age, and a quadratic function of age; using the quantile
likelihood fitting described in the Probability Distribution Development for the Clive PA white
paper (Appendix 14). None of these models adequately characterized the data, as the 16-20 age
group is significantly different from the 21-30 age group. As such, the 16-20 age group was fit
separately from the remaining data, and a linear fit was adequate for the remaining age ranges.

Figure 3. Geometric means for ventilation rate, as a function of age and gender.
Dose Assessment for the Clive DU PA
6 November 2015 52

Figure 4. Examples of ventilation rate distributions for different activities (20-year-old
male).

Dose Assessment for the Clive DU PA
6 November 2015 53
Soil Ingestion Rate: EPA (2009a) reports soil ingestion for adults only as a mean, median, and
standard deviation. The distribution derived here is based upon the only careful study of adult
ingestion that has been conducted to date (Davis and Mirick 2006), identified as a key study in
EPA (2009a). Three tracer elements (aluminum, silicon, and titanium) used in Davis and Mirick
(2006) provide different bases for quantifying soil ingestion rate. The data distribution is
significantly different for the three tracer elements. Thus, rather than combine data across the
three tracers, a separate distribution of soil ingestion is established for each tracer. Because there
was no significant difference between genders, males and females were combined. Given the
significant skew in the data (means much larger than the medians), a lognormal model was fit to
the combined data based using maximum likelihood estimates.

Figure 5. Distributions for soil ingestion, representing different tracers.

Dose Assessment for the Clive DU PA
6 November 2015 54
Ingestion Rates, Home-produced Meat (beef): EPA (2009a) reports quantiles of the body-weight-
adjusted average intake per day of home-produced meat, broken down by age and type of meat.
The age groups given do not correspond perfectly to the range of ages considered in this PA
model. Thus, the 20-39 age group was used to represent the 16-39 age group, and the 40-69 age
group was used to represent the 40-60 age group. The distributions were significantly different
for the two age groups, so they were fit separately. The lognormal distribution provided a good
fit to the center of the data, but had poor tail behavior in each case. Thus, a gamma distribution
was chosen instead, which provided a better overall fit.

Figure 6. Distributions for home-produced meat ingestion rates.

Dose Assessment for the Clive DU PA
6 November 2015 55
Activity-Based Exposure Time: EPA (2009a) document reports average time per day spent at
different levels of activity as quantiles for adults, broken down by age and gender. The quantiles
are reported independently for each activity level, and thus no information regarding the
correlation between the times is available. Correlation must exist, as an individual's daily
averages must exist on the simplex that sums to 24 hours. Dirichlet distributions are the only
standard statistical model that provides a distribution on a simplex. However, Dirichlet
distributions could not achieve the long tails observed in the distributions for the more active
levels. In order to achieve the tail behavior, the following approach was used. A lognormal
model was fitted for combined sleeping and sedentary time (constrained to be no more than 24
hours). Sleeping time alone was also fitted as a lognormal model and constrained to be smaller
than combined sleeping and sedentary time. Remaining average time per day was then
partitioned into light, medium, and heavy activities. A lognormal distribution was fit to each, but
for simulation purposes, the three values are simulated and then normalized to sum to time per
day remaining. The resulting distribution induces moderate negative correlation amongst the
time spent in each activity level; the greatest negative correlation existing between light and
medium activity durations. The tail behavior of medium and heavy activity durations is reduced
from that observed in the data (i.e., the upper percentiles are slightly lower than observed).
However, without the detailed correlation structure of the data, a simple model is unlikely to
both meet the constraints of the simplex and match the tail behavior.

Figure 7. Example distributions for sedentary plus sleeping time/day and sleeping time/day
(30-year-old female).
Dose Assessment for the Clive DU PA
6 November 2015 56

Figure 8. Distributions for light, medium, and heavy activity time/day (30-year-old female).

Dose Assessment for the Clive DU PA
6 November 2015 57
Numbers of Individuals in Vicinity of Site – Personal communication with BLM staff (Salt Lake
Field Office) provided 100 and 500 as bounds and 350 as a best guess. These might be
interpreted as 5th and 95th percentiles, along with a mean or median. However, due to the
informal nature of the conversation and a programming need to have a fixed upper bound on this
distribution, these will be treated as bounds, making a triangular distribution a reasonable
representation of the information.

Figure 9. Distribution for the total number of individuals at the site during a given year.

Receptor Type – The individuals in the vicinity of the site are partitioned into Ranchers, Hunters,
and Sport OHVers. The distribution for the number of Ranchers was based upon professional
judgment and the size of leases, and is independent of the total number of individuals within
vicinity of the site. The remaining individuals are then partitioned into Hunters and Sport
OHVers by utilizing a binomial distribution with the proportion of hunters equal to 0.25, the
value reported from the large survey in USFS (2005).

Dose Assessment for the Clive DU PA
6 November 2015 58
Sport OHVer Day-Trip Time in Area – The only reported value from the Sport OHVer survey
was a mean of 6.3 hr/day. The standard deviation is not reported, so professional judgment was
used to choose a standard deviation.

Figure 10. Distribution for the average day-trip time.

Dose Assessment for the Clive DU PA
6 November 2015 59
OHV Dust Loading – Summary data from EPA (2008) are available both for ambient conditions
(CCMA) and near ATV riders. Means are given, and standard errors for the mean can be
approximated from the upper confidence limit (UCL) values, by assuming a t-UCL. The standard
errors are high relative to the mean, so each of these distributions was treated as lognormal.
These two distributions were then simulated and a ratio taken, to obtain a distribution on the
ratio. The resulting distribution is also approximately lognormal. Figure 11 shows the simulated
values, along with the fitted distribution.

Figure 11. Distribution for dust loading (overlaid on a histogram of simulated values).

Dose Assessment for the Clive DU PA
6 November 2015 60
Rancher Exposure Frequency – Grazing leases are granted for 180 days each year, giving a natural upper
bound for the distribution. There is little other information available to develop a distribution, so
professional judgment was used, and a distribution was chosen that has most Ranchers spending a high
proportion of the allotted 180 days on site, but allows for Ranchers that spend weekends off-site, do not
utilize their full lease, etc.

Figure 12. Distribution for Rancher exposure frequency.

Dose Assessment for the Clive DU PA
6 November 2015 61
Sport OHVer Exposure Frequency – The USFS (2005) document reports a confidence interval for the
mean exposure frequency, which can be used to calculate the standard deviation of the exposure
frequency. Because the standard deviation is larger than the mean, a lognormal model was used to match
the observed mean and standard deviation from the survey data.

Figure 13. Distribution for Sport OHVer exposure frequency.

Dose Assessment for the Clive DU PA
6 November 2015 62
Hunter Exposure Frequency – The USFWS (2006) provides a mean estimate of 10 d/yr, but does not
provide any other summary information. It may be reasonable to assume that this distribution has a
similar shape as the exposure frequency for Sport OHVers; i.e., a right-skewed distribution that has most
of the population spending a relatively small amount of time, with a few individuals who dedicate a great
deal of time to the activity. Thus, a lognormal distribution was chosen with a mean of 10 d/yr, and a
geometric standard deviation that matches the Sport OHVer geometric standard deviation.

Figure 14. Distribution for Hunter exposure frequency.

Dose Assessment for the Clive DU PA
6 November 2015 63
Rest Area Caretaker Exposure Frequency – The distribution for this parameter was based on professional
judgment. The maximum was conservatively set to the maximum possible exposure of 365 days per year.
The mode is set to the EPA default exposure value of 350 days per year, and the minimum allows for 28
days of vacation plus 10 holidays for which the caretaker would be off-site.

Figure 15. Distribution for rest area caretaker exposure frequency.

Dose Assessment for the Clive DU PA
6 November 2015 64
Meat Loss – EPA (1997b) provides information on the amount of meat lost in preparation and in post-
cooking. An average and a standard deviation are reported for the mean loss. As the distribution of
interest represents uncertainty about the mean, the average and standard deviation were used for a normal
distribution.

Figure 16. Distributions for meat loss (preparation and post-cooking).

Dose Assessment for the Clive DU PA
6 November 2015 65
Cattle Range Acreage – There are only four data points available (the four leases in the Clive
area), but because the distribution of the mean acreage is desired, the mean and standard error of
the mean are used to define a normal distribution.

Figure 17. Distribution for the average cattle range acreage.

Dose Assessment for the Clive DU PA
6 November 2015 66
Miscellaneous Uniform Distributions – For many of the parameters, little information is available that is
specific to the Clive facility site. A default distribution in such a case was a uniform distribution over a
range of theoretical values, or from the minimum and maximum values found in literature. The uniform
distribution is generally a poor representation of uncertainty but has the advantage of spreading its mass
across a range of possible values. These uniform distributions are used as defaults until a sensitivity
analysis can be performed to demonstrate whether further data collection is needed to construct a better
representation of uncertainty.
REF Distributions - Kocher (2005) utilized lognormal distributions to represent the uncertainty in REF
parameters. Thus, lognormal distributions were fit to the reported 2.5th, 50th, and 97.5th percentiles of
these distributions.

Figure 18. Distribution for alpha particle REF.
Dose Assessment for the Clive DU PA
6 November 2015 67

Figure 19. Distribution for electron and photon REFs.

Uranium oral reference dose – EPA has two published values for this value: EPA (2011) and EPA
(2000). These two sources are considered equally viable, so each is selected with 50% probability.

NAC-0028_R2

Decision Analysis Methodology
for Assessing ALARA Collective
Radiation Doses and Risks

Clive DU PA Model v1.4
6 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 ii
1. Title: Decision Analysis Methodology for Assessing ALARA Collective Radiation
Doses and Risks
2. Filename: Decision Analysis v1.4.docx
3. Description: This White Paper describes the details of the Clive DU PA model ALARA
analysis, which is based on population dose.

Name Date
4. Originator Robert Lee 31 October 2015
5. Reviewer Paul Black 6 November 2015
6. Remarks
Original v1.0 in May 2010

Subsequent v1.2 in May 2014

Revisions made for v1.4 included changing the dollar cost per person rem per updated NRC
guidance.

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 iii
CONTENTS
1.0 Introduction ............................................................................................................................ 1
2.0 ALARA .................................................................................................................................. 3
3.0 Development of Current ALARA Cost per Person-Rem Estimates ...................................... 4
3.1 History .............................................................................................................................. 4
3.2 Estimating Value of a Statistical Life ............................................................................... 7
3.3 Risk Coefficient Estimates ............................................................................................... 8
3.4 Current NRC Recommendations .................................................................................... 10
3.5 Approach for the Clive ALARA Analysis ..................................................................... 11
4.0 Decision Analysis ................................................................................................................. 11
5.0 Scope of ALARA Decision Analysis for the Clive Depleted Uranium Performance
Assessment ........................................................................................................................... 13
6.0 References ............................................................................................................................ 15

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 iv

This page is intentionally blank, aside from this statement.

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 1
1.0 Introduction
The safe storage and disposal of depleted uranium (DU) waste is essential for mitigating releases
of radioactive materials and reducing exposures to humans and the environment. Currently, a
radioactive waste facility located in Clive, Utah (the “Clive facility”) operated by the company
EnergySolutions Inc. is being considered to receive and store DU waste that has been declared
surplus from radiological facilities across the nation. The Clive facility has been tasked with
disposing of the DU waste in a manner that protects humans from future radiological releases.
To assess whether the proposed Clive facility location and containment technologies are suitable
for protection of human health, specific performance objectives for land disposal of radioactive
waste set forth in Utah Administrative Code (UAC) Rule R313-25-9 and Title 10 of the Code of
Federal Regulations (CFR) Part 61 (10 CFR 61) Subpart C, promulgated by the Nuclear
Regulatory Commission (NRC), must be met. In order to support the required radiological
performance assessment (PA), a detailed computer model has been developed to evaluate the
doses to human receptors that would result from the disposal of DU and associated radioactive
compounds (collectively termed “DU waste”), and conversely to determine how much DU waste
can be safely disposed at the Clive facility.
The Neptune and Company, Inc. (Neptune) white paper Dose Assessment (Appendix 11) details
the methods for estimating radiation doses to future human receptors associated with DU waste
and its decay products. Both the NRC and UAC Rule R313-25-9 specify clear performance goals
of 25 mrem/yr for individual members of the public (MOP) and 500 mrem/yr for inadvertent
human intruders (IHI) within a 10,000-year compliance period. These goals are the result of a
complex balance of risk and feasibility, and are not specifically addressed here because they are
(at present and in a practical sense) inflexible and non-negotiable.
However, the CFR (Section 61.42) and UAC Rule R313-25-9 also define a second decision rule
that pertains to populations as well as individuals. The CFR regulation states "reasonable effort
should be made to maintain releases of radioactivity in effluents to the general environment as
low as is reasonably achievable" (or ALARA). Ionizing radiation protection limits have been
utilized since the 1920s, but the concept of keeping radiation doses as low as practicable or
achievable was an outgrowth of worker safety in the nuclear weapons development industry
(Hendee and Edwards, 1987).
The ALARA process is described in DOE regulations and associated guidance documents such
as 10 CFR Part 834 and DOE 5400.5 ALARA (10 CFR 834; DOE 1993, 1997), in NRC
regulations (10 CFR 20.1003, 10 CFR 61.42), and in other NRC documents (NRC, 1995, 2000a,
2015). The definitions in each case are very similar; indicating that exposures should be
controlled so that releases of radioactive material to the environment are as low as is reasonable
taking into account social, technical, economic, practical, and public policy considerations. It is
also noted that ALARA is not a dose limit, but rather a process, which has the objective of
attaining doses as far below the applicable limit of this part as is reasonably achievable.
The ALARA concept was first described in publication in ICRP (1973), following similar
concepts that date back to ICRP publications at least as early as 1959 (ICRP, 1959). Updates
have been provided by the ICRP in 1977 (ICRP, 1977), and more recently in 2006 (ICRP, 2006).
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 2
In this latest report, the ICRP focuses more on expanding the optimization process. This includes
evaluating different relatively homogeneous population groups, stakeholder involvement in
addressing receptor scenarios, site-specific evaluation of exposure, intergenerational equity, and
many other aspects. The ICRP report provides a comprehensive list of factors that should be
considered for optimization. However, the ICRP stops short of describing a methodology for
implementation, even suggesting that full quantification of all relevant factors is not possible.
However, with modern decision analysis methods this need not be the case (e.g., Keeney, 1992;
Gregory et al., 2012). The Office of Management and Budget (OMB, 1992) also provides a road
map for applying a decision analysis approach to policy analysis that could be adapted to PA.
Another obstacle that is recognized in ICRP (2006), is that lack of regulatory support for such an
approach. However, the ALARA principle exists in both DOE and NRC regulations and
guidance, decision analysis methods exist to implement the intended optimization, and there
appears to be some traction now with both DOE and NRC regarding decision analysis methods
for optimization, or ALARA.
In terms of the ALARA analysis performed for the Clive DU PA, it does not achieve all that the
ICRP calls for. This is primarily because the regulatory support for doing so does not clearly
exist. However, as ICRP has made clear, this is an approach that will help focus decision-making
on finding optimal solutions. To implement this approach to ALARA a paradigm shift is needed
in the industry, starting with the regulators, so that the focus is on optimal use of the US’s limited
disposal resources as opposed to somewhat arbitrary compliance decisions. ICRP (2006)
recognizes this same need. For the current PA the approach has included evaluation of specific
relatively homogeneous receptor groups, and has included a metric for evaluating potential costs
for the simulated doses. It has not engaged many of the other recommendations of the ICRP.
The words "reasonably" and "achievable" in ALARA are not precise. The two words imply some
degree of consideration of tradeoffs, but no clear definition is published. Assuming that there are
trade-offs, then this implies that an analysis should be performed that explicitly evaluates the
trade-offs and how different disposal options, designs, or sites may differentially satisfy the
objectives and resource constraints (e.g., a decision or economic analysis). Yet, at present, there
is limited specific guidance on how to apply ALARA principles to the PA process.
The ALARA concept can be thought of as a cost-benefit trade-off that requires an evaluation of
human health risk and the costs of achieving those risks. In the context of the Clive DU PA,
calculations that would be needed to support a more complete ALARA analysis are performed
for collective doses germane to the receptor populations described in Dose Assessment (Appendix
11). That is the costs of the population doses are calculated based on the modeled doses and the
cost per person rem specified in the relevant NRC and DOE guidance.
The remainder of this discussion will focus upon the concepts of population dose/risk and
ALARA, and how these can be integrated into a Bayesian decision analysis (DA) for application
to the Clive facility.

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 3
2.0 ALARA
The ALARA concept, as germane to radiation protection for both individual and population
(collective) levels, was described as follows by the ICRP in 1977 (ICRP, 1977):
"Most decisions about human activities are based on an implicit form of balancing of
costs and benefits leading to the conclusion that the conduct of a chosen practice is
'worthwhile.' Less generally, it is also recognized that the conduct of the chosen practice
should be adjusted to maximize the benefit to the individual or to society. In radiation
protection, it is becoming possible to formalize these broad decision-making procedures."
The ICRP (1977) basically recommended a system of radiation protection that included the
following principles:
 No practice shall be adopted unless its introduction produces a positive net benefit –
justification of the practice.
 All exposures shall be kept as low as reasonably achievable, economic and social factors
being taken into account – optimization of radiation protection.
 The dose equivalent to individuals shall not exceed the limits recommended for the
appropriate circumstances by the Commission – the limits of individual dose assessment.
In other words, ICRP defined radiation protection in the context of decision analysis, at least in
terms of the first two principles, considering health, economic, and social objectives; and invoked
the concept of net benefit. The third principle can, instead, be interpreted as a compliance
objective, so that the decision analysis can only be performed for decision options that first
comply with regulatory performance objectives.
The ALARA process is also described in DOE regulations and associated guidance documents
such as 10 CFR Part 834 and DOE 5400.5 ALARA (10 CFR 834; DOE 1993, 1997), and in
various NRC documents such as NRC, 1995, 2000a, and 2015. The definitions in each case are
very similar; indicating that exposures should be controlled so that releases of radioactive
material to the environment are as low as is reasonable taking into account social, technical,
economic, practical, and public policy considerations. 10 CFR 834 further describes the ALARA
process as a “logical procedure for evaluating alternative operations, processes, and other
measures, for reducing exposures to radiation and emissions of radioactive material into the
environment, taking into account societal, environmental, technological, economic, practical and
public policy considerations to make a judgment concerning the optimum level of public health
protection”. Although 10 CFR 834 is not aimed specifically at disposal of radioactive waste, the
basic goals are protection of the public from DOE activities, of which radioactive waste disposal
is one such activity.
NRC also provides guidance on application of the principle of ALARA. For example, although
the context is different, 10 CFR Part 20 provides guidance that suggests – “Reasonably
achievable” is judged by considering the state of technology and the economics of improvements
in relation to all the benefits from these improvements (NRC, 2008). NRC also notes that “...a
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 4
comprehensive consideration of risks and benefits will include risks from non-radiological
hazards”.
The overall implication of the various Agency regulations and guidance documents regarding
ALARA is that many factors should be taken into account when considering the potential
benefits of different options for disposal of radioactive waste. In order to implement ALARA in a
logical system, and so that economic factors are taken into consideration, a decision analysis is
implied. Decision analysis is the appropriate mechanism for evaluating and optimizing disposal,
closure and long term monitoring and maintenance of a radioactive waste disposal system.
Decision options for disposal at Clive might include engineering options and waste placement.
More generally, if decision analysis is applied, then a much wider range of options can be
factored into the decision model, such as transportation of waste, risk to workers, and effect on
the environment. However, for the Clive DU PA, the focus is on understanding the dose-based
costs associated with different options for waste disposal within the current proposed
configuration of the Federal DU Cell.
3.0 Development of Current ALARA Cost per Person-Rem
Estimates
3.1 History
The decision analysis context for radioactive waste disposal is essentially a benefit-cost analysis,
within which the dose costs associated with different options for the placement of waste are
ideally evaluated. In practice, for each option the PA model predicts doses to the array of
receptors, and the consequences of those doses are assessed as part of an overall cost model,
which also includes the costs of disposal of waste for each option. The goal is to find the best
option, which is the option that provides the greatest overall benefit.
The concept of assigning a monetary value to radiation dose in regulatory decision-making arose
in 1974 during a hearing for a rulemaking addressing routine effluent releases from nuclear
power reactors. The subsequent rule was Title 10 of the Code of Federal Regulations (10 CFR),
Part 50, “Domestic Licensing of Production and Utilization Facilities,” Appendix I, “Numerical
Guides for Design Objectives and Limiting Conditions for Operation To Meet the Criterion ‘As
Low As Is Reasonably Achievable’ for Radioactive Material in Light-Water-Cooled Nuclear
Power Reactor Effluents.” In adopting design criteria for limiting routine effluent releases from
power plants, NRC promoted the use of a cost-benefit test (NRC, 1975a):
“Such a cost-benefit analysis requires that both the costs and the benefits from the
reduction in dose levels to the population be expressed in commensurate units, and it
seems sound that these units be units of money. Accordingly, to accomplish the cost-
benefit balancing, it is necessary that the worth of a decrease of a person-rem be assigned
monetary values.”
NRC stated that “the record, in our view, does not provide an adequate basis to choose a specific
dollar value for the worth of decreasing the population dose by a man-rem.” Published studies
that were reviewed provided values ranging from $10 to $980 per person-rem. NRC concluded
that “there is no consensus in this record or otherwise regarding the proper value for the worth of
a man-rem,” and “we also recognize that selection of such values is difficult since it involves, in
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 5
addition to actuarial considerations that are commonly reduced to financial terms, aesthetic,
moral, and human values that are difficult to quantify” (NRC, 1975a). The final outcome was a
decision to adopt the value of $1,000 per person-rem as an interim measure (NRC, 1975a).
Two executive orders (EO) issued in 1977 (EO 11821 and EO 11949) encouraged Federal
agencies to perform value-impact (now called cost-benefit or benefit-cost) evaluations of
proposed regulatory requirements to demonstrate adequate justification for new requirements.
The NRC adopted this type of evaluation and issued their “Value-Impact Analysis Guidelines”
(NRC, 1977). This document referred to the techniques and detailed consequence analyses used
in the “Reactor Safety Study: An Assessment of Accident Risks in U.S. Commercial Nuclear
Power Plants (WASH-1400),” and recommended that the person-rem avoided as a result of
proposed changes should be multiplied by $1,000 per person-rem in order to place the benefit in
the same units as the costs (NRC, 1975b). Also in 1977, Congress added Section 210 to the
Energy Reorganization Act of 1974, directing the NRC to develop a plan for the identification
and analysis of unresolved safety issues relating to nuclear reactors. In response, the NRC
developed a program for the prioritization and resolution of unresolved safety issues and generic
issues. In 1982, the NRC issued guidance relating to the assignment of priorities with the
publication of “A Prioritization of Generic Safety Issues,” NUREG-0933 (NRC, 1982). NUREG-
0933 used $1,000 per person-rem value in setting the priority of unresolved safety issues and
more generic issues. Issues identified as high priority were then subject to resolution employing a
more detailed cost-benefit analysis that also applied the $1,000 per person-rem value.
In February 1981, EO 12291 was issued, which directed executive agencies to prepare a
regulatory impact analysis for all major rules and stated that regulatory actions should be based
on adequate information concerning the need for and consequences of any proposed actions. EO
12291 directed that actions were not to be undertaken unless they resulted in a net positive
benefit to society. As an independent agency, the NRC was not required to comply with EO
12291. NRC, however, noted that its established regulatory review procedures included an
evaluation of proposed and existing rules in a manner consistent with the regulatory impact
analysis provisions of EO 12291. NRC determined that clarifying and formalizing the existing
NRC cost-benefit procedures for the analysis of regulatory actions would advance the purposes of
regulatory decision-making.
In January 1983, the NRC published NUREG/BR-0058, “Regulatory Analysis Guidelines of the
US Nuclear Regulatory Commission”, followed in December 1983 by publication of
NUREG/CR-3568, “A Handbook for Value-Impact Assessment” (NRC, 1983a and 1983b,
respectively). These documents were issued to formalize NRC’s policies and procedures for
analyzing the costs and benefits of proposed regulatory actions. The $1,000 per person-rem figure
was not mentioned in the first revision of the Guidelines issued in May 1984, however,
NUREG/CR-3568 recommended that a range of values should be used, one of which should be
the $1,000 per person-rem value. As NUREG/CR-3568 provides implementation guidance for
performing regulatory analyses, it became standard practice of the NRC staff to apply this
guidance whenever a quantitative regulatory analysis or cost-benefit analysis was performed.
In 1983, NRC issued an interim Policy Statement on Safety Goals for the Operation of Nuclear
Power Plants for use during a two-year trial period (NRC, 1983c). In this statement, NRC
adopted qualitative and quantitative design goals for limiting individual and societal risks from
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 6
severe accidents. Also in this policy statement, NRC stated the benefit of an incremental
reduction of societal mortality risks should be compared with the associated costs on the basis of
$1,000 per person-rem averted as one consideration in decisions on safety improvements. The
value proposed was in 1983 dollars and was to be modified to reflect general inflation in the
future. As a result of comments on this interim policy statement, the $1,000 per person-rem value
was deleted in the Final Policy Statement on Safety Goals when published in August 1986 (NRC,
1986).
In 1985, the NRC staff revisited the $1,000 per person-rem valuation and its use in regulatory
analyses of nuclear power plant improvements designed to enhance safety. Although the
monetary value of averted person-rem of radiation exposure up to that time referred only to
averted health effects (such as averted latent cancer fatalities), the use of $1,000 per person-rem
was evaluated and defined at that time as a surrogate for all averted offsite losses, such as health
and property. The basis for this determination is documented in a memorandum from the NRC
Executive Director for Operations dated October 23, 1985 (NRC, 1985).
In 1995, the NRC revisited the $1,000 per person-rem value again and issued “Reassessment of
NRC’s Dollar per Person-Rem Conversion Factor Policy,” NUREG-1530 (NRC, 1995a). This
report updated the dollar per person-rem conversion factor to $2,000 per person-rem. The $2,000
per person-rem conversion factor served only as a dollar proxy for the health effects associated
with a person-rem of dose. Offsite property damage costs were no longer included within the
$2,000 per person-rem value. Separate estimates of the offsite costs were now necessary to
account for impacts beyond human health impacts. The dollar per person-rem estimate was
derived from a value of a statistical life (VSL; see below) of $3 million in 1995 dollars,
multiplied by a risk coefficient for stochastic health effects (see below) of 7.3 x 10-4 per person-
rem rounded to the nearest thousand. The VSL amount was derived using a willingness-to-pay
(WTP) method that reflected median values estimated in numerous studies. This process was
similar to the approaches used by other Federal agencies responsible for public health and safety
(NRC, 1995a). The risk coefficient for stochastic health effects as a result of radiation exposure
was taken from the International Commission on Radiation Protection (ICRP) Publication No. 60
(ICRP, 1991). This risk coefficient includes both mortality (e.g., fatal cancers) and morbidity
(e.g., nonfatal cancers and hereditary effects).
In July 2000, the NRC issued revision 3 to the “Regulatory Analysis Guidelines of the US
Nuclear Regulatory Commission” (NRC, 2000b), and in September 2004, the NRC issued
revision 4 (NRC, 2004). This revision reflects economic evaluation guidance provided in Office
of Management and Budget’s (OMB) Circular A-4, published in September 2003 (OMB, 2003).
In 2010, as discussed in “Consideration of Economic Consequences within the US Nuclear
Regulatory Commission’s Regulatory Framework,” NRC staff recommended updating numerous
guidance documents, including NUREG-1530 (NRC, 2012). This was approved in 2013. NRC
has routinely used the $2,000 per person-rem value from the original revision of NUREG-1530
and, on a case-by-case basis, used other dollar per person-rem values to understand the sensitivity
of this parameter on the resulting cost and benefit estimates.
Application of discount rates, which assume that present individuals and populations assign less
“worth” to future benefits, risks, and costs, has been inconsistent and controversial in radioactive
waste regulation. Typically, economists apply discount rates for short-term decisions, as there is
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 7
ample experimental evidence to support this. However, discounting for the extreme time horizons
associated with radioactive waste disposal has not been fully evaluated. If even small (e.g., 3%,
which is a typical lower bound currently employed by OMB in their economic analyses) discount
rates are applied to the problem of radioactive waste disposal, the “value” of future lives reduces
to essentially zero in a few hundred years. It is unclear, without conducting extensive surveys and
research, whether stakeholders truly believe that peoples’ lives a few hundred years from now are
essentially worthless. NRC, in its latest (2015) guidance, does not mention discount rates, likely
because application of such would be highly controversial. The assumption made for the Clive
DU PA model is that the discount rate is zero, thus assigning as much worth to future populations
as to the present population. This is likely a highly conservative assumption.
3.2 Estimating Value of a Statistical Life
The dollar per person-rem conversion factor for health effects is calculated as the product of the
value of a statistical life (VSL) and the risk coefficient for stochastic radiation effects.
The VSL (and therefore the associated dollar per person-rem conversion factor) corresponds to
society’s willingness-to-pay (WTP) for small reductions in a particular mortality risk. VSL is not
a measurement or valuation of a human life. OMB Circular A-4 states (OMB, 2003):
“Some describe the monetized value of small changes in fatality risk as the “value of
statistical life” (VSL) or, less precisely, the “value of a life.” The latter phrase can be misleading because it suggests erroneously that the monetization exercise tries to place a
“value” on individual lives. You should make clear that these terms refer to the
measurement of willingness to pay for reductions in only small risks of premature death.
They have no application to an identifiable individual or to very large reductions in
individual risks. They do not suggest that any individual’s life can be expressed in monetary terms. Their sole purpose is to help describe better the likely benefits of a regulatory action.
Confusion about the term “statistical life” is also widespread. This term refers to the sum of
risk reductions expected in a population. For example, if the annual risk of death is reduced
by one in a million for each of two million people, that is said to represent two “statistical
lives” extended per year (2 million people x 1/1,000,000 = 2). If the annual risk of death is reduced by one in 10 million for each of 20 million people, that also represents two
statistical lives extended.”
VSL is estimated using revealed- or stated-preference methods, or meta-analysis. These methods
can include statistical analysis of markets, wage statistics, surveys, and the like. NRC (2015)
provides further explanation. NRC chose to align its current VSL recommendations with those of
other Federal agencies. NRC’s current best estimate of $9.0 million is derived from the average
of the US Department of Transportation’s (DOT’s) estimate of $9.3 million and the US
Environmental Protection Agency’s (EPA’s) estimate of $8.7 million (in 2014 dollars). For the
purpose of sensitivity analysis, NRC adopted low and high median VSLs from other agencies
that have published ranges, per below:
Agency Low High
DOT $5.3 million $13.2 million
DHS $6.8 million $10.8 million
OMB $1.3 million $13.3 million
Median $5.3 million $13.2 million

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 8
3.3 Risk Coefficient Estimates
For the purposes of the Clive DU PA model, although regulatory agencies have adopted and
applied clear dose limits for individuals, evaluation of ALARA is restricted to collective doses
and risks. This is appropriate in the context of design and siting of radioactive waste facilities; as
it is likely, if any substantial future risks occur, that health concerns will be at a population level.
Further, it is assumed that facility workers will be protected under existing health and safety
regulations and guidance, and not evaluated as part of ALARA. In a complete decision analysis,
however, many other factors could be considered, including health and safety of workers,
transportation, etc.
Applying formal decision analysis to ALARA implies evaluation of the trade-off between risk
reduction and the costs associated with the actions that can be taken to reduce risk and the
benefits of the risk reduction. Risk in a PA is assessed through radiation dose.
Ionizing radiation protection limits have changed over time as more information regarding the
negative biological effects of radiation has become available (especially after World War II).
Concurrently, therapeutic and diagnostic (i.e., beneficial) uses of radiation have increased
dramatically, and nuclear fission is an important source of power in most of the developed world.
Thus, a tradeoff is immediately apparent; radiation can be both harmful and helpful, with the
balance depending upon the dose and the context.
An additional consideration are the biological endpoints of concern. Radiation in high doses kills
cells (so-called 'deterministic' effects), which can be harmful or beneficial to the receptor of the
doses (e.g., in the latter case, radiation is used to kill cancer cells). The effects of low doses of
radiation are more uncertain. There is ample evidence that ionizing radiation can damage DNA
and enhance cell proliferation in doses below those that kill cells, and thus can potentially cause
cancer (so-called 'stochastic' effects).
However, it is uncertain at what low doses carcinogenicity becomes a concern (also, note that
different tissues have different susceptibility to the effects of ionizing radiation). For many years,
there has been a presumption in radiation protection, based upon statistical analysis of animal and
human data, that ionizing radiation has a linear dose-response curve at low doses and that there
essentially is no threshold of effect; i.e. any dose of radiation can result in an increased
probability of cancer (this is termed the linear no-threshold, or LNT, hypothesis). This is not
borne out by experimental and clinical observation. Additionally, the fact that radiation is
associated with a large number of natural sources, ranging from sunlight to radon, and the fact
that multiple highly-efficient molecular and cellular defense and repair mechanisms exist, must
be considered (Scott 2008). Regardless, this LNT hypothesis is the basis for most regulatory
standards today. Consequently, if a PA uses the LNT approach to develop dose estimates, then
the ALARA analysis essentially assumes no carcinogenic threshold of radiation carcinogenesis.
A threshold of dose effect model is, arguably, more realistic than the LNT model, and could be
used to estimate dose and in the ensuing ALARA analysis. If ALARA is applied in the case of a
threshold or “target” concentration, then the threshold would be treated as a limit on the amount
of risk reduction that can be achieved by a particular management alternative. Proper evaluation
of uncertainty associated with the LNT hypothesis would be a large task in itself, but the
influence of a LNT assumption could still be evaluated within the decision analysis framework.
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 9
A different sort of threshold exists with regard to natural background levels of radiation. The
doses that the public receives from all environmental sources (e.g., local geology, extraterrestrial,
etc.) can be quite variable. For example, people who live at a location in the US with high levels
of uranium compounds in the local soil and rocks may have a much higher level of annual
exposure (due to radon) than people who live at sea level with little uranium compound content
of the soil and rocks (http://www.epa.gov/radon/zonemap.html). Similarly, individuals who
reside at higher elevations are exposed to higher levels of cosmic radiation that individuals
residing at sea level. From an ALARA perspective, it might be reasonable to consider that the
incremental population dose is of interest as well as the magnitude of the incremental dose
relative to dose from natural background radiation.
Uranium and many other metals are also associated with non-radiological toxicity; e.g. kidney or
liver damage. In such cases, toxicology has developed concepts such as the reference dose and
benchmark dose to account for the clear thresholds of effect that are associated with non-
carcinogenic toxicity (Filipsson et al., 2003). In these cases the threshold can be viewed as a
target, below which health effects are not of substantial concern.
For the purposes of ALARA, it is assumed that the LNT hypothesis is valid, despite the likely
conservatism of doing so. For NRC’s radiation risk coefficient, NRC’s previous dollar per
person-rem conversion factor was based upon the recommendations in the International
Commission on Radiological Protection (ICRP) Publication 60, published in 1991 (ICRP, 1991).
For doses to a population, the ICRP recommendation is a risk coefficient value of 7.3 x 10-4 per
rem. This coefficient accounts for the probability of occurrence of a harmful health effect plus a
judgment of the severity of the effect. The coefficient includes allowances for fatal and nonfatal
cancers and for severe hereditary effects. The nonfatal cancers and hereditary effects are
translated into loss-of-life measures based upon an assumed relationship between quality of life
and loss of life. Thus, the VSL is theoretically applicable across all contributors to the total health
risk coefficient.
In the subsequent ICRP Publication Number 103, (ICRP, 2007), the ICRP total risk coefficient
decreased by about 20 percent, from 7.3 x 10-4 per rem to 5.7 x 10-4 per rem. ICRP states that this
change was due primarily to improved methods in the calculation of heritable risks, as well as
advances in understanding of the mutational process. Also, the ICRP calculated its values
differently in ICRP 103 compared to ICRP 60. ICRP 103 states:
“It is important to note that the detriment-adjusted nominal risk coefficient for cancer
estimated here has been computed in a different manner from that of Publication 60. The
present estimate is based upon lethality/life-impairment-weighted data on cancer incidence
with adjustment for relative life lost, whereas in publication 60 detriment was based upon
fatal cancer risk weighted for non-fatal cancer, relative life lost for fatal cancers and life
impairment for non-fatal cancer. In this respect it is also notable that the detriment-
unadjusted nominal risk coefficient for fatal cancer in the whole population that may be
projected from the cancer incidence-based data of Table A.4.1a is around 4% per Sv [per
100 rem] as compared with the Publication 60 value of 5% per Sv [per 100 rem]. The
corresponding value using cancer mortality-based models is essentially unchanged at
around 5% per Sv [per 100 rem].”
As discussed above, WTP approaches are meant for application to small reductions in only
mortality risk. ICRP Publication No. 60 and ICRP Publication No. 103 combine both morbidity
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 10
and mortality into their risk coefficient numbers (ICRP, 1991, 2007). In contrast, EPA uses a
mortality-only risk coefficient with a value of 5.8 x 10-4 per rem (EPA, 2011). Using EPA’s value
would align the cancer risk coefficient with the underlying definition of WTP, and the value is
slightly greater than the ICRP risk coefficient. The US National Academies of Sciences estimated
the total risk for all classes of genetic diseases to be about 3,000-4,700 cases per million first-
generation progeny per gray of low dose rate low-LET radiation (NAS, 2006). This numerical
estimate (0.4 x 10-4 per rem) is defined relative to the “genetically significant dose” (i.e., the
combined dose received by both parents prior to conception). Thus, the EPA value may be
adjusted to account for heritable effects (i.e., adding 5.8 x 10-4 per rem and 0.4 x 10-4 per rem, to
result in 6.2 x 10-4 per rem). However, changing the risk coefficient from total detriment to a
mortality/heritable effects coefficient may still not adequately consider the full range of
consequences associated with public radiation exposure. This EPA adjusted factor (6.2 x 10-4 per
rem) may thus underestimate an appropriate risk coefficient because it is not weighted to include
cancer incidence data weighted for lethality and life impairment. Thus, by not accounting for
cancer morbidity, the benefits of a proposed action (e.g., medical costs averted, value of lost
production, etc.) may be underestimated by as much as another 20 percent.
NRC (2015) regardless chose to use the ICRP 103 value of 5.7 x 10-4 per rem for use in dollar per
person-rem estimates with the understanding this coefficient may underestimate US population
risk. The reason provided was consistency with their other regulatory programs. The final dollar
per person-rem estimate calculated using either the EPA or ICRP values is not substantially
different, due to the relatively large value of the VSL multiplier. Thus, as a practical matter in
estimation of dollars per person-rem, the ICRP and EPA values are similar.
3.4 Current NRC Recommendations
Per above, draft NRC (2015) guidance currently recommends use of a VSL of $9.0 million, and
the ICRP 103 risk coefficient of 5.7 x 10-4 per rem. The dollar conversion factor as a result of
multiplying these values is therefore equal to a rounded $5,100 in per person-rem in 2014 dollars.
NRC also recommends a low value of $3,000 per person-rem and $7,500 for a high value, based
upon variation in VSL estimates across agencies (NRC chose to use one risk coefficient for
unclear reasons, but perhaps because the risk coefficients would have to vary considerably in
order to make a difference in final estimates). NRC states that this value is to be used for “routine
effluent releases, accidental releases, 10 CFR Part 20 “as low as is reasonably achievable”
(ALARA) programs, regulatory analyses, backfit analyses, and environmental analyses”. NRC
suggests using the recommended best estimate of $5,100 per person-rem, and use of the low and
high estimates in sensitivity analysis.
NRC (2015) notes that the dollar per person-rem conversion factor is for stochastic effects only,
and is not to be applied to deterministic effects (e.g., organ failure as a result of high radiation
doses). It should also not be applied to any individual dose that could result in an early fatality.
These omissions are consistent with NRC's view that the monetizing of mortality effects as it
relates to the value of any single individual's life is not appropriate. Rather, the use of dollars per
person-rem is as an estimate of the value of small reductions in the probability of total detriment
for a given population.
DOE guidance (DOE, 1997) suggests that:
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 11
“In general, if the maximum individual dose is less than 1 mrem in a year and
collective dose is less than 100 person-rem in a year, only a qualitative or semi-
quantitative ALARA assessment can be justified. However, if individual doses are
significant, say 10s of mrem in a year, or collective dose exceeds 100 person-rem
in a year, quantitative ALARA analyses are recommended”.
As estimated collective doses from the Clive DU PA are much less than 100 person-rem per year.
Consequently, the semi-quantitative approach using the NRC (2015) value of $5,100 per person-
rem is applied here.
3.5 Approach for the Clive ALARA Analysis
For the Clive DU PA model Version 1.2, the individual doses and the population doses are small,
justifying a semi-quantitative analysis. Consequently, current the NRC value of $5,100 per
person rem per year is used in the ALARA analysis, assuming a zero discount rate. This is a
highly conservative approach when applied to a 10,000-year time frame, considering the potential
exponential effects of discounting. However, it is considered sufficient considering the low
individual and population doses, and hence low dose-based costs, which are estimated by the
Clive DU PA model.
Version 1.4 of the Clive DU PA model evaluates doses to several site-specific receptor groups for
the disposal option that all the DU waste is disposed below grade. Although comparisons are
made with the results from Version 1.0 of the Clive DU PA model, the cap design and erosion
model for Version 1.4 are very different than for the Version 1.0 model. Direct comparison of
waste disposal options is, hence, confounded by the different engineered systems. Consequently,
the focus of the ALARA analysis for the Version 1.4 model is simply to evaluate the dose costs
associated with disposal of DU waste below grade, including the evapo-transpiration cover and a
revised erosion model. The dose-based costs are projected to support at ALARA analysis for the
disposal of DU at the Clive site. Prior to describing the specific application, a more generic
discussion of decision analysis is provided.
4.0 Decision Analysis
A generic process for decision analysis has been described in many references, and includes the
following basic steps (cf., Berry, 1995, Clemen, 1996):
1. State a problem
2. Identify objectives (and measures of those objectives – i.e., attributes or criteria)
3. Identify decision alternatives or options
4. Gather relevant information, decompose and model the problem (structure, uncertainty,
preferences)
5. Choose the ‘best’ alternative (the option that maximizes the overall benefit)
6. Conduct uncertainty analysis, sensitivity analysis and value of information analysis to
determine if the decision should be made, or if more data/information should be collected
to reduce uncertainty and, hence, increase confidence in the decision
7. Go back if more data/information are collected

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 12
This framework is iterative and flexible; e.g., sensitivity analysis can also be performed before
choosing alternatives. Value-of-information analysis can be performed to help determine where
further data collection will be most informative. In the case of ALARA as described in Section 2,
the only disposal and design options that can be considered are those that first demonstrate
compliance. If no options are identified that comply after the first pass through the decision
analysis, then it might be necessary to redefine the options, or the problem. In this sense, the
decision analysis process is constrained.
Generally, in a decision analysis, there are many considerations for successful applications
including identifying the decision makers and stakeholders, the objectives of interest for all
parties involved in the decision making process, their preference structures (which attributes of
the decision problem do they prefer), characterization of uncertainty in the model, and measures
of the probable consequences of the different decision options. The spatial and temporal
constraints on the decision are also important.
There are many technical approaches that have been used to provide some form of numerical
decision support for a wide variety of decision problems (cf., Kiker et al, 2005, Linkov et al,
2009), however, only one is commonly recognized as rational and logical: Bayesian statistical
decision theory, although other names have been used. The main components of Bayesian
decision analysis include probability distributions that are used to capture what is known and
uncertain about the underlying process, and specification of cost and value functions to capture
the costs of each decision option that is being considered.
For an ALARA analysis of a PA, implementation of a Bayesian decision analysis requires
development of a PA model for different options (e.g., different disposal options, closure
options). This includes specification of probability distributions for each input parameter in the
PA model so that both the best estimate and its uncertainty is accounted for, subsequent
estimation of population doses from the model, and characterization of the costs of implementing
each option. The cost-benefit trade-off is performed by comparing options for the risks to human
health (as measured through dose), and the costs of each option considered. For Version 1.4 of
the Clive DU PA model only one set of conditions is evaluated, hence the comparison is between
the consequences of disposing of the DU waste versus not disposing of the waste.
In general, Bayesian decision analysis is a powerful means of facilitating decisions under
uncertainty. Decision analysis models, developed properly, are transparent and easy to use, even
for complex decisions. Decision analysis is also amenable to sensitivity and value-of-information
analyses, which can be used to inform decision makers regarding uncertainty in the decision.
That is, if the uncertainty is low enough, then confidence is high enough, and a decision can be
made. However, if greater confidence is needed, then further data collection is indicated, and this
is informed by the sensitivity analysis and a value of information analysis (i.e., which variables
are most uncertain and have the most influence on ranking of decision alternatives). The idea is to
reduce uncertainty cost-effectively. At some point the cost of collecting more data outweighs the
benefit from the reduction in uncertainty. Then the best decision option should be selected.
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 13
5.0 Scope of ALARA Decision Analysis for the Clive
Depleted Uranium Performance Assessment
Decision analysis in the context of ALARA has been simplified for application to the Clive DU
PA. There is one primary objective, which is to maximize human health in the context of disposal
of the DU waste. The attribute of interest is radiation dose to the receptors, which is measured in
terms of millirem in a year. Note that groundwater concentrations are also of concern, but a
simplification similar to the dose costs per person rem are not available for groundwater, hence,
an ALARA assessment for the groundwater pathway is not evaluated for the current Clive DU
PA model. However, it is noted that groundwater at Clive is not considered potable because it is
more saline than seawater. The cost consequences to human health are, consequently, negligible
or non-existent.
The Clive DU PA model evaluates dose for the three types of receptors evaluated – ranchers,
hunters and OHV enthusiasts. For the current, Version 1.4, model the DU waste is buried below
grade, and the cover is an evapo-transpiration design. The ALARA analysis evaluates the per
person rem costs of disposal of the DU waste. The results can be compared to those provided in
Version 1.0 of the model, but Version 1.0 includes DU waste disposal above grade, a rip-rap
cover, and other changes to the conceptual and probabilistic model. Given the changes, a direct
comparison of the Version 1.0 and Version 1.4 models is not appropriate. The ALARA analysis
for Version 1.4 of the model involves a cost analysis for the population risks (doses) associated
with the disposal of the DU waste.
The goal is to estimate the dose-related costs for Version 1.4 of the Clive DU PA model; that is,
assuming all DU waste is disposed below grade. As noted above, a discount rate could be applied
to the analysis. However, DU has a characteristic that is different than most forms of radioactive
waste; i.e., its decay dynamics result in higher radioactivity (and therefore dose) of the waste over
time, as opposed to lower radioactivity associated with many other types of radionuclide decay.
This perhaps has implications for whether to include a discounting factor for future benefits,
risks, and costs. Intergenerational issues are also considered in the decision to not use a discount
factor in the approach to ALARA estimation. A further consideration is the low population dose
estimates.
As noted in the introduction, specific performance objectives for land disposal of radioactive
waste are set forth in Utah Administrative Code (UAC) Rule R313-25-9 and Title 10 of the Code
of Federal Regulations (CFR) Part 61 (10 CFR 61) Subpart C, promulgated by the Nuclear
Regulatory Commission (NRC). These require a quantitative individual dose assessment over the
next 10,000 years. In effect, a decision is intended for all possible receptors over the course of the
next 10,000 years, and dose-based decisions are not made beyond that point. From the
perspective of an economic analysis this corresponds to a zero discount rate for the next 10,000
years followed by a zero value thereafter, at least from the perspective of dose. This also means
that decisions are made for possible receptors 10,000 years from now, apparently obviating the
need for any further decision making. An alternative is to couple a decision analysis approach
that perhaps includes discounting coupled with a financial plan to address continued evaluation
of the disposal system. There are other arguments for considering shorter compliance periods,
such as the reasonableness of evaluating dose far into the future, and the uncertainty that should
increase with time. However, for the current ALARA analysis a simple approach was taken: A
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 14
per person rem cost of $5,100 was assigned, and zero discounting was assumed for the next
10,000 years.
The overall decision scenario can be stated as in terms of the ‘best’ decision alternative with
regard to long-term disposal of DU. The decision evaluated for Version 1.4 of the Clive DU PA
model essentially is whether to dispose of the DU waste below grade, or to not dispose of the
waste. The decision analysis was confined to the disposal site itself, and did not address other
potentially important life-cycle issues such as interim storage, transportation, etc. However, note
that the decision analysis framework could be easily expanded to address these other issues. For
this decision analysis the 'best' decision was defined in terms of overall benefit-cost in the context
of the costs involved in reducing risk, the cost consequences of the risk, and the uncertainty
associated with choosing the best option. That is, the decision problem was framed as a benefit-
cost problem, but constrained by the requirement that each decision option considered must
comply with the performance objectives.

Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 15
6.0 References
Berry, D.A., 1995. Statistics: A Bayesian Perspective. Wadsworth Press.
Clemen RT. 1996. Making Hard Decisions. Duxbury, Pacific Grove.
DOE 1993. Radiation Protection of the Public and the Environment, DOE Order 5400.5 (January
1993).
DOE 1997. Applying the ALARA Process for Radiation Protection of the Public and
Environmental Compliance with 10 CFR 834 and DOE 5400.5 ALARA Program
Requirements, draft DOE standard (April 1997).
EPA. 2011. EPA Radiogenic Cancer Risk Models and Projections for the U.S. Population. April,
2011. Accessed at http://epa.gov/rpdweb00/docs/bluebook/bbfinalversion.pdf.
Filipsson AF, Sand S, Nilsson J, et al. 2003. The benchmark dose method - review of available
models, and recommendations for application in health risk assessment. Crit Rev Toxicol
33:505-542.
Gregory, R., Failing, L., Harstone, M., Long, G., McDaniels, T., and Ohlson, D. 2012. Structured
Decision Making: A Practical Guide to Environmental Management Choices. John Wiley
& Sons, Ltd, Chichester, UK.
Hendee WR, Edwards FM. 1986. ALARA and an integrated approach to radiation protection.
Seminars in Nuclear Medicine 16:142-150.
ICRP. 1959. Recommendations of the International Commission on Radiological Protection.
International Commission of Radiological Protection Publication No. 1. Permagon, NY.
ICRP. 1973. Implications of Commission Recommendations that Doses be Kept as Low as
Readily Achievable. International Commission of Radiological Protection Publication No.
22. Permagon, NY.
ICRP. 1977. Radiation Protection. International Commission of Radiological Protection
Publication No. 26. Permagon, NY.
ICRP. 1983. Cost-Benefit Analysis in the Optimization of Radiation Protection. International
Commission of Radiological Protection Publication No. 37. Permagon, NY.
ICRP. 1991. 1990 Recommendations of the International Commission on Radiological
Protection. ICRP Publication No. 60, Pergamon, NY.
ICRP. 2006. The Optimisation of Radiological Protection - Broadening the Process. ICRP
Publication No. 101b. Elsevier, Amsterdam.
ICRP. 2007. The 2007 Recommendations of the ICRP. ICRP Publication No. 103, Elsevier,
Amsterdam.
Keeney, R.L., 1992. Value-Focused Thinking: A Path to Creative Decision-making. Harvard
University Press, Cambridge, MA 1992
Kiker GA, Bridges TS, Varghese A, Seager PT, Linkov I. 2005. Application of multicriteria
decision analysis in environmental decision making. Integrated Environmental Assessment
and Management 1:95-108.
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 16
Linkov I, Loney D, Cormier S, Satterstrom FK, Bridges T. 2009. Weight-of-evidence evaluation
in environmental assessment: review of qualitative and quantitative approaches. Science of
the Total Environment 407:5199-5205.
NAS. 2006. Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII Phase 2.
Washington, D.C. National Academy Press
NRC. 1975a. 40 FR 19439, Appendix I to 10 CFR Part 50: Numerical Guides for Design
Objectives and Limiting Conditions for Operation to meet the Criterion “As Low as is
Reasonably Achievable” for Radioactive Material in Light-Water-Cooled Nuclear Power
Reactor Effluents. Federal Register, U.S. Nuclear Regulatory Commission, Washington,
D.C.
NRC. 1975b. Reactor Safety Study: An Assessment of Accident Risks in U.S. Commercial
Nuclear Power Plants (WASH-1400). NUREG-75/014, U.S. Nuclear Regulatory
Commission, Washington, D.C. ADAMS Accession No. ML072350618
NRC. 1977. SECY-77-388A, Value-Impact Analysis Guidelines, U.S. Nuclear Regulatory
Commission, Washington, D.C. ADAMS Accession No. ML12234B122.
NRC. 1982. Resolution of Generic Safety Issues (formerly entitled A Prioritization of Generic
Safety Issues): Main Report with Supplements. NUREG-0933. U.S. Nuclear Regulatory
Commission, Washington, D.C.
NRC. 1983a. Regulatory Analysis Guidelines of the U.S. Nuclear Regulatory Commission.
NUREG/BR-0058, revision 0, U.S. Nuclear Regulatory Commission, Washington, D.C.
ADAMS Accession No. ML15027A412.
NRC. 1983b. A Handbook for Value-Impact Assessment. NUREG/CR-3568, U.S. Nuclear
Regulatory Commission, Washington, D.C. ADAMS Accession No. ML062830096.
NRC. 1983c. Safety Goals for Nuclear Power Plant Operation. NUREG-0880, revision 1. U.S.
Nuclear Regulatory Commission, Washington, D.C ADAMS Accession No.
ML071770230.
NRC. 1985. Memorandum, W.J. Dirks to Commission, Basis for Quantifying Off-Site Property
Losses, dated October 23, 1985. U.S. Nuclear Regulatory Commission, Washington, D.C.
ADAMS Accession No. ML15050A141.
NRC 1986. 51 FR 28044 (August 4, 1986), as revised by FR 30028 (August 21, 1986). Safety
Goals for the Operations of Nuclear Power Plants; Policy Statement. Federal Register,
U.S. Nuclear Regulatory Commission, Washington, D.C.
NRC. 1995. Reassessment of NRC's Dollar Per Person-Rem Conversion Factor Policy. NUREG-
1530. December 1995. U.S. Nuclear Regulatory Commission, Washington, D.C.
NRC. 2000a. ALARA Analyses. NUREG-1727, Appendix D. September 2000. U.S. Nuclear
Regulatory Commission, Washington, D.C.
NRC. 2000b. Regulatory Analysis Guidelines of the U.S. Nuclear Regulatory Commission.
NUREG/BR-0058, revision 3, U.S. Nuclear Regulatory Commission, Washington, D.C.
ADAMS Accession No. ML023290519.
Decision Analysis Methodology for Assessing ALARA Collective Radiation Doses and Risks
6 November 2015 17
NRC. 2004. Regulatory Analysis Guidelines of the U.S. Nuclear Regulatory Commission.
NUREG/BR-0058, revision 4, U.S. Nuclear Regulatory Commission, Washington, D.C.
ADAMS Accession No. ML042820192
NRC. 2008. Generic FSAR Template Guidance for Ensuring that Occupational Radiation
Exposures are as Low as is Reasonably Achievable (ALARA), Revision 3. NEI 07-08.
November 2008. U.S. Nuclear Regulatory Commission, Washington, D.C.
NRC. 2012. SECY-12-0110, Consideration of Economic Consequences within the U.S. Nuclear
Regulatory Commission’s Regulatory Framework, U.S. Nuclear Regulatory Commission,
Washington, D.C. ADAMS Accession No. ML12173A478.
NRC. 2015. Reassessment of NRC’s Dollar per Person-Rem Conversion Factor Policy. Draft
report for comment. NUREG-1530, Rev. 1. U.S. Nuclear Regulatory Commission,
Washington, D.C.
OMB. 1992. Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs.
Circular No. A-94 Revised. U.S. Office of Management and Budget, Washington, D.C.
OMB. 2003. Regulatory Analysis, Circular No. A-4. September 17, 2003. Accessed at
http://www.whitehouse.gov/omb/circulars_a004_a-4/. U.S. Office of Management and
Budget, Washington, D.C.
Scott BR. 2008. It's time for a new low-dose radiation risk assessment paradigm. Dose-Response
6:333-351.

- NAC-0032_R4

Deep Time Assessment for the
Clive DU PA

Deep Time Assessment for the Clive DU PA
Model v1.4
22 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Deep Time Assessment for the Clive DU PA
22 November 2015 ii
1. Title: Deep Time Assessment for the Clive DU PA
2. Filename: Deep Time Assessment v1.4.docx
3. Description: This report describes details of the “deep time” component of the Clive DU PA Model. The “deep
time” model addresses long term effects (beyond 10,000 years post-closure) of disposal of DU at the Clive
facility.
Name Date
4. Originator Bruce Crowe, Robert Lee 3 Sep 2015
5. Reviewer Kate Catlett, Paul Black, Dan Levitt 22 Nov 2015
6. Remarks

3 Jul 2014; R2: Accepted track changes from R1 and added “a” and “b” to identify two Oviatt et al. (1994)
references – D. Levitt

30 Jul 2014: Updates and corrections for v1.2. White Paper now at rev 3. — R. Lee and J. Tauxe

27 Aug 2015: Merged “Deep Time Supplemental Analysis. . .” white paper with the Deep Time white paper – R.
Lee

03 Sep 2015: Edits- B. Crowe and R. Lee.

09 Sep 2015: Edits – B. Crowe and J. Oviatt

15 Oct 2015: Thorough edits and revisions to add latest GoldSim modeling and model simplification justification. –
K. Catlett

1 Nov 2015: Added to Table 1 dose parameters. Revised and added text relevant to latest model (v1.4)
consolidation and further CSM clarification.

4 Nov 2015: Added information to section 7, especially regarding dose calcs. K.Catlett and R. Perona

Deep Time Assessment for the Clive DU PA
22 November 2015 iii

This page is intentionally blank, aside from this statement.
Deep Time Assessment for the Clive DU PA
22 November 2015 iv
CONTENTS
TABLES ........................................................................................................................................ vi
1.0 Deep Time Model Distribution Summary ..............................................................................1
2.0 Introduction .............................................................................................................................3
3.0 Deep Time Model Overview ..................................................................................................3
4.0 Background on Pluvial Lake Formation in the Bonneville Basin ..........................................7
4.1 Long-term Climate ............................................................................................................7
4.2 Prehistorical Deep Lake Cycles ......................................................................................10
4.3 Shallow and Intermediate Lake Cycles ...........................................................................14
4.4 Sedimentation ..................................................................................................................17
4.5 Eolian Deposition ............................................................................................................18
5.0 Conceptual Overview of Modeling Future Lake Cycles ......................................................18
5.1 Introduction .....................................................................................................................18
5.2 Future Scenarios ..............................................................................................................19
6.0 A Heuristic Model for Relating Deep Lakes to Climate Cycles from Ice Core
Temperature ..........................................................................................................................21
6.1 Introduction .....................................................................................................................21
6.2 Glaciation ........................................................................................................................21
6.3 Precipitation ....................................................................................................................24
6.4 Evaporation .....................................................................................................................24
6.5 Simulations ......................................................................................................................26
7.0 Deep Time Modeling Approach ...........................................................................................28
7.1 Introduction .....................................................................................................................28
7.2 Deep Lake Characteristics ...............................................................................................28
7.3 Intermediate Lake Characteristics ...................................................................................30
7.4 Sedimentation Rates ........................................................................................................30
7.5 Eolian Depositional Parameters ......................................................................................35
7.5.1 Field Studies ..............................................................................................................35
7.5.2 Probability Distributions for the Depth and Age of Eolian Deposition ....................36
7.6 Destruction of the Federal DU Cell.................................................................................39
7.7 Radionuclide Concentration in DU Waste ......................................................................43
7.8 Radionuclide Concentration in Sediment ........................................................................43
7.9 Radioactivity in Lake Water ...........................................................................................44
7.10 Modeling of 222Rn Flux ...................................................................................................46
7.10.1 Waste and Sediment Water Content ..........................................................................47
7.11 Human Health Exposure and Dose Assessment .............................................................48
8.0 References .............................................................................................................................49
Appendix A ....................................................................................................................................54
Appendix B ....................................................................................................................................56

Deep Time Assessment for the Clive DU PA
22 November 2015 v
FIGURES

Figure 1. Comparison of delta deuterium (black line) from the European Project for Ice
Coring in Antarctica (EPICA) Dome C ice core and benthic (marine) oxygen-18
record (blue line) for the past 900 ky [from Jouzel et al. (2007)] ..................................5
Figure 2. Benthic oxygen isotope record for 700 ka (from Lisiecki and Raymo, 2005) ...............13
Figure 3. Temperature deviations for the last 810 k (from Jouzel et al., 2007) .............................22
Figure 4. Glacial change as a function of temperature for the coarse conceptual model ..............25
Figure 5. Two example simulated lake elevations as a function of time, with Clive facility
elevation represented by green line ..............................................................................27
Figure 6. Probability density functions for the start and end times for a deep lake, in yr prior
to the 100-ky mark and yr after the 100-ky mark, respectively. ..................................29
Figure 7. Probability density function for sedimentation rate for the deep-water phase of a
deep lake ......................................................................................................................32
Figure 8. Historical elevations of the Great Salt Lake ...................................................................33
Figure 9. Simulated transgressions of a deep lake including short-term variations in lake
elevations .....................................................................................................................34
Figure 10. Probability density function for the total sediment thickness associated with an
intermediate lake (or the transgressive of regressive phase of a deep lake) ................35
Figure 11. Eolian deposition rate results for 1,000 realizations (m/yr). ........................................40
Figure 12. Probability density function for the area over which the waste embankment is
dispersed upon destruction ...........................................................................................42

Deep Time Assessment for the Clive DU PA
22 November 2015 vi
TABLES
Table 1. Summary of distributions for the Deep Time Model container .........................................1
Table 2. Lake cycles in the Bonneville basin during the last 700 ky1 ...........................................12
Table 3. Lake cycles and sediment thickness from Clive pit wall interpretation (C. G. Oviatt,
personal communication) 1 ..........................................................................................17
Table 4. Thickness measurements from field studies of eolian silt near Clive..............................37
Deep Time Assessment for the Clive DU PA
22 November 2015 1
1.0 Deep Time Model Distribution Summary
A summary of parameter values used in the Deep Time Model component of the Clive DU PA
Model is provided in Table 1. For the purpose of this white paper, deep time refers to the period
between 10 thousand yr to 2.1 million yr; approximately when the progeny of 238U reach secular
equilibrium with 238U and peak activity.
For distributions, the following notation is used:
• N( μ, σ, [min, max] ) represents a normal distribution with mean μ and standard deviation
σ, and optional min and max if truncation is needed,
• LN( GM, GSD, [min, max] ) represents a log-normal distribution with geometric mean
GM and geometric standard deviation GSD, and optional min and max if truncation is
needed,
• U( [min, max] ) represents a uniform distribution with minimum min, and maximum max,
• Beta( μ, σ, [min, max] ) represents a generalized beta distribution with mean μ, standard
deviation σ, minimum min, and maximum max, and
• Gamma( μ, σ ) represents a gamma distribution with mean μ and standard deviation σ.

Table 1. Summary of distributions for the Deep Time Model container
Model Parameter Value or Distribution Units Reference
DepthEolianDeposition
long-term eolian deposition
depths
N(μ=72.7, σ=5 min=Small,
max=Porosity_Unit4)
cm Section 7.5
AgeEolianDeposition
long-term eolian deposition
ages
Beta(μ=13614,
σ=263.3,min=13000,max=15000)
yr Section 7.5
EolianCorrelationFactor
correlation between eolian
deposition depth and Eolian
deposition age
U(0.5,1.0) — Section 7.5
LakeDelayTime
time at which the
intermediate lake calculations
are allowed to occur
50,000 yr Section 4.1

IntermediateLakeDuration
length of time that Clive is
covered by an intermediate
lake
LN(GM=500, GSD=1.5,min=0,
max=2500)
yr Section 7.3
Deep Time Assessment for the Clive DU PA
22 November 2015 2
Model Parameter Value or Distribution Units Reference
IntermediateLakeSedimentA
mount
total depth of sediment laid
down by an intermediate lake
LN(GM=2.82, GSD=1.71) m Section 7.4
DeepLakeStart
time before the end of the
100,000-year climate cycle
LN(GM=14000, GSD=1.2,min=0,
max=50000 )
yr Section 7.2
DeepLakeEnd
time after the most recent
cold peak within the 100,000-
year climate cycle
LN(GM=6000, GSD=1.2,min=0,
max=50000)
yr Section 7.2
DeepLakeSedimentationRate
rate of the sedimentation
during the open water phase
of a deep lake
LN(GM=1.2E-4, GSD=1.2) m/yr Section 7.4
SiteDispersalArea
the area across which the
destroyed site is spread
Gamma(mean=24.2332,
stdev=11.43731)
Km2 Section 7.6

IntermediateLakeDepth
depth of an intermediate lake
at Clive
Beta(μ=30, σ=18,min=0, max=100) m Section 7.9
DeepLakeDepth
depth of a deep lake at Clive
Beta(μ=150, σ=20,min=100,
max=200)
m Section 7.9
TotalEmbankmentVolume
original total volume of the
embankment
3,231,556 m3 Section 7.8
DiffusionLength
Diffusion length for the deep
time sediments
N(μ=0.5, σ=0.16 min=0.0,
max=Large)
m Section 7.9
external_DCF_modifiers See table in ES external DCF
modifiers.xlsx Excel file
— Section 7.11
DCFs and parameters within
the DCFs container
See Dose Assessment white paper
for parameter values and reference
— See Dose
Assessment
white paper
Deep Time Assessment for the Clive DU PA
22 November 2015 3
Model Parameter Value or Distribution Units Reference
Rn_flux_ratio
ratio of Rn-222 flux at
different sediment thickness
to flux with no overlaying
cover
Thickness
0.001
0.5
1.0
1.5
2.0
3.0
4.0
5.0
6.0
7.5
Rn-222 flux
1.00000
4.392E-1
1.972E-1
8.750E-2
4.000E-2
8.140E-3
1.656E-3
3.371E-4
6.881E-5
1.00E-30
— Section 7.10
* “Large” is a very large number, and “Small” is a very small number, as defined by GoldSim.
2.0 Introduction
This white paper provides documentation of the development of parameter values and
distributions used for modeling scenarios of the fate of Federal DU Cell waste for the Clive DU
PA model in deep time. Data sources are identified and the rationale applied for developing
distributions is described. The intent of this white paper is to describe the characteristics and
potential processes of deep time and the subsequent effects on waste disposed at the Clive site.
3.0 Deep Time Model Overview
There are two major components of the Clive DU PA Model. The first component addresses
quantitative contaminant fate and transport and subsequent dose assessment for 10,000 yr
(10 ky). That modeling is based upon projections of current societal conditions into the future
and assumes no substantial change in climatic conditions. The second component addresses
“deep time” scenario calculations from 10 ky until the time of peak radioactivity. For this PA,
peak radioactivity associated with the ingrowth of progeny from 238U occurs at about
2.1 million yr in the future (2.1 My).
The initial Deep Time Models for this site, the Deep Time container of the Clive DU PA v1.0 and
v1.2 Models and the Deep Time Supplemental Analysis (DTSA) Model (Clive DU PA Model
vDTSA.gsm), addressed DU waste stored above and below the surrounding grade in an
embankment. The DTSA model is a standalone model, not directly linked to the PA model. The
models assume destruction of the embankment via wave action from a possible return of a lake
to the Clive area under future glacial period conditions, and subsequent dispersal of waste. With
a review of this modeling, a decision was made by the State of Utah to require EnergySolutions
to dispose of all DU waste below the surrounding grade, and thus no waste per se would be
exposed or dispersed upon return of a lake (SC&A, 2015). The only possible mechanisms for
dissolution and dispersal of radionuclides would then be associated with radon emanation into
the embankment materials and diffusion of dissolved radionuclides upwards. The current PA
model (v1.4) retains this assumption, and the 10 ky model and the revised Deep Time Model are
now integrated. Additional factors such as eolian (i.e., wind-borne) deposition are also now
included. Below is a brief summary of the current conceptual site model (CSM) for the Deep
Time Model. These terms and details are explained and discussed further in this report.
Deep Time Assessment for the Clive DU PA
22 November 2015 4
 Time scale of interest: 10 ky to 2.1 My post-closure.
 Waste placement: All DU waste is buried below grade in five waste cells, with a cover
embankment.
 Pluvial (i.e., caused by increased precipitation) lake occurrence: This is driven largely by
glacial cycles of cooler and wetter climate conditions. “Deep” lakes occur no more than
once per 100-ky cycle. “Intermediate” lakes can occur independent of a deep lake, or as
transitory events during the transgressive (rising lake) or regressive (falling lake) phases
of a deep lake. An intermediate lake will not occur at the elevation of the Clive site
without a return to pluvial conditions.
 Destruction of embankment: The embankment will be eroded to the level of the former
Lake Bonneville surface (current grade at the time of the first lake return) by wave action
and sediment churning during the first return of a deep or intermediate lake.
Radionuclides present in the above-grade part of the embankment (as a result of transport
processes) will be dispersed and mixed with sediments during active lake erosion across
the area of the lake. The waste itself will not be exposed.
 Release of radionuclides: Radionuclides in the dispersed sediments will be released to
lake water upon destruction of the embankment via diffusion. Radon is allowed to diffuse
upward through the sediment when a lake is not present.
 Fate of radionuclides: Radionuclides will partition between water and sediments
according to their solubility and sorption properties. Insoluble DU will be buried by lake
sediments. Radionuclides settle out in sediments after lakes recede.
 Sedimentation: Eolian deposition occurs while lake levels are below the Clive site and
are incorporated with lake sedimentation rates after the first lake returns. Clastic
sedimentation will dominate during formation of intermediate lakes with transitions to
carbonate precipitation when there are deep lakes.
The basic Deep Time Model scenario involves projecting the future environment based upon the
Pleistocene and Holocene record of climate variations and lake formation in the Bonneville
Basin. The conceptual model of the past environment is based upon scientific records (sediment
borehole logs, ice cores, deep ocean cores) of the past eight glacial/climate cycles that have
lasted approximately 100 ky each. The model considers cycles from the beginning of an
interglacial period onwards. In the past 100-ky cycles, after an interglacial period, the average
temperature drops and average precipitation increases throughout the glacial cycle, until the
relatively cold period (typically an ”ice age”) ends and the next interglacial period begins (Figure
1). The Earth is currently in an interglacial period. The first 10 ky of the Clive DU PA Model is
projected under interglacial conditions, and the Deep Time Model calculations include an
evaluation of the effect on the Federal DU Cell of future 100-ky glacial cycles for the next
2.1 My. The critical aspect of a glacial period is the potential return of a pluvial lake to the
elevation of the Clive site with accompanying lakeshore wave activity that would destroy the
Federal DU embankment. Thus, the objective of the Deep Time Model is to assess the potential
impact of glacial period pluvial lake events upon and associated with radionuclide
release/dispersal from the Federal DU Cell from 10 ky through 2.1 My post-closure.
Deep Time Assessment for the Clive DU PA
22 November 2015 5
Figure 1. Comparison of delta deuterium (black line) from the European Project for Ice
Coring in Antarctica (EPICA) Dome C ice core and benthic (marine) oxygen-18
record (blue line) for the past 900 ky [from Jouzel et al. (2007)]
The approximate historical 100-ky glacial cycles are depicted in Figure 1. The current
interglacial period is shown on the left edge of the figure. The last ice age finished between
12,000 and 20,000 yr ago (12 ka and 20 ka, indicated as “ky B.P.” in the figure). In the last
glacial maximum (represented as a trough on the far-left side of Figure 1), the major Western
United States water body Lake Bonneville, which covered much of Utah, reached its maximum
extent. Antarctic ice core data as well as benthic marine isotope data (described below) show
similar patterns for the past 800 ky. These 100-ky cycles are used as the basis for modeling the
return and recurrence of lake events in the Clive area.
The Deep Time Model should be regarded as conceptual and stylized and is not intended as a
prediction of expected future conditions at the Clive site. The intent is to estimate potential future
radionuclide releases from the remains of the Federal DU Cell, rather than to provide a
quantitative, temporally-specific prediction of future conditions, or an assessment of exposure or
doses to possible humans. Doses to potential human receptors and the presence and
characteristics of human populations in the Clive area during this time period are entirely
speculative. When a lake inundates the waste site, there will be no receptors at that location.
Additionally, calculation of radiological dose to human at times beyond 10 ky is not required by
Utah state regulations (Utah 2015). Instead, these regulations specify a “qualitative” assessment
with radionuclide release simulations for this period. Organizations such as the International
Atomic Energy Agency (IAEA 2012) have indicated that calculating doses beyond a few hundred
yr is not defensible; thus, quantitative dose assessment, particularly subsequent to lake events
related to the interglacial cycle, is insupportable from scientific and technical perspectives.
However, if the Deep Time Model results such as radon flux are considered in the context of
gauging system performance, such results may provide limited insight into the behavior of the
Deep Time Assessment for the Clive DU PA
22 November 2015 6
disposal system in deep time. Based on potential future radon fluxes, a rancher dose was
calculated in the Deep Time Model to provide a context for the radon flux results.
A “deep” lake is defined here as a large glacial-period lake on the scale of the prehistoric Lake
Bonneville (present in the area from about 32 ka to 14 ka). Such lakes have occurred in several
of the past 100-ky climate cycles. An “intermediate” lake is defined as a lake that reaches the
elevation of Clive (described further below). These lakes are assumed to occur in the
transgressive and regressive phases of a deep lake, but evidence of such lakes is difficult to
identify and interpret because lake deposits are reworked during their transgressive and
regressive lake phases. It is assumed that the first deep or intermediate lake that reaches the
elevation of Clive will destroy and disperse the Federal DU Cell embankment via wave action.
This dispersal mixes radionuclides with lake sediments. The characteristics of these mixed
sediments are dependent upon the duration and intensity of the lakeshore processes (e.g., wave
sediment churning, and formation of spits and bars from longshore drift).
Wave action associated with transgressing and regressing intermediate lakes will rework the
lake-sediment interface to a depth that is controlled by the dynamics of the wave action.
Evidence of wave action and sedimentary processes for past levels of Lake Bonneville is
preserved in the area’s sedimentary and geomorphic features. This evidence includes
paleoshorelines, fan and river deltas, wave-cut cliffs, bayhead barriers and spits (Sack, 1999;
Schofield et al., 2004; Nelson, 2012). The most relevant lake features from the geologic record
are paleoshorelines. Schofield et al. (2004) divide Lake Bonneville shorelines into erosion-
dominated and deposition-dominated. The elevation difference between shoreline bench deposits
and shoreline fronts from these studies provides a time-integrated analog for the dynamics of
wave action during shoreline transgressions and regressions (see Figure 3 in Schofield et al.
2004). These elevation differences are about 90 cm for erosion-dominated shorelines and 40 to
65 cm for deposition-dominated shorelines. Thus, the process of wave action is assumed to
remove approximately the same thickness of sediment (0.5 to 1 m) as the residual embankment
thickness (<1 m).
Any periods in which a lake does not exist are assumed to experience eolian (i.e., wind-borne)
deposition. Although some removal of embankment materials and sediment via wave action is
expected, this is not modeled explicitly. Instead, these effects are assumed to be relatively small
compared to eolian and lake deposition effects, and are assumed to have roughly a net zero effect
on overall sedimentation before and after the return of an intermediate or deep lake (remaining
embankment thickness is about 0.5 m and removal depth is about 0.5 m). The current model thus
explicitly considers eolian and lake deposition only as contributors to sedimentation thickness.
Other major geologic or climatic events could also occur in the next 2.1 My. Events such as
major meteorite impacts, and volcanic activity such as eruptions associated with the Yellowstone
Caldera could also be considered. Such future catastrophic events are often screened from
consideration in PAs on the basis of a low probability of occurrence and/or limited
consequences. In this case, a major meteorite impact and a future volcanic eruption at
Yellowstone were not screened. Instead, the impacts of these events are considered to be so
catastrophic on a global scale that their effects would far outweigh any potential radionuclide
releases from the Federal DU Cell. The same applies to major climate changes outside of those
associated with glacial cycles, although impacts of anthropogenic climate change on future lake
events are partially considered here.
Deep Time Assessment for the Clive DU PA
22 November 2015 7
4.0 Background on Pluvial Lake Formation in the Bonneville
Basin
4.1 Long-term Climate
Large-scale climatic fluctuations over the last 2.6 My (the Quaternary Period, the current and
most recent of the three periods of the Cenozoic as defined in the geologic time scale;
http://www.geosociety.org/science/timescale/) have been studied extensively in order to
understand the mechanisms underlying those changes (Hays et al., 1976, Berger, 1988, Paillard,
2001, Berger and Loutre, 2002). These climatic signals have been observed in marine sediments
(Lisiekcki and Raymo, 2005), land records (Oviatt et al., 1999), and ice cores (Jouzel et al.,
2007). These large-scale fluctuations in climate have resulted in glacial and interglacial cycles,
which have waxed and waned throughout the Quaternary Period. The causes of the onset of the
last major northern hemisphere glacial cycles 2.6 million yr ago (Ma) remain uncertain, but
several studies suggest that the closing of the Isthmus of Panama caused a marked reorganization
of ocean circulation patterns that resulted in continental glaciation (Haug and Tiedemann, 1998,
Driscoll and Haug, 1998). Future glacial events are likely to be caused by a combination of the
Earth’s orbital parameters as well as increases in freshwater inputs to the world’s oceans
resulting in a disruption to oceanic thermohaline circulation (Driscoll and Haug, 1998).
Changes in the periodicity of glacial cycles have been linked to variations in Earth’s orbit around
the Sun. These variations were described by the Serbian scientist Milutin Milankovitch in the
early 1900s, and are based upon changes that occur due to the eccentricity (i.e., orbital shape) of
Earth’s orbit every 100-ky, the obliquity (i.e., axial tilt) of Earth’s axis every 41 ky, and the
precession of the equinoxes (or solstices) (i.e., wobbling of the Earth on its axis) every 21 ky
(Berger, 1988). For the first 2 My of the Pleistocene (the first major Epoch of the Quaternary
Period), Northern Hemispheric glacial cycles occurred every 41 ky, while the last million yr have
indicated glacial cycles occurring every 100-ky, with strong cyclicity in solar radiation every 23
ky (Berger and Loutre, 2002; Paillard, 2006). The shift from shorter to longer cycles is one of the
greatest uncertainties associated with utilizing the Milankovitch orbital theory alone to explain
the onset of glacial cycles (Paillard, 2006).
Hays et al. (1976), who analyzed changes in the isotopic oxygen (δ18O) composition of deep-sea
sediment cores, suggest that major climatic changes have followed both the variations in
obliquity and precession through their impact on planetary insolation (i.e., the measure of solar
radiation energy received on a given surface area in a given time). In its most common form,
oxygen is composed of eight protons and eight neutrons (giving it an atomic weight of 16). This
is known as ”light” oxygen because a small fraction of oxygen atoms have two extra neutrons
and a resulting atomic weight of 18 (18O), which is then known as ”heavy” oxygen. 18O is a rare
form and is found in only about 1 in 500 atoms of oxygen. The ratio of these two oxygen
isotopes has changed over the ages and these changes are a proxy to changing climate that have
been used in both ice cores from glaciers and ice caps, and cores of deep sea sediments. Thus,
variations in δ18O reflect changes in oceanic isotopic composition caused by the waxing and
waning of Northern Hemispheric ice sheets, and are thus used as a proxy for previous changes in
climate (cf. Figure 1).
Deep Time Assessment for the Clive DU PA
22 November 2015 8
Slightly different external forcing and internal feedback mechanisms can lead to a wide range of
responses in terms of the causes of glacial-interglacial cycles. The collection of longer ice core
records, such as the European Project for Ice Coring in Antarctica (EPICA) Dome C core located
in Antarctica, has highlighted the clear distinctions between different interglacial-glacial cycles
(Jouzel et al., 2007). Variation in climatic conditions appears to be sufficient that large
differences have occurred in each of the past several 100-ky cycles. At the present time, the
EPICA Dome C core is the longest (in duration) Antarctic ice core record available, covering the
last 800 ky (Jouzel et al., 2007).
There is considerable uncertainty associated with the number, timing, and recurrence interval of
glacially-influenced pluvial lakes in the Bonneville Basin. The 100-ky glacial cycle is roughly
correlated with the occurrence of deep lakes (Balch et al. 2005, Davis 1998), and there appear to
be smaller, millennial scale (“Dansgaard-Oeschger”) cycles within this larger cycle that are not
necessarily uniform (Madsen, 2000). For example, the Little Valley lake cycle peaked in
elevation at about 135 ka, the Cutler Dam lake cycle peaked about 65 ka, and the Bonneville lake
cycle peaked about 18 ka (Machette et al., 1992).
Many studies highlight the importance of past atmospheric composition in the dynamics of
glaciations across the Northern Hemisphere, in addition to orbital influences (Masson-Delmotte
et al., 2010; Clark et al., 2009; Paillard, 2006). Carbon dioxide (CO2) is a well-known influence
on the atmospheric “greenhouse effect” (i.e. warming due to trapping of solar heat), and is a
globally well-mixed gas in the atmosphere due to its long lifetime. Therefore, measurements of
this gas in Antarctic ice are globally representative and provide long-term data important for
understanding past climatic changes. Direct measurement of CO2 trapped in the EPICA Dome C
core indicates that atmospheric CO2 concentrations decreased during glacial periods due to
greater storage in the deep ocean, thereby causing cooler temperatures from a reduction of the
atmosphere’s greenhouse effect (EPICA, 2004). Warmer temperatures resulting from elevated
concentrations of CO2 released from the ocean contribute to further warming and could support
hypotheses of rapid wasting at the end of glacial events (Hays et al., 1976). Earlier interglacial
events (prior to 420 ka), however, are thought to have been cooler than the most recent
interglacial events (since 420 ka) (Masson-Delmotte et al., 2010).
The predicted effect of anthropogenic CO2 on glacial cycles has evolved over time. For example,
Berger and Loutre (2002) conducted simulations including orbital forcing (i.e., cycles largely
driven by orbital variables) coupled with insolation and CO2 variations over the next 100-ky.
Their results indicated that the current interglacial period could last another 50 ky with the next
glacial maximum occurring about 100 ky from now. The scientific record (cf. Figure 1) supports
this pattern of variability across the 100-ky glacial cycles. Berger and Loutre (2002) effectively
indicate that the current 100-ky cycle will not be as glacially intense as some of the previous
cycles. They also quote J. Murray Mitchell (Kukla et al, 1972, p. 436) who predicts that “the net
impact of human activities on climate of the future decades and centuries is quite likely to be one
of warming and therefore favorable to the perpetuation of the present interglacial.” Archer and
Ganopolski (2005) conducted simulations suggesting that the combination of relatively weak
orbital forcing and the long atmospheric lifetime of carbon release from fossil fuel and methane
hydrate deposits could prevent glaciation for the next 500 ky over two glacial cycle eccentricity
minima. Cochelin et al. (2006) used a paleoclimate model to simulate the next glacial inception
under orbital and atmospheric CO2 forcings. Three scenarios were modeled: an impending
Deep Time Assessment for the Clive DU PA
22 November 2015 9
glacial inception under low CO2 levels; a glacial inception in 50 ky for CO2 levels of 280 to
290 ppm; and no glacial inception for the next 100-ky for CO2 levels of 300 ppm or higher.
Tzedakis et al. (2012a) defined interglacial periods as episodes where global climate is
incompatible with the wide global extent of glaciers, and examined differences in such
interglacial durations over the last 800 ky. They noted that the onset of interglacials occurs within
2 ky of the boreal summer insolation maximum consistent with Milankovitch forcing, whereas
the end of interglacials does not occur consistently on a similar part of the insolation curve.
Reduction in summer insolation is identified as a primary trigger for glacial inception, but
multiple other feedbacks including atmospheric CO2 concentrations combine to modify the
timing of glacial inception. They further recognized two main groups for mean duration of
interglacials: 13±3 ky and 28±2 ky. In a related paper, Tzedakis et al. (2012b) suggest that the
end of the current interglacial could occur within the next 1,500 yr if atmospheric CO2
concentrations were reduced to about 240 ppm, but no glacial inception is projected to occur at
current atmospheric CO2 concentrations of 400 ppm, consistent with the conclusions of Archer
and Granopolski (2005).
Jansen et al. (2007) in Chapter 6 of the fourth assessment report of the Intergovernmental Panel
on Climate Change (IPCC) concluded that “it is very unlikely that the Earth would naturally
enter another ice age for at least 30 ky.” These conclusions were updated and strengthened in
Chapter 5 of the fifth IPCC assessment report (Masson-Delmotte et al., 2013).
“Since orbital forcing can be accurately calculated for the future…, efforts can be
made to predict the onset of the next glacial period. However, the glaciation
threshold depends not only on insolation but also on the atmospheric CO2
concentration… Models of different complexity have been used to investigate the
response to orbital forcing in the future for a range of atmospheric CO2 levels.
These results consistently show that a glacial inception is not expected to happen
within the next approximate 50 ky if either atmospheric CO2 concentration
remains above 300 ppm or cumulative carbon emissions exceed 1000 PgC
[petagrams of carbon—one petagram is 1015 g]. Only if atmospheric CO2 content
was [sic] below the pre-industrial level would a glaciation be possible under
present orbital configuration… Simulations with climate–carbon cycle models
show multi-millennial lifetime of the anthropogenic CO2 in the atmosphere…
Even for the lowest [emissions] scenario, atmospheric CO2 concentrations will
exceed 300 ppm until the year 3000. It is therefore virtually certain [i.e., a
greater-than 99% probability] that orbital forcing will not trigger a glacial
inception before the end of the next millennium.”
Current CO2 levels are approximately 400 ppm (http://co2now.org/images/stories/data/co2-mlo-
monthly-noaa-esrl.pdf), and have been steadily rising over the past 150 yr due to anthropogenic
sources. Preindustrial levels of CO2 were about 280 ppm, and CO2 levels associated with glacial
periods tend to be about 240 ppm (Tzedakis et al., 2012b). It would require major reductions in
CO2 emissions worldwide in order to return to preindustrial levels, and/or engineering solutions
(e.g., “scrubbing” on a massive scale) to remove CO2 from the atmosphere so that pre-industrial
levels are attained. However, the Clive DU PA Model projects current knowledge as a
fundamental assumption, therefore it is assumed here that no major anthropogenic CO2-reduction
interventions will occur, and that CO2 levels will continue to rise, or at least will not attain
preindustrial levels within the next 50 ky or longer.
Deep Time Assessment for the Clive DU PA
22 November 2015 10
The Bonneville basin watershed is large and integrates runoff from the eastern Great Basin and
transition region of the Colorado plateau. Long-term changes in evaporation and precipitation
over a large region are required to sustain rising of a lake to the Clive elevation. These conditions
may be expected to occur only with a return to glacial conditions given climate model forecasts
of increased aridity for the southwest United States. Climate change risks to municipal water
supplies in Utah have been modeled using watershed hydrology models that explore temporal
changes in average conditions (temperature, precipitation, runoff), and severe drought and water
supply scenarios (e.g., the Salt Lake City Department of Public Utilities, Bardsley et al., 2013).
These types of studies are both prudent and timely, but future projections of decade scale data
are highly uncertain. Indeed, projection of the global climate change model results to regional
models has been a developing topic in the succession of IPCC reports. Warming temperatures
associated with anthropogenic climate effects will likely have appreciable impacts on the
Southwestern United States, but current drought projections do not exceed paleoclimate records
of droughts over the last two millennia (Woodhouse et al., 2010; Morgan and Pomerleau, 2012).
Multi-model ensemble studies of future climate projections from 16 global climate models show
both decreases and increases in streamflow projections for the upper Colorado River Basin
(Harding et al., 2012). Cook et al. (2010) suggest caution in projecting climate model projections
for the arid Southwest.
Regardless, the weight of evidence reviewed and summarized in the sequence of IPCC reports is
considered to be substantive and persuasive, and this information supports the current modeling.
It is assumed that CO2 levels will continue to rise for the foreseeable future, or will not decrease
below pre-industrial levels. It is also assumed that the IPCC and associated climate projection
studies are valid, with a high degree of confidence, including their conclusion that the inception
of the next glacial period will probably not occur for at least 50 ky.
The following sub-sections present an overall background on past events in the Bonneville basin
that are driven by major shifts in climate, and that are presumed to occur in the distant future as
well.
4.2 Prehistorical Deep Lake Cycles
The Bonneville basin is the largest drainage basin in the Great Basin of the Western United
States. It is a hydrologically closed basin of over 134,000 km2, and has previously been occupied
by deep pluvial lakes. Pluvial lakes typically form when warm air from arid regions meets
chilled air from glaciers, creating cloudy, cool, rainy weather beyond the terminus of the glacier.
The increase in rainfall and moisture can fill the drainage basin, forming a lake. This kind of
climate was evident during the last glacial period in North America, and resulted in more
precipitation than evaporation, hence the rise of Lake Bonneville.
Numerous studies have investigated previous lake cycles in the Bonneville Basin. These include
studies of Lake Bonneville shoreline geomorphology (Currey et al., 1984), palynological (i.e.,
pollen) studies of deep boreholes (Davis, 1998), and studies of the geochemistry of deep-water
lacustrine depositional sequences (Eardley et al., 1973; Oviatt et al., 1999, Balch et al., 2005).
Analysis of these sediment cores can be used to help understand previous lake levels and
characteristics as well as establish the approximate age of previous lake cycles (e.g., Oviatt et al.,
1999).
Deep Time Assessment for the Clive DU PA
22 November 2015 11
Oviatt et al. (1999) analyzed hydrolysate amino acid enantiomers for aspartic acid, which is
abundant in ostracode protein. Ostracodes are small crustaceans that are useful indicators of
paleo-environments because of their widespread occurrence and because they are easily
preserved. Ostracodes are highly sensitive to water salinity and other limnologic changes.
Therefore, portions of sediment cores that contain ostracodes indicate fresher, and hence
probably deeper, lake conditions than the modern Great Salt Lake (Oviatt et al., 1999). An
important exception to the deep lake interpretation inferred from the presence of ostracodes is
wetland/spring discharge areas. While wetland sites contain abundant ostracodes, the sites can
generally be discriminated from deep lake carbonates by their lithology and stratigraphic position
of the former within transgressive and regressive lake cycles. To establish the approximate
timing of previous lake cycles, Oviatt et al. (1999) examined sediments from the Burmester
sediment core originally collected in the early 1970s near Burmester, Utah (Eardley et al., 1973).
Burmester is approximately 65 km east of Clive on the southern edge of the Great Salt Lake, at
an elevation of 1286 m above mean sea level (amsl). The Clive area has an elevation of
1307 m amsl. Oviatt has also collected sediment data from Knolls (to the west of Clive) and at
Clive itself (described further below). These data are largely consistent with the more recent
layers from Burmester, indicating similar sedimentation processes at work at least during these
time periods.
Data from the 307-m Burmester core suggest that a total of four deep-lake cycles occurred
during the past 780 ky (Table 2. ). Oviatt et al. (1999) found that the four lake cycles correlated
with marine δ18O stages 2 (Bonneville lake cycle: ~24 to 12 ky), 6 (Little Valley lake cycle:
~186 to 128 ky), 12 (Pokes Point lake cycle: ~478 to 423 ky), and 16 (Lava Creek lake cycle:
~659 to 620 ky).
Oxygen isotope stages are alternating warm and cool periods in the Earth’s paleoclimate which
are deduced from oxygen isotope data (Figure 2). These stages suggest that deep pluvial lake
formation in the Bonneville basin occurred in the past only during the most extensive Northern
Hemisphere glaciations. There are many interacting mechanisms that could control or ‘force’
glaciation and deglaciation. For example, Oviatt (1997) and Asmerom et al. (2010) suggested
that these extensive glaciations were controlled by the mean position of storm tracks throughout
the Pleistocene, which were in turn controlled by the size and shape of the ice sheets. Other
glaciation forcing mechanisms have been suggested. The review by Ruddiman (2006) suggests
that insolation changes due to orbital tilt and precession, greenhouse gas concentrations, changes
in Pacific Ocean circulation, and possibly other interacting mechanisms could contribute to
glaciation and deglaciation cycles in North America, and thus pluvial lake existence and size.
Lyle et al. (2012) suggests that lake levels in the Pleistocene western US were influenced by
stronger spring/summer precipitation fed by tropical Pacific air masses, rather than higher
numbers of westerly winter storms.
Balch et al (2005) conducted a more recent detailed study on ostracode fossils in Great Salt Lake
sediment (i.e., under the lake). Other fossil invertebrates were also used as paleoecological
indicators in this study. Both brine shrimp and brine fly fossils are indicators of hypersaline
environments because they have a much higher salinity tolerance than most other invertebrates.
This study’s findings were consistent with Oviatt et al.’s (1999) later cycles, but as the core was
not as deep the findings are not as useful for the present purpose as the Burmester data. The
Burmester core data are more germane to the present modeling effort because they represent a
relatively long time period in which to establish the occurrence of pluvial lakes in the region.
Deep Time Assessment for the Clive DU PA
22 November 2015 12
Table 2. Lake cycles in the Bonneville basin during the last 700 ky1
Lake Cycle Approximate
Age2
Maximum
Elevation
Lake Level Influences
Great Salt
Lake (current
level)
present 1284 m (4212 ft) in
1873
Interglacial climate; human intervention
Bonneville
(Gilbert
Episode)
11.6 ka 1295 m (4250 ft) Beginning of interglacial climate;
Bonneville
(Provo
Shoreline)
17.4 to 15.0 ka 1445 m (4740 ft) Glacial climate; new threshold at Red Rock
Pass, Idaho (natural dam collapse)
Bonneville
(Bonneville
Shoreline)
18.0 ka 1552 m (5090 ft) Glacial climate; threshold at Zenda near Red
Rock Pass, Idaho
Bonneville
Transgression
~30 to 18.0 ka Glacial climate
Bonneville
(Stansbury
Shoreline)
26 to 24 ka 1372 m (4500 ft) Glacial climate; transgressive phase of Lake
Bonneville
Cutler Dam ~80 to 40 ka <1380 m (<4525 ft) Glacial climate
Little Valley ~128 to 186 ka 1490 m (4887 ft) Glacial climate
Pokes Point 417 to 478 ka 1428 m (4684 ft) Glacial climate
Lava Creek ~620 to 659 ka 1420 m (4658 ft) Glacial climate
Elevations are not corrected for isostatic variations.
1 Note the various levels of the last major lake cycle, Lake Bonneville. 2 Approximate ages derived from Currey, et al. (1984) Link et al. (1999) and Oviatt et al. (1999). Bonneville cycle
approximate age presented as calibrated yr.
However, note that there is considerable uncertainty associated with the number, timing, and
recurrence interval of lakes in the Bonneville Basin. The 100-ky glacial cycle is roughly
correlated with the occurrence of deep lakes (Balch et al., 2005; Davis, 1998), and there appear
to be smaller, millennial-scale cycles within this larger cycle that are not necessarily uniform
(Machette et al., 1992; Madsen, 2000). It is likely that intermediate lakes have also occurred in
each glacial period, but the shorelines have been destroyed by later lakes. Sediment mixing that
occurs during lake formation can also mask the existence of previous intermediate lakes. Thus, it
is impossible to have complete confidence in historical lake formation characteristics and
formation.
Lake Bonneville is the last major deep lake cycle that took place in the Bonneville basin and is
widely described in the literature (Hart et al., 2004; Oviatt and Nash, 1989; Oviatt et al., 1994a,
1999). Lake Bonneville was a pluvial lake that began forming approximately 28 to 30 ka,
forming various shorelines throughout its existence and covering over 51,000 km2 at its highest
level (Matsubara and Howard, 2009).
Deep Time Assessment for the Clive DU PA
22 November 2015 13
Figure 2. Benthic oxygen isotope record for 700 ka (from Lisiecki and Raymo, 2005)1
Most studies indicate that the high-stand (i.e., the highest level reached) of the lake at the Zenda
threshold (1,552 m amsl), located north of Red Rock Pass, occurred approximately 18.0 ka. The
high-stand of the lake was followed by an abrupt drop in lake level due to the catastrophic failure
(landslide) of a natural dam composed of unconsolidated material at approximately 17.4 ka. As a
result of this flood, the lake dropped to a level of 1,430 m amsl, called the Provo level (Miller et
al. 2013). The Provo level is the maximum level that any future deep lake is likely to reach
without major regional tectonic changes (Currey et al., 1984; Oviatt et al., 1999).
A more recent study (Miller et al., 2013), using radiocarbon dating for Provo shoreline gastropod
deposits, estimates that the dam collapse and Bonneville flood event occurred between 18.0 and
18.5 ka, and therefore the high-stand may have occurred earlier. However, Miller et al. (2013)
indicate that “uncertainties in [gastropod] shell ages may be as large as thousands of yr, and the
major shorelines of Lake Bonneville and the Bonneville flood require more work to establish a
reliable chronology.” The lake regressed rapidly during the last deglaciation, then increased
again to form the Gilbert episode ~ 11.6 ka, which remained below the elevation of Clive
(Oviatt, 2014). The lake then receded to levels of the current Great Salt Lake at approximately
10 ka for the remainder of the Holocene Epoch.

1Red (warm periods) and blue (cool periods) values correspond to marine isotope stages based upon Lisiecki and
Raymo (2005). Lake stages identified by Oviatt et al. (1999) are also included in blue text.
Deep Time Assessment for the Clive DU PA
22 November 2015 14
Glacial cycles can be discerned in Figure 2 by considering each cycle from the beginning of the
interglacial period and ending each cycle at the peaks that correspond to deep lake occurrence.
Using this approach, the current glacial cycle started around 12 ka, Lake Bonneville occurred at
the end of the last complete cycle, and Cutler Dam occurred in the middle of the last 100-ky
cycle. The previous 100-ky cycle resulted in the Little Valley lake. The Pokes Point lake
occurred five cycles ago, and The Lava Creek lake seven cycles ago. These deep lakes have been
identified in sediment cores and in shorelines around the Bonneville Basin. However, it is likely
that many more shallow lakes have also occurred in each glacial period, but the shorelines have
been destroyed by subsequent deeper lakes.
The types of sediment resulting from the formation and long-term presence of lakes in the
Bonneville basin are diverse and can be divided into three components (Schnurrenberger et al.,
2003): 1) chemical sediment (inorganic materials formed within the lake), 2) biogenic sediment
(fossil remains of former living organisms), and 3) terrigenous or clastic sediments (grains and
clasts that are mechanically and chemically fragmented from existing material, transported and
deposited by sedimentary processes). A fourth type of associated sediment, not formed by lakes,
includes eolian deposits consisting of windblown grains of sand, silt or dust (i.e., loess). These
deposits can locally be interbedded with lake sediments and may be affected by soil-forming
processes (i.e., pedogenesis) during prolonged periods of subaerial exposure. All four types of
sediments can be intermixed by lake-wave action or bioturbation, and deposited as clastic
sediments during transgressive and regressive lake cycles.
There is considerable uncertainty in the number of lakes, particularly lakes of intermediate size
that might have existed in the Bonneville basin. However, the main focus of the Deep Time
Model is to evaluate the presence of lakes that inundate Clive in future glacial cycles, and to
approximately match the net sedimentation of the past glacial cycles. In order to inform the
potential for radionuclide releases, the high-level, conceptual modeling of lake cycles that was
conducted here did not assume any particular mechanism of glaciation and deglaciation. For
example, the modeling simply assumed a 100-ky cycle, regardless of the mechanism. The model
addresses deep lakes by allowing them to return in some glacial cycles, and by allowing
intermediate lakes to occur as part of the transgressive and regressive phases of deep lake
development.
4.3 Shallow and Intermediate Lake Cycles
The current Great Salt Lake is an example of a shallow lake, as is the reinterpreted Gilbert
episode lake that has been shown to have not reached the elevation of the Clive site (Oviatt,
2014, contrasted with the map of Currey et al., 1984). The specific depths of lakes are not
important in the Deep Time Model, aside from calculations with regard to lake chemistry and
dominant processes of sedimentation. Under current climate conditions, only shallow lakes will
occur. Under future climate conditions, some glacial cycles will produce deep lakes in the
Bonneville basin, and intermediate lakes will occur during the transgressive and regressive
phases of deep lakes, or during glacial cycles that do not produce deep lakes. The approximate
timing of the return of the first intermediate lake is important in the Deep Time Model, because it
is assumed that the Federal DU Cell embankment is destroyed upon the occurrence of the first
intermediate lake.
Deep Time Assessment for the Clive DU PA
22 November 2015 15
A key assumption of the Deep Time Model, based upon core sediment studies, is that the net
depositional rate of deep lakes is lower than the sediment depositional rate for intermediate
lakes. The conceptual basis for this assumption is that sedimentation rates are dependent on basin
location, presence or absence of fluvial deposition, wave dynamics, availability of local sediment
sources, slope, water chemistry and biological activity. Biogenic carbonate deposition is likely to
occur under a wide range of lake conditions, but the ratio of carbonate deposition to clastic
sedimentation will increase as the lake deepens because of the reduction in sedimentary influx
with increased distance from shoreline processes and decreased wave activity.
There are recognized trends in carbonate mineralogy that can be correlated with lake volume and
indirectly lake depth (cf., Oviatt, 2002; Oviatt et al., 1994b; Benson et al., 2011). The transitions
from low-magnesium calcite to high-magnesium calcite to aragonite generally reflect increasing
lake salinity and increasing magnesium concentration, which occurs with decreasing lake
volume. Similarly, for a hydrologically closed pluvial lake system, the relative concentration of
total inorganic carbon should typically decrease as lake size increases. The δ18O of deposited
carbonate can be correlated with rising lake levels because of the interplay between the δ18O
value of river discharge entering a lake and the δ18O value of water vapor exiting the system via
evaporation (Benson, et al., 2011). The mineralogy and isotopic composition of carbonate
composition can be obtained from sediment cores. Interpretation of the data is complicated by
multiple processes, including: local groundwater discharge; introduction of glacial rock flour;
and, reworking of lake sediments during transgressive and regressive lake cycles.
Intermediate lake events are known to have occurred in the Clive area. These are documented in
Table 3 (C.G. Oviatt, Professor of Geology, Kansas State University, personal communication
December 2010, January 2011, and various email communication referred to as “C.G. Oviatt,
personal communication.”). These events are evident from a pit wall interpretation at the Clive
site (Appendix A; C.G. Oviatt, unpublished data) as well as at the ostracode and snail record
present in the Knolls sediment core (12 km west of Clive near the Bonneville Salt Flats;
Appendix B; C.G. Oviatt, unpublished data). In 1985 Lake Bonneville sediments were described
and measured in a pit wall during early development of the Clive disposal facility (Oviatt, 1985).
Lake sediments of intermediate and deep lakes were briefly studied during field studies at Clive
in the winter of 2014 (Neptune, 2015a). These studies confirmed:
1. The pit walls described by C.G. Oviatt in 1985 have been removed during quarrying
and/or disposal operations at the Clive site.
2. Soil-modified eolian silt (mean thickness 73 cm) was observed in the upper part of quarry
walls throughout the Clive site.
3. The stratigraphy of sediments of Lake Bonneville in modern quarry-wall exposures are
consistent with the 1985 pit wall interpretations (Appendix A).
4. Quarry-wall deposits of gravel and sand at the Clive site contain distinctive volcanic
clasts of black andesite derived from the Grayback Hills north of Clive. These deposits
are part of the transgressive Lake Bonneville sedimentary sequence.
5. Pre-Lake Bonneville lake sediments with interbedded-soils and eolian sands were
observed in one deep quarry wall at the north end of the Clive site. These sediments are
consistent with the 1985 pit-wall interpretations but the new exposures were
insufficiently studied to established sediment correlations and the deposition chronology.
Deep Time Assessment for the Clive DU PA
22 November 2015 16
Stratigraphic correlations between 1985 studies and the new field studies (Neptune, 2015a) are
shown in Appendix A.
From the Clive pit wall interpretation, it is presumed that at least three intermediate lake cycles
occurred prior to the Bonneville cycle, although there is uncertainty associated with this
estimate. For example, these intermediate cycles could be part of the transgressive phase (i.e.,
rising lake level) of the Lake Bonneville cycle (C.G. Oviatt, personal communication). By
analyzing the Knolls Core interpretation, the Little Valley cycle is present at approximately
16.8 m from the top of the core. Given that the pit wall at Clive was 6.1 m high and does not
capture the Little Valley cycle, it is possible that other smaller lake cycles occurred in the Clive
region in addition to the three intermediate lake events noted in Table 3 (labeled as Pre-
Bonneville Lacustrine Cycles).
There are few data to support the specific number of lakes that might have reached Clive or the
rate of sedimentation. There is also uncertainty associated with the particular times that these
cycles occur, as age dating (e.g., via radiocarbon dating) has not been performed in the Great Salt
Lake area. Most studies examine the degree of lake salinity using fossil records, and are
associated with cores that are in or near the Great Salt Lake. For example, Balch et al. (2005;
Fig. 6) estimated that there were six “saline/hypersaline” (i.e., shallow to intermediate) lake
cycles that occurred between the Lake Bonneville and Little Valley cycles, and approximately
that same number between the Little Valley cycle and the maximum age evaluated (300 ky).
However, this work does not inform the question of whether these lakes may have reached the
elevation of Clive, nor does similar work such as Davis (1998).
It is also possible that intermediate lakes could reach the elevation of Clive under unusual
conditions not necessarily associated with a return to a glacial cycle. The areal extent of lakes is
not only determined by elevation, but also by local topography, precipitation, temperature,
characteristics of inflow and outflow sources, and other factors. For instance, the Great Salt Lake
‘spilled’ over a 1285-m (4217-ft) amsl topographic barrier to the west of the present lake into the
area of the present Great Salt Desert as recently as the 1700s (Currey et al., 1984). This expanded
lake was about 15 m lower than the Clive site, and slightly higher than the current surface
elevation of the Great Salt Lake.
Precise dating of shorelines for the Great Salt Lake and variants is unfortunately lacking.
Radiocarbon dating for the Pyramid Lake area in Nevada indicates that this lake’s levels have
lowered approximately 35 m from the late Holocene Epoch (3.5 to 2.0 ky) to today (Briggs et al.,
2005). Radiocarbon and tree-ring dating to determine lake levels in the Carson Sink area in
Nevada indicates that lake elevations have risen approximately 20 m twice in the last 2000 yr
(Adams, 2003). It is not possible at this time to interpolate from these studies to the Great Salt
Lake area. However, given the lack of empirical evidence that under present climate conditions
(as opposed to cooler, wetter conditions) an intermediate lake would reach the Clive site, this
condition is not addressed in the Deep Time Model.

Deep Time Assessment for the Clive DU PA
22 November 2015 17
Table 3. Lake cycles and sediment thickness from Clive pit wall interpretation (C. G.
Oviatt, personal communication) 1
Lake Cycle Thickness of
Sediment Layer (m)
Depth Below
Ground Surface (m)
Soil-modified eolian silt1 1.05 1.05
Lake Bonneville Regressive Phase (reworked marl) 0.43 1.48
Lake Bonneville Open Water (white marl) 1.29 2.77
Lake Bonneville Transgressive (littoral facies) 0.76 3.53
Pre-Bonneville Lacustrine Cycle 3 (possible shallow lake) 0.71 4.24
Pre-Bonneville Lacustrine Cycle 2 (possible shallow lake) 0.62 4.86
Pre-Bonneville Lacustrine Cycle 1 (possible shallow lake) 1.14 6.00
1 The upper sedimentary sequence is no longer interpreted as a Gilbert lake phase (Oviatt, 2014). It is surficial
eolian deposits and soils based on recent field studies (Neptune, 2015a). The pit wall described in the 1985 studies
has been removed during quarrying and/or disposal operations.
4.4 Sedimentation
During all pluvial lake cycles, evaporites are deposited, as well as carbonates in the form of
tufas, marls, and mudstones. These sediments may contain varying components of shells (e.g. of
mollusks), and ostracodes (Hart et al., 2004). Terrigenous sedimentation however, accounts for
the major thickness of sediment observed throughout the Clive area sediment core record (C.G.
Oviatt, personal communication). The geomorphological evidence in the form of depositional
and erosional landforms produced at lake shorelines are carved into the landscape in the
Bonneville basin and provide examples of the erosional capacity of lake systems over long time
periods. Given the difficulty in separating chemical, biogenic, and terrigenous sediment deposits
in cores and natural exposures, the estimates reported below are assumed to be representative of
cumulative sedimentation from all causes during a lake event.
Brimhall and Merritt (1981) reviewed previous studies that analyzed sediment cores of Utah
Lake, a freshwater remnant of Lake Bonneville that formed at approximately 10 ka. They
suggest that up to 8.5 m of sediment has accumulated since the genesis of Utah Lake, implying
an average sedimentation rate of 0.85 mm/y or 850 mm/ky. Within the Bonneville basin as a
whole the major lake cycles resulted in substantial accumulations of sediment based upon the
depth of the cores analyzed (e.g., a 110 m core that corresponds to the past 780 ky, or four deep
lake cycles [Oviatt et al., 1999]). This accumulation averages about 140 mm/ky. Einsele and
Hinderer (1997) indicate that sediment accumulation in the Bonneville basin occurred at a rate of
120 mm/ky during the past 800 ky. The Knolls Core suggests that there has been 16.8 m of
sediment formed in the last glacial cycle, or nearly 170 mm/ky.
Interpretations of the Clive pit wall (C.G. Oviatt, unpublished data) indicate that the
sedimentation rate at the Clive site for the Lake Bonneville cycle is on the order of 2.75 m over a
17 to 19 ky time period (140 to 160 mm/ky). By contrast, shallow lacustrine cycles that occurred
prior to Lake Bonneville (but after the Little Valley cycle) indicate that the amount of sediment
deposited during each cycle is approximately 1/3 that of the Bonneville sediment deposited. The
timing of these shallow lake cycles is uncertain, however it can be approximated when
Deep Time Assessment for the Clive DU PA
22 November 2015 18
comparing the Clive pit wall interpretation to the Knolls Core (C.G. Oviatt, personal
communication). The Little Valley lake cycle is exhibited in the Knolls Core at a depth of
approximately 17 m, which is roughly 14 m deeper than the beginning of the transgressive phase
of the Bonneville lake cycle event noted on the Clive pit wall interpretation. Given the Little
Valley event occurred 150 ka, a sedimentation rate can be approximated for the depth between
this event and the transgressive phase of the Bonneville cycle of 110 mm/ky.
4.5 Eolian Deposition
Post-Lake Bonneville eolian deposition has occurred and will continue to occur at the Clive site
under current conditions. The expected primary mode of eolian deposition at the Clive site is
deposition of fine-grained silt from suspension fallout during episodic wind storms.
Exceptionally strong surface winds could potentially transport sand-sized material by saltation.
Evidence supporting these conclusions include (Neptune, 2015a):
 The presence of soil-modified eolian silt in the upper part of quarry-wall exposures at
multiple locations in the Clive site. The presence of these deposits requires continuing
eolian activity in the region and long-term maintenance of stable surfaces that promotes
preservation of the eolian deposits (suspension fallout) and soil-forming processes.
 Holocene dune deposits of eolian sand and silt in road cut exposures within 0.5 km of the
Clive site.
 Active gypsum sand dunes located approximately 13.5 km west of the Clive site.
 Active dune fields in the Lake Bonneville basin west and southwest of the Clive site
(Jewell and Nicoll, 2011).
Replicate measurements of the thickness of eolian deposits located in quarry wall exposures in
the Clive site are presented in Neptune (2015), and are used below to develop input probability
distributions for the Deep Time Model. These deposits are relevant to expected future eolian
sedimentation before the first return of an intermediate or deep lake; with the rise of a future lake
to the elevation of the Clive site, wave activity will rework the eolian sediments and intermix
them with clastic lakeshore sediments.
5.0 Conceptual Overview of Modeling Future Lake Cycles
5.1 Introduction
There is a lack of data and peer-reviewed literature that would allow accurate and precise
prediction of the direct effects of future climate change on intermediate and deep lake formation
in the Bonneville basin. However, assuming no major changes from prehistorical climate cycles,
there is a possibility of another major lake cycle occurring in the Bonneville basin within the
next few million yr. Variations in the Earth’s orbital parameters in combination with increases in
inputs of freshwater into the oceans could lead to another major ice age and could alter long-term
climatic patterns in the Bonneville basin, resulting in deep lake formation. The Clive site might
be subjected to deep lake formation in the future, unless anthropogenic effects on atmospheric
CO2 concentrations cause major long-term changes in glacial cycles and climatic patterns.
Deep Time Assessment for the Clive DU PA
22 November 2015 19
An overview of the Deep-Time CSM was presented at the beginning of this report. The basic
intent of the Deep Time Model is to allow a lake to exist that is sufficiently large that the above-
ground embankment of the Federal DU Cell will be destroyed. It assumes that the sedimentation
rates for each glacial cycle are similar. The exact timing of the recurring lakes is not important,
the current 100-ky cycle excepted. The Deep Time Model allows the possibility of a deep lake to
return in each 100-ky cycle. It also allows intermediate lakes to recur at a frequency that allows
the assumed 100-ky sedimentation rate to be satisfied. The current 100-ky cycle is not modeled
explicitly. It is possible that the current interglacial period will last for at least another 50 ky due
to anthropogenic influences, which is unusually long compared to the interglacial period for
recent 100-ky ice age cycles.
5.2 Future Scenarios
Representative lake occurrence scenarios for deep time are described below. Note that there are
two components of the models used to represent these scenarios. The first is modeling lake
formation and dynamics, based upon the scientific record, literature, and expert opinion. The
second is modeling the fate of the Federal DU Cell.
The Great Salt Lake represents the current condition of a shallow lake in the Bonneville Basin.
Lakes such as this are likely to exist in all future climatic cycles, but will not reach the elevation
of the DU waste embankment at Clive and thus will not affect the waste embankment. For the PA
model, it is assumed that destruction of the waste embankment will result from the effects of
wave action from an intermediate or deep lake. This assumption separates intermediate and
shallow lakes. In this destruction scenario, the embankment material above grade is assumed to
disperse through a combination of wave action/churning and dissolution into the water column
above the waste dispersal area. Radionuclides present in the embankment dissolve into the lake
and eventually return to the lakebed via precipitation or evaporation as the lake regresses.
Some radionuclides in the water column will bind with carbonate ions and precipitate as
chemical and biogenic sediment, while radionculides bound to embankment materials will
remain within the clastic sediment as the lake eventually recedes. Wave action during the lake
recession is expected to rework and mix the chemical, biogenic and clastic lake deposits. The
combined complexity of processes affecting the compositional and sedimentary features of
lacustrine deposits (Fritz, 1996) and the mixing of lake sediments during regressive and
transgressive lake cycles makes it difficult to develop quantitative models of chemical and
physical processes affecting the distribution of waste radionuclides in lake waters and sediments.
In reality, waste radionuclides dissolved in lake waters will mix and be diluted by lake
circulation driven by prevailing winds and geostrophic balances (Jewel, 2010). Waste-sediment
mixes will be dispersed by wave action and longshore drift. Sediment concentrations will
decrease over time because the amount of waste does not change other than through decay and
ingrowth, whereas more sediment is added over time. The model makes two simplifying
assumptions. First, sediments are thoroughly mixed throughout the total sediment depth. In the
Deep Time Model the sediment layers are considered to be a single mixing cell. Second,
diffusion can occur into the lake through this mixing cell, throughout the total sediment depth.
The mixing cell allows for radionuclides to diffuse through a short diffusion length, relative to
the depth of the mixing cell (sediment depth). Although sediment concentrations will decrease
Deep Time Assessment for the Clive DU PA
22 November 2015 20
over time and lake concentrations would be expected to do so concurrently, lake concentrations
do not necessarily decrease over time in the Deep Time Model because of the single mixing cell.
The Deep Time Model assumes that changes in climate will continue to cycle in a similar fashion
to the climate cycles that have occurred since the onset of the Pleistocene Epoch. These changes
follow those observed in the marine oxygen isotope record (Figure 2). The record captures major
climate regime shifts on a global scale and is used in this scenario in conjunction with expert
opinion (C.G. Oviatt, personal communication) plus site-specific sediment core and Clive pit
wall information to determine the approximate periodicity of lake events. However, uncertainties
exist due to the limitations related to the quality of the sediment core data.
It is assumed that during the 100-ky climatic cycles intermediate or deep lakes will reach the
elevation of Clive. Although a definitive distinction is not made, lakes that reach the elevation of
Clive but do not develop into a deep lake are considered intermediate lakes. These intermediate
lakes are also assumed to be large enough that their wave action will destroy the embankment.
Intermediate lakes might occur during the transgression and regression of a deep lake, or might
occur during a glacial cycle that does not produce a deep lake, perhaps in conjunction with
glacial cycles that are shorter and less severe than the 100-ky glacial cycles previously discussed.
In general, variation in lake elevation is assumed to be associated with all types of lakes. The
variation is due to local temporal changes in temperature, evaporation and precipitation. For
example, the Great Salt Lake has seen elevation changes of several meters in the past 30 to 40 yr.
The Great Salt Lake has also seen greater elevation changes in the past 10 ky, but in no cases
since the Younger Dryas has the Great Salt Lake reached the elevation of Clive (Oviatt, 2015).
Sedimentation is assumed to occur during these intermediate lake events at higher annual rates
than is assumed to occur for the open-water phase of deep lakes. This is based upon the pre-
Bonneville lacustrine cycles that are documented in Table 3 (Clive pit wall interpretation, see
Appendix A). The lake is assumed to recede after some period of time, at which point a shallow
lake (e.g., similar to the Great Salt Lake) will occupy Bonneville basin until the next
intermediate or deep lake cycle.
In the deep lake scenario, a deep lake forms throughout the Lake Bonneville basin in response to
major glaciation in North America and the Northern Hemisphere, following the ongoing 100-ky
glacial cycle. Increases in precipitation and decreases in evaporation over the long term, and
subsequent increases in discharge to the Bonneville basin via rivers that drain high mountains
along the eastern side of the basin have resulted in lakes that are more than 30 m deep and cover
an area similar to that of the most recent deep lake episode (e.g., Lake Bonneville, Provo
Shoreline). A similar extent of lake formation (geographic area, lake depth) is assumed to occur
in the future. Under such a scenario, the depth of a lake at the location of the Clive facility could
be many tens of meters. Resulting lake sedimentation at the Clive site will be high rates of
deposition of clastic sediments during intermediate lake events and much slower rates of
carbonate deposition during deep lake events. A key difference between the deep lake scenario
and the intermediate lake scenario is that both the transgressive and regressive phases of lake
formation are considered with the intermediate lake. Transgressive and regressive phases of lake
formation can lead to brief periods of rising and falling water levels in both phases. These phases
of transgression and regression are also assumed to have higher sedimentation rates than the
open-water phase. Upon the complete regression of a deep lake, it is assumed that only
intermediate lakes will form until the deep lake associated with the next climate cycle occurs.
Deep Time Assessment for the Clive DU PA
22 November 2015 21
6.0 A Heuristic Model for Relating Deep Lakes to Climate
Cycles from Ice Core Temperature
6.1 Introduction
In this section, a model is presented for estimating lake elevation that uses surface temperature
deviations from the EPICA Dome C ice core data (Jouzel et al., 2007), which is used to support
the modeling of future intermediate and deep lakes in the Deep Time Model. The model of lake
elevation is not intended to be highly accurate, but rather is aimed at capturing the major lake-
cycle features as shown in the studies conducted by Oviatt et al. (1999), Link et al. (1999), and
the sediment core and pit wall interpretations (C.G. Oviatt, personal communication). This model
is not used as a predictive model but rather to form a basis for the character and dynamics of lake
events in the Deep Time Model.
The deep-sea benthic δ18O record is in excellent agreement with the EPICA Dome C deuterium
measurements for the last ~810 ky (Jouzel et al., 2007). Temperature anomaly data for the past
810 ky were obtained from the World Data Center for Paleoclimatology, National Oceanic and
Atmospheric Administration/National Climate Data Center. These data are made available based
on calculations described in Jouzel et al. (2007), and are plotted in Figure 3. From the 810 ky of
data, the temperature deviations range from Tmin = –10ºC to Tmax = +5ºC. This range is used to
bound extreme events.
Water balance in the Bonneville basin is affected by many complex processes, so modeling water
balance simply as a function of temperature alone is not expected to produce precise results, but
instead provides a coarse representation. The conceptual model is based upon a water balance
reservoir model of precipitation versus evaporation. If precipitation outpaces evaporation, the
lake elevation increases. If evaporation outpaces precipitation, then the lake elevation decreases.
Precipitation and evaporation are affected directly by temperature, but long-term patterns of
precipitation are affected more greatly by the presence or absence of continental glaciation in
North America. Thus, glaciation is modeled first using a simple reservoir model depending on
temperature.
6.2 Glaciation
The water balance model begins by constructing a “continental glacier”; an artificial construct
that represents a glacier large enough to affect precipitation levels in the Bonneville Basin. The
extent of glaciation in proximity to the Bonneville basin is assumed to be zero initially, which is
a reasonable approximation for the start time of 785 ka, a start time chosen because it
corresponds to a warmer climate phase (data from Jouzel, et al., 2007; see Figure 3). For each
time step of 500 yr, an increase in glacial magnitude is dependent on temperature deviation (ΔT)
as scaled in Jouzel (see Figure 3):
Deep Time Assessment for the Clive DU PA
22 November 2015 22

Figure 3. Temperature deviations for the last 810 k (from Jouzel et al., 2007)

Deep Time Assessment for the Clive DU PA
22 November 2015 23
𝐺𝑙𝑎𝑐𝑖𝑎𝑙𝑎𝑑𝑑𝑖𝑡𝑖𝑜𝑛(Δ𝑇)={
0 if Δ𝑇≥Δ𝑇𝐺𝑀𝑎𝑥1
𝑁𝐺𝐴
((𝑒𝑅𝐺𝐴⋅(Δ𝑇𝐺𝑀𝑎𝑥–Δ𝑇)–1))if Δ𝑇<Δ𝑇𝐺𝑀𝑎𝑥
(1)
where
NGA is a normalizing constant:
𝑁𝐺𝐴=𝑒𝑅𝐺𝐴⋅(Δ𝑇𝐺𝑀𝑎𝑥−Δ𝑇min )(2)
RGA is a rate parameter (yr-1), and
TGMax is a threshold temperature (degrees Celsius).
As glaciation here is an artificial construct for modeling purposes, the units and scale of the
glacial “magnitude” are arbitrary. The parameters of the precipitation model described below
must be calibrated appropriately to the scale of the glaciation model.
For each time step, the decrease in glacial magnitude is also modeled as a function of
temperature:
𝐺𝑙𝑎𝑐𝑖𝑎𝑙𝑠𝑢𝑏𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛(Δ𝑇)={
0 if Δ𝑇≤Δ𝑇𝐺𝑀𝑖𝑛𝑆𝐺𝑆
𝑁𝐺𝑆
(𝑒𝑅𝐺𝑆⋅(Δ𝑇−Δ𝑇𝐺𝑀𝑖𝑛)–1)if Δ𝑇>Δ𝑇𝐺𝑀𝑖𝑛(3)
where
NGS is a normalizing constant:
𝑁𝐺𝑆=𝑒𝑅𝐺𝑆⋅(Δ𝑇𝑚𝑎𝑥−Δ𝑇𝐺𝑀𝑎𝑥) (4)
RGS is a rate parameter (yr-1), and
TGMin is a threshold temperature (degrees Celsius).
The change in glacial magnitude for a time step is thus:
𝐺𝑙𝑎𝑐𝑖𝑒𝑟𝑡=𝑚𝑎𝑥[0,𝐺𝑙𝑎𝑐𝑖𝑒𝑟𝑡−1 +𝐺𝑙𝑎𝑐𝑖𝑎𝑙𝑎𝑑𝑑𝑖𝑡𝑖𝑜𝑛(Δ𝑇𝑡)−𝐺𝑙𝑎𝑐𝑖𝑎𝑙𝑠𝑢𝑏𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛(Δ𝑇𝑡)](5)
where the t subscript is a time step index. The time step used for the model is 500 yr.
The parameters of the model were calibrated heuristically to compute parameters that produced a
glacial cycle that appeared reasonable for this coarse model. The set of parameters computed
was:
Δ𝑇𝐺𝑀𝑎𝑥=−6
𝑅𝐺𝐴=0.25
Δ𝑇𝐺𝑀𝑖𝑛=−6.0
𝑅𝐺𝑆=0.2
𝑆𝐺𝑆=5.0
(6)
Deep Time Assessment for the Clive DU PA
22 November 2015 24
The change in the glacial magnitude for a particular time step as a function of temperature is
shown in Figure 4. These values lead to slow growth during the very cold phases (Jouzel
temperature deviations of less than –6°C) of the glacial cycle, and rapid recession during warm
phases (temperature deviations of greater than –6°C).
6.3 Precipitation
A coarse model for precipitation in the Bonneville basin was developed dependent on global
temperature (as precipitation generally increases with global temperature), lake surface area
(which affects recharged evaporation), and an additional effect that depends of the magnitude of
the continental glacier. The precipitation in meters of annual rainfall is modeled as:
𝑃𝑡(Δ𝑇𝑡,𝐿𝑡−1,𝐺𝑡−1)=𝐵𝑃+𝑅𝑃𝑇⋅Δ𝑇+𝑅𝑃𝐿𝑆𝐴⋅𝑆𝐴(𝐿𝑡−1)+𝑆𝑃𝐺⋅𝑒𝑅𝑃𝐺⋅𝐺𝑡−1 (7)
where
BP is a baseline precipitation,
RPT is a coefficient of linear effect of global temperature,
RPLSA is a coefficient of linear effect of the surface area of the lake, and
SA(L) is the surface area in km2 associated with lake elevation L.
The effect of temperature and lake surface area are modeled as linear, while the glacial effect is
exponential with respect to glacier size. The set of parameters calibrated to the glacial magnitude
model are:
𝐵𝑃=0.30
𝑅𝑃𝑇=0.005
𝑅𝑃𝐿𝑆𝐴=2 × 10−6
𝑆𝑃𝐺=0.06
𝑅𝑃𝐺=0.03
(8)
The precipitation is then converted to a volume by multiplying by the area of Bonneville basin
(approximately 47,500 km2).
6.4 Evaporation
Evaporation rate in the region is modeled as a function of temperature:
𝐸𝑡(Δ𝑇𝑡)=𝐵𝐸+𝑆𝐸
𝑁𝐸
⋅𝑒𝑅𝐸⋅(Δ𝑇−Δ𝑇𝑚𝑖𝑛)(9)
where 𝑁𝐸is a normalizing constant:
𝑁𝐸=𝑒𝑅𝐸⋅(Δ𝑇𝑚𝑎𝑥−Δ𝑇min )(10)
The evaporation is then converted to a volume by multiplying by the area of the basin.

Deep Time Assessment for the Clive DU PA
22 November 2015 25
The calibrated parameters are:
𝐵𝐸=0.32;
𝑆𝐸=0.3
𝑅𝐸=0.05
Δ𝑇𝑚𝑖𝑛=−10
Δ𝑇𝑚𝑎𝑥=5
(11)
If the precipitation volume exceeds the evaporation volume, then the difference is added to the
lake volume, and the lake elevation is calculated from the total lake volume.

Figure 4. Glacial change as a function of temperature for the coarse conceptual model

Deep Time Assessment for the Clive DU PA
22 November 2015 26
If the evaporation volume is greater than the precipitation volume, then the total evaporation is
adjusted downward to adjust for the actual surface area exposed (rather than the full surface area
of the basin as used in the initial calculation). The difference between the adjusted evaporation
and the precipitation is then subtracted from the lake volume, and the lake surface elevation is
calculated from the total lake volume.
Δ𝑉𝑜𝑙𝑢𝑚𝑒𝑡={
[𝑃𝑡(Δ𝑇𝑡)−𝐸𝑡(Δ𝑇𝑡)]⋅𝑆𝐴𝑏𝑎𝑠𝑖𝑛if 𝐸𝑡(Δ𝑇𝑡)<𝑃𝑡(Δ𝑇𝑡)
[𝑃𝑡(Δ𝑇𝑡)−𝐸𝑡(Δ𝑇𝑡)]⋅𝑆𝐴(𝐿𝑡−1)
𝑆𝐴𝑏𝑎𝑠𝑖𝑛
if 𝐸𝑡(Δ𝑇𝑡)≥𝑃𝑡(Δ𝑇𝑡)(12)
6.5 Simulations
For simplicity, lake volume and glacial magnitude are assumed to be zero at the first time step
(785 ka), as that time step corresponds to a warm climate phase. The values for the parameters
given above are calibrated graphically to produce reasonable precipitation versus evaporation
values. Several lake elevation histories were simulated by simulating the parameter values of the
model probabilistically. The distributions for the parameters were lognormal with medians equal
to the parameter values listed in Equations (6), (8), and (11). The simulations provide a variety of
behaviors depending on the combination of parameters simulated.
A few common features are apparent in the simulated results. The largest lakes tend to occur at
the times of Lake Bonneville, Little Valley, and Lava Creek, and the smallest 100-ky cycle lake
occurs in δO18 cycle 14 (~533 ka), which matches the scientific record. When the simulated
glaciation effects are small (RGA and RGS), precipitation change in the model is due primarily to
temperature change. In this case, deep lakes form with few intermediate lakes, as the lake
elevation history in the top graph in Figure 5 shows. When glaciation effects are larger, then deep
lakes tend to last longer, and intermediate lakes form, as the lake elevation history in the lower
graph of Figure 5 shows.
The simulation models were then calibrated further by combining the simulated lake histories
with sedimentation rates seen in sediment cores. Based upon the results of this coarse model
calibration, some assumptions are carried forward to the Deep Time Model.
1. The 100-ky cycle in global temperature is a strong indicator of the return of a deep lake.
While not all simulations showed a lake returning to the Clive elevation in every 100-ky cycle
(particularly δO18 cycle 14), the results were consistent enough to treat as systematic behavior for
a heuristic model.
2. Intermediate lakes should be a part of the Deep Time Model, because sedimentation rates
did not calibrate well with simulations that produce only deep lakes.
3. Intermediate lakes are more likely to occur in the later stages of the 100-ky cycle than in
the early stages, primarily in conjunction with the slowly decreasing temperatures across the
cycle (as opposed to the relatively rapid warming period that occurs at the end of a 100-ky
cycle).
Deep Time Assessment for the Clive DU PA
22 November 2015 27

Figure 5. Two example simulated lake elevations as a function of time, with Clive facility
elevation represented by green line
Deep Time Assessment for the Clive DU PA
22 November 2015 28
7.0 Deep Time Modeling Approach
7.1 Introduction
The GoldSim systems analysis software (GTG, 2011) is used to construct the Clive DU PA
Model v1.4. The same Species list of contaminants, material properties, and site geometry are
retained from the Clive DU PA Model v1.2. The standalone DTSA Model is combined with the
deep time container of the Clive DU PA Model v1.2 in the Clive DU PA Model v1.4 deep time
container.
The DU waste inventory for the start of deep time is taken from the Clive DU PA Model v1.4
Federal DU Cell Disposal container at the time the first lake returns, which changes for each
realization. The DU waste is disposed below current grade. Contaminant fate and transport are
captured in the Federal DU Cell until the first lake returns. Radionuclides above grade when the
first lake returns are dispersed across the lake area and assumed to be available to diffuse into
any lake that appears. “Above ground” radionuclides are assumed to be at least 2 m above the
original ground surface, where eolian processes deposit at least 2 m of material in the 50,000
years or more before a lake returns. Remaining radioactivity in the lowest six waste layers (about
the lowest 2.5 m of the embankment) at the time the first lake appears is used as the Rn flux
inventory for the Deep Time Model.
The Deep Time Model is largely a heuristic representation of deep time. The underlying concepts
are that a lake will return to the elevation of Clive at some point in the future, and new lake
sediments will be sufficiently thick after the first lake recedes that radon flux will meet
regulatory guidelines. Contaminant fate and transport after the first lake returns are not evaluated
in the Deep Time Model, excepting radioactive decay and the ingrowth of progeny.
As previously discussed, the depth of lake and eolian sediments removed at the Clive location
due to wave action and the residual material from the destroyed embankment are expected to be
approximately equal, and their effects essentially cancel. Therefore, the thickness of residual
embankment material and sediment overlying the disposed DU waste at the time when the first
intermediate lake recedes will be effectively equivalent to the thickness of eolian sediments
deposited up until that point in time, represented by the rising elevation of the surrounding grade.
The Deep Time Model calculates radon ground surface flux from radionuclides in the disposed
DU waste buried beneath this layer. Dose to a rancher from this radon flux is calculation to
provide a reference point to interpret the significance of the radon flux.
7.2 Deep Lake Characteristics
The 100-ky climate cycle is treated as a sufficiently robust effect to create a hypothetical lake
that will reach and exceed the elevation of the Clive site during each glacial cycle. The exact
time of occurrence is not a crucial parameter, due to the slowly-changing concentrations during
deep time. Thus, the lake is set to be present during each 100-ky interval, with time beginning at
10 ky (the end of the performance period for the quantitative dose assessment component of the
PA).

Deep Time Assessment for the Clive DU PA
22 November 2015 29
There is limited information from the Quaternary geologic record for the duration of time that the
Clive location has been under water. Lake Bonneville has been estimated to have been present at
the elevation of Clive for an interval of approximately 16 ky (Oviatt et al., 1999). Durations of
pre-Lake Bonneville deep lakes are uncertain. Thus, a conservative choice was made to allow
deep lakes to remain an average of about 20 ky (conservative in the sense that more
radionuclides will migrate into the water column). The occurrence time for each deep lake is set
by choosing a start time some number of yr prior to the 100-ky mark. The start time is
represented by a lognormal distribution with geometric mean of 14 ky prior to the 100-ky mark,
and a geometric standard deviation of 1.2. The end time is represented by a lognormal
distribution with geometric mean of 6 ky after the 100-ky mark, and a geometric standard
deviation of 1.2. These distributions are depicted in Figure 6.

Figure 6. Probability density functions for the start and end times for a deep lake, in yr
prior to the 100-ky mark and yr after the 100-ky mark, respectively.

Deep Time Assessment for the Clive DU PA
22 November 2015 30
7.3 Intermediate Lake Characteristics
Intermediate lakes are modeled as potentially occurring during the transgressive and regressive
phases of deep lakes and at any time between deep lake events. In order to reflect the slow
decrease in temperature over the 100-ky cycle, the occurrence time for intermediate lakes is
modeled as a Poisson process with a rate that increases linearly over the cycle time, from a rate
of 0 to 7.5 lakes per 100 ky. This process produces an average of about 3 intermediate lakes per
100 ky. There is little recorded basis for this number, but it matches reasonably with the heuristic
model and was chosen so that long-term sedimentation rates matched the average clastic
sediment thickness observed in studies of lake cores from previous lake cycles.
There is virtually no information for the duration of intermediate lakes, due to the high mixing
rate of lake sediments, which prohibits establishing the chronology of individual stratigraphic
layers from studies of cores of intermediate lake sediments. Thus, a distribution was chosen to
roughly calibrate with the heuristic model: lognormal with geometric mean of 500 y and
geometric standard deviation of 1.5.
7.4 Sedimentation Rates
As previously mentioned, the Deep Time Model makes a distinction between deep and
intermediate lakes with regard to sedimentation.
 The sedimentation patterns of deep lakes are assumed to be similar to observed intervals
of carbonate marl from Lake Bonneville or Lake Provo, and are assumed to occur no
more than once per 100-ky glacial cycle. The depth of deep lakes is significantly greater
than the depth of wave action and slow precipitation of carbonate is assumed to be the
dominant sedimentation process.
 Intermediate lakes are defined as lakes that reach and exceed the altitude of the Clive site
but are not large (or deep) enough that carbonate sedimentation is the dominant mode of
lake sedimentation. The transgressive and regressive phases of the Bonneville and Provo
shoreline lakes represent intermediate lakes formed during transient lake cycles where
the lake levels exceeded the elevation of Clive and lake sedimentation was dominated by
clastic deposits associated with wave activity and reworking of pre-existing lake and
eolian sediments (see Table 2 for the chronology of the lake cycles).
 Shallow lakes, similar to the modern Great Salt Lake, are assumed to exist at all other
times, but these are irrelevant to the geomorphology of the Clive site and thus are not
explicitly modeled.
Deposition of eolian and lake sediments in the area of the Clive facility is a continuous process
that occurs during shallow, intermediate and deep lake periods. During shallow lake periods, as
observed in present-day conditions, eolian deposition of sand, and silt/loess is the primary
sedimentary mechanism. However, eolian deposits are rarely observed in sediment cores,
presumably because of reworking of the depositions during lake transgressions and mixing with
lake-derived sediments. Note however that the upper part of the Clive quarry exposure is now
known to be of eolian origin (Neptune, 2015a) and paleosoils and eolian deposits have been
observed in the pre-Lake Bonneville sedimentary deposits at Clive and described in the
Burmester core indicating prolonged periods of subaerial exposure. Intermediate lake sediments
Deep Time Assessment for the Clive DU PA
22 November 2015 31
include chemical, biogenic, and terrigenous sediments, with their proportions dependent on the
size and duration of the lake and the interplay between deep lake deposition and near-shore
sedimentary processes. Schofield et al. (2004) note that the large fetch of Lake Bonneville
(distance of wave forming winds over the water) produced a variety of wave-dominated
erosional and depositional sedimentary and geomorphic features. They identified cross-sections
of erosion-dominated and deposition-dominated shorelines and the composite sedimentation
rates of shoreline profiles will be dependent on local process of wind/wave erosion and
deposition and supply of sediments from alluvial fans flanking pluvial lakes (Schofield et al.,
2004). Moreover, eolian depositional layers are not commonly observed in the sediment cores, so
the model effectively combines eolian deposits with lake sediments. The mixing probably occurs
during intermediate lake cycles, which are likely to be the first lakes after interglacial periods.
These assumptions require that there is a mixing depth associated with each lake recurrence.
However, the mixing process itself makes it difficult to assign mixing depths for the different
layers in the sediment cores. Mixing depths are probably determined by the dynamics of wave
activity and resulting erosion/deposition during lake transgressions and regressions.
Deep lakes, in contrast, have similar sediment deposition rates to intermediate lakes in their
transgressive and regressive phases, but have slower rates of sedimentation when the lake is deep
enough that the dominant process is predominantly precipitation of chemical and biogenic
material from the lake waters. Studies of the sediment cores are able to distinguish between
layers associated with intermediate lakes with predominant sediment mixing, and sedimentary
layers associated with a deep lake that are dominated by carbonate layers (marl).
For deep lakes, a sedimentation rate is modeled as a lognormal distribution with geometric mean
of 120 mm/ky and geometric standard deviation of 1.2, a distribution that covers the range of
observed values for deep lakes. This distribution is represented in Figure 7. The sedimentation
rate is applied for the simulated duration of the deep lake. In addition, sedimentation is added at
the beginning of the lake cycle as well as the end that represents the shallow phase of the
transgressive and regressive lakes. This additional sediment mimics the behavior of an
intermediate lake.
For intermediate lakes (and shallow phases of deep lakes), there is high likelihood of multiple
short-term transgressions and regressions with respect to the elevation of Clive. For example, the
Clive pit wall (Appendix A) shows three distinct lakes after the deep-water phase of Lake
Bonneville and three distinct lakes prior to the deep-water phase of Lake Bonneville. Without
further systematic study of sediment cores and trench sections in and around the Clive site,
including chronology studies, it is impossible to determine if these distinct lakes were separated
by a few years or a few hundred years; i.e., whether they are distinct lake events or simply part of
the transgression and regression of Lake Bonneville. However, based upon current behavior of
the lake, some year-to-year variation in the lake elevation occurs, in addition to the longer-term
trends in lake elevation.

Deep Time Assessment for the Clive DU PA
22 November 2015 32

Figure 7. Probability density function for sedimentation rate for the deep-water phase of a
deep lake
Another heuristic model was constructed to evaluate the effect of the short-term variation. The
lake elevation for the years 1848 through 2009 is available from the Saltair Boat Harbor
monitoring site (USGS, 2001), as shown in Figure 8. The year-to-year variation can be modeled
as a second-order autoregressive process AR(2) (Brockwell and Davis, 1991), a model that
accounts for year-to-year temporal correlations in the variation. An AR(2) process was simulated
and added to a transgressive or regressive curve based upon the simplified model previously
presented. Examples of these simulations are given in Figure 9. As can be seen in the figure, the
short-term variation can result in lakes covering the Clive elevation for a short time, receding for
a short time, then rising again, often multiple times in a single transgression cycle. A similar
simulation was performed for simulated intermediate duration lakes as well. The transgressive
and regressive phases of a deep lake are assumed to behave similarly to the intermediate lakes in
that they averaged about four total occurrences of “mini-lakes;” i.e., occurrences of a rise above
the elevation of Clive followed by a drop below for at least one year.

Deep Time Assessment for the Clive DU PA
22 November 2015 33

Figure 8. Historical elevations of the Great Salt Lake
The distribution for sediment thickness for intermediate lakes was thus based upon simulating
this multiple mini-lake behavior. First, the number of mini-lakes associated with an intermediate
lake was simulated as 1 plus a Poisson random variable with rate 3 (the “plus 1” being necessary
to ensure at least one event in order to match the definition of a lake event). The sedimentation
for each mini-lake was simulated using a distribution based upon the sedimentary deposits of
mini-lakes exposed in the Clive pit wall, using the six distinct “mini-lakes” in Table 3 (all layers
except the one that corresponds to the deep-water phase of Lake Bonneville). These data are
represented in a lognormal distribution of sediment thickness with geometric mean 0.75 m and
geometric standard deviation 1.4.

Deep Time Assessment for the Clive DU PA
22 November 2015 34

Figure 9. Simulated transgressions of a deep lake including short-term variations in lake
elevations
The total sedimentation for all mini-lakes associated with a simulated intermediate lake cycle
was then added together to produce a total sedimentation for the intermediate lake. A distribution
was then based upon all simulated intermediate lake sedimentations, a lognormal distribution
with geometric mean 2.82 m and geometric standard deviation 1.71, as presented in Figure 10.
Note that the sedimentation pattern for intermediate lakes is represented as a distribution of
composite sediment thickness and contrasts with a distribution of sedimentation rates assumed
for deep lakes. The net effect is that the sedimentation rates are on the order of 15 to 20 m per
glacial cycle (100-ky). For the duration of the model (2.1 My), this implies sedimentation of
more than 300 m. The Basin and Range system accommodates this rate of sedimentation because
it is an extensional system; i.e., sedimentation continues as the basins expand and subside,
maintaining similar elevation in each cycle.
Deep Time Assessment for the Clive DU PA
22 November 2015 35

Figure 10. Probability density function for the total sediment thickness associated with an
intermediate lake (or the transgressive of regressive phase of a deep lake)
7.5 Eolian Depositional Parameters
Studies of eolian deposits in multiple quarry exposures at the Clive site and in surface exposures
west and southwest of the site show that deposition of eolian sand and silt is now occurring and
will continue to occur in the future as long as the at grade site elevation is exposed at the surface
(above the elevation of lake levels; Neptune, 2015a).
7.5.1 Field Studies
Field studies of the eolian depositional history at the Clive Disposal Site were conducted in
December 2014 to provide information for characterizing eolian deposits and establishing eolian
depositional rates for the original DTSA Model (Neptune, 2015a). The primary goals of the field

Deep Time Assessment for the Clive DU PA
22 November 2015 36
studies were to evaluate the modern geological and depositional setting of the Clive site, and to
assess the stratigraphy of the Holocene and Pleistocene lake sedimentation of Lake Bonneville
and post-lake depositional processes within the Clive site including the following:
1. Re-evaluating the stratigraphic section previously described by Oviatt (1985, cited in
Neptune, 2015b).
2. Describing the eolian sediments and processes affecting the sediments.
3. Measuring variations in thickness of the deposits across the site.
4. Providing sufficient replicate measurement at multiple sites to estimate eolian sediment
thicknesses and the variation in eolian sediment thicknesses at the Clive site.
The field studies achieved these primary goals, and the replicate measurements of the thickness
of eolian deposits located in the upper part of the stratigraphic section were made at multiple
locations on and in the vicinity of the Clive Disposal Site. The data are presented in Neptune
(2015) and are used below to develop input probability distributions for the Deep Time Model.
7.5.2 Probability Distributions for the Depth and Age of Eolian Deposition
The Deep Time Model requires specification of input probability distributions for the depth of
eolian deposition and the age of the eolian deposits. Together, these two variables provide the
information needed to estimate the rate of eolian deposition. The distribution for the depth of
eolian deposition is based on the field data described above (Neptune, 2015a), whereas the
distribution for the age of the eolian deposits are derived from a summary paper by Oviatt
(2015).
An assumption is made that the described eolian deposits at the Clive site represent an integrated
time interval of eolian sediment accumulation, modification by processes of soil formation and
minor modifications by processes of surface erosion. These deposits approximate a steady-state
representation of eolian processes since the regression of Lake Bonneville and these processes
should continue into the future until conditions at the site change considerably (e.g., natural
climate change). The distributions are based on the depth of eolian deposition since Lake
Bonneville regressed below the elevation of Clive and estimations of the age at which regression
below the Clive elevation occurred (Neptune, 2015a) These distributions are used to model
future eolian deposition until the return of a lake at the elevation of Clive.
The data presented in Table 4 from Neptune (2015a) are the measured thicknesses of eolian silt
in quarry walls and excavated surfaces for the Clive Disposal Site. The mean of the deposits is
72.7 cm, and the standard deviation is 16.6 cm. There are 11 data points, and the data are
reasonably symmetric about the mean. Consequently, a normal distribution is specified for the
Deep Time Model with a mean of 72.7 cm and a standard error of 5.0 cm. A reasonable
simulation range considering ± 3 standard errors would be 57.5 to 87.5 cm. The minimum of the
normal distribution was set to a very small number and the maximum was set to a very large
number so that the distribution was not unnecessarily restricted.
This distribution represents spatio-temporal scaling, so that the distribution is of the average
depth of eolian deposition at the Clive site since Lake Bonneville regressed below the site. This
provides the best representation of the future eolian depositional rates over the long timeframes
and spatial scales of the Deep Time Model.
Deep Time Assessment for the Clive DU PA
22 November 2015 37
Table 4. Thickness measurements from field studies of eolian silt near Clive
Neptune Field Studies December 2014
Site GPS Coord GPS Coord Silt Thick Date
UTM E UTM N (cm) (mm/dd/yy)
Clive 29-1 321354 4508262 90.0 12/16/14
Clive 29-2 321390 4508256 80.0 12/16/14
Clive 29-3 321423 4508248 80.0 12/16/14
Clive 29-4 321502 4508236 60.0 12/16/14
Clive 29-5 321239 4508283 110.0 12/16/14
Clive 5-1 320813 4504729 55.0 12/16/14
Clive 5-2 320869 4504730 70.0 12/16/14
Clive 5-3 320914 4504731 60.0 12/16/14
Clive 5-4 321041 4504732 70.0 12/16/14
Clive Hand-Dug-1 322093 4507482 70.0 12/17/14
Clilve hand-Dug-2 320445 4507035 55.0 12/17/14

Mean 72.7

Std Error 5.0

Note that several replicate measurements were taken at each location (usually three or four), and
the results represent the average thickness at each location. These data are also supported by
previous data collected from shallow core studies at Clive, which also are presented in Neptune
(2015a). The documentation and uncertainty in the measurements of the eolian sediment
thickness from the core studies data is not as precise as those made in the Neptune field study;
however, the data are supportive of the results of the field study, indicating very similar patterns
of eolian thickness data. These data provide another 21 data points that have an average of 71 cm
depth of eolian deposits, with a standard error of 4 cm. Because of the uncertain pedigree and
lesser precision of the data from the core studies, they were not used in the distribution
development. Their use would have resulted in a much tighter distribution because of the scaling
effects of spatio-temporal averaging.
Ages of the deposits were determined from radiocarbon dating. The summary paper by Oviatt
(2015) provides the most recent compilation and interpretation of radiocarbon ages for the
chronology of Lake Bonneville. Based on information summarized in Figure 2 of Oviatt (2015)
and supported by the supplemental radiocarbon data referenced in the paper, the preferred
estimate for the age of the final regression of Lake Bonneville below the altitude of the Clive site
is about 13.5 ka. (Clive elevation 1304 m). A reasonable lower bound on the youngest or
minimum age for this event is 13.3 ka based on radiocarbon ages determined from organic
material collected in post-Bonneville wetland deposits (Oviatt, 2015). The reasonable oldest or
maximum age of lake regression at the Clive site is constrained by the age of the Provo shoreline
and reliable radiocarbon ages for sites above the altitude of the Clive site and below the Provo
shoreline. This reasonable maximum age is estimated to be about 14.5 ka. A distribution was
developed based on these values from Oviatt (2015) and on expert elicitation of Oviatt. Oviatt
suggested that values around 13.5 ka were more likely. Based on this information a beta
distribution was fit to approximate elicited quantiles. The following quantile inputs were used:
Deep Time Assessment for the Clive DU PA
22 November 2015 38
 Absolute minimum possible age – 13,000 yr
 Reasonable minimum age – 13,300 yr
 Most likely age – 13,500 yr
 Reasonable maximum age – 14,500 yr
 Absolute maximum possible age – 15,000 yr
After considering possible quantiles for the middle three terms, a beta distribution fit was agreed
upon with the following parameters:
 Minimum – 13,000 yr
 Maximum – 15,000 yr
 α (shape 1 parameter) – 3.318
 β (shape 2 parameter) – 7.498
This beta distribution has a mean of approximately 13,600 yr and a standard deviation of
approximately 270 yr. The mean is reasonably close to the specified most likely age of 13,500
yr. Quantiles of this beta distribution are provided below:
 2.5% – 13,174 yr
 10% – 13,284 yr
 20% – 13,378 yr
 50% –13,592 yr
 80% – 13,846 yr
 90% – 13,988 yr
 97.5% – 14,207 yr
The distribution is slightly positively skewed, hence the median is slightly less than the mean,
and the difference between the maximum and the median is greater than the difference between
the minimum and the median. Note that averaging is not employed for this distribution. The
distribution simply reflects the age over which eolian deposition has occurred. The rate of eolian
deposition is averaged for spatio-temporal scaling by dividing the depth of deposition by the age
over which deposition has occurred as described in the next section.
In principle, the rate of eolian deposition is the deposition thickness divided by the age over
which deposition occurs. However, an assumption is made that greater ages imply greater depths,
in which case there is a correlation between depth and age of eolian deposition. There are no data
to inform a correlation between these two variables. Although elicitation could be performed to
develop a correlation, the approach taken is to specify the correlation as uncertain across a range
of 0.5 to 1. In a sense, this distribution is chosen to indicate that the “data are more likely to be
correlated than not-correlated.” A uniform distribution is used across this range, but this input
will be tracked specifically in sensitivity analysis to determine if it is an important predictor of
the Deep Time Model output.
Deep Time Assessment for the Clive DU PA
22 November 2015 39
Using the input distributions and the correlation described above, the resulting distribution of
rate of eolian deposition in the model has a mean of approximately 5.3 × 10-5 m/yr, (roughly 53
cm every 10 ky) with a standard deviation of approximately 3.0 × 10-6 m/yr. A histogram of the
eolian deposition rate for 1,000 realizations is depicted in Figure 11. Quantiles from these
simulated data include:
 5% – 4.84E-05 m/yr
 10% – 4.96E-05 m/yr
 20% – 5.10E-05 m/yr
 50% – 5.34E-05 m/yr
 80% – 5.58E-05 m/yr
 90% – 5.71E-05 m/yr
 95% – 5.81E-05 m/yr
The distribution is symmetric, as evidenced by the normal distribution fit that is laid over the
histogram. The normal distribution has the mean and standard deviation as specified above, and
the quantiles, which show similar differences between the 95% quantile and the median and the
5th quantile and the median.
Overall, this intermediate product of the Deep Time Model suggests eolian deposition rates of
slightly more than 0.5 m every 10,000 yr.
7.6 Destruction of the Federal DU Cell
Destruction of the Federal DU Cell embankment was modeled assuming future lakes have
sufficient wave energy to destroy the above-ground portions of the cell. The precise lake
elevation needed for this to happen is not considered for the model, but the intermediate lakes
that occur in the model are intended to match this definition.
The first lake in the time period assessed is more likely to be an intermediate lake but can be
either an intermediate or a deep lake. The destructive energy is equivalent in either case, as the
conceptual model treats the transgressive phase of a deep lake as behaving similarly to an
intermediate lake.
The mass of material that is within the embankment above the grade of the surrounding land is
assumed to be eroded to grade and dispersed by wave action. This volume of above grade
material in the embankment, including fill material and cap material, is assumed to be mixed
with the sediment associated with the intermediate lake, and subsequently spread across a
dispersal area determined by the dynamics of wave activity.
The dispersal area parameter used in the original Deep Time model was estimated for a projected
area where the above grade embankment material could be spread by wave action using different
assumptions for the final dispersal thickness of the volume of embankment material. The
dispersal area was designed to be conservative (small sediment dispersal areas) giving higher
waste concentrations in sediment allowing increased dissolution of waste in lake water.
Deep Time Assessment for the Clive DU PA
22 November 2015 40

Figure 11. Eolian deposition rate results for 1,000 realizations (m/yr).
With below grade disposal of DU, the approach to estimating the dispersal area is revised and
based on a conceptual model for processes affecting the Clive disposal site with the return of a
lake. The following assumptions are used for the revised lake return scenario:
1. The Clive site will be affected by the return of a lake at some time in the future. The
lake event will be either an intermediate lake or the transgressive phase of a deep lake
with the lake processes the same for either event (degradation of the site by near-
shoreline wave action).
Histogram with Normal Curve
Eolian Deposition Rate
Fr
e
q
u
e
n
c
y
4.5e-05 5.0e-05 5.5e-05 6.0e-05 6.5e-05
0
2
0

4
0

6
0

8
0

1
0
0

1
2
0
Deep Time Assessment for the Clive DU PA
22 November 2015 41
2. Eolian deposition will occur during the interval after waste emplacement and before the
first return of a lake to the elevation of the Clive site.
3. Wave action associated with the lake return is assumed to completely remove the
above-grade embankment material above the DU waste.
4. Wave action will churn (rework) the eolian deposits and lake sediments. The maximum
depth of reworking of the eolian deposits is assumed to be about 1 meter based on the
geometry of shoreline deposits for Lake Bonneville.
5. Radionuclides within the above grade embankment will be dispersed by wave action
and mixed with eolian deposits and lake sediments.
6. The alternative models used for estimating sediment dispersal areas include:
a. Analogue sites of modern sedimentary processes dispersing sediments at
shorelines of the Great Salt Lake;
b. Field assessments of sediment dispersal during the transgressive phase of Lake
Bonneville at and around the Clive Disposal Area (Neptune, 2015a);
c. Assessment of wind directions from dune forms west and southwest of Clive
(Jewell and Nicoll, 2011)
Google Earth© imagery was used to identify and determine the areas of active shoreline
sedimentation for the Great Salt Lake assuming these patterns provide analogues for wave action
and sediment dispersal for the lake return scenario at Clive. Dispersal area estimations assumed
no longshore drift (minimum areas) and one dominant direction of longshore drift (maximum
areas). The Great Salt Lake analogue may be somewhat conservative (underestimate sediment
dispersal) for two reasons. First, the fetch length for a lake return at the Clive elevation would be
longer than the fetch length for the Great Salt Lake. Second, the observed sedimentation patterns
of the Great Salt Lake represent relatively short term dynamics of lakeshore processes – the
dispersal area of sediments for the return of a lake at the Clive site and erosion of the
embankment would likely develop over a timescale of multiple decades.
Google Earth© imagery was used to estimate alternative sediment dispersal areas using
constraints from field observations of the distribution of conglomerate and sand deposits of the
transgressive phase of Lake Bonneville. These estimations combined data from surface
landforms and quarry-wall exposures of lake sediments at Clive.
Finally, alternative sediment dispersal patterns were estimated using Google Earth© imagery for
Clive by centering the sediment dispersal at the Clive embankment and adjusting the dispersal
patterns for the topographic features of the Clive area.
The following percentiles were assigned to the composite data to establish a distribution for the
sediment dispersal parameter:
 1%: 4 km2 from smallest measured dispersal area
 5%: 10 km2 assuming only west-east wind directions
 15%: 15 km2 averaging dispersal areas for no longshore drift
Deep Time Assessment for the Clive DU PA
22 November 2015 42
 30%: 16 km2 averaging dispersal areas for N-S and SW-NE longshore drift
 50% 24 km2 assuming multidirectional winds and longshore drift
 75% 36 km2 averaging all single direction longshore drift dispersal areas
95% 55 km2 from maximum measured dispersal area
A gamma distribution was used to fit the percentages above, with mean of 24.2332 and standard
deviation of 11.43731. A typical probability density function of this distribution is shown in
Figure 12.

Figure 12. Probability density function for the area over which the waste embankment is
dispersed upon destruction
Deep Time Assessment for the Clive DU PA
22 November 2015 43
7.7 Radionuclide Concentration in DU Waste
After a lake recedes, radionuclides in the original DU waste disposal volume are not likely to
move to the surface in any significant amounts via diffusion or other processes. This section
discusses processes that are likely to occur in deep time relating to the original DU waste source.
Infiltration rates will increase over time, moving material downward via advection, counteracting
potential upward diffusion of radionuclides. The climate will become cooler and wetter, entering
a glacial period, resulting in the lake return. Estimates of future net infiltration at Clive are
supported by work for the Yucca Mountain Project. Faybishenko (2006) developed models
predicting infiltration rates for future climate states based on factors including predicted
precipitation, evapotranspiration, and temperature. The meteorological stations at Simpson and
Spokane in Faybishenko (2006, Table 3) provide a reasonable range of infiltration rates for Clive
of 40 mm/yr to 73 mm/yr, for a glacial phase.
An external, finely-discretized GoldSim model was used to test diffusion behavior at Clive,
along with higher infiltration rates. The model results showed that if infiltration increased even
to 10 mm/yr, downward advection would dominate upward diffusion in the model. Infiltration of
40 mm/yr to 73 mm/yr would move radionuclides that had diffused above the original grade
back below grade.
Dry periods during the inner-glacial timeframes would be expected to behave like current
conditions. Because of the uncertain nature of the deep time future conditions and timing and
because it is important to keep the Deep Time Model simple, it was assumed that until the first
lake returns, radionuclides migrate upwards via the processes of diffusion and plant and animal
transport and that the associated material and radionuclides above grade is spread across the site
dispersal area and is available to diffuse into an intermediate or deep lake. These simplifying
assumptions ignore increases in infiltration during wetter periods in the climate cycle, which is a
conservative approach.
7.8 Radionuclide Concentration in Sediment
The radioactivity per unit volume of sediment following the dispersal of the waste is estimated
using Equation 13 below.
The model calculates radioactivity by volume in the sediment layers, after the embankment has
been destroyed. The current implementation always mixes sediment with the full amount of
waste, and does not consider a mixing depth; i.e., the waste is always fully mixed and not
covered by sediment. Thus, radioactivity concentration in sediment is initially calculated under
the assumption that all of the waste in the waste embankment is mixed evenly with the sediment
that forms as a result of the lake destroying the embankment.
Concentration in sediment is initially calculated under the assumption that all of the waste that
was above grade in the waste embankment is mixed evenly with the sediment that forms with the
lake that destroys the embankment.
𝐶sediment =𝑅embankment
𝑉material above grade+𝑉sediment
.(13)

Deep Time Assessment for the Clive DU PA
22 November 2015 44
where
Rembankment is all remaining radioactivity in the embankment,
Vmaterial above grade is the volume of material in the above grade portion of the embankment
(estimated as 3,231,556 m3), and
Vsediment is calculated as the depth of sediment due to lake processes multiplied by
the area over which the waste is dispersed.
This calculation assumes that there is no loss of waste from the initial dispersal region. While
this calculation is counter to the modeling of dissolution into the water column of the lake, a
simplifying assumption is that all waste that dissolves into the lake precipitates back into the
sediment upon recession of the lake.
The concentrations in sediment are modeled as constant, except for decay and ingrowth, until a
new lake occurs. When a new lake occurs, the sedimentation associated with that lake is likely to
mix with some portion of the top layer of existing sediment and leave the lower layers of the
sediment buried beneath. However, for simplicity, a conservative approach is to mix all sediment
that contains waste, effectively keeping some portion of the waste near-surface. The
concentration is again the total radioactivity divided by the volume containing waste, but the
volume that contains waste now has the additional volume of sediment associated with the
current lake.
7.9 Radioactivity in Lake Water
When lake water is present, radionuclides will partition between the water phase and the solid
phase depending on element-specific solubility and sorption properties. Radionuclides remaining
in the pore water will then diffuse into the lake. The waste is likely to mix over a wide area of the
lake, and many forms of the waste are likely to bind with carbonate ions in the water, ultimately
precipitating into carbonate sediments. As a conservative assumption, upon recession of the lake,
all waste is assumed to precipitate back into the local sediments, meaning that all radionuclides
in the sediments are returned to the sediments when the lake regresses.
When a lake returns, the sediments are assumed to be fully saturated, and radionuclides are
partitioned from the sediment to the pore water within the sediment using the same partitioning
coefficients (Kd) used for other sedimentary soils in the model. An important difference between
the assumptions for this model and the model for transport from the embankment in the 10-ky
model is that the lake water is assigned a different solubility for uranium for the Deep Time
Model. While solubilities for all other radionuclides remain the same, the solubility for uranium
is reduced to that of U3O8 which is appreciably lower than other forms of uranium originally
present in the waste. This change in solubility for uranium is adopted because it is expected that
by the time the first lake returns, soluble uranium forms (UO3) either will have been leached
from the embankment into the shallow aquifer or will have been converted into U(IV), which is
also very insoluble.
As radionuclides associated with the sediments dissolve into the pore water, they diffuse into the
lake water using a constant flux model based upon Fick’s first law, with the following
assumptions:
Deep Time Assessment for the Clive DU PA
22 November 2015 45
 The concentration in sediment remains constant over the deep time period. The sediment
concentration should in fact diminish over time if enough mass is migrated into the water,
but for simplicity, the sediment concentrations are kept constant across time steps.
 The diffusion length from the radionuclides in the sediment diffusing into the lake is
about 0.5 m. This diffusive length value assumes the mixing depths of the sediment
correspond to diffusive processes from the sediment into the lake.
 Mixing depths are expected to be between 0 and 1 m, with 0.5 m being most likely. The
distribution was set up as a normal distribution with mean of 0.5 m and standard
deviation of 0.16 m so that 99% of the distribution will be between 0 and 1 m. The
distribution is truncated at 0 m so that no negative diffusion lengths are chosen.
Fick’s law for this case estimates the mass diffusing from a given volume of sediment into the
lake with time. The mass (or activity) per area per time is the flux. Fick’s law states that this flux
is given by the difference in mass concentration over a distance (the concentration gradient)
multiplied by a free-water diffusion coefficient, across a diffusive area. The calculation assumes
that there is a stagnant interface boundary layer of water between the sediment and the open
water that is the thickness of the diffusion length (~0.5 m). The assumption is also made that the
mass concentration is zero in the open water. The difference in concentration across the stagnant
layer is then the concentration in the sediment Cv minus the concentration in the open water or
Cv – 0 g/mL. Fick’s law applied to diffusion is used to define the mass (or activity) flux J:
𝐽= 𝑅
∆𝑡 𝐴= 𝐷𝑚 𝐶𝑣
𝑏𝑏𝑑𝑦.(14)
where
R is the mass (M) activity (T-1),
ΔT is the length of the time period (T),
A is the area of the sediment that contains the waste (L2),
Dm is the diffusion coefficient for the radionuclide in water (L2/T), and
bbdy is the thickness of the boundary layer.

Multiplying both sides of the equation by ΔT·A gives
𝑅=Δ𝑇⋅𝐷𝑚⋅𝐶𝑉
0.1 m ⋅𝐴.(15)
Concentration in lake water is calculated based upon the conservative assumption that the
radioactive material does not dilute in a large basin of the lake but rather remains in the water
column immediately above the dispersed area. The activity concentration in the lake water is
then calculated by dividing the total activity, R, by the volume of lake water. The volume of lake
water is the product of the lake depth and the dispersal area:
𝐶𝑣=𝑅
𝐷⋅𝐴(16)
Deep Time Assessment for the Clive DU PA
22 November 2015 46
where
Cv is concentration (M/L3 or T-1/ L3),
R is the mass (M) or radioactivity (T-1),
A is the area of the sediment that contains waste (the dispersed area, as L2), and
D is the depth of the lake (L).
There is an insufficient record of lake elevations to construct a data-based distribution for lake
depth. Thus, the distributions for lake depth are chosen based upon the conceptual model. Depths
for intermediate lakes have a Beta distribution with mean of 30 m, standard deviation 18 m,
minimum of 0 m, and maximum of 100 m. Depths for deep lakes have a Beta distribution with
mean 150 m, standard deviation 20 m, minimum of 100 m, and maximum of 200 m.
For intermediate lakes, the time step is about the duration of the intermediate lake. For deep
lakes, the lake may exist for several time steps in the GoldSim model, in which case the time step
is the portion of the time step for which the lake is present. When deep lakes cross multiple time
steps, the concentration in sediment is allowed to change between time steps (only due to decay
and ingrowth) and the activity in the lake water is accumulated over those time steps.
7.10 Modeling of 222Rn Flux
Radon-222 flux through the overlying sediment is calculated using the approach described in the
Nuclear Regulatory Commission (NRC) Regulatory Guide 3.64 Calculation of Radon Flux
Attenuation by Earthen Uranium Mill Tailings Covers (NRC, 1989). These equations were
developed for estimating radon flux from uranium mill tailings buried under a monofill cover.
For the Deep Time Model, an assumption is made that the material above the below-grade DU
waste and the additional lake sedimentation is homogenous material with properties similar to
those of the surrounding Unit 3 sediments. The use of an analytical model such as that described
in NRC (1989) allows radon flux to be estimated through a homogeneous cover of varying
thickness with minimal complexity.
The increasing depth of material covering the disposed DU waste over time will result in
attenuation of radon flux. However, this rate of attenuation will be partly offset by the slowly
increasing activity of the radioactive progeny of 238U. Previous modeling results, such as those
from the Clive DU PA Model v1.2, indicated that sediment accumulation overwhelms the
influence of progeny ingrowth.
Although the median and mean sediment thickness track closely, the mean radon ground surface
flux is much larger than the median. This strongly skewed result for radon flux is a consequence
of the non-linearities inherent in the NRC radon ground surface flux calculation. These are
equations (9) through (12) in NRC (1989):
Deep Time Assessment for the Clive DU PA
22 November 2015 47

(17)
The definitions of variables are available in the NRC Regulatory Guide (1989), but the salient
point is that these equations will produce a highly non-linear result, Jc, which is the ground
surface flux of radon. Although all of the inputs to the calculation are essentially normal
distributions, the division calculations, exponents, etc. in the equations produce non-linear
results.
Modeling of radon transport to the surface of the intact Federal DU Cell in the Clive DU PA
Model v1.2 does not lend itself to such simplified analytical solutions, because the cover is
constructed of layers with widely-varying properties. Radon diffusive flux is therefore integrated
with other transport processes employing a column of well-mixed cells, allowing for the vertical
redistribution of radionuclides over time throughout the disposal system by diffusive, advective,
and biotic processes. Because the above-ground part of the Federal DU Cell is assumed to be
dispersed by wave action from the first intermediate lake, these processes are not relevant to the
Deep Time Model except insofar as they affect radionuclide concentrations in the below-grade
waste cells.
7.10.1 Waste and Sediment Water Content
Volumetric water contents are defined for the DU waste, and for sediments overlying the waste,
in order to support radon diffusive flux calculations through these sediments.
In the 10,000-year model, the waste material is assumed to be Unit 3 material. In the Deep Time
Model it is also assumed to be Unit 3, for both the mound material that is directly above the
waste but below grade when the first lake returns and for the sediment material that is deposited
from deep and intermediate lakes in deep time. Sediment porosity is assumed to be the same as
Unit 3 porosity.
The Deep Time Model water contents for the cover materials after the first lake recedes are
based on concentrations of waste materials just above the DU waste, in Waste Cells 17 – 21 and
the upper-most waste cell containing DU waste, Waste Cell 22. This cell may not be completely
full of DU waste, because the discretization of the model may not match exactly the
discretization of the disposed wastes, so Cell 22 was included in these calculations for moisture
content. Because the Deep Time Model is now fully integrated in the v1.4 model, these values
are taken directly from the waste properties and align with those directly for each model
realization.
Deep Time Assessment for the Clive DU PA
22 November 2015 48
7.11 Human Health Exposure and Dose Assessment
In the Deep Time component of the GoldSim model, external radiation dose and radon inhalation
dose are evaluated for the time period after a lake returns. Specifically, this special analysis
evaluates dose at a time immediately after the first intermediate lake has formed and
subsequently receded. The wave action of the lake is assumed to have destroyed the
embankment. The DU wastes at this point in time when the intermediate lake has receded are
covered by a thickness of material equal to the thickness of the eolian sediments that have been
continually deposited at a constant rate over time, plus the deposition of lake sediments while the
intermediate lake exists. The lower stratum of material of thickness equivalent to the eolian
sediments is comprised of the waste layers that existed above the DU waste in the embankment.
Although these wastes at one time contained radionuclides that had migrated upwards from the
DU, these radionuclides are assumed to have been dissolved and dispersed during the time when
the intermediate lake was present. Therefore, both these materials as well as the lacustrine
sediments are assumed to be practically free of DU-related radionuclides in this modeling.
The purpose of the dose calculations the Deep Time component of the GoldSim model is to
determine whether hypothetical doses in Deep Time may be higher or lower than doses
calculated for the 10,000-year performance period. The Deep Time dose calculation results are
not considered to have independent validity. Rather, they are a tool for evaluating the relative
radiation dose during these two time periods. For the dose assessment during the first 10,000
years of the Clive DU PA v1.4 Model, two future use exposure scenarios are identified for the
Clive site: ranching and recreation. However, only ranching receptors are evaluated for the Deep
Time component of the model because their utilization of the area including the Clive site is far
greater than that of recreational users and their doses are therefore higher.
The radiological assessment method for the Deep Time the Deep Time component of the
GoldSim model calculates total effective dose equivalent (TEDE) as the product of exposure
(behavioral) parameters, dose conversion factors (DCFs), and the concentrations of radium and
gamma-emitting radionuclides in the DU waste. The calculations are analogous to those
described for the Ranching scenario during the 10,000-year performance period with two
exceptions:
1. Radon flux is calculated using the approach described in the Nuclear Regulatory
Commission (NRC) Regulatory Guide 3.64 Calculation of Radon Flux Attenuation by
Earthen Uranium Mill Tailings Covers (NRC, 1989). These equations were developed
for estimating radon flux from uranium mill tailings buried under a monofill cover, and
the properties of the overlying materials are homogenous material with properties similar
to those of the surrounding Unit 4 sediments.
2. The external DCFs are multiplied by radionuclide-specific modifying factors to account
for the attenuation of external gamma radiation due to the material that overlies the DU
waste. The modifying factors were calculated using the RESRAD computer code by
evaluating the ratio of external dose at different cover thicknesses to external dose with
no overlying material.
Deep Time Assessment for the Clive DU PA
22 November 2015 49
8.0 References
Adams, K.D., 2003, Age and paleoclimatic significance of late Holocene lakes in the Carson
Sink, NV, USA, Quaternary Research, Vol. 60, pp. 294–306, 2003.
Archer, D. and A. Ganopolski, 2005. A movable trigger: fossil fuel CO2 and the onset of the next
glaciation. Geochemistry, Geophysics, Geosystems, 6(5), doi:10.1029/2004GC000891.
Asmerom, Y., Polyak, V. J., and S. J. Burns, 2010. Variable winter moisture in the southwestern
United States linked to rapid glacial climate shifts. Nature Geoscience, 3, 114-117.
Balch D.P., A.S. Cohen, D.W. Schnurrenberger, B.J. Haskell, B.L.V. Garces, J.W. Beck, H.
Cheng, and R.L. Edwards, 2005. Ecosystem and paleohydrological response to Quaternary
climate change in the Bonneville basin, Utah, Palaeogeography, Palaeoclimatology,
Palaeoecology, Vol. 221, pp. 99–121.
Berger, A., 1988. Milankovitch theory and climate. Reviews of Geophysics, 26(4): 624-657.
Berger, A. and M. F. Loutre, 2002. An exceptionally long interglacial ahead? Science, 297: 1287-
1288.
Benson, L.V., S.P. Lund, J.P. Smoot, D.E. Rhode, R.J. Spencer, K.L. Verosub, L.A. Louderback,
C.A. Johnson, R.O Rye, R.M. Negrini, 2011, The Rise and Fall of Lake Bonneville Between
45 and 10.5 ka, Quaternary International, Vol. 235, p. 57-69.
Briggs, R.W., S.G. Wesnousky, and K.D. Adams, 2005, Late Pleistocene and late Holocene lake
highstands in the Pyramid Lake subbasin of Lake Lahontan, Nevada, USA, Quaternary
Research, Vol. 64, pp. 257–263, 2005.
Brimhall W. H. and L. B. Merritt, 1981. The geology of Utah Lake – implications for resource
management. Great Basin Naturalist Memoirs, 5: 24-42.
Brockwell, P.J. and R.A. Davis, 1991. Time Series: Theory and Methods. Springer-Verlag, New
York, NY.
Clark, P. U., Dyke, A. S., Shakun, J. D., Carlson, A. E., Clark, J., Wohlfarth, B., Mitrovica, J. X.,
Hostetler, S. W., and A. M. McCabe, 2009. The Last Glacial Maximum. Science, 325: 710-
714.
Cochelin, Anne-Sophie B., L.A. Mysak, Z. Wang, 2006, Simulation of long-term future climate
changes with the green McGill paleoclimate model: The next glacial inception. Climate
Change, 79, 381-401.
Cook, E.R., R. Seager, R.R. Helm Jr, R. S. Vose, C. Herweijer, and C. Woodhouse, 2010.
Megadroughts in North America: Placing IPCC Projection of Hydroclimatic Change in a
Long-Term Palaeoclimate Context, Journal of Quaternary Science, 25, Issue 1, 48-
61.Currey, D.R., G. Atwood, and D.R. Mabey, Map 73 Major Levels of Great Salt Lake and
Lake Bonneville, Utah Geological and Mineral Survey, Salt Lake City, UT, May 1984
Davis, O.K., 1998. Palynological evidence for vegetation cycles in a 1.5 million yr pollen record
from the Great Salt Lake, Utah, U.S.A., Palaeogeography, Palaeoclimatology,
Palaeoecology, Vol. 138, pp. 175–185.
Deep Time Assessment for the Clive DU PA
22 November 2015 50
Driscoll, N. W. and G. H. Haug, 1998. A short circuit in thermohaline circulation: a cause for
Northern Hemisphere glaciation. Science, 282: 436-438.
Einselle, G. and M. Hinderer, 1997. Terrestrial sediment yield and the lifetimes of reservoirs,
lakes, and larger basins. Geologische Rundschau, 86: 288-310.
EPICA community members, 2004. Eight glacial cycles from an Antarctic ice core. Nature, 429:
623-628.
Faybishenko, B. 2006. Climatic Forecasting of Net Infiltration at Yucca Mountain Using
Analouge Meteorological Data. Lawrence Berkeley National Laboratory. DT#46762
QA:NA. MOL.20061204.0226.
Fritz, S.C., 1996, Paleolimnological records of climatic change in North America, Limnol.
Oceanogr. 45, 882-889.
GTG (GoldSim Technology Group), 2011. GoldSim: Monte Carlo Simulation Software for
Decision and Risk Analysis, http://www.goldsim.com
Harding, B.L., A.W. Wood, and J.R. Prarie, 2012. The Implications of Climate Change Scenario
Selection for Future Streamflow Projection in the Upper Colorado River Basin, Hydrol.
Earth Syst. Sci, 16, 3989-4007.
Hart, W. S., Quade, J., Madsen, D. B., Kaufmann, D. S., and C. G. Oviatt, 2004. The 87Sr/86Sr
Ratios of Lacustrine Carbonates and Lake-level History of the Bonneville Paleolake System.
GSA Bulletin, 116: 1107-1119.
Haug, G. H. and R. Tiedemann, 1998. Effect of the Formation of the Isthmus of Panama on
Atlantic Ocean Thermohaline Circulation. Nature, 393: 673-676.
Hays, J. D., Imbire, J., and N. J. Shackleton, 1976. Variations in the Earth's Orbit: Pacemaker of
the Ice Ages. Science, 194: 1121-1132.
IAEA. The Safety Case and Safety Assessment for the Disposal of Radioactive Waste. Vienna,
Austria: International Atomic Energy Agency; Specific Safety Guide No. SSG-23;
2012.Jansen, E. J. Overpeck, K.R. Briffa, J.-C. Duplessy, F. Joosv, V. Masson-Delmotte, D.
Olago, B. Otto-Bliesner, W.R. Peltier, S. Rahmstorf, R. Ramesh, D.. Raynaud, D. Rind, O.
Solomina, R. Villalba, and D. Zang, 2007, Palaeoclimate. In: Climate Change 2007: The
Physical Science Basis. Contribution of Working Group I to the fourth Assessment Report of
the Intergovernmental Panel on Climate Change [Solomon, s. D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tigor and H.L Miller (eds.)] Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
Jewell, P.W. 2010, River incision, circulation, and wind regime of Pleistocene Lake Bonneville,
USA, Palaeogeography, Palaeoclimatology, Palaeoecology, 293, 41-50.
Jewell, P.W., and K. Nicoll, Wind Regimes and Eolian Transport in the Great Basin, U.S.A.,
Geomorphology 129, 1-13 (2011).
Jouzel, J., Masson-Delmotte, V., Cattani, O., Dreyfus, G., et al., 2007. Orbital and millennial
Antarctic climate variability over the past 800,000 yr. Science, 317: 793-796.
Deep Time Assessment for the Clive DU PA
22 November 2015 51
Kukla, G. J., R. K. Matthews, J. M. Mitchell Jr., Quat. Res. 2, 261 (1972)
Lisiecki, L. E. and M. E. Raymo, 2005. A Pliocene-Pleistocene stack of 57 globally distributed
benthic δ18O records. Paleoceanography, 20, PA1003, doi:10.1029/2004PA001071.
Link, P. K., Kaufmann, D. S., and Thackray, G. D., 1999. Field guide to Pleistocene lakes
Thatcher and Bonneville and the Bonneville flood, southeastern Idaho, In: Hughes S. S. and
G. D Thackray (eds.), Guidebook to the Geology of Eastern Idaho, Idaho Museum of Natural
History, p. 251-266.
Lyle, M., L. Heusser, C. Rovelo, M. Yamamoto, J. Barron, N.S. Diffenbaugh, T. Herbert,
D. Andreasen, 2012, Out of the Tropics: The Pacific, Great Basin Lakes, and Late
Pleistocene Water Cycle in the Western United States, Science, Vol. 337, p. 1629–1633.
Machette, M.N., S.F. Personius, and A.R. Nelson, 1992. Paleoseismology of the Wasatch fault
zone: A summary of recent investigations, interpretations, and conclusions, in Gori, P.L., and
W.W. Hays, eds., Assessment of regional earthquake hazards and risk along the Wasatch
Front, Utah, U.S. Geological Survey Professional Paper 1500-A-J, pp. A1–A71, 1992.
Madsen, D.B., 2000. Late Quaternary Paleoecology in the Bonneville Basin, Utah Geological
Survey Bulletin 130, 2000.
Masson-Delmotte, V., Stenni, B., Pol, K., Braconnot, P., et al., 2010. EPICA Dome C record of
glacial and interglacial intensities. Quaternary Science Reviews, 29: 113-128.
Masson-Delmotte, V., M. Schulz, A. Abe-Ouchi, J. Beer, A. Ganopolski, J.F. Gonzalez Rouco, E.
Jansen, K. Lambeck, J. Luterbacher, T. Naish, T. Osborn, B. Otto-Bliesner, T. Quinn, R..
Ramesh, M. Rojas, X. Shao, and A. Timmermann, 2013, Information from Paleoclimate
Archive. In: Climate Change 2013: The Physical Science Basis. Contribution of Working
group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change,
Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia,
V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA.
Matsubara, Y. and A. D. Howard, 2009. A spatially explicit model of runoff, evaporation, and
lake extent: application to modern and late Pleistocene lakes in the Great Basin region,
western United States. Water Resources Research, 45, W06425,
doi:10.1029/2007WR005953.
Miller, D.M., C.G. Oviatt, and J.P. Mcgeehin, 2013. Stratigraphy and chronology of Provo
shoreline deposits and lake-level implications, Late Pleistocene Lake Bonneville, eastern
Great Basin, U.S.A., Boreas, Vol. 42, pp. 342–361, 2013.
Nash, W.P., 1990, Black Rock Desert, Utah, in C.A Woods and J. Kienle, eds. Volcanoes of
North America, Cambridge University Press, Cambridge p. 271-273.
Nelson, D.T., 2012. Geomorphic and Stratigraphic Development of Lake Bonneville’s
Intermediate Paleoshorelines during the Late Pleistocene, PhD Dissertation, University of
Utah, 237 p.
Deep Time Assessment for the Clive DU PA
22 November 2015 52
Neptune and Company, Inc., 2015a, Neptune field studies, December, 2014, Eolian depositional
history Clive Disposal Site.
Neptune and Company, Inc., 2015b, Final report the Clive DU PA Model, Clive DU PA Model
v1.4, Neptune and Company In., (November, 2015).
NRC (Nuclear Regulatory Commission), 1989. Calculation of Radon Flux Attenuation by
Earthen Uranium Mill Tailings Covers, U.S. Nuclear Regulatory Commission, Office of
Nuclear Regulatory Research, Task WM 503-4, June 1989.
Oviatt, C. G. and W. P. Nash, 1989. Late Pleistocene basaltic ash and volcanic eruptions in the
Bonneville basin, Utah. Geological Society of America Bulletin, 101: 292-303.
Oviatt, C. G., McCoy, W. D., and Nash, W. P., 1994a. Sequence stratigraphy of lacustrine
deposits: a Quaternary example from the Bonneville basin, Utah. Geological Society of
America Bulletin, 106: 133-144.
Oviatt, C.G., G.D. Habiger and J.E. Hay, 1994b. Variation in the Composition of Lake Bonneville
Marl: A Potential Key to Lake-Level Fluctuations and Paleoclimate, Journal of
Paleolimnology, Vol 11, p. 19-30.
Oviatt, C. G., 1997. Lake Bonneville fluctuations and global climate change. Geology, 25(2):
155-158.
Oviatt, C. G., Thompson, R. S., Kauffman, D. S., Bright, J., and R. M. Forester, 1999.
Reinterpretation of the Burmester core, Bonneville Basin, Utah. Quaternary Research, 52:
180-184.
Oviatt, C.G., 2002, Comparing the Shoreline and Offshore Sedimentary Records of Lake
Bonneville. Abstracts with Programs - Geological Society of America 34(6): 292.
Oviatt, C. G., D. M. Miller, J. P. McGeehin, C. Zachary, and S. Mahan, 2005. The Younger Dryas
phase of Great Salt Lake, Utah, USA, Palaeogeography, Palaeoclimatology, Palaeoecology,
219: 263-284.
Oviatt, C.G., and B. P. Nash, 2014, The Pony Express Basaltic Ash: A Stratigraphic Marker in
Lake Bonneville Sediments, Utah, Miscellaneous Publication 14-1, Utah Geological Survey
10 p.
Oviatt, C.G., Chronology of Lake Bonneville, 30,000 to 10,000 yr B.P. Quaternary Science
Reviews 110, 166-171, (2015).
Paillard, D., 2001. Glacial cycles: toward a new paradigm. Reviews of Geophysics, 39(3): 325-
346.
Paillard, D., 2006. What drives the Ice Age cycle? Science, 313: 455-456.
Ruddiman, W.F., 2006. Orbital Changes and Climate, Quaternary Science Reviews, Vol. 25,
pp. 3092–3112.Sack, D., 1999.The composite nature of the Provo Level of Lake Bonneville,
Great Basin, Western North America, Quaternary Research, v. 52, 316-327.
Deep Time Assessment for the Clive DU PA
22 November 2015 53
SC&A, Inc. 2015. Safety Evaluation Report: Volume I. for Utah Department of Environmental
Quality. April 2015.
Schnurrenber, D. J. Russell, and K Kelts, 2003, Classification of lacustrine sediments based on
sedimentary components, Journal of Paleolimnology 29, p. 141-154.
Schofield, I., P. Jewell, M. Chan, D. Currey, and M. Gregory, 2004, Shoreline development,
longshore transport, and surface water dynamics, Pleistocene Lake Bonneville, Utah, Earth
Surf. Process. Landforms, 29, 167-1690.
Tzedakis, P.C., E.W. Wolff, L.C. Skinner, V. Brovkin, d.A. Hodell, J.F. McManus, and D.
Raynaud, 2012a, Can we predict the duration of an interglacial? Climate Past 8, 1473-1485.
Tzedakis, P.C., J.E.T. Channel, D.A. Hodell, H.F. Kleiven, and L.C. Skinner, 2012b,
Determining the natural length of the current interglacial, Nature Geoscience, 5, 138-141.
United States Geological Survey (USGS), 2001. National Water Information System data (Water
Data for the Nation), accessed December, 2010 URL: http://waterdata.usgs.gov
Utah, State of, 2015, Utah Administrative Code Rule R313-25. License Requirements for Land
Disposal of Radioactive Waste - General Provisions. As in effect on September 1, 2015.
(http://www.rules.utah.gov/publicat/code/r313/r313-025.htm, accessed 5 Nov 2015).
Woodhouse, C.A., D. M. Meko, G. M. MacDonald, D.W. Stahle, and E. R. Cook, 2010. A
1,200-year Perspective of 21st Century Drought in Southwestern North America, Proc. Natl.
Acad. Of Sciences, 107, 21283-21288.

Deep Time Assessment for the Clive DU PA
4 November 2015 54
Appendix A
A.1 Clive Pit Wall Interpretation (C. G. Oviatt, unpublished data) and
stratigraphic comparison with quarry wall studies from Neptune
(2015a).

Deep Time Assessment for the Clive DU PA
4 November 2015 55

Deep Time Assessment for the Clive DU PA
4 November 2015 56
Appendix B
B.1 Knolls Core Interpretation (C. G. Oviatt, unpublished data)
Deep Time Assessment for the Clive DU PA
4 November 2015 57

 NAC-0031_R1

Fitting Probability Distributions

Clive DU PA Model v1.4
26 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Fitting Probability Distributions
26 November 2015 ii
1. Title: Fitting Probability Distributions
2. Filename: Probability Distributions v1.4.docx
3. Description:
Name Date
4. Originator Daniel Levitt 5 Nov 2015
5. Reviewers Paul Black 26 Nov 2015
6. Remarks

6/5/2014: No changes since v1.0 except for document formatting.
5 Nov 2015: Updated v1.2 to v1.4. – D.Levitt.
13 Nov 2015: Added discussion of upscaling and model abstraction. – D.Levitt.

Fitting Probability Distributions
26 November 2015 iii

This page is intentionally blank, aside from this statement.
Fitting Probability Distributions
26 November 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

1.0
 Introduction ............................................................................................................................ 1

2.0
 Types of Parameters ............................................................................................................... 1

3.0
 Fitting Distributions to Data ................................................................................................... 3

3.1
 Distributions Representing Epistemic Uncertainty ........................................................... 3

3.2
 Distributions Representing Aleatory Variability .............................................................. 4

4.0
 Fitting Distributions to Reported or Elicited Quantiles .......................................................... 8

4.1
 Quantiles ........................................................................................................................... 9

4.2
 Likelihood Functions ........................................................................................................ 9

4.3
 Example: Gaussian Distribution ..................................................................................... 11

5.0
 Parameter Relationships and Conditioning .......................................................................... 13

6.0
 Summary ............................................................................................................................... 13

7.0
 References ............................................................................................................................ 18

Fitting Probability Distributions
26 November 2015 v
FIGURES
Figure 1. Examples of normal probability density functions .......................................................... 5

Figure 2. Examples of lognormal probability density functions ..................................................... 6

Figure 3. Examples of gamma probability density functions .......................................................... 7

Figure 4. Examples of beta probability density functions ............................................................... 8

Figure 5. Fitted distribution to the quantiles of the example data ................................................ 12

Fitting Probability Distributions
26 November 2015 vi
TABLES
Table 1. Example data, reported only as quantiles ........................................................................ 11

Table 2. Calculation of quantities for the log-likelihood .............................................................. 11

Fitting Probability Distributions
26 November 2015 1
1.0 Introduction
In the Clive DU PA Model, most of the input parameters are treated as probabilistic. The term
parameter is used to refer to any numerical quantity in the PA model. This document provides an
overview of the approach to construction of probability distributions for parameters.
Note that the term parameter is used here because it is in common use in the PA and modeling
community. However, since probability distributions are applied to these parameters, from a
statistics perspective they should be termed variables, or even random variables.
2.0 Types of Parameters
Parameters of the PA model are mathematical constructs that represent a variety of different
concepts. Assignment of a probabilistic distribution must consider the use of the parameter
within the PA model.
The probabilistic behavior associated with the input may also represent a variety of different
concepts. The variation may represent aleatory variability, epistemic uncertainty, or some
combination of those two. The appropriate probabilistic representation for the parameter can
differ greatly depending on the appropriate representation.
• Epistemic uncertainty represents lack of knowledge about the true value of the parameter.
Hypothetically, data could be collected to reduce the uncertainty, which would then result
in a distribution with less variation.
• Aleatory variability represents inherent randomness in the “outcome” of the parameter.
The outcome may represent changes through time or space or the characteristics of
individual members of a population. Given assumptions about the population or modeling
assumptions underlying the parameter, further information gathering does not reduce
aleatory variability. Changing the modeling or population assumptions can lead to a
change in the variability (e.g. changing the spatial extent a soil porosity distribution is
applied to).
Many parameters in the Clive DU PA contain at least some element of both epistemic
uncertainty and aleatory variability, though the probabilistic construction is typically based on
assuming one or the other. Although there are exceptions, for the most part, distributions
developed assuming aleatory uncertainty are contained in the individual dose model (see the
Dose Assessment white paper). Most other input distributions are developed based on epistemic
uncertainty, although as noted, most parameters contain some element of both. It is often
difficult to completely separate epistemic and aleatory uncertainty. Another, and perhaps better,
way of framing the distinction is with respect to the spatial and/or temporal scale of each
parameter. Most parameters in the Clive DU PA model represent long time frames or large areas,
and the distribution of the average of the trait of interest is needed for the model. These cases are
aligned more with the concept of epistemic uncertainty. However, the dose parameters are
specific to individuals, representing points, space, and time frames that are specific to the
Fitting Probability Distributions
26 November 2015 2
available data. These cases are aligned more with the concept of aleatory variability. In effect, in
this model, epistemic uncertainty, upscaling and distribution of the average are related, and
aleatory variability, and distribution of the data are related without the need for upscaling.
These are important distinctions in the development of complex PA models, not just for model
building purposes, but also for model interpretation and comparison with performance
objectives. The PA model is constructed so that raw output doses are provided for each
hypothetical individual included in the model, in each year of the model. Typically, risk
assessment is based on the average risk. In that context, the average dose to the individuals in
each year is the relevant statistic for each receptor group (ranchers, hunters, OHV enthusiasts).
Since 5,000 simulations are performed, there are 5,000 estimates of the average dose in each year
of the model. If the input distributions are specified as epistemic at the appropriate spatio-
temporal scale, then, by analogy with typical approaches to risk assessment, the 95th percentile of
the average dose in each year is the relevant statistic of interest. This has the added advantage of
properly representing uncertainty in the average dose, and hence the uncertainty can be reduced
through further data collection.
Typically, doses generated from a PA are compared to performance objectives by using the
“peak of the means”, however, this does not adequately address the issue of dose in a year
(unless the peak of the mean dose is in the same year for every simulation). There are also 5,000
estimates of the peak of the mean, however, it is not clear how to match a statistic from that
distribution to the performance objectives. This model will allow exploration of this issue, to
evaluate possible approaches to comparison of output doses to performance objectives.
There are other sources of uncertainty that should also be considered in a PA model. These do
not fall easily into either the epistemic or aleatory categories.
• Conceptual uncertainty is typically not associated with a parameter, except in conjunction
with the model as a whole.
• Numerical uncertainty is similar to model uncertainty, except that it typically relates only
to the mathematical aspect of the model, and whether or not a single number can
adequately represent the process.
These latter sources of uncertainty are largely ignored when constructing probabilistic
distributions for parameters. These uncertainties are typically explored, to limited extent, with
sensitivity analyses. However, where expert judgment is utilized in construction of a probability
distribution, the presence of conceptual or numerical uncertainty may cause the expert to
increase the variation associated with a parameter in order to (perhaps) produce a broader range
of model outputs.
More generally, the development of distributions for model input parameters in a PA model
needs to accommodate a wide range of options that address spatio-temporal scales, correlation
structures, available data, secondary data, literature review information, expert opinion and
abstraction from more complex sub-models. Statistical methods that can be considered in each
case are described in the following sections. This is a critical component of model development.
Fitting Probability Distributions
26 November 2015 3
If not performed properly then the PA model runs the risk of the “garbage in – garbage out”
syndrome, uncertainty and sensitivity analysis are compromised, and the results of the model are
potentially meaningless. If performed properly, then everything falls into place regarding model
results, comparison with performance objectives, and useful uncertainty and sensitivity analysis.
3.0 Fitting Distributions to Data
3.1 Distributions Representing Epistemic Uncertainty
When data are available, whose distribution depends on a parameter of interest, then a Bayesian
approach can be used to combine any available prior information with information from the data.
The posterior distribution on the parameter represents the uncertainty about the value of the
parameter. Prior information could be obtained through expert elicitation, but for nearly every
parameter in the Clive DU PA model for which data are available, a non-informative prior is
used.
Most parameters in the Clive DU PA model correspond to physical quantities that represent an
average of some type. Some parameters represent averages over time, as they represent typical
behavior that will be used throughout the 10,000 year performance period, such as annual
precipitation. Other parameters represent averages over space. For example, properties of
vegetation represent an average vegetation effect across a model area, while soil properties
represent an average across a volume of material represented by a model cell. When data are
available that represent small amounts of time relative to the 10,000 years, or small
areas/volumes relative to the model cells, then it is the mean of the data distribution that needs to
be modeled. Under most regularity conditions (such as finite variance and the true parameter not
on the border of the parameter space), the asymptotic distribution of a posterior distribution of a
parameter is normally distributed (Gelman 2004). When a non-informative prior is used, the
posterior distribution is generally well-approximated by the sample distribution of the statistic
used to estimate the parameter. Thus, the posterior distribution for a mean µ is generally well-
approximated by a normal distribution, according to the Central Limit Theorem, if the sample
size n is sufficiently large:
𝜇∣𝑋~𝑁𝑋,𝑠
𝑛 (1)

where 𝑋 is the sample mean, and s is the sample standard deviation. This approximation can be
generalized to most other types of parameters, with the posterior distribution well-approximated
by:
𝑁  𝜃,𝑠.𝑒.(𝜃) (2)

Fitting Probability Distributions
26 November 2015 4
where 𝜃 is an estimate of the parameter of interest θ, and 𝑠.𝑒.(𝜃) is the standard error associated
with the estimate. Stricter regularity conditions may be required for the general approximation to
hold, and larger sample sizes may be needed for the posterior distribution to converge to
normality.
For parameters whose sampling distributions are difficult to calculate, due to the type of
parameter or the small sample size, a bootstrap approach can be utilized to simulate a sampling
distribution (Efron and Tibshirani 1994). The bootstrap method simulates a sampling distribution
for a parameter by simulating new sets of data of the same size and structure as the existing data.
The data simulation may be either parametric, assuming an underlying distribution for the data,
or non-parametric, simulating from the empirical distribution of the data. The simulated
bootstrap samples of the parameter are then fit to a distribution following the guidelines of fitting
presented in Section 3.2, since the bootstrap data represent hypothetical data that can be
processed similarly to the processing of data that represent aleatory uncertainty.
3.2 Distributions Representing Aleatory Variability
For cases where the goal is to find a distribution that reflects the variability in the data, a
goodness-of-fit approach is used. When the complete data set is available, the Akaike
Information Criterion (AIC) is used to choose a distribution (Akaike 1974). The special case of
data that are reported only as quantiles is address in Section 4.0.
AIC provides a measure of fit based on the likelihood function that attempts to discourage over-
fitting by penalizing models with larger numbers of fitted parameter values. AIC could be used
directly to choose a distribution by selecting the distribution that minimizes AIC. However, in
order to allow for scientific judgment to choose between models that are close in fit, Akaike
weights can be used for model selection (Burnham 2002). Akaike weights can be interpreted as
conditional probabilities for each model when all models are treated as equally likely a priori.
The Akaike approach is the following:
• Choose a set of distributions to be considered: M1, M2, …, Mk.
• Fit each distribution via maximum likelihood, and calculate the AIC for each model: A1,
A2, …, Ak.
• Calculate the Akaike weights for each model:
𝑊!=𝑒!!.!⋅(!!!!!"#)
∑!!!!𝑒!!.!⋅(!!!!!"#) (3)

where 𝐴!"# is the smallest AIC amongst the models being considered. Distributions with low
weights are removed from consideration, and scientific considerations are used to choose
between distributions with similarly high weights.
Fitting Probability Distributions
26 November 2015 5
The following descriptions and figures (Figures 1 through 4) provide a list of distributions that
are commonly considered for parameters in the Clive DU PA model: Normal, Lognormal,
Gamma, Beta. Note that the uniform distribution is special case of the Beta distribution. Many
other distributions are considered for special cases, but these four are adequate for most
purposes. Log-uniform distributions are used for Kd and solubility as described in the
Geochemistry white paper, and triangular distributions are used for a few parameters in the dose
model, which represent aleatory variability, when there was insufficient data and expert
elicitation has not yet been performed.
• Normal – N( m, s ), where m is the mean, and s is the standard deviation. This
distribution is unimodal and symmetric and has support on the entire real line. This
distribution occurs naturally in many settings and is generally preferred for parameters
representing averages or sums. Since the normal distribution has infinite support, the
distribution must be left-truncated at 0 (or some other natural boundary) for certain types
of parameters.

Figure 1. Examples of normal probability density functions

Fitting Probability Distributions
26 November 2015 6
• Lognormal – LN( m, s, θ ), where m is the geometric mean, and s is the geometric
standard deviation, and θ is a location parameter specifying the minimum. This
distribution is unimodal and right-skewed and has support on all real values greater than
θ. When the geometric standard deviation is near 1, the lognormal distribution closely
approximates the normal distribution. Physical quantities can often be modeled well with
a lognormal distribution, and typically θ=0, forcing those quantities to be positive.

Figure 2. Examples of lognormal probability density functions

Fitting Probability Distributions
26 November 2015 7
• Gamma – Gamma( m, s, θ ), m is the mean, s is the standard deviation, and θ is a location
parameter specifying the minimum. This distribution is unimodal and right-skewed and
has support on all real values greater than θ. Fitted gamma distribution and lognormal
distributions often appear quite similar, and the lognormal is typically preferred for
physical quantities. However, the gamma distribution can fit certain types of tail behavior
that the lognormal distribution cannot.

Figure 3. Examples of gamma probability density functions

Fitting Probability Distributions
26 November 2015 8
• Beta - Beta( m, s, l, u ), where m is the mean, s is the standard deviation, l is the lower
bound, and u is the upper bound. The beta distribution can take on a variety of shapes. It
is typically unimodal, but can be bimodal, with modes at the lower and upper bounds.
The beta distribution is sufficiently flexible that it might provide a reasonable fit where
other distributions cannot, and it is the only standard distribution that has finite support.
For many parameters, finite support does not make good sense, so the beta distribution is
typically only chosen when it is the only distribution that provides a reasonable fit, or
when there is a natural lower and upper bound.

Figure 4. Examples of beta probability density functions

4.0 Fitting Distributions to Reported or Elicited Quantiles
In many cases, data are available only in the form of reported quantiles of the distribution. A
formal method for fitting a distribution and choosing amongst possible distributions is needed.
While the focus here is on empirical quantiles, the same approach may also apply to quantiles
achieved via expert elicitation, though some assumptions about the expert's knowledge base must
be considered. This section begins with a definition of quantiles, and follows up with a
likelihood estimation method for estimating distributions based on quantile input, and ends with
an example.
Fitting Probability Distributions
26 November 2015 9
4.1 Quantiles
Let X be a random variable whose distribution is of interest. Suppose that a random sample of n
observations from this distribution has been collected, 𝑋=𝑋!!!!!, but that the reported
summaries of this sample are restricted to a set of k empirical quantiles, 𝑞!!!!
!, corresponding
to a set of proportions 𝑝!!!!
! (considered to be given in increasing order for convenience; i.e.,
𝑝!<𝑝!!!).
The empirical cumulative distribution function (CDF) is defined as:
𝐹!𝑥=# of sample values less than  𝑥
𝑛=  𝐼𝑋!<𝑥!!!!
𝑛 (4)

where I is the indicator function. An empirical quantile corresponds to the inverse of the
empirical distribution function:
.  𝑞!=𝐹!!!𝑝! (5)

Since 𝐹! is a step function, the inverse is not uniquely defined. However, there are many
common methods for defining a unique quantile (Hyndman and Fan, 1996). In practice, the exact
method of defining the quantile is rarely cited. Thus, there is some potential error associated with
a reported quantile. The relative size of the error is dependent on the underlying distribution and
the quantile of interest. When sample sizes are large and/or the underlying distributions are
smooth (as is the case with named families of distributions that one is likely to fit), the error
associated with non-uniqueness should be small, though sensitivity analysis to this error should
be performed in assessing fits based on reported quantiles. For the purposes of this document, 𝑞!
will be considered to be uniquely defined.
4.2 Likelihood Functions
If the original data set were available, then a reasonable choice for fitting the parameters of a
distribution is maximum likelihood. Suppose that the random variable of interest, X, is assumed
to come from a parametric family of distributions (e.g. Gaussian, gamma, etc.), that are uniquely
defined by a set of parameters θ. The likelihood function for a sample X is defined as:
𝐿𝜃∣𝑋=𝑓!  𝑥!∣𝜃!!!!, (6)

where fX is the probability density (or mass) function corresponding to the parametric family of
distributions. The maximum likelihood estimator (MLE) of the parameters is then defined by:
Fitting Probability Distributions
26 November 2015 10
𝜃=arg max!𝐿𝜃∣𝑋, (7)

or equivalently when maximizing the log-likelihood:
𝜃=arg max!ln𝐿𝜃∣𝑋=arg max!𝑙𝜃∣𝑋. (8)

When the sample has been summarized by quantiles, the likelihood function for the data takes a
different form. The reported data are effectively 𝑌=𝑌!!!!
!!!, where Yj is the number of
observations between qj-1 and qj.
𝑌!=𝐼𝑞!!!<𝑋!≤𝑞!!!!!, (9)

where 𝑞!=−∞ and 𝑞!!!=∞ for notational convenience.
The reported data thus follow a multinomial distribution:
𝑌~Multinomial!!!𝑛,𝜋𝜃, (10)
where
𝜋!𝜃=𝐹!𝑞!∣𝜃−𝐹!𝑞!!!∣𝜃, (11)

and FX represents the CDF for X.
The likelihood function associated with the reported data is then:
𝐿𝜃∣𝑌=𝑛!!!!!!
!!!
!!!!!!∝𝜋!𝜃!!!!!!!!, (12)

Where proportionality is with respect to the parameters of interest, θ. Maximizing the log-
likelihood is thus equivalent to maximizing:
𝑙*𝜃∣𝑌=𝑌!ln 𝜋!𝜃!!!!!!=𝑛𝜋!ln 𝜋!𝜃!!!!!!∝𝜋!ln 𝜋!𝜃!!!!!!, (13)
where
𝜋!=!!
!. (14)

Fitting Probability Distributions
26 November 2015 11
Note that maximizing Equation (13) does not depend on knowing the sample size n, which may
not be available for some data reports, and is only an abstract concept if the quantiles represent
elicited values.
For most parametric families, πj(θ) does not have a functional form that lends itself to analytical
maximization of Equation (13). However, the CDF for most parametric families is sufficiently
smooth that maximization routines work robustly.
Note also that the use of maximum likelihood estimation is similar to intent to using Bayesian
statistical methods with some types of non-informative prior distributions. This approach,
therefore, is similar in intent for quantile data as the methods described in Section 3.1. Use of
least squares minimization instead is not recommended, because the underlying assumptions will
probably not be met (e.g., normality, independence, identically distributed data).
4.3 Example: Gaussian Distribution
Suppose that data are reported as in Table 1:
Table 1. Example data, reported only as quantiles
p1 = 0.05 = 5% p2 = 0.25 =
25%
p3 = 0.5 = 50% p2 = 0.75 =
75%
p5 = 0.95 = 95%
q1 = 31 q2 = 58 q3 = 76 q4= 89 q5 = 120

Five quantiles are reported, and thus the data are separated into 6 bins. In fitting a Gaussian
distribution to these quantiles, π can be expressed in terms of the standard Gaussian CDF, Φ, as
in Table 2.
Table 2. Calculation of quantities for the log-likelihood
𝜋!=0.05 −0 =0.05 𝜋!=𝛷31 −𝜇
𝜎
𝜋!=0.25 −0.05 =0.2 𝜋!=𝛷58 −𝜇
𝜎−𝛷31 −𝜇
𝜎
𝜋!=0.50 −0.25 =0.25 𝜋!=𝛷76 −𝜇
𝜎−𝛷58 −𝜇
𝜎
𝜋!=0.75 −0.50 =0.25 𝜋!=𝛷89 −𝜇
𝜎−𝛷76 −𝜇
𝜎
𝜋!=0.95 −0.75 =0.2 𝜋!=𝛷120 −𝜇
𝜎−𝛷89 −𝜇
𝜎
𝜋!=1 −0.95 =0.05 𝜋!=1 −𝛷120 −𝜇
𝜎

Fitting Probability Distributions
26 November 2015 12
Maximum likelihood estimators can thus be calculated: 𝜇=74.6 and 𝜎=25.8, resulting in a
value of -1.65 for the (right portion of) Equation (13). The CDF and probability density function
(pdf) for the fitted distribution are plotted in Figure 5.
Figure 5. Fitted distribution to the quantiles of the example data
Fitting Probability Distributions
26 November 2015 13
5.0 Parameter Relationships and Conditioning
Many parameters in the Clive DU PA model are related to one another. One parameter may be
physically constrained by the value of another parameter, or they may simply tend to vary
together. Information about the joint behavior is often unavailable, but where it is, the preferred
approach is to construct joint distributions for the parameters.
When joint data are available, a simple approach is to simply calculate the sample correlation of
the parameters in the data and apply the same correlation to the parameters in the model to
induce a joint distribution. A simple correlation structure may not fully capture the relationship
between two parameters but often provides a reasonable first approximation. Where a correlation
structure is used in the Clive DU PA model, the correlation algorithms implemented in GoldSim
for Gaussian copula are used (Iman and Conover 1982, Embrechts et al. 2001).
Where data and expertise are available, it is generally preferable to construct joint distributions
for the parameters by constructing a marginal distribution for one parameter and conditional
distributions for the remaining parameters. By fitting a distinct conditional distribution of the
second parameter for each possible value of the first parameter, a more realistic relationship
might be constructed than can be achieved through simple correlation.
For example, for the population of American males the distribution of body weight changes as a
function of age, even after reaching adulthood. Beyond age 20, the median body weight tends to
increase as a function of age, until middle-age, after which median body weight decreases. The
variation in body weight across the population also changes with the mean. Thus, a reasonable
approach might be to model body weight as:
𝐵𝑊!"#$%~LN 𝑒!!!⋅!"#!!⋅!"#!,𝑒!. (15)
where a, b, c, and σ are estimated from data. This general approach was utilized for the Clive DU
PA model (including for this body weight example), by using the fitting techniques outlined in
Section 4.0 to quantile data available for age and body weight.
6.0 Scaling and Model Abstraction
Development of appropriate probability distributions is critical for ensuring that model results
are useful. The input distributions are based on expectation and uncertainty. Bias is not
introduced in general. If the input distributions are not based on expectation and associated
uncertainty, then the model is compromised as is the sensitivity analysis and more general model
evaluation. This is in part because the model is fully coupled, so that biases would propagate
through the entire model. Many aspects of distribution fitting are described above, but there are
two further aspects that are critical for understanding how these types of models are specified.
The first is model scaling, and the second is model abstraction.
Fitting Probability Distributions
26 November 2015 14
6.1 Upscaling
Upscaling of Clive DU PA model is required to ensure that the model is specified at the
appropriate spatial and temporal scale. The Clive disposal systems covers a large area, and the
model is run for 10,000 years (or longer) into the future. However, the models are also set up so
that inputs for a single realization of the simulation are sampled randomly at the beginning of
time, and those values that represent the input distributions are used throughout time. They are
also applied throughout the spatial domain of the model. This has broad implications for the
correct structure of the model from the perspective of spatio-temporal scaling (upscaling).
For example, suppose data for a specific input such as near surface moisture content has a
minimum of 0 and a maximum value of 30, and suppose there are 100 data points. Suppose a
distribution fit to these data is Gamma(1,10), so the mean is 10%, and the standard deviation is
10%. For a single realization it would be possible to draw a random number of very near 0%, or
of 30%. However, the nature of the model is such that this value would be applied to the entire
spatial domain of the disposal facility for the entire duration of the model. The variability of data
at points in time and space is not the appropriate representation of variance in a model that is
constructed and run this way. If, instead, the model could be constructed to choose, for a single
realization, a new random number every very small time step, then the net effect of 10,000 years
would be that the system is represented by the average of the many values drawn at each very
small time step (very small because the data in this example represent points in time and space,
or very small volumes over a very short amount of time). This assumes that the system is fairly
stable over the modeling time frame, and is not evolving in unexpected ways that would cause a
major shift in moisture content for this example. If such as a model could be constructed, then
the net effect on the system is the average moisture content over 10,000 years. (Also note that if
sufficient attention is not paid to the interaction between the number of time steps and the
variability in the data, such an approach would lead to a very tight (very little variance) average
(for example, 1-yr time steps would result in 10,000 random draws over time, and the variance of
the average would be an inverse function of the square root of that number of time steps). Also,
autocorrelation across time should be addressed in a model system constructed that way, which
would not be straightforward without the supporting data. A further disadvantage of constructing
a system with such small time steps is the computational complexity of doing so.
The correct approach is to upscale the data to reflect the spatial and temporal scale of the model.
In this particular case of moisture content, averaging provides a reasonable approximation to
addressing the underlying issues. That is, the input distribution represents the average moisture
content and the associated uncertainty in the average. In this case, since there were 100 data
points, this might result in a distribution that is N(10%, 1%). That is, the mean is 10% moisture
content, just as it was for the Gamma distribution of the data, and the standard deviation is 1%,
which is 1/10th of the variance of the data (because there are 100 data points).
Upscaling is critical for properly addressing uncertainty in the fate and transport model.
However, it is not always as straightforward as, using as a reasonable approximation, simple
averaging. For example, plant root depth matters because of the deeper roots that might uptake
radionuclides. Simple averaging across plant root depth could completely miss the effect that
Fitting Probability Distributions
26 November 2015 15
matters. Instead averaging across the plant root depth function is needed, so that, in effect, an
average is formed at each point (or interval) in the depth profile. Similarly, averaging across rain
events might not be appropriate if a subsequent action or response of interest is non-linear. In
general, averaging (expectation) is a linear construct. If the response is non-linear, then
consideration needs to be given to how the model must be constructed so that important non-
linear effects are captured. In the context of upscaling, this is usually an issue of breaking down
each component of the model until averaging can be reasonably applied (until there are
approximately linear responses).
Without temporal and spatial scaling the fate and transport model would carry variance
components that are far too large, do not represent the system effect, and lead to unnecessarily
wide ranges in the output. They would not adequately describe the system. Upscaling needs to be
done with care, and needs to address to the extent reasonable non-linear effects, autocorrelation,
and correlation between variables.
Upscaling applies to many of the contaminant transport processes in the Clive DU PA model.
However, they are not also applied to the exposure assessment. This is because the variability
between the hypothetical receptors is captured. The population of receptors is modeled explicitly,
and each receptor has different characteristics, which capture the variability between receptors.
The exposure and dose assessment evaluates dose to each receptor rather than does to an
“average individual”, in which case upscaling is not appropriate.
Because upscaling is a form of averaging, this also means that the model output represents
uncertainty in the system (average) response, rather than variability from data collected at (near)
points in time and space. This is also a necessary construct for understanding uncertainty, and
managing uncertainty through a decision process. That is, model evaluation might suggest data
collection is needed to reduce uncertainty in the model. This makes no sense without upscaling
so that the input distributions are averages (in the appropriate spatial and temporal scale, etc.). It
is not possible to reduce the variance term in the moisture content data example by collecting
more moisture content data. This also provides an indication that in the context of reducing
uncertainty, models that are built without upscaling are inappropriate.
A further note is that human health risk (exposure/dose) assessment is based on average
concentrations and uncertainty about those averages. Again, this implies the need to construct
models that are scaled correctly to the endpoint of interest.
Upscaling is performed for many of the input distributions in the Clive DU PA model, and some
attention is paid in each case to non-linearity, auto-correlation and correlation. This leads to a
model that appropriately addresses the endpoints of interest. In each case the details are provided
in the appropriate White Paper.

Fitting Probability Distributions
26 November 2015 16
6.2 Model Abstraction
There are several ways in which input distributions can be developed depending on the nature of
the data or information available. If data are available, then simple distribution fitting can be
performed as described in previous sections. This might also be the case when literature review
information is the primary source of information, but in this case, some meta analysis is often
performed as well. Elicitation of expert opinion is another option. A final option is model
abstraction, which is used when a more detailed process model is performed externally to the
Clive DU PA model developed in GoldSim, the results of which need to be incorporated into the
Clive DU PA model.
Model abstraction has been defined as “the intelligent capture of the essence of the behavior of a
model, without all the details (and therefore, run-time complexities) of how that behavior is
implemented in code” (Frantz and Ellor, 1996). Caughlin and Sisti (1997) describe model
abstraction methods as “techniques that derive simpler conceptual models while maintaining the
validity of the simulation results. These methods include variable resolution modeling, combined
modeling, multimodeling, and metamodeling. In addition, some taxonomies include
approximation, aggregation, linear function interpolation, and look up tables as model
abstraction methods.”
More generally the intent is to simplify the complex process model without losing much
information, so that the simpler form can be used in the probabilistic simulation that supports the
system model for the Clive DU PA. Perhaps all approaches can be thought of as creating a
“response surface” that adequately captures the essence of the process model. This is usually
supported by some statistical experimental design across the inputs of interest. Model abstraction
has been applied to the Clive DU PA model to components such as infiltration, erosion,
tortuosity, and air dispersion.
An example of model abstraction in the Clive DU PA Model v1.4 is the calculation of net
infiltration using a regression model that was developed from results of 50 simulations using
HYDRUS-1D. The HYDRUS methods and results are described in the Unsaturated Zone white
paper (see Appendix 5 of the Clive DU PA Model v1.4 Final Report). An experimental design
was set up across the expected sensitive inputs of interest, and a regression equation was
developed to predict the HYDRUS net infiltration results using the sensitive input parameters.
Figure 6 shows a comparison of the net infiltration results calculated using HYDRUS and using
the regression equation in the Clive DU PA Model v1.4 GoldSim model. Clearly, the comparison
shows an excellent fit to the HYDRUS results, demonstrating that the use of a regression
equation to approximate the HYDRUS simulations resulted in a successful model abstraction in
this case.
Fitting Probability Distributions
26 November 2015 17

Figure 6. Comparison of 1,000 realizations of net infiltration using the linear model in
GoldSim with the results of the 50 HYDRUS simulations of infiltration.

7.0 Summary
For the Clive DU PA considerable effort has been expended to provide statistical rigor and
defense for the PA model. There are few, if any, previous examples of PA for low-level waste
that have achieved this level of statistical support, except others developed by Neptune. This
should also be regarded as a critical quality assurance aspect of this type of modeling.
Regulations and guidance that could be used are sadly lacking in sufficient definition of how PA
models should be constructed and the role that statistics should play to ensure proper
construction, despite the fact that these are probabilistic models. When insufficient attention is
paid to proper development of input distributions the ensuing models are potentially worthless.
The Clive DU PA model provides an opportunity for others who perform PA for low level
radioactive waste to follow this path, and improve the statistical defensibility for PA more
generally.

Fitting Probability Distributions
26 November 2015 18
8.0 References
Akaike, H. (1974). “A New Look at the Statistical Model Identification,” IEEE Transactions on
Automatic Control 19 (6): 716-723.
Burnham, K.P., Anderson, D.R. (2002). “Understanding AIC and BIC in Model Selection.”
Sociological Methods and Research. Sociological Methods and Research, 33 (2): 261-304.
Caughlin, D. and A.F. Sisti (1997). Summary of model abstraction techniques. Proc. SPIE 3083,
Enabling Technology for Simulation Science, 2 (June 20, 1997).
Efron, B. and Tibshirani, R.J. (1994). Introduction to the Bootstrap. CRC Press LLC, Boca
Raton, FL.
Embrechts, P., Lindskog, F., and McNeil, A. (2001). Modelling Dependence with Copulas and
Applications to Risk Management, Department of Mathematics, Swiss Federal Institute of
Technology, Zurich.
Frantz, F.K., and A.J. Ellor (1996). Model Abstraction Techniques. Report for Rome Laboratory,
Air Force Materiel Command, Griffiss Air Force Base, New York. RL-TR-96-87.
Gelman, A., Carlin, J.B., Stern, H.S., and Rubin, D.B. (2004). Bayesian Data Analysis, 2nd
Edition. Chapman and Hall/CRC, Boca Raton, FL.
Hyndman, R.J., and Fan, Y. (1996). “Sample Quantiles in Statistical Packages,” American
Statistician, 50: 361-365.
Iman, R.L., and Conover, W.J. (1982). “A Distribution-Free Approach to Inducing Rank
Correlation Among Input Variables,” Communications in Statistics: Simulation and
Computation,11 (3): 311-334.

 NAC-0029_R2

Machine Learning for
Sensitivity Analysis of
Probabilistic Environmental
Models
8 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 ii
1. Title: Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
2. Filename: Sensitivity Analysis (Appendix 15) v1.4.docx
3. Description: Sensitivity Analysis methods with GW example from v1.4 model

Name Date
4. Originator Paul Duffy 7 November 2015
5. Reviewer Paul Black 8 November 2015
6. Remarks

5 Nov 2015: Updated from v1.2 to v1.4. – D.Levitt

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 iii

This page is intentionally blank, aside from this statement.

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 iv
CONTENTS
FIGURES ........................................................................................................................................ v

TABLES ......................................................................................................................................... vi

Executive Summary ......................................................................................................................... 1

1.0
 Introduction ............................................................................................................................ 2

2.0
 Sensitivity Analysis Approaches ............................................................................................ 2

2.1
 Analytical Approach: Sobol Design of Experiment and Fourier Amplitude
Sensitivity Test (FAST) .................................................................................................... 4

2.2
 Meta-models: Regression Based Methods ........................................................................ 4

2.3
 Meta-models: Machine Learning Approaches .................................................................. 5

2.3.1
 Multivariate Adaptive Regression Splines (MARS) ................................................... 6

2.3.2
 Gradient Boosting Machines (GBM) .......................................................................... 6

2.4
 Example: Comparison of SA methods .............................................................................. 8

2.4.1
 “Sobol g-function” ...................................................................................................... 8

2.4.2
 Visualization ............................................................................................................... 9

3.0
 References ............................................................................................................................ 17

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 v
FIGURES
Figure 1. Sensitivity and Partial Dependence Plots for the GBM fit to the Sobol Function. ........ 11

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 vi
TABLES
Table 1. Sensitivity Indices by Sensitivity Analysis Method for Sobol g-function application
with p = 8. ...................................................................................................................... 9

Table 2. Peak Groundwater Well Concentrations within 500 years - Tc99 .................................. 12

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 1
Executive Summary
The purpose of this document is to explain the development and application of the method used
for sensitivity analysis (SA) of Performance Assessment (PA) models constructed in GoldSim.
The overarching goal of the SA is to determine which explanatory variables (e.g. Kd in Sand for
Tc) have the largest impact on specific endpoints of interest (e.g. Peak Ground Surface Flux of
Radon-222). The SA procedure implemented for the Clive DU PA assesses the importance of
every explanatory variable (input parameters) used in the GoldSim PA model. In practice, this
means that for each endpoint of interest, every explanatory variable in the model has a
quantitative measure of importance associated with it. For a given explanatory variable, the
quantitative measure of importance depends on the endpoint of interest. All input parameters are
included and are essentially varied simultaneously. This is very different from the one-at-a-time
SA approaches that are used on deterministic models. This global SA approach allows all levels
of interactions to be evaluated, so that the variation in the output can be measured at every level
of possible interaction. The effects are collected together during SA processing, and then
separated out to attribute the overall contribution of each input parameter to the model output.
As described below, some global SA procedures have the ability to characterize non-linear and
non-monotonic relationships between explanatory variables (input parameters) and endpoints of
interest (output parameters). This is critically important because of the need to characterize
complex interactions among multiple explanatory variables in the PA models. These interactions
are often non-linear and non-monotonic.
The approach to simulating the PA model affects how the SA should be set up. Each PA
simulation is set up to draw random numbers from the input distributions at the beginning of
time, and then those random numbers are used throughout time in that simulation. This is one of
the reasons why the input distributions are set up to describe the mean of the factor of interest. If,
instead, random numbers were drawn at every time step, then the net effect over a long period of
time (e.g., 10,000 years) is to create an overall averaging effect. The SA is essentially a
regression model that uses the simulated inputs as observations of the input parameters, and the
simulated outputs as observations of the output parameters. The form of regression used
accommodates non-linear and non-monotonic effects, which are inherent in PA models.
The results of the SA provide a clear indication of the explanatory variables that most strongly
influence a given endpoint, since the result of the SA, applied to a given PA endpoint, is a
quantitative metric of importance for each explanatory variable. This information can be used in
a number of ways. During model development it is a very useful tool for model evaluation, often
leading to a better understanding of model constructs and modifications to the model structure as
necessary. This leads to iterative model development. Also, if there is an unacceptable level of
uncertainty associated with an endpoint of interest (for decision making purposes), the sensitive
parameters can be targeted for effective uncertainty reduction; that is, further data or information
should be collected to reduce the uncertainty on these sensitive input parameters. Another
possibility is to use the results of the SA to simplify the PA model, although the simplification
would depend on each specific endpoint.
The remainder of this document provides some background information on SA methods, leading
to the choice of SA methods that are used for the Clive DU PA model. This starts with one-at-a-
time SA methods for deterministic models, and moves through linear modeling approaches
before finishing with global SA approaches.
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 2
1.0 Introduction
Decision making for the management of complex systems in the presence of uncertainty requires
an explicit characterization of the current state of knowledge. In this context, a model is a
valuable tool for understanding the interactions and influence of explanatory variables on the
response of interest (e.g. media concentrations or future potential doses). The quantitative
assessment of the importance of inputs is critical, and this is especially true when uncertainty in
the response is deemed to be unacceptable for the decision at hand. Sensitivity analysis (SA) can
be used to help identify those inputs with the greatest impact on uncertainty in the model
response (Saltelli et al. 1999, Marrell et al. 2009, Nossent et al. 2011, Morris et al. 2014).
Specifically, SA helps quantify the benefit of subsequent data collection through the
identification of the explanatory variables for which uncertainty reduction through further
information collection will yield the most effective decrease in uncertainty for the response of
interest. Both analytical and simulation approaches can be implemented to develop SA models
that characterize the state of knowledge for a system, and the approach selected depends on the
application in question. Analytical representations (e.g. systems of differential equations) have
the advantage of allowing more straightforward analysis of model effects; however, some
systems possess sufficient complexity that simulation approaches need to be utilized.
Performance assessment (PA) is an integral part of post-closure for radioactive waste disposal
facilities and it requires characterization of the fate and transport of waste through space and
time. The main goal of a PA is to provide reasonable assurance that performance objectives for
radioactive waste disposal will be met; hence it is necessary to depict the relevant dynamics of
the facility of interest. This requires the characterization and simulation of multiple system
components including: hydrologic, edaphic, radiologic, biotic, and structural features. The
overarching model used for a PA is an aggregation of information from multiple sources
including: field studies, literature review, and output from other models. PAs are a modeling
application where analytical approaches alone are insufficient and the use of a high-dimensional
simulation model is essential. SA of high dimensional probabilistic models can be
computationally challenging; however, these challenges can be met through the application of
machine learning methods applied to probabilistic simulation results. This approach is sometimes
referred to as meta-modeling (Marrell et al. 2010, Coutts and Yokomizo 2014).
2.0 Sensitivity Analysis Approaches
This section provides a brief review of the common approaches to SA and reviews their pros and
cons in the context of application to PA models. Generally speaking, SA deals with the
estimation of influence measures for input variables that are components of a given model. In the
application to PA models, it is of interest to determine which input variables are driving the
uncertainty associated with an endpoint of interest (dose, flux, concentration, etc.). This can be
accomplished with either a qualitative (Melbourne-Thomas et al. 2012) or quantitative (Liu et al.
2006, Storlie et al. 2009) approach applied across a spectrum ranging from local (McKay et al.
1979) to global (Sobol 2001, Friedman 2002) analyses.
Qualitative SA provides a relative ranking of the importance (sensitivity) of input factors without
incurring the computational cost of quantitatively estimating the percentage of the response (e.g.
media concentrations or future potential doses) uncertainty accounted for by each input factor. It
was considered more useful prior to the availability of the computational capability needed for
quantitative SA approaches. Currently, for the purpose of PA modeling, qualitative SA is of little
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 3
utility beyond perhaps the initial preliminary model development stages. It is not considered
further in this document.
A local SA varies one input factor while holding all other input factors constant and assesses the
impact on the corresponding model response. This is often accomplished in an analytical context
by examining partial derivatives evaluated at the solution of interest locally. This analysis is
local in the sense that only a minimal portion of the full volume of the input factor space is
explored (i.e., the point at which all but one of the input factors are held constant). Although
local sensitivity analysis is useful in some applications, the region of possible realizations for the
model of interest is left largely unexplored. Global sensitivity analysis attempts to explore the
possible realizations of the model more completely. The space of possible realizations for the
model can be explored through the use of search curves or evaluation of multi-dimensional
integrals using Monte Carlo methods.
An example of quantitative local SA approach is differential analysis based on the partial
derivatives of the model with respect to each input factor. Given a model of the form y = f (X),
the local relative sensitivity measure, Si, of each input factor, xi, on model response y can be
calculated as:
Si =Ex
∂f (X)
∂xi
"
#$
%
&'
2
varx[xi ]
(
)*
+*
,
-*
.*
1/2
(1)
Typically, evaluation of this derivative at a specific solution is performed; hence the analysis is
relevant in a small local neighborhood around the solution of interest.
Quantitative global SA attempts to explore the full hyper-volume defined by the collective
ranges of possible values for the input factors. Sensitivity indices (SIs) for a single value are
obtained by averaging over the variation of all other input factors to provide an estimate of
sensitivity:
Si =varxi[E(y |xi )]
var(y) (2)
The degree of success for this type of analysis is measured using the quantity,∑
=
p
i
ixS
1
, where p is
the number of model parameters. If this sum is approximately 1, then the analysis is considered
successful in terms of depicting (and hence allowing the ability to decompose) the observed
variability in the response. Given the complexity of PA models, and the rigor that needs to be
applied in order to provide reasonable assurance that PA goals being met, the global, quantitative
SA is the most effective approach.
Quantitative global SA approaches can be partitioned into two groups: analytical; and, meta-
model. Several approaches have been proposed to implement analytical SA for nonlinear,
nonmonotonic models. Two of the analytical approaches that are considered here are the Fourier
Amplitude Sensitivity Test (FAST) (Saltelli et al. 1999) and Sobol’s design of experiment
(SDOE) approach (Sobol 1993). Application of the meta-model approach (Borgonovo et al.
2012, Coutts and Yokomizo 2014) consists of the development of a statistical model that
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 4
quantifies the relationship between the response of interest and all of the explanatory variables
that enter into the PA model.
The following is a brief review of these approaches. The goal of this review is to articulate the
logic for the implementation of the preferred approach, which is the application of the meta-
model using gradient boosting machines (GBM).
2.1 Analytical Approach: Sobol Design of Experiment and Fourier
Amplitude Sensitivity Test (FAST)
Several analytical approaches have been proposed to handle nonlinear, nonmonotonic models.
Two of these approaches include the Fourier Amplitude Sensitivity Test (FAST) (Saltelli et al.
1999) and Sobol’s design of experiment (SDOE) approach (Sobol 1993). These methods provide
an estimate of the proportion of the variation in the model response from an explanatory variable,
by using an analysis of variance (ANOVA) type of decomposition of the variability in the model
response. These two analytical methods use a different computational strategy for decomposing
the partial variances corresponding to increased dimensionality (main-effects, two-way
interactions, three-way interactions, etc.).
FAST and SDOE have been shown to be effectively equivalent with respect to SA application
(Saltelli et al. 1999). In the context of SA, the ANOVA decomposition can be described in terms
of total sensitivity indices for each input factor, STi. The STi for explanatory variable i is
calculated as the sum across all main and interaction sensitivities that involve the ith input
explanatory variable:
!+++=∑∑≠≠
n
ikj ijk
n
ij ijiT SSSSi, (3)
where Si is the first-order or (main effect) sensitivity index and Sij is the second-order (two- way
interaction effect) sensitivity index and so on. For a single STi,, the total number of interactive
terms for sensitivity indices is 2n – 1, where n is the number of explanatory variables. Because
SDOE requires multi-dimensional integration to estimate the sensitivity indices, this method can
be prohibitive computationally for moderately complex models. These approaches become more
computationally intensive as the dimensionality of the model (i.e., the number of model
parameters) increases and can be prohibitive for models that include hundreds or thousands of
stochastic explanatory variables.
2.2 Meta-models: Regression Based Methods
Regression based approaches are an option for the quantitative global SA of PA models. These
approaches include squared standardized regression coefficients (SSRC) and squared
standardized rank regression coefficients (SSRRC) (Storlie et al. 2009, Cea et al. 2011). Some of
the benefits of these methods are ease of implementation and relative familiarity of the basic
output of regression models for most members of a target audience. However, one of the main
drawbacks of these approaches is that the methods assume a monotonic and linear relationship
between the input factors and the model response. The degree to which the relationships between
explanatory and response variables follow these assumptions impacts the validity of the results
of these simple linear regression-based analyses. That is, this method does not provide reliable
SA for systems with a high degree of non-linearity or moderate lack of monotonicity.
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 5
A linear regression model has the following form
∑=++=p
i iixy10 εββ (4)
where y represents the output from a PA model (i.e. the response of interest), and the  𝑥!’s
corespond to each of the explanatory variables built into the PA model. The variance of the
linear regression model in Equation 4 can be estimated as
𝑣(𝑦)=𝛽!!!
!!!𝑣𝑎𝑟(𝑥!) (5)
where 𝛽!!  is the square of the standard parameter estimate for the ith variable in the regression
model. Equation [5] requires an assumption that the input factors are independent, which is often
not the case. For example, many of the explanatory variables in a PA model are correlated as a
direct consequence of the nature of the system. If the model response and explanatory variables
are standardized to a mean of 0 and a variance of 1 then the square of the regression coefficients
(i.e.  𝛽!!) provides an estimate of Si (in the form of Equation 3).
Regressing the ranks of the model response on explanatory variables can help to mitigate (but
not obviate) the impact of nonlinearities in the model on the lack of validity of output for the
regression model. The coefficient of determination associated with the regression model in
Equation [4], R2, measures the percent of variability in the ranks of the model response that is
explained by the linear combination of the explanatory variables multiplied by their respective
parameters. The closer R2 is to one, the less unexplained variability there is in the rank response
and the better the regression model performs as a meta-model estimating the dynamics of the
more complex PA model.
Alternatively, the quantity 1- R2 represents the percent of rank response variation not accounted
for by the SA method. As this percent increases, confidence in the analysis is reduced although
the resulting relative ranking may still be of value. For models with low enough values of R2, the
validity of the relative rankings also comes into question. In this context, it is worth recalling that
SSRC and SSRRC assume a monotonic linear relationship between the explanatory variables and
the response of interest for the PA model output. A low R2 might be reflective of a model
structure that does not meet this assumption.
2.3 Meta-models: Machine Learning Approaches
Because of the computational cost, SA of high-dimensional probabilistic PA models requires
efficient algorithms for practical application. Machine learning is a general approach that
provides tools for the construction of meta-models that can be used for subsequent SA. These
models allow for the partitioning of the variance in the model response in a manner that allocates
a proportion to the explanatory variables. Two common machine learning approaches that could
be used for SA are Multivariate Adaptive Regression Splines (MARS) (Friedman 1991) and the
gradient boosting machine (GBM) (Friedman 2001, Friedman 2002). Several of the most
important advantages of machine learning approaches are: the ability to fit non-monotonic and
non-linear effects; the ability to fit parameter interaction effects; and, the ability to visualize
these effects and their interaction across the range of the response and input parameters. MARS,
boosting and other machine learning approaches typically produce similar results for noisy data.
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 6
In the case of realizations from a probabilistic process model, each realization is a deterministic
evaluation of the model and all the stochastic predictor variables are available. As such there is
no unexplainable variation in the process model response, which is a stark contrast to the case
with observed data. Hence, machine learning algorithms should theoretically be able to construct
models that have R2 values very close to one. Given that, it might be tempting to conclude that
machine learning SA methods are all effectively equivalent; however, with respect to the ease of
implementation and interpretability of results, GBM has an advantage. The remainder of this
section articulates this with a high level review of both MARS and GBM.
2.3.1 Multivariate Adaptive Regression Splines (MARS)
MARS is a recursive partitioning approach that directly addresses the ANOVA decomposition
“curse of dimensionality”, making estimation of sensitivity indices computationally achievable
for large n (Friedman, 1991). MARS accomplishes this by optimally partitioning or, splitting the
model response and explanatory variables into subsets, from which splines (i.e. piecewise
smooth polynomial functions) are fit. The recursive nature of the algorithm results in
increasingly local splits of the model response in which all significant interaction effects in sub-
regions are found. MARS is able to find and fit significant nonlinear and threshold relationships
between the model input and explanatory variables. An explanatory variable’s influence is
calculated using MARS as the sum of the partial residuals removing all main and interaction
effects that variable enters.
∑∑∑∑⎥
⎦
⎤
⎢
⎣
⎡++++−=
≠≠≠
2
1,,1,1
),,(),()(!
kji
kji
ji
ji
i
iox xxxfxxfxfafSi (6)
2.3.2 Gradient Boosting Machines (GBM)
Boosting of regression trees provides a technique that adds to the flexibility offered by recursive
partitioning methods such as MARS. The GBM (Friedman 2001, Friedman 2002) approach
utilizes boosting of binary recursive partitioning algorithms that deconstructs a response into the
relative influence from a given set of explanatory variables (i.e. PA model input parameters).
The deconstruction breaks the PA model into separate parts (braches of the regression tree), and
each part is examined separately. This process is repeated with smaller and smaller parts, each
analyzed for the relationship between the explanatory variables and the PA model output (i.e. the
response of interest). The deconstructed parts are then collected together to provide estimates of
the sensitivity of each exploratory variable for a specific response variable from the PA model.
Hence, GBM provides a method for constructing a statistical meta-model of the more complex
PA simulation model. The GBM approach identifies the most influential explanatory variables in
the context of the observed uncertainty in the PA model output or response. Critically, the GBM
method also identifies the range over which the influence is strongest, which can be used to
better understand the full effect of a sensitive explanatory variable on the output results.
Variance decomposition of the GBM fit is then used to estimate SIs. Under this decomposition
approach, the goal is to identify the most influential explanatory variables that are identified
within a model. The necessary degree of model complexity can be assessed using validation
metrics, based on comparison of model predictions, with randomly selected subsets of the data.
This approach uses the “deviance” of the model as a measure of goodness of fit. The concept of
deviance is fundamental to classical statistical hypothesis tests (e.g., the common t-test can be
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 7
derived using a deviance-based framework) and guides the model selection process applied here.
The deviance for a model given a set of data, y, is defined as
𝐷𝑦=  −2(log 𝑝𝑦𝜃!"#−log 𝑝𝑦𝜃!"## (7)
Where 𝜃!"# is the vector of fitted values from the model of interest, and 𝜃!"## is the vector of
fitted values from the saturated model. Equation 7 is effectively the log likelihood ratio of the
fitted versus the full model. In this sense, it measures the deviance of the fitted model from the
full model.
Given that the PA models are deterministic simulations, where the only stochasticity comes from
the distributions assigned to each of the explanatory variables, the variance of the response from
the PA is solely attributable to the model uncertainty. Hence, these models can’t be overfit in the
same way that they would when applied to observational data. Because of this, the determination
of sufficient GBM model complexity should be focused on the minimal amount necessary to
explain approximately the maximal amount of the observed variance in the PA model response.
The GBM fitting approach is based on finding the values of each explanatory variable that result
in the greatest difference in the mean for the corresponding subsets of the response. For example,
if there were only a single explanatory variable, the GBM would identify the value of the
explanatory variable that corresponds to a split of the response into two parts for which the
means are more different than those corresponding to any other split of the response into two
subsets. When multiple explanatory variables are present, multiple splits are made corresponding
to each of the explanatory variables and the collection of splits is referred to as a “tree”. Each
tree results in an estimate (e.g., prediction) of the response. As multiple potential trees are
evaluated, they are compared to the observed data using a loss function. The selection of the loss
function is an important aspect of the GBM process, and depends on the distribution of the
response variable. For data that are sufficiently skewed (e.g., non-normal), experience has shown
the absolute error loss function typically produces more reliable results. This is often the case
with realizations of responses generated from PA models since there are often many cases of
small response (e.g., dose, flux, concentration) and only a few for which large responses are
simulated by the PA model.
There is a trade-off that exists when considering which loss function to use. The squared-error
loss function results in better fitting models, but can do so at the expense of introducing spurious
variables into the model selection process when the response distribution is sufficiently skewed.
The absolute error loss function produces model predictions with more variability, but is less
likely to result in the selection of spurious variables in the model. This is due to the squared-error
loss function methods increased sensitivity to results from the tails of distributions.
When considering the utility of the GBM approach and its increased computational burden
relative to simpler linear regression based approaches, it is important to recall that linear
regression techniques (e.g. SSRC) assume that the relationship between the response and the
explanatory variable is a constant (i.e. the statistical model is linear in the parameter space). With
the GBM approach, this relationship is not constrained by assumptions of linearity, and partial
dependence plots are used to show the estimate of the relationship between the response (i.e. the
output from the PA model) and the explanatory variables.
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 8
Partial dependence plots are used to describe and interpret the results of the SA. The partial
dependence curve depicts the change in the value of an endpoint as a function of the values of
the response variable (see blue curves in Figure 1). It is conceptually equivalent to the slope in a
linear regression model, but shows the non-linear impact across the range of input values. In a
linear regression model, this relationship is constrained to be constant. That is, the relationship
between a change in the explanatory variable and the endpoint of interest does not change across
the range of values for the explanatory variable. In the GBM approach, the relationship between
the explanatory variable and the endpoint of interest is allowed to change across the range of
values for the explanatory variable. For example, the x2 explanatory variable in Figure 1 displays
a different relationship to the response between the values of 0 and 0.5 than it does between 0.5
and 1.0. Specifically, for the range of explanatory variable values between 0 and 0.5 there is a
decrease in the response; however for the range of values between 0.5 and 1.0, there is an
increase in the response. This is a powerful distinction between the GBM approach and other
meta-modeling applications that do not allow this functional flexibility in the relationship
between the explanatory variable and response to be evaluated.
2.4 Example: Comparison of SA methods
2.4.1 “Sobol g-function”
The Sobol g-function (Saltelli et. al. 1999) provides an analytic non-monotonic test function for
evaluating the performance of various SA methods. This function is defined as:
∏
=
=
p
i
iixgf
1
)( (7)
where p is the total number of input factors and gi(xi) is given by

i
iiiia
axxg+
+−=1
|24|)(, (8)
with
xi =1
2 +1
π
arcsin(sin(ωis +φi)), (9)
and s varying along (-π,π), ϕi ~ U [0,2π), and ωi are specified frequencies.
The Sobol g function was simulated for p = 8 and frequencies {ωi} = {23, 55, 77, 97, 107, 113,
121, 125} for a specific set of ai’s. Table 1 provides a comparison of sensitivity indices, S,
calculated analytically (Saltelli et. al. 1999) versus those computed using GBM, MARS, FAST,
differential analysis (DERIV), squared standardized regression coefficients (SSRC), and, squared
standardized rank regression coefficients (SSRRC). DERIV is used to represent the calculus
based approach based on derivatives evaluated at a point.
Note that the GBM, MARS and FAST methods all return sensitivity indices that are close to the
actual sensitivities for the Sobol function (S). The Sobol function is highly non-linear; hence the
standardized regression approaches (i.e. SSRC and SSRRC) do not work very well. As described
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 9
above, FAST is computationally challenging. The difference between MARS and GBM is small,
but preference based on the results in this example is given overall to the GBM approach.
A goodness-of-fit statistic is also presented in the bottom row of Table 1. This is calculated as
the standard chi-square goodness-of-fit statistic (i.e. the sum of the square of the observed (SA
method) minus the expected (S value)) all divided by the expected value, in which case a small
value implies a better fit. These goodness-of-fit statistics show that the GBM method
outperforms the other methods, although the difference is small for GBM and FAST.
Table 1. Sensitivity Indices by Sensitivity Analysis Method for Sobol g-function application
with p = 8.
a S GBM MARS FAST DERIV SSRC SSRRC
x1 99 0.0001 0.0003 0.0000 0.0043 0.0037 0.6880 0.7805
x2 0 0.4227 0.4146 0.4397 0.4287 0.3151 0.0137 0.0036
x3 9 0.0058 0.0011 0.0084 0.0190 0.0401 0.0003 0.0000
x4 0 0.4227 0.4200 0.4239 0.4269 0.3169 0.0163 0.0098
x5 99 0.0001 0.0001 0.0000 0.0006 0.0037 0.0350 0.1152
x6 4.5 0.0182 0.0335 0.0239 0.0141 0.0787 0.0012 0.0554
x7 1 0.1304 0.1303 0.1041 0.1063 0.2382 0.0574 0.0344
x8 99 0.0001 0.0000 0.0000 0.0002 0.0037 0.1881 0.0012
Goodness-of-Fit statistic 3.3 14.8 3.6 470 7,250 535

GBM is run on the realizations themselves, whereas FAST requires set up in terms of an
embedded signal. This makes FAST relatively cumbersome to deal with. Also, GBM
outperforms MARS, which is not as flexible and takes more time to implement computationally.
GBM tends to provide the best fit, is flexible and is applied directly to the realizations from the
PA model. Consequently, GBM is the preferred method, and the one that is used for the
sensitivity analyses for the Clive DU PA.
2.4.2 Visualization
Once a GBM has been is constructed, every explanatory variable in the PA model has a
corresponding sensitivity index (SI). Experience has shown that for PA models with hundreds of
parameters, the majority of them will have SI values that are very near zero. That is, for a given
response from the PA model, the majority of the uncertainty in the response values simulated by
the PA model will be attributable to a handful of explanatory variables. The collection of
important variables will change as different responses from the PA model are considered.
The SI is obtained through variance decomposition and can be interpreted as the percent of
variability in the PA model output explained by a given explanatory variable. The sum of the
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 10
SI’s across the entire set of explanatory variables in the PA model will approximately equal the
R2 of the linear regression of the realizations from the PA model on the GBM predictions.
For a GBM model, the partial dependence of the response on each explanatory variable is
determined through the integration across the joint density of the explanatory variables to obtain
a marginal distribution for each explanatory variable. That is, if there are n explanatory variables
in a given PA model, the partial dependence for a single explanatory variable is obtained by
integrating across the other n-1 explanatory variables in the PA model. The integration is
performed using a “weighted tree traversal” measure that is analogous to more common
integration procedures performed with Riemann or Lebesgue measures (Friedman 2001).
The vertical axis of a partial dependence plot has units corresponding to those of the response
variable of interest for the meta-model built on the PA model output response. The partial
dependence shows the change in the response variable as a function of the changes in the
explanatory variable. If the underlying relationship between the response from the PA model and
the explanatory variable of interest is linear in the parameters (as is the assumption for a linear
regression model), then the partial dependence curve will be a line with slope equal to that of the
corresponding regression model.
In order to assess the relationship between an individual explanatory variable and the response of
interest, partial dependence plots are used (an example is provided in Figure 1). The first panel
depicts a density estimate of the simulated response from the PA model as well as the R2 and
summary statistics for the response. The percentiles of the response distribution in this panel are
shaded to provide a context for the partial dependence plots presented in the remaining panels.
The colors indicate the percentile range of the response as follows:
1. The 0th - 25th percentile region is shaded orange-brown
2. The 25th - 50th percentile region is shaded dark yellow-green
3. The 50th - 75th percentile region is shaded light green
4. The 75th - 100th percentile region is shaded light blue
To reiterate, the y-axis of the partial dependence plots is in units of the response distribution
(which is the x axis of the first panel in the upper left). Given that each parameter has a different
range and strength of influence on the response, the y axes of the partial dependence panels have
been constructed to depict only the range of the response over which a particular parameter is
influential. This provides the most meaningful presentation of the variables and their
relationships. In contrast, if the original scale of the response were maintained on each partial
dependence panel, then the influence of the least influential parameter would not be visible in
many cases. To mediate issues associated with this change in range of the vertical axes among
the different partial dependence plots, the background of the partial dependence panels is colored
to depict the percentiles of the response over which the parameter is influential. For example, if
the background of the partial dependence plot under the partial dependence line is light blue,
then that indicates the parameter’s influence on the upper end of the response distribution (i.e.,
the 75th to 100th percentile of the response).
The partial dependence panels in each figure show the distributions of the explanatory variables
(black line), and the partial dependence curve (blue line) shows changes in the response as a
function of each explanatory variable.
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 11

Figure 1. Sensitivity and Partial Dependence Plots for the GBM fit to the Sobol Function.
The plots show that the distributions for the three input parameters are uniform, and that the
effects show sensitivity across the entire range of the inputs. The effects are first negative, and
then positive, which is to be expected given Equation 15. Also note that the linear regression
methods would not be able to track the non-linearity, and instead fits a straight, horizontal line
for these parameters, which shows them to be non-sensitive. This is a prime example of why
methods such as GBM are advantageous.
NA
Pr
o
b
a
b
i
l
i
t
y

D
e
n
s
i
t
y
0 2 4 6
y
R² = 0.95
Mean: 1 50%: 0.7 95%: 3 99%: 4.3
0−25% 25−50% 50−75% 75−100%
5.0e−01
1.0e+00
1.5e+00
0.0 0.2 0.4 0.6 0.8 1.0
x2
SI = 42
5.0e−01
1.0e+00
1.5e+00
0.0 0.2 0.4 0.6 0.8 1.0
x4
SI = 41.7
8.0e−01
1.0e+00
1.2e+00
0.0 0.2 0.4 0.6 0.8 1.0
x7
SI = 12.2
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 12
Table 2 shows a specific example for one of the Clive DU PA endpoints of interest – the peak
groundwater concentrations within 500 years for technetium-99. As can be seen, all explanatory
variables (input parameters) are included in the GBM-based global SA. The dependent variable
(output variable or response variable) of interest is the peak groundwater concentration of 99Tc in
500 years. The most sensitive input parameter is the van Genuchten alpha parameter. This is
consistent with the conceptual understanding of the model. That is, infiltration is low and
diffusion dominates infiltration as a mechanism of movement. Note that, although inventory of 99Tc is the fourth most sensitive input parameter, its impact on the output is dominated by the
van Genuchten alpha parameter. This means that the output is relatively less affected by the
inventory of 99Tc because the inventory uncertainty is swamped by the uncertainty of the impact
of the van Genuchten alpha parameter. This is true only for the range considered in the 99Tc
inventory distribution.
Further examples are provided in the Sensitivity Analysis Results (Appendix 19) v1.4 White
Paper.
Table 2. Peak Groundwater Well Concentrations within 500 years - Tc99
Explanatory
 Variable
 Sensitivity
 Index

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 31.97

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 24.96

Kd
 Sand
 for
 Tc
 (mL/g)
 13.97

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 10.59

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 3.83

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 1.26

Saturated
 Zone
 Water
 Table
 Gradient
 1.20

OHV
 Dust
 Adjustment
 0.55

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.38

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.34

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.34

Fine
 CobbleMix
 Porosity
 0.28

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.26

Kd
 Silt
 for
 Ra
 (mL/g)
 0.22

Surface
 Atmosphere
 Thickness
 (m)
 0.21

Unit
 4
 Compacted
 Hb
 (cm)
 0.20

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.20

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.19

Kd
 Silt
 for
 U
 (mL/g)
 0.17

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.16

Kd
 Clay
 for
 Sr
 (mL/g)
 0.15

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.14

Kd
 Sand
 for
 Ac
 (mL/g)
 0.14

Kd
 Clay
 for
 Ra
 (mL/g)
 0.13

Forb
 Root
 Shape
 Parameter
 b
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.12

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 13
Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.12

Kd
 Silt
 for
 Cs
 (mL/g)
 0.11

Unit
 4
 Compacted
 Porosity
 0.11

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.11

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.11

Unit
 2
 Porosity
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.10

Shrub
 Root.Shoot
 Ratio
 0.10

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.10

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.10

Grass
 Root.Shoot
 Ratio
 0.09

Shrub
 Root
 Shape
 Parameter
 b
 0.09

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.09

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.09

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.08

Deep
 Time
 DCF
 Alpha
 REF
 0.08

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.08

Saturated
 Zone
 Thickness
 (m)
 0.08

Kd
 Sand
 for
 Am
 (mL/g)
 0.08

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Kd
 Silt
 for
 Sr
 (mL/g)
 0.07

RipRap
 Bulk
 Density
 (g/cm3)
 0.07

Kd
 Clay
 for
 Cs
 (mL/g)
 0.07

Deep
 Time
 DCF
 Photon
 2
 REF
 0.07

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.07

Kd
 Clay
 for
 Ac
 (mL/g)
 0.07

Unit
 3
 Residual
 Water
 Content
 0.07

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.07

Deep
 Time
 Lake
 Start
 (yr)
 0.07

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.06

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.06

Grass
 Root
 Shape
 Parameter
 b
 0.06

Ant
 Nest
 Volume
 (m3)
 0.06

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.06

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.06

Ant
 Nest
 Shape
 Parameter
 b
 0.06

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.06

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.06

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 14
Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.06

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.06

Deep
 Time
 DCF
 Photon
 1
 REF
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.05

Fine
 Gravel
 Mix
 Porosity
 0.05

RipRap
 Porosity
 0.05

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.05

Vegetation
 Association
 Selector
 0.05

Plant.Soil
 Conc
 Ratio
 for
 U
 0.05

Surface
 Wind
 Speed
 (m/s)
 0.05

Soil
 Ingestion
 Tracer
 Element
 0.05

Tortuosity
 Water
 Content
 Exponent

 0.05

Intermediate
 Lake
 Depth
 (m)
 0.05

Kd
 Sand
 for
 Pa
 (mL/g)
 0.05

Unit
 3
 Porosity
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.05

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.05

Kd
 Silt
 for
 Th
 (mL/g)
 0.05

Kd
 Sand
 for
 U
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.05

Kd
 Clay
 for
 Am
 (mL/g)
 0.05

Site
 Dispersal
 Area
 (km2)
 0.05

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.05

Random
 Gully
 Selector
 0.04

Kd
 Sand
 for
 Th
 (mL/g)
 0.04

Antelope
 Range
 Area
 (acre)
 0.04

Kd
 Clay
 for
 Pu
 (mL/g)
 0.04

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.04

Kd
 Sand
 for
 Np
 (mL/g)
 0.04

Tortuosity
 Porosity
 Exponent
 0.04

Ant
 Colony
 Lifespan
 (yr)
 0.04

Plant
 Fresh
 Weight
 Conversion
 0.04

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.04

Meat
 Post-­‐Cooking
 Loss
 0.04

Kd
 Sand
 for
 Cs
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.04

Body
 Weight
 Factor
 for
 Antelope
 0.04

Kd
 Silt
 for
 Pa
 (mL/g)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.04

Kd
 Sand
 for
 Ra
 (mL/g)
 0.04

Kd
 Silt
 for
 Np
 (mL/g)
 0.04

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 15
Kd
 Clay
 for
 Pb
 (mL/g)
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.04

Receptor
 Area
 (ha)
 0.04

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.04

DCF
 Alpha
 REF
 0.04

Biomass
 %
 Cover
 Selector
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.04

Unit
 4
 ET
 Layers
 Porosity
 0.04

Deep
 Time
 Aeolian
 Correlation
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 I
 0.04

Kd
 Sand
 for
 Pu
 (mL/g)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.03

DCF
 Beta
 REF
 0.03

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.03

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.03

Mammal
 Burrow
 Shape
 Parameter
 b
 0.03

Forb
 Root.Shoot
 Ratio
 0.03

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.03

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.03

Kd
 Silt
 for
 Am
 (mL/g)
 0.03

Deep
 Time
 DCF
 Beta
 REF
 0.03

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.03

DCF
 Photon1
 REF
 0.03

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.03

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.03

Kd
 Clay
 for
 Pa
 (mL/g)
 0.03

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.03

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.03

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.03

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.03

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.03

Kd
 Silt
 for
 Ac
 (mL/g)
 0.03

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.03

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.03

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.03

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 16
Tree
 Root.Shoot
 Ratio
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.03

Kd
 Silt
 for
 Pu
 (mL/g)
 0.03

Deep
 Time
 Diffusion
 Length
 (m)
 0.02

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.02

Silt
 Sand
 Gravel
 Porosity
 0.02

Deep
 Lake
 Depth
 (m)
 0.02

Resuspended
 Particle
 Fraction
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.02

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.02

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.02

Kd
 Sand
 for
 Sr
 (mL/g)
 0.02

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.02

Kd
 Clay
 for
 Th
 (mL/g)
 0.02

Soil
 Temperature
 (°C)
 0.02

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.02

Meat
 Preparation
 Loss
 0.02

Kd
 Sand
 for
 Pb
 (mL/g)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.02

DCF
 Photon2
 REF
 0.02

Kd
 Silt
 for
 Pb
 (mL/g)
 0.02

Greasewood
 Root.Shoot
 Ratio
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.02

Greasewood
 Root
 Shape
 Parameter
 b
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.02

Kd
 Sand
 for
 I
 (mL/g)
 0.02

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Kd
 Clay
 for
 U
 (mL/g)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 17
3.0 References
Borgonovo E., Castaings W., and Tarantols S. (2012) “Model emulation and moment-
independent sensitivity analysis: An application to environmental modelling”,
Environmental Modelling and Software 34:105-115.
Cea, L., Bermudez M., and Puertas J. (2011) “Uncertainty and sensitivity analysis of a depth-
averaged water quality model for evaluation of Escherichia Coli concentration in shallow
estuaries” Environmental Modelling and Software 26:1526-1539.
Coutts S.R. and Yokomizo H. (2014) “Meta-models as a straightforward approach to the
sensitivity analysis of complex models”, Population Ecology 56:7-19.
Friedman J.H. (1991) “Multivariate Adaptive Regression Splines”, Annals of Statistics 19(1):1-
67.
Friedman J.H. (2001) “Greedy function approximation: A gradient boosting machine”, Annals of
Statistics 29(5):1189-1232.
Friedman J.H. (2002) “Stochastic gradient boosting”, Computational Statistics and Data
Analysis 38:367-378.
Liu H., Chen W., and Sudjianto A. (2006) “Relative Entropy Based Method for Probabilistic
Sensitivity Analysis in Engineering Design” Journal of Mechanical Design 128:326-336.
Marrel A.M., Iooss B., Laurent B., and Roustant O. (2009), “Calcluations of Sobol indices for
the Gaussian process metamodel” Reliability Engineering and Safety System 94:742-751.
Marrel A.M., Iooss B., Jullien M., Laurent B., and Volkova E. (2011), “Global sensitivity
analysis for models with spatially dependent outputs” Environmetrics 22:383-397.
McKay M. D., Beckman R.J., and Conover W.J. (1979) “A comparison of three methods for
selecting values of input variables in the analysis of output from a computer code”
Technometrics 21(2): 239-245.
Melbourne-Thomas J., Wotherspoon S., Raymond B., and Constable A. (2012) “Comprehensive
evaluation of model uncertainty in qualitative network analyses” Ecological Monographs
82(4):505-519.
Morris D.J., Speirs D.C., Cameron A.I., Heeath M.R. (2014) “Global sensitivity analysis of an
end-to-end marine ecosystem model of the North Sea: Factors affecting the biomass of
fish and benthos”, Ecological Modelling 273:251-263.
Nossent J., Elsen P., and Bauwens W. (2011) “Sobol’s sensitivity analysis of a complex
environmental model”, Environmental Modelling and Software 26:1515-1525.
Saltelli A., Tarantola S., and Chan K.P.-S. (1999) “A quantitative model-independent method for
global sensitivity analysis of model output”, Technometrics 41:39-55.
Sobol I.M. (1993) “Sensitivity analysis for nonlinear mathematical models”, Mathematical
Modeling & Computational Experiment 1:407-414.
Sobol, I.M. (1993), “Sensitivity Analysis for Nonlinear Mathematical Models,” Mathematical
Modeling & Computational Experiment, 1, 407-414.
Sobol I.M. (2001) “Global sensitivity indices for nonlinear mathematical models and their Monte
Carlo estimates”, Mathematics and Computers in Simulation 55:271-280.
Machine Learning for Sensitivity Analysis of Probabilistic Environmental Models
5 November 2015 18
Storlie C.B., Swiler L.P., Helton J.C., and Sallaberry C.J. (2009) “Implementation and evaluation
f nonparametric regression procedures for sensitivity analysis of computationally
demanding models.” Reliability Engineering and Safety System 94:1735-1763.

 NAC-0026_R4

Model Parameters for the
Clive DU PA Model

Clive DU PA Model v1.4
8 November 2015

Prepared by

NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Model Parameters for the Clive DU PA Model
25 November 2015 ii
1. Title: Model Parameters for the Clive DU PA Model
2. Filename: Clive PA Model Parameters v1.4.docx
3. Description: This white paper provides documentation of all the parameters in the Clive DU
PA Model and references for their values and input distributions.

Name Date
4. Originator Gregg Ochiogrosso 27 October 2015
5. Reviewer Katie Catlett 8 November 2015
6. Remarks

17 Jul 2015 Amir Mokhtari updated terminology from “Class A South” to “Federal DU.”
29 Jul 2015 K. Catlett QA’d those changes.
11 August 2015 Amir Mokhtari updated deep time parameters
30 Sep 2015 G. Occhiogrosso Updated water content and infiltration regression coefficients
as well as saturated hydraulic conductivity distribution based on Calc Sheet ES 006 Rev 1.
27 Oct 2015 G. Occhiogrosso Updated with new geometry for the DU cell for v1.4 in
Section 7.2, with associated updates elsewhere. Revised several sections in Materials section to
clarify values for materials derived from Unit 4 materials. Other minor revisions for updated
values.
8 Nov 2015 K Catlett. Accepted changes and minor edits on a few references in the Deep Time
dose assessment tables.

Model Parameters for the Clive DU PA Model
25 November 2015 iii

This page is intentionally blank, aside from this statement.
Model Parameters for the Clive DU PA Model
25 November 2015 iv
CONTENTS
FIGURES ...................................................................................................................................... vii

TABLES ....................................................................................................................................... viii

1.0
 Introduction ............................................................................................................................ 1

2.0
 Distribution Specification ....................................................................................................... 1

3.0
 \SimulationSettings ................................................................................................................. 1

3.1
 Simulation Settings (the GoldSim dialog) ........................................................................ 2

3.2
 \SimulationSettings\Chronology ....................................................................................... 3

3.3
 \SimulationSettings\Switches ........................................................................................... 4

4.0
 \Materials ................................................................................................................................ 4

4.1
 \Materials\DecayChains .................................................................................................... 5

4.2
 \Materials\Loess_Properties .............................................................................................. 8

4.3
 \Materials\Unit4_Compacted_Properties .......................................................................... 9

4.4
 \Materials\Unit4_ETLayers_Properties ............................................................................ 9

4.5
 \Materials\Unit3_Properties ............................................................................................ 10

4.6
 \Materials\Unit2_Properties ............................................................................................ 11

4.7
 \Materials\RipRap_Properties ......................................................................................... 11

4.8
 Materials\FineCobbleMix_Properties ............................................................................. 12

4.9
 Materials\SiltSandGravel_Properties .............................................................................. 12

4.10
 Materials\FineGravelMix_Properties ............................................................................. 12

4.11
 Materials\UpperRnBarrierClay_Properties .................................................................... 13

4.12
 Materials\LowerRnBarrierClay_Properties .................................................................... 13

4.13
 Materials\LinerClay_Properties ...................................................................................... 14

4.14
 \Materials\UO3_Waste_Properties ................................................................................. 14

4.15
 \Materials\Waste_U3O8_Properties ............................................................................... 14

4.16
 \Materials\Generic_Waste_Properties ............................................................................ 14

4.17
 \Materials\Water_Properties ........................................................................................... 14

4.18
 \Materials\Kd .................................................................................................................. 15

4.18.1
 \Materials\Kd\Kd_Sand_Values ............................................................................... 15

4.18.2
 \Materials\Kd\Kd_Silt_Values .................................................................................. 16

4.18.3
 \Materials\Kd\Kd_Clay_Values ................................................................................ 16

4.19
 \Materials\WaterSolubility .............................................................................................. 17

4.19.1
 \Materials\WaterSolubility\Solubilities_Saltwater ................................................... 17

4.20
 \Materials\Air_Properties ................................................................................................ 18

5.0
 \Processes ............................................................................................................................. 18

5.1
 \Processes\AirTransport .................................................................................................. 18

5.2
 \Processes\AnimalTransport ........................................................................................... 20

5.2.1
 \Processes\AnimalTransport\AntData ....................................................................... 20

5.2.2
 \Processes\AnimalTransport\MammalData .............................................................. 21

5.3
 \Processes\PlantTransport ............................................................................................... 21

5.3.1
 \Processes\PlantTransport\PlantCR .......................................................................... 22

5.3.2
 \Processes\PlantTransport\BiomassCalcs ................................................................. 22

5.3.3
 \Processes\PlantTransport\GreasewoodData ............................................................ 23

5.3.4
 \Processes\PlantTransport\GrassData ....................................................................... 23

5.3.5
 \Processes\PlantTransport\ForbData ......................................................................... 23

5.3.6
 \Processes\PlantTransport\TreeData ......................................................................... 23

5.3.7
 \Processes\PlantTransport\ShrubData ....................................................................... 24

5.4
 \Processes\WaterTransport ............................................................................................. 24

Model Parameters for the Clive DU PA Model
25 November 2015 v
5.5
 \Processes\ErosionTransport ........................................................................................... 24

6.0
 \Inventory ............................................................................................................................. 25

6.1
 \Inventory\SRS_DU_Inventory ...................................................................................... 25

6.2
 \Inventory\GDP_DU_Inventory ..................................................................................... 26

6.3
 \Inventory\Other_DU_Inventory .................................................................................... 26

6.4
 \Inventory\ClassA_LLW_Inventory ............................................................................... 26

7.0
 \Disposal ............................................................................................................................... 27

7.1
 \Disposal\AtmosphericDispersion .................................................................................. 27

7.1.1
 \Disposal\AtmosphericDispersion\AirConc_Onsite ................................................. 27

7.1.2
 \Disposal\AtmosphericDispersion\MediaConc_Offsite ............................................ 27

7.1.3
 \Disposal\AtmosphericDispersion\AirConc_Remote ............................................... 28

7.2
 \Disposal\FederalDUCell ................................................................................................ 29

7.2.1
 \Disposal\FederalDUCell\FederalDU_Cell_Dimensions ......................................... 30

7.2.2
 \Disposal\FederalDUCell\NaturalSystemGeometry ................................................. 31

7.2.3
 \Disposal\FederalDUCell\CapCell_Thickness .......................................................... 31

7.2.4
 \Disposal\FederalDUCell\TopSlope ......................................................................... 31

7.2.4.1
 \Disposal\FederalDUCell\TopSlope\Column_Transport ........................ 31

7.2.4.1.1
 \Disposal\FederalDUCell\TopSlope\Column_Transport \WaterTransport 32

7.2.4.2
 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile .............. 32

7.2.4.2.1
\Disposal\FederalDUCell\TopSlope\Column_MoistureProfile\WaterContent
Calcs_ETCover ................................................................................. 32

7.2.4.2.2
 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_RnBarrier ......................................................... 33

7.2.4.2.3
 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_Waste ............................................................... 33

7.2.4.2.4
 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_Liner ................................................................ 34

7.2.4.2.5
 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_Unsat ............................................................... 34

7.2.4.3
 \Disposal\FederalDUCell\TopSlope\Cap_Layers ................................... 35

7.2.4.3.1
 \Disposal\FederalDUCell\TopSlope\CapLayers\CapCell_Dimensions35

7.2.4.4
 \Disposal\FederalDUCell\TopSlope\Liner .............................................. 35

7.2.4.5
 \Disposal\FederalDUCell\TopSlope\UnsatLayer .................................... 36

7.2.4.6
 \Disposal\FederalDUCell\TopSlope\WasteLayers ................................. 36

7.2.4.6.1
 \Disposal\FederalDUCell\TopSlope\WasteLayers\ WasteCell_Dimensions
........................................................................................................... 36

7.2.5
 \Disposal\FederalDUCell\SideSlope ......................................................................... 36

7.2.5.1
 \Disposal\FederalDUCell\SideSlope\Column_Transport ....................... 36

7.2.5.1.1
 \Disposal\FederalDUCell\SideSlope\Column_Transport \WaterTransport 36

7.2.5.2
 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile ............. 37

7.2.5.2.1
 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_RnBarrier ......................................................... 37

7.2.5.2.2
 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_Waste ............................................................... 37

7.2.5.2.3
 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_Liner ................................................................ 37

7.2.5.2.4
 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_Unsat ............................................................... 37

Model Parameters for the Clive DU PA Model
25 November 2015 vi
7.2.5.3
 \Disposal\FederalDUCell\SideSlope\Cap_Layers .................................. 38

7.2.5.3.1
 \Disposal\FederalDUCell\SideSlope\CapLayers\CapCell_Dimensions 38

7.2.5.4
 \Disposal\FederalDUCell\SideSlope\Liner ............................................. 39

7.2.5.5
 \Disposal\FederalDUCell\SideSlope\UnsatLayer ................................... 39

7.2.5.6
 \Disposal\FederalDUCell\SideSlope\WasteLayers ................................. 39

7.2.5.6.1
 \Disposal\FederalDUCell\SideSlope\WasteLayers\ WasteCell_Dimensions
........................................................................................................... 39

7.2.6
 \Disposal\FederalDUCell\ErosionCalcs .................................................................... 39

7.2.6.1
 \Disposal\FederalDUCell\ErosionCalcs\SiberiaErosionCalcs ................ 39

7.3
 \Disposal\SatZone ........................................................................................................... 39

7.3.1
 \Disposal\SatZone\SatZone_Parameters ................................................................... 40

7.3.2
 \Disposal\SatZone\SZ_FederalDUFootprint ............................................................. 40

7.3.2.1
 \Disposal\SatZone\SZ_FederalDUFootprint\Waste_to_Footprint .......... 40

7.3.3
 \Disposal\SatZone\SZ_ToWell ................................................................................. 40

7.4
 \Disposal\EngineeredSystemGeometry .......................................................................... 40

8.0
 \Exposure_Dose .................................................................................................................... 41

8.1
 \Exposure_Dose\Media_Concs ....................................................................................... 41

8.1.1
 \Exposure_Dose\Media_Concs\Exposure_Areas ..................................................... 41

8.1.2
 \Exposure_Dose\Media_Concs\Animal_Concentrations .......................................... 41

8.1.2.1
 \Exposure_Dose\Media_Concs\Animal_Concentrations\Beef_TFs ....... 42

8.2
 \Exposure_Dose\DCFs .................................................................................................... 43

8.2.1
 \Exposure_Dose\DCFs\Stochastic_REFs ................................................................. 43

8.3
 \Exposure_Dose\OuterLoop_Exposure_Parameters ...................................................... 45

8.4
 \Exposure_Dose\Dose_Calculations ............................................................................... 45

8.4.1
 \Exposure_Dose\Dose_Calculations\Physiology_Rancher ...................................... 46

8.4.2
 \Exposure_Dose\Dose_Calculations\Physiology_SportOHV .................................. 47

8.4.3
 \Exposure_Dose\Dose_Calculations\Physiology_Hunter ......................................... 48

8.4.4
 \Exposure_Dose\Dose_Calculations\ExposureTime_Rancher ................................. 49

8.4.5
 \Exposure_Dose\Dose_Calculations\ExposureTime_SportOHV ............................. 50

8.4.6
 \Exposure_Dose\Dose_Calculations\ExposureTime_Hunter ................................... 51

8.4.7
 \Exposure_Dose\Dose_Calculations\Population_Size_Variables ............................ 52

8.4.8
 \Exposure_Dose\Dose_Calculations\UraniumHazard .............................................. 53

8.4.9
 \Exposure_Dose\Dose_Calculations\OffSite_Receptors .......................................... 53

8.4.10
 \Exposure_Dose\Screening_Calculations ................................................................. 54

9.0
 \GWPLs ................................................................................................................................ 54

10.0
 \DeepTimeScenarios ............................................................................................................. 55

10.1
 \DeepTimeScenarios\LakeReturnCalcs .......................................................................... 56

10.2
 \DeepTimeScenarios\LakeChemistry ............................................................................. 56

10.3
 \DeepTimeScenarios\RadonFlux_NRC364 .................................................................... 57

10.4
 \DeepTimeScenarios\ExposureDose_DeepTime ............................................................ 57

10.4.1
 \DeepTimeScenarios\ExposureDose_DeepTime\Exposure_Areas .......................... 58

10.4.2
 \DeepTimeScenarios\ExposureDose_DeepTime\DCFs ........................................... 58

10.4.2.1
 DeepTimeScenarios\ExposureDose_DeepTime\DCFs\Stochastic_REFs58

11.0
 References ............................................................................................................................ 59

Model Parameters for the Clive DU PA Model
25 November 2015 vii
FIGURES
Figure 1. Decay chains modeled in the Clive DU PA Model, part 1 of 2. ...................................... 6

Figure 2. Decay chains modeled in the Clive DU PA Model, part 2 of 2. ...................................... 7

Figure 3. Details of the actinide decay chains modeled in the Clive DU PA Model, showing
which species are omitted, in gray. ............................................................................... 8

Model Parameters for the Clive DU PA Model
25 November 2015 viii
TABLES
Table 1. Statistical distribution types used in the parameter specifications. ................................... 1

Table 2. Generic constants used in simulations ............................................................................... 2

Table 3. Monte Carlo simulation settings ........................................................................................ 2

Table 4. Times Phase Settings for the full 2.1-million year run ...................................................... 3

Table 5. Global events and their probability of occurrence ............................................................ 3

Table 6. Atomic mass of Species .................................................................................................... 5

Table 7. Atomic masses of other elements ...................................................................................... 5

Table 8. Unit 4 compacted material properties ................................................................................ 9

Table 9. Unit 4 ET Layers material properties .............................................................................. 10

Table 10. Unit 3 material properties .............................................................................................. 10

Table 11. Unit 2 material properties .............................................................................................. 11

Table 12. Rip rap material properties ............................................................................................ 11

Table 13. Fine cobble mix material properties .............................................................................. 12

Table 14. Silt sand gravel material properties ............................................................................... 12

Table 15. Fine gravel mix material properties ............................................................................... 13

Table 16. Upper radon barrier clay material properties ................................................................. 13

Table 17. Lower radon barrier clay material properties ................................................................ 13

Table 18. Liner clay material properties ........................................................................................ 14

Table 19. Properties of water, the reference fluid. ........................................................................ 14

Table 20. Soil/water partition coefficients (Kds) for sand ............................................................. 15

Table 21. Soil/water partition coefficients (Kds) for silt ................................................................ 16

Table 22. Soil/water partition coefficients (Kds) for clay .............................................................. 16

Table 23. Aqueous solubilities in saltwater, by chemical element ................................................ 17

Table 24. Parameters relevant to diffusion in air. ......................................................................... 18

Table 25. Radon diffusive transport parameters. ........................................................................... 18

Table 26. Atmospheric transport parameters. ................................................................................ 19

Table 27. Model parameters for ants. ............................................................................................ 20

Table 28. Model parameters for small mammals. ......................................................................... 21

Table 29. Parameters general to all plants. .................................................................................... 22

Table 30. Plant/soil concentration ratio parameters. ..................................................................... 22

Table 31. Biomass calculation parameters. ................................................................................... 22

Table 32. Greasewood parameters. ............................................................................................... 23

Table 33. Grass parameters. .......................................................................................................... 23

Table 34. Forb parameters. ............................................................................................................ 23

Table 35. Tree parameters. ............................................................................................................ 23

Table 36. Other shrub parameters. ................................................................................................ 24

Table 37. Water transport parameters. ........................................................................................... 24

Table 38. Water transport parameters. ........................................................................................... 25

Table 39. SRS DU inventory parameters. ..................................................................................... 25

Table 40. GDP DU inventory parameters. .................................................................................... 26

Table 41. Atmosphere dispersion parameters for on-site exposures. ............................................ 27

Model Parameters for the Clive DU PA Model
25 November 2015 ix
Table 42. Atmosphere dispersion parameters for off-site exposures (in the “air dispersion”
area.) ............................................................................................................................ 28

Table 43. Atmosphere dispersion parameters for remote off-site exposures. ............................... 28

Table 44. Interior (waste) dimensions of the Federal Cell, Federal DU section. .......................... 30

Table 45. Natural system geometry parameters for the Federal DU cell. ..................................... 31

Table 46. Dimensions of the cap cells for the Federal DU cell. .................................................... 31

Table 47. Infiltration parameters for cap cells. .............................................................................. 32

Table 48. Parameters for moisture profile calculations for the ET Cover. .................................... 32

Table 49. Parameters for moisture profile calculations for the radon barrier. ............................... 33

Table 50. Parameters for moisture profile calculations for the waste. .......................................... 33

Table 51. Parameters for moisture profile calculations for the clay liner. .................................... 34

Table 52. Parameters for moisture profile calculations for the unsaturated zone below the
clay liner. ..................................................................................................................... 34

Table 53. Cap layering dimensions for the top slope. ................................................................... 35

Table 54. Number of liner cells. .................................................................................................... 35

Table 55. Number of unsaturated zone cells. ................................................................................ 36

Table 56. Top slope waste cell dimensions. .................................................................................. 36

Table 57. Parameters for moisture profile calculations for the radon barrier. ............................... 37

Table 58. Parameters for moisture profile calculations for the waste. .......................................... 37

Table 59. Parameters for moisture profile calculations for the clay liner. .................................... 37

Table 60. Cap layering dimensions for the side slope. .................................................................. 38

Table 61. Side slope waste cell dimensions. ................................................................................. 39

Table 62. SIBERIA erosion parameters. ................................................................................... 39

Table 63. Saturated zone parameters. ............................................................................................ 40

Table 64. Total number of cells in the saturated footprint zone. ................................................... 40

Table 65. Total number of cells in both footprint ends. ................................................................ 40

Table 66. Total number of cells from footprint to well. ................................................................ 40

Table 67. Engineered system geometry parameters. ..................................................................... 40

Table 68. Mechanically generated dust ......................................................................................... 41

Table 69. Exposure areas used in the calculation of exposure media concentrations ................... 41

Table 70. Animal tissue concentrations for the recreational and ranching scenarios ................... 41

Table 71. Parameters related to beef transfer factors ................................................................... 42

Table 72. Dose conversion factors ................................................................................................ 43

Table 73. Stochastic radiation effectiveness factors ...................................................................... 43

Table 74. Exposure parameters, sampled once per realization ...................................................... 45

Table 75. Attributes of inter-individual uncertainty in physiological characteristics for
rancher receptors (ranch hands) ................................................................................... 46

Table 76. Attributes of inter-individual uncertainty in physiological characteristics for Sport
OHV receptors ............................................................................................................. 47

Table 77. Attributes of inter-individual uncertainty in physiological characteristics for Hunter
receptors ...................................................................................................................... 48

Table 78. Attributes of inter-individual uncertainty in physiological characteristics for
Rancher receptors – Exposure Time ............................................................................ 49

Model Parameters for the Clive DU PA Model
25 November 2015 x
Table 79. Attributes of inter-individual uncertainty in physiological characteristics for Sport
OHV receptors – Exposure Time ................................................................................ 50

Table 80. Attributes of inter-individual uncertainty in physiological characteristics for Hunter
receptors – Exposure Time .......................................................................................... 51

Table 81. Attributes of population variability. .............................................................................. 52

Table 82. Uranium hazard for Rancher and Recreationists. .......................................................... 53

Table 83. Inhalation dose for off-site receptors. ............................................................................ 53

Table 84. Parameters used in screening dose calculations. ........................................................... 54

Table 85. Groundwater protection limits. ...................................................................................... 54

Table 86. Deep time scenario parameters. ..................................................................................... 55

Table 87. Parameters for the lake return calculations. .................................................................. 56

Table 88. Parameters for calculating the dispersal of the embankment and subsequent lake
and sediment concentrations. ....................................................................................... 56

Table 89. Parameters for the deep time human exposure and dose assessment. ........................... 57

Table 90. Exposure areas used in the calculation of exposure media concentrations. .................. 58

Model Parameters for the Clive DU PA Model
25 November 2015 1
1.0 Introduction
This document, along with the complementary Excel workbook, Clive PA Model Parameters.xls,
is a collection of all the input parameters used in the Clive DU PA GoldSim model. The
workbook contains those parameters that are most conveniently stored in arrays (such as
collections of values by contaminant Species or by chemical Elements), and this document
contains individual parameter values and distributions, organized by Containers in the model.
Expressions and other operators that do not have model inputs are not represented in these
documents. Some input distributions refer to other expression for part of their specification.
Rather than writing in those expressions, these are generally noted here as simply “f(x)”.
2.0 Distribution Specification
Distributions in this document are specified according to the notation shown in Table 1.
Table 1. Statistical distribution types used in the parameter specifications.
distribution type value or distribution
discrete value
uniform U( minimum, maximum )
log uniform LU( minimum, maximum )
triangular Tri( minimum, mode, maximum )
normal N( mean µ, standard deviation σ )
truncated normal N( mean µ, standard deviation σ, minimum, maximum )
log-normal LN( geometric mean GM, geometric standard deviation GSD )
truncated log-normal LN( GM, GSD, minimum, maximum )
beta (generalized) beta( mean µ, standard deviation σ, minimum, maximum )
Weibull W( minimum, Weibull slope, mean - minimum )
Gamma Gamma( mean µ, standard deviation σ )

3.0 \SimulationSettings
The SimulationSettings container has two primary subcontainers, Chronology and Switches. A
standard set of simulation settings is suggested in order to control intercomparisons between
various runs. The standard set includes Simulation Settings and the values of the various
Switches.

Model Parameters for the Clive DU PA Model
25 November 2015 2
Table 2. Generic constants used in simulations
GoldSim
element
value units reference / comment
Small 1 × 10–30 — arbitrarily small number for use in modeling constructs
Large 1 × 1030 — arbitrarily large number for use in modeling constructs
U_mask vector by
species of
1's for U
species, 0's
for non-U
species
Modeling construct: All uranium isotopes have a value
of 1, and all other radionuclides have a value of 0.

3.1 Simulation Settings (the GoldSim dialog)
The GoldSim Simulation Settings dialog (accessed through the F2 key, or from the menu as Run
| Simulation Settings...) controls a number of settings controlling the probabilistic and
deterministic modeling runs (Table 3) as well as the specification of time steps (Table 4). Time
steps are specific so that values of time-varying outputs are recorded at various times during the
simulation. These values, the saving of which is identified by checking the “FV” column, are
then available for post-processing analysis. Users of GoldSim are able to modify these time
steps, but GoldSim Player users may not. Do not modify the 2500-yr time step length in the later
time steps, as these are assumed to exist for the deep time assessment.
If the user desires to run a shorter simulation than the full 2.1 My, this should be done using the
model’s Control Panel dashboard—not by entering in a shorter duration in the Simulation
Settings dialog. See the Clive PA Model User Guide for more details on the use of model
controls and dashboards.
Table 3. Monte Carlo simulation settings
setting value comments
Time
Time Display Units yr This is a fixed model setting.
Duration 21000000 yr 2.1 million years is required for U-238 to reach
secular equilibrium with its decay products.
Start-time / End-time — These are ignored.
Probabilistic Simulation
# Realizations variable Set by user.
# Histories to save variable Set to # Realizations for viewing all realizations;
set to zero for sensitivity analysis.
Model Parameters for the Clive DU PA Model
25 November 2015 3
setting value comments
Use Latin Hypercube
Sampling
checked Use of LHC sampling is advisable in order to
evenly sample distributions.
Repeat Sampling Sequences checked Check to ensure reproducibility.
Random Seed variable This is a user-selected value.
Deterministic Simulation
Solve Simulation
deterministically using:
Element
Deterministic
Values

The Time Phase Settings are set on the Time tab of Simulation Settings. The table of values is
shown below, but there are also related settings accessed with the Advanced... button. These
settings should be as follows:
Uncheck “Allow events to occur between timesteps”
Check “Allow dynamic reduction in timestep length”, and set “Maximum timestep length to
allow:” to be if( ETime < 10 yr then 0.1 yr else if( ETime < 1e5 yr then 40 yr else 1000 yr )).
Set “Time to use for Edit Model updates:” to 0 s.
Table 4. Times Phase Settings for the full 2.1-million year run
time range (y) # steps time step length
(y)
plot every FV
0 - 1 10 0.1 10
1 - 10 9 1 9
10 - 100 18 5 2
100 - 1000 45 20 1
1000 - 10000 36 250 2 X
10000 - 100000 36 2500 2
100000 - 2100000 800 2500 4 X

3.2 \SimulationSettings\Chronology
The model chronology is documented in this container, referenced throughout the model (Table
5).
Table 5. Global events and their probability of occurrence
GoldSim element value or
distribution
units reference / comment
Model Parameters for the Clive DU PA Model
25 November 2015 4
GoldSim element value or
distribution
units reference / comment
ModelTimeZero
time at which calculations start
2012 Assumed date for first disposal of DU
in the Federal embankment.
IC_Period
time since time zero of loss of
institutional control
discrete, 100 yr Assumed duration of active
institutional control, per regulatory
language.
CapNaturalization_Time
time since time zero to when the
cap is fully naturalized
discrete, 1 yr Assume the ET Cover becomes
naturalized at the beginning of the
simulation.
Dose_Simulation_Duration
time since time zero that dose
user-selected yr User can set this value, up to 10,000
yr, per UAC R313-28-8

3.3 \SimulationSettings\Switches
Switches that control the model are not model inputs documented here, as they are user-
selectable via the Control Panel and other dashboards.
4.0 \Materials
Most of the Species-specific properties are defined in the Excel workbook, Clive DU PA Model
Parameters.xls, since they are tabulated lists and therefore better suited to a spreadsheet format
from which values can be electronically transferred to the model. A number of parameters,
however, as well as the overall decay chain scheme, are presented in the decay chain diagrams,
shown in Figure 1 and Figure 2. Radionuclides in black are modeled for contaminant transport
and dose contributions, those in green are modeled for dose contributions only, and those in gray
are not modeled. Figure 3 shows details of those actinide decay chains where some radionuclides
are omitted from the model calculations. These are radionuclides with exceedingly small
branching fractions and/or no dose conversion factors, so they could not possibly affect model
results or decisions based on those results.
One value defined for each contaminant species in the Species element cannot be referenced to
an array: the molecular (or in this case, atomic) mass, also called the molecular or atomic weight.
GoldSim assumes the same atomic mass for all isotopes for a given chemical element. For
example, all isotopes of uranium are assigned the atomic mass of the first isotope encountered — 232U in this case. Therefore, the atomic masses shown in Table 6 are defined for each element,
not for each radionuclide. These values are entered manually into the Species element in the
\Materials container of the model. In all cases, the most abundant isotope is used, based on
inventory mass as developed in \Inventory\Total_DU_Inventory for disposed radionuclides, and
the corresponding decay products for radionuclides that ingrow. For example, the disposed mass
of thorium is reported as zero, but since most of the thorium would be ingrowing from the large
mass of 238U, the corresponding thorium isotope of greatest mass would be 230Th. This ignores
Model Parameters for the Clive DU PA Model
25 November 2015 5
the half-life of the decay products, but any error in averaged or presumed atomic masses is
expected to be quite minor, since 230Th and 232Th have very similar atomic masses anyway.
Table 6. Atomic mass of Species
Species ID atomic mass
(g/mol)
Species ID atomic mass
(g/mol)
Sr90 90 Ac227 227
Tc99 99 Th228, Th229, Th230, Th232 230
I129 129 Pa231 231
Cs137 137 U232, U233, U234, U235, U236, U238 238
Pb210 210 Np237 237
Rn222 222 Pu238, Pu239, Pu240, Pu241, Pu242 239
Ra226, Ra228 226 Am241 241

Other chemical elements used in the model have their atomic masses listed in Table 7.
Table 7. Atomic masses of other elements
GoldSim element value or
distribution
units reference / comment
Fluorine_AtomicMass 19.0 g/mol Chart of the Nuclides, 16th Edition
Oxygen_AtomicMass 16.0 g/mol ibid.

4.1 \Materials\DecayChains
Decay chains are illustrated in this container and reproduced below in Figures 1 through 3.
Model Parameters for the Clive DU PA Model
25 November 2015 6
Figure 1. Decay chains modeled in the Clive DU PA Model, part 1 of 2.

Model Parameters for the Clive DU PA Model
25 November 2015 7
Figure 2. Decay chains modeled in the Clive DU PA Model, part 2 of 2.

Model Parameters for the Clive DU PA Model
25 November 2015 8
Figure 3. Details of the actinide decay chains modeled in the Clive DU PA Model, showing
which species are omitted, in gray.

4.2 \Materials\Loess_Properties
Since loess (windblown sediment) is derived from the surrounding Unit 4 surface soils, the
material properties for Loess are redirected to other containers. The particle density used for the
loess is the same particle density that is used for all materials derived from Unit 4, given in Table
8. Unit 4 compacted material properties. The bulk density is redirected to be the same as the
evapotranspiration layers (Table 9) because the loess is similar in that it is an uncompacted
material derived from Unit 4.

Model Parameters for the Clive DU PA Model
25 November 2015 9
4.3 \Materials\Unit4_Compacted_Properties
Unit 4 is a silty clay, the uppermost unit deposited in the region by ancestral lakes. Certain parts
of the engineered system are constructed using Unit 4 material which is subjected to compaction;
in its compacted form, Unit 4 has the properties listed in Table 8. The particle density in Table 8
is common to all materials derived from Unit 4 (including compacted engineered layers,
uncompacted evapotranspiration layers, and Aeolian deposition layers). All Unit 4 materials are
assigned Kd values for silt (see Section 4.18.2).
Table 8. Unit 4 compacted material properties
GoldSim element value or distribution units reference / comment
ParticleDensity_Unit4
particle density of Unit 4
material
2.65 g/cm3 see Unsaturated Zone Modeling
white paper
Porosity_Unit4Compa
cted
porosity of Unit 4
material
N( µ=0.428, σ=9.08e-3,
min=Small, max=1-Small )
— ibid., truncated just above 0 and
just below 1
BulkDensity_Unit4Co
mpacted
dry bulk density of
Unit 4 material
N( µ =f(x), σ=0.1,
min=Small, max=Large )
g/cm3 ibid., truncated just above 0
D_Unit4Compacted
Brooks-Corey fractal
dimension parameter for
Unit 4 material
N( µ=2.81, σ=9.93e-5,
min=0, max=3 )
— ibid., truncated at 0 and 3
Hb_Unit4Compacted
bubbling pressure head
of Unit 4 material
N( µ=104., σ=1.72,
min=Small, max=Large )
correlated to D_Unit4
as -0.66
cm ibid., truncated just above 0
MCres_Unit4Compact
ed
residual moisture content
for Unit 4 material
N( µ=0.108, σ=8.95e-4,
min=Small, max=Large )
— ibid., truncated just above 0

4.4 \Materials\Unit4_ETLayers_Properties
The properties defined in this container are used for materials derived from Unit 4 that have not
been subjected to compaction.

Model Parameters for the Clive DU PA Model
25 November 2015 10
Table 9. Unit 4 ET Layers material properties
GoldSim element value or distribution units reference / comment
Porosity_Unit4ETLaye
rs
porosity of Unit 4
material
N( µ=0.481, σ=0.015,
min=Small, max=1-Small )
— see Unsaturated Zone Modeling
white paper, truncated just
above 0 and just below 1
BulkDensity_Unit4ETL
ayers
dry bulk density of
Unit 4 material
N( µ =f(x), σ=0.1,
min=Small, max=Large )
g/cm3 ibid., truncated just above 0
log_vG_Alpha N( µ= -1.79, σ=0.121,
min= -Large, max=0 )
— ibid., truncated at 0
log_vG_n N( µ=0.121, σ=0.019,
min=0, max=Large )
— ibid., truncated at 0

4.5 \Materials\Unit3_Properties
Material properties for the unsaturated zone below the liner of the disposal embankment,
comprised of stratigraphic Unit 3, a silty sand, are provided in Table 10. Unit 3 is assigned Kd
values for sand.
Table 10. Unit 3 material properties
GoldSim element value or distribution units reference / comment
ParticleDensity_Unit3
particle density of
Unit 3 material
2.65 g/cm3 see Unsaturated Zone
Modeling white paper
Porosity_Unit3
porosity of Unit 3
material
N( µ=0.393, σ=6.11e-3,
min=Small, max=1-Small )
— ibid., truncated just above 0
and just below 1
BulkDensity_Unit3
dry bulk density of
Unit 3 material
N( µ =f(x), σ=0.1,
min=Small, max=Large )
g/cm3 ibid., truncated just above 0
D_Unit3
Brooks-Corey fractal
dimension parameter for
Unit 3 material
N( µ=2.73, σ=5.21e-3,
min=0, max=3 )
— ibid., truncated at 0 and 3
Hb_Unit3
bubbling pressure head
of Unit 3 material
N( µ=8.85, σ=0.929,
min=Small, max=Large );
[correlated to D_Unit3 -0.85]
cm ibid., truncated just above 0
Model Parameters for the Clive DU PA Model
25 November 2015 11
GoldSim element value or distribution units reference / comment
MCres_Unit3
residual moisture
content for Unit 3
material
N( µ=6.78e-3, σ=2.05e-3,
min=Small, max=Large )
— ibid.
Ksat_Unit3
saturated hydraulic
condictivity for Unit 3
material
N( µ=5.14e-5, σ=5.95e-6,
min=Small, max=Large);
[correlated to D_Unit3 -0.98]
cm/s ibid., truncated just above 0

4.6 \Materials\Unit2_Properties
Material properties for the saturated zone, comprised of stratigraphic Unit 2, a silty clay, are
provided in Table 11. Unit 2 is assigned Kd values for clay.
Table 11. Unit 2 material properties
GoldSim element value or distribution units reference / comment
BulkDensity_Unit2
dry bulk density for
Unit 2 material
N( µ=1.57, σ=0.05,
min=Small, max=Large )
g/cm3 see Saturated Zone Modeling
white paper
truncated just above 0
Porosity_Unit2
porosity for Unit 2
material
N( µ=0.29, σ=0.05,
min=Small, max=1-Small )
— ibid., truncated just above 0
and just below 1
Ksat_Unit2
saturated hydraulic
conductivity for Unit 2
N( µ=9.6e-4, σ=9.67e-5,
min=Small, max=Large )
cm/s ibid., truncated just above 0

4.7 \Materials\RipRap_Properties
Rip Rap was used to construct the uppermost layer: Armor. It quickly becomes infilled with
Loess. The Rip Rap itself is assumed to be an inert material. It is not used in Model v1.4, but it
is left in the model for now.
Table 12. Rip rap material properties
GoldSim element value or distribution units reference / comment
ParticleDensity_RipRap 2.65 g/cm3 see Unsaturated Zone Modeling
white paper
BulkDensity_RipRap N( µ=f(x), σ=0.1,
min=Small,
max=Large )
g/cm3 ibid., truncated just above 0
Model Parameters for the Clive DU PA Model
25 November 2015 12
GoldSim element value or distribution units reference / comment
Porosity_RipRap N( µ=0.18, σ=0.01,
min=Small,
max=1-Small )
— ibid., truncated just above 0 and
just below 1

4.8 Materials\FineCobbleMix_Properties
Fine Cobble Mix is used to construct the upper filter layer in the Model v1.0. It also becomes
quickly infilled with Loess. The Fine Cobble Mix itself is assumed to be an inert material. It is
not used in Model v1.4, but it is left in the model for now.
Table 13. Fine cobble mix material properties
GoldSim element value or distribution units reference / comment
ParticleDensity_
FineCobbleMix
2.65 g/cm3 see Unsaturated Zone
Modeling white paper
BulkDensity_
FineCobbleMix
N( µ=f(x), σ=0.1,
min=Small, max=Large )
g/cm3 ibid., truncated just above 0
Porosity_
FineCobbleMix
N( µ=0.19, 0.01, min=Small,
max=1-Small)
— ibid., truncated just above 0
and just below 1

4.9 Materials\SiltSandGravel_Properties
Silt Sand Gravel is used to construct the Sacrificial Soil layer in Model v1.0 and the Frost
Protection Layer in Model v1.2 and Model v1.4.
Table 14. Silt sand gravel material properties
GoldSim element value or distribution units reference / comment
ParticleDensity_
SiltSandGravel
2.65 g/cm3 see Unsaturated Zone
Modeling white paper
BulkDensity_
SiltSandGravel
N( µ=f(x), σ=0.1, min=Small,
max=Large )
g/cm3 ibid., truncated just above 0
Porosity_
SiltSandGravel
N( µ=0.41, 0.0026,
min=Small, max=1-Small)
— ibid., truncated just above 0
and just below 1

4.10 Materials\FineGravelMix_Properties
Fine Gravel Mix is used to construct the lower filter layer in Model v1.0. The Fine Gravel Mix
itself is assumed to be an inert material. It is used as an inert filler material in the Surface Layer
of Model v1.2 and Model v1.4.
Model Parameters for the Clive DU PA Model
25 November 2015 13
Table 15. Fine gravel mix material properties
GoldSim element value or distribution units reference / comment
ParticleDensity_
FineGravelMix
2.65 g/cm3 see Unsaturated Zone
Modeling white paper
BulkDensity_
FineGravelMix
N( µ=f(x), σ=0.01,
min=Small, max=Large )
g/cm3 ibid., truncated just above 0
Porosity_
FineGravelMix
N( µ=0.28, 0.01, min=Small,
max=1-Small)
— ibid., truncated just above 0
and just below 1

4.11 Materials\UpperRnBarrierClay_Properties
The Radon Barrier layers are divided into upper and lower layers. Both are constructed of local
Unit 4 clay, compacted to different hydraulic conductivities. UpperRnBarrierClay represents the
upper of the two layers, and has significantly lower Ksat (see Table 16). Other material properties
for this material are redirected to those of compacted Unit 4 material (see Table 8).

Table 16. Upper radon barrier clay material properties
GoldSim element value or distribution units reference / comment
UpperRnBarrierKsat_As
Built
4e-3 cm/day see Unsaturated Zone Modeling
white paper
RnBarrierKsat_Natdist LN( 3.37, 3.23) cm/day Ibid., right shift of 0.00432 is added
after a value is pulled from the
distribution

4.12 Materials\LowerRnBarrierClay_Properties
The Lower Radon Barrier underlies the Upper Radon Barrier and is constructed of compacted
local Unit 4 clay, but has its own Ksat (see Table 17). The naturalized Ksat value is set equal to
that of the UpperRnBarrier_Clay_Properties container; see Unsaturated Zone Modeling white
paper. Other material properties for this material are set equal to those of compacted Unit 4
material (see Table 8).
Table 17. Lower radon barrier clay material properties
GoldSim element value or distribution units reference / comment
LowerRnBarrierKsat_
Asbuilt
8.6e-2 cm/day see Unsaturated Zone Modeling
white paper

Model Parameters for the Clive DU PA Model
25 November 2015 14
4.13 Materials\LinerClay_Properties
The Liner is constructed of compacted local Unit 4 clay, but has its own Ksat (see Table 18).
Other material properties for this material are redirected to those of compacted Unit 4 material
(see Table 8).
Table 18. Liner clay material properties
GoldSim element value or distribution units reference / comment
Ksat_LinerClay LN( GM=1e-6,
GSD=1.2 )
cm/s see Unsaturated Zone Modeling
white paper

4.14 \Materials\UO3_Waste_Properties
UO3 waste is typical of the Savannah River Site DU waste stream. Note, however, that given that
the DU-containing waste layer is overwhelmingly inert fill by volume, the material properties for
this layer as modeled are set to those of Unit 3 (see Table 10).
4.15 \Materials\Waste_U3O8_Properties
U3O8 waste is typical of the gaseous diffusion plant DU waste streams. Like the UO3 waste, the
material properties for this layer as modeled are set to those of Unit 3 (see Table 10).
4.16 \Materials\Generic_Waste_Properties
The current Clive DU PA Model has no generic waste inventory, but this material is defined as a
placeholder. Any layers to be filled with generic LLW borrow material properties from Unit 3
(see Table 10).
4.17 \Materials\Water_Properties
Water is the reference fluid in GoldSim.
Table 19. Properties of water, the reference fluid.
GoldSim element value or
distribution
units reference / comment
RefDiffusivity_Water
reference diffusivity in Water
1 × 10–9 m2/s as given in the GoldSim manual
Dm
molecular diffusivity in Water
U( 3e-6, 2e-5 ) cm2/s see the Geochemical Modeling white
paper

Model Parameters for the Clive DU PA Model
25 November 2015 15
4.18 \Materials\Kd
Since the Kd distribution for each element and each material can be defined independently, with a
different distributional form, the Model Parameters workbook does not lend itself to listing these
as a vector. Instead, each chemical element is listed in the following tables, one table for each
material. Materials are limited to sand, silt, and clay, which spans the gross material properties
found at the site. Since the depleted uranium is assumed to be dispersed in a large volume of fill
material of as yet unspecified characteristics, the material properties of the disposed waste
generally assumes the properties of this fill material. For now, then, the uranium oxide wastes are
not assigned their own chemical properties.

4.18.1 \Materials\Kd\Kd_Sand_Values
Table 20. Soil/water partition coefficients (Kds) for sand
chemical
element
value or distribution units reference / comment
Ac LU( min=16.8, max=535 ) mL/g see Geochemical Modeling
white paper
Am LU( min=43.2, max=811 ) mL/g ibid.
Cs LU( min=2.70, max=22.2 ) mL/g ibid.
I_dist N( 0.428, 0.605 ), with values less
than 0 set to 0.
mL/g ibid.; Values sampled below 0
are set to 0, within the
Expression I.
Np LU( min=0.392, max=51 ) mL/g ibid.
Pa LU( min=8.32, max=331 ) mL/g ibid.
Pb LU( min=2.70, max=22.2 ) mL/g ibid.
Pu LU( min=66.9, max=2390 ) mL/g ibid.
Ra LU( min=0.387, max=64.6 ) mL/g ibid.
Rn 0 mL/g ibid.
Sr LU( min=2.7, max=22.2 ) mL/g ibid.
Tc_dist N( 0.102, 0.145 ), with values less
than 0 set to 0.
mL/g ibid.; Values sampled below 0
are set to 0, within the
Expression Tc.
Th LU( min=19.2, max=41.6 ) mL/g ibid.
U LU( min=0.344, max=6.77 ) mL/g ibid.

Model Parameters for the Clive DU PA Model
25 November 2015 16
4.18.2 \Materials\Kd\Kd_Silt_Values
Table 21. Soil/water partition coefficients (Kds) for silt
chemical
element
value or distribution units reference / comment
Ac LU( min=15.7, max=1910 ) mL/g see Geochemical Modeling
white paper
Am LU( min=88.0, max=1140 ) mL/g ibid.
Cs LU( min=4.23, max=118 ) mL/g ibid.
I Equal to Kd for I in Sand mL/g ibid.
Np LU( min=0.805, max=62.1 ) mL/g ibid.
Pa LU( min=184, max=978 ) mL/g ibid.
Pb LU( min=4.23, max=118 ) mL/g ibid.
Pu LU( min=80.5, max=6210 ) mL/g ibid.
Ra LU( min=0.797, max=75.3 ) mL/g ibid.
Rn 0 mL/g ibid.
Sr LU( min=4.23, max=118 ) mL/g ibid.
Tc Equal to Kd for Tc in Sand mL/g ibid.
Th LU( min=34.4, max=697 ) mL/g ibid.
U LU( min=0.880, max=11.4 ) mL/g ibid.

4.18.3 \Materials\Kd\Kd_Clay_Values
Table 22. Soil/water partition coefficients (Kds) for clay
chemical
element
value or distribution units reference / comment
Ac LU( min=83.6, max=2990 ) mL/g see Geochemical Modeling
white paper
Am LU( min=88.0, max=1140 ) mL/g ibid.
Cs LU( min=6.69, max=239 ) mL/g ibid.
I Equal to Kd for I in Sand mL/g ibid.
Np LU( min=4.32, max=81.1 ) mL/g ibid.
Pa LU( min=180, max=1560 ) mL/g ibid.
Pb LU( min=6.69, max=239 ) mL/g ibid.
Pu LU( min=914, max=5470 ) mL/g ibid.
Ra LU( min=1.42, max=1410 ) mL/g ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 17
chemical
element
value or distribution units reference / comment
Rn 0 mL/g ibid.
Sr LU( min=6.69, max=239 ) mL/g ibid.
Tc Equal to Kd for Tc in Sand mL/g ibid.
Th LU( min=84.7, max=2360 ) mL/g ibid.
U LU( min=9.05, max=66.3 ) mL/g ibid.

4.19 \Materials\WaterSolubility
Since the aqueous solubility distribution for each element and each material could be defined
independently, with a different distributional form, the Model Parameters workbook does not
lend itself to listing these as a vector. Instead, each chemical element is listed in the following
table.
4.19.1 \Materials\WaterSolubility\Solubilities_Saltwater

Table 23. Aqueous solubilities in saltwater, by chemical element
chemical
element
value or distribution units reference / comment
Ac LU( min=6.81e-9, max=1.47e-5 ) mol/L see Geochemical Modeling
white paper
Am LU( min=6.81e-10, max=1.47e-6 ) mol/L ibid.
Cs LU( min=6.81e-3, max=1.47e1 ) mol/L ibid.
I LU( min=5.99e-5, max=1.67e0 ) mol/L ibid.
Np LU( min=6.81e-6, max=1.47e-2 ) mol/L ibid.
Pa LU( min=6.81e-9, max=1.47e-5 ) mol/L ibid.
Pb LU( min=6.81e-9, max=1.47e-5 ) mol/L ibid.
Pu LU( min=5.27e-11, max=1.90e-5 ) mol/L ibid.
Ra LU( min=5.99e-10, max=1.67e-5 ) mol/L ibid.
Rn LU( min=7.74e-4, max=1.29e-1 ) mol/L ibid.
Sr LU( min=6.81e-7, max=1.47e-3 ) mol/L ibid.
Tc LU( min=7.74e-5, max=1.29e-2 ) mol/L ibid.
Th LU( min=7.74e-9, max=1.29e-6 ) mol/L ibid.
UO3 LU( min=3.58e-6, max=2.79e-3 ) mol/L ibid.
U3O8 LU( min=1e-16, max=6.5e-10 ) mol/L ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 18
4.20 \Materials\Air_Properties
Currently, the only gaseous radionuclide in the model is 222Rn, which diffuses in the air phase.
Table 24. Parameters relevant to diffusion in air.
GoldSim element value or
distribution
units reference / comment
RefDiffusivity_Air 1 cm2/s arbitrary value in GoldSim, as it falls out in
math
Da_Rn 0.11 cm2/s see Radon Modeling white paper
SoilTemp
average soil
temperature
N( µ=12, σ=1 ) °C Estimated from the Clive Test Cell
temperature data “Temp and Dose Data
9-19-01 to 1-15-09.xls” provided by
EnergySolutions.
Khcp_Rn
parameter used in
devising Henry’s Law
constant
9.3e-3 mol/L·atm Sander (1999), table 7, page 13

5.0 \Processes
Physical process parameters global in scope (general to the entire model) are defined in this
container.
5.1 \Processes\AirTransport
Contaminant transport in air includes both pore air in porous media, and the dispersion into and
within the atmosphere. Chi/Q values for gas and particles that are specific to the Federal DU
embankment are listed in Table 43(for the \Disposal\AtmosphericDispersion\AirConc_Remote
container).
Table 25. Radon diffusive transport parameters.
GoldSim element value or distribution units reference / comment
EPRatio_Radon
radon escape/production
ratio
beta( 0.290, 0.156,
min=0, max=1 )
— see Radon Modeling white paper

ThicknessAtm
mixing thickness of the
atmosphere, for
purposes of diffusion
from soil layers
N( µ=2.0, σ=0.5,
min=Small,
max=Large )
m see Unsaturated Zone Modeling
white paper

Model Parameters for the Clive DU PA Model
25 November 2015 19
GoldSim element value or distribution units reference / comment
WindSpeed
average wind speed, for
purposes of diffusion
from soil layers
N( µ=3.14, σ=0.5,
min=Small,
max=Large )
m/s ibid.
AtmDiffusionLength
diffusion length for the
atmosphere, for
purposes of diffusion
from soil layers
N( µ=0.1, σ=0.02,
min=Small,
max=Large )
m ibid.

Table 26. Atmospheric transport parameters.
GoldSim element value or distribution units reference / comment
Dust_mask
logical mask to identify
PM-10 particles
Rn = 0, all others = 1
(see workbook)
— masks Species with 0/1 to be those
found in dust particles
Gas_mask
logical mask to identify
gases
Rn = 1, all others = 0
(see workbook)
— masks Species with 0/1 to be those
found in gaseous phase
ResuspensionFlux
mass flux of soil
particles into
atmosphere
LU( Small, 0.3 ) kg/m2-yr see Atmospheric Modeling white
paper
Particle_Fraction
the fraction of PM-10
particles in the 0 to
2.5 µm size bin
U(0,1) — based on physical limits
Frac_OffSite_
Deposition
fraction of all particles
that migrate off site that
are deposited in the off-
site air dispersion area.
a lookup table based on
Particle_Fraction
0 0.11 — see Atmospheric Modeling white
paper 0.05 0.11 —
0.1 0.11 —
0.2 0.099 —
0.4 0.086 —
0.6 0.072 —
0.8 0.057 —
1.0 0.041 —
OnSiteRedepos_
Ratio_bySize
a lookup table based on
0 4.224e-7 g/m2-yr
per g/yr
ibid.
0.05 4.114e-7
0.1 4.002e-7
Model Parameters for the Clive DU PA Model
25 November 2015 20
GoldSim element value or distribution units reference / comment
Particle_Fraction 0.2 3.776e-7
0.4 3.311e-7
0.6 2.827e-7
0.8 2.321e-7
1.0 1.794e-7

5.2 \Processes\AnimalTransport
Burrowing animals have the potential to exhume waste or contaminated cap materials. All
burrowers are collected into one of two types: ants and small mammals.
5.2.1 \Processes\AnimalTransport\AntData
Table 27. Model parameters for ants.
GoldSim element value or distribution units reference / comment
NestVolume
volume of each nest
N( µ=0.161, σ=0.024, min=0,
max=Large )
m3 see Biological Modeling
white paper
ColonyLifespan
lifespan of each
colony
N( µ=20.2, σ=3.6, min=Small,
max=Large )
yr ibid.
ColonyDensity
area density of
colonies on the
ground
see below for each field study plot 1/ha ibid.
_Plot1 Gamma( 33,1, min=0, max=Large ) 1/ha ibid.
_Plot2 Gamma( 2, 1, min=0, max=Large ) 1/ha ibid.
_Plot3 Gamma( 7, 1, min=0, max=Large ) 1/ha ibid.
_Plot4 Gamma( 17, 1, min=0, max=Large ) 1/ha ibid.
_Plot5 Gamma( 6, 1, min=0, max=Large ) 1/ha ibid.
MaxDepth
maximum depth for
any colony
212 cm ibid.
b
fitting parameter for
nest shape
N( µ=10, σ=0.71, min=1,
max=Large )
— ibid.

Model Parameters for the Clive DU PA Model
25 November 2015 21
5.2.2 \Processes\AnimalTransport\MammalData
Table 28. Model parameters for small mammals.
GoldSim
element
value or distribution units reference / comment
MoundDensity
area density of
mounds on the
ground

see below for each plot 1/ha see Biological Modeling
white paper
_Plot1 Gamma( 235, 1, min=0, max=Large ) 1/ha ibid.
_Plot2 Gamma( 239, 1, min=0, max=Large ) 1/ha ibid.
_Plot3 Gamma( 1.33, 1, min=0, max=Large ) 1/ha ibid.
_Plot4 Gamma( 1.33, 1, min=0, max=Large ) 1/ha ibid.
_Plot5 Gamma ( 1.33, 1, min=0, max=Large ) 1/ha ibid.
ExcavationRate
volumetric rate of
a single burrow
excavation
N( µ=0.0006, σ=0.00015, min=Small,
max=Large )
m3/yr ibid.
MaxDepth
maximum depth
for any burrow
200 cm ibid.
b
fitting parameter
for burrow shape
N( µ=4.5, σ=0.84, min=1, max=Large ) — ibid.

5.3 \Processes\PlantTransport
Plants have the potential to translocate contaminants in waste or contaminated cap materials. All
plants are collected into one of five types: greasewood, grasses, forbs, trees, and shrubs. Each of
these plant types is characterized in each of the five plot locations that were studied,
corresponding to five vegetation associations:
• Plot 1: Mixed Grassland
• Plot 2: Juniper - Sagebrush
• Plot 3: Black Greasewood
• Plot 4: Halogeton - Disturbed
• Plot 5: Shadscale - Gray Molly

Each of these vegetation associations is picked at random for a given realization.
Model Parameters for the Clive DU PA Model
25 November 2015 22
Table 29. Parameters general to all plants.
GoldSim element value or
distribution
units reference / comment
BiomassProductionRate U( 300, 1500 ) kg/ha-yr see Biological Modeling white paper
VegetationAssociationPicker discrete( 1, 2,
3, 4, 5 )
— ibid.

5.3.1 \Processes\PlantTransport\PlantCR
Table 30. Plant/soil concentration ratio parameters.
GoldSim element value or
distribution
units reference / comment
CR_GM tabulated in
Clive PA Model
Parameters.xls
workbook
— see Biological Modeling white paper
CR_GSD tabulated in
Clive PA Model
Parameters.xls
workbook
— ibid.
CR_GM_radon Small — ibid.
CR_GSD_radon 1 — ibid.

5.3.2 \Processes\PlantTransport\BiomassCalcs
Table 31. Biomass calculation parameters.
GoldSim element value or
distribution
units reference / comment
percent cover tables, such as
PctCover_Plot4_Forb
tabulated in
Clive PA Model
Parameters.xls
workbook
% These are 25 tables, one for each
Plot and for each plant type. Source:
plant.cover.percent.simulations.xlsx
in Clive PA Model Parameters.xls
workbook
PctCoverRandomSelector probability of
0.001 assigned
to discrete
values from 1
to 1000
% An index generator used to pick
correlated sets of percent cover

Model Parameters for the Clive DU PA Model
25 November 2015 23
5.3.3 \Processes\PlantTransport\GreasewoodData
Table 32. Greasewood parameters.
GoldSim element value or distribution units reference / comment
RootShoot_Ratio U( 0.30, 1.24 ) — see Biological Modeling white paper
MaxDepth 570 cm ibid.
b N( µ=14.6, σ=0.0807,
min=1, max=Large )
— ibid.

5.3.4 \Processes\PlantTransport\GrassData
Table 33. Grass parameters.
GoldSim element value or distribution units reference / comment
RootShoot_Ratio T( 1, 1.2, 2 ) — see Biological Modeling white paper
MaxDepth 150 cm ibid.
b N( µ=2.19 σ=0.036,
min=1, max=Large )
— ibid.

5.3.5 \Processes\PlantTransport\ForbData
Table 34. Forb parameters.
GoldSim element value or distribution units reference / comment
RootShoot_Ratio U( 0.40, 1.80 ) — see Biological Modeling white paper
MaxDepth 51 cm ibid.
b N( µ=23.9 σ=0.313,
min=1, max=Large )
— ibid.

5.3.6 \Processes\PlantTransport\TreeData
Table 35. Tree parameters.
GoldSim element value or distribution units reference / comment
RootShoot_Ratio U( 0.55, 0.76 ) — see Biological Modeling white paper
MaxDepth 450 cm ibid.
b N( µ=14.6 σ=0.0807,
min=1, max=Large )
— ibid.

Model Parameters for the Clive DU PA Model
25 November 2015 24
5.3.7 \Processes\PlantTransport\ShrubData
Table 36. Other shrub parameters.
GoldSim element value or distribution units reference / comment
RootShoot_Ratio U( 0.4, 1.8 ) — see Biological Modeling white paper
MaxDepth 110 cm ibid.
b N( µ=23.9 σ=0.313,
min=1, max=Large )
— ibid.

5.4 \Processes\WaterTransport
Flow within moving water (advection) and diffusion within water are typically significant
contaminant transport mechanisms. Global parameters for water transport are located here. Other
parameters specific to a modeled column are located within that column’s modeling container
(e.g. Section Table 47) or that material property’s modeling container (e.g Section 4.11, Table
16).
Table 37. Water transport parameters.
GoldSim element value or
distribution
units reference / comment
Water tortuosity water content
exponent
N( µ=7/3, σ=0.01) — Ibid.
Water tortuosity porosity
exponent
N( µ=2.0, σ=0.01 — Ibid.

5.5 \Processes\ErosionTransport
Erosion through the formation of gullies can be a significant mechanism for exposing waste to
the environment. Global parameters for erosion are located here. Other parameters specific to an
embankment are located within that embankment’s modeling container (e.g. Section
\Disposal\FederalDUCell\ErosionCalcs\SiberiaErosionCalcs
SIBERIA modeling results were used to create 1000 realization inputs for gully density for each
modeling cell layer.
Table 62. SIBERIA erosion parameters.
GoldSim element value or distribution units reference / comment
FractionGully Lookup table with 1000
realizations
— See Erosion Modeling white
paper.
Model Parameters for the Clive DU PA Model
25 November 2015 25
GoldSim element value or distribution units reference / comment
GullyRandomSelector discrete( 1, ..., 1000 )
equal probability (0.001)
ibid.

). These parameters are for the initial gully screening calculations from Model v.1.0. Those
screening calculations are still in the model, although they are not used in the dose assessment.
Table 38. Water transport parameters.
GoldSim element value or
distribution
units reference / comment
AngleOfRepose_Gully
angle of repose for gully walls
N( µ=38, σ=5,
min=Small,
max=90-Small )
degrees see Erosion Modeling white paper
Gully_b_parameter
shape parameter for gully
thalweg
N( µ=-0.4,
σ=0.15,
min=-0.75,
max=-0.05 )
— ibid.

6.0 \Inventory
The DU waste is characterized by analysis of the SRS DU. To date, insufficient information
exists to thoroughly characterize the DU wastes expected to arrive from the gaseous diffusion
plants (GDPs).
6.1 \Inventory\SRS_DU_Inventory
The SRS DU, which consists of several thousand 208-L (55-gal) drums of powdered DUO3, has
been subjected to laboratory analysis, so activity concentrations are based on that information.
Table 39. SRS DU inventory parameters.
GoldSim element value or
distribution
units reference / comment
ActivityConc_DUWaste_Mean See parameters
workbook, sheet
“Inventory”
pCi/g see Waste Inventory white paper
ActivityConc_DUWaste_StdDev See parameters
workbook, sheet
“Inventory”
pCi/g see Waste Inventory white paper
SRS_DU_Drums_Disposed 21000 — (not considered in this PA)
SRS_DU_Drums_ProposedUT 5408 — see Waste Inventory white paper
Model Parameters for the Clive DU PA Model
25 November 2015 26
GoldSim element value or
distribution
units reference / comment
SRS_DU_Drums_ProposedEW 5408 × 2 — (not considered in this PA)
Drum_Mass 20 kg see Waste Inventory white paper
ShippedMass_Proposed_UT 3577 Mg see Waste Inventory white paper

6.2 \Inventory\GDP_DU_Inventory
Since insufficient information exists to exactly characterize the DU wastes expected to arrive
from the GDPs, the activity concentrations and other waste material characteristics are borrowed
from the SRS DUO3 waste, as a proxy.
Table 40. GDP DU inventory parameters.
GoldSim element value or
distribution
units reference / comment
Num_DUF6_Cylinders_PGDP 36191 — see Waste Inventory white paper
Num_DUF6_Cylinders_PORTS 16109 — ibid.
Num_DUF6_Cylinders_K25 4822 — ibid.
Mass_DUF6_PGDP 436400 Mg ibid.
Mass_DUF6_PORTS 195800 Mg ibid.
Mass_DUF6_K25 54300 Mg ibid.
CylinderDiameter 4 ft ibid.
CylinderLength 12 ft ibid.
Num_CylindersDisposed 48628 --- ibid.
FractionGDP_Contaminated Beta( 0.0392,
0.0025, 0, 1 )
— ibid.
CleanDU_Mask see workbook — simply a mask for uranium

6.3 \Inventory\Other_DU_Inventory
This is a placeholder container. No other DU inventory is assumed in the model.
6.4 \Inventory\ClassA_LLW_Inventory
This is a placeholder container. No other LLW inventory is assumed in the model.
Model Parameters for the Clive DU PA Model
25 November 2015 27
7.0 \Disposal
The Disposal container hosts all the actual contaminant calculations, including atmospheric
transport, transport mechanisms within each column of each embankment (water, air, biological,
etc.) and the saturated zone. While global transport parameters are defined in the \Processes
container (Section ), parameters and calculations specific to local mechanisms are defined here.
7.1 \Disposal\AtmosphericDispersion
The values for the ratio of airborne contaminant concentration to source release rate into the
atmosphere are known as Χ/Q (Chi/Q) values. These are implemented as lookup tables on
Particle_Fraction.
7.1.1 \Disposal\AtmosphericDispersion\AirConc_Onsite
OnSite air concentrations are used for exposures to receptors that traverse the embankment itself.
Table 41. Atmosphere dispersion parameters for on-site exposures.
GoldSim element value or
distribution
units reference / comment
ChiQ_Embankment
_538m
0 222 (µg/m3)/(g/s) see Atmospheric Modeling white
paper 0.05 223
0.1 224
0.2 225
0.4 228
0.6 231
0.8 234
1.0 238
ChiQ_Gas_Onsite
(Embankment)
234 same ibid.

7.1.2 \Disposal\AtmosphericDispersion\MediaConc_Offsite
OffSite air concentrations are used for exposures to receptors that traverse the area surrounding
the embankment. These receptors also have access to the embankment itself. Functionally, the air
concentrations are set to those same values used for OnSite air.
Model Parameters for the Clive DU PA Model
25 November 2015 28
Table 42. Atmosphere dispersion parameters for off-site exposures (in the “air dispersion”
area.)
GoldSim element value or
distribution
units reference / comment
ChiQ_Dust_Offsite set equal to
ChiQ_Dust
_Onsite
(µg/m3)/(g/s) see Atmospheric Modeling white
paper
ChiQ_Gas_Offsite 0.38 same ibid.

7.1.3 \Disposal\AtmosphericDispersion\AirConc_Remote
Various receptors are at specific geographic locations farther away from the site, including
Interstate-80, the rail road, the Grassy Rest Area, the Knolls OHV Recreation Area, and the
UTTR access road.
Table 43. Atmosphere dispersion parameters for remote off-site exposures.
GoldSim element value or
distribution
units reference / comment
ChiQ_RestArea_1K 0 0.0069 (µg/m3)/(g/s) see Atmospheric Modeling
white paper 0.05 0.0069
0.1 0.0069
0.2 0.0070
0.4 0.0071
0.6 0.0072
0.8 0.0073
1.0 0.0074
ChiQ_Gas_RestArea 0.0088 same ibid.
ChiQ_Knolls 0 0.043 same ibid.
0.05 0.044
0.1 0.044
0.2 0.046
0.4 0.049
0.6 0.052
0.8 0.055
1.0 0.058
ChiQ_Gas_Knolls 0.053 same ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 29
GoldSim element value or
distribution
units reference / comment
ChiQ_I80_1K 0 0.26 same ibid.
0.05 0.26
0.1 0.26
0.2 0.27
0.4 0.27
0.6 0.28
0.8 0.28
1.0 0.28
ChiQ_Gas_I80 0.28 same ibid.
ChiQ_Railroad_1K 0 0.43 same ibid.
0.05 0.43
0.1 0.43
0.2 0.43
0.4 0.43
0.6 0.44
0.8 0.44
1.0 0.44
ChiQ_Gas_Railroad 0.44 same ibid.
ChiQ_UTTRaccess_1K 0 222 same ibid.
0.05 223
0.1 224
0.2 225
0.4 228
0.6 231
0.8 234
1.0 238
ChiQ_Gas_UTTRaccess 234 same ibid.

7.2 \Disposal\FederalDUCell
This PA model considers only the Federal DU cell, part of the Federal Waste Cell embankment.
Model Parameters for the Clive DU PA Model
25 November 2015 30
7.2.1 \Disposal\FederalDUCell\FederalDU_Cell_Dimensions
Exact dimensions of the embankment are somewhat irregular, so the shape of the cell has been
somewhat idealized to facilitate calculations. See Embankment Modeling white paper for details.
Table 44. Interior (waste) dimensions of the Federal Cell, Federal DU section.
GoldSim element value or
distribution
units reference / comment
OriginalGrade
Average original grade elevation
4272 ft
amsl
see Embankment Modeling.pdf
WasteTopHeight_Ridge
Height of top of the waste at the
ridgeline
47.5 ft ibid.
WasteTopHeight_Break
Height of top of the waste at the
break in slope
35 ft ibid.
WasteBottomElev
Elevation of the bottom of the
waste
4264 ft
amsl
ibid.
LengthOverall
Length overall
1318 ft ibid.
WidthOverall
Width overall
1775 ft ibid.
LengthToBreak
Length from edge to the break in
slope
175 ft ibid.
WidthToBreak
Width from edge to the break in
slope
175 ft ibid.
BreakToRidge_Width
With from break in slope to the
ridge
521 ft ibid.
BreakToRidge_Length_West
Width from break in slope to the
ridge, west side of cell
521 ft ibid
BreakToRidge_Length_East
Length from break in slope to the
ridge, east side of cell
447 ft

Model Parameters for the Clive DU PA Model
25 November 2015 31
7.2.2 \Disposal\FederalDUCell\NaturalSystemGeometry
Table 45. Natural system geometry parameters for the Federal DU cell.
GoldSim element value or distribution units reference / comment
UZ_Thickness
thickness of the
unsaturated zone
below the FDU cell
N( µ=12.9, σ=0.25,
min=Small, max=Large )
ft see Unsaturated Zone Modeling
white paper

7.2.3 \Disposal\FederalDUCell\CapCell_Thickness
Table 46. Dimensions of the cap cells for the Federal DU cell.
GoldSim element value or distribution units reference / comment
TSurface
Total thickness of the
surface soil layer
6 in see Embankment Modeling.pdf
TEvap
Total thickness of
the evaporative
zone
12 in ibid.
TFrostProt
Total thickness of
frost protection layer
18 in ibid.
TUpperRadon
Total thickness of
upper radon barrier
12 in ibid.
TLowerRadon
Total thickness of
lower radon barrier
12 in ibid.
TopCell_Thickness 1 cm ibid.

7.2.4 \Disposal\FederalDUCell\TopSlope
No input elements are defined at this level.
7.2.4.1 \Disposal\FederalDUCell\TopSlope\Column_Transport
No input elements are defined at this level.
Model Parameters for the Clive DU PA Model
25 November 2015 32
7.2.4.1.1 \Disposal\FederalDUCell\TopSlope\Column_Transport \WaterTransport
Water flow calculations for the top slope column are performed here. These parameters are
clones for the side slope.
Table 47. Infiltration parameters for cap cells.
GoldSim element value or distribution units reference / comment
B_0
Regression parameter
-0.32921 — see Unsaturated Zone
Modeling white paper,
units of mm/yr are added
after the regression is
calculated
B_2
Regression parameter
5.56826 — ibid.
B_3
Regression parameter
0.19538 — ibid.

7.2.4.2 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
7.2.4.2.1 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile\WaterConte
ntCalcs_ETCover
These elements are cloned in the corresponding side slope container.
Table 48. Parameters for moisture profile calculations for the ET Cover.
GoldSim element value units reference / comment
B_0
Regression parameter vector SurfaceSoil 0.48155
EvapLayer 0.57947
FrostLayer 0.04282
UpperRnBarrier 0.14737
LowerRnBarrier 0.14740

see Unsaturated Zone
Modeling white paper
B_1
Regression parameter vector SurfaceSoil 0.00000
EvapLayer 0.00000
FrostLayer 0.00000
UpperRnBarrier -0.00076
LowerRnBarrier -0.00076

day/cm Ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 33
GoldSim element value units reference / comment
B_2
Regression parameter vector SurfaceSoil 0.54920
EvapLayer 0.73997
FrostLayer 0.43297
UpperRnBarrier 1.70702
LowerRnBarrier 1.70648

Ibid.
B_3
Regression parameter vector SurfaceSoil -0.20020
EvapLayer -0.24790
FrostLayer 0.01617
UpperRnBarrier 0.06353
LowerRnBarrier 0.06351

Ibid.
WaterContentResidual
SurfaceSoil 0.11
EvapLayer 0.11
FrostLayer 0.065
UpperRnBarrier 0.1
LowerRnBarrier 0.1

Ibid.

7.2.4.2.2 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_RnBarrier
These calculations are no longer used in the model. They are currently present for reference
only.
Table 49. Parameters for moisture profile calculations for the radon barrier.
GoldSim element value units reference / comment
NumNodes 5 this is the number of modeled radon barrier
layers +1
UpperRn_NodeNumber 2 middle node in part of column
LowerRn_NodeNumber 4 middle node in part of column

7.2.4.2.3 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_Waste
Table 50. Parameters for moisture profile calculations for the waste.
GoldSim element value units reference / comment
NumNodes 28 — this is the number of modeled waste layers +1

Model Parameters for the Clive DU PA Model
25 November 2015 34
7.2.4.2.4 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_Liner
Table 51. Parameters for moisture profile calculations for the clay liner.
GoldSim element value units reference / comment
NumNodes 5 this is the number of modeled liner layers +1
MiddepthNodeNumber 3 middle node in column

7.2.4.2.5 \Disposal\FederalDUCell\TopSlope\Column_MoistureProfile
\WaterContentCalcs_Unsat
Table 52. Parameters for moisture profile calculations for the unsaturated zone below the
clay liner.
GoldSim element value or
distribution
units reference / comment
NumNodes 24 see Unsaturated Zone Modeling
white paper
ZoneThickness
specified from the bottom up
-0.0204
-0.0204
-0.0204
-0.0204
-0.0204
-0.0510
-0.0510
-0.0510
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
-0.2550
0

m ibid.
MiddepthNodeNumber 16 middle node in column

Model Parameters for the Clive DU PA Model
25 November 2015 35
7.2.4.3 \Disposal\FederalDUCell\TopSlope\Cap_Layers
7.2.4.3.1 \Disposal\FederalDUCell\TopSlope\CapLayers\CapCell_Dimensions
Table 53. Cap layering dimensions for the top slope.
GoldSim element value or distribution units reference / comment
TArmor
Type B rip rap thickness
18 in see Embankment Modeling white
paper
TUpperFilter
Type A filter zone
thickness
6 in ibid.
TSacrificialSoil
Sacrificial soil thickness
12 in ibid.
TLowerFilter
Type B filter zone
thickness
6 in ibid.
TUpperRadon
upper radon barrier clay
thickness
12 in ibid.
TLowerRadon
lower radon barrier clay
thickness
12 in ibid.
NArmorCells 3 — modeling construct
NUpperFilterCells 1 — modeling construct
NSacrificialSoilCells 2 — modeling construct
NLowerFilterCells 1 — modeling construct
NUpperRadonCells 2 — modeling construct
NLowerRadonCells 2 — modeling construct
TopCell_Thickness U( 1 cm, TArmor –
NArmorCells × 1 cm )
cm modeling construct
This allows the thickness of the
topmost cell to vary between 1 cm
and the maximum so that the other
cells in this layer are at least 1 cm.

7.2.4.4 \Disposal\FederalDUCell\TopSlope\Liner
Table 54. Number of liner cells.
GoldSim element value or
distribution
units reference / comment
Model Parameters for the Clive DU PA Model
25 November 2015 36
GoldSim element value or
distribution
units reference / comment
NumLinerCells 4 — modeling construct

7.2.4.5 \Disposal\FederalDUCell\TopSlope\UnsatLayer
Table 55. Number of unsaturated zone cells.
GoldSim element value or
distribution
units reference / comment
NumUnsatCells 10 — modeling construct

7.2.4.6 \Disposal\FederalDUCell\TopSlope\WasteLayers
No input elements are defined at this level.
7.2.4.6.1 \Disposal\FederalDUCell\TopSlope\WasteLayers\ WasteCell_Dimensions
Table 56. Top slope waste cell dimensions.
GoldSim element value or
distribution
units reference / comment
NumWasteCells_TS 27 — modeling construct

7.2.5 \Disposal\FederalDUCell\SideSlope
No input elements are defined at this level.
7.2.5.1 \Disposal\FederalDUCell\SideSlope\Column_Transport
No input elements are defined at this level.
7.2.5.1.1 \Disposal\FederalDUCell\SideSlope\Column_Transport \WaterTransport
No input elements are defined at this level.
Model Parameters for the Clive DU PA Model
25 November 2015 37
7.2.5.2 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
7.2.5.2.1 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_RnBarrier
Table 57. Parameters for moisture profile calculations for the radon barrier.
GoldSim element value units reference / comment
NumNodes 5 this is the number of modeled radon barrier
layers +1
UpperRn_NodeNumber 2 middle node in part of column
LowerRn_NodeNumber 4 middle node in part of column

7.2.5.2.2 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_Waste
Table 58. Parameters for moisture profile calculations for the waste.
GoldSim element value units reference / comment
NumNodes 13 — this is the number of modeled waste layers +1

7.2.5.2.3 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_Liner
Table 59. Parameters for moisture profile calculations for the clay liner.
GoldSim element value units reference / comment
NumNodes 5 this is the number of modeled liner layers +1
MiddepthNodeNumber 3 middle node in column

7.2.5.2.4 \Disposal\FederalDUCell\SideSlope\Column_MoistureProfile
\WaterContentCalcs_Unsat
Parameters for moisture profile calculations for the unsaturated zone below the clay liner in the
side slope are identical to those for the top slope, as listed in Table 52.
Model Parameters for the Clive DU PA Model
25 November 2015 38
7.2.5.3 \Disposal\FederalDUCell\SideSlope\Cap_Layers
7.2.5.3.1 \Disposal\FederalDUCell\SideSlope\CapLayers\CapCell_Dimensions
Table 60. Cap layering dimensions for the side slope.
GoldSim element value or distribution units reference / comment
TArmor
Type A rip rap thickness
18 in see Embankment Modeling white
paper
TUpperFilter
Type A filter zone
thickness
6 in ibid.
TSacrificialSoil
Sacrificial soil thickness
12 in ibid.
TLowerFilter
Type B filter zone
thickness
18 in ibid. (Note how this is different from
the TopSlope value.)
TUpperRadon
upper radon barrier clay
thickness
12 in ibid.
TLowerRadon
lower radon barrier clay
thickness
12 in ibid.
NArmorCells 3 — modeling construct
NUpperFilterCells 1 — modeling construct
NSacrificialSoilCells 2 — modeling construct
NLowerFilterCells 1 — modeling construct
NUpperRadonCells 2 — modeling construct
NLowerRadonCells 2 — modeling construct
TopCell_Thickness U( 1 cm, TArmor –
NArmorCells × 1 cm )
cm modeling construct
This allows the thickness of the
topmost cell to vary between 1 cm
and the maximum so that the other
cells in this layer are at least 1 cm.

Model Parameters for the Clive DU PA Model
25 November 2015 39
7.2.5.4 \Disposal\FederalDUCell\SideSlope\Liner
Parameters in this section are identical to those defined for the Top Slope in Section .
7.2.5.5 \Disposal\FederalDUCell\SideSlope\UnsatLayer
Parameters in this section are identical to those defined for the Top Slope in Section .
7.2.5.6 \Disposal\FederalDUCell\SideSlope\WasteLayers
No input elements are defined at this level.
7.2.5.6.1 \Disposal\FederalDUCell\SideSlope\WasteLayers\
WasteCell_Dimensions
Table 61. Side slope waste cell dimensions.
GoldSim element value or distribution units reference / comment
NumWasteCells 12 — modeling construct

7.2.6 \Disposal\FederalDUCell\ErosionCalcs
The calculation of the volume, depth, and potential to expose waste by gullies is examined here.
This work includes the preliminary calculations, designed to evaluate whether more sophisticated
landform evolution modeling is warranted, as well as the more sophisticated erosion modeling
using SIBERIA, a landscape evolution model.
7.2.6.1 \Disposal\FederalDUCell\ErosionCalcs\SiberiaErosionCalcs
SIBERIA modeling results were used to create 1000 realization inputs for gully density for each
modeling cell layer.
Table 62. SIBERIA erosion parameters.
GoldSim element value or distribution units reference / comment
FractionGully Lookup table with 1000
realizations
— See Erosion Modeling white
paper.
GullyRandomSelector discrete( 1, ..., 1000 )
equal probability (0.001)
ibid.

7.3 \Disposal\SatZone
The saturated zone underlies and accepts recharge from all the embankments at the Clive
Facility. All contaminated recharge flows down-gradient to a monitoring well.
Model Parameters for the Clive DU PA Model
25 November 2015 40
7.3.1 \Disposal\SatZone\SatZone_Parameters
Table 63. Saturated zone parameters.
GoldSim element value or distribution units reference / comment
SZ_Thickness N( µ=16.2, σ=0.25,
min=0.1, max=Large )
ft see Saturated Zone
Modeling white paper
MonitoringWellDistance 90 ft ibid.
WaterTableGradient N( µ=6.94e-4, σ=1.27e-4,
min=0, max=Large )
— ibid.
7.3.2 \Disposal\SatZone\SZ_FederalDUFootprint
Table 64. Total number of cells in the saturated footprint zone.
GoldSim element value or
distribution
units reference / comment
NumCells_Footprint 25 modeling construct

7.3.2.1 \Disposal\SatZone\SZ_FederalDUFootprint\Waste_to_Footprint
Table 65. Total number of cells in both footprint ends.
GoldSim element value or
distribution
units reference / comment
NumCells_Footprint_Ends 4 modeling construct

7.3.3 \Disposal\SatZone\SZ_ToWell
Table 66. Total number of cells from footprint to well.
GoldSim element value or
distribution
units reference / comment
NumCells_ToWell 20 modeling construct

7.4 \Disposal\EngineeredSystemGeometry
Table 67. Engineered system geometry parameters.
GoldSim element value units reference / comment
ClayLiner_Thickness 2 ft see Embankment Modeling white paper

Model Parameters for the Clive DU PA Model
25 November 2015 41
8.0 \Exposure_Dose
The Data element Dose_Timestep_Length is controlled by the user, and so has no set value.
8.1 \Exposure_Dose\Media_Concs
Concentrations of contaminants in environmental media to which receptors may be exposed are
collected and calculated in this container.
Table 68. Mechanically generated dust
GoldSim element value or
distribution
units reference / comment
OHV_DustAdjustment
OHV dust loading
LN( GM=98.1,
GSD=1.65,
min=Small,
max=Large )
— See Dose Assessment white paper

8.1.1 \Exposure_Dose\Media_Concs\Exposure_Areas
Table 69. Exposure areas used in the calculation of exposure media concentrations
GoldSim element value or distribution units reference / comment
Receptor_Area
Receptor area (exposure
area)
U( 16,000, 64,000 ) acres See Dose Assessment white paper
AntelopeRange_Area
Pronghorn range area
U( 995, 9192 ) acres ibid.

8.1.2 \Exposure_Dose\Media_Concs\Animal_Concentrations
Table 70. Animal tissue concentrations for the recreational and ranching scenarios
GoldSim element value or
distribution
units reference / comment
TF_Beef_GM
Beef transfer factor,
geometric mean
Tabulated in
workbook
day/kg “Clive PA Model Parameters.xls”,
Elements worksheet; see also Dose
Assessment white paper
TF_Beef_GSD
Beef transfer factor,
geometric standard deviation
Tabulated in
workbook
— ibid.
WaterIngRate_Cattle
Cattle water ingestion rate
U( 33, 53 ) kg/day See Dose Assessment white paper
Model Parameters for the Clive DU PA Model
25 November 2015 42
GoldSim element value or
distribution
units reference / comment
ForageIngRate_Cattle
Cattle forage ingestion rate
U( 8.85, 14.75 ) kg/day ibid.
SoilIngRate_Cattle
Cattle soil ingestion rate
U( 0.05, 0.95 ) kg/day ibid.
GrazingTimeFrac_Cattle
Cattle time fraction in
exposure area
1 — ibid.
WaterIngRate_Antelope
Pronghorn water ingestion
rate
U( 0.1, 1 ) kg/day ibid.
BodyWtFactor_Antelope
Pronghorn body weight, as a
unitless factor for allometric
scaling
U ( 38,000,
41,000 )
— ibid. Body mass in Dose Assessment
white paper reported in units of kg.
ForageIngRate_Antelope
Pronghorn forage ingestion
rate
0.577 ×
BodyWtFactor
_Antelope0.727 ×
0.001
kg/day ibid.
SoilIngRate_Antelope
Pronghorn soil ingestion rate
U( 0.005, 0.095 ) kg/day ibid.

8.1.2.1 \Exposure_Dose\Media_Concs\Animal_Concentrations\Beef_TFs
The beef transfer factors are tabulated in the Parameters Workbook, but some values in those
tables point to fixed values in the GoldSim model. These are tabulated here:
Table 71. Parameters related to beef transfer factors
GoldSim element value or
distribution
units reference / comment
BeefTF_GM_radon
Beef transfer factor for radon,
geometric mean
Small day/kg See Dose Assessment white paper
BeefTF_GSD_radon
Beef transfer factor for radon,
geometric standard deviation
1 — ibid.
BeefTF_GSD_generic
Generic beef transfer factor,
geometric standard deviation
1.475 — ibid.

Model Parameters for the Clive DU PA Model
25 November 2015 43
8.2 \Exposure_Dose\DCFs
Table 72. Dose conversion factors
GoldSim element value or
distribution
units reference / comment
BranchingFractions
Radionuclide branching
fractions
Tabulated in
workbook
— “Dose Assessment Appendix II.xls”,
see also Dose Assessment white
paper
DCF_Inh_Dust_determ
Dose conversion factor,
inhalation dust
Tabulated in
workbook
Sv/Bq ibid.
DCF_Inh_Gas_determ
Dose conversion factor,
inhalation gas
Tabulated in
workbook
Sv/Bq ibid.
DCF_Ing_determ
Dose conversion factor,
ingestion
Tabulated in
workbook
Sv/Bq ibid.
DCF_Ext_Imm_determ
Dose conversion factor,
immersion
Tabulated in
workbook
( Sv-m3 )/(
Bq-s )
ibid.
DCF_Ext_Soil_determ
Dose conversion factor,
external
Tabulated in
workbook
( Sv-m3 )/(
Bq-s )
ibid.
Rn222_EffectiveDose
Effective dose for Radon-
222
6 ( mSv-m3 )/ (
mJ-hr )
See Dose Assessment white paper
Rn_progeny_equil
energy per Bq of radon at
equilibrium
5.56E-06 mJ/Bq ibid.
Rn_Inh_rate
Breathing rate for a
standard worker
1.2 m3/hr ibid.

8.2.1 \Exposure_Dose\DCFs\Stochastic_REFs
Table 73. Stochastic radiation effectiveness factors
GoldSim element value or distribution units reference / comment
Alpha_GM
Alpha radiation effectiveness
factor, geometric mean
18.1 — “Dose Assessment
Appendix II.xls”,
see also Dose
Assessment white paper
Model Parameters for the Clive DU PA Model
25 November 2015 44
GoldSim element value or distribution units reference / comment
Alpha_GSD
Alpha radiation effectiveness
factor, geometric standard
deviation
2.37 — ibid.
Alpha_REF
Alpha radiation effectiveness
factor, distribution
LN( GM=Alpha_GM,
GSD=Alpha_GSD )
— ibid.
Beta_GM
Electron radiation
effectiveness factor, geometric
mean
2.41 — ibid.
Beta_GSD
Electron radiation
effectiveness factor, geometric
standard deviation
1.44 — ibid.
Beta_REF
Electron radiation
effectiveness factor,
distribution
LN( GM=Beta_GM, GSD=
Beta_GSD )
— ibid.
Photon1_GM
Photon radiation effectiveness
factor (30-250 keV), geometric
mean
1.96 — ibid. (>0.03 and <=0.25
MeV)
Photon1_GSD
Photon radiation effectiveness
factor (30-250 keV), geometric
standard deviation
1.48 — ibid.
Photon1_REF
Photon radiation effectiveness
factor (30-250 keV),
distribution
LN( GM=Photon1_GM,
GSD= Photon1_GSD )
— ibid.
Photon2_GM
Photon radiation effectiveness
factor (< 30 keV), geometric
mean
2.45 — ibid.
Photon2_GSD
Photon radiation effectiveness
factor (< 30 keV), geometric
standard deviation
1.55 — ibid. (<=0.03 MeV)
Photon2_REF
Photon radiation effectiveness
factor (< 30 keV), distribution
LN( GM=Photon2_GM,
GSD=Photon2_GSD )
— ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 45
GoldSim element value or distribution units reference / comment
Deterministic_REF
Deterministic radiation
effectiveness factor
1 — See Dose Assessment
white paper
WeightingFactor_Alpha
Weighting factor for alpha
radiation
20 — ibid.
WeightingFactor_Beta
Weighting factor for beta
radiation
1 — ibid.
WeightingFactor_Gamma
Weighting factor for gamma
radiation
1 — ibid.

8.3 \Exposure_Dose\OuterLoop_Exposure_Parameters
Table 74. Exposure parameters, sampled once per realization
GoldSim element value or
distribution
units reference / comment
SoilIngestionTracerElement
Adult incidental soil ingestion
rate tracer elements
Probability Value — See Dose Assessment white
paper
0.3333 0 Tracer element: silicon
0.3334 1 Tracer element: aluminum
0.3333 2 Tracer element: titanium
EF_food
Exposure frequency, food
365 day/yr See Dose Assessment white
paper
Meat_PrepLoss
Meat preparation loss
N( µ=0.27, σ=0.07,
min = 0.01, max =
1 )
— ibid.
Meat_PostCookLoss
Meat post-cooking loss
N( µ=0.24, σ=0.09,
min = 0.01, max =
1 )
— ibid.

8.4 \Exposure_Dose\Dose_Calculations
This looping container performs calculations on a finer time step than the outer model, and has
parameters that are sampled on the inner time steps.
Model Parameters for the Clive DU PA Model
25 November 2015 46
8.4.1 \Exposure_Dose\Dose_Calculations\Physiology_Rancher
Table 75. Attributes of inter-individual uncertainty in physiological characteristics for
rancher receptors (ranch hands)
GoldSim element value or distribution units reference / comment
Age N( µ=25.7, σ=20.3,
min = 16, max = 60 )
yr See Dose
Assessment white
paper
Gender Male 60.8%, Female
39.2%
— ibid.
BodyWeight
Body mass
LN( GM=f(x), GSD=f(x) ) kg Inputs denoted as f(x)
are calculated based
on other outputs from
the model and are
documented in the
Dose Assessment
white paper
SoilIngestionRate
Adult incidental soil ingestion
rate
LN( GM=f(x), GSD=f(x),
Min=0, Max=f(x) )
mg/day ibid.
BeefIngestionRate_BWA
Ingestion rate: “home-
produced” beef
Gamma( µ=f(x) , σ=f(x) ) g/kg-day ibid.
VentilationRateSleep_BWA
Ventilation rate: sleeping
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationSleep_dist
Daily exposure time: sleeping
LN( GM=f(x), GSD=f(x),
Min=1, Max=24 )
hr/day ibid.
VentilationRateSedentary_
BWA
Ventilation rate: sedentary
activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationSedSleep
Daily exposure time:
sedentary+sleeping
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
VentilationRateLight_BWA
Ventilation rate: light activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
VentilationRateMedium_BWA
Ventilation rate: moderate
activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
VentilationRateHeavy_BWA
Ventilation rate: high activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 47
GoldSim element value or distribution units reference / comment
ActivityDurationLight_UN
Daily exposure time: light
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
ActivityDurationMedium_UN
Daily exposure time: moderate
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
ActivityDurationHeavy_UN
Daily exposure time: high
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.

8.4.2 \Exposure_Dose\Dose_Calculations\Physiology_SportOHV
Table 76. Attributes of inter-individual uncertainty in physiological characteristics for
Sport OHV receptors
GoldSim element value or distribution units reference /
comment
Age N( µ=25.7, σ=20.3,
min = 16, max = 60 )
yr See Dose
Assessment white
paper
Gender Male 60.8%, Female
39.2%
— ibid.
BodyWeight
Body mass
LN( GM=f(x), GSD=f(x) ) kg Inputs denoted as f(x)
are calculated based
on other outputs from
the model and are
documented in the
Dose Assessment
white paper
SoilIngestionRate
Adult incidental soil ingestion
rate
LN( GM=f(x), GSD=f(x),
Min=0, Max=f(x) )
mg/day ibid.
VentilationRateSleep_BWA
Ventilation rate: sleeping
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationSleep_dist
Daily exposure time: sleeping
LN( GM=f(x), GSD=f(x),
Min=1, Max=24 )
hr/day ibid.
VentilationRateSedentary
_BWA
Ventilation rate: sedentary
activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 48
GoldSim element value or distribution units reference /
comment
ActivityDurationSedSleep
Daily exposure time:
sedentary+sleeping
LN( GM=f(x),GSD=f(x) 1.09
or 1.08 )
hr/day ibid.
VentilationRateLight_BWA
Ventilation rate: light activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
VentilationRateMedium_BWA
Ventilation rate: moderate
activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
VentilationRateHeavy_BWA
Ventilation rate: high activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationLight_UN
Daily exposure time: light
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
ActivityDurationMedium_UN
Daily exposure time: moderate
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
ActivityDurationHeavy_UN
Daily exposure time: high
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.

8.4.3 \Exposure_Dose\Dose_Calculations\Physiology_Hunter
Table 77. Attributes of inter-individual uncertainty in physiological characteristics for
Hunter receptors
GoldSim element value or distribution units reference /
comment
Age
Age
N( µ=25.7, σ=20.3,
min = 16, max = 60 )
yr See Dose
Assessment white
paper
Gender
Gender
Male 60.8%, Female
39.2%
— ibid.
BodyWeight
Body weight
LN( GM=f(x), GSD=f(x) ) kg Inputs denoted as
f(x) are calculated
based on other
outputs from the
model and are
documented in the
Dose Assessment
white paper, Section
1.0.
Model Parameters for the Clive DU PA Model
25 November 2015 49
GoldSim element value or distribution units reference /
comment
SoilIngestionRate
Adult incidental soil ingestion rate
LN( GM=f(x), GSD=f(x),
Min=0, Max=f(x) )
mg/day ibid.
function of age
GameIngestionRate_BWA
Ingestion rate: “home-produced”
game
Gamma( µ=f(x), σ=f(x) ) g/kg-day ibid.
VentilationRateSleep_BWA
Ventilation rate: sleeping
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationSleep_dist
Daily exposure time: sleeping
LN( GM=f(x), GSD=f(x),
Min=1, Max=24 )
hr/day ibid.
VentilationRateSedentary
_BWA
Ventilation rate: sedentary activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationSedSleep
Daily exposure time:
sedentary+sleeping
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
VentilationRateLight_BWA
Ventilation rate: light activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
VentilationRateMedium_BWA
Ventilation rate: moderate activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
VentilationRateHeavy_BWA
Ventilation rate: high activity
LN( GM=f(x),GSD=f(x) ) m3/min-kg ibid.
ActivityDurationLight_UN
Daily exposure time: light activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
ActivityDurationMedium_UN
Daily exposure time: moderate
activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.
ActivityDurationHeavy_UN
Daily exposure time: high activity
LN( GM=f(x),GSD=f(x) ) hr/day ibid.

8.4.4 \Exposure_Dose\Dose_Calculations\ExposureTime_Rancher
Table 78. Attributes of inter-individual uncertainty in physiological characteristics for
Rancher receptors – Exposure Time
GoldSim element value or
distribution
units reference / comment
ET_Ranch_DayTrip
Ranchers; day trip time in
exposure area
U( min=4,
max=12 )
hr/day See Dose Assessment white
paper
Model Parameters for the Clive DU PA Model
25 November 2015 50
GoldSim element value or
distribution
units reference / comment
ET_Overnight
Exposure frequency, overnight
trips
24 hr/day ibid.
ET_Camp_OnsiteFrac
All receptors; fraction of camp
trip exposure time on disposal
cell
U( min=0.25,
max=0.75 )
— ibid.
OHV_timeFrac_Camper
All receptors; camp trip time
spent OHVing
U( min=2, max=8 ) hr/day ibid.
OHV_timeFrac_HuntRanch_
DayTrip
Hunter/Rancher; fraction of day
trip time spent OHVing
U( min=0.1,
max=0.75 )
hr/day ibid.
EF_Ranch_dist
Rancher; exposure frequency
beta( µ=135,
σ=34.9, min = 0,
max = 180 )
day/yr ibid.
Frac_Ranch_Overnight_dist
Ranchers; fraction of exposure
frequency related to overnight
trips
U( min=0.5,
max=0.67 )
— ibid.

8.4.5 \Exposure_Dose\Dose_Calculations\ExposureTime_SportOHV
Table 79. Attributes of inter-individual uncertainty in physiological characteristics for
Sport OHV receptors – Exposure Time
GoldSim element value or
distribution
units reference / comment
ET_Rec_DayTrip
Sport OHVers; day trip time in
exposure area
beta( µ=6.3,
σ=2.11, min = 1,
max = 20 )
hr/day See Dose Assessment white
paper
ET_Overnight
Exposure frequency, overnight
trips
24 hr/day ibid.
ET_Camp_OnsiteFrac
All receptors; fraction of camp
trip exposure time on disposal
cell
U( min=0.25,
max=0.75 )
— ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 51
GoldSim element value or
distribution
units reference / comment
OHV_timeFrac_Camper
All receptors; camp trip time
spent OHVing
U( min=2, max=8 ) hr/day ibid.
EF_Recreational_dist
Sport OHVer; exposure
frequency
LN( GM=11.3,
GSD=3.45, Min=1,
Max=200 )
d/yr ibid.
Frac_recOHV_Overnight_dist
Sport OHVers; fraction of
exposure frequency related to
overnight trips
U( min=0, max=1 ) — ibid.

8.4.6 \Exposure_Dose\Dose_Calculations\ExposureTime_Hunter
Table 80. Attributes of inter-individual uncertainty in physiological characteristics for
Hunter receptors – Exposure Time
GoldSim element value or distribution units reference / comment
ET_Rec_DayTrip
Sport OHVers; day trip time in
exposure area
beta( µ=6.3, σ=2.11,
min = 1, max = 20 )
hr/day See Dose
Assessment white
paper
ET_Overnight
Exposure frequency, overnight
trips
24 hr/day ibid.
ET_Hunt_DayTrip_OnsiteFrac
Hunter; fraction of hunting day
trip exposure time on disposal
cell
U( min=0.02, max=0.17 ) — ibid.
ET_Camp_OnsiteFrac
All receptors; fraction of camp
trip exposure time on disposal
cell
U( min=0.25, max=0.75 ) — ibid.
OHV_timeFrac_Camper
All receptors; camp trip time
spent OHVing
U( min=2, max=8 ) hr/day ibid.
OHV_timeFrac_HuntRanch_
DayTrip
Hunter/Rancher; fraction of day
trip time spent OHVing
U( min=0.1, max=0.75 ) — ibid.
EF_Hunting_dist
Hunter; exposure frequency
LN( GM=4.66, GSD=3.45,
min=1, max=100 )
day/yr ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 52
GoldSim element value or distribution units reference / comment
Frac_Hunt_Overnight_dist
Hunters; fraction of exposure
frequency related to overnight
trips
U( min=0, max=1 ) — ibid.
EF_Recreational_dist
Sport OHVer; exposure
frequency
LN( GM=11.3, GSD=3.45,
min=1, max=200 )
day/yr ibid.

8.4.7 \Exposure_Dose\Dose_Calculations\Population_Size_Variables
Table 81. Attributes of population variability.
GoldSim element value or distribution units reference / comment
Number_Individuals_Total
Total number of individuals in
vicinity of site, per year
Tri( 100, 350, 500 ) — See Dose
Assessment white
paper
Ranch_Hands_dist
Number of ranchers in vicinity
of site, per year
U( 1, 20 ) — ibid.
Ranchers_Picker
This element is used to identify
the number of ranch receptors
present.
Binomial( Batch Size = 1,
Probability = f(x)/20 )
— For probability, the
denominator
corresponds to the
size of the receptor
array and f(x) to the
value of
Ranch_Hands_dist.
Number_Hunter
Number of hunters in vicinity of
site, per year
Binomial( Batch Size =
round(
Number_Individuals_Total –
Number_Ranch_Hands ),
Probability = 0.25 )
— See Dose
Assessment white
paper
Hunters_Picker
This element is used to identify
the number of hunter receptors
present.
Binomial( Batch Size = 1,
Probability =
Number_Hunter/175 )
— Analogous to
Ranchers_Picker.
Number_Recreationalists
Number of recreationalists in
vicinity of site
f(x) =
Number_Individuals_Total -
Ranch_Hands_Dist
— See Dose
Assessment white
paper
Number_SportOHV
Number of OHVers in vicinity
of site
f(x) =
Number_Recreationalists -
Number_Hunter
— See Dose
Assessment white
paper
Model Parameters for the Clive DU PA Model
25 November 2015 53
GoldSim element value or distribution units reference / comment
SportOHVers_Picker
This element is used to identify
the number of SportOHV
receptors present.
Binomial( Batch Size = 1,
Probability =
Number_SportOHV/424 )
— Analogous to
Ranchers_Picker.

8.4.8 \Exposure_Dose\Dose_Calculations\UraniumHazard
Table 82. Uranium hazard for Rancher and Recreationists.
GoldSim element value or
distribution
units reference / comment
Uranium_RfD
Reference dose for uranium
Probability Value See Dose Assessment white
paper 0.5 0.0006 mg/kg-
day
0.5 0.0030 mg/kg-
day

8.4.9 \Exposure_Dose\Dose_Calculations\OffSite_Receptors
Table 83. Inhalation dose for off-site receptors.
GoldSim element value or
distribution
units reference / comment
ET_RestArea
Exposure time rest area caretaker
24 hr/day See Dose Assessment white
paper
EF_RestArea
Exposure frequency rest area
caretaker
Tri( 327, 350, 365 ) day/yr ibid.
ET_Knolls
Exposure time for day trip,
Knolls OHVer
Beta( µ=6.3,
σ=2.11, min=1,
max=20)
hr/day ibid.
EF_Knolls
Exposure frequency, Knolls
OHVer
LN( µ=11.3,
σ=3.45, min=1,
max=200 )
day/yr ibid.
ET_Traveller
Exposure time travelers on I-80
and train
U( 2.3, 7.2 ) min/day ibid.
EF_Traveller
Exposure frequency I-80 and
west-side access road traveller
U( 250, 365 ) day/yr ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 54
GoldSim element value or
distribution
units reference / comment
ET_UTTR_Road
Exposure time cars on west-side
access road (Utah Test and
Training Range access)
U( 2.4, 4.0 ) min/day ibid.

8.4.10 \Exposure_Dose\Screening_Calculations
Table 84. Parameters used in screening dose calculations.
GoldSim element value or
distribution
units reference / comment
NativePlant_Ing_Rate 1 kg/yr See Dose Assessment white
paper
FreshWeightConversion U( 0.05, 0.3 ) — ibid.
OffsiteWater_Ing_Rate 1 L/yr ibid.

9.0 \GWPLs
The model estimates concentrations in a hypothetical monitoring well down gradient of the
waste embankment. Certain radionuclides are of interest, and their concentrations are displayed
for comparison to Ground Water Protection Limits (GWPLs) as specified in State of Utah (2010)
Table 1A.
Table 85. Groundwater protection limits.
GoldSim element value or
distribution
units reference / comment
MaxTime_WellConcs 500 yr State of Utah (2010)
GWPL_Sr90 42 pCi/L ibid.
GWPL_Tc99 3790 pCi/L ibid.
GWPL_I129 21 pCi/L ibid.
GWPL_Th230 83 pCi/L ibid.
GWPL_Th232 92 pCi/L ibid.
GWPL_Np237 7 pCi/L ibid.
GWPL_U233 26 pCi/L ibid.
GWPL_U234 26 pCi/L ibid.
GWPL_U235 27 pCi/L ibid.
Model Parameters for the Clive DU PA Model
25 November 2015 55
GoldSim element value or
distribution
units reference / comment
GWPL_U236 27 pCi/L ibid.
GWPL_U238 26 pCi/L ibid.

10.0 \DeepTimeScenarios
Deep time scenarios are developed to provide information for a qualitative analysis of effects
from the Clive Facility on future conditions after 10,000 years.
Table 86. Deep time scenario parameters.
GoldSim element value or
distribution
units reference / comment
NumDUCells
Number of waste cells that
contain DU
6 — See Deep Time white paper

NumCells_BelowGrade
Number of waste layer cells
below grade
10 --- ibid.
DepthAeolianDeposition
long-term aeolian deposition
depths
N(µ=72.7, σ=5
min=Small,
max=Porosity_
Unit4)
cm ibid.
AgeAeolianDeposition
long-term aeolian deposition
ages
Beta(µ=13614,
σ=263.3,min=1
3000,max=150
00)
Correlated to
DepthAeolianD
eposition:
AeolianCorrela
tionFactor
yr ibid.
AeolianCorrelationFactor
correlation between aeolian
deposition depth and Aeolian
deposition age
U(0.5,1.0) — ibid.

Model Parameters for the Clive DU PA Model
25 November 2015 56
10.1 \DeepTimeScenarios\LakeReturnCalcs
Table 87. Parameters for the lake return calculations.
GoldSim element value or
distribution
units reference / comment
LakeDelayTime
time at which the intermediate
lake calculations are allowed
to occur
50,000 yr See Deep Time white paper

IntermediateLakeDuration
length of time that Clive is
covered by an intermediate
lake
LN(GM=500,
GSD=1.5,min=0
, max=2500)
yr ibid.
IntermediateLakeSedimentA
mount
total depth of sediment laid
down by an intermediate lake

LN(GM=2.82,
GSD=1.71)
m ibid.
DeepLakeStart
time before the end of the
100,000-year climate cycle
LN(GM=14000,
GSD=1.2,min=0
, max=50000 )
yr ibid.
DeepLakeEnd
time after the most recent
cold peak within the 100,000-
year climate cycle
LN(GM=6000,
GSD=1.2,min=0
, max=50000)
yr ibid.
DeepLakeSedimentationRate
rate of the sedimentation
during the open water phase
of a deep lake
LN(GM=1.2E-4,
GSD=1.2)
m/yr ibid.

10.2 \DeepTimeScenarios\LakeChemistry
Table 88. Parameters for calculating the dispersal of the embankment and subsequent lake
and sediment concentrations.
GoldSim element value or
distribution
units reference / comment
SiteDispersalArea
the area across which the
destroyed site is spread
gamma(µ=24.23
32,
σ=11.43731)
Km2 See Deep Time white paper

Model Parameters for the Clive DU PA Model
25 November 2015 57
GoldSim element value or
distribution
units reference / comment
IntermediateLakeDepth
depth of an intermediate lake
at Clive
Beta(µ=30,
σ=18,min=0,
max=100)
m ibid.
DeepLakeDepth
depth of a large lake at Clive
Beta(µ=150,
σ=20,min=100,
max=200)
m ibid.
DiffusionLength
Diffusion length for the deep
time sediments
N(µ=0.5, σ=0.16
min=0.0,
max=Large)
m ibid.

10.3 \DeepTimeScenarios\RadonFlux_NRC364
No input elements are defined at this level.
10.4 \DeepTimeScenarios\ExposureDose_DeepTime
Many dose assessment input parameters for deep time were taken from the dose assessment
container for the 10,000-year analysis. Some dose parameters for deep time were chosen to be
deterministic rather than stochastic, with the assumption that deep time already has many
uncertainties. More information is provided in the tables below.
Table 89. Parameters for the deep time human exposure and dose assessment.
GoldSim element value or
distribution
units reference / comment
ET_ranch
length of a work day for a
ranching receptor
8 hr/d Used the mean of the similar
distribution from Section 8.4.4

EF_ranch
upper bound yearly exposure
frequency for the ranching
receptor
180

d/yr ibid.
ChiQ_Gas_Onsite
Chi/Q value for the gases at
the on-site receptor location
234 (ug/m3)/
(g/s)
See Section 7.1 above.
InhRate
moderate intensity short-term
inhalation rate
0.03 m3/min 0.03 m³/min is slightly above the
mean value for ages 21 - 30 (0.026)
through 51 - 60 (0.029) in Table 6-2
of EFH 2011
Model Parameters for the Clive DU PA Model
25 November 2015 58
GoldSim element value or
distribution
units reference / comment
external_DCF_modifiers
RESRAD-derived multipliers
for infinite-source external
DCFs to account for
attenuation by overlaying
sediments
Tabulated in
workbook
— See Deep Time white paper and
workbook “ES external DCF
modifiers.xlsx”
Rn_flux_ratio
ratio of Rn-222 flux at
different sediment thickness
to flux with no overlaying
cover
0.001
0.5
1.0
1.5
2.0
3.0
4.0
5.0
6.0
7.5

1.00000
4.392E-1
1.972E-1
8.750E-2
4.000E-2
8.140E-3
1.656E-3
3.371E-4
6.881E-5
1.00E-30
— See Deep Time white paper and
workbook “ES radon dose.xlsx”

10.4.1 \DeepTimeScenarios\ExposureDose_DeepTime\Exposure_Areas
Table 90. Exposure areas used in the calculation of exposure media concentrations.
GoldSim element value or
distribution
units reference / comment
Receptor_Area
size of ranching activities
area
U(min=16000,max
=64000)
acre
See Dose Assessment white paper

10.4.2 \DeepTimeScenarios\ExposureDose_DeepTime\DCFs
This container is identical to the \Exposure_Dose\DCFs container described in Section 8.2.
10.4.2.1 DeepTimeScenarios\ExposureDose_DeepTime\DCFs\Stochastic_REFs
This container is identical to the \Exposure_Dose\DCFs\Stochastic_REFs container described in
Section 8.2.1

Model Parameters for the Clive DU PA Model
25 November 2015 59
11.0 References
EPA. 2011. Exposure Factors Handbook. Office of Research and Development. US
Environmental Protection Agency, Washington, DC.
Kocher, D.C., 1981. Radioactive Decay Data Tables, DOE/TIC-11026, Technical Information
Center, U.S. Dept. of Energy, Washington, DC.
Tuli, J.K., 2005, Nuclear Wallet Cards, National Nuclear Data Center. Brookhaven National
Laboratory. Seventh edition, April 2005.

NAC-0055_R2

Quality Assurance Project Plan

Clive DU PA Model v1.4
11 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 110, Lakewood, CO 80215
Quality Assurance Project Plan
11 November 2015 ii

Solutions

Quality Assurance Project Plan
11 November 2015 1
CONTENTS
CONTENTS .....................................................................................................................................1
FIGURES .........................................................................................................................................2
TABLES ..........................................................................................................................................3
1.0 Introduction .............................................................................................................................4
2.0 Project Management and Organization ...................................................................................4
3.0 Personnel Qualifications and Training ...................................................................................5
4.0 Project Description .................................................................................................................6
4.1 DU PA Model Consolidation ............................................................................................6
4.2 Model Embankment Terminology Change .......................................................................7
4.3 SER Unresolved Issues Responses ...................................................................................7
4.4 Class A West ET Cover Model Revisions ......................................................................10
5.0 Quality Objectives and Model Performance Criteria ...........................................................11
6.0 Documentation and Records .................................................................................................11
7.0 Data Acceptance Criteria ......................................................................................................11
8.0 Data Management and Software Configuration ...................................................................12
9.0 Model Assessment and Response Actions ............................................................................12
10.0 Model Requirements Assessment .........................................................................................13
Appendix A: Subversion SOP .......................................................................................................14
Appendix B: GoldSim Model Development SOP .........................................................................26
Appendix C: Neptune Check Print SOP ........................................................................................51

Quality Assurance Project Plan
11 November 2015 2
FIGURES
Figure 1. Neptune Organizational Chart ..........................................................................................5

Quality Assurance Project Plan
11 November 2015 3
TABLES
Table 1. Roles, Responsibilities, and Training ................................................................................6

Quality Assurance Project Plan
11 November 2015 4
1.0 Introduction
This document describes the Quality Assurance Project Plan (QAPP) for modeling services
provided for the development of a Performance Assessment (PA) model for the disposal of
depleted uranium (DU) by EnergySolutions at the Clive, Utah facility. Throughout this
document, the term Quality Assurance (QA) refers to a program for the systematic monitoring
and evaluation of the various aspects of PA model development to ensure that the models and
analyses are of the type and quality of that needed and expected by the client.
2.0 Project Management and Organization
Neptune and Company, Inc. (Neptune) has developed this QAPP for conducting work for
EnergySolutions. This QAPP is based on the Environmental Protection Agency (EPA) QA/G-5M
Guidance for Quality Assurance Project Plans for Modeling, and our company’s 23-year history
working in the environmental quality arena. A tiered approach is used that includes specific
procedures developed by Neptune that have been developed for modeling projects. This project-
specific QAPP will work as an umbrella plan that ensures quality across all tasks.
The Neptune quality program includes:
 Experienced and trained personnel who understand the QA requirements of each task.
 An experienced Project Manager.
 A corporate Quality Assurance Officer
 Task planning, tracking, and operation via internal SOPs.
 Emphasis on continuous improvement via internal reviews and customer feedback.
It is the policy of Neptune to implement a quality program designed to generate products or
services that meet or exceed the expectations established by our clients. This quality policy
addresses all products delivered to our EnergySolutions client under the contract. We will ensure
quality through the use of a quality program that includes program and project management,
systematic planning, work and product assessment and control along with continuous
improvement to ensure that data and work products are produced of acceptable quality to support
the intended use.
To achieve this goal, Neptune will assign appropriately qualified and trained staff and ensure that
all products are carefully planned. Tasks will be conducted according to the QAPP or applicable
SOP and any and all problems affecting quality will be brought to the immediate attention of the
project or task manager for resolution. All products will be reviewed by another technical expert.
Adequate budget and time will be planned to execute the quality system.
As indicated on Figure 1, the Neptune organizational structure ensures direct reporting between
the Neptune Project QA Officer and the Project Manager. This structure requires that all Neptune
technical staff report to the Neptune Project Manager who is responsible for the work.
Quality Assurance Project Plan
11 November 2015 5

Figure 1. Neptune Organizational Chart

The Neptune Quality Assurance Officer has the authority and responsibility to ensure that the
project-specific QAPP is implemented by Neptune staff. Roles and Responsibilities for this
project are detailed in Table 1. The QA aspects of the project are handled by those project
members responsible for any particular part of the project. The lead modeler is responsible for
QA for the GoldSim models. For probabilistic models, the lead statistician is responsible for QA
of statistical routines and products that feed into the model. The responsibility for other QA tasks
may be assigned to other project members at the direction of the lead modeler or lead statistician.
The model custodian is responsible for configuration control of the model. The role of model
custodian may be assumed by any project team member, but only one person at a time may be
the custodian.
3.0 Personnel Qualifications and Training
Neptune technical staff is composed of highly qualified chemists, engineers, statisticians, IT
professionals, QA specialists, and biologists with advanced degrees in their fields and direct
training experience. Many of the Neptune staff have participated in GoldSim training courses
and GoldSim User Conferences. Qualifications for the staff are shown in Table 1. Each Neptune
employee or contractor involved with this project will be required to read this QAPP and
associated standard operating procedures (SOPs).

Quality Assurance Project Plan
11 November 2015 6
Table 1. Roles, Responsibilities, and Training
Roles and
Responsbilities
Personnel Training
Project Manager Paul Black Ph.D. Statistics
QA Officer,
Hydrologist
Mike Sully Ph.D. Soil Science
GoldSim Training
Technical Lead John Tauxe Ph.D. Civil Engineer, Professional Engineer
(New Mexico), GoldSim Training
Modeler,
Geochemist
Katie Catlett Ph.D. Soil Science
GoldSim Training
Modeler,
Hydrologist
Dan Levitt Ph.D. Soil Science
GoldSim Training
Statistician Stephanie Fitchett Ph.D. Statistics
Modeler,
Hydrologist
Amy Jordon Ph.D. Hydrology
GoldSim Training
Modeler,
Statistician
Tom Stockton Ph.D. Environmental Modeling
GoldSim Training
Modeler,
Exposure and Dose
Assessment
Ralph Perona M.S. Environmental Health, DABT
GoldSim Training
Modeler,
Engineer
Gregg
Occhiogrosso
M.S. Environmental Engineering
GoldSim Training
Risk analyst Robert Lee M.S. Environmental Health
Modeler,
Ecologist
Greg McDermott M.S. Entomology
Statistician Matt Pocernich M.S. Environmental Engineering
M.S. Applied Mathematics (Statistics)
Statistician Will Barnett M.S. Ecological and Environmental Statistics
Technical Writer Annette Devlin M.A. English

4.0 Project Description
Current scope under this QAPP includes four major elements: 1) DU PA model consolidation; 2)
Model embankment terminology change; 3) Responses to unresolved issues in the April, 2015
Safety Evaluation Report (SER); and 4) Modification of the Class A West (CAW)
evapotranspiration (ET) model to address certain SER issues. A description of the activities for
each element are described below in more detail.
4.1 DU PA Model Consolidation
Several different models have been developed to date including: the initial DU PA v1.0; v1.2
developed in response to interrogatories; the original Deep Time Supplemental Analysis (DTSA)
Quality Assurance Project Plan
11 November 2015 7
model; and revisions to the DTSA model that addressed changes in sedimentation rates and did
not disperse the DU waste upon destruction of the mound (upon return of a lake to the Clive
elevation). Consolidating these models into a single model will help respond to the SER issues
more efficiently, address any future reviews more efficiently, and will bring all current models
under one roof, which will be needed in the future if the PA is to be expanded to address other
wastes and/or embankments.
The v1.2 model will also be updated to v11 of GoldSim – the DTSA model is already in v11.
The consolidated model will need to be rerun, and results produced, including sensitivity
analysis. The primary changes will be in the deep time part of the model, but the entire model
will be rerun.
The new consolidated version will be labeled the Clive DU PA v1.4 model. Changes will be
noted in the version change log of the GoldSim model. Supporting documentation also needs to
be updated with this model revision (again, mostly for the deep time aspects of the model). This
will include the v1.2 Conceptual Site Model (CSM) and Features, Events and Processes (FEPs)
reports, the white papers, the parameter list document, and the final report including a revised
sensitivity analysis.
4.2 Model Embankment Terminology Change
The terminology for the CAS Cell needs to be changed to the Federal DU Cell, and the
dimensions need to be updated.
The nomenclature will be changed for the embankment from Class A South Cell to Federal DU
Cell. In the model, all references to "_CAS_" will be replaced with "_FDU_". These references
run throughout the model, and will require many changes to parameter names and to in-model
documentation. Any change to a parameter name will require a coordinated change in the
Parameters Document, and an update to QA (at least in the version change notes and Note Panes
for each element changed). The terminology changes are likely to require text modifications on
many elements, and nearly every dashboard and result element.
EnergySolutions will provide the most recent engineering drawings and the Engineering white
paper will be revised accordingly. This will require including references to any new engineering
drawings that EnergySolutions may have of the Federal DU Cell.
4.3 SER Unresolved Issues Responses
The scope of this work involves modifying v1.4 of the model to address the SER issues. New
model versions will be created to address each of the issues. These model versions will be
labeled with v1.4XXX, where the XXX is used to denote that these models are not sanctioned by
Neptune, but rather, were developed in order to respond to SER issues. The SER issues will be
investigated using four XXX models:
 Clive DU PA Model v1.4XXX Benson.gsm
 Clive DU PA Model v1.4XXX Benson Clay Liner.gsm
 Clive DU PA Model v1.4XXX Benson Deep Time.gsm
Quality Assurance Project Plan
11 November 2015 8
 Clive DU PA Model v1.4XXX Benson Erosion.gsm
Documentation will include the results of these four XXX models, and a discussion of the basis,
or lack thereof, of the modifications included in the XXX models. These four models will help to
investigate the following SER issues:
a. UAC R313-25-8(2) and (3): Evapotranspiration Cover (lack of correlation between the
alpha and hydraulic conductivity values, etc.)
The Hydrus 1D model that is used as the basis for infiltration and water balance parameters in
GoldSim will be modified so that the cap is naturalized. Input parameters for these infiltration
models are derived from the distributions and methods described by Dr. Craig Benson in Volume
2, Appendix E, of the SER. Fifty of these parameter sets will be used as inputs to the naturalized
Hydrus 1D model. The infiltration and moisture content results of these runs will be statistically
abstracted to provide inputs to the modified GoldSim model. Tables showing average water
balance components for the last 100 years of the Hydrus simulations will be prepared for 5 of the
parameter sets with model results that span the observed range of net infiltration.
b. UAC R313-25-8(2): Infiltration (lack of correlation between the alpha and hydraulic
conductivity values, etc.)
This issue is resolved as part of issue 4.3 a above.
c. UAC R313-25-25: Erosion of Cover (clarification of certain issues relating to Appendix
10 to the DU PA version 1.2, June 5, 2014)
Appendix 10 of the DU PA Model Final report will be revised to more clearly explain the
SIBERIA model (v1.2). Figure 2 in Appendix 10 will be revised to include all realizations that
were performed, or a new figure will be added with all the realizations for clarification (the
figure currently shows the first 5 simulations, not all 1,000 that were run).
The influence of cover thinning on net infiltration will be investigated using Hydrus 1D models
of the cover system.
Clarification will be provided in Appendix 10, Figure 2 that the distribution of cover area
associated with a channel depth is unaffected when all the realizations are considered.
Cover thinning (erosion) will be included in Hydrus and, consequently, in the GoldSim model.
The CSM document and other supporting documents will be updated to further explain the
conceptual model underlying the v1.4 model. The v1.4XXX Benson Erosion model will be run,
results obtained, sensitivity analysis performed, and a technical memorandum written to
document the results and compare results to the v1.4 model results.
d. UAC R313-25-25(3) and (4): Frost Damage (need to resolve concerns with assumed
recurrence intervals, estimated frost penetration depths, and hydraulic property
estimates)
Quality Assurance Project Plan
11 November 2015 9
The SER issue indicates that EnergySolutions should account in modeling for substantial
disruption of near-surface layers above and within the radon barriers by frost, with
accompanying decreases in ET and increases for initially low-permeability soil in both hydraulic
conductivity and correlated values, which could affect modeled infiltration and radon release
rates. These are the types of processes accounted for by using the naturalized cover material
properties for the modeling to be provided for issue 4.3 a. This issue will be addressed through
the analysis to be done in 4.3 a.
e. UAC R313-25-24(3) and (4): Effect of Biologicals on Radionuclide Transport (need to
account for natural increases in cover permeability over time)
The SER issue indicates that an increase in cover permeability will occur in response to biotic
activity. These are the types of processes accounted for by using the naturalized cover material
properties for the modeling to be provided for issue 4.3 a. This issue will be addressed through
the analysis to be done in 4.3 a.
The response should also indicate that the maximum rooting depth currently used in the DU PA
model extends below the lower radon barrier.
f. UAC R313-25-8(2): Clay Liner (lack of increase in Ksat values over time; lack of
correlation between the alpha and hydraulic conductivity values)
These changes will be implemented in the GoldSim model in conjunction with model v1.4XXX
Benson Clay Liner. There could be some small change in GoldSim model results because the
saturated hydraulic conductivity (Ksat) for the clay liner affects water content in the clay liner
layer, but not significantly. The model will be run and a table of results produced to show that
there is no significant difference.
g. UAC R313-25-8(10): GoldSim Quality Assurance [the relationship between the process
level model (i.e., HYDRUS) abstractions and the primary model (i.e., GoldSim) results
needs to be demonstrated]:
Table 4-1 in the SER shows that the HYDRUS and GoldSim infiltration rates are different. The
GoldSim and HYDRUS infiltration rates need to be compared and some investigation performed
to fully address the SER issue. This might also require running the GoldSim model to provide a
basis for discussion.
More generally, a discussion of scaling for PAs needs to be included up front and center in the
PA documentation explaining why use of standard errors of data is more appropriate than
standard deviations for parameter distributions.
h. UAC R313-25-9(5)(a): Deep Time Analysis
The v1.4 model will be modified, which will incorporate the latest DTSA model, as follows:
1. The material above the DU waste will be modeled as Unit 3 to account for the expected
grain-size characteristics of intermediate lake sediments and an expected southern flux of
long-shore drift sand from the Grayback Hills southward toward the Clive site.
Quality Assurance Project Plan
11 November 2015 10
2. The intermediate lake sedimentation rate will be changed to 10 times the large lake
sedimentation rate.
3. The standard deviation of the eolian deposition rate will be used instead of the standard
error of the mean.
This will result in a v1.4XXX Benson Deep Time version of the model. A technical
memorandum will be prepared to discuss the results, show sensitivity analysis, and compare
results to the current deep time model results. The results will focus on sediment and water
concentrations, but will use receptor scenarios from the main model as well to provide dose
estimates for comparison.
4.4 Class A West ET Cover Model Revisions
The ET cover model for Class A West also needs to be modified to accommodate some of the
SER issue requirements. In particular, SER issues a, b, and c need to be addressed (while issues
d and e will be included implicitly).
The current Hydrus 1D models applied to the ET cover will be revised to accommodate input
from Dr. Craig Benson on the correlation between hydrologic input parameters, the cap will be
naturalized, and some thinning of the cap will be accommodated if infiltration is found to be
affected by erosion. These changes effectively cover the frost and biotic issues (d and e).
Assuming that the 12 in. ET layer cover model will be used as the basis for the SER issue
modifications, the following Hydrus 1D models will be run:
1) the current based model with an updated leaf area index;
2) the current model modified to a naturalized cap, and using input values suggested in the
SER issue responses; and
3) the model modified again to allow for a thinning cap, which will be run by thinning the
entire cap by the same amount (if erosion is found to impact net infiltration for a
naturalized cover).
Note also that the thinned cap will be implemented at time 0 – otherwise Hydrus would need to
be applied to different cap thicknesses over time, which would be computationally
intensive/expensive.
For each model, 50 simulations will be run, and the resulting output will be made available for
revising the RESRAD models. Because the RESRAD modeling is deterministic, a reasonable
deterministic statistic (e.g., the mean or upper confidence bound on the mean) will be selected
for each output parameter from Hydrus that is used in RESRAD. The purpose of running many
simulations is to evaluate the conditions under which the Hydrus output results change.
The results of the different models will be described in a technical memorandum report. If any
other changes are needed to the current report, those will be made as well. The technical
memorandum will address the changes that have been requested through the SER, and why/how
these changes are counter to the conceptual model.
Quality Assurance Project Plan
11 November 2015 11
5.0 Quality Objectives and Model Performance Criteria
Systematic planning to identify required GoldSim model components will be accomplished
through the development of a CSM for the disposal of depleted uranium at the Clive facility. The
CSM describes the physical, chemical, and biological characteristics of the Clive facility.
The CSM encompasses everything from the inventory of disposed wastes, the migration of
radionuclides contained in the waste through the engineered and natural systems, and the
exposure and radiation doses to hypothetical future humans. These site characteristics are used to
define variables for the quantitative PA model that is used to provide insights and understanding
of the future potential human radiation doses from the disposal of DU waste. The content of the
CSM provides the basis for selection of the significant regional and site-specific features, events
and processes that need to be represented mathematically in the PA model. A report describing
the CSM will be developed as part of Task 1.
As described in Section 4.0 the objective of the PA is to provide a tool for determining if specific
performance objectives will be met for land disposal of radioactive waste set forth in Title 10
Code of Federal Regulations Part 61 (10 CFR 61) Subpart C, and promulgated by the NRC. The
quality objective for the model is to provide results that are consistent with the site
characteristics, the waste characteristics, and the CSM. If data are available, the demonstration of
consistency will be supported by available site monitoring data and other field investigations.
The model predictions of transport of radionuclides and the inadvertent intrusion into the
disposal facility, and the sensitivity and uncertainty of the calculated results should be
comprehensive representations of the existing knowledge of the site and the disposal facility
design and operations.
6.0 Documentation and Records
Subversion version-control software will be used to maintain records of ownership and
traceability of all project-specific files and database contents. Original data are stored in version-
controlled repositories. Additions, deletions and file modifications within the repository are
tracked by the version control software, which documents the file user and the date and time of
modification. The version control software also offers a “compare between revisions” feature for
text files that denotes line-by-line changes between previous and current versions of a file. User-
entered comments are also maintained by the version control software as files are added, deleted,
or modified. Version control of records is described in more detail in the Subversion SOP in
Appendix A.
Internal documentation of the GoldSim model, version change notes, change log, model
versioning, and model error reporting and resolution are described in the GoldSim Model
Development SOP in Appendix B and the Issue Tracking SOP in Appendix C.
7.0 Data Acceptance Criteria
The choice of data sources depends on data availability and data application in the model. The
following hierarchy outlines different types of information and their application. The information
Quality Assurance Project Plan
11 November 2015 12
becomes increasingly site-specific and parameter uncertainty is generally reduced moving down
the list.
 Physical limitations on parameter ranges, used for bounding values when no other
supporting information is available. Example: Porosity must be between 0 and 1 by
definition.
 Generic information from global databases or review literature, used for bounding values
and initial estimates in the absence of site-specific information. Example: A common
value for porosity of sand is 0.3.
 Local information from regional or national sources, used to refine the above
distributions, but with little or no site-specific information. Example: Sandy deposits in
the region have been reported to have porosities in the range of 0.30 to 0.37, based on
drilling reports.
 Information elicited from experts regarding site-specific phenomena that cannot be
measured. Example: The likelihood of farming occurring on the site sometime within the
next 1000 years is estimated at 50% to 90%.
 Site-specific information gathered for other purposes. Example: Water well drillers
report the thickness of the regional aquifer to be 10 to 12 meters.
 Site-specific modeling and studies performed for site-specific purposes. Example: The
infiltration of water through the planned engineered cap is estimated by process
modeling to be between 14 and 22 cm/yr.
 Site-specific data gathered for specific purposes in the models. Example: The density of
Pogonomyrmex ant nests adjacent to the site is counted and found to be 243 nests per
hectare.
The determination of data adequacy is informed by a sensitivity analysis of the model, which
identifies those parameters most significant to a given model result. Such parameters are
candidates for improved quality. As the model development cycle proceeds, sensitive parameters
are identified, and their sources are evaluated to determine the cost/benefit of reducing their
uncertainty.
8.0 Data Management and Software Configuration
The acquired data, developed statistical distributions and results generated by the GoldSim
model and the uncertainty and sensitivity analyses will be archived in a version-control
repository as described in Section 6.0 above. Configuration management for the GoldSim model
is described in GoldSim Model Development SOP in Appendix B.
9.0 Model Assessment and Response Actions
During model development, assessments will be conducted using a graded approach with the
level of testing proportional to the importance of the model feature. Assessments will consist of:
 reviews of model theory
 reviews of model algorithms
 reviews of model parameters and data
 sensitivity analysis
Quality Assurance Project Plan
11 November 2015 13
 uncertainty analysis
 tests of individual model modules using alternate methods of calculation such as analytic
solutions or spreadsheet calculations
 reasonableness checks
Response actions including error reporting and resolution processes are described in the GoldSim
SOP and the Issue Tracker SOP.
10.0 Model Requirements Assessment
The purpose of these assessments is to confirm that the modeling process was able to produce a
model that meets project objectives. Model results will be reviewed to ensure that results are
consistent with the site characteristics, the waste characteristics, and the CSM as described in
Section 5.0. Model results will be assessed to determine that the requirements of
EnergySolutions for the use of the model have been met. Any limitations on the use of the model
results will be reported to the project manager and discussed with EnergySolutions.

Quality Assurance Project Plan
11 November 2015 14
Appendix A: Subversion SOP

N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 2
 of
 10

Revision
 0

3. RESPONSIBILITIES

3.1. N&C Corporate Quality Assurance Officer (QAO): Maintains current record of this SOP
and may modify as the developments in contracts or internal N&C procedures warrant.
Conveys proposed modifications to contract-specific quality manager and program
manager.

3.2. N&C Contract Specific Quality Manager: Recommends modifications to this SOP when
appropriate and as needed to meet contract specific QA requirements, and drafts
recommended changes for review.

3.3. N&C Program (Contract) Manager: Ensures all Technical Staff working on the contract
are trained to the internal quality SOPs that pertain to the contract and that these procedures
are implemented. Reviews and approves SOPs that relate to the contract. Works with
Contract Specific QA Manager to execute contract specific modifications to this SOP.

3.4. N&C Technical Staff: Maintain current training on, and implement this SOP. Recommend
modifications to this procedure when appropriate by discussing their ideas with the contract
QAM and/or Program Manager, to maximize their effectiveness. Participates in any and all
assessments related to work under the contract, to ensure the Quality Management Plan and
related SOPs are routinely implemented.
4. DEFINITIONS

Definitions relevant to this SOP are provided in the following section.
5. PROCEDURE

As the Subversion online manual (http://svnbook.red-bean.com/) states, Subversion is a centralized
system for sharing information. At its core is a repository, which is a central store of data. The
SVN repositories live on a central server, SVN.neptuneinc.org. New repositories can be created on
the server at any time. To the user, a repository appears as a collection of files and directories
(although they are not actually stored that way on the SVN server).

Users access the contents of a repository by “checking out” a local copy of the repository. This
process copies files from the repository to the user’s computer, creating a local “working copy” of
the repository. The user can then make changes to their local copy and “commit” these changes
back to the repository, so they become part of the centralized data store. To get the latest changes
committed by others, the user should always “update” their repository before working on a given
file. Updating pulls down any new changes from the server that are not yet part of the user’s
working copy (see Section 5.4.3).
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 3
 of
 10

Revision
 0

Repositories have typically been created on a per-project basis, but some have instead been created
to house all the data associated with a particular client (for example, the EPA repository). The
latter approach produces very large repositories, which can make downloading the whole
repository time consuming, especially for users outside the Los Alamos office where the server
resides. However, this can be worked around by the user checking out only the sub-folders they
need from a given repository. This will be discussed in more detail later in this SOP.
5.1 Accessing Repositories

To access Neptune’s subversion repositories, you will need two things:
1) a subversion user account on the server
2) a client program running on your computer which can interact with the subversion server to
allow you to check out, update, and commit files
5.2 Obtaining a Subversion Account

This should be done automatically as part of your new-employee setup but any member of the IT
team can also set yours up. You will receive a username and a password, both of which need to be
submitted for most SVN transactions. Fortunately, all SVN clients provide the opportunity to
cache your identity so that you do not have to repeatedly enter your credentials.
5.3 Subversion Clients
5.3.1 Windows GUI

On Windows machines, the main client we use is Tortoise SVN, which is available from
Tigris.org. Its home page is http://tortoisesvn.tigris.org/. Downloading, installing and periodically
upgrading Tortoise SVN is a straightforward process, but IT staff will always be glad to offer
assistance if needed. Tortoise works as a plugin to Windows Explorer (NOT Internet Explorer the
web browser, but the file explorer); once you have Tortoise installed, you will see special icons
next to files that are part of working copies, and you will have access to SVN commands via right-
clicking on any file or folder in Windows Explorer.

Other clients are available – the other client that software developers use is a plugin to the Eclipse
development environment called Subclipse (also from Tigris).
5.3.2 Mac GUI

There are two main Mac clients currently in use at Neptune, SCplugin (http://scplugin.tigris.org/),
which mimics some of the Tortoise functionality but unfortunately does not have all features
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 4
 of
 10

Revision
 0

enabled on the latest OS version (Snow Leopard), and svnX
(http://www.lachoseinteractive.net/en/community/subversion/svnx/), which has a richer feature set
but a very different UI concept. Both clients are useful and can even coexist on the same machine.
As is the case on Windows, plugins are also available for various development environments (e.g.,
Netbeans, Eclipse).
5.3.3 Command Line

On Linux and other Unix-based systems (including the Mac), there is a command-line client
program called SVN. The command line client is the most flexible and powerful way to interact
with subversion, and may be needed in special situations to address issues that the GUI clients
cannot handle. In these cases, IT personnel can lead you through the necessary steps.
5.4 Getting Started with Subversion

Your first experience with subversion will likely involve someone on your project team telling you
to check out a repository (or sub-section of a repository) so you can examine and/or modify files.
You will need the URL of the repository (or sub-directory) to be able to check it out. All Neptune
SVN URLs will begin with http://SVN.neptuneinc.org/repos followed by the repository name. So
if I wanted to check out the entire Neptune repository (not recommended, as it is very large), I
would use the URL http://SVN.neptuneinc.org/repos/neptune.
5.4.1 Trunk, Branches, and Tags

Most repositories have three top-level directories called trunk, branches, and tags. The trunk
represents the main line of work in the repository – the branches and tags folders have specialized
uses, which will be discussed later (they are mainly relevant to programmers). When someone asks
you to check out the “project1” repository, and that repository has a trunk, the URL you will want
to use is http://SVN.neptuneinc.org/repos/project1/trunk. However, the name of the directory you
will create to check the files out into should be called project1, so you will know what repository
you are working with.
5.4.2 Checking Out

Once you have been given the URL of the repository you want to check out, you will enter that
URL into your subversion client as part of a “checkout” operation. Depending on your client, you
may need to create the containing directory first, or the client may do it for you if you indicate a
directory that does not yet exist. Either way, the files you have requested will be copied from the
SVN server to the location you have specified. Subversion does NOT CARE where on your
machine you chose to store your files. Subversion keeps hidden “metadata” folders inside each
folder of your working copy. One of the things these metadata folders keep track of is what URL
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 5
 of
 10

Revision
 0

on the server the current directory corresponds to. This means that you can move the location of
the working copy on your computer, and this will not affect subversion at all – it still knows where
to go on the server to get updates for those files, or commit changes to those files.

If the repository is large, and especially if you are not in the Los Alamos office where the SVN
server resides, this initial checkout could take a long time. Your client will show you a running
progress display, usually listing each file that is pulled down from the server. If the listing seems to
get “stuck” on a particular file, that probably means that the next file in the list is very large, as the
files are not listed until their download is complete. Occasionally, you will some kind of “timeout”
error message during a long checkout. In this case, it almost always works to simply update your
working copy to get the rest of the files (see the next section for updating).
5.4.3 Updating

As time passes, other team members may make changes to files in the repository you have checked
out. The only way for you to see these changes is to update your working copy of the repository.
Your SVN client will allow you to select any directory or file in a working copy and request that it
be updated. Usually, you will want to pick the top-level directory, so you can get all the updates at
once. As with checking out, your client will give you a listing of files, but in this case it will only
be files that have versions newer than the one you already have in your working copy. If nothing
has changed, you will see a message confirming that your working copy is already at the latest
version, for example “at revision 258.”
5.4.3.1 Conflicts

If you have changed a file in your working copy, and someone else has changed the same file in their
working copy and committed (uploaded) their change back to the server, you may get a conflict
notification. If the file is plain text, and the changes in the repository are in a different part of the file
than the changes you made, you will see a notification that those changes have been merged into
your version of the file (there will be a G after the file name in the list of changes). However, if your
text changes conflict with the changes from the repository, or if the file is a binary file, you will get a
conflict. We will talk about resolving conflicts later in this document.
5.4.4 Committing

When you have made changes to one or more files and want to publish those changes back to the
repository, you need to commit them. Your SVN client will allow you to select a file or directory
and issue the commit command. The client will show you a list of the changed files it found, and
offer you the option of unselecting any files that might have changes you are not ready to commit.
It will also provide you a space to enter a comment describing the changes made to the file(s) in
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 6
 of
 10

Revision
 0

question. It is critical that a meaningful comment always be filled in. This requirement will be
discussed in more detail later in the document.
5.4.4.1 Why Commits Can Fail

The main reason that a commit will fail is if one of the files to be committed is not the latest
version from the repository. Subversion will not allow you to potentially overwrite someone else’s
changes. For example, you cannot commit a file that is based on an earlier version than the latest
version from the repository. When a commit fails for this reason, the only thing to do is to update.
If the file is a text file, you may find that the changes in the repository are simply merged into your
file. However, the most likely scenario is that you will get a conflict, which you will then have to
resolve (see Resolving Conflicts later in this document).

Practically speaking, this means that just before you begin editing a file, you need to do an update
to make sure you have the latest version. Also, if the file is binary (e.g. a MS Word document),
you will want to let other members of your team know that you are editing the document, so that
they won’t start editing in parallel. Of course, for large documents, there are strategies that allow
for editing files in parallel when you know that your changes will not conflict with your
colleagues’ (for example when two people are editing different sections of the document). These
strategies will be discussed later under the Workflow section.
5.4.4.2 Reverting Changes

Sometimes you may be working on a file and wish to discard all your changes an return to the base
revision from the repository. This might happen if you were to realize that you had been modifying
the wrong file, or for a variety of other reasons. The revert command will discard all local changes
and restore your working copy with a “pristine” version of the last version of the file or files you
checked out.

Sometimes reverting is the best way to resolve a conflict. You can always save your version of the
changed file to a different location and then revert the conflicted file. This will give you the latest
file from the repository, and allow you to examine that file and see how it differs from yours, so
you can incorporate your changes into the new version.
5.4.5 Adding New Files or Folders

Generally, there are two kinds of new files we add to a repository. The first are new Neptune-
created files, which may become work products or simply supporting project information. In these
cases, it is REQUIRED to enter a comment describing the purpose of the file and perhaps its initial
content.

N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 7
 of
 10

Revision
 0

The second type of files we add to repositories are files received from outside sources – reports,
data, communications from clients, meeting minutes, etc. In these cases it is CRUCIAL that the
comment contain as much detail as possible about the provenance of the file. Being able to track
down exactly where we got the file and from whom is crucial to the QA process. So the comment
“adding new Eco data” is fairly useless, whereas “adding new mammal field data received from
Brett Tiller via email on 7/21/2008” gives us solid backward traceability to the source of the data.

If you create a new file or folder inside a directory that is part of your working copy, it has no
effect on the repository until you first add the file to the working copy and then commit that
addition. Most GUI clients allow you to combine these operations by including new files in the list
of changes when you begin the process of committing a directory. New files will usually appear
with a question mark next to them. If you check the box next to a new file, you are telling the
client program to first add the file to the containing directory and then include that addition in the
final commit operation. Some GUIs will have a check box that allows you to toggle whether or not
new files are shown in the commit list.
5.6 Subversion Workflow
This section describes the workflow process involved in using Subversion.
5.6.1 Repository Creation

A repository can be created at any time by a member of the IT staff. Repository names must
conform to the following requirements (not that not all existing repositories conform):
-­‐ all lower case
-­‐ no spaces – use underscores instead
-­‐ alphanumeric characters only – no special characters

Repositories are created on an as-needed basis. Once again, communication is key – team
members should decide if their project needs a new repository or if it best fits inside an existing
repository.

The structure of the files within the repository is also a team decision. Several templates have been
used on different types of projects. Specific template examples may be made available in the future
to use as starting points for new projects.
5.6.2 Working with Existing Repositories

You always have the option to check out an entire repository, or just a subsection of a repository.
The only difference between the two is the URL that is passed to the checkout command. To check
out an entire repository, your URL will look like this:
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 8
 of
 10

Revision
 0

http://SVN.neptuneinc.org/repos/repository_name/trunk
or, in the case of a repository with no trunk,
http://SVN.neptuneinc.org/repos/repository_name
If you only want to check out a sub-section of the repository, you simply include the path to the
sub-section in your URL. Here is an example of how to check out just the QA folder (containing
the new company QA plan documents) from the Neptune repository:
http://SVN.neptuneinc.org/repos/neptune/trunk/QA
This way you only get a folder with three documents rater than an entire repository.
5.6.3 Repository Browsing

Many of the GUI clients include a feature that allows you to “browse” the repository on the server.
By entering the base URL of the repository (for example,
http://SVN.neptuneinc.org/repos/neptune) in the browser window, you can view the structure of
the repository as it is on the server without having to download anything. This is a great way to
figure out what you might need to check out for a given purpose. For example, the browser will
show you that under the trunk of the Neptune repository there is a Business Development folder,
which in turn contains a proposals folder. If you are just interested in seeing the proposal work
done for DOD, you can just check out the DOD folder from inside the proposals folder. Most
repository browser GUIs allow you to select a sub-folder from within a repository and ask to check
it out. At worst, you can use the browser view to see how to build the URL you will need to check
out the sub-folder you are interested in.

One thing that a repository browser GUI will NOT do is allow you to see all the different
repositories on the server. To see a list of all repositories, visit to the password-protected web page
at http://repositories.neptuneinc.org/index.php. You can get the username and password from one
of the IT staff.
5.6.4 Making Changes

There are three kinds of changes you can make to a repository:
1) Modify existing files in a repository
2) Add new files to a repository
3) Reorganize the structure of a repository
5.6.4.1 Modifying Existing Files

As noted earlier (Section 5.4.1), to make sure that you are working on the latest versions, always
do an update before you begin modifying files. Also, especially in the case of binary files, notify
other team members that you will be modifying the file(s).
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 9
 of
 10

Revision
 0

5.6.4.2 Using Locks to Enforce Serial Editing of Binary Documents

The best way to avoid conflicts when editing files is to use subversion’s locking feature. Both
svnX on the Mac and Tortoise on Windows give you access to this feature. Locking a file is
simple. First be sure you have the latest version of the file by running an update. Then use the GUI
(or command line) to invoke the lock command (you will get an error message if a more recent
version of the file exists in the repository). Once a file is locked, no one else can commit changes
to that file – they will receive an error when trying to commit, telling them the file is locked and
the name of the user who has the lock.

Therefore, when editing a binary file, one should ALWAYS lock the file first. If someone else
already has the file locked, you will get an error with the lock owner’s username, and you know
that you need to wait for that team member to finish his or her edits before you can work on the
file. If you successfully gain the lock, you can be sure that no one will commit a new version that
will then cause a conflict when you try to commit yours. When you commit your version of the
file, the lock is automatically released.

In case someone locks a file and then forgets about it and goes on vacation, locks can be broken
(you may need help from an IT staff member to do this). Locks are not a strict enforcement
mechanism – rather they are a way to enhance team communication.
5.6.4.3 Editing Binary Documents in Parallel

In cases of large binary documents with many sections, team members may work on a file in
parallel, with the understanding that the different team members are working on different sections
of the file. When one team member is ready to commit their changes, they may do so, and the
other member(s) then need to update their versions. Before doing so, they should save their
versions with changes to a location outside of their working copy, or save their changes to a new
filename, perhaps with their initials appended (for example, save Report1.docx as
Report1_WH.docx). This way, before the other members update, they can revert their changes in
the repository to avoid a conflict when they updated to get their colleague’s changes (the revert
operation can also happen after the conflict – this will discard all local changes and leave the
working copy with the latest version from the repository). The next team member to finish their
edits can then copy just their section into the new version of the document and commit those
changes. As discussed in the previous section, locks can be used to enforce the order in which
changes are made to the document. Needless to say, this process requires good communication
among team members to make sure that no ones changes are unintentionally overwritten.

In all cases it is a REQUIREMENT of N&C QA policy that a comment be entered summarizing
the changes to the file as part of the commit process. This is essential to leveraging the full power
of Subversion to provide support for Quality Assurance by providing a clear trail of comments
N&C
 Internal
 Procedure

 Confidential

General
 Procedure

Standard
 Operating
 Procedure

Document
 No.

NAC-­‐0003

Revision:
 0

Document
 Status:

 Final

Title:
 Subversion
 SOP
 Page
 10
 of
 10

Revision
 0

explaining how documents evolve over time. If the project is using Bugzilla to track tasks, the
comment should include references to Bugzilla task numbers where appropriate (for more details
see the Bugzilla SOP; NAC-0004_R0).
5.6.4.4 Reorganizing the Structure of a Repository

This operation is the one most likely to lead to confusion and errors if it is done incorrectly. As
mentioned earlier in the document, each directory in a working copy keeps hidden metadata about
how it corresponds to the data in the repository on the server. This means that moving directories
around on your computer has NO EFFECT on the structure of the repository on the server. You
must move a special “SVN move” command to let the working copy know that you want to
modify the directories in the working copy by adding or removing files from the (a move operation
will delete files from one directory and add them to another). The actual effect on the repository
will not take place until you commit your changes which include the moved files.

Similarly, deleting files from your working copy will have NO EFFECT on those files in the
repository. You must use a special “SVN delete” command to let the directory containing those
files that they are scheduled for deletion. The actual deletion of the files will not take place until
you commit your changes that include the SVN deletes. It is important to realize that deleting a file
does NOT delete the file from the repository. It simply deletes the file from the latest version of the
repository. It is always possible to go back to earlier versions of the repository to “resurrect”
deleted files.

Finally, because deleting files from your hard disk does not affect the repository, this can be a
good last-ditch solution for solving SVN problems. Occasionally, the metadata in some part of a
working copy may become corrupted, leading to error messages when you try to update the
repository or delete files. You can always delete the directory to which the error message refers
and then run an update on the containing directory to get a fresh copy of the data pulled down from
the repository. Of course, if you have changed files in the problem directory or any of its sub-
directories, you should first copy the changed files to a location outside your working copy before
deleting the problem directory. Then once you have done the update to get a clean copy of the
directory, you can copy your changed files back into their appropriate locations in the working
copy, and they will once again show up as changed files that you can commit.

Receipt
 and
 Acknowledgement
 of

Neptune
 and
 Company,
 Inc.

Subversion
 SOP,
 Revision
 0

Please
 read
 the
 following
 statements
 and
 sign
 below
 to
 indicate
 your
 receipt
 and
 acknowledgement

of
 the
 Neptune
 and
 Company,
 Inc.
 Subversion
 SOP
 (NAC-­‐0003_R0).

• I
 have
 received
 and
 read
 a
 copy
 of
 the
 Neptune
 and
 Company,
 Inc.
 Subversion
 SOP.

• I
 understand
 that
 my
 signature
 below
 indicates
 that
 I
 have
 read,
 understand,
 and
 will
 adhere
 to

the
 Neptune
 and
 Company,
 Inc.
 Subversion
 SOP.

Employee’s
 Printed
 Name

Employee’s
 Signature

 Date

Quality Assurance Project Plan
11 November 2015 26
Appendix B: GoldSim Model Development SOP

Neptune and
Company, Inc.
GoldSim Model Development SOP Page 2 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 2
CONTENTS
1.0 Introduction .............................................................................................................................5
2.0 Modeling Lifecycle .................................................................................................................5
2.1 Model Objectives and Context ..........................................................................................5
2.2 Conceptual Model Development .......................................................................................7
2.3 Model Requirements Evaluation .......................................................................................7
2.4 Verification of Software Installation .................................................................................7
2.5 GoldSim Model Development ..........................................................................................7
2.6 Model Data Inputs .............................................................................................................8
2.6.1 Input Data Selection ....................................................................................................8
2.6.2 Input Data Placeholders ...............................................................................................8
2.6.3 Data Acceptance Criteria .............................................................................................8
2.6.4 Records of Parameter Values ......................................................................................9
2.6.5 The Parameter List ......................................................................................................9
2.6.6 Check Prints ..............................................................................................................10
2.7 Model Evaluation ............................................................................................................10
2.7.1 Scientific Basis ..........................................................................................................11
2.7.2 Computational Infrastructure ....................................................................................11
2.7.3 Assumptions and Limitations ....................................................................................11
2.7.4 Peer Review ...............................................................................................................11
2.7.5 Quality Assurance and Quality Control ....................................................................11
2.7.6 Data Availability and Quality ....................................................................................11
2.7.7 Comparison with Analytical or Empirical Solutions ................................................11
2.7.8 Benchmarking against Other Models ........................................................................12
2.7.9 Corroboration of Model Results with Observations ..................................................12
2.7.10 Sensitivity Analyses ..................................................................................................12
2.7.11 Reasonableness Checking .........................................................................................12
2.8 Model Review .................................................................................................................13
3.0 Model Documentation ..........................................................................................................13
3.1 Documentation Components ...........................................................................................13
3.2 Model Element Note Panes .............................................................................................14
4.0 Model Configuration Management .......................................................................................14
4.1 Model Custody ................................................................................................................14
4.1.1 Experimental Module Development .........................................................................15
4.1.2 Criteria for Making Changes .....................................................................................15
4.2 Documentation of Changes .............................................................................................16
4.2.1 Version Change Notes ...............................................................................................16
4.2.2 The Change Log ........................................................................................................17
4.3 GoldSim Versioning ........................................................................................................18
4.3.1 Model Version Numbers ...........................................................................................19
4.4 Model Testing .................................................................................................................21
4.5 Model Backup .................................................................................................................21
4.6 Error Reporting and Resolution ......................................................................................21
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 3 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 3
4.6.1 Reporting Error Candidates .......................................................................................21
4.6.2 Assessing Error Candidates .......................................................................................22
4.6.3 Resolving Errors ........................................................................................................22
4.6.4 Error Resolution Verification ....................................................................................22
4.6.5 Error Impact Assessment ...........................................................................................22
4.7 Model Distribution ..........................................................................................................22
5.0 References .............................................................................................................................24

Neptune and
Company, Inc.
GoldSim Model Development SOP Page 4 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 4
FIGURES
Figure 1. Model development work process flow diagram .............................................................6
Figure 2. GoldSim provides for annotation regarding any change in an element's definition
through the Version Change Note ................................................................................17
Figure 3. The model’s Change Log can be maintained using a formatted text box ......................18
Figure 4. GoldSim's Version Manager ..........................................................................................19

Neptune and
Company, Inc.
GoldSim Model Development SOP Page 5 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 5
1.0 Introduction
This standard operating procedure (SOP) describes the development of GoldSim-based computer
models. These models are used to perform contaminant transport and dose assessment
calculations as the computational basis for radiological Performance Assessments (PA). They are
developed using the GoldSim systems analysis software, developed by the GoldSim Technology
Group (GTG), as a principal platform, commonly in conjunction with various supporting
computer programs and data sources. Throughout this document, the term Quality Assurance
(QA) refers to a program for the systematic monitoring and evaluation of the various aspects of
GoldSim model development to ensure that standards of quality are being met.
2.0 Modeling Lifecycle
GoldSim model development follows a structured process or lifecycle that requires a graded
approach to QA at each phase. The lifecycle for GoldSim model development is described below
and correlates with the work process shown in Figure 1. Model documentation is associated with
each step of the work process.
2.1 Model Objectives and Context
The regulatory modeling process is seen by the National Research Council (NRC, 2007) as
beginning when “…decision makers, model developers, and other analysts must consider
regulatory needs and whether modeling could contribute to the regulatory process.” With
consensus on the value of developing a model the next step is to specify the objectives and
context of the GoldSim model.
Defining the objectives of the model includes establishing who will use the model, what
decisions the model will be designed to support and what model calculations are required to
support these decisions.
Model context includes components such as the following (NRC, 2007):
• Determination of spatial and temporal scales,
• Determination of the appropriate level of detail for process representation,
• Identification of the proposed users, their expertise, and any constraints,
• Determination of sources and required quality of input data,
• Determination of sources and required quality of data for model evaluation,
• Definition of the inputs and outputs needed and whether they will be deterministic or
probabilistic,
• Determination of the level of reliability required, and
• Determination of appropriate evaluation criteria required to demonstrate that the model is
sufficiently accurate for its intended use.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 6 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 6

Figure 1. Model development work process flow diagram
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 7 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 7
Model objectives and context are documented in the requirements document described in Section
2.3.
2.2 Conceptual Model Development
Model development continues with the development of a Conceptual Site Model (CSM). The
CSM identifies important features and processes of the system being modeled that are consistent
with the existing data. Development of the CSM along with the model objectives and context
form the basis for the GoldSim model design.
The CSM is documented in a Conceptual Site Model document, which explains and provides
justification for the mathematical approaches for modeling geological, hydrogeological,
contaminant fate and transport, demographic, and other component processes of the overall
model. Existing data and literature and expert opinion are used to support the modeling approach
described by the CSM.
2.3 Model Requirements Evaluation
The CSM provides information to determine the attributes and capabilities of the software
required to meet the project objectives. These requirements and those determined in the
definition of the “model objectives and context” step are compiled in a model requirements
document. Model requirements also include consideration of modeling objectives to determine
the reliability, certainty, and accuracy needed in predicting the performance measures for the
decision process. This evaluation also includes a review conducted to verify that the GoldSim
modeling platform is capable of providing these required attributes and capabilities. The model
requirements document is archived in the project repository.
2.4 Verification of Software Installation
The GoldSim software is installed and registered as described in the GoldSim User's Guide
(GTG 2010a et seq.). Following the installation and registration the user runs the example model
“FirstModel.gsm” located in the “General Examples” directory and verifies that the output
obtained matches the chart shown on page 26 of the User's Guide (GTG 2010a et seq.).
The GoldSim User's Guide (GTG 2010a et seq.) and the GoldSim Contaminant Transport
Module User's Guide (GTG 2010b et seq.) provide complete descriptions of the features and
capabilities of GoldSim and the Contaminant Transport Module.
2.5 GoldSim Model Development
During model development individual modelers work in parallel to model specific sub processes
described in the CSM. For example, existing mathematical models are translated into specific
algorithms to be used in the modeling process. GoldSim offers a level of model structure that can
closely resemble a conceptual model, so the structural implementation of the GoldSim model
will follow the CSM developed by the project team. As the different components of the model
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 8 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 8
are developed in GoldSim, they are integrated to form a coherent representation of the overall
process being studied. GoldSim's object-oriented structure facilitates this process, often allowing
independently developed sub-modules to be copied and pasted into the main model. GoldSim's
“self-documenting” features allow the graphical user interface (GUI) design to incorporate
documentation of modeling concepts and parameter derivation, so that it is relatively easy to
crosswalk between individual GoldSim pages and sections of the CSM document.
2.6 Model Data Inputs
2.6.1 Input Data Selection
The development of appropriate definitions of input parameters is guided by model sensitivity
analyses, which identify those parameters most important in determining the model results. In
some cases, the definition of an input value matters little to the results and in these cases less
effort is expended in developing distributions. Sensitive parameters, however, warrant a closer
investigation, and their input distributions are devised with great care where possible. All
parameters in the model are based on some sort of information source, be it a “literature value,”
the result of a site-specific data collection campaign, or the result of expert professional
judgment.
2.6.2 Input Data Placeholders
On occasion, a modeling element must be added to the model in order to proceed with
construction, but no value has yet been developed. In this case, an ad hoc placeholder value is
chosen so that model development may continue, and the parameter is noted as a placeholder.
Before the model can be relied upon for any purpose, however, all such placeholder values must
be replaced with suitably-derived and documented values.
2.6.3 Data Acceptance Criteria
The sources of input data for the model are various, and the quality of the source is a
compromise between model sensitivity (identifying the need for high-quality data), availability,
appropriateness, and the ability (budget and/or practicality) to generate data of sufficient quality.
Input parameters that have a strong influence on the model results as determined by sensitivity
analyses are given higher priority than those with little influence.
The choice of data sources depends on the availability and application of the data in the model.
The following hierarchy outlines different types of information and their application. The
information becomes increasingly site-specific and parameter uncertainty is generally reduced
moving down the list.
• Physical limitations on parameter ranges, used for bounding values when no other
supporting information is available. Example: Porosity must be between 0 and 1 by
definition.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 9 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 9
• Generic information from global databases or review literature, used for bounding values
and initial estimates in the absence of site-specific information. Example: A generic value
for porosity of sand is 0.3.
• Local information from regional or national sources, used to refine the above
distributions, but with little or no site-specific information. Example: Sandy deposits in
the region have been reported to have porosities in the range of 0.30 to 0.37, based on
drilling logs or reports.
• Information elicited from experts regarding site-specific phenomena that cannot be
measured. Example: The likelihood of farming occurring on the site at some time within
the next 1000 years is estimated at 50% to 90%.
• Site-specific information gathered for other purposes. Example: Water well drillers
report the thickness of the regional aquifer to be 10 to 12 meters.
• Site-specific modeling and studies performed for site-specific purposes. Example: The
infiltration of water through the planned engineered cap is estimated by process
modeling to be between 14 and 22 cm/yr.
• Site-specific data gathered for specific purposes in the models. Example: The density of
Pogonomyrmex ant nests adjacent to the site is counted, and found to be 43 nests per
hectare.
The determination of data adequacy is informed by a sensitivity analysis of the model, which
identifies those parameters most significant to a given model result. Such parameters are
candidates for additional measurements or more deliberate estimation. As the model
development cycle proceeds, sensitive parameters are identified and their sources are evaluated
in order to determine the cost/benefit of reducing their uncertainty.
2.6.4 Records of Parameter Values
One limitation of the GoldSim platform is that there is no straightforward way to examine all the
values of inputs (data and stochastic elements) in one place. The user must search the model and
open (or “mouse-over”) each input element individually in order to see its value. In order to
overcome this inconvenience, all the parameter inputs are stored external to the model, in the
Parameter List document.
2.6.5 The Parameter List
The Parameter List is a complete list of the input parameters for the model, and may consist of a
text document, a workbook of spreadsheets, a database, or a combination of these, depending on
the changing capabilities of the GoldSim modeling platform. Each parameter is listed in only one
place, so that there is no ambiguity about the proper value of a parameter. Accompanying the
listing of the parameter value in the Parameter List is a traceable reference to its origin, which
may be in a White Paper or literature reference. Any change to a parameter is made to the
Parameter List first, and then the change is made to the model: The value in the Parameter List is
cross-checked to its source via a check print (see below), and the value in the model is then
changed, noted in the Version Change Note for the modified element, and in the Change Log.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 10 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 10
2.6.6 Check Prints
Whenever information (e.g. a parameter distribution) is transferred from one record to another
(e.g. from a White Paper to the Parameter List) a QA Check Print process is invoked. This
process is intended to positively and unambiguously document the source of information for
each model input parameter or distribution.
The flow of information is from primary sources (field data, literature, expert elicitations, etc.) to
White Papers that develop the input distributions (this step may not apply to all cases, and may
have supplemental calculation sheets), to the Parameter List to the GoldSim model. QA check
prints are maintained in all but the final step—transferring input values to the model. The check
print process consists of obtaining paper copies of the data source and its destination, such as a
paper from the literature and the Parameter List, for example. A comment field in the Parameter
List (either a column in a table, a comment attached to a spreadsheet cell, or other location
unambiguously associated with the data) identifies the value’s origin. A paper copy of that page
or pages of the Parameter List is attached to a paper copy of the data source (which may be
simply the page from the identified source), and the QA reviewer annotates each page. Typically,
a yellow highlighter is used to indicate each positively-checked value, and a red pen identifies
any value that does not match. After checking each value against its source, the check print is
documented with the date and the signature of the checker. Errors discovered in the process are
noted, the errors are corrected in the destination document, and the values are rechecked with a
subsequent check print, which is attached to the original. This process is repeated until the check
prints can document that information transfers are error-free. Check prints are stored as hard
copy at Neptune.
The final step of information transfer—from the Parameter Document to the GoldSim model—
does not lend itself to paper check printing. However, traceability of parameter information can
be maintained using GoldSim’s internal QA tools, such as Note Panes and Version Change
Notes discussed in Sections 3.0 and 4.0.
2.7 Model Evaluation
Evaluation of the proper operation of the Model is done on two levels. The overall model, as
represented in the results, is subjected to benchmarking with process model results if a process
model is available, and is compared to previous versions of the Model to assure that incremental
changes are in line with those expected from modifications to the Model. On a submodel scale,
particular parts of the Model may be evaluated independently. Many computer models that
attempt to predict the outcomes of processes and events can be validated (verified) with
measurable results (e.g. environmental media concentrations). Due to the nature of performance
assessments, which attempt to estimate concentrations and fluxes of materials in environmental
media and the potential doses or risks resulting from exposure to those materials far into the
future, the results are not amenable to this type of validation. It is not possible to “test” the model
at a system level to see if it has done a good job of predicting the dose to a hypothetical
individual 10,000 years from now. GoldSim model evaluation includes elements recommended
by the NRC (2007) that provide the evidence used to demonstrate that the models are sufficiently
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 11 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 11
accurate for their intended use. These model evaluation methods used by Neptune are described
in the following sections.
2.7.1 Scientific Basis
Documentation of the theory represented in the models in the topical White Papers demonstrates
the sound scientific basis for the models.
2.7.2 Computational Infrastructure
Verification of the computer program GoldSim is provided in the GoldSim Verification Plan
(GTG, 2010c et seq.). The definition of verification in the Plan is taken from Sandia National
Laboratory (1995): “The process of demonstrating that a computer software program performs
its numerical and logical operations correctly.”
2.7.3 Assumptions and Limitations
Detailed descriptions of the assumptions and limitations of the models necessary for model
evaluation are documented in the model or in the associated White Papers.
2.7.4 Peer Review
Peer review described more fully in Section 2.8 below is a documented review of the model and
its application to determine if the model is sufficiently accurate for its intended use, properly
documented, and meets specified quality assurance requirements.
2.7.5 Quality Assurance and Quality Control
The model development process is conducted under a documented QA plan that includes training
and assessment of implementation of QA processes through internal reviews by staff not
associated with the project.
2.7.6 Data Availability and Quality
The availability of data and the quality of data described in more detail in Section 2.6 are
important elements in the model evaluation process.
2.7.7 Comparison with Analytical or Empirical Solutions
To evaluate the reasonableness of the results of a particular algorithm, the modeler may set up
equation(s) both as an element in GoldSim and also using another tool, such as a Microsoft Excel
Spreadsheet. This allows the modeler to compare results using two different calculation methods
to provide a higher level of confidence that the algorithm has been implemented correctly in the
GoldSim model.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 12 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 12
2.7.8 Benchmarking against Other Models
Benchmarking consists of reproducing the deterministic results of the process model calculations
using an established process model and GoldSim. This “benchmarking” is a fundamental high-
level corroboration of the model implementation and calculations. Agreement between the two
models serves to build confidence in the validity of the GoldSim model. Model benchmarking is
documented using spreadsheet based test plans. Test plans include documentation of
• Prerequisite conditions
• Inputs
• Assumptions and constraints
• Software requirements
• Test descriptions
• Requirements—tests crosswalk
• Test log
The test plan includes the rationale for the plan, a review of the test results and identifies
reviewers and their specific responsibilities.
2.7.9 Corroboration of Model Results with Observations
Field and laboratory data applicable to the model predictions are rarely available but for some
long-term processes geologic data may be useful for corroboration.
2.7.10 Sensitivity Analyses
A sensitivity analysis (SA) is conducted by Neptune for system models such as performance
assessments and radiological risk assessment models constructed in GoldSim. The goal of the SA
is to determine which explanatory variables have the largest impact on specific endpoints of
interest. That is, the SA provides the overall contribution of each input parameter to the model
output. A global approach is used where all input parameters are essentially varied
simultaneously. SA results are used for model evaluation leading to better understanding of
model constructs and modifications to the model structure. In addition, if there is an
unacceptable level of uncertainty associated with an endpoint of interest (for decision making
purposes), the sensitive parameters can be targeted for effective uncertainty reduction; that is,
further data or information should be collected to reduce the uncertainty on these sensitive input
parameters.
2.7.11 Reasonableness Checking
A model can incorporate several tools for checking the reasonableness of certain inputs and
results. Examples follow:
• Intermediate results are provided where they are useful for checking calculations.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 13 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 13
• Mass balance checks demonstrate that the mass of materials (soil, water, air) and
radionuclides is preserved. This is a fundamental requirement of physical environmental
models (GTG 2010c et seq.).
2.8 Model Review
Model development is subject to review by a modeler different from the one who did the original
model building. As parts of the model are revised, with changes in parameters, expressions, or
other functional elements, or model structure, these changes are reviewed for accuracy and
completeness. Any accompanying text on the model pages is also reviewed for clarity and
accuracy. The modeler making the changes identifies which parts of the model are subject to
review, and another Neptune GoldSim modeler examines these in detail, providing review
comments to the originating modeler. The entire model is subjected to review before release to
the client (see Section 4.7). This review is documented and archived in the project repository.
3.0 Model Documentation
3.1 Documentation Components
GoldSim models are documented both internally and externally. Internal documentation includes
the Change Log, Version Change Notes, modeling element Note Panes, and GoldSim’s internal
versioning capability. External documentation includes White Papers, check prints, and a
Parameter List. White Papers document the development of specific algorithms and other inputs
to the GoldSim model and are intended to explain and justify the approach taken.
The scope of a given White Paper is normally concentrated on a specific topic such as
engineering or processes within a subject matter area such as biotic transport. The White Paper
contains more detailed discussion of literature and available data describing mathematical
representations of the CSM in the model. The White Paper is used to document the source of the
data used for model parameters and the methods used for developing statistical distributions for
parameters. White Papers are also used to document the model used for a parameter or a process
in the model if it is not a part of the GoldSim platform and explain how the process is
implemented in the GoldSim model.
A typical page in the model consists of model elements and explanatory text. Illustrations such as
drawings or photographs may also be used. Each page represents a modeling concept, and the
model is logically divided into parts that will fit onto pages. Text at the top of each page explains
the function of the page, and text juxtaposed with the model elements explains the function of
the element. Each element also has a description field that is used for a short descriptive
identifier. Additional details may be provided in each element’s Note Pane.
The influence of one model element on another can be easily traced through the model using the
“Show All Links” function attached to the triangle-shaped arrows on each side of the element
graphic. The left triangle, pointing into the element, shows the other elements referenced by the
current one, and the right triangle, pointing out of the element, shows the other elements that are
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 14 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 14
dependent on it. By following these links, the complete interdependency of elements can be
traced through the model.
3.2 Model Element Note Panes
Associated with each GoldSim modeling element is the optional Note Pane feature. If an element
has a note, it is identified by an underlined element name. Note panes have a dual purpose in the
model. They are used for general information, describing the purpose of a container or element.
They also serve the QA process, as a convenient place to make notes about the source of
information or the status of QA review. While most of the note pane is free-format, the QA
related notes are to include a date (which can be cross-indexed to a version number using the
Change Log, described below), the name or initials of the person making the note, and a
description about the nature of the QA check. For example, a QA note for an entire container
might read:
1 Apr 05 JT QA for this container completed
13 Apr 05 KC QA updated with cross-check of water tortuosity exponent
parameter values
and one for an individual element:
13 Aug 04 KC Verified source of these data: Each value was checked against
the 15th edition of the Chart of the Nuclides (General Electric Co.
and Knolls Atomic Power Laboratory, 1996), wall chart version.
28 Sep 04 JT Updated and verified source of these data: Each value was
checked against the 16th edition of the Chart of the Nuclides
(General Electric Co. and Knolls Atomic Power Laboratory, 1996),
booklet version.
If an element is actually changed in the process of a QA review (or for any other reason), such
change is noted in the Version Change Note associated with that element. This is part of
GoldSim’s internal QA process, and the text of all Version Change Notes is kept in an internal
database of changes, indexed to model versions.
4.0 Model Configuration Management
Managing the model configuration through its various versions is critical to the production of a
usable modeling product that meets client requirements. The following sections discuss various
topics relevant to model modification and control.
4.1 Model Custody
During model development, the baseline model is tracked by the lead modeler. In the event that
another modeler needs to have custody of the model for development purposes, the custody will
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 15 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 15
be passed to that modeler and returned when the work is finished. The current custodian is
always known, and is recorded on the topmost page of the model (except in released versions).
Modelers make use of the internal GoldSim versioning and Change Log in order to document
changes made to the model.
A GoldSim model differs from many other software development projects in that it exists in a
single binary file (with the “.gsm” extension). There are no separate files for subroutines as in a
more traditional programming language like C, FORTRAN, or even Java. Therefore, the model
cannot be edited by more than a single person at a time. At any given time, there is a single
“main” model file. The custody of the main model must be explicitly passed from the lead
modeler to another, and the custody is always known by the lead modeler, who is also the default
custodian. The lead modeler may assign custody to another for a particular modeling task, but
will resume custody when that task is completed. Upon return of custody, the returned model is
inspected and one of two paths is chosen: 1) The returned model is maintained as the baseline
model, or 2) the baseline model is modified appropriately to incorporate changes made in the
returned model, and the modified baseline model is retained as the new baseline model. The
baseline model resides on the custodian’s computer, and is backed up by several methods,
including off-site media (see Section 4.5).
4.1.1 Experimental Module Development
On occasion, model development requires some experimentation that may not be desirable in the
main model. In such cases, a copy of the main model is made and given a unique file name in
order to keep it distinct from the main model. This “branch copy” is used for module
development and prototyping of modeling methods. Once the prototype of a specific module is
complete, tested, and accepted, the new model parts are re-integrated into the main model, either
by copying model containers and elements from the branch copy into the main model (the
preferred method), or by re-entering elements directly into the main model in cases where
GoldSim will not allow copying between model files. Either way, the additions and/or changes to
the main model are cross checked for accuracy (by a modeler other than the one implementing
the change), and the modifications are noted in the Change Log (see Section 4.2.2). At all times,
however, there is only one main model file.
4.1.2 Criteria for Making Changes
Changes to the model occur at different levels. Minor changes to internal documentation
language, including clarifications of text and correction of inconsequential typographical errors,
are made as they are identified, and without formal documentation. Changes involving any type
of data input or calculation that could potentially affect the modeling results are documented in
all affected supporting documents (calc sheets, white papers, etc.) and in the model’s Change
Log.
A change to an input parameter (e.g. a distribution) may be precipitated by the following:
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 16 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 16
• QA review, in which model parameters are found to not match their values as
documented outside the model. In such a case, the value and its QA records (e.g. check
prints) are reviewed, and errors are corrected as appropriate.
• A decision by a subject matter expert (SME), generally in consultation with other project
team members, that a value should be changed for some technical reason, such as the
availability of new data on which a distribution is based. This would be considered an
update, and the change would cascade through the proper sequence, from an update to the
data set, through development of an updated distribution, updating of the documentation
in a White Paper (if applicable) and in the Parameter List and finally an update to the
model itself, with an accompanying entry in the Change Log and in the parameter
element’s Version Change Note (see Section 4.2.1). Each step in the change sequence is
reviewed by an individual other than the person implementing the change.
• Major changes to the model, such as changing the GoldSim Species list, adding a
contaminant transport process, a waste configuration, or an exposure scenario, are
discussed and planned by Team SMEs.
4.2 Documentation of Changes
The documentation of changes made to the model is done at a level appropriate to the changes. If
individual parameters are modified or added, this is documented with a note provided in the
model element’s Version Change Note, referencing the nature of the change, who made the
change, and date of the change. The name of the changed element is noted in the Change Log,
along with the model version number, date of the change, the name of the person executing the
change, and the name of the reviewer of the change process. Such changes may also be noted in
the element’s Note Pane or that of its container.
4.2.1 Version Change Notes
Version Change Notes (Figure 2) are automatically attached by GoldSim to any model element
that has been modified, and are used to store information about changes in any particular
element. GoldSim keeps a versioning database within the Model, consisting of a list of all
changes to the model between version-stamps, and the text supplied in the Version Change
Notes. At any time, GoldSim can generate a report of changes made between versions. Once a
model version number has been incremented, all Version Change Notes are “reset” and a new set
begins for that version. Any information that is to be maintained through versions for viewing by
users or reviewers, such as QA reviews, is kept in the Note Panes associated with model
elements or containers. Any time an element is edited, a log entry is generated internally by
GoldSim documenting the event. Note that this happens even if nothing is actually changed in
the element when the “OK” button is chosen in the dialog. Use of the “Cancel” button does not
signal a change.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 17 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 17

Figure 2. GoldSim provides for annotation regarding any change in an element's definition
through the Version Change Note
4.2.2 The Change Log
Neptune’s GoldSim Models have a Change Log, in a block of paragraph text in the
Documentation container, as shown in Figure 3. This log is maintained by the modelers, and
documents when a change was made, who made it, the model version number, and descriptive
details. Modifications that could potentially change modeling results are noted to the level of the
element changed, with more detail included in the element’s Version Change Note or Note Pane.
Modifications to explanatory text and changes to diagrams and other supporting material are
noted in broad terms, such as “Modified figures depicting waste cell geometries.” Typographical
corrections are generally not noted.
All of these documentation techniques are used in model development. If a change was made to
the model, or if part of the model was reviewed, this will be noted in the Change Log. A note
regarding the QA review (and details, if necessary) will be made in the element's note pane or in
its container's note pane. The container's note pane is appropriate if there are many similar
elements in the container. If a change is made to an element, either from a QA review or for
another reason, GoldSim will automatically provide the element with a Version Change Note,
which is used for recording the change.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 18 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 18

Figure 3. The model’s Change Log can be maintained using a formatted text box
4.3 GoldSim Versioning
Introduced specifically as a model QA feature, GoldSim has model-level and element-level
versioning built in to the Version Manager.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 19 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 19
4.3.1 Model Version Numbers
At the model level, illustrated in Figure 4, version numbers are incremented at the modeler’s
discretion. The model version number is incremented as described below. GoldSim keeps track
of changes made to the model in any given version, and can generate a report of changes made.

Figure 4. GoldSim's Version Manager
Neptune GoldSim models use versioning at two levels: Release versions and development
versions. Major revisions to the model, resulting in planned releases, generally proceed in
increments of X.Y, with a change in X signifying a more significant model evolution than a
change in Y. The assignment of these values is subjective, and may be decided upon in
coordination with the client.
Model development uses GoldSim’s minor version definition, which increments the Y in the
three digits following the decimal point. For example, development following the release of
version 2.1 starts with version 2.101. After making some changes to the model, a modeler
decides to preserve the incremental version. At this point, the version number is incremented to
2.102 and the work proceeds, with 2.101 being archived.
Day-to-day and hour-to-hour development versions are noted with letters appended to the
version number, such as 2.010a, 2.010b, etc. This is done so that during the process of editing the
model, any change can be easily undone. When a specific modeling task is accomplished, the
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 20 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 20
model is saved with the next letter in the sequence. As the changes are tested and accepted, the
letter suffixes are dropped, and these intermediate versions are generally not archived. If a
problem is found during testing of daily builds, or if the model file becomes corrupted, then the
modeler can easily revert to a previously saved version of the model file and rebuild the part that
caused the problem. This is preferable to attempting to “undo” the work, which takes time, can
be prone to error, and clutters the internal versioning record.
4.3.1.1 Incrementing the Version Number
The following example illustrates the documentation of incrementing development versions, as
recorded in the Change Log:
1) Make a final entry in the Change Log under version 1.034 that you are incrementing the
version number:
29 Jun 02 1.034 JJ Versioning counter updated to 1.034, and model
saved.
2) Immediately change the internal versioning to 1.034 using “Model | Versioning...” (see Figure
4)
3) Save the model as "name v1.034.gsm", (any name plus the version number) overwriting all
previous versions of that name.
4) Change the file attributes to “read only” so that the model file will not be inadvertently
overwritten.
5) Change the front page and the Change Log entries to 1.035.
29 Jun 02 1.035 JJ Begin work on v1.035.
6) Save the model as "name v1.035a.gsm" (or similar)
7) Begin work on version 1.035, starting at intermediate development version 1.035a.
8) After developing using intermediates 1.035a, 1.035b, 1.035c, etc., determine when to save the
model as 1.035, and return to step 1) using the new version number.
4.3.1.2 Creating a Versioning Report
A report can be generated from GoldSim (using the “Generate Report...” button shown in Figure
4), listing all changes to the model for a particular version. The report is a text file with global
changes as well as changes to individual elements, including the text from the Version Change
Notes.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 21 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 21
4.4 Model Testing
Any time a change is made to the model calculations that could change the results, the effects of
the change are assessed. Model testing is relatively easy using GoldSim, since the results of any
element in the model can be examined through a time series or final value. This enables
straightforward parallel calculations to be done in order to verify correct and consistent
operation. The modeling environment also allows the simple creation of temporary elements to
perform calculations parallel to any others in the model.
Model testing is most readily done on discrete parts of the model, where results of a small
number of straightforward calculations can be examined. Confirmation of discrete parts of the
model is done by constructing a test model in GoldSim that is focused in its analysis. Ideally, this
test model is excised directly from the main model, so that all relationships and definitions are
preserved. For example, to confirm that GoldSim is performing internal diffusion calculations as
expected, a simple GoldSim model can be constructed to examine the diffusion of materials
between various media in two cells, and the results can be compared to an analytical solution to
the diffusion equation. Calculations verified in the test model give confidence in the correct
operation in the model.
4.5 Model Backup
Preservation of electronic model files is paramount in any software development project. Several
redundant methods are employed for backup of the GoldSim model files and all other files and
documentation. Foremost are project files, maintained on a Neptune server, which are backed up
on a separate hard drive. Incremental versions of the model are likewise backed up locally and in
addition, the lead modeler keeps a copy on his/her computer, and backs that copy up to a
Neptune server. Off-site backups are also maintained.
4.6 Error Reporting and Resolution
As errors are discovered, they must be identified, reported, and resolved. This section discusses
the handling of errors in the development of a model. Formal tracking of errors, bugs, and other
issues is done using an issue-tracking system maintained by the QA manager and lead modeler.
4.6.1 Reporting Error Candidates
Errors such as inconsequential typographical errors in supporting text are not considered in this
process. Errors considered for this process include errors in parameter data entry or GoldSim
programming. If an error is suspected, it is to be reported to the lead modeler along with any
supporting information. It is the responsibility of the lead modeler to evaluate the error candidate
and see that the issue is resolved, invoking the issue-tracking system as appropriate.
Data entry errors may be discovered in input elements (Data or Stochastic GoldSim model
elements). These are also brought to the attention of the lead modeler. These or any other
modeling issues are to be entered into the issue-tracking system.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 22 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 22
4.6.2 Assessing Error Candidates
Once an error candidate has been brought to the attention of the lead modeler via the issue
tracking system an assessment is made to determine if the candidate is in fact an error. This is
usually a simple process, involving examining a mathematical expression or entered data. Real
errors are subject to resolution. False errors are identified as such, noting the resolution in the
issue-tracking system. If, however, the problem was due to some other cause, such as an
ambiguity in documentation, the causes of the identification of a false error may require
attention.
4.6.3 Resolving Errors
Errors, once discovered and confirmed, are usually easily remedied. Like other changes to the
model, fixing an error is documented at least in Version Change Notes and the Change Log.
Resolution is also noted in the issue-tracking system.
4.6.4 Error Resolution Verification
Checking the error resolution may be as simple as cross-checking an input value with the value
in the Parameters List to ensure it is correct. Alternatively, a modification to an expression may
involve an independent check of the calculation, using a spreadsheet, calculator, or a separate
GoldSim model.
4.6.5 Error Impact Assessment
Each resolved error is assessed regarding its potential effect on the results. If the effect is
anything more than negligible, its discovery and resolution are reported to the project
participants via email. Similarly, if the error could have had an effect on the results of previous
versions of the model, this is also reported.
4.7 Model Distribution
GoldSim models, like other computer model software, are open to modification. This is a benefit
for modelers and researchers, since the logic is transparent and the model is easily maintained.
This is a potential detriment to model integrity for the same reason. There are ways to tell if a
model has been modified, however, as discussed above. Versioning and the tracking of all
changes between versions are important. Nevertheless, developers and clients alike need to know
the configuration status of the model they are using, and the read-only media-released versions
always provide unambiguous starting points.
Release versions of the model(s) are ideally delivered to the client on read-only media (such as a
CD-ROM), which inherently precludes modification of the models and supporting files. Using
this method of delivery ensures that there is no ambiguity about the model and supporting
documentation that constitutes the deliverable.
Neptune and
Company, Inc.
GoldSim Model Development SOP Page 23 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 23
The GoldSim modeling software allows for complete construction and editing of models. The
companion GoldSim Player, however, is currently available at no cost and can run GoldSim
models that have been specifically “exported” as Player versions. The Player version of the
model is not editable. For distribution to the general public, a GoldSim Player version of a Model
can be provided as part of the deliverable. The Player model cannot be modified in its significant
parts, though the user can still operate switches and controls to evaluate various effects.

Neptune and
Company, Inc.
GoldSim Model Development SOP Page 24 of 24
NAC_0040_R1 Revision 1 Effective date 1 Jan 2015

2 Feb 2015 24
5.0 References
GTG (GoldSim Technology Group), 2010a, GoldSim User's Guide: Volumes 1 and 2, GoldSim
Technology Group LLC, Issaquah, WA.
GTG (GoldSim Technology Group), 2010b, GoldSim Contaminant Transport Module User's
Guide, GoldSim Technology Group LLC, Issaquah, WA.
GTG (GoldSim Technology Group), 2010c, Verification plan GoldSim version 10.50 SP1,
GoldSim Technology Group LLC, Issaquah, WA.
NRC (National Research Council), 2007, Models in regulatory decision making, National
Research Council, The National Academies Press, Washington, D.C.
Sandia National Laboratories, 1995, WIPP computer software requirements, QAP19-1, Revision
0, Sandia National Laboratories Waste Isolation Pilot Plant, Carlsbad, NM.
Quality Assurance Project Plan
11 November 2015 51
Appendix C: Neptune Check Print SOP

Neptune and
Company, Inc.
Neptune Check Print SOP Page 2 of 7
NAC-0041_R2 Revision 2 Effective date 1 Jan 2015

2 Feb 2015 2
Neptune Check Print SOP
for Verification of Data Entry

Purpose This procedure describes the method for providing a check
for the completeness and accuracy of data entry processes.

Scope This procedure applies to manual or electronic data entry including data
documentation packages developed for model input, databases or
spreadsheets supporting models, and data/results tables included in
reports.

In this This procedure addresses the following major topics:
procedure
Topic See Page
General information about this procedure 1
Check print process 2
Records resulting from this procedure 3

General information about this procedure

Attachments This procedure has the following attachments:

Number Attachment Title No. of pages
1 Check print 1 example 1
2 Check print 1 example data source document 1
3 Check print 2 example 1

Neptune and
Company, Inc.
Neptune Check Print SOP Page 3 of 7
NAC-0041_R2 Revision 2 Effective date 1 Jan 2015

2 Feb 2015 3
History of This table lists the revision history and effective dates of this procedure
revision
Revision Date Description of Changes
0 8 Sep 2004 New document
1 21 Dec 2010 Revised signature page

Who requires Personnel verifying data entry processes.
training to
this
procedure?

Training The training method for this procedure is on-the-job training by a
method previously trained individual and is documented by signature on training
form and archived with project records

Prerequisites None.

Check print process

Overview
This procedure applies to work processes requiring the manual entry or
electronic transfer of data. Examples of entities that receive data include
data documentation packages for model input parameters, external
spreadsheets and databases used to provide input parameters for modeling,
and tables of data/results in documents. Using this procedure, data entry or
transfer is verified by comparing values in the receiving entity with values
in the source documents/files to insure accuracy and completeness of the
data entry or transfer. An individual other than the one compiling the data
in the receiving entity should perform this check. For manual data entry
100 percent of the entries are checked. For electronic data transfer, 10
percent of the entries are checked. Inputs are checked using the check
print process described below. This process can be used to verify most
data entry tasks. Large files may require a modified procedure.

Neptune and
Company, Inc.
Neptune Check Print SOP Page 4 of 7
NAC-0041_R2 Revision 2 Effective date 1 Jan 2015

2 Feb 2015 4
Check print To check print manually entered or electronically transferred data
process perform the following steps:

1. Obtain a paper copy of the receiving entity and a copy of the data
source document. For example, see attachments 1 and 2.
2. Compare the parameter value in the source document including units
with the value in the receiving entity to determine if it was entered
accurately and completely.
3. If the value is correct, mark with a highlighter.
4. If the value is incorrect, circle in red ink and note the correct value.
5. Verify that the cited reference for the value is correct and complete
with page number, table number, or other reference as required.
6. If the reference is accurate and complete, mark with a highlighter.
7. If the reference is inaccurate or incomplete, note corrections in red
ink.
8. Label the checked receiving entity as “Check Print 1”, sign, date and
return to the author for corrections.
9. When the corrections to the receiving entity are completed follow
the same process as described in Steps 1 through 7, however, only
the corrected values/references identified in check print 1 need to be
checked. See attachment 3.
10. Label this check print as “Check Print 2”. Date and sign.
11. Repeat this process until all data/references entered are accurate and
complete. The check print number is incremented for each iteration.
Keep all iterations for archiving.

Records resulting from this procedure

Records The following records are created as a result of this procedure. Paper or
electronic copies are maintained at Neptune and Company as described in
the QAPP.
• All check prints
• Data source documents (or relevant sections thereof)

Neptune and
Company, Inc.
Neptune Check Print SOP Page 5 of 7
NAC-0041_R2 Revision 2 Effective date 1 Jan 2015

2 Feb 2015 5
Attachment 1

An Example GoldSim Parameter List – Check Print 1

1.0 \DoseAssessment\PlantCRFood

Plant/soil concentration ratios are taken from Kennedy and Strenge (1992) [Table 6.16 p. 6-25].
All values in the table are defined as geometric means. The following table presents geometric
mean values for four different plant parts and for each chemical element. These values are also
used in plant-induced contaminant transport calculations (see the container
\TransportProcesses\PlantTransport\PlantCRTransport).

element Leafy Veg Root Fruit Grain
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
C 7.00E-01 7.00E-01 7.00E-01 7.00E-01
Cl 7.11E+01
7.00E+01
7.00E+01 7.00E+01 7.00E+01
Ar 0.00E+00 0.00E+00 0.00E+00 0.00E+00

Reference
Kennedy, W.E. Jr., and D.L. Strenge, 1992. Residual Radioactive Contamination From
Decommissioning, NUREG\CR-5512, Vol. 1 Pacific Northwest Laboratory, Richland,
Washington.

Check Print 1 8 Sep 2004

______________________
Mary Jones

An error was found for the entry for Cl for Leafy veg. The incorrect value was marked in red and the
correct value was noted directly below it. The reference was determined to be incomplete since the data
source was a single table in a 376 page document. The specific location of the table used as the data
source was noted in red ink. The copy is labeled as Check Print 1 and is signed and dated by the
reviewer.

Neptune and
Company, Inc.
Neptune Check Print SOP Page 6 of 7
NAC-0041_R2 Revision 2 Effective date 1 Jan 2015

2 Feb 2015 6
Attachment 2

Source Document Referenced in the Parameter List
Kennedy and Strenge (1992)

Table 6.16 Soil-to-plant concentration factors
Soil-to-plant concentration factors (pCi/kg dry weight per pCi/kg soil)
Element/atomic
number
Leafy vegetables Root vegetables Fruit Grain
H 1 (-)* (-)* (-)* (-)*
Be 4 1.0E-2 1.5E-3 1.5E-3 1.5E-3
C 6 7.0E-1 7.0E-1 7.0E-1 7.0E-1
N 7 3.0E+1 3.0E+1 3.0E+1 3.0E+1
F 9 6.0e-2 6.0E-3 6.0E-3 6.0E-3
Na 11 7.5E-2 5.5E-2 5.5E-2 5.5E-2
Mg 12 1.0E+0 5.5E-1 5.5E-1 5.5E-1
Si 14 3.5E-1 7.0E-2 7.0E-2 7.0E-2
P 15 3.5E+0 3.5E+0 3.5E+0 3.5E+0
S 16 1.5E+0 1.5E+0 1.5E+0 1.5E+0
Cl 17 7.0E+1 7.0E+1 7.0E+1 7.0E+1
Ar 18 (-)** (-)** (-)** (-)**
* Concentration factors for 3H are not needed because a special model is used to determine 3H uptake
in plants.
** Noble gas radionuclides are not assumed to be taken up by plants.

Neptune and
Company, Inc.
Neptune Check Print SOP Page 7 of 7
NAC-0041_R2 Revision 2 Effective date 1 Jan 2015

2 Feb 2015 7
Attachment 3

An example GoldSim Parameter List—Check Print 2

2.0 \DoseAssessment\PlantCRFood

Plant/soil concentration ratios are taken from Kennedy and Strenge (1992) [Table 6.16, p. 6-25].
All values in the table are defined as geometric means. The following table presents geometric
mean values for four different plant parts and for each chemical element. These values are also
used in plant-induced contaminant transport calculations (see the container
\TransportProcesses\PlantTransport\PlantCRTransport).

element Leafy Veg Root Fruit Grain
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
(Ci/kg dry Plant)
per
(Ci/kg dry Soil)
C 7.00E-01 7.00E-01 7.00E-01 7.00E-01
Cl 7.00E+01 7.00E+01 7.00E+01 7.00E+01
Ar 0.00E+00 0.00E+00 0.00E+00 0.00E+00

Reference
Kennedy, W.E. Jr., and D.L. Strenge, 1992. Residual Radioactive Contamination From
Decommissioning, NUREG\CR-5512, Vol. 1 Pacific Northwest Laboratory, Richland,
Washington.

Check Print 2 8 Sep 2004

______________________
Mary Jones

The correction of the entry for Cl for Leafy veg and the additional data source information are verified and
marked. The check print number is incremented and the copy is signed and dated by the reviewer. This is
the final check print since the document is now accurate and complete.

 NAC-0033_R1

Radon Diffusion Modeling
for the Clive DU PA

Clive DU PA Model v1.4
5 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 ii

1. Title: Radon Diffusion Modeling for the Clive DU PA
2. Filename: Radon Modeling v1.4.docx
3. Description: This documents diffusive transport of radon in the Clive DU PA Model.

Name Date
4. Originator Gregg Ochiogrosso 27 October 2015
5. Reviewer(s) Kate Catlett, Dan Levitt 5 November 2015
6. Remarks

5 Nov 2015: Updated from v1.2 to v1.4. – D.Levitt

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 iii
CONTENTS
1.0
 Radon Input Summary Distribution Summary ....................................................................... 1

2.0
 Introduction ............................................................................................................................ 1

3.0
 GoldSim Diffusion of Radon .................................................................................................. 2

3.1
 Background ....................................................................................................................... 2

3.2
 Radon Diffusion ................................................................................................................ 3

4.0
 Diffusion Calculations in the Clive DU PA Model ................................................................ 3

4.1
 Diffusion in Discretized Models ....................................................................................... 3

4.2
 Scaling the GoldSim Calculations to the NRC’s Analytical Method ............................... 5

5.0
 Radon Escape/Production Ratio ............................................................................................. 6

5.1
 Background ....................................................................................................................... 6

5.2
 Cover Materials ................................................................................................................. 7

5.3
 Depleted Uranium Wastes ................................................................................................ 7

6.0
 Model Implementation ........................................................................................................... 7

6.1
 Calibration of Air Diffusion to Counteract Numerical Dispersion ................................... 7

6.2
 Implementation of the E/P Ratio ...................................................................................... 8

7.0
 References ............................................................................................................................ 10

Appendix A: NRC Method for Calculating Radon Flux ............................................................. A-1
Appendix B: Ratio Method for Correcting GoldSim Radon Flux .............................................. B-1

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 iv
FIGURES
Figure 1. One-cell interface and two-cell interface diffusion model schematics. ........................... 4

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 v
TABLES
Table 1. Summary of distributions for radon diffusion modeling ................................................... 1

Table 2. Comparison of exact and GoldSim radon ground surface flux calculations. .................... 4

Table 3. Exact and corrected GoldSim flux comparisons. .............................................................. 5

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 1
1.0 Radon Input Summary Distribution Summary
A summary of parameter values and distributions employed in the radon diffusion modeling
component of the Clive Depleted Uranium Performance Assessment Model (the Clive DU PA
Model) is provided in Table 1. Additional information on the derivation and basis for these
inputs is provided in subsequent sections of this report.
For distributions, the following notation is used:
• Beta( µ, σ, min, max ) represents a generalized beta distribution with mean µ, standard
deviation σ, minimum min, and maximum max.
Table 1. Summary of distributions for radon diffusion modeling
GoldSim Model Parameter Distribution or Value Units Notes
free-air diffusivity of radon gas 0.11 cm2/s See Section 4.1
E/P ratio for DU waste Beta( 0.290, 0.156, 0, 1 ) — See Section 5.2
E/P ratio for other materials 1 — See Section 5.3
2.0 Introduction
This white paper documents the modeling of diffusive transport of the radioactive noble gas
radon-222 (222Rn) in the Clive Depleted Uranium (DU) Performance Assessment (PA) Model
(the Clive DU PA Model, or the Model) for the EnergySolutions’ radioactive waste disposal site
at Clive, Utah. The Clive DU PA Model is dominated by a single waste form: depleted uranium.
This material predominantly consists of 238U. This uranium isotope is a parent of 222Rn, its rate of
production controlled by the decay rate of the 238U and its escape from the solid waste form.
Using a basic representation of fate and transport, radon emanation (production and escape) is
accounted for, as is Henry’s Law partitioning into water, and the gas is allowed to diffuse in pore
air using GoldSim’s internal diffusion processes, as corrected for unsaturated media and for
numerical dispersion. Radon emanation and diffusion are discussed below.
The diffusion of radon is a special modeling problem, due to the coarse discretization (in time
and space) of the Model relative to the diffusivity of radon and the short half-life of 222Rn of
3.8 days. This combination causes the Model, which is set up as a finite difference model (i.e. a
one-dimensional stack of compartments representing a column), to potentially over predict the
diffusive flux of 222Rn through the ground surface.
A solution is found in “calibrating” the GoldSim calculations to radon ground surface flux
estimates resulting from an analytic solution presented in the Nuclear Regulatory Commission
(NRC) Regulatory Guide 3.64 Calculation of Radon Flux Attenuation by Earthen Uranium Mill
Tailings Covers (NRC 1989), and in NUREG/CR-3533 Radon Attenuation Handbook for
Uranium Mill Tailings Cover Design (NRC 1984), referenced in NRC (1989). By varying the
diffusivity of radon, the ground surface flux estimated by the Clive DU PA Model can be
matched (assuming an initially uncontaminated cover and in the absence of other transport
processes) to the NRC model calculation. Once that is done, other contaminant transport
processes, such as advective water transport and biotically-induced transport, are enabled and the
model runs as a fully-coupled system.
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 2
An additional parameter required to model the transport of radon is the escape/production (E/P)
ratio (also known as the emanation factor and by other terms). In this white paper, factors
influencing the magnitude of E/P ratios are discussed and the input parameter distributions for
DU waste materials are presented.
3.0 GoldSim Diffusion of Radon
3.1 Background
Radon-222, a gaseous radioactive isotope of radon, is produced by the alpha decay of
radium-226 (226Ra). The NRC models estimate the diffusive flux of 222Rn as a function of the
concentration of 226Ra in a waste. A two-step method for estimating the cover thickness required
to attenuate the radon flux from uranium mill tailings is described in NRC (1984 and 1989) and
has been used for calculating radon flux at the Area 5 Radioactive Waste Management Site
(RWMS) at the Nevada National Security Site (NNSS, formerly the Nevada Test Site) (Shott et
al. 1998, p. 3-107). This approach provides an analytic solution for radon flux at steady state and
is summarized in Appendix A. This approach was used in two RWMS PA models developed by
Neptune and Company, and was readily applied at the NNSS sites, since the waste cover was a
homogeneous monofill of local alluvium (Neptune, 2005). Radon transport through the waste
and a clean cover was modeled in GoldSim using a series of diffusive flux links between cells.
This approach is mathematically identical to a finite difference model of the system, as detailed
in the GoldSim Contaminant Transport Manual, Chapter 4 (GTG, 2011).
The cover system for the portion of the Federal Cell that is allocated for disposal of DU (the
Federal DU Cell), however, is more complex. Instead of using a single material in the cover, the
layering above the DU waste is as follows, from the bottom up: Directly atop the DU waste lies
generic Class A waste, which is represented in the DU model as Unit 4 material with no
inventory. Above that is a specially-constructed radon barrier, made of two layers of compacted
clay. Intact compacted clay is an especially effective barrier to radon diffusion, as it has a low
permeability and high moisture content, and radon partitions between air and water. Above this
clay is an evapotranspirative (ET) cover consisting of a frost protection layer, and an evaporative
zone and surface layer made up of Unit 4 soils with varying admixtures of gravel in the surface
layer. In essence there are three materials above the radon-producing DU waste, so the simple
monolayer analytical models presented in the NRC Regulatory Guide (NRC 1989) are
insufficient. For systems involving more than two layers, the practitioner is referred in that Guide
to the previously published NRC Handbook (NRC 1984). This is the source of the analytical
equations built into the Clive DU PA Model for purposes of calibration of the native GoldSim
diffusion calculations.
While these analytical solutions are specifically acceptable to the NRC, use of this approach does
not account for the coupling of other processes that may contribute to the radon flux at the
ground surface. For example, diffusion of radium and other radon parents through the clay in the
liquid phase is possible, and biota can translocate radon and its parents. These processes can be
discretely disabled in the Clive DU PA Model for the purposes of examining their significance in
determining radon ground surface flux.
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 3
3.2 Radon Diffusion
The transport of radon presents special modeling challenges. The diffusion of gas phase
constituents follows concentration gradients, and these vary along the diffusive path. Local
concentrations are subject to other transport processes, notably partitioning into water (governed
the Henry’s Law constant) and encountering sinks like the atmosphere. In the PA modeling of
radioactive constituents, most radionuclides have relatively long half-lives, and the concentration
gradients are not much affected by decay and ingrowth. Radon isotopes, however, have short
half-lives relative to the rate of diffusive transport processes, can move quickly in pore air, and
decay to a chain of radionuclides that can be significant in terms of dose and risk. In the Clive
DU PA Model, the radon isotope of interest is 222Rn, with a half-life of about 3.8 days. With this
short half-life, 222Rn decays away quickly enough that the decay alone can produce strong
concentration gradients, causing additional challenges in numerical simulation.
Chemical engineers are faced with similar issues in process plants, where chemicals in a process
that moves through the plant are simultaneously undergoing chemical transformation to other
substances. The quantification of this effect is called the Damköhler number. The value can be
expressed in a number of different ways for different applications, and in the case of this Model,
it is the ratio of the decay rate to the diffusive mass transport rate. For 222Rn, with its high rate of
decay, the Damköhler number is also high, indicating that diffusive transport will be over-
predicted in a coarsely-discretized model such as the Clive DU PA Model.
4.0 Diffusion Calculations in the Clive DU PA Model
4.1 Diffusion in Discretized Models
Compartment models (and all spatially-discretized models) tend to overestimate diffusive flux.
The discrepancy in estimated flux is due to the discretized model’s inability to accurately
represent the fine spatial distribution of radon concentration using the model’s coarse spatial
discretization—a limitation shared by other finite difference models. Since the compartments are
mathematically modeled as being fully mixed at all times, they tend to move radon through the
system faster than it would in reality—a phenomenon known to modelers as numerical
dispersion.
One approach for correcting or calibrating the flux is based on developing a ratio of the
calculated flux to the exact flux (an analytical solution). The derivation for the exact flux and the
ratio is given in Appendix B. The problem is illustrated for a simple case of a steady-state system
with a one-dimensional diffusion pathway for radon in air with no porous media. For the “1 cell
interface” problem the concentration in the first cell is held to a constant value and the last cell is
advectively “flushed” to maintain a concentration of zero. Three GoldSim models were built
using 1, 2, and 10 cell interfaces to model a 1-meter column. Configurations for the 1- and 2-cell
interface models are shown in Figure 1.

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 4

Figure 1. One-cell interface and two-cell interface diffusion model schematics.
Steady-state fluxes were calculated for diffusivities of 11 × 10-6 m2/s (or 0.11 cm2/s, the free air
diffusivity of radon from Rogers and Nielson, 1991), 1.1 × 10-6 m2/s, and 0.11 × 10-6 m2/s and
compared to the exact fluxes calculated from analytical solutions of the diffusion equation. These
results are shown in Table 2.
Table 2. Comparison of exact and GoldSim radon ground surface flux calculations.
No. of Cells Diffusivity Exact Flux GoldSim Flux Error
(× 10-6 m2/s) (× 10-8 g/yr) (× 10-8 g/yr) (percent)
1 11 3.499 3.614 3.3
1 1.1 1.095 1.479 35.1
1 0.11 0.0237 0.2142 803.8
2 11 3.724 3.756 0.9
2 1.1 1.484 1.618 9.0
2 0.11 0.04231 0.113 167.1
10 11 3.883 3.887 0.1
10 1.1 1.910 1.918 0.4
10 0.11 0.08764 0.09288 6.0
Comparison of exact and GoldSim fluxes in Table 2 shows that for any given number of cells
representing the system, the error in the GoldSim flux increases as the diffusivity decreases. In
addition, for a given value of diffusivity, the error decreases as the number of cells is increased.
The ratio of the linear flux calculated by GoldSim to the exact flux R (discussed in Appendix B)
is
𝑅=sinh  (𝑏𝐿)
𝑏𝐿 (1)
where
L   =  diffusion  path  length  and  
b  =  a  function  of  the  decay  coefficient  and  the  diffusivity.  
For a one cell model, the exact flux is obtained from the GoldSim flux by dividing the GoldSim
flux by R. Table 3 shows the close agreement between exact and corrected GoldSim fluxes for a
one cell model.
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 5
Table 3. Exact and corrected GoldSim flux comparisons.
No. of Cells Diffusivity Exact Flux GoldSim Flux Error
(× 10-6 m2/s) (× 10-8 g/yr) (× 10-8 g/yr) (percent)
1 11 3.4980 3.5000 0.1
1 1.1 1.0948 1.0956 0.1
1 0.11 0.02368 0.02370 0.1
2 11 3.7436 3.7532 0.3
2 1.1 1.5259 1.5499 1.6
2 0.11 0.04491 0.04626 3.0
10 11 3.8491 3.8528 0.1
10 1.1 1.8043 1.8095 0.3
10 0.11 0.073252 0.072097 -1.6

To obtain a value of R for a multiple cell model, R is calculated using an effective length,
𝐿!""=L
1.15𝑛 (2)
where n is the number of cells. The diffusive areas for each GoldSim cell are then reduced by
dividing by R. Exact and corrected GoldSim fluxes using this method for two- and five-cell
interface models are shown in Table 3. In all cases, the error is reduced to 3 percent or less.
A similar approach was applied to correct the radon fluxes for the NNSS Area 5 RWMS
GoldSim Model without success. The ratios obtained from the simple problem described above
with a constant known concentration at one end of the diffusion path were not applicable to the
system at the RWMS where the radon source is distributed spatially throughout the depth of the
waste. For such shallow land burial waste configurations, without correction, the flux calculated
by GoldSim was 11 times the flux calculated using the NRC method. Reducing the diffusive
areas by R resulted in only a 13 percent reduction in the flux calculated by GoldSim. The
behavior of radon flux at the Clive facility would be expected to suffer the same limitation, and
would result in an unrealistic over-prediction of the flux at the ground surface.
4.2 Scaling the GoldSim Calculations to the NRC’s Analytical
Method
An empirical approach was adopted in which a scaling factor was obtained that would reduce the
value of the flux calculated in GoldSim to the value of flux calculated using various
implementations of the NRC method. While this technique was originally developed for the
NNSS RWMSs, it applies equally well for the Clive Facility. This approach is implemented in
the Clive DU PA Model by multiplying the bulk diffusivity for radon by a scaling factor unique
to each material in the cover system. The scaling factor is determined by calculating radon flux
using both the radon flux estimated “natively” by GoldSim in the Model and a selected variant of
the NRC method (coded into the Model itself) and adjusting the scaling factor until the time
histories of ground surface flux matched sufficiently well. This is done using the Radon
Calibration procedure (and the associated dashboard) built into the model, and comparing the
resulting flux estimates graphically. In the Clive DU PA Model, there is a different effective
radon diffusivity calibration factor for each of the different materials comprising the cover.
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 6
Calibration factors are developed for the various cover layers, making use of some special
GoldSim functionality: In most GoldSim models the fluid properties of air, including the relative
diffusivity of radon in air, are defined globally. In such a case, scaling the relative diffusivity
would scale it everywhere in the model. To circumvent this global influence, local definitions of
the fluid Air have been created for each modeled layer above the DU waste. This allows unique
values of the bulk diffusion coefficient to be used for each layer (GoldSim Contaminant
Transport Manual Chapter 4). Unique scaling factors for the generic Class A waste, clay radon
barrier, and the upper cover materials are implemented and are to be calibrated independently.
5.0 Radon Escape/Production Ratio
5.1 Background
The fraction of 222Rn produced by decay of radium-226 (226Ra) that is released from the solid
matrix is known as the E/P ratio, as well as the emanation coefficient, the emanation factor, or
emanating power (Nielson and Sandquist, 2011). When 226Ra decays a small fraction of the
decay energy, 0.1 MeV, is carried by the recoiling 222Rn atom. This is sufficient energy for the
recoiling atom to travel about 45 nm in a mineral matrix, 0.1 µm in water, and about 63 µm in
air. Recoiling atoms with just sufficient energy to stop in the air or water filled pore space will be
released from the matrix and become available for transport. If there is too little energy available,
the atom will remain trapped in the solid matrix. If there is too much energy, the atom will cross
the pore space and be embedded in the solid matrix of a nearby grain. The E/P ratio describes
that fraction of 222Rn that stops in the air or water-filled pore space and is free to diffuse. The E/P
ratio can physically vary from 0, implying no escape, to 1, where all radon escapes.
Predicting the E/P ratio for a material is difficult as numerous factors have been identified that
affect it: The E/P ratio is inversely related to grain size. The closer decaying atoms are to the
surface of a grain, the more likely they will be released to the pore space. The adsorption or
coprecipitation of 226Ra on surficial coatings increases emanation, as will cracks, fissures, or
pitting of grains. In contrast, the E/P ratio is directly related to pore size. As the pore size
increases, it is more likely that recoiling atoms will stop in the pore space, increasing emanation.
The presence of water in the pore space also increases emanation, because the reduced particle
range in water increases the likelihood that the recoiling atom will stop in the pore space.
Predicting the E/P ratio of a material is particularly difficult because it requires detailed
knowledge of the microscopic physical structure of the material, microscopic distribution of 226Ra in the material, and water content.
The E/P ratios for different types of common geologic materials have been reported. From
geometrical considerations, the maximum emanation expected from a thick slab source is 0.5 and
from a thin film, 1.0. The maximum E/P ratio of natural materials will lie somewhere between
these two extremes. The maximum value reported for common materials is approximately 0.7 to
0.8. Reported E/P ratios for soils and rocks range from 0.02 to 0.7 (UNSCEAR 2000; NCRP
1988). The emanation factor of a single material may vary over a substantial portion of this range
depending on the water content. Rock and uncrushed ores usually have lower emanation factors
ranging from 0.02 to 0.26 (Nazaroff 1992). Concrete emanation factors may range from 0 to 0.3
(Rogers et al. 1994; Cozmuta et al. 2003).
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 7
5.2 Cover Materials
There are probably insufficient data to assign E/P ratios to 226Ra in earthen covers, and assigning
a value of 1 takes no credit for attenuation of radon in the soil or clay materials. Radium-226
may be transported to the cover by liquid advection and diffusion, plant uptake, animal
burrowing, and intruder activities. Radium transported by liquid advection and plant uptake, after
sufficient time for degradation, may exist in surface coatings that are reported to have high E/P
ratios. Radium transported by animal burrowing or intruder activities may be in forms similar to
the original waste form.
5.3 Depleted Uranium Wastes
Nielson and Sandquist (2011) report that the E/P ratio of depleted uranium oxides should be
expected to mimic that of natural western uranium ores, which are on average about 0.29. The
bulk of the DU waste considered in the Clive DU PA Model is in the form of U3O8, which is
similar to that of natural ores. Based on escape/production data presented in Table 1 of Nielson
and Sandquist (2011), and the physical limitations of 0 and 1, a stochastic input parameter
distribution to represent the E/P ratio of DU waste is fit with a beta distribution: beta( µ = 0.290,
σ = 0.156, min = 0, max = 1 ).
6.0 Model Implementation
6.1 Calibration of Air Diffusion to Counteract Numerical Dispersion
In the Clive DU PA Model, where wastes and other materials with a variety of porous medium
properties can be intermixed, calibration to analytical solutions is challenging. The most
effective way to refine the solution is by refining the grid, which is the approach that is taken in
the Model. This refinement reduces numerical dispersion, providing a more realistic simulation
of the diffusion process. To refine the Top Slope and Side Slope columns, which integrate all the
contaminant transport processes, would introduce an unreasonable computational burden, since
most of the processes would not appreciably benefit from the finer discretization. By taking
advantage of a clever side calculation, the model can benefit from the increased accuracy of the
finer discretization, without significant computational effort.
Within the GoldSim container for the Class A South Cell (now the Federal DU Cell) Top Slope
contaminant transport calculations, an additional container called RnCalibCalcs is added, with a
switch for enabling and disabling its calculations. This container is devoted to determining the
appropriate measure by which the radon diffusivity of 0.11 cm2/s should be reduced in order to
exactly counter the effects of numerical dispersion. The calculation consists of two columns: one
coarsely and one finely discretized. The coarse discretization matches that of the main Top Slope
column, with GoldSim Cell pathway elements representing layers in the cover of about 15 cm (6
in) thickness, and layers in the waste about 50 cm (20 in) thick. Each of the Cells is populated
with air, water, porous media, and initial inventory just as is the main column, but the only
processes represented include retardation, solubility, and air diffusion. This protects the coarse
column from all the confounding factors of water advection and diffusion, biotic processes, etc.
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 8
The fine column is built the same way, but with 15 times as many Cells. That is, the cover cells
in the fine column are about 1 cm thick, and those in the waste are about 3.4 cm. This fine
column has significantly less numerical dispersion for the air diffusion calculations, and is used
for calibration of the coarse column. The calibrated diffusion coefficients for radon in the coarse
column, then, are also applied to the main Top Slope and Side Slope columns, thereby
counteracting the effects of numerical dispersion there as well.
Both the coarse and fine columns are populated with porous materials that exactly match those in
the Top Slope column. These reside in three distinct zones. From the bottom up, these zones are:
1) the waste zone, where radioactive wastes are disposed, and are of similar material properties
as far as air diffusion is concerned, 2) the radon barrier clay layers, which consist of tightly
compacted clays overlying the wastes, and 3) the upper cover materials, which are uncompacted
fill materials. The radon fluxes (mass flow per area, with dimensions of M/L2·T and units of
g/m2·s or pCi/m2·s) are recorded at the top of each of these three zones in both the coarse and
fine columns. With less numerical dispersion, the finer column always has a lower rate of
diffusion out the top of the zone.
If the coarse column’s diffusion coefficient for radon is adjusted downward, it can be forced to
match the finer column, producing a more accurate flux. This correction is performed
sequentially, from the bottom up, and a different correction factor is applied for each material
within the three zones. This results in one radon diffusion correction factor for the waste zone,
one for the clays, and one for the upper cover layers. The correction factor for wastes is applied
to the waste layers in the Top Slope and Side Slope contaminant transport columns. Likewise,
the correction factors for the clay layers is applied to the radon barrier clay and liner clay layers
in the Top Slope and Side Slope columns, and the correction factors for the upper cover layers is
applied to both columns as well.
This radon calibration need be done only once, unless the layer geometries and/or material
properties change significantly. Fortunately, performing the calibration in deterministic mode is
sufficient, as it is robust and holds quite well even using stochastic inputs in probabilistic mode.
Once the calibration has been completed, the radon calibration container may be disabled, so that
it does not impose further computational burden on the model.
6.2 Implementation of the E/P Ratio
When an atom of 226Ra decays to 222Rn, it does one of two things: If it escapes, it can migrate
away, subject to various contaminant transport processes. If it does not escape to a point where it
may migrate away, then it will decay in place to polonium-218 and other short-lived progeny
until it gets to lead-210 (210Pb) and eventually to stable lead. Of these progeny, only 210Pb is
long-lived enough to be modeled for potential transport in the Clive DU PA Model.
In constructing the GoldSim Species element in the Model, the model developer defines decay
chains according to branching fractions, or “stoichiometry”, as GoldSim call it. Clever use of the
branching fraction allows the Model to emulate the binary behavior of the decay of 226Ra, so that
some decays to 222Rn and can migrate away, while the remainder does not travel as radon, since
it did not escape, and decays in place to 210Pb. The decay of 226Ra is therefore assigned two
branches: The first branch is decay to 222Rn, and the fraction of decays that follow that branch is
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 9
the E/P ratio, since these atoms have escaped. The second branch is decay directly to 210Pb, so
that the atoms that have not escaped do not have the opportunity to migrate away as radon. The
error in skipping the step of decay of 222Rn to 210Pb is negligibly small, since the half-life of 226Ra (1599 yr) and that of 210Pb (22.3 yr) are both much larger than the 3.8-day half-life of 222Rn.
In the DU waste, the E/P ratio is selected from the distribution presented in Section 5.3, and this
fraction of decays is allowed to decay to 222Rn. The balance (1 – E/P ratio) decays directly to 210Pb. Outside the DU waste, where the 226Ra is no longer in the crystalline matrix of the waste,
the E/P ratio is 1, so all decays produce 222Rn, and none (1 – 1) decay directly to 210Pb.

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 10
7.0 References
Cozmuta, I., E. R. van der Graaf, and R. J. de Meijer, 2003. Moisture Dependence of Radon
Transport in Concrete: Measurements and Modeling. Health Physics 85(4): 438 – 456.
GTG (GoldSim Technology Group), 2011. User's Guide: GoldSim Contaminant Transport
Module, GoldSim Technology Group, Issaquah, WA, December 2010
Nazaroff, W. W., 1992. Radon Transport from Soil to Air. Rev. of Geophysics 30(2): 137 – 162.
NCRP (National Council on Radiation Protection and Measurements). 1988. Measurements of
Radon and Radon Daughters in Air. NCRP Report No. 97, NCRP Bethesda, Maryland,
November 1988.
Nielson, K.K., and G.M. Sandquist. 2011. Radon Emanation from Disposal of Depleted Uranium
at Clive, Utah. Report for EnergySolutions by Applied Science Professionals, LLC.
February 2011.
NRC (U.S. Nuclear Regulatory Commission). 1984. Radon Attenuation Handbook for Uranium
Mill Tailings Cover Design. NUREG/CR-3533. V.C. Rogers, K.K. Nielson, and D.R.
Kalkwarf for Nuclear Regulatory Commission, Washington, DC, April 1984
NRC. 1989. Calculation of Radon Flux Attenuation by Earthen Uranium Mill Tailings Covers,
Regulatory Guide 3.64, Nuclear Regulatory Commission, June 1989.
Rogers, V.C., and K.K. Nielson. 1991. Correlations for Predicting Air Permeabilities and
Rn- 222 Diffusion Coefficients of Soils, Health Physics (61) 2
Rogers, V.C., K. K. Nielson, M. A. Lehto, and R. B. Holt, 1994. Radon Generation and
Transport Through Concrete Foundations. EPA/600/SR-94/175, U. S. Environmental
Protection Agency, Research Triangle Park, North Carolina, November 1994.
Shott, G.J., L.E. Barker, S.E. Rawlinson, M.J. Sully, and B.A. Moore, 1998, Performance
Assessment for the Area 5 Radioactive Waste Management Site at the Nevada Test Site,
Nye County, Nevada (Rev. 2.1), U.S. DOE DOE/NV/11718-176.
UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation). 2000.
UNSCEAR Report to the General Assembly – Sources and Effects of Ionizing Radiation.

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 A-1
Appendix A: NRC Method for Calculating Radon Flux
The following discussion is a synopsis of the NRC method for calculating the ground surface
flux of radon from a covered source of uranium mill tailings. The original notation has been
preserved, and the units conversion factors in the original equations have been removed. As in
any application, be sure to use a consistent set of units, such as the cgs system used here. The
one-dimensional steady-state diffusion equation describing radon flux is given in NRC (1989) as
𝐷𝑑!𝐶
𝑑𝑥!=−𝜆  𝐶+𝑅  𝜌  𝐸  𝜆
𝑛=0 (A-1)
where
D  =  diffusion  coefficient  for  radon  in  the  total  pore  space  (cm2  s-­‐1),  
C   =  radon  concentration  in  the  total  pore  space  (pCi  cm-­‐3),  
λ   =  radon  decay  constant  (2.1  ×  10-­‐6  s-­‐1),  
R    =  specific  activity  of  the  parent  material,  radium-­‐226  (pCi  g-­‐1),  
ρ     =  dry  bulk  mass  density  of  soil  or  tailings  (g  cm-­‐3),  
E     =  radon  emanation  coefficient  (dimensionless),  and  
n     =  soil  or  tailings  porosity  (dimensionless).  

The radon flux is obtained from the radon concentration gradient by (NRC 1989):
𝐽=−𝐷𝑛𝑑𝐶
𝑑𝑥 (A-2)
where
J   =  radon  flux  (pCi  m-­‐2  s-­‐1).  

The radon diffusion coefficient can be estimated using a correlation in (NRC 1989):
𝐷=0.07  exp −4 𝑚−𝑚𝑛!+𝑚! (A-3)
where
m  =  moisture  saturation  fraction,  the  volumetric  fraction  of  saturation  of  pore  space.  
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 A-2

Alternatively, the correlation developed by Rogers and Nielson (1991) can be used:
𝐷=𝐷!𝑛  exp −6𝑚𝑛−6𝑚!"! (A-4)
where
D0  =  radon  free  air  diffusion  coefficient  (0.11  cm2  s-­‐1).  

In the first step of the calculation, the flux from the tailings source is calculated using (A-1) and
assuming no cover material over the tailings. Boundary conditions are (Rogers and Nielson,
1984):
(a) zero radon concentration at the ground surface: 𝐶𝑥=0 =0 (A-5)
(b) zero radon flux at the base of the source (tailings): !"
!"𝑥=𝑥!=0 (A-6)
where
x  =  thickness,  and  the  subscript  t  refers  to  tailings.  

The flux is given by NRC (1989) as
𝐽!=𝑅!  𝜌!  𝐸!𝜆  𝐷!  tanh 𝑥!𝜆𝐷! (A-7)
This flux is used in the second step to solve for the surface flux of radon assuming continuity of
flux and continuity of concentration across the tailings-cover interface:
𝐽!=2  𝐽!  exp −𝑏!𝑥!
1 +𝑎!𝑎!tanh 𝑏!𝑥!+1 −𝑎!𝑎!tanh 𝑏!𝑥!  exp −2𝑏!𝑥!
(A-8)
where the subscript c refers to the cover. The inverse relaxation lengths are defined as
𝑏!=𝜆  /𝐷!   and 𝑏!=𝜆  /𝐷! (A-9)
and the interface constants are defined as
𝑎!=𝑛!!  𝐷!  1 −1 −𝑘𝑚!!   and 𝑎!=𝑛!!  𝐷!  1 −1 −𝑘𝑚!! (A-10)
where
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 A-3
k   =  equilibrium  distribution  coefficient  for  radon  in  water  and  air  (pCi  cm-­‐3  water  
per  pCi  cm-­‐3  air).  

This solution is presented without reference in NRC (1989) and Rogers and Nielson (1984). For
sources thicker than 200 cm and thick covers (NRC, 1989), (A-8) reduces to
𝐽!=2  𝐽!  exp −𝑏!𝑥!
1 +𝑎!𝑎!
(A-11)

For the Clive DU PA Model, porosities, diffusion coefficients, saturations, and distribution
coefficients are the same for the waste and the cover such that ac and at in (A-10) above are
equal. For this case, the surface flux can be further simplified to
𝐽!=  𝐽!  exp −𝑏!𝑥! (A-12)

This equation indicates that the radon flux decreases exponentially with distance through the
cover.

Radon Diffusion Modeling for the Clive DU PA
5 November 2015 B-1
Appendix B: Ratio Method for Correcting GoldSim Radon Flux
Consider one-dimensional steady-state diffusion in a cell of length L. The concentration is held
at one end at a value C0 and at the other end at zero. The concentration distribution is given by
𝐷𝜕!𝐶
𝜕𝑥!−𝜆  𝐶=0 (B-1)
and the flux by
𝐽=−𝐷𝐴  𝜕𝐶
𝜕𝑥 (B-2)
where A is the cross-sectional area.
Boundary conditions are
C(x=0) = C0 (the initial concentration is maintained constant at x=0), and
C(x=L) = 0 (the concentration is maintained at zero at x=L).
Assume a solution for the concentration at any position x in the cell of
𝐶𝑥=𝐶!sinh 𝑏𝑥+𝐶!cosh 𝑏𝑥  . (B-3)
Where b is some as yet unknown constant for which we intend to solve. Applying the first
boundary condition,
𝐶!=𝐶!sinh 0 +𝐶!cosh 0  .
Since sinh(0) = 0 and cosh(0) = 1, C2 = C0. Applying the second boundary condition,
0 =𝐶!sinh 𝑏𝐿+𝐶!cosh 𝑏𝐿  
𝐶!=−𝐶!
cosh(𝑏𝐿)
sinh(𝑏𝐿)  .
Using B-2, B-3, and the values of C1 an C2 with
𝜕𝐶
𝜕𝑥=𝑏  𝐶!cosh 𝑏𝑥+𝑏  𝐶!sinh 𝑏𝑥  ,
the flux at the end of the cell (x=L) is
𝐽=−𝐴𝐷𝑏−𝐶!
cosh(𝑏𝐿)
sinh(𝑏𝐿)cosh(𝑏𝐿)+𝐶!sinh 𝑏𝐿
Radon Diffusion Modeling for the Clive DU PA
5 November 2015 B-2
=−𝐴𝐷𝑏𝐶!−cosh!(𝑏𝐿)+sinh!(𝑏𝐿)
sinh(𝑏𝐿)
=𝐴𝐷𝑏𝐶!
cosh!(𝑏𝐿)−sinh!(𝑏𝐿)
sinh(𝑏𝐿)
=𝐴𝐷𝑏𝐶!
sinh(𝑏𝐿)  ,
since cosh2 t − sinh2 t = 1. Substituting B-3 in B-1 gives
𝐷𝑏!  𝐶!𝑠𝑖𝑛ℎ𝑏𝑥+𝐶!𝑐𝑜𝑠ℎ𝑏𝑥−𝜆  𝐶!𝑠𝑖𝑛ℎ𝑏𝑥+𝐶!𝑐𝑜𝑠ℎ𝑏𝑥=0  

 𝐶!𝑠𝑖𝑛ℎ𝑏𝑥+𝐶!𝑐𝑜𝑠ℎ𝑏𝑥𝐷𝑏!−𝜆=0  

Since    𝐶!𝑠𝑖𝑛ℎ𝑏𝑥+𝐶!𝑐𝑜𝑠ℎ𝑏𝑥≠0  ,  

𝑏=𝜆𝐷  .
The value of b is now known. The linear flux as calculated in the Clive DU PA Model is
therefore
𝐽!=−𝐴𝐷0 −𝐶!
𝐿    ,
so the ratio of the linear flux to the exact flux is
𝐽!
𝐽=sinh(𝑏𝐿)
𝑏𝐿    .
 NAC-0030_R1

Sensitivity Analysis Results
for the Clive DU PA

Clive DU PA Model v1.4
25 November 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 ii
1. Title: Sensitivity Analysis Results for the Clive DU PA
2. Filename: Sensitivity Analysis Results (Appendix 19) v1.4.docx
3. Description: Sensitivity indices for explanatory variables presented for each of the
endpoints considered.
Name Date
4. Originator Paul Duffy 22 November 2015
5. Reviewer Paul Black 25 November 2015
6. Remarks

5 Nov 2015: Updated from v1.2 to v1.4. – D.Levitt

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 iii

This page is intentionally blank, aside from this statement.

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 iv
CONTENTS
TABLES .......................................................................................................................................... v

Summary .......................................................................................................................................... 1

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 v
TABLES
Table 1: Peak Groundwater Well Concentrations within 500 years – I129 .................................... 2

Table 2: Peak Groundwater Well Concentrations within 500 years – Np237 ................................. 7

Table 3: Peak Groundwater Well Concentrations within 500 years - Tc99 .................................. 12

Table 4: Peak Groundwater Well Concentrations within 500 years – Th230 ............................... 17

Table 5: Peak Groundwater Well Concentrations within 500 years – Th232 ............................... 22

Table 6: Peak Groundwater Well Concentrations within 500 years – U233 ................................. 27

Table 7: Peak Groundwater Well Concentrations within 500 years – U234 ................................. 32

Table 8: Peak Groundwater Well Concentrations within 500 years – U235 ................................. 37

Table 9: Peak Groundwater Well Concentrations within 500 years – U236 ................................. 42

Table 10: Peak Groundwater Well Concentrations within 500 years – U238 ............................... 47

Table 11: Dose summed over 10,000 years - Population .............................................................. 52

Table 12: Peak Dose within 10,000 years - Hunter ....................................................................... 57

Table 13: Peak Dose within 10,000 years – I-80 ........................................................................... 62

Table 14: Peak Dose within 10,000 years – Knolls ....................................................................... 67

Table 15: Peak Dose within 10,000 years – Railroad ................................................................... 72

Table 16: Peak Dose within 10,000 years – Rest Area ................................................................. 77

Table 17: Peak Dose within 10,000 years – Sport OHV ............................................................... 82

Table 18: Peak Dose within 10,000 years – UTTR Dose .............................................................. 87

Table 19: Peak Dose within 10,000 years – Rancher .................................................................... 92

Table 20: Peak Uranium Hazard within 10,000 years - Hunter .................................................... 97

Table 21: Peak Uranium Hazard within 10,000 years - Rancher ................................................ 102

Table 22: Peak Uranium Hazard within 10,000 years – Sport OHV .......................................... 107

Table 23: Benson Peak Groundwater Well Concentrations within 500 years –Tc99 ................. 112

Table 24: Benson Peak Dose within 10,000 years – Rancher ..................................................... 117

Table 25: Benson Erosion Peak Groundwater Well Concentrations within 500 years –Tc99 .... 122

Table 26: Benson Erosion Peak Dose within 10,000 years – Rancher ....................................... 127

Table 27: Benson Clay Liner Peak Groundwater Well Concentrations within 500 years –
Tc99 ........................................................................................................................... 132

Table 28: Benson Clay Liner Peak Dose within 10,000 years – Rancher ................................... 137

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 1
Summary
This appendix presents tabular summaries of the sensitivity analysis results for each of the 13
endpoints considered within the Clive DU Performance Assessment (PA) Version 1.4
(November 2015). As described in the Sensitivity Analysis (SA) methods White Paper, for each
endpoint, every explanatory variable (input parameter) in the PA model is included in the SA.
The SA calculates a sensitivity index (SI) for each explanatory variable. Each SI represents the
portion of total statistical variance in the output that is attributed to the corresponding
explanatory variable for a specific model output of interest. This global SA approach essentially
allows all input parameters to be varied simultaneously to find the input parameters that explain
most of the output parameter variance and, hence, are identified as the most important, or
sensitive, predictors of the model output.
In effect, the simulated data are treated as observations of the explanatory variables (input
parameters) and the dependent variables (model output – dose or concentration). The SA
performs a regression analysis of the input on the output, but does so in a way that
accommodates the non-linear and non-monotonic aspects of this complex model. All explanatory
variables are included in this “regression”. The specific statistical, non-linear regression-based,
approach taken to SA is explained in the SA methods White Paper. It relies on a gradient
boosting machines (GBM) method that utilizes boosting of binary recursive partitioning
algorithms that deconstruct a model output, or response, into the relative influence from a given
set of explanatory variables (input parameters).
A table of SIs is presented for each model endpoint (Tables 1 – 22). For a given endpoint, the
sum of the SIs across the explanatory variables is 100%. The diagnostic goodness-of-fit statistic
R-squared is used to indicate goodness-of-fit of the SA model. If R-squared is near 100% then
the SA model explains nearly all of the variation in the model output, and only a very small
portion is unexplained by the input parameters. In general, this suggests that the SA provides a
good fit for the model output, which provides greater confidence in the results of the SA.
The underlying concept is that only a few input parameters are likely to explain most of the
output variance. In general, it is rare that more than 4 or 5 input parameters can be classed as
sensitive. It is difficult to share 100% variation across many more parameters and still have a
reasonably predictive model. The goal of the global SA is to identify those few sensitive input
parameters. If uncertainty needs to be reduced to support decision making, then reduction in
uncertainty in the sensitive input parameters is likely to prove most beneficial.
The identification of important explanatory variables is done within the context of the ability of
the GBM model to explain the observed variance in the endpoint of interest from the GoldSim
model. If the R-squared of regressing the observed values on the predictions is close to 100%,
indicating a good fit, then experience suggests that a SI of 5% is a reasonable threshold for
identification of a sensitive parameter. For example, the endpoint ‘Peak Groundwater Well
Concentrations within 500 years - Tc99’ has a GBM model with the R-squared of the linear
model regressing the observed values on the GBM predictions very close to one (99% – Table
3). Consequently, any of the explanatory variables with a SI less than 5% suggest random noise
rather than a predictive input parameter.

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 2
Table 1: Peak Groundwater Well Concentrations within 500 years – I129
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 35.78

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 30.15

Kd
 Sand
 for
 I
 (mL/g)
 17.86

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 6.22

Saturated
 Zone
 Water
 Table
 Gradient
 1.95

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.63

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.48

Unit
 2
 Porosity
 0.33

Kd
 Sand
 for
 Pu
 (mL/g)
 0.31

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.21

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.17

Random
 Gully
 Selector
 0.15

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.15

Saturated
 Zone
 Thickness
 (m)
 0.12

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.12

Kd
 Silt
 for
 Np
 (mL/g)
 0.11

Surface
 Atmosphere
 Thickness
 (m)
 0.10

Kd
 Silt
 for
 Ac
 (mL/g)
 0.10

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.09

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.07

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.07

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.07

Tortuosity
 Water
 Content
 Exponent

 0.07

Surface
 Wind
 Speed
 (m/s)
 0.07

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.07

Fine
 CobbleMix
 Porosity
 0.07

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.06

Greasewood
 Root
 Shape
 Parameter
 b
 0.06

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.06

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.06

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.06

Intermediate
 Lake
 Depth
 (m)
 0.06

Ant
 Nest
 Volume
 (m3)
 0.06

Ant
 Nest
 Shape
 Parameter
 b
 0.06

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.05

Kd
 Sand
 for
 U
 (mL/g)
 0.05

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 3
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.05

Plant
 Fresh
 Weight
 Conversion
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.05

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.05

Kd
 Sand
 for
 Ac
 (mL/g)
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.04

DCF
 Alpha
 REF
 0.04

Mammal
 Burrow
 Shape
 Parameter
 b
 0.04

Site
 Dispersal
 Area
 (km2)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.04

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.04

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.04

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.04

Deep
 Time
 DCF
 Photon
 2
 REF
 0.04

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.04

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.04

Kd
 Silt
 for
 U
 (mL/g)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.03

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.03

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.03

DCF
 Photon1
 REF
 0.03

Kd
 Silt
 for
 Th
 (mL/g)
 0.03

Unit
 3
 Residual
 Water
 Content
 0.03

Silt
 Sand
 Gravel
 Porosity
 0.03

RipRap
 Bulk
 Density
 (g/cm3)
 0.03

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.03

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.03

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.03

Kd
 Silt
 for
 Am
 (mL/g)
 0.03

Deep
 Time
 Diffusion
 Length
 (m)
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.03

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.03

DCF
 Photon2
 REF
 0.03

Kd
 Clay
 for
 U
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.03

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.03

Shrub
 Root
 Shape
 Parameter
 b
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 4
Kd
 Sand
 for
 Sr
 (mL/g)
 0.03

Fine
 Gravel
 Mix
 Porosity
 0.03

Kd
 Silt
 for
 Pb
 (mL/g)
 0.03

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.03

DCF
 Beta
 REF
 0.03

Meat
 Preparation
 Loss
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.03

Kd
 Sand
 for
 Am
 (mL/g)
 0.03

Kd
 Clay
 for
 Sr
 (mL/g)
 0.03

Unit
 3
 Porosity
 0.03

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.03

Deep
 Time
 Aeolian
 Correlation
 0.03

Greasewood
 Root.Shoot
 Ratio
 0.03

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.03

Kd
 Silt
 for
 Cs
 (mL/g)
 0.03

Kd
 Clay
 for
 Pa
 (mL/g)
 0.03

Kd
 Sand
 for
 Np
 (mL/g)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.02

Deep
 Time
 DCF
 Photon
 1
 REF
 0.02

Kd
 Clay
 for
 Th
 (mL/g)
 0.02

Kd
 Clay
 for
 Am
 (mL/g)
 0.02

RipRap
 Porosity
 0.02

Kd
 Sand
 for
 Cs
 (mL/g)
 0.02

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.02

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.02

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.02

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.02

Deep
 Time
 DCF
 Alpha
 REF
 0.02

Deep
 Lake
 Depth
 (m)
 0.02

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.02

Kd
 Silt
 for
 Sr
 (mL/g)
 0.02

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.02

Biomass
 %
 Cover
 Selector
 0.02

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.02

Unit
 4
 ET
 Layers
 Porosity
 0.02

Shrub
 Root.Shoot
 Ratio
 0.02

Tree
 Root.Shoot
 Ratio
 0.02

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.02

Deep
 Time
 Receptor
 Area
 (ac)
 0.02

Kd
 Sand
 for
 Pb
 (mL/g)
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 5
Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.02

Grass
 Root
 Shape
 Parameter
 b
 0.02

Kd
 Clay
 for
 Ac
 (mL/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.02

Unit
 4
 Compacted
 Porosity
 0.02

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.02

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.02

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.02

Kd
 Sand
 for
 Tc
 (mL/g)
 0.02

Tortuosity
 Porosity
 Exponent
 0.02

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.02

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.02

Grass
 Root.Shoot
 Ratio
 0.02

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.02

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.02

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.02

Unit
 4
 Compacted
 Hb
 (cm)
 0.02

Resuspended
 Particle
 Fraction
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.02

Receptor
 Area
 (ha)
 0.02

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.02

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.02

Kd
 Sand
 for
 Pa
 (mL/g)
 0.02

Meat
 Post-­‐Cooking
 Loss
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Deep
 Time
 Lake
 Start
 (yr)
 0.01

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.01

OHV
 Dust
 Adjustment
 0.01

Kd
 Clay
 for
 Pu
 (mL/g)
 0.01

Kd
 Sand
 for
 Th
 (mL/g)
 0.01

Kd
 Silt
 for
 Ra
 (mL/g)
 0.01

Kd
 Silt
 for
 Pa
 (mL/g)
 0.01

Deep
 Time
 DCF
 Beta
 REF
 0.01

Kd
 Clay
 for
 Pb
 (mL/g)
 0.01

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.01

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Body
 Weight
 Factor
 for
 Antelope
 0.01

Plant.Soil
 Conc
 Ratio
 for
 U
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 6
Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.01

Soil
 Temperature
 (°C)
 0.01

Kd
 Clay
 for
 Cs
 (mL/g)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.01

Antelope
 Range
 Area
 (acre)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Kd
 Sand
 for
 Ra
 (mL/g)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Kd
 Clay
 for
 Ra
 (mL/g)
 0.01

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.01

Ant
 Colony
 Lifespan
 (yr)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 I
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.01

Kd
 Silt
 for
 Pu
 (mL/g)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.01

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.01

Forb
 Root
 Shape
 Parameter
 b
 0.01

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.01

Forb
 Root.Shoot
 Ratio
 0.01

Vegetation
 Association
 Selector
 0.01

Soil
 Ingestion
 Tracer
 Element
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 7
Table 2: Peak Groundwater Well Concentrations within 500 years – Np237
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Plant.Soil
 Conc
 Ratio
 for
 Tc
 19.05

Kd
 Sand
 for
 Ra
 (mL/g)
 9.87

Kd
 Clay
 for
 Np
 (mL/g)
 2.27

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 1.46

Silt
 Sand
 Gravel
 Porosity
 1.10

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 1.07

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 1.05

Kd
 Silt
 for
 Cs
 (mL/g)
 0.94

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.94

Unit
 4
 Compacted
 Hb
 (cm)
 0.90

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.90

Unit
 3
 Residual
 Water
 Content
 0.90

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.88

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.87

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.84

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.80

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.80

Plant.Soil
 Conc
 Ratio
 for
 U
 0.80

Unit
 3
 Porosity
 0.80

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.78

Unit
 4
 ET
 Layers
 Porosity
 0.78

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.76

Kd
 Clay
 for
 Pu
 (mL/g)
 0.76

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.74

Kd
 Sand
 for
 Th
 (mL/g)
 0.74

Fine
 Gravel
 Mix
 Porosity
 0.74

Unit
 4
 Compacted
 Porosity
 0.70

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.68

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.67

Kd
 Silt
 for
 Np
 (mL/g)
 0.67

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.67

Kd
 Clay
 for
 Sr
 (mL/g)
 0.66

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.64

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.64

Saturated
 Zone
 Water
 Table
 Gradient
 0.61

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.60

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.58

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.58

Kd
 Sand
 for
 U
 (mL/g)
 0.58

RipRap
 Porosity
 0.57

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 8
Fine
 CobbleMix
 Porosity
 0.56

Kd
 Sand
 for
 Cs
 (mL/g)
 0.55

Tree
 Root.Shoot
 Ratio
 0.55

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.53

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.53

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.52

Kd
 Sand
 for
 Np
 (mL/g)
 0.52

Kd
 Clay
 for
 Th
 (mL/g)
 0.50

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.50

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.50

RipRap
 Bulk
 Density
 (g/cm3)
 0.50

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.49

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.49

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.49

Deep
 Lake
 Depth
 (m)
 0.48

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.47

Ant
 Nest
 Volume
 (m3)
 0.47

Kd
 Clay
 for
 U
 (mL/g)
 0.46

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.46

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.46

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.45

Ant
 Colony
 Lifespan
 (yr)
 0.45

DCF
 Photon1
 REF
 0.45

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.45

Soil
 Temperature
 (°C)
 0.45

Deep
 Time
 DCF
 Photon
 2
 REF
 0.44

Kd
 Clay
 for
 Am
 (mL/g)
 0.43

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.43

Deep
 Time
 Lake
 Start
 (yr)
 0.43

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.43

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.43

OHV
 Dust
 Adjustment
 0.43

Kd
 Silt
 for
 Pa
 (mL/g)
 0.42

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.42

Kd
 Clay
 for
 Cs
 (mL/g)
 0.42

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.42

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.41

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.40

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.40

Kd
 Clay
 for
 Ra
 (mL/g)
 0.40

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.40

Grass
 Root.Shoot
 Ratio
 0.40

Kd
 Silt
 for
 U
 (mL/g)
 0.40

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 9
Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.38

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.37

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.37

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.37

Kd
 Silt
 for
 Ra
 (mL/g)
 0.36

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.36

Shrub
 Root
 Shape
 Parameter
 b
 0.35

Mammal
 Burrow
 Shape
 Parameter
 b
 0.35

Receptor
 Area
 (ha)
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.34

Surface
 Atmosphere
 Thickness
 (m)
 0.34

Saturated
 Zone
 Thickness
 (m)
 0.34

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.34

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.33

Kd
 Silt
 for
 Pu
 (mL/g)
 0.33

Kd
 Sand
 for
 Sr
 (mL/g)
 0.33

Kd
 Clay
 for
 Pa
 (mL/g)
 0.32

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.32

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.32

Kd
 Sand
 for
 Pb
 (mL/g)
 0.31

Kd
 Silt
 for
 Pb
 (mL/g)
 0.31

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.30

Body
 Weight
 Factor
 for
 Antelope
 0.29

Resuspended
 Particle
 Fraction
 0.28

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.28

Unit
 2
 Porosity
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.27

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.27

Grass
 Root
 Shape
 Parameter
 b
 0.26

Plant
 Fresh
 Weight
 Conversion
 0.25

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.25

Meat
 Post-­‐Cooking
 Loss
 0.24

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.23

Kd
 Sand
 for
 Pa
 (mL/g)
 0.23

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.23

Greasewood
 Root.Shoot
 Ratio
 0.23

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.23

Tortuosity
 Water
 Content
 Exponent

 0.22

Deep
 Time
 Diffusion
 Length
 (m)
 0.22

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.22

Kd
 Sand
 for
 Am
 (mL/g)
 0.22

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.21

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 10
Kd
 Silt
 for
 Sr
 (mL/g)
 0.21

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.20

Plant.Soil
 Conc
 Ratio
 for
 I
 0.19

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.19

Forb
 Root.Shoot
 Ratio
 0.18

Vegetation
 Association
 Selector
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.18

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.18

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.18

Biomass
 %
 Cover
 Selector
 0.18

DCF
 Alpha
 REF
 0.17

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.17

Meat
 Preparation
 Loss
 0.16

Kd
 Silt
 for
 Am
 (mL/g)
 0.16

Kd
 Clay
 for
 Ac
 (mL/g)
 0.15

Kd
 Sand
 for
 I
 (mL/g)
 0.15

Kd
 Sand
 for
 Pu
 (mL/g)
 0.15

Kd
 Sand
 for
 Ac
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.14

Site
 Dispersal
 Area
 (km2)
 0.14

Surface
 Wind
 Speed
 (m/s)
 0.13

Deep
 Time
 DCF
 Photon
 1
 REF
 0.13

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.13

Kd
 Clay
 for
 Pb
 (mL/g)
 0.13

Greasewood
 Root
 Shape
 Parameter
 b
 0.12

DCF
 Beta
 REF
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.12

Tree
 Root
 Shape
 Parameter
 b
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.12

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.12

Kd
 Silt
 for
 Th
 (mL/g)
 0.11

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.10

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.10

Soil
 Ingestion
 Tracer
 Element
 0.10

Forb
 Root
 Shape
 Parameter
 b
 0.10

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.09

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.09

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.09

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.08

Deep
 Time
 Aeolian
 Correlation
 0.08

Random
 Gully
 Selector
 0.08

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 11
Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.08

Ant
 Nest
 Shape
 Parameter
 b
 0.07

Kd
 Silt
 for
 Ac
 (mL/g)
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.07

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.07

Shrub
 Root.Shoot
 Ratio
 0.07

Deep
 Time
 DCF
 Beta
 REF
 0.07

Intermediate
 Lake
 Depth
 (m)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.06

Tortuosity
 Porosity
 Exponent
 0.06

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.05

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.05

Deep
 Time
 Receptor
 Area
 (ac)
 0.04

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.04

Kd
 Sand
 for
 Tc
 (mL/g)
 0.04

Contaminated
 Fraction
 of
 GDP
 DU
 0.04

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.04

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.03

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Deep
 Time
 DCF
 Alpha
 REF
 0.03

Antelope
 Range
 Area
 (acre)
 0.03

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.02

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.02

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.02

DCF
 Photon2
 REF
 0.01

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Tc
 19.05

Kd
 Sand
 for
 Ra
 (mL/g)
 9.87

Kd
 Clay
 for
 Np
 (mL/g)
 2.27

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 1.46

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 12
Table 3: Peak Groundwater Well Concentrations within 500 years - Tc99
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 31.97

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 24.96

Kd
 Sand
 for
 Tc
 (mL/g)
 13.97

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 10.59

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 3.83

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 1.26

Saturated
 Zone
 Water
 Table
 Gradient
 1.20

OHV
 Dust
 Adjustment
 0.55

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.38

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.34

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.34

Fine
 CobbleMix
 Porosity
 0.28

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.26

Kd
 Silt
 for
 Ra
 (mL/g)
 0.22

Surface
 Atmosphere
 Thickness
 (m)
 0.21

Unit
 4
 Compacted
 Hb
 (cm)
 0.20

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.20

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.19

Kd
 Silt
 for
 U
 (mL/g)
 0.17

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.16

Kd
 Clay
 for
 Sr
 (mL/g)
 0.15

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.14

Kd
 Sand
 for
 Ac
 (mL/g)
 0.14

Kd
 Clay
 for
 Ra
 (mL/g)
 0.13

Forb
 Root
 Shape
 Parameter
 b
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.12

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.12

Kd
 Silt
 for
 Cs
 (mL/g)
 0.11

Unit
 4
 Compacted
 Porosity
 0.11

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.11

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.11

Unit
 2
 Porosity
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.10

Shrub
 Root.Shoot
 Ratio
 0.10

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.10

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.10

Grass
 Root.Shoot
 Ratio
 0.09

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 13
Shrub
 Root
 Shape
 Parameter
 b
 0.09

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.09

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.09

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.08

Deep
 Time
 DCF
 Alpha
 REF
 0.08

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.08

Saturated
 Zone
 Thickness
 (m)
 0.08

Kd
 Sand
 for
 Am
 (mL/g)
 0.08

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Kd
 Silt
 for
 Sr
 (mL/g)
 0.07

RipRap
 Bulk
 Density
 (g/cm3)
 0.07

Kd
 Clay
 for
 Cs
 (mL/g)
 0.07

Deep
 Time
 DCF
 Photon
 2
 REF
 0.07

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.07

Kd
 Clay
 for
 Ac
 (mL/g)
 0.07

Unit
 3
 Residual
 Water
 Content
 0.07

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.07

Deep
 Time
 Lake
 Start
 (yr)
 0.07

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.06

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.06

Grass
 Root
 Shape
 Parameter
 b
 0.06

Ant
 Nest
 Volume
 (m3)
 0.06

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.06

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.06

Ant
 Nest
 Shape
 Parameter
 b
 0.06

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.06

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.06

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.06

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.06

Deep
 Time
 DCF
 Photon
 1
 REF
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.05

Fine
 Gravel
 Mix
 Porosity
 0.05

RipRap
 Porosity
 0.05

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.05

Vegetation
 Association
 Selector
 0.05

Plant.Soil
 Conc
 Ratio
 for
 U
 0.05

Surface
 Wind
 Speed
 (m/s)
 0.05

Soil
 Ingestion
 Tracer
 Element
 0.05

Tortuosity
 Water
 Content
 Exponent

 0.05

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 14
Intermediate
 Lake
 Depth
 (m)
 0.05

Kd
 Sand
 for
 Pa
 (mL/g)
 0.05

Unit
 3
 Porosity
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.05

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.05

Kd
 Silt
 for
 Th
 (mL/g)
 0.05

Kd
 Sand
 for
 U
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.05

Kd
 Clay
 for
 Am
 (mL/g)
 0.05

Site
 Dispersal
 Area
 (km2)
 0.05

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.05

Random
 Gully
 Selector
 0.04

Kd
 Sand
 for
 Th
 (mL/g)
 0.04

Antelope
 Range
 Area
 (acre)
 0.04

Kd
 Clay
 for
 Pu
 (mL/g)
 0.04

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.04

Kd
 Sand
 for
 Np
 (mL/g)
 0.04

Tortuosity
 Porosity
 Exponent
 0.04

Ant
 Colony
 Lifespan
 (yr)
 0.04

Plant
 Fresh
 Weight
 Conversion
 0.04

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.04

Meat
 Post-­‐Cooking
 Loss
 0.04

Kd
 Sand
 for
 Cs
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.04

Body
 Weight
 Factor
 for
 Antelope
 0.04

Kd
 Silt
 for
 Pa
 (mL/g)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.04

Kd
 Sand
 for
 Ra
 (mL/g)
 0.04

Kd
 Silt
 for
 Np
 (mL/g)
 0.04

Kd
 Clay
 for
 Pb
 (mL/g)
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.04

Receptor
 Area
 (ha)
 0.04

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.04

DCF
 Alpha
 REF
 0.04

Biomass
 %
 Cover
 Selector
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.04

Unit
 4
 ET
 Layers
 Porosity
 0.04

Deep
 Time
 Aeolian
 Correlation
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 I
 0.04

Kd
 Sand
 for
 Pu
 (mL/g)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 15
DCF
 Beta
 REF
 0.03

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.03

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.03

Mammal
 Burrow
 Shape
 Parameter
 b
 0.03

Forb
 Root.Shoot
 Ratio
 0.03

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.03

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.03

Kd
 Silt
 for
 Am
 (mL/g)
 0.03

Deep
 Time
 DCF
 Beta
 REF
 0.03

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.03

DCF
 Photon1
 REF
 0.03

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.03

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.03

Kd
 Clay
 for
 Pa
 (mL/g)
 0.03

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.03

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.03

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.03

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.03

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.03

Kd
 Silt
 for
 Ac
 (mL/g)
 0.03

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.03

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.03

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.03

Tree
 Root.Shoot
 Ratio
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.03

Kd
 Silt
 for
 Pu
 (mL/g)
 0.03

Deep
 Time
 Diffusion
 Length
 (m)
 0.02

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.02

Silt
 Sand
 Gravel
 Porosity
 0.02

Deep
 Lake
 Depth
 (m)
 0.02

Resuspended
 Particle
 Fraction
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.02

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.02

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 16
Kd
 Sand
 for
 Sr
 (mL/g)
 0.02

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.02

Kd
 Clay
 for
 Th
 (mL/g)
 0.02

Soil
 Temperature
 (°C)
 0.02

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.02

Meat
 Preparation
 Loss
 0.02

Kd
 Sand
 for
 Pb
 (mL/g)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.02

DCF
 Photon2
 REF
 0.02

Kd
 Silt
 for
 Pb
 (mL/g)
 0.02

Greasewood
 Root.Shoot
 Ratio
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.02

Greasewood
 Root
 Shape
 Parameter
 b
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.02

Kd
 Sand
 for
 I
 (mL/g)
 0.02

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Kd
 Clay
 for
 U
 (mL/g)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 17
Table 4: Peak Groundwater Well Concentrations within 500 years – Th230
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 U
 (mL/g)
 6.82

DCF
 Beta
 REF
 6.30

Ant
 Nest
 Volume
 (m3)
 3.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 3.30

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 3.28

DCF
 Photon1
 REF
 3.12

Kd
 Silt
 for
 Np
 (mL/g)
 2.65

Kd
 Clay
 for
 Pb
 (mL/g)
 2.65

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 2.61

Deep
 Time
 Diffusion
 Length
 (m)
 2.56

Kd
 Sand
 for
 Pa
 (mL/g)
 2.49

Resuspension
 Flux
 (kg.m2-­‐yr)
 2.49

Meat
 Preparation
 Loss
 2.45

Grass
 Root.Shoot
 Ratio
 2.36

Kd
 Clay
 for
 U
 (mL/g)
 2.36

Meat
 Post-­‐Cooking
 Loss
 2.35

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 2.25

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 2.16

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 2.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 1.99

Unit
 4
 Compacted
 Residual
 Water
 Content
 1.97

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.77

Plant
 Fresh
 Weight
 Conversion
 1.70

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 1.61

Forb
 Root.Shoot
 Ratio
 1.56

Plant.Soil
 Conc
 Ratio
 for
 U
 1.52

Plant.Soil
 Conc
 Ratio
 for
 Pu
 1.52

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 1.49

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 1.49

Unit
 2
 Bulk
 Density
 (g/cm3)
 1.49

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 1.40

Plant.Soil
 Conc
 Ratio
 for
 Tc
 1.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 1.32

Plant.Soil
 Conc
 Ratio
 for
 I
 1.13

RipRap
 Bulk
 Density
 (g/cm3)
 1.12

Kd
 Clay
 for
 Th
 (mL/g)
 1.09

Surface
 Atmosphere
 Thickness
 (m)
 0.99

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.91

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.89

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.88

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 18
Greasewood
 Root.Shoot
 Ratio
 0.78

Deep
 Lake
 Depth
 (m)
 0.77

Kd
 Sand
 for
 Np
 (mL/g)
 0.68

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.63

Kd
 Clay
 for
 Am
 (mL/g)
 0.62

Forb
 Root
 Shape
 Parameter
 b
 0.55

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.54

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.52

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.47

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.47

Kd
 Sand
 for
 Ac
 (mL/g)
 0.45

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.44

Receptor
 Area
 (ha)
 0.40

OHV
 Dust
 Adjustment
 0.37

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.36

Saturated
 Zone
 Water
 Table
 Gradient
 0.30

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.24

Surface
 Wind
 Speed
 (m/s)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.20

Deep
 Time
 Aeolian
 Correlation
 0.19

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.18

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.16

Kd
 Sand
 for
 Th
 (mL/g)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.13

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.13

Tortuosity
 Water
 Content
 Exponent

 0.13

Vegetation
 Association
 Selector
 0.11

Kd
 Clay
 for
 Pa
 (mL/g)
 0.11

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.08

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.08

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.08

Mammal
 Burrow
 Shape
 Parameter
 b
 0.07

RipRap
 Porosity
 0.07

Unit
 3
 Residual
 Water
 Content
 0.07

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.07

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.07

Fine
 Gravel
 Mix
 Porosity
 0.06

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.06

Kd
 Silt
 for
 U
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 19
Unit
 4
 ET
 Layers
 Porosity
 0.04

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.04

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.04

Unit
 3
 Porosity
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.03

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.03

Ant
 Colony
 Lifespan
 (yr)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.03

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.03

Unit
 2
 Porosity
 0.03

Unit
 4
 Compacted
 Hb
 (cm)
 0.03

Silt
 Sand
 Gravel
 Porosity
 0.03

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.03

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.03

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.03

Kd
 Silt
 for
 Ra
 (mL/g)
 0.03

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.03

Kd
 Clay
 for
 Ac
 (mL/g)
 0.03

Kd
 Clay
 for
 Sr
 (mL/g)
 0.03

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.03

Fine
 CobbleMix
 Porosity
 0.02

Random
 Gully
 Selector
 0.02

Unit
 4
 Compacted
 Porosity
 0.02

Tree
 Root.Shoot
 Ratio
 0.02

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.02

Kd
 Silt
 for
 Am
 (mL/g)
 0.02

Intermediate
 Lake
 Depth
 (m)
 0.02

Kd
 Silt
 for
 Th
 (mL/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.02

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.02

Resuspended
 Particle
 Fraction
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.02

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.02

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.02

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.02

Kd
 Silt
 for
 Sr
 (mL/g)
 0.02

Kd
 Silt
 for
 Ac
 (mL/g)
 0.02

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.02

Site
 Dispersal
 Area
 (km2)
 0.02

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.01

Saturated
 Zone
 Thickness
 (m)
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 20
Kd
 Sand
 for
 Am
 (mL/g)
 0.01

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.01

Soil
 Temperature
 (°C)
 0.01

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.01

Kd
 Silt
 for
 Pu
 (mL/g)
 0.01

Biomass
 %
 Cover
 Selector
 0.01

Kd
 Sand
 for
 Cs
 (mL/g)
 0.01

Tortuosity
 Porosity
 Exponent
 0.01

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.01

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.01

Kd
 Silt
 for
 Cs
 (mL/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.01

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Kd
 Sand
 for
 Pb
 (mL/g)
 0.01

Shrub
 Root.Shoot
 Ratio
 0.01

Kd
 Sand
 for
 Sr
 (mL/g)
 0.01

Kd
 Sand
 for
 Ra
 (mL/g)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.01

Kd
 Clay
 for
 Pu
 (mL/g)
 0.01

Greasewood
 Root
 Shape
 Parameter
 b
 0.01

Grass
 Root
 Shape
 Parameter
 b
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.01

Ant
 Nest
 Shape
 Parameter
 b
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.01

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Kd
 Clay
 for
 Ra
 (mL/g)
 0.01

Deep
 Time
 Lake
 Start
 (yr)
 0.01

Kd
 Clay
 for
 Cs
 (mL/g)
 0.01

Kd
 Silt
 for
 Pa
 (mL/g)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Body
 Weight
 Factor
 for
 Antelope
 0.01

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 21
Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.01

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.01

Antelope
 Range
 Area
 (acre)
 0.00

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.00

Deep
 Time
 DCF
 Photon
 1
 REF
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.00

Deep
 Time
 DCF
 Photon
 2
 REF
 0.00

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.00

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.00

Deep
 Time
 DCF
 Alpha
 REF
 0.00

Kd
 Sand
 for
 Pu
 (mL/g)
 0.00

Kd
 Silt
 for
 Pb
 (mL/g)
 0.00

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.00

DCF
 Photon2
 REF
 0.00

Contaminated
 Fraction
 of
 GDP
 DU
 0.00

Deep
 Time
 DCF
 Beta
 REF
 0.00

Kd
 Sand
 for
 I
 (mL/g)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.00

DCF
 Alpha
 REF
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.00

Soil
 Ingestion
 Tracer
 Element
 0.00

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.00

Kd
 Sand
 for
 Tc
 (mL/g)
 0.00

Shrub
 Root
 Shape
 Parameter
 b
 0.00

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 22
Table 5: Peak Groundwater Well Concentrations within 500 years – Th232
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

DCF
 Beta
 REF
 15.55

Kd
 Sand
 for
 U
 (mL/g)
 5.77

Kd
 Clay
 for
 Pb
 (mL/g)
 3.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 2.93

Ant
 Nest
 Volume
 (m3)
 2.93

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 2.83

DCF
 Photon1
 REF
 2.43

Deep
 Time
 Diffusion
 Length
 (m)
 2.27

Kd
 Sand
 for
 Pa
 (mL/g)
 2.20

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 2.14

Kd
 Silt
 for
 Np
 (mL/g)
 2.07

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 2.07

Kd
 Clay
 for
 U
 (mL/g)
 2.05

Grass
 Root.Shoot
 Ratio
 1.91

Meat
 Post-­‐Cooking
 Loss
 1.91

Meat
 Preparation
 Loss
 1.83

Resuspension
 Flux
 (kg.m2-­‐yr)
 1.80

Plant
 Fresh
 Weight
 Conversion
 1.78

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 1.74

Unit
 2
 Bulk
 Density
 (g/cm3)
 1.61

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 1.61

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.54

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 1.51

Unit
 4
 Compacted
 Residual
 Water
 Content
 1.47

Plant.Soil
 Conc
 Ratio
 for
 U
 1.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 1.36

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 1.36

Plant.Soil
 Conc
 Ratio
 for
 Tc
 1.29

Forb
 Root.Shoot
 Ratio
 1.27

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 1.26

RipRap
 Bulk
 Density
 (g/cm3)
 1.23

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 1.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 1.16

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 1.06

Plant.Soil
 Conc
 Ratio
 for
 Pu
 1.03

Kd
 Clay
 for
 Th
 (mL/g)
 1.02

Plant.Soil
 Conc
 Ratio
 for
 I
 0.93

Surface
 Atmosphere
 Thickness
 (m)
 0.91

Greasewood
 Root.Shoot
 Ratio
 0.91

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.88

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 23
Kd
 Sand
 for
 Np
 (mL/g)
 0.66

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.66

Deep
 Lake
 Depth
 (m)
 0.63

Kd
 Sand
 for
 Ac
 (mL/g)
 0.58

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.57

Kd
 Clay
 for
 Am
 (mL/g)
 0.56

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.52

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.51

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.37

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.36

Forb
 Root
 Shape
 Parameter
 b
 0.34

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.34

Saturated
 Zone
 Water
 Table
 Gradient
 0.31

OHV
 Dust
 Adjustment
 0.29

Receptor
 Area
 (ha)
 0.29

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.29

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.28

Fine
 Gravel
 Mix
 Porosity
 0.26

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.25

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.25

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.24

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.24

RipRap
 Porosity
 0.23

Unit
 3
 Residual
 Water
 Content
 0.23

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.23

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.21

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.19

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.18

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.18

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.18

Surface
 Wind
 Speed
 (m/s)
 0.17

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.16

Unit
 4
 ET
 Layers
 Porosity
 0.15

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.14

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.13

Unit
 3
 Porosity
 0.13

Tortuosity
 Water
 Content
 Exponent

 0.13

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.11

Deep
 Time
 Aeolian
 Correlation
 0.09

Fine
 CobbleMix
 Porosity
 0.09

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.09

Unit
 4
 Compacted
 Porosity
 0.08

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 24
Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.08

Unit
 4
 Compacted
 Hb
 (cm)
 0.07

Unit
 2
 Porosity
 0.07

Silt
 Sand
 Gravel
 Porosity
 0.07

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.06

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.06

Kd
 Sand
 for
 Th
 (mL/g)
 0.06

Deep
 Time
 DCF
 Beta
 REF
 0.06

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.06

Vegetation
 Association
 Selector
 0.06

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.06

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.05

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.05

Kd
 Clay
 for
 Pa
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.04

Kd
 Sand
 for
 Am
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.04

Kd
 Silt
 for
 Ra
 (mL/g)
 0.04

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.04

Kd
 Sand
 for
 Cs
 (mL/g)
 0.04

Kd
 Silt
 for
 Sr
 (mL/g)
 0.04

Kd
 Silt
 for
 Pu
 (mL/g)
 0.04

Kd
 Silt
 for
 Ac
 (mL/g)
 0.03

Kd
 Silt
 for
 Cs
 (mL/g)
 0.03

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.03

Kd
 Sand
 for
 Pu
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.03

Kd
 Silt
 for
 Th
 (mL/g)
 0.03

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Mammal
 Burrow
 Shape
 Parameter
 b
 0.03

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.03

Kd
 Clay
 for
 Sr
 (mL/g)
 0.03

Kd
 Clay
 for
 Ac
 (mL/g)
 0.03

Kd
 Silt
 for
 U
 (mL/g)
 0.03

Kd
 Sand
 for
 Ra
 (mL/g)
 0.03

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.03

Kd
 Silt
 for
 Am
 (mL/g)
 0.03

Kd
 Sand
 for
 Sr
 (mL/g)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Kd
 Sand
 for
 Pb
 (mL/g)
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 25
Ant
 Colony
 Lifespan
 (yr)
 0.02

Kd
 Clay
 for
 Cs
 (mL/g)
 0.02

Shrub
 Root.Shoot
 Ratio
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.02

Kd
 Clay
 for
 Ra
 (mL/g)
 0.02

Ant
 Nest
 Shape
 Parameter
 b
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.02

Kd
 Clay
 for
 Pu
 (mL/g)
 0.02

Grass
 Root
 Shape
 Parameter
 b
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.02

Tree
 Root.Shoot
 Ratio
 0.02

Soil
 Temperature
 (°C)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Tortuosity
 Porosity
 Exponent
 0.01

Site
 Dispersal
 Area
 (km2)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.01

Resuspended
 Particle
 Fraction
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.01

Body
 Weight
 Factor
 for
 Antelope
 0.01

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.01

Deep
 Time
 DCF
 Photon
 2
 REF
 0.01

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Biomass
 %
 Cover
 Selector
 0.01

Random
 Gully
 Selector
 0.01

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Saturated
 Zone
 Thickness
 (m)
 0.01

Intermediate
 Lake
 Depth
 (m)
 0.01

Kd
 Silt
 for
 Pa
 (mL/g)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.01

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.01

Kd
 Silt
 for
 Pb
 (mL/g)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 26
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.01

DCF
 Photon2
 REF
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.01

Deep
 Time
 DCF
 Alpha
 REF
 0.01

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.01

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.01

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Deep
 Time
 Lake
 Start
 (yr)
 0.01

Deep
 Time
 DCF
 Photon
 1
 REF
 0.00

Antelope
 Range
 Area
 (acre)
 0.00

Greasewood
 Root
 Shape
 Parameter
 b
 0.00

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.00

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.00

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.00

Contaminated
 Fraction
 of
 GDP
 DU
 0.00

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.00

Shrub
 Root
 Shape
 Parameter
 b
 0.00

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.00

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.00

Kd
 Sand
 for
 I
 (mL/g)
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.00

Soil
 Ingestion
 Tracer
 Element
 0.00

DCF
 Alpha
 REF
 0.00

Kd
 Sand
 for
 Tc
 (mL/g)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 27
Table 6: Peak Groundwater Well Concentrations within 500 years – U233
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

DCF
 Photon1
 REF
 18.86

Plant.Soil
 Conc
 Ratio
 for
 Tc
 5.50

Grass
 Root.Shoot
 Ratio
 5.38

Ant
 Nest
 Volume
 (m3)
 3.02

Kd
 Clay
 for
 Cs
 (mL/g)
 2.85

Kd
 Sand
 for
 Ra
 (mL/g)
 2.46

Kd
 Sand
 for
 U
 (mL/g)
 2.45

DCF
 Beta
 REF
 2.38

Kd
 Clay
 for
 Pb
 (mL/g)
 1.86

Kd
 Clay
 for
 U
 (mL/g)
 1.66

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 1.64

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 1.54

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.50

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 1.03

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.99

Kd
 Silt
 for
 Np
 (mL/g)
 0.99

Plant
 Fresh
 Weight
 Conversion
 0.96

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.95

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.88

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.82

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.80

Kd
 Sand
 for
 Pa
 (mL/g)
 0.78

Kd
 Sand
 for
 Np
 (mL/g)
 0.76

Deep
 Time
 Diffusion
 Length
 (m)
 0.76

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.74

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.72

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.70

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.70

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.68

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.66

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.64

Greasewood
 Root.Shoot
 Ratio
 0.60

Plant.Soil
 Conc
 Ratio
 for
 U
 0.60

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.58

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.56

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.55

Forb
 Root.Shoot
 Ratio
 0.55

Meat
 Post-­‐Cooking
 Loss
 0.54

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.53

Kd
 Clay
 for
 Np
 (mL/g)
 0.52

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 28
Unit
 4
 ET
 Layers
 Porosity
 0.52

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.48

Kd
 Clay
 for
 Th
 (mL/g)
 0.48

Kd
 Silt
 for
 Cs
 (mL/g)
 0.44

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.44

Plant.Soil
 Conc
 Ratio
 for
 I
 0.44

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.43

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.43

RipRap
 Bulk
 Density
 (g/cm3)
 0.42

Fine
 Gravel
 Mix
 Porosity
 0.41

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.40

Silt
 Sand
 Gravel
 Porosity
 0.40

Unit
 4
 Compacted
 Porosity
 0.39

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.39

Kd
 Sand
 for
 Ac
 (mL/g)
 0.39

Meat
 Preparation
 Loss
 0.38

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.38

Tortuosity
 Porosity
 Exponent
 0.38

Saturated
 Zone
 Water
 Table
 Gradient
 0.38

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.36

Surface
 Atmosphere
 Thickness
 (m)
 0.36

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.36

Kd
 Silt
 for
 Am
 (mL/g)
 0.36

Unit
 3
 Porosity
 0.35

RipRap
 Porosity
 0.34

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.32

Kd
 Sand
 for
 Sr
 (mL/g)
 0.32

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.32

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.31

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.30

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.30

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.29

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.29

Unit
 4
 Compacted
 Hb
 (cm)
 0.29

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.29

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.29

Deep
 Lake
 Depth
 (m)
 0.27

Kd
 Silt
 for
 Pu
 (mL/g)
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.27

Receptor
 Area
 (ha)
 0.26

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.25

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.25

Kd
 Clay
 for
 Pu
 (mL/g)
 0.24

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 29
Unit
 3
 Residual
 Water
 Content
 0.24

OHV
 Dust
 Adjustment
 0.24

Unit
 2
 Porosity
 0.24

Forb
 Root
 Shape
 Parameter
 b
 0.23

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.23

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.23

Kd
 Sand
 for
 Am
 (mL/g)
 0.22

Kd
 Silt
 for
 Pa
 (mL/g)
 0.22

Kd
 Clay
 for
 Am
 (mL/g)
 0.22

Kd
 Silt
 for
 U
 (mL/g)
 0.21

Kd
 Clay
 for
 Sr
 (mL/g)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.21

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.20

Kd
 Sand
 for
 Th
 (mL/g)
 0.20

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.20

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.20

Kd
 Sand
 for
 Cs
 (mL/g)
 0.20

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.19

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.19

Tree
 Root.Shoot
 Ratio
 0.19

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.18

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.18

Body
 Weight
 Factor
 for
 Antelope
 0.18

Fine
 CobbleMix
 Porosity
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.18

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.17

Biomass
 %
 Cover
 Selector
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.17

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.17

Kd
 Clay
 for
 Ra
 (mL/g)
 0.17

Tree
 Root
 Shape
 Parameter
 b
 0.17

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.17

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.16

Mammal
 Burrow
 Shape
 Parameter
 b
 0.16

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.16

Tortuosity
 Water
 Content
 Exponent

 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.16

Kd
 Silt
 for
 Ra
 (mL/g)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.15

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 30
Random
 Gully
 Selector
 0.15

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.15

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.15

Kd
 Sand
 for
 Pb
 (mL/g)
 0.15

Ant
 Colony
 Lifespan
 (yr)
 0.14

Soil
 Temperature
 (°C)
 0.14

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.14

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.13

Surface
 Wind
 Speed
 (m/s)
 0.13

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.13

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.13

Deep
 Time
 DCF
 Photon
 2
 REF
 0.13

Deep
 Time
 Lake
 Start
 (yr)
 0.12

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.12

Saturated
 Zone
 Thickness
 (m)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.12

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.11

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.11

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.11

Shrub
 Root
 Shape
 Parameter
 b
 0.11

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.10

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.10

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.10

DCF
 Photon2
 REF
 0.09

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.09

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.09

Kd
 Silt
 for
 Pb
 (mL/g)
 0.09

Kd
 Clay
 for
 Pa
 (mL/g)
 0.09

DCF
 Alpha
 REF
 0.09

Deep
 Time
 DCF
 Photon
 1
 REF
 0.09

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.08

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.08

Vegetation
 Association
 Selector
 0.08

Kd
 Silt
 for
 Th
 (mL/g)
 0.08

Resuspended
 Particle
 Fraction
 0.08

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.08

Greasewood
 Root
 Shape
 Parameter
 b
 0.08

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.07

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.07

Ant
 Nest
 Shape
 Parameter
 b
 0.07

Kd
 Silt
 for
 Ac
 (mL/g)
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.07

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 31
Deep
 Time
 Aeolian
 Correlation
 0.07

Antelope
 Range
 Area
 (acre)
 0.07

Kd
 Sand
 for
 Tc
 (mL/g)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.07

Site
 Dispersal
 Area
 (km2)
 0.06

Soil
 Ingestion
 Tracer
 Element
 0.06

Deep
 Time
 DCF
 Alpha
 REF
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.06

Grass
 Root
 Shape
 Parameter
 b
 0.06

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.06

Kd
 Sand
 for
 I
 (mL/g)
 0.06

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.06

Kd
 Sand
 for
 Pu
 (mL/g)
 0.06

Kd
 Silt
 for
 Sr
 (mL/g)
 0.05

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.04

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.04

Kd
 Clay
 for
 Ac
 (mL/g)
 0.04

Shrub
 Root.Shoot
 Ratio
 0.03

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.03

Deep
 Time
 Receptor
 Area
 (ac)
 0.03

Intermediate
 Lake
 Depth
 (m)
 0.03

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Contaminated
 Fraction
 of
 GDP
 DU
 0.02

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.02

Deep
 Time
 DCF
 Beta
 REF
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 32
Table 7: Peak Groundwater Well Concentrations within 500 years – U234
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 U
 (mL/g)
 6.29

DCF
 Beta
 REF
 5.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 4.25

Kd
 Clay
 for
 Pb
 (mL/g)
 4.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 3.22

Ant
 Nest
 Volume
 (m3)
 3.05

DCF
 Photon1
 REF
 2.97

Kd
 Clay
 for
 U
 (mL/g)
 2.88

Kd
 Silt
 for
 Np
 (mL/g)
 2.58

Plant
 Fresh
 Weight
 Conversion
 2.33

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 2.27

Kd
 Sand
 for
 Pa
 (mL/g)
 2.27

Deep
 Time
 Diffusion
 Length
 (m)
 2.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 2.26

Meat
 Post-­‐Cooking
 Loss
 2.21

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 2.16

Grass
 Root.Shoot
 Ratio
 2.00

Meat
 Preparation
 Loss
 1.95

Resuspension
 Flux
 (kg.m2-­‐yr)
 1.89

Unit
 2
 Bulk
 Density
 (g/cm3)
 1.84

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.76

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 1.75

Unit
 4
 Compacted
 Residual
 Water
 Content
 1.66

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 1.60

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 1.56

Forb
 Root.Shoot
 Ratio
 1.51

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 1.49

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 1.38

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 1.35

Plant.Soil
 Conc
 Ratio
 for
 U
 1.35

Plant.Soil
 Conc
 Ratio
 for
 Pu
 1.35

Plant.Soil
 Conc
 Ratio
 for
 Tc
 1.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 1.22

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 1.20

Kd
 Clay
 for
 Th
 (mL/g)
 1.17

Plant.Soil
 Conc
 Ratio
 for
 I
 1.11

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.08

RipRap
 Bulk
 Density
 (g/cm3)
 1.02

Kd
 Sand
 for
 Np
 (mL/g)
 0.98

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.92

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 33
Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.91

Surface
 Atmosphere
 Thickness
 (m)
 0.89

Greasewood
 Root.Shoot
 Ratio
 0.88

Deep
 Lake
 Depth
 (m)
 0.84

Kd
 Sand
 for
 Ac
 (mL/g)
 0.76

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.62

Saturated
 Zone
 Water
 Table
 Gradient
 0.61

Kd
 Clay
 for
 Am
 (mL/g)
 0.54

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.51

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.49

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.48

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.48

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.47

Forb
 Root
 Shape
 Parameter
 b
 0.46

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.39

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.32

OHV
 Dust
 Adjustment
 0.31

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.30

Tortuosity
 Water
 Content
 Exponent

 0.26

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.20

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.17

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.17

Receptor
 Area
 (ha)
 0.17

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.16

Surface
 Wind
 Speed
 (m/s)
 0.14

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.13

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.12

Deep
 Time
 Aeolian
 Correlation
 0.10

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.10

RipRap
 Porosity
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.09

Ant
 Colony
 Lifespan
 (yr)
 0.08

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.08

Unit
 3
 Residual
 Water
 Content
 0.08

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.08

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.07

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.07

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.07

Fine
 Gravel
 Mix
 Porosity
 0.07

Unit
 4
 ET
 Layers
 Porosity
 0.07

Unit
 3
 Porosity
 0.06

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.06

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 34
Kd
 Silt
 for
 Ra
 (mL/g)
 0.06

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.06

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.05

Unit
 4
 Compacted
 Porosity
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.05

Kd
 Clay
 for
 Pa
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.05

Fine
 CobbleMix
 Porosity
 0.05

Tree
 Root.Shoot
 Ratio
 0.05

Kd
 Sand
 for
 Th
 (mL/g)
 0.04

Unit
 2
 Porosity
 0.04

Kd
 Clay
 for
 Pu
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.04

Unit
 4
 Compacted
 Hb
 (cm)
 0.04

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.04

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.04

Silt
 Sand
 Gravel
 Porosity
 0.04

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.04

Mammal
 Burrow
 Shape
 Parameter
 b
 0.03

Ant
 Nest
 Shape
 Parameter
 b
 0.03

Kd
 Silt
 for
 Sr
 (mL/g)
 0.03

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.03

Shrub
 Root.Shoot
 Ratio
 0.03

Body
 Weight
 Factor
 for
 Antelope
 0.03

Kd
 Silt
 for
 Am
 (mL/g)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.03

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.02

Resuspended
 Particle
 Fraction
 0.02

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.02

Vegetation
 Association
 Selector
 0.02

Kd
 Silt
 for
 Pu
 (mL/g)
 0.02

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.02

Kd
 Silt
 for
 Cs
 (mL/g)
 0.02

Kd
 Silt
 for
 Th
 (mL/g)
 0.02

Soil
 Temperature
 (°C)
 0.02

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.02

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.02

Kd
 Silt
 for
 Ac
 (mL/g)
 0.02

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.02

Kd
 Sand
 for
 Am
 (mL/g)
 0.02

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.02

Kd
 Sand
 for
 Cs
 (mL/g)
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 35
Kd
 Clay
 for
 Ac
 (mL/g)
 0.02

Saturated
 Zone
 Thickness
 (m)
 0.02

Kd
 Clay
 for
 Ra
 (mL/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.02

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.02

Deep
 Time
 DCF
 Photon
 2
 REF
 0.02

Kd
 Clay
 for
 Sr
 (mL/g)
 0.02

Kd
 Sand
 for
 Ra
 (mL/g)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.01

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Deep
 Time
 Lake
 Start
 (yr)
 0.01

Kd
 Silt
 for
 Pb
 (mL/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.01

Kd
 Sand
 for
 Sr
 (mL/g)
 0.01

DCF
 Photon2
 REF
 0.01

Random
 Gully
 Selector
 0.01

Kd
 Clay
 for
 Cs
 (mL/g)
 0.01

Kd
 Silt
 for
 U
 (mL/g)
 0.01

Grass
 Root
 Shape
 Parameter
 b
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Kd
 Sand
 for
 Pu
 (mL/g)
 0.01

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.01

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.01

Intermediate
 Lake
 Depth
 (m)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.01

Tortuosity
 Porosity
 Exponent
 0.01

Kd
 Silt
 for
 Pa
 (mL/g)
 0.01

Kd
 Sand
 for
 Pb
 (mL/g)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.01

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 36
Greasewood
 Root
 Shape
 Parameter
 b
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.01

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.01

Site
 Dispersal
 Area
 (km2)
 0.01

Deep
 Time
 DCF
 Photon
 1
 REF
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Kd
 Sand
 for
 Tc
 (mL/g)
 0.01

Biomass
 %
 Cover
 Selector
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Deep
 Time
 DCF
 Beta
 REF
 0.01

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.01

Deep
 Time
 DCF
 Alpha
 REF
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.01

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Antelope
 Range
 Area
 (acre)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.00

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.00

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.00

Shrub
 Root
 Shape
 Parameter
 b
 0.00

Kd
 Sand
 for
 I
 (mL/g)
 0.00

DCF
 Alpha
 REF
 0.00

Soil
 Ingestion
 Tracer
 Element
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 37
Table 8: Peak Groundwater Well Concentrations within 500 years – U235
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 U
 (mL/g)
 6.13

DCF
 Beta
 REF
 5.90

Kd
 Clay
 for
 Pb
 (mL/g)
 5.36

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 4.91

Kd
 Clay
 for
 U
 (mL/g)
 4.21

Plant
 Fresh
 Weight
 Conversion
 2.88

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 2.73

DCF
 Photon1
 REF
 2.65

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 2.63

Ant
 Nest
 Volume
 (m3)
 2.55

Deep
 Time
 Diffusion
 Length
 (m)
 2.18

Unit
 2
 Bulk
 Density
 (g/cm3)
 2.13

Meat
 Post-­‐Cooking
 Loss
 2.05

Kd
 Silt
 for
 Np
 (mL/g)
 2.02

Kd
 Sand
 for
 Pa
 (mL/g)
 1.94

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 1.88

Resuspension
 Flux
 (kg.m2-­‐yr)
 1.81

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 1.75

Grass
 Root.Shoot
 Ratio
 1.71

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 1.66

Meat
 Preparation
 Loss
 1.53

Plant.Soil
 Conc
 Ratio
 for
 Pu
 1.51

Forb
 Root.Shoot
 Ratio
 1.48

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 1.41

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.40

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 1.31

Unit
 4
 Compacted
 Residual
 Water
 Content
 1.30

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 1.25

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 1.24

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 1.23

Greasewood
 Root.Shoot
 Ratio
 1.18

Plant.Soil
 Conc
 Ratio
 for
 U
 1.14

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 1.14

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 1.11

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.09

Plant.Soil
 Conc
 Ratio
 for
 Tc
 1.03

Plant.Soil
 Conc
 Ratio
 for
 I
 1.02

Kd
 Clay
 for
 Th
 (mL/g)
 0.93

RipRap
 Bulk
 Density
 (g/cm3)
 0.88

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.85

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 38
Kd
 Sand
 for
 Np
 (mL/g)
 0.84

Surface
 Atmosphere
 Thickness
 (m)
 0.77

Saturated
 Zone
 Water
 Table
 Gradient
 0.67

Forb
 Root
 Shape
 Parameter
 b
 0.66

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.66

Deep
 Lake
 Depth
 (m)
 0.65

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.65

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.63

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.61

Kd
 Sand
 for
 Ac
 (mL/g)
 0.56

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.51

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.44

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.44

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.39

Kd
 Clay
 for
 Am
 (mL/g)
 0.37

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.35

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.32

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.31

OHV
 Dust
 Adjustment
 0.30

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.24

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.24

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.24

Tortuosity
 Water
 Content
 Exponent

 0.23

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.19

Ant
 Colony
 Lifespan
 (yr)
 0.17

Deep
 Time
 Aeolian
 Correlation
 0.16

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.16

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.16

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.16

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.15

Receptor
 Area
 (ha)
 0.15

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.13

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.12

RipRap
 Porosity
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.11

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.10

Unit
 4
 Compacted
 Porosity
 0.10

Tree
 Root.Shoot
 Ratio
 0.10

Fine
 Gravel
 Mix
 Porosity
 0.09

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.09

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.09

Fine
 CobbleMix
 Porosity
 0.09

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 39
Unit
 3
 Porosity
 0.09

Unit
 3
 Residual
 Water
 Content
 0.09

Kd
 Silt
 for
 Ra
 (mL/g)
 0.09

Kd
 Clay
 for
 Pa
 (mL/g)
 0.09

Surface
 Wind
 Speed
 (m/s)
 0.09

Unit
 4
 ET
 Layers
 Porosity
 0.08

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.08

Unit
 2
 Porosity
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.07

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.07

Unit
 4
 Compacted
 Hb
 (cm)
 0.06

Body
 Weight
 Factor
 for
 Antelope
 0.06

Kd
 Silt
 for
 U
 (mL/g)
 0.06

Shrub
 Root.Shoot
 Ratio
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.06

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.06

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.05

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.05

Vegetation
 Association
 Selector
 0.05

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.05

Silt
 Sand
 Gravel
 Porosity
 0.04

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.04

Kd
 Sand
 for
 Pb
 (mL/g)
 0.04

Kd
 Sand
 for
 Th
 (mL/g)
 0.04

Kd
 Silt
 for
 Sr
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.04

Kd
 Sand
 for
 Cs
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.04

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.04

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.04

Kd
 Silt
 for
 Cs
 (mL/g)
 0.04

Kd
 Silt
 for
 Am
 (mL/g)
 0.04

Kd
 Clay
 for
 Pu
 (mL/g)
 0.03

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.03

Resuspended
 Particle
 Fraction
 0.03

Kd
 Silt
 for
 Th
 (mL/g)
 0.03

Kd
 Silt
 for
 Pu
 (mL/g)
 0.03

Deep
 Time
 DCF
 Photon
 2
 REF
 0.03

Saturated
 Zone
 Thickness
 (m)
 0.03

Kd
 Clay
 for
 Ac
 (mL/g)
 0.03

Kd
 Sand
 for
 Sr
 (mL/g)
 0.03

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 40
Soil
 Temperature
 (°C)
 0.03

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.03

Kd
 Sand
 for
 Am
 (mL/g)
 0.03

Deep
 Time
 Lake
 Start
 (yr)
 0.03

Kd
 Silt
 for
 Ac
 (mL/g)
 0.02

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.02

Random
 Gully
 Selector
 0.02

Deep
 Time
 Receptor
 Area
 (ac)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.02

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.02

Kd
 Clay
 for
 Cs
 (mL/g)
 0.02

Kd
 Sand
 for
 Ra
 (mL/g)
 0.02

Kd
 Silt
 for
 Pa
 (mL/g)
 0.02

Mammal
 Burrow
 Shape
 Parameter
 b
 0.02

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.02

Deep
 Time
 DCF
 Photon
 1
 REF
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Ant
 Nest
 Shape
 Parameter
 b
 0.02

Kd
 Clay
 for
 Sr
 (mL/g)
 0.02

Greasewood
 Root
 Shape
 Parameter
 b
 0.02

Kd
 Sand
 for
 Pu
 (mL/g)
 0.02

Kd
 Clay
 for
 Ra
 (mL/g)
 0.02

Kd
 Silt
 for
 Pb
 (mL/g)
 0.02

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.02

Contaminated
 Fraction
 of
 GDP
 DU
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.01

Tortuosity
 Porosity
 Exponent
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.01

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.01

Grass
 Root
 Shape
 Parameter
 b
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.01

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.01

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 41
Deep
 Time
 DCF
 Beta
 REF
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.01

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.01

DCF
 Photon2
 REF
 0.01

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Site
 Dispersal
 Area
 (km2)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Intermediate
 Lake
 Depth
 (m)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.01

Biomass
 %
 Cover
 Selector
 0.01

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Shrub
 Root
 Shape
 Parameter
 b
 0.01

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.01

Deep
 Time
 DCF
 Alpha
 REF
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.01

DCF
 Alpha
 REF
 0.01

Kd
 Sand
 for
 Tc
 (mL/g)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Antelope
 Range
 Area
 (acre)
 0.01

Kd
 Sand
 for
 I
 (mL/g)
 0.00

Soil
 Ingestion
 Tracer
 Element
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 42
Table 9: Peak Groundwater Well Concentrations within 500 years – U236
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

DCF
 Beta
 REF
 12.79

Kd
 Clay
 for
 U
 (mL/g)
 5.66

Kd
 Sand
 for
 U
 (mL/g)
 5.15

Kd
 Clay
 for
 Pb
 (mL/g)
 4.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 4.20

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 2.57

DCF
 Photon1
 REF
 2.47

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 2.41

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 2.26

Ant
 Nest
 Volume
 (m3)
 2.15

Deep
 Time
 Diffusion
 Length
 (m)
 1.93

Plant
 Fresh
 Weight
 Conversion
 1.91

Saltwater
 Solubility
 for
 Am
 (mol/L)
 1.78

Unit
 2
 Bulk
 Density
 (g/cm3)
 1.77

Kd
 Silt
 for
 Np
 (mL/g)
 1.69

Grass
 Root.Shoot
 Ratio
 1.66

Kd
 Sand
 for
 Pa
 (mL/g)
 1.66

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 1.53

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 1.44

Resuspension
 Flux
 (kg.m2-­‐yr)
 1.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 1.39

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.37

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 1.32

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 1.32

Unit
 4
 Compacted
 Residual
 Water
 Content
 1.26

Greasewood
 Root.Shoot
 Ratio
 1.18

Meat
 Post-­‐Cooking
 Loss
 1.16

Meat
 Preparation
 Loss
 1.12

Plant.Soil
 Conc
 Ratio
 for
 U
 1.11

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 1.07

RipRap
 Bulk
 Density
 (g/cm3)
 1.06

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 1.01

Forb
 Root.Shoot
 Ratio
 1.00

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.97

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.96

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.95

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.85

Plant.Soil
 Conc
 Ratio
 for
 I
 0.84

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.82

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.75

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 43
Kd
 Clay
 for
 Th
 (mL/g)
 0.71

Surface
 Atmosphere
 Thickness
 (m)
 0.70

Kd
 Sand
 for
 Np
 (mL/g)
 0.66

Saturated
 Zone
 Water
 Table
 Gradient
 0.63

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.57

Kd
 Clay
 for
 Am
 (mL/g)
 0.53

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.53

Deep
 Lake
 Depth
 (m)
 0.52

Kd
 Sand
 for
 Ac
 (mL/g)
 0.47

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.46

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.45

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.45

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.44

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.33

Forb
 Root
 Shape
 Parameter
 b
 0.31

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.31

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.31

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.30

RipRap
 Porosity
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.28

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.27

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.25

Unit
 4
 ET
 Layers
 Porosity
 0.24

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.21

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.20

Unit
 3
 Porosity
 0.20

Tortuosity
 Water
 Content
 Exponent

 0.19

Unit
 3
 Residual
 Water
 Content
 0.18

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.18

Kd
 Silt
 for
 Ra
 (mL/g)
 0.18

Fine
 Gravel
 Mix
 Porosity
 0.18

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.18

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.17

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.17

Surface
 Wind
 Speed
 (m/s)
 0.17

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.16

Fine
 CobbleMix
 Porosity
 0.15

Receptor
 Area
 (ha)
 0.14

OHV
 Dust
 Adjustment
 0.14

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.14

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.13

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.12

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 44
Unit
 4
 Compacted
 Porosity
 0.12

Unit
 4
 Compacted
 Hb
 (cm)
 0.11

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.11

Unit
 2
 Porosity
 0.11

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.10

Silt
 Sand
 Gravel
 Porosity
 0.10

Deep
 Time
 Aeolian
 Correlation
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.10

Kd
 Sand
 for
 Cs
 (mL/g)
 0.10

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.09

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.09

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.09

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.08

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.08

Kd
 Silt
 for
 Am
 (mL/g)
 0.07

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.07

Kd
 Silt
 for
 Cs
 (mL/g)
 0.07

Kd
 Silt
 for
 Sr
 (mL/g)
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.06

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.06

Kd
 Sand
 for
 Am
 (mL/g)
 0.06

Kd
 Clay
 for
 Pu
 (mL/g)
 0.05

Kd
 Silt
 for
 U
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.05

Resuspended
 Particle
 Fraction
 0.05

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.05

Ant
 Colony
 Lifespan
 (yr)
 0.05

Vegetation
 Association
 Selector
 0.05

Kd
 Silt
 for
 Pu
 (mL/g)
 0.05

Kd
 Clay
 for
 Ac
 (mL/g)
 0.05

Shrub
 Root.Shoot
 Ratio
 0.05

Body
 Weight
 Factor
 for
 Antelope
 0.05

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Kd
 Silt
 for
 Th
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.05

Saturated
 Zone
 Thickness
 (m)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.04

Kd
 Clay
 for
 Ra
 (mL/g)
 0.04

Kd
 Silt
 for
 Ac
 (mL/g)
 0.04

Kd
 Sand
 for
 Pu
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.04

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.04

Kd
 Clay
 for
 Sr
 (mL/g)
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 45
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.04

Kd
 Sand
 for
 Th
 (mL/g)
 0.04

Kd
 Sand
 for
 Ra
 (mL/g)
 0.04

Random
 Gully
 Selector
 0.04

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.03

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.03

Kd
 Sand
 for
 Pb
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.03

Kd
 Clay
 for
 Np
 (mL/g)
 0.03

Grass
 Root
 Shape
 Parameter
 b
 0.03

Kd
 Clay
 for
 Cs
 (mL/g)
 0.03

DCF
 Photon2
 REF
 0.03

Kd
 Clay
 for
 Pa
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.03

Kd
 Sand
 for
 Sr
 (mL/g)
 0.03

Deep
 Time
 Lake
 Start
 (yr)
 0.03

Tree
 Root.Shoot
 Ratio
 0.03

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.03

Deep
 Time
 DCF
 Photon
 2
 REF
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.02

Kd
 Silt
 for
 Pa
 (mL/g)
 0.02

Soil
 Temperature
 (°C)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.02

Kd
 Silt
 for
 Pb
 (mL/g)
 0.02

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.02

Tortuosity
 Porosity
 Exponent
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.02

Site
 Dispersal
 Area
 (km2)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.02

Greasewood
 Root
 Shape
 Parameter
 b
 0.02

Deep
 Time
 DCF
 Beta
 REF
 0.02

Ant
 Nest
 Shape
 Parameter
 b
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.02

Mammal
 Burrow
 Shape
 Parameter
 b
 0.02

Kd
 Sand
 for
 Tc
 (mL/g)
 0.02

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 46
Deep
 Time
 DCF
 Alpha
 REF
 0.02

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.02

Biomass
 %
 Cover
 Selector
 0.02

DCF
 Alpha
 REF
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.01

Intermediate
 Lake
 Depth
 (m)
 0.01

Deep
 Time
 DCF
 Photon
 1
 REF
 0.01

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.01

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Soil
 Ingestion
 Tracer
 Element
 0.01

Shrub
 Root
 Shape
 Parameter
 b
 0.01

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.01

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.01

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.01

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.01

Antelope
 Range
 Area
 (acre)
 0.00

Kd
 Sand
 for
 I
 (mL/g)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 47
Table 10: Peak Groundwater Well Concentrations within 500 years – U238
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 U
 (mL/g)
 6.34

DCF
 Beta
 REF
 6.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 4.19

Kd
 Clay
 for
 Pb
 (mL/g)
 3.95

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 3.17

Kd
 Clay
 for
 U
 (mL/g)
 3.08

Ant
 Nest
 Volume
 (m3)
 3.00

DCF
 Photon1
 REF
 2.75

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 2.44

Kd
 Silt
 for
 Np
 (mL/g)
 2.38

Deep
 Time
 Diffusion
 Length
 (m)
 2.21

Grass
 Root.Shoot
 Ratio
 2.17

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 2.09

Plant
 Fresh
 Weight
 Conversion
 2.08

Meat
 Preparation
 Loss
 2.02

Kd
 Sand
 for
 Pa
 (mL/g)
 2.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 1.96

Meat
 Post-­‐Cooking
 Loss
 1.94

Resuspension
 Flux
 (kg.m2-­‐yr)
 1.83

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 1.80

Unit
 2
 Bulk
 Density
 (g/cm3)
 1.68

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 1.60

Unit
 4
 Compacted
 Residual
 Water
 Content
 1.57

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.56

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 1.54

Forb
 Root.Shoot
 Ratio
 1.53

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 1.45

Plant.Soil
 Conc
 Ratio
 for
 Pu
 1.45

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 1.39

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 1.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 1.37

Plant.Soil
 Conc
 Ratio
 for
 Tc
 1.33

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 1.23

Plant.Soil
 Conc
 Ratio
 for
 U
 1.11

Plant.Soil
 Conc
 Ratio
 for
 I
 1.11

Kd
 Clay
 for
 Th
 (mL/g)
 1.03

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 1.03

Greasewood
 Root.Shoot
 Ratio
 1.02

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.02

Kd
 Sand
 for
 Np
 (mL/g)
 0.88

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 48
RipRap
 Bulk
 Density
 (g/cm3)
 0.86

Surface
 Atmosphere
 Thickness
 (m)
 0.84

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.75

Kd
 Sand
 for
 Ac
 (mL/g)
 0.70

Deep
 Lake
 Depth
 (m)
 0.68

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.67

Kd
 Clay
 for
 Am
 (mL/g)
 0.61

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.60

OHV
 Dust
 Adjustment
 0.58

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.57

Saturated
 Zone
 Water
 Table
 Gradient
 0.50

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.50

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.42

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.39

Forb
 Root
 Shape
 Parameter
 b
 0.37

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.36

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.28

Tortuosity
 Water
 Content
 Exponent

 0.26

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.23

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.22

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.22

Receptor
 Area
 (ha)
 0.19

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.17

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.16

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.15

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.13

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.13

Deep
 Time
 Aeolian
 Correlation
 0.12

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.11

RipRap
 Porosity
 0.10

Ant
 Colony
 Lifespan
 (yr)
 0.10

Kd
 Clay
 for
 Pa
 (mL/g)
 0.10

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.09

Unit
 4
 Compacted
 Porosity
 0.09

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.09

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.08

Kd
 Silt
 for
 Ra
 (mL/g)
 0.08

Fine
 Gravel
 Mix
 Porosity
 0.08

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.08

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 49
Unit
 3
 Residual
 Water
 Content
 0.08

Surface
 Wind
 Speed
 (m/s)
 0.08

Unit
 3
 Porosity
 0.07

Kd
 Sand
 for
 Pb
 (mL/g)
 0.07

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.06

Tree
 Root.Shoot
 Ratio
 0.06

Unit
 4
 ET
 Layers
 Porosity
 0.06

Fine
 CobbleMix
 Porosity
 0.06

Kd
 Silt
 for
 Am
 (mL/g)
 0.06

Kd
 Silt
 for
 U
 (mL/g)
 0.06

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.06

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.05

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Unit
 2
 Porosity
 0.05

Kd
 Sand
 for
 Th
 (mL/g)
 0.05

Unit
 4
 Compacted
 Hb
 (cm)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.05

Body
 Weight
 Factor
 for
 Antelope
 0.05

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.05

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.04

Silt
 Sand
 Gravel
 Porosity
 0.04

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.04

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.04

Vegetation
 Association
 Selector
 0.04

Shrub
 Root.Shoot
 Ratio
 0.03

Kd
 Clay
 for
 Pu
 (mL/g)
 0.03

Kd
 Silt
 for
 Sr
 (mL/g)
 0.03

Kd
 Sand
 for
 Cs
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.03

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.03

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.03

Kd
 Sand
 for
 Am
 (mL/g)
 0.03

Soil
 Temperature
 (°C)
 0.03

Kd
 Silt
 for
 Cs
 (mL/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.02

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.02

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.02

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.02

Kd
 Silt
 for
 Pu
 (mL/g)
 0.02

Kd
 Silt
 for
 Pb
 (mL/g)
 0.02

Saturated
 Zone
 Thickness
 (m)
 0.02

Kd
 Silt
 for
 Th
 (mL/g)
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 50
Mammal
 Burrow
 Shape
 Parameter
 b
 0.02

Kd
 Sand
 for
 Ra
 (mL/g)
 0.02

Grass
 Root
 Shape
 Parameter
 b
 0.02

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.02

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.02

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.02

Resuspended
 Particle
 Fraction
 0.02

Kd
 Clay
 for
 Sr
 (mL/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.02

Kd
 Clay
 for
 Cs
 (mL/g)
 0.02

Kd
 Sand
 for
 Pu
 (mL/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.02

Random
 Gully
 Selector
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.02

DCF
 Photon2
 REF
 0.02

Deep
 Time
 DCF
 Photon
 2
 REF
 0.02

Kd
 Clay
 for
 Ra
 (mL/g)
 0.02

Kd
 Silt
 for
 Ac
 (mL/g)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Kd
 Sand
 for
 Sr
 (mL/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.02

Deep
 Time
 DCF
 Beta
 REF
 0.02

Ant
 Nest
 Shape
 Parameter
 b
 0.01

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

Kd
 Clay
 for
 Ac
 (mL/g)
 0.01

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.01

Kd
 Silt
 for
 Pa
 (mL/g)
 0.01

Deep
 Time
 Receptor
 Area
 (ac)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.01

Deep
 Time
 Lake
 Start
 (yr)
 0.01

Greasewood
 Root
 Shape
 Parameter
 b
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.01

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.01

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.01

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.01

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Intermediate
 Lake
 Depth
 (m)
 0.01

Biomass
 %
 Cover
 Selector
 0.01

Tortuosity
 Porosity
 Exponent
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 51
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.01

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Kd
 Sand
 for
 Tc
 (mL/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.01

Site
 Dispersal
 Area
 (km2)
 0.01

DCF
 Alpha
 REF
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.01

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.01

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.01

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.01

Kd
 Clay
 for
 Np
 (mL/g)
 0.01

Antelope
 Range
 Area
 (acre)
 0.01

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.01

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.01

Shrub
 Root
 Shape
 Parameter
 b
 0.01

Deep
 Time
 DCF
 Photon
 1
 REF
 0.01

Deep
 Time
 DCF
 Alpha
 REF
 0.01

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.01

Kd
 Sand
 for
 I
 (mL/g)
 0.01

Soil
 Ingestion
 Tracer
 Element
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 52
Table 11: Dose summed over 10,000 years - Population
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 80.28

Kd
 Sand
 for
 Ra
 (mL/g)
 9.63

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 4.60

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.23

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.12

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.48

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.35

Unit
 4
 ET
 Layers
 Porosity
 0.33

Unit
 4
 Compacted
 Porosity
 0.33

Unit
 3
 Residual
 Water
 Content
 0.25

Unit
 3
 Porosity
 0.23

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.22

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.10

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.10

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.08

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.05

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.03

Resuspended
 Particle
 Fraction
 0.02

Unit
 2
 Porosity
 0.02

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.02

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.01

Kd
 Sand
 for
 Pb
 (mL/g)
 0.01

Deep
 Time
 Lake
 Start
 (yr)
 0.01

Kd
 Clay
 for
 Sr
 (mL/g)
 0.01

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.01

RipRap
 Bulk
 Density
 (g/cm3)
 0.01

Grass
 Root.Shoot
 Ratio
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

RipRap
 Porosity
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.01

Kd
 Silt
 for
 Cs
 (mL/g)
 0.01

Biomass
 %
 Cover
 Selector
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.01

Contaminated
 Fraction
 of
 GDP
 DU
 0.01

Deep
 Time
 DCF
 Beta
 REF
 0.01

Tortuosity
 Water
 Content
 Exponent

 0.01

Kd
 Sand
 for
 Np
 (mL/g)
 0.01

DCF
 Photon2
 REF
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 53
Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.01

Kd
 Silt
 for
 Ra
 (mL/g)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.01

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.01

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.01

Shrub
 Root.Shoot
 Ratio
 0.01

Deep
 Time
 DCF
 Photon
 2
 REF
 0.00

Kd
 Clay
 for
 Pu
 (mL/g)
 0.00

Kd
 Clay
 for
 U
 (mL/g)
 0.00

Kd
 Sand
 for
 Tc
 (mL/g)
 0.00

Kd
 Silt
 for
 Ac
 (mL/g)
 0.00

Saturated
 Zone
 Water
 Table
 Gradient
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.00

Kd
 Silt
 for
 Pa
 (mL/g)
 0.00

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.00

Meat
 Post-­‐Cooking
 Loss
 0.00

Plant.Soil
 Conc
 Ratio
 for
 I
 0.00

Deep
 Lake
 Depth
 (m)
 0.00

Deep
 Time
 DCF
 Alpha
 REF
 0.00

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.00

Silt
 Sand
 Gravel
 Porosity
 0.00

Tortuosity
 Porosity
 Exponent
 0.00

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.00

Kd
 Silt
 for
 Sr
 (mL/g)
 0.00

Soil
 Temperature
 (°C)
 0.00

Surface
 Atmosphere
 Thickness
 (m)
 0.00

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.00

OHV
 Dust
 Adjustment
 0.00

Kd
 Sand
 for
 Th
 (mL/g)
 0.00

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.00

Deep
 Time
 Receptor
 Area
 (ac)
 0.00

Ant
 Nest
 Shape
 Parameter
 b
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.00

Meat
 Preparation
 Loss
 0.00

Kd
 Clay
 for
 Pb
 (mL/g)
 0.00

Tree
 Root.Shoot
 Ratio
 0.00

Kd
 Sand
 for
 Pa
 (mL/g)
 0.00

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.00

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 54
Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.00

Body
 Weight
 Factor
 for
 Antelope
 0.00

Kd
 Silt
 for
 Pb
 (mL/g)
 0.00

Mammal
 Burrow
 Shape
 Parameter
 b
 0.00

Kd
 Sand
 for
 Am
 (mL/g)
 0.00

Fine
 CobbleMix
 Porosity
 0.00

Plant.Soil
 Conc
 Ratio
 for
 U
 0.00

Kd
 Sand
 for
 Cs
 (mL/g)
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.00

Kd
 Silt
 for
 Np
 (mL/g)
 0.00

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.00

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.00

Kd
 Silt
 for
 Pu
 (mL/g)
 0.00

Unit
 4
 Compacted
 Hb
 (cm)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.00

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.00

Kd
 Silt
 for
 U
 (mL/g)
 0.00

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.00

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.00

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.00

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.00

Shrub
 Root
 Shape
 Parameter
 b
 0.00

Deep
 Time
 DCF
 Photon
 1
 REF
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.00

Receptor
 Area
 (ha)
 0.00

Deep
 Time
 Aeolian
 Correlation
 0.00

Antelope
 Range
 Area
 (acre)
 0.00

DCF
 Beta
 REF
 0.00

Fine
 Gravel
 Mix
 Porosity
 0.00

Plant
 Fresh
 Weight
 Conversion
 0.00

Random
 Gully
 Selector
 0.00

Ant
 Colony
 Lifespan
 (yr)
 0.00

Kd
 Clay
 for
 Ac
 (mL/g)
 0.00

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.00

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.00

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.00

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.00

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.00

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.00

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 55
DCF
 Alpha
 REF
 0.00

Saturated
 Zone
 Thickness
 (m)
 0.00

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.00

Ant
 Nest
 Volume
 (m3)
 0.00

Kd
 Silt
 for
 Th
 (mL/g)
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.00

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.00

Kd
 Clay
 for
 Cs
 (mL/g)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.00

Kd
 Clay
 for
 Ra
 (mL/g)
 0.00

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.00

Kd
 Sand
 for
 Ac
 (mL/g)
 0.00

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.00

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.00

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.00

Kd
 Sand
 for
 Sr
 (mL/g)
 0.00

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.00

Kd
 Sand
 for
 I
 (mL/g)
 0.00

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.00

Forb
 Root
 Shape
 Parameter
 b
 0.00

Kd
 Sand
 for
 U
 (mL/g)
 0.00

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.00

Surface
 Wind
 Speed
 (m/s)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.00

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.00

Kd
 Clay
 for
 Am
 (mL/g)
 0.00

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.00

Kd
 Silt
 for
 Am
 (mL/g)
 0.00

Kd
 Clay
 for
 Th
 (mL/g)
 0.00

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.00

Deep
 Time
 Diffusion
 Length
 (m)
 0.00

Intermediate
 Lake
 Depth
 (m)
 0.00

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.00

Kd
 Clay
 for
 Pa
 (mL/g)
 0.00

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.00

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.00

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.00

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 56
Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.00

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.00

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.00

Vegetation
 Association
 Selector
 0.00

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.00

Forb
 Root.Shoot
 Ratio
 0.00

Greasewood
 Root.Shoot
 Ratio
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.00

Kd
 Sand
 for
 Pu
 (mL/g)
 0.00

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.00

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.00

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.00

Kd
 Clay
 for
 Np
 (mL/g)
 0.00

Site
 Dispersal
 Area
 (km2)
 0.00

Grass
 Root
 Shape
 Parameter
 b
 0.00

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.00

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.00

DCF
 Photon1
 REF
 0.00

Tree
 Root
 Shape
 Parameter
 b
 0.00

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.00

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.00

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.00

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.00

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.00

Greasewood
 Root
 Shape
 Parameter
 b
 0.00

Soil
 Ingestion
 Tracer
 Element
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 57
Table 12: Peak Dose within 10,000 years - Hunter
R-squared = 93%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 60.78

Kd
 Sand
 for
 Ra
 (mL/g)
 9.93

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 5.89

Resuspension
 Flux
 (kg.m2-­‐yr)
 4.50

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.31

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.11

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.01

Unit
 2
 Porosity
 0.63

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.51

Unit
 4
 ET
 Layers
 Porosity
 0.36

Unit
 3
 Porosity
 0.36

Unit
 3
 Residual
 Water
 Content
 0.35

Unit
 4
 Compacted
 Porosity
 0.34

Kd
 Sand
 for
 Np
 (mL/g)
 0.33

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.29

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.22

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.18

Silt
 Sand
 Gravel
 Porosity
 0.18

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.16

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.16

Kd
 Silt
 for
 Np
 (mL/g)
 0.13

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.13

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.12

Kd
 Sand
 for
 Pu
 (mL/g)
 0.12

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.12

DCF
 Photon2
 REF
 0.12

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.12

Deep
 Time
 Lake
 Start
 (yr)
 0.11

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.11

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.10

Kd
 Sand
 for
 Sr
 (mL/g)
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.10

Deep
 Time
 DCF
 Photon
 1
 REF
 0.10

Contaminated
 Fraction
 of
 GDP
 DU
 0.10

Kd
 Sand
 for
 Pa
 (mL/g)
 0.10

Kd
 Sand
 for
 Pb
 (mL/g)
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.10

Kd
 Sand
 for
 Ac
 (mL/g)
 0.10

Grass
 Root.Shoot
 Ratio
 0.10

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.09

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 58
Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.09

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.09

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.09

Forb
 Root.Shoot
 Ratio
 0.09

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.09

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.09

Deep
 Time
 Receptor
 Area
 (ac)
 0.08

RipRap
 Porosity
 0.08

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.08

Kd
 Sand
 for
 Cs
 (mL/g)
 0.08

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.08

Greasewood
 Root
 Shape
 Parameter
 b
 0.08

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.08

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.08

Surface
 Atmosphere
 Thickness
 (m)
 0.08

Kd
 Sand
 for
 Tc
 (mL/g)
 0.08

Kd
 Silt
 for
 Pu
 (mL/g)
 0.08

Kd
 Silt
 for
 Pa
 (mL/g)
 0.08

Deep
 Lake
 Depth
 (m)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.08

Fine
 CobbleMix
 Porosity
 0.08

Ant
 Colony
 Lifespan
 (yr)
 0.08

Kd
 Silt
 for
 Ra
 (mL/g)
 0.08

Random
 Gully
 Selector
 0.08

Kd
 Clay
 for
 Pb
 (mL/g)
 0.07

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.07

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.07

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.07

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.07

Kd
 Silt
 for
 Am
 (mL/g)
 0.07

Kd
 Clay
 for
 Ac
 (mL/g)
 0.07

Mammal
 Burrow
 Shape
 Parameter
 b
 0.07

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.07

RipRap
 Bulk
 Density
 (g/cm3)
 0.07

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.07

Kd
 Sand
 for
 U
 (mL/g)
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.07

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.07

Receptor
 Area
 (ha)
 0.07

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.07

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 59
Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.07

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.07

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Kd
 Sand
 for
 Th
 (mL/g)
 0.07

Tortuosity
 Porosity
 Exponent
 0.07

Kd
 Clay
 for
 Th
 (mL/g)
 0.07

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.07

Kd
 Silt
 for
 Pb
 (mL/g)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.07

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.07

Antelope
 Range
 Area
 (acre)
 0.07

Meat
 Post-­‐Cooking
 Loss
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.06

Plant.Soil
 Conc
 Ratio
 for
 U
 0.06

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.06

Kd
 Silt
 for
 Sr
 (mL/g)
 0.06

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.06

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.06

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.06

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.06

Deep
 Time
 Diffusion
 Length
 (m)
 0.06

Saturated
 Zone
 Water
 Table
 Gradient
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.06

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.06

Kd
 Clay
 for
 Pa
 (mL/g)
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.06

Ant
 Nest
 Volume
 (m3)
 0.06

Greasewood
 Root.Shoot
 Ratio
 0.06

Plant.Soil
 Conc
 Ratio
 for
 I
 0.06

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.06

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.06

DCF
 Beta
 REF
 0.06

Shrub
 Root
 Shape
 Parameter
 b
 0.06

DCF
 Photon1
 REF
 0.06

Kd
 Silt
 for
 Th
 (mL/g)
 0.06

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.06

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.06

Resuspended
 Particle
 Fraction
 0.06

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.06

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.06

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 60
Kd
 Clay
 for
 Np
 (mL/g)
 0.06

Deep
 Time
 Aeolian
 Correlation
 0.05

Site
 Dispersal
 Area
 (km2)
 0.05

Deep
 Time
 DCF
 Photon
 2
 REF
 0.05

Fine
 Gravel
 Mix
 Porosity
 0.05

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.05

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.05

Kd
 Clay
 for
 Ra
 (mL/g)
 0.05

Deep
 Time
 DCF
 Beta
 REF
 0.05

Tree
 Root.Shoot
 Ratio
 0.05

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.05

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.05

Kd
 Clay
 for
 Am
 (mL/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.05

Kd
 Sand
 for
 Am
 (mL/g)
 0.05

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Kd
 Silt
 for
 Cs
 (mL/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.05

Surface
 Wind
 Speed
 (m/s)
 0.05

Soil
 Temperature
 (°C)
 0.05

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.05

Ant
 Nest
 Shape
 Parameter
 b
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.05

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.05

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.05

OHV
 Dust
 Adjustment
 0.05

Body
 Weight
 Factor
 for
 Antelope
 0.05

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.05

Kd
 Clay
 for
 U
 (mL/g)
 0.05

Unit
 4
 Compacted
 Hb
 (cm)
 0.05

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.05

Deep
 Time
 DCF
 Alpha
 REF
 0.05

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

DCF
 Alpha
 REF
 0.05

Meat
 Preparation
 Loss
 0.05

Saturated
 Zone
 Thickness
 (m)
 0.05

Kd
 Clay
 for
 Pu
 (mL/g)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 61
Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.04

Shrub
 Root.Shoot
 Ratio
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.04

Forb
 Root
 Shape
 Parameter
 b
 0.04

Grass
 Root
 Shape
 Parameter
 b
 0.04

Tortuosity
 Water
 Content
 Exponent

 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.04

Kd
 Sand
 for
 I
 (mL/g)
 0.04

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.04

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.04

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.04

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Kd
 Silt
 for
 U
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.04

Kd
 Clay
 for
 Sr
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.04

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.04

Biomass
 %
 Cover
 Selector
 0.04

Plant
 Fresh
 Weight
 Conversion
 0.04

Tree
 Root
 Shape
 Parameter
 b
 0.04

Kd
 Silt
 for
 Ac
 (mL/g)
 0.03

Intermediate
 Lake
 Depth
 (m)
 0.03

Vegetation
 Association
 Selector
 0.03

Kd
 Clay
 for
 Cs
 (mL/g)
 0.03

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Soil
 Ingestion
 Tracer
 Element
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 62
Table 13: Peak Dose within 10,000 years – I-80
R-squared = 81%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 37.73

Kd
 Sand
 for
 Ra
 (mL/g)
 6.84

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 3.53

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.49

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.02

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 1.02

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.01

Unit
 3
 Porosity
 0.80

RipRap
 Bulk
 Density
 (g/cm3)
 0.65

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.62

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.54

Kd
 Silt
 for
 Pu
 (mL/g)
 0.54

Unit
 3
 Residual
 Water
 Content
 0.54

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.52

Unit
 4
 ET
 Layers
 Porosity
 0.49

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.49

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.47

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.46

Unit
 4
 Compacted
 Porosity
 0.45

Deep
 Time
 Receptor
 Area
 (ac)
 0.44

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.44

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.43

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.43

Saturated
 Zone
 Thickness
 (m)
 0.41

Shrub
 Root.Shoot
 Ratio
 0.41

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.41

RipRap
 Porosity
 0.40

Ant
 Nest
 Shape
 Parameter
 b
 0.39

Kd
 Sand
 for
 Th
 (mL/g)
 0.39

Fine
 CobbleMix
 Porosity
 0.39

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.39

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.38

Surface
 Atmosphere
 Thickness
 (m)
 0.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.36

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.36

Kd
 Silt
 for
 Pa
 (mL/g)
 0.36

Kd
 Clay
 for
 Pb
 (mL/g)
 0.35

Plant
 Fresh
 Weight
 Conversion
 0.34

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.34

Kd
 Silt
 for
 Cs
 (mL/g)
 0.34

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 63
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.33

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.33

Kd
 Sand
 for
 U
 (mL/g)
 0.33

Grass
 Root.Shoot
 Ratio
 0.32

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.32

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.32

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.31

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.31

Unit
 4
 Compacted
 Hb
 (cm)
 0.29

Ant
 Nest
 Volume
 (m3)
 0.29

Kd
 Sand
 for
 Cs
 (mL/g)
 0.28

Kd
 Clay
 for
 Cs
 (mL/g)
 0.28

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.28

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.28

Kd
 Clay
 for
 Ra
 (mL/g)
 0.28

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.28

Kd
 Sand
 for
 Pa
 (mL/g)
 0.27

Surface
 Wind
 Speed
 (m/s)
 0.27

Kd
 Clay
 for
 Pu
 (mL/g)
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.27

Kd
 Sand
 for
 Am
 (mL/g)
 0.27

OHV
 Dust
 Adjustment
 0.27

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.26

DCF
 Photon1
 REF
 0.26

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.26

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.26

Deep
 Time
 DCF
 Photon
 2
 REF
 0.26

Kd
 Sand
 for
 Ac
 (mL/g)
 0.26

Kd
 Clay
 for
 Np
 (mL/g)
 0.26

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.26

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.25

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.25

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.25

Receptor
 Area
 (ha)
 0.25

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.25

Random
 Gully
 Selector
 0.25

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.25

Kd
 Clay
 for
 Pa
 (mL/g)
 0.25

Deep
 Time
 DCF
 Alpha
 REF
 0.24

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 64
Meat
 Post-­‐Cooking
 Loss
 0.24

DCF
 Photon2
 REF
 0.24

Kd
 Silt
 for
 Np
 (mL/g)
 0.24

Forb
 Root
 Shape
 Parameter
 b
 0.24

Kd
 Sand
 for
 Sr
 (mL/g)
 0.24

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.24

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.24

Mammal
 Burrow
 Shape
 Parameter
 b
 0.24

Meat
 Preparation
 Loss
 0.23

Kd
 Silt
 for
 Sr
 (mL/g)
 0.23

Greasewood
 Root.Shoot
 Ratio
 0.23

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.23

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.23

Kd
 Clay
 for
 Am
 (mL/g)
 0.23

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.23

Kd
 Silt
 for
 Ac
 (mL/g)
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.22

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.22

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.22

Kd
 Sand
 for
 Np
 (mL/g)
 0.22

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.22

Deep
 Time
 DCF
 Beta
 REF
 0.22

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.21

Intermediate
 Lake
 Depth
 (m)
 0.21

Kd
 Silt
 for
 Th
 (mL/g)
 0.21

Kd
 Silt
 for
 Pb
 (mL/g)
 0.21

Unit
 2
 Porosity
 0.21

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.21

Biomass
 %
 Cover
 Selector
 0.21

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.21

Deep
 Lake
 Depth
 (m)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.21

Deep
 Time
 Lake
 Start
 (yr)
 0.21

Kd
 Clay
 for
 Ac
 (mL/g)
 0.21

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.21

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.20

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.20

Kd
 Sand
 for
 Tc
 (mL/g)
 0.20

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.20

Site
 Dispersal
 Area
 (km2)
 0.20

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 65
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.20

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.20

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.20

Kd
 Silt
 for
 Ra
 (mL/g)
 0.19

Tree
 Root.Shoot
 Ratio
 0.19

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.19

Kd
 Clay
 for
 Sr
 (mL/g)
 0.19

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.19

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.19

Body
 Weight
 Factor
 for
 Antelope
 0.19

Ant
 Colony
 Lifespan
 (yr)
 0.19

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.19

Grass
 Root
 Shape
 Parameter
 b
 0.19

Kd
 Clay
 for
 U
 (mL/g)
 0.19

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.19

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.19

Contaminated
 Fraction
 of
 GDP
 DU
 0.19

DCF
 Alpha
 REF
 0.19

Plant.Soil
 Conc
 Ratio
 for
 I
 0.19

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.18

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.18

Shrub
 Root
 Shape
 Parameter
 b
 0.18

Deep
 Time
 DCF
 Photon
 1
 REF
 0.18

Kd
 Silt
 for
 Am
 (mL/g)
 0.18

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.18

Soil
 Temperature
 (°C)
 0.18

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.18

Kd
 Silt
 for
 U
 (mL/g)
 0.17

Resuspended
 Particle
 Fraction
 0.17

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.17

Tortuosity
 Porosity
 Exponent
 0.17

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.17

Kd
 Sand
 for
 Pb
 (mL/g)
 0.17

DCF
 Beta
 REF
 0.17

Tortuosity
 Water
 Content
 Exponent

 0.17

Fine
 Gravel
 Mix
 Porosity
 0.17

Silt
 Sand
 Gravel
 Porosity
 0.16

Antelope
 Range
 Area
 (acre)
 0.16

Deep
 Time
 Diffusion
 Length
 (m)
 0.16

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.16

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.16

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.16

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 66
Saturated
 Zone
 Water
 Table
 Gradient
 0.16

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.16

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.16

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.16

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.15

Kd
 Clay
 for
 Th
 (mL/g)
 0.15

Kd
 Sand
 for
 Pu
 (mL/g)
 0.15

Deep
 Time
 Aeolian
 Correlation
 0.15

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.15

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.15

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.14

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.14

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.14

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.13

Greasewood
 Root
 Shape
 Parameter
 b
 0.13

Forb
 Root.Shoot
 Ratio
 0.12

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.12

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.12

Kd
 Sand
 for
 I
 (mL/g)
 0.12

Vegetation
 Association
 Selector
 0.12

Plant.Soil
 Conc
 Ratio
 for
 U
 0.11

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.11

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.10

Tree
 Root
 Shape
 Parameter
 b
 0.06

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 67
Table 14: Peak Dose within 10,000 years – Knolls
R-squared = 74%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 10.20

Plant.Soil
 Conc
 Ratio
 for
 Th
 2.65

Biomass
 %
 Cover
 Selector
 2.64

Unit
 3
 Porosity
 1.52

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 1.50

Kd
 Sand
 for
 Ra
 (mL/g)
 1.35

Kd
 Sand
 for
 Th
 (mL/g)
 1.23

Kd
 Silt
 for
 Np
 (mL/g)
 1.23

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 1.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 1.09

Deep
 Time
 DCF
 Alpha
 REF
 1.07

Kd
 Clay
 for
 Cs
 (mL/g)
 0.96

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.96

Unit
 4
 ET
 Layers
 Porosity
 0.94

Kd
 Silt
 for
 Ac
 (mL/g)
 0.93

Kd
 Sand
 for
 Tc
 (mL/g)
 0.93

Kd
 Sand
 for
 U
 (mL/g)
 0.91

Kd
 Clay
 for
 Pu
 (mL/g)
 0.91

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.89

Kd
 Sand
 for
 Am
 (mL/g)
 0.87

RipRap
 Bulk
 Density
 (g/cm3)
 0.86

Meat
 Preparation
 Loss
 0.85

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.80

Kd
 Silt
 for
 Pa
 (mL/g)
 0.80

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.78

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.74

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.72

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.70

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.70

Forb
 Root.Shoot
 Ratio
 0.70

RipRap
 Porosity
 0.67

Unit
 3
 Residual
 Water
 Content
 0.67

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.64

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.64

Plant.Soil
 Conc
 Ratio
 for
 I
 0.64

Antelope
 Range
 Area
 (acre)
 0.64

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.63

DCF
 Beta
 REF
 0.62

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.61

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.59

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 68
Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.59

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.59

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.59

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.57

Kd
 Silt
 for
 Th
 (mL/g)
 0.56

Kd
 Silt
 for
 Pu
 (mL/g)
 0.56

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.55

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.55

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.54

Tree
 Root.Shoot
 Ratio
 0.53

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.53

Deep
 Time
 DCF
 Photon
 1
 REF
 0.53

Kd
 Clay
 for
 Pa
 (mL/g)
 0.51

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.51

Kd
 Clay
 for
 Ac
 (mL/g)
 0.50

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.50

Deep
 Lake
 Depth
 (m)
 0.49

Kd
 Silt
 for
 Cs
 (mL/g)
 0.49

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.49

OHV
 Dust
 Adjustment
 0.48

Kd
 Clay
 for
 Am
 (mL/g)
 0.48

Contaminated
 Fraction
 of
 GDP
 DU
 0.48

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.48

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.47

Fine
 Gravel
 Mix
 Porosity
 0.47

Mammal
 Burrow
 Shape
 Parameter
 b
 0.47

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.47

Surface
 Atmosphere
 Thickness
 (m)
 0.46

Kd
 Clay
 for
 Pb
 (mL/g)
 0.46

Kd
 Sand
 for
 Pu
 (mL/g)
 0.46

Kd
 Clay
 for
 Th
 (mL/g)
 0.45

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.45

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.45

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.45

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.45

Silt
 Sand
 Gravel
 Porosity
 0.44

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.44

Unit
 4
 Compacted
 Porosity
 0.43

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.42

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.42

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.42

Kd
 Sand
 for
 Sr
 (mL/g)
 0.41

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.41

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 69
Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.41

Kd
 Silt
 for
 U
 (mL/g)
 0.40

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.40

Kd
 Sand
 for
 Cs
 (mL/g)
 0.40

Saturated
 Zone
 Thickness
 (m)
 0.40

Shrub
 Root
 Shape
 Parameter
 b
 0.39

Greasewood
 Root
 Shape
 Parameter
 b
 0.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.38

DCF
 Photon2
 REF
 0.38

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.38

Deep
 Time
 Receptor
 Area
 (ac)
 0.38

Kd
 Sand
 for
 Pa
 (mL/g)
 0.37

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.37

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.37

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.36

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.36

Fine
 CobbleMix
 Porosity
 0.36

Unit
 4
 Compacted
 Hb
 (cm)
 0.36

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.36

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.35

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.35

Kd
 Silt
 for
 Pb
 (mL/g)
 0.35

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.34

Surface
 Wind
 Speed
 (m/s)
 0.34

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.34

Forb
 Root
 Shape
 Parameter
 b
 0.34

Kd
 Silt
 for
 Am
 (mL/g)
 0.33

Resuspended
 Particle
 Fraction
 0.33

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.33

Plant
 Fresh
 Weight
 Conversion
 0.33

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.32

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.32

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.32

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.32

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.31

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.31

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.31

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.31

Kd
 Silt
 for
 Sr
 (mL/g)
 0.30

Kd
 Clay
 for
 Sr
 (mL/g)
 0.30

Ant
 Nest
 Volume
 (m3)
 0.30

Plant.Soil
 Conc
 Ratio
 for
 U
 0.30

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 70
Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.30

Kd
 Clay
 for
 Ra
 (mL/g)
 0.30

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.29

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.29

Unit
 2
 Porosity
 0.29

Kd
 Sand
 for
 Ac
 (mL/g)
 0.29

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.29

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.29

Site
 Dispersal
 Area
 (km2)
 0.29

Shrub
 Root.Shoot
 Ratio
 0.29

Receptor
 Area
 (ha)
 0.29

DCF
 Alpha
 REF
 0.28

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.28

Kd
 Sand
 for
 Pb
 (mL/g)
 0.28

Deep
 Time
 Diffusion
 Length
 (m)
 0.27

Soil
 Temperature
 (°C)
 0.27

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.26

Kd
 Sand
 for
 Np
 (mL/g)
 0.26

Ant
 Nest
 Shape
 Parameter
 b
 0.26

Grass
 Root.Shoot
 Ratio
 0.26

Tortuosity
 Water
 Content
 Exponent

 0.25

Kd
 Silt
 for
 Ra
 (mL/g)
 0.25

Grass
 Root
 Shape
 Parameter
 b
 0.25

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.25

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.24

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.24

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.24

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.24

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.23

Random
 Gully
 Selector
 0.23

Deep
 Time
 DCF
 Beta
 REF
 0.23

Deep
 Time
 DCF
 Photon
 2
 REF
 0.23

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.23

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.22

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.22

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.22

Deep
 Time
 Lake
 Start
 (yr)
 0.22

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.22

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.22

Tree
 Root
 Shape
 Parameter
 b
 0.21

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 71
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.21

DCF
 Photon1
 REF
 0.21

Kd
 Clay
 for
 U
 (mL/g)
 0.21

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.21

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.21

Kd
 Clay
 for
 Np
 (mL/g)
 0.20

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.20

Deep
 Time
 Aeolian
 Correlation
 0.20

Kd
 Sand
 for
 I
 (mL/g)
 0.20

Vegetation
 Association
 Selector
 0.20

Ant
 Colony
 Lifespan
 (yr)
 0.20

Intermediate
 Lake
 Depth
 (m)
 0.19

Body
 Weight
 Factor
 for
 Antelope
 0.19

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.19

Saturated
 Zone
 Water
 Table
 Gradient
 0.18

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.18

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.18

Meat
 Post-­‐Cooking
 Loss
 0.18

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.17

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.17

Tortuosity
 Porosity
 Exponent
 0.16

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.14

Greasewood
 Root.Shoot
 Ratio
 0.13

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.13

Soil
 Ingestion
 Tracer
 Element
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 72
Table 15: Peak Dose within 10,000 years – Railroad
R-squared = 81%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 37.73

Kd
 Sand
 for
 Ra
 (mL/g)
 6.82

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 3.53

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.47

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 1.02

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.02

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.00

Unit
 3
 Porosity
 0.76

RipRap
 Bulk
 Density
 (g/cm3)
 0.64

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.63

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.57

Unit
 3
 Residual
 Water
 Content
 0.54

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.54

Kd
 Silt
 for
 Pu
 (mL/g)
 0.50

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.49

Unit
 4
 ET
 Layers
 Porosity
 0.48

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.46

Deep
 Time
 Receptor
 Area
 (ac)
 0.46

Unit
 4
 Compacted
 Porosity
 0.45

Shrub
 Root.Shoot
 Ratio
 0.45

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.45

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.44

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.42

Ant
 Nest
 Shape
 Parameter
 b
 0.40

Kd
 Sand
 for
 Th
 (mL/g)
 0.40

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.40

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.39

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.39

Fine
 CobbleMix
 Porosity
 0.39

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.39

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.39

Surface
 Atmosphere
 Thickness
 (m)
 0.38

Saturated
 Zone
 Thickness
 (m)
 0.38

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.38

RipRap
 Porosity
 0.36

Kd
 Silt
 for
 Pa
 (mL/g)
 0.35

Kd
 Clay
 for
 Pb
 (mL/g)
 0.35

Kd
 Silt
 for
 Cs
 (mL/g)
 0.34

Plant
 Fresh
 Weight
 Conversion
 0.34

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.33

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 73
Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.33

Ant
 Nest
 Volume
 (m3)
 0.33

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.33

Grass
 Root.Shoot
 Ratio
 0.33

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.32

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.32

Kd
 Sand
 for
 U
 (mL/g)
 0.31

Kd
 Sand
 for
 Cs
 (mL/g)
 0.31

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.30

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.30

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.30

Unit
 4
 Compacted
 Hb
 (cm)
 0.29

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.29

Kd
 Sand
 for
 Am
 (mL/g)
 0.28

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.28

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.28

Kd
 Clay
 for
 Pu
 (mL/g)
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.27

Kd
 Clay
 for
 Ra
 (mL/g)
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.27

Kd
 Sand
 for
 Pa
 (mL/g)
 0.27

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.27

Kd
 Sand
 for
 Ac
 (mL/g)
 0.27

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.26

Surface
 Wind
 Speed
 (m/s)
 0.26

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.26

DCF
 Photon1
 REF
 0.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.26

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.26

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.26

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.26

OHV
 Dust
 Adjustment
 0.26

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.26

Kd
 Silt
 for
 Np
 (mL/g)
 0.26

Receptor
 Area
 (ha)
 0.25

Deep
 Time
 DCF
 Alpha
 REF
 0.25

Random
 Gully
 Selector
 0.25

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.25

Kd
 Clay
 for
 Cs
 (mL/g)
 0.25

Kd
 Sand
 for
 Sr
 (mL/g)
 0.25

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.25

Deep
 Time
 DCF
 Photon
 2
 REF
 0.25

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 74
Resuspension
 Flux
 (kg.m2-­‐yr)
 0.24

Kd
 Clay
 for
 Np
 (mL/g)
 0.24

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.24

Meat
 Preparation
 Loss
 0.24

Meat
 Post-­‐Cooking
 Loss
 0.24

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.24

DCF
 Photon2
 REF
 0.24

Intermediate
 Lake
 Depth
 (m)
 0.23

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.23

Kd
 Silt
 for
 Sr
 (mL/g)
 0.23

Kd
 Sand
 for
 Np
 (mL/g)
 0.23

Deep
 Time
 DCF
 Beta
 REF
 0.23

Deep
 Lake
 Depth
 (m)
 0.23

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.23

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.23

Forb
 Root
 Shape
 Parameter
 b
 0.23

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.23

Kd
 Silt
 for
 Ac
 (mL/g)
 0.23

Mammal
 Burrow
 Shape
 Parameter
 b
 0.22

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.22

Unit
 2
 Porosity
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.22

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.22

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.22

Kd
 Silt
 for
 Pb
 (mL/g)
 0.22

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.21

Kd
 Clay
 for
 Ac
 (mL/g)
 0.21

Kd
 Clay
 for
 Pa
 (mL/g)
 0.21

Deep
 Time
 Lake
 Start
 (yr)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.21

Kd
 Clay
 for
 U
 (mL/g)
 0.21

Greasewood
 Root.Shoot
 Ratio
 0.21

Kd
 Silt
 for
 Th
 (mL/g)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.21

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.21

Kd
 Silt
 for
 Ra
 (mL/g)
 0.21

Body
 Weight
 Factor
 for
 Antelope
 0.20

Kd
 Clay
 for
 Am
 (mL/g)
 0.20

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.20

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.20

Kd
 Clay
 for
 Sr
 (mL/g)
 0.20

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 75
Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.20

Tree
 Root.Shoot
 Ratio
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.20

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.20

Biomass
 %
 Cover
 Selector
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.20

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.20

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.20

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.19

Contaminated
 Fraction
 of
 GDP
 DU
 0.19

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.19

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.19

Kd
 Sand
 for
 Tc
 (mL/g)
 0.19

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.19

Site
 Dispersal
 Area
 (km2)
 0.19

DCF
 Alpha
 REF
 0.19

Deep
 Time
 DCF
 Photon
 1
 REF
 0.19

Ant
 Colony
 Lifespan
 (yr)
 0.19

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.18

Plant.Soil
 Conc
 Ratio
 for
 I
 0.18

Silt
 Sand
 Gravel
 Porosity
 0.18

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.18

Shrub
 Root
 Shape
 Parameter
 b
 0.18

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.18

Kd
 Silt
 for
 Am
 (mL/g)
 0.18

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.18

Tortuosity
 Water
 Content
 Exponent

 0.17

Tortuosity
 Porosity
 Exponent
 0.17

Kd
 Sand
 for
 Pb
 (mL/g)
 0.17

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.17

Saturated
 Zone
 Water
 Table
 Gradient
 0.17

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.17

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.17

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.17

DCF
 Beta
 REF
 0.17

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.17

Antelope
 Range
 Area
 (acre)
 0.16

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.16

Resuspended
 Particle
 Fraction
 0.16

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.16

Kd
 Silt
 for
 U
 (mL/g)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 76
Kd
 Sand
 for
 Pu
 (mL/g)
 0.16

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.16

Soil
 Temperature
 (°C)
 0.16

Grass
 Root
 Shape
 Parameter
 b
 0.15

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.15

Deep
 Time
 Diffusion
 Length
 (m)
 0.15

Deep
 Time
 Aeolian
 Correlation
 0.15

Fine
 Gravel
 Mix
 Porosity
 0.15

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.15

Greasewood
 Root
 Shape
 Parameter
 b
 0.15

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.14

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.14

Kd
 Clay
 for
 Th
 (mL/g)
 0.14

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.14

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.14

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.13

Forb
 Root.Shoot
 Ratio
 0.13

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.13

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.12

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.12

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.12

Kd
 Sand
 for
 I
 (mL/g)
 0.12

Plant.Soil
 Conc
 Ratio
 for
 U
 0.11

Vegetation
 Association
 Selector
 0.11

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.09

Tree
 Root
 Shape
 Parameter
 b
 0.08

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 77
Table 16: Peak Dose within 10,000 years – Rest Area
R-squared = 87%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 49.19

Kd
 Sand
 for
 Ra
 (mL/g)
 7.80

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 3.89

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.63

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 1.19

Surface
 Atmosphere
 Thickness
 (m)
 1.08

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.02

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.98

Unit
 3
 Porosity
 0.61

Unit
 4
 Compacted
 Porosity
 0.44

Mammal
 Burrow
 Shape
 Parameter
 b
 0.43

Unit
 3
 Residual
 Water
 Content
 0.41

Greasewood
 Root
 Shape
 Parameter
 b
 0.38

Unit
 4
 ET
 Layers
 Porosity
 0.38

RipRap
 Bulk
 Density
 (g/cm3)
 0.37

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.35

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.33

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.33

Shrub
 Root.Shoot
 Ratio
 0.33

Saturated
 Zone
 Thickness
 (m)
 0.32

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.30

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.30

Kd
 Clay
 for
 Am
 (mL/g)
 0.30

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.28

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.28

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.27

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.26

Kd
 Clay
 for
 Np
 (mL/g)
 0.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.26

Kd
 Clay
 for
 Ra
 (mL/g)
 0.25

Surface
 Wind
 Speed
 (m/s)
 0.25

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.25

Greasewood
 Root.Shoot
 Ratio
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.24

Fine
 CobbleMix
 Porosity
 0.24

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.24

Kd
 Silt
 for
 Np
 (mL/g)
 0.24

Ant
 Nest
 Shape
 Parameter
 b
 0.24

Kd
 Silt
 for
 Ra
 (mL/g)
 0.24

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 78
Ant
 Nest
 Volume
 (m3)
 0.23

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.23

Kd
 Sand
 for
 U
 (mL/g)
 0.23

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.23

Kd
 Sand
 for
 Th
 (mL/g)
 0.23

RipRap
 Porosity
 0.23

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.23

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.23

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.22

Kd
 Clay
 for
 Pb
 (mL/g)
 0.22

Unit
 4
 Compacted
 Hb
 (cm)
 0.22

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.22

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.22

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.22

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.21

Kd
 Silt
 for
 Pu
 (mL/g)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.21

Kd
 Sand
 for
 Am
 (mL/g)
 0.21

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.21

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.20

Kd
 Clay
 for
 Pa
 (mL/g)
 0.20

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.20

Plant
 Fresh
 Weight
 Conversion
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.20

Kd
 Silt
 for
 Th
 (mL/g)
 0.20

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.20

Kd
 Sand
 for
 Cs
 (mL/g)
 0.19

Unit
 2
 Porosity
 0.19

Grass
 Root
 Shape
 Parameter
 b
 0.19

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.19

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.19

DCF
 Photon1
 REF
 0.19

Kd
 Silt
 for
 Pa
 (mL/g)
 0.19

Kd
 Silt
 for
 Ac
 (mL/g)
 0.18

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.18

Intermediate
 Lake
 Depth
 (m)
 0.18

Soil
 Temperature
 (°C)
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.18

Meat
 Post-­‐Cooking
 Loss
 0.18

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.18

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.18

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 79
Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.17

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.17

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.17

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.17

Receptor
 Area
 (ha)
 0.17

Kd
 Silt
 for
 Cs
 (mL/g)
 0.17

Kd
 Clay
 for
 Cs
 (mL/g)
 0.17

Kd
 Sand
 for
 Sr
 (mL/g)
 0.17

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.17

Resuspended
 Particle
 Fraction
 0.17

Deep
 Time
 DCF
 Alpha
 REF
 0.17

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.17

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.17

Kd
 Sand
 for
 Ac
 (mL/g)
 0.16

Deep
 Time
 Receptor
 Area
 (ac)
 0.16

Plant.Soil
 Conc
 Ratio
 for
 I
 0.16

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.16

Deep
 Time
 DCF
 Photon
 1
 REF
 0.16

Plant.Soil
 Conc
 Ratio
 for
 U
 0.16

Kd
 Clay
 for
 Pu
 (mL/g)
 0.15

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.15

Kd
 Sand
 for
 Np
 (mL/g)
 0.15

Deep
 Time
 DCF
 Photon
 2
 REF
 0.15

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.15

Kd
 Silt
 for
 U
 (mL/g)
 0.15

Tree
 Root.Shoot
 Ratio
 0.15

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.15

Meat
 Preparation
 Loss
 0.15

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.15

Site
 Dispersal
 Area
 (km2)
 0.15

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.14

DCF
 Alpha
 REF
 0.14

Grass
 Root.Shoot
 Ratio
 0.14

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.14

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.14

Deep
 Time
 DCF
 Beta
 REF
 0.14

Kd
 Silt
 for
 Pb
 (mL/g)
 0.14

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.14

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.14

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 80
Deep
 Time
 Lake
 Start
 (yr)
 0.14

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.14

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.14

DCF
 Beta
 REF
 0.13

Deep
 Time
 Diffusion
 Length
 (m)
 0.13

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.13

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.13

Kd
 Sand
 for
 Pa
 (mL/g)
 0.13

Ant
 Colony
 Lifespan
 (yr)
 0.13

Random
 Gully
 Selector
 0.13

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.13

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.13

OHV
 Dust
 Adjustment
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.13

Body
 Weight
 Factor
 for
 Antelope
 0.13

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.13

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.13

DCF
 Photon2
 REF
 0.12

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.12

Forb
 Root
 Shape
 Parameter
 b
 0.12

Kd
 Clay
 for
 U
 (mL/g)
 0.12

Shrub
 Root
 Shape
 Parameter
 b
 0.12

Kd
 Clay
 for
 Ac
 (mL/g)
 0.12

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.12

Kd
 Clay
 for
 Sr
 (mL/g)
 0.12

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.12

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.12

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.12

Fine
 Gravel
 Mix
 Porosity
 0.12

Kd
 Silt
 for
 Am
 (mL/g)
 0.12

Deep
 Time
 Aeolian
 Correlation
 0.11

Tortuosity
 Porosity
 Exponent
 0.11

Saturated
 Zone
 Water
 Table
 Gradient
 0.11

Kd
 Sand
 for
 Tc
 (mL/g)
 0.11

Biomass
 %
 Cover
 Selector
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.11

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.11

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.11

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.11

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.11

Kd
 Sand
 for
 Pb
 (mL/g)
 0.11

Kd
 Clay
 for
 Th
 (mL/g)
 0.11

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 81
Silt
 Sand
 Gravel
 Porosity
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.10

Contaminated
 Fraction
 of
 GDP
 DU
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.10

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.10

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.10

Kd
 Sand
 for
 I
 (mL/g)
 0.10

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.10

Forb
 Root.Shoot
 Ratio
 0.10

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.10

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.10

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.10

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.10

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.10

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.09

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.09

Deep
 Lake
 Depth
 (m)
 0.09

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.09

Antelope
 Range
 Area
 (acre)
 0.09

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.08

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.08

Tortuosity
 Water
 Content
 Exponent

 0.08

Kd
 Sand
 for
 Pu
 (mL/g)
 0.08

Kd
 Silt
 for
 Sr
 (mL/g)
 0.08

Tree
 Root
 Shape
 Parameter
 b
 0.05

Vegetation
 Association
 Selector
 0.05

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.04

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 82
Table 17: Peak Dose within 10,000 years – Sport OHV
R-squared = 97%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 68.90

Kd
 Sand
 for
 Ra
 (mL/g)
 11.81

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 6.86

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.44

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.19

Unit
 4
 Compacted
 Porosity
 0.35

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.35

Unit
 4
 ET
 Layers
 Porosity
 0.32

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.30

Unit
 3
 Residual
 Water
 Content
 0.29

Unit
 3
 Porosity
 0.29

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.19

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.18

Shrub
 Root.Shoot
 Ratio
 0.17

Resuspended
 Particle
 Fraction
 0.14

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.13

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.10

Kd
 Sand
 for
 Pb
 (mL/g)
 0.10

Biomass
 %
 Cover
 Selector
 0.09

Unit
 2
 Porosity
 0.09

RipRap
 Bulk
 Density
 (g/cm3)
 0.08

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.08

Kd
 Clay
 for
 Sr
 (mL/g)
 0.08

Saturated
 Zone
 Water
 Table
 Gradient
 0.06

Kd
 Silt
 for
 Th
 (mL/g)
 0.06

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.06

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.06

Kd
 Clay
 for
 Pu
 (mL/g)
 0.06

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.06

Deep
 Time
 DCF
 Beta
 REF
 0.05

Soil
 Temperature
 (°C)
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.05

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.05

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.05

Kd
 Clay
 for
 Am
 (mL/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.05

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.05

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 83
Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.04

Deep
 Time
 Lake
 Start
 (yr)
 0.04

Deep
 Time
 DCF
 Photon
 1
 REF
 0.04

Kd
 Sand
 for
 Pa
 (mL/g)
 0.04

Kd
 Silt
 for
 Pb
 (mL/g)
 0.04

Deep
 Time
 DCF
 Alpha
 REF
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.04

Kd
 Sand
 for
 Np
 (mL/g)
 0.04

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.04

Kd
 Clay
 for
 Ra
 (mL/g)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.04

Contaminated
 Fraction
 of
 GDP
 DU
 0.04

RipRap
 Porosity
 0.04

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Meat
 Post-­‐Cooking
 Loss
 0.04

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.04

Intermediate
 Lake
 Depth
 (m)
 0.04

Deep
 Time
 Aeolian
 Correlation
 0.04

DCF
 Beta
 REF
 0.04

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 I
 0.04

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.04

Kd
 Silt
 for
 Ra
 (mL/g)
 0.04

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.04

Meat
 Preparation
 Loss
 0.04

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.04

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.04

OHV
 Dust
 Adjustment
 0.04

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.03

Grass
 Root.Shoot
 Ratio
 0.03

Kd
 Silt
 for
 Pa
 (mL/g)
 0.03

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.03

Deep
 Time
 DCF
 Photon
 2
 REF
 0.03

Tree
 Root.Shoot
 Ratio
 0.03

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.03

Kd
 Silt
 for
 Ac
 (mL/g)
 0.03

Deep
 Time
 Receptor
 Area
 (ac)
 0.03

Saturated
 Zone
 Thickness
 (m)
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.03

Kd
 Sand
 for
 Tc
 (mL/g)
 0.03

DCF
 Photon1
 REF
 0.03

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 84
Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.03

Kd
 Clay
 for
 Pb
 (mL/g)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.03

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.03

Kd
 Sand
 for
 Am
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.03

DCF
 Alpha
 REF
 0.03

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.03

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.03

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.03

Kd
 Clay
 for
 Cs
 (mL/g)
 0.03

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.03

Surface
 Wind
 Speed
 (m/s)
 0.03

Kd
 Sand
 for
 Cs
 (mL/g)
 0.03

Kd
 Clay
 for
 U
 (mL/g)
 0.03

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.03

Vegetation
 Association
 Selector
 0.03

Body
 Weight
 Factor
 for
 Antelope
 0.03

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.03

Forb
 Root
 Shape
 Parameter
 b
 0.03

Ant
 Nest
 Shape
 Parameter
 b
 0.03

Shrub
 Root
 Shape
 Parameter
 b
 0.03

Surface
 Atmosphere
 Thickness
 (m)
 0.03

Kd
 Sand
 for
 U
 (mL/g)
 0.03

Kd
 Silt
 for
 Np
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.03

Kd
 Silt
 for
 Pu
 (mL/g)
 0.03

Kd
 Silt
 for
 Cs
 (mL/g)
 0.03

Receptor
 Area
 (ha)
 0.03

Mammal
 Burrow
 Shape
 Parameter
 b
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.03

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 U
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.03

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.03

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.03

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.03

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.03

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.03

Kd
 Clay
 for
 Pa
 (mL/g)
 0.03

Kd
 Silt
 for
 Sr
 (mL/g)
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 85
Kd
 Silt
 for
 Am
 (mL/g)
 0.03

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.02

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.02

Kd
 Sand
 for
 Pu
 (mL/g)
 0.02

Kd
 Clay
 for
 Np
 (mL/g)
 0.02

DCF
 Photon2
 REF
 0.02

Deep
 Lake
 Depth
 (m)
 0.02

Site
 Dispersal
 Area
 (km2)
 0.02

Kd
 Sand
 for
 Th
 (mL/g)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.02

Greasewood
 Root
 Shape
 Parameter
 b
 0.02

Tortuosity
 Water
 Content
 Exponent

 0.02

Ant
 Colony
 Lifespan
 (yr)
 0.02

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.02

Kd
 Sand
 for
 Sr
 (mL/g)
 0.02

Fine
 CobbleMix
 Porosity
 0.02

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.02

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.02

Random
 Gully
 Selector
 0.02

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.02

Greasewood
 Root.Shoot
 Ratio
 0.02

Antelope
 Range
 Area
 (acre)
 0.02

Silt
 Sand
 Gravel
 Porosity
 0.02

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.02

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.02

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.02

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.02

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.02

Ant
 Nest
 Volume
 (m3)
 0.02

Kd
 Clay
 for
 Th
 (mL/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.02

Kd
 Silt
 for
 U
 (mL/g)
 0.02

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.02

Plant
 Fresh
 Weight
 Conversion
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 86
Fine
 Gravel
 Mix
 Porosity
 0.02

Grass
 Root
 Shape
 Parameter
 b
 0.02

Kd
 Sand
 for
 Ac
 (mL/g)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.02

Kd
 Sand
 for
 I
 (mL/g)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.02

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.02

Deep
 Time
 Diffusion
 Length
 (m)
 0.02

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.02

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.02

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.02

Forb
 Root.Shoot
 Ratio
 0.02

Unit
 4
 Compacted
 Hb
 (cm)
 0.02

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.02

Tortuosity
 Porosity
 Exponent
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.02

Kd
 Clay
 for
 Ac
 (mL/g)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.02

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.02

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.01

Tree
 Root
 Shape
 Parameter
 b
 0.01

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.01

Soil
 Ingestion
 Tracer
 Element
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 87
Table 18: Peak Dose within 10,000 years – UTTR Dose
R-squared = 85%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 44.58

Kd
 Sand
 for
 Ra
 (mL/g)
 7.08

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 3.32

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 1.76

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 1.08

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.03

Surface
 Atmosphere
 Thickness
 (m)
 0.91

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.80

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.77

Unit
 3
 Porosity
 0.76

RipRap
 Bulk
 Density
 (g/cm3)
 0.62

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.54

Unit
 3
 Residual
 Water
 Content
 0.43

Unit
 4
 Compacted
 Porosity
 0.43

Surface
 Wind
 Speed
 (m/s)
 0.41

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.39

Shrub
 Root.Shoot
 Ratio
 0.39

Kd
 Sand
 for
 U
 (mL/g)
 0.39

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.37

Unit
 4
 ET
 Layers
 Porosity
 0.37

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.37

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.36

Fine
 CobbleMix
 Porosity
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.34

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.34

Kd
 Clay
 for
 Am
 (mL/g)
 0.34

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.33

Kd
 Silt
 for
 Np
 (mL/g)
 0.33

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.33

Kd
 Sand
 for
 Np
 (mL/g)
 0.33

Intermediate
 Lake
 Depth
 (m)
 0.32

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.31

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.31

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.30

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.30

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.29

Kd
 Silt
 for
 Th
 (mL/g)
 0.29

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.28

Kd
 Sand
 for
 Am
 (mL/g)
 0.28

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 88
Soil
 Temperature
 (°C)
 0.28

Kd
 Clay
 for
 Np
 (mL/g)
 0.27

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.27

Ant
 Nest
 Volume
 (m3)
 0.26

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.26

Saturated
 Zone
 Thickness
 (m)
 0.25

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.24

Kd
 Clay
 for
 Pb
 (mL/g)
 0.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.24

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.24

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.24

Unit
 4
 Compacted
 Hb
 (cm)
 0.24

Kd
 Silt
 for
 Pb
 (mL/g)
 0.24

Ant
 Nest
 Shape
 Parameter
 b
 0.23

DCF
 Alpha
 REF
 0.23

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.23

Greasewood
 Root
 Shape
 Parameter
 b
 0.23

Kd
 Silt
 for
 Pa
 (mL/g)
 0.23

Kd
 Clay
 for
 Pa
 (mL/g)
 0.23

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.23

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.23

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.23

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.23

Plant
 Fresh
 Weight
 Conversion
 0.23

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.23

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.22

DCF
 Photon1
 REF
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.22

Kd
 Silt
 for
 Ac
 (mL/g)
 0.22

Deep
 Time
 Lake
 Start
 (yr)
 0.22

Plant.Soil
 Conc
 Ratio
 for
 I
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.22

Resuspended
 Particle
 Fraction
 0.22

RipRap
 Porosity
 0.22

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.22

Kd
 Sand
 for
 Th
 (mL/g)
 0.22

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.22

Kd
 Silt
 for
 Ra
 (mL/g)
 0.21

Mammal
 Burrow
 Shape
 Parameter
 b
 0.21

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.21

Kd
 Silt
 for
 Cs
 (mL/g)
 0.21

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 89
Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.21

Kd
 Silt
 for
 U
 (mL/g)
 0.21

Unit
 2
 Porosity
 0.20

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.20

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.20

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.20

Random
 Gully
 Selector
 0.20

Kd
 Sand
 for
 Cs
 (mL/g)
 0.20

Kd
 Silt
 for
 Pu
 (mL/g)
 0.20

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.20

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.20

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.20

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.19

Kd
 Clay
 for
 Ra
 (mL/g)
 0.19

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.19

Meat
 Post-­‐Cooking
 Loss
 0.19

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.19

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.19

Site
 Dispersal
 Area
 (km2)
 0.19

Kd
 Sand
 for
 Pa
 (mL/g)
 0.19

Deep
 Time
 DCF
 Photon
 2
 REF
 0.19

Silt
 Sand
 Gravel
 Porosity
 0.19

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.19

Meat
 Preparation
 Loss
 0.18

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.18

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.18

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.18

Ant
 Colony
 Lifespan
 (yr)
 0.18

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.18

Greasewood
 Root.Shoot
 Ratio
 0.18

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.18

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.18

Kd
 Sand
 for
 Sr
 (mL/g)
 0.17

Kd
 Clay
 for
 Cs
 (mL/g)
 0.17

Kd
 Clay
 for
 Pu
 (mL/g)
 0.17

Forb
 Root.Shoot
 Ratio
 0.17

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.17

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.17

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.17

Grass
 Root.Shoot
 Ratio
 0.16

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 90
Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.16

Shrub
 Root
 Shape
 Parameter
 b
 0.16

Kd
 Sand
 for
 I
 (mL/g)
 0.16

Receptor
 Area
 (ha)
 0.16

Deep
 Time
 Diffusion
 Length
 (m)
 0.16

Plant.Soil
 Conc
 Ratio
 for
 U
 0.16

Deep
 Time
 DCF
 Alpha
 REF
 0.16

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.16

Fine
 Gravel
 Mix
 Porosity
 0.16

Deep
 Time
 DCF
 Photon
 1
 REF
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.15

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.15

Body
 Weight
 Factor
 for
 Antelope
 0.15

Grass
 Root
 Shape
 Parameter
 b
 0.15

Kd
 Sand
 for
 Pb
 (mL/g)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.15

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.14

Kd
 Clay
 for
 Ac
 (mL/g)
 0.14

Biomass
 %
 Cover
 Selector
 0.14

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.14

Deep
 Time
 Receptor
 Area
 (ac)
 0.14

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.14

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.14

Kd
 Clay
 for
 U
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.14

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.14

Tree
 Root.Shoot
 Ratio
 0.14

Kd
 Sand
 for
 Ac
 (mL/g)
 0.14

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.14

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.14

DCF
 Beta
 REF
 0.14

Deep
 Time
 DCF
 Beta
 REF
 0.14

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.14

Kd
 Sand
 for
 Pu
 (mL/g)
 0.14

Kd
 Clay
 for
 Sr
 (mL/g)
 0.14

Deep
 Time
 Aeolian
 Correlation
 0.13

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.13

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.13

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.13

Forb
 Root
 Shape
 Parameter
 b
 0.13

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 91
OHV
 Dust
 Adjustment
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.13

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.12

DCF
 Photon2
 REF
 0.12

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.12

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.12

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.12

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.12

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.12

Kd
 Silt
 for
 Am
 (mL/g)
 0.11

Kd
 Silt
 for
 Sr
 (mL/g)
 0.11

Kd
 Sand
 for
 Tc
 (mL/g)
 0.11

Saturated
 Zone
 Water
 Table
 Gradient
 0.11

Deep
 Lake
 Depth
 (m)
 0.11

Tortuosity
 Porosity
 Exponent
 0.11

Antelope
 Range
 Area
 (acre)
 0.11

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.11

Tortuosity
 Water
 Content
 Exponent

 0.11

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.11

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.11

Kd
 Clay
 for
 Th
 (mL/g)
 0.10

Contaminated
 Fraction
 of
 GDP
 DU
 0.10

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.08

Vegetation
 Association
 Selector
 0.08

Tree
 Root
 Shape
 Parameter
 b
 0.05

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.05

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 92
Table 19: Peak Dose within 10,000 years – Rancher
R-squared = 85%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 43.78

Kd
 Sand
 for
 Ra
 (mL/g)
 7.92

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 4.08

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 1.19

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.95

Unit
 4
 Compacted
 Porosity
 0.84

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.74

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.71

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.70

Unit
 3
 Porosity
 0.68

Unit
 3
 Residual
 Water
 Content
 0.67

Greasewood
 Root
 Shape
 Parameter
 b
 0.63

Kd
 Clay
 for
 Sr
 (mL/g)
 0.57

Kd
 Sand
 for
 Pa
 (mL/g)
 0.55

Kd
 Sand
 for
 Pb
 (mL/g)
 0.53

Kd
 Silt
 for
 Pu
 (mL/g)
 0.52

Kd
 Clay
 for
 Am
 (mL/g)
 0.43

Unit
 4
 ET
 Layers
 Porosity
 0.43

Intermediate
 Lake
 Depth
 (m)
 0.42

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.42

Kd
 Silt
 for
 Pa
 (mL/g)
 0.40

Grass
 Root
 Shape
 Parameter
 b
 0.39

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.39

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.38

Meat
 Preparation
 Loss
 0.38

Deep
 Time
 Receptor
 Area
 (ac)
 0.36

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.33

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.33

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.32

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.32

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.32

Deep
 Time
 Lake
 Start
 (yr)
 0.31

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.31

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.30

Forb
 Root
 Shape
 Parameter
 b
 0.30

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.30

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.30

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.29

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.29

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 93
Plant.Soil
 Conc
 Ratio
 for
 I
 0.28

Contaminated
 Fraction
 of
 GDP
 DU
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.28

Kd
 Silt
 for
 Np
 (mL/g)
 0.28

Deep
 Time
 Diffusion
 Length
 (m)
 0.28

Ant
 Nest
 Volume
 (m3)
 0.27

Tree
 Root.Shoot
 Ratio
 0.27

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.27

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.27

DCF
 Photon2
 REF
 0.26

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.26

Surface
 Atmosphere
 Thickness
 (m)
 0.26

Kd
 Silt
 for
 Pb
 (mL/g)
 0.25

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.25

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.24

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.24

RipRap
 Bulk
 Density
 (g/cm3)
 0.24

Kd
 Silt
 for
 Sr
 (mL/g)
 0.24

Unit
 2
 Porosity
 0.24

Kd
 Sand
 for
 Ac
 (mL/g)
 0.24

RipRap
 Porosity
 0.23

Kd
 Silt
 for
 Cs
 (mL/g)
 0.23

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.23

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.23

Kd
 Silt
 for
 Ac
 (mL/g)
 0.23

Kd
 Clay
 for
 Pu
 (mL/g)
 0.22

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.22

Kd
 Sand
 for
 Pu
 (mL/g)
 0.22

Surface
 Wind
 Speed
 (m/s)
 0.22

Grass
 Root.Shoot
 Ratio
 0.22

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.22

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.22

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.22

Mammal
 Burrow
 Shape
 Parameter
 b
 0.21

Deep
 Time
 DCF
 Photon
 1
 REF
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.21

Kd
 Clay
 for
 U
 (mL/g)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 U
 0.21

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.21

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.21

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.20

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.20

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 94
Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.20

Kd
 Clay
 for
 Cs
 (mL/g)
 0.20

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.20

Deep
 Lake
 Depth
 (m)
 0.20

Kd
 Silt
 for
 Am
 (mL/g)
 0.19

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.19

Deep
 Time
 DCF
 Photon
 2
 REF
 0.19

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.19

Kd
 Sand
 for
 Sr
 (mL/g)
 0.19

Kd
 Sand
 for
 Tc
 (mL/g)
 0.19

Saturated
 Zone
 Water
 Table
 Gradient
 0.19

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.18

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.18

Kd
 Sand
 for
 Cs
 (mL/g)
 0.18

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.18

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.18

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.18

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.18

Kd
 Silt
 for
 Th
 (mL/g)
 0.18

Soil
 Temperature
 (°C)
 0.18

Deep
 Time
 DCF
 Alpha
 REF
 0.18

Kd
 Silt
 for
 Ra
 (mL/g)
 0.18

Ant
 Colony
 Lifespan
 (yr)
 0.18

OHV
 Dust
 Adjustment
 0.18

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.18

Shrub
 Root
 Shape
 Parameter
 b
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.17

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.17

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.17

Resuspended
 Particle
 Fraction
 0.17

Shrub
 Root.Shoot
 Ratio
 0.17

Tortuosity
 Porosity
 Exponent
 0.17

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.17

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.17

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.16

Plant
 Fresh
 Weight
 Conversion
 0.16

Tortuosity
 Water
 Content
 Exponent

 0.16

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.16

Ant
 Nest
 Shape
 Parameter
 b
 0.16

Kd
 Sand
 for
 Np
 (mL/g)
 0.16

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 95
Body
 Weight
 Factor
 for
 Antelope
 0.16

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.16

Site
 Dispersal
 Area
 (km2)
 0.16

Meat
 Post-­‐Cooking
 Loss
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Greasewood
 Root.Shoot
 Ratio
 0.16

Kd
 Clay
 for
 Pb
 (mL/g)
 0.16

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.15

DCF
 Beta
 REF
 0.15

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.15

Silt
 Sand
 Gravel
 Porosity
 0.15

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.15

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.15

Random
 Gully
 Selector
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.15

Kd
 Silt
 for
 U
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.14

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.14

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.14

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.14

Biomass
 %
 Cover
 Selector
 0.14

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.14

Fine
 CobbleMix
 Porosity
 0.14

Kd
 Clay
 for
 Np
 (mL/g)
 0.14

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.14

Fine
 Gravel
 Mix
 Porosity
 0.14

Kd
 Clay
 for
 Ra
 (mL/g)
 0.13

Deep
 Time
 DCF
 Beta
 REF
 0.13

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.13

Saturated
 Zone
 Thickness
 (m)
 0.13

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.13

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.13

Kd
 Sand
 for
 U
 (mL/g)
 0.13

DCF
 Photon1
 REF
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.13

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.12

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.12

Kd
 Clay
 for
 Th
 (mL/g)
 0.12

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 96
Kd
 Clay
 for
 Ac
 (mL/g)
 0.12

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.12

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.12

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.11

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.11

Kd
 Sand
 for
 Th
 (mL/g)
 0.11

Unit
 4
 Compacted
 Hb
 (cm)
 0.11

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.11

Kd
 Clay
 for
 Pa
 (mL/g)
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.11

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.10

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.10

Forb
 Root.Shoot
 Ratio
 0.10

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.10

DCF
 Alpha
 REF
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.10

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.09

Receptor
 Area
 (ha)
 0.09

Vegetation
 Association
 Selector
 0.09

Kd
 Sand
 for
 Am
 (mL/g)
 0.09

Tree
 Root
 Shape
 Parameter
 b
 0.09

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.09

Antelope
 Range
 Area
 (acre)
 0.08

Deep
 Time
 Aeolian
 Correlation
 0.08

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.07

Kd
 Sand
 for
 I
 (mL/g)
 0.06

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 97
Table 20: Peak Uranium Hazard within 10,000 years - Hunter
R-squared = 96%
Explanatory
 Variable
 Sensitivity
 Index

Unit
 3
 Porosity
 8.65

Contaminated
 Fraction
 of
 GDP
 DU
 8.55

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 7.42

Tree
 Root.Shoot
 Ratio
 7.07

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 3.32

Kd
 Sand
 for
 U
 (mL/g)
 2.49

Kd
 Sand
 for
 Am
 (mL/g)
 2.21

Unit
 3
 Residual
 Water
 Content
 1.72

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 1.66

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 1.59

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.53

DCF
 Alpha
 REF
 1.47

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 1.45

Shrub
 Root
 Shape
 Parameter
 b
 1.17

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 1.16

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 1.00

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.87

Saturated
 Zone
 Thickness
 (m)
 0.85

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.84

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.83

Kd
 Sand
 for
 Tc
 (mL/g)
 0.82

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.79

Grass
 Root
 Shape
 Parameter
 b
 0.74

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.73

Unit
 4
 Compacted
 Porosity
 0.72

DCF
 Beta
 REF
 0.70

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.69

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.68

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.67

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.67

Biomass
 %
 Cover
 Selector
 0.66

Vegetation
 Association
 Selector
 0.64

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.59

Kd
 Sand
 for
 Sr
 (mL/g)
 0.59

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.57

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.57

Kd
 Sand
 for
 Ra
 (mL/g)
 0.57

Soil
 Temperature
 (°C)
 0.56

Kd
 Silt
 for
 U
 (mL/g)
 0.56

Kd
 Silt
 for
 Pa
 (mL/g)
 0.56

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 98
Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.55

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.55

Kd
 Clay
 for
 Am
 (mL/g)
 0.53

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.52

Kd
 Sand
 for
 Ac
 (mL/g)
 0.52

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.51

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.50

Kd
 Silt
 for
 Th
 (mL/g)
 0.50

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.49

Kd
 Clay
 for
 Np
 (mL/g)
 0.48

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.48

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.48

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.48

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.48

Unit
 4
 ET
 Layers
 Porosity
 0.46

Greasewood
 Root
 Shape
 Parameter
 b
 0.45

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.44

Fine
 CobbleMix
 Porosity
 0.43

Deep
 Time
 DCF
 Photon
 1
 REF
 0.43

Deep
 Time
 DCF
 Beta
 REF
 0.43

Kd
 Silt
 for
 Cs
 (mL/g)
 0.42

Kd
 Silt
 for
 Am
 (mL/g)
 0.41

Surface
 Wind
 Speed
 (m/s)
 0.41

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.40

Deep
 Time
 Lake
 Start
 (yr)
 0.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.39

Kd
 Silt
 for
 Sr
 (mL/g)
 0.39

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.38

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.37

Intermediate
 Lake
 Depth
 (m)
 0.37

Kd
 Sand
 for
 Pb
 (mL/g)
 0.37

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.36

Surface
 Atmosphere
 Thickness
 (m)
 0.36

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.36

Site
 Dispersal
 Area
 (km2)
 0.36

Deep
 Time
 Receptor
 Area
 (ac)
 0.35

Kd
 Silt
 for
 Pu
 (mL/g)
 0.35

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.35

Unit
 4
 Compacted
 Hb
 (cm)
 0.35

RipRap
 Bulk
 Density
 (g/cm3)
 0.35

Soil
 Ingestion
 Tracer
 Element
 0.34

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 99
Plant.Soil
 Conc
 Ratio
 for
 I
 0.33

Kd
 Sand
 for
 Np
 (mL/g)
 0.33

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.32

Kd
 Clay
 for
 Pu
 (mL/g)
 0.31

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.30

Deep
 Time
 Diffusion
 Length
 (m)
 0.29

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.29

Kd
 Sand
 for
 I
 (mL/g)
 0.28

Kd
 Silt
 for
 Ra
 (mL/g)
 0.28

Ant
 Nest
 Shape
 Parameter
 b
 0.27

RipRap
 Porosity
 0.26

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.26

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.25

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.24

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.23

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.22

Tortuosity
 Porosity
 Exponent
 0.22

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.22

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.21

Kd
 Clay
 for
 Ra
 (mL/g)
 0.21

Silt
 Sand
 Gravel
 Porosity
 0.21

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.21

Forb
 Root
 Shape
 Parameter
 b
 0.21

Forb
 Root.Shoot
 Ratio
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.20

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.19

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.18

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.18

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.17

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.16

Antelope
 Range
 Area
 (acre)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.15

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.15

Ant
 Nest
 Volume
 (m3)
 0.15

Plant.Soil
 Conc
 Ratio
 for
 U
 0.15

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.14

Kd
 Silt
 for
 Pb
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.13

Kd
 Clay
 for
 Pa
 (mL/g)
 0.12

Kd
 Clay
 for
 Sr
 (mL/g)
 0.12

Tortuosity
 Water
 Content
 Exponent

 0.12

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.12

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.12

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 100
OHV
 Dust
 Adjustment
 0.12

Meat
 Post-­‐Cooking
 Loss
 0.12

Kd
 Silt
 for
 Np
 (mL/g)
 0.11

Random
 Gully
 Selector
 0.11

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.11

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.10

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.09

Mammal
 Burrow
 Shape
 Parameter
 b
 0.09

Deep
 Lake
 Depth
 (m)
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.09

Kd
 Sand
 for
 Th
 (mL/g)
 0.09

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.09

Grass
 Root.Shoot
 Ratio
 0.09

DCF
 Photon1
 REF
 0.08

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.08

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.08

Greasewood
 Root.Shoot
 Ratio
 0.07

Tree
 Root
 Shape
 Parameter
 b
 0.07

Kd
 Silt
 for
 Ac
 (mL/g)
 0.07

Kd
 Clay
 for
 Pb
 (mL/g)
 0.07

Meat
 Preparation
 Loss
 0.07

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.07

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.06

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.06

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.06

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.06

Unit
 2
 Porosity
 0.06

Deep
 Time
 DCF
 Alpha
 REF
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.06

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.05

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.05

Kd
 Sand
 for
 Pa
 (mL/g)
 0.05

Kd
 Sand
 for
 Pu
 (mL/g)
 0.05

Fine
 Gravel
 Mix
 Porosity
 0.05

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.05

Receptor
 Area
 (ha)
 0.05

Kd
 Clay
 for
 Cs
 (mL/g)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.05

DCF
 Photon2
 REF
 0.04

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 101
Resuspended
 Particle
 Fraction
 0.04

Kd
 Clay
 for
 U
 (mL/g)
 0.04

Kd
 Sand
 for
 Cs
 (mL/g)
 0.04

Saturated
 Zone
 Water
 Table
 Gradient
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.04

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.04

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.04

Ant
 Colony
 Lifespan
 (yr)
 0.03

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.03

Kd
 Clay
 for
 Th
 (mL/g)
 0.02

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.02

Shrub
 Root.Shoot
 Ratio
 0.02

Deep
 Time
 Aeolian
 Correlation
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.02

Plant
 Fresh
 Weight
 Conversion
 0.02

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.02

Kd
 Clay
 for
 Ac
 (mL/g)
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.01

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.01

Deep
 Time
 DCF
 Photon
 2
 REF
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.01

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.01

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.01

Body
 Weight
 Factor
 for
 Antelope
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 102
Table 21: Peak Uranium Hazard within 10,000 years - Rancher
R-squared = 94%
Explanatory
 Variable
 Sensitivity
 Index

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 44.11

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 6.68

Kd
 Sand
 for
 U
 (mL/g)
 2.86

Unit
 3
 Residual
 Water
 Content
 2.47

Ant
 Nest
 Shape
 Parameter
 b
 2.08

DCF
 Alpha
 REF
 1.95

Saltwater
 Solubility
 for
 I
 (mol/L)
 1.83

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.79

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 1.39

Contaminated
 Fraction
 of
 GDP
 DU
 1.38

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 1.37

Saturated
 Zone
 Thickness
 (m)
 0.92

Soil
 Temperature
 (°C)
 0.86

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.85

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.71

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.66

Deep
 Time
 DCF
 Photon
 2
 REF
 0.58

Deep
 Time
 Diffusion
 Length
 (m)
 0.57

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.50

Unit
 4
 ET
 Layers
 Porosity
 0.47

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.46

Kd
 Sand
 for
 Am
 (mL/g)
 0.46

Tree
 Root.Shoot
 Ratio
 0.45

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.44

Kd
 Silt
 for
 U
 (mL/g)
 0.44

Silt
 Sand
 Gravel
 Porosity
 0.43

Kd
 Silt
 for
 Ra
 (mL/g)
 0.43

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.42

Unit
 4
 Compacted
 Porosity
 0.39

Kd
 Sand
 for
 Cs
 (mL/g)
 0.38

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.38

Fine
 CobbleMix
 Porosity
 0.38

Kd
 Sand
 for
 Ra
 (mL/g)
 0.37

DCF
 Photon1
 REF
 0.36

Kd
 Silt
 for
 Th
 (mL/g)
 0.35

Tortuosity
 Water
 Content
 Exponent

 0.35

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.33

Ant
 Colony
 Lifespan
 (yr)
 0.33

Kd
 Sand
 for
 Np
 (mL/g)
 0.31

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.30

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 103
Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.30

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.29

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.29

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.29

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.29

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.28

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.28

Kd
 Sand
 for
 Pu
 (mL/g)
 0.26

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.26

Plant.Soil
 Conc
 Ratio
 for
 U
 0.25

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.24

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.24

Kd
 Sand
 for
 Sr
 (mL/g)
 0.24

Mammal
 Burrow
 Shape
 Parameter
 b
 0.24

Deep
 Time
 DCF
 Alpha
 REF
 0.24

Kd
 Clay
 for
 Cs
 (mL/g)
 0.23

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.22

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.21

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.20

OHV
 Dust
 Adjustment
 0.20

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.19

Unit
 3
 Porosity
 0.19

Fine
 Gravel
 Mix
 Porosity
 0.19

Greasewood
 Root
 Shape
 Parameter
 b
 0.19

RipRap
 Bulk
 Density
 (g/cm3)
 0.19

Kd
 Silt
 for
 Sr
 (mL/g)
 0.19

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.19

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.18

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.18

Deep
 Time
 DCF
 Photon
 1
 REF
 0.18

Surface
 Atmosphere
 Thickness
 (m)
 0.18

Kd
 Sand
 for
 Pb
 (mL/g)
 0.17

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.17

RipRap
 Porosity
 0.17

Kd
 Clay
 for
 Am
 (mL/g)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.16

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.16

Kd
 Sand
 for
 Pa
 (mL/g)
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.15

Kd
 Sand
 for
 Ac
 (mL/g)
 0.15

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 104
Kd
 Clay
 for
 Pa
 (mL/g)
 0.15

Biomass
 %
 Cover
 Selector
 0.15

Kd
 Silt
 for
 Ac
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.14

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.14

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.14

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.14

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.14

Plant.Soil
 Conc
 Ratio
 for
 I
 0.14

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.14

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.13

Kd
 Silt
 for
 Am
 (mL/g)
 0.13

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.13

Greasewood
 Root.Shoot
 Ratio
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.13

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.13

Unit
 4
 Compacted
 Hb
 (cm)
 0.13

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.13

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.13

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.13

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.13

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.12

Kd
 Clay
 for
 Pb
 (mL/g)
 0.12

Surface
 Wind
 Speed
 (m/s)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.12

Ant
 Nest
 Volume
 (m3)
 0.12

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.11

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.11

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.11

Deep
 Time
 Lake
 Start
 (yr)
 0.11

Meat
 Preparation
 Loss
 0.11

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.11

Kd
 Sand
 for
 Th
 (mL/g)
 0.11

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.11

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.11

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.10

Resuspended
 Particle
 Fraction
 0.10

Kd
 Silt
 for
 Np
 (mL/g)
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.10

Kd
 Clay
 for
 Pu
 (mL/g)
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.10

Shrub
 Root.Shoot
 Ratio
 0.10

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 105
Unit
 3
 Bulk
 Density
 (g/cm3)
 0.10

Tree
 Root
 Shape
 Parameter
 b
 0.10

Kd
 Silt
 for
 Pa
 (mL/g)
 0.09

Kd
 Silt
 for
 Pb
 (mL/g)
 0.09

Kd
 Clay
 for
 Np
 (mL/g)
 0.09

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.09

Kd
 Clay
 for
 Ac
 (mL/g)
 0.09

Soil
 Ingestion
 Tracer
 Element
 0.09

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.09

Unit
 2
 Porosity
 0.09

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.09

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.09

Antelope
 Range
 Area
 (acre)
 0.08

Tortuosity
 Porosity
 Exponent
 0.08

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.08

Deep
 Lake
 Depth
 (m)
 0.08

Grass
 Root
 Shape
 Parameter
 b
 0.08

Site
 Dispersal
 Area
 (km2)
 0.08

Kd
 Silt
 for
 Cs
 (mL/g)
 0.08

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.08

Vegetation
 Association
 Selector
 0.08

Kd
 Clay
 for
 U
 (mL/g)
 0.08

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.07

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.07

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.07

Deep
 Time
 Receptor
 Area
 (ac)
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.06

Grass
 Root.Shoot
 Ratio
 0.06

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.06

Random
 Gully
 Selector
 0.06

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.05

Saturated
 Zone
 Water
 Table
 Gradient
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.05

Kd
 Silt
 for
 Pu
 (mL/g)
 0.05

Intermediate
 Lake
 Depth
 (m)
 0.05

Kd
 Clay
 for
 Sr
 (mL/g)
 0.05

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.05

Shrub
 Root
 Shape
 Parameter
 b
 0.05

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.05

DCF
 Beta
 REF
 0.05

Kd
 Clay
 for
 Ra
 (mL/g)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 106
GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.04

Forb
 Root
 Shape
 Parameter
 b
 0.04

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.04

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.04

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.03

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Kd
 Clay
 for
 Th
 (mL/g)
 0.03

Forb
 Root.Shoot
 Ratio
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.03

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.02

Deep
 Time
 Aeolian
 Correlation
 0.02

Kd
 Sand
 for
 Tc
 (mL/g)
 0.02

Plant
 Fresh
 Weight
 Conversion
 0.02

Kd
 Sand
 for
 I
 (mL/g)
 0.02

Meat
 Post-­‐Cooking
 Loss
 0.02

Deep
 Time
 DCF
 Beta
 REF
 0.02

Body
 Weight
 Factor
 for
 Antelope
 0.02

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.01

Receptor
 Area
 (ha)
 0.01

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.01

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.01

DCF
 Photon2
 REF
 0.00

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 107
Table 22: Peak Uranium Hazard within 10,000 years – Sport OHV
R-squared = 96%
Explanatory
 Variable
 Sensitivity
 Index

Contaminated
 Fraction
 of
 GDP
 DU
 13.40

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 11.80

Tree
 Root.Shoot
 Ratio
 5.93

Unit
 3
 Porosity
 4.09

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 3.06

Kd
 Sand
 for
 U
 (mL/g)
 2.63

Kd
 Sand
 for
 Am
 (mL/g)
 2.06

Unit
 3
 Residual
 Water
 Content
 1.93

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 1.92

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 1.85

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 1.76

DCF
 Alpha
 REF
 1.14

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 1.14

Shrub
 Root
 Shape
 Parameter
 b
 0.96

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.96

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.88

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.88

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.83

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.80

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.78

Kd
 Sand
 for
 Tc
 (mL/g)
 0.73

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.71

Saturated
 Zone
 Thickness
 (m)
 0.69

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.68

Unit
 4
 Compacted
 Porosity
 0.67

Biomass
 %
 Cover
 Selector
 0.66

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.63

Unit
 4
 ET
 Layers
 Porosity
 0.63

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.61

Kd
 Silt
 for
 Pb
 (mL/g)
 0.60

DCF
 Beta
 REF
 0.59

Kd
 Silt
 for
 U
 (mL/g)
 0.59

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.57

Grass
 Root
 Shape
 Parameter
 b
 0.55

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.55

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.53

Kd
 Silt
 for
 Th
 (mL/g)
 0.53

Vegetation
 Association
 Selector
 0.52

Kd
 Sand
 for
 Sr
 (mL/g)
 0.52

Kd
 Sand
 for
 Ra
 (mL/g)
 0.51

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 108
Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.51

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.51

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.51

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.50

Kd
 Clay
 for
 Np
 (mL/g)
 0.48

Kd
 Sand
 for
 Pb
 (mL/g)
 0.46

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.45

Kd
 Clay
 for
 Am
 (mL/g)
 0.45

Kd
 Sand
 for
 Ac
 (mL/g)
 0.45

Kd
 Silt
 for
 Pa
 (mL/g)
 0.44

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.43

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.42

Kd
 Silt
 for
 Sr
 (mL/g)
 0.42

Soil
 Temperature
 (°C)
 0.41

Surface
 Wind
 Speed
 (m/s)
 0.41

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.41

Deep
 Time
 Lake
 Start
 (yr)
 0.40

Kd
 Sand
 for
 Np
 (mL/g)
 0.40

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.39

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.39

Intermediate
 Lake
 Depth
 (m)
 0.39

Greasewood
 Root
 Shape
 Parameter
 b
 0.38

Deep
 Time
 DCF
 Beta
 REF
 0.38

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.37

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.37

Surface
 Atmosphere
 Thickness
 (m)
 0.36

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.35

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.35

Fine
 CobbleMix
 Porosity
 0.34

Kd
 Silt
 for
 Cs
 (mL/g)
 0.34

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.34

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.32

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.32

Site
 Dispersal
 Area
 (km2)
 0.32

Ant
 Nest
 Shape
 Parameter
 b
 0.31

RipRap
 Bulk
 Density
 (g/cm3)
 0.31

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.30

Deep
 Time
 DCF
 Photon
 1
 REF
 0.30

Unit
 4
 Compacted
 Hb
 (cm)
 0.30

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.30

Plant.Soil
 Conc
 Ratio
 for
 I
 0.29

Soil
 Ingestion
 Tracer
 Element
 0.29

Kd
 Silt
 for
 Pu
 (mL/g)
 0.28

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 109
Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.28

RipRap
 Porosity
 0.26

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.26

Kd
 Clay
 for
 Sr
 (mL/g)
 0.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.25

Kd
 Silt
 for
 Np
 (mL/g)
 0.25

Forb
 Root
 Shape
 Parameter
 b
 0.25

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.24

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.24

Deep
 Time
 Receptor
 Area
 (ac)
 0.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.24

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.23

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.23

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.23

Kd
 Clay
 for
 Pu
 (mL/g)
 0.22

Forb
 Root.Shoot
 Ratio
 0.22

Kd
 Silt
 for
 Am
 (mL/g)
 0.21

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.21

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.21

Deep
 Time
 Diffusion
 Length
 (m)
 0.21

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.21

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.21

Silt
 Sand
 Gravel
 Porosity
 0.20

Tortuosity
 Porosity
 Exponent
 0.20

Kd
 Clay
 for
 Pa
 (mL/g)
 0.19

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.19

Random
 Gully
 Selector
 0.19

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.19

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.19

Kd
 Clay
 for
 Ra
 (mL/g)
 0.17

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.17

Kd
 Silt
 for
 Ra
 (mL/g)
 0.17

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.16

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.15

Kd
 Sand
 for
 I
 (mL/g)
 0.15

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.15

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.14

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.14

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.13

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.13

Plant.Soil
 Conc
 Ratio
 for
 U
 0.13

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 110
Antelope
 Range
 Area
 (acre)
 0.12

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.11

Tortuosity
 Water
 Content
 Exponent

 0.11

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.11

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.11

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.11

Grass
 Root.Shoot
 Ratio
 0.11

OHV
 Dust
 Adjustment
 0.10

Kd
 Sand
 for
 Pa
 (mL/g)
 0.10

Unit
 2
 Porosity
 0.10

Meat
 Post-­‐Cooking
 Loss
 0.09

DCF
 Photon1
 REF
 0.09

Fine
 Gravel
 Mix
 Porosity
 0.09

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.09

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.09

Deep
 Lake
 Depth
 (m)
 0.09

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.09

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.08

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.08

Meat
 Preparation
 Loss
 0.08

Receptor
 Area
 (ha)
 0.08

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.08

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.08

Kd
 Clay
 for
 U
 (mL/g)
 0.08

Kd
 Sand
 for
 Cs
 (mL/g)
 0.08

Ant
 Nest
 Volume
 (m3)
 0.08

Mammal
 Burrow
 Shape
 Parameter
 b
 0.07

Kd
 Sand
 for
 Th
 (mL/g)
 0.07

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.07

Deep
 Time
 DCF
 Alpha
 REF
 0.07

Greasewood
 Root.Shoot
 Ratio
 0.07

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.07

Kd
 Clay
 for
 Cs
 (mL/g)
 0.06

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.06

Kd
 Sand
 for
 Pu
 (mL/g)
 0.05

Kd
 Silt
 for
 Ac
 (mL/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.05

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.05

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 111
Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.05

Kd
 Clay
 for
 Pb
 (mL/g)
 0.05

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.04

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.04

Kd
 Clay
 for
 Th
 (mL/g)
 0.03

Ant
 Colony
 Lifespan
 (yr)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.03

Kd
 Clay
 for
 Ac
 (mL/g)
 0.03

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.03

Body
 Weight
 Factor
 for
 Antelope
 0.03

DCF
 Photon2
 REF
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.02

Deep
 Time
 Aeolian
 Correlation
 0.02

Resuspended
 Particle
 Fraction
 0.02

Saturated
 Zone
 Water
 Table
 Gradient
 0.02

Tree
 Root
 Shape
 Parameter
 b
 0.02

Deep
 Time
 DCF
 Photon
 2
 REF
 0.02

Shrub
 Root.Shoot
 Ratio
 0.02

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.02

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.02

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.02

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.02

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.02

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.02

Plant
 Fresh
 Weight
 Conversion
 0.02

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.01

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.01

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.01

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 112
Table 23: Benson Peak Groundwater Well Concentrations within 500 years –Tc99
R-squared = %
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 Tc
 (mL/g)
 42.71

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 15.8

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 13.92

VG_n_Benson
 5.42

Porosity_Benson
 1.79

Saturated
 Zone
 Water
 Table
 Gradient
 1.6

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.73

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.59

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.5

alpha_Benson
 0.42

Deep
 Time
 DCF
 Photon
 1
 REF
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.32

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.29

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.26

Unit
 3
 Porosity
 0.25

Kd
 Sand
 for
 Ra
 (mL/g)
 0.25

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.23

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.22

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.2

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.19

Kd
 Sand
 for
 U
 (mL/g)
 0.18

Meat
 Post-­‐Cooking
 Loss
 0.18

Kd
 Clay
 for
 Ac
 (mL/g)
 0.18

Ks_Benson
 0.17

Antelope
 Range
 Area
 (acre)
 0.17

OHV
 Dust
 Adjustment
 0.17

Kd
 Sand
 for
 Th
 (mL/g)
 0.16

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.15

Deep
 Time
 Receptor
 Area
 (ac)
 0.13

Tortuosity
 Porosity
 Exponent
 0.13

Tortuosity
 Water
 Content
 Exponent

 0.13

Kd
 Sand
 for
 Cs
 (mL/g)
 0.13

Kd
 Clay
 for
 Th
 (mL/g)
 0.13

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.13

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.13

Deep
 Time
 Aeolian
 Correlation
 0.13

Grass
 Root
 Shape
 Parameter
 b
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.12

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 113
DCF
 Photon2
 REF
 0.12

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.12

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.12

Unit
 2
 Porosity
 0.12

Plant.Soil
 Conc
 Ratio
 for
 I
 0.12

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.12

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.12

Deep
 Time
 DCF
 Beta
 REF
 0.11

Kd
 Silt
 for
 Pa
 (mL/g)
 0.11

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.11

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.11

Kd
 Silt
 for
 Am
 (mL/g)
 0.11

Kd
 Sand
 for
 Pb
 (mL/g)
 0.11

Deep
 Time
 Diffusion
 Length
 (m)
 0.11

Tree
 Root.Shoot
 Ratio
 0.11

Kd
 Silt
 for
 Np
 (mL/g)
 0.11

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.1

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.1

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.1

Kd
 Sand
 for
 Am
 (mL/g)
 0.1

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.1

Kd
 Clay
 for
 Np
 (mL/g)
 0.1

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.1

Soil
 Temperature
 (°C)
 0.1

Kd
 Sand
 for
 Sr
 (mL/g)
 0.1

Kd
 Sand
 for
 Pa
 (mL/g)
 0.1

Kd
 Sand
 for
 Pu
 (mL/g)
 0.1

Kd
 Clay
 for
 U
 (mL/g)
 0.09

Greasewood
 Root
 Shape
 Parameter
 b
 0.09

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.09

Deep
 Time
 Lake
 Start
 (yr)
 0.09

Site
 Dispersal
 Area
 (km2)
 0.09

Ant
 Nest
 Shape
 Parameter
 b
 0.09

Surface
 Atmosphere
 Thickness
 (m)
 0.09

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.09

Kd
 Silt
 for
 Cs
 (mL/g)
 0.09

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.09

Deep
 Time
 DCF
 Alpha
 REF
 0.09

Kd
 Silt
 for
 Ra
 (mL/g)
 0.09

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.09

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.09

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.08

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 114
Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.08

Resuspended
 Particle
 Fraction
 0.08

Fine
 CobbleMix
 Porosity
 0.08

Kd
 Sand
 for
 Ac
 (mL/g)
 0.08

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.08

Kd
 Silt
 for
 Pu
 (mL/g)
 0.08

RipRap
 Porosity
 0.08

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.08

Meat
 Preparation
 Loss
 0.08

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.08

RipRap
 Bulk
 Density
 (g/cm3)
 0.08

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.08

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.08

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.07

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.07

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.07

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.07

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.07

Plant
 Fresh
 Weight
 Conversion
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.07

Kd
 Clay
 for
 Ra
 (mL/g)
 0.07

Greasewood
 Root.Shoot
 Ratio
 0.07

Shrub
 Root.Shoot
 Ratio
 0.07

Kd
 Clay
 for
 Pa
 (mL/g)
 0.07

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.07

DCF
 Photon1
 REF
 0.07

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.07

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.07

Saturated
 Zone
 Thickness
 (m)
 0.06

Silt
 Sand
 Gravel
 Porosity
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.06

Kd
 Clay
 for
 Pu
 (mL/g)
 0.06

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.06

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.06

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.06

Kd
 Clay
 for
 Am
 (mL/g)
 0.06

Fine
 Gravel
 Mix
 Porosity
 0.06

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.06

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.06

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 115
Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.06

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.06

Unit
 4
 Compacted
 Porosity
 0.06

Kd
 Silt
 for
 Sr
 (mL/g)
 0.06

Kd
 Silt
 for
 U
 (mL/g)
 0.06

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.06

Ant
 Nest
 Volume
 (m3)
 0.06

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.06

Deep
 Lake
 Depth
 (m)
 0.06

Ant
 Colony
 Lifespan
 (yr)
 0.06

Contaminated
 Fraction
 of
 GDP
 DU
 0.06

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.05

Shrub
 Root
 Shape
 Parameter
 b
 0.05

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.05

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.05

Surface
 Wind
 Speed
 (m/s)
 0.05

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.05

Kd
 Silt
 for
 Pb
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.05

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.05

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.05

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.05

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.05

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.05

Unit
 3
 Residual
 Water
 Content
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.05

Kd
 Sand
 for
 Np
 (mL/g)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.05

Kd
 Clay
 for
 Sr
 (mL/g)
 0.05

Grass
 Root.Shoot
 Ratio
 0.05

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.05

Kd
 Clay
 for
 Pb
 (mL/g)
 0.04

Biomass
 %
 Cover
 Selector
 0.04

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.04

Tree
 Root
 Shape
 Parameter
 b
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.04

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 116
Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.04

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.04

Unit
 4
 ET
 Layers
 Porosity
 0.04

Intermediate
 Lake
 Depth
 (m)
 0.04

Mammal
 Burrow
 Shape
 Parameter
 b
 0.04

DCF
 Beta
 REF
 0.04

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.04

Vegetation
 Association
 Selector
 0.04

DCF
 Alpha
 REF
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.04

Kd
 Sand
 for
 I
 (mL/g)
 0.04

Unit
 4
 Compacted
 Hb
 (cm)
 0.04

Random
 Gully
 Selector
 0.04

Deep
 Time
 DCF
 Photon
 2
 REF
 0.04

Kd
 Silt
 for
 Th
 (mL/g)
 0.04

Kd
 Clay
 for
 Cs
 (mL/g)
 0.04

Kd
 Silt
 for
 Ac
 (mL/g)
 0.04

Forb
 Root
 Shape
 Parameter
 b
 0.04

Receptor
 Area
 (ha)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 U
 0.04

Body
 Weight
 Factor
 for
 Antelope
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.04

Forb
 Root.Shoot
 Ratio
 0.03

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.03

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.03

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.03

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.02

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 117
Table 24: Benson Peak Dose within 10,000 years – Rancher
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 37.68

Kd
 Sand
 for
 Ra
 (mL/g)
 3.61

Plant.Soil
 Conc
 Ratio
 for
 Ra
 1.15

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.12

Porosity_Benson
 0.79

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.74

Kd
 Silt
 for
 Th
 (mL/g)
 0.74

VG_n_Benson
 0.71

Ks_Benson
 0.70

Grass
 Root.Shoot
 Ratio
 0.69

Soil
 Temperature
 (°C)
 0.66

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.61

Kd
 Sand
 for
 Am
 (mL/g)
 0.59

Deep
 Time
 DCF
 Photon
 2
 REF
 0.59

Kd
 Clay
 for
 Ra
 (mL/g)
 0.59

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.59

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.56

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.54

Kd
 Clay
 for
 U
 (mL/g)
 0.54

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.54

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.53

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.53

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.52

Receptor
 Area
 (ha)
 0.50

Kd
 Silt
 for
 Np
 (mL/g)
 0.50

Kd
 Sand
 for
 Cs
 (mL/g)
 0.50

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.49

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.48

Deep
 Time
 Diffusion
 Length
 (m)
 0.47

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.46

Fine
 CobbleMix
 Porosity
 0.45

Plant.Soil
 Conc
 Ratio
 for
 U
 0.45

Ant
 Nest
 Shape
 Parameter
 b
 0.44

Silt
 Sand
 Gravel
 Porosity
 0.44

Kd
 Silt
 for
 Cs
 (mL/g)
 0.44

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.44

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.43

Unit
 4
 Compacted
 Hb
 (cm)
 0.42

Shrub
 Root.Shoot
 Ratio
 0.42

DCF
 Photon2
 REF
 0.42

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 118
Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.41

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.40

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.39

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.38

Antelope
 Range
 Area
 (acre)
 0.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.37

Ant
 Nest
 Volume
 (m3)
 0.37

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.37

alpha_Benson
 0.37

Body
 Weight
 Factor
 for
 Antelope
 0.37

Kd
 Silt
 for
 Pu
 (mL/g)
 0.37

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.36

Biomass
 %
 Cover
 Selector
 0.35

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.35

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.34

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.34

Kd
 Silt
 for
 U
 (mL/g)
 0.34

Unit
 2
 Porosity
 0.33

OHV
 Dust
 Adjustment
 0.33

Kd
 Silt
 for
 Pb
 (mL/g)
 0.33

Unit
 4
 Compacted
 Porosity
 0.32

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.32

Deep
 Time
 Lake
 Start
 (yr)
 0.32

Plant
 Fresh
 Weight
 Conversion
 0.31

Grass
 Root
 Shape
 Parameter
 b
 0.31

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.31

Kd
 Sand
 for
 U
 (mL/g)
 0.31

Forb
 Root.Shoot
 Ratio
 0.31

Deep
 Time
 Aeolian
 Correlation
 0.31

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.30

Meat
 Preparation
 Loss
 0.30

Kd
 Silt
 for
 Ra
 (mL/g)
 0.30

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.30

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.30

Contaminated
 Fraction
 of
 GDP
 DU
 0.29

Kd
 Clay
 for
 Pa
 (mL/g)
 0.29

Kd
 Clay
 for
 Pu
 (mL/g)
 0.29

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.29

Kd
 Sand
 for
 I
 (mL/g)
 0.28

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.28

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.28

Plant.Soil
 Conc
 Ratio
 for
 I
 0.28

Kd
 Silt
 for
 Sr
 (mL/g)
 0.28

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 119
Resuspension
 Flux
 (kg.m2-­‐yr)
 0.27

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.27

Kd
 Clay
 for
 Ac
 (mL/g)
 0.27

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.27

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.26

Ant
 Colony
 Lifespan
 (yr)
 0.26

Intermediate
 Lake
 Depth
 (m)
 0.26

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.26

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.26

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.26

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.26

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.26

Saturated
 Zone
 Thickness
 (m)
 0.26

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.25

Deep
 Time
 DCF
 Alpha
 REF
 0.25

Kd
 Clay
 for
 Cs
 (mL/g)
 0.25

Greasewood
 Root.Shoot
 Ratio
 0.25

RipRap
 Bulk
 Density
 (g/cm3)
 0.25

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.25

Kd
 Sand
 for
 Pu
 (mL/g)
 0.24

Unit
 3
 Residual
 Water
 Content
 0.24

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.24

Meat
 Post-­‐Cooking
 Loss
 0.24

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.24

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.24

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.24

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.23

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.23

Shrub
 Root
 Shape
 Parameter
 b
 0.23

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.23

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.23

DCF
 Beta
 REF
 0.23

Kd
 Silt
 for
 Pa
 (mL/g)
 0.23

Kd
 Silt
 for
 Am
 (mL/g)
 0.23

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.23

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.22

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.22

Tree
 Root
 Shape
 Parameter
 b
 0.22

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.22

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.22

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.22

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 120
Random
 Gully
 Selector
 0.22

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.22

Kd
 Clay
 for
 Np
 (mL/g)
 0.22

Kd
 Sand
 for
 Sr
 (mL/g)
 0.22

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.22

Kd
 Silt
 for
 Ac
 (mL/g)
 0.21

Unit
 4
 ET
 Layers
 Porosity
 0.21

Site
 Dispersal
 Area
 (km2)
 0.21

Surface
 Atmosphere
 Thickness
 (m)
 0.21

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.20

DCF
 Photon1
 REF
 0.20

Deep
 Time
 DCF
 Beta
 REF
 0.20

Kd
 Sand
 for
 Th
 (mL/g)
 0.20

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.20

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.20

Resuspended
 Particle
 Fraction
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.20

Fine
 Gravel
 Mix
 Porosity
 0.20

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.20

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.20

Kd
 Clay
 for
 Th
 (mL/g)
 0.20

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.20

Forb
 Root
 Shape
 Parameter
 b
 0.19

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.19

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.19

Kd
 Sand
 for
 Np
 (mL/g)
 0.19

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.19

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.19

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.18

Kd
 Clay
 for
 Am
 (mL/g)
 0.18

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.18

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.18

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.18

Mammal
 Burrow
 Shape
 Parameter
 b
 0.17

Surface
 Wind
 Speed
 (m/s)
 0.17

Saturated
 Zone
 Water
 Table
 Gradient
 0.17

Vegetation
 Association
 Selector
 0.17

Deep
 Lake
 Depth
 (m)
 0.17

Kd
 Sand
 for
 Pb
 (mL/g)
 0.17

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.17

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.17

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.16

RipRap
 Porosity
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 121
Kd
 Clay
 for
 Pb
 (mL/g)
 0.16

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.16

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.16

Deep
 Time
 DCF
 Photon
 1
 REF
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.16

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.15

Kd
 Sand
 for
 Ac
 (mL/g)
 0.15

DCF
 Alpha
 REF
 0.15

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.15

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.15

Tree
 Root.Shoot
 Ratio
 0.14

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.14

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.14

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.14

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.14

Kd
 Clay
 for
 Sr
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.13

Deep
 Time
 Receptor
 Area
 (ac)
 0.13

Kd
 Sand
 for
 Pa
 (mL/g)
 0.12

Unit
 3
 Porosity
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.12

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.12

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.12

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.12

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.11

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.11

Greasewood
 Root
 Shape
 Parameter
 b
 0.11

Kd
 Sand
 for
 Tc
 (mL/g)
 0.09

Tortuosity
 Water
 Content
 Exponent

 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.09

Tortuosity
 Porosity
 Exponent
 0.08

Soil
 Ingestion
 Tracer
 Element
 0.03

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 122
Table 25: Benson Erosion Peak Groundwater Well Concentrations within 500 years –Tc99
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 Tc
 (mL/g)
 42.71

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 15.80

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 13.92

VG_n_Benson
 5.42

Porosity_Benson
 1.79

Saturated
 Zone
 Water
 Table
 Gradient
 1.60

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.73

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.59

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.50

alpha_Benson
 0.42

Deep
 Time
 DCF
 Photon
 1
 REF
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.32

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.29

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.28

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.26

Unit
 3
 Porosity
 0.25

Kd
 Sand
 for
 Ra
 (mL/g)
 0.25

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.24

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.23

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.22

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.20

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.19

Kd
 Sand
 for
 U
 (mL/g)
 0.18

Meat
 Post-­‐Cooking
 Loss
 0.18

Kd
 Clay
 for
 Ac
 (mL/g)
 0.18

Ks_Benson
 0.17

Antelope
 Range
 Area
 (acre)
 0.17

OHV
 Dust
 Adjustment
 0.17

Kd
 Sand
 for
 Th
 (mL/g)
 0.16

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.15

Deep
 Time
 Receptor
 Area
 (ac)
 0.13

Tortuosity
 Porosity
 Exponent
 0.13

Tortuosity
 Water
 Content
 Exponent

 0.13

Kd
 Sand
 for
 Cs
 (mL/g)
 0.13

Kd
 Clay
 for
 Th
 (mL/g)
 0.13

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.13

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.13

Deep
 Time
 Aeolian
 Correlation
 0.13

Grass
 Root
 Shape
 Parameter
 b
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.12

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 123
DCF
 Photon2
 REF
 0.12

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.12

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.12

Unit
 2
 Porosity
 0.12

Plant.Soil
 Conc
 Ratio
 for
 I
 0.12

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.12

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.12

Deep
 Time
 DCF
 Beta
 REF
 0.11

Kd
 Silt
 for
 Pa
 (mL/g)
 0.11

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.11

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.11

Kd
 Silt
 for
 Am
 (mL/g)
 0.11

Kd
 Sand
 for
 Pb
 (mL/g)
 0.11

Deep
 Time
 Diffusion
 Length
 (m)
 0.11

Tree
 Root.Shoot
 Ratio
 0.11

Kd
 Silt
 for
 Np
 (mL/g)
 0.11

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.10

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.10

Kd
 Sand
 for
 Am
 (mL/g)
 0.10

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.10

Kd
 Clay
 for
 Np
 (mL/g)
 0.10

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.10

Soil
 Temperature
 (°C)
 0.10

Kd
 Sand
 for
 Sr
 (mL/g)
 0.10

Kd
 Sand
 for
 Pa
 (mL/g)
 0.10

Kd
 Sand
 for
 Pu
 (mL/g)
 0.10

Kd
 Clay
 for
 U
 (mL/g)
 0.09

Greasewood
 Root
 Shape
 Parameter
 b
 0.09

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.09

Deep
 Time
 Lake
 Start
 (yr)
 0.09

Site
 Dispersal
 Area
 (km2)
 0.09

Ant
 Nest
 Shape
 Parameter
 b
 0.09

Surface
 Atmosphere
 Thickness
 (m)
 0.09

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.09

Kd
 Silt
 for
 Cs
 (mL/g)
 0.09

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.09

Deep
 Time
 DCF
 Alpha
 REF
 0.09

Kd
 Silt
 for
 Ra
 (mL/g)
 0.09

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.09

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.09

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.08

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 124
Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.08

Resuspended
 Particle
 Fraction
 0.08

Fine
 CobbleMix
 Porosity
 0.08

Kd
 Sand
 for
 Ac
 (mL/g)
 0.08

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.08

Kd
 Silt
 for
 Pu
 (mL/g)
 0.08

RipRap
 Porosity
 0.08

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.08

Meat
 Preparation
 Loss
 0.08

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.08

RipRap
 Bulk
 Density
 (g/cm3)
 0.08

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.08

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.08

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.07

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.07

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.07

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.07

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.07

Plant
 Fresh
 Weight
 Conversion
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.07

Kd
 Clay
 for
 Ra
 (mL/g)
 0.07

Greasewood
 Root.Shoot
 Ratio
 0.07

Shrub
 Root.Shoot
 Ratio
 0.07

Kd
 Clay
 for
 Pa
 (mL/g)
 0.07

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.07

DCF
 Photon1
 REF
 0.07

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.07

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.07

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.07

Saturated
 Zone
 Thickness
 (m)
 0.06

Silt
 Sand
 Gravel
 Porosity
 0.06

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.06

Kd
 Clay
 for
 Pu
 (mL/g)
 0.06

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.06

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.06

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.06

Kd
 Clay
 for
 Am
 (mL/g)
 0.06

Fine
 Gravel
 Mix
 Porosity
 0.06

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.06

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.06

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 125
Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.06

Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.06

Unit
 4
 Compacted
 Porosity
 0.06

Kd
 Silt
 for
 Sr
 (mL/g)
 0.06

Kd
 Silt
 for
 U
 (mL/g)
 0.06

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.06

Ant
 Nest
 Volume
 (m3)
 0.06

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.06

Deep
 Lake
 Depth
 (m)
 0.06

Ant
 Colony
 Lifespan
 (yr)
 0.06

Contaminated
 Fraction
 of
 GDP
 DU
 0.06

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.05

Shrub
 Root
 Shape
 Parameter
 b
 0.05

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.05

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.05

Surface
 Wind
 Speed
 (m/s)
 0.05

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.05

Kd
 Silt
 for
 Pb
 (mL/g)
 0.05

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.05

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.05

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.05

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.05

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.05

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.05

Unit
 3
 Residual
 Water
 Content
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.05

Kd
 Sand
 for
 Np
 (mL/g)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.05

Kd
 Clay
 for
 Sr
 (mL/g)
 0.05

Grass
 Root.Shoot
 Ratio
 0.05

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.05

Kd
 Clay
 for
 Pb
 (mL/g)
 0.04

Biomass
 %
 Cover
 Selector
 0.04

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.04

Tree
 Root
 Shape
 Parameter
 b
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.04

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.04

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 126
Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.04

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.04

Unit
 4
 ET
 Layers
 Porosity
 0.04

Intermediate
 Lake
 Depth
 (m)
 0.04

Mammal
 Burrow
 Shape
 Parameter
 b
 0.04

DCF
 Beta
 REF
 0.04

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.04

Vegetation
 Association
 Selector
 0.04

DCF
 Alpha
 REF
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.04

Kd
 Sand
 for
 I
 (mL/g)
 0.04

Unit
 4
 Compacted
 Hb
 (cm)
 0.04

Random
 Gully
 Selector
 0.04

Deep
 Time
 DCF
 Photon
 2
 REF
 0.04

Kd
 Silt
 for
 Th
 (mL/g)
 0.04

Kd
 Clay
 for
 Cs
 (mL/g)
 0.04

Kd
 Silt
 for
 Ac
 (mL/g)
 0.04

Forb
 Root
 Shape
 Parameter
 b
 0.04

Receptor
 Area
 (ha)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 U
 0.04

Body
 Weight
 Factor
 for
 Antelope
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.04

Forb
 Root.Shoot
 Ratio
 0.03

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.03

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.03

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.03

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.03

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.03

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.03

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.02

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 127
Table 26: Benson Erosion Peak Dose within 10,000 years – Rancher
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 39.33

Kd
 Sand
 for
 Ra
 (mL/g)
 3.81

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.03

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.87

Kd
 Silt
 for
 Th
 (mL/g)
 0.84

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.81

Grass
 Root.Shoot
 Ratio
 0.72

Soil
 Temperature
 (°C)
 0.66

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.62

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.58

Kd
 Clay
 for
 Ra
 (mL/g)
 0.58

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.54

Receptor
 Area
 (ha)
 0.54

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.54

Porosity_Benson
 0.54

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.53

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.51

Kd
 Silt
 for
 Cs
 (mL/g)
 0.51

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.50

Ant
 Nest
 Shape
 Parameter
 b
 0.50

Deep
 Time
 DCF
 Photon
 2
 REF
 0.48

Plant.Soil
 Conc
 Ratio
 for
 U
 0.47

Kd
 Sand
 for
 Cs
 (mL/g)
 0.46

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.46

Kd
 Sand
 for
 Am
 (mL/g)
 0.45

Shrub
 Root.Shoot
 Ratio
 0.44

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.44

Kd
 Silt
 for
 Np
 (mL/g)
 0.43

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.43

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.43

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.43

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.41

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.41

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.40

Silt
 Sand
 Gravel
 Porosity
 0.40

Fine
 CobbleMix
 Porosity
 0.40

Deep
 Time
 Diffusion
 Length
 (m)
 0.40

DCF
 Photon2
 REF
 0.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.39

Kd
 Clay
 for
 U
 (mL/g)
 0.39

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 128
Liner
 Clay
 Saturated
 Hyd
 Cond
 (cm/s)
 0.39

Biomass
 %
 Cover
 Selector
 0.38

Kd
 Silt
 for
 Pu
 (mL/g)
 0.37

Unit
 4
 Compacted
 Hb
 (cm)
 0.37

Kd
 Sand
 for
 U
 (mL/g)
 0.37

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.36

OHV
 Dust
 Adjustment
 0.36

Kd
 Silt
 for
 Ra
 (mL/g)
 0.36

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.35

Body
 Weight
 Factor
 for
 Antelope
 0.35

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.35

Ant
 Nest
 Volume
 (m3)
 0.35

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.35

Forb
 Root.Shoot
 Ratio
 0.35

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.35

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.35

Meat
 Preparation
 Loss
 0.34

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.34

Deep
 Time
 Lake
 Start
 (yr)
 0.34

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.34

Unit
 2
 Porosity
 0.33

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.33

Plant
 Fresh
 Weight
 Conversion
 0.33

Antelope
 Range
 Area
 (acre)
 0.33

Intermediate
 Lake
 Depth
 (m)
 0.32

Grass
 Root
 Shape
 Parameter
 b
 0.32

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.32

Kd
 Clay
 for
 Pa
 (mL/g)
 0.32

Kd
 Sand
 for
 I
 (mL/g)
 0.32

Contaminated
 Fraction
 of
 GDP
 DU
 0.31

Kd
 Silt
 for
 Pb
 (mL/g)
 0.31

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.31

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.31

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.30

Unit
 4
 Compacted
 Porosity
 0.30

Kd
 Silt
 for
 U
 (mL/g)
 0.30

Kd
 Clay
 for
 Cs
 (mL/g)
 0.30

Kd
 Silt
 for
 Sr
 (mL/g)
 0.30

RipRap
 Bulk
 Density
 (g/cm3)
 0.30

Kd
 Clay
 for
 Pu
 (mL/g)
 0.29

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.29

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.29

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.29

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 129
Deep
 Time
 DCF
 Alpha
 REF
 0.28

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.28

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.28

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.28

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.28

Kd
 Clay
 for
 Ac
 (mL/g)
 0.27

Greasewood
 Root.Shoot
 Ratio
 0.27

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.27

Ant
 Colony
 Lifespan
 (yr)
 0.27

Deep
 Time
 Aeolian
 Correlation
 0.27

Deep
 Time
 DCF
 Beta
 REF
 0.26

Kd
 Silt
 for
 Pa
 (mL/g)
 0.26

Shrub
 Root
 Shape
 Parameter
 b
 0.25

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.25

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.25

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.25

Unit
 3
 Residual
 Water
 Content
 0.25

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.25

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.24

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.24

Kd
 Sand
 for
 Pu
 (mL/g)
 0.24

Kd
 Silt
 for
 Ac
 (mL/g)
 0.24

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.24

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.24

Surface
 Atmosphere
 Thickness
 (m)
 0.24

Tree
 Root
 Shape
 Parameter
 b
 0.24

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.24

Meat
 Post-­‐Cooking
 Loss
 0.24

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.24

Plant.Soil
 Conc
 Ratio
 for
 I
 0.24

Kd
 Clay
 for
 Th
 (mL/g)
 0.24

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.23

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.23

Kd
 Clay
 for
 Np
 (mL/g)
 0.23

DCF
 Beta
 REF
 0.23

Unit
 4
 ET
 Layers
 Porosity
 0.23

Forb
 Root
 Shape
 Parameter
 b
 0.23

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.22

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.22

Saturated
 Zone
 Thickness
 (m)
 0.22

Deep
 Lake
 Depth
 (m)
 0.22

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 130
Kd
 Silt
 for
 Am
 (mL/g)
 0.22

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.22

DCF
 Photon1
 REF
 0.22

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.22

Random
 Gully
 Selector
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.21

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.21

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.21

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.21

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.20

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.20

Site
 Dispersal
 Area
 (km2)
 0.20

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.19

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.19

Kd
 Sand
 for
 Np
 (mL/g)
 0.19

Kd
 Sand
 for
 Th
 (mL/g)
 0.19

Ks_Benson
 0.19

Fine
 Gravel
 Mix
 Porosity
 0.19

Resuspended
 Particle
 Fraction
 0.19

Kd
 Sand
 for
 Pb
 (mL/g)
 0.18

Deep
 Time
 DCF
 Photon
 1
 REF
 0.18

alpha_Benson
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.18

DCF
 Alpha
 REF
 0.18

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.18

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.18

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.18

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.18

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.18

Kd
 Sand
 for
 Sr
 (mL/g)
 0.18

Mammal
 Burrow
 Shape
 Parameter
 b
 0.17

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.17

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.17

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.17

Tree
 Root.Shoot
 Ratio
 0.17

Kd
 Clay
 for
 Sr
 (mL/g)
 0.17

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.17

Vegetation
 Association
 Selector
 0.16

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.16

Saturated
 Zone
 Water
 Table
 Gradient
 0.16

RipRap
 Porosity
 0.16

VG_n_Benson
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 131
Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.16

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.16

Kd
 Clay
 for
 Am
 (mL/g)
 0.16

Surface
 Wind
 Speed
 (m/s)
 0.16

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.16

Unit
 3
 Porosity
 0.15

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.15

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.15

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.14

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.14

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.14

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.14

Kd
 Sand
 for
 Ac
 (mL/g)
 0.14

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.13

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.13

Deep
 Time
 Receptor
 Area
 (ac)
 0.13

Kd
 Clay
 for
 Pb
 (mL/g)
 0.12

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.12

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.12

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.12

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.12

Kd
 Sand
 for
 Pa
 (mL/g)
 0.11

Tortuosity
 Porosity
 Exponent
 0.10

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.10

Greasewood
 Root
 Shape
 Parameter
 b
 0.09

Tortuosity
 Water
 Content
 Exponent

 0.09

Kd
 Sand
 for
 Tc
 (mL/g)
 0.08

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 132
Table 27: Benson Clay Liner Peak Groundwater Well Concentrations within 500 years –
Tc99
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Kd
 Sand
 for
 Tc
 (mL/g)
 42.60

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 15.65

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 13.33

X.Added.VG_n_Benson
 5.29

X.Added.Porosity_Benson
 2.06

Saturated
 Zone
 Water
 Table
 Gradient
 1.59

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.74

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.71

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.50

X.Added.alpha_Benson
 0.41

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.41

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.36

Deep
 Time
 DCF
 Photon
 1
 REF
 0.36

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.28

Unit
 3
 Porosity
 0.27

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.26

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.26

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.25

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.22

Kd
 Sand
 for
 Ra
 (mL/g)
 0.22

Meat
 Post-­‐Cooking
 Loss
 0.22

Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.19

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.19

Antelope
 Range
 Area
 (acre)
 0.18

Kd
 Sand
 for
 U
 (mL/g)
 0.17

OHV
 Dust
 Adjustment
 0.16

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.16

Deep
 Time
 Receptor
 Area
 (ac)
 0.16

Kd
 Sand
 for
 Th
 (mL/g)
 0.15

Kd
 Clay
 for
 Ac
 (mL/g)
 0.13

X.Added.Ks_Benson
 0.13

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.13

Kd
 Silt
 for
 Pa
 (mL/g)
 0.13

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.12

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.12

Site
 Dispersal
 Area
 (km2)
 0.12

RipRap
 Porosity
 0.12

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.12

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 133
Grass
 Root
 Shape
 Parameter
 b
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.12

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.12

Unit
 2
 Porosity
 0.12

Kd
 Sand
 for
 Pa
 (mL/g)
 0.11

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.11

Deep
 Time
 Aeolian
 Correlation
 0.11

Kd
 Sand
 for
 Cs
 (mL/g)
 0.11

Kd
 Silt
 for
 Np
 (mL/g)
 0.11

Kd
 Clay
 for
 Th
 (mL/g)
 0.11

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.11

Tortuosity
 Water
 Content
 Exponent

 0.11

Tree
 Root.Shoot
 Ratio
 0.11

Deep
 Time
 Lake
 Start
 (yr)
 0.11

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.11

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.11

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.10

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.10

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.10

Kd
 Sand
 for
 Sr
 (mL/g)
 0.10

DCF
 Photon2
 REF
 0.10

Tortuosity
 Porosity
 Exponent
 0.10

Deep
 Time
 DCF
 Beta
 REF
 0.10

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.10

Kd
 Clay
 for
 Ra
 (mL/g)
 0.10

Kd
 Silt
 for
 Cs
 (mL/g)
 0.10

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.10

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.09

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.09

Silt
 Sand
 Gravel
 Porosity
 0.09

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.09

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.09

RipRap
 Bulk
 Density
 (g/cm3)
 0.09

Kd
 Clay
 for
 U
 (mL/g)
 0.09

Plant.Soil
 Conc
 Ratio
 for
 I
 0.09

Shrub
 Root.Shoot
 Ratio
 0.09

Kd
 Sand
 for
 Pu
 (mL/g)
 0.09

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.09

Kd
 Silt
 for
 Ra
 (mL/g)
 0.09

Kd
 Sand
 for
 Pb
 (mL/g)
 0.09

Meat
 Preparation
 Loss
 0.09

Surface
 Atmosphere
 Thickness
 (m)
 0.09

Kd
 Sand
 for
 Am
 (mL/g)
 0.09

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 134
Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.08

Kd
 Sand
 for
 Np
 (mL/g)
 0.08

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.08

Greasewood
 Root
 Shape
 Parameter
 b
 0.08

Kd
 Sand
 for
 Ac
 (mL/g)
 0.08

Kd
 Clay
 for
 Pu
 (mL/g)
 0.08

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 0.08

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.08

Kd
 Clay
 for
 Np
 (mL/g)
 0.08

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.08

Unit
 4
 Compacted
 Porosity
 0.08

Kd
 Silt
 for
 Sr
 (mL/g)
 0.08

Plant
 Fresh
 Weight
 Conversion
 0.08

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.08

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.08

Kd
 Clay
 for
 Pa
 (mL/g)
 0.08

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.08

DCF
 Beta
 REF
 0.07

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.07

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.07

Greasewood
 Root.Shoot
 Ratio
 0.07

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.07

Kd
 Silt
 for
 Am
 (mL/g)
 0.07

Deep
 Time
 DCF
 Alpha
 REF
 0.07

Unit
 4
 ET
 Layers
 Porosity
 0.07

Fine
 CobbleMix
 Porosity
 0.07

Soil
 Temperature
 (°C)
 0.07

Resuspended
 Particle
 Fraction
 0.07

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.07

Vegetation
 Association
 Selector
 0.07

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.07

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.07

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.07

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.07

Kd
 Silt
 for
 Pu
 (mL/g)
 0.07

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.07

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.07

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.07

Ant
 Nest
 Shape
 Parameter
 b
 0.07

Fine
 Gravel
 Mix
 Porosity
 0.07

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.07

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 135
Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.07

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.07

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.07

DCF
 Photon1
 REF
 0.07

Kd
 Silt
 for
 U
 (mL/g)
 0.06

Deep
 Lake
 Depth
 (m)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.06

Kd
 Clay
 for
 Sr
 (mL/g)
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.06

Ant
 Colony
 Lifespan
 (yr)
 0.06

Contaminated
 Fraction
 of
 GDP
 DU
 0.06

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.06

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.06

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.06

Kd
 Clay
 for
 Am
 (mL/g)
 0.06

Mammal
 Burrow
 Shape
 Parameter
 b
 0.06

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.06

Radon
 Escape.Production
 Ratio
 for
 Waste
 0.06

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.06

Deep
 Time
 Diffusion
 Length
 (m)
 0.06

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.06

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.06

DCF
 Alpha
 REF
 0.06

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.06

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.06

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.05

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.05

Shrub
 Root
 Shape
 Parameter
 b
 0.05

Kd
 Silt
 for
 Pb
 (mL/g)
 0.05

Saturated
 Zone
 Thickness
 (m)
 0.05

Surface
 Wind
 Speed
 (m/s)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 Ra
 0.05

Ant
 Nest
 Volume
 (m3)
 0.05

Plant.Soil
 Conc
 Ratio
 for
 U
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.05

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.05

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.05

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.05

Forb
 Root.Shoot
 Ratio
 0.05

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 136
GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.05

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.05

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.05

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.05

Grass
 Root.Shoot
 Ratio
 0.05

Kd
 Silt
 for
 Th
 (mL/g)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.04

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.04

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.04

Kd
 Clay
 for
 Pb
 (mL/g)
 0.04

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.04

Biomass
 %
 Cover
 Selector
 0.04

Deep
 Time
 DCF
 Photon
 2
 REF
 0.04

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Unit
 4
 Compacted
 Hb
 (cm)
 0.04

Unit
 3
 Residual
 Water
 Content
 0.04

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.04

Tree
 Root
 Shape
 Parameter
 b
 0.04

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.04

Kd
 Clay
 for
 Cs
 (mL/g)
 0.04

Body
 Weight
 Factor
 for
 Antelope
 0.04

Forb
 Root
 Shape
 Parameter
 b
 0.04

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.04

Kd
 Sand
 for
 I
 (mL/g)
 0.04

Intermediate
 Lake
 Depth
 (m)
 0.03

Receptor
 Area
 (ha)
 0.03

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.03

Kd
 Silt
 for
 Ac
 (mL/g)
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.03

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.02

Random
 Gully
 Selector
 0.02

Soil
 Ingestion
 Tracer
 Element
 0.02

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 137
Table 28: Benson Clay Liner Peak Dose within 10,000 years – Rancher
R-squared = 99%
Explanatory
 Variable
 Sensitivity
 Index

Radon
 Escape.Production
 Ratio
 for
 Waste
 37.67

Kd
 Sand
 for
 Ra
 (mL/g)
 3.57

Plant.Soil
 Conc
 Ratio
 for
 Ra
 1.18

Saltwater
 Solubility
 for
 Ra
 (mol/L)
 1.09

X.Added.Porosity_Benson
 0.75

Kd
 Silt
 for
 Th
 (mL/g)
 0.74

Beef
 Transfer
 Factor
 for
 Pu
 (day/kg)
 0.73

Grass
 Root.Shoot
 Ratio
 0.73

X.Added.Ks_Benson
 0.72

X.Added.VG_n_Benson
 0.72

Soil
 Temperature
 (°C)
 0.66

Ant
 Colony
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.64

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U234
 (pCi/g)
 0.61

Kd
 Sand
 for
 Am
 (mL/g)
 0.61

Saltwater
 Solubility
 for
 Pb
 (mol/L)
 0.58

Kd
 Clay
 for
 Ra
 (mL/g)
 0.58

Deep
 Time
 DCF
 Photon
 2
 REF
 0.56

Natural
 Rn
 Barrier
 Clay
 Sat
 Hyd
 Cond
 (cm/s)
 0.55

Receptor
 Area
 (ha)
 0.53

Ant
 Colony
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.53

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 n
 0.53

Mammal
 Mound
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.52

Kd
 Silt
 for
 Np
 (mL/g)
 0.52

Kd
 Sand
 for
 Cs
 (mL/g)
 0.51

Saltwater
 Solubility
 for
 Rn
 (mol/L)
 0.50

Beef
 Transfer
 Factor
 for
 I
 (day/kg)
 0.50

Kd
 Clay
 for
 U
 (mL/g)
 0.49

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U238
 (pCi/g)
 0.46

Plant.Soil
 Conc
 Ratio
 for
 U
 0.46

Silt
 Sand
 Gravel
 Porosity
 0.45

Kd
 Silt
 for
 Cs
 (mL/g)
 0.45

Unit
 4
 Compacted
 Hb
 (cm)
 0.44

Fine
 CobbleMix
 Porosity
 0.43

Deep
 Time
 Diffusion
 Length
 (m)
 0.43

Ant
 Nest
 Volume
 (m3)
 0.43

Ant
 Nest
 Shape
 Parameter
 b
 0.43

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu238
 (pCi/g)
 0.42

Soil
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.41

Beef
 Transfer
 Factor
 for
 U
 (day/kg)
 0.41

Silt
 Sand
 Gravel
 BulkDensity
 (g/cm3)
 0.41

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 138
DCF
 Photon2
 REF
 0.40

Antelope
 Range
 Area
 (acre)
 0.40

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U233
 (pCi/g)
 0.39

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Tc99
 (pCi/g)
 0.39

Shrub
 Root.Shoot
 Ratio
 0.38

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Cs137
 (pCi/g)
 0.38

Body
 Weight
 Factor
 for
 Antelope
 0.37

OHV
 Dust
 Adjustment
 0.37

Deep
 Time
 Intermediate
 Lake
 Duration
 (yr)
 0.37

Kd
 Silt
 for
 Pu
 (mL/g)
 0.36

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu241
 (pCi/g)
 0.35

Plant.Soil
 Conc
 Ratio
 for
 Pb
 0.35

Forage
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.34

Saltwater
 Solubility
 for
 I
 (mol/L)
 0.34

Biomass
 %
 Cover
 Selector
 0.34

Contaminated
 Fraction
 of
 GDP
 DU
 0.34

Kd
 Sand
 for
 U
 (mL/g)
 0.34

Beef
 Transfer
 Factor
 for
 Sr
 (day/kg)
 0.34

X.Added.alpha_Benson
 0.33

Plant.Soil
 Conc
 Ratio
 for
 Am
 0.33

Kd
 Silt
 for
 Ra
 (mL/g)
 0.33

Kd
 Silt
 for
 Pb
 (mL/g)
 0.32

Deep
 Time
 Deep
 Lake
 Sedimentation
 Rate
 (m/yr)
 0.32

Plant.Soil
 Conc
 Ratio
 for
 Tc
 0.32

Unit
 2
 Porosity
 0.32

Plant
 Fresh
 Weight
 Conversion
 0.32

Deep
 Time
 Lake
 Start
 (yr)
 0.32

Forb
 Root.Shoot
 Ratio
 0.31

Deep
 Time
 Aeolian
 Correlation
 0.31

Kd
 Silt
 for
 U
 (mL/g)
 0.31

Kd
 Clay
 for
 Pa
 (mL/g)
 0.30

Grass
 Root
 Shape
 Parameter
 b
 0.30

Unit
 4
 Compacted
 Residual
 Water
 Content
 0.30

Deep
 Time
 Aeolian
 Deposition
 Depth
 (m)
 0.29

Meat
 Preparation
 Loss
 0.29

Ant
 Colony
 Lifespan
 (yr)
 0.29

Beef
 Transfer
 Factor
 for
 Pb
 (day/kg)
 0.29

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U235
 (pCi/g)
 0.28

Unit
 3
 Bubbling
 Pressure
 Head
 (cm)
 0.28

Kd
 Clay
 for
 Pu
 (mL/g)
 0.28

Kd
 Sand
 for
 I
 (mL/g)
 0.28

Saltwater
 Solubility
 for
 UO3
 (mol/L)
 0.28

Plant.Soil
 Conc
 Ratio
 for
 Cs
 0.27

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 139
Fine
 Cobble
 Mix
 BulkDensity
 (g/cm3)
 0.27

Unit
 3
 Saturated
 Hyd
 Cond
 (cm/s)
 0.27

Ant
 Colony
 Density
 -­‐
 Plot
 3
 (1/ha)
 0.27

RipRap
 Bulk
 Density
 (g/cm3)
 0.27

Kd
 Clay
 for
 Ac
 (mL/g)
 0.27

Ant
 Colony
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.27

Unit
 4
 ET
 Layers
 Bulk
 Density
 (g/cm3)
 0.27

Biomass
 Production
 Rate
 (kg.ha.yr)
 0.27

Plant.Soil
 Conc
 Ratio
 for
 Sr
 0.27

Deep
 Time
 DCF
 Alpha
 REF
 0.27

Kd
 Clay
 for
 Cs
 (mL/g)
 0.27

Intermediate
 Lake
 Depth
 (m)
 0.27

Plant.Soil
 Conc
 Ratio
 for
 I
 0.26

Unit
 4
 Compacted
 Porosity
 0.26

Surface
 Atmosphere
 Diffusion
 Length
 (m)
 0.26

Kd
 Silt
 for
 Sr
 (mL/g)
 0.26

Saltwater
 Solubility
 for
 Pa
 (mol/L)
 0.26

Unit
 4
 Compacted
 Bulk
 Density
 (g/cm3)
 0.26

Deep
 Time
 DCF
 Beta
 REF
 0.26

Plant.Soil
 Conc
 Ratio
 for
 Np
 0.26

Shrub
 Root
 Shape
 Parameter
 b
 0.26

Unit
 3
 Residual
 Water
 Content
 0.25

Greasewood
 Root.Shoot
 Ratio
 0.25

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu240
 (pCi/g)
 0.25

Unit
 2
 Bulk
 Density
 (g/cm3)
 0.25

Fine
 Gravel
 Mix
 BulkDensity
 (g/cm3)
 0.25

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Pu239
 (pCi/g)
 0.25

Plant.Soil
 Conc
 Ratio
 for
 Pa
 0.25

Molecular
 Diffusivity
 in
 Water
 (cm2/s)
 0.24

Unit
 3
 Bulk
 Density
 (g/cm3)
 0.24

Saltwater
 Solubility
 for
 U3O8
 (mol/L)
 0.24

Water
 Ingestion
 Rate
 for
 Cattle
 (kg/day)
 0.24

Saturated
 Zone
 Thickness
 (m)
 0.23

Kd
 Silt
 for
 Pa
 (mL/g)
 0.23

DCF
 Photon1
 REF
 0.23

Saltwater
 Solubility
 for
 Th
 (mol/L)
 0.23

Kd
 Silt
 for
 Ac
 (mL/g)
 0.23

Fine
 Gravel
 Mix
 Porosity
 0.23

Beef
 Transfer
 Factor
 for
 Tc
 (day/kg)
 0.23

Resuspension
 Flux
 (kg.m2-­‐yr)
 0.23

Kd
 Silt
 for
 Am
 (mL/g)
 0.23

Beef
 Transfer
 Factor
 for
 Pa
 (day/kg)
 0.23

Beef
 Transfer
 Factor
 for
 Ra
 (day/kg)
 0.23

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 140
Kd
 Sand
 for
 Pu
 (mL/g)
 0.22

Meat
 Post-­‐Cooking
 Loss
 0.22

Random
 Gully
 Selector
 0.22

Site
 Dispersal
 Area
 (km2)
 0.22

Unit
 4
 ET
 Layers
 Porosity
 0.21

Beef
 Transfer
 Factor
 for
 Np
 (day/kg)
 0.21

Kd
 Clay
 for
 Np
 (mL/g)
 0.21

Plant.Soil
 Conc
 Ratio
 for
 Pu
 0.21

Kd
 Sand
 for
 Sr
 (mL/g)
 0.21

Activity
 Conc
 in
 SRS
 DU
 Waste:
 I129
 (pCi/g)
 0.21

Forb
 Root
 Shape
 Parameter
 b
 0.21

Kd
 Sand
 for
 Np
 (mL/g)
 0.21

Saltwater
 Solubility
 for
 Pu
 (mol/L)
 0.20

Kd
 Sand
 for
 Th
 (mL/g)
 0.20

DCF
 Beta
 REF
 0.20

Deep
 Lake
 Depth
 (m)
 0.20

Ant
 Colony
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.20

Tree
 Root
 Shape
 Parameter
 b
 0.20

Mammal
 Mound
 Density
 -­‐
 Plot
 4
 (1/ha)
 0.20

Kd
 Clay
 for
 Th
 (mL/g)
 0.20

Plant.Soil
 Conc
 Ratio
 for
 Ac
 0.20

Soil
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.19

Kd
 Sand
 for
 Pb
 (mL/g)
 0.19

Surface
 Atmosphere
 Thickness
 (m)
 0.19

Plant.Soil
 Conc
 Ratio
 for
 Th
 0.19

Mammal
 Mound
 Density
 -­‐
 Plot
 5
 (1/ha)
 0.19

Beef
 Transfer
 Factor
 for
 Th
 (day/kg)
 0.19

Deep
 Time
 Aeolian
 Deposition
 Age
 (yr)
 0.18

Beef
 Transfer
 Factor
 for
 Am
 (day/kg)
 0.18

Saltwater
 Solubility
 for
 Ac
 (mol/L)
 0.18

Vegetation
 Association
 Selector
 0.18

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Np237
 (pCi/g)
 0.17

Mammal
 Burrow
 Shape
 Parameter
 b
 0.17

Kd
 Clay
 for
 Am
 (mL/g)
 0.17

Saltwater
 Solubility
 for
 Cs
 (mol/L)
 0.17

Tree
 Root.Shoot
 Ratio
 0.17

Unit
 4
 ET
 Layers
 log
 of
 van
 Genuchten’s
 α
 0.17

Resuspended
 Particle
 Fraction
 0.17

Mammal
 Mound
 Density
 -­‐
 Plot
 2
 (1/ha)
 0.16

Mammal
 Mound
 Density
 -­‐
 Plot
 1
 (1/ha)
 0.16

Saltwater
 Solubility
 for
 Am
 (mol/L)
 0.16

Surface
 Wind
 Speed
 (m/s)
 0.16

Activity
 Conc
 in
 SRS
 DU
 Waste:
 U236
 (pCi/g)
 0.16

Sensitivity Analysis Results for the Clive DU PA
5 November 2015 141
RipRap
 Porosity
 0.16

Saltwater
 Solubility
 for
 Tc
 (mol/L)
 0.16

Saltwater
 Solubility
 for
 Sr
 (mol/L)
 0.15

DCF
 Alpha
 REF
 0.15

Kd
 Clay
 for
 Pb
 (mL/g)
 0.15

Kd
 Clay
 for
 Sr
 (mL/g)
 0.15

Federal
 DU
 Cell
 Unsaturated
 Zone
 Thickness
 (m)
 0.15

Saturated
 Zone
 Water
 Table
 Gradient
 0.15

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Sr90
 (pCi/g)
 0.15

GDP
 DU
 Inventory
 Storage
 Dead
 Space
 (m2)
 0.15

Kd
 Sand
 for
 Ac
 (mL/g)
 0.15

Beef
 Transfer
 Factor
 for
 Cs
 (day/kg)
 0.14

Beef
 Transfer
 Factor
 for
 Ac
 (day/kg)
 0.14

Deep
 Time
 DCF
 Photon
 1
 REF
 0.13

Deep
 Time
 Deep
 Lake
 End
 (yr)
 0.13

Water
 Ingestion
 Rate
 for
 Antelope
 (kg/day)
 0.13

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Ra226
 (pCi/g)
 0.13

Kd
 Sand
 for
 Pa
 (mL/g)
 0.13

Unit
 2
 Saturated
 Hyd
 Cond
 (cm/s)
 0.13

Intermediate
 Lake
 Sed
 Thickness
 (m)
 0.12

Saltwater
 Solubility
 for
 Np
 (mol/L)
 0.12

Deep
 Time
 Receptor
 Area
 (ac)
 0.12

Activity
 Conc
 in
 SRS
 DU
 Waste:
 Am241
 (pCi/g)
 0.11

Unit
 3
 Porosity
 0.11

Mammal
 Burrow
 Excavation
 Rate
 (m3/yr)
 0.10

Greasewood
 Root
 Shape
 Parameter
 b
 0.10

Tortuosity
 Water
 Content
 Exponent

 0.09

Tortuosity
 Porosity
 Exponent
 0.09

Kd
 Sand
 for
 Tc
 (mL/g)
 0.08

Unit
 3
 Brooks-­‐Corey
 Fractal
 Dimension
 0.08

Soil
 Ingestion
 Tracer
 Element
 0.03

 NAC-0059_R0

Model Comparisons

Clive DU PA Model v1.4
November 11, 2015

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544

Model Comparisons
November 11, 2015 ii

1. Title: Model Comparisons
2. Filename: Model comparison (Appendix 20).docx
3. Description: This document, Appendix 20 to Clive Model Final Report; describes model changes from v1.0 to the
present model version.
Name Date
4. Originator Gregg Occhiogrosso 10 November 2015
5. Reviewer Kate Catlett 11 November 2015
6. Remarks

Model Comparisons
November 11, 2015 iii
CONTENTS
1.0
 Introduction ............................................................................................................................ 1

2.0
 Model Updates ....................................................................................................................... 1

3.0
 Results Comparison ............................................................................................................... 1

3.1
 Groundwater Concentration and Rancher Dose ............................................................... 1

3.2
 Deep Time ........................................................................................................................ 3

Model Comparisons
November 11, 2015 iv
Tables
Table 1. Comparison or groundwater concentrations and receptor dose results for PA model
iterations. ....................................................................................................................... 2

Table 2. Deep time peak lake and sediment U-238 concentrations within 100,000 years for
v1.0 and v1.2. ................................................................................................................ 3

Table 3. Deep time lake and sediment U-238 concentrations at model year 90,000 for v1.4. ....... 3

Model Comparisons
November 11, 2015November 11, 2015 1
1.0 Introduction
The Clive DU PA Model has undergone several revisions in the development and review process.
Three major versions have been released: v1.0, v1.2, and v1.4. This document highlights major
changes made in these models and compares results.
2.0 Model Updates
There are three main versions of the Clive DU PA Model: Clive DU PA Model v1.0 (v1.0), Clive
DU PA Model v1.2 (v1.2), and Clive DU PA Model v1.4 (v1.4). Several major updates occurred
in each new version.
The differences in results between v1.0 and v1.2 are primarily from evaluating the DU waste
disposed below grade instead of above grade. However, the DU waste was still dispersed in the
deep time model as a consequence of intermediate lake return. Other changes include differences
in cover design from a riprap cover to an evapo-transpirative cover, a change in approach to
erosion calculations using a landscape evolution model (SIBERIA), and a few other model
changes, such as changes in the tortuosity exponent distributions.
In addition to various minor changes, model v1.4 includes several notable updates, including:
• Substantial revision to the deep time model that no longer disperses below grade waste
based upon return of an intermediate lake, accounting for eolian deposition until a lake
returns (based on recently collected field data), and addressing lake dispersal area and
diffusion into the lake based on lake dynamics.
• Updated embankment geometry per the latest design drawings (see Appendix 3,
Embankment Modeling for the Clive DU PA white paper)
• Updated infiltration modeling based on changes to hydraulic properties in the cover
system, resulting in lower net infiltration (see Appendix 5, Unsaturated Zone Modeling
for the Clive DU PA white paper)
3.0 Results Comparison
Results from the Clive DU PA Model v1.4 are compared to results from the Clive DU PA
Models v1.2 and 1.0 below. For comparison purposes, the Clive DU PA Model v1.0 was updated
(at the time of v1.2 release) into GoldSim version 10.5 service pack 4 (sp4) and was rerun for
10,000 realizations with the waste configured as it is in the current model: all the waste buried
below grade. Gullies were allowed to form in all models and were included in receptor scenarios.
The Clive DU PA Model v1.4 was updated into GoldSim version 11.1.2.
3.1 Groundwater Concentration and Rancher Dose
Peak groundwater concentrations of 99Tc are shown for model comparison purposes in Table 1,
along with rancher dose and total population dose.
Model Comparisons
November 11, 2015November 11, 2015 2
Table 1. Comparison or groundwater concentrations and receptor dose results for PA
model iterations.

Mean Median 95th Percentile
v1.0 v1.2 v1.4 v1.0 v1.2 v1.4 v1.0 v1.2 v1.4
Peak 99Tc
groundwater
concentration within
500 yr (GWPL =
3790 pCi/L) (pCi/L)
3.4E4 7.4E2 2.6E1 2.0E3 2.0E1 4.3E-2 1.8E5 4.5E3 1.5E2
Peak rancher dose
within 10,000 yr
(mrem/yr)
3.9E-3 1.6E-2 6.2E-2 2.9E-3 1.4E-2 5.1E-2 9.4E-3 3.7E-2 1.5E-1
Total population dose
within 10,000 yr
(rem)
3.5E-1 1.6E0 1.2E1 2.8E3 1.4E0 1.1E1 8.7E3 3.3E0 2.6E1

The 99Tc groundwater summary addresses the peak concentrations for the 500-yr period, for
which the peak occurs at 500 yr. Hence, these are summary statistics are 500 yr. Similarly the
rancher does and population dose summaries represents the doses at 10,000 yrs.
Concentrations of 99Tc in v1.2 and v1.4 are much lower than those in v1.0, primarily because of
the reduction in the infiltration rate with the new ET cover and potentially the narrowing of the
tortuosity coefficient distributions. In v1.0 and v1.2, 99Tc concentrations exceed the groundwater
protection limits (GWPLs) in the 95th percentile, and the mean for v1.0 also exceeds the GWPL
for 99Tc. Note that the median did not exceed the GWPL in any model. In the current version of
the model, v1.4, the mean, median, and 95th percentile values are much lower than the GWPLs
for all radionuclides of concern, including 99Tc as shown in Table 1. Reductions in groundwater
concentrations are tied to lower infiltration rates which resulted from less conservatism in
unsaturated zone modeling (see the Unsaturated Zone Modeling for the Clive PA white paper,
Appendix 5). This is despite the waste being placed below grade (i.e., closer to groundwater) in
the v1.2 and v1.4 models. The infiltration rate is much lower, which mitigates the effect of lower
waste placement.
Doses to the ranch worker receptor are also compared for Clive DU PA Model v1.0, v1.2, and
v1.4 in Table 1. Doses for Models v1.2 and v1.4 are slightly higher than for v1.0, indicating that
model revisions for infiltration and tortuosity resulted in increased rates of radon migration to the
ground surface. Total population doses (used as the basis for ALARA cost calculations in the
Final Report) also increased due to increased radon flux at the surface. However, the doses from
all three model iterations are much less than the performance objectives, and the total ALARA
population doses are very small.
Specifically in the infiltration model, the single value saturated hydraulic conductivities of the
radon barriers were replaced by statistical distributions developed from the range of hydraulic
conductivities recommended by Benson et al. (2011). In addition, distributions were developed
for van Genuchten hydraulic parameters alpha and n for both the surface and evaporative zone
Model Comparisons
November 11, 2015November 11, 2015 3
layers of the ET cover. Thus, net infiltration was modeled using a wide range of hydraulic input
parameters. As a result, on average, volumetric water contents in the radon barriers are generally
drier in Model v1.2 and v1.4 than in Model v1.0. So there is more air-filled porosity available for
the radon to diffuse through. Looking at the sensitivity analyses for v1.2 and v1.4, the most
sensitive parameter to dose endpoints is the radon escape/production ratio. This suggests that
radon is the primary driver for dose. These sensitivity analyses results indicate that the
characteristics of the radon barriers have an influence on radon flux.
3.2 Deep Time
Deep time lake and sediment concentrations for the three model releases are summarized in Table
2 and Table 3. Note that the results for v1.4 are based on a specific model timestep of 90,000
years, because that timestep coincides with a greater chance of a deep lake being present at the
site. This is considered a more representative value for the lake and sediment concentrations,
because it is corresponds to a time at which a large lake is most likely to occur according to
10,000 realizations of the model. It avoids the occasional high values that drive the
concentrations reported for Model V1.0 and Model v1.2. Nevertheless, the sediment
concentrations are much smaller in Model v1.4, and the lake water concentrations are smaller.
Lake and sediment concentrations decreased considerably in each successive new model version.

Table 2. Deep time peak lake and sediment U-238 concentrations within 100,000 years for
v1.0 and v1.2.

Mean Median 95th Percentile
v1.0 v1.2 v1.0 v1.2 v1.0 v1.2
Peak lake water uranium-238
concentration within 100,000
yr (pCi/L)
5.2E-1 1.2E-1 2.0E-3 6.6E-4 2.5E1 7.9E-1
Peak sediment uranium-238
concentration within 100,000
yr (pCi/g)
1.5E3 7.7E2 1.2E3 5.2E2 3.5E3 2.6E3

Table 3. Deep time lake and sediment U-238 concentrations at model year 90,000 for v1.4.
25th Percentile Median Mean 95th Percentile
U-238 lake
concentration (pCi/L) 1.4E-7 2.1E-5 1.8E-2 1.1E-1
U-238 sediment
concentration (pCi/g) 1.7E-4 1.8E-3 2.0E-2 9.5E-2

Model Comparisons
November 11, 2015November 11, 2015 4
In Model v1.2 the DU waste was still dispersed upon intermediate lake return. The primary
difference between Model v1.0 and v1.2 was related to changes in the way different versions of
GoldSim handled time steps. Since the peak concentrations across time are the reported results,
time steps impact when the peaks occur, and affect mixing (averaging) in a time period. The
differences are somewhat artificial in this sense.
The lake water and sediment concentrations are orders of magnitude lower in v1.4 of the model.
This is largely because the DU waste is not dispersed upon return of an intermediate lake, the
dispersal area is increased based on site-specific data, diffusion takes into account both arid and
moist periods, and eolian deposition is included in the modeling process. The sediment
concentrations in the deep time model are much less than background concentrations of U238 in
the area (average about 2 pCi/g). (Note that sediment concentrations for Ra226 are more similar to
background because of ingrowth – see Page 9 of the Main Report.) This model allows diffusion
of the waste into the lake, rather than covering the waste, although some covering the waste is
likely during inter-glacial periods. Also, note that the current inter-glacial is expected to last for
many tens of not several hundreds of thousands of year because of the concentration of carbon-
dioxide in the atmosphere. The longer the inter-glacial period, the more covering of the waste is
likely from eolian deposition, rather than diffusion-based mixing that is assumed under more
moist conditions.
Note also that radon flux at the time of the first lake recession was calculated in the Deep Time
Supplemental Analysis model and in v1.4. For reference, deep time rancher doses were
calculated based on the radon flux results. These results are presented in the Main Report but are
not presented here because previous model versions did not make these calculations.

Safe
25 Nov
Prepare
NEPTUN
1435 Garri
y Eva
mber 201
d for Ener
AND CO
on St, Suite 1
luatio
5
ySolutio
MPANY, INC.
0, Lakewood
n Rep
ns by

CO 80215
ort R
NAC-0053
espon
_R0
se
Safety Evaluation Report Response
25 November 2015 ii
1. Title: Safety Evaluation Report Response
2. Filename: SER Response.docx
3. Description: Response to Utah Dept. of Environmental Quality, Division of Waste
Management and Radiation Control, Safety Evaluation Report (SC&A 2015).
Name Date
4. Originator Mike Sully 15 September 2015
5. Reviewer Kate Catlett, Paul Black 25 November 2015
6. Remarks

Safety Evaluation Report Response
25 November 2015 iii
CONTENTS
CONTENTS ................................................................................................................................... iii
FIGURES ....................................................................................................................................... iv
TABLES ..........................................................................................................................................v
ACRONYMS AND ABBREVIATIONS ...................................................................................... vi
1.0 Introduction .............................................................................................................................1
2.0 Evapotranspiration and Infiltration .........................................................................................1
2.1 Methods .............................................................................................................................2
2.2 Surface Boundary Conditions ...........................................................................................2
2.3 Hydraulic Properties ..........................................................................................................3
2.4 HYDRUS Simulation Results ...........................................................................................5
2.4.1 Water Balance Results .................................................................................................5
2.5 Regression Model Development .......................................................................................7
2.5.1 Exploratory Data Plots ................................................................................................7
2.5.2 Linear Regression Models ...........................................................................................7
2.6 Implementation in GoldSim ..............................................................................................8
2.7 Results ...............................................................................................................................8
2.8 Sensitivity Analysis of GoldSim v1.4XXX Benson .......................................................10
2.9 Discussion .......................................................................................................................10
3.0 GoldSim Quality Assurance — Comparison with GoldSim Results ...................................13
4.0 Frost Damage ........................................................................................................................14
5.0 Effect of Biotic Activity .......................................................................................................14
6.0 Erosion ..................................................................................................................................15
6.1 Influence of Cover Erosion on Net Infiltration ...............................................................16
6.2 Influence of Cover Erosion on Contaminant Transport and Receptor Dose ..................17
6.2.1 Implementation in GoldSim ......................................................................................17
6.2.2 Results .......................................................................................................................18
6.2.3 Sensitivity Analysis ...................................................................................................18
6.2.4 Discussion .................................................................................................................18
7.0 Clay Liner .............................................................................................................................19
7.1 GoldSim Implementation ................................................................................................20
7.2 Results .............................................................................................................................20
7.3 Sensitivity Analysis of v.1.4XXX Benson Clay Liner ...................................................20
7.4 Discussion .......................................................................................................................21
8.0 Deep Time .............................................................................................................................21
8.1 GoldSim Implementation ................................................................................................22
8.2 Results .............................................................................................................................22
8.3 Discussion .......................................................................................................................23
8.3.1 Eolian deposition standard error ................................................................................23
8.3.2 Intermediate lake sedimentation rates .......................................................................24
9.0 References .............................................................................................................................26
Appendix A HYDRUS Simulation Results ................................................................................ A-1
Appendix B Flow Model Development Plots ............................................................................. A-4

Safety Evaluation Report Response
25 November 2015 iv
FIGURES
Figure 1. Model layers, root density, and observation nodes for naturalized H1D model. .............3
Figure 2. Net infiltration rate estimated by HYDRUS-1D for 50 combinations of hydraulic
properties generated using the method described by Benson in Appendix E,
Volume 2, of SC&A (2015). ..........................................................................................5
Figure 3. Water balance components for naturalized cover simulations in order of increasing
net infiltration.................................................................................................................6
Figure 4. Eolian silt in trench located at Clive Pit 29 overlying Lake Bonneville sedimentary
deposits (Neptune 2015). .............................................................................................12
Figure 5. An example of upper soil-modified eolian silt in Pit 29. Basal contact of the silt is
approximately located at the middle of the pick handle. It is a gradational contact
between eolian silt intermixed with regressive Lake Bonneville marl (bottom of
the pick handle). ...........................................................................................................13
Figure 6. Comparison of 1,000 realizations of net infiltration using the linear model in
GoldSim with the results of the 50 HYDRUS simulations for the naturalized
cover. ............................................................................................................................14
Figure 7. Relationship between model hydraulic parameters and modeled volumetric water
content in the upper 6 inches of the cover. ............................................................... A-4
Figure 8. Relationship between model hydraulic parameters and modeled volumetric water
content from 6 inches to 18 inches deep in the cover. .............................................. A-5
Figure 9. Relationship between model hydraulic parameters and modeled volumetric water
content from 18 inches to 36 inches deep in the cover. ............................................ A-6
Figure 10. Relationship between model hydraulic parameters and modeled volumetric water
content from 36 inches to 48 inches deep in the cover. ............................................ A-7
Figure 11. Relationship between model hydraulic parameters and modeled volumetric water
content from 48 inches to 60 inches deep in the cover. ............................................ A-8
Figure 12. Relationship between model hydraulic parameters and modeled net infiltration at
the the top of the waste. ............................................................................................ A-9
Figure 13. HYDRUS volumetric water contents plotted with linear model values for the
surface through the upper radon barrier of the cover. ............................................. A-10
Figure 14. HYDRUS volumetric water content for the lower radon barrier and net infiltration
into the top of the waste plotted with linear model values. .................................... A-11

Safety Evaluation Report Response
25 November 2015 v
TABLES
Table 1. Recommended mean values and standard deviations for hydraulic parameters from
Appendix E of SC&A (2015). .......................................................................................4
Table 2. Hydraulic parameter sets generated using “Hyd Props Calculator.xls” as described in
Appendix E of SC&A (2015). .......................................................................................4
Table 3. Water balance components for five of the 50 homogeneous cover hydraulic property
simulations. ....................................................................................................................7
Table 4. Fitted model coefficients. ..................................................................................................8
Table 5. Groundwater and ranch dose results for v1.4XXX Benson compared to v1.4. .................9
Table 6. Comparison of deep time results at model year 90,000 for v1.4XXX Benson with
v1.4. All results based on 1000 realizations. .................................................................9
Table 7. Sensitive input parameters for v1.4XXX Benson. ...........................................................10
Table 8. Comparison of net infiltration for eroded and non-eroded cases, for three sets of
hydraulic properties. ....................................................................................................17
Table 9. Model results for v1.4XXX Benson Erosion. ..................................................................18
Table 10. Sensitive input parameters for v1.4XXX Benson Erosion. ...........................................19
Table 11. Model results for v1.4XXX Benson Clay Liner. ...........................................................20
Table 12. Sensitive input parameters for v1.4XXX Benson Clay Liner. ......................................21
Table 13. Comparison of deep time results at 90,000 yr for v1.4XXX Benson Deep Time and
v1.4XXX Benson models. ...........................................................................................23
Table 14. Thickness measurements from field studies of eolian silt near Clive ............................24
Table 15. Water content and infiltration results from 50 HYDRUS simulations using
naturalized (homogenous) hydraulic properties. ....................................................... A-1

Safety Evaluation Report Response
25 November 2015 vi
ACRONYMS AND ABBREVIATIONS
DEQ Utah Department of Environmental Quality
DRC Division of Radiation Control
DU depleted uranium
DWMRC Division of Waste Management and Radiation Control
ES EnergySolutions
NRC U.S. Nuclear Regulatory Commission
PA performance assessment
SER safety evaluation report

Safety Evaluation Report Response
25 November 2015 1
1.0 Introduction
Based on its review of Round 3 Interrogatories, the Utah Department of Environmental Quality
(DEQ) had additional questions regarding the performance of the evapotranspiration (ET) cover
system and deep time modeling as part of the Clive Depleted Uranium (DU) Performance
Assessment (PA) Model (the Clive DU PA Model) constructed by Neptune and Company, Inc.
(Neptune). These concerns were discussed with EnergySolutions (ES) and, on August 11, 2014,
DEQ submitted additional interrogatories for ES to address. DEQ also requested that ES conduct
some additional bounding calculations with HYDRUS to provide greater transparency as to how
the percolation model performed. ES’ replies are documented in its August 18, 2014, “Responses
to August 11, 2014 – Supplemental Interrogatories Utah LLRW Disposal License RML UT
2300249 Condition 35 Compliance Report”.
DEQ reviewed the responses in ES (2014) and determined that the information provided was not
sufficient to resolve the supplemental interrogatories. Their review is documented in the Safety
Evaluation Report (SER) (SC&A 2015, Volume 2, Appendix B). In general, DEQ decided that
there needs to be much more description of how the analysis proceeded from the input data to the
results. Appendix B of SC&A (2015, Volume 2) includes specific examples from the ES
response where DEQ believes that additional information and explanations are necessary.
This document provides the additional information and explanation requested by DEQ. Note that
DEQ and DRC (Division of Radiation Control) are used interchangeably within this document.
In addition, in July 2015, the DRC was merged with another division and renamed the Division
of Waste Management and Radiation Control (DWMRC). Furthermore, the SER was prepared
by SC&A, Inc., so references to the SER are cited as “SC&A, 2015”.
2.0 Evapotranspiration and Infiltration
This section provides an alternative net infiltration and volumetric water content model for the
Clive DU PA Model v1.4 that represents a cover system with different hydraulic properties and
clearly correlates the hydraulic parameters alpha and hydraulic conductivity. This alterative
model is named the “v1.4XXX Benson” model. The flow models used were developed to
account for changes in hydraulic properties due to plant and animal activity and frost action that
might affect the net infiltration rate and water content status based on the conceptual model of
cover “naturalization” described in the work of Benson et al. (2011) and in Appendix E of SC&A
(2015), Volume 2. This approach to modeling of flow takes into account changes in hydraulic
properties due to biological activity and freeze/thaw cycles predicted by the conceptual model of
cover naturalization.
For this set of unsaturated zone flow models, the cover system is considered to be entirely
homogeneous with respect to hydraulic properties other than a minor adjustment to a parameter
(gravel adjustment) for the surface layer. Input parameters for these infiltration models are
derived from the distributions and methods described by Dr. Craig Benson in Volume 2,
Appendix E, of the safety evaluation report (SER) prepared by SC&A (SC&A 2015), consistent
with the request of DEQ to use this approach (SC&A 2015).
Safety Evaluation Report Response
25 November 2015 2
These models represent modifications to previous models required in response to the SER issues.
These models are conservative and do not represent the likely evolution of the cover system
Differences between the homogenous cap and the Clive DU PA Model v1.4 conceptual models
are described in the discussion following the model results.
2.1 Methods
The evapotranspiration (ET) cover design, unsaturated zone and shallow aquifer characteristics,
climate, and vegetation are described in detail in Appendix 5 of the Final Report for the Clive
DU PA Model v1.4. The infiltration models include 50 HYDRUS-1D (H1D) simulations using
homogeneous properties (except for the gravel correction in the surface layer) and the method for
developing hydraulic property values provided in Appendix E of the SER (SC&A 2015). H1D
was used to estimate water contents with depth in the ET cover, and to estimate average annual
drainage out of the bottom of the cover into the waste zone (net infiltration). Simulation
durations were 1,000 years. Mean water contents and infiltration rates for each parameter set
were calculated from the last 100 years of the 1,000-year simulations.
For the homogenized cap model, the following changes were made from previous ET cover
simulations:
• Deeper rooting depth (1.5 m), with constant root density throughout the 1.5-m cover
profile
• Homogenous hydraulic properties (except for gravel correction of saturated water content
in surface layer).
Figure 1 illustrates the homogeneous material distribution, constant root density, and locations of
observation nodes used for the naturalized H1D model.
2.2 Surface Boundary Conditions
The WGEN model (Richardson and Wright 1984) was used to generate a 100-year synthetic
daily record of precipitation, maximum temperature, and minimum temperature for the site. Use
of the WGEN model, a component of the HELP model (Schroeder et al. 1994a; Schroeder et al.
1994b), is consistent with U.S. NRC guidance (Meyer et al. 1996). The 100-year record was
generated using the monthly average values from measurements at the site based on 17 years of
observations. Simulations were run for 1,000 years repeating the 100-year daily boundary
conditions. The model is deliberately run for a long period of time (1,000 years) in order to reach
a near-steady state net infiltration rate that is not influenced by the initial conditions. Long-term
variations in climate are addressed in the Deep Time Model.

Safety E
25 Nove
Figure 1
2.3
Hydraul
Benson i
“Hyd Pr
and satu
paramet
0.48 pro
chosen f
below, t
addition
sets use
generate
layer.

valuation Rep
ber 2015
. Model la
Hydraulic
ic property
n Appen ix
ps Calculat
ated hydra
rs, the gene
ided in SC
om the “Lo
keep the i
l upscaling
for the mo
value of t
ort Respons
yers, root d
Properti
values for th
E, Volume
or.xls,” 50
lic conducti
ated Ks an
A (2015).
” column
put parame
of the para
els are sho
e saturated
e
ensit , and
es
ese simulati
, of SC&A
ombination
ity (Ks) w
alpha valu
ecommen
f Table 2 i
ers within t
eter values
n in Table
ater conte
observation
ons were ge
(2015). Us
of the alp
re generate
s were corr
ed standar
Appendix
e ranges re
was done f
below. Th
t corrected
nodes for
nerated usin
ng the EX
a, n, saturat
. For the 5
elated usin
deviations
(SC&A 2
ommended
r these par
e parameter
for the addi
naturalized
g the metho
EL spreads
d water co
combinati
a correlati
or the four
15), show
in Benson
meter sets.
theta_s* co
ion of grav
H1D mode
d described
eet provide
tent (theta
ns of
n coefficie
arameters
in Table 1
t al. (2011)
he 50 para
responds t
l in the sur
3

l.
by
d,
s),
t of
ere
No
eter
the
ace
Safety Evaluation Report Response
25 November 2015 4
Table 1. Recommended mean values and standard deviations for hydraulic parameters
from Appendix E of SC&A (2015).
Parameter Base Units Mean Standard
Deviation
lnKs m/s -14.51 0.59
lnalpha 1/kPa -1.609 0.12
n - 1.3 0.04
theta_s - 0.4 0.013

Table 2. Hydraulic parameter sets generated using “Hyd Props Calculator.xls” as
described in Appendix E of SC&A (2015).
Realization theta_s* theta_s alpha (1/cm) n Ks (cm/d)
1 0.331 0.389 0.0255 1.24 15.49
2 0.358 0.421 0.0184 1.29 12.08
3 0.323 0.380 0.0216 1.39 9.77
4 0.345 0.406 0.0230 1.37 10.09
5 0.339 0.399 0.0195 1.31 4.42
6 0.347 0.409 0.0209 1.32 2.38
7 0.352 0.414 0.0177 1.28 2.11
8 0.349 0.411 0.0254 1.29 3.64
9 0.333 0.392 0.0191 1.36 6.00
10 0.341 0.401 0.0203 1.39 3.45
11 0.345 0.405 0.0175 1.31 3.22
12 0.343 0.403 0.0218 1.32 4.89
13 0.336 0.396 0.0187 1.29 2.48
14 0.345 0.405 0.0241 1.30 7.31
15 0.348 0.410 0.0201 1.24 6.70
16 0.346 0.407 0.0149 1.27 1.59
17 0.330 0.388 0.0185 1.32 3.55
18 0.339 0.399 0.0184 1.33 1.10
19 0.342 0.402 0.0204 1.29 12.83
20 0.340 0.401 0.0212 1.30 2.85
21 0.361 0.425 0.0179 1.20 4.76
22 0.342 0.402 0.0244 1.27 4.15
23 0.345 0.406 0.0193 1.34 3.89
24 0.335 0.394 0.0205 1.33 3.94
25 0.325 0.382 0.0171 1.30 1.39
26 0.336 0.396 0.0220 1.30 5.59
27 0.340 0.400 0.0190 1.25 5.22
28 0.338 0.398 0.0245 1.28 9.18
29 0.353 0.416 0.0173 1.33 2.76
30 0.340 0.400 0.0152 1.19 2.22
31 0.343 0.403 0.0183 1.31 1.94
32 0.332 0.390 0.0195 1.31 2.00
33 0.308 0.362 0.0199 1.27 5.75
34 0.334 0.393 0.0214 1.28 3.31
35 0.337 0.397 0.0183 1.35 3.08
36 0.336 0.395 0.0241 1.23 4.71
37 0.349 0.411 0.0197 1.31 8.69
38 0.325 0.382 0.0206 1.31 4.47
39 0.331 0.389 0.0178 1.27 4.20
40 0.328 0.386 0.0222 1.28 6.48
Safety Evaluation Report Response
25 November 2015 5
Realization theta_s* theta_s alpha (1/cm) n Ks (cm/d)
41 0.355 0.418 0.0192 1.22 3.41
42 0.355 0.418 0.0227 1.33 3.46
43 0.341 0.401 0.0172 1.33 3.50
44 0.335 0.394 0.0241 1.29 11.49
45 0.343 0.403 0.0186 1.30 4.96
46 0.346 0.407 0.0213 1.24 4.65
47 0.345 0.405 0.0200 1.27 7.47
48 0.327 0.384 0.0194 1.32 2.33
49 0.337 0.397 0.0178 1.33 6.93
50 0.340 0.401 0.0209 1.29 6.38

2.4 HYDRUS Simulation Results
Net infiltration for each of the 50 replicate input parameter sets is plotted in Figure 2 below. Net
infiltration ranged from 0.57 mm/yr to 1.31 mm/yr with a mean value of 0.91 mm/yr. Results for
volumetric water content and net infiltration for each realization are provided in Appendix A.

Figure 2. Net infiltration rate estimated by HYDRUS-1D for 50 combinations of hydraulic
properties generated using the method described by Benson in Appendix E, Volume
2, of SC&A (2015).
2.4.1 Water Balance Results
Five of the 50 naturalized cover simulations were selected for summarizing the water balances.
The five selected are realizations #41, 7, 25, 20, and 24 of the 50 simulations, which correspond
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 10 20 30 40 50
Ne
t

i
n
f
i
l
t
r
a
t
i
o
n

(
m
m
/
y
r
)
Realization
naturalized Cover
Safety E
25 Nove
to the 10
for the
of the w

Figure
in

Table 3
Total w
0.27 per
Changes
also neg
alance:

valuation Rep
ber 2015
th, 30th, 50th,
ater balanc
ter balance
. Water ba
reasin ne
summarizes
ter balance
ent of the t
in storage
igible. The
precipitatio
ort Respons
70th and 90
analyses r
for the five
lance comp
infiltratio
the water b
rrors for th
tal annual
re zero whe
ater balan
, evaporati
e
th percentile
nged from
imulations
onents for
.
alance comp
five simul
recipitation
averaged
e plots in F
n, transpir
infiltration
.711 mm/y
are shown i
naturalized
onents from
tions are o

over the last
gure 3 sho
tion, and ne
alues, resp
to 1.038 m
Figure 3
cover simu
the five si
average ab
100 years o
the remai
infiltratio
ctively. Ne
/yr. The
elow.
lations in o
mulations in
out 0.56 m
f the simula
ing compo
.
infiltratio
ajor compo
rder of
more detail.
/yr or
ions. Runo
ents of wat
6
n rates
ents
f is
r
Safety Evaluation Report Response
25 November 2015 7

Table 3. Water balance components for five of the 50 homogeneous cover hydraulic
property simulations.
Replicate 41 Replicate 7 Replicate 25 Replicate 20 Replicate 24
Water
Balance
Component
mm/yr % of
Precip
mm/y
r
% of
Precip mm/yr % of
Precip mm/yr % of
Precip mm/yr % of
Precip
Precipitation 212.3 212.0 212.3 212.0 212.1
Evaporation 210.6 99.20 209.7 98.92 210.1 98.96 209.1 98.63 209.4 98.73
Transpiration 0.389 0.18 0.982 0.46 0.625 0.29 1.444 0.68 1.127 0.53
Net
Infiltration 0.711 0.34 0.848 0.40 0.865 0.41 0.955 0.45 1.038 0.49
Runoff 1.56
E-04 0.00
1.37
E-04 0.00
1.36
E-04 0.00
1.02
E-04 0.00
9.75
E-05 0.00
Storage 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Total 211.7 99.72 211.5 99.78 211.6 99.67 211.5 99.76 211.6 99.75
Mass
balance error 0.599 0.28 0.470 0.22 0.710 0.33 0.501 0.24 0.535 0.25

2.5 Regression Model Development
2.5.1 Exploratory Data Plots
Exploratory scatter plots for each depth zone showed generally linear relationships between the
covariates alpha, n, Ks, and theta_s and the response variables water content and net infiltration
flux model results. The saturated water content of the surface layer (theta_s*) was not included
as a parameter in the linear regression since it is derived from adjusting the value of theta_s.
These relationships are shown in Figure 7 through Figure 12 in Appendix B for zones in the
naturalized cover corresponding to depths of the surface (WC1), evaporative (WC2), frost
protection (WC3), upper radon barrier (WC4), and lower radon barrier (WC5) layers for water
content and the net infiltration at the top of the waste.

2.5.2 Linear Regression Models
Multiple linear regression models were fit to the HYDRUS simulations for each layer. The
general form of the regression was:
= + ∗ + ∗ ℎ + ∗ + ∗ ℎ _
Net infiltration was converted to units of cm/day and volumetric water content was
dimensionless. The regressions were fit using the “lm()” function in the software package R,
which uses least squares for estimating parameters.
The statistics underlying linear regression assume that the coefficients are distributed as normal,
so the coefficient estimates and their associated standard error estimates represent the mean and
Safety Evaluation Report Response
25 November 2015 8
standard deviation from a normal distribution. Regression coefficients are shown in Table 4
below.
Table 4. Fitted model coefficients.
Response
Surface WC 0.37326 -0.00309 -0.19961 -0.26633 0.32691
Evap WC 0.45616 -0.00365 -0.27057 -0.32052 0.39271
Frost WC 0.47409 -0.00341 -0.38131 -0.33119 0.38654
Rn1 WC 0.48466 -0.00325 -0.45964 -0.33817 0.38318
Rn2 WC 0.48888 -0.00319 -0.49211 -0.34102 0.38190
Flux -0.00029 -3.5389E-6 0.00574 0.00065 -0.00100

2.6 Implementation in GoldSim
The following changes are made to the Clive Model v1.4; the resulting model iteration is referred
to as v1.4XXX Benson.
Using Table 4, the resulting equations for infiltration and water content in GoldSim become:
= + ∗ + ∗ ℎ + ∗ + ∗ ℎ _

= ,+ ,∗ + ,∗ ℎ + ,∗ + ,∗ ℎ _
where Infil is net infiltration in cm/day, WC is the average volumetric water content, and β
values are the linear regression coefficients with the subscript i corresponding to Surface,
Evaporative, Frost protection, Upper radon barrier, and Lower radon barrier depth zones listed in
Table 4.
After the water content is calculated, GoldSim expression elements are used to enforce physical
bounds of water content as the residual water content (if the water content is less than the
residual water content) and as the porosity (if water content is greater than porosity).
The input parameters Ks, alpha, n, and theta_s for each realization are obtained from a lookup
table of 1000 realizations generated using the method described by Benson in Appendix E,
Volume 2 of SC&A (2015). A lookup table is used for the inputs rather than stochastic elements
in GoldSim to force a correlation between ln alpha and ln Ks since GoldSim does not include a
multivariate normal distribution for representing correlation.
2.7 Results
Results of this simulation are compared to those of the Clive DU PA Model v1.4 in Table 5 and
Table 6.
Safety Evaluation Report Response
25 November 2015 9
Groundwater concentrations of Tc-99 and Rancher doses are compared in Table 5. The greater
infiltration of the homogenized cap leads to higher groundwater concentrations. The Tc-99
median concentration is below the groundwater protection limit (GWPL) of 3790 pCi/L, while
the mean and 95th percentile results exceed the GWPL. Rancher doses are slightly lower in the
v1.4XXX Benson model because the increased infiltration suppresses upward radon flux.
Table 5. Groundwater and ranch dose results for v1.4XXX Benson compared to v1.4.

Mean Median 95th Percentile
v1.4* v1.4XXX
Benson v1.4* v1.4XXX
Benson v1.4* v1.4XXX
Benson
Peak Tc-99 groundwater
concentration within 500
yr (pCi/L)
2.6E1 7.6E3 4.3E-2 3.0E2 1.5E2 4.1E4
Peak rancher dose within
10,000 yr (mrem/yr) 6.2E-2 5.1E-2 5.1E-2 4.5E-2 1.5E-1 1.2E-1
* v1.4 results are based on 10,000 realizations, while other results in this table are based on 1,000 realizations.

Deep time results are compared in Table 6. The homogenized cap model produces lower lake
and sediment concentrations because increased infiltration suppresses upward diffusion of
radionuclides in the model.
Table 6. Comparison of deep time results at model year 90,000 for v1.4XXX Benson with
v1.4. All results based on 1000 realizations.
25th Percentile Median Mean 95th Percentile
v1.4 v1.4XXX
Benson v1.4 v1.4XXX
Benson v1.4 v1.4XXX
Benson v1.4 v1.4XXX
Benson
U-238 lake
concentration
(pCi/L) 1.4E-7 1.4E-7 2.1E-5 1.3E-5 1.8E-2 3.9E-3 1.1E-1 1.5E-2
Ra-226 lake
concentration
(pCi/L) 8.5E-3 2.0E-4 1.5E-1 9.4E-3 5.4E-1 6.2E-2 2.4E0 3.0E-1
U-238
sediment
concentration
(pCi/g)
1.7E-4 1.3E-7 1.8E-3 5.3E-6 2.0E-2 2.4E-4 9.5E-2 1.1E-3
Ra-226
sediment
concentration
(pCi/g)
6.9E-5 1.7E-6 1.2E-3 7.1E-5 5.0E-3 5.8E-4 2.2E-2 2.9E-3

Safety Evaluation Report Response
25 November 2015 10
2.8 Sensitivity Analysis of GoldSim v1.4XXX Benson
A sensitivity analysis of the 99Tc groundwater concentrations with 500 years and rancher doses
within 10,000 years was performed in order to determine which modeling parameters are most
significant in predicting these results. The most sensitive parameters for these endpoints are
presented in Table 3.

The soil-water partition coefficient (Kd) was the most sensitive parameter for the groundwater
concentration of 99Tc. Kd controls sorption to the solid phase, with higher Kd resulting in
increased sorption which retards migration of the radionuclides. In model version 1.4, the most
sensitive parameter for groundwater concentrations of 99Tc was van Genuchten’s α, which is
involved in the water content and infiltration regression equations. In v1.4XXX Benson, the
homogenized cover leads to much high infiltration rates, and the model is thus not as sensitive to
the regression inputs compared to v1.4.
The most sensitive input parameter for rancher dose is the radon E/P ratio, which defines the
fraction of 222Rn that escapes into the mobile environment when formed by radioactive decay
from its parent, 226Ra. Radon that does not escape but remains within the matrix of the radium-
containing waste material stays in place and decays to polonium and then to 210Pb. Note that the
higher the E/P ratio, the higher the dose.
Table 7. Sensitive input parameters for v1.4XXX Benson.
SI rank Input parameter Sensitivity
index (SI)
Peak 99Tc groundwater
concentration within 500
years
1 Kd for Tc 43
2 Activity Concentration of Tc-99 in SRS DU
Waste
16
3 Molecular Diffusivity in Water 14
4 Van Genuchten’s n 5
Peak rancher dose within
10,000 years
1 Radon Escape/Production Ratio for Waste 38
2 Kd for Ra in sand 3.61
1 For technetium, the same Kd value was used for all materials.

2.9 Discussion
The hydraulic property recommendations and cover material naturalization present in Benson et
al. (2011) and in Appendix E (SC&A, 2015) are inappropriate for the Clive site. When included
in the model, they produce a model that does not make sense for the site conditions of Clive.
This model can be considered “conservative” in terms of modeling groundwater concentrations
Safety Evaluation Report Response
25 November 2015 11
but dose results are lower for this model implementation than for the Clive DU PA Model v1.4,
which does not imply “conservative.” The rationale for not using these homogenized cap
properties in the Clive DU PA Model v1.4 are presented in this section.
The hydraulic property recommendations provided in Benson et al. (2011) are based on
measurements for samples from in-service covers made at 12 sites throughout the continental
United States. One element of the characterization of a site’s climate is the ratio of mean annual
precipitation to mean annual potential evapotranspiration. The magnitude of this ratio is
estimated to be 0.17 for Clive. Only one of the sites sampled by Benson et al. (2011) was
considered to be arid, having a ratio of 0.06. The mean value of this ratio for all sites sampled
was 0.51, with a highest value of 1.10. At two of the sites rainfall exceeded potential
evaporation, which is completely inappropriate for the arid conditions at Clive. All but one of the
sites that form the basis for the hydraulic property recommendations have much wetter
conditions than Clive.
The conceptual model of cover material “naturalization” for Clive based on the work of Benson et
al. (2011) is described in Appendix E of SC&A (2015) as including changes in the hydraulic
behavior of the material following construction. These changes are characterized by increasing
values of hydraulic properties such as Ks and the hydraulic function alpha parameter that begin
soon after cover completion. These changes are commonly attributed to pedogenic processes
including wet-dry and freeze-thaw cycles, activity of roots and soil animals, decomposition of
organic matter by microbes producing compounds that tend to bind soil particles into aggregates,
and changes in cations adsorbed onto soil particle surfaces. In this conceptual model these
processes lead to the development of soil structure but not soil horizons.
Under the wetter conditions considered by Benson et al. (2011), plant and animal activity are
greater than in an arid setting. These wetter conditions promote a faster rate of disruptive
processes due to plant and animal activity and in some cases freeze-thaw activity that were
shown by Benson et al. (2011) to lead to formation of an aggregated soil structure and natural
mixing of soil layers at their study sites. Most importantly, the sites considered by Benson et al.
(2011) also lack significant eolian deposition. This is not the case for a site like Clive. Recent
field studies (Neptune 2015) provide evidence for a site-specific conceptual model of weak
development of soil profiles (limited pedogenesis) in a setting influenced by low rates of
deposition of eolian silt in the Holocene history. The Site is within a region of significant eolian
activity evidenced by locally thick accumulation of gypsum dunes west and southwest of the site
and a laterally continuous layer of suspension fallout silts preserved beneath the modern surface
throughout the Clive site. Clive quarry exposures examined in a field study (Neptune 2015)
showed sections of eolian silts immediately below a modern vegetated surface (Figure 4). The
bottom of the eolian silt formed a gradational but definable contact with the lake muds and marl
below. The upper vegetated surface at the top of the eolian section was distinct and noted as
being partially indurated. In addition, buried soils were found in the eolian and lake sediments
below the Lake Bonneville lacustrine sequence.
The eolian deposits in the upper part of the stratigraphic section shown in Figure 4 represent a
10,000-year-old record of deposition and soil formation (Neptune 2015). Primary soil features
developed over this time interval include an indurated Av-zone, and slight reddening of the silt
profile with local platy structure from formation of clays (Figure 5). These observations are
Safety E
25 Nove
consiste
continui
eolian si
hydrauli
O serva
long-ter
deposits
Holocen
wet play
extremel
vegetati
the lake
increase
sands).
with agg
slightly
Climate
the hom
site.
Figure
se
valuation Rep
ber 2015
nt with slow
g suppressi
t. There is
properties.
ions of Hol
eolian de
which are o
deposition
sources o
low gradi
n and limit
evels re ai
as future la
ecurring la
ading clast
tabilize clo
onditions a
genized ca
4. Eolian sil
imentar
ort Respons
processes o
n and buri
o evidence
cene eolia
osition on a
ly slightly
l condition
eolian sour
nt paleo-L
d surface e
below the
es approac
es during i
c lake sedi
ure covers
d soil for
and demo
t in trench l
eposits (Ne
e
f pedogenes
l of develo
f soil struc
silt throug
stable surfa
odified by
at the Cliv
e material t
ke Bonnevi
osion. Thes
site elevati
the site wi
e ages (cli
ents. The e
ntil future l
ation proce
strate the in
ocated at C
ptune 2015
is in a high
ing soils b
ure develo
out the Cli
ce that pro
slow proce
e site are pr
the west a
le surface s
conditions
n. Rates of
h increased
ate cycles)
pectation i
akes return
ses at the C
applicabilit
live Pit 29
.
elevation se
a relativel
ment exten
e site supp
otes and pr
ses of soil
moted by
d southwe
rrounding
will persist
olian depo
formation o
ill rework
that eolian
o the Clive
live site co
of the ho
overl in L
mi-arid setti
low rate of
ive enough
rt a concept
se ves con
ormation.
combinati
t of the Cli
he site with
at the Clive
ition woul
f dunes (de
and mix th
deposits wi
site.
tradict the
ogenized c
ake Bonne
ng and
deposition
to influence
ual model o
urrent eoli
he past
n of extensi
e site and t
sparse surf
site as long
be expecte
osition of e
eolian dep
l drape and
ssumptions
p to the Cli
ville
12
of
soil

n
e
e
ce
as
to
lian
sits
of
e

Safety E
25 Nove
Figure
ap
be
pi
3.0
This sec
directly
HYDR
simulati
second s
Benson).
infiltrati
paramet
SC&A (
The HY
match cl
HYDR
Net infil
ranged f

valuation Rep
ber 2015
5. An exam
roximatel
ween eolia
k handle).
GoldSi
oldSi
tion provide
rom the H
S results. T
ns using na
t consists o
These wer
n, and inpu
r sets gener
015).
RUS and l
sely, assur
S simulatio
ration calc
om 0.57 m
ort Respons
ple of upper
located at
silt inter
m Qualit
Result
s a compari
DRUS mo
e first set c
uralized m
f 1,000 real
calculated
parameter
ted using t
near model
ng consiste
s ranged fr
lated from t
/yr to 1.29
e
soil-modif
the middle
ixed with
y Assur

son between
els and the
nsists of n
terial prope
zations of n
using the li
alues for
e method d
net infiltrat
cy in the re
m 0.57 m
e linear m
mm/yr with
ied eolian si
of the pick
e ressive
ance —
two sets of
ther from l
t infiltratio
ties and co
et infiltratio
ear model
s, alpha, n,
scribed by
on rates are
sults of thes
/yr to 1.31
del used in
a mean of
lt in Pit 29.
handle. It
ake Bonne
Compa
net infiltrati
near model
rates obtai
related val
n from the
ith coeffic
nd theta_s
Benson in
compared i
e two mode
m/yr with
the GoldSi
.93 mm/yr.
Basal cont
s a radati
ille marl (
rison w
on rates, on
abstracted
ed from th
es of ln Ks
oldSim M
ents fro
from a look
ppendix E,
Figure 6.
s. Net infilt
mean val
v.1.3 Mod
act of the si
nal contac
ottom of t
ith
e calculated
from the
50 HYDR
nd ln alpha
del (v1.4X
able 4 for n
p table of
Volume 2,
These resul
ation from
e of 0.91 m
l coefficie
13

lt is

e
US
. The
X
t
,000
f
s
he
/yr.
ts
Safety E
25 Nove
Figure
G
4.0
The infil
disrupti
paramet
describe
(SC&A
state foll
properti
completi
Changes
GoldSi
correcti
conducti
to accou
5.0
The SE
due to bi
cover sy
hydrauli
valuation Rep
ber 2015
6. Comparis
ldSim with
Frost D
tration mod
n of nea -s
rs used for
by Dr. Cr
015) that r
wing cons
s such as K
on and are
in hydrauli
v1.4XXX
n of saturat
ity values,
t for effect
Effect o
R (SC&A 20
otic activity
tems, incre
conductivi
ort Respons
on of 1,000
the results
amage
eling descri
rface layers
he infiltrati
ig Benson i
present cha
ruction. Th
and the hy
ommonly a
properties
enson Mo
d water co
and correlat
of frost da
f Biotic
15) describ
. The report
se infiltrati
y of cove -
e
realization
of the 50
bed in Secti
above and
n modelin
Volume 2,
ges in hyd
se changes
raulic func
tributed to
f the cover
el include h
tent in the s
ion of the h
age on net
Activity
es concerns
comments t
n, and hast
ystem soils
s of net infi
YDRUS si
on 2.0 takes
ithin the ra
were deriv
Appendix
aulic behav
re characte
ion alpha p
edogenic p
used to dev
omogeneou
rface layer
draulic con
infiltration
ith increas
at: “Bioint
n migratio
penetrated
ltration usi
ulations f
into accoun
don barrier
d from the
, of the SE
or of cover
rized by inc
rameter th
ocesses inc
elop the inf
hydraulic
, increased
ductivity a
ates.
d flow and
usion by pl
of contam
y roots. Th
ng the line
r the natu
t the asserte
by frost.”
istribution
prepared
materials to
easing val
t begin soo
uding frost
ltration dist
roperties (
saturated h
d alpha hyd
transport in
ant roots ca
nants by in
is can be es
ar model in
alized cove
d “substanti
ydraulic
and metho
y SC&A
a naturaliz
es of hydra
after cove
penetration.
ibutions fo
xcept for g
draulic
aulic para
the cover s
also dama
reasing the
ecially
14

r.
al
s
d
lic

the
avel
eters
stem
e
Safety Evaluation Report Response
25 November 2015 15
problematic at clay radon barriers.” The comment continues with references to a wide range of
reported rooting depths for woody plants at Clive, from a variety of studies and literature
references. The comment notes that the Utah Radioactive Material License - Condition 35 (RML
UT2300249) Compliance Report (Revision 2) of July 8, 2014, submitted with the performance
assessment, cites maximum rooting depth for woody plants at Clive Site of 1.3 to 2.3 ft. The
comment further notes that information presented in Envirocare (2000) and Hoven et al. (2000)
indicates site-specific rooting depths for greasewood of 13 ft, and that observations by DRC staff
in the past have suggested that plant roots observed in borrow pits at Clive extend 10 or more
feet below ground surface. These last two references are used to refute the shallow site rooting
depths reported in SWCA (2013) and the Compliance Report. The comment then cites Waugh
and Smith (1998) as evidence that roots can penetrate the compact clay radon barrier that occurs
at 3 ft bgs in the Clive cover. Though we could not readily obtain the Waugh and Smith (1998)
reference cited by the reviewer, the same information is presented in Waugh et al. (1999). Roots
that penetrate the radon barrier can provide preferential pathways for infiltration.
It is important to recognize how the range of rooting depths discussed in the comment actually
relates to what was used as a maximum rooting depth in GoldSim Models v1.2 and v1.4. A
maximum root depth of 5.7 meters (18.7 ft) (Robertson 1983) is used in the Model, so the Model
already assumes that roots extend beyond the radon barrier. In addition, v1.4 of the GoldSim
Model assumes increased permeability, correlation between saturated hydraulic conductivity and
the hydraulic function alpha parameter, and homogenization of the cover materials, with no
physical barriers to either plant roots or infiltration.
Greasewood has been reported to extend taproots up to ~60 ft to reach groundwater (Meinzer
1927). This is not likely to occur on the Clive disposal cells, where the distance from the top of
the cover to groundwater is greater than 65 ft. With groundwater beyond the reach of the taproot,
the functional rooting depth of greasewood will be much shallower, and the growth of the
greasewood plants will be controlled by precipitation infiltration in the upper soil horizon
(Branson et al. 1976). The zone of infiltration at the site does not extend to groundwater;
therefore, the use of a 60 ft maximum rooting depth is not warranted for the Clive GoldSim
Model. The use of a 5.7 m maximum rooting depth is appropriately conservative because it
allows for root penetration of the entire cover system, including the radon barrier, and v1.4 of the
GoldSim Model assumes a naturalized cover for purposes of modeling infiltration of the cover.
Note that this naturalized cover is different from the Benson homogenized cover described in
Section 2.0.
6.0 Erosion
The SER (SC&A 2015) describes issues with erosion modeling documentation as unclear and a
need for demonstrating the simplification of erosion modeling processes in the Clive DU PA
Model. Updates were made to Erosion Modeling for the Clive DU PA Model (Appendix 10 of
the Final Report for the Clive DU PA Model) based on clarification requested in Section 4.4.2 of
the SER (SC&A 2015). Appendix 10 includes a detailed description of the conceptual erosion
model and its implementation in the Clive DU PA Model. To assess the effects of erosion of the
cover an additional model scenario was developed constructed based on the v1.4XXX Benson
model described in Section 2.0. This model named “v1.4XXX Benson Erosion” includes
Safety Evaluation Report Response
25 November 2015 16
consideration of the receptor doses which would result from widespread erosion of the cover, as
well as changes in infiltration resulting from cover erosion, described in Sections 6.1 and 6.2.
6.1 Influence of Cover Erosion on Net Infiltration
The conceptual model of cover “naturalization” described in Appendix E of the SER (SC&A
2015) is that plant and animal activity and freeze-thaw cycles result in disturbance and mixing of
soil layers in the upper portion of the cover system subject to their influences. The extent of the
influence of these processes decreases with depth of roots, animal burrowing, and frost
penetration. This conceptual model does not maintain the designed functions of store and release
layers and barrier layers to reduce net infiltration. Using this conceptual model, the upper portion
of the soil profile subject to naturalization processes is considered to be homogeneous with
respect to the hydraulic properties affecting net infiltration. For the Clive Site, the hydraulic
properties of the waste below the cover are modeled as Unit 3 material and would be subject to
the same naturalization processes as the materials used to construct the cover.
With this conceptual model, the depth to the waste would be reduced by erosion but the net
infiltration will not vary. The net infiltration is determined by climate and hydraulic properties. If
the hydraulic properties are assumed to be homogeneous and determined by climate and biotic
activity, loss of material from the surface of the cover will not change the net infiltration.
A series of HYDRUS simulations were completed to demonstrate this concept. Input parameters
for infiltration models representing two states of erosion loss were derived from the distributions
and methods described by Dr. Craig Benson in Volume 2, Appendix E, of the safety evaluation
report (SER) prepared by SC&A (SC&A 2015). Fifty realizations of parameters were generated
and three, representing the lowest, mid-range, and highest net infiltration rates, were selected for
the models. The two eroded cases chosen from the SIBERIA Model output had a loss of 6 inches
(15.2 cm) of cover consisting of the surface layer, and a loss of 4 feet (122 cm) of cover
corresponding to the surface layer, ET layer, frost protection layer, and the upper radon barrier.
These two erosion depths were chosen based on SIBERIA Model output. SIBERIA Model
results showed that 75% of the cap area has gullies that ended at 6 inches or shallower.
Similarly, the results showed that 98% of the cap area has gullies that ended at 4 feet or
shallower. These depths are good depth representations to explore erosion behavior.
An assumption of the one-dimensional HYDRUS model is that ponding does not occur in any
channels that have been formed on the cover. Infiltration in a channel is subject to the same
surface boundary condition as non-eroded portions of the cover. Given the assumptions that the
hydraulic properties of the cover are homogeneous and that there is no focusing of infiltration in
channels, root water uptake below channels will also be the same as in the cover outside the
channel, as there will be no variation in material properties that would affect root extension or
moisture distribution. All cases used 1,000-year durations for the simulations, approaching
steady state. Net infiltration rates were calculated as the mean of daily simulated values from the
last 100 years of the 1,000-year simulations. These results are compared with net infiltration
rates from previous simulations of non-eroded covers using the same uniform hydraulic
properties. For a given set of hydraulic properties, net infiltration rates are independent of the
cover thickness. Small differences between the eroded and non-eroded cases are due to
numerical grid differences in the non-eroded models.
Safety Evaluation Report Response
25 November 2015 17
Table 8. Comparison of net infiltration for eroded and non-eroded cases, for three sets of
hydraulic properties.
Erosion Depth
(cm)
Net Infiltration
from Hydraulic
Properties Set
Net Infiltration
(mm/yr)
0 High 1.1
15.2 High 1.02
122 High 1.02

0 Mid-Range 0.77
15.2 Mid-Range 0.73
122 Mid-Range 0.73

0 Low 0.47
15.2 Low 0.44
122 Low 0.44

With this conceptual model of soil naturalization and the representation of waste as Unit 3
material, as soil is lost due to erosion, disturbance due to biotic activity and freeze-thaw extend
to maintain the same thickness of “naturalized” soil characterized by the same ranges of
hydraulic properties and thus there is no variation in the net infiltration.
6.2 Influence of Cover Erosion on Contaminant Transport and
Receptor Dose
An additional model scenario was constructed to assess the effects of side-wide erosion. For this
scenario, the gully formation model described in Appendix 10 was not used; instead, the entire
cover was assumed to be eroded by a fixed depth throughout the simulation to assess how a
thinner cover affects contaminant transport and the resulting receptor doses. Assuming the entire
cover erodes produces a bounding case on the effects of erosion on risk. This erosion model is
referred to as v1.4XXX Benson Erosion, and was built starting with v1.4XXX Benson. As such,
this scenario assumes the homogenized parameters for the cover layers described in Section 2.6.
6.2.1 Implementation in GoldSim
The cover cell thicknesses in v1.4XXX Benson were reduced to arrive at v1.4XXX Benson
Erosion. Two simulations were modeled: one in which the total cover thickness is reduced by 6
inches, and another where the cover thickness was reduced by 4 feet. Each cover cell thickness,
except the top cell which remains at 1 cm, was reduced by a fixed fraction which resulted in the
desired cover thickness. As the original cover was 5 feet thick, the resulting cover thicknesses in
these simulations were 4.5 feet (6 inches of erosion) and 1 foot (4 feet of erosion). Plant root and
animal burrowing depths were extended deeper into the cell column to account for the thinner
cover. A switch is used to choose between the two erosion depths.
Safety Evaluation Report Response
25 November 2015 18
6.2.2 Results
Results for v1.4XXX Benson Erosion are provided in Table 9, where “6 in” and “4 ft” indicate
the erosion depth. Results comparing the v1.4XXX Benson model to the erosion models indicate
that downward migration of contaminants to the water table is not affected by erosion of the
cover layer. This makes sense because net infiltration is not appreciably influenced by cover
thickness as demonstrated in Section 6.1. Doses to the rancher receptor are increased due to a
thinner amount of material above the DU waste. The thinner cover results in increased radon flux
at the surface. The scenario with 4 feet of erosion showed a larger increase, as expected.
However, even 4 feet of erosion across the entire cover produced less than an order of
magnitudes increase, and the 95th percentile doses still remain less than 0.5 mrem/year. These
results demonstrate that while receptor doses do increase with an eroded cover, doses still remain
low despite the assumption of site-wide erosion of the cover.
6.2.3 Sensitivity Analysis
Sensitive parameters for v1.4XXX Benson Erosion are presented in Table 10. The sensitive
parameters are the same as for v1.4 XXX Benson as described in Section 2.8.
6.2.4 Discussion
The subject modifications to the cover erosion model do not appreciably affect endpoint
contaminant transport and dose. Changes to the erosion model do not need to be made to the
Clive DU PA Model v1.4.
Table 9. Model results for v1.4XXX Benson Erosion.

Mean Median 95th Percentile
v1.4XXX
Benson
v1.4XXX Benson
Erosion v1.4XX
X
Benson
v1.4XXX Benson
Erosion v1.4XX
X
Benson
v1.4XXX Benson
Erosion
6 in 4 ft 6 in 4 ft 6 in 4 ft
Peak Tc-99
groundwater
concentration
within 500 yr
(pCi/L)
7.6E3 7.6E3 7.6E3 3.0E2 3.0E2 3.0E2 4.1E4 4.1E4 4.1E4
Peak rancher dose
within 10,000 yr
(mrem/yr)
5.1E-2 5.7E-2 1.2E-1 4.5E-2 5.0E-2 1.0E-1 1.2E-1 1.4E-1 2.8E-1
* v1.4 results are based on 10,000 realizations, while other results in this table are based on 1,000 realizations.

Safety Evaluation Report Response
25 November 2015 19
Table 10. Sensitive input parameters for v1.4XXX Benson Erosion.
SI rank Input parameter Sensitivity
index (SI)
Peak 99Tc groundwater
concentration within 500
years
1 Kd for Tc 43
2 Activity Concentration of Tc-99 in SRS DU
Waste
16
3 Molecular Diffusivity in Water 14
4 Van Genuchten’s n 5
Peak rancher dose within
10,000 years
1 Radon Escape/Production Ratio for Waste 38
2 Kd for Ra in sand 4
1 For technetium, the same Kd value was used for all materials.

7.0 Clay Liner
The SER (SC&A 2015) describes concern with the modeling of water flow through the clay liner
below the waste in the DU PA GoldSim Model. The problems stated are:
• Flow modeling does not include a correlation between the hydraulic function parameters
α and Ks.
• Flow modeling does not account for “naturalization” of the cover, which will change
hydraulic function parameters.
The GoldSim software platform cannot directly model flow. A water flow rate is assigned in the
GoldSim cell network for every realization based on simulations using a variably saturated flow
model that is run external to GoldSim. The development of linear models for water content and
net infiltration (flow rate) is described in Section 2.0 above. Net infiltration values for the entire
unsaturated portion of the model were calculated using a flow model. These flow model net
infiltration results were based on hydraulic function parameters for homogenized materials using
the method provided in Appendix E, Volume 2, of the SER (SC&A 2015). This included use of
the “Hyd Props Calculator.xls” for generating 50 hydraulic parameter sets for the HYDRUS
simulations, where the values of ln(α) and ln(Ks) were correlated with a correlation coefficient of
0.48 provided in SC&A (2015). The net infiltration rate through the clay liner used for a
realization of the DU PA Model represents behavior that accounts for homogenization of
materials and correlation of the lnα and ln(Ks) parameters. The flow rate of water through the
unsaturated cells of the GoldSim model is the same in the clay liner as it is in the radon barriers,
so the above concerns are addressed through using this modeling approach.
Safety Evaluation Report Response
25 November 2015 20
An addition GoldSim simulation was created to assess the effects of using homogenized
properties for the clay liner.
7.1 GoldSim Implementation
The following changes are implemented to model v1.4XXX Benson. The resulting model
iteration is referred to as v1.4XXX Benson Clay Liner.
Porosity, bulk density, and Ks for the clay liner layers were set equal to those of the naturalized
cover, which are obtained from a lookup table for each realization of the model as described in
Section 2.6. The model was run for 1000 realization, and the results are summarized in
Section 7.2.
7.2 Results
Results from the 1.4XXX Benson Clay Liner simulation are summarized in Table 11. The
v1.4XXX models produce similar results. These results indicate that changing the clay liner
properties to those of the homogenized cover does not appreciably affect endpoint contaminant
transport and dose.
Table 11. Model results for v1.4XXX Benson Clay Liner.

Mean Median 95th Percentile
v1.4XXX
Benson
1.4XXX
Benson
Clay Liner
v1.4XXX
Benson
1.4XXX
Benson
Clay Liner
v1.4XXX
Benson
1.4XXX
Benson
Clay Liner
Peak Tc-99 groundwater
concentration within 500
yr (pCi/L)
7.6E3 7.9E3 3.0E2 3.0E2 4.1E4 4.2E4
Peak rancher dose within
10,000 yr (mrem/yr) 5.1E-2 5.2E-2 4.5E-2 4.5E-2 1.2E-1 1.2E-1
* These results are based on 1,000 realizations of the models.

7.3 Sensitivity Analysis of v.1.4XXX Benson Clay Liner
Sensitive parameters for v1.4XXX Benson Erosion are presented in Table 12. The sensitive
parameters are the same as those for v1.4 XXX Benson as described in Section 2.8.
Safety Evaluation Report Response
25 November 2015 21
Table 12. Sensitive input parameters for v1.4XXX Benson Clay Liner.
SI rank Input parameter Sensitivity
index (SI)
Peak 99Tc groundwater
concentration within 500
years
1 Kd for Tc 43
2 Activity Concentration of Tc-99 in SRS DU
Waste
16
3 Molecular Diffusivity in Water 13
4 Van Genuchten’s n 5
Peak rancher dose within
10,000 years
1 Radon Escape/Production Ratio for Waste 38
2 Kd for Ra in sand 4
1 For technetium, the same Kd value was used for all materials.
7.4 Discussion
Modifications to the clay liner properties do not appreciably affect endpoint contaminant
transport and dose. Changes to the clay liner properties do not need to be made to the Clive DU
PA Model v1.4.
8.0 Deep Time
The SER (SC&A 2015) describes issues with deep time modeling and requests model changes.
Three changes are requested in the SER:
• The material above the DU waste be modeled as Unit 3 for consistency with other
Model processes that characterize waste layers as Unit 3.
• The standard deviation of the eolian deposition rate be used instead of the standard
error of the mean.
• The intermediate lake sedimentation rate be changed to 10 times the large lake
sedimentation rate.
In the Clive DU PA Model v1.4, Unit 3 properties were used in deep time above the DU waste
layers for consistency with other near time model processes. As well, the expected grain-size
characteristics of intermediate lake sediments and an expected southern flux of long-shore drift
sand from the Grayback Hills southward toward the Clive site share those characteristics. This
model change is acceptable.
The second and third requests in the SER were modeled to demonstrate the effects of those
changes on results; however, these changes are not acceptable to implement in the Clive DU PA
Model v1.4. The modeling, results and discussion are provided in the following sections.
Safety Evaluation Report Response
25 November 2015 22
8.1 GoldSim Implementation
The following changes are implemented to model v1.4XXX Benson; the resulting model
iteration is referred to as v1.4XXX Benson Deep Time.
The depth of eolian deposition layers was set as a normal distribution with a mean of 72.7 cm
and the standard deviation in the distribution was changed from 5.0 cm to 16.6 cm, as discussed
above. Additionally, the sedimentation rate for intermediate lakes was calculated as ten times the
deep lake sedimentation rate.
The results of this simulation are presented in Section 8.2.
8.2 Results
Endpoint results for the v1.4XXX Benson Deep Time model and the v1.4XXX Benson model
are presented Table 13. Sediment concentrations increase by about double with these deep time
model changes due to thinner sediment thicknesses. These sediment concentrations are still quite
low.
Lake concentrations do not change much with these model changes. The amount of material
above grade when the first lake returns is not affected by the model changes requested for deep
time. There are sufficient amounts of radionuclides in the sediments that lake concentrations are
controlled by diffusion rather than by sediment concentrations. These lake concentrations are
still quite low.
Safety Evaluation Report Response
25 November 2015 23
Table 13. Comparison of deep time results at 90,000 yr for v1.4XXX Benson Deep Time
and v1.4XXX Benson models.

25th Percentile Median Mean 95th Percentile
v1.4XXX
Benson
v1.4XXX
Benson
Deep
Time
v1.4XXX
Benson
v1.4XXX
Benson
Deep
Time
v1.4XXX
Benson
v1.4XXX
Benson
Deep
Time
v1.4XXX
Benson
v1.4XXX
Benson
Deep
Time
U-238 lake
concentration
(pCi/L)
1.4E-7 1.4E-7 1.3E-5 1.3E-5 3.9E-3 3.9E-3 1.5E-2 1.5E-2
Ra-226 lake
concentration
(pCi/L)
2.0E-4 2.0E-4 9.4E-3 9.2E-3 6.2E-2 6.1E-2 3.0E-1 2.9E-1
U-238
sediment
concentration
(pCi/g)
1.3E-7 2.3E-7 5.3E-6 9.5E-6 2.4E-4 4.1E-4 1.1E-3 1.9E-3
Ra-226
sediment
concentration
(pCi/g)
1.7E-6 2.8E-6 7.1E-5 1.3E-4 5.8E-4 9.4E-4 2.9E-3 4.6E-3

Peak radon flux averages in deep time are 18 pCi/m2s for the Clive DU PA Model v1.4 with an
associated rancher dose of 0.14 mrem/yr. For v1.4XXX Benson Deep Time model they increase
to 160 pCi/m2s, with an associated rancher dose of 2 mrem/yr. This dose is still in an acceptable
value for protection of human health.
The deep time model changes requested in the SER (SC&A, 2015) for eolian deposition and
intermediate lake sedimentation are overly conservative and contradict the deep time conceptual
model.
8.3 Discussion
8.3.1 Eolian deposition standard error
The measured thicknesses of eolian silt in quarry walls and excavated surfaces for the Clive
Disposal Site can be found in Table 14 from field research (Neptune, 2015). The mean of the
deposits is 72.7 cm, and the standard deviation is 16.6 cm. There are 11 data points, and the data
are reasonably symmetric about the mean. Consequently, a normal distribution is specified for
the Deep Time Model with a mean of 72.7 cm and a standard error of 5.0 cm. A reasonable
simulation range considering ± 3 standard errors would be 57.5 to 87.5 cm. The minimum of the
normal distribution was set to a very small number and the maximum was set to a very large
number, so that the distribution was not unnecessarily restricted.
Safety Evaluation Report Response
25 November 2015 24
Table 14. Thickness measurements from field studies of eolian silt near Clive
Neptune Field Studies December 2014
Site GPS Coord GPS Coord Silt Thick Date
UTM E UTM N (cm) (mm/dd/yy)
Clive 29-1 321354 4508262 90.0 12/16/14
Clive 29-2 321390 4508256 80.0 12/16/14
Clive 29-3 321423 4508248 80.0 12/16/14
Clive 29-4 321502 4508236 60.0 12/16/14
Clive 29-5 321239 4508283 110.0 12/16/14
Clive 5-1 320813 4504729 55.0 12/16/14
Clive 5-2 320869 4504730 70.0 12/16/14
Clive 5-3 320914 4504731 60.0 12/16/14
Clive 5-4 321041 4504732 70.0 12/16/14
Clive Hand-Dug-1 322093 4507482 70.0 12/17/14
Clilve hand-Dug-2 320445 4507035 55.0 12/17/14
Mean 72.7
Std Dev. 16.6

This distribution represents spatio-temporal scaling, so that the distribution is of the average
depth of eolian deposition at the Clive site since Lake Bonneville regressed below the site. This
provides the best representation of the future eolian depositional rates over the long time frames
and spatial scales of the Deep Time Model.
8.3.2 Intermediate lake sedimentation rates
The intermediate lake sedimentation rate used in the v1.4XXX Benson Deep Time model is set
at 10 times the large lake sedimentation rate per review guidance. However, the assignment of a
sedimentation rate for intermediate lakes derived from a deep lake sedimentation rate is not
conceptually valid. The explanation for this conclusion requires re-examination of the definitions
of shallow, intermediate, and deep lakes used in the Deep Time Assessment for the Clive DU PA
white paper.
Deep lakes in the cyclical, climate driven Deep Time Assessment are similar to the Lake
Bonneville stage where the dominant mode of sedimentation is deposition of carbonate (for
example, the marl sedimentary facies of the Bonneville and Provo lakes). These carbonate
sedimentation rates are dependent on rates of precipitation of chemical sediment (inorganic
materials precipitated from the lake) and biogenic sediment (fossil remains of former living
organisms). In order to form carbonate-dominated lake sediments with subordinate clastic
sedimentary deposits, the elevation of a deep lake must be sufficiently higher than the elevation
of the Clive site (lake depth above the site) to exclude sedimentation associated with shoreline
processes and/or wave activity.

Safety Evaluation Report Response
25 November 2015 25
Intermediate lakes are transitory features that, by definition, reach the elevation of the Clive site.
The sedimentation rate for an intermediate lake is dependent on basin location, shoreline
processes, wave dynamics, presence or absence of fluvial deposition and/or local sedimentary
sources, basin slope, water chemistry and biological activity. Inorganic and biogenic deposition
occurs during intermediate lake cycles but is secondary to clastic sedimentation. Intermediate
lakes may rise and fall about the Clive elevation but they do not reach sufficient elevations (lake
depth) to deposit only carbonate sediments. Dependent on local conditions, sedimentation rates
of intermediate lakes can be significantly higher than carbonate depositional rates of large lakes.
Shallow lakes are equivalent to the modern Great Salt Lake and by definition do not reach the
elevation of the Clive site.
The duration of intermediate lakes is difficult to establish because they are transitory, their
deposits are reworked by wave activity, and they do not preserve prominent shoreline features
that can be used to establish lake chronology. Because of these limitations, the intermediate lake
sedimentation parameter used in the Deep Time Assessment is a sediment thickness per lake
cycle where thickness data are obtained for clastic sedimentary lake sequences using lake-core
data from multiple locations in the Lake Bonneville basin. The intermediate lake thickness is an
average thickness obtained from composite Lake Bonneville and pre-Lake Bonneville clastic
sedimentary deposits. The deep lake sedimentation rate is established from dated cycles of deep
lake marl deposits from both field observations and core studies. The deep lake sediments and
the intermediate lake sedimentation thickness are controlled by different processes (carbonate
precipitation versus lake-shoreline processes, respectively). The intermediate lake sedimentation
rate or the thickness of intermediate lake sedimentary cycles cannot be established from the
sedimentation record of deep lakes.
Iterative refinement of performance assessment models is a well-established methodology for
improving the quality and information content of model results. Initial models are developed and
sensitivity analysis is used to identify model parameters that most strongly affect model results.
These sensitive parameters are refined, usually through focused data gathering. The performance
assessment model is then rerun with refined parameters and the model results are re-examined
for impact on decision objectives. Iterative updates of the Clive DU PA deep time modeling have
been used to improve the usefulness of model results. However, as discussed above, using an
intermediate lake sedimentation rate based on large lake sedimentation rates is not conceptually
valid and degrades the model results. The initial GoldSim model results demonstrated clearly the
importance of the timing and characteristics of the first return of an intermediate lake to the Clive
site on resulting waste/sediment concentrations. A more effective approach to model
improvements for intermediate lakes would be to focus the model structure and results on the
characteristic of lake sediments at the Clive site. The intermediate lake sedimentary thickness
used in the deep time model is based on patterns of sedimentation in the Lake Bonneville basin
and is applied to the 2.1 million year cycle of the deep time analysis. Future model
improvements should shift to the timing and characteristics of the first return of an intermediate
lake at the Clive site. Deposits of the transgressive phase of Lake Bonneville and clastic
sedimentary sequences below the Lake Bonneville deposits can be studied to develop
sedimentation patterns of intermediate lakes specific to the Clive site.
Safety Evaluation Report Response
25 November 2015 26
9.0 References
Benson, C.H., et al., 2011. Engineered Covers for Waste Containment: Changes in Engineering
Properties and Implications for Long-Term Performance Assessment, NUREG/CR-7028,
United States Nuclear Regulatory Commission, Washington DC, December 2011
Branson, F.A., et al., 1976. Moisture Relationships in Twelve Northern Desert Shrub
Communities Near Grand Junction, Colorado, Ecology 57 (6) 1104–1124
Envirocare, 2000. Application for License Amendment for Class B & C Waste, Envirocare of
Utah Inc., North Salt Lake UT, December 2000
Hoven, H.M., et al., 2000. Assessment of Vegetative Impacts on LLRW, prepared for Envirocare
of Utah Inc., SWCA Inc. Environmental Consultants, Salt Lake City UT, November 2000
Meinzer, O.E., 1927. Plants as Indicators of Ground Water, Water-Supply Paper 577, United
States Geological Survey, Washington DC, 1927
Meyer, P.D., et al., 1996. Hydrologic Evaluation Methodology for Estimating Water Movement
Through the Unsaturated Zone at Commercial Low-Level Radioactive Waste Disposal
Sites, NUREG/CR-6346, PNL-10843, prepared for United States Nuclear Regulatory
Commission, Pacific Northwest Laboratory, Richland WA, January 1996
Neptune, 2015. Neptune Field Studies, December, 2014, Eolian Depositional History Clive
Disposal Site, NAC-0044_R0, Neptune and Company Inc., Los Alamos NM, March 2015
Richardson, C.W., and D.A. Wright, 1984. WGEN: A Model for Generating Daily Weather
Variables, United States Department of Agriculture, Washington DC, August 1984
Robertson, J.H., 1983. Greasewood (Sarcobatus vermiculatus (Hook.) Torr.), Phytologia 54 (5)
309–324
SC&A, 2015. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
2, SC&A, Vienna VA, April 2015
Schroeder, P.R., et al., 1994a. The Hydrologic Evaluation of Landfill Performance (HELP)
Model, Engineering Documentation for Version 3, EPA/600/R-94/168b, United States
Environmental Protection Agency, Office of Research and Development, Washington
DC, 1994
Schroeder, P.R., et al., 1994b. The Hydrologic Evaluation of Landfill Performance (HELP)
Model, User's Guide for Version 3, EPA/600/R-94/168a, United States Environmental
Protection Agency, Office of Research and Development, Washington DC, September
1994
Safety Evaluation Report Response
25 November 2015 27
SWCA, 2013. EnergySolutions Updated Performance Assessment—SWCA’s Response to First
Round DRC Interrogatories, SWCA Environmental Consultants, September 2013
Waugh, W.J., et al., 1999. Plant Encroachment on the Burrell, Pennsylvania, Disposal Cell:
Evaluation of Long-Term Performance and Risk, GJO–99–96–TAR, United States
Department of Energy, Grand Junction CO, June 1999
Waugh, W.J., and G.M. Smith, 1998. Root Intrusion of the Burrell, Pennsylvania, Uranium Mill
Tailings Cover (invited paper), Proceedings: Long-Term Stewardship Workshop, CONF–
980652, 1998, Grand Junction CO

Safety Evaluation Report Response
25 November 2015 A-1
Appendix A HYDRUS Simulation Results
Table 15 provides the results from the 50 HYDRUS simulations using naturalized values of
hydraulic function parameters. Volumetric water contents are listed for zones in the naturalized
cover corresponding to depths of the surface (WC1), evaporative (WC2), frost protection (WC3),
upper radon barrier (WC4), and lower radon barrier (WC5) layers and the net infiltration at the
top of the waste.
Table 15. Water content and infiltration results from 50 HYDRUS simulations using
naturalized (homogenous) hydraulic properties.
Replicate WC1
[-]
WC2
[-]
WC3
[-]
WC4
[-]
WC5
[-]
Net Infiltration
(mm/yr)
1 0.129 0.161 0.163 0.163 0.163 0.74
2 0.125 0.157 0.159 0.159 0.159 0.63
3 0.099 0.126 0.127 0.127 0.126 1.31
4 0.109 0.138 0.139 0.139 0.138 1.16
5 0.132 0.166 0.167 0.166 0.166 0.99
6 0.147 0.183 0.182 0.180 0.179 0.93
7 0.161 0.200 0.201 0.200 0.199 0.86
8 0.148 0.184 0.183 0.182 0.181 0.97
9 0.114 0.144 0.145 0.145 0.144 1.10
10 0.122 0.152 0.151 0.150 0.149 1.10
11 0.140 0.175 0.176 0.176 0.175 0.89
12 0.130 0.163 0.164 0.163 0.162 1.04
13 0.148 0.185 0.185 0.184 0.184 0.95
14 0.131 0.164 0.165 0.165 0.164 0.97
15 0.150 0.187 0.189 0.189 0.189 0.68
16 0.166 0.207 0.207 0.207 0.206 0.81
17 0.130 0.164 0.164 0.164 0.163 1.00
18 0.157 0.194 0.190 0.187 0.185 0.78
19 0.121 0.152 0.154 0.154 0.154 0.72
Safety Evaluation Report Response
25 November 2015 A-2
20 0.145 0.182 0.181 0.180 0.180 0.97
21 0.175 0.217 0.219 0.219 0.219 0.57
22 0.148 0.184 0.184 0.184 0.183 0.96
23 0.131 0.164 0.165 0.164 0.164 1.01
24 0.129 0.161 0.162 0.161 0.161 1.05
25 0.153 0.189 0.188 0.187 0.186 0.93
26 0.130 0.164 0.165 0.164 0.164 1.00
27 0.145 0.181 0.183 0.183 0.183 0.76
28 0.130 0.163 0.165 0.165 0.165 0.89
29 0.141 0.177 0.177 0.176 0.176 0.93
30 0.182 0.226 0.227 0.227 0.227 0.66
31 0.151 0.187 0.187 0.185 0.185 0.92
32 0.147 0.182 0.182 0.180 0.180 0.96
33 0.126 0.158 0.160 0.160 0.160 0.95
34 0.145 0.181 0.182 0.181 0.181 0.97
35 0.129 0.162 0.163 0.162 0.161 1.04
36 0.156 0.194 0.195 0.195 0.195 0.88
37 0.125 0.157 0.159 0.159 0.159 0.83
38 0.128 0.161 0.161 0.161 0.161 1.06
39 0.142 0.177 0.179 0.178 0.178 0.83
40 0.130 0.163 0.164 0.164 0.164 0.95
41 0.173 0.214 0.216 0.216 0.216 0.72
42 0.141 0.175 0.174 0.173 0.172 0.98
43 0.133 0.167 0.168 0.167 0.167 0.94
44 0.120 0.152 0.153 0.153 0.153 0.92
45 0.135 0.169 0.170 0.170 0.170 0.89
Safety Evaluation Report Response
25 November 2015 A-3
46 0.155 0.193 0.194 0.194 0.194 0.80
47 0.137 0.171 0.172 0.172 0.172 0.76
48 0.138 0.172 0.171 0.170 0.169 1.02
49 0.118 0.149 0.151 0.151 0.151 0.92
50 0.133 0.167 0.168 0.168 0.168 0.89

Safety E
25 Nove
Appe
The foll
and hyd
Figure
co

valuation Rep
ber 2015
ndix B
owing explo
aulic param
7. Relations
tent in the
ort Respons
Flow Mo
ratory data
ter inputs
hip betwee
upper 6 in
e
del Dev
plots show t
r the 50 pa
n model h
hes of the
elopme
he relations
ameter sets

draulic par
over.
nt Plot
hip between
.
ameters an
s
HYDRUS
d modeled v
model outpu
olumetric
A-4
ts
water
Safety E
25 Nove
Figure
co

valuation Rep
ber 2015
8. Relations
tent from
ort Respons
hip betwee
inches to
e
n model h
8 inches d

draulic par
ep in the c
meters an
ver.
modeled volumetric
A-5

water
Safety E
25 Nove
Figure
co

valuation Rep
ber 2015
9. Relations
tent from
ort Respons
hip betwee
8 inches t
e
n model h
36 inches
raulic par
eep in the
meters an
over.
modeled volumetric
A-6

water
Safety E
25 Nove
Figure 1
w

valuation Rep
ber 2015
0. Relation
ter content
ort Respons
ship betwe
from 36 in
e
en model h
hes to 48 i

ydraulic pa
ches deep
ameters a
n the cove
d modeled
.
olumetric
A-7
Safety E
25 Nove
Figure 1
w

valuation Rep
ber 2015
1. Relation
ter content
ort Respons
ship betwe
from 48 in
e
en model h
hes to 60 i
draulic pa
ches deep
ameters a
n the cove
d modeled
.
olumetric
A-8

Safety E
25 Nove
Figure 1
at

valuation Rep
ber 2015
2. Relation
he the top
ort Respons
ship betwe
f the wast

e
en model h
.
draulic pa ameters a d modeled net infiltra
A-9
tion
Safety E
25 Nove
The foll
content
for speci
Figure 1
su
valuation Rep
ber 2015
owing plots
nd net infil
ying water
3. HYDRU
face throu
ort Respons
show the fit
ration and t
flow.
S volumetr
h the uppe
e
between th
e linear mo
ic water co
r radon ba
e HYDRUS
dels uilt in

ntents plott
rier of the
model resul
to the Gold
ed with line
cover.
ts for volu
im v1.4X
ear model v
metric water
X Benson
alues for th
A-10
Model

e
Safety E
25 Nove
Figure 1
in

valuation Rep
ber 2015
4. HYDRU
iltration in
ort Respons
S volumetr
o the top o
e
ic water co
the waste

ntent for th
lotted wit
lower rad
linear mo
n barrier
el values.
nd net
A-11

NAC-0115_R0

Introduction to DU PA Model
Version 1.4 Interrogatory
Responses
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 ii

Introduction to DU PA Model Version 1.4 Interrogatory Responses
Introduction to DU PA Model Version 1.4 Interrogatory Responses.docx
Introductory document providing context for seven associated topic-based
response documents.

Sean McCandless 12 February 2018
Mike Sully 12 February 2018

Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
ACRONYMS AND ABBREVIATIONS ................................................................................... iv
1.0 Introduction ........................................................................................................................ 1
2.0 Introduction to Probabilistic Modeling ............................................................................... 2
3.0 References .......................................................................................................................... 4

Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 iv
ACRONYMS AND ABBREVIATIONS
CSM Conceptual Site Model
DEQ (Utah) Department of Environmental Quality
DU depleted uranium
ET evapotranspiration
GIGO garbage in, garbage out
LLRW low-level radioactive waste
PA performance assessment
PPA probabilistic performance assessment
SER Safety Evaluation Report
UDEQ Utah Department of Environmental Quality

Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 1
1.0 Introduction
Beginning in 2009, EnergySolutions contracted Neptune and Company, Inc. (Neptune) to create
a probabilistic performance assessment (PPA) for the disposal of large quantities of depleted
uranium (DU) at their Clive, Utah low-level radioactive waste (LLRW) disposal facility.
The initial model was submitted as version 1.0 on June 1, 2011 (Neptune 2011) and was revised
to version 1.2 on June 5, 2014 (Neptune 2014). A Safety Evaluation Report (SER) based on
review of version 1.2 was issued by the Utah Department of Environmental Quality (UDEQ) in
April 2015 (SC&A 2015).
On November 25, 2015, EnergySolutions submitted Radioactive Material License UT2300249:
Safety Evaluation Report for Condition 35.B Performance Assessment; Response to Issues
Raised in the April 2015 Draft Safety Evaluation Report (EnergySolutions 2015). This document
included version 1.4 of the DU PA (Neptune 2015), prepared in response to open primary and
supplemental interrogatories included in Appendix C and Appendix B, respectively, of the SER.
On May 11, 2017, UDEQ provided Amended and New Interrogatories Related to Clive DU PA
Modeling Report Version 1.4 Dated November 2015 (Utah DEQ 2017). This document contains
revised and new interrogatories regarding version 1.4 of the DU PA.
UDEQ has arranged the interrogatories and SER comments in the Amended and New
Interrogatories Related to Clive DU PA Modeling Report Version 1.4 (Utah DEQ 2017) into
seven general categories. In order to retain focus on each subject area, responses are grouped into
individual papers by these seven categories. Each category is addressed within a standalone
Neptune response document as follows:
1. NAC_0106: Evapotranspiration (ET) Cover Design
2. NAC_0108: Erosion
3. NAC_0105: Deep Time Supplemental Analysis
4. NAC_0102: Other Wastes
5. NAC_0104: Groundwater Exposure
6. NAC_0103: Recycled Uranium
7. NAC_0101: Federal Cell Design
Note that, in an attempt to improve readability, the complete interrogatory number is cited as
each is first introduced and in headings, while the abbreviated number is used during discussion.
For example, Interrogatory CR R313-25-7(2)-05/2 is introduced by its full number, then is
abbreviated to Interrogatory 05/2, since the interrogatory numbering system employed by UDEQ
applies a unique number after the last hyphen in the sequence.
Note also that some interrogatories are linked to more than one section of the Utah
Administrative Code; thus, a single interrogatory can appear at first glance to be two. For
example, “Interrogatory CR R313-25-3 and R313-25-8-195/1” is a single interrogatory
referencing two parts of the Utah Administrative Code related to the subject at hand. The
shorthand for this interrogatory becomes “195/1.”
Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 2
Full text of each interrogatory, with UDEQ images, tables, and references, is available in Utah
DEQ (2017). Within these responses, the interrogatory is briefly quoted or paraphrased to
identify what is understood to be the issue of concern. When quoting longer passages of
interrogatory text, blue font in Arial size 10.5 is used and indented to visually distinguish the
interrogatory from the response. An example is shown below:
Sample format for quoting interrogatory text.
When quoting shorter passages within a paragraph, italics are used to distinguish the
interrogatory text being quoted.
The attached responses address issues raised in the interrogatories point by point, providing a
defensible basis for closing all open review items.
2.0 Introduction to Probabilistic Modeling
The Clive DU PA Model is a probabilistic model of the potential long-term risk consequences
from disposing DU in the Clive facility. This means that the inputs to the Model are specified as
probability distributions. The PA results depend not only on the model structure, but also on the
model specification (input probability distributions). The Clive DU PA Model structure is a
complex fully coupled organization of features, events, processes, and scenarios as described in
the Conceptual Site Model (CSM). Each of the inputs to this Model needs to be specified with
the same degree of care and attention that is given to the model structure. The PA Model results
depend on the probabilistic inputs; consequently, it is critical that the specification addresses
underlying statistical principles that relate to such long-term dynamic modeling. Note also that
the sensitivity analysis for the PA Model depends on the specific probabilistic input, in which
case identification of the appropriate important, or sensitive, parameters also requires correct
specification of the input probabilistic distributions.
The input distributions must address uncertainty in the data/information available for
specification, and must address spatial and temporal scaling. A well-formed probabilistic model
includes distributions that: 1) are based on what is thought to be known (expectation) and the
uncertainty about that expectation; 2) address spatial and temporal scaling, and 3) address
correlation between inputs where possible.
There is a tendency in PA modeling to ignore all three facets and instead to perform modeling
that is based on notions associated with conservatism (often in a deterministic model instead of a
probabilistic model), to ignore spatial and temporal scaling, and to ignore correlation. This leads
to projected human health risks or doses that are biased high and are far more uncertain than
appropriate, which in turn leads to poor decision making.
It is fine to make conservative decisions, but it is not fine to make important decisions based on
purposefully erroneous models. This is the same basic message that was attributed to Charles
Babbage after his invention of a “difference machine” in the late 1800s. Babbage wrote, “On two
occasions I have been asked ‘Pray, Mr. Babbage, if you put into the machine the wrong figures,
will the right answers come out’ I am not able rightly to comprehend the kind of confusion of
ideas that could provoke such a question.” This is the genesis of the term “garbage in, garbage
Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 3
out” (GIGO), which also applies to today’s modeling paradigm if the “wrong figures” are used.
It is crucial to pay attention to the statistical details of specifying a probabilistic model if
reasonable output is desired.
The Clive DU PA Model has been built to the extent possible based on expectation and
uncertainty. Correlation is included when and if supporting data/information are available. And,
considerable effort has been put into ensuring that the spatial and temporal scales of the input
distributions used in the PA Model are appropriate. There are several aspects of the supporting
data/information, supporting models (process level models such as Hydrus, SIBERIA, and
AERMOD, which were used to support the PA Model), and the PA Model itself that must be
considered when establishing the correct spatial and temporal domain of the PA Model and the
supporting data. Often data that are useful for specifying input distributions for a PA are
available at a very small spatial and temporal scale. Process level models need to address scaling
at a refined level, and the systems level PA model must address scaling at yet another level.
Statistical scaling must also consider the endpoint of interest. For PA models, an intermediate
endpoint is concentration of radionuclides in some media thousands of years (or in the case of
the Clive DU PA, millions of years) into the future. Some assumptions are critical, such as that a
PA model will project current knowledge into the future, so that changes to many aspects of the
environmental system are considered unlikely (for example, the basin contents are silt/clay, and
this is assumed to not change throughout time). Scaling can address temporal (or spatial)
changes, but the point here is to explain why scaling is needed, and relatively simple examples
are used for demonstration purposes. Consequently, for the purpose of this explication, the
parameters of interest are assumed stable in time and space (e.g., chemical/physical
characteristics).
Other considerations relate to the computational set up and the computational complexity of a
PA model. Suppose, for simplicity, that the model looks forwards 1,000 years, and that each time
step is one year. In principle a new random number could be drawn from an input distribution
every year; however, in practice this adds greatly to the computational complexity. Instead, an
approach is taken that draws a random number at the beginning of time, and applies that random
number in each of the 1,000 time steps. The trick is to find the appropriate distribution for this
approach to mimic pulling a random number every time step. If the response to the input
parameter is linear and the system with respect to that input parameter is stationary (or steady
state), then the input distribution that is needed at the beginning of time to mimic the effect of
drawing new random numbers every year (time step) is the distribution of the average of the
distribution that would be used for 1,000 random draws over time. For non-linear responses, and
non-stationary systems, the statistical scaling issues are more complex, but they are conceptually
the same.
In effect, the spatial and temporal scaling that is needed must address the differences in scale
between the data/information available and the models that are used, including the difference in
scale of the supporting process level models and the systems level PA model. Scaling in this
context is largely an averaging process. If the response is linear and stationary, then pure
averaging works directly, and this is the approach that has been taken in the Clive DU PA Model
for projecting future concentrations to which human receptors are exposed.
Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 4
A simple example of scaling considers the proportion of green M&Ms in a packet as a predictor
of the proportion of green M&Ms that are produced. The distribution of green M&Ms from
many packets might range from near 0% up to about 33%, but that is not the correct distribution
to use to predict the proportion of green M&Ms in 1,000 packets. Scaling takes over, and a better
estimate for 1,000 packets would be very close to 16% overall (the production rate of green
M&Ms is about 16%). One could estimate the number of green M&Ms in 1,000 packets by
adding up how many are in each packet (similar to choosing a new random number each time
step), or by taking the distribution of the average of the packet specific distribution. To make this
more concrete, suppose the proportion of green M&Ms in a packet follows a normal distribution
with mean 16% and standard deviation 5%; then, simulating by drawing a new random number
from that distribution for each one of the 1,000 packets gives the same result (distributional
estimate of overall proportion) as that obtained from simulating directly from a normal
distribution with mean 16% and standard deviation 5%/sqrt(1,000), where the 1,000 represents
the 1,000 packets. However, the wrong thing to do would be to simulate from the N(16%, 5%)
distribution and apply that to the overall proportion of green M&Ms in 1,000 packets. This
describes the essence of the scaling problem, which depends on the endpoint of interest for the
problem, and the scale of both the data and the model.
The basic approach to developing input probability distributions is described in Appendix 14 of
the initial Clive DU PA submittal. Further explanation of some of the details is provided in
response to interrogatories concerning the ET Cover, because there are more interrogatories
about input distributions for the ET Cover than for other aspects of the Clive DU PA Model. The
critical issue is the importance of spatial and temporal scaling so that parameter uncertainties are
appropriately captured in the Model.
3.0 References
EnergySolutions, 2015. Radioactive Material License UT2300249: Safety Evaluation Report for
Condition 35.B Performance Assessment; Response to Issues Raised in the April 2015
Draft Safety Evaluation Report, EnergySolutions LLC, Salt Lake City UT, November
2015
Neptune, 2011. Final Report for the Clive DU PA Model version 1.0, Neptune and Company
Inc., Los Alamos NM, June 2011
Neptune, 2014. Final Report for the Clive DU PA Model, Clive DU PA Model v1.2, NAC-
0024_R2, Neptune and Company, Inc., Los Alamos NM, August 2014
Neptune, 2015. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
SC&A, 2015. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
1, SC&A Inc., Vienna VA, April 2015
Introduction to DU PA Model Version 1.4 Interrogatory Responses
23 Feb 2018 5
Utah DEQ, 2017. Division of Waste Management and Radiation Control, EnergySolutions Clive
LLRW Disposal Facility License No: UT2300249; RML #UT 2300249, Amended and
New Interrogatories Related to Clive DU PA Modeling Report Version 1.4 Dated
November 2015, Utah Department of Environmental Quality (DEQ), Salt Lake City UT,
May 2017

NAC-0106_R0

ET Cover Design Responses for the
Clive DU PA Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 ii
ET Cover Design Responses for the Clive DU PA Model
ET Cover Design Responses for the Clive DU PA Model.docx
Responses to UDEQ Interrogatories and Safety Evaluation Report Comments
received May 11, 2017.
Mike Sully and Sean McCandless 12 Feb 2018
Dan Levitt

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
FIGURES ................................................................................................................................... v
TABLES ..................................................................................................................................viii
ACRONYMS AND ABBREVIATIONS ................................................................................... ix
1.0 Overview and Conceptual Model........................................................................................ 1
1.1 Modeling for Probabilistic Performance Assessments ................................................... 1
1.2 Setting .......................................................................................................................... 9
1.2.1 Disposal Cell Design ............................................................................................... 9
1.2.2 Unsaturated Zone and Shallow Aquifer ................................................................. 11
1.2.3 Climate .................................................................................................................. 14
1.2.4 Vegetation ............................................................................................................. 14
1.3 HYDRUS-1D Software Package ................................................................................. 15
2.0 UDEQ Interrogatory Responses........................................................................................ 17
2.1 Interrogatory CR R313-25-7(2)-05/2: Radon Barriers ................................................. 17
2.1.1 Interrogatory Response .......................................................................................... 19
2.2 Interrogatory CR R313-22-32(2)-10/3: Effect of Biologicals on Radionuclide
Transport .................................................................................................................... 37
2.2.1 Interrogatory Response .......................................................................................... 38
2.3 Interrogatory CR R317-6-2.1-20/2: Groundwater Concentrations ............................... 39
2.3.1 Interrogatory Response .......................................................................................... 41
2.4 Interrogatory CR R313-25-8(4)(d)-21/2: Infiltration Rates .......................................... 42
2.4.1 Interrogatory Response .......................................................................................... 47
2.5 Interrogatory CR R313-25-8(4)(a)-28/3: Bioturbation Effects and Consequences ....... 61
2.5.1 Interrogatory Response .......................................................................................... 62
2.6 Interrogatory CR R313-25-7(2)-59/2: Bathtub Effect .................................................. 63
2.6.1 Interrogatory Response .......................................................................................... 64
2.7 Interrogatory CR R313-25-7(3)-60/2: Modeled Radon Barriers .................................. 64
2.7.1 Interrogatory Response .......................................................................................... 64
2.8 Interrogatory CR R313-25-7(1–2)-90/2: Calibration of Infiltration Rates .................... 64
2.8.1 Interrogatory Response .......................................................................................... 65
2.9 Interrogatory CR R313-25-7(2)-150/3: Plant Growth and Cover Performance ............ 65
2.9.1 Interrogatory Response .......................................................................................... 66
2.10 Interrogatory CR R313-25-8(4)(d)-153/2: Impact of Pedogenic Processes on the
Radon Barrier ............................................................................................................. 66
2.10.1 Interrogatory Response .......................................................................................... 66
2.11 Interrogatory CR R313-25-7(2)-175/1: Infiltration Rates for the Federal Cell Versus
the Class A West Cell ................................................................................................. 67
2.11.1 Interrogatory Response .......................................................................................... 68
2.12 Interrogatory CR R313-25-8(5)(a)-176/1: Representative Hydraulic Conductivity
Rates........................................................................................................................... 68
2.12.1 Interrogatory Response .......................................................................................... 68
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 iv
2.13 Interrogatory CR R313-25-7(2)-189/3: Modeling Impacts of Changes in Federal Cell
Cover-System Soil Hydraulic Conductivity and Alpha Values .................................... 70
2.13.1 Interrogatory Response .......................................................................................... 71
2.14 Interrogatory CR R313-25-7(2)-192/3: Implications of Great Salt Lake Freezing on
Federal Cell Performance............................................................................................ 72
2.14.1 Interrogatory Response .......................................................................................... 72
2.15 SER B.1 Supplemental Interrogatory Comment 1 ....................................................... 73
2.15.1 Interrogatory Response .......................................................................................... 74
2.16 SER B.2 Supplemental Interrogatory Comment 2 ....................................................... 77
2.16.1 Interrogatory Response .......................................................................................... 78
2.17 SER B.3 Supplemental Interrogatory Comment 3 ....................................................... 82
2.17.1 Interrogatory Response .......................................................................................... 83
2.18 SER B.4 Supplemental Interrogatory Comment 4 ....................................................... 86
2.18.1 Interrogatory Response .......................................................................................... 88
2.19 SER B.5 Supplemental Interrogatory Comment 5 ....................................................... 90
2.19.1 Interrogatory Response .......................................................................................... 92
2.20 SER B.6 Supplemental Interrogatory Comment 6 ....................................................... 97
2.20.1 Interrogatory Response .......................................................................................... 97
2.21 SER B.7 Supplemental Interrogatory Comment 7 ..................................................... 100
2.21.1 Interrogatory Response ........................................................................................ 101
2.22 SER B.8 Supplemental Interrogatory Comment 8 ..................................................... 106
2.22.1 Interrogatory Response ........................................................................................ 107
2.23 SER B.9 Supplemental Interrogatory Comment 9 ..................................................... 108
2.23.1 Interrogatory Response ........................................................................................ 108
2.24 SER B.11 Supplemental Interrogatory Comment 11 ................................................. 110
2.24.1 Interrogatory Response ........................................................................................ 111
3.0 Conclusion ..................................................................................................................... 116
4.0 References ...................................................................................................................... 119

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 v
FIGURES
Figure 1. Evapotranspiration (ET) cover profile showing materials, observation nodes, and root
distribution used in the HYDRUS-1D models. ......................................................... 11
Figure 2. Stratigraphic profile showing ET cover, waste zone, and stratigraphy below the Federal
DU Cell. .................................................................................................................. 12
Figure 3. Eolian silt in trench located at Clive Pit 29 overlying Lake Bonneville sedimentary
deposits (Neptune 2015b). ....................................................................................... 21
Figure 4. An example of upper soil-modified eolian silt in Pit 29. Basal contact of the silt is
approximately located at the middle of the pick handle. It is a gradational contact
between eolian silt intermixed with regressive Lake Bonneville marl (bottom of the
pick handle). ............................................................................................................ 22
Figure 5. Figure 3 from Benson and Gurdal (2013) showing the data requested by
EnergySolutions. ...................................................................................................... 24
Figure 6. Estimated linear relationships between α and Ks for all observations (solid line) and
without the high-leverage points making up the clusters of points in the upper right
half of the plot with α greater than 0.10 kPa-1 (dotted line). The estimated correlation
changes from 0.627 for all the data to 0.384 for the restricted range. ........................ 25
Figure 7. Estimated linear relationships within each soil type for the data provided by Benson
(2017). Pearson’s correlation coefficients, and 95% confidence intervals, are shown
in Table 2 for the soil textural classes, and the individual soil classes are shown in
panels in Figure 8. ................................................................................................... 26
Figure 8. Estimated linear relationships by soil textural class for the data plotted in Figure 6 and
Figure 7. Note the x and y axes are allowed to change among panels and are on the
log10 scale. .............................................................................................................. 29
Figure 9. Estimated linear relationship on the log scale between α and Ks for Silty Clay.
Pearson’s correlation coefficient is -0.2, with a very wide 95% confidence interval of
(-0.97, 0.94). ............................................................................................................ 30
Figure 10. Estimated linear relationships based on 16 realizations of six random pairs of
observations that come from statistically independent variables (x and y were drawn
independently from two standard normal distributions). ........................................... 31
Figure 11. Borrow soil cross-section below a greasewood plant shows the compacted clay layer
at approximately 60-cm depth. Roots extend laterally and do not penetrate the
compacted layer (SWCA 2011)................................................................................ 35
Figure 12. Table 16 from the Clive PA Model Parameters v1.4 document providing the
distribution used. A description of the method used to select the geometric mean,
geometric standard deviation, and minimum reported in this table, and the associated
parameterization of the log normal were also provided in Appendix 14 of the Clive
DU PA Model Final Report for v1.4. ....................................................................... 48
Figure 13. R Statistical Software (R Core Team 2017) code and output for getting quantiles from
the distribution of Ks used in v1.4. ........................................................................... 50
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 vi
Figure 14. The lognormal distribution used for the Ks in v1.4. It is parameterized by a geometric
mean of 3.37 (log-scale mean 1.215), a geometric standard deviation of 3.23 (log-
scale standard deviation of 1.17), and a minimum of 0.00432 implemented through a
shift of the distribution. The target 1st, 50th, and 99th percentiles are shown by the
vertical dotted lines. ................................................................................................. 51
Figure 15. The average annual net infiltration values obtained from HYDRUS compared to the
predictions from the linear regression model abstraction. The results from the 50
HYDRUS realizations were used to develop the regression model abstraction, and
therefore this plot depicts in-sample predictive performance. The one-to-one line is
shown for reference. ................................................................................................ 57
Figure 16. The average annual net infiltration values obtained from HYDRUS compared to the
predictions obtained from test datasets via 3-fold cross validation. The original
HYDRUS values predicted were not used to fit the regression equations used to
obtain the predictions. This approximates out-of-sample predictive performance of
the linear regression model abstraction. The one-to-one line is shown on each plot for
reference. ................................................................................................................. 60
Figure 17. Comparison of 1,000 realizations of net infiltration using the linear model in GoldSim
with the results of the 50 HYDRUS simulations of infiltration. ................................ 61
Figure 18. Relationship between hydraulic conductivity and water content used for Unit 4
material.................................................................................................................... 69
Figure 19. Comparison of 1,000 realizations of net infiltration using the linear model in GoldSim
with the results of the 50 HYDRUS simulations of infiltration. ................................ 76
Figure 20. A comparison of predictions (fitted values) from the linear and quadratic regression
models, against the HYDRUS results for net infiltration used in the model fitting. ... 94
Figure 21. A comparison of predictions from the linear and quadratic regression models based on
the input values used for the 50 HYDRUS runs. ...................................................... 95
Figure 22. Comparison of predicted net infiltration rates at the inputs used with the 50 HYDRUS
runs for the linear regression model and the “exponential” model suggested by
UDEQ. .................................................................................................................... 96
Figure 23. Re-creation of information in UDEQ Figure B-2 showing complementary cumulative
distribution functions (CDFs) for the 50 α values used in the HYDRUS realizations
and 1000 values drawn from the distribution used in GoldSim for the Clive DU PA
Model v1.4. ........................................................................................................... 103
Figure 24. Re-creation of information in UDEQ Figure B-3 showing complementary cumulative
distribution functions (CDFs) for the 50 n values used in the HYDRUS realizations
and 1000 values drawn from the distribution used in GoldSim for the Clive DU PA
Model v1.4. ........................................................................................................... 103
Figure 25. Re-creation of information in UDEQ’s Figure B-4 showing complementary
cumulative distribution functions (CDFs) for the 50 net infiltration values from the
HYDRUS realizations and 1000 values drawn from the distribution used in GoldSim
for the Clive DU PA Model v1.4. ........................................................................... 104
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 vii
Figure 26. Comparison of Bingham Environmental (1991) water content data with water content
calculated using the regression equation for the DU PA GoldSim model and with the
results of the 20 HYDRUS simulations. ................................................................. 110
Figure 27. Time series of infiltration into the waste zone for one of the 20 HYDRUS-1D
simulations. ........................................................................................................... 115

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 viii
TABLES
Table 1. Texture class, thickness range, and average thickness for the hydrostratigraphic units
underlying the Clive site. ......................................................................................... 12
Table 2. Estimated Pearson’s correlation coefficients between Ks and α on the log scale by soil
textural class, and associated 95% confidence intervals, calculated using the data
provided by Benson in 2017 data. ............................................................................ 28
Table 3. Percentiles associated with elicited information for the Ks distribution (cm/day), and the
distribution actually used. These are based on a lognormal distribution with geometric
mean of 3.37 cm/day and a geometric sd of 3.23 cm/day, with shifts for the minimum
associated with each row. See Figure 13 for example R code to get the percentiles. . 50
Table 4. Minerals in Unit 4 soil clays. ....................................................................................... 67
Table 5. Coefficients calculated from multiple linear regression models. ................................... 75
Table 6. Parameter sets of van Genuchten α, n, and Ks used for HYDRUS modeling. ............... 80
Table 7. Results of 50 flow realizations described in Appendix 5 of DU PA Model v1.4. .......... 98
Table 8. Water Content Data from Table 6 of Bingham Environmental (1991). ....................... 109

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 ix
ACRONYMS AND ABBREVIATIONS
ACAP Alternative Cover Assessment Program
bgs below ground surface
BSC biological soil crust
CCD complementary cumulative distribution
CSU Colorado State University
DEQ (Utah) Department of Environmental Quality
DOE (United States) Department of Energy
DU depleted uranium
DWMRC Division of Waste Management and Radiation Control
DWR (Utah) Division of Water Rights
ET evapotranspiration
HAL Hansen, Allen, and Luce
HSU hydrostratigraphic unit
LLRW low-level radioactive waste
LM DOE’s Office of Legacy Management
MAUP modifiable areal unit problem
NOAA National Oceanographic and Atmospheric Administration
NRC (United States) Nuclear Regulatory Commission
NRCS Natural Resources Conservation Service
PA performance assessment
PPA probabilistic performance assessment
PDF probability density function
PET potential evapotranspiration
RML Radioactive Material License
RMSE root mean squared error
sd standard deviation
SER Safety Evaluation Report
SWCC soil water characteristic curve
UDEQ Utah Department of Environmental Quality
USDA United States Department of Agriculture
WSS (NRCS) Web Soil Survey

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 1
1.0 Overview and Conceptual Model
This document begins with three sections that provide necessary background information for the
interrogatories and responses on the subject of the proposed evapotranspiration (ET) cover
design. Section 1.1 offers a discussion of the basis for the statistical approaches applied to
probabilistic PA modeling. This includes fitting distributions to data or elicited information,
scaling, and model abstraction. Section 1.2 is a summary of the setting of the Clive Site that
includes a description of the cell design, local hydrogeology, climate, and vegetation. Section 1.3
is a brief summary of the features of the variably saturated flow model HYDRUS-1D used to
develop models for net infiltration and water content in the GoldSim DU PA model. The
implementation of the software for modeling flow in the cover is also summarized in this section.
Recharge is an important process in controlling the release of contaminants to the groundwater
pathway. Site characteristics influencing movement of water from precipitation through the
vadose zone to the water table at the Clive Site include climate, soil characteristics, and native
vegetation. Engineered barriers are used at the Clive Site to control the flow of water into the
waste.
A hydrologic model of the waste disposal system must realistically represent precipitation, the
source of water to the system, runoff, evaporation, transpiration, and changes in storage to
estimate the flow through the system. Under natural conditions, plants remove water from the
upper soil zone through root uptake and transpiration, reducing the water available for seepage
deeper into the profile. The same processes occur in an engineered cover layer that has been
revegetated. Seepage through a cover system can occur when soils become wet enough to
increase their conductivity to water. Cover surface layers with adequate storage capacity can
hold the water in the near surface until it can move back into the atmosphere through
evaporation, reducing the seepage of water to the waste (net infiltration).
These processes would be expected to show temporal variability at the Clive Site on the time
scale of minutes to hours in the near surface and days to years deeper in the disposal cell.
Processes that tend to change cover properties such as plant and animal activity and climate
influences (e.g. frost heave, erosion) are expected to be slowed by the effects of eolian
deposition.
1.1 Modeling for Probabilistic Performance Assessments
Introduction
The Clive DU PA Model is a probabilistic model; consequently, its inputs are specified as
probability distributions. The PA Model results depend on both the model structure and the
model specification through the input probability distributions; consequently, it is critical that the
specification addresses underlying statistical principles that relate to this type of long-term
dynamic modeling. The sensitivity analysis for the PA Model also depends on the specific
probabilistic input, in which case identification of the appropriate important, or sensitive,
parameters also requires correct specification of the input probabilistic distributions. The
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 2
statistical aspects of this type of dynamic probabilistic modeling are critical to the success and
defensibility of the model results.
A crucial piece of developing and evaluating distributions is understanding what the values from
the distributions should, or do, represent. This understanding is necessarily linked to how values
of the input parameters are used in the PA Model, and to the assumptions needed to use available
data/information to inform the distributions. A challenge for evaluating and developing
distributions is identifying the appropriate temporal and spatial scales that need to be
represented. The data/information to support specification are typically available at different
scales than those required by the model. The following provides a more detailed discussion of
the implications of this “scaling” on the development of input distributions for the Clive DU PA
Model, and is intended to supplement the information provided in Appendix 14 of the Clive DU
PA Final Report (particularly Section 6.0, titled Scaling and Model Abstraction). Appendix 14 of
the Clive DU PA Final Report contains supporting information for the probability distributions—
and methods to obtain those distributions—used to develop the distributions for the Clive PA.
Probabilistic PA
The probabilistic Clive DU PA Model uses probability distributions to represent the current state
of knowledge in the PA Model inputs, with the goal of running the PA Model at many sets of
plausible inputs. The approach is designed to provide information regarding the expected
performance of the Site. In general, the modeling effort should be focused on running the PA
Model under different realistic scenarios defined by different combinations of values of the input
parameters. Judging what is realistic is the fundamental challenge, and it depends on the spatial
and temporal scale represented by the values of the input in the PA Model.
This is in contrast to testing the potential performance of the Site “deterministically” by using a
single value to represent each input parameter. In addition, probabilistic modeling is aimed at
modeling what is thought to be known and the uncertainty in that knowledge, whereas
deterministic models are often run using extreme, and likely unrealistic, input parameter values
that are usually described as a conservative or protective. A problem with so-called conservatism
in PA models is that the direction of conservatism is often difficult to justify. Conservative
models are also difficult to explain because they are knowingly wrong, and thus use of
conservatism leads to poor decision making. If there is a desire to make a conservative
(protective) decision, then those value judgments should be separated from the PA model, and
should be based on a model that explains the system response as best possible.
The PA model uses time steps to propagate movement of radionuclides through the engineered
and environmental system to project concentrations in various media at various future points in
time. However, for a single simulation, random numbers are drawn from the input distributions
at the beginning of time, and are used throughout time (thousands of years, or more). This means
that the input distributions must represent long time periods, and the data/information available
must be scaled appropriately to represent those long time periods. The same is the case for the
spatial scale of the model; the available data/information often represent small spatial scales, and
must be adjusted to represent the large areas or volumes that underlie the PA model. Each
simulation is, essentially, a deterministic run of the PA model. Uncertainty in the output is
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 3
obtained by running many (thousands of) simulations, each one of which has a distinct set of
input values drawn at random from the underlying distributions of the input parameters.
For example, using a value that is extreme at a small temporal scale (such as a daily soil
temperature in the summer) implies that it is realistic to hold that input constant at such an
extreme for the entire time period represented by a run of the PA model (thousands of years or
more). Therefore, it is crucial to question whether a value is “plausible” or “realistic” to
represent a constant condition over a long period of time or a large spatial area. A probabilistic
PA provides a natural framework for running the model at sets of values that are deemed realistic
for representing long-term conditions over the spatial extents represented in the model. If this is
accomplished, the model outputs can be interpreted as indicative of expected performance of the
entire site under realistic long-term conditions, given the current state of knowledge in input
parameters and the assumptions underlying the PA model.
The Importance of Scaling and Its Challenges
All models are approximations of reality, and the assumptions should always be evaluated and
discussed. However, given the assumptions of a PA model, the distributions should be developed
to explicitly match, as much as possible, the assumptions and design of the model. The use of
distributions representing values at a different scale than that represented by the model is not
consistent with the fundamental goals underlying probabilistic PA modeling. Therefore, model
developers and reviewers must ensure they are considering the problem relative to the same
spatial and temporal scales in order to have a meaningful and productive conversation and to
ultimately improve the distributions, and hence models, used in a probabilistic PA.
Developing an appropriate distribution for a PA model input parameter is not simply an exercise
in approximating a histogram of available data thought to be relevant to the site. It is unlikely
that available data, information from the literature, and/or expert knowledge exist at spatial and
temporal scales consistent with the design of the PA model. Therefore, an additional step in
distribution development is taking the available information and “scaling” it appropriately; this
typically means taking available values that are representative of relatively small scales (e.g.,
locations in space and/or points in time) and combining them to represent larger scales, typically
through averaging.
Identifying a scale consistent with the PA model requires in-depth understanding of how the
values from the distribution will ultimately be used in the PA model. That is, there should be an
explicit connection between what the values are assumed to represent within the PA model and
the distribution developed to produce those values. To accomplish this, the values coming from
the distribution should be aligned with the spatial and temporal scales upon which the PA model
is built; that is, they should represent the spatial and temporal scales represented by a value of the
input. Tangible examples are provided later in this discussion. If the scale represented in the
distribution does not match that used for the model, then the results from the model cannot be
interpreted as representing the performance of the site under realistic conditions consistent with
the model assumptions.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 4
The Change-in-Scale Problem
A mismatch in spatial and/or temporal scale between available data and what is needed to inform
a model, or to make predictions from a model, is a problem that arises in many disciplines. It is
not a new problem or a problem unique to PA work, and it has different names in different
disciplines such as change-of-support, change-of-scale, downscaling and/or upscaling, and the
modifiable areal unit problem (MAUP) in the geography literature (see Cressie and Wikle
(2011)). The need to match the scale of available data to that needed for the model or predictions
is fundamental to problems with temporal and/or spatial components. Higgs et al. (2017) offer a
general description of the problem within the context of PA work.
Understanding the Scales Represented in the PA Model
The first step in understanding what spatial and temporal scales a distribution should represent is
asking questions about how a value drawn from that distribution will ultimately be used in the
PA model. In the current implementation of the Clive DU PA Model in GoldSim, volumes of
material at the Site are represented as a network of “cells,” and the material represented by a
single cell is assumed to be completely mixed (homogeneous). For a single realization of the
Model, a property of the entire volume of material/soil associated with a cell is represented by
the value of input parameter drawn from a distribution for that property. Therefore, a value from
a distribution used to specify a property of material within a cell should represent the entire
volume of completely mixed materials within the cell. If the value is used to represent properties
of multiple cells, then the value should represent the larger spatial scale associated with the
volume of the combined cells. What values are appropriate to represent the entire volume of
material? How should available information, collected at various scales from the Site or
elsewhere, be used to inform the distribution of such values?
A value drawn from a distribution is typically held constant over the entire time period of a PA
model run or realization; it is usually not changed over time within a particular realization.
Therefore, the distribution should provide plausible values for summarizing the input parameter
over thousands of years (or longer), such as a long-term average. For example, even if values of
the input parameter may vary seasonally, that does not mean it is reasonable to run the model as
if a seasonal extreme is held constant for 1000 years. It is most realistic, and thus informative of
expected performance, to run the model at plausible long-term averages, not at short-term
extremes. Running the model at values representative of smaller time scales (such as individual
seasons) results in many unrealistically extreme realizations that will not adequately characterize
expected site performance over the period of interest. Over a long period of time, the input
variable is expected to be below average and above average during small periods of time, but the
performance is assessed over such large time scales that the extremes should “balance each other
out” given the model structure. Data may only be available over short time periods, such as
months, but distributions for longer-term averages can be developed using the short-term data
and assumptions regarding the future.
The relatively large spatial and temporal scales of input parameters represented in the PA model
result in challenges for development and evaluation of distributions. It is often not intuitive how
to use the information from available data, or from subject matter knowledge of “possible
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 5
values” for an input parameter, to develop a distribution representing the scale consistent with
the PA model. For example, it is tempting to assume that any value that could possibly be
obtained from a single sample from one location at one point in time should be represented in the
distribution. However, such an approach is generally misaligned with the scale of the PA model,
as described previously. This idea is further explored in the following discussion, with spatial
and temporal dimensions considered separately, and in more detail.
Spatial Scaling Example
Suppose data are collected on material properties from 20 soil cores taken at random locations
from around the actual site. The variability in the resulting values from the samples will reflect
the heterogeneity across the site. An individual sample represents the spatial scale of a point
location (or small core), and a distribution fit directly to the observed values from all the cores
could be used to represent possible values that could be obtained from other potential cores taken
from the same site. Now, suppose values that reflect the entire volume of soil of the site are of
interest, and need to be obtained from the 20 soil cores. Is it reasonable to use a value
representing a single core as if it represents the larger volume of soil? In general, it is not
reasonable for a value from an individual sample to represent the larger volume; instead, an
average (or estimated mean) would typically be used as a representative value of the larger
volume. The idea can be thought of in terms of prediction; what value is reasonable to use to
predict the same property of a different site of the same size: the value from an individual core or
the average over all the cores sampled? Choosing the estimated mean will clearly give better
predictions and should have smaller errors on average (uncertainty) and less variability in errors.
A single data set has an associated estimate of the mean, but different estimates would be
obtained from different sets of 20 locations sampled. That is, there is uncertainty in the value of
the mean because the entire volume of material represented in the PA model will never be
sampled. Therefore, a probability distribution is used to incorporate uncertainty in the mean
rather than variability among values from different point locations; the distribution should
represent plausible means for the entire volume of material, rather than variability among
measurements from cores taken from individual locations within a site. Using values from a
distribution describing variability among individual small-scale locations to represent the entire
volume of material leads to extreme, and unrealistic, model runs. Practically, this often translates
into using standard errors for the estimated mean over a larger scale to define distributions, rather
than standard deviation of available small-scale measurements.
Another way to think about the problem is through the assumption of complete mixing used in
the conceptual construction of the PA model cells. First, envision the 20 samples described in the
previous paragraph. Now, suppose a huge mixer is employed to completely mix the entire
volume of material/soil before the 20 samples are taken. What would variability among the
values from these new 20 samples look like compared to variability among the first 20 samples
from the heterogeneous material? Because the soil is thoroughly mixed, the relatively large
values and small values are now dispersed through the whole volume so that the samples will
now all have values relatively close together (the more completely the material is mixed, the
closer they will be). The average value of the 20 samples in both cases (heterogeneous vs.
mixed) should be the same (or close to the same), but the variability may be drastically different.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 6
The point estimate of the mean from the original 20 samples can be used as the center for the
desired distribution, with the variance adjusted to reflect uncertainty in that mean because it is
based on a limited number of samples.
Temporal Scaling Example
The time scale associated with the input parameters in a PA model is the length of time over
which a value from a distribution is held constant in a run of the model. The Clive DU PA Model
most often uses a single value, drawn from an input distribution at the beginning of a model
realization, over the whole time period represented by the Model. Therefore, it must represent
the spatial volume (described in the previous section) over thousands of years (or longer). This is
obviously challenging because available data exist at much, much smaller temporal scales (less
than a day to perhaps 50 years) and an assumption must be made that current knowledge is
projected into the future. The lack of knowledge about the future requires making assumptions
about future conditions, and typically the assumption is made that the conditions under which
data were collected will hold into the future (i.e., stationarity). This implies the mean and
associated variance will stay constant into the future represented by the PA Model. Under this
assumption, information in the available data and expert knowledge can be used to directly
inform the distribution; uncertainty and lack of trust in assumptions can also be incorporated.
As discussed for the spatial scale, the distribution should represent the current state of knowledge
about the input at the PA model scale, meaning random draws from the distribution should
capture variability among plausible long-term means for the volume of material represented by
the PA model cell(s). Using soil temperature as an example, consider the variability among daily
average soil temperatures in a region with seasonal variation, with extremes captured by the
daily average soil temperatures during the coldest time of the year, and the daily average soil
temperatures during the warmest part of the year. Now, contrast that with the extremes expected
across different annual average soil temperatures; one extreme is a year that is colder on average
over all seasons and the other extreme is a year that is warmer on average over all seasons. The
difference between these two annual extremes is expected to be far less than the difference
between the two extremes defined on the daily scale. Is it reasonable to use an extreme daily
average to represent an extreme annual average soil temperature? Using a daily average as an
annual average is equivalent to representing incredibly extreme conditions that are not supported
by available data—it amounts to assuming that conditions in the system are going to change so
drastically that daily averages become representative of annual averages.
These ideas continue to larger scales, where, at each larger scale, less variability is expected
among averages at that scale. In other words, we expect less variability among decadal averages
than annual averages, and less variability among 50-year averages than decadal averages, and so
on. The degree to which the distributions representing the current state of knowledge in larger
and larger scales get narrower does depend on the trust in the assumption of stationarity of
conditions into the future. However, regardless of the proposed difference in variance between a
distribution of annual averages and a distribution of 100- or 1000-yr averages, it is clear that
seasonal variation occurs at a scale much smaller than that represented by the PA model. It is not
appropriate to assume that a value measured at a location at one point in time should be
represented in the collection of draws from a distribution for that input used for the PA model
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 7
when the value drawn from the distribution is held constant over thousands of years (or longer).
When evaluating the reasonableness of an input distribution, given the modeling conditions, one
should ask “Are these values plausible given they will be held constant in the model over
hundreds or thousands of years?”
Combining Spatial and Temporal Scaling
As described above, from a temporal perspective the values captured in a distribution should be
judged reasonable in the context of being held constant over long time periods, and therefore
should not reflect extremes that are only realistic at smaller time scales. Thinking about plausible
long-term averages for the values representing the spatial volumes of the PA model is the easiest
way to judge reasonableness of the distribution relative to time. However, this is challenging
because it is clearly impossible to collect data over hundreds or thousands of years into the future
to inform the distribution of a long-term average.
Therefore, temporal scaling is often more abstract than the methods discussed for spatial scaling,
and often must actually use mainly spatial variability as a surrogate for variability that might be
observed over time. For example, using data from different locations and regions captures
conditions in different materials that may be observed at one location over time. Therefore, the
heterogeneity over space is used as a surrogate for heterogeneity over time for the volumes
represented by the PA model.
This is necessary if distributions are to be based on empirical information because it is not
possible to collect the data needed to capture the heterogeneity over time into the future. While
the scaling does appear to be simply “spatial scaling” on the surface, it is not a misnomer to call
it spatio-temporal scaling because both scales are always simultaneously considered even if most
of the information available is spatial. The resulting distributions should be evaluated relative to
both dimensions by asking “does the distribution provide values that are plausible when applied
to the spatial volumes of PA cells over long periods of time?”
Choice of Statistical Methods
There are many statistical methods that can be used to estimate a mean, along with associated
uncertainty; the choice depends on the method of data collection (e.g., sampling design) and
characteristics of the data such as dependence (e.g., correlation) among values from different
samples. The average (arithmetic mean) is the simplest, and is often appropriate, as long as
samples do not severely violate underlying linearity and stationarity assumptions.
If direct averaging applies, then the scaled distribution can often be represented by a normal
distribution (depending on the nature of the underlying data). This is because sampling
distributions of averages tend to be approximately normally distributed, even when the
population of values from which the samples are taken are not normal, and normality improves
with increasing sampled sizes. The t-distribution is also an option to account for uncertainty in
estimating the standard deviation of the population, but the t-distribution is barely
distinguishable from the normal distribution for sample sizes greater than 30, and the normal
distribution is simpler to implement and communicate about for a PA model.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 8
The data distribution has a mean and a standard deviation, whereas the scaled distribution
obtained as the distribution of the average has the same mean and a standard deviation of the
mean, which is usually called the standard error of the mean.
Scaling Challenges Related to Sources of Information
If samples are available directly from the volume of material of interest (represented by a PA
model cell), then the development of the distribution at the appropriate scale based on smaller-
scale measurements is a straightforward statistical exercise. However, if the information
available to inform the distribution is composed of measurements from other sites, under
different experimental conditions, representing predictions from other estimated relationships,
from expert judgment, etc., then distribution development becomes much more challenging and
must rely on professional and expert judgment decisions to factor in uncertainty regarding what
measurements might be from the site of interest and to gauge how different sources of
information should be weighted against each other in the distribution development process.
In such cases, assumptions and decisions should be explicitly stated, and it is not reasonable that
every rational expert would provide the same distribution. It is rare that many samples are taken
from the site to inform a distribution, and the variety of types of information and “data” available
certainly add to the challenges of representing current state of knowledge in the mean of a spatial
volume at the site. Different values and summaries presented in the literature may represent
different spatial scales, and this information should be used, if possible and if available.
Additionally, development of distributions is often complicated by limited site-specific
information over space and over time. This limited information must be transformed to represent
the scales consistent with the model, after taking into account the quality of the different data
sources and their relevancy to the site, and the assumptions of the PA model.
Conclusions
While all rational experts may not develop the exact same distribution, they should all be
critically examining the distributions with the same spatial and temporal scales in mind, given
the PA model in question. The relevant question to ask is not “Does the distribution cover all
plausible values of the input variable that could be observed from a single sample?” but instead
“Does the distribution cover all plausible values of the input variable that could be representative
of the large volume of material represented by the PA model cell over long periods of time?”
These questions have different answers because they are tied to different spatial scales, and
answers depend on the current state of knowledge regarding site conditions, including linearity
and stationarity.
Development of distributions for PA model inputs would certainly be easier if each distribution
could simply represent possible values that have been observed in data collected from single
locations and single points in time. However, running the PA model with such distributions
would result in many of the realizations of a PA model capturing unrealistically extreme
scenarios (e.g., an extreme daily average temperature of 90 degrees assumed to be held constant
over not only a year, but thousands of years). The problem as described for a single distribution
magnifies as the number of distributions with spatial and temporal scales misaligned with the PA
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 9
model increases, possibly to the point where only a small proportion of the PA model
realizations represent realistic and useful outcomes.
The goal of a probabilistic PA is to represent expected outcomes of the site. Using distributions
at temporal and spatial scales misaligned with the representations built into the PA model makes
this goal unattainable.
If there is interest in running the model at extreme, unrealistic scenarios to test hypothetical
limits of the model and the associated performance of the site, then this can be done in a separate
investigation, but this should not be built into the distributions meant to capture the current state
of knowledge at a scale consistent with the current PA model. Distributional goals are
necessarily tied to spatial and temporal scale, and therefore developers and reviewers must
ensure they are considering the problem relative to the same spatial and temporal scales in order
to have a meaningful and productive conversation about the input distributions. This leads to a
defensible model, and a sensitivity analysis that yields meaningful results that can ultimately be
used to direct further data/information collection to improve, or refine, the distributions used in a
probabilistic PA.
Further examples of scaling in the Clive DU PA are provided in Appendix 14 to the Clive DU
PA Final Report.
1.2 Setting
1.2.1 Disposal Cell Design
The design of the Federal DU Cell is a covered embankment, with relatively steeper sloping
sides nearer the edges. The upper part of the embankment, known as the top slope, has a
moderate slope (2.4%), while the side slope is markedly steeper (20%). The embankment is also
constructed such that a portion of it lies below-grade. The overall length of the embankment is
1317.8 ft and the overall width is 1775.0 ft. A detailed description of embankment dimensions
and a discussion of the representation of the Federal DU Cell in the GoldSim model are provided
in Embankment Modeling for the Clive DU PA Model (Neptune 2015g).
Disposal involves placing waste on a prepared clay liner that is approximately 8 ft below the
ground surface. For the Federal DU Cell design, the depth of the waste below the top slope is a
maximum of 47.5 ft (14.5 m). A cover system is constructed above the waste. The objective of
the cover system is to limit contact of water with the waste, limit biointrusion, and protect the
lower layers of the cover from freezing. The cover is sloped to promote runoff and is designed to
limit water flow by increasing evapotranspiration (ET). The arrangement of the layers used for
the ET cover design is shown in Figure 1. Beginning at the top of the cover, the layers above the
waste used for the ET cover design are:
• Surface Layer: This layer is composed of native vegetated Unit 4 silty clay material with
15 percent gravel mixture on the top slope and 50 percent gravel mixture for the side
slope. This layer is 6 inches thick. The functions of this layer are to control runoff,
minimize erosion, and maximize water loss from ET. This layer provides storage for
water accumulating from precipitation events, enhances losses due to evaporation, and
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 10
provides a rooting zone for plants that will further decrease the water available for
downward movement.
• Evaporative Zone Layer: This layer is composed of Unit 4 material. The thickness of
this layer is 12 inches. The purpose of this layer is to provide additional storage for
precipitation and additional depth for the plant rooting zone to maximize ET.
• Frost Protection Layer: This material ranges in size from 16-inch cobbles to clay-size
particles. This layer is 18 inches thick. The purpose of this layer is to protect layers below
it from freeze/thaw cycles and wetting/drying cycles, and to inhibit plant, animal, or
human intrusion.
• Upper Radon Barrier Layer: This layer consists of 12 inches of compacted clay with a
low hydraulic conductivity. This layer has the lowest conductivity of any layer in the
cover system. This is a barrier layer that reduces the downward movement of water to the
waste and the upward movement of gas out of the disposal cell. The design specification
for saturated hydraulic conductivity (Ks) of this layer is 5.00 × 10-8 cm/s (Whetstone
Associates (2011), Table 15).
• Lower Radon Barrier Layer: This layer consists of 12 inches of compacted clay with a
low hydraulic conductivity. This is a barrier layer placed directly above the waste that
reduces the downward movement of water and the upward movement of gas out of the
disposal cell. The design specification for Ks of this layer is 1.00 × 10-6 cm/s, from Table
15 of Whetstone Associates (2011).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 11

Figure 1. Evapotranspiration (ET) cover profile showing materials, observation nodes, and
root distribution used in the HYDRUS-1D models.
1.2.2 Unsaturated Zone and Shallow Aquifer
The following description of the Clive Site hydrology is taken from the review prepared by
Envirocare (2004). The Site is described as being located on lacustrine (lake bed) deposits
associated with the former Lake Bonneville. The sediments underlying the facility are principally
interbedded silt, sand, and clay. Sediments at the Site are described by Bingham Environmental
(1991, 1994) and Envirocare (2000, 2004) as being classified into four hydrostratigraphic units
(HSU). Predominant sediment textural class, layer thickness range, and average layer thickness
for each unit are listed in Table 1. A diagram of the unsaturated zone is shown in Figure 2.
Unit 4: This unit begins at the ground surface and extends to between 6 ft and 16.5 ft below the
ground surface (bgs). The average thickness of this unit is 10 ft. This unit is composed of finer
grained low permeability silty clay and clay silt.
Unit 3: Unit 3 underlies Unit 4 and ranges from 7 ft to 25 ft in thickness. The average thickness
of this unit is 15 ft. Unit 3 is described as consisting of silty sand with occasional lenses of silty
to sandy clay.
Unit 2: Unit 2 underlies Unit 3 and ranges from 2.5 ft to 25 ft in thickness. The average
thickness of this unit is 15 ft. Unit 2 is described as being composed of clay with occasional silty
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 12
sand interbeds. A structure map was prepared by Envirocare (2004) (their Figure 5) with
contours representing the elevations of the top of the unit. This map shows that the top surface of
Unit 2 slopes downward gradually from east to west in the vicinity of the Class A South cell.
Unit 1: Unit 1 underlies Unit 2 and is saturated beneath the facility, containing a locally confined
aquifer. Unit 1 extends from approximately 45 ft bgs and contains the deep aquifer. The deep
aquifer is reported to be made up of lacustrine deposits consisting of deposits of silty sand with
some silty clay layers. One or possibly more silty clay layers overlie the aquifer (Bingham
Environmental 1994).
Table 1. Texture class, thickness range, and average thickness for the hydrostratigraphic
units underlying the Clive site.
Unit Sediment Texture Class Thickness
Range (ft)
Average
Thickness (ft)
4 silt and clay 6–16.5 10
3 silty sand with interbedded silt and clay layers 7–25 15
2 clay with occasional silty sand interbeds 2.5–25 15
1 silty sand with interbedded clay and silt layers >620 >620

Figure 2. Stratigraphic profile showing ET cover, waste zone, and stratigraphy below the
Federal DU Cell.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 13
The aquifer system in the vicinity of the Clive Site is described by Bingham Environmental
(1991, 1994) and Envirocare (2000, 2004) as consisting of unconsolidated basin-fill and alluvial
fan aquifers. Characterization of the aquifer system is based on subsurface stratigraphy
observations from borehole logs and from potentiometric measurements. The aquifer system is
described as being composed of two aquifers: a shallow, unconfined aquifer and a deep confined
aquifer. The shallow unconfined aquifer extends from the water table to a depth of
approximately 40 ft to 45 ft bgs. The water table in the shallow aquifer is reported to be located
in Unit 3 on the west side of the Site and in Unit 2 on the east side.
The deep confined aquifer is encountered at approximately 45 ft bgs and extends through the
valley fill (Bingham Environmental 1994). The boring log from a water supply well drilled in
adjoining Section 29 indicates continuous sediments to a depth of 620 ft bgs as described in Utah
Division of Water Rights water right number 16-816 and associated well log 11293 (DWR
2014).The deepest portion of the basin in the Clive area is believed to be north of Clive in Ripple
Valley, where the basin fill was estimated to be 3,000 ft thick (Baer and Benson as cited in Black
et al. (1999)).
Deeper saturated zones in Unit 1 below approximately 45 ft bgs are reported to show higher
potentiometric levels than the shallow unconfined aquifer. Differences in potentiometric levels
are attributed to the presence of the Unit 2 clays. These observations are interpreted as indicating
that the shallow unconfined aquifer below the Site does not extend into Unit 1 but is contained
within Units 2 and 3 (Bingham Environmental 1994). The aquifer systems are described in more
detail in the Saturated Zone Modeling white paper (Neptune 2015a).
Recharge to the shallow aquifer in the vicinity of Clive is thought to consist of three components:
a small amount is due to vertical infiltration from the surface, some small amount of lateral flow
is from recharge areas to the east of the Site, and the majority of recharge is believed to be from
upward vertical leakage from the deeper confined aquifer (Bingham Environmental 1994).
Average annual groundwater recharge from the surface in the southern Great Salt Lake Desert in
the precipitation zone typical of Clive was estimated by Gates and Kruer (1981). An estimated
300 acre-feet per year were recharged to lacustrine deposits and other unconsolidated sediments
over an area of 47,100 acres. This is a recharge rate of approximately 0.08 in/yr (2 mm/yr).
Groundwater recharge from lateral flow occurs due to infiltration at bedrock and alluvial fan
deposits away from the Site, which moves laterally through the unconfined and confined aquifers
(Bingham Environmental 1994). This is evidenced by the increasing salinity of the groundwater
due to dissolution of evaporate minerals as water moves from the recharge area to the aquifers
below the Site facility (Bingham Environmental 1994). The majority of recharge to the shallow
aquifer is believed by Bingham Environmental (1994) to be due to vertical leakage upward from
the deep confined aquifer due to the presence of upward hydraulic gradients.
The higher potentiometric levels in the deeper saturated zones in Unit 1 described previously are
attributed to the presence of the Unit 2 clays (Bingham Environmental 1994). Vertical gradients
between shallow and deeper screened intervals in the monitor well clusters were calculated by
Bingham Environmental (1994). An upward vertical gradient was observed ranging in magnitude
from 0.02 to 0.04 based on the distance between the screen centers. For a vertical hydraulic
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 14
conductivity of 1 × 10-6 cm/s (Bingham Environmental 1994), this corresponds to a recharge
range from 0.25 in/yr to 0.5 in/yr (6.35 mm/yr to 12.7 mm/yr).
1.2.3 Climate
Precipitation measurements taken at the Site over the 17-year period 1992 to 2009 show a mean
annual value of 8.53 inches (21.7 cm) (Whetstone Associates 2011). Precipitation exceeds the
annual average from January through June and again in October and is below average for the
remaining months. The nearest National Oceanographic and Atmospheric Administration
(NOAA) station with a long-term record is located in Dugway, Utah, approximately 40 miles to
the south. The mean annual precipitation for the same 17-year period measured at the Dugway
station is 8.24 inches (20.9 cm). A comparison of the Dugway precipitation data for the 17-year
period 1992 to 2009 with the long-term average for Dugway was made by Whetstone Associates
(2011). This comparison indicated that annual average precipitation at the Site during this 17-
year period has been greater than the long-term average at Dugway by 8 percent. Whetstone
Associates (2011) concluded that simulations of cover performance using precipitation data from
this 17-year period might be overestimating this component of the site water balance.
The HYDRUS-1D modeling performed is based on the 17-year record for consistency with the
modeling results reported in Whetstone Associates (2011). However, an additional 2 years of
monthly precipitation data are available from MSI (2012). The 19-year average precipitation is
8.62 inches (21.9 cm). This difference is driven primarily by the 4.28 inches of rainfall in May
2011. The small change in the overall average suggests that the modeling results presented for
this analysis would not change significantly if the 19-year precipitation record had been used
instead of the 17-year record.
The close correspondence between mean monthly temperatures measured at the Clive Site and
the Dugway NOAA station was demonstrated by Whetstone Associates (2011). Average
monthly temperatures measured at the Clive Site over the 17-year period 1992–2009 ranged
from 27.7 °F in December to 79.5 °F in July.
1.2.4 Vegetation
Actual transpiration is dependent on the characteristics of the plant communities at the Site.
Vegetation cover at the Site is less than 20 percent, with soils supporting a range of native and
invasive shrubs. Excavations at the Site have shown plant rooting depths extending to
approximately 31 inches (80 cm) below the ground surface, with root density decreasing with
depth (SWCA 2011).
Vegetation surveys of three field plots on or adjacent to the Clive Site were conducted by SWCA
(2011). The three low desert vegetation associations were characterized as: black greasewood,
Plot 3; halogeton-disturbed, Plot 4; and shadscale-gray-molly, Plot 5. The dominant shrub in Plot
3 was black greasewood with a percent cover of 4.5% and the dominant forb was halogeton with
a percent cover of 0.7%. In Plot 4, the dominant shrub was shadscale saltbush with a percent
cover of 2.3%, and the dominant forb was halogeton with a percent cover of 3.3%. In Plot 5, the
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 15
dominant shrub was shadscale saltbush with a percent cover of 12.5%, and the dominant forb
was halogeton with percent cover of 0.9%.
Black greasewood, shadscale saltbush, and halogeton are all classified as facultative halophytes
(Anderson 2004; Pavek 1992; Simonin 2001). Facultative halophytes are known to benefit from
high salt concentrations in their growth media (Shabala 2013). Halophytes are able to adjust to
saline environments through various physiological adaptations such as compartmentalization of
ions in cell vacuoles, succulence, and the elimination of salt through salt-secreting glands and
bladders (Shabala 2013). Optimal growth for halophytes has been demonstrated by Shabala
(2013) to occur in media with a concentration of approximately 50 mM NaCl for monocots, and
between 100 and 200 mM for dicots. For the optimum range for dicots of 100 to 200 millimoles
per liter (mM), the corresponding range of electrical conductivity for a NaCl solution is 9.5 to
18.4 mmho/cm (https://www.wolframalpha.com/).
Depending on the extent of the area defined on and adjacent to the Clive Site, approximately 80
to 90 percent of the soils are mapped by the Natural Resources Conservation Service (NRCS) as
the Skumpah on 0 to 2 percent slopes (NRCS 2016). This Unit is characterized as having
maximum salinity ranging from 8.0 to 16.0 mmhos/cm. The top end of this range of maximum
salinity does not exceed the maximum of the range of salinity considered optimum for halophyte
growth of 18.4 mmho/cm. Given the similarity in ranges of salinity in the surface soils at the
Clive Site and that needed for optimum halophyte growth, the influence of the osmotic head
reduction in the root-water uptake water stress response function is considered negligible and
was, consequently, not included in the model.
1.3 HYDRUS-1D Software Package
HYDRUS-1D (Šimůnek et al. 2013) was selected for simulating the performance of the ET
cover proposed for the DU waste cell. The HYDRUS-1D platform was selected for this project
because of its ability to simulate processes known to have a significant role in water flow in
landfill covers in arid regions. HYDRUS includes the capabilities to simulate:
• water flow in variably saturated porous media,
• material hydraulic property functions,
• atmospheric surface boundary conditions including precipitation and evapotranspiration,
• root water uptake, and
• free-drainage boundary conditions.
The flow component of unsaturated flow and transport software packages with atmospheric
boundary conditions such as HYDRUS solves modified forms of the Richards equation for
variably saturated water flow. The flow equation incorporates a sink term to account for water
uptake by plant roots. HYDRUS can be applied to one-, two-, and three-dimensional problems.
The HYDRUS software includes grid generators for structured and unstructured finite element
meshes. Programs such as HYDRUS require detailed data to represent the atmospheric boundary
conditions and plant responses that are the dominant influences on flow in the cover in arid and
semi-arid conditions. These programs use the infiltration capacity of the soil at any time as
calculated in the model to partition precipitation into infiltration and overland flow. HYDRUS
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 16
has been used for many applications for unsaturated zone modeling and has received numerous
favorable reviews such as Scanlon’s (Scanlon et al. 2002) review of HYDRUS-1D, Diodato’s
(Diodato 2000) review of HYDRUS-2D, and McCray’s (McCray 2007) review of the most
recent program, HYDRUS (2D/3D).
HYDRUS-1D was selected for simulating flow in the Federal DU Cell ET cover since previous
numerical modeling of flow in the similar ET cover design for the Class A West cover
demonstrated that subsurface lateral flow was not significant (EnergySolutions 2012). To test the
importance of 2-D flow effects in the ET cover design, 2-D transient flow simulations were
conducted for representative sections of the cover. The approach taken was to model a section of
the side slope in two dimensions. Representative hydraulic properties were assigned to the ET
cover layers and the models were run with daily atmospheric boundary conditions for 100 years.
The results of these 2-D simulations demonstrated that water flow in the cover system for both
designs is predominantly vertical with no significant horizontal component. These results
demonstrate that 1-D models can be used to provide a defensible analysis of cover performance
for the ET cover design due to the lack of lateral flow.
HYDRUS-1D models were developed for the evapotranspiration cover design for the Federal
DU Cell (Figure 1). Model development requires construction of a computational grid based on
the geometry of the model domain. Hydraulic properties for each layer required for the model are
available from previous studies at the Site or can be estimated from site-specific measurements
such as particle size distributions. HYDRUS requires daily values of precipitation, potential
evaporation, and potential transpiration to represent the time-variable boundary conditions on the
upper surface of the cover. Representative boundary conditions were developed from records of
nearby meteorological observations. Parameters for describing root water uptake were available
from the literature.
HYDRUS implements the soil-hydraulic functions of van Genuchten (1980), who used the
statistical pore-size distribution model of Mualem (1976) to obtain a predictive equation for the
unsaturated hydraulic conductivity function in terms of soil water retention parameters. The
expressions of van Genuchten (1980) are given by
𝜃(ℎ)=&𝜃'+𝜃)−𝜃'
[1 +|𝛼ℎ|/]1 ℎ<0

𝜃)ℎ≥0
𝐾(ℎ)=𝐾7𝑆9:[1 −;1 −𝑆9
<
1=
1
]>
𝑆9 = 𝜃−𝜃'
𝜃7 −𝜃'
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 17
where
θ is volumetric water content (–),
h is the matric potential (l),
θr is residual volumetric water content (–),
θs is saturated volumetric water content (–),
α is a van Genuchten fitting parameter (l-1),
n is a van Genuchten fitting parameter (–),
m is 1-1/n with n > 1,
K is hydraulic conductivity (l/t), and
Ks is the saturated hydraulic conductivity l/t.

The above equations contain five independent parameters: θr, θs, α, n, and Ks. The pore-
connectivity parameter “l” (lower-case L) in the hydraulic conductivity function was estimated
(Mualem, 1976) to be about 0.5 as an average for many soils. The value for l is commonly taken
to be 0.5, and this value was used for all simulations for all soil types.
2.0 UDEQ Interrogatory Responses
UDEQ has arranged the interrogatories and SER comments in the Amended and New
Interrogatories; Clive DU PA Modeling Report Version 1.4 (Utah DEQ 2017) into seven groups.
The first group named by UDEQ as ET Cover Design will be addressed in this document. This
includes Open Interrogatories CR R313-25-7(2)-05/2, CR R313-22-32(2)-10/3, CR R317-6-2.1-
20/2, CR R313-25-8(4)(d)-21/2, CR R313-25-8(4)(a)-28/3, CR R313-25-7(2)-59/2, CR R313-
25-7(3)-60/2, CR R313-25-7(1–2)-90/2, CR R313-25-7(2)-150/3, CR R313-25-8(4)(d)-153/2,
CR R313-25-7(2)-175/1, CR R313-25-8(5)(a)-176/1, CR R313-25-7(2)-189/3, and CR R313-25-
7(2)-192/3, and Supplemental Comments from Appendix B of the SER (SC&A 2015) 1, 2, 3, 4,
5, 6, 7, 8, 9, and 11.
2.1 Interrogatory CR R313-25-7(2)-05/2: Radon Barriers
DEQ Conclusion from April 2015 SER, Appendix C:
Based on several unresolved issues related to the evapotranspiration (ET) cover, DEQ indicated
in the DU PA SER that the cover design was deficient. Therefore, this interrogatory remains
open. The unresolved issues are as follows:
Evapotranspiration Cover—There are still a number of unresolved issues with respect to the
selection of parameter ranges, distributions, and correlations, as well as the modeling approach
and predicted sensitivities. These concerns are detailed in Appendix B. Further, because the
model-predicted infiltration rates will be sensitive to the hydraulic properties assigned to each ET
layer, DEQ recommends that EnergySolutions develop hydraulic properties for the cover system
based on the approach outlined by Dr. Craig H. Benson in Appendix F to this SER. Issues
related to this portion of the performance assessment cannot be closed until these concerns
have been resolved. (All references in prior interrogatories to Appendices of “the SER” refer to
the April 2015 SER (SC&A 2015)).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 18
Clay Liner—As with the ET cover, there is still an unresolved concern that Ksat values will
increase greatly over time, and that the α and Ksat values assumed for modeling flow through
the liner must either be correlated or a sensitivity analysis be conducted to demonstrate that the
lack of correlation assumed does not adversely affect the modeling results. In addition, there are
problems with assumed liner hydraulic conductivity values. Furthermore, the DU PA Model v1.2
does not account for liner degradation over time. These issues must be resolved before DEQ
can determine the adequacy of this portion of the DU PA.
DEQ Critique of DU PA v1.4, Appendix 21:
Modeling conducted for the clay liner beneath the waste should employ hydraulic parameters
representative of a compacted clay liner. Typical α, n, and Ɵs for compacted clays can be found
in Tinjum et al. (1997). Typical saturated hydraulic conductivities for clay liners can be found in
Benson et al. (1994).
Infiltration—Before the adequacy of the DU PA can be determined, additional modeling of the
ET cover infiltration rates must be conducted based on in-service hydraulic properties and
correlated log(α) and log(Ksat) values as described in Appendix E. Without this information, DEQ
is unable to conclude if the infiltration rates predicted by the DU GoldSim model are reliable or
representative of future conditions (i.e., ≥ 10,000 years).
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatories 21, 175, 176, and 189 for
discussions regarding the relationship between infiltration and the in-service hydraulic properties.
Erosion of Cover—Before the adequacy of the DU PA can be determined, EnergySolutions
needs to clarify certain issues relating to Appendix 10 to the DU PA Model v1.2 (June 5, 2014;
Neptune 2014g) as described in Section 4.4.2 of the SER. The Division of Waste Management
and Radiation Control (DWMRC) is currently reviewing a proposed ET cover test request as part
of a Stipulation and Consent Agreement to use a cover of similar design to that proposed for the
Federal Cell in the DU PA. Any recommendations and conclusions from that review will need to
be applied to the proposed Federal Cell as well.
DEQ Critique, v1.4, Appendix 21: See Interrogatories 20, 28, 160, and 191 for discussions
regarding cover erosion.
Effect of Biologicals on Radionuclide Transport—EnergySolutions has not shown that the
cover system is sufficiently thick or designed with adequate materials to protect the cover system
or the underlying bulk waste in the embankments against deep rooting by indigenous
greasewood (a species known to penetrate soils at other sites down to 60 feet) or other plants, or
against biointrusion by indigenous ants or mammals (e.g., with maximum documented burrowing
depths greater than the proposed cover thickness). Higher rates of infiltration are typically
associated with higher contaminant transport rates. Under Utah rules, infiltration should be
minimized [see UAC Rule R313-25-25(3) and (4)]. DEQ cannot determine the adequacy of the
DU PA until EnergySolutions accounts for greater infiltration through the cover system at the
proposed Federal Cell embankment due to biointrusion by plant roots and by animals.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatories 10, 20, 28, and 71 for discussions
regarding enhanced transport due to biological processes.
Frost Damage—With the current proposed Federal Cell design, EnergySolutions should account
in modeling for substantial disruption of near-surface layers above and within the radon barriers
by frost, with accompanying decreases in ET and increases for initially low-permeability soil in
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 19
both hydraulic conductivity and correlated α values, which could affect modeled infiltration rates
and radon release rates. UAC R313-25-25(3) and (4) require a licensee to minimize infiltration;
therefore, EnergySolutions must model infiltration under realistic long-term assumed site
conditions before DEQ can determine that this requirement has been met.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 192 for discussions regarding
depth of potential frost impacts.
2.1.1 Interrogatory Response
Evapotranspiration Cover (ET Cover)
UDEQ argues that modeling of infiltration rates “must be conducted based on in-service
hydraulic properties and correlated log(α) and log(Ksat) values” as described by Dr. Craig Benson
in Volume 2, Appendix E, of the safety evaluation report (SER) prepared by SC&A (SC&A
2015a) [see interrogatories and comments 05/2, 21/2, 60/2, 90/2, 176/1 and B.4].
However, the hydraulic property recommendations and cover material naturalization presented in
Benson et al. (2011) and in Appendix E of SC&A (2015a) are inappropriate for the Clive Site.
When included in the model, they produce a model inconsistent with the observed site conditions
of Clive. This model can be considered “conservative” in terms of modeling groundwater
concentrations but dose results are lower for this model implementation than for the Clive DU
PA Model v1.4 (Neptune 2015d), which does not imply “conservative.” The rationale for not
using these naturalized or in-service cover properties in the Clive DU PA Model v1.4 (Neptune
2015d) are presented in this section.
The hydraulic property recommendations provided in Benson et al. (2011) are based on
measurements for samples from in-service covers made at 12 sites throughout the continental
United States. One element of the characterization of a site’s climate is the ratio of mean annual
precipitation to mean annual potential evapotranspiration. The magnitude of this ratio is
estimated to be 0.17 for Clive. Only one of the sites sampled by Benson et al. (2011) was
considered to be arid, having a ratio of 0.06. The mean value of this ratio for all sites sampled
was 0.51, with a highest value of 1.10. At two of the sites rainfall exceeded potential
evaporation, which is completely inappropriate for the arid conditions at Clive. All but one of the
sites that form the basis for the hydraulic property recommendations have much wetter
conditions than Clive.
The conceptual model of cover material “naturalization” for Clive based on the work of Benson et
al. (2011) is described in Appendix E of SC&A (2015a) as including changes in the hydraulic
behavior of the material following construction. These changes are characterized by increasing
values of hydraulic properties such as Ks and the hydraulic function α parameter (see Equation
(1)) that begin soon after cover completion. These changes are commonly attributed to soil
forming processes including wet-dry and freeze-thaw cycles, activity of roots and soil animals,
decomposition of organic matter by microbes producing compounds that tend to bind soil
particles into aggregates, and changes in cations adsorbed onto soil particle surfaces. In this
conceptual model these processes lead to the development of soil structure but not soil horizons.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 20
Under the wetter conditions considered by Benson et al. (2011), plant and animal activity are
greater than in an arid setting. These wetter conditions promote a faster rate of disruptive
processes due to plant and animal activity and in some cases freeze-thaw activity that were
shown by Benson et al. (2011) to lead to formation of an aggregated soil structure and natural
mixing of soil layers within a 5 to 10 year period at their study sites. Most importantly, the sites
considered by Benson et al. (2011) also lack significant eolian deposition.
This is not the case for a site like Clive. Recent field studies (Neptune 2015b) provide evidence
for a site-specific conceptual model of weak development of soil profiles (limited soil formation)
in a setting influenced by deposition of eolian silt in Holocene history. The Site is within a region
of significant eolian activity evidenced by locally thick accumulation of gypsum dunes west and
southwest of the Site and a laterally continuous layer of suspension fallout silts preserved
beneath the modern surface throughout the Clive Site. Clive quarry exposures examined in a
field study (Neptune 2015b) showed sections of eolian silts immediately below a modern
vegetated surface (Figure 3). The bottom of the eolian silt formed a gradational but definable
contact with the lake muds and marl below. The upper vegetated surface at the top of the eolian
section was distinct and noted as being partially indurated. In addition, buried soils were found in
the eolian and lake sediments below the Lake Bonneville lacustrine sequence.
The eolian deposits in the upper part of the stratigraphic section shown in Figure 3 represent a
10,000-year-old record of deposition and soil formation (Neptune 2015b). Primary soil features
developed over this time interval include an indurated Av-zone, and slight reddening of the silt
profile with local platy structure from formation of clays (Figure 4). These observations are
consistent with slow processes of soil formation in a high elevation semi-arid setting and
continuing suppression and burial of developing soils by a relatively low rate of deposition of
eolian silt. There is no evidence of soil structure development extensive enough to influence soil
hydraulic properties.
Observations of Holocene eolian silt throughout the Clive Site support a conceptual model of
long-term eolian deposition on a stable surface that promotes and preserves concurrent eolian
deposits which are only slightly modified by slow processes of soil formation. The past
Holocene depositional conditions at the Clive Site are promoted by a combination of extensive
wet playa sources of eolian source material to the west and southwest of the Clive Site and the
extremely low gradient paleo-Lake Bonneville surface surrounding the Site with sparse surface
vegetation and limited surface erosion. These conditions will persist at the Clive Site as long as
the lake levels remain below the site elevation. Rates of eolian deposition would be expected to
increase as future lakes approach the Site with increased formation of dunes (deposition of eolian
sands). Recurring lakes during ice ages (climate cycles) will rework and mix the eolian deposits
with aggrading clastic lake sediments. The expectation is that eolian deposits will drape and
slightly stabilize closure covers until future lakes return to the Clive Site. These studies at the
Clive Site were conducted by two highly experienced Ph.D. geologists, one of whom, Dr. Jack
Oviatt (https://www.k-state.edu/geology/faculty-staff/Oviatt.html), has decades of experience
working in the Lake Bonneville region.
Climate conditions and soil formation processes at the Clive Site contradict the assumptions of
rapid soil structure formation in the cover layers observed by Benson et al. (2011) at other sites
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 21
and demonstrate the inapplicability of the conceptual model to the Clive Site. The one-size-fits-
all generalization of hydraulic properties does not appropriately represent the unique conditions
at the Clive Site. As described by Neuman et al. (2003) in NUREG/CR-6805:
Hydrogeologic models are by nature site-specific. Though there is an established (and
evolving) set of general hydrogeologic principles that apply to many sites on many scales,
they are insufficient to either describe (conceptualize) or quantify (model) the hydrogeology
of a particular site over a given range of scales. Because each site is unique, general
principles must always be supplemented by regional and site-specific data to be useful for
conceptualizing and modeling subsurface flow and transport at a site, regardless of purpose.
The soils at the Clive Site used for construction of the cover have been mapped, described,
sampled, and their properties measured. These soils have hydraulic properties that derive from
their origin and their environment, which are unique to the Clive Site. These properties are
therefore already “naturalized.”

Figure 3. Eolian silt in trench located at Clive Pit 29 overlying Lake Bonneville
sedimentary deposits (Neptune 2015b).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 22

Figure 4. An example of upper soil-modified eolian silt in Pit 29. Basal contact of the silt is
approximately located at the middle of the pick handle. It is a gradational contact
between eolian silt intermixed with regressive Lake Bonneville marl (bottom of the
pick handle).
UDEQ asserts that “the photographs in Figures 4 and 5 of Appendix 21 [reproduced herein as
Figure 3 and Figure 4] are inconclusive and provide no quantitative basis to support inferences
that structural development and alterations in hydraulic properties do not occur at Clive.
Structural development that occurs in covers due to pedogenesis generally is not visible at the
scale represented in these photographs. Moreover, the smearing that occurs in test pits can
obscure structure that is present.” The interpretation of the expert geologists that recent soil
formation is minimal at the Site was based on their direct field observation of the exposures, not
on examination of the photographs.
UDEQ states further, “If EnergySolutions wishes to use these analogs as evidence to support
hydraulic properties representing long-term conditions significantly different from NUREG/CR-
7028 (Benson et al. 2011), EnergySolutions should conduct appropriate measurements on these
in-place materials to demonstrate that the hydraulic properties are indeed different from the
abundance of data in NUREG/CR-7028.” The profiles excavated and examined by the geologists
were not analogs but observations of the Unit 4 material at the Clive Site while the “abundance
of data” referred to by UDEQ was acquired from other (nearly all wetter) sites throughout the
continental United States.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 23
Correlation Between Saturated Hydraulic Conductivity, Ks, and the van Genuchten α
Parameter
The hydraulic parameters α and Ks used in the Clive DU PA Model v1.4 are considered to be
statistically independent, implying they are theoretically uncorrelated and distributions can be
developed separately for each parameter. UDEQ believes this assumption is not valid based on
summaries of data presented by Benson and Gurdal (2013) as presented in Appendix E of the
Safety Evaluation Report (SER) (SC&A 2015a), Sections 4.1.1.1 and 4.4.1. Both references
examine data aggregated over soil textural classes, while the Clive DU PA Model v1.4 (Neptune
2015c) focuses on a single primary texture class (silty clay), and distributions for α and Ks are
developed specifically for that soil textural class. In this response, EnergySolutions provides new
analysis of data associated with Benson and Gurdal (2013) to demonstrate the importance of
considering soil textural class when attempting to quantify correlation between α and Ks.
In addition to providing detail on the different patterns seen in aggregated data versus
subpopulations defined by soil textural classes, several other fundamental statistical concepts
support a lack of evidence for building strong positive correlation into the distribution
development for α and Ks. There are several fundamental statistical concepts that should be taken
into account when considering generalization of a correlation observed in a particular data set to
a larger population or context: (1) correlation is designed as a summary of the linear association
between two random variables, and therefore linearity should be assessed before relying too
heavily on correlation as a summary of a relationship, (2) data aggregated over many
subpopulations with different ranges for the variables may exhibit a drastically different
correlation from the correlations observed within the subpopulations, (3) even when random
variables are statistically independent, random realizations of data will have estimated
correlations that are different from zero, and (4) points with extreme values for both variables
have high leverage and can be very influential on estimated correlations and estimated slopes.
Investigations into observed correlations between α and Ks are typically carried out on the
logarithmic scale because the relationship is expected to be more linear on the log scale than the
original scale. All plots and correlations provided in this response are based on the log10 scale for
both α and Ks.
A closer look at the Benson and Gurdal (2013) data
The data from the Benson and Gurdal (2013) paper are used to investigate the evidence of a
correlation. Benson (C. Benson, personal communication, October 11, 2017) supplied
EnergySolutions with the 253 pairs of Ks and α values presented in the 2013 paper, along with
information regarding the percent sand/clay/silt, as measured for the material for each sample.
The plot provided in Benson and Gurdal (2013), and included here as Figure 5, does not display
or take into account the soil texture associated with the measurements (i.e., data are aggregated
over soil textural classes). They reported a sample correlation coefficient for the aggregated data
of 0.462 and did not provide a measure of uncertainty or information about what correlation type
was calculated.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 24

Figure 5. Figure 3 from Benson and Gurdal (2013) showing the data requested by
EnergySolutions.
Using the USDA soil classification system (Jury and Horton 2004), each data point in the data
provided by Benson (2017) was assigned a soil textural class. This information is then used to re-
visualize the data (see Figure 6 and Figure 7) and to further investigate estimated correlation
coefficients. Pearson’s correlation coefficients are used unless otherwise stated.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 25

Figure 6. Estimated linear relationships between α and Ks for all observations (solid line)
and without the high-leverage points making up the clusters of points in the upper
right half of the plot with α greater than 0.10 kPa-1 (dotted line). The estimated
correlation changes from 0.627 for all the data to 0.384 for the restricted range.
●●●●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●●
●
●
●●
●●
●
●●
●
●●
●
●●●●●●
●●
●
●●
●
●●
●●
●●●●
●
●●
●
●
●
●
●●
●●●●●●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●●
●
●●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●●
●●
●
●●●●
●
●
●
●
●●
●●●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●
●
●
●●●●●●●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
r = 0.627
0.001
0.010
0.100
1.000
1e−07 1e−05 1e−03
Ks (cm/sec)
a (
1
/
k
P
a
)
SoilType
●
●
●
●
●
●
●
●
●
●
ClLo
Lo
LoSa
Sa
SaCl
SaClLo
SaLo
SiCl
SiClLo
SiLo
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 26

Figure 7. Estimated linear relationships within each soil type for the data provided by
Benson (2017). Pearson’s correlation coefficients, and 95% confidence intervals, are
shown in Table 2 for the soil textural classes, and the individual soil classes are shown
in panels in Figure 8.
A few important observations from Figure 6 are:
1. The clusters of observations in the upper right with large value for Ks and α belong to
three distinct soil textural classes (sand, sandy-loam, and loamy-sand); they do not
represent pairs of Ks and α across a range of soil types. The estimated linear relationship
and correlation are heavily influenced by the location of the clusters relative to the cloud
of other points because they have high leverage on the estimated relationship (see the
difference in slope between the solid and dotted lines in Figure 6).
2. The relationship displayed in Figure 5 and the reported correlation coefficient based on
the aggregated soil textural classes are not indicative of the relationships observed within
individual soil textural classes. In fact, the observed relationships between Ks and α vary
widely across soil textural classes, as seen in Figure 7. The estimated correlation for the
aggregated data does not appear to be useful for individual soil textural classes and,
therefore, should not be generally applied to distribution development for PA models
meant to apply to specific soil textural classes.
3. The Pearson’s correlation coefficient and the estimated relationship shown in the plot in
Figure 5 do not match those for the data provided to EnergySolutions by Benson in 2017
●●●●
●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●●
●
●
●●
●●
●
●●
●
●●
●
●●●●●●
●●
●
●●
●
●●
●●
●●●●
●
●●
●
●
●
●
●●
●●●●●●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●●
●
●●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●●
●●
●
●●●●
●
●
●
●
●●
●●●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●
●
●
●●●●●●●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
0.001
0.010
0.100
1.000
1e−07 1e−05 1e−03
Ks (cm/sec)
a (
1
/
k
P
a
)
SoilType
●
●
●
●
●
●
●
●
●
●
ClLo
Lo
LoSa
Sa
SaCl
SaClLo
SaLo
SiCl
SiClLo
SiLo
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 27
(compare Figure 5 to Figure 6). The data presented in Figure 6 are based on 253 pairs of
Ks and α, which matches the number of pairs expected based on Table 2 in Benson &
Gurdal (2013). However, it is clear the data do differ slightly (e.g., there are two points
near α of 0.1 kPa-1 in Figure 5 that do not appear in Figure 6). This discrepancy could not
be reconciled in this investigation. Regardless, the general conclusions and arguments do
not depend on the value of the correlation coefficient reported or small differences
between the two data sets.
Patterns in aggregated data do not apply to individual soil textural classes
For the correlation based on data aggregated over all soil textural classes to be an appropriate
description for individual soil textural classes, the relationships within soil textural classes
should look similar to the pattern observed in the aggregated data. The existence of a different
relationship for aggregated data (or a population) as compared to data separated by subgroups (or
subpopulations) is a common and well-studied statistical phenomenon known by names such as
Simpson’s Paradox and ecological fallacy (e.g., Wagner (1982) and Ch. 9 of Ramsey and
Schafer (2013)). This observation is often expected when, as shown in Figure 6 and Figure 7, the
ranges of values for each variable vary by group, sometimes greatly. This creates a situation
where subgroups with little correlation, or even negative correlation, can stack together to induce
positive correlation over the population. Potentially misleading conclusions and predictions can
result through the assumption that subgroups behave similarly to the aggregated population with
respect to the association between the two variables. The assumed correlation within a
population need not be the same as the correlations within its subpopulations.
The observed relationships between Ks and α within the data provided by Benson in 2017 clearly
vary in magnitude and sign for different soil textural classes. The data for each soil textural class
are plotted in their own panels (with different x and y axis scales) in Figure 8 to supplement
information in Figure 7. Table 2 provides the estimated correlation coefficients and their
associated 95% confidence intervals for each soil textural class. There is no evidence in these
data that the pattern in the aggregated data across soil types should apply at the scale of a
particular waste cover primarily made up of one soil textural class. Benson and Gurdal (2013)
suggest a large positive correlation exist for data aggregated across soil textural classes, but, as
shown above and below, it is important to consider the soil textural class when introducing the
assumption of correlation into distribution development for the PA model.
Correlation coefficients can be useful statistics to provide the direction and strength of a linear
relationship between two random variables (Ramsey and Schafer 2013). Few of the soil textural
classes in the data from Benson (2017) seem to exhibit clear linear associations, thus limiting the
usefulness of the correlation coefficient as a summary measure of the relationship. Statistical
hypothesis tests can be helpful if evidence against a particular null hypothesis is of interest, but
confidence intervals provide more information about uncertainty and are not tied to a particular
null hypothesis. The 95% confidence interval can be interpreted as a range of plausible
correlations that could be obtained from different random datasets of the same size from the
same population.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 28
Table 2. Estimated Pearson’s correlation coefficients between Ks and α on the log scale by
soil textural class, and associated 95% confidence intervals, calculated using the data
provided by Benson in 2017 data.
Soil Textural
Class
Estimated
correlation
95% lower
limit
95% upper
limit
Clay Loam -0.01 -0.64 0.62
Loam -0.15 -0.49 0.23
Loamy Sand 0.79 -0.06 0.98
Sand -0.28 -0.79 0.48
Sandy Clay 0.96 -0.02 1.00
Sandy Clay Loam 0.25 -0.03 0.49
Sandy Loam 0.80 0.63 0.90
Silty Clay -0.20 -0.97 0.94
Silty Clay Loam 0.39 0.10 0.62
Silty Loam 0.36 0.12 0.55

On the log scale, seven of the ten soil textural classes represented in the data from Benson and
Gurdal (2013) have negative estimated correlations or have 95% confidence intervals that
overlap zero. The widths of the confidence intervals demonstrate the substantial uncertainty in
the correlations of the underlying populations, given the available data.
It is also important to combine the information in Table 2 with that in Figure 7 and Figure 8. For
example, the sandy loam soil class (SaLo) has a point estimate with a large magnitude (0.80) and
a confidence interval suggesting a positive correlation between Ks and α (0.63, 0.90), but a closer
look at the plot of the data for that soil textural class in Figure 8 calls into question the validity of
that conclusion based on the cluster of points with high leverage driving the correlation.
For the Clive DU PA Model v1.4, the objective is to develop distributions for the Unit 4 soil
described as Silty Clay, which shows no evidence of a positive correlation in the four samples in
the Benson and Gurdal (2013) data (see Figure 8, Figure 9, and Table 2).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 29

Figure 8. Estimated linear relationships by soil textural class for the data plotted in Figure
6 and Figure 7. Note the x and y axes are allowed to change among panels and are on
the log10 scale.
●
●
●●●
●
●
●●●
●
●
●●
●
●
●
●
●●
●●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●●
●
●
●
●
●
●●●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●●●
●
●●
●
●●
●
●
●●
●
●
●
●
●●
●
●●
●
●●
●
●
●
●
●
●
●
●
●
●
●●●
●●
●●
●●
●●
●●
●
●
●
●
●●●
●
●●
●●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●●
●●
●
●
●
●●
●
●
●●
●●●●
●
●●●●
●
●
●
●
●●●
●
●●
●●
●
●
●
●
●
●
●
●
●
●●
●●●●●●●●●●●●●●
●
●●●●●
●●
●●●●
●
●●
●
●
●
●●●●
●●
●
●
●●
SiClLo SiLo
SaCl SaClLo SaLo SiCl
ClLo Lo LoSa Sa
1e−06
1e−04
1e−07
1e−06
1e−05
1e−07 1e−07 1e−05 1e−06 1e−04 1e−08 1e−07 1e−06
1e−07 1e−05 1e−06 1e−04 1e−05 1e−04 0.001 0.010
0.001
0.001
0.010
0.100
1.000
0.001
0.010
0.001
0.010
0.001
0.010
0.001
0.010
0.100
0.001
0.001
0.010
Ks (cm/sec)
a (
1
/
k
P
a
)
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 30

Figure 9. Estimated linear relationship on the log scale between α and Ks for Silty Clay.
Pearson’s correlation coefficient is -0.2, with a very wide 95% confidence interval of
(-0.97, 0.94).
Observed sample correlations from independent random variables
Assessing evidence of a theoretical correlation between two variables based on relatively few
samples, as is the case for the separate soil textural classes, is very difficult. Figure 10 illustrates
estimated linear relationships that could be observed even when the two variables are truly
independent. The panels are created by taking random draws from independent standard normal
distributions with sample sizes of 6. Note the different conclusions one might draw about
correlation between the two random variables just from looking at one of the sixteen panels. The
difficulty of assessing evidence for a theoretical correlation due to small sample sizes can lead to
the desire to increase the number of samples used by aggregating the data of immediate interest
with that from related groups. As discussed above, this should be done with caution.
●
●
●●
r = −0.2
0.001
1e−08 1e−07 1e−06
Ks (cm/sec)
a (
1
/
k
P
a
)
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 31

Figure 10. Estimated linear relationships based on 16 realizations of six random pairs of
observations that come from statistically independent variables (x and y were drawn
independently from two standard normal distributions).
Summary
In summary, there is not evidence of a strong linear relationship between α and Ks that should be
incorporated into PA distributions, particularly for the Unit 4 soil within the Silty Clay soil
textural class found at the Clive Site. Additionally, incorporating correlation into distribution
development without adequate empirical and theoretical evidence should be done with care to
avoid excluding combinations of values of the two variables that are actually plausible (i.e., to
avoid inadvertently excluding realistic scenarios). The argument for correlation between Ks and
α suffers from some common statistical misconceptions and misuses of correlation, as described
in detail in the response. It is clear from the exploratory analysis conducted here that soil textural
classes should be taken into account when assessing possible correlation between Ks and α for a
particular site. The two variables exhibit very different relationships within different soil textural
classes—a clear indication that the aggregated correlation should not be applied to the individual
soil textural classes. In the particular soil textural class used in the Clive DU PA Model v1.4,
Silty Clay, there are very few samples available in Benson and Gurdal (2013) and no evidence of
a positive linear relationship between the Ks and α parameters.
●
●
●●●
●
●
●
●●●
●
●
●●●●
●
●●
●●
●
●
●
●●
●
●●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●●●
●●●●
●●
●
●
●●●
●
●
●●
●●●
●●●●●●
●
●●●
●
●
●
●●
●●●
●
●
●●
●●
●
●●
●●●
13 14 15 16
9 10 11 12
5 6 7 8
1 2 3 4
−2 −1 0 1 2 −2 −1 0 1 2 −2 −1 0 1 2 −2 −1 0 1 2
−2.5
0.0
2.5
5.0
−2.5
0.0
2.5
5.0
−2.5
0.0
2.5
5.0
−2.5
0.0
2.5
5.0
X
Y
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 32
Clay Liner
The Federal Cell is constructed on top of a compacted clay liner. Disposal involves placing waste
on a prepared clay liner that is approximately 2.5 m (8 ft) below the ground surface. The liner is
constructed of compacted local clay, i.e., Unit 4 material. Porosity and bulk density values for
the clay liner are assumed to be the same as for the Radon Barrier Clays, as these clays are all
compacted, unlike the surface and ET layer Unit 4 material. Properties of nine Unit 4 cores
acquired at the Clive Site were determined by Bingham Environmental (1991, 1994). Grain size
distributions were determined for nine cores and saturated hydraulic conductivities and water
retention relations determined for two of the cores.
The GoldSim software platform cannot calculate flow of water from the cover surface to the
water table using the unsaturated flow equation the way that variably saturated numerical flow
models (e.g. HYDRUS) can. GoldSim includes flow by assigning a flow rate or flux for a
realization from a statistical distribution of fluxes. In this context a GoldSim realization is “a
single model run within a Monte Carlo simulation. It represents one possible path the system
could follow through time” (GTG 2014). This water flux distribution is developed from a
collection of results obtained from simulations using more complex models that incorporate daily
changing weather conditions at the surface and that allow for many flow models to be run
encompassing a range of hydraulic input parameter values. The resulting statistical distribution
represents the net infiltration or water flux at the top of the waste. For any given realization, the
same value of the flux drawn from the distribution is assigned to each cell representing the
pathway through the waste as inflow to the cell and outflow from the cell. The clay liner is
represented in the model by four cells below the waste. The outflow from the bottom cell
representing the waste is connected to the inflow of the top cell of the clay liner. Each of the clay
liner cells are assigned the same flux as the waste.
The layer above the clay liner with the lowest hydraulic conductivity is the Upper Radon Barrier.
As described in Section 2.4.1, the distribution for the hydraulic conductivity of this layer has a
minimum value of 4.32 × 10-3 cm/day, which is the engineering design specification. The
distribution has 1st, 50th, and 99th percentile values of 0.65 cm/day, 3.8 cm/day, and 52 cm/day,
respectively, which are from a range of in-service (“naturalized”) clay barrier Ks values
described in Section 6.4, p. 6-12 of Benson et al. (2011). Given this distribution, any Ks value
chosen from the distribution other than the minimum is a value greater than the initial
construction specification. As described above, each cell from the bottom of the waste to the top
of the upper aquifer is assigned the same water flux drawn from the distribution for that
realization, including the clay liner cells. As larger fluxes are drawn representing the variability
due to naturalization of the cover materials, those same fluxes are applied to the clay liner
representing naturalization of that layer.
UDEQ has argued that the modeling approach described above does not account for liner
degradation over time and that there are problems with the assumed liner hydraulic conductivity
values (see the responses to UDEQ Interrogatories 05/2 and 90/2). With respect to the first issue,
liner degradation is accounted for by using a statistical distribution of water fluxes for the
modeling as described above. This distribution takes into account a range of possible values for
saturated hydraulic conductivities greater than the engineering specification. The next issue is
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 33
immaterial because the water flux through the clay liner cells depends only on the value of the
water flux at the top of the waste drawn from the water flux distribution. This means that liner
hydraulic properties are assumed to be the same as the radon barriers.
UDEQ states that modeling of the clay liner “should employ hydraulic parameters representative
of a clay liner.” They go further to recommend a reference for clay hydraulic properties not
based on samples from the Clive Site. As described above, hydraulic properties for the clay liner
were obtained from testing of two cores acquired from the Clive Site of Unit 4 silty clay
(Bingham Environmental (1991, 1994)).
The last issue raised by UDEQ with respect to the clay liner is that van Genuchten hydraulic
function α parameter and the saturated hydraulic conductivity values must be correlated (also
Interrogatories 05/2, 21/2, 60/2, 90/2, 153/2, and Comment B.4). See the response to
Interrogatory 05/2.
Infiltration
See the response to Interrogatory 05/2 Evapotranspiration Cover (ET Cover).
Erosion of Cover
UDEQ points to issues on erosion raised in Section 4.4.2 of the SER (SC&A 2015b). These
issues are addressed in the response to Interrogatory 71/1.
Calculations to evaluate the stability of the design with respect to gully erosion for the ET cover
of the Class A West cell were provided in Appendix D of EnergySolutions (2015). Similar
calculations for the Federal Cell are presented in the response to Interrogatory 71/1.
Effect of Biologicals on Radionuclide Transport
SWCA Environmental Consultants (SWCA) was contracted by EnergySolutions to acquire field
data at the Clive Site to identify representative study areas and to collect data to “document the
diversity and composition of plant and animal species and to quantify soil mixing by burrowing
mammals and ants associated with each vegetation community” (SWCA 2013). In addition, six
excavations were completed and surveyed to quantify the aboveground and belowground size of
the dominant plant species, and the maximum rooting depth and width of root of the dominant
species. This work was focused on characterizing the biotic aspects of the ET cover system at the
Clive Site for the DU PA Model (SWCA 2011). SWCA is a nationwide firm of environmental
professionals operating since 1981 and is well known for their work in multiple environmental
sectors, including ecology.
Field studies and analyses were conducted by SWCA at the Clive Site including quantifying
vegetation, small mammal distributions, and mammal burrow and ant mound size and densities
at the Clive Site and nearby ecologically analogous sites. Additionally, several excavations in the
Unit 4 soils were completed to estimate ant mound depth and soil volume transported to the
surface by ants for the DU PA Model (SWCA 2011).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 34
SWCA describes the ecological context for the ET cover design at the Clive Site in their
responses to Round 1 Interrogatories from UDEQ. They note that the design for Clive is
consistent with both published ET cover-system recommendations (ITRC 2003; Peace et al.
2004; Scanlon et al. 2005) and with the site-specific climate and ecology documented through
the studies conducted by SWCA (2011, 2012). As summarized from SWCA (2013), the ecology
at the Clive Site can be characterized by the following:
• The dominant shrub species are small in stature compared to those of the same species in
deeper more fertile soils and/or in areas of greater annual precipitation.
• The distributions, densities, and stature of plant species at the Site are determined by high
soil salinity, soil pH, low fertility, and aridity.
Based on data collection and analysis and the proposed cover design, SWCA evaluated the
potential for disturbance by plant roots, mammals, and ants to result in increased infiltration
(SWCA 2013). The Surface and Evaporative Zone layers are composed of native Unit 4 material
and revegetated with native, locally adapted plant species. The target vegetation community is
designed to represent the diversity and density of the native vegetation (SWCA 2013). With this
combination of local materials and plants, the hydrology and ecology of the upper layers of the
cover will be similar to native undisturbed areas at the Site.
Based on evidence from analogous sites detailed in SWCA (2013), the densities of plant roots
and burrows will be sparse in these layers and infiltration would be minimally affected. Total soil
disturbance due to mammal burrowing was estimated by SWCA (2013) for two cases of
vegetation conditions. These cases are considered because these authors noted an association
between vegetation conditions and mammal activity. Under expected vegetation conditions, soil
volume disturbed by mammal activity was less than 1/100th of a percent of the total soil volume.
Under a worst-case scenario of vegetation dominated by greasewood, the soil volume disturbed
was 1/10th of a percent of the total soil volume of the ET cover. Their conclusions, supported by
detailed evidence presented in SWCA (2013), were that the Surface and Evaporative Zone layers
would be deep enough to allow for ant activity and some mammal burrowing and soil movement
without compromising the functions of the lower layers.
SWCA concludes that, despite the low density and small size of plants and mammals at the Site,
local plants and animals could penetrate the Frost Protection layer but “the amount of soil
disturbance that could potentially occur on the ET cover is minute, compared to the total soil
volume on the ET cover. Similarly, the amount of water infiltration that could occur in
association with biointrusion of the ET cover is also minute.”
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 35

Figure 11. Borrow soil cross-section below a greasewood plant shows the compacted clay
layer at approximately 60-cm depth. Roots extend laterally and do not penetrate the
compacted layer (SWCA 2011).
The potential for greasewood roots to extend to depths greater than the cover layers is recognized
by SWCA. SWCA (2013) notes, however, that “the aboveground mass of greasewood plants on
and near Clive is consistent with low water availability.” Roots are expected to follow water
availability and not penetrate clay layers. This behavior was documented by SWCA (2013) in an
excavation at the Clive Site showing lateral root growth of a greasewood plant at the Site. Roots
were seen to extend vertically downward to the top of a clay layer at approximately 60 cm depth
(Figure 11). At the clay layer roots extended only laterally and did not penetrate the layer. The
explanation for this rooting behavior from the observation of SWCA, consistent with other
research, is that infiltrating water would tend to perch on lower permeability clay layers causing
plant roots to grow laterally following water availability (Groeneveld 1989).
SWCA (2013) conclude from their observations at the Clive Site and ecological analog sites that
the low fertility and alkalinity of the soils and the aridity of the climate limit the growth of plants
at the Site and would inhibit the development of large deep-rooted plants in the cover. Their
opinion is that, while greasewood will eventually become established at the Site, roots will
follow available water and remain within the Surface, Evaporative Zone, and Frost Protection
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 36
Layers. Additional evidence supporting the projection of limited greasewood root penetration at
the Site presented by SWCA (2013) includes:
• The water table depth exceeds maximum levels accessible by greasewood.
• There is little or no capillary rise above the water table.
• The root to shoot stature of greasewood at Clive is not sufficient to root deeply.
• Regardless of plant stature, the majority of root biomass is concentrated in the upper soil.
• Nutrients are concentrated near the surface in desert soils.
Local Unit 4 material is used for the upper layers of the cover, and these layers will be vegetated
with native, locally adapted plant species. The hydraulic properties of these layers are expected
to be similar to those of undisturbed areas of the Site, which will experience the same biotic
activity as the upper layers of the cover. Using hydraulic conductivity measurements of native
Unit 4 samples from the Site as inputs to the model accounts for the influence of biotic activity
on infiltration.
Frost Damage
Calculations of frost depth at the Clive Site were made by Hansen, Allen, and Luce (HAL), a
professional engineering firm. These calculations were provided to UDEQ in Appendix E of the
Updated Site-Specific Performance Assessment (Revision 2) (EnergySolutions 2015).
HAL chose to use the modified Berggren equation to calculate frost depth because of its long-
established use and acceptability by the engineering community. This method, first presented by
Berggren in 1943 and further refined by Aldrich and Paynter in 1953, was later adopted by the
US Army Corps of Engineers and other agencies as their preferred method for frost depth
determination (Departments of the Army and the Air Force 1988).
The frost depths calculated as part of this analysis give results that are in line with the depths of
cover and frost protection proposed in the EnergySolutions ET Cover system design. The
proposed radon barrier begins at depths ranging from 30 inches to 42 inches, which provides
frost protection for the calculated 100-year frost penetration depth of 22.4 inches to 27.8 inches
for the top slope and side slope, respectively.
A statistical distribution for saturated hydraulic conductivity of the radon barriers used for
modeling flow in the cover was developed from in-service properties based on the results of
Benson et al. (2011). Use of these properties as inputs to the flow models takes into account
changes in hydraulic properties due to freeze/thaw cycles assumed by Benson et al. (2011). See
the response to Interrogatory 21/2 for a discussion of the development of the saturated hydraulic
conductivity distribution for the radon barriers (Section 2.4.1).
Early degradation of cover layer properties described by Benson et al. (2011) is not consistent
with soil formation observations at the Clive Site. Recent field studies (Neptune 2015b) provide
evidence for a site-specific conceptual model of weak development of soil profiles (limited soil
formation) in a setting influenced by low rates of deposition of eolian silt in the Holocene
history. See the response to Interrogatory 05/2 Evapotranspiration Cover (ET Cover) (Section
2.1.1) for a detailed discussion of soil formation at the Clive Site.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 37
2.2 Interrogatory CR R313-22-32(2)-10/3: Effect of Biologicals on
Radionuclide Transport
DEQ Conclusion from April 2015 SER, Appendix C:
As discussed in the DU PA SER (Section 4.4.3), EnergySolutions has not shown that the cover
system is sufficiently thick or designed with adequate materials to protect the cover system or the
underlying bulk waste in the embankments against deep rooting by indigenous greasewood (a
species known to penetrate soils at other sites down to 60 feet) or other plants, or against
biointrusion by indigenous ants or mammals (e.g., with maximum documented burrowing depths
greater than the proposed cover thickness). Higher rates of infiltration are typically associated
with higher contaminant transport rates. Under Utah rules, infiltration should be minimized [see
UAC Rule R313-25-25(3) and (4)]. DEQ cannot determine the adequacy of the DU PA until
EnergySolutions accounts for greater infiltration through the cover system at the Federal Cell
embankment due to biointrusion by plant roots and by animals. Therefore, this interrogatory
remains open.
DEQ Critique of DU PA v1.4, Appendix 5: EnergySolutions/Neptune retain the same
assumptions with respect to biointrusion depths and potential impact on infiltration in v1.4 as
were provided in v1.2.
In v1.4 Appendix 5 (p. 33), EnergySolutions indicates that root water uptake was modeled
assuming the roots extended to the bottom of the evaporative zone layer and that rooting density
decreased with depth. This text seems to contradict the statement in v1.4 Appendix 5 (p. 33) that
root distribution was modeled as extending into the frost protection layer with a maximum depth
of 31 inches (80 cm). The base of the evaporative zone would be at 18 inches. Figure 1 indicates
that the roots cease within the frost protection layer. The impact of the rooting depth in v1.4 is to
remove water from the system and thereby reduce the infiltration rates. The concern raised by
the interrogatory is related to the roots creating preferential pathways and thereby increasing the
infiltration.
DEQ Critique of DU PA Appendix 21: EnergySolutions/Neptune state (p. 15):
It is important to recognize how the range of rooting depths discussed in the comment
actually relates to what was used as a maximum rooting depth in GoldSim Models v1.2
and v1.4. A maximum root depth of 5.7 meters (18.7 ft) (Robertson 1983) is used in the
Model, so the Model already assumes that roots extend beyond the radon barrier. In
addition, v1.4 of the GoldSim Model assumes increased permeability, correlation
between saturated hydraulic conductivity and the hydraulic function alpha parameter,
and homogenization of the cover materials, with no physical barriers to either plant roots
or infiltration.
It is unclear how the specification of the rooting depth in GoldSim is particularly relevant to the
concern expressed in the comment pertaining to potential increased infiltration rates due to
biointrusion of plants and animals. The rooting depth in GoldSim is related to the depth of
contaminant uptake, redistribution of contamination, and assimilation of contaminants once the
plant dies rather than changes to the hydraulic properties that would allow greater infiltration.
Plant roots will almost certainly extend downward and into the radon barrier. These roots will
then penetrate into the underlying waste if water is available in the waste. As described Benson
et al. (2008), roots were observed in the radon barrier in the caisson lysimeters exhumed at
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 38
Monticello in 2008. These were at depths of 1.6–1.9 m bgs (see Figure 10-1 below). The roots
desiccated the radon barrier, causing large cracks and an increase in Ks.
Furthermore, EnergySolutions has used a homogeneous cover profile in the most recent
simulations. This was not the intent of the previous comments and approach outlined in
Appendix E to the April 2015 SER and was misconstrued from the parameter recommendations
provided in Appendix E. The cover profile should retain a layered structure representative of the
materials planned for each layer, but with the hydraulic properties of each layer adjusted to
reflect pedogenesis. The parameters in the 2015 recommendations were presented as a guide
for reasonable ranges consistent with the recommendations in NUREG/CR-7028 (Benson et al.
2011).
2.2.1 Interrogatory Response
As described in the response to Interrogatory 05/2, SWCA acquired field data at the Site and
conducted extensive literature reviews to “document the diversity and composition of plant and
animal species and to quantify soil mixing by burrowing mammals and ants associated with each
vegetation community” (SWCA 2013). Based on data collection and analysis and the proposed
cover design, SWCA evaluated the potential for disturbance by plant roots, mammals, and ants
to result in increased infiltration (SWCA 2013) and determined that soil disturbance and
increased infiltration due to biotic activity would be minute. Their evaluation does not support
UDEQ’s assertion that biointrusion of plants and animals will substantially increase infiltration.
See the response to Interrogatory 05/2: Radon Barriers.
UDEQ correctly points out an error in the description of the maximum modeled rooting depth in
Section 12.1, p. 33 of Appendix 5 of the Clive DU PA Model v1.4 (Neptune 2015c). The correct
description is on p. 36 of Section 12.3 of the same document: “Root distribution was modeled as
extending into the frost protection layer with a maximum depth of 31 inches (80 cm). Root
density was modeled as decreasing linearly with depth.” This description is consistent with the
maximum rooting depths and rooting density shown for the model (Figure 1 of Neptune (2015c))
and observed in excavations at the Site (Figure 9 of Neptune (2015c)).
UDEQ makes the inference that observed root penetration and cracking of clay layers at the
Monticello Site means that similar root penetration and degradation of the radon barriers will be
likely at Clive. Monticello and Clive are not comparable sites with respect to rooting depths.
SWCA (2013) describes the important ecological differences between Monticello and Clive:
• Monticello receives approximately 50% greater average annual precipitation than Clive
(15.4 in).
• The Monticello ET cover consists of clay-loam to sandy-loam soils that are less alkaline
and more fertile than the saline, alkaline silty-clay soils at Clive (Waugh et al. 2008).
• The native vegetation at Monticello is dominated by big sagebrush shrublands and
grasslands that are more diverse and of larger stature—with greater target plant densities
and cover for the ET cover—than those proposed at Clive.
The last issue of this interrogatory raised by UDEQ is that the hydraulic parameter distributions
and correlations used for modeling of flow in the cover system based on the recommendations of
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 39
Benson et al. (2011) were not how the recommendations were intended by UDEQ to be
implemented.
Furthermore, EnergySolutions has used a homogeneous cover profile in the most recent
simulations. This was not the intent of the previous comments and approach outlined in
Appendix E to the April 2015 SER and was misconstrued from the parameter
recommendations provided in Appendix E. The cover profile should retain a layered
structure representative of the materials planned for each layer, but with the hydraulic
properties of each layer adjusted to reflect pedogenesis.
The proposed cover system shown in Figure 1 is composed of layers that either store and release
water or act as a barrier. The proposed system does not include geomembranes. UDEQ argues
that input parameter distributions for the α, n, and Ks hydraulic parameters be developed using
recommended ranges for hydraulic properties from NUREG/CR-7028 (Benson et al. 2011).
Benson et al. (2011) provide these recommendations in Section 10.2:
• The saturated hydraulic conductivity of fine-textured earthen storage and barrier layers
can be assumed to range between 1 × 10-7 m/s and 5 × 10-6 m/s.
• The porosity of earthen storage and barrier layers will likely range between 0.35 and
0.45.
• The α-parameter in the van Genuchten equation, which is used to describe the SWCC
(soil water characteristic curve) for hydrologic simulations, varies between 0.01 and
0.33 kPa-1 for field-scale barrier and storage layers.
• The n-parameter in van Genuchten’s equation, which is used to describe the SWCC for
hydrologic simulations, varies over a very small range (typically between 1.2 to 1.4).
Note that in assigning hydraulic properties no distinction is made by Benson et al. (2011)
between storage and barrier layers. Based on their conceptual model of formation of soil
structure in cover systems, there is no difference in the in-service properties of what were
constructed as storage layers or barrier layers. The only distinctions in properties between depth
intervals in the flow model used for the cover simulations are the initial conditions, which are
minimized by long simulation times, root density (which is maintained as a constant throughout
the cover depth), and the hydraulic properties of the intervals. Given their conceptual model that
makes no distinction between the properties of storage and barrier layers, the cover can no longer
be represented by a layered system. UDEQ objects to this homogeneity in hydraulic properties in
this interrogatory, but that is the only logical outcome of applying the Benson et al. (2011)
method to the ET cover.
2.3 Interrogatory CR R317-6-2.1-20/2: Groundwater Concentrations
DEQ Critique of DU PA v1.4, Appendix 10: See Interrogatory 201 for further discussion.
DEQ Critique of DU PA Appendix 21: EnergySolutions/Neptune state (p. 16):
The conceptual model of cover “naturalization” described in Appendix E of the SER
(SC&A 2015) is that plant and animal activity and freeze-thaw cycles result in
disturbance and mixing of soil layers in the upper portion of the cover system subject to
their influences. The extent of the influence of these processes decreases with depth of
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 40
roots, animal burrowing, and frost penetration. This conceptual model does not maintain
the designed functions of store and release layers and barrier layers to reduce net
infiltration. Using this conceptual model, the upper portion of the soil profile subject to
naturalization processes is considered to be homogeneous with respect to the hydraulic
properties affecting net infiltration. For the Clive Site, the hydraulic properties of the
waste below the cover are modeled as Unit 3 material and would be subject to the same
naturalization processes as the materials used to construct the cover.
With this conceptual model, the depth to the waste would be reduced by erosion but the
net infiltration will not vary. The net infiltration is determined by climate and hydraulic
properties. If the hydraulic properties are assumed to be homogeneous and determined
by climate and biotic activity, loss of material from the surface of the cover will not
change the net infiltration.
EnergySolutions has used a homogeneous cover profile in the most recent simulations. This was
not the intent of our previous comments and approach outlined in Appendix E to the April 2015
SER and was misconstrued from the parameter recommendations provided in Appendix E. The
cover profile should retain a layered structure representative of the materials planned for each
layer, but with the hydraulic properties of each layer adjusted to reflect pedogenesis. The
parameters in the 2015 recommendations were presented as a guide for reasonable ranges
consistent with the recommendations in NUREG/CR-7028.
EnergySolutions has conducted a series of analyses to evaluate the impact of erosion on
percolation rates from the cover. In one case, the simulation included loss of 1.2 m of cover soil.
EnergySolutions reports that percolation rates obtained for the full thickness cover and a cover
eroded by 1.2 m are essentially the same.
This is not logical given that the soil in the cover is required to store the water during cooler and
wetter periods, and then release the water during drier periods. The proposed cover is 1.52 m
thick. If the cover thickness is reduced by 1.2 m via erosion, then the soil water storage capacity
of the cover will be reduced by approximately 80%, and the percolation should change
accordingly. This result without supporting analysis makes all of the HYDRUS modeling suspect.
Additional quantitative and mechanistic evidence is needed to support the outcomes in this part
of the report. Water balance graphs, which depict the temporal variation in water balance
quantities (rather than a water balance quantity chart) could be used to illustrate whether the
outcomes are reasonable. Water balance graphs typically are created using daily output
predicted from a water balance model and show the seasonal variation in each water balance
quantity. Examples of water balance graphs are shown in Figure 20-1. These graphs depict
actual water balance data; water balance graphs from a model prediction would be similar. The
soil water storage record in the water balance graph would be compared to the soil water storage
capacity of the eroded profile.
Significantly higher technetium-99 (99Tc) concentrations were obtained for percolation rates
predicted using the hydraulic properties EnergySolutions developed with the recommended
approach (Appendix E, April 2015 SER) relative to the percolation rates predicted in their
previous analyses (Figure 20-2). The differences are very large, which is difficult to understand
given that the percolation rates predicted for the cover are on the order of 1 mm/yr and are
consistent with percolation rates measured for covers placed at other sites in the region.
If the impact on groundwater concentrations is this sensitive to percolation rates on the order of
1 mm/yr, then detailed assessment and proof of the cover design should be particularly
important. EnergySolutions should consider installing a lysimeter to confirm that the cover
modeling is reliable.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 41
2.3.1 Interrogatory Response
In this interrogatory UDEQ proposes contaminant transport from the Federal Cell beginning with
the formation of gullies on the cover leading to increased infiltration rate and leaching of
contaminants into the upper aquifer.
The first part of this interrogatory points to soil erosion issues raised in Section 4.4.2 of the
Safety Evaluation Report (SC&A 2015a). UDEQ expresses concerns that gullies will form and
enhance radon diffusion, deep infiltration, and contaminant transport. EnergySolutions, however,
has provided plans for both ecological and engineering measures to address these concerns by
minimizing gully formation. Demonstration of the stability of the cover is addressed in the
response to interrogatory 71/1.
At the request of UDEQ, EnergySolutions has simulated the impact of erosion on net infiltration
using the conceptual model of in-service (naturalized) covers described in Appendix E of the
SER (SC&A 2015b). These simulations showed that a loss of 1.2 m would have little effect on
the net infiltration rate. UDEQ argues that “This is not logical given that the soil in the cover is
required to store the water during cooler and wetter periods, and then release the water during
drier periods.”
The conceptual model of the in-service cover (naturalization) described in Appendix E of the
SER (SC&A 2015b) and Benson et al. (2011) is that plant and animal activity and freeze-thaw
cycles result in disturbance and mixing of soil layers in the upper portion of the cover system
subject to their influences. In this conceptual model, these processes lead to the development of
soil structure but not soil horizons. The extent of the influence of these processes decreases with
depth of roots, animal burrowing, and frost penetration. This alternative conceptual model of soil
formation is discussed in the response to interrogatory 20/2.
UDEQ maintains that input parameter distributions for the α, n, and Ks hydraulic parameters for
water flow modeling be developed using recommended ranges for hydraulic properties from
NUREG/CR-7028 (Benson et al. 2011). However, these ranges are considered unrealistic. Based
on UDEQ’s conceptual model of formation of soil structure in cover systems, there is no
difference in the in-service properties of what were constructed as storage layers or barrier layers
(Benson et al. 2011). Normally, the distinction in properties between depth intervals in the flow
model that constitute layers used for the cover simulations are the hydraulic properties of the
intervals. The UDEQ conceptual model does not maintain the differences in hydraulic properties
that provide the designed functions of store and release layers and barrier layers. Thus, the cover
can no longer be represented by a layered system. UDEQ objects to this homogeneity in
hydraulic properties in this interrogatory, but that is the only logical outcome of applying the
Benson et al. (2011) method to the ET cover.
Using this conceptual model, the upper portion of the soil profile subject to naturalization
processes is considered to be homogeneous with respect to the hydraulic properties affecting net
infiltration. For the Clive Site, the hydraulic properties of the waste below the cover are modeled
as Unit 3 material and would be subject to the same naturalization processes as the materials
used to construct the cover. As soil is lost due to erosion, disturbance due to biotic activity and
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 42
freeze-thaw extends deeper to maintain the same thickness of “naturalized” soil characterized by
the same ranges of hydraulic properties. The net infiltration is determined by climate and
hydraulic properties. If the hydraulic properties are assumed to be homogeneous and determined
by climate and biotic activity, the loss of material from the surface of the cover due to erosion is
compensated by a deepening of the naturalized soil profile. The thickness of the naturalized soil
remains the same and net infiltration will not change. Given the approximately 10-meter depth of
burial of the DU, the logical projection of this conceptual model is that this process of soil loss
from the surface and deeper extension of soil formation would continue for a very long time
before the soil above the waste was thin enough to influence the net infiltration. The hydraulic
property recommendations and cover material naturalization presented in Benson et al. (2011)
and in Appendix E of SC&A (2015a) are not reasonably applicable for the Clive Site (see the
response to Interrogatory 05/2). Further, the cover design is demonstrated to be stable with
respect to sheet erosion and gully formation. See the response to Interrogatories 71/1 and 191/3
for further discussion on erosion analyses.
UDEQ requests that daily water balance plots of the flow simulation results be provided. UDEQ
has been provided with annual averages for water balance components of precipitation, runoff,
evapotranspiration, storage, and deep drainage. Daily water balance is not the appropriate scale
to evaluate a performance assessment model. See the response to Interrogatory 21/2 for a
discussion of evaluation of flow model water balance.
UDEQ notes that higher technetium-99 concentrations resulted from the simulations using the
unrealistic ranges for hydraulic properties from NUREG/CR-7028 (Benson et al. 2011) requested
by UDEQ. It is not unexpected that unrealistic results will come from a model using unrealistic
input parameters.
2.4 Interrogatory CR R313-25-8(4)(d)-21/2: Infiltration Rates
DEQ Critique of DU PA v1.4, Appendix 5: EnergySolutions/Neptune describe their approach to
parameterizing the radon barriers for v1.4 as follows (pp. 39–40):
An expanded assessment of the performance of the radon barriers was made possible
by developing a distribution for the saturated hydraulic conductivity (Ks) of the radon
barriers to use for the modeling. The Ks values for the radon barriers were sampled from
a distribution developed from a minimum value of 4.32×10-3 cm/day corresponding to the
design specification for the upper radon barrier (Whetstone 2007, Table 8), and 1st,
50th, and 99th percentile values of 0.65 cm/day, 3.8 cm/day, and 52 cm/day,
respectively, which are from a range of in-service (“naturalized”) clay barrier Ks values
described by Benson et al. (2011, Section 6.4, p. 6-12). A shifted lognormal distribution
was fit to the 1st, 50th, and 99th percentiles, and the minimum value of 4.32E-3 cm/day
was used as a shift. The resulting distribution is:
𝐾𝑠 ~ 𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙 𝑔𝑒𝑜𝑚. 𝑚𝑒𝑎𝑛: 3.37 𝑐𝑚/𝑑𝑎𝑦, 𝑔𝑒𝑜𝑚. 𝑠𝑑: 3.23 𝑐𝑚/𝑑𝑎𝑦, with a right shift of
0.00432 cm/day
For all HYDRUS simulations, the same Ks value was applied to both the upper and
lower radon barriers.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 43
Correlations between α and n were investigated by analyzing the combinations of α and
n for the 12 textural classes in Rosetta (Schaap, 2002), and no correlations were
evident. There were also no statistically significant correlations between Ks and α or n.
The developed 50 sets of uncertain parameters for α, n, and Ks were then used as
hydraulic property inputs to 50, 1000-year simulations using HYDRUS-1D.
This approach varies from that taken in DU PA v1.2 as described below (Neptune 2014,
Appendix 5, pp. 41–41):
An expanded assessment of the performance of the radon barriers was made possible
by developing a distribution for the saturated hydraulic conductivity (Ks) of the radon
barriers to use for the modeling. The Ks values for the radon barriers were sampled from
a distribution developed from a minimum value of 4 × 10-3 cm/day corresponding to the
design specification for the upper radon barrier (Whetstone 2007, Table 8), and 50th and
99th percentile values of 0.7 cm/day and 52 cm/day, respectively, which are from a
range of in-service (“naturalized”) clay barrier Ks values described by Benson et al.
(2011, Section 6.4, p. 6-12). A normal distribution was fit to the 50th and 99th
percentiles, and the minimum value of 4E-3 cm/day was used as a shift. For all
HYDRUS simulations, the same Ks value was applied to both the upper and lower radon
barriers.
Correlations between α and n were investigated by analyzing the combinations of α and
n for the 12 textural classes in Rosetta (Schaap, 2002), and no correlations were
evident. There were also no correlations between Ks and α or n.
The developed 20 sets of uncertain parameters for α, n, and Ks were then used as
hydraulic property inputs to 20 1000 year simulations using HYDRUS-1D.
The infiltration results for v1.4 are presented on p. 45 of Appendix 5:
The 50 HYDRUS-1D simulations resulted in a distribution of average annual infiltration
into the waste zone, and average volumetric water contents for each ET cover layer.
Infiltration flux into the waste zone ranged from 0.0067 to 0.1 mm/yr, with an average of
0.024 mm/yr, and a log mean of 0.01 mm/yr for the 50 replicates.” These fluxes are
significantly lower than those calculated in v1.2 and provided on p.45 (Appendix 5)
“Infiltration flux into the waste zone ranged from 0.007 to 2.9 mm/yr, with an average of
0.42 mm/yr, and a log mean of 0.076 mm/yr for the 20 replicates.
Since it appears that the greatest change between v1.2 and v1.4 is that the Ksat values were
increased in v1.4, it is not clear why the infiltration rates would decrease since increasing Ksat
values are typically accompanied by increasing infiltration rates. However, deciphering why the
predictions differ is nearly impossible with the output provided. Understanding the outcome
requires water balance graphs showing the seasonal hydrologic cycle and the dynamics of water
throughout the year. The difference in the predictions may have to do with the shape of the
normal distributions that were used. They are similar, but as described below using the lower
bound constraint may have affected the distribution of K values that are predicted.
Probability density functions (PDFs) are shown in Figure 21-1 that were used to describe
uncertainty and spatial variability in the saturated hydraulic conductivity of the radon barrier in the
Unsaturated Zone Modeling reports submitted in June 2014 and October 2015. A PDF is
analogous to a histogram, describing the probability density associated with a particular value of
the random variable for a defined probability distribution (in this case, the three-parameter log-
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 44
normal distribution). The distributions for 2014 and 2015 were parameterized to the extent
practical using the methodology described in the 2014 and 2015 Unsaturated Zone Modeling
reports. A three-parameter log-normal distribution was used given that the 2014 and 2015
reports indicate that a lower bound > 0 was stipulated in the 2014 and 2015 reports. A
description of the three-parameter log-normal distribution can be found in Zhai and Benson
(2006).
For 2014, the distribution was parameterized using a lower bound (x) = 0.004 cm/d, a log-mean
(µ) of -0.357 corresponding to a 50thpercentile of 0.7 cm/d, and a log-standard deviation (s) of
1.85. The lower bound and log-mean are equal to the values stipulated in the 2014 report. The
log-standard deviation was obtained iteratively by ensuring that the 99th percentile equaled
52 cm/d, as described in the 2014 report.
Two PDFs are shown for 2015 in Figure 21-1 below because the fitting methodology and
parameters cited in the 2015 report lead to ambiguity. The PDF marked “2015 reported”
corresponds to x = 0.00432 cm/d (lower bound indicated in 2015 report), µ = 1.215
(corresponding to geometric mean of 3.37 cm/d indicated in 2015 report), and s = 1.17
(corresponding to 3.23 cm/d referred to in 2015 report as the “geom. sd”). These parameters
(“2015 reported”), however, do not yield a 1st percentile of 0.65 cm/d and a 99th percentile of
52 cm/d as indicated in the report (mathematically impossible). Thus, a second parameter set
was selected (referred to as “2015 reported and fit”). This parameter set has x = 0.00432 cm/d
(lower bound indicated in 2015 report), µ = 1.215 (corresponding to geometric mean of 3.37 cm/d
indicated in 2015 report), and s = 1.17. The log-standard deviation (s) was selected by iteration
so that the 99th percentile equaled 52 cm/d, as indicated in the report. However, the 1st percentile
could not be matched along with the 99th percentile (mathematically impossible). The 1st
percentile hydraulic conductivity for the distribution “2015 reported and fit” is 0.1 cm/d.
The PDFs in Figure 21-1 provide insight into the unexpected outcomes for the percolation rates
predicted in 2014 and 2015, the latter percolation rates being lower despite substantially higher
geometric mean saturated hydraulic conductivity. For the PDF marked “2015 reported and fit,”
which seems to be the PDF most likely used as input to the model, the upper tail of the
distribution is much lighter than for the 2014 PDF (e.g., the probability of high hydraulic
conductivities is lower in the 2015 modeling). Consequently, the percolation rates tend to be
lower in the 2015 report relative to those in the 2014 report. This would not be the case if the
parameters corresponding to “2015 reported” were used as input to the model, as the PDF for
this case generally has a heavier upper tail relative to the PDF used as input to the 2014 model.
This ambiguity highlights an important issue: reports issued by EnergySolutions should include
sufficient information for an independent party to reproduce the outcomes without ambiguity. At a
minimum, probabilistic descriptions should show a mathematical description of the distribution
employed (e.g., probably distribution and definition of parameters) and a list of the values
assigned to each parameter for each case being analyzed.
DEQ Critique of DU PA v1.4, Appendix 21:
Naturalized Cover
Significant disagreement remains regarding appropriate hydraulic properties to represent
“naturalized” conditions (EnergySolutions nomenclature) for the cover. EnergySolutions states
correctly that hydraulic properties they developed with the approach recommended in Appendix
E to the 2015 SER are significantly different from those used in their previous analyses for the
DU PA. This is not surprising, as the hydraulic properties EnergySolutions had used in previous
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 45
analyses (Bingham Environmental 1991) were obtained nearly three decades ago using poorly
documented sampling and testing methods. Techniques for undisturbed sampling and
measurement of unsaturated hydraulic properties have improved dramatically since the Bingham
Environmental data set was created. The quality and relevancy of the Bingham Environmental
data used by EnergySolutions is suspect, and there is good reason for hydraulic properties
obtained using the approach recommended in Appendix E (April 2015 SER) to differ significantly
from those EnergySolutions has used in past analyses.
EnergySolutions also states that the parameters sets obtained with the approach recommended
in Appendix E (2015 SER) “are conservative” and “do not represent the likely evolution of the
cover system.” EnergySolutions also states that the model predictions “do not make sense.”
EnergySolutions will need to provide quantitative evidence to support these assertions. The
photographs in Figures 4 and 5 of Appendix 21 are inconclusive and provide no quantitative
basis to support inferences that structural development and alterations in hydraulic properties do
not occur at Clive. Structural development that occurs in covers due to pedogenesis generally is
not visible at the scale represented in these photographs. Moreover, the smearing that occurs in
test pits can obscure structure that is present. If EnergySolutions wishes to use these analogs as
evidence to support hydraulic properties representing long-term conditions significantly different
from NUREG/CR-7028 (Benson et al. 2011), EnergySolutions should conduct appropriate
measurements on these in-place materials to demonstrate that the hydraulic properties are
indeed different from the abundance of data in NUREG/CR-7028.
EnergySolutions goes on to argue that the Clive location is not represented properly using the
data set in NUREG/CR-7028, and indicates that less extensive pedogenic change should be
expected at Clive relative to the sites in NUREG/CR-7028. They attribute more extensive
pedogenic change to a greater abundance of biota as well as surface and subsurface biomass at
sites in humid climates, which is incorrect. Changes in hydraulic properties due to pedogenesis
are predominantly caused by cycling in state of stress due to seasonal changes in pore water
suction. Those cycles tend to be larger in arid regions than in humid regions, which promotes
greater volume change and more rapid pedogenesis. In fact, conceptually, pedogenesis should
occur more rapidly, and be more extensive, in a more arid climate such as Clive relative to a
more humid climate. However, as shown in NUREG/CR-7028, climate effects are not significant
over time, as structure develops and hydraulic properties are altered in essentially all climates.
EnergySolutions also suggests that the Clive site is outside the range of sites represented in the
data included in NUREG/CR-7028. DEQ does not agree with the suggestion that the semi-arid
climate at Clive is greatly different from the climate at sites in Apple Valley, California, Monticello,
Utah, or Boardman, Oregon. Each of these sites is semi-arid to arid and not greatly different from
Clive. To further address this issue, data from other sites in the region should be considered as
discussed in Interrogatory 189. These sites include the Monticello Uranium Mill Tailings
Repository, the Blue Water Uranium Mill Tailings Reclamation Site near Grants, New Mexico,
and the Cheney Disposal Facility near Grand Junction, Colorado. While none of these sites has
the same climate as Clive, they are sufficiently similar to be considered reasonable analogs. An
argument against the relevancy of these analogs, especially without data, is not logical.
Homogeneous Cover
EnergySolutions has used a homogeneous cover profile in the most recent simulations. This was
not the intent of our previous comments, and was misconstrued from the parameter
recommendations provided in Appendix E of the 2015 SER. The cover profile should retain a
layered structure representative of the materials planned for each layer, but with the hydraulic
properties of each layer adjusted to reflect pedogenesis. The parameters in the 2015 SER
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 46
recommendations were presented as a guide for reasonable ranges consistent with the
recommendations in NUREG/CR-7028.
Correlation and Range of Hydraulic Properties
The hydraulic properties EnergySolutions developed based on the multivariate normal random
generator as recommended by DEQ/SC&A are consistent with those in NUREG/CR-7028 for
“naturalized” conditions. The cross-correlation structure between Ks and α, based on ln Ks and
ln alpha, is also consistent with the literature, as shown in Figure 21-2.
The scatter in this correlation is characteristic of real data, and the correlation is realistic.
However, the range is constrained for both Ks and α because EnergySolutions used the lower-
end standard deviation provided in the 2015 Appendix E SER recommendations. A broader
range would have been obtained using the typical and high-end recommendations for the
standard deviation.
EnergySolutions indicates that the lower end standard deviation was used “to keep the input
parameters within the ranges” of the 2015 Appendix E SER recommendation, which was not the
intent of the recommendation. EnergySolutions should conduct their simulation using a typical
standard deviation for each parameter. This will likely affect only the tails in the percolation data
(high and low percolation rates in Figure 2 of Appendix 21) but likely will affect the 95th percentile
doses (reported in Table 5 of Appendix 21).
Furthermore, the NUREG/CR-7028 recommended range of α values utilizes averaged values for
the entire cover system for each embankment studied in the NUREG, not individual sampling
points, or small parts of an embankment. The information is already presented at the scale
needed for application to a single cover system on a single embankment. Therefore, either
upscaling, or sub-sampling of the data, by Neptune to get a narrower range of α values for an
embankment cover system would be neither necessary nor appropriate.
For all sets of realizations, the mean and the standard deviation (or ln std deviation for Ks and
alpha) should be cited.
Unsaturated Flow Model Output
Percolation rates predicted with the hydraulic properties developed by EnergySolutions using the
procedure recommended in Appendix E to the 2015 SER are reasonable and consistent with
percolation rates measured and predicted for other final covers in regions of similar aridity, as
reported in NUREG/CR-7028. EnergySolutions predicts percolation rates ranging from 0.57 to
1.31 mm/yr using hydraulic properties developed with the procedure recommended by
DEQ/SC&A. As a comparison, percolation ranging from 0.0 to 3.8 mm/yr have been measured
using an ACAP lysimeter at the U.S. Department of Energy’s (DOE’s) Monticello U-Mill Tailings
Disposal Facility in Monticello, Utah, over the period 2000–2016. Percolation rates at other arid
or semi-arid sites described in NUREG/CR-7028 with comparable cover profiles include Apple
Valley, California (0–1.8 mm/yr), Boardman, Oregon (0 mm/yr), and Underwood, North Dakota
(1.9–9.4 mm/yr).
As in past reports from EnergySolutions, the model predictions are difficult to interpret and
evaluate with the level of detail provided. We have requested water balance graphs (see CR
R317-6-2.1-20/2, Figure 20-1), which depict the important interplay between the water balance
quantities throughout the water year. EnergySolutions has included an annualized water balance
chart (Figure 3, Appendix 21), but this chart does not provide the additional information or insight
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 47
that is necessary for a proper evaluation of the model predictions. Water balance graphs should
be provided.
Regression Model
The regression model used in GOLDSIM was updated using predictions obtained with the
hydraulic properties EnergySolutions developed based on the method recommended in
Appendix E to the April 2015 SER. This model relates the average annual percolation rate into
the waste to the hydraulic properties of the cover soils. The regression method is not described
in Appendix 21, but is likely the same method used by EnergySolutions in the past. Appendix 21
does not include supporting statistics confirming the significance of the regression and each of
the independent variables included in the regression model. Thus, the efficacy of the regression
cannot be evaluated.
Percolation rates predicted with the regression model and obtained directly from HYDRUS show
a good comparison (see Figure 6 of Appendix 21). This is expected, because the regression
model is based on the HYDRUS output. A concern raised before, and yet unresolved, is whether
good agreement would exist between percolation rates predicted with the regression model and
an independent set of predictions from HYDRUS using the same underlying inputs (e.g., a blind
forward comparison). That type of evaluation is needed to confirm the validity of the regression
model. For example, if an analysis was conducted with the typical standard deviations to obtain a
broader range in outcomes, would the comparison between the predictions from the regression
model and predictions from HYDRUS be in comparable agreement?
At a minimum, EnergySolutions should conduct an independent set of simulations where
percolation is predicted with HYDRUS and then compared with predictions obtained with the
regression model. This is the only fair means to evaluate the efficacy of the regression model.
These predictions should be conducted with the typical standard deviations to get a realistic
representation of the tails of the distribution of percolation.
2.4.1 Interrogatory Response
Radon Barrier Ks Distribution
UDEQ raises a number of issues in this interrogatory. The first is a request for clarification of
what probability distribution was used in v1.4 for the saturated hydraulic conductivity, Ksat, of
the radon barriers. The distribution is clearly provided in Table 16 of the Clive PA Model
Parameters v1.4 document (shown below in Figure 12), and is stated correctly in the
interrogatory as a three-parameter log-normal distribution with “x = 0.00432 cm/d (lower bound
indicated in 2015 report), µ = 1.215 (corresponding to geometric mean of 3.37 cm/d indicated in
2015 report), and s = 1.17 (corresponding to 3.23 cm/d referred to in 2015 report as the “geom.
sd”).” It appears the confusion regarding the distribution stems from the difference between
actual quantiles (percentiles) of the final distribution and the target quantiles that were used as a
basis to inform the development of the distribution. Therefore, this response describes why this
difference exists and provides more detail regarding how the final distribution was selected
based on the available information.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 48

Figure 12. Table 16 from the Clive PA Model Parameters v1.4 document providing the
distribution used. A description of the method used to select the geometric mean,
geometric standard deviation, and minimum reported in this table, and the associated
parameterization of the log normal were also provided in Appendix 14 of the Clive
DU PA Model Final Report for v1.4.
For DU PA v1.4, the Ks distribution was based on 1st, 50th, and 99th target percentiles elicited
from the relevant literature (Benson et al. 2011) to capture plausible values of Ks that could apply
to the entire cover described by the distribution. A lognormal distribution was chosen as the
distributional form to capture skew toward larger values. Parameters of the lognormal
distribution were chosen such that the percentiles of the final distribution were close to the target
percentiles.
In this case, the following information was used to obtain target quantiles for the distribution of
Ks, where the values included in the distribution are meant to represent possible values for the
entire cover (as opposed to small cores from different locations within the cover):
• Per Section 6.4, p. 6-12 (Benson et al. 2011), “For all but one site, Ks falls within the
range of 7.5 × 10-8 and 6.0 × 10-6 m/s regardless of cover type…”. The lower value of
7.5 × 10-8 m/s is 0.648 cm/day after conversion. This was rounded to 0.65 cm/day, and is
used as the elicited target for the 1st percentile.
• The upper value quoted above, 6.0 × 10-6 m/s, is 51.84 cm/day after conversion. This is
rounded to 52 cm/day, and was used as the elicited target for the 99th percentile.
• Per Section 6.4, p. 6-12 (Benson et al. 2011), “The geometric mean in-service hydraulic
conductivity is 4.4 × 10-7 m/s.” This geometric mean is 3.8016 cm/day after conversion
from m/s to cm/day, and this was rounded to 3.8 cm/day. The geometric mean of a
lognormal distribution is equal to the median, so this value was used as the elicited target
for the 50th percentile of the distribution.
• A minimum value of 0.00432 cm/day was also specified to correspond to the design
specification for the layers (Whetstone Associates 2011). This minimum value was used
for the final distribution, but not to define a target percentile.
Neptune uses the open-source statistical software package R (R Core Team 2017), and functions
to facilitate fitting of distributions to target quantiles, or percentiles. The function implements an
optimization algorithm to find the log scale mean, log scale standard deviation, and, for a three-
parameter distribution, a minimum value that together provide a lognormal distribution with
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 49
quantiles close to the targets. Due to the restrictions on the shape of the lognormal distribution, it
is not possible to exactly match percentiles using only two parameters and the usual lower bound
of 0; the goal is finding a distribution with percentiles close to, but not exactly equal to, the
targets.
As a starting point, a three-parameter lognormal distribution was fit to provide a close match to
the target percentiles. With three parameters to vary, the target quantiles were achieved, but this
corresponded to a minimum value (implemented as a shift) of 0.43 to go with log-scale mean and
standard deviation of 1.215 and 1.17, respectively. That is, the minimum value chosen shifts the
lognormal distribution defined by a lower bound of zero by 0.43, and after shifting the
distribution quantiles are very close to the targets (see Table 3). However, the minimum value
needed to match the target quantiles is greater than the 0.00432 minimum deemed reasonable
from expert knowledge of the site design, and, therefore, the smaller minimum was used for the
final distribution for the PPA Model to ensure that is it possible to obtain random draws between
0.00432 and 0.42. The mean and standard deviation for the pre-shifted distribution were not
changed. The change in minimum from 0.43 to 0.00432 changes the percentiles of the final
distribution, but these were judged to be close enough to the target quantiles to be reasonable
(see Table 3); the 1st percentile of the final distribution is 0.22 instead of 0.65 and the 99th
percentile is 51.5 instead of 52. Note that a shift of 0.00432 is so small relative to the spread of
the entire distribution that it is not discernible from a lower bound of zero in a plot of the
probability density function (Figure 14) and it only changes the percentiles in the third decimal
place.
Table 3 provides the percentiles from the distributions discussed above, compared to the target
percentiles used to inform the distribution, and Figure 14 shows the associated probability
density function (PDF) for the distribution used for v1.4 and reported in Figure 12. As is clear
from Table 3, percentiles of the selected lognormal distribution are close to the elicited
percentiles, while also adhering to the minimum chosen based on design knowledge of the Site.
The interrogatory presents Figure 21-1, along with curves meant to represent PDFs associated
with different possible distributions used. It is not clear from the description how the curves were
created for the “2015 reported” and “2015 reported and fit.” By definition, the area under a PDF
must integrate to 1, and this is not the case for the curves displayed in Figure 21-1. It appears
from the text that the “2015 reported” curve should match the PDF displayed in Figure 14, but it
does not (the y-axis values between the two plots do not match). Because they are not truly
probability density functions, it is difficult to evaluate them and to attempt to match them to
actual probability distributions. For additional clarification of the percentiles associated with the
final distribution, the code to get the percentiles from the final lognormal distribution using R
statistical software is provided in a screenshot in Figure 13; the code also explains how the
percentiles reported in Table 3 were obtained.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 50

Figure 13. R Statistical Software (R Core Team 2017) code and output for getting quantiles
from the distribution of Ks used in v1.4.

Table 3. Percentiles associated with elicited information for the Ks distribution (cm/day),
and the distribution actually used. These are based on a lognormal distribution with
geometric mean of 3.37 cm/day and a geometric sd of 3.23 cm/day, with shifts for the
minimum associated with each row. See Figure 13 for example R code to get the
percentiles.
1st Percentile 50th Percentile 99th Percentile
Elicited Percentiles (Target) 0.65 3.8 52
Percentiles from distribution
(minimum=0, i.e., no shift)
0.220 3.370 51.548
Percentiles from distribution
with minimum allowed to
vary to match quantiles
(minimum = 0.43)
0.650 3.800 51.978

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 51
Figure 14. The lognormal distribution used for the Ks in v1.4. It is parameterized by a
geometric mean of 3.37 (log-scale mean 1.215), a geometric standard deviation of 3.23
(log-scale standard deviation of 1.17), and a minimum of 0.00432 implemented
through a shift of the distribution. The target 1st, 50th, and 99th percentiles are shown
by the vertical dotted lines.
A log-triangular distribution was not used for this fit. Please see the response to Comment B.3
for a description of problems with using a log-triangular distribution.
UDEQ requests that “At a minimum, probabilistic descriptions should show a mathematical
description of the distribution employed (e.g., probably [sic] distribution and definition of
parameters) and a list of the values assigned to each parameter for each case being analyzed.”
The description of hydraulic parameter distributions and the values of each of the parameters
used for each flow simulation case were provided in Appendix 5 of the Clive DU PA v1.4
(Neptune 2015c). Values for parameters that were held constant were provided in Table 8 of
Neptune (2015c). Development of a probability distribution for Ks of the radon barriers is
provided here again in the response to Interrogatory 21/2. Values of each parameter varied for
the 50 simulations are provided again in the response to Comment B.2.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 52
Naturalized Cover
The next issue raised by UDEQ in this interrogatory is objection to the use of site-specific
hydraulic property data provided in a report by Bingham Environmental (1991). UDEQ has
criticized the use of hydraulic property results from testing of samples of Unit 3 and Unit 4
materials at the Clive Site by Colorado State University that were published in a report prepared
by Bingham Environmental (see also UDEQ Comment B.4). UDEQ states that “Techniques for
undisturbed sampling and measurement of unsaturated hydraulic properties have improved
dramatically since the Bingham Environmental data set was created.”
This statement is simply not true as there has been little change to the standard methods used to
determine saturated hydraulic conductivity, volumetric water content, and water retention
relations. The method commonly used to estimate the van Genuchten hydraulic function
parameters from water retention data is the RETC software published in 1991 by van Genuchten
et al. (1991). UDEQ states further “The quality and relevancy of the Bingham Environmental
data used by EnergySolutions is suspect.” And, with respect to van Genuchten parameters
derived from the testing, “This α is based in part on historic measurements made at Colorado
State University on core samples obtained at the Clive site by Bingham Environmental (1991),
which are known to be too small and too disturbed to adequately represent in-service conditions.
The relevancy of this historic data from Bingham Environmental is dubious, at best.”
There is no indication in the text of Bingham Environmental (1991) or in the test result sheets
from Colorado State University that the data were considered irrelevant, suspect, or dubious.
Colorado State University hosts an internationally recognized Civil and Environmental
Engineering Department. Their facilities include the Groundwater and Porous Media Laboratory,
where testing of the samples from the Clive Site was conducted. UDEQ has provided no basis
for disparaging the technical capabilities of this laboratory.
Following these statements, UDEQ reiterates their position that hydrogeologic models are not
site specific. They argue that statistical distributions for hydraulic property parameters used for
flow modeling of cover material described in Benson et al. (2011); SC&A (2015a, 2015b) should
be universally applied irrespective of climate, ecology, or geologic setting of the site. See the
responses to Interrogatory 05/2 and Interrogatory 153/2. UDEQ lists a number of waste disposal
sites that they consider to be analogs to the Clive Site: Monticello Uranium Mill Tailings
Repository, the Blue Water Uranium Mill Tailings Reclamation Site near Grants, New Mexico,
and the Cheney Disposal Facility near Grand Junction, Colorado. The response to Interrogatory
192/3 discusses why these sites are not suitable analogs for the Clive Site.
Homogeneous Cover
UDEQ raises objections in this interrogatory regarding how simulations were conducted using
in-service cover properties as requested by UDEQ. An alternative set of 50 HYDRUS
simulations was conducted using input parameters derived from the distributions and methods
described by Dr. Craig Benson in Volume 2, Appendix E, of the Safety Evaluation Report (SER)
prepared by SC&A (SC&A 2015a), consistent with the request of UDEQ to use this approach.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 53
These models represent modifications to previous models required in response to the SER issues.
These models are conservative and do not represent the likely evolution of the cover system.
UDEQ objects to use of an essentially hydrologically homogeneous cover for these alternative
simulations, stating that:
This was not the intent of our previous comments, and was misconstrued from the parameter
recommendations provided in Appendix E of the 2015 SER. The cover profile should retain
a layered structure representative of the materials planned for each layer, but with the
hydraulic properties of each layer adjusted to reflect pedogenesis. The parameters in the
2015 SER recommendations were presented as a guide for reasonable ranges consistent with
the recommendations in NUREG/CR-7028.
Given the UDEQ conceptual model of making no distinction between the properties of storage
and barrier layers, the cover can no longer be represented by a layered system. Please see the
response to Interrogatory 10/3 for a discussion of the consequence of using the Benson et al.
(2011) method for developing hydraulic input parameters.
The hydraulic property recommendations and cover material naturalization presented in Benson
et al. (2011) and in Appendix E (SC&A 2015a) are inappropriate for the Clive Site. See the
response to Interrogatory 05/2.
Correlation and Range of Hydraulic Properties
See the response to Interrogatory 05/2 regarding the lack of correlation of hydraulic properties of
Unit 4 soil.
In the application of the Benson et al. (2011) and Benson and Gurdal (2013) method in Appendix
E (SC&A 2015a), the low variance option was selected for the spreadsheet calculations specified
in Appendix E for creating sets of parameters for flow modeling. This choice was made because
the “typical” and “high” variance options produced parameter values outside the range of
recommended values. Though not clear in Appendix E (SC&A 2015a), from the references cited
the variances used in the spreadsheet appeared to be estimated from point data, not from mean
values. A distribution of means, associated with a lower variance than a distribution of point
values, is better aligned with the scale of the PA Model. Neither upscaling nor sub-sampling of
the data was used in developing the hydraulic parameter distributions for the alternative set of 50
HYDRUS simulations described in Appendix 21 (Neptune 2015f).
For Clive DU PA Model v1.4 (Neptune 2015c), estimates of the uncertainty in the hydraulic
properties of Unit 4 materials that compose the surface and evaporative zone layers of the ET
cover were needed. These were obtained by using the α and n values from the distributions
(mean and standard deviation) for each parameter from the Rosetta database of hydraulic
parameters for the textural class of silty clay (Schaap 2002). Details of the development of
distributions for α and n parameters for the surface and evaporative zone layers are given in the
response to Comment B.2.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 54
The α and n distributions in Appendix 5 of Clive DU PA Model v1.4 (Neptune 2015c) are based
on the standard error of the average from the Rosetta database, rather than the sample standard
deviation of the data. The reason for the use of the standard error of the average is to develop the
distribution at the spatial scale consistent with how the values will be used in the PA Model (see
Section 1.1 and Appendix 14 of the Clive DU PA Model v1.4 (Neptune 2015e) for more
discussion of scaling). The Rosetta database is comprised of measurements from point locations
or single cores, thus representing a smaller scale than the site-scale values needed for the PA
Model. Averages of the small-scale values are used to represent larger site-scales consistent with
the scale represented by GoldSim cells in the model. The standard error of the average represents
the expected variability among averages from different random datasets of 28 measurements
(i.e., the approximate sampling distribution of the average), or can be thought of as representing
uncertainty in the value of the true population mean, so that values from the distribution
represent possible site-scale averages (or means).
Unsaturated Flow Model Output
UDEQ compares the net infiltration rates simulated using the alternative distributions based on
the hydraulic property recommendations and cover material naturalization presented in Benson
et al. (2011) and in Appendix E (SC&A, 2015) with rates from a number of sites they regard as
analogs to the Clive Site. These hydraulic property recommendations are inappropriate for the
Clive Site. See the response to Interrogatory 189/3 for a discussion on the applicability of these
sites as analogs.
UDEQ requests plots of flow model water balance components on a daily basis. These
components are precipitation, runoff, infiltration, evaporation, transpiration, storage, and net
infiltration (deep drainage).
Steady-state annual averages of net infiltration and water content from the HYDRUS simulations
are the model results used to develop statistical distributions of these parameters for inputs to the
GoldSim model for the Clive DU PA. The GoldSim model samples inputs randomly at the
beginning of time for a single realization of the simulation, and those values that represent the
input distributions are used throughout the realization time. The model was deliberately run for a
long period of time (1,000 years) in order to reach a near-steady state net infiltration rate that is
not influenced by the initial conditions. Flow model results for use in the GoldSim DU PA model
consisted of the daily fluxes from the bottom of the Lower Radon Barrier and the water contents
for the Surface Layer, Evaporative Zone Layer, Frost Protection Layer, and Upper and Lower
Radon Barriers averaged over the years 900 to 1000 (36,525 days) of the simulation. Changes in
storage were zero when averaged over the last 100 years of the simulations. Runoff was also
negligible. The water balance plots showed the remaining components of water balance:
precipitation, evaporation, transpiration, and net infiltration as annual averages.
Water balance plots were provided to UDEQ for a range of net infiltration rates in Appendix 21
of the Clive DU PA v1.4 (Neptune 2015f). The results of five simulations were shown
representing the 10th, 30th, 50th, 70th, and 90th percentile net infiltration values from 50 HYDRUS
flow model simulations. Average annual values for each nonzero water balance component
(precipitation, evaporation, transpiration, and net infiltration) were presented. Using annual
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 55
average parameter values and their associated uncertainty reflects the spatial and temporal scale
of the net infiltration and water content in the DU PA Model. The variability of data at points in
time and space is not the appropriate representation of variance in a PA model.
Regression Model
In this section of the interrogatory UDEQ raises four different issues regarding the model
abstraction that are intermingled in the interrogatory comments. This section of the response
begins by reiterating the goal of the model abstraction, and then addresses the issues separately
to avoid further conflating them.
Goal of model abstraction
As described in Section 1.1 and Appendix 14 (Neptune 2015e), the goal of “model abstraction”
in the context of PA modeling is to build a relatively simple statistical model to approximate the
relationship between the input parameters and an output of interest obtained from a sophisticated,
and usually computationally intensive, process-based model. For example, net infiltration may be
calculated using a process-based model like HYDRUS for a given set of input parameters.
The process model, HYDRUS in this case, is run at many combinations of input parameters,
chosen to explore the relevant region of the multi-dimensional parameter space of the inputs. The
goal is to explore the parameter space of the input parameters as much as is possible for a given
number of HYDRUS runs in order to gain as much information as possible to predict the
response for new combinations of input parameter values that may be used for a PA model run. It
is often the case that a relatively simple statistical model can be used to predict the output of a
very complicated process model, not by re-approximating the physical processes represented in
the process model, but just by capturing the overall relationship, perhaps as a function of few
input parameters and simple functional forms. The statistical model will predict a reasonable
value that would have been obtained from HYDRUS and is not expected to match the HYDRUS
result exactly. HYDRUS is an approximation of reality, and the model abstraction is a simple
approximation to HYDRUS. Uncertainty in predictions from the statistical model can also easily
be incorporated into distribution development if desired.
The success of the statistical model in predicting the output depends on the information
contained in the original runs of the HYDRUS model used to fit the model. There are three
settings to consider when evaluating the predictive ability of the statistical model for use in a PA
model: (1) predictive success at the same input parameter values used in the original HYDRUS
runs (in-sample prediction), (2) predictive success at new input parameter values within the
region explored in the original HYDRUS runs (out-of-sample prediction based on interpolation),
and (3) prediction success at input parameter values beyond the region explored in the original
HYDRUS runs (extrapolation).
Need for “significance of the regression” for assessing its efficacy
The goal of building a statistical model for prediction differs from the goal of developing a
model to understand and explain relationships (e.g., Shmueli (2010); Ramsey and Schafer
(2013); Rawlings et al. (1998)). The different goals lead to different strategies and statistical
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 56
methods. Statistical models developed for the goal of prediction are often different from those
developed for the goal of explanation, and some of the tools developed for use when the goal is
explanation should not be used in the same way.
The linear regression models developed for predicting water content and net infiltration are
based on the results of modeling performed using the HYDRUS software. The model for net
infiltration is the focus of the interrogatory comments. Even when the goal is prediction,
statistical software typically provides p-values for individual regression coefficients; this does
not imply they should be used to assess the “efficacy of a regression” for prediction as suggested
in the interrogatory comments. In this case, the model for net infiltration contains α and n as
predictors, with p-values of 6.36×10-12 and 0.000108, respectively. The hypothesis tests
associated with these p-values for individual coefficients should not be used simultaneously to
assess predictive ability of the model (despite common misuses in practice). Instead, they simply
provide a summary of the degree of statistical evidence that the slope of the linear relationship
between the predictor (α or n) and the mean net infiltration differs from zero, while holding the
other predictor constant. For the goal of prediction, any variable thought to be useful for
prediction can be included in the model (Kutner et al. 2005; Ramsey and Schafer 2013). Use of a
large number of predictors, including some that contribute little to predictive performance, can
lead to larger standard errors of prediction, but in this case the standard errors of prediction are
not ultimately used and the number of predictors is so small (two or three) that there is no danger
in including too many predictors. Instead of relying on p-values, the usefulness of the regression
model should be primarily judged by the plot of the predicted values from the regression vs. the
original values obtained from HYDRUS, as well as by measures of out-of-sample predictive
performance, such as those obtained from hold-out datasets and/or cross-validation, both of
which are provided in this response.
A comparison of the predictions from the regression model for net infiltration rate (mm/yr) to the
original results from HYDRUS are shown in Figure 15. For simplicity, the linear regression
model was developed using the inputs and net infiltration results on the scales ultimately used as
inputs to the PA Model, rather than on a logarithmic scale. It is expected that the linear
regression will miss some curvature found in the Clive DU PA Model v1.4 HYDRUS results
(Neptune 2015c), but it is developed as a simple and easy-to-implement approximation that can
be easily incorporated in the PA Model using GoldSim. The R-squared value indicates the
proportion of variance in HYDRUS results explained by the linear regression predictions; 0.67 is
deemed reasonable given the observed relationship in Figure 15. The bulk of the net infiltration
predictions from the regression (between approximately .01 and 0.05 mm/yr) are slightly larger
than the values from HYDRUS, providing a conservative approximation within that range.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 57

Figure 15. The average annual net infiltration values obtained from HYDRUS compared to
the predictions from the linear regression model abstraction. The results from the 50
HYDRUS realizations were used to develop the regression model abstraction, and
therefore this plot depicts in-sample predictive performance. The one-to-one line is
shown for reference.
The existence of “typical standard deviations”
The standard deviation used for a distribution representing the current state of knowledge in an
input parameter for a PA model should be evaluated relative to the temporal and spatial scale it is
meant to represent in the PA model (see Section 1.1 and Appendix 14 (Neptune 2015e)) for more
discussion on scaling). Therefore, it is not productive to define a “typical standard deviation”
without reference to the scale it is meant to apply to. A “typical” standard deviation representing
large scales will not be the same as a “typical” standard deviation representing small scales. That
is, the variability among individual measurements taken at particular points in space and time is
not expected to be the same as that among quantities meant to describe large volumes of
materials over long periods of time. Hence, the adequacy of one distribution should not be
judged relative to a “typical” distribution developed for a different goal.
●●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●●●
●
●
●
●
●●
●
●
●
●
●●
●●
●
●●●
●
●●
r^2: 0.67
0.00
0.05
0.10
0.15
0.20
0.00 0.05 0.10 0.15 0.20
Predictions of Net Infiltration from Regression (mm/yr)
HY
D
R
U
S

N
e
t

I
n
f
i
l
t
r
a
t
i
o
n

(
m
m
/
y
r
)
Comparison of HYDRUS Results and Regression Predictions
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 58
Evaluation of out-of-sample prediction outside the range needed for the PA Model
(extrapolation)
For the most efficient use of information and the best predictions within the context that the
model abstraction will be used, the abstraction should be developed based on the distributions
used in the PA model and not on different distributions that might be wider to reflect possible
smaller-scale values or different assumptions. Developing the model abstraction for wider
distributions is a waste of the limited resources available (mainly computational time running
HYDRUS in this case) to effectively explore the region of the parameter space needed for the PA
model. That is, it is more efficient to do a better job exploring the specific region of the input
parameter space that will actually be needed for the PA model than to spread the effort out over
values that will not be sampled in the PA model (thereby decreasing the predictive ability within
the region needed for the PA model). Statistical modeling for prediction should always be tied as
closely as possible to the region within which it will ultimately be used to obtain predictions.
Models cannot be assumed to perform well outside of the parameter space they were developed
for unless assumptions that the relationships should continue into other regions of the parameter
space can be justified (e.g., a linear relationship estimated for one region is expected to continue
into neighboring space). In the context of model abstraction for PA models, the original
HYDRUS runs should explore the parameter space covered by the distributions for inputs that
will ultimately be used for the PA model; this avoids the second issue as long as the number of
original runs of the HYDRUS model is large enough to adequately explore the multi-
dimensional space. The model abstraction is not designed to be used with different distributions
for input parameters, particularly if new distributions are wider than those used to inform the
original HYDRUS runs. As described above, it is actually a waste of resources to spread effort
out over a wider range than needed, at the expense of less information in the range of interest.
With this in mind, evaluation of out-of-sample prediction outside the range of inputs needed for
the PA model (i.e., extrapolation) is not considered a meaningful goal. The interrogatory
comments suggest this should be considered, but because the regression model is not designed to
be used with different input distributions, this is not undertaken. The values of the inputs for the
50 HYDRUS runs associated with v1.4 are based on the same distributions used in the PA
model.
Evaluation of out-of-sample prediction within range modeled (interpolation)
The model abstraction should be evaluated with respect to out-of-sample prediction within the
range covered by the original inputs to HYDRUS. The best-case performance of the regression
model will be its ability to predict the output for the values of the input parameters used to obtain
the original HYDRUS results and used to fit the regression model (termed “in-sample”
prediction) (see Figure 15). Therefore, it is important to check the ability to predict at new sets of
input values that fall within the region explored by the original HYDRUS runs; this is how the
model will be used to obtain predictions for the PA Model (out-of-sample prediction via
interpolation). The best method for evaluating “out-of-sample prediction” is to run the HYDRUS
model for an additional number of runs, use the model abstraction to predict the output, then
compare the predictions to the actual HYDRUS outputs (evaluation based on an independent
sample). This is often described as using a “hold-out” data set—the model is fit using a subset of
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 59
the data available and then tested independently on the “hold-out” data set. Predictive
performance is not expected to be as good as for in-sample prediction, but if the model is
effective at predicting within the space explored in the original HYDRUS runs, it should be
similar. The downside of the “hold-out” method is that information potentially available to
inform the building of the regression model is not used because it is withheld only to be used for
model evaluation. Another method, k-fold cross-validation, allows use of information from all
the HYDRUS runs while still approximating out-of-sample prediction performance (described in
many statistics textbooks such as Hastie et al. (2009)). Cross-validation proceeds by repeatedly
splitting the available data into a “training set” and a “test set,” where the model is repeatedly fit
for each training set and predictive ability is tested for each test set. Then, the results from all the
predictions are combined to estimate out-of-sample predictive ability. The “k-fold” refers to the
size of the test sets relative to the training sets and also relates to the number of different splits,
and subsequent model fits, needed to carry out the procedure. For each split into a training and
test set, predictive accuracy can be summarized, and then combined across the splits. Each
observation will have a prediction, but the prediction will come from a model fit to a random
subset of 2/3 of the data instead of all the data. Then, typical summaries can be calculated, such
as root mean squared error (RMSE), and R-squared between the original values from HYDRUS
and the predictions.
For the net infiltration linear regression model, 3-fold cross-validation performed with the 50
v1.4 HYDRUS runs (Neptune 2015c) yields an RMSE for out-of-sample prediction of 0.018 and
an R-squared value of 0.57 (Figure 16). The regression model using all of the data for in-sample
prediction has an RMSE of 0.016 and an R-squared value of 0.67 (Figure 15). The in-sample
RMSE is expected to be less than that obtained via cross-validation to approximate out-of-
sample. In this case, the values are judged close enough to trust the out-of-sample prediction
implemented for new sets of input parameters used for the PA Model. A further check compares
the 1000 predictions of net infiltration actually used in the v1.4 PA Model with the 50 net
infiltration rates obtained from the original HYDRUS runs. Figure 17 displays these two sets of
values ordered from smallest to largest and spaced so that the 50 HYDRUS results cover the
same x-axis as the 1000 regression predictions from different inputs. The two lines are expected
to diverge near the minimum and maximum of the x-axis because the results based on 1000
different sets of inputs are expected to explore more extreme regions of the multivariate space
than are observed in the 50, even with the same input distributions. The goal of the model
abstraction is to produce a simple statistical model to provide reasonable predictions quickly and
easily within the PA Model, and the comparison in Figure 17 shows that the predictions are
consistent with those from the HYDRUS model, with a similar range and percentiles. This is
consistent with the cross-validation results. It is easy to think of HYDRUS results as the “truth,”
but it is important for the general context of the modeling to remember that they too are
approximations of reality and the goal of the model abstraction is reasonable predictions of net
infiltration.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 60

Figure 16. The average annual net infiltration values obtained from HYDRUS compared to
the predictions obtained from test datasets via 3-fold cross validation. The original
HYDRUS values predicted were not used to fit the regression equations used to
obtain the predictions. This approximates out-of-sample predictive performance of
the linear regression model abstraction. The one-to-one line is shown on each plot for
reference.
●●
●
●
●
●
●
●●●
●
●
●●●●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●●
●
●
●●
●
●
●●
●
●●
1 2 3
0.00 0.03 0.06 0.00 0.03 0.06 0.00 0.03 0.06
0.00
0.05
0.10
0.15
3−Fold Cross−Validation Predictions of Net Infiltration (mm/yr)
HY
D
R
U
S

N
e
t

I
n
f
i
l
t
r
a
t
i
o
n

(
m
m
/
y
r
)
Net Infiltration Regression Model Cross−Validation Results
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 61

Figure 17. Comparison of 1,000 realizations of net infiltration using the linear model in
GoldSim with the results of the 50 HYDRUS simulations of infiltration.
2.5 Interrogatory CR R313-25-8(4)(a)-28/3: Bioturbation Effects and
Consequences
DEQ Critique of DU PA v1.4, Appendix 5: EnergySolutions/Neptune retain the same
assumptions with respect to biointrusion depths and potential impact on infiltration in v1.4 as
were provided in v1.2.
DEQ Critique of DU PA Appendix 21: EnergySolutions has conducted a series of analyses to
evaluate the impact of erosion on percolation rates from the cover. In one case, the simulation
included loss of 1.2 m of cover soil. EnergySolutions reports that percolation rates obtained for
the full thickness cover and a cover eroded by 1.2 m are essentially the same.
This is not logical, given that the soil in the cover is required to store the water during cooler and
wetter periods, and then to release the water during drier periods. The proposed cover is 1.52 m
thick. If the cover thickness is reduced by 1.2 m via erosion, then the soil water storage capacity
of the cover will be reduced by approximately 80%, and the percolation should change
accordingly. This result without supporting analysis makes all of the HYDRUS modeling suspect.
Additional quantitative and mechanistic evidence is needed to support the outcomes in this part
of Appendix 21. Water balance graphs, which depict the temporal variation in water balance
quantities (rather than a water balance quantity chart) could be used to illustrate whether the
outcomes are reasonable. Water balance graphs typically are created using daily output
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 62
predicted from a water balance model and show the seasonal variation in each water balance
quantity.
Examples of water balance graphs are shown in Figure 20-1 (CR R317-6-2.1-20/2). These
graphs depict actual water balance data; water balance graphs from a model prediction would be
similar. The soil water storage record in the water balance graph would be compared to the soil
water storage capacity of the eroded profile.
Clive lies in an area having a semi-arid climate. Only certain types of plants grow readily at Clive.
Very little grass grows there. It’s difficult to see how the limited variety and density of plants will
provide adequate vegetative cover for erosion protection on an embankment. EnergySolutions
should find and document natural analogs in the area that support their predictions, particularly
since the predicted erosion rates appear too low to be realistic.
A related concern is the importance of the biological soil crust for sustaining plant growth and the
high uncertainty regarding its characteristics at the Clive site. EnergySolutions should provide
examples with quantitative data from sites in similar climate and with similar soils. These
examples should show how biological soil crust is preserved or re-established, the timeline for
re-establishment, and how presence (or not) of the biological soil crust affected erosion.
2.5.1 Interrogatory Response
This interrogatory begins by pointing to Section 4.4.3 of the SER (SC&A 2015b), stating that
EnergySolutions has not demonstrated that the ET cover design is adequate to protect against
intrusion by plants, animals, or ants. Field studies and analyses conducted by SWCA at the Clive
Site and nearby ecologically analogous sites demonstrate that disturbance of the cover by plants,
mammals, and ants will be negligible. See the response to Interrogatory 05/2.
UDEQ then raises concerns with the results of their requested simulations of the impact of
erosion on net infiltration using the conceptual model of in-service (naturalized) covers described
in Appendix E of the SER (SC&A 2015a). These simulations showed that a loss of 1.2 m would
have little effect on the net infiltration rate. UDEQ argues that “This is not logical given that the
soil in the cover is required to store the water during cooler and wetter periods, and then release
the water during drier periods.” This outcome is the result of applying UDEQ’s conceptual
model of soil formation processes in the cover materials. See the response to Interrogatory 20/2.
The next matter raised by UDEQ in this interrogatory is a request for daily water balance plots of
the flow model results. UDEQ has been provided with annual averages for water balance
components of precipitation, runoff, evapotranspiration, storage, and deep drainage. UDEQ
claims that these water balance plots are not adequate and have produced numerous examples of
daily water balance plots. Daily water balance is not the appropriate scale for evaluating a
performance assessment model. See the response to Interrogatory 21/2 for a discussion of the
evaluation of flow model water balance, and also see Section 1.1, Modeling for Probabilistic
Performance Assessment.
Next UDEQ argues that “It’s difficult to see how the limited variety and density of plants will
provide adequate vegetative cover for erosion protection on an embankment.” UDEQ requests
that natural analogs in the area be identified and documented to support predictions of adequate
plant cover. As described in the response to Interrogatory 71/1, SWCA Environmental
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 63
Consultants (SWCA) assessed erosion under undisturbed conditions at Clive in June 2012
(SWCA 2012). Their observations were that the effects of wind and water erosion were limited.
There was minimal evidence found of water erosion even on the sloped study plots. This is
consistent with the Hansen, Allen, and Luce calculations of projected minimal sheet and rill
erosion loss described in the response to Interrogatory 191/3 and the evaluation of the potential
for gully erosion on the Federal Cell described in the response to Interrogatory 71/1.
The response to Interrogatory 71/1 describes the sequence of reclamation measures for quickly
re-establishing natural conditions on the cover developed by SWCA (2013). Ecological and
engineering measures described in the response to 171/1 will rapidly stabilize the cover in the
short term and continue to provide long-term stabilization. As SWCA (2013) states, “functioning
native ecosystems comprised of the borrow soils at the Clive site do not show erosion as the
DRC suggests.”
UDEQ expresses their concern about biological soil crust (BSC) at the Site. In particular, they
are concerned with “high uncertainty regarding its characteristics at the Clive site.” UDEQ
requests that EnergySolutions “provide examples with quantitative data from sites in similar
climate and with similar soils. These examples should show how biological soil crust is
preserved or re-established, the timeline for re-establishment, and how presence (or not) of the
biological soil crust affected erosion.”
BSC is common at the Clive Site. An average of nearly 80 percent of the soil surface in the
vicinity of the Clive Site is covered by BSC according to surveys done by SWCA (2013).
EnergySolutions has developed a comprehensive plan (SWCA 2013) for establishment of BSC
on the Federal Cell based on consideration of the literature (Belnap et al. 2001; Bowker 2007).
Plans for recovery of the BSC will not be passive. Active measures will be taken to accelerate
establishment, including topsoil inoculation with BSC organisms (Belnap et al. 2001; Bowker
2007) and stabilization of the soil surface by seeding with fast growing species. SWCA (2013)
describes the inoculation process used to spread BSC organisms collected from undisturbed
native soils on the topsoil (SWCA 2013), and the stabilization of the soil surface with fast
growing species shown to enhance BSC rehabilitation (Bowker 2007). Recovery time for BSC at
the Clive Site is projected by SWCA (2013) to be 3–5 years based on their Site analyses and the
use of active assistance measures.
Lastly, UDEQ requests information from similar sites regarding “how presence (or not) of the
biological soil crust affected erosion.” SWCA (2013) has provided discussion of the literature
describing the beneficial role of BSC in controlling soil erosion.
2.6 Interrogatory CR R313-25-7(2)-59/2: Bathtub Effect
DEQ Conclusion from April 2015 SER, Appendix C:
Until the issues are resolved regarding the design of the cover and infiltration rates (see the DU
PA SER, Section 4.1.1.1 and Appendix B) the potential for bathtubbing effects cannot be ruled
out. Therefore, this interrogatory remains open.
DEQ Critique of DU PA v1.4, Appendix 21: No further analysis has been performed since v1.2.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 64
2.6.1 Interrogatory Response
Net infiltration in the Clive DU PA Model v1.4 is calculated using stochastic inputs. The
implementation in the DU PA Model is described in detail in Appendix 5 of the Final Report for
the Clive DU PA Model, Clive DU PA Model v1.4 (Neptune 2015d). To evaluate the likelihood
of the bathtub effect occurring, a distribution of net infiltration rates for the ET cover was
developed from 1,000 realizations of the net infiltration model. The 99th percentile value of this
distribution was 0.106 mm/yr. The design value for the saturated hydraulic conductivity of the
clay liner below the waste is 1.0 × 10-6 cm/s (316 mm/yr) (Whetstone Associates 2011). At
steady state under unit gradient conditions this hydraulic conductivity corresponds to the flux of
water through the saturated clay liner.
Given the much greater capacity of the clay liner to allow water to flow through it in comparison
to the 99th percentile of net infiltration rates, the bathtub effect is not possible. Any increase in
saturated hydraulic conductivity of the clay liner below the waste due to naturalization will make
the bathtub effect even less likely. The largest net infiltration rate at the base of the cover
modeled using the Benson et al. (2011) approach for in-service cover properties was 1.31 mm/yr.
Even for this unlikely conceptual model of clay liner properties, the maximum net infiltration is
over 240 times smaller than the 316 mm/yr that can flow through the cover with its design value
saturated hydraulic conductivity.
2.7 Interrogatory CR R313-25-7(3)-60/2: Modeled Radon Barriers
DEQ Conclusion from April 2015 SER, Appendix C:
As described under Interrogatory 05, based on several unresolved issues related to the ET
cover, DEQ indicated in the DU PA SER Section 4.1.1.1 that the cover design was deficient and
that it cannot determine the adequacy of this portion of the Clive DU PA. (See the description
under Interrogatory 05 above for specific details.) Therefore, this interrogatory remains open.
DEQ Critique of DU PA v1.4, Appendix 5, Appendix 21: See Interrogatory 21 for discussion
regarding approach and concerns related to modeling the radon barriers.
2.7.1 Interrogatory Response
This interrogatory points to issues raised in Interrogatories 05/2, 21/1, and 90/2. See the
responses to Interrogatories 05/2, 21/1, and 90/2.
2.8 Interrogatory CR R313-25-7(1–2)-90/2: Calibration of Infiltration
Rates
DEQ Conclusion from April 2015 SER, Appendix C:
As noted in Sections 4.1.1.1, 4.1.1.3, and 4.4 of the DU PA SER, several issues (including
infiltration rates) regarding the ET cover remain unresolved. Therefore, this interrogatory remains
open.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 65
DEQ Critique of DU PA v1.4, Appendix 21: No further analysis has been performed on
calibration of infiltration rates since v1.2.
2.8.1 Interrogatory Response
Issues regarding infiltration rates raised in Sections 4.1.1.1, 4.1.1.3, and 4.4 of the DU PA SER
(SC&A 2015b) are:
Sensitivity analysis proposed by UDEQ
See the response to UDEQ Comment B.11.
Correlation between α and Ks parameters
See the response to Interrogatory 05/2.
Development of hydraulic parameter distributions
See the response to UDEQ Comment B.2.
Increased hydraulic conductivity of the clay liner over time is not accounted for in the
model.
See the discussion on the clay liner properties in the response to Interrogatory 05/2.
Only a single value of Ks was used for the Surface and Evaporative Zone layers and a
single value of alpha was used for the radon barriers.
See the response to UDEQ Comment B.2.
Disruption of cover due to erosion, frost, and biointrusion
See the discussion on biointrusion (Effect of Biologicals on Radionuclide Transport) in the
response to Interrogatory 05/2.
See the discussion on calculations of frost depth in the response to Interrogatory 05/2.
See the discussion on cover disruption due to erosion in the response to Interrogatory 05/2.
2.9 Interrogatory CR R313-25-7(2)-150/3: Plant Growth and Cover
Performance
DEQ Conclusion from April 2015 SER, Appendix C:
As discussed in the DU PA SER (Section 4.4.3), concerns remain regarding the potential
impacts of biointrusion on infiltration and this interrogatory is open.
DEQ Critique of DU PA v1.4 and Appendix 21: See responses to Interrogatories 10 and 28 for
further discussion.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 66
2.9.1 Interrogatory Response
In this interrogatory, UDEQ references issues in Section 4.4.3 of the SER (SC&A 2015b). This
section of the SER is titled “Effect of Biological Activity on Radionuclide Transport.” SC&A
(2015b) cite several examples of increased infiltration due to activity of ants and mammals.
These examples, however, are not specific to the Clive Site. SC&A (2015b) criticizes the number
of excavations and the mapping of roots at the Site conducted by SWCA (2013). Practical
considerations, however, dictate that sampling be limited.
Based on site-specific information collected by SWCA (2013), their opinion is that, while
greasewood will eventually become established at the Site, roots will follow available water and
will remain within the Surface, Evaporative Zone, and Frost Protection Layers growing laterally
along the clay barriers. For the flow modeling through the cover layers, a statistical distribution
for the saturated hydraulic conductivity based on the in-service cover properties described by
Benson et al. (2011) was used for the radon barriers. Use of these in-service hydraulic properties
accounts for changes in the hydraulic properties of the radon barriers due to biotic activity.
For additional discussion of these issues, see the responses to Interrogatories 05/2 and 10/3.
2.10 Interrogatory CR R313-25-8(4)(d)-153/2: Impact of Pedogenic
Processes on the Radon Barrier
DEQ Critique, v1.4 and Appendix 21: See responses to interrogatories 10 and 28 for further
discussion. In addition, alterations in the hydraulic properties of cover soils are due primarily to
changes in the size, shape, and connectivity of the pores in response to volume change.
Changes in hydrologic conditions within the cover profile (e.g., wetting or drying, freezing or
thawing) induce changes in pore water potential (aka pore water suction) that cause volume
change. Decreases in pore water due to drying or freezing cause the soil to shrink, resulting in
tensile stresses that form cracks and other macropores. Formation of macropores causes the
saturated hydraulic conductivity and the van Genuchten α parameter to increase. The
“macropores” formed by volume change are larger than the pores in the soil on completion of
construction, but generally are not large cracks that would be visible in a transect or test pit
excavated with a clay spade or similar tool.
Cover soils in more arid regions have a greater propensity for volume change and alterations in
hydraulic properties because very large changes in pore water potential occur seasonally. Plants
in arid regions have the ability to extract water to much higher potentials than plants in humid
regions (Gee et al. 1999), resulting greater volume change and more significant structural
changes. However, over time, cycling of pore water potential combined with the effects of biota
intrusion result in similar alterations in hydraulic properties regardless of climate (Benson et al.
2007, 2011).
2.10.1 Interrogatory Response
See the response to Interrogatory 05/2: Evapotranspiration Cover (ET Cover), which includes a
discussion of the conceptual model of soil formation at Clive based on field data analysis.
In this interrogatory, UDEQ attributes changes in hydraulic properties of a cover material to
volume changes of the material: “alterations in the hydraulic properties of cover soils are due
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 67
primarily to changes in the size, shape, and connectivity of the pores in response to volume
change.”
Changes in volume due to expansion and shrinkage are characteristic of expansive clays. These
clays that develop a blocky structure due to vertical shrinkage cracks formed during dry periods
are described by Handy and Spangler (2007). Moisture cycles form a subangular blocky structure
characterized by aggregates approximately 1 cm in size. Expansive clays are composed of a
significant fraction of smectite group minerals. Volumetric shrinkage strain during drying was
measured by Albrecht and Benson (2001) on eight natural clay soils through cycles of wetting
and drying. These authors found that samples containing the larger percentages of smectite or
mixed illite/smectite had the highest volumetric shrinkage strains and showed the most extensive
cracking. The smallest shrinkage strains were measured for the soils with less smectite and more
illite, kaolinite, and quartz.
The mineralogy of Unit 4 soil clay composition was determined by x-ray diffraction through
analysis conducted by the University of Utah (Bingham Environmental 1996). The results shown
in Table 4 indicate a zero composition of minerals in the smectite group or, as it was formerly
named, montmorillonite. The absence of smectite minerals in the Unit 4 soil makes changes in
soil structure attributed to wetting and drying cycles as suggested by UDEQ unlikely.
Table 4. Minerals in Unit 4 soil clays.
Quartz 12
Plagioclase 2
K-Feldspar 3
Dolomite 4
Calcite 8
Aragonite 53
Kaolinite 2
Illite/Mica 1

2.11 Interrogatory CR R313-25-7(2)-175/1: Infiltration Rates for the
Federal Cell Versus the Class A West Cell
DEQ Conclusion from April 2015 SER, Appendix C:
As DEQ noted in the Round 3 Interrogatories:
ES notes that this interrogatory is no longer relevant since the Federal Cell will use an
ET cover. We agree with this position. However, a thorough discussion of the modeling
of infiltration rates, with soil hydraulic conductivity values as provided in NUREG/CR-
7028 (Benson et al., 2011), is expected in the report on the ET cover system.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 68
The role of hydraulic conductivity on infiltration rates is extensively discussed in the DU PA SER.
See Section 4.1.1.1 and Appendix B. As specifically noted in Section 4.1.1.1:
There are still a number of unresolved issues with respect to the selection of parameter
ranges, distributions, and correlations, as well as the modeling approach and predicted
sensitivities. These concerns are detailed in Appendix B. Further, because the model-
predicted infiltration rates may be sensitive to the hydraulic properties assigned to each
ET layer, the α and Ksat values assumed for modeling moisture in each soil layer within
the cover system must be correlated based on experimental data. Also, additional
justification is required for the soil property values used in the model by
EnergySolutions. Therefore, DEQ does not consider this portion of the performance
assessment resolved.
Therefore, this interrogatory remains open.
DEQ Critique of DU PA v1.4 and Appendix 21: See responses to Interrogatories 10, 21, 28, and
153 for further discussion.
2.11.1 Interrogatory Response
This interrogatory states that there are issues detailed in Appendix B. See the responses to SER
Appendix B, Comments B.1 through B.9 and B.11.
For discussion of the use of correlated α and Ks parameter values for flow modeling, see the
response to Interrogatory 05/2.
For discussion of the use of hydraulic parameter selection and distribution development, see the
response to UDEQ Comment B.2.
2.12 Interrogatory CR R313-25-8(5)(a)-176/1: Representative Hydraulic
Conductivity Rates
DEQ Conclusion from April 2015 SER, Appendix C:
At this time, DEQ does not accept the EnergySolutions position that infiltration results are
insensitive to radon barrier changes. As discussed under Interrogatory CR R313-25-7(2)-05/2:
Radon Barrier, an appropriate modeling analysis needs to be performed with DEQ agreement as
to values of in-service hydraulic conductivity and correlation between Ksat and α (see Appendix
E to the DU PA SER). Until that study is performed and the results analyzed, this interrogatory
remains open. (See also Appendix B to the DU PA SER.)
DEQ Critique of DU PA v1.4 and Appendix 21: See responses to Interrogatories 10, 21, 28, and
153 for further discussion.
2.12.1 Interrogatory Response
The saturated hydraulic conductivity, Ks, was not included in the regression equation for net
infiltration because it was found not to be a predictor (that is, not close to statistical significance).
Ks is, however, included in the regression models for volumetric water content of the radon
barriers.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 69
Saturated hydraulic conductivity is the ratio of water flux to hydraulic gradient under saturated
conditions. The hydraulic conductivity of an unsaturated soil is strongly dependent on the
volumetric water content. As air enters the pores of a saturated soil, the hydraulic conductivity
decreases rapidly. For example, a plot showing the relationship of average hydraulic conductivity
to water content for the Unit 4 silty clay material modeled for the Surface and Evaporative Zone
Layers is shown in Figure 18. This plot uses the mean value of Ks determined from site-specific
core data and hydraulic function properties from the United States Department of Agriculture
(USDA) Rosetta database (USDA 2017).

Figure 18. Relationship between hydraulic conductivity and water content used for Unit 4
material.
For this example, if the water content is slightly reduced from its saturated water content of
0.481 to a water content of 0.4, the hydraulic conductivity is greatly reduced from its saturated
value of 4.46 cm/day to a value of 0.04 cm/day. This reduction of hydraulic conductivity with
reduced water content is even more pronounced with coarser textured soils. In this example, a
small reduction in water content was shown to produce a 100-fold reduction in hydraulic
conductivity. Since most of the flow in the cover layers occurs under unsaturated conditions, it is
not surprising that net infiltration is not sensitive to the value of saturated hydraulic conductivity.
Another reason for the lack of sensitivity of the value of the saturated hydraulic conductivity to
net infiltration is likely the low moisture availability at the Site. Due to the climate there is not
enough available water for saturated conditions to occur extensively. This can be illustrated with
a simple comparison of precipitation volume to soil storage volume. If all of the water from the
entire mean annual precipitation at Clive was instantly infused into the cover, the porosity of the
layers could contain the saturation within the upper 19 inches of the cover, which is to the top
one inch of the Frost Protection Layer.
Changes in the hydraulic conductivity of the radon barrier were accounted for in the flow
modeling. A statistical distribution for the saturated hydraulic conductivity, Ks, of the radon
1.0E-15
1.0E-13
1.0E-11
1.0E-09
1.0E-07
1.0E-05
1.0E-03
1.0E-01
1.0E+01
0.1 0.2 0.3 0.4 0.5Hy
d
r
a
u
l
i
c

C
o
n
d
u
c
t
i
v
i
t
y

(
c
m
/
d
a
y
)
Volumetric Water Content (-)
Hydraulic Conductivity
silty clay
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 70
barriers was developed using values from a range of in-service (“naturalized”) clay barrier Ks
values described by Benson et al. (2011, Section 6.4, p. 6-12). Details of the hydraulic property
distribution development of this distribution are described in the response to Interrogatory 21/2.
Correlation between the α and Ks parameters is discussed in the response to Interrogatory 05/2.
2.13 Interrogatory CR R313-25-7(2)-189/3: Modeling Impacts of
Changes in Federal Cell Cover-System Soil Hydraulic
Conductivity and Alpha Values
DEQ Critique of DU PA v1.4 and Appendix 21: See responses to Interrogatories 10, 21, 28, and
153 for further discussion. In addition, data from other facilities in the region near the Clive site
also confirm that changes in the hydraulic properties of cover soils occur, and the effectiveness
of a cover can change in response to changes in the hydraulic properties.
For example, Benson et al. (2008) report on an assessment of hydraulic properties in the fine-
textured layers in the cover over the uranium mill tailings facility in the Monticello, Utah. The
investigators found that the saturated hydraulic conductivity of the cover soils in the upper 1.5
meters increased by approximately 10x. Similarly, α increased by approximately 5x. Excavation
of caisson lysimeters at the site also showed roots and cracks present in the radon barrier, which
was 1.6–1.9 m bgs (Figure 10-1).
The radon barrier at the Grants, New Mexico, reclamation site was evaluated in the summer of
2016, 20 years after completion, by investigators sponsored by the U.S. Nuclear Regulatory
Commission (NRC) and DOE’s Office of Legacy Management (LM). At this site, the radon barrier
is closer to the surface, with 12 inches of riprap and a sand bedding layer placed directly over the
radon barrier. The capillary break provided by the riprap and the sand bedding layer were
believed to prevent drying and cracking of the radon barrier.
Large block samples were collected from the radon barrier at Grants, New Mexico, for
assessment of field-scale saturated hydraulic conductivity in the laboratory. Block samples were
also collected from an analog site representing conditions anticipated in the long term. A
summary of the hydraulic conductivities reported to date is included in Figure 189-1 below. All of
the saturated hydraulic conductivities are greater than 10-6 cm/s. Most are within or close to the
range described in NUREG/CR-7028 and are approaching the saturated hydraulic conductivity
measured at the analog site. None are less than 1 × 10-7 cm/s as assumed for the lower radon
barrier at Clive.
At the Cheney Disposal Facility near Grand Junction, Colorado, data from two large-scale
lysimeters indicate that the percolation rate from the cover profile has increased substantially
over time, most likely due to structural development within the frost protection layer and the
radon barrier at the site. A summary of the water balance data from these lysimeters is shown in
Table 189-1. This cover employs a rock armor layer, a sand bedding layer, and a frost protection
layer over the radon barrier. Herbicide is used to prevent plant intrusion and root development.
Thus, conditions at this site should minimize the possibility for pedogenesis and alterations in
hydraulic properties. Initially, percolation was on the order of 1 mm/y and less than about 1% of
precipitation. In less than a decade, however, the percolation rate has risen substantially and
was nearly 20% of precipitation in Water Year 2016.
As illustrated in NUREG/CR-7028, changes in hydraulic properties occur at sites more arid and
more humid than Clive. At the hyperarid Apple Valley site in the arid High Plains desert in
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 71
southern California, the saturated hydraulic conductivity of a clay barrier similar to the radon
barrier at Clive increased from 1.5 × 10-8 to 1.2 × 10-5 cm/s, or 800x (Benson et al. 2011).
These examples illustrate that structural changes, alterations in hydraulic properties, and
alterations in the water balance occur at other sites in the region near Clive, Utah, and at more
arid locations. Accordingly, changes in the hydraulic properties should be anticipated in the cover
proposed for the Clive site.
2.13.1 Interrogatory Response
In this interrogatory, UDEQ again states their perception that a correlation between the van
Genuchten α parameter and the saturated hydraulic conductivity of cover materials should be
included in the flow model. The lack of correlation between the α and Ks parameters for the Unit
4 soil is discussed in the response to Interrogatory 05/2.
In addition, UDEQ maintains that statistical distributions of hydraulic properties developed by
Benson et al. (2011) should be used to represent the degradation of cover performance with time.
The basis for this assertion is saturated hydraulic conductivity data and water balance estimates
from other sites provided by UDEQ that in their words “confirm” the general application of a
conceptual model of soil structure formation described in Benson et al. (2011).
UDEQ argues that, since soil structural changes and resulting changes in hydraulic properties
and deep drainage are observed at these sites “in the region near Clive Utah,” that these same
changes should be expected at the Clive Site. The sites listed are the Monticello Mill Tailings
Repository south of the town of Monticello, Utah (290 miles from Clive), the Blue Water
disposal site near Grants, New Mexico (485 miles from Clive), the Cheney disposal site near
Grand Junction Colorado (270 miles from Clive), and the Apple Valley Alternative Cover
Assessment Program (ACAP) site referenced by Benson et al. (2011).
Of these sites, two do not have ET covers. At the Cheney site the upper layer is described as rock
armor treated with herbicide to eliminate vegetation. At the Blue Water site the tailings are
covered by a 1.7 to 2.6 ft radon barrier (DOE 2017). The radon barrier is overlain by a 4 to
12-inch thick layer of riprap. Neither of these sites features layers that would function as an ET
cover.
No information was found on the full facility name, identity of the owner/operator, setting of the
site, construction details, or operation of the site named Apple Valley in Benson et al. (2011).
Searches were unable to find more than a general location for this site near Apple Valley,
California, provided in Benson et al. (2011). The relevance of measurements of this site’s
performance to the Clive Site is uncertain without more information.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 72
SWCA (2013) describe the Monticello site as having similar seasonal precipitation, rainfall
patterns, and vegetation conditions (Waugh et al. 2008) to the Clive Site. However, SWCA
(2013) describe three important differences between the sites:
1. Monticello receives approximately 50% greater average annual precipitation
(15.4 in) than Clive.
2. The Monticello ET cover is comprised of clay-loam to sandy-loam soils that are
less alkaline and more fertile than the saline, alkaline silty-clay soils at Clive
(Waugh et al. 2008).
3. The native vegetation at Monticello is dominated by big sagebrush shrublands
and grasslands that are more diverse and of larger stature—with greater target
plant densities and cover for the ET cover—than those proposed at Clive.
Site-specific observations of soil formation at the Clive Site that differ significantly from those
described in Benson et al. (2011) are discussed in the response to Interrogatory 05/2.
2.14 Interrogatory CR R313-25-7(2)-192/3: Implications of Great Salt
Lake Freezing on Federal Cell Performance
DEQ Critique of DU PA v1.4, Appendix 2: In the Updated Site-Specific Performance Assessment
(EnergySolutions 2013), Appendix E, EnergySolutions presents a calculation of frost depth at the
Clive site based on the modified Berggren equation, which first presented by Berggren (1943),
refined by Aldrich and Paynter in 1953, and later adopted by the U.S. Army Corps of Engineers
and other agencies as their preferred method for frost depth determination (Departments of the
Army and Airforce, 1988).
In their July 8, 2014 (EnergySolutions 2014), response to this interrogatory, EnergySolutions
points to Appendix E to the Updated Site-Specific Performance Assessment (EnergySolutions
2013) for the calculation of the potential frost depth; however, that reference (nor any other
estimation of frost depth) is not provided in v1.4, Appendix 2.
Therefore, this interrogatory will remain open until an estimate of the potential frost depth has
been incorporated into DU PA Appendix 2, either by reference to or reproducing EnergySolutions
2013, Appendix E, or by providing a similar calculation of the potential frost depth. Additionally, if
EnergySolutions 2013, Appendix E, is referenced or reproduced, any open interrogatories
against Appendix E must be resolved before it is incorporated into DU PA Appendix 2.
DEQ Critique of DU PA v1.4, Appendix 21: An estimate of the potential frost depth has not been
provided in Appendix 21.
2.14.1 Interrogatory Response
See the response to Interrogatory 05/2 for a description of frost depth calculations for the Clive
Site.

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 73
2.15 SER B.1 Supplemental Interrogatory Comment 1
1) Demonstrate why 20 HYDRUS runs are sufficient to capture the parameter uncertainty.

DEQ Critique from April 2105 SER, Appendix B:
EnergySolutions’ response provided to this comment did not address the comment satisfactorily.
DEQ understands that the regressions [Equations 39 and 40 of Appendix 5 to the depleted
uranium performance assessment (DU PA) (Neptune 2014b)] were created as simplified
surrogate models that relate percolation from the base of the cover and water content in each
layer of the cover profile to hydraulic properties of the cover soils. This regression model was
developed based on output from HYDRUS from 20 sets of input parameters.
Because only 20 cases were used for the simulations, the tails of the distributions describing the
hydraulic properties are poorly sampled, and more extreme cases may be inadequately
represented. Consequently, the regressions may represent average or mean conditions
sufficiently but may not adequately represent the more extreme cases. No information has been
provided to demonstrate that the extreme cases in the tails of the distributions are adequately
represented by the regression, or that 20 cases are sufficient to capture the effects of the tails of
the distributions. For heavy-tailed distributions such as those used for hydraulic properties, many
more simulations would be needed to adequately represent events driven by properties
associated with the tails of the distributions.
The predictions in EnergySolutions (2014) Figure 5 (see the discussion on Comment 7 below)
suggest that the process of developing the regression model has resulted in predictions that are
centered more around the mean behavior and that are insensitive to the tails. The percolation
predicted from the regression varies within a narrow range of around 0.3 millimeters per year
(mm/yr), whereas percolation predicted by HYDRUS predictions for all realizations ranges from
approximately 0.01 mm/yr to 10 mm/yr. The response suggests that this insensitive behavior is
due to the variance reduction in the hydraulic properties to account for spatial averaging, but
another plausible reason is that the regression is based on mostly mean behavior and is
relatively insensitive to extremes represented by the hydraulic properties in the tails of the
distributions.
A well-documented justification is needed that demonstrates that Equations 39 and 40, based on
predictions from 20 simulations using 20 sets of randomly sampled properties, adequately
predict the percolation rate and the water contents for cases near the mean and more extreme
cases in the tails of the distributions.
In addition, the analysis fails to adequately account for (1) correlations between parameters α
and Ksat in the same soil layer, and (2) correlations between the values of each parameter within
different soil layers. These deficiencies need to be resolved. DEQ also notes that the
EnergySolutions response contains no substantive discussion of how and why scaling was
conducted and how it impacts the results. This discussion must be provided.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
HYDRUS model parameter uncertainty.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 74
2.15.1 Interrogatory Response
In response to supplemental interrogatories submitted to EnergySolutions on August 11, 2014
(EnergySolutions 2014), additional capability was developed to allow for more HYDRUS
simulations of the cover system to be completed within a practical time period. Fifty HYDRUS-
1D simulations were conducted to evaluate the uncertainty in infiltration flux into the waste zone
and water content within each ET cover layer as a function of hydraulic property uncertainty;
these simulations were documented in DU PA v1.4 Appendix 5 (Neptune 2015c). Equation
numbers 39 and 40 from Appendix 5 of DU PA v1.2 (Neptune 2015c) correspond to equation
numbers 41 and 42 from Appendix 5 of DU PA v1.4 (Neptune 2015c).
The development of hydraulic property input statistical distributions for this modeling is
described in the response to UDEQ Comment B.2. The values of α, n, and Ks used as inputs for
each of the 50 simulations are listed in Table 9 of DU PA v1.4 Appendix 5 (Neptune 2015c) and
are reproduced here in Table 6.
Simulations were run for 1,000 years. The mean of the fluxes into the top of the waste layer and
the mean water contents for the Surface Layer, Evaporative Zone Layer, Frost Protection Layer,
and Upper and Lower Radon Barriers over years 900 to 1000 were calculated.
The 50 HYDRUS-1D simulations resulted in a distribution of average annual infiltration into the
waste zone, and average volumetric water contents for each ET cover layer. Infiltration flux into
the waste zone ranged from 0.0067 to 0.18 mm/yr, with an average of 0.024 mm/yr, and a log
mean of 0.018 mm/yr for the 50 replicates.
Multiple linear regression models were fit to the HYDRUS infiltration results, and water
contents for each ET cover layer. The general form of the regression was:
𝑌=βV +β<∗𝐾)+β>∗α +βX ∗𝑛
Net infiltration is in units of mm/yr and volumetric water content is dimensionless. For the net
infiltration flux regressions, Ks was dropped as a predictor due to poor fit of the models. The
regressions were fit using the ‘lm()’ function in the software package R (R Core Team 2015),
which uses least squares optimization for estimating parameters. All values of β coefficients are
summarized in Table 5.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 75
Table 5. Coefficients calculated from multiple linear regression models.
Coefficient βo β1 β2 β3
SurfaceWC 0.48155 0.00000 0.54920 -0.20020
EvapWC 0.57947 0.00000 0.73997 -0.24790
FrostWC 0.04282 0.00000 0.43297 0.01617
Rn1WC 0.14737 -0.00076 1.70702 0.06353
Rn2WC 0.14740 -0.00076 1.70648 0.06351
Flux (mm/yr) -0.32921 N/A 5.56826 0.19538

Implementation in GoldSim
Average annual infiltration flux into the waste zone, and the volumetric water content of each ET
cover layer, were calculated using Equations 41 and 42, developed from HYDRUS-1D
simulation results (Neptune 2015c). GoldSim calculates values using Equations 41 and 42 from
Appendix 5 of DU PA v1.4 (Neptune 2015c) for each ET cover layer.
These equations for solving infiltration and water content in GoldSim are:
𝐼𝑛𝑓𝑖𝑙=βV +β>∗α +βX ∗𝑛
𝑊𝐶=𝛽_,V +𝛽_,<∗𝐾)+𝛽_,>∗𝛼+𝛽_,X ∗𝑛
where Infil is net infiltration in mm/yr, WC is average volumetric water content, and β values are
linear regression coefficients with the subscript i corresponding to Surface, Evaporative Zone,
Frost Protection, Upper Radon Barrier, and Lower Radon Barrier Layers. The necessary
distributions in GoldSim are VG_logAlpha, VG_logN, and RnBarrierKsat_Natdist.
α and n are calculated from values drawn from distributions using:
𝛼=10VG_logAlpha,𝑤ℎ𝑒𝑟𝑒 VG_logAlpha ~ 𝑁𝑜𝑟𝑚𝑎𝑙(𝑚𝑒𝑎𝑛: −1.79,𝑠𝑒: 0.121)
𝑛=10VGlogN,𝑤ℎ𝑒𝑟𝑒 VGlogN~ 𝑁𝑜𝑟𝑚𝑎𝑙(𝑚𝑒𝑎𝑛: 0.121,𝑠𝑒: 0.019).
Ks is sampled using:
RnBarrierKsat_Natdist =𝐾7,~𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙(𝑔𝑒𝑜𝑚.𝑚𝑒𝑎𝑛:3.37 𝑐𝑚/
𝑑𝑎𝑦,𝑔𝑒𝑜𝑚.𝑠𝑑: 3.23 𝑐𝑚/𝑑𝑎𝑦),
See the response to Interrogatory 05/2 regarding correlation of the van Genuchten α parameter
and saturated hydraulic conductivity.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 76
Figure 19 shows a comparison of the net infiltration results calculated using HYDRUS and using
the regression equation in the Clive DU PA Model v1.4. Clearly, the comparison shows an
excellent fit to the HYDRUS results, demonstrating that the use of a regression equation to
approximate the HYDRUS simulations resulted in a successful model abstraction in this case.

Figure 19. Comparison of 1,000 realizations of net infiltration using the linear model in
GoldSim with the results of the 50 HYDRUS simulations of infiltration.
As described in EnergySolutions (2014), the discrepancy between the net infiltration rates
obtained from the 20 HYDRUS simulations and the GoldSim estimated net infiltration using the
regression model was due to not scaling the α and n distributions used as inputs for the
HYDRUS simulations. This discrepancy has been resolved for the 50 HYDRUS simulations
described in Appendix 5 of the DU PA Model v1.4 (Neptune 2015c), and the comparison
between the HYDRUS results and the 1,000 realizations using the regression equation are shown
here in Figure 19.
The 50 cases were run at sets of values chosen to explore the part of the input parameter space
used for the DU PA Model, thus putting the effort and resources into the portion of the parameter
space most important to support the model. The values chosen are directly from the distributions
developed for Clive DU PA Model v1.4, and the use of values is adequate to expect to sample
from the tails of the distributions. The model abstraction and connections to input parameter
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 77
distributions were described in detail earlier in this response, as well as the reasons for
developing distributions on a scale appropriate for the PA Model.
UDEQ concludes this interrogatory stating that the “EnergySolutions response contains no
substantive discussion of how and why scaling was conducted and how it impacts the results.”
An explanation regarding how and why scaling was conducted in multiple sections of the
responses. In addition, Section 1.1 of this document provides a high-level discussion, with
examples.
In summary, the reason for “scaling” is to take the available information about the input
parameters and use it to develop a distribution at a scale consistent with the scale represented by
the PA Model. In other words, the random draws from the distributions should be values that are
consistent with the scales they are supposed to represent (e.g., an extreme value possible from a
point location at one point in time should not be used to represent a large spatial volume over a
long period of time).
Available data often represent much smaller spatial volumes and time periods than are needed
for the PA Model; therefore, the data should not be used directly to develop a distribution, but
instead “scaling” should be employed to attempt to build a distribution representing the PA
Model scale. Averaging is one simple method of scaling information up by aggregating
information available at smaller scales.
Larger scales are typically associated with less variability, and thus narrower distributions, but
the statistical method used for scaling should depend on the properties of the available data and
on uncertainty in assumptions needed to justify the scaling. The end goal is distributions that
produce draws representing reasonably realistic values to apply to the spatial volumes
represented in the PA Model and long time periods over which they are held constant within a
run of the PA Model.
2.16 SER B.2 Supplemental Interrogatory Comment 2
2) The Table 9 HYDRUS parameters do not appear to “bound” the α, n, and Ksat
distributions. For example, in the distribution, Ksat ranges from 0.0043 to 52 cm/day,
but in the 20 HYDRUS runs Ksat only ranged from 0.16 to 10.2 cm/day.
DEQ Critique from April 2015 SER, Appendix B:
EnergySolutions’ response indicates that the input “values are considered sufficiently extreme to
evaluate the influence of Ks on the HYDRUS model outputs, and hence to determine the
influence of Ks on the water content and infiltration model outputs.” The basis for the conclusion
“considered sufficiently extreme” needs to be demonstrated rather than stipulated.
As cited in the response to Comment 1 (above), a well-documented justification is needed that
demonstrates that Equations 39 and 40, based on predictions from 20 simulations using 20 sets
of randomly sampled properties, adequately predict the percolation rate and the water contents
for cases near the mean and more extreme cases in the tails of the distributions. This
demonstration should also provide a physical basis for excluding some of the variability in key
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 78
hydraulic properties normally considered to affect percolation strongly, such as Ksat in the
shallow cover-system layers (i.e., the Surface Layer and the Evaporative Zone Layer). Any
exclusion of this parameter or its full range of variability from other aspects of modeling,
correlation, or sensitivity analysis should also be justified. Although the Clive DU PA v1.2
appears superficially to have illustrated that the output was not sensitive to Ksat, this conclusion
may be the result of predictions from a cover hydrology model for which unrealistic parameters
were used as input (e.g., changing some parameter values but not others for a given soil layer).
A separate quantitative demonstration is needed showing that Equations 39 and 40, based on
the 20 sets of hydraulic properties used as input, are representative.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
HYDRUS model input distribution, ranges and bounds.
2.16.1 Interrogatory Response
In response to supplemental interrogatories submitted to EnergySolutions on August 11, 2014,
additional capability was developed to allow for more HYDRUS simulations of the cover system
to be completed within a practical time period. Fifty HYDRUS-1D simulations were completed
to evaluate the uncertainty in infiltration flux into the waste zone and water content within each
ET cover layer as a function of hydraulic property uncertainty and were documented in DU PA
v1.4 Appendix 5 (Neptune 2015c). See the response to UDEQ Comment B.1.
HYDRUS Hydraulic Property Input Parameters
Parameter values and parameter statistical distributions for the hydraulic properties of the cover
layers shown in Figure 1 were developed based on site-specific data, engineering specifications,
widely used soil hydraulic property databases, and consideration of the function of the layer.
These are the parameters used for the hydraulic conductivity model described in Equations (1),
(2), and (3) in Section 1.3 needed for the HYDRUS flow modeling.
Surface Layer and Evaporative Zone Layer
These two layers are composed of the same Unit 4 silty clay. The source of site-specific material
properties for Unit 4 is Bingham Environmental (1991) (pp. B-20 and B-26). They report results
from measurement of water retention for two of nine cores sampled from Unit 4 at the Site. The
water retention relation is the correspondence between the tension of water held in the pores of
the material and the water content of the material. These data are used to estimate the parameters
of the van Genuchten-Mualem hydraulic conductivity model.
The functions of these layers, beyond supporting vegetation and providing erosion control, are to
store water from precipitation events within the layers and slow deeper drainage so that the water
is available for release to the atmosphere through evaporation and plant transpiration. The α and
n values of the van Genuchten water retention model (Section 1.3) influence the capacity of the
Unit 4 material used for the Surface and Evaporative Zone Layers to hold water. To provide a
better estimate of the uncertainty of these Unit 4 properties, statistical distributions were
developed for α and n to be used for inputs for the flow modeling.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 79
To develop the distributions for flow model input, α and n values were taken from the
distributions of the mean and the standard deviation for each parameter from the Rosetta model
database of hydraulic parameters for the textural class of silty clay (USDA 2017). The standard
deviations were converted to standard errors by dividing by √n, where n is the number of
samples (28 in this case).
The Rosetta model provided estimates of van Genuchten water retention parameters, saturated
hydraulic conductivity, and unsaturated hydraulic conductivity parameters. The Rosetta model is
widely used and has been successful in many applications. In addition to other predictive
features, the Rosetta model provides a database containing the class average values of soil
hydraulic function parameters for the 12 USDA soil textural classifications. These average
values are based on 2,134 soil samples for water retention and 1,306 soil samples for saturated
hydraulic conductivity (USDA 2017). These data were obtained from the RAWLS, AHUJA, and
UNSODA databases (USDA 2017).
To provide consistency with the conceptual model, probability distributions needed to be
specified that matched the spatio-temporal scale of the model. The fitted regression models were
used in the Clive DU PA GoldSim model, but the distributions of α and n were re-scaled to
match the structure of the GoldSim model. Scaling in this way is inherently an averaging
process, although some care needs to be taken to ensure that the immediate response reacts
linearly to the inputs (expectation is a linear operator). The Rosetta database indicates that 28
samples were used to develop the mean and standard deviation estimates. Consequently, scaling
was performed by dividing the standard deviation by the square root of 28, which represents
using the standard error of the Rosetta data for the parameter distributions implemented in
GoldSim. This provides an appropriate distribution for the Clive DU PA GoldSim model given
the structure and scale of that model.
The distributions for A and N given in terms of log transforms of α and n were described in the
response to UDEQ Comment B.1. Normal distributions of A and N were sampled 50 times, and
then transposed from log space by calculating 10A and 10N to provide 50 sampled values of α
and n. In addition, N was truncated such that it could not be less than or equal to 0.0. A
correlation between Ks and α was not used. (See the response to Interrogatory 05/2.)
To manage computational burden, the saturated hydraulic conductivity, Ks, for the Surface Layer
and Evaporative Zone Layer was assigned a single deterministic value of 4.46 cm/day based on
the mean value determined from analysis of the Unit 4 cores (Equation 29 of Neptune (2015c)).
See the response to Comment B.5.
Radon Barriers
An expanded assessment of the performance of the radon barriers was made possible by
developing a distribution for the saturated hydraulic conductivity (Ks) of the radon barriers to use
for the modeling. Development of this distribution is described in the response to Interrogatory
21/2.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 80
For all HYDRUS simulations, the same Ks value was applied to both the Upper and Lower
Radon Barriers. To manage computational burden, deterministic values for θr, θs, α, and n from
Table 17 of Whetstone Associates (2011) were used for the radon barriers and are listed in Table
8 of Neptune (2015c). A correlation between Ks and α was not used (see the response to
Interrogatory 05/2).
The equivalent of Table 9 of Appendix 5 of DU PA Model v1.2 (Neptune 2014) cited in the
interrogatory is reproduced here from Appendix 5 of DU PA Model v1.4 (Neptune 2015c). This
table, Table 6 below, contains the values of α, n, and Ks drawn from the distributions described
above and used in 50 HYDRUS simulations of net infiltration and volumetric water content.
Table 6. Parameter sets of van Genuchten α, n, and Ks used for HYDRUS modeling.
α (1/cm)
1 0.013091 1.359766 3.285794
2 0.014317 1.371086 12.497148
3 0.010969 1.357776 3.736272
4 0.018089 1.342287 5.162964
5 0.019954 1.316356 2.325706
6 0.010797 1.279182 4.168751
7 0.016004 1.396199 2.595876
8 0.012816 1.308572 0.838501
9 0.014744 1.372326 2.055096
10 0.014791 1.360367 5.052781
11 0.020639 1.276159 3.234858
12 0.019501 1.327968 2.194697
13 0.015766 1.334194 1.307280
14 0.019048 1.373538 1.719640
15 0.018539 1.338996 1.635838
16 0.017045 1.267606 1.749758
17 0.019983 1.413655 5.126214
18 0.012494 1.326223 10.753272
19 0.019503 1.356646 1.845171
20 0.028186 1.378016 3.643845
21 0.010929 1.244500 6.738214
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 81
Realization α (1/cm) n Ks (cm/d)
22 0.020973 1.282170 6.943533
23 0.017971 1.372107 1.099495
24 0.016549 1.467656 3.648668
25 0.012120 1.330512 6.338780
26 0.011984 1.382991 0.792890
27 0.012782 1.382761 7.005276
28 0.017094 1.275082 4.768674
29 0.013032 1.382671 9.861743
30 0.024165 1.349583 7.758327
31 0.016054 1.386282 1.478986
32 0.024889 1.310637 2.501489
33 0.017247 1.320670 2.459523
34 0.014338 1.265236 66.503659
35 0.016633 1.286526 31.683457
36 0.014343 1.383885 1.005712
37 0.022207 1.236303 3.733521
38 0.012511 1.317326 4.565641
39 0.018395 1.333180 6.167757
40 0.013735 1.294514 2.206236
41 0.015243 1.229113 4.106400
42 0.018063 1.282922 3.299065
43 0.017010 1.326811 32.484809
44 0.020072 1.323515 31.128008
45 0.015950 1.357247 2.326748
46 0.018944 1.252554 2.976567
47 0.015677 1.301147 1.241111
48 0.024293 1.287802 4.617869
49 0.018819 1.264178 0.737824
50 0.017781 1.263628 2.880623

ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 82
The 50 random draws from the distribution for Ks have a minimum of 0.73 cm/day, a maximum
of 66.5 cm/day, and a mean value of 6.76 cm/day. The maximum of 66.5 cm/day is greater than
the maximum of the range reported by Benson et al. (2011) of 52 cm/day (6.0 × 10-6 m/s). Also
note that the lowest Ks value drawn was 0.738 cm/day, a value more than 100 times larger than
the minimum of the distribution.
The sensitivity of the modeled net infiltration rate to the value of the saturated hydraulic
conductivity is discussed in the response to Interrogatory 176/1.
2.17 SER B.3 Supplemental Interrogatory Comment 3
3) NUREG/CR-7028 (Benson et al. 2011) gives the “in-service hydraulic conductivity” as
ranging from 7.5 × 10-8 to 6.0 × 10-6 m/s [0.7 to 52 cm/day], with a mean of
4.4 × 10-7 m/s [3.8 cm/day]. Instead of using the provided distribution (i.e., log-triangular
with a minimum, maximum, and most likely), ES/Neptune constructed a lognormal
distribution with a mean and standard deviation of 0.691 and 6.396 cm/day,
respectively. Provide the justification for this approach. For example, the selection of
0.0043 cm/day as the lower end of the Ksat distribution requires justification (Appendix
5, p.41). It is not clear why a design parameter value should be used when adequate
field data are available. The number chosen by the Licensee for the lower end of the
distribution range in the GoldSim implementation is 163 times lower than the lowest
value in the range specified within the NUREG guidance (see Section 13.0 of Appendix
5, Unsaturated Zone Modeling to the Clive DU PA). We believe that use of the design
parameter biases the Ksat distribution in a non-conservative manner.
DEQ Critique from April 2015 SER, Appendix B:
EnergySolutions’ response to Comment 3 has not demonstrated that the distribution of Ksat
used for the HYDRUS modeling adequately represents the range of conditions that might be
realized for a “naturalized” cover, i.e., one that has undergone pedogenesis as described in
NUREG/CR-7028 (Benson et al. 2011). To account for the higher Ksat in NUREG/CR-7028
(Benson et al. 2011), the lognormal distribution for Ksat was re-fit by the Licensee using an
abnormally large log(s) of 6.396. This provides an unrealistic distribution of Ksat that
substantially overweights Ksat in the lower range.
This, in turn, has the general effect of artificially increasing apparent capillary barrier effects in
the DU PA Model v1.2, i.e., at the interface between a relatively lower-permeability zone (the
combined Surface Layer and the Evaporative Zone Layer, having a mean Ksat value in the DU
PA Model v1.2 of 4.46 cm/day) and a relatively higher-permeability zone (the Frost Protection
Layer, having a mean Ksat value in the DU PA Model v1.2 of 106.1 cm/day). When
EnergySolutions assumes in HYDRUS that the Ksat value for the lower-permeability zone can be
as small as 0.0042 cm/day, the ratio in hydraulic conductivity between the higher-permeability
zone and the lower-permeability zone can thus be as large as 25,000. This creates in the model
an extremely potent artificial, non-realistic capillary barrier at the Evaporative Zone Layer/Frost
Protection Layer interface that, in an unrealistic way, reduces infiltration below that interface to
extremely small or even negligible values.
The primary model hydraulic conductivity value for the higher-permeability zone in the DU PA
Model v1.2, 106.1 cm/day, may already be unrealistic, since the assemblage of soil particles in
the Frost Protection Layer is proposed to be a random, poorly-sorted mixture of grain sizes, with
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 83
smaller grains being as small as clay. The Frost Protection Layer is not characterized in terms of
actual grain size distribution in the DU PA Model v1.2, other than to say that particle sizes can
range from 16-inch diameter to clay size. The hydraulic conductivity assigned to it is arbitrary.
The assigned value is representative of a sandy loam, which is a very poor representation of the
proposed Frost Protection Layer. A mixture of poorly-sorted grain sizes, as found in the Frost
Protection Layer, tends to greatly diminish the hydraulic conductivity of a soil compared to a
relatively well-sorted mixture. Further exacerbating the problem in the DU PA Model v1.2 is that
the hydraulic conductivity values assumed in HYDRUS for the lower-permeability zone are
additionally allowed to be 163 times lower than the lowest specified value in the NUREG range
for in-service hydraulic conductivity (Benson et al. 2011).
The rationale for dramatically increasing log(s) to account for the higher Ksat associated with
pedogenesis or “naturalization” has not been provided and is counterintuitive. The log(s) should
at least be similar for as-built and naturalized covers and may, in fact, be lower for naturalized
covers because pedogenic processes ameliorate hydraulic anomalies inherent in the cover from
construction. NUREG/CR-7028 (Benson et al. 2011) indicates that pedogenesis tends to
transform in-service hydraulic conductivity values to as-built values found in a much higher, but a
more restricted, range. The mean should shift upward during naturalization as structure
develops, reflecting overall increase in Ksat and α rather than a broader range.
As noted previously, while the Clive DU PA Model v1.2 may have illustrated that the output was
not sensitive to Ksat, this conclusion may be the result of predictions from a cover hydrology
model for which unrealistic parameters were used as input. Insensitivity of infiltration to hydraulic
conductivity would be expected if inappropriate input parameter values are used so as to create
in the model an unjustified, artificial capillary barrier effect. Normally, in the absence of a capillary
barrier, infiltration is very sensitive to hydraulic conductivity. As stated by Alvarez-Acosta et al.
(2012):
A soil hydraulic property that is often a required input to simulation models is the
saturated hydraulic conductivity, Ks. It is one of the most important soil physical
properties for determining infiltration rate and other hydrological processes…. In
hydrologic models, this is a sensitive input parameter and is one of the most problematic
measurements at field-scale in regard to variability and uncertainty.
Thus, the insensitivity of deep infiltration to Ksat reported in the Clive DU PA is not sufficient to
dismiss the need for demonstrating the efficacy of the parameters used for the HYDRUS input in
Appendix 5 to the DU PA Model v1.2.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
EnergySolutions assumptions regarding the in-service versus naturalized parameters.
2.17.1 Interrogatory Response
In this interrogatory UDEQ questions the validity of the lognormal distribution fit by Neptune to
the in-service barrier saturated hydraulic conductivity measurements from Benson et al. (2011).
See the response to Interrogatory 21/2 for a description of the methods used for fitting the
distribution for DU PA Model v1.4 (Neptune 2015c).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 84
UDEQ states “Instead of using the provided distribution (i.e., log-triangular with a minimum,
maximum, and most likely), ES/Neptune constructed a lognormal distribution with a mean and
standard deviation of 0.691 and 6.396 cm/day, respectively.”
Saturated hydraulic conductivity, Ks, is generally modeled with a lognormal distribution, as
demonstrated through the figures and the reporting of the geometric mean, rather than the
arithmetic mean, for this parameter in Benson et al. (2011). Therefore, developing a lognormal
distribution for this parameter is consistent with previous work. Benson et al. (2011) provide a
median and range of likely Ks values, but do not explicitly state that the distribution should be
log-triangular. The information provided has been used for the median and the likely range, but
fit to a lognormal distribution. This assumes that the ends of the “likely range” should be close to
the 1st and 99th percentiles of the distribution, rather than a strict minimum and maximum that
would be assumed using the suggested log-triangular distribution.
The lognormal distribution has many advantages over the log-triangular distribution. More mass
of the distribution is concentrated close to the median; for the log-triangular distribution, more
mass is located in the tails where values should be less likely. In practice, this means that the
lognormal distribution will lead to draws that are closer to the median, while draws from the log-
triangular distribution would lead to more draws that are at the extreme edges of the expected Ks
values. The lognormal distribution does not exclude plausible Ks values.
Although Benson et al. (2011) provide a range of likely Ks values, they never state that it would
be impossible for Ks to be smaller or greater than the ends of the reported range. In fact,
throughout the document, the authors provide examples of other Ks values that are sometimes
beyond the given range, i.e., the in-service Ks for Cedar Rapids is 0.06 cm/day. The lognormal
distribution does not completely exclude values outside the range deemed plausible, while a log-
triangular distribution completely excludes them. This is true for the low end of the distribution,
as well as the high end, where the lognormal distribution has a long right tail, allowing for
approximately 1% of Ks values to be greater than the upper end of the range specified by Benson
et al. (2011). Finally, as stated previously, Benson et al. (2011) suggest a value for the median of
the distribution. It is often assumed the mode of a triangular distribution is equal to the median,
but this is not the case. The log of the median can be used as the mode on the log scale, but then
the median on the original scale will be close to 6, instead of the goal of 3.8.
UDEQ also has concerns with the use of 0.00432 as a minimum Ks value. The lognormal
distribution was not fit with the value of 0.0043 but this value was used to truncate the
distribution after fitting so that lower values could not be drawn at random. The lower bound
excludes Ks values below the design specification, and it barely affects the distribution. Without
the lower bound and distribution shift, the probability of a draw being at or below the boundary
of 0.0043 cm/day is less than 10-8. Setting a lower bound does not mean that the distribution
approaches that value, but it is a precaution to ensure it never goes below it, which is the use of
the lower bound here.
UDEQ argues that “EnergySolutions’ response to Comment 3 has not demonstrated that the
distribution of Ksat used for the HYDRUS modeling adequately represents the range of
conditions that might be realized for a “naturalized” cover, i.e., one that has undergone
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 85
pedogenesis as described in NUREG/CR-7028 (Benson et al. 2011).” In general, it appears that
the distribution developed for Ks is not entirely understood. UDEQ claims that the work
“provides an unrealistic distribution of Ks,” but the distribution stems directly from the Benson et
al. (2011) field data that they would like used. The lognormal distribution has a similar median
value, 3.37 cm/day, and a similar range of values for the 1st to 99th percentile, 0.22 cm/day to
51.5 cm/day, compared to their range of 0.65 to 52 cm/day.
UDEQ then states “To account for the higher Ksat in NUREG/CR-7028 (Benson et al. 2011), the
lognormal distribution for Ksat was re-fit by the Licensee using an abnormally large log(s) of
6.396. This provides an unrealistic distribution of Ksat that substantially overweights Ksat in the
lower range.” See the response to Interrogatory 21/2 for a description of the lognormal
distribution used for Ks in Appendix 5 of the Clive DU PA Model v1.4 (Neptune 2015c)
developed based on the information in NUREG/CR-7028 (Benson et al. 2011). The distribution
has a geometric mean of 3.37 and a geometric standard deviation of 3.23 (log(3.23) = 1.17).
Therefore, the lognormal distribution used for the HYDRUS modeling to obtain the 50
realizations used for the model abstraction associated with the DU PA Model v1.4 has a log(s)
of 1.17, not 6.396. The 6.396 was used for the 20 HYDRUS realizations used for the model
abstraction in Appendix 5 of the DU PA Model v1.2 (Neptune 2014). The distribution was
revised when capability was developed to run many more HYDRUS simulations. See the
response to UDEQ Comment B.2.
UDEQ then raises the issue of a capillary barrier at the interface between the bottom of the
Evaporative Zone Layer and the top of the Frost Protection Layer. For the Frost Protection
Layer, hydraulic properties for a sandy loam were used as taken from the HYDRUS-1D pull-
down menu, which includes properties from the database of Carsel and Parrish (1988). The
parameters for a sandy loam were chosen because this texture represents a mixture of particle
sizes consisting of sand, silt, and clay in fractions representing the range seen in the Unit 3
material at the Site. This soil component fills the voids between the cobbles in the Frost
Protection Layer and determines the water flow properties of this layer.
Single deterministic values were used for the hydraulic properties θr, θs, α, n, and Ks. The
saturated hydraulic conductivity for the Frost Protection Layer was assigned a single,
deterministic value of 106.1 cm/day based on the sandy loam textural class average from the
Carsel and Parrish database. Statistical distributions were not developed for the hydraulic
properties of this layer since the function of this layer is protection against freezing and
biointrusion, and modeling uncertainty in the hydraulic properties of this layer would add
significant computational burden to the modeling.
UDEQ states that the properties of this layer contribute to the layer behaving as a capillary
barrier. A capillary barrier is a two-layered system consisting of a fine-grained layer overlying a
coarse-grained layer. In their comments UDEQ is referring to the Surface and Evaporative Zone
Layers as the upper layer and the Frost Protection Layer as the lower layer. They argue that the
hydraulic properties of these layers lead to “an extremely potent artificial, non-realistic capillary
barrier” at the interface between the Evaporative Zone Layer and the Frost Protection Layer.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 86
The capillary barrier phenomenon occurs due to differences in the relationship between the
hydraulic conductivity and the water tension of the two materials. Water moving in the
unsaturated zone is at a negative pressure that can be referred to as tension. For water applied to
the top of the fine layer in an initially dry profile at an initially large tension the fine-grained
material will have a hydraulic conductivity larger than the coarse-grained material. Due to this
difference, water can move down through the fine-grained layer but cannot enter the coarse-
grained layer immediately. As more water moves into the fine-grained layer the water content in
this layer increases and the corresponding water tension decreases. Water cannot enter the air-
filled pores of the coarse-grained layer until the tension decreases to the water entry value of the
coarse-grained layer. When this tension is reached, water infiltrates into the coarse-grained layer.
This phenomenon is commonly used in cover design to temporarily hold water in the upper
layers by capillarity until it is released to the atmosphere through evaporation and transpiration
(Morris and Stormont 1999). This is the same concept as the “store-and-release” cover described
by Benson et al. (2011): “The storage layer resides below the topsoil (approx. 300 mm thick) and
above a capillary break or foundation layer (if present).”
The hydraulic property values and statistical distributions assigned to the Evaporative Zone and
the Frost Protection Layers are reasonable estimates based on site-specific information and
commonly used soils databases. Given these property assignments, the Frost Protection Layer
behaves hydraulically by enhancing storage in the Evaporative Zone in a realistic manner, not as
“an extremely potent artificial, non-realistic capillary barrier.”
UDEQ states further in Comment B.3 with respect to capillary barriers, “When EnergySolutions
assumes in HYDRUS that the Ksat value for the lower-permeability zone can be as small as
0.0042 cm/day, the ratio in hydraulic conductivity between the higher-permeability zone and the
lower-permeability zone can thus be as large as 25,000.” The basis of this comment is not clear.
The saturated hydraulic conductivity of the Frost Protection Layer in the flow model is
106.1 cm/day, while the saturated hydraulic conductivity of the Surface and Evaporative Zone
layers is 4.46 cm/day. This is a ratio of 24 to 1 not 25,000 to 1.
UDEQ comments on the analysis of flow modeling results that the magnitude of net infiltration
is not sensitive to the magnitude of the saturated hydraulic conductivity of the radon barriers.
There are a number of factors that contribute to this. The sensitivity of the modeled net
infiltration rate to the value of the saturated hydraulic conductivity is discussed in the response to
Interrogatory 176/1.
2.18 SER B.4 Supplemental Interrogatory Comment 4
4) Provide justification for using the Rosetta database, as appropriate for an
engineering earthen cover.
DEQ Critique from April 2015 SER, Appendix B:
This interrogatory asked for justification for using the Rosetta database for an engineered
earthen cover. The response goes to great length comparing the attributes of the Rosetta
database to other databases, none of which are populated with data for engineered earthen
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 87
covers. Most of the databases are for agricultural soils, many of which have been tilled. Their
relevance to an engineered earthen cover has not been demonstrated. The response has
shown, however, that many of the mean values of hydraulic properties used as input are, to
some extent, in reasonable agreement with those associated with engineered earthen covers, as
described in NUREG/CR-7028 (Benson et al. 2011). On the other hand, as discussed in the
Supplemental Interrogatory Comment 3 (see Section B.3), the low-end value in the range of
hydraulic conductivity used in the GoldSim model is 163 times lower than the lowest specified
value in NUREG/CR-7028 for in-service hydraulic conductivity. The low-permeability tail of the
distribution is overweighted, and variability is not properly accounted for.
One response to the interrogatory, if it could be substantiated using data, would be that the
Rosetta database is not based on engineered earthen cover soils and should not be assumed to
be representative, but point-wise comparisons between hydraulic recommended properties in
Rosetta and those in NUREG/CR-7028 demonstrate that the mean hydraulic properties are
similar in both cases. However, as pointed out above, the variability assumed in the hydraulic
properties chosen to represent the soils in the DU PA Model v1.2 is not appropriately
characterized, and this limitation in the model biases the modeling results greatly.
While it is true that engineered soils undergo pedogenesis and become more like natural soils
over time, it is important to follow NUREG/CR-7028 guidelines. The fact that the GoldSim model
uses values for its Ksat distribution that, at the low end, are two orders of magnitude lower than
specified in NUREG/CR-7028, and that the low-permeability range of values is overweighted,
does not lead to confidence that the GoldSim model is set up appropriately.
Furthermore, in the GoldSim model as implemented, it is assumed for the input parameter values
that there is no correlation between log(α) and log(Ksat). When databases based on natural soils
are used, it is important to account for correlation between these two parameters. Strong
correlation between log(α) and log(Ksat) (with R2 = 0.9) has been established for the largest
database in North America, as well as for the largest database in Europe [see Sections 4.1.1.1
and 4.4.1 of the safety evaluation report (SER)]. The two correlation equations are quite similar.
Furthermore, a mathematical relationship similar to the correlation equations has been
developed from fundamental soil physics theory by Guarracino (2007).
Failure to account for this correlation, or other, significant correlations (e.g., correlation in
individual parameter values between different cover-system soil layers), leads to unrealistic
modeling. As stated in GoldSim’s User Manual, Appendix A: Introduction to Probabilistic
Simulation (GTG 2013):
Ignoring correlations, particularly if they are very strong (i.e., the absolute value of the
correlation coefficient is close to 1) can lead to physically unrealistic simulations. In the
above example, if the solubilities of the two contaminants were positively correlated
(e.g., due to a pH dependence), it would be physically inconsistent for one
contaminant’s solubility to be selected from the high end of its possible range while the
other’s was selected from the low end of its possible range. Hence, when defining
probability distributions, it is critical that the analyst determine whether correlations need
to be represented.
The response has also clarified that the Surface Layer and Evaporative Zone Layer were each
assigned a geometric mean hydraulic conductivity of 5 ´ 10-7 meters per second (m/s). This
hydraulic conductivity is considered unrealistically low for in-service near-surface layers (e.g.,
< 10 feet deep) that will be densely structured due to wet-dry cycling, freeze-thaw cycling, and
biota intrusion by roots, insects, etc. This unrealistically low Ksat at or near the surface may have
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 88
choked off infiltration in the HYDRUS model and exacerbated runoff, thereby limiting deeper
ingress of meteoric water in the profile and under-predicting percolation. As discussed in Section
4.1.1.1 of the SER, the unrealistically low near-surface Ksat value, combined with the
unrealistically high Frost Protection Layer Ksat value, which is inputted into the model, would
tend to create in the model an unrealistic, artificial capillary barrier at the top of the higher
permeability layer that would inappropriately render modeled values of infiltration extremely low.
Soils at the surface develop significant structure and generally are much more permeable than
those much deeper in the profile. EnergySolutions will need to provide additional evidence that
this assumed hydraulic conductivity did not artificially bias the HYDRUS modeling.
The response to Comment 4 also indicates that NUREG/CR-7028 recommends using a single
measurement from a single site to define α. This is an incorrect interpretation of the design
recommendations in NUREG/CR-7028. The recommendation in NUREG/CR-7028 to use α = 0.2
1/kilopascal (kPa) applies when reliable site-specific information is not available and when a
single typical value (not a range of values) is desired. It is based on an interpretation of the
dataset presented in NUREG/CR-7028 as accounting for scale-dependent hydraulic properties.
The HYDRUS modeling in Appendix 5 used an α that is approximately one order of magnitude
lower than the recommendation in NUREG/CR-7028. This α is based in part on historic
measurements made at Colorado State University on core samples obtained at the Clive site by
Bingham Environmental (1991), which are known to be too small and too disturbed to adequately
represent in-service conditions. The relevancy of this historic data from Bingham Environmental
is dubious, at best.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
EnergySolutions assumptions regarding the in-service versus naturalized parameters.
2.18.1 Interrogatory Response
The class average values of soil hydraulic function parameters for the 12 soil textural
classifications in Rosetta were based on 2,134 soil samples for water retention, and 1,306 soil
samples for saturated hydraulic conductivity (Schaap et al. 2001). These data were obtained from
the RAWLS, AHUJA, and UNSODA databases (Schaap et al. 2001). Given the stronger
economic incentive for characterizing agricultural land than for rangeland, the more extensive
soils databases are derived from data obtained from agricultural lands. Soil textural
classifications are determined by particle size distributions, not by land use, so these databases
have utility for non-agricultural application. Since the objective of the revegetation plan (SWCA
2013) is to develop a sustainable, steady-state condition that mirrors the natural system, few
inputs used in the hydrologic modeling would be related to engineered properties.
The available Unit 4 soil texture data indicate that the sample represents an extreme of the range
of particle sizes that compose the silty clay textural class. Distributions were developed for the
van Genuchten α and n parameters for the Surface and Evaporative Zone Layers that represented
the entire range of the silty clay class by using the mean and standard deviation values provided
by the Rosetta database.
The Benson et al. (2011) report published by the NRC (NUREG/CR-7028) provides
recommendations for ranges of hydraulic parameters that may be used to represent in-service
conditions of store-and-release and barrier layers in covers.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 89
The Surface and Evaporative Zone Layers in the Clive ET cover system correspond to store-and-
release layers. The α and n values of the van Genuchten water retention model strongly influence
the capacity of the Unit 4 material in the Surface and Evaporative Zone Layers to hold water. To
provide a better estimate of the uncertainty of these Unit 4 properties, statistical distributions
were developed for α and n to be used for inputs for the flow modeling. To develop the
distributions for flow model input, α and n values were taken from the distributions of the mean
and the standard deviation for each parameter from the Rosetta model database of hydraulic
parameters for the textural class of silty clay (USDA 2017). The standard deviations were
converted to standard errors by dividing by √n, where n is the number of samples (28 in this
case). Values of the van Genuchten parameter α for these two layers were drawn from a
statistical distribution with a mean of 0.016 1/cm. The value for α recommended for in-service
layers by Benson et al. (2011) (p. 10-4) is 0.2 1/kPa, which corresponds to a value of 0.02 1/cm,
similar to the mean used for the infiltration simulations. The distribution used for the van
Genuchten n parameter for the flow model simulations had a mean of 1.32. The value for n
recommended for in-service layers by Benson et al. (2011) (p. 10-4) is 1.3.
To manage computational burden, the saturated hydraulic conductivity, Ks, for the Surface Layer
and Evaporative Zone Layer was assigned a single deterministic value of 4.46 cm/day
(5.16 × 10-5 cm/s) based on the mean value determined from analysis of the Unit 4 cores
(Equation 29 of Neptune (2015c)). The hydraulic properties for Units 3 and 4 are based on
laboratory measurements by the Colorado State University (CSU) Porous Media Laboratory for
the moisture retention and hydraulic conductivity of core samples from Units 3 and 4 at the Clive
Site (Bingham Environmental 1991). The Unit 4 material used for the Surface and Evaporative
Zone Layers is classified as a silty clay. The average saturated hydraulic conductivity assigned to
a silty clay is 0.48 cm/day from the Carsel and Parrish (1988) database and 9.61 cm/day from the
Rosetta database (USDA 2017). The value used for Ks is clearly realistic as it was derived from
site-specific data and falls between estimates from two widely used soil hydraulic parameter
databases. Comparison with results obtained by Benson et al. (2011) can be made with mean
values of the Ks of store-and-release layers of in-service covers listed in Table 6.6 of Benson et
al. (2011). The geometric mean of these results is 8.7 × 10-7 m/s or 7.5 cm/day. This value is less
than twice the value used for the infiltration modeling.
A distribution for the saturated hydraulic conductivity (Ks) of the radon barriers was developed
to use for the modeling. The resulting distribution described in the response to Interrogatory 21/2
is:
𝐾𝑠 ~ 𝐿𝑜𝑔𝑛𝑜𝑟𝑚𝑎𝑙() with a right
shift of 0.00432 cm/day.
The value for Ks recommended by Benson et al. (2011), p. 10-3, for modeling “typical
conditions” of in-service cover layers is 5 × 10-7 m/s (4.32 cm/day), which is well within the
distribution used for the Clive DU PA infiltration modeling.
The minimum value of 0.00432 cm/day, corresponding to the design specification for the radon
barriers (Whetstone Associates 2011), was used as a shift. Note that the minimum value was not
used to fit the distribution, but simply to constrain the distribution by not allowing Ks values
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 90
smaller than that. UDEQ suggests that truncating the distribution at the design Ks value
“overweights” the low end of the distribution. This is not correct, as the shift only makes it
impossible to choose a value for Ks that is less than the design value.
Single values of α and n determined from site-specific measurements were used for the radon
barrier in the infiltration modeling to manage computational burden. A value of 0.003 1/cm was
used for α and a value of 1.17 was used for n. These values were used in previous modeling by
Whetstone Associates (2011). Benson et al. (2011) describe α as varying between 0.001 1/cm
and 0.032 1/cm. A range from 1.2 to 1.4 is recommended by Benson et al. (2011) for the n
parameter. The value used for the infiltration modeling is slightly below the low end of that
range.
Correlations between hydraulic parameters were not included in the distribution development.
See the response to Interrogatory 05/2 for a discussion of why hydraulic parameter correlations
are not included.
In this comment UDEQ has raised a number of objections to the choice of parameter values and
statistical distributions used for the Clive DU PA infiltration modeling. These objections appear
to be based on the belief that the recommendations contained in NUREG/CR-7028 (Benson et al.
2011) should be rigidly applied to the Clive Site. Hydrogeologic models, however, are site-
specific. Climate conditions and soil formation processes at the Clive Site contradict the
assumptions of rapid soil structure formation in the cover layers observed by Benson et al.
(2011) at other sites and demonstrate the inapplicability of the conceptual model to the Clive
Site. The one-size-fits-all generalization of hydraulic properties does not appropriately represent
the unique conditions at the Clive Site. See the response to Interrogatory 05/2:
Evapotranspiration Cover (ET Cover) and the response to Interrogatory 153/2.
For discussion of capillary barrier behavior in the cover system, see the response to Comment
B.3.
In this interrogatory UDEQ again questions the quality of the hydraulic property measurements
made by the Groundwater and Porous Material Laboratory at Colorado State University. See the
discussion under Naturalized Cover in the response to Interrogatory 21/2.
2.19 SER B.5 Supplemental Interrogatory Comment 5
5) a) Provide additional explanation/justification for the assumed surface boundary
condition and the sensitivity of the HYDRUS results to the boundary conditions.
b) Also, why is a linear regression the optimal surface response for the design?
DEQ Critique from April 2015 SER, Appendix B:
The interrogatory asked for additional justification for the assumed surface boundary condition.
EnergySolutions’ response explains how the boundary condition was created but does not
provide justification for the boundary condition. Two shortcomings need to be addressed
explicitly.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 91
First, the repetition of the same 100-year periods 10 times to represent the climatic conditions
over a 1000-year period of climatic input will need to be justified quantitatively. For all practical
purposes, this simulation strategy will provide essentially the same output for each 100-year
period in the record. This demonstration should show that the meteorological conditions over a
1000-year period, including extreme events expected over a 1000-year period, can be
represented adequately using a sequence of repeated 100-year records. Normally, longer
periods of time involve greater variability in the data. This requested demonstration should also
show that the impacts of these extremes on the hydrological response of the cover are
adequately represented.
Second, the justification should show that the hydrological behavior at the upper boundary (i.e.,
surface of the cover) is reasonable and within expected norms. This has not been demonstrated
in Appendix 5 (Neptune 2014b), and the unrealistically low Ksat assigned to the Surface Layer
(see Comment 4) in combination with likely capillary-barrier effect artifacts in the model may
have choked off infiltration into the cover profile. At a minimum, water-balance graphs should be
presented for typical and wet years showing the temporal behavior of each of the primary
cumulative water-balance variables for the cover (e.g., precipitation, runoff, soil water storage,
evapotranspiration, percolation). These graphs, and their associated discussion, should
demonstrate that the surface boundary is represented adequately and that predictions are within
expected norms.
The absence of climate change considerations should also be presented in the context of the
most recent climate science, which does show systematic shifts in climate throughout North
America within the next 10,000 years, if not sooner. An explanation should also be provided as to
why climate change is not relevant at the Clive site when it has been considered in performance
assessments for other disposal facilities in the region (e.g., the Monticello U mill tailings disposal
facility).
EnergySolutions’ response also provides an extensive discussion to justify the efficacy of
Equations 39 and 40 in Appendix 5. However, these outcomes may have been biased by the
unrealistically low Ksat assigned to the Surface Layer and Evaporative Zone Layer (see
Comment 4), which, in combination with likely capillary-barrier effect artifacts in the model, may
have choked off infiltration into the cover profile. The efficacy of Equations 39 and 40 should be
revisited once the impacts of the unrealistically low Ksat assigned to the Surface Layer and
Evaporative Zone Layer (see Comment 4) have been investigated.
As an alternative to the linear regression, DEQ/SC&A fit an exponential equation to the van
Genuchten α, n, and Ksat input data and the HYDRUS-calculated fluxes (Figure B-1). The
triangles shown in Figure B-1 are the fluxes calculated using the following exponential fit:
Flux = 45.465 ´ α1.4408 ´ n-1.332 ´ Ksat-0.445. For large fluxes, the exponential fit does not appear to
be much better than the linear fit, but for small fluxes (which tend to result when the van
Genuchten α is small), the exponential fit is much better than the linear fit.
DEQ Critique of DU PA v1.4, See Interrogatory 21 for a description of the EnergySolutions
assumptions regarding the linear regression of the GoldSim versus HYDRUS infiltration rates.
DEQ Critique of DU PA V1.4, Appendix 21: No changes have been made with respect to the
treatment of the surface boundary conditions.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 92
2.19.1 Interrogatory Response
a) Atmospheric Boundary Conditions
The methods used to develop atmospheric boundary conditions applied to the top of the Surface
Layer in the HYDRUS models were described in Unsaturated Zone Modeling for the Clive PA,
Clive DU PA Model v1.4 (Neptune 2015c). Precipitation, surface runoff, and evaporation under
time-varying climate conditions are boundary condition information required by HYDRUS. The
data required for the atmospheric boundary condition includes daily values of precipitation,
potential evaporation, and potential transpiration to represent the time-variable boundary
conditions on the upper surface of the cover. The long-term evaluation period for this analysis
makes it necessary to generate a representative climate record with a longer term than any
existing daily data record. The model is deliberately run for a long period of time (1,000 years)
in order to reach a near-steady state net infiltration rate that is not influenced by the initial
conditions.
The WGEN model (Richardson and Wright 1984) was used to generate a 100-year synthetic
precipitation record for the Site. The WGEN model is a component of the widely used HELP
model (Schroeder et al. 1994a; Schroeder et al. 1994b). A 100-year precipitation record was
generated using the monthly average values from measurements at the Site based on 17 years of
observations. Meteorological measurements at the Site have been shown to be consistent with
observations at National Oceanographic and Atmospheric Administration (NOAA) stations
(Whetstone Associates 2011). Use of the WGEN model is consistent with U.S. NRC guidance
(Meyer et al. 1996). These authors found a good comparison between observed and WGEN
simulated monthly average precipitation and histograms of daily precipitation for an arid site
example.
Daily potential evapotranspiration (PET) was calculated with values of daily maximum (Tmax),
minimum (Tmin), and mean (Tmean) temperatures and extraterrestrial radiation using the
Hargreaves method. This approach is used extensively in accepted modeling platforms such as
the Soil and Water Assessment Tool (Neitsch et al. 2011) and is also documented in the
HYDRUS manuals (Šimůnek et al. 2013). The use of appropriate meteorological data and
accepted methods for estimating daily boundary condition values supports the confidence in the
application of these boundary conditions. Repetition of a long-term daily record to provide a
boundary condition for an even longer-term simulation is often necessary for the large timescales
required by PA models. Examples can be found in EPA (2002) and Levitt (2011).
UDEQ argues that “an explanation should also be provided as to why climate change is not
relevant at the Clive site when it has been considered in performance assessments for other
disposal facilities in the region.” This response provides a brief discussion of potential climate
change impacts on unsaturated zone hydrology taken from a more detailed consideration of
issues involved with representing global climate change in performance assessment models by
Crowe et al. (2017).
In evaluating the influence of climate change on the unsaturated zone, the standard approach in
past performance assessments is to assume bounding or worse-case effects of future climate
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 93
changes through model evaluations of the wettest and coolest glacial climate in order to
maximize future precipitation and infiltration. While potentially useful, these
bounding/conservative approaches do not consider the range of possible site-specific effects
from both long-term and abrupt anthropogenic driven climate change. Instead, climate scenarios
used in performance assessments evaluate natural variations in past climate associated with
glacial and interglacial conditions. Further, the timing of future climate change or climate-driven
events are not assessed; transition to a glacial climate is assumed during the isolation period of
disposed waste.
The earth is currently in an interglacial period and the timing of the inception of the next glacial
period is critically important for forecasting future climate states. Climate modeling studies
coupling changes in the earth’s orbital configurations with variable atmospheric CO2 content
demonstrate conclusively that the earth is very unlikely to return to a glacial state when CO2
concentrations remain above pre-industrial levels (~ 280 parts per million; current levels exceed
400 parts per million). A return to pre-industrial concentrations will likely require in excess
of 50,000 years and may require hundreds of thousands of years even if current emissions were
to drop dramatically.
The most likely scenario for the future global climate is continuation of the current interglacial
climate under conditions of variable but progressive global warming. Previously used
assumptions in performance assessments of bounding glacial climates (coldest/wettest
conditions) are no longer applicable and under some conditions may require design of overly
protective and costly closure covers to reduce infiltration.
UDEQ maintains that “unrealistically low Ks” values have been assigned to the surface layer
material in the flow model. This value comes from testing of core samples from the Clive Site.
See the response to UDEQ Comment B.4 for further discussion. UDEQ argues that “likely
capillary-barrier effect artifacts in the model, may have choked off infiltration into the cover
profile.” The hydraulic property values and statistical distributions assigned to the Evaporative
Zone and the Frost Protection Layers are reasonable estimates based on site-specific information
and commonly used soils databases. Given these property assignments, the Frost Protection
Layer behaves hydraulically by enhancing storage in the Evaporative Zone in a realistic manner.
See the response to UDEQ Comment B.3 for further discussion on capillary barriers.
UDEQ has been provided with annual averages for water balance components of precipitation,
runoff, evapotranspiration, storage, and deep drainage. UDEQ claims that these water balance
plots are not adequate and have produced numerous examples of daily water balance plots. Daily
water balance is not the appropriate scale for evaluating a performance assessment model. See
the response to Interrogatory 21/2 for a discussion of evaluation of flow model water balance.
b) Use of Linear Model for Abstraction
UDEQ then discusses comparisons between linear and quadratic regression models developed
previously from the results of the 20 HYDRUS realizations.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 94
In response to supplemental interrogatories submitted to EnergySolutions on August 11, 2014
(EnergySolutions 2014), additional capability was developed to allow for more HYDRUS
simulations of the cover system to be completed within a practical time period. Fifty
HYDRUS-1D simulations were conducted to evaluate the uncertainty in infiltration flux into the
waste zone and water content within each ET cover layer as a function of hydraulic property
uncertainty; these simulations were documented in DU PA v1.4 Appendix 5 (Neptune 2015c).
Therefore, modeling of net infiltration and cover layer water content in Clive DU PA v1.4 is
based on 50 realizations/cases, not the 20 referred to in the interrogatory.
Predictions of average annual net infiltration from regression models with only linear terms, or
with quadratic terms, are compared to the net infiltrations actually obtained from HYDRUS
(Figure 20). The predictions from the regression models use the same input values used for the
50 HYDRUS cases (i.e., the predictions are simply the fitted values from the fitting of the
regression model to the data from HYDRUS).

Figure 20. A comparison of predictions (fitted values) from the linear and quadratic
regression models, against the HYDRUS results for net infiltration used in the model
fitting.
As seen in Figure 20, there is curvature in the HYDRUS net infiltrations that is not fully captured
in the fitted regression models, even with quadratic terms. However, HYDRUS is a complicated
process model, and an abstraction of reality itself, and the goal of the abstraction is not to capture
●●●
●●
●
●
●●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●●●●
●
●
●
●
●●●
●
●
●
●●
●●
●
●●●
●
●●●●●
●●
●
●
●●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●●●●
●
●
●
●
●●●
●
●
●
●●
●●
●
●●●
●
●●
0.00
0.05
0.10
0.15
0.20
0.00 0.05 0.10 0.15 0.20
Net Infiltration Predictions(mm/yr)
HY
D
R
U
S

N
e
t

I
n
f
i
l
t
r
a
t
i
o
n

(
m
m
/
y
r
)
Model
●
●
Linear
Quadratic
Linear and Quadratic Regression Fits
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 95
the HYDRUS net infiltration response surface perfectly, but to provide reasonable infiltration
rates into GoldSim.
The ranges of the predictions from the two regression models are very similar, with most values
between 0 and 0.05 (Figure 21). The predictions from the model with only linear terms tend to be
slightly larger than those from the quadratic model by about 0.02 to 0.05, but the differences are
so small that they are not practically meaningful. The small differences in predictions suggest
both models are reasonable; the linear regression model was chosen for implementation in DU
PA Model v1.4 because it is simpler and matches model form and is also used for water content
across layers. Additional discussion of the predictions from the regression model with only linear
terms is provided in the response to Interrogatory 21/2.

Figure 21. A comparison of predictions from the linear and quadratic regression models
based on the input values used for the 50 HYDRUS runs.
As described in Section 1.1 and in the responses to Interrogatory 21/2 and Comment B.1, the
goal of the model abstraction is to develop linear regression models to predict reasonable net
infiltration rates and water contents for use in the DU PA Model v1.4. This comment (B.5) refers
to an “exponential fit” based on the results from an original 20 HYDRUS runs associated with
DU PA Model v1.2; the specific fit referenced is not relevant to the DU PA Model v1.4 because
of changes to distributions and the use of 50 HYDRUS cases for the model abstractions
described previously in this response.
●●●●
●
●
●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●●
●
●
●
●
0.00
0.05
0.10
0.15
0.20
0.00 0.05 0.10 0.15 0.20
Model Predictions with Linear Terms (mm/yr)
Mo
d
e
l

P
r
e
d
i
c
t
i
o
n
s

w
i
t
h

Q
u
a
d
r
a
t
i
c

T
e
r
m
s

(
m
m
/
y
r
)
Linear and Quadratic Regression Comparison
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 96
The model form suggested by UDEQ is one of many possible reasonable alternatives and, as
expected, does capture some of the curvature in the HYDRUS results not captured by the linear
regression model used for DU PA Model v1.4. However, as described in other responses, the
goal of the model abstraction is to produce reasonable values of net infiltration and water content
to use for GoldSim realizations, not to match the HYDRUS results as closely as possible;
HYDRUS is itself a model of reality with its own approximations and assumptions.
For comparison, model form suggested by UDEQ was refit using the 50 HYDRUS realizations
for v1.4. The predicted values for net infiltration are very similar between the two models
(Figure 22), with the linear regression model predictions covering a slightly larger range of
infiltrations and having slightly higher infiltrations in the middle of the range covered. The small
differences in predictions suggest both models are reasonable; the linear regression model was
chosen for implementation in DU PA Model v1.4 because it is simpler and matches the model
form also used for water content across layers. Additional discussion of the predictions from the
regression model with only linear terms is provided in the response to Interrogatory 21/2.

Figure 22. Comparison of predicted net infiltration rates at the inputs used with the 50
HYDRUS runs for the linear regression model and the “exponential” model
suggested by UDEQ.

●●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●●●●
●
●
●
●
●
●●
●
●
●
●●
●●
●
●●●
●
●●
0.00
0.05
0.10
0.15
0.20
0.00 0.05 0.10 0.15 0.20
Model Predictions with Linear Terms (mm/yr)
Mo
d
e
l

P
r
e
d
i
c
t
i
o
n
s

w
i
t
h

E
x
p
o
n
e
n
t
i
a
l

T
e
r
m
s

(
m
m
/
y
r
)
Linear and Exponential Fits
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 97
2.20 SER B.6 Supplemental Interrogatory Comment 6
6) To summarize the 20 HYDRUS results, Appendix 5, Section 12.9 states: “Infiltration flux
into the waste zone ranged from 0.007 to 2.9 mm/yr, with an average of 0.42 mm/yr, and
a log mean of 0.076 mm/yr for the 20 replicates.” In addition to this statement, provide
the results for each HYDRUS run so that the results can be matched to the input data.
DEQ Critique from April 2015 SER, Appendix B:
This interrogatory requested that the results be provided for each HYDRUS run so that the
results can be matched to the input data. The response included a spreadsheet summarizing
percolation from the base of the cover and water contents from the HYDRUS analysis. However,
the output from HYDRUS was not provided.
The output from HYDRUS should be included in the report and presented in a manner consistent
with the practice associated with design and evaluation of water-balance covers (i.e., ET covers).
Water-balance graphs should be reported showing the key water-balance quantities, and
discussion should be provided that demonstrates that the predictions are within expected norms
for water-balance covers. This type of presentation and discussion has not been provided in
Appendix 5 or in subsequent responses to interrogatories.
EnergySolutions’ response also discusses graphs in an attached spreadsheet and indicates that
these graphs demonstrate that there is no relationship between percolation from the base of the
cover and Ksat of the radon barrier. This finding may have been biased by the unrealistically low
Ksat assigned to the Surface Layer and Evaporative Zone Layer (see Comment 4), which, in
combination with likely capillary-barrier effect artifacts in the model, may have choked off
infiltration into the cover profile. This issue needs to be reevaluated once the impact of the Ksat
assigned to the near-surface layers has been addressed.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
adequacy of the HYDRUS model output.
2.20.1 Interrogatory Response
In response to supplemental interrogatories submitted to EnergySolutions on August 11, 2014,
additional capability was developed to allow for more HYDRUS simulations of the cover system
to be completed within a practical time period. Fifty HYDRUS-1D simulations were completed
to evaluate the uncertainty in infiltration flux into the waste zone, and water content within each
ET cover layer as a function of hydraulic property uncertainty, and were documented in DU PA
v1.4 Appendix 5 (Neptune 2015c). See the response to UDEQ Comment B.1 (Section 2.15.1).
The input parameters for the 50 realizations were provided in Appendix 5 of DU PA Model v1.4
(Neptune 2015c) and are reproduced in this document in Table 6. Flow model results for use in
the GoldSim model consisted of the average annual fluxes into the top of the waste layer and
average annual water contents for the Surface Layer, Evaporative Zone Layer, Frost Protection
Layer, and Upper and Lower Radon Barriers averaged over the years 900 to 1000 of the
simulation. The outputs from HYDRUS for each realization are provided in Table 7. The results
of these 50 realizations were used to fit regression equations for net infiltration and water content
for the DU PA Model v1.4.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 98
Table 7. Results of 50 flow realizations described in Appendix 5 of DU PA Model v1.4.
1 0.0107 0.22 0.25 0.07 0.25 0.25
2 0.0123 0.21 0.25 0.07 0.24 0.24
3 0.0085 0.21 0.25 0.07 0.25 0.25
4 0.0236 0.22 0.26 0.07 0.25 0.25
5 0.0279 0.23 0.27 0.07 0.27 0.27
6 0.0079 0.23 0.27 0.07 0.24 0.24
7 0.0230 0.21 0.25 0.07 0.26 0.26
8 0.0098 0.23 0.26 0.07 0.26 0.26
9 0.0146 0.21 0.25 0.07 0.26 0.26
10 0.0131 0.22 0.25 0.07 0.25 0.25
11 0.0221 0.24 0.28 0.07 0.26 0.26
12 0.0281 0.23 0.26 0.07 0.27 0.27
13 0.0144 0.22 0.26 0.07 0.26 0.26
14 0.0380 0.22 0.25 0.07 0.27 0.27
15 0.0254 0.22 0.26 0.07 0.27 0.27
16 0.0122 0.24 0.28 0.07 0.26 0.26
17 0.0663 0.21 0.25 0.07 0.27 0.27
18 0.0084 0.22 0.26 0.07 0.24 0.24
19 0.0364 0.22 0.26 0.07 0.27 0.27
20 0.1832 0.22 0.26 0.08 0.28 0.28
21 0.0075 0.24 0.28 0.07 0.24 0.24
22 0.0248 0.24 0.28 0.07 0.25 0.25
23 0.0301 0.22 0.25 0.07 0.28 0.28
24 0.0458 0.20 0.24 0.07 0.27 0.27
25 0.0084 0.22 0.26 0.07 0.24 0.24
26 0.0109 0.21 0.24 0.07 0.27 0.27
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 99
Realization
Net
Infiltration
(mm/yr)
Volumetric Water Content (-)
Surface Evaporative
Zone
Frost
Protection
Upper
Radon
Lower
Radon
27 0.0104 0.21 0.25 0.07 0.24 0.24
28 0.0117 0.24 0.28 0.07 0.25 0.25
29 0.0106 0.21 0.25 0.07 0.24 0.24
30 0.0821 0.22 0.26 0.08 0.26 0.26
31 0.0220 0.21 0.25 0.07 0.27 0.27
32 0.0649 0.23 0.27 0.07 0.27 0.27
33 0.0168 0.23 0.26 0.07 0.26 0.26
34 0.0067 0.24 0.28 0.07 0.22 0.22
35 0.0101 0.23 0.27 0.07 0.23 0.23
36 0.0149 0.21 0.25 0.07 0.27 0.27
37 0.0210 0.25 0.29 0.07 0.26 0.26
38 0.0086 0.22 0.26 0.07 0.24 0.24
39 0.0233 0.22 0.26 0.07 0.25 0.25
40 0.0096 0.23 0.27 0.07 0.25 0.25
41 0.0088 0.25 0.29 0.07 0.25 0.25
42 0.0151 0.23 0.27 0.07 0.25 0.25
43 0.0155 0.22 0.26 0.07 0.23 0.23
44 0.0301 0.23 0.27 0.07 0.24 0.24
45 0.0168 0.22 0.25 0.07 0.26 0.26
46 0.0142 0.24 0.28 0.07 0.25 0.25
47 0.0122 0.23 0.27 0.07 0.26 0.26
48 0.0471 0.24 0.28 0.07 0.26 0.26
49 0.0157 0.24 0.28 0.07 0.27 0.27
50 0.0130 0.24 0.28 0.07 0.25 0.25

UDEQ requests plots of flow model water balance components on a daily basis. See the
discussion of the applicability of daily water balance plots to PA models in the response to
Interrogatory 21/2.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 100
UDEQ questions the outcome from the fitting of the regression equation that net infiltration rates
are not sensitive to the magnitude of the saturated hydraulic conductivity of the radon barriers.
See the response to UDEQ Comment B2 (Section 2.16.1) for discussion of this issue.
2.21 SER B.7 Supplemental Interrogatory Comment 7
7) The HYDRUS and GoldSim calculated infiltration rates (and perhaps other intermediary
results) need to be provided in the report, so that the reviewers do not have to delve into
the code’s output files. For example, provide dot plots of the infiltration rates through the
surface layer and/or provide a statistical summary of the infiltration rates that were
sampled in GoldSim.
DEQ Critique from April 2015 SER, Appendix B:
This interrogatory requested that the percolation rates reported by HYDRUS be presented
directly in the report. The response includes Figure 4, which shows “infiltration” in mm/yr for
various layers in the cover and Figure 5, which shows “infiltration” (interpreted as percolation
from the base of the cover) from HYDRUS and predicted with the regression equation, i.e.,
Equation 39 in Appendix 5.
The quantities shown in Figure 4 need more explanation. Infiltration is defined as the flux of
water across the atmosphere-soil interface in response to precipitation. Water movement below
the surface is a volumetric flux, and the flux from the base of the cover and into the waste is the
percolation rate for the cover. Do these quantities represent the net flux from the base of each
layer in the cover? The “infiltration” for the surface layer report in Figure 4 also raises concern, as
the results indicate that the unrealistically low Ksat assigned to the Surface Layer and
Evaporative Zone Layer (see Comment 4), in combination with likely capillary-barrier effect
artifacts in the model, may have choked off infiltration into the cover profile and unrealistically
limited downward movement of water. A discussion of the HYDRUS predictions in the context of
cumulative water-balance quantities and expected norms for water-balance covers could
address this issue.
As indicated in the discussion associated with Comment 1, the predictions shown in Figure 5
illustrate that the percolation rate from the regression used in GoldSim is considerably different
from the predictions made with HYDRUS and is essentially insensitive to the hydraulic properties
used as input. The lack of sensitivity is attributed to the reduction in log-variance to address
spatial averaging, but another plausible explanation is that Equation 39 reflects central conditions
adequately but extreme conditions in the tailings inadequately. Yet another plausible explanation
is the likely capillary-barrier effect artifacts in the model, which would minimize or possibly even
exclude infiltration of water to greater depths, so long as evaporation could remove it from the
upper two soil layers. Furthermore, evapotranspiration rates in the model are likely too high,
since they do not account for accumulation of gravel at the surface over time, which would tend
to greatly diminish evaporation. A quantitative demonstration and explanation is needed to
address this issue.
The response should also indicate how and why temporal scaling was incorporated into the
hydraulic properties, as indicated by the term “spatio-temporal” used in the response to the
interrogatory. Temporal scaling should account explicitly for the temporal evolution of the
distribution of hydraulic properties due to pedogenic effects. No discussion has been provided
regarding a temporal evolution of hydraulic properties. If temporal scaling has not been
incorporated, then scale matching should be described as spatial rather than spatio-temporal.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 101
EnergySolutions’ response should also indicate why conventional spatial averaging procedures
for correlated hydraulic properties were not used in the spatial scaling process from point scale
measurements in the Rosetta database to grid scale in the model. Spatial scaling from a point
measurement to model grid scale will need to account for upscaling of the mean to address
measurement bias as well as downscaling of the log-variance in a manner consistent with the
spatial correlation structure of engineered but degraded-over-time in-service earthen cover soils.
The response should indicate how these factors are addressed by reducing the log-variance by
the square root of the sample size in the Rosetta database.
The discussion below illustrates DEQ’s mathematical (as opposed to hydrogeologic) concerns
with the way infiltration is being abstracted into GoldSim from the HYDRUS results.
1) The linear regression equation that has been programmed into GoldSim does not give
results that are consistent with what is calculated by HYDRUS (i.e., for a given pair of α and
n, the regression equation result in GoldSim does not approximate the HYDRUS result).
This is demonstrated by Figure B-1 (See DEQ Critique to Supplemental Interrogatory
Comment 5).
2) As acknowledged by EnergySolutions in its responses to Supplemental Interrogatories 1
and 2, due to scaling effects the ranges for α and n that have been programmed into
GoldSim are more narrow than those in HYDRUS (i.e., in HYDRUS, α ranges from
0.001883 to 0.3021, but in GoldSim, α only ranges from 0.005 to 0.0493; likewise, in
HYDRUS, n ranges from 1.029 to 1.883, but in GoldSim n only ranges from 1.060 to 1.540).
See Figure B-2 and Figure B-3 for complementary cumulative distribution (CCD)
comparisons that were prepared by SC&A utilizing EnergySolutions HYDRUS results and
the Neptune (2014b), Table 1 GoldSim α and n distributions.

The CCD comparison in Figure B-4 shows the effect of these two mathematical considerations
on the resulting GoldSim infiltration rate. This infiltration CCD is very similar to Figure 5 of the
EnergySolutions Response to Supplemental Interrogatories, except that it is rotated 90 degrees.
Note that GoldSim was not re-run for these analyses. Instead, the GoldSim equations were
programmed into an Excel Crystal Ball file, and 10,000 realizations were run. Also, the reason
the GoldSim CCDs are smoother than the HYDRUS CCDs is that the GoldSim CCDs have
10,000 points, whereas the HYDRUS CCDs have only 20.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
adequacy of the GoldSim and HYDRUS model output.
2.21.1 Interrogatory Response
In this comment UDEQ first references the regression models for net infiltration developed from
the HYDRUS simulations with 20 realizations from Appendix 5 of v1.2 of the Clive DU PA
Model (Neptune 2014). As described previously, in response to supplemental interrogatories
submitted to EnergySolutions on August 11, 2014 (EnergySolutions 2014), additional capability
was developed to allow for more HYDRUS simulations of the cover system to be completed
within a practical time period. Fifty HYDRUS-1D simulations were conducted to evaluate the
uncertainty in infiltration flux into the waste zone and water content within each ET cover layer
as a function of hydraulic property uncertainty; these simulations were documented in DU PA
v1.4 Appendix 5 (Neptune 2015c). Therefore, net infiltration and cover water content values
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 102
used in Clive DU PA v1.4 are based on 50 realizations/cases, not the 20 referred to in this
comment.
UDEQ requests that HYDRUS and GoldSim calculated infiltration rates be provided in a report
and that a statistical summary of the infiltration rates that were sampled in GoldSim be provided.
The input parameters for the 50 realizations were provided in Appendix 5 of DU PA Model v1.4
(Neptune 2015c). Infiltration rates used in GoldSim can be calculated using the regression model
in Appendix 5 (Neptune 2015c). The net infiltration rates calculated in the 50 HYDRUS
realizations are given in Table 7 of this document. A statistical summary of net infiltration
calculated in the 50 realizations is given in Section 12.9 of Appendix 5 (Neptune 2015c).
UDEQ shows comparisons between the van Genuchten α and n values generated using the
distributions for these parameters from the GoldSim model and the values used for the 20
HYDRUS simulations in their Figures B-2 and B-3. In addition, UDEQ shows a comparison in
their Figure B-4 between values of net infiltration generated using the regression model and the
values resulting from the 20 HYDRUS realizations. These comparisons for input parameters and
net infiltration clearly differ.
UDEQ has suggested a number of causes for these differences but, as described in
EnergySolutions (2014), these differences were due to not scaling the α and n distributions used
as inputs for the HYDRUS simulations. The proper scaling has been applied for the 50
HYDRUS simulations described in Appendix 5 of the DU PA Model v1.4 (Neptune 2015a).
The same comparisons using v1.4 distributions for α and n are shown in Figure 23 and Figure 24,
corresponding to the v1.2 comparisons in UDEQ’s Figures B-2 and B-3. Comparisons between
the complementary cumulative distribution functions for the 50 net infiltration values from the
HYDRUS realizations and 1000 values generated from the regression model used in GoldSim
are shown in Figure 25. These results demonstrate that proper scaling of the input parameter
distributions has resulted in consistency between HYDRUS inputs and outputs and the simplified
model used in GoldSim for net infiltration.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 103

Figure 23. Re-creation of information in UDEQ Figure B-2 showing complementary
cumulative distribution functions (CDFs) for the 50 α values used in the HYDRUS
realizations and 1000 values drawn from the distribution used in GoldSim for the
Clive DU PA Model v1.4.

Figure 24. Re-creation of information in UDEQ Figure B-3 showing complementary
cumulative distribution functions (CDFs) for the 50 n values used in the HYDRUS
realizations and 1000 values drawn from the distribution used in GoldSim for the
Clive DU PA Model v1.4.
0.00
0.25
0.50
0.75
1.00
0.00 0.01 0.02 0.03 0.04 0.05
α(1/cm)
Co
m
p
l
e
m
e
n
t
a
r
y

C
D
F
Model
HYDRUS
GoldSim
Complementary CDF for α: GoldSim and HYDRUS
0.00
0.25
0.50
0.75
1.00
1.2 1.3 1.4 1.5
n
Co
m
p
l
e
m
e
n
t
a
r
y

C
D
F
Model
HYDRUS
GoldSim
Complementary CDF for n: GoldSim and HYDRUS
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 104

Figure 25. Re-creation of information in UDEQ’s Figure B-4 showing complementary
cumulative distribution functions (CDFs) for the 50 net infiltration values from the
HYDRUS realizations and 1000 values drawn from the distribution used in GoldSim
for the Clive DU PA Model v1.4.
UDEQ then argues “Furthermore, evapotranspiration rates in the model are likely too high, since
they do not account for accumulation of gravel at the surface over time, which would tend to
greatly diminish evaporation. A quantitative demonstration and explanation is needed to address
this issue.”
The accumulation of gravel at the surface at the Clive Site is highly unlikely. Recent field studies
(Neptune 2015b) provide evidence that the Site is within a region of significant eolian activity
(wind driven) evidenced by locally thick accumulation of gypsum dunes west and southwest of
the Site and a laterally continuous layer of suspension fallout silts preserved beneath the modern
surface throughout the Clive Site. The eolian deposits in the upper part of the stratigraphic
section represent a 10,000-year record of deposition and soil formation (Neptune 2015b).
Observations of Holocene eolian silt throughout the Clive Site support a conceptual model of
long-term eolian deposition on a stable surface that promotes and preserves concurrent eolian
deposits which are only slightly modified by slow processes of soil formation. See the response
to Interrogatory 05/2 for further discussion of eolian deposits and soil formation at the Site.
UDEQ requests clarification of the term “spatio-temporal scaling” used in reference to
distribution development for hydraulic properties. UDEQ suggests that “temporal scaling should
account explicitly for the temporal evolution of the distribution of hydraulic properties due to
pedogenic effects,” and “If temporal scaling has not been incorporated, then scale matching
should be described as spatial rather than spatio-temporal.”
0.00
0.25
0.50
0.75
1.00
0.00 0.05 0.10 0.15
Net Infiltration (mm/yr)
Co
m
p
l
e
m
e
n
t
a
r
y

C
D
F
Model
HYDRUS
GoldSim
CCDF for Net Infiltration: GoldSim and HYDRUS
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 105
UDEQ’s conceptual model of significant soil structural changes occurring during the
performance period of the Clive DU PA Model is inconsistent with field observations at the Site.
For a discussion of soil formation at the Site, see the response to Interrogatory 05/2.
As described in Section 1.1 and Appendix 14 of the DU PA Model v1.4 (Neptune 2015e), the
temporal scale represented in a distribution should align with the scale represented by a value
from the distribution in the PA Model. The PA Model takes a single draw from each of the
relevant hydraulic property distributions at the beginning of a model realization and typically
uses that value for the entire realization of the model (over all time represented by the model).
From a temporal perspective, the values should be judged to be reasonable and to be held
constant over long time periods, and therefore should not reflect extremes that are only realistic
at smaller time scales.
Thinking about plausible long-term averages for the values representing the spatial volumes of
the PA Model is the easiest way to judge reasonableness of the distribution relative to time.
However, this is challenging because it is clearly impossible to collect data over 100s or 1000s of
years into the future to capture potential soil forming effects when developing the distribution of
a long-term average. Therefore, temporal scaling is more abstract than the methods discussed for
spatial scaling, and actually uses spatial variability as a surrogate for variability that might be
observed over time. For example, using data from different locations and regions captures
conditions in different materials that may be observed at one location over time through soil
forming effects. Therefore, the heterogeneity over space is used as a surrogate for heterogeneity
over time for the volumes represented by the PA Model; this is necessary if distributions are to
be based on empirical information because it is not possible to collect the data needed to capture
the heterogeneity over time into the future. Therefore, while the scaling does appear to be simply
“spatial scaling” on the surface, it is not a misnomer to call it spatio-temporal scaling because
both scales are always simultaneously considered even if most of the information available is
spatial. The resulting distributions should be evaluated relative to both dimensions by asking
“does the distribution provide values that are plausible when applied to the spatial volumes of
PA cells over long periods of time?”
The “conventional spatial averaging procedures” referred to by UDEQ are useful when extensive
spatially referenced point-scale data are collected from the area of interest (the Site) and the goal
is to aggregate or “upscale” the point-scale data to reflect information at many larger grid cells
representing aggregated material over the same spatial area as covered by the point-scale data.
There are two main reasons this approach is not realistic for distribution development for the DU
PA Model: (1) the Rosetta data are not site-specific (i.e. not from Clive) and therefore are not
spatially referenced to the volumes of soil represented by cells of the PA Model, and (2) the
volumes and time periods represented by cells of the PA Model are very large relative to the
areas over which data are collected.
The Rosetta database represents a diverse collection of measurements from samples at different
locations and different points in time, with no spatial reference to the Clive Site. As described in
Section 1.1 and Appendix 14 of the DU PA Model v1.4 (Neptune 2015e), and in other responses
related to “scaling,” the goal is to capture values in the distributions that are deemed plausible
considering they must represent large volumes of soil and long periods of time. Measurements
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 106
from single locations included in the Rosetta database will capture extreme values that would not
be representative of aggregated measurements representing large volumes of material, and
therefore the standard deviation of the raw measurements should not be used as the standard
deviation of the final distribution for the PA Model. Instead, the distributional goals for the PA
Model require aggregation over potential values from single locations and times that could be
observed at the Site to develop a distribution reflecting plausible aggregated values given the
current state of knowledge.
Given the lack of site-specific data and spatially explicit information, a reasonable and simple
way to proceed is to base the distribution on the average over the values from a variety of
conditions found in the Rosetta database. The goal is then to develop a distribution capturing
plausible values of averages of the input variables to apply over the spatial areas and time
periods used in the PA Model. An elegant way to obtain such a distribution is use of fundamental
statistical results regarding the distribution of plausible averages that could have been obtained
under many different random samples of point-locations and times (the sampling distribution of
the average), and its estimated standard deviation (the standard error of the average). That is,
based on the available data, statistical theory can be used to develop a distribution describing
other averages that could have been obtained had sampling differed. The sampling distribution
can also be thought of as communicating the uncertainty in the estimated mean. Greater
variability in the raw data or less information in available data (e.g., smaller sample sizes) leads
to more uncertainty, and therefore a larger standard error of the average.
A reasonable distribution is a normal distribution with standard deviation equal to the standard
error of the average based on the Rosetta database, calculated as the sample standard deviation
divided by the square root of the sample size. This is not the only approach that could be used to
distribution development, but its simplicity is appropriate given the distributional goal and
available information. The approach essentially assumes that the available data represent a
collection of measurements from point locations randomly selected (in space and time) from the
Site over the next 10,000 years. The sampling locations and occasions are assumed to be far
enough apart that they can be assumed to be independent. Then, the sampling distribution of the
average reflects plausible spatio-temporal averages that are reasonable to apply over the Site and
over the length of time the model is run. For the Clive DU PA Model v1.4 distributions, the
distributions for α, n, and Ks are developed on the log scale where the underlying population
distribution of Rosetta values is more symmetric, further justifying the use of the normal
distribution to approximate the sampling distribution of the average.
2.22 SER B.8 Supplemental Interrogatory Comment 8
8) a) Demonstrate that the fitted equations for water content and infiltration (Appendix 5,
Equations 39 and 40, and Table 10) give “reasonable” results when compared to
HYDRUS.
b) For example, provide an explanation for why Ksat is insensitive to the infiltration rates.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 107
DEQ Critique from April 2015 SER, Appendix B:
This interrogatory asked for demonstration that Equations 39 and 40 provide realistic predictions
relative to the predictions from HYDRUS. EnergySolutions’ response provides a number of
graphs showing that the predictions in the Clive DU PA Model v1.2 using Equations 39 and 40
are similar to those from HYDRUS in the sense of the mean but exhibit less variability than the
predictions in HYDRUS. The reduced variability in the percolation predicted by Equation 39 is
attributed to the reduction in log-variance to address spatial averaging, but another plausible
explanation is that Equation 39 reflects central conditions adequately, but extreme conditions in
the tailings inadequately. A quantitative demonstration and explanation is needed to resolve this
issue.
This interrogatory also asked for an explanation of the lack of sensitivity of percolation rate to
Ksat. The response on pages 25 and 26 (un-numbered figures) shows that water is isolated in
the surface layer. However, using an unrealistically low Ksat for the Surface Layer and
Evaporative Zone Layer, in combination with likely capillary-barrier effect artifacts in the model
(see Comment 4), may have choked off infiltration into the cover profile and trapped water at the
surface, thereby limiting downward movement of water unrealistically and artificially impacting
the significance of Ksat of the radon barrier. A discussion of the HYDRUS predictions in the
context of cumulative water-balance quantities and expected norms for water-balance covers
could address this issue.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
verification of the HYDRUS results.
2.22.1 Interrogatory Response
UDEQ requests a comparison between the regression models in the GoldSim DU PA Model for
net infiltration and water content with the HYDRUS simulation results for these parameters. The
request refers to “Appendix 5, Equations 39 and 40, and Table 10.” Equations 39 and 40 are
found in Appendix 5 of the DU PA Model v1.2 (Neptune 2014). As described in the response to
UDEQ Comment B. 1, improved regression models were provided based on 50 HYDRUS
simulations in Appendix 5 of the DU PA Model v1.4 (Neptune 2015c). These models correspond
to Equations 41 and 42 in v1.4. The regression equation coefficients are provided in Table 10 of
both documents. This comparison of net infiltration between the regression models and
HYDRUS results is discussed in the response to Comment B.1.
A comparison between measurements of the water content of the Evaporative Zone Layer Unit 4
soil and water contents obtained from the regression model is discussed in the response to UDEQ
Comment B. 9.
UDEQ raises a concern about the sensitivity of modeled net infiltration to the value of the
saturated hydraulic conductivity. See the response to Interrogatory 176/1 for a discussion of the
sensitivity of the net infiltration at the Clive Site to the values of the saturated hydraulic
conductivity of the cover layers.
In this interrogatory UDEQ argues that the values of the saturated hydraulic conductivity
assigned to the Surface and ET Zone Layers are unrealistically low and that hydraulic properties
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 108
used create an artificial and unrealistic capillary barrier. See the response to UDEQ Comment
B.4 for a discussion of Surface and ET Zone Layer saturated hydraulic conductivities.
See the response to UDEQ Comment B.3 for a discussion of capillary barrier effects.
2.23 SER B.9 Supplemental Interrogatory Comment 9
9) Compare the moisture contents calculated using the fitted equations to the Bingham
(1991, Table 6 and/or Appendix B) Clive site measured Unit 4 moisture contents,
and rationalize any differences.
DEQ Critique from April 2015 SER, Appendix B:
The comparison with HYDRUS is remarkably good. However, the comparison with Equation 39
is not good. Equation 39 seems to predict q between 0.27 and 0.31 for nearly all cases, whereas
the data are over a much broader range.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
adequacy of the goodness of fit against the Bingham (1991) data.
2.23.1 Interrogatory Response
UDEQ requested a comparison between the regression model for water content in Appendix 5 of
DU PA v1.2 (Neptune 2014) with the water content data for Unit 4 material acquired by
Bingham Environmental (1991). Comparisons were made between the regression model for the
Evaporative Zone Layer and Unit 4 water contents listed in Table 6 of Bingham Environmental
(1991). The Surface Layer was not selected for this comparison because it has a reduced porosity
due to the gravel admixture.
Gravimetric water contents for Unit 4 soils, at depths less than or equal to 2 feet (near the depth
of the Evaporative Zone Layer (0.5 to 1.5 ft)), were taken from Bingham Environmental (1991),
(Table 6, pdf p. 42-43). Six values matched the depth constraint and those data are presented in
Table 8. Volumetric water contents for these six samples were calculated by multiplying the
gravimetric values by the bulk density of 1.397 g/cm3 reported on the Adobe pdf page 174 of
Bingham Environmental (1991) for sample GW19A-B1 (Unit 4 sample).
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 109
Table 8. Water Content Data from Table 6 of Bingham Environmental (1991).

GW-17A L-1 2 4 27.8 0.39
GW-19B L-1 2 4 17.5 0.24
SLC-203 NA 2 4 21.7 0.30
SLC-204 NA 2 4 15.3 0.21
SLC-205 NA 2 4 20.7 0.29
SLC-206 NA 2 4 19.6 0.27

Avg 20.43 0.285
Min 15.30 0.214
Max 27.80 0.388

Volumetric water contents calculated using the regression model were extracted from the
GoldSim DU PA Model v1.2 by adding a result element for the Expression
“WaterContentETCover_regr”. Then the model was run for 1,000 simulations to generate 1,000
values of water content for the Evaporative Zone Layer (Unit 4 soil).
Volumetric water contents from the regression model used in GoldSim (1000 replicates), from
HYDRUS (20 replicates), and from the six measured values from Table 8, are shown in Figure
26. For the x-axis, each of the 6 values in Table 8 were plotted at increments of ~ 167 in order to
show the data on the x-axis with 1,000 values (for the GoldSim results). Similarly, the
HYDRUS-1D values were plotted at increments of 50. As shown in Figure 26, the volumetric
water contents calculated with the fitted equation in GoldSim are well-bounded by the Bingham
data from Table 8. The mean volumetric water content value in Table 8 is 0.285 while the mean
from the GoldSim model 1,000 replicates is slightly higher at 0.294. The mean value of the 20
HYDRUS-1D replicates is 0.286, effectively identical to the Bingham Environmental (1991)
samples.
UDEQ states that, while the comparison of the sampled water contents with the HYDRUS
results is remarkably good, the comparison with Equation 39 (regression model) is not good.
They state that “Equation 39 seems to predict q (water content) between 0.27 and 0.31 for
nearly all cases, whereas the data are over a much broader range.” This observation on their
part is incorrect. Examination of the results from the regression model (labeled “GoldSim”) in
Figure 26 show water content values ranging from slightly greater than 0.1 to a maximum of
0.35. It is also important to keep in mind that the parameter input distributions to the regression
model have been upscaled to specify the model at the appropriate spatial scale of the Site and
the temporal scale of the model. Using proper upscaling the regression model output represents
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 110
uncertainty in the system’s average response rather than the variability of data collected at
points in time and space.

Figure 26. Comparison of Bingham Environmental (1991) water content data with water
content calculated using the regression equation for the DU PA GoldSim model and
with the results of the 20 HYDRUS simulations.
2.24 SER B.11 Supplemental Interrogatory Comment 11
DWMRC provided EnergySolutions with an Excel file, “Clive Hydrus Sensitivity Recommend
REV2.xlsx,” which contains suggested or proposed combinations of input values for the
HYDRUS runs used to support the Clive DU PA.
DEQ Critique from April 2015 SER, Appendix B:
DEQ requested a sensitivity analysis for a reasonable range of parameters to evaluate whether
the model responds within expected norms for a water-balance cover. This request has been
made in part because Appendix 5 provides inadequate documentation to demonstrate the
efficacy of the HYDRUS model and its realism relative to expected norms for a water-balance
cover.
Moreover, Appendix 5 indicates that predictions made by the model are insensitive to hydraulic
parameters (notably Ksat) generally known to have a strong influence on predictions made by
HYDRUS and similar models. For example, the unrealistically low Ksat for the Surface Layer and
Evaporative Zone Layer (see Comment 4) may have choked off infiltration into the cover profile
and trapped water at the surface, thereby limiting downward movement of water unrealistically
and artificially impacting the significance of Ksat of the radon barrier. As explained throughout
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 111
this document, there are significant concerns that the HYDRUS model may not be realistic and
may be biasing the analyses in the performance assessment. An assessment of the efficacy of
the HYDRUS model in the context of expected norms is essential to resolve this issue.
EnergySolutions’ response goes to great length to dismiss the requested sensitivity analysis as
not based on reasonable soil properties and as being inconsistent with a performance
assessment approach. The response justifies the criticism of the soil properties by citing
databases for soil properties unrelated to engineered earthen covers (e.g., the National
Resource Conservation Service database) or data reports known to contain measurements on
samples that are too small to represent in-service conditions and collected with antiquated
techniques that are known to cause disturbance of soil structure (e.g., the 1991 Bingham
Environmental report).
Despite these criticisms, the requested analyses apparently were conducted, but the output was
not included or presented comprehensively in the responses. The findings from these
simulations should be tabulated and reported, and water-balance graphs should be prepared and
discussed in the context of the mechanisms known to influence the hydrology of water-balance
covers. A thoughtful discussion would help justify the use of the HYDRUS model and build
confidence in the output.
DEQ Critique of DU PA v1.4, Appendix 21: See Interrogatory 21 for a description of the
adequacy of the range, distribution and bounds on the HYDRUS input data. The type of output
that should be provided is also presented.
2.24.1 Interrogatory Response
UDEQ requested responses to 10 supplemental interrogatories on August 11, 2014. Comment 11
in Appendix B of the SER (SC&A 2015a) contains Comment 10 of the 10 supplemental
interrogatories.
In this interrogatory, UDEQ provided a table with sets of flow model input parameters for the
cover layers. Single values were specified for the van Genuchten n parameter, the residual water
content, and the saturated water content. These same values were to be applied to all layers.
Values of the van Genuchten α parameter and the saturated hydraulic conductivity, Ks, were
given for each layer in categories labeled “low,” “typical,” and “high.” The values for α and Ks
for the radon barrier and the Frost Protection Layer were identical. UDEQ requested that
HYDRUS simulations be conducted using multiple combinations of parameter values for a total
of nine simulations.
For each simulation, UDEQ instructed the modelers to “run ‘warm up’ simulation 5 times back
to back beforehand using meteorological year having annual precipitation closest to long-term
average. Use heads from end of this 5 yr simulation as initial conditions for the performance
simulation.”
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 112
EnergySolutions expressed their concern with this approach but did fulfill UDEQ’s request to
run the requested simulations. EnergySolutions provided the following response
(EnergySolutions 2014).
In general, EnergySolutions strongly disagrees with the request of running highly
speculative, unsupported, one-off cases suggested in the subject request. This is not
consistent with the intent of the Utah regulation nor the meaning or application of a
“sensitivity analysis.” In practice, an appropriate sensitivity analysis would consider only
combinations of input values that are plausibly visible at the site under study. Whereas
the concept of plausibility in this context is applied based on available data and
professional judgment, the values that are suggested in the subject document (and
repeated above) are not plausible for this site.
EnergySolutions also disagrees with the intent given that the site will return to natural
conditions. In fact, the Division-suggested input values do not match natural conditions,
whereas the probability distributions used in the Clive DU PA Model provide reasonable
bounds for site conditions projected into the future given the available information and
data.
There are significant limitations in assessing the effects of parameter and conceptual
uncertainty using deterministic modeling with specified (discrete) cover designs and
bounding transport parameters and assumptions. Any more comprehensive sensitivity
analysis for the infiltration modeling should not be based on selective, unrepresentative,
and non-systematic changes in physical properties of cover materials.
In accordance with well-documented NRC guidance, the probabilistic approach models
future conditions by projecting current knowledge/conditions as reasonably as possible
while capturing uncertainty in the parameters or assumptions of the model. This is
distinctly differentiated from “conservative” (i.e., supposedly biased towards safety)
modeling that is occasionally seen, typically using point values for parameters (implying
a great deal of confidence; i.e., no uncertainty, or conditioning). This type of conservative
modeling is often termed “deterministic” modeling, and has often been used to support
compliance decisions. However, supposed conservatism in parameter estimates (or
distributions) is often difficult to judge in fully coupled models in which all transport
processes are contained in the same overall PA model. More importantly perhaps, actual
conservative dose results from PA models do not support the full capability of a disposal
facility, which leads to sub-optimal decisions for disposal of legacy waste and for the
nuclear industries that need a disposal option. Conservative, deterministic models may
have utility at a “screening” level, and they are often useful during probabilistic model
building, but they do not provide the full range of information that is necessary for
important decisions such as compliance or rule-making (cf., Bogen (1994), Cullen
(1994)).
Analysis of non-representative, arbitrarily selected one-off cases that are based on
unrealistic conditions easily lead to misinterpretation of the performance of the disposal
system.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 113
What is proposed by the Division is not a sensitivity analysis. Rather, the Division
proposes an analysis of separate implausible combinations of input parameter values.
Some details are provided below:
Saturated Hydraulic Conductivity (Ks)—Surface Layer
The surface layer in the ET cover functions as a store and release layer. Proposed values
for this layer are 86.4 cm/day for a low value, 864 cm/day for a typical value, and
864 cm/day for a high value. The typical and high values proposed exceed the values for
the Ks of sand provided in both the Rosetta (USDA 2017) and the Carsel and Parrish
(1988) databases (712.8 cm/day and 643 cm/day respectively) and are not appropriate for
characterizing a silty clay.
These values are also inconsistent with the measurements provided by Benson et al.
(2011) for store and release covers. Table 6.6 of Benson et al. (2011) contains geometric
mean values of measurements of in-service Ks for store and release covers at 10 sites.
The geometric mean values of Ks ranged from 0.65 cm/day to 45.79 cm/day with a
geometric mean of all sites of 7.5 cm/day. The proposed low value is an order of
magnitude larger and the typical and high values are more than two orders of magnitude
larger than the mean of the measured values.
The National Resources Conservation Service (NRCS) Web Soil Survey (WSS)
(http://websoilsurvey.sc.egov.usda.gov/App/HomePage.htm) provides online access to
Soil Reports containing soil property data. The most extensive surface soil type at the
Clive Site corresponding to Unit 4 is classified as the Skumpah by NRCS. NRCS assigns
Ks values for the upper 5 feet of the Skumpah ranging from 3.6 cm/day to 35 cm/day.
These Ks values represent natural conditions.
Again, these values are orders of magnitude lower than the proposed values.
α Values—Surface Layer
The α values recommended for the low, typical, and high cases for the Surface Layer are
0.3 1/kPa (0.03 1/cm). These values are too large when compared to the values of
0.00295 1/cm and 0.0012 1/cm measured by Bingham Environmental (1991) on two
cores taken from Unit 4 at the site.
All Hydraulic Model Parameters —Frost Protection Layer
All hydraulic parameter values for the Frost Protection Layer are set to the identical
values recommended for the radon barriers. These two materials are quite different, and
treating them as identical is unrealistic; even after naturalization the Frost Protection
Layer will not reach the conditions of the current radon barrier. This would artificially
induce more flow through the frost layer, but would not represent the naturalized system.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 114
Added Gravel
For the HYDRUS simulations a mean value of 0.48 for the porosity of the Unit 4 silty
clay used for the Surface Layer was obtained from the Rosetta database (USDA 2017).
The effect of the addition of 15 percent gravel to the Surface Layer on porosity was
calculated using ideal packing equations (Koltermann and Gorelick 1995), giving a value
of 0.41. If adding gravel and naturalizing the layer have compensating effects, then the
saturated water content should have remained 0.48. Their recommended value is 0.4,
nearly identical to what we used.
The influence of change in soil structure on saturated hydraulic conductivity of the radon
barriers was included in the Clive DU PA Model by sampling from a distribution of
saturated hydraulic conductivity developed from measurements of barrier layers in
service covers (Benson et al. 2011).
Warm-up Simulations
UDEQ included requests for “warm up” simulations. Specifically, the request was “For
each case above, run ‘warm up’ simulation 5 times back to back beforehand using
meterological year having annual precipitation closest to long-term average. Use heads
from end of this 5 yr simulation as initial conditions for the performance simulation.”
The 20 HYDRUS-1D simulations were conducted with, essentially, a 900-year warm-up
period, which is a considerably longer time period than the 5 average years requested by
the Division. Neptune used a 100-year synthetic record that was repeated 10 times to
generate a 1,000-year record of atmospheric boundary conditions. All 20 simulations
were run for 1,000 years but only the time series of average water content and infiltration
for the last 100 years were used as results. This was done because the initial conditions
for all simulations were set to a water potential of -200 cm, which is wetter than steady-
state conditions. The long simulation time allowed for equilibration to steady-state. So,
essentially, there is a 900-year warm up period. Figure 27 shows the time series of
infiltration through the ET cover and into the waste zone. It is apparent that, even after
900 years, the line is not quite flat, indicating that our infiltration estimates are slightly
over-estimated.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 115

Figure 27. Time series of infiltration into the waste zone for one of the 20 HYDRUS-
1D simulations.
Nevertheless, the nine HYDRUS-1D simulations requested by the Division were run and
results showing the range from minimum to maximum infiltration (into the waste zone),
along with the results from the original 20 HYDRUS-1D simulations, are shown in
Figure 27. Despite the implementation of the high Ks values requested by the Division,
infiltration in the new nine simulations is generally lower than for the original 20
HYDRUS-1D simulations. This is largely due to setting residual water content to zero,
which effectively increases the water holding capacity of each soil layer.
Overall, the Clive DU PA Model provides a reasonable range for the input parameters for
the hydraulic properties given the currently available data and information, and the
HYDRUS runs for the nine additional combinations of single values for inputs adds no
further insight.

Additional issues raised in Supplemental Interrogatory Comment B.11 are: observed insensitivity
of net infiltration to the value of the saturated hydraulic conductivity, a capillary barrier formed
by the upper cover layers, criticism of the NRCS soil properties data base, and criticism of
sampling and testing methods used by the CSU Porous Media laboratory.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 116
Sensitivity of Net Infiltration to Saturated Hydraulic Conductivity
See the response to Interrogatory 176/1.
Capillary Barrier Effects
See the response to UDEQ Comment B.3.
Antiquated Techniques for Sampling and Analysis of Core
See the response to Interrogatory 21/2.
3.0 Conclusion
Many issues relating to ET cover design and modeling are interwoven throughout the UDEQ
interrogatories discussed in this document. The following summarizes critical aspects of these
responses:
Geologic and Depositional Setting
• Observations of eolian (wind carried) silt deposited in the last 10,000 years across the
Clive Site support a conceptual model of long-term eolian deposition on a stable surface
that promotes and preserves concurrent eolian deposits that are only slightly modified by
slow processes of soil formation. Based on these observations, modeling substantial
cover material hydraulic property changes within 5 to 10 years after construction as
presented in Benson et al. (2011) and in Appendix E of SC&A (2015a) is not appropriate
for the Clive Site.
• UDEQ attributes changes in hydraulic properties of a cover material to volume changes
of the material. Smectite minerals responsible for shrink/swell behavior in clays
attributed to wetting and drying cycles are absent from the Unit 4 clay soils at the Site.
• Sites listed by UDEQ as being comparable to Clive are the Monticello Mill Tailings
Repository south of the town of Monticello, Utah (290 miles from Clive), the Blue Water
disposal site near Grants, New Mexico (485 miles from Clive), the Cheney disposal site
near Grand Junction Colorado (270 miles from Clive), and the Apple Valley Alternative
Cover Assessment Program (ACAP) site referenced by Benson et al. (2011). These sites
are not comparable for a number of reasons including different cover types, different
soils, and different ecology.
Disruptive Processes
• Based on data collection and analysis and the proposed cover design, SWCA evaluated
the potential for disturbance by plant roots, mammals, and ants to result in increased
infiltration (SWCA 2013) and determined that soil disturbance and increased infiltration
due to biotic activity would be minute. Their evaluation does not support UDEQ’s
assertion that biointrusion of plants and animals will substantially increase infiltration.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 117
• SWCA (2013) conclude from their observations at the Clive Site and ecological analog
sites that the low fertility and alkalinity of the soils and the aridity of the climate limit the
growth of plants at the Site and would inhibit the development of large deep-rooted plants
in the cover. The Clive and Monticello Sites are not comparable with respect to rooting
depths.
• The frost depths calculated as part of this analysis show results that are consistent with
the depths of cover and frost protection proposed in the ET Cover design. Disruption of
the cover due to freeze/thaw is not likely.
Flow Model Input Parameter Distribution Development
• Probability distributions used in a PA model incorporate uncertainty in the mean rather
than variability among values from different point locations and times. The distribution
should represent plausible means for the entire volume of material over the time frame of
the model, rather than variability among measurements taken from individual locations
within a site. Practically, this often translates into using standard errors for the estimated
mean over a larger scale to define distributions, rather than standard deviation of
available small-scale measurements. The relevant question to ask is “Does the
distribution cover all plausible values of the input variable that could be representative of
the large volume of material represented by the PA Model cell over long periods of
time?”
• Parameter values and parameter statistical distributions for the hydraulic properties of the
cover layers were developed based on site-specific data, engineering specifications,
widely used soil hydraulic property databases, and consideration of the function of the
layer.
• Estimates of the uncertainty in the hydraulic function α and n values of Unit 4 materials
used for the surface and evaporative zone layers of the ET cover were obtained by using
α and n values from distributions (mean and standard deviation) for each parameter from
the Rosetta database of hydraulic parameters for the textural class of silty clay.
• The radon barrier saturated hydraulic conductivity distribution was based on 1st, 50th, and
99th target percentiles elicited from Benson et al. (2011) to capture plausible values for
in-service (degraded) barriers that could apply to the entire cover described by the
distribution. To manage computational burden, deterministic values for θr, θs, α, and n
from previous site-specific modeling were used for the radon barriers.
• Distributions were developed using the open-source statistical software package R (R
Core Team 2017), which functions to facilitate fitting of distributions to target quantiles,
or percentiles.
• The net infiltration through the clay liner in the flow model is identical to the value of the
net infiltration at the bottom of the radon barriers. This value is drawn from the net
infiltration distribution which is based on measurements of in-service barriers from
Benson et al (2011). Thus, the modeling approach does take into account clay liner
degradation.
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 118
Model Abstraction—Development of Simplified Models for GoldSim
• The goal of “model abstraction” in the context of PA modeling is to build a relatively
simple statistical model to approximate the relationship between the input parameters and
an output of interest obtained from a sophisticated, and usually computationally
intensive, process-based model. Using the flow modeling as an example, the goal of the
statistical effort is to build a statistical model to capture the main structure in the
relationship between input parameters and results produced by HYDRUS (needed as an
input to the GoldSim model). The statistical model is then used to predict HYDRUS-like
outputs within GoldSim.
• Fifty HYDRUS-1D flow simulations were conducted to evaluate the uncertainty in net
infiltration into the waste zone and water content within each ET cover layer as a
function of hydraulic property uncertainty. Model input parameter distributions and
HYDRUS flow model results were fit with multiple linear regression models to develop
simplified models for the GoldSim PA model.
• Comparisons showed consistency between α and n inputs to the HYDRUS model and α
and n generated from the distributions for these parameters used in the GoldSim model.
Similarly, the net infiltration values from the HYDRUS realizations were consistent with
values generated using the regression equation in the GoldSim model.
• A linear regression model was chosen to represent net infiltration for implementation in
the DU PA Model because it was simpler and matched the model form also used for
water content across layers.
• The saturated hydraulic conductivity, Ks, was not included in the regression equation for
net infiltration because it was found not to be a predictor (that is not close to statistical
significance).
• Contrary to the opinion of UDEQ, there is not evidence of a strong linear relationship
between α and Ks that should be incorporated into PA distributions, particularly for the
Unit 4 soil within the silty clay soil textural class found at the Clive Site. These
parameters are demonstrated to be statistically independent, implying they are
theoretically uncorrelated and distributions can be developed separately for each
parameter.
Other Issues
• The hydraulic property values and statistical distributions assigned to the Evaporative
Zone and the Frost Protection Layers are reasonable estimates based on site-specific
information and commonly used soils databases. Given these property assignments, the
Frost Protection Layer behaves hydraulically by enhancing storage in the Evaporative
Zone in a realistic manner, not as described by UDEQ as “an extremely potent artificial,
non-realistic capillary barrier.”
• Steady-state annual averages of net infiltration and water content from the HYDRUS
simulations are the model results used to develop statistical distributions of these
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 119
parameters for inputs to the GoldSim model for the Clive DU PA. Daily water balance is
not the appropriate scale to evaluate a performance assessment model.
• Common, accepted methods were used for developing atmospheric boundary conditions
for the HYDRUS flow model.
• Given the much greater capacity of the clay liner to allow water to flow through it in
comparison to the 99th percentile of net infiltration rates, the bathtub effect is not
possible.

4.0 References
Albrecht, B.A., and C.H. Benson, 2001. Effect of Desiccation on Compacted Natural Clays,
Journal of Geotechnical and Geoenvironmental Engineering 127 (1) 67–75
Anderson, M.D., 2004. Sarcobatus vermiculatus. In: Fire Effects Information System [Online],
United States Department of Agriculture (USDA), Forest Service, Rocky Mountain
Research Station, Fire Sciences Laboratory, accessed 2/11/2017, available from
https://www.feis-crs.org/feis/
Belnap, J., et al., 2001. Biological Soil Crusts: Ecology and Management, Technical Reference
1730-2, United States Department of the Interior, Bureau of Land Management, Denver
CO, 2001
Benson, C.H., et al., 2011. Engineered Covers for Waste Containment: Changes in Engineering
Properties and Implications for Long-Term Performance Assessment, NUREG/CR-7028,
United States Nuclear Regulatory Commission, Washington DC, December 2011
Benson, C.H., and T. Gurdal, 2013. Hydrologic Properties of Final Cover Soils. In Foundation
Engineering in the Face of Uncertainty, Honoring Fred H. Kulhawy, Geotechnical
Special Publication No. 229, edited by J.L. Withiam, et al., pp. 283–297, American
Society of Civil Engineers (ASCE), Reston VA
Bingham Environmental, 1991. Hydrogeologic Report, Envirocare Waste Disposal Facility,
South Clive, Utah, prepared for Envirocare of Utah, Bingham Environmental, Salt Lake
City UT, October 1991
Bingham Environmental, 1994. Hydrogeologic Report, Mixed Waste Disposal Area, Envirocare
Waste Disposal Facility, South Clive, Utah, prepared for Envirocare of Utah, Bingham
Environmental, Salt Lake City UT, November 1994
Bingham Environmental, 1996. Revised Hydrogeologic Report, prepared for Envirocare of Utah
Inc., Bingham Environmental Inc., Salt Lake City UT, February 1996
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 120
Black, B.D., et al., 1999. Geology and Geologic Hazards of Tooele Valley and the West Desert
Hazardous Industry Area, Tooele County, Utah, Special Study 96, Utah Geological
Survey, Utah Department of Natural Resources, Salt Lake City UT, 1999
Bogen, K.T., 1994. A Note on Compounded Conservatism, Risk Analysis 14 (4) 379–381
Bowker, M.A., 2007. Biological Soil Crust Rehabilitation in Theory and Practice: An
Underexploited Opportunity, Restoration Ecology 15 (1) 13–23
Carsel, R.F., and R.S. Parrish, 1988. Developing Joint Probability Distributions of Soil Water
Retention Characteristics, Water Resources Research 24 (5) 755–769
Cressie, N., and C.K. Wikle, 2011. Statistics for Spatio-temporal Data, Wiley Series in
Probability and Statistics, John Wiley & Sons Inc., Hoboken NJ
Crowe, B., et al., 2017. Representation of Global Climate Change in Performance Assessment
Models for Disposal of Radioactive Waste—17183, proceedings of the WM2017
Conference, March 5–9, Phoenix AZ, 2017
Cullen, A.C., 1994. Measures of Compounding Conservatism in Probabilistic Risk Assessment,
Risk Analysis 14 (4) 389–393
Departments of the Army and the Air Force, 1988. Arctic and Subarctic Construction
Calculation Methods for Determination of Depths of Freeze and Thaw in Soils, Technical
Manual No. 5-852-6, Air Force Regulation AFR 88-19, Volume 6, Departments of the
Army and the Air Force, Washington DC, January 1988
Diodato, D.M., 2000. Computer Spotlights in Ground Water (38(1) 10–11), PC-Progress,
available from https://www.pc-progress.com/en/Default.aspx?h3d-reviews-2000
DOE, 2017. UMTRCA Title II, Bluewater, New Mexico, Disposal Site Fact Sheet, United States
Department of Energy, Office of Legacy Management, Grand Junction CO, 2017
DWR, 2014. Water Right Records and Well Log Database, Utah Division of Water Rights,
accessed 3/18/2014, available from http://waterrights.utah.gov/wrinfo/query.asp
EnergySolutions, 2012. Utah Low-Level Radioactive Material License (RML UT2300249)
Updated Site-Specific Performance Assessment, prepared for Utah Division of Radiation
Control, EnergySolutions LLC, Salt Lake City UT, October 2012
EnergySolutions, 2014. Responses to August 11, 2014—Supplemental Interrogatories, Utah
LLRW Disposal License RML UT 2300249 Condition 35 Compliance Report, prepared
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 121
for Utah Division of Radiation Control, EnergySolutions LLC, Salt Lake City UT,
August 2014
EnergySolutions, 2015. Utah Radioactive Material License (RML UT2300249) Updated Site-
Specific Performance Assessment, Revision 2, prepared for Utah Division of Radiation
Control, EnergySolutions LLC, Salt Lake City UT, January 2015
Envirocare, 2000. Application for License Amendment for Class B & C Waste, Envirocare of
Utah Inc., North Salt Lake UT, December 2000
Envirocare, 2004. Revised Hydrogeologic Report for the Envirocare Waste Disposal Facility,
Clive, Utah, Version 2.0, Envirocare of Utah Inc., August 2004
EPA, 2002. Simulating Radionuclide Fate and Transport in the Unsaturated Zone: Evaluation
and Sensitivity Analyses of Select Computer Models, EPA/600/R-02/082, United States
Environmental Protection Agency, National Risk Management Research Laboratory,
Cincinnati OH, July 2002
Gates, J.S., and S.A. Kruer, 1981. Hydrologic Reconnaissance of the Southern Great Salt Lake
Desert and Summary of the Hydrology of West-Central Utah, Technical Publication No.
71, prepared by the United States Geological Survey, State of Utah Department of
Natural Resources, Salt Lake City UT, 1981
Groeneveld, D.P., 1989. Shrub Rooting and Water Acquisition on Threatened Shallow
Groundwater Habitats in the Owens Valley, California, Symposium on Cheatgrass
Invasion, Shrub Die-off, and Other Aspects of Shrub Biology and Management, 1989, Las
Vegas NV
GTG, 2014. GoldSim Contaminant Transport Module User's Guide, GoldSim Technology
Group, Issaquah WA, May 2014
Handy, R.L., and M.G. Spangler, 2007. Geotechnical Engineering: Soil and Foundation
Principles and Practice, Fifth Edition, McGraw-Hill Companies, New York NY
Hastie, T., et al., 2009. The Elements of Statistical Learning: Data Mining, Inference, and
Prediction, Second Edition, Springer-Verlag, New York NY
Higgs, M., et al., 2017. Statistical Methods for Effective Spatial and Temporal Scaling in
Support of Probabilistic Performance Assessments—17340, proceedings of the WM2017
Conference, March 5–9, Phoenix AZ, 2017
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 122
ITRC, 2003. Technical and Regulatory Guidance for Design, Installation, and Monitoring of
Alternative Final Landfill Covers, The Interstate Technology & Regulatory Council,
Alternative Landfill Technologies Team, Washington DC, December 2003
Jury, W., and R. Horton, 2004. Soil Physics, Sixth Edition, John Wiley & Sons Inc., Hoboken NJ
Koltermann, C.E., and S.M. Gorelick, 1995. Fractional Packing Model for Hydraulic
Conductivity Derived from Sediment Mixtures, Water Resources Research 31 (12) 3283–
3297
Kutner, M.H., et al., 2005. Applied Linear Statistical Models, Fifth Edition, McGraw-Hill/Irwin,
New York NY
Levitt, D.G., 2011. Modeling the Movement of Transient Moisture through Disposal Units at Los
Alamos National Laboratory Technical Area 54, Area G, LA-UR-11-05424, prepared for
United States Department of Energy, Los Alamos National Laboratory, Los Alamos NM,
September 2011
McCray, J.E., 2007. Review of HYDRUS-2D/3D, Southwest Hydrology January/February
(2007) p. 41
Meyer, P.D., et al., 1996. Hydrologic Evaluation Methodology for Estimating Water Movement
Through the Unsaturated Zone at Commercial Low-Level Radioactive Waste Disposal
Sites, NUREG/CR-6346, PNL-10843, prepared for United States Nuclear Regulatory
Commission, Pacific Northwest Laboratory, Richland WA, January 1996
Morris, C.E., and J.C. Stormont, 1999. Parametric Study of Unsaturated Drainage Layers in a
Capillary Barrier, Journal of Geotechnical and Geoenvironmental Engineering 125 (12)
1057–1065
MSI, 2012. January 2011 through December 2011 and January 1993 through December 2011
Summary Report of Meteorological Data Collected at EnergySolutions' Clive, Utah
Facility, prepared for EnergySolutions Inc., Meteorological Solutions Inc., Salt Lake City
UT, February 2012
Mualem, Y., 1976. A New Model for Predicting the Hydraulic Conductivity of Unsaturated
Porous Media, Water Resources Research 12 (3) 513–522
Neitsch, S.L., et al., 2011. Soil and Water Assessment Tool Theoretical Documentation: Version
2009, Texas Water Resources Institute Technical Report No. 406, Texas A&M University
System, accessed 11/2/2017, available from http://swat.tamu.edu/media/99192/swat2009-
theory.pdf
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 123
Neptune, 2014. Unsaturated Zone Modeling for the Clive PA, Clive DU PA Model v1.2, NAC-
0015_R1, prepared for EnergySolutions, Neptune and Company Inc., Los Alamos NM,
June 2014
Neptune, 2015a. Saturated Zone Modeling for the Clive DU PA, Clive DU PA Model v1.4, NAC-
0016_R4, Neptune and Company Inc., Los Alamos NM, October 2015
Neptune, 2015b. Neptune Field Studies, December, 2014, Eolian Depositional History Clive
Disposal Site, NAC-0044_R0, Neptune and Company Inc., Los Alamos NM, March 2015
Neptune, 2015c. Unsaturated Zone Modeling for the Clive PA, Clive DU PA Model v1.4, NAC-
0015_R4, prepared for EnergySolutions, Neptune and Company Inc., Los Alamos NM,
October 2015
Neptune, 2015d. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015e. Fitting Probability Distributions, Clive DU PA Model v1.4, NAC-0031_R1,
Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015f. Safety Evaluation Report Response, NAC-0053_R0, prepared for
EnergySolutions, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015g. Embankment Modeling for the Clive DU PA, Clive DU PA Model v1.4, NAC-
0019_R4, Neptune and Company Inc., Los Alamos NM, October 2015
Neuman, S.P., et al., 2003. A Comprehensive Strategy of Hydrogeologic Modeling and
Uncertainty Analysis for Nuclear Facilities and Sites, NUREG/CR-6805, United States
Nuclear Regulatory Commission, Washington DC, July 2003
NRCS, 2016. Web Soil Survey (WSS), Natural Resources Conservation Service, accessed
2/29/16, available from http://websoilsurvey.sc.egov.usda.gov/App/WebSoilSurvey.aspx
Pavek, D.S., 1992. Halogeton glomeratus. In: Fire Effects Information System, [Online], United
States Department of Agriculture (USDA), Forest Service, Rocky Mountain Research
Station, Fire Sciences Laboratory, accessed 2/11/17, available from https://www.feis-
crs.org/feis/
Peace, J.L., et al., 2004. Vegetation Study in Support of the Design and Optimization of
Vegetative Soil Covers, Sandia National Laboratories, Albuquerque, New Mexico,
SAND2004-6144, Sandia National Laboratories, Albuquerque NM, November 2004
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 124
R Core Team, 2015. R: A Language and Environment for Statistical Computing, R Foundation
for Statistical Consulting, available from https://www.r-project.org/
R Core Team, 2017. R: A Language and Environment for Statistical Computing, R Foundation
for Statistical Computing, Vienna, Austria
Ramsey, F.L., and D.W. Schafer, 2013. The Statistical Sleuth: A Course in Methods of Data
Analysis, Third Edition, Brooks/Cole, Cengage Learning, available from
https://www.cengage.com/c/the-statistical-sleuth-a-course-in-methods-of-data-analysis-
3e-ramsey
Rawlings, J.O., et al., 1998. Applied Regression Analysis: A Research Tool, Second Edition,
Springer-Verlag, New York NY
Richardson, C.W., and D.A. Wright, 1984. WGEN: A Model for Generating Daily Weather
Variables, United States Department of Agriculture, Agricultural Research Service,
Washington DC, August 1984
SC&A, 2015a. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
2, SC&A, Vienna VA, April 2015
SC&A, 2015b. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
1, SC&A Inc., Vienna VA, April 2015
Scanlon, B.R., et al., 2002. Intercode Comparisons for Simulating Water Balance of Surficial
Sediments in Semiarid Regions, Water Resources Research 38 (12) 59-1–59-16 doi:
10.1029/2001WR001233
Scanlon, B.R., et al., 2005. Evaluation of Evapotranspirative Covers for Waste Containment in
Arid and Semiarid Regions in the Southwestern USA, Vadose Zone Journal 4 (2005) 55–
71
Schaap, M.G., et al., 2001. ROSETTA: A Computer Program for Estimating Soil Hydraulic
Parameters with Hierarchical Pedotransfer Functions, Journal of Hydrology 251 (2001)
163–176
Schroeder, P.R., et al., 1994a. The Hydrologic Evaluation of Landfill Performance (HELP)
Model, Engineering Documentation for Version 3, EPA/600/R-94/168b, United States
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 125
Environmental Protection Agency, Office of Research and Development, Washington
DC, September 1994
Schroeder, P.R., et al., 1994b. The Hydrologic Evaluation of Landfill Performance (HELP)
Model, User's Guide for Version 3, EPA/600/R-94/168a, United States Environmental
Protection Agency, Office of Research and Development, Washington DC, September
1994
Shabala, S., 2013. Learning from Halophytes: Physiological Basis and Strategies to Improve
Abiotic Stress Tolerance in Crops, Annals of Botany 112 (2013) 1209–1221 doi:
10.1093/aob/mct205
Shmueli, G., 2010. To Explain or to Predict?, Statistical Science 25 (3) 289–310 doi:
10.1214/10-STS330
Simonin, K.A., 2001. Atriplex confertifolia. In: Fire Effects Information System, [Online],
United States Department of Agriculture (USDA), Forest Service, Rocky Mountain
Research Station, Fire Sciences Laboratory, accessed 2/11/2017, available from
https://www.feis-crs.org/feis/
Šimůnek, J., et al., 2013. The HYDRUS-1D Software Package for Simulating the One-
Dimensional Movement of Water, Heat, and Multiple Solutes in Variably-Saturated
Media, Version 4.16, University of California Riverside, Riverside CA, March 2013
SWCA, 2011. Field Sampling of Biotic Turbation of Soils at the Clive Site, Tooele County, Utah,
prepared for EnergySolutions, SWCA Environmental Consultants, Salt Lake City UT,
January 2011
SWCA, 2012. Vegetated Cover System for the EnergySolutions Clive Site: Literature Review,
Evaluation of Existing Data, and Field Studies Summary Report, prepared for
EnergySolutions, SWCA Environmental Consultants, Salt Lake City UT, August 2012
SWCA, 2013. EnergySolutions Updated Performance Assessment—SWCA’s Response to First
Round DRC Interrogatories, SWCA Environmental Consultants, September 2013
USDA, 2017. ROSETTA Class Average Hydraulic Parameters, United States Department of
Agriculture, Agricultural Research Service, available from
https://www.ars.usda.gov/pacific-west-area/riverside-ca/us-salinity-
laboratory/docs/rosetta-class-average-hydraulic-parameters
Utah DEQ, 2017. Division of Waste Management and Radiation Control, EnergySolutions Clive
LLRW Disposal Facility License No: UT2300249; RML #UT 2300249, Amended and
ET Cover Design Responses for the Clive DU PA Model
23 Feb 2018 126
New Interrogatories Related to Clive DU PA Modeling Report Version 1.4 Dated
November 2015, Utah Department of Environmental Quality (DEQ), Salt Lake City UT,
May 2017
van Genuchten, M.T., 1980. A Closed-form Equation for Predicting the Hydraulic Conductivity
of Unsaturated Soils, Soil Sci. Soc. Am. J. 44 (1980) 892–898
van Genuchten, M.T., et al., 1991. The RETC Code for Quantifying the Hydraulic Functions of
Unsaturated Soils, EPA/600/2-91/065, United States Environmental Protection Agency,
Ada OK, December 1991
Wagner, C.H., 1982. Simpson’s Paradox in Real Life, The American Statistician 36 (1) 46–48
Waugh, W.J., et al., 2008. Monitoring the Performance of an Alternative Landfill Cover at the
Monticello, Utah, Uranium Mill Tailings Disposal Site, Waste Management 2008
Conference, Phoenix AZ
Whetstone Associates, 2011. EnergySolutions: Class A West Disposal Cell Infiltration and
Transport Modeling Report, prepared for EnergySolutions, Whetstone Associates Inc.,
Gunnison CO, November 2011

NAC-0108_R0

Erosion Responses for the Clive DU
PA Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Erosion Responses for the Clive DU PA Model
23 Feb 2018 ii
1. Title: Erosion Responses for the Clive DU PA Model
2. Filename: Erosion Responses for the Clive DU PA Model.docx
3. Description: Responses to UDEQ Interrogatories and Safety Evaluation Report Comments
received May 11, 2017

Name Date
4. Originator Mike Sully and Sean McCandless 12 Feb 2018
5. Reviewer Dan Levitt
6. Remarks

Erosion Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
FIGURES .................................................................................................................................. iv
TABLES ..................................................................................................................................... v
ACRONYMS AND ABBREVIATIONS ................................................................................... vi
1.0 Overview and Conceptual Model........................................................................................ 7
2.0 UDEQ Interrogatory Responses.......................................................................................... 8
2.1 Interrogatory CR R313-25-8(4)(a)-71/1: Biotic Processes in Gully Formation .............. 8
2.1.1 Interrogatory Response ............................................................................................ 8
2.2 Interrogatory CR R313-25-7(2)-191/3: Effect of Gully Erosion .................................. 13
2.2.1 Interrogatory Response .......................................................................................... 14
2.3 Interrogatory CR R313-25-25(4)-197/1: Properties of Embankment Side Slope
Materials..................................................................................................................... 20
2.3.1 Interrogatory Response .......................................................................................... 21
2.4 Interrogatory CR R313-25-25(4)-198/1: Gravel Content of Embankment Materials .... 21
2.4.1 Interrogatory Response .......................................................................................... 21
2.5 Interrogatory CR R313-25-25(4)-199/1: Uncertainties in Erosion Modeling ............... 25
2.5.1 Interrogatory Response .......................................................................................... 25
2.6 Interrogatory CR R313-25-25(4)-200/1: Use of RHEM to Develop Parameters for
SIBERIA .................................................................................................................... 25
2.6.1 Interrogatory Response .......................................................................................... 26
2.7 Interrogatory CR R313-25-25(4)-201/1: Estimating Rainfall Intensity ........................ 27
2.7.1 Interrogatory Response .......................................................................................... 27
2.8 Interrogatory CR R313-25-25(4)-202/1: Use of SIBERIA to Model Federal Cell
Erosion ....................................................................................................................... 30
2.8.1 Interrogatory Response .......................................................................................... 30
2.9 Interrogatory CR R313-25-25(4)-205/1: Erosion Analysis .......................................... 30
2.9.1 Interrogatory Response .......................................................................................... 31
3.0 Conclusion ....................................................................................................................... 34
4.0 References ........................................................................................................................ 35
Appendix A. HAL 2018 ............................................................................................................ 40
DRC Interrogatory CR R313-25-7(2)-191/3 ................................................................ 41
Appendix B. HAL 2017a........................................................................................................... 47
DRC Interrogatory CR R313-25-25(4)-198/1 .............................................................. 48
Appendix C. HAL 2017b .......................................................................................................... 53
DRC Interrogatory CR R313-25-25(4)-201/1 .............................................................. 54

Erosion Responses for the Clive DU PA Model
23 Feb 2018 iv
FIGURES
Figure 1. Utah-specific Iso-Erodent (R) mapping provided in the Utah Water Research
Laboratory report (Israelsen et al. 1984). .................................................................. 16
Figure 2. Wischmeier nomograph (Haan et al. 1994; Israelsen et al. 1984). ............................... 17
Figure 3. Table 9 and Table 10 of the US Department of Agriculture Handbook Number 537,
“Predicting Rainfall Erosion Losses, A Guide to Conservation Planning” (USDA
1978). ...................................................................................................................... 18
Figure 4. Empirical cumulative distribution functions (ecdf) of the values across realizations
found in the GoldSim look-up table for the seven deepest depth intervals. Note
that, in general, the deeper intervals have larger proportions of realizations with
smaller fractions. The two green curves associated with depth intervals (1.37m,
1.52m] and (1.52m, 2.03m] do not follow the order with depth, which can be
interpreted as consistent with the conclusion from Figure 205-1............................... 32

Erosion Responses for the Clive DU PA Model
23 Feb 2018 v
TABLES
Table 1. Gully erosion potential—permissible velocity analysis for Federal Cell. ...................... 12
Table 2. Effects of erosion—average soil loss analysis using USLE. Soil loss is calculated for
the top slope with Unit 4 clay with 15% gravel admixture and side slope with Unit
4 clay and 50% gravel admixture. ............................................................................ 18
Table 3. Calculated Time of Concentration Using TR-55 Method for Class A West Cell. .......... 28
Table 4. Calculated Time of Concentration Using Tr-55 Method for Federal Cell. .................... 28
Table 5. PMP Rainfall Intensity for the Class A West Cell. ....................................................... 29
Table 6. PMP Rainfall Intensity for the Federal Cell. ................................................................ 29
Table 7. Gully Erosion Potential—Velocity Analysis for Class A West Cell. ............................ 30

Erosion Responses for the Clive DU PA Model
23 Feb 2018 vi
ACRONYMS AND ABBREVIATIONS
CERCLA Comprehensive Environmental Response, Compensation, and Liability Act
CWCB Colorado Water Conservation Board
DEQ (Utah) Department of Environmental Quality
DOE (United States) Department of Energy
DU depleted uranium
DWMRC Division of Waste Management and Radiation Control
ecdf empirical cumulative distribution function
EPA (United States) Environmental Protection Agency
ET evapotranspiration
HAL Hansen, Allen, and Luce
HMR Hydrometeorological Report
LEM landscape evolution model
LLRW low-level radioactive waste
MPV maximum possible velocity
NOAA National Oceanographic and Atmospheric Administration
NRC (United States) Nuclear Regulatory Commission
NRCS Natural Resources Conservation Service
PA performance assessment
PDF probability density function
PMP probable maximum precipitation
PPA probabilistic performance assessment
PSB Prototype Surface Barrier
RCRA Resource Conservation and Recovery Act
RHEM Rangeland Hydrology and Erosion Model
RML Radioactive Material License
RUSLE Revised Universal Soil Loss Equation
sd standard deviation
SER Safety Evaluation Report
SWCA SWCA Environmental Consultants
SWCC soil water characteristic curve
UDEQ Utah Department of Environmental Quality
UMTRCA Uranium Mill Tailings Radiation Control Act
USDA United States Department of Agriculture
USLE Universal Soil Loss Equation

Erosion Responses for the Clive DU PA Model
23 Feb 2018 7
This document addresses Category 2, Erosion of the DU PA Version 1.4 Interrogatories (Utah
DEQ 2017). There are two open interrogatories and seven new interrogatories within this
category. Background information on the statistical approaches applied for the probabilistic PA
model is provided in Section 1.1 of NAC_0108, ET Cover Design. Section 1.2 of that document
also provides a summary of the cell design, climate, and vegetation.
1.0 Overview and Conceptual Model
There are three water erosion processes considered for the Site: sheet erosion, rill erosion, and
gully erosion (channel formation). Sheet erosion is the detachment of soil particles by water
flowing overland as a sheet. During rainfall events when rainfall exceeds infiltration, runoff can
occur, acting to erode cover materials. Sheet erosion is a process that extends over the entire area
of the cover. If flow concentrates on the surface, shallow channels called rills can also be
formed. Gully erosion is a process that occurs when flowing water forms larger narrow channels.
Gully erosion typically results in a gully that has an approximate “V” cross section that widens
and deepens with time until the gully stabilizes. The rate of erosion loss from all of these
processes depends on the steepness and length of the slope, soil texture, vegetation, and cover
surface characteristics, as well as rainfall intensity.
The stability of the cover with respect to erosion is evaluated for the Clive DU PA using several
modeling approaches. The Universal Soil Loss Equation (USLE) (USDA 1978) is used for
estimating losses due to sheet and rill erosion, and the permissible velocity method described in
NUREG-1623 (NRC 2002) is used for evaluating the potential for gully erosion. The USLE
provides estimates of long-term annual soil loss produced by rain drop impact and runoff. This
method is recommended by the EPA for evaluation of cover stability (EPA 1991) and is a
standard used across the United States. The NRC has recommended several methods for
evaluating the potential for gully erosion on soil covers for waste embankments (NRC 2002).
Modeling of sheet and rill erosion losses from the cover was done with USLE. The USLE uses a
conceptual model to evaluate stability that is described by Israelsen et al. (1984) as “if adequate
protection is provided to control sheet erosion, then rills and gullies will never form from
rainfall.” The USLE procedure is based on the Universal Soil Loss Equation—a combination of
parameters representing rainfall intensity, soil erodibility, slope length, slope steepness, cover
and management factors, and conservation practices. Values for the parameters are obtained
from tables and charts (or the computer program) applicable to the location, site-specific
materials, and cover design.
The annual erosion loss rates estimated by USLE for the cover were corroborated using the
Rangeland Hydrology Erosion Model (RHEM) (Nearing et al. 2011). RHEM is a process-based
runoff and erosion model developed specifically for rangeland applications.
The evaluation of design elements for long-term cover stability by the NRC (NRC 2002) is based
on the conceptual model that gully erosion differs from other design considerations in that the
development of gullies and the process of erosion are cumulative and can progress with
subsequent storm events. The method adopted by the NRC to provide long-term stability is to
design stable slopes that prevent the initiation of gullies by a single, very large storm event. By
Erosion Responses for the Clive DU PA Model
23 Feb 2018 8
designing to an unusually large single event, smaller more frequent events will have no
significant cumulative impact on stability.
NUREG-1623 (NRC 2002) details methods, guidelines, and procedures for designing erosion
protection for earthen covers at uranium mill tailings sites that apply to the Federal Cell at the
Clive Site. The stability of a design is evaluated by comparing the velocity of flow from an
extreme precipitation event on the embankment slopes to the maximum flow a channel can carry
without causing gullies. Site-specific information used to calculate the velocity on the
embankment includes the probable maximum precipitation intensity, catchment area, slope, and
a roughness coefficient.
Modeling of erosion in a borrow pit at the Site provided a line of evidence in addition to the
erosion modeling using USLE and the NRC gully projection method (NRC 2002) that
demonstrates the stability of the proposed ET cover design. Although the borrow pit is an
approximation to the embankment, this analysis using the landscape evolution model SIBERIA
provides insight into erosion projected out to 10,000 years.
2.0 UDEQ Interrogatory Responses
This section contains responses for Open Interrogatories CR R313-25-8(4)(a)-71/1 and CR
R313-25-7(2)-191/3, as well as New Interrogatories CR R313-25-25(4)-197/1, CR R313-25-
25(4)-198/1, CR R313-25-25(4)-199/1, CR R313-25-25(4)-200/1, CR R313-25-25(4)-201/1, CR
R313-25-25(4)-202/1, and CR R313-25-25(4)-205/1.
2.1 Interrogatory CR R313-25-8(4)(a)-71/1: Biotic Processes in Gully
Formation
DEQ Conclusion from April 2015 SER, Appendix C:
…the DU PA SER, Section 4.4.2 concluded the following:
Before the DU PA can be determined to be adequate, EnergySolutions needs to clarify
certain issues relating to Appendix 10 to the DU PA Model v1.2 (…)... DRC is currently
reviewing a license amendment request5 to use an ET cover of similar design to that
proposed for the Federal Cell in the DU PA. Any recommendations and conclusions
from that review must be applied to the proposed Federal Cell as well.
Therefore, this interrogatory remains open.
DEQ Critique of DU PA v1.4, Appendix 5, Appendix 21: No further analysis has been performed
pertaining to biotic processes in gully formation since v1.2.
2.1.1 Interrogatory Response
This interrogatory points to soil erosion issues raised in Section 4.4.2 of the Safety Evaluation
Report (SC&A 2015a). UDEQ expresses concerns that gullies will form and enhance radon
diffusion, deep infiltration, and contaminant transport. EnergySolutions plans both ecological
and engineering measures to address these concerns by minimizing gully formation.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 9
Note that this interrogatory also attempts to link two distinct cell design reviews. Since issuance
of the 2015 SER, EnergySolutions has begun construction of rock armor cover on the CAW cell.
More importantly, the DU PA models the Federal Cell’s ET cover as designed. CAW cover
licensing is separate and distinct from the Federal Cell. Since their geometry differs, any attempt
to couple or join the reviews is unsupported and arbitrary. Therefore, this aspect of the question
is moot.
Ecological Measures
SWCA Environmental Consultants (SWCA) assessed erosion under undisturbed conditions at
Clive in June 2012 (SWCA 2012). Eight study plots and one soil borrow area were identified and
the location of wind and water erosion features were mapped. Each study plot was 0.25 ac
(1000 m2) with three plots located on hills. All plots were within the elevational range of the
Clive Site.
SWCA observed that the effects of wind and water erosion were limited. There was minimal
evidence found of water erosion even on the sloped study plots. This is consistent with the
Hansen, Allen, & Luce (HAL) calculations of projected minimal sheet and rill erosion loss
described in the response to UDEQ Interrogatory 191/3. The site-specific SWCA studies
indicated that a cover would remain stable with the established native vegetation and biological
soil crust mimicking undisturbed conditions.
SWCA has developed a sequence of reclamation measures for quickly re-establishing natural
conditions for the cover that will minimize erosion (SWCA 2013). During the first two years the
cover will be stabilized by seeding with fast-growing species. Establishment of these species will
also exclude invasive annuals. The planned seed mix includes native perennial shrubs, forbs, and
bunchgrasses that will become established during years three through five. The soil surface will
be treated with local biological soil crust to speed up crust recovery during these years.
Biological soil crusts are known to reduce wind and water erosion (Belnap and Gillette 1998;
Belnap et al. 2001). The upper Unit 4 soil layer of the cover will be mixed with 15 percent gravel
and the side slopes with 50 percent gravel as additional measures to prevent erosion.
Observations at a number of other sites have shown that a gravel admixture reduces erosion and
promotes evapotranspiration (SWCA 2013). SWCA (2013) concludes:
The proposed ET cover designs and associated installation procedures comply with UAC
R313-25-8(4)(a) and (b); UAC R313-25-18; UAC R313-25-19; UAC R313-25-20; UAC
R313-25-8(4)(d); and UACR313-25-22 by establishing a stable and functioning system
comprised of native vegetation and soil biota that minimizes any near-term episodic
erosional exposure of contaminated materials. A functioning native ecosystem also provides
long-term soil stabilization via soil development, plant roots in upper soil layers, and
biological soil crusts.
Engineering Measures
Calculations to evaluate the stability of the cover design with respect to rill and gully erosion for
the Class A West cell were provided in Appendix D of EnergySolutions (2015). Similar
calculations for gully erosion potential for the Federal Cell are presented here.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 10
Methods for evaluating the design basis for erosion of soil covers for waste embankments are
given in NUREG-1623 (NRC 2002). This report details methods, guidelines, and procedures for
designing erosion protection for earthen covers at uranium mill tailings sites that apply to the
Federal Cell at the Clive Site.
The evaluation of design elements for long-term cover stability is based on the view taken by
NRC (2002) that gully erosion differs from other design considerations in that the development
of gullies and the process of erosion are cumulative and can progress with subsequent storm
events. The method adopted by the NRC to provide long-term stability is to design stable slopes
that prevent the initiation of gullies by a single, very large storm event. By designing to an
unusually large single event, smaller more frequent events will have no significant cumulative
impact on stability.
The potential for gully erosion on the Federal Cell was evaluated using the permissible velocity
method described in NUREG-1623 (NRC 2002).
Slope Description
The Federal Cell is designed as a covered embankment with relatively steeper sloping sides
nearer the edges. The upper part of the embankment referred to as the top slope has a slope of 2.4
percent, while the side slope is steeper with a slope of 20 percent. The length of the side slope is
175 ft. Top slopes of the embankment have different lengths. The longest top slope length is
521 ft.
Flow Concentration
The peak flow unit discharge, Q (cubic feet per second per foot [cfs/ft]), is calculated using the
Rational Formula (NRC 2002):
!=#× %× &× '
where
F is flow concentration factor,
c is dimensionless runoff coefficient,
i is rainfall intensity (inches/hour [in/hr]), and
A is catchment area (acres).
A default value of 3 is recommended in NRC (2002) for the flow concentration factor, F. A
value for the runoff coefficient of 0.5 is recommended for a graveled surface in Table 4.6 of
NUREG 4620 (NRC 1986). The rainfall intensity used for the projection is 18.4 inches for the
top slope and 19.8 inches for the side slope at the Clive Site. See the response to UDEQ
Interrogatory 201 for a description of the calculation of these values. The catchment area is the
area of a 1-ft wide strip along the length of the slope. Using these values, the peak flow unit
discharge, Q, for the top slope and side slope are found to be 0.328 and 0.122 cfs/ft respectively.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 11
Flow Depth
The flow depth, y, is then calculated using the Manning equation for normal depth on a one-foot-
wide strip of the slope. This equation is given by NRC (2002) as
()*+=!,
-1.486 45 6+78
where
y is flow depth (ft),
n is Manning n, and
S is slope (ft/ft).
A value of 0.05 is used for the Manning’s n based on the calculation method of Bray for natural
channels described in Coon (1998).
Using the previously calculated values for Q and the Manning’s n, flow depths were calculated
to be 0.205 ft for the top slope and 0.060 ft for the side slope.
Maximum Permissible Velocity
A value of 5.0 ft/s was chosen as the maximum permissible velocity (MPV) based on the
characteristics of the channel. This is the value listed for gravel in Table CH13-T103 of Colorado
Water Conservation Board (CWCB 2006) and in Table 4.7 of Nelson et al. (1986). The NRC
(2002) method requires that the MPV be adjusted to account for the flow depth. Correction
factors developed by Chow are provided in Appendix A of NRC (2002). The correction factor
for flows less than 0.25 ft in depth is 0.5. The adjusted MPV values for both the top slope and the
side slope are adjusted to 2.5 ft/s.
Actual Flow Velocity
The actual flow velocity is determined by dividing the discharge by the flow depth:
9:=!(+
Using this equation, the top slope and side slope velocities are 1.60 ft/s and 2.03 ft/s.
These velocities for the top slope and side slope do not exceed the adjusted MPV, so the design
is acceptable (see Table 1). Both slope scenarios using the ET cover system provide acceptable
protection against gully erosion using these criteria by limiting the potential of gully formation
from high velocity channelization.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 12
Table 1. Gully erosion potential—permissible velocity analysis for Federal Cell.
Slope
Description
Lengt
h (ft)
Slop
e
(ft/ft)
i
(in/hr
)
c Q
(cfs/ft
)
y
(ft)
Va
(ft/s
)
MP
V
(ft/s
)
Reductio
n Factor
Adjuste
d MPV
(ft/s)
Top Slope 521 0.02
4
18.3 0.
5
0.328 0.20
5
1.60 5.0 0.5 2.5
Side Slope 178.5 0.20 19.8 0.
5
0.122 0.06
0
2.03 5.0 0.5 2.5

Borrow Pit Model
Modeling of gully formation in a borrow pit at the Site provides another line of evidence in
addition to the sheet erosion modeling using USLE and the NRC gully projection modeling using
the permissible velocity method (NRC 2002) that demonstrate the stability of the proposed ET
cover design. (See the response to UDEQ Interrogatory 191/3.) Although the borrow pit is an
approximation to the embankment, the intent of this analysis is to provide insight into erosion
projected out to 10,000 years.
The SER (SC&A 2015a) states in Section 4.4.2 that results from the borrow pit modeling
described in Appendix 10 of Neptune (2014) indicate that at 10,000 years about 1 percent of the
surface will have gullies greater than or equal to 1 meter and suggest that, if this is the correct
interpretation of the results, the influence of gully formation on radon release and infiltration
should be minimal. Appendix 10 of the DU PA Model v1.2 presented a sampling of the results of
the 1,000 simulations, while the results from all 1,000 simulations were provided in Appendix 10
of the DU PA Model v1.4 (Neptune 2015a).
See the response to UDEQ Interrogatory 199 regarding issues raised in Section 4.4.2 of the SER
on the comparability of geometry and materials of the borrow pit and the embankment.
Section 4.4.2 of the SER noted that the description of how the results of the borrow pit erosion
modeling were used in the DU PA Goldsim Model needed clarification. A detailed description of
the analysis of erosion modeling results and their implementation in the GoldSim model was
included in Appendix 10 of the DU PA Model v1.4 (Neptune 2015b).
Lack of evidence of significant erosion activity at the Site, planned ecological remediation of the
soils used to construct the cover, and erosion models that demonstrate the stability of the cover
support the prediction that erosion of the embankment will be minimal and will not enhance
radon fluxes or deep infiltration.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 13
2.2 Interrogatory CR R313-25-7(2)-191/3: Effect of Gully Erosion
DEQ Conclusion from April 2015 SER, Appendix C:
In conclusion, the analysis performed by HAL may or may not be correct, but before DEQ can
accept it, each value selected and used in the analysis needs to be justified.
EnergySolutions/HAL also needs to address how the embankment will be re-vegetated, how
much re-vegetation is necessary and how much is expected, and how long is it expected to take.
Therefore, this interrogatory remains open.
DEQ Critique of DU PA v1.4 and Appendix 21:
Gravel Surface
Gravel embedded in the upper layer may migrate upward over time due to environmental effects,
such as freeze/thaw or wet/dry cycling phenomena, bringing some particles to the surface. At the
same time, eolian erosion and deposition is likely to occur on the cover, potentially “silting in”
gravel particles that move to the surface. Formation of a “gravel mulch” layer (i.e., a clean coarse
layer of gravel at the surface) that would impede evaporation is unlikely. A more likely
phenomenon is formation of a desert pavement, with finer sands, silts, and clay particles
embedding around gravel particles. These finer materials provide a capillary conduit for
evaporation.
This phenomenon is observed at sites where riprap or cobbles are used as cover. Fines deposit
in the pores between the large particles, gradually accumulating and filling the pores. These fines
serve as a seed bed and as a capillary conduit, allowing water to flow upward. This was very
clear in the armored surfaces at both Uranium Mill Tailings Radiation Control Act (UMTRCA)
covers studied in 2016 for the NRC and DOE-LM. An example from the Uranium Mill Tailings
Reclamation site in Grants, New Mexico, is shown in Figure 191-1 below. Roots and structure in
the radon barrier are being mapped as shown in Figure 191-1(a) and brush growing in an
adjacent area of the riprap surface layer is shown in Figure 191-1(b).
The best approach to understand this issue, and to develop a suitable conceptual model for
Clive, is to seek out analogs in the area where undisturbed fluvial surficial soils exist with
appreciable gravel. Studying the surface of these soils will provide evidence regarding the long-
term surface characteristics that can be anticipated at Clive.
Gravel Fraction to Address Erosion
The appropriate gravel fraction necessary to prevent erosion has not been defined with
precision, nor has a validated methodology been developed to determine the appropriate gravel
fraction as a function of site-specific conditions. Models have been developed, but they have not
been validated in the field. For example, Smith and Benson (2016) used the model SIBERIA to
evaluate erosion from a top deck with a gravel amendment, but the model was not validated in
the field.
The gravel admixtures used at Hanford and Monticello have been effective in controlling erosion.
No major erosion issues have been encountered at either site on the shallow top decks. Riprap
is used on the steeper side slopes on both sites. There have been no quantitative field studies to
evaluate the reduction in erosion achieved with the gravel admixture on the top deck at either
site.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 14
Gravel Fraction to Control or Prevent Biointrusion
There should be no expectation that 15% gravel, or even 50% gravel, will preclude biointrusion.
As noted previously, plants readily germinate and root in riprap layers when silt accumulates in
the pores (Figure 191-1). Vegetation is likely to be more robust in a gravel-amended surface
layer with smaller particles and more fine-textured particles.
Burrowing animals will readily penetrate a layer containing gravel particles, and plants will readily
grow in a fine-textured layer with as much as 50% gravel. Biointrusion design to prevent
burrowing requires particles larger than the breadth of the animal (precludes particles from being
moved through a burrow), and a gradation that results in pore sizes smaller than the breadth of
the animal (prevents burrowing between particles).
Homogenization
Pedogenic phenomena are known to create structure and alter the hydraulic properties of
earthen cover materials. There is no evidence in the literature that layering in covers diminishes
with time or that a homogeneous profile develops. For example, distinct layering has been
observed in recent excavations into UMTRCA covers that are 20 years old (Figure 191-2).
Structure has developed in these layers, and the hydraulic properties have changed, but the
profile is not homogeneous. A model for Clive should include a layered profile with appropriate
hydraulic properties assigned to each layer that reflect realistic development of structure.
2.2.1 Interrogatory Response
UDEQ raises a number of issues in this interrogatory, beginning with design for slope stability.
Erosion Analyses
The first request from UDEQ is to provide the rationale for selection of the Revised Universal
Soil Loss Equation (USLE) (USDA 1978) and the Rangeland Hydrology and Erosion Model
(RHEM) (Nearing et al. 2011) parameters used for erosion modeling. A detailed response to this
request was developed by Hansen, Allen, and Luce (HAL 2018). A summary of their final report
is provided here. The complete report is provided in Appendix A.
Universal Soil Loss Equation (USLE) Parameters
The USLE estimates average annual soil losses from erosion. While the USLE equations are the
same as the Revised Universal Soil Loss Equations (RUSLE), the use of the acronym RUSLE
could imply the use of the software program developed by the USDA for factor determination. In
order to clarify the methodology used in the analysis, all references are to USLE. The software
package was not used because factors from a Utah-specific publication were available for the
analysis as described below.
The USLE methodology is commonly used to determine the long-term stability of slopes and is
an industry-standard method for design of erosion control. Guidance given by the EPA states that
“The U.S. Department of Agriculture’s (USDA’s) Universal Soil Loss Equation is recommended
as the tool to evaluate erosion potential” (EPA 1991). The basis for this approach is that “if
adequate protection is provided to control sheet erosion, then rills and gullies will never form
from rainfall” (Israelsen et al. 1984). The USLE equation is defined as:
Erosion Responses for the Clive DU PA Model
23 Feb 2018 15
A = R*K*LS*C*P
where
A is the average soil loss per unit area, expressed in tons/acre/year,
R is the rainfall/runoff factor, which is the number of rainfall units for rainfall energy
and runoff and snowmelt,
K is the soil erodibility factor in tons per acre per year per unit of R,
LS is the topographic factor (length and steepness of the slope),
C is the cover and management factor, which is the ratio of soil loss from an area
with a given cover and management relative to that from an identical area in
continuous fallow, and
P is the supporting conservation practice factor, in this case assumed to be equal to 1.
This procedure and site-specific factors are described in “Erosion and Sedimentation in Utah—A
Guide for Control” (Israelsen et al. 1984) and “Design Hydrology and Sedimentation for Small
Catchments” (Haan et al. 1994).
R Factor
The R factor (6) is selected based on the Utah-specific Iso-Erodent (R) mapping provided in the
Utah Water Research Laboratory report (Israelsen et al. 1984) which is provided in Figure 1.
K Soil Erodibility Factor
The K values (0.18 and 0.07 for top slope and side slope, respectively) are based on the Unit
4-specific material characteristics with the top slope gravel admixture and characteristics for the
side slope gravel admixture, together with the Wischmeier nomograph as described in the
methodology presented in the Israelsen and Hann procedures and shown in Figure 2. A particle
size analysis using the hydrometer method was performed on the Unit 4 clay to determine the
percent silt and very fine sand (see Figure 2) for use with the Wischmeier nomograph.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 16

Figure 1. Utah-specific Iso-Erodent (R) mapping provided in the Utah Water Research
Laboratory report (Israelsen et al. 1984).
Erosion Responses for the Clive DU PA Model
23 Feb 2018 17

Figure 2. Wischmeier nomograph (Haan et al. 1994; Israelsen et al. 1984).

C Factor (VM Factor)
The C parameters used in the equation for both the 4% and 20% slopes were based on Table 9
and Table 10 of the US Department of Agriculture Handbook Number 537, “Predicting Rainfall
Erosion Losses, A Guide to Conservation Planning” (USDA 1978), and are shown in Figure 3.
The C factor for the top slopes (0.2) is based on the sparse vegetative cover naturally found in
the areas immediately surrounding the Clive facility (Table 10 of Handbook Number 537, No
Appreciable Canopy, Type G with 20% ground cover) and the Unit 4 gravel admixture. The C
factor for the side slope is based on the higher percentage of gravel in the Unit 4 gravel
admixture (50% gravel). The 50% gravel admixture on the side slopes results in a pseudo-gravel
mulch once some of the fines have been removed. Therefore, a C factor of 0.02 was selected
(Table 9 Handbook Number 537, Crushed Stone ¼ to 1½ in, Mulch Rate of 240 tons/acre, Land
Slope of 21–33%, with a length limit of 200 feet).
Results
Soil loss calculated using these parameters in the USLE equation for the top slope and side slope
are shown in Table 2.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 18

Figure 3. Table 9 and Table 10 of the US Department of Agriculture Handbook Number
537, “Predicting Rainfall Erosion Losses, A Guide to Conservation Planning” (USDA
1978).

Table 2. Effects of erosion—average soil loss analysis using USLE. Soil loss is calculated for
the top slope with Unit 4 clay with 15% gravel admixture and side slope with Unit 4
clay and 50% gravel admixture.
Slope
Segment
R
(ft tons/
ac/hr)
K
(tons/ac/EI)
L
(ft)
S
(%)
C A
(tons/ac/yr)
Total Soil
Loss
(mm/year)
Top Slope
(4%)
6 0.18 942 (4%) 4% 0.2 0.25 0.24
overall
0.026
Side Slope
(20%)
6 0.07 188 (20%) 20% 0.02 0.19

Erosion Responses for the Clive DU PA Model
23 Feb 2018 19
The Rangeland Hydrology and Erosion Model (RHEM) Model Parameters
The RHEM model was used in HAL’s interrogatory response entitled “DRC RFI Section 4.0
Erosion” completed in 2013 to corroborate the results of the USLE analysis (Appendix D of
EnergySolutions (2013)). This comparison showed a total loss from the embankment of 0.026
mm/yr from the USLE analysis and 0.016 mm/yr from the RHEM analysis. Since 2013, the
RHEM on-line model has been updated. The latest version of the model (Version 2.3) lacks
some of the functionality of the earlier model.
Specifically, slope lengths cannot be directly input as they were previously, so a direct
comparison with the calculation using USLE is not possible. This appears to be due to an error
within the program since the descriptions of the model available on the website include slope
length as an input parameter option. Attempts to reach the contacts listed for the RHEM model in
order to resolve the issue were unsuccessful. The model defaults to a slope length of 50 meters
(164 feet).
Additional parameter options for cover characteristics appear to have been added since 2013,
including options for a distinction between foliar cover and basal plant cover as well as rock
cover, litter cover, and biological soil crust cover. Since the tool is a web-based model and not
stand-alone software, the version of the online tool used in the 2013 analysis is no longer
available. The parameters described in the 2013 interrogatory for vegetative conditions, slope
grades, climate station, soil texture class, and slope lengths are still valid for the analysis
completed at the time but the results reported therein can no longer be validated.
Gravel Surface
Next UDEQ discusses the use of local analogs to study soils with gravel surfaces. There are no
questions or requests raised.
Gravel Fraction to Address Erosion
Then UDEQ goes on to argue that the appropriate gravel fraction necessary to prevent erosion
has not been defined with precision. Further, a validated methodology has not been developed to
determine this. Refer to the response to UDEQ Interrogatory 198/1 for further discussion.
Gravel Fraction to Control or Prevent Biointrusion
The next issue raised is a statement by UDEQ that gravel will not prevent biointrusion. This
observation is accurate but not relevant, as gravel used in the cover design is not assumed or
intended to inhibit biointrusion. In fact, as described in the revegetation plan developed by
SWCA (2013), one function of gravel mulch over the soil surface is to increase moisture
retention and seed establishment.
The proposed cover design uses the frost protection zone to provide the function of a
biointrusion barrier. This layer is described by SWCA (2013) as “18 inches (46 cm) of 10–16
inches (25–41 cm) gravel and cobble mixture in-filled with small gravel, sand, and other fines
(cobble and gravel to 16 inches [40.6 cm] diameter).” Please see the response to UDEQ
Interrogatory 05/2 for discussion of evaluation of the extent of expected biointrusion at the Site.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 20
Homogenization
The last issue of this interrogatory raised by UDEQ is that the hydraulic parameter distributions
and correlations used for modeling of flow in the cover system based on the recommendations of
Benson et al. (2011) were not implemented in the way that was intended by UDEQ.
Contrary to UDEQ’s claim in this alternative modeling of net infiltration and water content of
the cover layers requested by UDEQ, EnergySolutions did in fact use the recommendations from
Section 10.2 “Parameters for Performance Assessments” in Benson et al. (2011), cited as the
source of recommendations in Appendix E of the SER (SC&A 2015b). These recommendations
for hydraulic parameters are:
• The saturated hydraulic conductivity of fine-textured earthen storage and barrier layers
can be assumed to range between 1 × 107 m/s and 5 × 106 m/s.
• The porosity of earthen storage and barrier layers will likely range between 0.35 and
0.45.
• The α-parameter in the van Genuchten equation, which is used to describe the soil water
characteristic curve (SWCC) for hydrologic simulations, varies between 0.01 and
0.33 kPa-1 for field-scale barrier and storage layers.
• The n-parameter in van Genuchten’s equation, which is used to describe the SWCC for
hydrologic simulations, varies over a very small range (typically between 1.2 to 1.4).
Except for the n-parameter, where UDEQ provides no reference as to which layer the
recommendation refers, parameter value recommendations for all other parameters are applied to
both barrier and storage layers without distinction. For the proposed Federal Cell cover design,
the layers of the evapotranspiration (ET) cover are all either storage or barrier layers; however,
UDEQ’s recommended method makes no differentiation between the layers with respect to
hydraulic parameters. The only distinguishing characteristics between depth zones of a variably
saturated flow model are the values of the hydraulic parameters. If a saturated hydraulic
conductivity value, for example, is drawn from a statistical distribution developed according to
the method of Benson et al. (2011), then that value will be assigned according to that method to
all cover layers in the model. Thus, the flow model of the cover will be homogeneous with
respect to hydraulic conductivity. UDEQ objects to this homogeneity in hydraulic properties in
this interrogatory, but that is the only viable outcome possible when applying the Benson et al.
(2011) method to the ET cover.
2.3 Interrogatory CR R313-25-25(4)-197/1: Properties of
Embankment Side Slope Materials
Interrogatory Statement:
Please explain and justify how, from an erosion perspective, the properties of Unit 4 material “are
sufficiently similar” to the Federal Cell side slope, which consists of a mixture of Unit 4 soil with
50% gravel, to support this “sufficiently similar” modeling assumption. Also, please explain how
the properties of Unit 4 material are sufficiently similar to Class A waste, which would be included
over the DU waste. See also Interrogatory 203/1 below.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 21
2.3.1 Interrogatory Response
In this interrogatory UDEQ preferentially extracts two from a number of assumptions provided
in Appendix 10 of the Clive DU PA Model Final Report v1.4 (Neptune 2015a) as the basis for
applying the results from modeling of a borrow pit face at the Site to the projection of erosion on
the Federal Cell embankment. While there are differences in the characteristics of the borrow pit
and the embankment, this analysis provides another line of evidence in addition to the more
conventional modeling used to demonstrate the stability of the proposed ET cover design
described below. Notably, the multiple lines of evidence provide good agreement that the
proposed ET cover design will provide adequate erosion resistance.
Landscape evolution models such as SIBERIA are still in an early stage of scientific
development. Accordingly, the methods for building confidence in model results that are
commonly used in environmental modeling, such as rigorous calibration, sensitivity analyses,
and uncertainty analyses, are difficult to apply. These models, however can be used in an
exploratory context to gain insight into processes (Skinner et al. 2017) and to provide additional
evidence to support the results of cover stability analyses using methods recommended by EPA
and NRC. Calculations to evaluate the stability of the cover design with respect to gully erosion
for the Class A West cell were provided in Appendix D of EnergySolutions (2015). Similar
calculations for the Federal Cell are presented in the response to UDEQ Interrogatory 71/1.
2.4 Interrogatory CR R313-25-25(4)-198/1: Gravel Content of
Embankment Materials
Interrogatory Statement:
Please provide the design bases and justification for the amount and sizing of the gravel in the
top and side slopes of the Federal Cell. The proposal for the gravel admixture in the top slope
(15%) appears too small. Also, please provide evidence for existing semi-arid or arid sites where
only 15% gravel has been added to form a successful cover-system surface layer for a landfill.
Please describe actual analog sites where 50% gravel for side slopes has been demonstrated to
be effective against erosion.
2.4.1 Interrogatory Response
A detailed response to this interrogatory was developed by Hansen, Allen, and Luce (HAL
2017a). A summary of their final report is provided here. The complete report is provided in
Appendix B.
The design basis for the amount of gravel to add to the soil in the top and side slopes was first
determined based on the analysis of the long-term sustainability of the slopes exposed to sheet
erosion over time. The methods that were used are the Universal Soil Loss Equation (USLE)
(USDA 1978) and the Rangeland Hydrology and Erosion Model (RHEM) (Nearing et al. 2011).
These methods produce an estimate of the average annual soil loss in terms of tons per acre per
year. With the gravel admixed into the Unit 4 clay soil, average annual soil losses were
calculated to be about 0.24 tons/acre/year using USLE, which was also closely matched by the
RHEM. This amount of loss is almost 10 times less than what has been recommended in EPA
Erosion Responses for the Clive DU PA Model
23 Feb 2018 22
guidance for covers for hazardous waste facilities. A more detailed discussion of the
methodologies and results is provided in the response to UDEQ Interrogatory 191/3.
Once the slope was determined to be stable from an average soil loss perspective, checks on the
gully erosion potential were completed based on the calculation of predicted velocities by
comparing them to the maximum permissible velocity according to the method presented in
NUREG-1623, “Design of Erosion Protection for Long-Term Stabilization” (NRC 2002). This
second assessment based on an extreme intense rainfall event is recommended in NUREG-4620
(Nelson et al. 1986) because of the potential for significant damage to cover systems from such
large events. In fact, this extreme intense rainfall event is known as the Probable Maximum
Precipitation (PMP) and is so extreme that “the point precipitation data base, even if maximized
for PMP moisture potential, shows no observed values even close to the 10-inch PMP estimate
(the generalized, local storm PMP estimates in Utah)” (Jensen 1995). Additionally, the
methodology applies a flow concentration factor (F) of three (NRC 2002). Therefore, the
extreme event that is many times the highest ever recorded rainfall intensity in the area is also
compounded by an additional safety factor of a multiplier of three. The end result of this process
is an analysis that is meant to produce an extremely conservative value.
Flow velocities on the top and side slopes of the CAW embankment during the PMP event were
predicted to be 2.37 and 2.07 ft/sec, respectively, as discussed in more detail in the response to
UDEQ Interrogatory 191/3. The acceptable Maximum Permissible Velocity (MPV) was selected
from tables provided in NUREG/CR-4620 (Nelson et al. 1986). Under this method the slope is
stable if the calculated velocity (V, the velocity resulting from a PMP in this application) is less
than the MPV. By contrast, if velocities exceed the MPV, the slope will experience excessive
erosion that will lead to the formation of gullies. HAL found that the calculated velocities
resulting from the PMP did not exceed the prescribed permissible velocity.
Reference Review
The methods discussed above have been published and/or accepted by the EPA (1989) and the
US Nuclear Regulatory Commission (Nelson et al. 1986; NRC 2002). Other methods referred to
in the interrogatory, on the other hand, have not been adopted, published, or referred to in
guidance by federal agencies. While the works cited are part of productive ongoing academic
research and study, they are not conclusive in their findings and should instead be the focus of
additional research. A summary of each is provided below:
• “Gravel Admixtures for Erosion Protection in Semi-Arid Climates,” Erosion of Soils and
Scour of Foundations (Anderson and Stormont 2005). This paper was published as part
of the proceedings of sessions of the Geo-Frontiers 2005 Congress held in Austin, Texas.
Along with other conference proceedings articles, these early attempts to define a process
for admixture design are helpful in that they outline practical steps to determine gravel
size, gravel percentage, and admixture thickness. This approach, however, has limitations
that result from the lack of empirical evidence to back up the steps and applications of
many of the equations presented in the paper. The paper states that “the design method
for gravel admixtures presented here may serve as an outline for further erosion
investigations and provide guidance for future designs of gravel-soil admixture layers.”
Therefore, the paper was not meant to be used as a proven method for admixture design
but rather as a starting point for further research and investigation.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 23
• “Design of Erosion Protection at Landfill Areas with Slopes Less than 10%” (Anderson
and Wall 2010a). Much of the background discussion of this paper is the same as
Anderson and Stormont (2005). The methodology presented is similar to the steps
provided in Anderson and Stormont (2005) with some changes. Again, the paper does not
claim this to be a proven method but indicates the authors are taking steps to test its
outcomes. Anderson and Wall (2010a) state “the procedure described here is being
applied to a landfill cover soon to be constructed in southern Nevada. The Nevada project
will provide the first large scale application of the procedure.” HAL was unable to find
information regarding empirical data on a smaller scale nor was final information on the
results of the large scale Nevada project able to be found. Without this information or
additional discussion in the paper, the limits of the equation and how the criteria for
success or failure are defined remain unknown. For example, assuming the Nevada
project site is shown to be successful with the 40% gravel admixture, there is nothing that
would indicate if a lower percentage of gravel admixture, such as 25%, would also be
adequate. Additionally, the suitability of the design and how success is defined is not
sufficiently described in the paper to know whether the absence of observed gully erosion
constitutes a successful design or if it is acceptable to have the formation of rills and
gullies as long as the depth does not exceed the admixture layer thickness calculated
using the method. This is an important distinction since the end result of the design
process presented in the paper is the calculation of the thickness of the admixture layer
which is presented as being dependent on the percentage of gravel in the admixture. This
specific paper does not provide limits or conclusive guidance to the reliability of the
method and therefore should serve as the starting point for further investigation and
research.
• “Erosion Protection at Landfill Slopes Greater than 10%” (Anderson and Wall 2010b).
This paper focuses on riprap on side slopes. The design proposed by EnergySolutions
does not include riprap side slopes; therefore, this paper does not apply.
• “Long-Term Cover Design for Low-Level Radioactive and Hazardous Waste Sites as
Applied to the Rocky Flats Environmental Technology Site Solar Evaporation Ponds”
(Stenseng and Nixon 1995). The paper describing the design used at the Rocky Flats
Environmental Technology Site was included as part of the proceedings from the 50th
Industrial Waste Conference held in May 1995. The paper includes a brief description of
the design basis for the 5% top slopes and 20% side slopes and the 40% gravel
admixture. The admixture gravel content appears to have been selected instead of
calculated as the result of a defined process. While instructive regarding the specific
cover design discussed in the paper, no additional guidance is provided to guide
application at other sites.
• “Ecology, Design and Long-Term Performance of Surface Barriers: Applications at a
Uranium Mill Tailings Site” (Waugh and Richardson 1997). This paper is similar to the
Stenseng and Nixon paper discussed previously in that it contains a summary of the
design parameters but does not provide a specific and applicable methodology for the
design of a soil gravel admixture with regard to erosion protection.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 24
Sites with Top Slopes Comprised of 15% Gravel Admixture
There are some examples of sites that have employed slopes with 15% admixture and that give
some measure of proof that a 15% admixture can be effective at controlling erosion. A test site
was established at Hanford that is referred to as the Hanford Prototype Surface Barrier (PSB).
The purpose of the PSB test site was to evaluate surface barrier constructability, construction
costs, and physical and hydrologic performance at field scale (DOE 2016). This field-scale test
cover was installed in 1994 and is comprised of a top erosion protection layer made up of a silt
loam admixed with 15% pea gravel. The top slope of 2% in this case is slightly less than the top
slopes of 2.5% and 4% proposed for the Federal Cell, but is an example of a cover system that
relies on the same amount of gravel admixture. A review of the data collected at the site through
2015 (DOE 2016) concludes that:
The 19-year PHB record showed practically no evidence of wind or water erosion of the
ETC barrier, despite 3 years of triple the mean annual precipitation; three simulated 1000-
year-return, 24-hour precipitation events; and an intense, controlled fire that burned off all
vegetation across half the barrier surface….Even in the absence of vegetation (e.g., following
a fire), the pea gravel added to the silt loam protected the barrier surface from wind and
water erosion….Overall, the monitoring results have confirmed that the PHB design is
resistant to water and wind erosion and that resistance is expected throughout the barrier’s
1000-year design life.
Another test site using the same cover system was established at Hill Air Force Base in 1994,
though little information has been published about that site to date.
Sites with Side Slopes Comprised of 50% Gravel Admixture
No sites were found that have used a gravel admixture on side slopes at or above 20%. Similarly,
there were no methodologies found that specifically address the calculation of gravel admixtures
for slopes greater than 10%, other than the general methods found in NUREG-1623 and
NUREG/CR-4620 (Nelson et al. 1986; NRC 2002). For slopes over 9%, Simanton et al. (1984)
found that the rate of water erosion decreases exponentially with increasing rock fragment cover.
The effect of biological soil crust is also difficult to quantify. These crusts are expected to
become established along the slopes as has been observed along natural slopes in the Clive area.
The required lengthy post-closure care period provides the greatest opportunity for verification
of the design methodology described previously by Simanton et al. (1984).
The design process completed to determine the acceptability of the 50% gravel admixture on the
20% side slopes is in accordance with EPA and Nuclear Regulatory Commission guidance.
These guidance procedures contain no requirements that existing sites be used as evidence to
support proposed designs. While it is acknowledged that such evidence would be helpful, each
site is different and contains unique design constraints based on physical layout, climate,
material availability, and project goals that make determinations based on sound methodologies
necessary.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 25
2.5 Interrogatory CR R313-25-25(4)-199/1: Uncertainties in Erosion
Modeling
Interrogatory Statement:
Please provide quantitative estimates of the uncertainties involved using the borrow pit model as
an analog for estimating erosion of the Federal Cell, including use of RHEM to develop input
parameters for SIBERIA, and modeling uncertainties inherent in the selection of SIBERIA.
2.5.1 Interrogatory Response
Landscape evolution models (LEMs) were developed and applied for a face of a borrow pit at the
Clive Site in order to predict the response of the pit face and upslope land surface to water
erosion during runoff events. The models provide a quantitative description of the evolution of
slopes and channels over time. While the embankment upper cover layers and the borrow pit
material were the same Unit 4 soil, there were differences between the embankment and the
borrow pit. The borrow pit was modeled as bare soil with no vegetation or gravel, with a much
smaller top slope but a much larger catchment area. The embankment will be vegetated, and will
have a surface layer with a gravel admixture but a steeper top slope. The catchment area of the
embankment is smaller and the side slope is much less steep than the borrow pit face.
Although the borrow pit is an approximation to the embankment, the intent of this analysis was
to provide insight into erosion projected out to 10,000 years. This analysis provides another line
of evidence in addition to the sheet erosion modeling using USLE and the gully projection
modeling using the permissible velocity method (NRC 2002) that demonstrate the stability of the
proposed ET cover design. (See the response to UDEQ Interrogatories 71/1, 191/3, and 197/1.)
This interrogatory requests that quantitative uncertainty estimates be provided for the use of the
borrow pit erosion model results as an analog to evaluate the influence of erosion on
embankment performance at 10,000 years. While potentially interesting, such estimates are not
necessary for demonstrating erosion resistance of the Federal Cell ET Cover. Furthermore,
LEMs have been used as exploratory models providing insight into landscape–climate processes
for many years, but they have not been developed to the level of other types of environmental
modeling (Skinner et al. 2017). Typical methods used for calibration, sensitivity analysis, and
uncertainty analysis are difficult to apply to LEMs since little data is available for calibration and
verification (Skinner et al. 2017); (Temme et al. 2009). LEMs, however, even at this early stage,
can be useful in providing insight into landscape-climate processes.
2.6 Interrogatory CR R313-25-25(4)-200/1: Use of RHEM to Develop
Parameters for SIBERIA
Interrogatory Statement:
Please remodel erosion of the Federal Cell cover using the newer version of the RHEM model
Al-Hamdan et al. 2015) applicable to disturbed soils and concentrated surface-water flow. The
SIBERIA model results in the DU PA v.1.4 should be compared with those of SIBERIA modeling
of erosion for the site based on the Grand Junction embankment modeling by Smith (2011).
Modeling of the latter embankment indicates that significant gullying can occur over time on side
Erosion Responses for the Clive DU PA Model
23 Feb 2018 26
slopes even with vegetated soil on the embankment having considerable (i.e., 40%) added
gravel (Smith 2011).
2.6.1 Interrogatory Response
UDEQ requests that modeling of erosion on the Federal Cell be redone using an updated version
of the RHEM model. As described by Al-Hamdan et al. (2014), “RHEM was initially developed
for functionally intact rangelands where concentrated flow erosion is minimal.” Al-Hamdan goes
on to describe the changes to RHEM as including accounting for changes in flow due to soil
disturbance: “Disturbance such as fire or woody plant encroachment can amplify overland flow
erosion by increasing the likelihood of concentrated flow formation.” UDEQ states that the
modeling for the Federal Cell should be redone because the previous version of RHEM used for
the modeling “has limited application to describing erosion by concentrated flow on disturbed
soils, as would be expected at Clive.”
The response to UDEQ Interrogatory 71/1 describes the sequence of reclamation measures for
quickly re-establishing natural conditions on the cover developed by SWCA (2013). Ecological
and engineering measures described in the response to 171/1 will rapidly stabilize the cover in
the short term and continue to provide long-stabilization. As SWCA (2013) states, “functioning
native ecosystems comprised of the borrow soils at the Clive Site do not show erosion as the
DRC suggests.” The cover surface will not have the characteristics of disturbed rangeland soils,
so the added features of RHEM for disturbed conditions are not relevant to modeling the Federal
Cell. Thus, redoing the SIBERIA erosion model calibration with RHEM will not add to the
information from the modeling results.
Erosion modeling of the Federal Cell and the Class A West Cell using methods in NRC- and
EPA-approved guidance has demonstrated the stability of the proposed covers. Sheet and rill
erosion modeling using USLE is described in the response to UDEQ Interrogatory 191/3.
Evaluation of the potential for gully erosion on the Federal Cell using the permissible velocity
method is performed as described in NUREG-1623 (NRC 2002) (response to UDEQ
Interrogatory 71/1).
In addition, changes to the RHEM model have made it currently inapplicable to modeling
scenarios at Clive. HAL (2018) note that slope length is no longer a functioning input variable to
RHEM; all simulations have a set slope length of 50 meters (164 ft). See the response to UDEQ
Interrogatory 191/3 for more information.
This interrogatory also requests a comparison of the borrow pit modeling results (DU PA v1.4
Appendix 10 (Neptune 2015c)) to the results of Smith and Benson (2016) for an embankment at
Grand Junction. Smith and Benson (2016) describe a modeling study comparing cover stability
for scenarios considering cover geometry, climate, rock fragment content of the surface layer,
and vegetation. The conditions considered by these authors were: semi-arid and humid climate;
surface layers of riprap, topsoil, and gravel admixture; and presence and absence of vegetation.
These scenarios were simulated to 1,000 years by Smith and Benson (2016).
There are a number of reasons why the borrow pit simulations (DU PA v1.4 Appendix 10
(Neptune 2015c)) and Smith and Benson (2016) are not comparable. The characteristics of the
topsoils used for the two covers are different. Smith and Benson described the properties of the
Erosion Responses for the Clive DU PA Model
23 Feb 2018 27
materials they used for the simulations in their Table 3.1. This table shows the particle size
distribution of the topsoil as 40 percent sand, 40 percent silt, and 20 percent clay. In contrast,
Unit 4 at the Clive Site is approximately 23 percent sand, 47 percent silt, and 30 percent clay (see
also the response to Interrogatory 191 and Appendix A).
The Grand Junction site is not ecologically analogous to the Clive Site (SWCA 2013).
Vegetation is an important factor in stabilizing covers. Smith and Benson (2016) assumed the
dominant vegetation for the semi-arid case to be Mountain Big Sagebrush. In contrast, SWCA
(2013) identifies the dominant vegetation community as Mountain Basins Mixed Salt Desert
Scrub on analogs to the Clive Site. The soils at Clive are saline; different chemistry and fertility
of soils at Grand Junction influence the characteristics of the native vegetation. SWCA (2013)
describe biological soil crusts as a “dominant feature of the vegetation communities in the Great
Salt Lake Basin.” Biological soil crusts, important in stabilizing soils, are not considered by
Smith and Benson (2016).
Clive borrow pit simulations projected to 10,000 years (DU PA v1.4 Appendix 10) considered
unvegetated, Unit 4 soils with no added gravel. Smith and Benson (2016) show erosion model
results for a cover in a semi-arid climate with a vegetated topsoil surface in Figure 4.1, and
results for a cover in a semi-arid climate with an unvegetated gravel admixure surface in Figure
4.6. No directly comparable results to the Clive borrow pit simulation conditions (semi-arid, no
vegetation, no gravel) were found in Smith and Benson (2016).
2.7 Interrogatory CR R313-25-25(4)-201/1: Estimating Rainfall
Intensity
Interrogatory Statement
It is not clear that the Probable Maximum Precipitation (PMP) was determined using the
procedures outlined in the National Oceanic and Atmospheric Administration and U.S. Army
Corps of Engineers publication Hydrometeorological Report No. 49 (HMR 49) (1977).
According to EnergySolutions, these procedures resulted in a “1-hour PMP rainfall intensity of
9.9 inches (Jones, 2012).” However, DWMRC finds that a value of 9.8 or 9.9 inches is not the
intensity, but rather the 1-hour PMP, or the maximum precipitation expected over 1 square mile
when averaged over an hour.
Please re-calculate the PMP using NUREG/CR-4620, as outlined below.
2.7.1 Interrogatory Response
A detailed response to this interrogatory was developed by Hansen, Allen, and Luce (HAL
2017b). A summary of their final report is provided here. The complete report is provided in
Appendix C.
Gully erosion potential was initially checked based on the calculation of permissible velocities
according to the method presented in NUREG-1623, “Design of Erosion Protection for Long-
Term Stabilization” (NRC 2002). As pointed out in the Interrogatory Statement, it is
acknowledged that the methodology utilized previously failed to incorporate the proper rainfall
intensity as outlined in NUREG/CR-4620 (Nelson et al. 1986).
Erosion Responses for the Clive DU PA Model
23 Feb 2018 28
In order to determine the probable maximum precipitation (PMP) intensity as outlined in the
guidance in NUREG/CR-4620 (Nelson et al. 1986), it is necessary to first calculate the time of
concentration for representative drainage areas for both the top and side slopes. The
methodology outlined in Technical Release 55, “Urban Hydrology for Small Watersheds”
(USDA 1986), was used to calculate the time of concentration for each representative slope.
TR-55 describes three types of drainage flow: sheet flow, shallow concentrated flow, and open
channel flow. Sheet flow is defined as flow over planar surfaces at very shallow depths for up to
300 feet. After a maximum of 300 feet, the flow transitions to shallow concentrated flow. Open
channels were not included in the calculations due to the absence of designed channels. The
same Manning’s roughness coefficient calculated to be 0.05 for use in the NUREG-1623 (NRC
2002) methodology using an empirical equation for channels with gravel beds with shallow flow
depths of Bray (Coon 1998) was also used to represent the sheet flow roughness. It was decided
to use the value calculated using the Bray method (0.05) instead of a higher value found in other
publications for overland flow using sparse vegetative cover in order to be more protective. The
above described methodology was applied to both the Class A West Cell and the Federal Cell to
calculate the time of concentrations shown in Table 3 and
Table 4.
Table 3. Calculated Time of Concentration Using TR-55 Method for Class A West Cell.
Slope Description Total
Length
(ft)
Slope
(ft/ft)
Sheet
Flow
Length
(ft)
Sheet Flow
Manning’s n
Shallow
Concentrated
Flow Length
(ft)
Total Time of
Concentration
(min)
Top Slope (4%) 942 0.04 300 0.05 642 17.4
Side Slope (20%) 188 0.20 188 0.05 0 4.6

Table 4. Calculated Time of Concentration Using Tr-55 Method for Federal Cell.
Slope Description Total
Length
(ft)
Slope
(ft/ft)
Sheet
Flow
Length
(ft)
Sheet Flow
Manning’s n
Shallow
Concentrated
Flow Length
(ft)
Total Time of
Concentration
(min)
Top Slope (2.5%) 521 0.025 300 0.05 221 16.9
Side Slope (20%) 178.5 0.20 178.5 0.05 0 4.4

The rainfall depth is 9.9 inches, determined using the methods outlined in Table 6.3a of the US
Army Corps of Engineers publication, Hydrometeorological Report No. 49 (HMR 49) (NOAA
1984), for determining an average 1-hour 1-square mile PMP. The steps were then followed from
NUREG/CR-4620 (Nelson et al. 1986) to determine rainfall intensity from the PMP. The
intensities for the Class A West and Federal Cells are shown in Table 5 and Table 6.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 29
Table 5. PMP Rainfall Intensity for the Class A West Cell.
Slope Description Rainfall
Duration (Tc)
(minutes)
% PMP
(%)
Rainfall
Intensity (i)
(inches/hr)
Top Slope (4%) 17.4 53.8% 18.4
Side Slope (20%) 4.6 15.3% 19.8
Table 6. PMP Rainfall Intensity for the Federal Cell.
Slope Description Rainfall
Duration (Tc)
(minutes)
% PMP
(%)
Rainfall
Intensity (i)
(inches/hr)
Top Slope (2.5%) 16.9 52.1% 18.3
Side Slope (20%) 4.4 14.7% 19.8

Gully Erosion Potential
As opposed to the projection of the long-term effects of precipitation over time due to sheet
erosion, the effects of gully erosion are determined by the consideration of a large single rainfall
event. The procedure described in NUREG-1623 (NRC 2002) begins by calculating the peak
runoff rate considering the PMP rainfall intensity, the slope length, and the multiplication factor
of three recommended by NRC (2002). A flow depth is then calculated using the runoff rate,
slope, and a roughness coefficient for channels with gravel beds and shallow depths. The flow
rate and depth are then used to estimate a flow velocity.
The results for both the top slope and the side slope of the Class A West Cell using the vegetated
slope condition are summarized in Table 7. Flow velocities on the top and side slopes of the
Class A West Cell during the PMP event are predicted to be 2.37 and 2.07 ft/sec, respectively.
The permissible velocity method is commonly applied to determine channel stability. The slope
is assumed stable if the calculated velocity (V, the velocity resulting from a PMP in this
application) is less than the maximum permissible velocity (MPV). By contrast, if velocities
exceed the MPV, it is assumed that the slope will experience excessive erosion that will lead to
the formation of gullies. An MPV of 5.0 ft/s, appropriate for gravel channels, was selected from
tables provided in NUREG/CR-4620 (Nelson et al. 1986). This methodology then directs that the
estimates for the MPV be adjusted downward to account for the influences of flow depth. For the
flow depths calculated for this cell the adjustment factor is 0.5, reducing the MPV to 2.5 ft/s. The
side slope gully analysis was completed independently of the top slope.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 30
Table 7. Gully Erosion Potential—Velocity Analysis for Class A West Cell.
Slope Description Length
(ft)
Slope
(ft/ft)
i
(in/hr)
c Q
(cfs/ft)
y
(ft)
V
(ft/sec)
Adjusted
MPV
(ft/sec)
Top Slope (4%) 942 0.04 18.4 0.5 0.60 0.25 2.37 2.5
Side Slope (20%) 188 0.20 19.8 0.5 0.13 0.06 2.07 2.5

Comparing calculated velocities to MPVs in Table 7 demonstrates that all slope scenarios using
the ET cover system provide acceptable protection against gully erosion using these criteria by
limiting the potential of gully formation from high velocity channelization.
2.8 Interrogatory CR R313-25-25(4)-202/1: Use of SIBERIA to Model
Federal Cell Erosion
Interrogatory Statement:
The Division is concerned that the SIBERIA model referenced in DU PA v.1.4 discussions
assumes a modeling-realm geometry inconsistent with that of the proposed Federal Cell. The
Federal Cell embankment is approximately 30 feet high (height of waste under top slope and
above grade) compared to the model analog height of 10 feet. Also, the SIBERIA model allows
for several hundred meters of ground surface upslope from the sloping pit face, but that ground
surface only has a 0.3% (0.003) grade in the model. By contrast, as described in Appendix 3 to
the DU PA v.1.4, the waste under the top slope above and upslope from the side slopes of the
embankment has a grade of up to 2.4%. This is about eight times greater. EnergySolutions
needs to explain how these differences affect the results and how the Federal Cell modeling
results can be reconciled against similar modeling studies conducted by Smith and Benson
(2016) for the Grand Junction Uranium Mill Tailings Disposal Site.
2.8.1 Interrogatory Response
SIBERIA is employed in the DU PA model as a supplementary line of evidence for embankment
stability. If SIBERIA results were the sole or primary basis for demonstrating embankment
stability, the distinctions noted in the interrogatory could be relevant; however, LEMs such as
SIBERIA are acknowledged to be subject to further development before their results should be
considered conclusive in licensing situations. See also the response to UDEQ Interrogatory
199/1.
2.9 Interrogatory CR R313-25-25(4)-205/1: Erosion Analysis
Interrogatory Statement:
As discussed below, there appears to be an issue with the FractionGully 1.52 m depth data.
Please explain why the 1.52 m depth percentages are smaller than the 1.97 m and 2.42 m depth
results.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 31
2.9.1 Interrogatory Response
The interrogatory creates Figure 205-2 to summarize the percentage of realizations within
particular intervals/categories of “fraction of cover area” for six depth intervals. In general,
Figure 205-2 and the associated Table 205-1 are used to conclude that the percentage of
realizations with different fractions of cover area (below 0.03) are generally smaller for the depth
interval they label at “1.52” than for deeper depth intervals. It was expected that deeper depths
would consistently have a larger percentage of the realizations with smaller fractions. This
behavior can be attributed both to how Figure 205-2 was created (how the summaries were
calculated to create the curves), and also to the stochasticity incorporated in the creation of the
fractions at each depth interval within a single realization.
First, while it is interesting to summarize over realizations within a depth category, it is
important to keep in mind that, for one realization of the GoldSim model, a single realization is
selected from 1000 and the fractions of cover for each depth category are used for that one
realization. The fractions within each realization (one row of the GoldSim look up table) are
created as described in Appendix 10 of the Clive DU PA Model v1.4 (Neptune 2015c); they
come from the same distribution and sum to one. The fractions associated with the six or seven
deepest depth intervals are all small (generally less than ~ 0.006) and are close together due to
the relatively flat tail of the lognormal distribution used to generate them. Therefore, due to the
stochasticity in generating the fractions from the lognormal distribution, there are realizations
with a slightly higher fraction of cover in deeper intervals than in a shallower interval. For
example, for one realization the fraction is 0 for depth interval (1.52, 2.03m], but then 0.0018 for
the depth interval (2.03m, 2.53m], followed by 0 for the deepest two intervals. This stochasticity,
along with different distributions used for each realization, contributes unexpected relative
differences among depth intervals within the range of very small fractions and deep depth
intervals.
The implications of this on the PA Model are expected to be negligible given how the
realizations are used in the PA Model and the magnitude of the differences in fractions involved
in the behavior pointed out in the interrogatory. The interrogatory provides an extreme case of
erosion in Figure 205-4 and Table 205-2, created from summarizing over realizations within
depth intervals. The resulting curves are presented as the minimum, mean, and maximum, but
they do not represent particular realizations used in the GoldSim model and no longer meet the
sum-to-one constraint within individual realizations. The summaries presented in Figures 205-4
and 205-2 are interesting if the results are not over-sold as representing actual erosion behavior
of individual GoldSim realizations. For example, it would not be possible for a single realization
to have the maximum fraction observed over all realizations for every depth interval, and
statements regarding “mean erosion” using these summaries are only valid to the extent that
behavior of individual realizations used in the GoldSim model reflects the averages per depth
interval.
As described above, the general idea presented in Figure 205-2 and Table 205-1 may stem
mainly from a concern that fractions within each realization are not non-increasing with depth at
deeper depths. However, Figure 205-2 is not easy to replicate without additional information. In
particular, it is not clear how the labels for the depth intervals were chosen or how many
realizations were used to create it (just 250 to match the other plots, or all 1000 in the GoldSim
Erosion Responses for the Clive DU PA Model
23 Feb 2018 32
look up table?). The depth intervals (in mm) created for the look up table are (0,10], (10,152],
(152,305], (305,457], (457,610], (610,762], (762,914] , (914,1.07e+03], (1.07e+03,1.22e+03],
(1.22e+03,1.37e+03], (1.37e+03,1.52e+03], (1.52e+03,2.03e+03], (2.03e+03,2.53e+03],
(2.53e+03,3.04e+03], and (3.04e+03,3.54e+03]. Two of the labels used in Figure 205-2
correspond to endpoints of these intervals (1.07m and 1.52m), but the additional depth labels
(1.97m, 2.42m, 2.87m, and 3.32m) do not. They are also not midpoints. It is possible that
interpolation was used to get these from Figure 205-1, but it is not clear why this would be done
when the fractions provided specifically apply to the stated depth intervals. It is also not clear
over what intervals of fraction the percentages of realizations were calculated. Figure 4 has been
created to display the same information that Figure 205-2 is meant to represent, although using
cumulative proportions as the interrogatory did for Figure 205-3 because this is a more natural
and repeatable way to display the values in the look-up table. It uses all 1000 realizations and
seven deepest depth intervals. The “trends” in the plot over depth intervals do not match those
presented in Figure 205-1, but the general idea that curves are not “in order” by depth is the
same.

Figure 4. Empirical cumulative distribution functions (ecdf) of the values across
realizations found in the GoldSim look-up table for the seven deepest depth intervals.
Note that, in general, the deeper intervals have larger proportions of realizations with
smaller fractions. The two green curves associated with depth intervals (1.37m,
1.52m] and (1.52m, 2.03m] do not follow the order with depth, which can be
interpreted as consistent with the conclusion from Figure 205-1.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 33
Using values of simulated cover area eroded by gullies from their Table 205-1, UDEQ states,
“An eroded area of 1,087 m2 means that the entire perimeter of the embankment has eroded back
0.6 m (1.9 feet), and a 5,433 m2 eroded area means the perimeter has eroded 2.9 m (9.5 feet).”
This statement is a gross misrepresentation of the modeled erosion processes. UDEQ is
presenting a false equivalence between areally distributed erosion losses due to sheet flow and
gullying on the embankment and losses due to cliff retreat, a mass wasting process.
Radon flux
UDEQ next presents comparisons between radon flux in the embankment predicted by the Clive
DU PA Model v1.2 and several analytical solutions.
Direct comparison of late-time results from the Clive DU PA Model v1.4 with steady-state
analytical solutions is made difficult by basic differences in the assumptions involved in the
approaches. Analytical solutions typically are restricted by simplifying assumptions that allow
for closed-form solutions. For example, a constant or uniform source concentration is a typical
assumption. The PA Model, conversely, allows for movement as well as decay of radium-226,
which amounts to a non-uniform source that analytical solutions do not account for.
The analytical solution procedure for multi-layered cover systems described in NRC’s
Regulatory Guide 3.64 (NRC 1989) and applied by UDEQ begins with calculating the flux from
the waste zone assuming no cover exists (Equation 1 in SER Section 4.2.1 (SC&A 2015a)), and
then repeating a similar calculation for each layer up the column. Reg. Guide 3.64 describes this
iterative solution procedure as an “approximate method” due to the assumed boundary conditions
at each step.
Typically, comparisons of analytical solutions to numerical solutions require that the numerical
model be carefully prepared to match the assumptions of the analytical solutions. Differences in
the assumed parameter values can also be important. For example, the SER states that since “the
diffusion coefficient for the Rogers 2002 curve is simply a fixed value, UDEQ believes that the
IAEA Equation radon flux is more representative of the conditions at the proposed Federal Cell.”
However, Table 205-3 of the interrogatory presents only values derived using the Rogers (2002)
method.
Nevertheless, it is acknowledged that cross-comparison with different modeling methods can be
valuable and instructive. The interrogatory states that radon flux, though at the ground surface,
compares well to UDEQ’s application of analytical solutions, but also mentions some
discrepancies at depth. It is noted that the results from the IAEA Equation are generally within
an order of magnitude of the DU PA Model v1.4 results at all depths despite possible differences
in assumptions. UDEQ correctly states that the DU PA Model v1.4 is calibrated to focus on
agreement with known analytical solutions of the flux at the ground surface rather than fluxes
deep in the waste zone, which are generally not important to dose calculations.
Lack of evidence of significant erosional activity at the Site, planned ecological remediation of
the soils used to construct the cover (response to Interrogatory 05/2), and erosion models that
demonstrate the stability of the cover (response to Interrogatory 71/1) support the prediction that
erosion of the embankment will be minimal and will not enhance radon fluxes. Given these site-
Erosion Responses for the Clive DU PA Model
23 Feb 2018 34
specific observations and modeling results, the differences in radon flux predictions between the
DU PA Model v1.4 and UDEQ’s application of analytical solutions at depths greater than 1 m
are not meaningfully different.
The April 2015 SER (SC&A 2015a) also notes that the embankment’s performance with respect
to radon flux would be adequate even with complete removal of the cover system, as predicted
doses are several orders of magnitude below regulatory limits. Thus, the comparison of the Clive
DU PA Model v1.4 with other radon flux predictions, which show only modest differences at
relevant depths, does not detract from the conclusion that radon fluxes are adequately attenuated
by the embankment due to the depth of burial of radon-generating wastes and the prevailing site
conditions.
3.0 Conclusion
UDEQ has raised many diverse issues in this set of interrogatories. The following points
summarize this response:
• Predicted erosion performance of the Federal Cell is reasonable. Lack of evidence of
significant erosional activity at the Site, planned ecological remediation of the soils used
to construct the cover, and erosion models that demonstrate the stability of the cover
support this position. Erosion of the embankment cover will be minimal and will not
enhance radon fluxes or deep infiltration.
• Using uniform hydraulic properties to model the entire ET cover represents the only
logical outcome of applying the Benson et al. (2011) method for assigning properties.
Given their conceptual model that makes no distinction between the hydraulic properties
of storage and barrier layers, the cover can no longer be represented by a layered system
in the flow model.
• The proper rainfall intensities for the embankment surfaces have been calculated
according to NUREG/CR-4620 (Nelson et al. 1986) and incorporated into the erosion
models.
• The design for the amount of gravel to add to the soil in the top and side slopes was
determined based on sheet erosion modeling using USLE and gully projection evaluation
using the permissible velocity method of the NRC.
• The results of the SIBERIA modeling of the borrow pit provide another line of evidence,
in addition to the more conventional modeling. This is useful in demonstrating the
stability of the proposed ET cover design despite differences in the characteristics of the
borrow pit and the embankment.
• The cover surface will not have the characteristics of disturbed rangeland soils, so the
added features of RHEM for disturbed conditions are not relevant to modeling the
Federal Cell. Thus, redoing the SIBERIA erosion model calibration with RHEM will not
add to the information from the modeling results.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 35
• The SIBERIA models of the borrow pit (DU PA v1.4 Appendix 10 (Neptune 2015c)) and
a Grand Junction embankment (Smith and Benson 2016) are not comparable. No
conditions directly comparable to the Clive borrow pit simulations (semi-arid, no
vegetation, no gravel) were found in the Smith and Benson (2016) report.
• Radon fluxes are adequately attenuated by the embankment due to the depth of burial of
radon-generating wastes and the prevailing site conditions.
4.0 References
Al-Hamdan, O.Z., et al., 2014. Rangeland Hydrology and Erosion Model (RHEM)
Enhancements for Applications on Disturbed Rangelands, Hydrological Processes (2014)
1–13 doi: 10.1002/hyp.10167
Anderson, C., and J. Stormont, 2005. Gravel Admixtures for Erosion Protection in Semi-Arid
Climates, Erosion of Soils and Scour of Foundations, Proceedings of Sessions of the
Geo-Frontiers 2005 Congress, 2005, Austin TX
Anderson, C., and S. Wall, 2010a. Design of Erosion Protection at Landfill Areas with Slopes
Less than 10%. In Scour and Erosion, Geotechnical Special Publication (GSP) No. 210,
Proceedings of the 5th Annual Conference on Scour and Erosion (ICSE-5) 2010,
November 7–10, edited by S.E. Burns, et al., pp. 1054–1063, American Society of Civil
Engineers, San Francisco CA
Anderson, C., and S. Wall, 2010b. Erosion Protection at Landfill Slopes Greater than 10%. In
Scour and Erosion, Geotechnical Special Publication (GSP) No. 210, Proceedings of the
5th Annual Conference on Scour and Erosion (ICSE-5) 2010, November 7–10, edited by
S.E. Burns, et al., pp. 1064–1073, American Society of Civil Engineers, San Francisco
CA
Belnap, J., and D.A. Gillette, 1998. Vulnerability of Desert Biological Soil Crusts to Wind
Erosion: The Influences of Crust Development, Soil Texture, and Disturbance, Journal of
Arid Environments 39 (2) 133–142
Belnap, J., et al., 2001. Biological Soil Crusts: Ecology and Management, Technical Reference
1730-2, United States Department of the Interior, Bureau of Land Management, Denver
CO, 2001
Benson, C.H., et al., 2011. Engineered Covers for Waste Containment: Changes in Engineering
Properties and Implications for Long-Term Performance Assessment, NUREG/CR-7028,
United States Nuclear Regulatory Commission, Washington DC, December 2011
Coon, W.F., 1998. Estimation of Roughness Coefficients for Natural Stream Channels with
Vegetated Banks, U.S. Geological Survey Water-Supply Paper 2441, prepared in
Erosion Responses for the Clive DU PA Model
23 Feb 2018 36
cooperation with the New York State Department of Transportation, U.S. Geological
Survey, U.S. Department of the Interior, Denver CO, 1998
CWCB, 2006. Chapter 13, Hydraulic Analysis and Design, Section 1 Open Channels. In
Colorado Floodplain and Stormwater Criteria Manual, pp. CH13-100–CH13-F124,
Colorado Water Conservation Board, Denver CO
DOE, 2016. Prototype Hanford Barrier 1994 to 2015, DOE/RL-2016-37, Revision 0, United
States Department of Energy, Richland Operations Office, Richland WA, March 2016
EnergySolutions, 2013. Utah Radioactive Material License (RML UT2300249) Updated Site-
Specific Performance Assessment, Revision 1, prepared for Utah Division of Radiation
Control, EnergySolutions LLC, Salt Lake City UT, December 2013
EnergySolutions, 2015. Utah Radioactive Material License (RML UT2300249) Updated Site-
Specific Performance Assessment, Revision 2, prepared for Utah Division of Radiation
Control, EnergySolutions LLC, Salt Lake City UT, January 2015
EPA, 1989. Technical Guidance Document: Final Covers on Hazardous Waste Landfills and
Surface Impoundments, EPA 530-SW-89-047, United States Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington DC, July 1989
EPA, 1991. Seminar Publication, Design and Construction of RCRA/CERCLA Final Covers,
EPA/625/4-91/025, United States Environmental Protection Agency, Washington DC,
May 1991
Haan, C.T., et al., 1994. Design Hydrology and Sedimentology for Small Catchments, 1st
Edition, Academic Press, San Diego CA
HAL, 2017a. Response to Interrogatory 198, Hansen, Allen, & Luce Inc., South Jordan UT,
December 2017
HAL, 2017b. Response to Interrogatory 201, Hansen, Allen, & Luce Inc, South Jordan UT,
December 2017
HAL, 2018. Response to Interrogatory 191, Hansen, Allen, & Luce Inc., South Jordan UT,
February 2018
Israelsen, C.E., et al., 1984. Erosion and Sedimentation in Utah: A Guide for Control, Reports.
Paper 372, Utah Water Research Laboratory, Utah State University, Logan UT, 1984
Erosion Responses for the Clive DU PA Model
23 Feb 2018 37
Jensen, D.T., 1995. Final Report, Probable Maximum Precipitation Estimates for Short-
Duration, Small-Area Storms in Utah, Utah Climate Center, Utah State University,
Logan UT, October 1995
Nearing, M.A., et al., 2011. A Rangeland Hydrology and Erosion Model, Transactions of the
ASABE 54 (3) 1–8
Nelson, J.D., et al., 1986. Methodologies for Evaluating Long-Term Stabilization Designs of
Uranium Mill Tailings Impoundments, NUREG/CR-4620, ORNL/TM-10067, United
States Nuclear Regulatory Commission (NRC), Washington DC, June 1986
Neptune, 2014. Final Report for the Clive DU PA Model, Clive DU PA Model v1.2, NAC-
0024_R2, Neptune and Company, Inc., Los Alamos NM, August 2014
Neptune, 2015a. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015b. FEPS Analysis for the Area G Radiological Risk Assessment, NAC-0038_R0,
Neptune and Company Inc., Los Alamos NM, March 2015
Neptune, 2015c. Erosion Modeling for the Clive DU PA, Clive DU PA Model v1.4, NAC-
0017_R4, Neptune and Company Inc., Los Alamos NM, October 2015
NOAA, 1984. Probable Maximum Precipitation Estimates, Colorado River and Great Basin
Drainages, Hydrometeorological Report No. 49, National Oceanic and Atmospheric
Administration, Silver Spring MD, 1984
NRC, 1986. Update of Part 61 Impacts Analysis Methodology, Methodology Report, Volume 1,
NUREG/CR-4370, United States Nuclear Regulatory Commission, Washington DC
NRC, 1989. Calculation of Radon Flux Attenuation by Earthen Uranium Mill Tailings Covers,
Regulatory Guide 3.64, United States Nuclear Regulatory Commission, Washington DC,
June 1989
NRC, 2002. Design of Erosion Protection for Long-Term Stabilization, NUREG-1623, United
States Nuclear Regulatory Commission, Washington DC, 2002
Rogers, T., 2002. A Change in Envirocare’s Disposal Cell Design, Waste Management 2002
Conference, February 2002, Tucson AZ
SC&A, 2015a. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Erosion Responses for the Clive DU PA Model
23 Feb 2018 38
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
1, SC&A Inc., Vienna VA, April 2015
SC&A, 2015b. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
2, SC&A, Vienna VA, April 2015
Simanton, J.R., et al., 1984. Effects of Rock Fragments on Erosion of Semiarid Rangeland Soils,
SSSA Special Publication No. 13, Soils Science Society of America, Madison WI, 1984
Skinner, C., et al., 2017. LEMSI—The Landscape Evolution Model Sensitivity Investigation,
Geophysical Research Abstracts 19 (EGU2017-15699)
Smith, C.L., and C.H. Benson, 2016. Influence of Coupling Erosion and Hydrology on the Long-
Term Performance of Engineered Surface Barriers, NUREG/CR-7200, United States
Nuclear Regulatory Commission (NRC), Washington DC, May 2016
Stenseng, S.E., and P.A. Nixon, 1995. Long-Term Cover Design for Low-Level Radioactive and
Hazardous Waste Sites as Applied to the Rocky Flats Environmental Technology Site
Solar Evaporation Ponds, 50th Purdue Industrial Waste Conference Proceedings, 1995
SWCA, 2012. Vegetated Cover System for the EnergySolutions Clive Site: Literature Review,
Evaluation of Existing Data, and Field Studies Summary Report, prepared for
EnergySolutions, SWCA Environmental Consultants, Salt Lake City UT, August 2012
SWCA, 2013. EnergySolutions Updated Performance Assessment—SWCA’s Response to First
Round DRC Interrogatories, SWCA Environmental Consultants, September 2013
Temme, A.J.A.M., et al., 2009. Can Uncertain Landscape Evolution Models Discriminate
Between Landscape Responses to Stable and Changing Future Climate? A Millennial-
Scale Test, Global and Planetary Change 69 (2009) 48–58
USDA, 1978. Predicting Rainfall Erosion Losses, A Guide to Conservation Planning,
Agriculture Handbook Number 537, United States Department of Agriculture,
Washington DC, December 1978
USDA, 1986. Urban Hydrology for Small Watersheds, Technical Release 55 (TR-55), United
States Department of Agriculture, Washington DC, June 1986
Utah DEQ, 2017. Division of Waste Management and Radiation Control, EnergySolutions Clive
LLRW Disposal Facility License No: UT2300249; RML #UT 2300249, Amended and
New Interrogatories Related to Clive DU PA Modeling Report Version 1.4 Dated
Erosion Responses for the Clive DU PA Model
23 Feb 2018 39
November 2015, Utah Department of Environmental Quality (DEQ), Salt Lake City UT,
May 2017
Waugh, W.J., and G.N. Richardson, 1997. Ecology, Design and Long-Term Performance of
Surface Barriers: Applications at a Uranium Mill Tailings Site. In Barrier Technologies
for Environmental Management: Summary of a Workshop, edited by Committee on
Remediation of Buried and Tank Wastes and National Research Council, pp. D-54–D-67,
National Academy Press, Washington DC

Erosion Responses for the Clive DU PA Model
23 Feb 2018 40
Appendix A. HAL 2018

Erosion Responses for the Clive DU PA Model
23 Feb 2018 41

DRC Interrogatory CR R313-25-7(2)-191/3
Selected Statements from DEQ Conclusion from April 2015 SER, Appendix C:
Interrogatory 191 requested EnergySolutions to provide additional information about the
ability of steep side slopes to resist gully erosion. After reviewing…DEQ believes that the key
analysis is….Appendix D uses both RUSLE and REHM to calculate rill or sheet erosion, with
similar results. Both are well below the EPA’s criteria for ……cover systems. One problem
with the Appendix D analysis is that it does not describe how the values for the various
RUSLE and REHM parameters were selected. For example, the RUSLE has R, K, L, S, and C
parameters, but only L and S are functions of the embankment’s design, so the basis for
selecting the other parameters is not clear.
In conclusion, the analysis performed by HAL may or may not be correct, but before DEQ can
accept it, each value selected and used in the analysis needs to be justified.
HAL Response: February 12, 2018
Universal Soil Loss Equation (USLE)
The USLE estimates average annual soil losses from erosion. In order to clarify the
methodology used in the analysis, all references to RUSLE have been change to refer to USLE.
While the equations are the same, the use of the acronym RUSLE could imply the use of the
software program developed by the USDA for factor determination. The software package was
not used because factors from a Utah-specific publication were relied on for the analysis as
described below.
The review from DEQ indicated that the previous interrogatory response was lacking in the
description of the selection of the parameters. Therefore, the summary of the methodology has
been enhanced to provide additional background information. The USLE methodology is
commonly used to determine the long-term stability of slopes and is an industry-standard means
for design of erosion control. Guidance given by the EPA states that “The U.S. Department of
Agriculture’s (USDA’s) Universal Soil Loss Equation is recommended as the tool to evaluate
erosion potential” (US EPA Seminar Publication, 1991). The basis for this approach comes from
the theory that “if adequate protection is provided to control sheet erosion, then rills and gullies
will never form from rainfall” (Israelsen et al, 1984). Generally, the USLE equation is defined
as:
A = R*K*LS*C*P
Where:
A = the average soil loss per unit area, expressed in tons/acre/year
R = the rainfall/runoff factor, which is the number of rainfall units for rainfall energy and runoff
and snowmelt
K = soil erodibility factor in tons per acre per year per unit of R
LS = topographic factor (length and steepness of the slope)
C = the cover and management factor (equivalent to the VM factor), which is the ratio of soil
loss from an area with a given cover and management relative to that from an identical area in
continuous fallow
P = the supporting conservation practice factor, in this case assumed to be equal to 1
This procedure and site-specific factors are described in “Erosion and Sedimentation in Utah – A
Guide for Control” (Israelsen, 1984) and “Design Hydrology and Sedimentation for Small
Catchments” (C.T. Haan et al, 1994). The computed average sheet erosion soil loss is presented
in Table 1.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 42
R Factor
The R factor (6) is selected based on the Utah-specific Iso-Erodent (R) mapping provided in the
Utah Water Research Laboratory report (Israelsen, 1984) which is also provided on the following
page.
K Soil Erodibility Factor
The K values (0.18 and 0.07 for top slope and side slope, respectively) are based on the Unit 4-
specific material characteristics with the top slope and side slope gravel admixtures together with
the Wischmeier nomograph as described in the methodology presented in the Israelsen and Hann
procedures and shown in the figure below. A particle size analysis using the hydrometer method
was performed on the Unit 4 clay to determine the percent silt and very fine sand (see figure
below) for use with the Wischmeier nomograph. The results of the hydrometer analysis are
attached to this response.
.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 43

Erosion Responses for the Clive DU PA Model
23 Feb 2018 44
C Factor (VM Factor)
The C parameters used in the equation for both the 4% and 20% slopes were based on Table 9
and Table 10 of the US Department of Agriculture Handbook Number 537 “Predicting Rainfall
Erosion Losses, A Guide to Conservation Planning” and are shown in the figures below. The C
factor for the top slopes (0.2) is based on the sparse vegetative cover naturally found in the areas
immediately surrounding the Clive facility (Table 10 of Handbook Number 537, No Appreciable
Canopy, Type G with 20% ground cover) and the Unit 4 gravel admixture. The C factor for the
side slope is based on the higher percentage of gravel in the side slope gravel admixture (50%
gravel). The 50% gravel admixture on the side slopes results in a pseudo-gravel mulch once
some of the fines have been removed. Therefore, a C factor of 0.02 was selected (Table 9
Handbook Number 537, Crushed Stone ¼ to 1 ½ in, Mulch Rate of 240 tons/acre, Land Slope of
21-33%, with a length limit of 200 feet).

Erosion Responses for the Clive DU PA Model
23 Feb 2018 45
TABLE 1
EFFECTS OF EROSION - AVERAGE SOIL LOSS ANALYSIS USING USLE
(TOP SLOPE WITH UNIT 4 CLAY W/ 15% GRAVEL ADMIXTURE AND SIDE
SLOPE WITH UNIT 4 CLAY AND 50% GRAVEL ADMIXTURE)

Slope Segment
R
(ft tons/ac/hr)
K
(tons/ac/EI)
L
(ft)
S
(%)
C A
(tons/ac/yr)
Total Soil
Loss
(mm/year)
Top Slope (4%) 6 0.18 942 (4%) 4% 0.2 0.25 0.24
overall 0.026
Side Slope (20%) 6 0.07 188 (20%) 20% 0.02 0.19

The Rangeland Hydrology and Erosion Model (RHEM) Model Parameters
The RHEM model was used in HAL’s interrogatory response entitled “DRC RFI Section 4.0
Erosion” completed in 2013 in order to corroborate the results of the USLE analysis. Since the
completion of that effort, the RHEM on-line model has been updated. The latest version of the
model (Version 2.3) lacks some of the functionality of the earlier model. More specifically,
slope lengths are not able to be directly input as they were previously and therefore a direct
comparison with the calculation completed using USLE is not possible. This appears to be an
error within the program since the descriptions of the model available on the website include
slope length as an input parameter option. Attempts to reach the contacts listed for the RHEM
model in order to resolve the issue were unsuccessful. The model defaults to a slope length of 50
meters (164 feet). Additional parameter options for cover characteristics appear to have been
added since 2013, including options for a distinction between foliar cover and basal plant cover
as well as rock cover, litter cover and biological soil crust cover. Since the tool is a web-based
model and not stand alone software, the version of the online tool used in the 2013 analysis is no
longer available. The parameters described in the 2013 interrogatory for vegetative conditions,
slope grades, climate station, soil texture class, and slope lengths are still valid for the analysis
completed at the time but the results reported therein can no longer be validated.

Literature Cited

Bray, D. I., 1979. Estimating Average Velocity in Gravel-Bed Rivers. Proc. Am. Soc. Civ. Engrs, J.
Hydraul. Div., 105 (HY9), 1103-1122.

Haan, C.T., Barfield, B.J., Hayes, J.C., 1994. Design Hydrology and Sedimentology for Small
Catchments. Academic Press, Inc, San Diego, California.

Israelsen, C. Earl, Fletcher, Joel E., Haws, Frank W., Kisraelsen, Eugene K., 1984. Erosion and
Sedimentation in Utah: A Guide for Control. Utah Water Research Laboratory, Logan,
Utah.
Johnson, T. L. 2002. Design of Erosion Protection for Long-Term Stabilization, Final Report,
NUREG-1623, U.S. Nuclear Regulatory Commission, Office of Nuclear Material Safety
and Safeguards, Washington, D.C.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 46

Jones, Gordon L., 2012. PMF and 100-year Storm Analysis for Clive, Utah. Hansen, Allen &
Luce, Midvale, Utah.

Mazor, G., G.J. Kidron, A. Vonshak, and A. Albeliovich, 1996. The role of cyanobacterial
exopolysaccharides in structuring desert microbial crusts. FEMS Microbiology Ecology.

Nearing, M.A., H. Wei, J.J. Stone, F.B. Pierson, K.E. Spaeth, M.A. Weltz, D.C. Flanagan and M.
Hernandez, 2011. A Rangeland Hydrology and Erosion Model. Soil & Water Division
of ASABE in March 2011 Edition, Vol. 54(3).

Nelson, J. D., R. L. Volpe, R. E. Wardwell, S. H. Schumm, and W. P. Staub. 1983. Design
Considerations for Long-Term Stabilization of Uranium Mill Tailings Impoundments.
NUREG/CR-3397 (ORNL-5979), U.S. Nuclear Regulatory Commission, Washington,
D.C.

U.S. Army Corps of Engineers, 1977. Probable Maximum Precipitation Estimates, Colorado
River and Great Basin Drainages, Hydrometeorological Report No. 49. US Departement
of Army Corps of Engineers, Silver Spring, Maryland.

U.S. Department of Agriculture, 1978. Predicting Rainfall Erosion Losses, A Guide to
Conservation Planning, Handbook 537. U.S. Department of Agriculture, Washington
D.C.

US EPA Seminar Publication, 1991. Design and Construction of RCRA/CERCLA Final Covers.

U.S. Geological Survey, 1973. Aragonite NW, Utah; Aragonite, Utah; Hastings Pass, Utah; and
Low, Utah Quadrangle Maps.

U.S. Geological Survey, 1987. Bonneville Salt Flats, Utah, 30 x 60 Minute Quadrangle Map.

U.S. Nuclear Regulatory Commission, 2002. NUREG-1623 - Design of Erosion Protection for
Long-Term Stabilization. Office of Nuclear Material Safety and Safeguards,
Washington, DC.

Erosion Responses for the Clive DU PA Model
23 Feb 2018 47
Appendix B. HAL 2017a
Erosion Responses for the Clive DU PA Model
23 Feb 2018 48
DRC Interrogatory CR R313-25-25(4)-198/1
DRC Interrogatory Statement:
Please provide the design bases and justification for the amount and sizing of the gravel in the
top and side slopes of the Federal Cell. The proposal for the gravel admixture in the top slope
(15%) appears too small. Also, please provide evidence for existing semi-arid or arid sites
where only 15% gravel has been added to form a successful cover-system surface layer for a
landfill. Please describe actual analog sites where 50% gravel for side slopes has been
demonstrated to be effective against erosion.
HAL Response: December 18, 2017
The design basis for the amount of gravel to add to the soil in the top and side slopes were first
determined based on the analysis of the long term sustainability of the slopes due to sheet erosion
over time. The methods that were used are the Revised Universal Soil Loss Equation (RUSLE)
and the Rangeland Hydrology and Erosion Model (RHEM). These methods produce an estimate
of the average annual soil loss in terms of tons per acre per year. With the gravel admixed into
the Unit 4 clay soil, average annual soil losses were calculated to be about 0.24 tons/acre/year
using RUSLE which was also closely matched by the RHEM. This amount of loss is almost 10
times less than what has been recommended in EPA guidance for covers for hazardous waste
facilities. A more detailed discussion of the methodologies and results is provided in the
response to DRC Interrogatory CR R313-25-7(2)-191-3.
Once the slope was determined to be stable from an average soil loss perspective, checks on the
gully erosion potential were completed based on the calculation of predicted velocities and
comparing them to permissible velocity according to the method presented in NUREG-1623
“Design of Erosion Protection for Long-Term Stabilization”. This second assessment based on
an extreme intense rainfall event is recommended in NUREG-1623 because of the potential for
significant damage to cover systems from such large events. In fact, this extreme intense rainfall
event is known as the Probable Maximum Precipitation (PMP) and is so extreme that “the point
precipitation data base, even if maximized for PMP moisture potential, shows no observed
values even close to the 10-inch PMP estimate (the generalized, local storm PMP estimates in
Utah)” (Jensen, 1995). Additionally, the methodology applies a flow concentration factor (F)
which NUREG-1623 recommends to be 3. Therefore, the extreme event that is multiples of the
highest ever recorded rainfall intensity in the area is compounded by an additional safety factor.
The end result of this process is an analysis that is meant to produce a highly conservative value.
Flow velocities on the top and side slopes of the CAW embankment during the PMP event were
predicted to be 2.37 and 2.07 ft/sec, respectively, as discussed in more detail in the response to
DRC Interrogatory CR R313-25-7(2)-191-3. The acceptable Maximum Permissible Velocity
(MPV) was selected from tables provided in NUREG/CR-4620. Under this method the slope is
stable if the calculated velocity (V, the velocity resulting from a PMP in this application) is less
than the MPV. By contrast, if velocities exceed the MPV, the slope will experience excessive
erosion that will lead to the formation of gullies. HAL found that the calculated velocities
resulting from the PMP did not exceed the prescribed permissible velocity.
Reference Review
Erosion Responses for the Clive DU PA Model
23 Feb 2018 49
The methods discussed above have been published and/or accepted by the US EPA and US
Nuclear Regulatory Commission. Other methods referred to in the interrogatory, on the other
hand, have not been adopted, published or referred to in guidance by federal agencies. While the
works cited are part of productive ongoing academic research and study, they are not conclusive
in their findings and should instead be the focus of additional research. A summary of each is
provided below:
“Gravel Admixture for Erosion Protection in Semi-Arid Climates, Erosion of Soils and Scour of
Foundations” (Anderson, C., and J. Stormont, 2005) – This paper was published as part of the
proceedings of sessions of the GeoFrontiers 2005 Congress held in Austin, Texas. Along with
some other conference proceedings articles, these early attempts to define a process for
admixture design is very helpful in that it outlines practical steps to determine gravel size, gravel
percentage and admixture thickness. This approach, however, has limitations that result from the
lack of empirical evidence to back up the steps and applications of many of the equations
presented in the paper. The paper states that “the design method for gravel admixtures presented
here may serve as an outline for further erosion investigations and provide guidance for future
designs of gravel-soil admixture layers”. Therefore, the paper as written was not meant to be
used as a proven method for admixture design but rather a starting point for further research and
investigation.
“Design of Erosion Protection at Landfill Areas with Slopes Less than 10%” (Anderson, C., and
Wall, S., 2010) – This paper is very similar to the aforementioned paper by Anderson and
Stormont where it comes from the proceedings of a conference, in this case the International
Conference on Scour and Erosion held in 2010. Much of the background language of this paper
is the same as the 2005 article by Anderson and Stormont. Similarly, the methodology presented
is very similar to the steps provided in 2005 with some changes. Again, the paper does not claim
this to be a proven method but indicates it is taking steps to test its outcomes. It states that “the
procedure described here is being applied to a landfill cover soon to be constructed in southern
Nevada. The Nevada project will provide the first large scale application of the procedure.”
HAL was unable to find information regarding empirical data on a smaller scale nor was final
information on the results of the large scale Nevada project able to be found. Without this
information or additional discussion in the paper, the limits of the equation and how the
parameters of success or failure are defined remain unknown. For example, assuming the
Nevada project site is shown to be successful with the 40% gravel admixture there is nothing that
would indicate if a lower percentage of gravel admixture, such as 25%, would also be adequate.
Additionally, the suitability of the design and how success is defined is not sufficiently described
in the paper to know whether the absence of observed gully erosion constitutes a successful
design or if it is acceptable to have the formation of rills and gullies as long as the depth does not
exceed the admixture layer thickness calculated using the method. This is an important
distinction since the end result of the design process presented in the paper is the calculation of
the thickness of the admixture layer which is presented as being dependent on the percentage of
gravel in the admixture. This specific paper does not provide limits or conclusive guidance to
the reliability of the method and therefore should serve as the starting point for further
investigation and research.
Erosion Responses for the Clive DU PA Model
23 Feb 2018 50
“Erosion Protection at Landfill Slopes Greater than 10%” (Anderson, C. and Wall, S.) – This
paper focuses on riprap on side slopes. The design proposed by EnergySolutions does not
include riprap side slopes; therefore, this paper does not apply.
“Long-Term Cover Design for Low-Level Radioactive and Hazardous Waste Sites as Applied to
the Rocky Flats Environmental Technology Site Solar Evaporation Ponds” (Stenseng, S.E. and
Nixon, P.A., 1997) – The paper describing the design used at the Rocky Flats Environmental
Technology Site was included as part of the proceedings from the 50th Industrial Waste
Conference held in May 1997. The paper includes a brief description of the design basis for the
5% top slopes and 20% side slopes and the 40% gravel admixture. The admixture gravel content
appears to have been selected instead of calculated as the result of a defined process. While
instructive regarding the specific cover design discussed in the paper, no additional guidance is
provided to guide application at other sites.
“Ecology, Design, and Long-Term Performance of Waste-Site Covers: Applications at a
Uranium Mill Tailings Site” (Waugh, W.J. and Richardson, G.N., 1997) – This paper is similar
to the Stenseng and Nixon paper discussed previously in that it contains a summary of the design
parameters but does not provide a specific and applicable methodology for the design of a soil
gravel admixture with regard to erosion protection.
Sites with Top Slopes Comprised of 15% Gravel Admixture
There are some examples of sites that have employed slopes with 15% admixture and give some
measure of proof that a 15% admixture can be effective at controlling erosion. A test site was
established at Hanford and is referred to as the Hanford Prototype Surface Barrier (PSB). The
purpose of the PSB test site was to evaluate surface barrier constructability, construction costs,
and physical and hydrologic performance at field scale (US Department of Energy, 2016). This
field-scale test cover was installed in 1994 and is comprised of a top erosion protection layer
made up of a silt loam admixed with 15% pea gravel. The top slope of 2% in this case is slightly
less than the top slopes of 2.5% and 4% proposed by EnergySolutions but is an example of a
cover system that relies on the same amount gravel admixture. A recent review completed by
the US Department of Energy in 2016 of the data collected at the site through 2015 concluded
that:
“The 19-year PHB record showed practically no evidence of wind or water erosion of the
ETC barrier, despite 3 years of triple the mean annual precipitation; three simulated
1000-year-return, 24-hour precipitation events; and an intense, controlled fire that
burned off all vegetation across half the barrier surface….Even in the absence of
vegetation (e.g., following a fire), the pea gravel added to the silt loam protected the
barrier surface from wind and water erosion….Overall, the monitoring results have
confirmed that the PHB design is resistant to water and wind erosion and that resistance
is expected throughout the barrier’s 1000-year design life”
Another test site using the same cover system was established at Hill Air Force Base in 1994
though little information has been published about that site to date.
Sites with Side Slopes Comprised of 50% Gravel Admixture
Erosion Responses for the Clive DU PA Model
23 Feb 2018 51
No sites were found that have used a gravel admixture on side slopes at or above 20%.
Similarly, there were no methodologies found that specifically address the calculation of gravel
admixtures for slopes greater than 10% other than the general methods found in NUREG-1623
and NUREG/CR-4620. For slopes over 9%, Simanton et al. found that the rate of water erosion
diminishes exponentially with increasing cover of rock fragments. The effect of biological soil
crust is also difficult to quantify but is expected to establish itself along the slopes as has been
observed along natural slopes in the Clive area. The observation of natural analogs in the Clive
vicinity could prove helpful in demonstrating the stability of steeper slopes although none would
be available with such a high gravel content (50%) as the proposed side slope design. The
required lengthy post-closure care period will perhaps provide the greatest opportunity for
verification of the design methodology described previously.
The design process that was completed to determine the acceptability of the 50% gravel
admixture on the 20% side slopes is in accordance with EPA and Nuclear Regulatory
Commission guidance. There was no requirement found in the guidance procedures that existing
sites be used as evidence to support proposed designs. While it is acknowledged that such
evidence would be helpful, each site is different and contains unique design constraints based on
physical layout, climate, material availability and project goals that make determinations based
on sound methodologies necessary.

Literature Cited

Anderson, C., and J. Stormont, 2005. Gravel Admixtures for Erosion Protection in Semi-Arid Climates,
Erosion of Soils and Scour of Foundations. Proceedings of Sessions of the Geo-Frontiers 2005
Congress, Austin, Texas, 1-12.

Anderson, C., and S. Wall, 2010. Design of Erosion Protection at Landfill Areas with Slopes Less than
10%. Scour and Erosion, Geotechnical Special Publication (GSP) No. 210, ed. S.E. Burns, S.K.
Bhatia, C.M.C. Avila, and B.E. Hunt, 1054-1063, American Society of Civil Engineers,
Proceedings of the 5th Annual Conference on Scour and Erosion (ICSE-5) 2010, November 7-10,
2010, San Francisco, California, United States.

Anderson, C. and S. Wall, 2010. Erosion Protection at Landfill Slopes Greater Than 10%. Scour and
Erosion, Geotechnical Special Publication (GSP) No. 210, ed. S.E. Burns, S.K. Bhatia, C.M.C.
Avila, and B.E. Hunt, 1064-1073, American Society of Civil Engineers, Proceedings of the 5th
Annual Conference on Scour and Erosion (ICSE-5) 2010, November 7-10, 2010, San Francisco,
California, United States.

Bray, D. I., 1979. Estimating Average Velocity in Gravel-Bed Rivers. Proc. Am. Soc. Civ. Engrs, J.
Hydraul. Div., 105 (HY9), 1103-1122.

Haan, C.T., Barfield, B.J., Hayes, J.C., 1994. Design Hydrology and Sedimentology for Small
Catchments. Academic Press, Inc, San Diego, California.

Jensen, Donald T., 1995. Probable Maximum Precipitation Estimates for Short-Duration, Small-
Area Storms in Utah. Utah Climate Center, Utah State University, Logan, Utah.

Erosion Responses for the Clive DU PA Model
23 Feb 2018 52
Johnson, T. L. 2002. Design of Erosion Protection for Long-Term Stabilization, Final Report,
NUREG-1623, U.S. Nuclear Regulatory Commission, Office of Nuclear Material Safety
and Safeguards, Washington, D.C.

Nelson, J. D., S. R. Abt, R. L. Voipe, D. van Zyl, N.E. Hikle, and W. P. Staub. 1986.
Methodologies for Evaluating Long-Term Stabilization Designs of Uranium Mill
Tailings Impoundments. NUREG/CR-4620, U.S. Nuclear Regulatory Commission,
Washington, D.C.

Simanton, J.R., E. Rawitz and E.D. Shirley, 1984. Effects of Rock Fragments on Erosion of
Semiarid Rangeland Soils. SSSA Special Publication No. 13 Soils Science Society of
America, Madison, Wisconsin.

Stenseng, S.E. and P.A. Nixon. Long-Term Cover Design for Low-Level Radioactive and
Hazardous Waste Sites as Applied to the Rocky Flats Environmental Technology Site
Solar Evaporation Ponds. Wukasch, R.F., 1995, Proceedings of the 50th Industrial Waste
Conference May 8-10, Purdue Research Foundation, Office of Technology Transfer,
West Lafayette, Indiana.

United States Environmental Protection Agency, 1989. Technical Guidance Document: Final
Covers on Hazardous Waste Landfills and Surface Impoundments. Office of Solid Waste
and Emergency Response, Washington, D.C.

Waugh, W.J. and G.N. Richardson, 1997. Ecology, Design and Long-Term Performance of
Surface Barriers: Applications at a Uranium Mill Tailings Site. Committee on
Remediation of Buried and Tank Wastes, National Research Council, eds. Barrier
Technologies for Environmental Management: Summary of a Workshop, The National
Academies Press, Washington, D.C.

Erosion Responses for the Clive DU PA Model
23 Feb 2018 53

Appendix C. HAL 2017b

Erosion Responses for the Clive DU PA Model
23 Feb 2018 54
DRC Interrogatory CR R313-25-25(4)-201/1
DRC Interrogatory Statement:
It is not clear that the Probable Maximum Precipitation (PMP) was determined using the
procedures outlined in the National Oceanic and Atmospheric Administration and U.S. Army
Corps of Engineers publication Hydrometeorological Report No. 49 (HMR 49) (1977).
According to EnergySolutions, these procedures resulted in a “1-hour PMP rainfall intensity
of 9.9 inches (Jones, 2012).” However, DWMRC finds that a value of 9.8 or 9.9 inches is not
the intensity, but rather the 1-hour PMP, or the maximum precipitation expected over 1
square mile when averaged over an hour.
Please re-calculate the PMP using NUREG/CR-4620 as outlined below.
HAL Response: December 19, 2017
Gully erosion potential was initially checked based on the calculation of permissible velocities
according to the method presented in NUREG-1623 “Design of Erosion Protection for Long-
Term Stabilization”. As pointed out in the Interrogatory Statement, it is acknowledged that the
methodology utilized previously failed to incorporate the proper rainfall intensity as outlined in
NUREG/CR-4620.
In order to determine the PMP intensity as outlined in the guidance in NUREG/CR-4620 it is
necessary to first calculate the time of concentration for representative drainage areas for both
the top and side slopes. The methodology outlined in Technical Release 55 “Urban Hydrology
for Small Watersheds” was used to calculate the time of concentration for each representative
slope.
TR-55 describes three types of drainage flow: sheet flow, shallow concentrated flow, and open
channel flow. Sheet flow is defined as flow over planar surfaces at very shallow depths for up to
300 feet. After a maximum of 300 feet, the flow transitions to shallow concentrated flow. Open
channels were not included in the calculations due to the absence of designed channels. The
same Manning’s roughness coefficient calculated to be 0.05 for use in the NUREG-1623
methodology using an empirical equation for channels with gravel beds with shallow flow depths
(Bray, 1979) was also used to represent the sheet flow roughness. It was decided to use the value
calculated using the Bray method (0.05) instead of a higher value found in other publications for
overland flow using sparse vegetative cover in order to be more protective. The above described
methodology was applied to both the Class A West Cell and the Federal Cell.
CALCULATED TIME OF CONCENTRATION USING TR-55 METHODOLOGY
FOR CLASS A WEST CELL

Slope Description Total
Length
(ft)
Slope
(ft/ft)
Sheet
Flow
Length
(ft)
Sheet Flow
Manning’s n
Shallow
Concentrated
Flow Length
(ft)
Total Time of
Concentration
(min)
Top Slope (4%) 942 0.04 300 0.05 642 17.4
Side Slope (20%) 188 0.20 188 0.05 0 4.6
CALCULATED TIME OF CONCENTRATION USING TR-55 METHODOLOGY
Erosion Responses for the Clive DU PA Model
23 Feb 2018 55
FOR FEDERAL CELL

Slope Description Total
Length
(ft)
Slope
(ft/ft)
Sheet
Flow
Length
(ft)
Sheet Flow
Manning’s n
Shallow
Concentrated
Flow Length
(ft)
Total Time of
Concentration
(min)
Top Slope (2.5%) 521 0.025 300 0.05 221 16.9
Side Slope (20%) 178.5 0.20 178.5 0.05 0 4.4

The rainfall depth determined using the methods outlined in Table 6.3a of the US Army Corps of
Engineers publication Hydrometeorological Report No. 49 (HMR 49) for determining an average
1-hour 1 square mile PMP is 9.9 inches. NUREG/CR-4620 gives the following steps for
determining rainfall intensity from the PMP:
PMP rainfall depth = (% PMP) x (PMP)

Where:
PMP rainfall depth = rainfall depth after adjusting the PMP for the duration, inches
% PMP = the region-specific % PMP based on the duration, %
PMP = Probable Maximum Precipitation, inches (9.9 inches using Table 6.3a from HMR 49)

The % PMP was obtained from a report commissioned by the State of Utah Division of Water
Resources and Division of Water Rights, State Engineer’s Office entitled “Probable Maximum
Precipitation Estimates for Short-Duration, Small-Area Storms in Utah” (Jensen, 1995). The
report states that “because the depth-duration values published in HMR 57 were very much like
those derived in this study, the published HMR 57 depth-duration values were accepted for
operational use in Utah.” Therefore, the % PMP was determined using the Utah-specific depth-
duration found in Figure 8 (Jensen, 1995) and HMR 57 instead of the general depth-duration
relationships described in HMR 49 and NUREG/CR-4620.
Rainfall intensity (i) for the purpose of use in the rational equation is then calculated as:
i = PMP rainfall depth x 60 / rainfall duration

Where:
i = rainfall intensity, inches/hour
rainfall duration = time of concentration calculated as described above, minutes
The results of the above calculation is summarized in the table below.

Erosion Responses for the Clive DU PA Model
23 Feb 2018 56
PMP RAINFALL INTENSITY
CLASS A WEST CELL

Slope Description Rainfall
Duration (Tc)
(minutes)
% PMP
(%)
Rainfall
Intensity (i)
(inches/hr)
Top Slope (4%) 17.4 53.8% 18.4
Side Slope (20%) 4.6 15.3% 19.8

PMP RAINFALL INTENSITY
FEDERAL CELL

Slope Description Rainfall
Duration (Tc)
(minutes)
% PMP
(%)
Rainfall
Intensity (i)
(inches/hr)
Top Slope (2.5%) 16.9 52.1% 18.3
Side Slope (20%) 4.4 14.7% 19.8

As opposed to the projection of the long-term effects of precipitation over time due to sheet
erosion, the effects of gully erosion are determined by the consideration of a large single rainfall
event. The equation given in NUREG-1623 for the procedure to determine the peak runoff flow
rate is:
Q = Fci A

Where:
Q = Runoff Rate, cfs/ft
F = Flow concentration factor (recommended to use a factor of 3 by NRC staff in NUREG-1623)
c = dimensionless runoff coefficient
i = rainfall intensity, in/hr
A = catchment area, acres (using a 1 foot wide strip along the length of the slope)
Using this flow rate, a flow depth is calculated by solving the Manning Equation for normal
depth on a one foot wide strip along the slope length. The Manning’s n value was calculated to
be 0.05 using an empirical equation for channels with gravel beds with shallow flow depths
(Bray, 1979). The derivation of the Manning Equation to solve for depth is given in NUREG-
1623 as:
y5/3 = Qn / (1.486 S1/2)
and V = Q/y where V is the flow velocity in ft/sec
The results for both the top slope and the side slope of the Class A West Cell using the vegetated
slope condition are summarized in the following table. Flow velocities on the top and side
slopes of the Class A West Cell during the PMP event are predicted to be 2.37 and 2.07 ft/sec,
Erosion Responses for the Clive DU PA Model
23 Feb 2018 57
respectively. The acceptable Maximum Permissible Velocity (MPV) was selected from tables
provided in NUREG/CR-4620. The permissible velocity method is a commonly applied method
to determine channel stability. Under this method the slope is assumed stable if the calculated
velocity (V, the velocity resulting from a PMP in this application) is less than the MPV. By
contrast, if velocities exceed the MPV, it is assumed that the slope will experience excessive
erosion that will lead to the formation of gullies. The methodology presented in NUREG-1623
then directs that the estimates for the MPV be adjusted downward to account for the influences
of flow depth. The side slope gully analysis was completed independently of the top slope.
GULLY EROSION POTENTIAL – VELOCITY ANALYSIS
CLASS A WEST CELL

Slope Description Length
(ft)
Slope
(ft/ft)
i
(in/hr)
c Q
(cfs/ft)
y
(ft)
V
(ft/sec)
Adjusted
MPV
(ft/sec)
Top Slope (4%) 942 0.04 18.4 0.5 0.60 0.25 2.37 2.5
Side Slope (20%) 188 0.20 19.8 0.5 0.13 0.06 2.07 2.5

Therefore, all slope scenarios using the ET cover system are assumed to provide acceptable
protection against gully erosion using these criteria by limiting the potential of gully formation
from high velocity channelization.

Literature Cited

Bray, D. I., 1979. Estimating Average Velocity in Gravel-Bed Rivers. Proc. Am. Soc. Civ. Engrs, J.
Hydraul. Div., 105 (HY9), 1103-1122.

Haan, C.T., Barfield, B.J., Hayes, J.C., 1994. Design Hydrology and Sedimentology for Small
Catchments. Academic Press, Inc, San Diego, California.

Jensen, Donald T., 1995. Probable Maximum Precipication Estimates for Short-Duration, Small-
Area Storms in Utah. Utah Climate Center, Utah State University, Logan, Utah.

Johnson, T. L. 2002. Design of Erosion Protection for Long-Term Stabilization, Final Report,
NUREG-1623, U.S. Nuclear Regulatory Commission, Office of Nuclear Material Safety
and Safeguards, Washington, D.C.

Jones, Gordon L., 2012. PMF and 100-year Storm Analysis for Clive, Utah. Hansen, Allen &
Luce, Midvale, Utah.

Nelson, J. D., S. R. Abt, R. L. Voipe, D. van Zyl, N.E. Hikle, and W. P. Staub. 1986.
Methodologies for Evaluating Long-Term Stabilization Designs of Uranium Mill
Tailings Impoundments. NUREG/CR-4620, U.S. Nuclear Regulatory Commission,
Washington, D.C.

Erosion Responses for the Clive DU PA Model
23 Feb 2018 58
U.S. Army Corps of Engineers, 1977. Probable Maximum Precipitation Estimates, Colorado
River and Great Basin Drainages, Hydrometeorological Report No. 49. US Departement
of Army Corps of Engineers, Silver Spring, Maryland.

U.S. Army Corps of Engineers, 1994. Probable Maximum Precipitation – Pacific Northwest
States Columbia River (including portions of Canada), Snake River and Pacific Coastal
Drainages, Hydrometeorological Report No. 57. US Department of Army Corps of
Engineers, Silver Spring, Maryland.

U.S. Department of Agriculture, 1986. Urban Hydrology for Small Watersheds, Technical
Release 55. Natural Resources Conservation Service, Conservation Engineering
Division, Washington D.C.

U.S. Nuclear Regulatory Commission, 2002. NUREG-1623 - Design of Erosion Protection for
Long-Term Stabilization. Office of Nuclear Material Safety and Safeguards,
Washington, DC.

NAC-0105_R0

Deep Time Supplemental Analysis
Responses for the Clive DU PA
Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 ii

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
Deep Time Supplemental Analysis Responses for the Clive DU PA Model.docx

Bruce Crowe 1 Feb 2018
Dan Levitt

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
FIGURES .................................................................................................................................. iv
TABLES ..................................................................................................................................... v
ACRONYMS AND ABBREVIATIONS ................................................................................... vi
1.0 Overview and Conceptual Model........................................................................................ 1
1.1 Aeolian Deposition ....................................................................................................... 1
1.2 Intermediate Lake Sedimentation .................................................................................. 1
2.0 UDEQ Interrogatory Responses.......................................................................................... 3
2.1 Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation Part One—Aeolian
Deposition .................................................................................................................... 3
2.1.1 Interrogatory Response ............................................................................................ 4
2.1.1.1 Factual Accuracy and Technical Applicability of the UDEQ Review
Comments ...................................................................................................... 4
2.1.1.2 The Effects of the Aeolian Deposition Rate on the Deep Time Model Results
and Uncertainty in the Results ........................................................................ 6
2.2 Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation Part Two—
Intermediate Lake Sedimentation Rate .......................................................................... 6
2.2.1 Interrogatory Response ............................................................................................ 6
2.2.1.1 Intermediate Lake Sedimentation .................................................................... 6
2.2.1.2 Expanded Documentation of Lake Sedimentation Rates ................................. 8
2.2.1.3 Evidence of Clastic Sedimentation at the Clive Site ...................................... 10
2.2.1.4 Impact of Intermediate Lake Sedimentation on the Deep Time Model Results
..................................................................................................................... 10
3.0 Conclusion ....................................................................................................................... 11
4.0 References ........................................................................................................................ 13
Appendix A. Sedimentary Deposits of the Transgressive and Regressive Phases of Lake
Bonneville Near and Within the Clive Site (Deposits of Intermediate Lakes) .................... 15

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 iv
FIGURES
Figure 1. Deep Time ground surface flux of Rn-222 per NRC Reg. Guide 3.64. ........................ 11

Figure A-1. Roadcut exposures of the upper contact of the transgressive phase of Lake
Bonneville. .............................................................................................................. 16
Figure A-2. Annotated Google Earth image showing the Clive area and buried spits
developed on the southeast edges of the Grayback Hills. The spits prograded
toward and across what is now the Clive Site. .......................................................... 17
Figure A-3. (From Neptune (2015a).) View looking south from the north wall of Pit 29 at the
Clive Site. The dashed line marks the contact between the laminated marl of Lake
Bonneville (deep lake) and underlying transgressive sands with lenses of volcanic
conglomerate (deposits of an intermediate lake). ...................................................... 18
Figure A-4. (From Neptune (2015a).) Gully exposure of a lens of sandy conglomerate from
the outcrop shown in Figure A-3. The conglomerate contains distinctive clasts of
black volcanic rock derived from bedrock exposures in the Grayback Hills,
northwest of the Clive Site. The dashed line is the base of the laminated marl of
Lake Bonneville (deep lake). The gravel lens is here approximately 1 meter thick.
It is correlative with the Lake Bonneville Transgressive unit (0.76 m thick) listed
in Table 3 of Neptune (2015b). ................................................................................ 18
Figure A-5. (From Neptune (2015a).) Intermediate lake deposits of the regressive phase of
Lake Bonneville exposed in the north quarry wall of the Clive Site. The lake
deposits are exposed in the interval extending from the top of the hammer head to
the middle of the handle and are overlain by aeolian silt with local reddening from
soil alteration. .......................................................................................................... 19

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 v
TABLES
Table 1. Compilation of Lake Sedimentation Rates. .................................................................... 9

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 vi
ACRONYMS AND ABBREVIATIONS
DU depleted uranium
ka kilo-annum (1,000 years)
UDEQ Utah Department of Environmental Quality (also referred to in quoted
interrogatory text as “DEQ/SC&A,” reflecting authorship of DEQ’s consultant,
SC&A, Inc.)

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 1
1.0 Overview and Conceptual Model
1.1 Aeolian Deposition
During warm interglacial climates like the present, the predominant process of sedimentation at
the Clive Site is aeolian. (Note: the terms “aeolian” and “eolian” are used interchangeably in the
geological literature and refer to wind-generated sedimentary processes. The names are derived
from Aeolus, the Greek god of wind.) Aeolian source materials are transported and deposited by
prevailing winds sweeping across barren dry lake beds and the wetted fringes of playas where
gypsum crystals are supplied continuously by evaporation above sites of shallow groundwater.
Aeolian sedimentation occurs in two modes: saltation or bouncing of sand-sized particles
forming dune landforms, and suspension fallout that produces a near-uniform blanket of fine-
grained silt deposited at the ground surface.
The primary mode of aeolian deposition at the Clive Site is suspension fallout. The average
thickness of these deposits is about 72 cm (Neptune 2015a). The age of the aeolian silt deposits
at the Clive Site is estimated to be between 14,500 and 10,000 years before present, based
primarily on the estimates of the timing of final retreat of Lake Bonneville below the elevation of
the Clive Site (Neptune 2015a). Data for the thickness variations and ages of aeolian sediments
at the Clive Site are used to develop a distribution of the range of past aeolian sedimentation
rates (mean rate of 59 mm per 1,000 years with a standard error of the mean of 5 mm). This
distribution is used in the Deep Time model to project thicknesses of aeolian sediments over
time. The sediment thickness numbers are calculated in the model by multiplying the aeolian
deposition rates by the time when a future lake rises to the elevation of the Clive Site. Using
computer simulation, the Deep Time model samples a range of values of the depositional rate
and the time to the first lake arrival. A longer time to the first lake arrival and a higher aeolian
deposition rate result in a thicker blanket of aeolian deposits over the Clive Site, which in turn
reduces the Deep Time model projections of the peak ground surface radon flux. The peak radon
flux is the critical model output for assessing the Deep Time model results.
UDEQ agrees with the thickness estimates and time constraints for aeolian deposition in the
Clive area (see Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation). They did not
comment on and presumably accept the model representation of the future time of the arrival of
the first lake. They disagree with the approach used to estimate the statistical variation in the
aeolian deposition rate (standard error of the mean; see Interrogatory CR R313-25-8(5)(A)-18/3:
Sediment Accumulation).
1.2 Intermediate Lake Sedimentation
The Lake Bonneville basin of Utah and Nevada has been occupied by fresh to saline lakes
ranging in size from the scale of Lake Bonneville during the Last Glacial Maximum to the
present day Great Salt Lake. Future glacial cycles will lead to the rise of the Great Salt Lake
above the elevation of the Clive Site. A glacial cycle refers to climate intervals associated with
long-term reduction in global temperature, reduction in atmospheric CO2, and growth of ice
sheets and glaciers with falling sea levels. An interglacial cycle refers to warmer periods between
glacial cycles with sea levels at or near current levels. The earth has been in an interglacial cycle
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 2
since the Holocene epoch, a subdivision of geologic time representing the last 12 ka years before
the present (ka is kilo-annum or 1,000 years).
The Deep Time model (Neptune 2015b) uses a combination of past patterns of glacial and
interglacial cycles over the last 2 million years, global ice core and oxygen isotopic data, and the
history of past lake cycles in the Lake Bonneville basin to develop model projections of future
lake cycles in the Clive region. These model projections are closely tied to a widely accepted
100 ka length of global glacial cycles during the last 1 million years (Crowe et al. 2017; Lisiecki
and Raymo 2005). Field observations and drill core studies of lake deposits from two locations in
the Lake Bonneville basin are used to correlate observed lake sediments to lake patterns during
glacial cycles.
Three processes of sedimentation will affect the Clive Site during future glacial and interglacial
cycles, and the integrated processes will result in progressive burial of the Site through time.
These processes are aeolian deposition, deep lake sedimentation, and intermediate lake
sedimentation.
Aeolian deposition: During warm interglacial cycles like present-day conditions, small lakes
like the modern Great Salt Lake are maintained (lake elevation about 1280 meters). Under these
conditions, the Clive Site (elevation about 1307 meters) is at the surface, and sedimentation at
the Site consists of deposition of aeolian silt as suspension fall-out from surface winds.
Deep lake sedimentation: During past glacial conditions, the Clive Site has been covered by
large/deep lakes with lake surface elevations ranging from about 1360 meters (Stansbury
Shoreline) to about 1620 meters (Bonneville Shoreline). These elevations are 53 to more than
300 meters above the elevation of the Clive Site. Sedimentation at the Site under these deep lake
conditions is primarily from slow precipitation of carbonate with secondary quantities of fine-
grained silt. The silt content of deep lake sediments varies depending on lake currents, distance
to shorelines or surface outcrops, and fluvial (stream) sources of sediment. The Deep Time
model uses a distribution of deep lake sedimentation rates that range from about 0.075 to
0.20 mm/yr (Neptune 2015b). The Clive Site is adjacent to areas of higher topography that are
likely to be local sources of clastic sediments during future deep lake cycles.
Intermediate lake sedimentation: Intermediate lakes are defined with respect to the elevation
of the Clive Site. They include lakes that rise to and slightly above the elevation of the Site.
Intermediate lakes transition to deep lakes when the water depth exceeds the depth of wave
agitation (arbitrarily defined as approximately 10 meters above the elevation of the Clive Site).
Thus defined, intermediate lakes form during periods of fluctuating lake levels and also during
transitory phases of transgressive and regressive deep lakes. Intermediate lake sedimentation is
affected by wave activity and sediment transport associated with wave activity (longshore drift).
Wave-driven currents transport clastic sediments. (Note: clastic sediments are defined as material
that is mechanically or chemically fragmented and transported/deposited by sedimentary
processes.) Intermediate lake sedimentation includes carbonate precipitation, but carbonate
deposits tend to be secondary in volume to clastic sediments in areas of significant clastic
sediment supply and/or wave activity.
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 3
Sedimentation associated with intermediate lakes in the Deep Time model is represented as a
sediment thickness per intermediate lake event. UDEQ argues that the basis for representation of
intermediate lake events is inadequately documented and that the depositional rates for the
intermediate lake events may be too high. Overestimation of depositional rates could lead to
Deep Time model projections that overestimate the depth of burial of the Clive Site over time.
2.0 UDEQ Interrogatory Responses
This section responds to Interrogatories CR R313-25-8(5)(A)-18/3 and CR R313-25-8(4)(D)-
132/2. As noted by UDEQ in Interrogatory 132/2, the issues raised therein “are similar to some
of the Interrogatory 18 responses. To reduce redundancy, consideration should be given to
closing this interrogatory.” Accordingly, this response is focused on the two main issues raised in
Interrogatory 18/3: aeolian deposition and intermediate lake sedimentation rate. The latter issue
covers the concerns raised in Interrogatory 132/2.
2.1 Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation
Part One—Aeolian Deposition
After much exposition on the subject of aeolian deposition under the heading DEQ Critique of
DU PA v1.4, Appendix 13, this interrogatory states:
In conclusion, the above discussion presented four concerns that DEQ/SC&A has identified with
the EnergySolutions/Neptune aeolian deposition model. In order of perceived importance, these
are:
1) Deposition in the sub-areas of the embankment is likely correlated, rather than independent. A
correlated model would produce results that are more conservative than the current
EnergySolutions/Neptune model. However, the degree of correlation is presently unknown (and
perhaps unknowable).
2) The sample results do not represent a “point in time,” as EnergySolutions/Neptune indicated in
their previous response. Rather, the samples are an accumulation over 13,000 to 15,000 years
(Appendix 13, p. 38). Thus, the sample results can be thought of as being time averages.
3) Using the EnergySolutions/Neptune model results in a dose calculation means that the dose
receptor spends an equal amount of time in each embankment sub-area. The more conservative
assumption is that the dose receptor spends all of his time in the sub-area with the least amount
of deposition. Alternatively, the sub-area in which the dose receptor spends his time could be
randomly selected.
4) Dividing the embankment into 11 sub-areas based on the number of samples, as was done for
the EnergySolutions/Neptune model, appears reasonable. However, EnergySolutions/Neptune
should provide the rationale for selecting this approach.
DEQ/SC&A continues to believe that for nuclear licensing purposes the mean and standard
deviation aeolian deposition model should be used.
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 4
2.1.1 Interrogatory Response
2.1.1.1 Factual Accuracy and Technical Applicability of the UDEQ Review
Comments
Each of the four concerns outlined by UDEQ is addressed individually below.
Concern 1: UDEQ has questioned the approach used to select sub-areas for the embankment and
has suggested that the number of sub-areas should be established initially and used to identify
collection sites for each sub-area. They argue that sub-areas of the embankment are correlated
rather than independent, and that a correlated model would produce more conservative results
than the current Deep Time model. (Note: we assume “conservative” means thinner aeolian
deposits that would allow higher ground-surface radon flux in the Deep Time model results.)
Response to Concern 1
The Clive Site was not divided into sub-areas. UDEQ has incorrectly equated “sub-areas” with
field measurement sites. Field measurement sites for aeolian sediments at the Clive Site met two
selection criteria (Neptune (2015a), Section 2.1.4.1):
1. Contained exposures of the base of the aeolian sediments and the underlying sequence of
Lake Bonneville lake deposits.
2. Preserved the top surface of aeolian deposits to ensure the deposits were not changed by
post-deposition erosion or construction activities at the Site.
Additionally, the measurement data for the aeolian sediments were designed to be used in the
Deep Time model. The model requires thickness data for the aeolian sediments at the scale of the
DU disposal cell—the model does not calculate aeolian deposition rates for sub-areas. Field data
that met the two selection requirements were collected throughout the Clive Site (see Figure 8
and Table 2 in Neptune (2015a)).
Concern 2: UDEQ has argued that the field measurements of the thickness of aeolian sediments
across the Site do not represent a “point in time” as inferred in the Deep Time report. They
suggest instead that the sample sites are time integrated averages of aeolian deposition over
13,000 to 15,000 years.
Response to Concern 2
We agree in principle with the review comment. However, the Deep Time report did not claim
that the thickness of aeolian sediments was a “point in time.” UDEQ’s confusion may result from
an incorrect interpretation of the last paragraph of Section 4.0 of the Deep Time supplemental
model (Neptune (2015b), Section 7.1):
Therefore, the thickness of residual embankment material and sediment overlying the
disposed DU waste at the time when the first intermediate lake recedes will be effectively
equivalent to the thickness of aeolian sediments deposited up until that point in time,
represented by the rising elevation of the surrounding grade (emphasis added).
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 5
Here the “point in time” refers to the time of return of the first lake to the Clive elevation, not
processes of aeolian sedimentation. Aeolian sediments at the Clive Site represent an integrated
response of multiple processes (aeolian deposition and erosion, soil formation) occurring over
time at the spatial scale of the Clive Site. This is described in Section 5.1 of Neptune (2015c):
The preserved record of eolian deposition at the Clive site integrates variations in eolian
parameters and processes of deposition and burial over thousands of years. This integrated
record is assumed to provide the most consistent information for forecasting eolian
depositional rates over tens of thousands of years.
The Deep Time model uses a probability distribution of aeolian depositional rates projected
forward in time. The development of the probability distribution for the aeolian depositional rate
is described in Neptune (2015a), and its application is analogous to the spatial averaging of
aeolian sediment thickness across the Clive Site discussed in the response to Concern 1 above.
Concern 3: UDEQ argues that the aeolian deposition rate and Deep Time model results used
multiple sub-areas and should have included implicit assumptions about the location and
exposure time of a dose receptor in each sub-area.
Response to Concern 3
UDEQ’s objection is an improper interpretation of the radon flux and the Deep Time model
results. Sub-areas were not used in the Deep Time model. More importantly, the ground surface
radon flux has dimensions of area and time (pico-Curies per square meter per second). This
performance metric is not a dose and does not include or require a receptor or a receptor location.
Concern 4: UDEQ has requested that a rationale be provided for dividing the embankment into
11 sub-areas.
Response to Concern 4
The rationale for selection of measurement sites for aeolian sediments is described in Neptune
(2015a) and is repeated in the response to Concern 1 above (availability of suitable outcrops and
the scale of representation of the Clive Site in the Deep Time model). The probability
distribution used for the aeolian depositional rate in the Deep Time model is based on field data
from the Clive Site (11 field measurement sites and 21 clay resource test pits). These data
provide a statistically representative sample of the population of thickness data for aeolian
deposits across the Clive Site.
The Deep Time model requires upscaling of the field measurement data to the model scale, a
technically appropriate and established practice in environmental modeling studies (for example,
Blöschl and Sivapalan (1995); Neuman et al. (2003); Zhang et al. (2004)). The field
measurement scale of aeolian sediment thickness is applicable for areas of several square meters
(individual measurement site); the application of the aeolian rate calculation in the Deep Time
model is the DU disposal unit, a scale of thousands of square meters. The technically established
statistic for upscaling is the mean thickness of the field measurement and the standard error of
the mean to represent the variance of the averaged measurement data. Using the standard
deviation of the measurement data as suggested by UDEQ would incorrectly distort the variance
structure (representation of the variation of the measurement sites) in the Deep Time model
application.
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 6
2.1.1.2 The Effects of the Aeolian Deposition Rate on the Deep Time Model
Results and Uncertainty in the Results
UDEQ/SC&A have noted that differences between the standard error of the mean versus the
standard deviation of the measurements have a small impact on the average model results
(Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation, p. 13). We agree. The
important Deep Time model result is the ground surface radon flux through time. This model
result is sensitive to the following: (a) the sub-model used to estimate the ground surface radon
flux, (b) the time to the return of the next lake at the elevation of the Clive Site, (c) the aeolian
deposition rate, and (d) lake sedimentation rates during glacial cycles. UDEQ comments critique
only the variance of the aeolian depositional rate, and we have demonstrated that their comments
are not applicable to the data used for aeolian sedimentation in the Deep Time model.
2.2 Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation
Part Two—Intermediate Lake Sedimentation Rate
After discussing intermediate lake sedimentation rates under the heading DEQ Critique of DU PA
Appendix 21, this interrogatory concludes:
For all of these reasons, EnergySolutions/Neptune needs to either (1) provide independent
documentation that a sedimentation rate of 1.2 mm/yr is plausible or (2) define a plausible,
defensible intermediate lake sedimentation rate and redo the deep time analysis.
2.2.1 Interrogatory Response
This response addresses the following topics related to UDEQ concerns on the intermediate lake
sedimentation rate:
1. The technical rationale for the representation of intermediate lake sedimentation in the
Deep Time model.
2. New information is compiled from the scientific literature that expands documentation of
overall lake sedimentation rates and their application in the Deep Time model.
3. Field observations are summarized on the processes of intermediate lake sedimentation at
the Clive Site during the last transgressive (rising lake) and regressive (declining lake)
stages of Lake Bonneville (Appendix A).
4. The effects of intermediate lake sedimentation rates on the Deep Time model results, and
by extension the limited value of additional studies in this area.
2.2.1.1 Intermediate Lake Sedimentation
UDEQ has questioned the intermediate lake sedimentation “rate” used in the Deep Time model,
including both a reference to a sedimentation “rate” that is 10 times greater than the large lake
sedimentation rate (Version 1.4 of the Deep Time model; Neptune (2015b)) and a sedimentation
“rate” that they estimate to be as high as 5.64 mm/yr. They suggest the intermediate lake
sedimentation “rates” used in the model are unrealistic and cite literature from the Eastern Great
Basin of Utah supporting lake sedimentation rates that range from 0.12 to 0.83 mm/yr. (Note:
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 7
these sedimentation rates apply to lake sedimentation in general and are not specific to
intermediate lake sedimentation.) UDEQ also argues, citing input from Dr. Paul Jewell,
University of Utah, that lake sedimentation patterns can be complicated depending on proximity
to active faults and local sources of stream sediment. They have requested enhanced
documentation of sedimentation rates of 1.2 mm/yr or greater and development of a more
defensible “rate” of intermediate lake sedimentation. UDEQ also suggests that revision of the
intermediate lake sedimentation “rates” could require revision of the Deep Time model.
Establishing sedimentation rates for intermediate lakes is problematic for two reasons. First, the
lake sediments are reworked by wave activity that intermixes deep lake sediments (carbonate
marl), aeolian sediments, and intermediate lake clastic sediments. It is often difficult or
impossible to distinguish the different types of deposits in core studies. Second, intermediate
lakes are transitory features of limited duration. It is very difficult and impractical to attempt to
obtain the data necessary to constrain the ages of intermediate lake sediments and to estimate
depositional rates.
Recognizing these difficulties, the Deep Time model uses a pragmatic approach that represents
sedimentation by intermediate lakes as a depositional thickness per lake event. Estimates of
sediment thickness of intermediate lakes in the Deep Time model are derived primarily from
conceptual models of the processes of lake sedimentation, studies of sediment cores from drill
holes in the Lake Bonneville basin (Knolls and Burmester lake cores), and interpretations of lake
sediments in a Clive pit wall (Neptune 2015b). The latter interpretations are from a single
quarry-wall exposure that was studied in 1985. This combined information base is used to
establish a probability distribution of sediment thickness per intermediate lake event (see Figure
11; Neptune (2015b)).
UDEQ derives their cited intermediate lake sedimentation “rate” of 5.64 mm/yr by dividing the
geometric mean of the distribution of intermediate lake sediment thickness used in the Deep
Time model (2.82 m) by the estimated duration of an intermediate lake (500 years; Neptune
(2015b)). However, the Deep Time model does not use a sedimentation rate for intermediate
lakes and, in fact, the Deep Time model report specifically cautions (Neptune (2015b), Section
7.1):
There is virtually no information for the duration of intermediate lakes, due to the high
mixing rate of shallow lake sediments, which makes dating of times within a single
stratigraphic layer of a shallow lake sediment core extremely difficult. Thus, a distribution
was chosen to roughly calibrate with the heuristic model: lognormal with geometric mean of
500 y and geometric standard deviation of 1.5.
UDEQ, in their comment, incorrectly used information from the Deep Time probability
distribution of sediment thickness per lake event to calculate a sedimentation “rate” that is non-
applicable and was not used in the Deep Time model results. The technically correct approach
that UDEQ should have evaluated is a comparison of Deep Time model results for sedimentation
rates during 100-ka glacial cycles. Under this approach, the composite sediment thickness of
intermediate lake events is combined with aeolian and deep lake sedimentation. This composite
sediment thickness is constrained in the model by observed sediment thicknesses of 15 to 20
meters per glacial cycle based on core studies at the Knolls and Burmester sites (Neptune
(2015b), Section 7.4). This model interpretation of sediment thickness per glacial cycle is
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 8
equivalent to sedimentation rates of 0.15 to 0.20 mm/yr, and these rates are consistent with lake
sedimentation rates in new studies reported herein.
2.2.1.2 Expanded Documentation of Lake Sedimentation Rates
This section examines lake sedimentation rates in response to UDEQ’s request for
documentation of rates greater than 1.2 mm/yr. This new information does not apply directly to
the Deep Time model representation of intermediate lake sediment thickness per lake event. As
noted above, the data from Table 1 rates should be compared with sedimentation rates used in
the Deep Time model for100 ka glacial cycles.

Table 1 is a compilation of lake sedimentation rates from literature publications and calculations
derived from unpublished data and/or interpreted data. Sedimentation rates in Table 1 range from
as low as < 0.01 mm/yr to over 6.0 mm/yr. This range of variation is caused by a combination of
factors, including:

1. Rates of carbonate precipitation (fine-grained calcareous (CaCO3) mud).
2. Local availability of clastic sediment sources (gravel, sand, silt, and clay-sized particles),
primarily from proximity to areas of active faulting, higher topography near lake edges,
and/or near areas of river or stream that discharge into the lake.
3. Processes of wave activity that may produce strong longshore currents and may locally
transport and deposit clastic sediments.
4. Mass movement of lake bottom deposits.

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 9
Table 1. Compilation of Lake Sedimentation Rates.
Lake Location Min
Rate
(mm/yr)
Max
Rate
(mm/yr)
TimeSpan
(years)
Long/Short
(rate
duration)
Reference
Great Salt
Lake
U.S. 0.57 0.57 11,500 Long Thompson et al.
(2016)
Great Salt
Lake
U.S. 1.1 4.1 < 100 Short Naftz et al. (2000)
Great Salt
Lake
U.S. 0.1 6.7 < 10,000 Short/Long Oliver et al. (2009)
Lake Baikal Asia 0.05 0.05 > 100,000 Long Colman et al. (1995)
Lake
Bonneville
U.S. 0.8 0.8 12,000 Long Thompson et al.
(2016)
Lake
Bonneville
U.S. 0.04 0.94 12,000 Long Oviatt (2018 [in
press])1
Lake
Bonneville
U.S. 0.04 > 3.0 12,000 Long Oviatt (2018 [in
press])2
Lake
Michigan
U.S. < 0.01 0.84 < 10,000 Long Robbins and
Edgington (1975)
Lake
Michigan
U.S. 3.0 3.0 < 100 Short Robbins and
Callender (1975)
Lake
Tanganyika
Zaire 0.07 1.84 > 100,000 Long Cohen et al. (1993)
Owens Lake U.S. 0.4 0.4 > 100,000 Long Bischoff et al. (1997)
Red Lake Romania 1.17 3.2 < 100 Short Begy et al. (2009)
Taihu Lake China 0.1 0.42 < 100 yr Short Wang et al. (2001)
1 Open lake; only carbonate marl
2 Including deltas

High clastic sediment flux can cause very high local lake sedimentation rates. Very low
sedimentation rates generally occur in lake interiors that are distant from shorelines and sources
of clastic sediment (Oviatt 2018 [in press]). Significant variations in lake sedimentation rates can
occur over short horizontal distances depending on the local variations in processes of sediment
transport. During intervals of high wave activity, when intermediate lakes are present at the
elevation of the Clive Site, sedimentation rates can vary from negative (erosion) to very high
(progradation of gravel spits by longshore drift). Additionally, the dynamics of wave activity will
vary through time, with short-term variations in lake levels and changes in wind direction and
wind velocity. Maximum sediment deposition for intermediate lakes will most likely occur
during storm events (storm surge).
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 10
2.2.1.3 Evidence of Clastic Sedimentation at the Clive Site
Field studies of the lake sediments at the Clive Site provide important evidence of significant
past clastic sedimentation at the Clive Site associated with formation of transitory intermediate
lakes. These clastic sediments have been studied primarily for the transgressive (rising lake) and
regressive (declining lake) phases of Lake Bonneville (Neptune 2015a). Supporting
documentation for these observations, including photographs of the lake sediments and larger-
scale sedimentary features of longshore drift, were provided previously to UDEQ (Neptune
2015) and are repeated in Appendix A of this report.
The Clive Site has been and is expected to be an area of highly variable sedimentation during
future glacial cycles; this includes sedimentation associated with aeolian activity and formation
of intermediate and deep lakes. The optimal approach for representing the multiple dynamic
processes of lake sedimentation is through model simulations using a distribution of composite
sediment thickness for 100 ka glacial cycles. The sedimentation “rate” derived by UDEQ for
intermediate lakes is not applicable to the Deep Time model.
2.2.1.4 Impact of Intermediate Lake Sedimentation on the Deep Time Model
Results
The Deep Time model output of significance for the supplemental analysis is the ground surface
flux of radon (Figure 1 this paper; derived from Neptune (2015c)). Figure 1 shows that the peak
flux of radon occurs at about 60 ka, the model-projected time of the first occurrence of an
intermediate lake at the Clive Site.
The important Deep Time model assumptions and parameters that control the timing and
magnitude of the peak surface radon flux are:
1. The model representation of radon diffusion through the embankment cover and
overlying sediments.
2. The time of the first lake event at the Clive Site.
3. The aeolian deposition rate.
4. The intermediate lake sediment thickness for individual lake events.
Generally, the longer the time to the first event, the thicker the mantle of aeolian sediments on
top of the Clive Site. An intermediate lake event adds sediments above the top of the Site; the
added amount is determined by randomly sampling (computer simulation) the sediment
thickness probability distribution (see Figure 11; Neptune (2015b)). Overall, the Deep Time
model projections of combined processes of sedimentation (aeolian, intermediate, and deep lake
processes) affect the thickness of sediments covering the Clive Site over time. Increased depth
of burial will result in a systematic reduction in the radon flux (see the curve of the surface
radon flux in Figure 1).
An additional consideration is that the intermediate lake sediment thickness is not an
independent parameter in the Deep Time model results. It covaries with both the aeolian
sedimentation rate and the deep lake sedimentation rate. The three model parameters are
constrained in the Deep Time model by the composite thickness of sediments during the 100-ka
glacial cycles (15 to 20 meters thickness; Neptune (2015c)).
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 11

Figure 1. Deep Time ground surface flux of Rn-222 per NRC Reg. Guide 3.64.
3.0 Conclusion
UDEQ has raised several issues in the Deep Time interrogatories. The following points
summarize this response:
• The aeolian sediment data reported in Neptune (2015a) were used to establish statistical
parameters (sediment thickness and variations in sediment thickness) in the Deep Time
model. These measurements are a statistically representative sample of the population of
aeolian thickness data for the Clive Site and the DU disposal unit. UDEQ’s assertions
concerning sub-areas and sub-area correlations are not applicable to either the aeolian
field studies or the aeolian sedimentation parameters used in the Deep Time model.
• UDEQ has incorrectly assumed that the embankment is divided into 11 sub-areas in the
Deep Time model. The rationale for using the standard error of the mean in the
probability distribution for the aeolian depositional rate in the Deep Time model was
described in Neptune (2015a). The standard error of the mean is the correct statistical
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 12
parameter for representing the variance of the aeolian deposition rate for the scale of the
Deep Time model application. The recommendations by UDEQ do not follow established
modeling practices for environmental studies.
• The correct approach and statistical parameters are used for the aeolian depositional rate
in the Deep Time model and the impact of the variation in this rate on model results is
insignificant.
• Contrary to the interpretations in the review responses by UDEQ, the Deep Time model
does not use an intermediate lake sedimentation “rate.” Intermediate lakes are represented
by a distribution of sediment thickness per lake event.
• The distribution of sediment thickness for intermediate lake events is integrated with the
aeolian and deep lake sedimentation rates during 100-ka glacial cycles. The integrated
sedimentation rates during the glacial cycles (0.15 to 0.20 mm/yr) are consistent with
documented lake sedimentation rates, including rates cited in the review responses by the
UDEQ and the newly developed data shown on Table 1 in this report.
• We agree with the UDEQ comment that lake sedimentation patterns can be complicated
and are dependent on locations within lakes, interaction with wave-generated currents,
proximity to active faults, and local sources of sediment. These issues are already
addressed appropriately in the Deep Time model by integrating multiple sources of
information on lake sedimentation, including observations of lake and aeolian
sedimentary deposits at the Clive Site.
• Table 1 provides expanded documentation of lake sedimentation rates in response to the
UDEQ review comments, but these rates do not apply to the representation of
intermediate lakes in the Deep Time model. The data summarized in the table support
wide ranges in lake sedimentation rates and include rates greater than 1.2 mm/yr.
• The model distribution used to represent intermediate lake sediment thickness is based on
1985 data at the Clive Site for primarily the Lake Bonneville sedimentation history, with
limited local data for sedimentation associated with pre-Lake Bonneville glacial cycles.
The integrated net sedimentation rates used in the model for the 100-ka glacial cycles are
consistent with drill core data from the Clive region and published information on lake
sedimentation rates (see Table 1).
• The range of sediment thicknesses for intermediate lake events and integrated lake
sedimentation rates during glacial cycles used in the Deep Time model decreases the
peak ground surface radon flux and the radon flux over time. The data used for the lake
sedimentation rates are derived from and are consistent with core studies of deposits of
Lake Bonneville for past glacial cycles, published studies of lake sedimentation rates, and
field observations at the Clive Site.
On the basis of the concerns expressed by UDEQ as responded to herein, revision of the Deep
Time model is not justified.
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 13
4.0 References
Begy, R., et al., 2009. Recent Changes in Red Lake (Romania) Sedimentation Rate Determined
from Depth Profiles of 210Pb and 137Cs Radioisotopes, Journal of Environmental
Radioactivity 100 (2009) 644–648
Bischoff, J.L., et al., 1997. A Time-Depth Scale for Owens Lake Sediments of Core OL-92:
Radiocarbon Dates and Constant Mass-Accumulation Rate. In An 800,000-Year
Paleoclimatic Record from Core OL-92, Owens Lake, Southeast California, Geological
Society of America Special Paper 317, edited by G.I. Smith and J.L. Bischoff, pp. 91–98,
Geological Society of America, Boulder CO
Blöschl, G., and M. Sivapalan, 1995. Scale Issues in Hydrological Modelling: A Review,
Hydrological Processes 9 (1995) 251–290
Cohen, A.S., et al., 1993. Estimating the Age of Formation of Lakes: An Example from Lake
Tanganyika, East African Rift System, Geology 21 (6) 511–518
Colman, S.M., et al., 1995. Continental Climate Response to Orbital Forcing from Biogenic
Silica Records in Lake Baikal, Nature 378 (1995) 769–771
Crowe, B., et al., 2017. Representation of Global Climate Change in Performance Assessment
Models for Disposal of Radioactive Waste—17183, proceedings of the WM2017
Conference, March 5–9, Phoenix AZ, 2017
Lisiecki, L.E., and M.E. Raymo, 2005. A Pliocene-Pleistocene Stack of 57 Globally Distributed
Benthic δ18O Records, Paleoceanography 20 (2005) 1–17 doi: 10.1029/2004PA001071
Naftz, D.L., et al., 2000. Reconstructing Historical Changes in the Environmental Health of
Watersheds by Using Sediment Cores from Lakes and Reservoirs in Salt Lake Valley,
Utah, USGS Fact Sheet FS–164–00, United States Geological Survey (USGS),
December 2000
Neptune, 2015a. Neptune Field Studies, December, 2014, Eolian Depositional History Clive
Disposal Site, NAC-0044_R0, Neptune and Company Inc., Los Alamos NM, March 2015
Neptune, 2015b. Deep Time Assessment for the Clive DU PA, Deep Time Assessment for the
Clive DU PA Model v1.4, NAC-0032_R4, Neptune and Company, Inc., Los Alamos NM,
November 2015
Neptune, 2015c. Deep Time Supplemental Analysis for the Clive DU PA, Clive DU PA Model
vDTSA1, NAC-0043_R0, Neptune and Company, Inc., Los Alamos NM, March 2015
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 14
Neuman, S.P., et al., 2003. A Comprehensive Strategy of Hydrogeologic Modeling and
Uncertainty Analysis for Nuclear Facilities and Sites, NUREG/CR-6805, United States
Nuclear Regulatory Commission, Washington DC, July 2003
Oliver, W., et al., 2009. Estimating Selenium Removal by Sedimentation from the Great Salt
Lake, Utah, Applied Geochemistry 24 (2009) 936–949
Oviatt, C.G., 2018 [in press]. Geomorphic Controls on Sedimentation in Pleistocene Lake
Bonneville, Eastern Great Basin, Chapter in Geological Society of America Special
Paper, 2018
Robbins, J.A., and E. Callender, 1975. Diagenesis of Manganese in Lake Michigan Sediments,
American Journal of Science 275 (5) 512–533
Robbins, J.A., and D.N. Edgington, 1975. Determination of Recent Sedimentation Rates in Lake
Michigan Using Pb-210 and Cs-137, Geochimica et Cosmochimica Acta 39 (3) 285–304
Thompson, R.S., et al., 2016. Late Quaternary Changes in Lakes, Vegetation, and Climate in the
Bonneville Basin Reconstructed from Sediment Cores from Great Salt Lake. In Lake
Bonneville: A Scientific Update. Developments in Earth Surface Processes Vol. 20,
edited by C.G. Oviatt and J.F. Shroder Jr., pp. 221–291, Elsevier, Amsterdam,
Netherlands
Wang, J., et al., 2001. Taihu Lake, Lower Yangtze Drainage Basin: Evolution, Sedimentation
Rate and the Sea Level, Geomorphology 41 (2) 183–193
Zhang, X., et al., 2004. Scaling Issues in Environmental Modelling. In Environmental
Modelling: Finding Simplicity in Complexity, edited by J. Wainwright and M. Mulligan,
pp. 319–334, John Wiley & Sons, Chichester, England

Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 15
Appendix A. Sedimentary Deposits of the Transgressive and
Regressive Phases of Lake Bonneville Near and Within the Clive Site
(Deposits of Intermediate Lakes)
Field studies of aeolian sediments and lake sediments in the Clive region (Neptune 2015a) have
described local exposures of gravel, sandy conglomerate, and sand in the Clive region. These
sediments were deposited primarily during the transgressive phase of Lake Bonneville, and
patterns of clastic sedimentation associated with past intermediate lake events were documented.
The cited studies were not focused on the lake events and thus detailed descriptions of the
sediments, including facies variations and thickness variations within and near the Clive Site,
have not been compiled.
The purpose of this brief summary is to document local characteristics of the deposits, including
evidence of energetic longshore drift with formation of local spits during the last transgressive
intermediate lake event near the Clive Site. The intermediate lake sediments contain distinctive
clasts of black volcanic rocks derived from local bedrock exposed in the Grayback Hills
northwest of the Clive Site. These relationships are consistent with southeast-directed wave
activity driving longshore drift that transported coarse sandy conglomerate and sand
southeastward from the Grayback Hills across the Clive Site.
Figure A-1 is a photograph of a roadcut exposure along Highway I-80 north of the Clive Site.
The outcrops expose the upper part of spit deposits of sandy conglomerate of the transgressive
phase of Lake Bonneville (deposited by longshore drift of an intermediate lake) overlain by Lake
Bonneville marl (deep lake deposits). The marl is overlain by Holocene aeolian silt.
Figure A-2 shows the locations of the Grayback Hills, the Clive Site, and the roadcut exposure of
Figure A-1. Buried ridges extend from the Grayback Hills toward the Clive Site and are upheld
by spit deposits of sandy conglomerate formed by longshore drift associated with an intermediate
lake. These ridges were interpreted originally as shorelines of the Gilbert phase of Lake
Bonneville. Recent field work has shown instead that the topography is upheld by spits buried
beneath the Lake Bonneville marl (Neptune 2015a).
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 16

Figure A-1. Roadcut exposures of the upper contact of the transgressive phase of Lake
Bonneville.
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 17

Figure A-2. Annotated Google Earth image showing the Clive area and buried spits
developed on the southeast edges of the Grayback Hills. The spits prograded toward
and across what is now the Clive Site.
Figure A-3 is from Neptune (2015a) and shows a quarry-wall exposure in the north part of the
Clive Site that contains exposures of the upper contact of the transgressive deposits of Lake
Bonneville.
Figure A-4 (also from Neptune (2015a) is a close-up view of the outcrop area shown in Figure
A-3. The sandy conglomerate was deposited by longshore drift that prograded sediment from the
northwest to southeast across the Clive Site. The thickness of this intermediate lake sediment is
expected to thin across the Clive Site (north to south) away from the sediment sources at the
Grayback Hills.
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 18

Figure A-3. (From Neptune (2015a).) View looking south from the north wall of Pit 29 at
the Clive Site. The dashed line marks the contact between the laminated marl of Lake
Bonneville (deep lake) and underlying transgressive sands with lenses of volcanic
conglomerate (deposits of an intermediate lake).

Figure A-4. (From Neptune (2015a).) Gully exposure of a lens of sandy conglomerate from
the outcrop shown in Figure A-3. The conglomerate contains distinctive clasts of
black volcanic rock derived from bedrock exposures in the Grayback Hills, northwest
of the Clive Site. The dashed line is the base of the laminated marl of Lake Bonneville
(deep lake). The gravel lens is here approximately 1 meter thick. It is correlative with
the Lake Bonneville Transgressive unit (0.76 m thick) listed in Table 3 of Neptune
(2015b).
Deep Time Supplemental Analysis Responses for the Clive DU PA Model
23 Feb 2018 19
Figure A-5 is a photograph of the deposits of the regressive phase of Lake Bonneville exposed in
the north wall of Pit 29 at the Clive Site. The sediments of this regressive intermediate lake are
preserved as a gradational zone between overlying aeolian silt (Holocene) and underlying
reworked marl of Lake Bonneville (light-colored sediments near the middle of the pick handle).
These regressive intermediate lake deposits are approximately 30 centimeters thick and illustrate
the variability of sedimentation patterns of intermediate lakes. The fine-grained silts of this
regressive intermediate lake are attributed to a combination of draping of the underlying
topography by fine-grained marl (the only available sediment source) and the probable rapid fall
of lake levels during the regressive phase of Lake Bonneville at the elevation of the Clive Site.

Figure A-5. (From Neptune (2015a).) Intermediate lake deposits of the regressive phase of
Lake Bonneville exposed in the north quarry wall of the Clive Site. The lake deposits
are exposed in the interval extending from the top of the hammer head to the middle
of the handle and are overlain by aeolian silt with local reddening from soil
alteration.

NAC-0102_R0

Other Wastes Responses for the
Clive DU PA Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 ii

Other Wastes Responses for the Clive DU PA Model
Other Wastes Responses for the Clive DU PA Model.docx
Summary of how wastes other than DU are addressed in GoldSim model
v. 1.4 and supporting documentation.

Sean McCandless 12 Feb 2018
Mike Sully 13 Feb 2018

Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
ACRONYMS AND ABBREVIATIONS ................................................................................... iv
1.0 Overview and Conceptual Model........................................................................................ 1
2.0 UDEQ Interrogatory Responses.......................................................................................... 4
2.1 Interrogatory CR R313-25-9(5)(A)-196/1: .................................................................... 4
2.1.1 Interrogatory Response ............................................................................................ 4
2.2 Interrogatory CR R313-25-9(5)(a)-203/1: ..................................................................... 4
2.2.1 Interrogatory Response ............................................................................................ 4
3.0 References .......................................................................................................................... 5

Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 iv
ACRONYMS AND ABBREVIATIONS
CAW Class A West embankment
CQA/QC Construction Quality Assurance/Quality Control
DEQ (Utah) Department of Environmental Quality
DOE (United States) Department of Energy
DU depleted uranium
DUO3 depleted uranium trioxide
ET evapotranspiration
GDP gaseous diffusion plant
LLRW low-level radioactive waste
NRC (United States) Nuclear Regulatory Commission
PA performance assessment
SER Safety Evaluation Report
SRS Savannah River Site
UDEQ Utah Department of Environmental Quality

Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 1
1.0 Overview and Conceptual Model
The Clive DU PA is limited to the disposal of DU wastes of two general waste types: 1) depleted
uranium trioxide (DUO3) waste from the Savannah River Site (SRS); and 2) anticipated DU
waste as U3O8 from gaseous diffusion plants (GDPs) at Portsmouth, Ohio and Paducah,
Kentucky. The quantity and characteristics of DU waste from other sources that has already been
disposed of at the Clive Facility was not included. Similarly, other Class A LLRW anticipated
for disposal within the Federal Cell was not included.
This choice was made for several reasons:
1. Probabilistic PAs, while a mature approach within the Nuclear Regulatory Commission
(NRC) and the Department of Energy (DOE), are new to Utah LLRW regulators. This
reality is reflected by Utah’s need to utilize external contractor review support. A
probabilistic PA is initially a more complex and expensive modeling approach than
deterministic modeling; however, once the foundation is established it can easily be
adapted for additional wastes and new data. Due to shorter half-lives, other Class A
radionuclides have much shorter time periods for evaluation. Therefore, a relatively
simplified approach was selected, with the intent that, once the basic model structure and
approach was developed, additional Class A wastes could be more explicitly modeled as
a follow-on activity. Given the commitment to place DU in the bottom of the Federal
Cell, additional Class A wastes do not need to be modeled in order for embankment
construction and initial DU placement to begin.
2. Class A LLRW had already been modeled multiple times for the Clive Site and basic cell
design. These models demonstrated the Site’s ability to achieve the performance
objectives, albeit using deterministic rather than probabilistic performance assessments.
Note that deterministic performance assessments for the Clive Site are consistently
structured to demonstrate compliance even under extreme bounding conditions. These
models also explicitly included DU at its specific activity (i.e., elemental DU), and
existing disposals were placed in full compliance with a UDEQ-approved license. The
Class A West (CAW) embankment differs from the Federal Cell in physical geometry
and cover system design; however, the horizontal travel distances to compliance points
are the same, the vertical vadose zone thicknesses are the same, and the precipitation
falling on the cells is the same. Therefore, if the modeled infiltration is the same or less
for the Federal Cell than for the Class A West embankment, then Federal Cell disposal of
wastes already approved for disposal in the CAW can be expected to meet the
performance criteria.
3. By definition (via the regulatory basis for 10 CFR Part 61), Class A LLRW decays to
radiation levels that protect health and safety within the 100-year institutional control
period. Therefore, the majority of Class A contaminants of concern have decayed away
millennia before ingrowth concerns associated with DU would be exhibited.
In the final report for model version 1.4 (Neptune 2015a), Section 6.0 documents this scoping
constraint: “The disposal volume above the DU waste is assumed to be backfilled with clean
material for the purposes of this DU analysis.” This constraint was established and has been
Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 2
carried forward unchanged from the initial submittal of version 1.0 of the DU PA Model
(Neptune 2011).
UDEQ acknowledged this approach in Section 1.0 of the 2015 SER (SC&A 2015):
It should be noted that this SER apples only to approval/disapproval of the DU PA as
required by UAC Rule R313-25-9(5)(a). That regulation does not consider “other wastes,”
which must be addressed in a separate performance assessment. If the DU PA performance
assessment is approved, it is expected that the next step in the regulatory process would be
submission of a proposed license amendment for review by DEQ.
Section 6.2.4 of the 2015 SER elaborates and provides Condition 4, which would apply to an
amended License that would permit DU disposal:
To meet the requirements of UAC R313-25-9(5)(a), EnergySolutions shall submit a revised
performance assessment that meets the requirements of that provision and addresses the total
quantities of concentrated DU and other radioactive wastes the facility now proposes to
dispose in the Federal Cell. This revised performance assessment shall be subject to notice
and comment and must be approved by the Director prior to the land disposal of other
radioactive waste [emphasis added].
EnergySolutions did not object to this aspect of the 2015 SER, as this accurately reflects their
intention and understanding of the regulatory process. Contrary to this agreed-upon regulatory
process, Utah DEQ (2017) introduced two new interrogatories relating to other Class A waste.
This is in spite of the basis for Interrogatory CR R313-25-9(5)(A)-196/1 correctly stating:
Therefore, to address the R313-25-9(5)(a) “other waste” requirement, a revised PA, or a
separate PA, must be prepared by EnergySolutions, and approved by DEQ, before any
“other waste” (understood to be DOE-generated Class A waste) is disposed of above the
DU [emphasis added].
EnergySolutions does not intend or request to dispose of other Class A LLRW above the DU
until such time as a PA accounting for the combined effects of DU and other Class A LLRW is
approved. This does not preclude approval to dispose of DU in the interim.
Current approved Clive deterministic PAs apply Unit 3 material properties to Class A waste. See
Whetstone’s Revised Western LARW Cell Infiltration and Transport Modeling Report (July 19,
2000) and their Class A West Disposal Cell Infiltration and Transport Modeling Report
(Whetstone Associates 2011), which both include discussion of model insensitivity to waste
thickness and other physical properties. This longstanding approach for the facility has been
acceptable to UDEQ, dating to the earliest deterministic PAs. In raising an objection, UDEQ
does not substantiate the change in approach from their prior approvals.
Modeling of waste layers as Unit 3 sandy soil has been accepted by the regulator precisely
because Class A unstable waste is heterogeneous and may have a variety of physical forms. It
would not be possible to model all possible waste form combinations; therefore, the model is
simplified to represent waste layers as more permeable than the overlying cover and the
underlying clay liner.
Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 3
Furthermore, waste form can often contribute to delayed release of radionuclides for
environmental transport; however, credit is not taken for this mechanism in current approved
deterministic PAs nor in the DU PA v.1.4. For example, radionuclides present in waste as
activated metal or within a grouted monolith would not be available for transport as early after
disposal as radionuclides present as contamination on debris. By assuming that all activity is
available for environmental transport immediately, Clive PAs bias environmental release to
occur earlier than it may actually occur.
It is not necessary to model the potential for preferential pathways for radon or precipitation
within or between actual waste placement layers in order to evaluate performance of the Federal
Cell. For an evapotranspiration (ET) cover, infiltration is influenced by properties of the Unit 4
soil surface layer, while percolation is mainly controlled by Unit 4 Surface and Evaporative Zone
layers and the Frost Protection layer, with some influence possible from the radon barrier. In
other words, rates of infiltration, percolation, and exhalation are controlled by the cover system
and the clay liner. Model performance will be essentially the same regardless of the modeled
properties within the waste layers. See also ET Cover Design Responses for the Clive DU PA
Model (Neptune 2018a).
Sandy soil properties were conservatively selected for modeling the waste layers in order to
minimize potential effects of absorption/retardation that could occur if the waste layer were
assumed to be clay. Note that, in practice, clay is often used as fill material within the
embankment.
Appendix 2 to Neptune (2015a), Conceptual Site Model for Disposal of Depleted Uranium at the
Clive Facility, Clive DU PA Model v. 1.4 (Neptune 2015d), Section 8.1, remains consistent with
Clive’s prior approved PAs in this respect: “All wastes are assumed to have the characteristics of
local Unit 3 sandy soil.”
See also Appendix 16 to Neptune (2015a), Model Parameters for the Clive DU PA Model, Clive
DU PA Model v. 1.4 (Neptune 2015c), Section 4.16: “The current Clive DU PA Model has no
generic waste inventory, but this material is defined as a placeholder. Any layers to be filled with
generic LLW borrow material properties from Unit 3 (see Table 10).”
As noted in Interrogatory CR R313-25-9(5)(A)-196/1, Section 3.1 of Appendix 18 to Neptune
(2015a), Radon Diffusion Modeling for the Clive DU PA, Clive DU PA Model v. 1.4 (Neptune
2015b), states that other wastes are assigned Unit 4 properties. Appendix 18 has a typographical
error and should state that other wastes are assigned Unit 3 properties, consistent with the
Conceptual Site Model, Appendix 16, and the GoldSim model.
Note that the Federal Cell, as currently configured, isolates DU (and, ultimately, overlying Class
A LLRW) from the adjacent 11e.(2) embankment. The Federal Cell has no existing waste
placement of any type. Refer to Federal Cell Design Responses for the Clive DU PA Model
(Neptune 2018b) for discussion of the designs and related issues in response to interrogatories
under the category “Federal Cell Design.”
Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 4
While not prohibiting placement of DU upon approval of the Clive DU PA Model v.1.4,
additional modeling is required by rule before Class A waste is disposed in the Federal Cell.
There are two reasons that this is the case:
1. The performance of the ET cover included in the DU PA should be verified for Class A
LLRW disposed in the same embankment as DU.
2. Incremental dose contributions from other Class A waste in the Federal Cell should be
quantified for a complete probabilistic PA.
Considering that current approved PAs for the Class A West, LARW, and Mixed Waste cells
include Class A LLRW in addition to DU, it is difficult to imagine a scenario where Class A
LLRW would require any limitation below the Class A limits when disposed with DU under a
probabilistic PA. Nonetheless, if demonstrated to be desirable under a combined probabilistic
PA, Federal Cell waste acceptance criteria can be set to limit sensitive Class A radionuclide
concentrations or placement location.
2.0 UDEQ Interrogatory Responses
This section contains responses for Interrogatories CR R313-25-9(5)(A)-196/1 and CR R313-25-
9(5)(a)-203/1.
2.1 Interrogatory CR R313-25-9(5)(A)-196/1:
Please provide an analysis to demonstrate that the DU PA v1.4 assumed homogeneous Unit 4
silty clay material used to model the layer above the DU is representative of the various types of
DOE-generated Class A waste EnergySolutions intends to dispose of in that layer. Density,
among other factors, should be considered.
2.1.1 Interrogatory Response
As discussed above, the reference to Unit 4 material is in error. The use of Unit 3 material
properties for Class A LLRW is reasonable and consistent with prior PAs for the facility.
2.2 Interrogatory CR R313-25-9(5)(a)-203/1:
Please describe how EnergySolutions proposes to address the requirements of R313-25-9(5)(a)
to demonstrate that PA requires consideration of the “total quantities of concentrated depleted
uranium and other wastes.”
2.2.1 Interrogatory Response
As provided in draft Condition 4 of the 2015 SER, prior to disposing of Class A LLRW other
than DU in the Federal Cell, EnergySolutions will obtain approval of additional modeling.
EnergySolutions does not intend or request to dispose of other Class A LLRW above the DU
until such time as a PA accounting for the combined effects of DU and other Class A LLRW is
approved. This does not preclude approval to dispose of DU in the interim.
Other Wastes Responses for the Clive DU PA Model
23 Feb 2018 5
3.0 References
Neptune, 2011. Final Report for the Clive DU PA Model version 1.0, Neptune and Company
Inc., Los Alamos NM, June 2011
Neptune, 2015a. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015b. Radon Diffusion Modeling for the Clive DU PA, Clive DU PA Model v1.4,
NAC-0033_R1, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015c. Model Parameters for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0026_R4, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015d. Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility,
Clive DU PA Model v1.4, NAC-0018_R4, Neptune and Company Inc., Los Alamos NM,
November 2015
Neptune, 2018a. ET Cover Design Responses for the Clive DU PA Model, NAC-0106_R0,
Neptune and Company Inc., Lakewood CO, February 2018
Neptune, 2018b. Federal Cell Design Responses for the Clive DU PA Model, NAC-0101_R0,
Neptune and Company Inc., Lakewood CO, February 2018
SC&A, 2015. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
1, SC&A Inc., Vienna VA, April 2015
Utah DEQ, 2017. Division of Waste Management and Radiation Control, EnergySolutions Clive
LLRW Disposal Facility License No: UT2300249; RML #UT 2300249, Amended and
New Interrogatories Related to Clive DU PA Modeling Report Version 1.4 Dated
November 2015, Utah Department of Environmental Quality (DEQ), Salt Lake City UT,
May 2017
Whetstone Associates, 2011. EnergySolutions: Class A West Disposal Cell Infiltration and
Transport Modeling Report, prepared for EnergySolutions, Whetstone Associates Inc.,
Gunnison CO, November 2011

NAC-0104_R0

Groundwater Exposure Responses
for the Clive DU PA Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 ii

Groundwater Exposure Responses for the Clive DU PA Model
Groundwater Exposure Responses for the Clive DU PA Model.docx
Summary of how the deep and basal aquifers are addressed in GoldSim
model v. 1.4 and supporting documentation.

Gregg Occhiogrosso 12 Feb 2018
Mike Sully and Dan Levitt 12 Feb 2018

Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
FIGURES .................................................................................................................................. iv
ACRONYMS AND ABBREVIATIONS .................................................................................... v
1.0 Overview and Conceptual Model........................................................................................ 1
1.1 Considerations for a Groundwater Dose Pathway.......................................................... 2
1.1.1 Water Quality .......................................................................................................... 2
1.2 Interaction of Shallow and Lower Aquifers ................................................................... 3
2.0 UDEQ Interrogatory Responses.......................................................................................... 3
2.1 Interrogatory CR R313-25-3 and R313-25-8-195/1: Aquifer Characterization and
Interrogatory CR R313-25-20-204/1: Exposure to Groundwater ................................... 3
2.1.1 Interrogatory Responses .......................................................................................... 4
2.1.2 Evaluation of Applicability of UDEQ Basal and Lower Aquifer Contamination
Scenarios ................................................................................................................. 4
2.1.3 Updated Groundwater Ingestion Tc-99 Concentration Estimate ............................... 9
3.0 Conclusion ....................................................................................................................... 10
4.0 References ........................................................................................................................ 10

Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 iv
FIGURES
Figure 1. Stratigraphic profile showing ET cover, waste zone, and stratigraphy below the
Federal Cell. The deep aquifer is subdivided into the lower confined aquifer and
the basal aquifer. ........................................................................................................ 1
Figure 2. Conceptual model of Johnson et al. (2011). .................................................................. 5
Figure 3. Wells near York, NE. Generated using Nebraska Department of Natural Resources
online mapping tools. The total area depicted is roughly 68 square miles. .................. 6
Figure 4. Production wells within a 5-mile radius (~78 square miles) of the Clive Site.
Generated using Utah Division of Water Rights online tools. Note that well 16-
816 was drilled on January 9, 1996 and abandoned on January 30, 1996. This well
was contemplated as a source of construction water for the EnergySolutions Clive
facility, but not used................................................................................................... 7

Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 v
ACRONYMS AND ABBREVIATIONS
DEQ (Utah) Department of Environmental Quality
DU depleted uranium
ECL Effluent Concentration Limits
EPA (United States) Environmental Protection Agency
ET evapotranspiration
GWPL groundwater protection limit
NRC (United States) Nuclear Regulatory Commission
PA performance assessment
PRA probabilistic risk assessment
PWS public water supply
RO reverse osmosis
SER Safety Evaluation Report
TDS total dissolved solids
UDEQ Utah Department of Environmental Quality

Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 1
1.0 Overview and Conceptual Model
This document provides a summary of the important context for a discussion of potential risks
associated with groundwater at the Clive Site. As discussed below, the hydrogeologic setting,
geochemical conditions, and known usage history of the Site largely preclude traditional
groundwater dose scenarios. The saline groundwater (discussed in Section 1.1.1) is of limited
utility without extensive treatment. Low well yield in the shallow subsurface due to the soil
texture necessitates that the withdrawal of any substantial quantity of water from beneath the Site
involves extraction from lower aquifers, providing dilution of exposure concentrations
(discussed in Section 1.2). Sections 2.0 considers the latest UDEQ interrogatories.
Figure 1 depicts the generalized stratigraphic profile of the Site subsurface; detailed discussion
of the hydrogeologic system beneath the Site can be found in ET Cover Design Responses for the
Clive DU PA Model (Neptune 2018). The water table aquifer which would be impacted by any
migration from the waste cells is referred to as the shallow unconfined aquifer. Groundwater
occurring at greater depths is referred to as the lower confined (~70–100 ft below ground
surface) and basal (450–750 ft below ground surface) aquifers.

Figure 1. Stratigraphic profile showing ET cover, waste zone, and stratigraphy below the
Federal Cell. The deep aquifer is subdivided into the lower confined aquifer and the
basal aquifer.

Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 2
Groundwater exposure as a dose pathway has been the subject of several past interrogatories.
Certain exposure scenarios have been evaluated in past rounds of interrogatories with UDEQ.
Round 2 Interrogatory Responses (EnergySolutions 2014) calculated doses for well drilling
scenarios (reflecting practices commonly seen in the Clive area drainage basin) based on results
from v1.2 of the Clive DU PA Model. The general characteristics of these aquifers that are
relevant to potential groundwater dose scenarios are described below.
1.1 Considerations for a Groundwater Dose Pathway
1.1.1 Water Quality
The State of Utah classifies groundwater resources into four classes according to total dissolved
solids (TDS) concentrations. Class I waters, referred to as “pristine,” “irreplaceable,” or
“ecologically important,” are most strongly protected and are subject to the most stringent
protections regarding TDS and contaminant levels. Class II and Class III waters are referred to as
“drinking water quality” and “limited use,” respectively, and are similarly subjected to specific
but less stringent standards. Class IV groundwater, defined by TDS concentrations greater than
10,000 mg/L, is referred to as “saline” and has no specific protection standards.
Typical TDS concentrations at the Clive Site range from about 10,000 mg/L to about
70,000 mg/L, and its water has been defined by UDEQ as Class IV (saline groundwater) (Utah
2014). For perspective, typical seawater TDS concentrations are about 35,000 mg/L, and the
Environmental Protection Agency (EPA) Secondary Standard for the TDS level in drinking
water is 500 mg/L. Poor water quality at all depths in the immediate vicinity of the Site limits the
utility of the groundwater without treatment. An EnergySolutions analysis (EnergySolutions
2013) concluded that consumption of untreated, native groundwater would result in 100%
mortality of the receptor due to salinity levels, making radiological risks irrelevant.
The lower confined aquifer exhibits lower, though still elevated, TDS levels and is also
considered Class IV. While treatment of the groundwater to potable TDS levels is possible in
principle by utilizing a desalination process such as reverse osmosis (RO), this practice is not
found in the Clive facility basin. It has been acknowledged that RO treatment would also reduce
contaminant concentrations, which is discussed in more detail below. Additionally, the
probability of such a practice is low due to the lack of population and the availability of less
costly and higher quality water sources throughout the west desert. Extraction wells in the area
around the Site do exist, but these are generally installed in more productive gravel zones in the
recharge areas of the surrounding foothills and they are generally used for industrial purposes.
As such, a groundwater ingestion scenario in the immediate vicinity of the Site is highly
unlikely, is outside of the known usage history of the Site, and has never been required in
EnergySolutions’ various other UDEQ-approved performance assessments (PA), nor is it
supported by the promulgation of any new regulatory requirement since prior PA approvals.

Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 3
1.2 Interaction of Shallow and Lower Aquifers
Groundwater level measurements and geochemical data suggest minimal flow from the shallow
aquifer to the basal and lower aquifers. This is important to any discussion of a groundwater dose
scenario because the shallow aquifer is not of sufficient yield to be a productive source and
would not produce a reliable water supply for an inadvertent intruder using treated groundwater
as a drinking water source. Therefore, a productive well in the vicinity of the Site would involve
pumping from the basal or lower confined aquifers. Significant contamination of the lower
aquifers would only occur with advective movement of radionuclides from the shallow aquifer
(i.e., via downward water flow), which would be possible via movement through the naturally
occurring material or some anthropogenic pathway like a multi-aquifer well. As discussed below,
an anthropogenic pathway is very unlikely based on the Site conditions and usage history.
Based on simultaneous water level measurements in well clusters GW-19A/B, I-1-30/50/100,
and I-3-30/50/100, Bingham Environmental (1996) reported an apparent upward hydraulic
gradient, with freshwater equivalent head differentials of about 1–2 ft between deep (~100 ft
depth) and shallower (~30–50 ft depth) well screens. Corresponding gradients (0.02 to 0.04) and
upward velocities were calculated (0.05 ft/yr to 0.10 ft/yr). Bingham Environmental (1996) also
noted that solute and isotopic data suggest “minimal or no vertical movement” from the shallow
unconfined aquifer to the deeper confined aquifer, stating that, “Lower TDS concentrations for
the few confined aquifer samples suggest the unconfined and confined aquifers may be
chemically as well as hydraulically distinct systems.” This evidence suggests that contamination
of the lower aquifer due to the natural flow from the shallow aquifer is unlikely. UDEQ has
acknowledged the upward gradient by including a performance standard in the Groundwater
Quality Discharge Permit (Utah 2014) for the Site which states that a neutral or upward
hydraulic gradient should be maintained in nested or paired monitoring wells.
2.0 UDEQ Interrogatory Responses
This section contains responses for Interrogatories CR R313-25-3 and R313-25-8-195/1 (a single
interrogatory referencing two parts of the Utah Administrative Code) and CR R313-25-20-204/1.
2.1 Interrogatory CR R313-25-3 and R313-25-8-195/1: Aquifer
Characterization and Interrogatory CR R313-25-20-204/1:
Exposure to Groundwater
• Interrogatory CR R313-25-3 and R313-25-8-195/1: Aquifer Characterization
- Statement: Please provide information assessing the aquifer hydraulic properties
and groundwater quality for the lower confined aquifer (e.g., at 70–100 ft) and
valley-fill or basal-aquifer-system aquifers (e.g., at 450–750 ft) at the Clive site.
Specific types of information include, for example, groundwater flow velocities,
aquifer transmissivities, water quality, sorption properties, and the degree of
hydraulic interconnection between the upper and basal aquifers. Calculations
should be shown for horizontal and vertical components of groundwater flow and
contaminant migration velocities.
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 4
• Interrogatory CR R313-25-20-204/1: Exposure to Groundwater
- Statement: Please revise your June 8, 2014, partial response to Interrogatory 182
by extending it to 10,000 years and including the groundwater consumption
pathway, and include the results of the extended analysis in the next revision of the
DU PA, including the Appendix 19 sensitivity analyses.
- Excerpt of Basis: While preparing the April 2015 Safety Evaluation Report (SER),
the Department of Environmental Quality (DEQ)/SC&A extended the
EnergySolutions Interrogatory 182 partial response to 10,000 years and included
the groundwater consumption pathway, as well as several postulated scenarios
including a leaking well casing, a nearby failed or abandoned well that presents a
direct path between the upper and lower aquifer, and fresh water in the lower
aquifer. The results of this DEQ/SC&A analysis are given in the white paper,
“Groundwater Pathway Doses, Part 2,” Revision 2 (Marschke 2015).
2.1.1 Interrogatory Responses
These interrogatories share a common theme in their suggestion that various anthropogenic
scenarios might cause cross-contamination of the lower aquifer, and therefore a perceived need
for further development of the lower aquifer hydraulic properties. Because both stem from these
postulated scenarios, they are addressed together rather than as separate responses.
2.1.2 Evaluation of Applicability of UDEQ Basal and Lower Aquifer
Contamination Scenarios
Interrogatory R313-25-20-204/1 suggests the need to calculate groundwater consumption doses
based on the exposure mechanisms conjectured in Marschke (2015), which relied on two
literature sources not specific to Clive area practices to support the idea of cross-contamination
of a confined aquifer by the overlying unconfined aquifer (Johnson et al. 2011; Zinn and
Konikow 2007). The applicability of the scenarios presented in the literature sources is largely
unaddressed by Marschke (2015). Both sources, which consider mechanisms for contamination
from an unconfined aquifer moving to a confined aquifer via wells screened through both
aquifers, exclude justification of human groundwater consumption and are discussed below.
Johnson et al. (2011) conducted flow modeling for a synthetic domain meant to mimic the
hydrogeologic setting near York, Nebraska, an area where irrigation wells are very common (and
hydrogeologically different from the Clive drainage basin). The analysis attempted to quantify
the impacts of nearby irrigation wells to a public water supply (PWS) well placed in the same
vicinity. This analysis used hydraulic properties and hydrogeologic unit thicknesses specific to
the Nebraska site, and included multiple multi-aquifer wells near the pumping well that provided
hydraulic connection between the upper unconfined aquifer and the lower confined aquifer.
Figure 2 from Johnson et al. (2011) is reproduced in Figure 2 below to provide some context
regarding the conceptual model used in the study.

Note that this paper is prominently watermarked “WORKING DRAFT—DO NOT CITE OR QUOTE” yet was
cited by UDEQ.
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 5
Figure 2. Conceptual model of Johnson et al. (2011).
The vicinity of the Clive Site does not have a similar hydrogeologic setting or a similar history of
closely spaced irrigation wells when compared to the study site in NebraskaError! Reference
source not found.. Figure 3, generated from well installation records of the Nebraska
Department of Natural Resources, shows hundreds of irrigation wells (purple). In the
hydrogeologic setting of York, NE, cross-contamination of aquifers via the dense network of
wells may indeed be a viable scenario warranting an analysis like that undertaken by Johnson et
al. (2011). The scenario is not readily applicable to the Clive Site, as production wells installed
in the Clive vicinity are comparatively rare and are primarily in upgradient gravel units within
the recharge zone of the surrounding foothills, as shown in Figure 4. The well nearest the Clive
Site, identified as 16-816 in Figure 4, was abandoned in 1996. The hydraulic gradient in the
shallow aquifer is generally toward the northeast.
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 6

Figure 3. Wells near York, NE. Generated using Nebraska Department of Natural
Resources online mapping tools. The total area depicted is roughly 68 square miles.
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 7

Figure 4. Production wells within a 5-mile radius (~78 square miles) of the Clive Site.
Generated using Utah Division of Water Rights online tools. Note that well 16-816
was drilled on January 9, 1996 and abandoned on January 30, 1996. This well was
contemplated as a source of construction water for the EnergySolutions Clive facility,
but not used.
The postulated mechanism for contamination of the lower aquifer with radionuclides from the
Clive Site would require not only a network of pre-existing multi-aquifer wells near a pumping
well, but also that the wells be in a specific geometry relative to one another and the Site. Multi-
aquifer wells might promote the flow of uncontaminated water to a pumping well for many
possible configurations. Additionally, given the low yields of the upper aquifer, it is unlikely that
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 8
it could sustain prolonged elevated flows to the lower aquifer near the hypothesized multi-
aquifer wellbore. In short, the list of suppositions necessary for this mechanism of contamination
is long and specific. Johnson et al. (2011) also noted that:
Predicting the impact of contaminants on a PWS well due to leaking multi-aquifer wells is,
in general, not possible. However, the modeling steps described here can provide site-
specific insight into the potential for multi-aquifer wells to affect a confined-aquifer PWS
well.
Marschke (2015) made no attempt to apply the modeling steps laid out in Johnson et al. (2011) to
the Clive Site. Nevertheless, Marschke (2015) stated that “Johnson et al. (2011) determined that
a dilution factor of 10 would be appropriate” to calculate the relative flows from the upper and
lower aquifer to a well screened in the lower aquifer. Considering that no attempt was made to
apply the methodology on a site-specific basis, the dilution factor of 10 cited by Marschke
(2015) is arbitrary at best.
The second study cited, Zinn and Konikow (2007), is similarly a hypothetical model domain
meant to simulate the effects of existing, non-pumping wells on nearby pumping wells. The
authors noted the relevance of the study to regions with agricultural history like Nebraska and
Alberta, where numerous irrigation wells potentially affect local flow and transport. The model
domain size and shape are described as “arbitrary,” with the boundary conditions “conceptually
similar to a river valley”:
These conditions led to a predevelopment steady state regional flow field that was
conceptually similar to the side of a river valley or other simple groundwater basin—a
groundwater divide with primarily downward infiltration of water at the upstream end of the
system, predominantly horizontal flow near the middle part of the system, and predominantly
upward flow near the constant-head boundary.
The hydrologic conditions considered are not similar to the Clive Site. The model domain is
assigned homogenous properties, dissimilar to the Site (e.g., hydraulic conductivity of
76.2 m/day compared with Clive Site values on the order of 0.1 m/day). As noted above, the
combination of hydraulic properties and boundary conditions produces vertical hydraulic
gradients on either side of the domain. For certain configurations of pumping and non-pumping
wells, vertical flows in the non-pumping wells were altered. Zinn and Konikow (2007) noted that
“the effect of intraborehole flow is highly dependent on the location of the borehole within the
system, particularly in relation to the head field of the system.” This mechanism for
contamination of the lower aquifer is thus extremely improbable at the Clive Site, as it requires
an unlikely combination of pumping wells and non-pumping wells in a prescribed geometry.
Furthermore, the expense to install such a well pumping network far exceeds the cost of
transporting potable water from other sources outside of the Clive drainage basin. Moreover, as
described above in the discussion of Johnson et al. (2011), UDEQ made no attempt to apply this
scenario using site-specific hydrogeological unit structure or parameter values. As before, many
conditions which are inconsistent with the current use of the Clive Site are necessary for this
mechanism of contamination.
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 9
In summary, the applicability of these contamination scenarios to the Clive Site is not well
supported upon closer examination of their bases. NRC’s Regulatory Guide 1.174 expresses the
need for site-specific information in probabilistic risk assessments (PRAs):
One overriding requirement is that the PRA should realistically reflect the actual design,
construction, operational practices, and operational experience of the plant and its owner.
Regulatory Guide 1.200 also states that a consensus PRA standard is that PRAs should use
“plant-specific information versus generic information to represent the as-designed, as-built and
as-operated plant.”
Interrogatories CR R313-25-3 and R313-25-8-195/1 express the perceived need for further
evaluation of aquifer properties and flow regimes for the aquifers below the shallow unconfined
aquifer, based on scenarios developed for markedly different hydrogeologic conditions rather
than on site-specific information. Given the improbability of any scenario at the Clive Site by
which deeper aquifers would become contaminated and thus impact the Federal Cell’s ability to
satisfy the required performance objectives, further evaluation is completely unnecessary.
2.1.3 Updated Groundwater Ingestion Tc-99 Concentration Estimate
A simple bounding calculation based on v1.4 of the Clive DU PA Model results shows that even
modest mixing with lower aquifer water and RO treatment process results in exposure
concentrations of Tc-99 associated with basal and deep groundwater below 3,760 pCi/L.
One thousand realizations of the Model were run with groundwater concentrations calculated
until 10,000 years, as requested in the interrogatory. The peak median concentration (i.e., the
median value of Tc-99 concentrations across the realizations at each time step) was about
4,000 pCi/L at the hypothesized shallow aquifer groundwater well location. The 95th percentile
values for all times were below 45,000 pCi/L. As such, a total dilution factor of ~12 would
reduce Tc-99 concentrations to below the limit of 3,760 pCi/L, even for the peak 95th percentile
concentration.
As discussed above, due to the high salinity of the groundwater, reverse osmosis or a similar
process would be required prior to consumption. Marschke (2015) assumed a conservative
dilution factor of 10 due to the RO treatment (i.e., 90% removal). RO Tc-99 removal rates as
high as 99% (dilution factor of 100) have been documented (Williamson 1992). As such,
removal of Tc-99 by RO is likely adequate to bring well water concentrations well below
3,760 pCi/L, even for the 95th percentile concentrations.
An additional dilution factor due to mixing with water from the lower aquifer was estimated in
the Round 2 Interrogatory Responses and was also utilized for calculations by Marschke (2015).
It was calculated using the Thiem-Dupuit equation for the most likely (though still highly
improbable) scenario of a well drilled near the Clive Site and screened in the lower aquifer. The
resulting dilution factor was about 300. Applying this factor along with the RO dilution factor
reduces Tc-99 concentrations to negligible levels, on the order of 10 pCi/L, even for the 95th
percentile case.
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 10
In summary, though highly unlikely given the history and usage of the Clive Site, use of the
groundwater from the lower aquifer units would involve dilution of the contamination to an
extent that the risks to human health in a groundwater ingestion scenario would be negligible.
3.0 Conclusion
The poor water quality (Class IV) of water at the Clive Site limits its utility without treatment,
which would reduce radionuclide concentrations along with TDS concentrations. Substantial
withdrawals from the shallow aquifer are difficult due to low well yields, and cross-
contamination of the underlying aquifers is unlikely due to the natural hydrogeologic separation
and observed upward hydraulic gradients. Scenarios proposed by UDEQ involving
anthropogenic hydraulic connections are not applicable to the Clive Site, as these scenarios are
based on literature studies of areas with starkly contrasting usage histories and hydrogeologic
conditions.
4.0 References
Bingham Environmental, 1996. Revised Hydrogeologic Report, prepared for Envirocare of Utah
Inc., Bingham Environmental Inc., Salt Lake City UT, February 1996
EnergySolutions, 2013. Utah Radioactive Material License (RML UT2300249) Updated Site-
Specific Performance Assessment, Revision 1, prepared for Utah Division of Radiation
Control, EnergySolutions LLC, Salt Lake City UT, December 2013
EnergySolutions, 2014. Responses to August 11, 2014—Supplemental Interrogatories, Utah
LLRW Disposal License RML UT 2300249 Condition 35 Compliance Report, prepared
for Utah Division of Radiation Control, EnergySolutions LLC, Salt Lake City UT,
August 2014
Johnson, R.L., et al., 2011. Modeling the Potential Impact of Seasonal and Inactive Multi-
Aquifer Wells on Contaminant Movement to Public Water-Supply Wells, Journal of the
American Water Resources Association 47 (3) 588–596
Marschke, S., 2015. Groundwater Pathway Doses, Part 2, Revision 2, prepared for Utah
Department of Environmental Quality, SC&A Inc., Vienna VA, May 2015
Neptune, 2018. ET Cover Design Responses for the Clive DU PA Model, NAC-0106_R0,
Neptune and Company Inc., Lakewood CO, February 2018
Utah, 2014. Ground Water Quality Discharge Permit, Permit No. UGW450005, State of Utah,
Division of Water Quality, Salt Lake City UT, 2014
Williamson, D., 1992. Bench-Scale Testing of Reverse Osmosis to Remove Technetium 99 and
Trichloroethylene from Groundwater, Proceedings of National Research and
Groundwater Exposure Responses for the Clive DU PA Model
23 Feb 2018 11
Development Conference on the Control of Hazardous Materials, February 1992, San
Francisco CA
Zinn, B.A., and L.F. Konikow, 2007. Effects of Intraborehole Flow on Groundwater Age
Distribution, Hydrogeology Journal 2007 (15) 633–643 doi: 10.1007/s10040-006-0139-8

NAC-0103_R0

Recycled Uranium Responses for
the Clive DU PA Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 ii

Recycled Uranium Responses for the Clive DU PA Model
Recycled Uranium Responses for the Clive DU PA Model.docx
Summary of how recycled uranium waste is addressed in GoldSim model
v. 1.4 and supporting documentation.

Gregg Occhiogrosso 13 Feb 2018
Mike Sully and Dan Levitt

Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
ACRONYMS AND ABBREVIATIONS ................................................................................... iv
1.0 Overview and Conceptual Model........................................................................................ 1
2.0 UDEQ Interrogatory Responses.......................................................................................... 2
2.1 Interrogatories CR R313-25-8(4)(a)-08/1 and CR R313-25-7(9)-51/3 ........................... 2
2.1.1 Interrogatory Response ............................................................................................ 2
3.0 Conclusion ......................................................................................................................... 4
4.0 References .......................................................................................................................... 5

Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 iv
ACRONYMS AND ABBREVIATIONS
CAW Class A West embankment
DEQ (Utah) Department of Environmental Quality
DU depleted uranium
ET evapotranspiration
GDP gaseous diffusion plant
GWPL groundwater protection limit
PA performance assessment
PGDP Paducah Gaseous Diffusion Plant
PORTS Portsmouth Gaseous Diffusion Plant
SER Safety Evaluation Report
UDEQ Utah Department of Environmental Quality
WAC waste acceptance criteria

Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 1
1.0 Overview and Conceptual Model
This document discusses issues related to the disposal of depleted uranium (DU) containing
recycled uranium in the Federal Cell. Recycled uranium waste contains fission products such as
Tc-99 which are not necessarily present in other DU wastes. UDEQ issued interrogatories related
to Tc-99’s source term concentration and the groundwater concentration at compliance
endpoints, but deemed the questions closed, stating that any license would be granted under the
condition that no DU containing recycled uranium would be disposed at the Site. Precedent
exists for disposal of Tc-99 at the Clive Site, as UDEQ has permitted Tc-99 disposal in the
adjacent Class A West (CAW) Disposal Cell. Issues surrounding disposal of DU containing
recycled uranium are discussed, including the mechanisms of exposure and responses to the
interrogatories (Section 2.0).
In the performance assessment (PA) model, all radionuclides, including Tc-99 and other fission
products, are transported via various transport processes which would cause contaminant
migration through the waste zone and into the underlying geologic units. Exposures are
evaluated at the hypothetical receptor locations (compliance points).
Contaminant transport includes transport media (water, air, soil), transport processes (advection-
dispersion, diffusion, plant uptake, soil translocation), and partitioning between phases.
Diffusion occurs in gas and water phases. Dilution occurs when mixing with less concentrated
water. Hydrodynamic dispersion is associated with water advection. Dissolution in water is
limited by aqueous solubility.
Infiltration of water through the cover, into wastes, and toward a hypothetical groundwater user
is the transport pathway for Tc-99 cited in UDEQ’s interrogatories related to recycled uranium.
Tc-99 is unique in that it is regarded as relatively long half-lived and mobile in the environment
because it does not generally sorb to the soil structure as readily as other radionuclides, meaning
it can be transported for longer distances.
The Clive DU PA Model calculates potential exposure concentrations in a probabilistic fashion
by incorporating uncertainty in each parameter value through Monte Carlo simulation, in which
the model is run thousands of times using different combinations of parameter values in each
realization. Details of the modeling processes can be found in the Conceptual Site Model for
Disposal of Depleted Uranium at the Clive Facility (Neptune 2015b).
Tc-99 has been accepted for disposal of waste with a limiting Tc-99 concentration in the adjacent
Class A West embankment, based on deterministic modeling using extreme bounding conditions.
The cited basis for the prohibition of DU containing recycled uranium is that the PA Model does
not adequately capture the range of possible waste concentrations. UDEQ goes on to claim that
the evapotranspiration (ET) cover performance may not adequately limit infiltration to assure
compliance with the groundwater protection limit (GWPL) at the hypothetical receptor location.
ET cover performance is addressed in ET Cover Design Responses for the Clive DU PA Model
(Neptune 2018a).
Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 2
2.0 UDEQ Interrogatory Responses
This section contains responses for Interrogatories CR R313-25-8(4)(a)-08/1 and CR R313-25-
7(9)-51/3. UDEQ’s conclusions on the interrogatories regarding the disposal of wastes
containing recycled uranium are reproduced below. A single response is provided since the
interrogatories are essentially identical.
2.1 Interrogatories CR R313-25-8(4)(a)-08/1 and CR R313-25-7(9)-51/3
Interrogatory CR R313-25-8(4)(a)-08/1: Groundwater Concentration Endpoints
DEQ has stated that no depleted uranium (DU) waste containing recycled uranium will be
allowed to be disposed at Clive, so this interrogatory is closed.
Interrogatory CR R313-25-7(9)-51/3: Nature of Contamination
This interrogatory is closed because any license amendment will contain a license condition that
disposal of recycled uranium is not allowed in the DU waste. Furthermore, the license condition
will indicate that DU-waste containers shall contain neither heels of enriched uranium at average
concentrations greater than that allowed in the license nor heels of transuranic compounds at
average concentrations greater than 10 pCi/g (the Class A limit).
2.1.1 Interrogatory Response
Interrogatories 08/1 and 51/3 were deemed closed under the unsubstantiated condition that no
DU waste containing recycled uranium would be disposed at the Site, to be dictated by License
condition. These interrogatories are linked by their concern with groundwater exposures to
radionuclides, principally fission product radionuclides like Tc-99 which may be present in DU
wastes containing recycled uranium. UDEQ cited uncertainty in Tc-99 concentrations in
groundwater relative to the GWPL of 3790 pCi/L over the compliance period of 500 years to
justify the prohibition of waste containing recycled uranium (April 2015 SER, Section 6.1.2):
Because there is significant uncertainty regarding the Tc-99 concentration in the DU3O8
to be produced from the GDP tailings, and because Tc-99 and other mobile isotopes
may exceed the GWPL at 500 years, DEQ approves this portion of the DU PA with the
condition that no DU waste containing recycled uranium be accepted for disposal inside
the Federal Cell at Clive. Based on this restriction, GWPLs for the 500-year compliance
period can easily be met regardless of uncertainties in the infiltration rate through the ET
cover.
The concerns expressed are thus related to the waste concentrations upon emplacement (as
opposed to hypothetical ranges of concentrations in DU being stored at the Clive Facility) and
the performance of the cover system and the associated infiltration rates, which impact potential
exposure concentrations at hypothetical receptor locations.
From a risk perspective, this condition is arbitrary and unsupported, given that modeling results
demonstrate that DU containing Tc-99 and other fission products can be safely disposed at the
Federal Cell. Furthermore, Tc-99 in other Class A wastes is licensed for disposal in other
comparable embankments at the Clive Site. The relevant modeling concepts are briefly
summarized below, followed by responses to the concerns expressed in the SER which
precipitated the closure of these interrogatories.
Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 3
Comparison of v1.4 Model Results to GWPL
Refinements to the PA Model incorporated in v1.4 have resulted in predicted Tc-99
concentrations in the upper aquifer well below the GWPL (3,790 pCi/L) over the 500-year
compliance period. As presented in the v1.4 Final Report for the Clive DU PA Model (Neptune
2015a), ten thousand realizations of the Model were run and the peak Tc-99 concentration at the
hypothetical groundwater compliance well was recorded for each run. The mean and median
peak values across the realizations were 26 pCi/L and 4.3E-2 pCi/L, respectively. The 95th
percentile peak value was 150 pCi/L. These results are a small fraction of the GWPL of
3790 pCi/L.
Due to these updated estimates, the prohibition of DU wastes containing recycled uranium is not
required to remain in compliance with the GWPL, as even the 95th percentile estimates of peak
Tc-99 concentrations are more than an order of magnitude below the GWPL for the 500-year
compliance period. Any refusal to remove a prohibition on disposal of recycled DU is arbitrary,
unsupported by the Model, and contrary to site precedence.
Responses to UDEQ Concerns
1) ET Cover Performance
Issues pertaining to the ET cover performance and net infiltration into the waste zone are
discussed in ET Cover Design Responses for the Clive DU PA Model (Neptune 2018a).
2) Tc-99 Waste Concentrations
Some uncertainty in the Tc-99 concentrations of wastes derived from gaseous diffusion plant
(GDP) operations is acknowledged, as direct measurements of fission product concentrations are
rare. However, a blanket prohibition of DU containing recycled uranium is not justifiable due to
uncertainties in Tc-99 concentrations. Tc-99 disposals at the Site in the CAW cell undergo
receipt and acceptance actions to confirm compliance with the waste acceptance criteria (WAC),
with WAC values established using a transport modeling approach not unlike the Federal Cell
PA Model. Federal Cell disposals would be subject to similar receipt and acceptance procedures,
and a WAC could be established in a similar fashion.
UDEQ’s contention that Tc-99 concentrations in the Model are not adequate is based on Section
4.1.2 of the April 2015 SER, which hypothesizes a possible Tc-99 concentration in cylinder
heels as high as 5,700,000 ppb. This value is based on a mass balance approach summarized in
Hightower et al. (2000); it should be interpreted as an extreme bounding value, and is described
as such by Hightower et al. (2000). As noted in NUREG/CR-2300 (NRC 1983):
The simplest quantitative measure of variability in a parameter or a measurable quantity is
given by an assessed range of the values the parameter or quantity can take. This measure
may be adequate for certain purposes (e.g., as input to a sensitivity analysis), but in general it
is not a complete representation of the analyst's knowledge or state of confidence and
generally will lead to an unrealistic range of results if such measures are propagated through
an analysis.
Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 4
As such, a bounding, hypothetical maximum value is of limited utility and should be discounted
in favor of measured values from similar waste streams. A more realistic, measured value for an
upper Tc-99 cylinder heel concentration is presented in Appendix C of Hightower et al. (2000),
where measurements of heel concentrations were taken from a cylinder used multiple times at
both Portsmouth and Paducah:
Several empty feed cylinders have been identified that contain heels of feed and/or product
from the periods when reactor returns were being fed to the cascades. Cylinder 003174 is
such a cylinder with 13 lb of feed material heel remaining. This cylinder had been filled at
the PGDP [Paducah Gaseous Diffusion Plant] UF6 feed plant with UF6 prepared from
reactor returns material sometime after the cylinder’s purchase date of June 1954. After the
original charge in this cylinder had been fed to the cascades, the cylinder was filled with
enriched product from Paducah and shipped to the PORTS [Portsmouth Gaseous Diffusion
Plant], where the enriched material was fed to the Portsmouth cascade for further
enrichment. This cylinder was similarly refilled with enriched product and emptied two more
times without any washing, leaving its original heel of reactor returns feed material
combined with the heels left from emptying enriched product three times into the cascades.
(Hightower et al. (2000), Appendix C)
The highest, “first wash” Tc-99 concentration for this cylinder is reported as 270,000 ppb, a
factor of 20 lower than UDEQ’s hypothetical extreme bounding value derived from the mass
balance approach. The April 2015 SER claimed that the PA Model concentrations were as much
as 3.7 times too low when compared to the value of 5,700,000 ppb. When compared to the
measurements from this heavily used cylinder, however, the uncertainty currently incorporated
in the Model for Tc-99 concentrations is representative.
It is also worth noting that mobilization of the heel material for concentration measurement
required dissolving the heel with an acidic wash solution. The Model conservatively takes no
transportation rate credit for this solid waste form or waste packaging, and assumes all
radionuclides are immediately available for transport in the liquid phase at the first model time
step, a simple and conservative approach compared to more complicated release models that
incorporate mass-transfer limitations due to, for example, solubility and/or sorption
(NUREG/CR-5532, Kozak et al. (1990)). Additional discussion of the potential for doses via
groundwater is discussed in the response to Interrogatory CR R313-25-20-204/1: Exposure to
Groundwater in Groundwater Exposure Responses for the Clive DU PA Model (Neptune 2018b).
3.0 Conclusion
UDEQ’s stated prohibition of disposal of DU containing recycled uranium is based on
uncertainties in the performance of the cover system and in the concentration of fission products
in the wastes. As shown above, the latter concern is based on an extreme theoretical value.
Furthermore, the prohibition stands in contrast to site precedence, as disposals of Tc-99 have
been permitted in the CAW Cell, subject to WAC limitations. A similar framework could be
applicable to the Federal Cell, though DU PA Model v1.4 results suggest the risk associated due
to disposal of DU containing recycled uranium is minimal, and that no such WAC limitation is
warranted.

Recycled Uranium Responses for the Clive DU PA Model
23 Feb 2018 5
4.0 References
Hightower, J.R., et al., 2000. Strategy for Characterizing Transuranics and Technetium
Contamination in Depleted UF6 Cylinders, ORNL/TM-2000/242, Oak Ridge National
Laboratory, Oak Ridge TN, October 2000
Kozak, M.W., et al., 1990. A Performance Assessment Methodology for Low-Level Waste
Facilities, NUREG/CR-5532, SAND90-0375, prepared for United States Nuclear
Regulatory Commission, Sandia National Laboratories, Albuquerque NM, July 1990
Neptune, 2015a. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015b. Conceptual Site Model for Disposal of Depleted Uranium at the Clive Facility,
Clive DU PA Model v1.4, NAC-0018_R4, Neptune and Company Inc., Los Alamos NM,
November 2015
Neptune, 2018a. ET Cover Design Responses for the Clive DU PA Model, NAC-0106_R0,
Neptune and Company Inc., Lakewood CO, February 2018
Neptune, 2018b. Groundwater Exposure Responses for the Clive DU PA Model, NAC-0104_R0,
Neptune and Company Inc., Lakewood CO, February 2018
NRC, 1983. PRA Procedures Guide: A Guide to the Performance of Probabilistic Risk
Assessments for Nuclear Power Plants, Volume 2, Chapters 9–13 and Appendices A–G,
NUREG/CR-2300, Vol. 2, prepared by The American Nuclear Society and The Institute
of Electrical and Electronics Engineers, United States Nuclear Regulatory Commission,
Washington DC, January 1983

NAC-0101_R0

Federal Cell Design Responses for
the Clive DU PA Model
23 February 2018

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 ii

Federal Cell Design Responses for the Clive DU PA Model
Federal Cell Design Responses for the Clive DU PA Model.docx
Summary of Federal Cell design assumptions and considerations in GoldSim
model v. 1.4.

Sean McCandless 12 Feb 2018
Mike Sully 12 Feb 2018

Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 iii
CONTENTS
CONTENTS ............................................................................................................................. iii
FIGURES .................................................................................................................................. iv
ACRONYMS AND ABBREVIATIONS .................................................................................... v
1.0 Overview and Conceptual Model........................................................................................ 1
2.0 UDEQ Interrogatory Responses.......................................................................................... 4
2.1 Interrogatory CR R313-25-7(2) and 7(6)-81/2: Comparison of Disposal Cell Designs .. 4
2.1.1 Interrogatory Response ............................................................................................ 4
2.2 Interrogatory CR R313-25-7(6)-84/3: Below-Grade Disposal of DU ............................. 4
2.2.1 Interrogatory Response ............................................................................................ 4
2.3 Interrogatory CR R313-25-7(2)-160/2: Comparison of Class A West and Federal Cell
Designs ......................................................................................................................... 5
2.3.1 Interrogatory Response ............................................................................................ 5
2.4 Interrogatory CR R313-25-22-162/2: Disposal Cell Stability ........................................ 6
2.4.1 Interrogatory Response ............................................................................................ 6
3.0 References .......................................................................................................................... 6

Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 iv
FIGURES
Figure 1. Federal Cell and 11e.(2) Cell engineering drawing 14004 V1A (Neptune 2015b). ........ 2
Figure 2. Federal Cell and 11e.(2) Cell engineering drawing 14004 V3A (west-east cross
section) (Neptune 2015b). .......................................................................................... 3

Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 v
ACRONYMS AND ABBREVIATIONS
CAW Class A West embankment
CQA/QC Construction Quality Assurance/Quality Control
DU depleted uranium
ET evapotranspiration
LLRW low-level radioactive waste
PA performance assessment
SER Safety Evaluation Report
UDEQ Utah Department of Environmental Quality

Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 1
1.0 Overview and Conceptual Model
Version 1.4 of the Clive DU PA is based on a Federal Cell design that physically isolates DU
and Class A LLRW from the adjacent 11e.(2) embankment. This is a change from the approach
modeled in versions 1.0 and 1.2, which consisted of a single cell with a barrier wall intended to
isolate the different waste types.
The choice to revise the Federal Cell design to a standalone embankment was made for several
reasons:
1. The Federal Cell is proposed to have an evapotranspiration (ET) cover design, while the
existing 11e.(2) cell has a rock armor cover design. Final cover has been constructed over
a large portion of the currently open 11e.(2) cell footprint. Joining the two cover designs
would be difficult to engineer and model.
2. Remediating the existing completed 11e.(2) cover to an ET design would be costly in
terms of time, labor, and materials. The relatively small loss of overall site capacity,
compared with a single-cell design, was not judged to warrant the activity.
Appendix 3 to Neptune (2015a), Embankment Modeling for the Clive DU PA, Clive DU PA
Model v. 1.4 (Neptune 2015b), explains that version 1.4 of the DU PA considers a single,
standalone Federal Cell. This cell is located to the west of the existing 11e.(2) mill tailings cell
and to the south of the existing Class A West (CAW) cell. EnergySolutions engineering drawing
series 14004 was referenced and included in Appendix 3. Drawings 14004 V1A and 14004 V3A
are reproduced from this report as Figure 1 and Figure 2, respectively, below.
The Federal Cell isolates DU (and, ultimately, overlying Class A LLRW) from the adjacent
11e.(2) embankment. The Federal Cell has no existing waste placement of any type. It appears
that at least one of the interrogatories on this topic fails to recognize the design revision from
version 1.2 to version 1.4 of the DU PA Model.
Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 2

Figure 1. Federal Cell and 11e.(2) Cell engineering drawing 14004 V1A (Neptune 2015b).
Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 3

Figure 2. Federal Cell and 11e.(2) Cell engineering drawing 14004 V3A (west-east cross
section) (Neptune 2015b).
Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 4
2.0 UDEQ Interrogatory Responses
This section contains responses for Interrogatories CR R313-25-7(2), CR R313-25-7(6)-81/2, CR
R313-25-7(6)-84/3, CR R313-25-7(2)-160/2, and CR R313-25-22-162/2.
2.1 Interrogatory CR R313-25-7(2) and 7(6)-81/2: Comparison of
Disposal Cell Designs
No further analysis has been performed on disposal cell designs since v1.2.
2.1.1 Interrogatory Response
This statement is incorrect. Version 1.2 of the DU PA proposed to dispose DU, Class A LLRW,
and 11e.(2) waste within a single embankment, with DU and Class A LLRW separated from
11e.(2) waste by a clay barrier wall. As discussed above, version 1.4 of the DU PA models a
distinct Federal Cell that is completely separate from the 11e.(2) cell.
The interrogatory basis furthermore states:
…EnergySolutions has advised that the proposed Federal Cell will be physically separated from
the 11e.(2) cell. EnergySolutions has provided only engineering drawings but no written
description of the new cell (i.e., Appendices 3 and 16 to the DU PA have not been revised).
This statement is also incorrect. Appendices 3 and 16 of the DU PA, v1.4, correctly incorporate
the Federal Cell design and dimensions as a single, standalone embankment separate from the
11e.(2) cell. There is no reason for this interrogatory to remain open.
2.2 Interrogatory CR R313-25-7(6)-84/3: Below-Grade Disposal of DU
No further analysis was performed in Appendix 21 on the below-grade disposal of DU.
2.2.1 Interrogatory Response
The interrogatory basis discusses a discrepancy between the reported potential number of DU
drums and cylinders in Appendices 3 and 4 to Neptune (2015a):
Please explain the difference between v1.4 Appendices 3 and 4 regarding the maximum number
of cylinders and drums, and demonstrate how the entire DU inventory can be disposed below
grade.
This discrepancy traces back to the use of different EnergySolutions drawings as points of
reference for the Neptune documents. Appendix 3 to Neptune (2015a) includes engineering
drawing 14004 L1A as Figure 7 and as the basis for the potential number of DU drums and
cylinders that could physically fit below grade in the Federal Cell. Appendix 4 to Neptune
(2015a) references drawing 14004 SK1 as its data source. Clearly, the different drawings are
intended to project different disposal scenarios.
While the difference between the scenarios is acknowledged, it is also irrelevant to the licensing
decision. Section 6.2.4 of the 2015 SER provides Condition 2 that would apply to an amended
Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 5
License approving DU disposal: “All DU waste must be disposed of below the original-grade
level of the proposed Federal Cell (i.e., 4,272 ft-amsl).” This constraint will control waste
placement operations to be consistent with the important model assumption that DU waste is
placed below the original grade of the Federal Cell, regardless of the number or type of cylinders
and/or drums disposed.
There is no Utah or Federal Rule whereby EnergySolutions must demonstrate that all DU in
storage (or that will ever be created at any site within the United States) must completely fit
within the permitted capacity of the Federal Cell. However, this interrogatory demands, without
regulatory basis, that EnergySolutions “demonstrate how the entire DU inventory can be
disposed below grade.” By refusing to close interrogatories associated with DU capacity under
various disposal configurations, UDEQ’s reviewers imply a requirement that the Site be able to
dispose of the full national inventory of DU available for disposal. This is without regulatory
basis and is thus arbitrary and capricious on the part of UDEQ. Note also that, if approved, the
EnergySolutions Clive facility would not be the only U.S. site licensed for DU disposal.
EnergySolutions will only place DU below grade within the approved top slope footprint of the
Federal Cell, and in accordance with Construction Quality Assurance/Quality Control
(CQA/QC) requirements applicable for the waste form and container. These controls, together
with the type and timing of shipments for disposal, will drive actual cell utilization. Disposal
operations will be performed to optimize cell utilization within License constraints, but will often
result in under-utilized capacity due to material handling and spacing needs, use of clean fill for
shielding and contamination control, and other inefficiencies.
For example, the figures used to estimate potential total cylinders and drums that could be
disposed under various configurations assume regular, tight spacing for each container. It will
not always be operationally practical to maintain tight spacing, perhaps due to rigging or
equipment needs. Regardless of these realities and of the actual volume ultimately disposed,
further disposal of DU in the Federal Cell will cease once the below grade capacity is exhausted.
2.3 Interrogatory CR R313-25-7(2)-160/2: Comparison of Class A
West and Federal Cell Designs
See also Interrogatory CR R313-25-25(4)-202/1: Use of SIBERIA to Model Federal Cell Erosion.
2.3.1 Interrogatory Response
A response to Interrogatory 202/1 regarding use of SIBERIA to model erosion is provided in
Erosion Responses for the Clive DU PA Model (Neptune 2018).
The basis for Interrogatory 160/2 goes on to discuss a prior request (SC&A 2015) to compare
similar ET cover designs between the Class A West and Federal Cell:
DRC is currently reviewing a license amendment request to use an ET cover of similar design to
that proposed for the Federal Cell in the DU PA. Any recommendations and conclusions from
that review must be applied to the proposed Federal Cell as well.
Federal Cell Design Responses for the Clive DU PA Model
23 Feb 2018 6
Since submitting the request for approval of an ET cover design on the CAW cell,
EnergySolutions has begun construction of a rock armor cover on the CAW cell. More
importantly, the DU PA models the Federal Cell’s ET cover as designed. CAW cover licensing
is separate and distinct from the Federal Cell. Since their geometry differs, any attempt to couple
or join the reviews is unsupported and arbitrary. Therefore, this aspect of the question is moot.
2.4 Interrogatory CR R313-25-22-162/2: Disposal Cell Stability
This interrogatory can be closed because the same issues are raised in Interrogatory CR R313-
25-7(2)-160/2: Comparison of Class A West and Federal Cell Designs, which remains open.
2.4.1 Interrogatory Response
We concur that this interrogatory should be closed. See above for discussion of Interrogatory CR
R313-25-7(2)-160/2.
3.0 References
Neptune, 2015a. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
Neptune, 2015b. Embankment Modeling for the Clive DU PA, Clive DU PA Model v1.4, NAC-
0019_R4, Neptune and Company Inc., Los Alamos NM, October 2015
Neptune, 2018. Erosion Responses for the Clive DU PA Model, NAC-0108_R0, Neptune and
Company Inc., Lakewood CO, February 2018
SC&A, 2015. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
1, SC&A Inc., Vienna VA, April 2015

NAC-0147_R0

Clive DU PA Model—Response
to Model Version 1.4 Amended
Interrogatories
24 April 2020

Prepared by

NEPTUNE AND COMPANY, INC.
1435 Garrison St, Suite 201, Lakewood, CO 80215
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 ii
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
Clive DU PA Model - Response to Model Version 1.4 Amended
Interrogatories.docx
Responses to UDEQ Letter “Depleted Uranium Performance Assessment
(DU PA); Clive Facility; Model Version 1.4 Amended Interrogatories,” dated July 25, 2019.
Dan Levitt, Paul Black, and Sean
McCandless
08 April 2020
Mike Sully, Gregg Occhiogrosso, Aharon
Fleury, Bruce Crowe
10 April 2020

Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 iii
CONTENTS
CONTENTS ................................................................................................................................... iii
FIGURES ....................................................................................................................................... vi
TABLES ....................................................................................................................................... vii
ACRONYMS AND ABBREVIATIONS .................................................................................... viii
Executive Summary ........................................................................................................................ 1
1.0 Introduction ............................................................................................................................ 4
2.0 UDEQ Interrogatory Responses ............................................................................................ 5
2.1 Interrogatory CR R313-25-7(2)-05/2: Radon Barriers .................................................... 5
2.1.1 Interrogatory Response ............................................................................................... 6
2.2 Interrogatory CR R313-22-32(2)-10/3: Effect of Biologicals on Radionuclide
Transport ........................................................................................................................ 12
2.2.1 Interrogatory Response ............................................................................................. 12
2.3 Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation .............................. 13
2.3.1 Interrogatory Response ............................................................................................. 14
2.4 Interrogatory CR R317-6-2.1-20/2: Groundwater Concentrations ................................ 16
2.4.1 Interrogatory Response ............................................................................................. 16
2.5 Interrogatory CR R313-25-8(4)(d)-21/2: Infiltration Rates ........................................... 19
2.5.1 Interrogatory Response ............................................................................................. 21
2.6 Interrogatory CR R313-25-8(4)(a)-28/3: Bioturbation Effects and Consequences ....... 22
2.6.1 Interrogatory Response ............................................................................................. 22
2.7 Interrogatory CR R313-25-7(3)-60/2: Modeled Radon Barriers ................................... 22
2.7.1 Interrogatory Response ............................................................................................. 22
2.8 Interrogatory CR R313-25-8(4)(a)-71/1: Biotic Processes in Gully Formation ............ 22
2.8.1 Interrogatory Response ............................................................................................. 23
2.9 Interrogatory CR R313-25-7(2) and 7(6)-81/2: Comparison of Disposal Cell Designs 24
2.9.1 Interrogatory Response ............................................................................................. 24
2.10 Interrogatory CR R313-25-7(1–2)-90/2: Calibration of Infiltration Rates .................... 24
2.10.1 Interrogatory Response ............................................................................................. 25
2.11 Interrogatory CR R313-25-7(1)-100/2: Groundwater Recharge from Precipitation ..... 25
2.11.1 Interrogatory Response ............................................................................................. 25
2.12 Interrogatory CR R313-25-8(4)(a)-108/2: Biointrusion ................................................ 25
2.12.1 Interrogatory Response ............................................................................................. 25
2.13 Interrogatory CR R313-25-8(4)(a)-112/2: Hydraulic Conductivity .............................. 25
2.13.1 Interrogatory Response ............................................................................................. 26
2.14 Interrogatory CR R313-25-8(4)(D)-132/2: Sedimentation Model ................................. 26
2.14.1 Interrogatory Response ............................................................................................. 26
2.15 Interrogatory CR R313-25-7(2)-150/3: Plant Growth and Cover Performance ............ 26
2.15.1 Interrogatory Response ............................................................................................. 26
2.16 Interrogatory CR R313-25-8(4)(d)-153/2: Impact of Pedogenic Processes on the
Radon Barrier ................................................................................................................. 26
2.16.1 Interrogatory Response ............................................................................................. 26
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 iv
2.17 Interrogatory CR R313-25-7(2)-160/2: Comparison of Class A West and Federal Cell
Designs ........................................................................................................................... 27
2.17.1 Interrogatory Response ............................................................................................. 27
2.18 Interrogatory CR R313-25-22-162/2: Disposal Cell Stability ....................................... 27
2.18.1 Interrogatory Response ............................................................................................. 27
2.19 Interrogatory CR R313-25-7(2)-175/1: Infiltration Rates for the Federal Cell Versus
the Class A West Cell .................................................................................................... 27
2.19.1 Interrogatory Response ............................................................................................. 27
2.20 Interrogatory CR R313-25-8(5)(a)-176/1: Representative Hydraulic Conductivity
Rates ............................................................................................................................... 27
2.20.1 Interrogatory Response ............................................................................................. 28
2.21 Interrogatory CR R313-25-7(2)-189/3: Modeling Impacts of Changes in Federal Cell
Cover-System Soil Hydraulic Conductivity and Alpha Values ..................................... 28
2.21.1 Interrogatory Response ............................................................................................. 28
2.22 Interrogatory CR R313-25-7(2)-191/3: Effect of Gully Erosion ................................... 28
2.22.1 Interrogatory Response ............................................................................................. 29
2.23 Interrogatory CR R313-25-7(2)-192/3: Implications of Great Salt Lake Freezing on
Federal Cell Performance ............................................................................................... 30
2.23.1 Interrogatory Response ............................................................................................. 30
2.24 Interrogatory CR R313-25-3 and R313-25-8-195/1: Aquifer Characterization ............ 30
2.24.1 Interrogatory Response ............................................................................................. 30
2.25 Interrogatory CR R313-25-9(5)(A)-196/1: Non-DU Waste Characteristics ................. 31
2.25.1 Interrogatory Response ............................................................................................. 31
2.26 Interrogatory CR R313-25-25(4) 197/1: Properties of Embankment Side Slope
Materials ......................................................................................................................... 31
2.26.1 Interrogatory Response ............................................................................................. 31
2.27 Interrogatory CR R313-25-25(4)-198/1: Gravel Content of Embankment Materials ... 32
2.27.1 Interrogatory Response ............................................................................................. 32
2.28 Interrogatory CR R313-25-25(4)-199/1: Uncertainties in Erosion Modeling ............... 32
2.28.1 Interrogatory Response ............................................................................................. 32
2.29 Interrogatory CR R313-25-25(4)-200/1: Use of RHEM to Develop Parameters for
SIBERIA ........................................................................................................................ 32
2.29.1 Interrogatory Response ............................................................................................. 33
2.30 Interrogatory CR R313-25-25(4)-201/1: Estimating Rainfall Intensity ........................ 33
2.30.1 Interrogatory Response ............................................................................................. 33
2.31 Interrogatory CR R313-25-25(4)-202/1: Use of SIBERIA to Model Federal Cell
Erosion ........................................................................................................................... 33
2.31.1 Interrogatory Response ............................................................................................. 33
2.32 Interrogatory CR R313-25-9(5)(a)-206/1: Temporal Uncertainty in Performance
Assessment ..................................................................................................................... 34
2.32.1 Interrogatory Response ............................................................................................. 34
2.33 Interrogatory CR R313-25-23-207/1: Stability of Disposal Site ................................... 36
2.33.1 Interrogatory Response ............................................................................................. 36
2.34 SER B.1 Supplemental Interrogatory Comment 1 ......................................................... 36
2.34.1 Interrogatory Response ............................................................................................. 37
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 v
2.35 SER B.2 Supplemental Interrogatory Comment 2 ......................................................... 40
2.35.1 Interrogatory Response ............................................................................................. 40
2.36 SER B.3 Supplemental Interrogatory Comment 3 ......................................................... 40
2.36.1 Interrogatory Response ............................................................................................. 41
2.37 SER B.4 Supplemental Interrogatory Comment 4 ......................................................... 42
2.37.1 Interrogatory Response ............................................................................................. 42
2.38 SER B.5 Supplemental Interrogatory Comment 5 ......................................................... 42
2.38.1 Interrogatory Response ............................................................................................. 42
2.39 SER B.6 Supplemental Interrogatory Comment 6 ......................................................... 45
2.39.1 Interrogatory Response ............................................................................................. 45
2.40 SER B.7 Supplemental Interrogatory Comment 7 ......................................................... 45
2.40.1 Interrogatory Response ............................................................................................. 46
2.41 SER B.8 Supplemental Interrogatory Comment 8 ......................................................... 46
2.41.1 Interrogatory Response ............................................................................................. 47
2.42 SER B.9 Supplemental Interrogatory Comment 9 ......................................................... 47
2.42.1 Interrogatory Response ............................................................................................. 47
2.43 SER B.11 Supplemental Interrogatory Comment 11 ..................................................... 47
2.43.1 Interrogatory Response ............................................................................................. 47
3.0 Attachments ......................................................................................................................... 48
4.0 Conclusion ........................................................................................................................... 48
5.0 References ............................................................................................................................ 48

Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 vi
FIGURES
Figure 1. Evapotranspiration (ET) cover profile showing materials, observation nodes, and root
distribution used in the HYDRUS-1D models. ............................................................. 7
Figure 2. Test pit texture results, textural triangle, and Rosetta properties for a loamy sand. ....... 8
Figure 3. HYDRUS results showing drainage through the bottom of the ET cover. ..................... 9
Figure 4. HYDRUS results (cumulative distribution function) showing drainage through the
bottom of the ET cover. .............................................................................................. 10
Figure 5. HYDRUS results showing impacts from tipping bucket error testing. ......................... 11
Figure 6. Daily water balance of one HYDRUS simulation showing cumulative fluxes (except
for storage) for a one-year wet period. ........................................................................ 17
Figure 7. Daily water contents for a 10-year period for one of the supplemental HYDRUS
simulations. ................................................................................................................. 18
Figure 8. Daily water contents for a 10-year period for one of the original 50 HYDRUS
simulations used with DU PA Model v1.4. ................................................................ 19
Figure 9. Types of uncertainties and their relative magnitudes in the near-surface disposal of
radioactive waste (from NRC (2011)). ....................................................................... 35
Figure 10. Sorted values of van Genuchten alpha and n, and Ksat, for the original 50 HYDRUS
runs (green line), and using the Hyd Props Calculator (blue line). ............................. 39
Figure 11. Comparison of old 100-yr precipitation record, generated with HELP, and new 1,000-
yr precipitation record, generated with SWAT. Precipitation is plotted from highest to
lowest daily precipitation. ........................................................................................... 43
Figure 12. Comparison of old 100-yr precipitation record, generated with HELP, and new 1,000-
yr precipitation record, generated with SWAT. Daily precipitation is shown for each
record. ......................................................................................................................... 44
Figure 13. Histograms showing the distribution of daily precipitation for the old 100-yr
precipitation record, generated with HELP, and the new 1,000-yr precipitation record,
generated with SWAT. ................................................................................................ 44

Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 vii
TABLES
Table 1. Summary statistics for the supplemental and DU PA Model v1.4 sets of HYDRUS
simulations. ................................................................................................................. 10

Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 viii
ACRONYMS AND ABBREVIATIONS
bgs below ground surface
CQA/QC Construction Quality Assurance/Quality Control
CWCB Colorado Water Conservation Board
DEQ (Utah) Department of Environmental Quality
DU depleted uranium
DWMRC Division of Waste Management and Radiation Control
ET evapotranspiration
GWPL groundwater protection limits
HELP Hydrologic Evaluation of Landfill Performance model
LLRW low-level radioactive waste
MOP member of the public
MPV maximum permissible velocity
NRC (United States) Nuclear Regulatory Commission
PA performance assessment
PAWG Performance Assessment Working Group
QA/QC quality assurance/quality control
SCS Soil Conservation Service
SER Safety Evaluation Report
SWAT Soil and Water Assessment Tool
TEDE total effective dose equivalent
UDEQ Utah Department of Environmental Quality

Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 1
Executive Summary
The Clive depleted uranium (DU) performance assessment (PA) evaluates the range of likely
impacts of disposal of DU in a new Federal Cell to be located in the southwest corner of the
licensed area. The DU PA is created as a systems-level model using the GoldSim probabilistic
modeling platform and is currently at version 1.4. The DU PA v1.4 model and supporting
documentation have been evaluated by the Utah Department of Environmental Quality (UDEQ)
and their contractor, SC&A Inc.
It is a truism when modeling complex systems such as radioactive waste disposal sites that no
model is perfect, but some models are useful. “Useful,” in this context, means that the model is a
reasonable representation of the system as currently understood and conceptualized, with the
acknowledgement that uncertainties will always remain. Important uncertainties are captured in
the probability distributions of the input parameters. Decisions can and should be made based on
the current model results. Standard PA practice calls for the model to be routinely reviewed and
updated as new information and data from monitoring programs or new relevant research
becomes available. This could include new information about site characteristics, the waste
itself, and the process models that have been abstracted into the systems-level probabilistic
model.
Updates to the model can lead to adaptive decision making if new model results indicate a need
to change a current decision. For the Clive site over the next few decades before final closure,
this could simply result, for example, in a change in cover design or placement of waste.
EnergySolutions is required to provide a surety fund that would accommodate changes under
such an adaptive management program. Alternatively, adaptive updates to the PA could also
demonstrate that initial constraints may safely be relaxed, such as that requiring DU waste be to
placed at an elevation below current native grade. What is described here might be called a PA
maintenance program, the details of which in terms of schedule would normally be captured by
License condition outlining the schedule and expectations for routine updates to the PA.
This concept is important in the context of DU PA v1.4 because a number of the outstanding
interrogatories rest all or in part on research that has emerged since completion of this version of
the model in 2015. This emerging research is certainly of interest to the model and appropriate to
incorporate in PA maintenance; but is shown in the attached responses to support rather than
change the fundamental conclusions of DU PA v1.4.
In this interrogatory response, Neptune and Company, Inc. (Neptune) addresses each issue raised
in the interrogatories and in the reviewers’ comments regarding prior responses. Items such as
those relating to deep time cannot be known with certainty. In fact, NRC (2000) guidance on
performance assessment methodology cautions that “…consideration given to the issue of
evaluating site conditions that may arise from changes in climate or the influences of human
behavior should be limited so as to avoid unnecessary speculation.” Deep time is of interest in
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 2
the context of the DU waste due to the ingrowth of progeny over geologic timeframes1. For an
aggrading site such as Clive, the DU will gradually be buried deeper over geologic time due to
geomorphic processes such as aeolian deposition. Deeper burial of DU over geologic time is
expected to more than counterbalance the ingrowth of progeny, as demonstrated in the Deep
Time model and discussed further in Section 2.3.1 below. Therefore, considering that the Deep
Time demonstrates that this basic premise is supported, it is reasonable for minor uncertainties
relating to the Deep Time model to be addressed via PA maintenance.
The Final Report for the Clive DU PA Model, Clive DU PA Model v1.4 (Neptune 2015) provides
the following summary of DU PA v1.4 results for the quantitative compliance period of 10,000
years. Additional work preparing interrogatory and comment responses after creation of version
1.4 (Neptune 2015) have not changed the principle analysis and reported conclusions.
Compliance with the performance objectives for the inadvertent intruder dose of 500
mrem in a year and for the MOP of 25 mrem in a year is clearly established for all three
types of potential future receptors. This indicates that for the disposal configuration
where DU wastes are placed below grade, doses are expected to remain well below
applicable dose thresholds…
Results are also available for the offsite (MOP) receptors. None of the 95th percentile
dose estimates for these receptors exceeds 1 mrem in a year, and all of the peak mean
dose estimates are at or below 0.1 mrem in a year.
Table ES-1. Peak TEDE: statistical summary
peak TEDE (mrem in a yr) within 10,000 yr
receptor mean median
(50th %ile) 95th %ile
ranch worker 6.2E-2 5.1E-2 1.5E-1
hunter 4.5E-3 3.8E-3 9.9E-3
OHV enthusiast 8.4E-3 7.5E-3 1.8E-2
Results are based on 10,000 realizations of the Model.
TEDE: Total effective dose equivalent
For those radionuclides for which GWPLs exist, as specified in the facility’s permit
(UWQB 2009), results are shown in Table ES-2. For all such radionuclides compliance
with the GWPLs is clearly demonstrated.

1 Note, however, that deep time represents a qualitative endpoint, in contrast with quantitative dose limits applied for
the 10,000-year compliance period. As noted in Utah DEQ (2019b), “…the qualitative evaluation could impact the
final decision based on its severity to human health and the environment.”
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 3
Table ES-2. Peak groundwater activity concentrations within 500 yr, compared to
GWPLs
peak activity concentration within 500 yr (pCi/L)
radionuclide GWPL1
(pCi/L) mean median
(50th %ile) 95th %ile
90Sr 42 0 0 0
99Tc 3790 26 4.3E-2 150
129I 21 1.7E-2 4.3E-11 1.1E-1
230Th 83 2.2E-28 0 0
232Th 92 1.4E-34 0 0
237Np 7 1.5E-19 0 3.7E-27
233U 26 5.6E-24 0 3.9E-28
234U 26 2.1E-23 0 2.2E-28
235U 27 1.6E-24 0 2.0E-29
236U 27 2.7E-24 0 3.3E-29
238U 26 1.5E-22 0 1.8E-27
1GWPLs are from UWQB (2009) Table 1A.
Results are based on 10,000 realizations of the Model.

DU PA v1.4 demonstrates compliance with the dose and groundwater protection requirements of
Utah regulations relating to DU disposal. The interrogatory and response process has added to
the record supporting these conclusions but has not caused the quantitative model demonstrating
compliance with UAC R313-25-9(5)(a) to require revision. Accordingly, DU PA v1.4 remains
the basis for demonstrating compliance of the disposal facility.
The Deep Time model has been revised to v1.5 to correct chronological errors identified in
Interrogatory CR R313-25-8(5)(A)-18/3. The revised model continues to demonstrate radon
fluxes that are comparable with those observed in Deep Time model v1.4.
Compliance with UAC R313-25-9(5)(a) is affirmed by DU PA v1.4, together with the supporting
documentation as supplemented by the interrogatory/response cycle.
As noted above, further revisions to the model can be made under a PA maintenance program as
new information, for example, from monitoring programs and the results of relevant research
become available.

Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 4
1.0 Introduction
Beginning in 2009, EnergySolutions contracted Neptune to create a probabilistic performance
assessment (PA) for the disposal of large quantities of depleted uranium (DU) at their Clive,
Utah low-level radioactive waste (LLRW) disposal facility.
The initial model was submitted as version 1.0 on June 1, 2011 (Neptune 2011) and was revised
to version 1.2 on June 5, 2014 (Neptune 2014). A Safety Evaluation Report (SER) based on
review of version 1.2 was issued by UDEQ in April 2015 (SC&A 2015b).
On November 25, 2015, EnergySolutions submitted Radioactive Material License UT2300249:
Safety Evaluation Report for Condition 35.B Performance Assessment; Response to Issues
Raised in the April 2015 Draft Safety Evaluation Report (EnergySolutions 2015). This document
included version 1.4 of the DU PA (Neptune 2015), prepared in response to open primary and
new interrogatories raised after development and DWMRC review of version 1.0; included in
Appendix C and Appendix B, respectively, of the SER.
On May 11, 2017, UDEQ provided Amended and New Interrogatories Related to Clive DU PA
Modeling Report Version 1.4 Dated November 2015 (Utah DEQ 2017). This document contains
clarifications to the original interrogatories from DU PA version 1.0 that remained open,
clarifications to the interrogatories newly raised with version 1.2 and new interrogatories
introduced with version 1.4 of the DU PA.
On April 2, 2018, EnergySolutions submitted Radioactive Material License UT2300249:
Responses to Amended and New Interrogatories Related to Clive DU PA Modeling Report
Version 1.4 Dated November 2015 (EnergySolutions 2018). As suggested by UDEQ, this
document included seven topical reports organized consistently with the themes expressed in the
interrogatory package (Utah DEQ 2017).
On July 25, 2019, UDEQ provided Depleted Uranium Performance Assessment (DU PA); Clive
Facility; Model Version 1.4 Amended Interrogatories; Radioactive Materials License #2300249
(Utah DEQ 2019a). This document contains amended interrogatories of open issues regarding
version 1.4 of the DU PA model, closes several interrogatories, and introduces two more new
interrogatories.
In an effort to ensure comprehensive resolution of the interrogatories from the Director’s initial
and subsequent reviews, this response document responds to each open interrogatory in Utah
DEQ (2019a) in the sequence presented therein. Each complete interrogatory number is cited
when it is first introduced and in headings, and abbreviated numbers are used during discussion.
For example, Interrogatory CR R313-25-7(2)-05/2 is introduced by its full number, and is then
abbreviated to Interrogatory 05/2, since the interrogatory numbering system employed by UDEQ
applies a unique number after the last hyphen in the sequence.
Note also that some interrogatories span more than one section of the Utah Administrative Code;
thus, a single interrogatory may cite multiple groups of otherwise isolated regulatory
requirements. For example, “Interrogatory CR R313-25-3 and R313-25-8-195/1” is a single
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 5
interrogatory drawing upon requirements from two parts of the Utah Administrative Code related
to the subject at hand. The shorthand for this interrogatory becomes “195/1.”
Full text of evaluation of each interrogatory through various rounds of review, with UDEQ
images, tables, and references, is available in Utah DEQ (2019a). Within this response
document, each open interrogatory is briefly quoted to summarize the issue. When passages of
interrogatory text are quoted, blue text in Arial font, size 10.5 pt, is used and is indented to
visually distinguish the interrogatory from the response. An example is shown below:
Sample format for quoting interrogatory text.
References within quoted text are not necessarily listed with this response document.
At times, in the interest of improved response clarity with this document, text or figures from the
discussion in Utah DEQ (2019a) are omitted. This is identified as follows:
[snip]
This response document does not comment on interrogatories and supplemental comments that
are considered “closed” in Utah DEQ (2019a).
2.0 UDEQ Interrogatory Responses
2.1 Interrogatory CR R313-25-7(2)-05/2: Radon Barriers
DEQ Discussion of NAC-0106_R0 – July 2019
Evapotranspiration Cover—As discussed in greater detail in Supplementary Interrogatory
Comment 2, EnergySolutions/Neptune predicted the infiltration through the cover with the
computer code HYDRUS utilizing two conceptual models. One conceptual model assumes that
the ET cover consists of five layers with distinct properties and functions (i.e., surface layer,
evaporative layer, frost protection layer, and two radon barrier layers). The parameter ranges
assigned to each of these layers, however, do not fall within the ranges of naturalized
parameters recommended in NUREG/CR-7028 (Benson et al. 2011). In the second
conceptualization, all five of the ET cover layers are combined into a single model layer. The
ranges of the hydraulic properties for this single layer model are within the ranges recommended
in NUREG/CR-7028. In our interrogatory discussions, these conceptual models are referred to
as the non-naturalized and naturalized models, respectively.
[snip]
Clay Liner—The methods that EnergySolutions/Neptune have used to evaluate this issue are
sufficient. The liners would be below ground in a humid and high-stress environment, in which
ideal conditions to maintain barrier integrity exist. The National Research Council study
conducted about a decade ago on waste containment systems is consistent with this position
and indicates that liners that are properly constructed continue to function well (National
Research Council 2007). Furthermore, over the long term, the cover properties would control the
flux through the system and the properties of the liner would become inconsequential. This is a
well-established principle in waste containment systems.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 6
Infiltration—Two concerns regarding infiltration were raised in this interrogatory, including the
appropriate evolutionary assumptions for cover properties and the selection of soil hydraulic
properties. Both of these aspects of infiltration are discussed below:
Appropriate Evolutionary Assumptions for Cover Properties. EnergySolutions/Neptune
have not adequately modeled site-specific soil parameters because they have not accounted
for any long-term changes in soil properties over time, including during the compliance period.
[snip]
Soil Hydrologic Properties. When selecting the input parameters for the unsaturated flow
model (HYDRUS), the relevant question to address is: “Does the distribution used for each
input parameter represent plausible and realistic values and distributions so that the
percolation rate predicted by the flow model represents the range of plausible percolation
rates into the waste with appropriate probabilities of occurrence?” [snip]
Erosion of Cover—EnergySolutions/Neptune indicate that calculations to evaluate the stability
of the design with respect to gully erosion for the ET cover of the Class A West cell were
provided in Appendix D to EnergySolutions (2015a), and that similar calculations for the Federal
Cell are presented in the response to Interrogatory 71/1.
The same issues are raised in Interrogatory CR R313-25-8(4)(a)-71/1: Biotic Processes in Gully
Formation, which remains open.
Effect of Biologicals on Radionuclide Transport—Since naturalized parameters incorporate
the effects of plants, animals, and insects on infiltration rates, the effect of biologicals are
considered under the naturalization issues discussed in Supplementary Interrogatory Comment 2
(see Section 4.1.2).
Frost Damage—Since naturalized parameters incorporate the effects of frost damage, the effect
of frost damage is considered under naturalization issues discussed in Supplementary
Interrogatory Comment 2 (see Section 4.1.2).
2.1.1 Interrogatory Response
Evapotranspiration Cover (ET Cover)
In order to facilitate the potential resolution of multiple infiltration-related interrogatories, a
series of eight meetings was conducted in the Fall of 2019 and early Spring of 2020 between
staff from Neptune and SC&A (including Dr. Craig Benson). There was also a meeting
conducted in Salt Lake City on January 7, 2020 with staff from Neptune, SC&A (including Dr.
Benson), and UDEQ. As a result of these meetings, Neptune and SC&A agreed to a
supplemental set of 100 HYDRUS runs that would likely alleviate concerns raised in most, if not
all, of the infiltration-related interrogatories.
The supplemental set of 100 HYDRUS simulations includes the following attributes:
1) A new 1,000-year precipitation record considered by SC&A to be more representative
of a 1,000-year period (e.g., includes extreme events), as opposed to repeating a 100-
year record 10 times (a method previously acceptable to UDEQ). [Refer to the response
to SER B.5 Supplemental Interrogatory Comment 5.]
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 7
2) Snowmelt hydrology, including snowmelt, sublimation, and storage modeled using the
HYDRUS snowmelt module. [Refer to the response to Interrogatory 10/3.]
3) The original five-layer model design, including the frost protection layer.
4) Hydraulic properties for layers 1, 2, 4, and 5 generated using the Hyd Props
Calculator.xls, developed by Dr. Benson, and using inputs for a fine-grained material.
5) Hydraulic properties for layer 3 (frost protection layer) generated using the Hyd Props
Calculator.xls, developed by Dr. Benson, and using inputs for a loamy sand.
6) Continue to apply a gravel correction to the surface layer (porosity × 0.85).
7) Increase the saturated hydraulic conductivity (Ksat) of the surface layer by one order of
magnitude (Ksat × 10).
8) Use a value of -2 for the pore interaction term for fine grained layers (layers 1, 2, 4, 5).
9) Use a residual water content of zero for all layers.
10) Additional test runs to assure that potential errors introduced by heated tipping buckets
are not causing drainage to be underestimated.
Hydraulic properties for layers 1, 2, 4, and 5 of the HYDRUS profile that represents the ET
cover (Figure 1) were generated using the Hyd Props Calculator.xls, developed by Dr. Benson,
and using inputs for a fine-grained material (see Tables 1 and 2 on page E-4 of the SER, Vol. 2
(SC&A 2015a)). A correlation coefficient of 0.462 was used for lnK-lna (Benson and Gurdal
(2013), Figure 3).

Figure 1. Evapotranspiration (ET) cover profile showing materials, observation nodes, and
root distribution used in the HYDRUS-1D models.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 8
Hydraulic properties for layer 3 (frost protection layer) were also generated using the Hyd Props
Calculator.xls and inputs for a loamy sand. In 2019, the Cover Test Cell at Clive was
deconstructed in order to conduct hydraulic tests to evaluate cover performance and
naturalization effects. The textural analysis shown in Figure 2 confirms the model’s
characterization of the frost protection layer as a loamy sand (assuming that the fraction of silt
and clay is 50/50). Frost protection layer hydraulic properties were then taken for a loamy sand
from the Rosetta soils database (https://www.ars.usda.gov/pacific-west-area/riverside-
ca/agricultural-water-efficiency-and-salinity-research-unit/docs/model/rosetta-model/) and used
as input in the Hyd Props Calculator.xls after adjusting the residual water content to zero per
direction from Dr. Benson and reducing porosity (thetaS) by 39.3 to account for the gravel
fraction.

Figure 2. Test pit texture results, textural triangle, and Rosetta properties for a loamy
sand.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 9
100 sets of five layers of hydraulic parameters generated from the Hyd Props Calculator.xls
were then input into HYDRUS and run for 1,000 years using a new 1,000-year precipitation
dataset (see response to SER B.5 Supplemental Interrogatory Comment 5) and using the
HYDRUS snowmelt hydrology module.
Results from the supplemental HYDRUS simulations are shown in Figure 3. Drainage out of the
bottom of the profile is shown in mm/yr, sorted from lowest to highest (blue line). Also shown in
this figure is sorted drainage from the original HYDRUS simulations that were abstracted and
used as input to the DU PA Model v1.4 (green line).
The original simulations incorporated into DU PA Model v1.4 (green line) show higher drainage
than the supplemental simulations (blue line). Figure 4 shows the same results as Figure 3,
except that the results are shown as a cumulative distribution function. Accordingly, the original
simulations are confirmed to provide a reasonably conservative model of the system compared
with the supplemental simulations. If the supplemental simulations were adopted as a new
baseline for modeling, in fact, embankment performance would more easily meet the
groundwater protection compliance criteria.

Figure 3. HYDRUS results showing drainage through the bottom of the ET cover.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 10

Figure 4. HYDRUS results (cumulative distribution function) showing drainage through
the bottom of the ET cover.
The reduction in system drainage is further observed in summary statistics from the sets of
simulations as shown in Table 1.
Table 1. Summary statistics for the supplemental and DU PA Model v1.4 sets of HYDRUS
simulations.
Drainage
(mm/yr)
Supplemental runs
(1,000yr Precip,
Snow, FP)
Original simulations
used in DU PA Model
v1.4
Min 0.0049 0.007
Avg 0.0073 0.024
Max 0.0121 0.183

The original HYDRUS simulations were used as input to the DU PA Model v1.4, producing
results well below applicable regulatory dose and groundwater protection limits. Because the
supplemental HYDRUS runs have lower drainage than the original simulations, they support the
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 11
degree of conservatism projected from DU PA Model v1.4. Accordingly, DU PA Model v1.4
remains a reasonable basis for demonstrating compliance of the disposal facility.
At the informal request of SC&A, ten additional test runs were conducted in HYDRUS. The
purpose of these test runs was to evaluate the significance associated with any potential errors
introduced by the Clive weather station. This audited and certified weather station uses a heated
tipping bucket to measure precipitation. SC&A questioned whether the heated tipping bucket
might cause precipitation, and therefore drainage, to be underestimated. For these ten
simulations, the new 1,000-year precipitation dataset was multiplied by 1.3 on every day in
which daily average temperature was less than zero deg C. This resulted in a total precipitation
increase of about 7%. The resulting increases in drainage ranged from virtually nothing to 3.4%,
with an average of about 1%. Results are shown in Figure 5 for these ten simulations, compared
to the ten simulations without this adjustment. Clearly, any impact on drainage from a heated
tipping bucket is negligible.

Figure 5. HYDRUS results showing impacts from tipping bucket error testing.
Clay Liner
As noted in Utah DEQ (2019a), this aspect of the interrogatory is closed.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 12
Infiltration
See response in Section 2.1.1 for ET Covers.
Erosion of Cover
Refer to Section 2.8.1 for response to Interrogatory 71/1.
Effect of Biologicals on Radionuclide Transport
See response in Section 2.1.1 for ET Covers.
Frost Damage
See response in Section 2.1.1 for ET Covers.
2.2 Interrogatory CR R313-22-32(2)-10/3: Effect of Biologicals on
Radionuclide Transport
DEQ Discussion of NAC-0106_R0 – July 2019
In their response in NAC-0106_R0, ET Cover Design Responses for the Clive DU PA Model
(Neptune 2018f), EnergySolutions/Neptune describe the literature reviews and field data
collected by SWCA (2013). EnergySolutions/Neptune contend that increased infiltration due to
biotic activity would be minute, based on the same conclusion reached by SWCA (2013). It was
not apparent, however, that SWCA (2013) based this conclusion on any quantitative data (e.g.,
lysimeters) from the immediate vicinity of the Clive site. However, SWCA (2013) did raise the
issue of high-precipitation events and snowmelt leading to the greatest infiltration rates.
Since naturalized parameters incorporate the effects of biologicals, their effects are considered
under naturalization issues discussed in Supplementary Interrogatory Comment 2 (see Section
4.1.2).
The issues related to the variability of infiltration rates during high-precipitation events and
snowmelt are discussed in Round 3 Interrogatory CR R313-25-8(4)(A)-28/3: Bioturbation Effects
and Consequences (ES 2014c) and in Supplemental Interrogatory Comment 3 in Appendix B of
2015 SER. Interrogatory 28/3 and Supplemental Interrogatory Comments 2 and 3 remain open.
See also Section 4.1 in this document.
2.2.1 Interrogatory Response
See response in Section 2.1.1 for ET Covers. Note that the Ksat of the surface layer was increased
by a factor of 10 for all simulations to account, in part, for the effects of biologicals, as
recommended by Dr. Benson.
Refer to Sections 2.6.1, 2.35.1, and 2.36.1 for responses to Interrogatory 28/3 and Supplemental
Interrogatory Comments 2 and 3, respectively.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 13
2.3 Interrogatory CR R313-25-8(5)(A)-18/3: Sediment Accumulation
DEQ Discussion of NAC-0105_R0 – July 2019
Due to the evolving nature of the deep time analysis, Interrogatory CR R313-25-8(5)(A)-18/3:
Sediment Accumulation has also evolved somewhat since it was first created in Round 1. In its
current form, it consists of two concerns: aeolian deposition and Intermediate Lake
sedimentation rate. Both of these processes result in material overlaying the DU once the
embankment has been “washed away” by the first returning lake.
Interrogatory CR R313-25-8(5)(A)-18/3: Part One—Aeolian Deposition
Based on samples collected on site, EnergySolutions/Neptune have developed an aeolian
deposition distribution with a mean, standard distribution, and standard error. To account for
spatial and temporal effects, EnergySolutions/Neptune propose to substitute the standard error
for the standard deviation when the aeolian deposition distribution is entered into the GoldSim
v1.4 model. While DEQ/SC&A agree with the concept of adjusting for spatial and temporal
effects, we believe that for the purposes of nuclear facility licensing, a more rigorous and
defendable approach to its implementation is required.
As we demonstrated in our previous response, using the standard error, rather than the standard
deviation, is mathematically equivalent to dividing the embankment into 11 subareas. As
EnergySolutions/Neptune state in their NAC-0105_R0 responses to concerns 1 through 3
(Neptune 2018e), we agree that the site was not divided into 11 subareas. Thus, their use of the
standard error seems to contradict their responses to concerns 1 to 3. EnergySolutions/Neptune
provide no basis for using the standard error in their response to concern 4, other than to state
that the “technically established statistic for upscaling is the mean thickness of the field
measurement and the standard error of the mean to represent the variance of the averaged
measurement data” (Neptune 2018e, p. 5).
We have reviewed the three references provided earlier in the response by
EnergySolutions/Neptune (i.e., Blöschl and Sivapalan (1995); Neuman and Wierenga (2003);
Zhang et al. (2004)), and, while the references describe the need for upscaling, they do not
“technically establish” the use of the standard error in this manner. Zhang et al. (2004) do state
that “scaling should only be applied over a limited range of scales and in specific situations,”
while Neuman and Wierenga (2003) state that “One approach has been to postulate more-or-
less ad hoc rules for upscaling based on numerically determined criteria of equivalence”
(emphasis added).
DEQ remains concerned that (1) the proposed embankment is not a specific situation that allows
for upscaling and (2) the approach taken by EnergySolutions/Neptune is “an ad hoc rule” rather
than a “technically established” approach. Although it is recognized that the impacts of this
interrogatory are limited and understood, it is important that all nuclear facility licensing
assumptions that result in less conservative results be well understood and documented, rather
than based on “ad hoc rules.” Therefore, until EnergySolutions/Neptune provide either a
technical basis that is sufficiently detailed and defendable for a nuclear licensing application for
using the standard error or revert to using the standard deviation, this portion of the interrogatory
remains open.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 14
Interrogatory CR R313-25-8(5)(A)-18/3: Part Two—Intermediate Lake Sedimentation Rate
EnergySolutions/Neptune have developed an Intermediate Lake sediment deposition model for
input to the GoldSim v1.4 model (Neptune 2015l, Section 7.4). It is DEQ/SC&A’s opinion that the
EnergySolutions/Neptune-developed Intermediate Lake sediment deposition model overpredicts
the amount of sediment that would be present once the lake recedes and thus underpredicts the
radon flux on the surface of the sediment.
[snip]
In conclusion, this interrogatory remains open, and the v1.4 Deep Time GoldSim model is
currently unacceptable. Three actions are needed to close this interrogatory: (1) a more realistic
Intermediate Lake total sedimentation or sedimentation rate needs to be developed,
(2) Appendix 13 of the DU PA needs to be revised to correctly describe the deep time model,
and (3) the logic bugs need to be removed from the GoldSim model.
2.3.1 Interrogatory Response
The GoldSim Deep Time model has been updated to v1.5 in response to the issues identified in
this interrogatory. In addition, the supporting white paper, Deep Time Assessment for the Clive
DU PA, has been updated to revision 5. The GoldSim model player is attached electronically; the
white paper is provided as Attachment 1.
Select results of the revised model are provided below. Attachment 2 updates the Deep Time
section of the Final Report for the Clive DU PA Model (Neptune 2015).
The overall sedimentation rate in the model matches historical data that suggests 13-26m per
100ky cycle. The first cycle has a smaller amount because of the impact of climate change as
described and interpreted from IPCC reports at the time the model was built in 2014. The IPCC
reports at that time suggested that the next ice age is likely not to occur, in which case the model
did not allow any intermediate lakes to develop at the elevation of Clive in the first 50ky of the
first cycle. To accommodate this difference in the first cycle, the model allows pure aeolian
deposition until the first lake arrives. The aeolian deposition rate is slow, but the long period of
time before the first lake necessitated its inclusion to provide a better model and understanding
of the evolution of the area in the broad vicinity of the Clive facility.
As noted in earlier interrogatory responses, a field study was performed to gather data on the
depth of aeolian deposition since the last ice age. The field data came from previous large-scale
excavations and from some new holes that dug to the bottom of the silt layers that were assumed
to represent aeolian deposition since the last ice age. In total, 11 measurements were made at
different locations in the vicinity of the Clive facility. The average of these data is 72.7 cm of
postglacial aeolian deposition. The data fit a normal distribution, and the appropriate distribution
to use in the model is a distribution that represents the average effect over the large area that
encompasses the Clive facility. The underlying conceptual model is that aeolian deposition will
occur in the area around the Clive facility so that the portion of the cell that is at the time of
closure about 45 ft above grade at its highest point, will be some amount less when the first
intermediate lake returns to the Clive elevation. Consequently, the aeolian deposition rate that is
of interest for the model is the average rate of deposition across an area that is much larger than
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 15
the Clive facility, albeit an undefined large area (it does not need to be defined – the point is
simply that aeolian deposition will occur in the general area so that the elevation of the waste cell
above grade gets smaller with time). The data are assumed to represent independent and
identically distributed (iid) locations in that area, in which case the technically correct
distribution of the spatially scaled distribution is the distribution of the average. This is not an ad
hoc choice. This is a technically correct choice.
There are two reasons why the approach used to scaling the aeolian deposition rate leads to a
conservative estimate of variance. The first is that 21 data points from a previous study were
ignored. These data points had an almost identical mean to the newer 11 data points, but the
underlying data did not have the same sensitivity. If the older data had been included then the
spatially scaled distribution would be narrower.
The second is that temporal scaling has been ignored, even though it would be reasonable to
account for this. Suppose the time since the last glacial is 13ky, and that the time from the last
glacial to the first lake is 65ky, and the deposition rate is roughly between 60 and 85cm per 13ky
(reasonable lower and upper bounds based on 11 data points from a normal(72.7, 5) distribution).
If a new random number is pulled from this distribution every 13ky, then the sum of the 5 pulls
would have a normal(363.5, 5√5) distribution. However, when pulling a single random number
from the normal(72.7, 5) distribution, and applying that 5 times, the simulated distribution is
normal(363.5, 25). The lack of temporal scaling has led to a considerably wider distribution than
might otherwise be used.
In general, both spatial and temporal scaling should be considered because the basic goal of
scaling is to re-scale data to the scale of the model, which has both spatial and temporal aspects.
When the scale of data is less than the scale of the model, then the term upscaling is usually
applied, and upscaling is a form of averaging. This has nothing directly to do with the
embankment. This is simply a reflection of what will happen in the future based on the available
data (with the exceptions that the scaled distribution should be narrower if the older data had
been used and if temporal scaling had been applied). The purpose is to avoid creating impossible
hypothetical futures, which becomes an even bigger problem if all variables in the model are not
scaled – this would create a situation where the model output would include many impossible
combinations of inputs. Scaling is absolutely necessary to properly accommodate uncertainty in
the model. It just needs to be done carefully so that the averaging that is performed does not have
the effect of diluting risk.
Note also that Neptune has now written a research paper for DOE on scaling, which has been
shared with SC&A. This paper provides examples that show why scaling is appropriate, but also
shows an example where scaling with insufficient attention to potential dilution of risk is not
appropriate. In the case of aeolian deposition, averaging is appropriate so that a reasonable range
of total aeolian deposition is realized in the model.
One particular example that Neptune has used to try to explain why scaling is important
(although Neptune understands that it must be applied carefully) uses a simple case of packets of
regular M&Ms. Suppose regular packets contain 50 M&Ms, and a large bag contains 1,000
M&Ms. The concept is that the number of green M&Ms in 20 packets should be roughly the
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 16
same as the number in a big bag. 20 packets make up a large bag. Based on experimentation the
number of green M&Ms in a packet can range from 0 to 20. If we do not scale to the big bag,
then we might predict that a big bag contains anywhere from 0 to 400 green M&Ms. But
sampling big bags shows that the number of green M&Ms is between 140 and 200 (expected
value is 160). This is because a big bag effectively represents 20 random packets; it does not
represent 20 packets that are all the same. This is exactly the same principle when applying 11
measurements of aeolian deposition to a much larger area than the Clive facility, which is what’s
going on in this DU PA model. There are 11 samples that range from 55 – 110 cm (in 13ky, say),
then we could have a realization that indicates that there are 110 cm of deposition across the
entire area, despite the data to the contrary (or even more if a normal (or other) distribution is fit
to the data, considering the standard deviation of the data is 16.6). What we really have instead is
an average of 72.7 and a reasonable range of that average from about 58-88 cm (using three
standard errors).
If there were spatial complications in this aspect of the model, then this would not be the right
way to upscale. But, in this case there are not.
2.4 Interrogatory CR R317-6-2.1-20/2: Groundwater Concentrations
DEQ Discussion of NAC-0106_R0 – July 2019
EnergySolutions/Neptune refer to Interrogatory 71/1 for their complete response to concerns
related to the formation of gullies on the cover leading to infiltration rates. This interrogatory
remains open.
[snip]
DEQ/SC&A disagrees that EnergySolutions/Neptune cannot build in layering into the naturalized
model and remain consistent with the recommendations in NUREG/CR-7028. For example, a
more permeable surface crust (upper 300 mm) could be input that would ensure that the surface
boundary (i.e., infiltration) is represented realistically. That is not outside the bounds of the
recommendations in NUREG/CR-7028. Lack of detailed (e.g., daily) water balance predictions
for any part of the simulation period has confounded an assessment of the realism of this
interface. For example, without careful attention to the surface layer properties, runoff can be
overpredicted and infiltration underpredicted. Consequently, SC&A is not able to evaluate the
efficacy of the surface boundary with information provided and cannot have confidence that
infiltration at the surface is represented reasonably.
EnergySolutions/Neptune need to provide a daily water balance record for several periods in the
time series that would allow an independent assessment of the realism of their predictions.
2.4.1 Interrogatory Response
Concerns related to the formation of gullies are addressed in the response to Interrogatory 71/1
(Section 2.8.1).
See response in Section 2.1.1 for ET Covers. As suggested by SC&A, the Ksat of the surface
layer was increased by a factor of 10 for all simulations to account, in part, for the effects of
biologicals, as recommended by Dr. Benson.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 17
Several daily water balance figures have been prepared to assist with visualizing the hydrology
of the ET cover. First, a one-year period is shown that includes a wet period of multiple high
precipitation events. Figure 6 shows a one-year period of cumulative precipitation, ET,
infiltration, and soil water storage (not cumulative) on the left y-axis, and cumulative drainage on
the right y-axis. Note that runoff for this period is zero, which is likely the result of the increased
Ksat for the surface layer.

Figure 6. Daily water balance of one HYDRUS simulation showing cumulative fluxes
(except for storage) for a one-year wet period.
Note that ET closely tracks infiltration into the profile, and both of these track behind
precipitation. Soil water storage increases, but returns to pre-precipitation levels quickly within
several months. Drainage continues at a steady state, unaffected by the high precipitation events
during this period.
A ten-year period of water contents showing the last ten years of the 1,000-year simulation
period for one of the supplemental HYDRUS simulations is shown in Figure 7. Note the increase
in water contents following each precipitation event for the surface and evaporative zone layers.
The overall low water contents are notable. These are a result of using the Benson-recommended
inputs into the Hyd Props Calculator.xls. Water contents for this same ten-year period are shown
for one simulation from the original 50 HYDRUS runs that are used with the DU PA Model v1.4
in Figure 8. Note the higher water contents in the original simulations.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 18
Water contents in the surface layer appear to be more dynamic in the original simulation (Figure
8) than in the new simulation (Figure 7). Both HYDRUS models have the observation nodes at
the same depths. Therefore, the higher Ksat of the new simulation likely allows infiltration (but
not to the depth of the observation node), followed by rapid ET, while the hydraulic properties of
the original simulation allow deeper infiltration to the observation node.

Figure 7. Daily water contents for a 10-year period for one of the supplemental HYDRUS
simulations.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 19

Figure 8. Daily water contents for a 10-year period for one of the original 50 HYDRUS
simulations used with DU PA Model v1.4.
2.5 Interrogatory CR R313-25-8(4)(d)-21/2: Infiltration Rates
DEQ Discussion of NAC-0106_R0 – July 2019
Radon Barrier Ksat Distribution
In response to a DEQ request for clarification with respect to what probability distribution of Ksat
was used in v1.4, EnergySolutions/Neptune provide a detailed explanation. However, exactly
how this distribution was incorporated into HYDRUS is still not clear. In response to Interrogatory
20/2: Groundwater Concentrations, EnergySolutions/Neptune indicate that all layers were set to
homogeneous, naturalized conditions, whereas the response to this interrogatory indicates that
the radon barriers were assigned parameter distributions distinct from the other layers. This
issue is discussed in greater detail in Section 4.1.2.
Naturalized Cover
[snip]
The primary issue with EnergySolutions/Neptune’s position regarding naturalization is that they
have not characterized the soil properties from near-surface layers that have undergone
weathering and pedogenesis. They have excavated test pits and examined profiles, but for their
non-naturalized HYDRUS modeling rely heavily on (1) old data from Bingham Environmental for
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 20
the radon barrier and (2) Rosetta properties for overlying layers, neither of which is relevant to
designing and evaluating an engineered cover.
Homogeneous Cover
[snip]
As described in Interrogatory 20/2: Groundwater Concentrations, it is DEQ/SC&A’s position that
layering can be included and remain consistent with the recommendations in NUREG/CR-7028.
Correlation and Range of Hydraulic Properties
EnergySolutions/Neptune contend that the saturated hydraulic conductivity (Ksat) and van
Genuchten’s α parameter used as input to the model are uncorrelated. That is an unlikely
occurrence that is inconsistent with other data sets in the literature and suggests an
inconsistency in the data set. However, independent simulations conducted by DEQ/SC&A have
shown that percolation rates predicted with the saturated hydraulic conductivity (Ksat) and α
parameter uncorrelated are higher than those predicted with correlation. Thus, ignoring
correlation between the Ksat and α parameter is acceptable.
With respect to the range of hydraulic properties, EnergySolutions/Neptune provide an
explanation for the selected variance and estimates of uncertainty and its origins in the Rosetta
database. Per the discussion provided in Supplemental Interrogatory Comment 2 (Section 4.1.2),
for their naturalized single layer HYDRUS model, EnergySolutions/Neptune are using the
hydraulic property recommendations and cover material naturalization presented in NUREG/CR-
7028. However, there are still related concerns remaining, as discussed in Section 4.1.2.
Unsaturated Flow Model Output
DEQ/SC&A requested plots of flow model water balance components on a daily basis. These
components are precipitation, runoff, infiltration, evaporation, transpiration, storage, and
percolation (deep drainage). In response to this request, EnergySolutions/Neptune discuss how
steady-state annual averages of net infiltration and water content from the HYDRUS simulations
are the model results used to develop statistical distributions of these parameters for inputs to
the GoldSim model for the Clive DU PA. However, as is described in greater detail in Section
4.1.5, annual averages can be misleading because they do not capture larger pulses of
percolation rate that can occur during wet periods, such as snow melt events. Assessing the
significance of these short-duration higher percolation events is necessary to adequately
evaluate the reliability of the DU PA.
Regression Model
A concern raised before, and yet unresolved, is whether good agreement would exist between
percolation rates predicted with the regression model and an independent set of predictions from
HYDRUS using the same underlying inputs (e.g., a blind forward comparison). That type of
evaluation is needed to confirm the validity of the regression model. DEQ/SC&A also requested
that, at a minimum, EnergySolutions/Neptune should conduct an independent set of simulations
where percolation is predicted with HYDRUS and then compared with predictions obtained with
the regression model. This is the only fair means to evaluate the efficacy of the regression
model. These predictions should be conducted with the typical standard deviations to get a
realistic representation of the tails of the distribution of percolation.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 21
In response to these comments, EnergySolutions/Neptune provide a detailed explanation for why
these model verification tests should not be performed. To reasonably demonstrate that the
performance objectives will be met, DEQ/SC&A maintains the position that these verification
tests need to be performed to gain adequate confidence that the linear model provides a
reasonable abstraction of the HYDRUS output. Furthermore, as discussed in greater detail in
Section 4.1.5, the linear model derived from the non-naturalized HYDRUS output and used in
GoldSim for percolation rates has been shown to provide a reasonable prediction of percolation
rates, except at the extreme case where the linear model underpredicts the percolation rate by
approximately a factor of 2. Given the limited number of realizations used in the non-naturalized
HYDRUS model (50), the tails of the distribution likely are underrepresented, and a much greater
deviation may exist for higher percolation rates. The significance of underprediction in the tails
needs greater documentation, and the impact of this underprediction needs to be quantified.
The results of comparable analysis with the naturalized HYDRUS model have not been
presented.
2.5.1 Interrogatory Response
Radon Barrier Ks Distribution
See response in Section 2.1.1 for ET Covers.
Naturalized Cover
See response in Section 2.1.1 for ET Covers.
Homogeneous Cover
See response in Section 2.1.1 for ET Covers.
Correlation and Range of Hydraulic Properties
See response in Section 2.1.1 for ET Covers. Note that a correlation is included for lnK-lna using
data from Benson and Gurdal (2013)).
Unsaturated Flow Model Output
See response in Section 2.1.1 for ET Covers, and Section 2.4.1 for daily water balance plots.
Regression Model
See response in Section 2.1.1 for ET Covers. The issue of forward prediction is now moot given
that the range of drainage calculated with the original HYDRUS simulations, and abstracted into
the DU PA Model v1.4, are considerably higher than the new set of 100 HYDRUS runs that
were run to accommodate the concerns raised in the infiltration-related interrogatories.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 22
2.6 Interrogatory CR R313-25-8(4)(a)-28/3: Bioturbation Effects and
Consequences
DEQ Discussion of NAC-0106_R0 – July 2019
[snip]
The concerns raised under Interrogatory 28/3: Bioturbation Effects and Consequences are the
same as those raised in Interrogatory 20/2: Groundwater Concentrations, Supplemental
Interrogatory Comment 2, and Interrogatory 71/1: Biotic Processes in Gully Formation, which all
remain open.
2.6.1 Interrogatory Response
Refer to Sections 2.4.1, 2.35.1, and 2.8.1 for responses to Interrogatory 20/2, Supplemental
Interrogatory Comment 2, and Interrogatory 71/1, respectively.
2.7 Interrogatory CR R313-25-7(3)-60/2: Modeled Radon Barriers
DEQ Discussion of NAC-0106_R0 – July 2019
For their response to this interrogatory, EnergySolutions/Neptune referred to other
interrogatories that include the same concerns.
This interrogatory raises the same issues as those included in Interrogatories 5/2: Radon Barrier
and 21/2: Infiltration Rates and Supplemental Interrogatory Comment 2, which all remain open.
2.7.1 Interrogatory Response
Refer to Sections 2.1.1, 2.5.1, and 2.35.1 for responses to Interrogatories 5/2, 21/2, and
Supplemental Interrogatory Comment 2, respectively.
2.8 Interrogatory CR R313-25-8(4)(a)-71/1: Biotic Processes in Gully
Formation
DEQ Discussion of NAC-0108_R0 – July 2019
[snip]
DEQ expressed concern that gullies will form and enhance radon diffusion, deep infiltration, and
contaminant transport. EnergySolutions/Neptune used the maximum permissible velocity
methodology developed by NRC in NUREG-1623 (NRC 2002), to determine whether or not
gullies would form in the most severe storm. According to EnergySolutions/Neptune: “A value of
5.0 ft/s was chosen as the maximum permissible velocity (MPV) based on the characteristics of
the channel. This is the value listed for gravel in Table CH13-T103 of Colorado Water
Conservation Board (CWCB 2006) and in Table 4.7 of Nelson et al. (1986)” (Neptune 2018g, p.
11). Based on the NRC methodology, this value must be adjusted downward depending on the
depth of flow across the embankment surfaces. In the case of the Federal Cell, the adjustment
factor is 0.5, which would set the maximum permissible velocity at 2.5 feet per second (ft/s). The
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 23
calculated flow velocities for the ET cover were 1.60 ft/s for the top slope and 2.03 ft/s for the
side slope. Thus, based on an adjusted MPV of 2.5 ft/s, gullies would not form.
However, the selection of 5.0 ft/s as the MPV is not consistent with NRC recommendations. The
NRC states in Appendix A of NUREG-1623 that “Additionally, based on examination of data from
the Soil Conservation Service (SCS, 1984), the staff recommends that the maximum permissible
velocity for grasses covers and channels be limited to about 2½ to 3 ft per second. This limit is
necessary because no credit may be taken for active maintenance in designing for long-term
stability.” Additionally, SCS 1984 states on page 7-7 that “A velocity of 3.0 ft per second should
be the maximum, where because of shade, soils, or climate, only a sparse cover can be
established or maintained.” These sources call into question the assumption used in v1.4 of the
DU PA of 5 ft/s. Using the flow depth adjustment factor of 0.5 indicates that the adjusted
maximum permissible flow velocity should be 1.5 ft/sec. With the maximum permissible velocities
recommended by NRC and SCS, and the NAC-0108_R0 calculated flow velocities, gullies can
be expected to form.
Because the permissible velocity assumptions are less conservative than NRC
recommendations, this interrogatory remains open.
2.8.1 Interrogatory Response
As discussed below, the design MPV is reasonable and is supported by guidance documents.
Nonetheless, as discussed in Section 3.27.1, EnergySolutions has chosen to rock armor the side
slopes using a design comparable to that on the taller Class A West embankment. Therefore,
existing erosion analyses of the longer slope lengths for that embankment provide a bounding
evaluation for performance of the Federal Cell.
This interrogatory relates to the Maximum Permissible Velocity (MPV) to be applied within the
context of NUREG-1623 calculations. The value of 5.0 ft/s reasonably reflects the side slope
cover design of 50% gravel as well as NRC recommendations in NUREG/CR-4620, Table 4.7,
for “Fine gravel.” Furthermore, Table 4.8 sets MPV at 4.00 to 5.00 for materials described as
“Stiff clay soil, gravel soil” (Nelson et al. 1986).
The interrogatory cites a section of NUREG-1623 that is stated to be applicable to grassed covers
as justification for setting MPV at 3.0. This selection would include no credit for the design with
50% gravel. Note also that an MPV of 3.0 as suggested by the interrogatory is comparable to that
allowed in Table 4.7 of NUREG/CR-4620 for un-armored, un-vegetated silty loam.
The interrogatory cites Chapter 7 of SCS (1984) as further justification for an MPV of 3.0;
however, Chapter 7, “Grassed Waterways,” is not applicable to the Federal Cell cover
evaluation2. For the 50% gravel side slope cover layer, Chapter 14, “Water Management
(Drainage)” (NRCS 2001) is more directly applicable. Table 14-3 (NRCS 2001) provides
permissible bare earth velocities for consideration in designing ditches. These velocities are

2 Note that Chapter 7 of the NRCS (formerly SCS) Engineering Field Handbook has been updated since the cited
version. The current version of Chapter 7 is dated December 2007 (NRCS 2007) and no longer includes a
permissible velocity approach for the engineering design of grassed waterways.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 24
comparable to those in Table 4.7 of NUREG/CR-4620, specifically for the soil texture of “stiff
clay, fine gravel, graded loam to gravel” with a maximum velocity of 5.0 ft/s.
2.9 Interrogatory CR R313-25-7(2) and 7(6)-81/2: Comparison of
Disposal Cell Designs
DEQ Discussion of NAC-0101_R0 – July 2019 2018a
At the time this interrogatory was generated, EnergySolutions was considering use of ET covers
on both the Federal Cell and the Class A West cell. The focus of this interrogatory was to
compare the design and expected functionality of the two ET covers. Since the time the
interrogatory was developed, EnergySolutions has dropped the ET cover from consideration and
now plans to use a rock-armor cover on the Class A West cell. Given this change in cover
selection, the focus of this interrogatory shifts from a comparison of the two ET covers to an
explanation of why a rock armor was determined to be the preferred choice for the Class A West
cell while the ET cover was selected for Federal Cell.
We also note that the interrogatory raised a question regarding the cover liner: No information
has been provided on the function of the 1-foot liner protective cover shown in Drawing No.
14002-L1A(0). What material is used? Was it included in performance assessment analyses?
EnergySolutions/Neptune have not provided an answer to these questions.
2.9.1 Interrogatory Response
EnergySolutions reverted to a rock armor design for the Class A West cell because portions of
that cell were nearing a compliance point for the open cell time limit, and the review process for
the ET cover design did not appear to be on track for successful conclusion with time to enable
cover construction to be completed before that compliance point was reached. EnergySolutions
and Neptune remain convinced that an ET cover design offers better performance than the rock
armor design in terms of reduced infiltration.
The interrogatory goes on to question the “cover liner”; which is presumed to be a reference to
embankment element “liner protective cover”. The liner protective cover is a one-foot layer of
clean soil placed on top of the clay liner before waste placement begins. It is described in the
facility’s Construction Quality Assurance/Quality Control (CQA/QC) Manual, specification 55,
as consisting of compacted native soils, free of debris (EnergySolutions 2020). Prior licensing
submittals involved in the development of the CQA/QC Manual discuss the purpose of this layer
as being to provide a cushion between bulk waste placement and the top of the specification clay
liner. The liner protective cover also serves to protect the clay liner from drying between the time
of construction and the start of waste placement.
2.10 Interrogatory CR R313-25-7(1–2)-90/2: Calibration of Infiltration
Rates
DEQ Discussion of NAC-0106_R0 – July 2019
For their response to this interrogatory, EnergySolutions/Neptune referred to other
interrogatories that include the same concerns.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 25
This interrogatory raises the same issues as those included in Interrogatory 5/2: Radon Barrier
and Supplemental Interrogatory Comment 2, which all remain open.
2.10.1 Interrogatory Response
Refer to Sections 2.1.1 and 2.35.1 for responses to Interrogatory 5/2 and Supplemental
Interrogatory Comment 2, respectively.
2.11 Interrogatory CR R313-25-7(1)-100/2: Groundwater Recharge
from Precipitation
DEQ Discussion of NAC-0105_R0 – July 2019
EnergySolutions/Neptune did not explicitly respond to this interrogatory. The same topics in this
interrogatory are discussed in Supplemental Interrogatory Comment 2 pertaining to naturalized
parameters in HYDRUS, which remains open (see Section 4.1.2).
2.11.1 Interrogatory Response
No response was provided to this interrogatory in NAC-0105_R0 because this interrogatory was
not included in Utah DEQ (2017). Refer to Section 2.35.1 for response to Supplemental
Interrogatory Comment 2.
2.12 Interrogatory CR R313-25-8(4)(a)-108/2: Biointrusion
DEQ Discussion of NAC-0105_R0 – July 2019
EnergySolutions/Neptune did not explicitly respond to this interrogatory. The same topics in this
interrogatory are discussed in Supplemental Interrogatory Comment 2 pertaining to naturalized
parameters in HYDRUS, which remains open (see Section 4.1.2).
2.12.1 Interrogatory Response
No response was provided to this interrogatory in NAC-0105_R0 because this interrogatory was
not included in Utah DEQ (2017). Refer to Section 2.35.1 for response to Supplemental
Interrogatory Comment 2.
2.13 Interrogatory CR R313-25-8(4)(a)-112/2: Hydraulic Conductivity
DEQ Discussion of NAC-0105_R0 – July 2019
EnergySolutions/Neptune did not explicitly respond to this interrogatory. All of the concerns
raised in this interrogatory are covered in open supplemental interrogatory comments discussed
in Section 4.1.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 26
2.13.1 Interrogatory Response
No response was provided to this interrogatory in NAC-0105_R0 because this interrogatory was
not included in Utah DEQ (2017). Refer to Section 2.35.1 for response to Supplemental
Interrogatory Comments.
2.14 Interrogatory CR R313-25-8(4)(D)-132/2: Sedimentation Model
DEQ Discussion of NAC-0105_R0 – July 2019
The concerns raised by this interrogatory are addressed in Interrogatory 18 and will no longer be
discussed under this interrogatory. However, this interrogatory remains open until Interrogatory
18 is closed.
2.14.1 Interrogatory Response
Refer to Section 2.3.1.
2.15 Interrogatory CR R313-25-7(2)-150/3: Plant Growth and Cover
Performance
DEQ Discussion of NAC-0106_R0 – July 2019
EnergySolutions/Neptune refer to their response provided in Interrogatories 05/2: Radon Barrier
and 10/03: Effect of Biologicals on Radionuclide Transport supporting their position that the ET
cover design is adequate to protect against intrusion by plants, animals, or ants. As discussed in
Supplemental Interrogatory Comment 2, EnergySolutions/Neptune have used naturalized
parameters in their naturalized HYDRUS modeling; however, a number of issues still remain and
are considered in Section 4.1.2.
2.15.1 Interrogatory Response
Refer to Section 2.35.1 for response to Supplemental Interrogatory Comment 2.
2.16 Interrogatory CR R313-25-8(4)(d)-153/2: Impact of Pedogenic
Processes on the Radon Barrier
DEQ Discussion of NAC-0106_R0 – July 2019
[snip]
As discussed in Supplemental Interrogatory Comment 2, EnergySolutions/Neptune have used
naturalized parameters in their naturalized HYDRUS modeling; however, a number of issues still
remain and are discussed under Supplemental Interrogatory Comment 2 (see Appendix B of the
April 2015 SER and Section 4.1.2 of this report).
2.16.1 Interrogatory Response
Refer to Section 2.35.1 for response to Supplemental Interrogatory Comment 2.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 27
2.17 Interrogatory CR R313-25-7(2)-160/2: Comparison of Class A
West and Federal Cell Designs
DEQ Discussion of NAC-0101_R0 – July 2019
[snip]
This interrogatory currently addresses the same issues as Interrogatory CR R313-25-7(2) and
7(6)-81/2: Comparison of Disposal Cell Designs, which remains open.
2.17.1 Interrogatory Response
Refer to Section 2.9.1.
2.18 Interrogatory CR R313-25-22-162/2: Disposal Cell Stability
DEQ Discussion of NAC-0101_R0 – July 2019
This interrogatory is not closed because the same issues are raised in Interrogatory CR R313-
25-7(2) and 7(6)-81/2: Comparison of Disposal Cell Designs, which remain open.
2.18.1 Interrogatory Response
Refer to Section 2.9.1.
2.19 Interrogatory CR R313-25-7(2)-175/1: Infiltration Rates for the
Federal Cell Versus the Class A West Cell
DEQ Discussion of NAC-0106_R0 – July 2019
EnergySolutions/Neptune refer to discussions in the April 2015 SER Appendix B interrogatories,
including Supplemental Interrogatory Comments 1 through 9 and 11. EnergySolutions/Neptune
also indicate that a discussion of the use of correlated α and Ksat parameter values for flow
modeling is included in the response to Interrogatory 05/2. The topics for Interrogatory 175/1 are
covered in these other interrogatories.
2.19.1 Interrogatory Response
Refer to Sections 2.34.1 through 2.43.1 and 2.1.1 for response to the various Supplemental
Interrogatory Comments and Interrogatory 05/2, respectively.
2.20 Interrogatory CR R313-25-8(5)(a)-176/1: Representative Hydraulic
Conductivity Rates
DEQ Discussion of NAC-0106_R0 – July 2019
In their response, EnergySolutions/Neptune describe the importance of water content and the
effect on hydraulic conductivity and state:
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 28
if the water content is slightly reduced from its saturated water content of 0.481 to a water
content of 0.4, the hydraulic conductivity is greatly reduced from its saturated value of
4.46 cm/day to a value of 0.04 cm/day. This reduction of hydraulic conductivity with
reduced water content is even more pronounced with coarser textured soils. In this
example, a small reduction in water content was shown to produce a 100-fold reduction in
hydraulic conductivity. [Neptune 2018f, p. 69]
This is not a logical argument, in that the unsaturated hydraulic conductivity in HYDRUS is
linearly proportional to the saturated hydraulic conductivity that is assigned to the model,
regardless of the water content at any particularly point or time. Thus, any change in saturated
hydraulic conductivity results in a comparable change in unsaturated hydraulic conductivity and a
corresponding change in water flux. Consequently, the lack of sensitivity of percolation rate to
saturated hydraulic conductivity is not logical and needs a quantitative and mechanistic
explanation. Since EnergySolutions/Neptune assume yearly average infiltration, the saturated
water content will always remain at average values—one of the main reasons DEQ/SC&A has
requested water balance plots of the flow simulation results based on daily output (see
Interrogatory CR R317-6-2.1-20/2: Groundwater Concentrations). Furthermore, the single θs and
Ksat values of 0.481 and 4.46 cm/day, respectively, specified in HYDRUS for the evaporative
zone (Neptune 2015e, Table 8) may be the reason for the lack of sensitivity of infiltration to Ksat
and require additional justification.
2.20.1 Interrogatory Response
This issue is no longer applicable given the supplemental HYDRUS modeling that has been
conducted. Now all hydraulic properties are varied for all layers, for every simulation. See
response in Section 2.1.1 for ET Covers and Section 2.4.1 for daily water balance plots.
2.21 Interrogatory CR R313-25-7(2)-189/3: Modeling Impacts of
Changes in Federal Cell Cover-System Soil Hydraulic
Conductivity and Alpha Values
DEQ Discussion of NAC-0106_R0 – July 2019
EnergySolutions/Neptune respond to this interrogatory by explaining why the sites described in
Benson et al. (2011) do not provide analogs to the Clive site, and that site-specific observations
of soil formation at the Clive site that differ significantly from those described in Benson et al.
(2011) are discussed in their response to Interrogatory 05/2. Similar concerns are raised in
Interrogatory 05/2, which remains open.
2.21.1 Interrogatory Response
Refer to Section 2.1.1 for response to Interrogatory 05/2.
2.22 Interrogatory CR R313-25-7(2)-191/3: Effect of Gully Erosion
DEQ Discussion of NAC-0108_R0 – July 2019
It should be noted that the modeling reported in EnergySolutions 2013a and EnergySolutions
2013c using the USLE and RHEM addresses sheet and rill erosion and not gully formation. Gully
formation will have a greater impact on long-term performance of the Federal Cell. Smith and
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 29
Benson (2016) have shown that water erosion can result in gullies up to about 24 feet deep on
fine-grained soil armored by the addition of up to 40 percent gravel, even in the presence of plant
cover, over a period of only 1,000 years. Such erosion would be devastating to the Clive
embankment.
Some questions remain regarding use of the USLE as one of the tools to evaluate the magnitude
of sheet erosion. In Table 2 of Neptune 2018g, the parameters used to calculate soil loss were
summarized for the top slope and the side slope of the Federal Cell embankment. Sheet erosion
was estimated to be 0.24 tons/acre/year for the top slope and 0.19 tons/acre/year for the side
slopes. It was possible to reproduce the annual soil loss for the top slope using the Table 2
parameters. However, it was not possible to reproduce the side slope soil loss rate from the
Table 2 parameters. Using the available data, the side slope soil loss rate was calculated to be
0.038 tons/acre/year. Incorrect parameters for the top slope gradient (4 percent instead of 2.4
percent) and for the length of the slope (942 feet instead of 521 feet) were used. When applying
the correct parameters, the erosion of the top slope is 0.107 tons/acre/year and for the side slope
is 0.04 tons/acre/year. However, we question the calculational approach of treating the top slope
and the side slope independently.
As quoted above from Appendix D (ES 2013c, p. D-5): “The C factor for the top slopes [0.2] is
based on the sparse vegetative cover naturally found in the areas immediately surrounding the
Clive facility.” Additionally, the Georgia Soil and Water Conservation Commission’s Erosion and
Sedimentation Control Manual (2000) indicates that a C factor of 0.2 is representative of land
with 20 percent ground cover. However, a recent EnergySolutions/Neptune report (Neptune
2015m) cites measured vegetative cover fractions on three plots at the Clive Site as 5.9, 9.1 and
14.4 percent. In light of these differences, EnergySolutions must justify the assumption of a 20
percent vegetative cover.
Although the Neptune 2018g USLE results are conservative relative to the results determined by
DEQ, the fact that there are errors in the calculations bring into question the quality
assurance/quality control (QA/QC) that was or was not performed for Neptune 2018g. This
interrogatory remains open until those QA/QC questions can be resolved. Please provide DEQ
with the NAC-0108 supporting calculations and the procedure under which they were performed,
so that the calculations can be audited. Assuming that these QA/QC issues can be satisfactorily
resolved, there are still some basic questions as to the appropriateness of the USLE model.
There are two slopes: the top slope, and the side slope. The USLE is poorly equipped to analyze
a dual-slope erosion line from the crest of the cell, over the top-slope/side-slope shoulder, down
to the cell base. This is needed, because top-slope water flow and erosion affects side-slope
water flow and erosion. The eroded materials ending up near the edge of the cell may travel in
total as much as half the total embankment width, or about 721 feet. This is considerably more
than 521 feet (which covers only the top slope). Assuming that, for side-slope erosion, the USLE
program is run for only the length of the side slope is not consistent with physical reality.
2.22.1 Interrogatory Response
As noted by DEQ, there are errors in the Neptune 2018g USLE calculation results. DEQ’s
calculation shows that these errors overstate the potential impact of sheet erosion by a factor of
two to five. In other words, the Neptune calculations are conservative in terms of over-estimating
the potential impact of sheet erosion. By Neptune’s calculations, sheet erosion amounts to only a
fraction of a millimeter per year total soil loss. Accordingly, sheet erosion is not likely to be a
critical phenomenon in terms of embankment performance.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 30
Interrogatory 198/1 takes the position that USLE is not a suitable tool for evaluating the potential
impact of gully erosion. As noted in Section 2.27.1, EnergySolutions has chosen to apply rock
armor to the embankment side slopes.
2.23 Interrogatory CR R313-25-7(2)-192/3: Implications of Great Salt
Lake Freezing on Federal Cell Performance
DEQ Discussion of NAC-0106_R0 – July 2019
EnergySolutions/Neptune refer to their response to Interrogatory 05/2 for a description of frost
depth calculations for the Clive site. See the discussion under Interrogatory 05/2: Radon Barrier,
since the same issues with respect to the frost depth calculations are considered there. The
interrogatory remains open.
2.23.1 Interrogatory Response
Refer to Section 2.1.1 for response to Interrogatory 05/2.
2.24 Interrogatory CR R313-25-3 and R313-25-8-195/1: Aquifer
Characterization
DEQ Discussion of NAC-0104_R0 – July 2019
[snip]
EnergySolutions/Neptune must obtain data to “include hydraulic conductivity and other
information necessary to estimate adequately the ground water travel distance” and “to assess
the quality of the ground water of all aquifers identified in the area” (emphasis added).
2.24.1 Interrogatory Response
In a letter dated March 16, 2020, EnergySolutions provided the Phase 1 Basal-Depth Aquifer
Study Report (Stantec 2020) to UDEQ. Field activities were performed consistent with a study
plan developed with and approved by UDEQ. Results indicate limited connectivity between the
shallow zones and the basal aquifer observed at 325 to 355 feet below ground surface (bgs), as
demonstrated by upward groundwater flow, low vertical hydraulic conductivity, observed dry
soils in the aquitard zones, and a lack of response in shallow (30 and 50 feet bgs) wells during
the aquifer test. These results reinforce the long-standing conceptual site groundwater model of
limited upward gradient, low groundwater volumes, and poor groundwater quality.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 31
2.25 Interrogatory CR R313-25-9(5)(A)-196/1: Non-DU Waste
Characteristics
DEQ Discussion of NAC-0102_R0 – July 2019
[snip]
This interrogatory remains open pending resolution of questions involving the density of the
waste (i.e., Unit 3) and how that density may affect infiltration.
2.25.1 Interrogatory Response
As provided in draft Condition 4 of the 2015 SER, EnergySolutions will not dispose of any waste
other than DU in the Federal Cell until such time as an amended PA accounting for the
additional Class A LLRW is approved. This does not preclude approval to dispose of DU in the
interim.
2.26 Interrogatory CR R313-25-25(4) 197/1: Properties of Embankment
Side Slope Materials
DEQ Discussion of NAC-0108_R0 – July 2019
[snip]
In NAC-0108_R0, EnergySolutions/Neptune provide several lines of evidence regarding the
stability of the Federal Cell cover. However, there are problems with each.
• Use of the USLE predicts losses that are below the EPA-recommended value of 2
tons/acre/year. However, there appear to be errors in the soil loss calculations that must
be corrected. (See Interrogatory CR R313-25-7(2)-191/3: Effect of Gully Erosion.)
• Use of RHEM predicts soil loss rates similar to those calculated with USLE. However, as
noted under Interrogatory 200, the RHEM code is no longer functional.
• The maximum permissible velocity method from NUREG-1623 uses a limiting velocity at
variance with NRC guidance. (See Interrogatory CR R313-25-8(4)(a)-71/1: Biotic
Processes in Gully Formation.)
• Use of the borrow pit at Clive to model land form evolution with SIBERIA is flawed
because no attempt is made rationalize the borrow pit parameters with those of the
Federal Cell. Additionally, the description of the borrow pit modeling in Appendix 10 to
DU PA v1.4 is confusing and lacking in detail.
2.26.1 Interrogatory Response
Refer to Sections 2.22.1, 2.29.1, and 2.8.1 for responses to Interrogatories 191/3, 200/1, and
71/1, respectively. As discussed in Section 2.27.1, EnergySolutions has chosen to apply rock
armor to the embankment side slopes; therefore, the discussion of SIBERIA and its applicability
no longer applies.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 32
2.27 Interrogatory CR R313-25-25(4)-198/1: Gravel Content of
Embankment Materials
Discussion of NAC-0108_R0 – July 2019
[snip]
EnergySolutions/Neptune have provided modeling and in-service information on covers with a 15
percent gravel admixture suggesting that such a cover material may function adequately.
However, little useful information on side slopes with a 50 percent gravel admixture was located.
This interrogatory remains open for several reasons, including:
• Use of USLE to model cover erosion does not consider effects of gully erosion.
• Inadequate justification is provided for using a 50 percent gravel mixture on the side
slopes, especially when in-service tests of riprap side slopes is promising.
2.27.1 Interrogatory Response
EnergySolutions has chosen to apply rock armor to the embankment side slopes. Please refer to
engineering drawing series 14004 in Attachment 3.
2.28 Interrogatory CR R313-25-25(4)-199/1: Uncertainties in Erosion
Modeling
DEQ Discussion of NAC-0108_R0 – July 2019
[snip]
EnergySolutions/Neptune have presented arguments for not providing the requested uncertainty
assessment, based on the current, immature state of LEM development. If that is a fair
evaluation of the status of model development, DEQ believes that the SIBERIA modeling adds
little to the ability to characterize the erosion behavior of the Federal Cell. This interrogatory
remains open.
2.28.1 Interrogatory Response
As discussed in Section 2.27.1, EnergySolutions has chosen to apply rock armor to the
embankment side slopes. Therefore, the SIBERIA modeling no longer applies.
2.29 Interrogatory CR R313-25-25(4)-200/1: Use of RHEM to Develop
Parameters for SIBERIA
DEQ Discussion of NAC-0108_R0 – July 2019
[snip]
Based on the inability to perform additional RHEM calculations with a model that considers such
factors as slope lengths, disturbed soils, and concentrated surface water flows,
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 33
EnergySolutions/Neptune should provide an alternative approach for modeling Federal Cell
erosion and to provide fluvial parameters for the SIBERIA model. These parameters then need to
be provided in Appendix 10 for DEQ review. Pending resolution of the cited problems with
RHEM, this interrogatory remains open.
[snip]
2.29.1 Interrogatory Response
As discussed in Section 2.27.1, EnergySolutions has chosen to apply rock armor to the
embankment side slopes. Therefore, the SIBERIA modeling no longer applies.
2.30 Interrogatory CR R313-25-25(4)-201/1: Estimating Rainfall
Intensity
Discussion of NAC-0108_R0 – July 2019
EnergySolutions/Neptune agreed that rainfall intensity was not calculated correctly and provided
revised estimates in NAC-0108_R0 (Neptune 2018g). The corrected values for the Federal Cell
are presented in Table 6 of that document. Under this interrogatory, EnergySolutions/Neptune
also discussed the NRC maximum permissible velocity (MPV) approach for gully formation (NRC
2002). See also Interrogatory 71. We believe that the EnergySolutions/Neptune-selected
maximum permissible velocity of 5 ft/s is not consistent with NRC recommendations. Using an
MPV of 2.5 to 3 ft/s as recommended by the NRC (2002, Appendix A, p. A-3) results in
calculated flow velocities that exceed the MPV.
This interrogatory remains open because the MPV used was not consistent with NRC guidance.
2.30.1 Interrogatory Response
Refer to Section 2.8.1 for response to Interrogatory 71/1.
2.31 Interrogatory CR R313-25-25(4)-202/1: Use of SIBERIA to Model
Federal Cell Erosion
DEQ Discussion of NAC-0108_R0 – July 2019
[snip]
EnergySolutions/Neptune chose not to discuss why the borrow pit modeling was applicable to
describing the Federal Cell cover performance and stated that use of SIBERIA in its present
state of development should not be considered to provide conclusive results for licensing
decisions. Consequently, we question the use of SIBERIA to demonstrate embankment stability.
This interrogatory remains open.
2.31.1 Interrogatory Response
As discussed in Section 2.27.1, EnergySolutions has chosen to apply rock armor to the
embankment side slopes. Therefore, the SIBERIA modeling no longer applies.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 34
2.32 Interrogatory CR R313-25-9(5)(a)-206/1: Temporal Uncertainty in
Performance Assessment
Interrogatory Statement
Please provide an explanation as to why the DU PA GoldSim Model v1.4 uncertainty seems to
be decreasing with increasing time.
2.32.1 Interrogatory Response
The interrogatory observation is correct. The noted phenomenon is a direct result of the model
structure, which is typical for this type of model. The DU PA model is a steady-state
probabilistic model, in which variability is applied in the initial values selected from the input
probability distributions for each parameter and is then held constant for the duration of the
model run. Uncertainty is assessed in this type of modeling by running many realizations and
evaluating the variability of results across realizations, as well as by performing sensitivity
analysis on many runs of the model.
These types of probabilistic models are not expected to capture the effects of increasing
uncertainty in societal, natural, or engineering systems over geologic timeframes, largely because
such changes in uncertainty are difficult at best to estimate – that is, it is essentially impossible to
predict the future of engineered systems, the environment, and characteristics of society this far
into the future.
PA models are set up mostly to project and evaluate performance from the long term effects of a
steady state system. In effect these types of models condition on current knowledge and project
that knowledge throughout time – they do not allow for increased uncertainty in time through the
input parameters. Uncertainty does change for different endpoints (concentration, dose, flux at
different locations) but only through the fate (decay and ingrowth) and transport of the
radionuclides through time.
The Radiation Control Board promulgated a 10,000-year compliance period directly in UAC
R313-25-9(a). Accordingly, these concerns are ultimately immaterial to evaluation of the DU
PA, and the DU PA model does not attempt to introduce changing uncertainty in any input
parameters for the 10,000-year compliance period. As noted in the interrogatory, NRC and others
have wrestled with this issue in relation to determining a suitable timeframe for the compliance
period in LLW performance assessments. Ultimately, this is why NRC has determined that the
compliance period should be no longer than 1,000 to 10,000 years. As quoted in the
interrogatory, “PAWG [Performance Assessment Working Group] also recognizes that the
uncertainties in calculations increase with time, thus for very long timeframes (such as beyond
10,000 years) such calculations are best used for making qualitative evaluations” [emphasis
added].
To reinforce the comments above, NRC (2011) includes a conceptual representation of types of
uncertainties and their relative magnitudes in the near-surface disposal of radioactive waste
(reproduced below as Figure 9).
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 35

Figure 9. Types of uncertainties and their relative magnitudes in the near-surface disposal
of radioactive waste (from NRC (2011)).
As noted in NRC (2011) and depicted on Figure 9, “If technology development is considered,
societal uncertainties are likely to dominate other sources of uncertainty… In terms of
development and deployment of technologies to identify, characterize, and remediate
environmental hazards, the growth rate has been exponential (and positive) over the last several
hundred years. The technologies that are employed today, in many cases, did not exist one
hundred years ago.” It is worth noting that, while Figure 9 depicts societal uncertainty as
increasing off the scale within no more than several hundred years, the scenarios considered for
compliance evaluations are held static. If anything, LLRW disposal regulations drive inadvertent
intruder scenarios that artificially assume knowledge about radioactive waste sites and hazards,
together with strategies for mitigating these hazards, are lost within 100 years or by the end of an
institutional control period.
Tauxe (2015) considers these questions as follows: “The peak radioactivity of a mass of refined
U-238 is about 2 million years. Consider that the genus Homo is roughly two million years old.
Are we to base decisions on estimates of risk to some future hominid? What assumptions must
be made in these estimates, and are those assumptions defensible? How are decision makers to
use this information intelligently? The role of performance assessment is to inform this type of
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 36
decision making. Radioactive waste decision makers need useful decision tools supported by
computer models of specific radioactive waste sites under their purview. Today’s performance
assessments explicitly include uncertainty, but that is not always quantifiable, especially when
considering deep time.”
The International Atomic Energy Agency states that “for above surface disposal facilities (e.g.
for waste from mining), the uncertainties in modeling results will already be substantial when
considering periods of several hundred years, and quantitative estimates may become
meaningless already beyond a period of a thousand years.” (IAEA 2012). The International
Commission on Radiation Protection (ICRP) (ICRP 1998) states that “doses and risks, as
measures of health detriment, cannot be forecast with any certainty for periods beyond around
several hundreds of years into the future”.
Accordingly, the DU PA model does not attempt to quantify increasing uncertainty over future
millenia. An appearance that uncertainty is modeled to decrease with increasing time is an
artifact of the model structure. Nonetheless, the DU PA model continues to provide decision-
makers with useful information about DU disposal at the Clive facility.
2.33 Interrogatory CR R313-25-23-207/1: Stability of Disposal Site
Interrogatory Statement
The footprint of the Federal Cell embankment is situated directly adjacent to the 11e.(2) Cell.
Please provide an analysis of the impact of the Federal Cell on the stability of the adjacent
11e.(2) Cell and demonstrate that consolidation settlement will not negatively affect the
performance of the Federal Cell.
2.33.1 Interrogatory Response
EnergySolutions has informed Neptune that it will respond to this issue under separate cover.
2.34 SER B.1 Supplemental Interrogatory Comment 1
1) Demonstrate why 20 HYDRUS runs are sufficient to capture the parameter uncertainty.

DEQ Discussion of NAC-0106_R0 – July 2019
[snip] EnergySolutions/Neptune need to submit the 50 naturalized HYDRUS simulations and
conduct similar infiltration comparisons, statistical testing, and resolve scaling issues for the
naturalized conceptual model as described below.
[snip]
Scaling of the van Genuchten α and n parameters it is an identical issue to the scaling concerns
raised under aeolian deposition in Interrogatory R313-25-8(5)(A)-18/3 (see Section 2.1.3). In
both cases, EnergySolutions/Neptune proposed replacement of the standard deviation in the
normal distribution with the standard error to account for spatial and temporal scaling. The only
difference is the number of samples used to convert the standard deviation to the standard error:
28 for the van Genuchten parameters and 11 for the aeolian deposition.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 37
Conceptually, we agree that “some” correction could legitimately be made to account for the
spatial and temporal variations. However, we do not believe that EnergySolutions/Neptune have
provided justification for the approach that they are taking. EnergySolutions/Neptune need to
provide a detailed, step-by-step description of how they conducted the averaging process, and
how this process compared to accepted procedures for spatial averaging of hydraulic properties
in spatially correlated geo-media.
[snip] In Section 2.1.3, our latest response to EnergySolutions/Neptune regarding aeolian
deposition under Interrogatory R313-25-8(5)(A)-18/3, we state that we have reviewed the three
references provided earlier in the response by EnergySolutions/Neptune (i.e., Blöschl and
Sivapalan (1995); Neuman and Wierenga (2003); Zhang et al. (2004)), and, while the references
describe the need for upscaling, they do not “technically establish” the use of the standard error
in this manner. Zhang et al. (2004) do state that “scaling should only be applied over a limited
range of scales and in specific situations,” while Neuman and Wierenga (2003) states that “One
approach has been to postulate more-or-less ad hoc rules for upscaling based on numerically
determined criteria of equivalence”. DEQ remains concerned that (1) the proposed embankment
is not a specific situation that allows for upscaling and (2) the approach taken by
EnergySolutions/Neptune is “an ad hoc rule” rather than a “technically established” approach.
[snip]
Until the approach taken to scaling is better justified and the sensitivities of scaling factors on
infiltration rates are evaluated, this interrogatory will remain open.
2.34.1 Interrogatory Response
These hydrologic issues are no longer applicable given the supplemental HYDRUS simulations
using hydraulic property inputs as recommended by Dr. Benson. Please see response in Section
2.1.1 for ET Covers.
Regarding the scaling of van Genuchten alpha and n parameters, it is notable that the ranges of
these parameters that were input into the original 50 HYDRUS runs (and used with DU PA v1.4)
are actually wider than those generated for the supplemental HYDRUS runs, which use the Hyd
Props Calculator.xls and recommendations from Dr. Benson. The only exception is the two
smallest values of van Genuchten n, where the Hyd Props Calculator.xls generated lower values
than the range in the original HYDRUS model. This can be seen in Figure 10.
Scaling is a critical aspect of these types of models and should be applied to both the spatial and
temporal domains. The underlying issue is that, often, data available to specify input
distributions are not obtained at the scale of the model. Often available data represent little more
than points in space and time, and yet they are expected to be applied to large area or volumes
and large time scales. There are not many other modeling paradigms that need to consider these
types of issues. For example, groundwater modeling often has very refined grid networks that
eliminate the need for spatial scaling for at least some input parameters, and they are not usually
run far into the future. The only other “industry” that needs to address scaling on the same order
as PA modeling is climate change modeling, for which scaling is a routine part of model
development. If scaling is not performed, then the specification of the model is simply wrong.
However, the methods by which scaling is performed require specific attention to ensure that risk
dilution does not occur.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 38
To provide another sense of scaling, when someone drives to work and look at their
speedometer, they might see 75 mph if they are on a highway. This does not mean that it takes
only 12 minutes to travel 15 miles to work. However, if a data distribution was used as input to a
probabilistic simulation similar to these PA models, then such a realization would be possible.
Similarly, if stopped at a traffic light, then the speedometer will read 0 mph, in which case the
driver will never get to work, unless the model is scaled properly. The point of scaling is not to
remove conservatism, it is to accurately reflect the system that is being modeled.
Neptune has sometimes used an example from the Nevada National Security Site, in which 400
or more data were collected for near surface water content. The data ranged from 3-30%, but it
would not be reasonable to pull a data point of 30% and apply it to the entire Mojave desert for
the next 10,000 years. Instead the distribution must be scaled to reflect the system level response.
Some have argued that perhaps the 30% might apply to some subset of space in the Mojave, and
if that is the case, then the scaling should accommodate that, and not simply scale (average)
away the effect of a comparatively wet area. However, that is not the case in the Mojave desert.
If such a comment is aimed at change in time instead, then the model should address that directly
(this is despite, in general, not accommodating increased uncertainty in time). For example, the
DU PA model incorporates such a change in time by allowing large lakes to return to the
Bonneville Basin in deep time.
Scaling is absolutely necessary if the scale of the data and the scale of the model do not match.
The method by which scaling is done needs to match the model structure so that risk dilution is
not a consequence. For example, averaging maximum plant root depth (a common parameter in
PA models) could result in plants not getting into a waste cell. However, averaging plant root
mass by depth layer allows an evaluation of the average root mass in the waste cell. The point
being, scaling is necessary, but needs to be done carefully.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 39

Figure 10. Sorted values of van Genuchten alpha and n, and Ksat, for the original 50
HYDRUS runs (green line), and using the Hyd Props Calculator (blue line).
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 40
2.35 SER B.2 Supplemental Interrogatory Comment 2
2) The Table 9 HYDRUS parameters do not appear to “bound” the α, n, and Ksat
distributions. For example, in the distribution, Ksat ranges from 0.0043 to 52 cm/day, but
in the 20 HYDRUS runs Ksat only ranged from 0.16 to 10.2 cm/day.

DEQ Discussion of NAC-0106_R0 – July 2019
[snip]
As discussed below, DEQ/SC&A has done an independent assessment of the tabular data
(Neptune 2018f, Table 6) provided for the 50 realizations of hydraulic properties relative to
NUREG/CR-7028. They are consistent with the recommendations in NUREG/CR-7028, and we
believe the Monte Carlo method used to generate this set of 50 realizations is appropriate.
However, concerns remain regarding hydraulic properties used for the five-layer model
(particularly the residual volumetric water content). These properties are not adequately justified.
[snip]
How these properties were specifically used in the modeling is not clear, however. Were these
hydraulic properties used for the radon barrier, the overlying layers, all layers? In Neptune
(2018f), there is discussion of the entire cover profile being treated as a single layer with one set
of properties, but other discussion of the radon barrier being separate from other layers in the
hydrologic model. Moreover, the statement quoted above from Neptune (2018f) is ambiguous.
Were the fluxes obtained from the HYDRUS runs made with these properties used in the final
GoldSim model? We do believe that EnergySolutions/Neptune have used properties consistent
with NUREG/CR-7028 in their modeling, but we are not clear on whether the outcomes from that
modeling were carried through into the GoldSim model.
[snip]
2.35.1 Interrogatory Response
These issues are now moot given the new HYDRUS simulations, which use hydraulic property
inputs as recommended by Dr. Benson. See response in Section 2.1.1 for ET Covers.
Yes, the outcomes from the homogeneous layered model were fully carried through in the
GoldSim model (identified as v1.4XXX). However, this issue is moot given that Dr. Benson
stated that this was a misunderstanding, and that a homogeneous cover was never SC&A’s and
Dr. Benson’s intent.
2.36 SER B.3 Supplemental Interrogatory Comment 3
3) NUREG/CR-7028 (Benson et al. 2011) gives the “in-service hydraulic conductivity” as
ranging from 7.5 × 10-8 to 6.0 × 10-6 m/s [0.7 to 52 cm/day], with a mean of
4.4 × 10-7 m/s [3.8 cm/day]. Instead of using the provided distribution (i.e., log-triangular
with a minimum, maximum, and most likely), ES/Neptune constructed a lognormal
distribution with a mean and standard deviation of 0.691 and 6.396 cm/day,
respectively. Provide the justification for this approach. For example, the selection of
0.0043 cm/day as the lower end of the Ksat distribution requires justification (Appendix 5,
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 41
p.41). It is not clear why a design parameter value should be used when adequate field
data are available. The number chosen by the Licensee for the lower end of the
distribution range in the GoldSim implementation is 163 times lower than the lowest
value in the range specified within the NUREG guidance (see Section 13.0 of Appendix
5, Unsaturated Zone Modeling to the Clive DU PA). We believe that use of the design
parameter biases the Ksat distribution in a non-conservative manner.
DEQ Discussion of NAC-0106_R0 – July 2019
The following discussion pertains to the layered HYDRUS modeling without naturalization. The
sensitivity of saturated hydraulic conductivity to the HYDRUS abstraction modeling with
naturalized parameters was not presented by EnergySolutions/Neptune.
The lack of significance of saturated hydraulic conductivity in the absence of saturated hydraulic
conductivity in the linear model remains a concern, as saturated hydraulic conductivity is known
to be a key factor affecting the percolation rate from water balance covers (e.g., Bohnhoff et al.
2009). The unsaturated hydraulic conductivity used in HYDRUS scales linearly with the saturated
hydraulic conductivity that is input to the model. Consequently, fluxes should scale directly with
saturated hydraulic conductivity. Adequate explanation and documentation are needed for this
outcome, which is inconsistent with expectations and the literature. This issue can be resolved
by presenting parametric simulations with the unsaturated flow model that illustrate
mechanistically why the saturated hydraulic conductivity is not important. Water balance graphs,
based on daily outputs, as requested previously, and other graphs showing hydrologic behavior
(e.g., temporal evolution of water content profiles) could provide the documentation needed to
demonstrate that the unsaturated flow model is not flawed, and that the saturated hydraulic
conductivity is not significant.
Water balance graphs and other graphs of hydrologic behavior would also resolve queries
regarding the saturated hydraulic conductivity assumed for the Surface Layer and the propensity
for a modeled capillary break between the Surface Layer and the Frost Protection Layer. These
graphs can be prepared using daily output from the unsaturated flow model for several one-year
periods to illustrate the hydrologic behavior.
These graphs and other reporting at shorter time scales may not be necessary as high-level
output for a performance assessment, but they are necessary to build confidence in the model
and to identify model shortcomings. Without confidence in the model, the high-level output
cannot be evaluated. Until the sensitivity of the saturated hydraulic conductivity to the naturalized
HYDRUS abstraction modeling is provided, in conjunction with sufficient backup material (e.g.,
water balance graphs) that adequately explain any anomalous behavior, this interrogatory will
remain open.
2.36.1 Interrogatory Response
See response in Section 2.1.1 for ET Covers, and Section 2.4.1 for daily water balance plots.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 42
2.37 SER B.4 Supplemental Interrogatory Comment 4
4) Provide justification for using the Rosetta database, as appropriate for an
engineering earthen cover.
DEQ Discussion of NAC-0106_R0 – July 2019
In addition to the non-naturalized HYDRUS simulations, EnergySolutions/Neptune have also
used the ranges of recommended naturalized parameters in HYDRUS. However, until the
ambiguities and other concerns that are described in Section 4.1.2 are resolved, this
interrogatory will remain open (see Supplemental Interrogatory Comment 2).
2.37.1 Interrogatory Response
Refer to Section 2.35.1 for response to Supplemental Interrogatory Comment 2.
2.38 SER B.5 Supplemental Interrogatory Comment 5
5) a) Provide additional explanation/justification for the assumed surface boundary
condition and the sensitivity of the HYDRUS results to the boundary conditions.
b) Also, why is a linear regression the optimal surface response for the design?
DEQ Discussion of NAC-0106_R0 – July 2019
The rationale for representing a 1,000-year meteorological record using 10 sequential runs with
the same 100-year record has not been clarified if the performance assessment is to represent a
1,000-year scenario. If the objective is solely to provide a prediction for 100 years that is
independent of the assumed initial condition, then this approach can be acceptable provided that
metrics are provided showing that 10 repetitions is acceptable to eliminate the effects of the
initial condition. However, if the objective is to provide predictions for a 1,000-year scenario, then
this approach will not represent extreme conditions adequately. For example, a 1,000-year storm
event is much different from a 100-year storm event. These issues need to be clarified.
The linear model used in GoldSim for percolation rates has been shown to provide a reasonable
prediction of percolation rates predicted by the non-naturalized HYDRUS model, except at the
extreme case where the linear model underpredicts the percolation rate by approximately a
factor of 2 (see Figure 20 in Neptune (2018f) responses, reproduced as Figure 34 below). The
prediction with the quadratic model is in closer agreement but is still an underprediction. Given
the limited number of realizations used in the HYDRUS model (50), the tails of the distribution
likely are underrepresented, and a much greater deviation may exist for higher percolation rates.
The significance of underprediction in the tails needs greater documentation and the impact of
this underprediction needs to be quantified. Furthermore, an analogous analysis needs to be
performed using the naturalized HYDRUS net infiltration rates. Therefore, this interrogatory
remains open.
2.38.1 Interrogatory Response
The response in Section 2.1.1 on ET Covers includes a revised 1,000-year precipitation record
that is more representative of a 1,000-year period (e.g., includes extreme events), as opposed to
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 43
repeating a 100-year record 10 times as was used with the HYDRUS modeling for the DU PA
Model v1.4.
The new 1,000-year precipitation record was developed using the WGEN sub-model within
SWAT (https://swat.tamu.edu/), and using site-specific data from Clive, Utah. The old 100-year
record was developed using the WGEN sub-model within HELP (https://www.epa.gov/land-
research/hydrologic-evaluation-landfill-performance-help-model), and using site-specific data
from Dugway, Utah, that was scaled to conditions measured in Clive, Utah.
A comparison of these two precipitation records is shown in Figure 11, where precipitation is
sorted from highest daily values to lowest non-zero values. The new record has more high
precipitation events: 136 days with precipitation over 1 inch/day, compared to 50 days for the old
record. The new record has fewer days with precipitation: 66,696 days for the new record ,versus
80,930 days for the old record.

Figure 11. Comparison of old 100-yr precipitation record, generated with HELP, and new
1,000-yr precipitation record, generated with SWAT. Precipitation is plotted from
highest to lowest daily precipitation.
Daily precipitation for the new record (1,000 years) and the old record (100 years, repeated 10
times) is shown in Figure 12. The x-axis is days, and the y-axis is daily precipitation in inches
per day. This figure clearly shows the increase in number of days with higher precipitation in the
new record compared to the old record.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 44

Figure 12. Comparison of old 100-yr precipitation record, generated with HELP, and new
1,000-yr precipitation record, generated with SWAT. Daily precipitation is shown for
each record.
Histograms illustrating the distribution of daily precipitation for the new and old records are
shown in Figure 13. Note that the y-axis for the new record (left) is for 1000 years of daily data
while the y-axis for the old record (right) is for 500 years of daily data.

Figure 13. Histograms showing the distribution of daily precipitation for the old 100-yr
precipitation record, generated with HELP, and the new 1,000-yr precipitation
record, generated with SWAT.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 45
Regarding item 5b, and the second paragraph above, these issues are no longer applicable given
the supplemental HYDRUS modeling that has been conducted.
2.39 SER B.6 Supplemental Interrogatory Comment 6
6) To summarize the 20 HYDRUS results, Appendix 5, Section 12.9 states: “Infiltration flux
into the waste zone ranged from 0.007 to 2.9 mm/yr, with an average of 0.42 mm/yr, and
a log mean of 0.076 mm/yr for the 20 replicates.” In addition to this statement, provide
the results for each HYDRUS run so that the results can be matched to the input data.
DEQ Discussion of NAC-0106_R0 – July 2019
EnergySolutions/Neptune have provided the non-naturalized HYDRUS input and output for the
50 simulations; however, EnergySolutions/Neptune have not provided the naturalized HYDRUS
input and output for the 50 simulations.
2.39.1 Interrogatory Response
These issues are no longer applicable given the supplemental HYDRUS simulations, which use
hydraulic property inputs as recommended by Dr. Benson. See response in Section 2.1.1 for ET
Covers.
The outcomes from the homogeneous layered model were fully carried through in the GoldSim
model (identified as v1.4XXX). However, this issue is moot given that Dr. Benson stated that
this was a misunderstanding, and that a homogeneous cover was never SC&A’s and Dr.
Benson’s intent.
2.40 SER B.7 Supplemental Interrogatory Comment 7
7) The HYDRUS and GoldSim calculated infiltration rates (and perhaps other intermediary
results) need to be provided in the report, so that the reviewers do not have to delve into
the code’s output files. For example, provide dot plots of the infiltration rates through the
surface layer and/or provide a statistical summary of the infiltration rates that were
sampled in GoldSim.
DEQ Discussion of NAC-0106_R0 – July 2019
In response to this interrogatory, EnergySolutions/Neptune explain their use of spatial scaling
and long-term averaging, and normal distribution of data in the Rosetta data base. As discussed
in Supplementary Interrogatory Comment 2, considerable ambiguity remains regarding which
hydraulic properties are the basis for the current water fluxes employed in the performance
assessment. As indicated previously, naturalized hydraulic properties should be used for all
layers in the cover, and those properties should be consistent with recommendations in
NUREG/CR-7028 (Benson et al. 2011). The graphs provided by EnergySolutions/Neptune in
response to this interrogatory do not provide the intermediate results requested in order to better
evaluate the consistency between the GoldSim and HYDRUS results.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 46
2.40.1 Interrogatory Response
This issue is no longer applicable given the supplemental HYDRUS modeling that has been
conducted. Now all hydraulic properties are varied for all layers, for every simulation. See
response in Section 2.1.1 for ET Covers.
2.41 SER B.8 Supplemental Interrogatory Comment 8
8) a) Demonstrate that the fitted equations for water content and infiltration (Appendix 5,
Equations 39 and 40, and Table 10) give “reasonable” results when compared to
HYDRUS.
b) For example, provide an explanation for why Ksat is insensitive to the infiltration rates.
DEQ Discussion of NAC-0106_R0 – July 2019
The linear model used in GoldSim for percolation rates has been shown to provide a reasonable
prediction of percolation rates, except at the extreme case where the linear model underpredicts
the percolation rate by approximately a factor of 2 (see Figure 20 in the Neptune (2018f)
responses, reproduced as Figure 34 above). The prediction with the quadratic model is in closer
agreement but is still an underprediction. Given the limited number of realizations used in the
HYDRUS model (50), the tails of the distribution likely are underrepresented, and a much greater
deviation may exist for higher percolation rates. The significance of underprediction in the tails
needs greater documentation, and the impact of this underprediction needs to be quantified.
The lack of significance of saturated hydraulic conductivity in the absence of saturated hydraulic
conductivity in the linear model remains a concern, as saturated hydraulic conductivity is known
to be a key factor affecting the percolation rate from water balance covers (e.g., Bohnhoff et al.
2009). The unsaturated hydraulic conductivity used in HYDRUS scales linearly with the saturated
hydraulic conductivity that is input to the model. Consequently, fluxes should scale directly with
saturated hydraulic conductivity. Adequate explanation and documentation are needed for this
outcome, which is inconsistent with expectations and the literature. This issue can be resolved
by presenting parametric simulations with the unsaturated flow model that illustrate
mechanistically why the saturated hydraulic conductivity is not important. Water balance graphs,
as requested previously, and other graphs showing hydrologic behavior (e.g., temporal evolution
of water content profiles) could provide the documentation needed to demonstrate that the
unsaturated flow model is not flawed, and that the saturated hydraulic conductivity is not
significant.
Water balance graphs and other graphs of hydrologic behavior showing [sic] would also resolve
queries regarding the saturated hydraulic conductivity assumed for the Surface Layer and the
propensity for a capillary break between the Surface Layer and the Frost Protection Layer. These
graphs can be prepared using daily output from the unsaturated flow model for several one-year
periods to illustrate the hydrologic behavior.
These graphs and other reporting at shorter time scales may not be necessary as high-level
output for a performance assessment, but they are necessary to build confidence in the model
and to identify model shortcomings. Without confidence in the model, the high-level output
cannot be evaluated.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 47
2.41.1 Interrogatory Response
Paragraph 1 of the DEQ Discussion above re-states comments made in response to Supplemental
Interrogatory Comment 5 above. Accordingly, refer to Section 2.38.1 for a response to these
concerns.
Paragraphs 2, 3, and 4 of the DEQ Discussion above re-state comments made in response to
Supplemental Interrogatory Comment 3 above. Accordingly, refer to Section 2.36.1 for a
response to these concerns.
2.42 SER B.9 Supplemental Interrogatory Comment 9
9) Compare the moisture contents calculated using the fitted equations to the Bingham
(1991, Table 6 and/or Appendix B) Clive site measured Unit 4 moisture contents,
and rationalize any differences.
DEQ Discussion of NAC-0106_R0 – July 2019
The volumetric water content provided in Bingham Environmental (1991) appears too high,
based on data from other similar sites. An unreasonably high volumetric water content leads to
infiltration being underestimated.
2.42.1 Interrogatory Response
This issue is no longer applicable given the supplemental HYDRUS modeling that has been
conducted. Now all hydraulic properties are varied for all layers, for every simulation. See
response in Section 2.1.1 for ET Covers.
2.43 SER B.11 Supplemental Interrogatory Comment 11
DWMRC provided EnergySolutions with an Excel file, “Clive Hydrus Sensitivity Recommend
REV2.xlsx,” which contains suggested or proposed combinations of input values for the
HYDRUS runs used to support the Clive DU PA.
DEQ Discussion of NAC-0106_R0 – July 2019
As described in Supplemental Interrogatory Comment 2, EnergySolutions/Neptune have used
naturalized parameters in one set of HYDRUS simulations. However, the results and other
relevant information have not been presented. Furthermore, water balance and other hydrologic
information is requested in Supplemental Interrogatory Comment 11; this information is also
required to close Supplemental Interrogatory Comment 8. Therefore, this supplemental
interrogatory will remain open until these other supplemental interrogatories are closed.
2.43.1 Interrogatory Response
This issue is no longer applicable given the supplemental HYDRUS modeling that has been
conducted. Now all hydraulic properties are varied for all layers, for every simulation. See
response in Section 2.1.1 for ET Covers and Section 2.4.1 for daily water balance plots.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 48
3.0 Attachments
1. Deep Time Assessment for the Clive DU PA Model v1.5
2. EnergySolutions DU PA Deep Time Results v1.5
3. Federal Waste Disposal Cell engineering drawings, series 14004
4.0 Conclusion
DU PA v1.4 demonstrates compliance with the dose and groundwater protection requirements of
Utah regulations relating to DU disposal. The interrogatory and response process has added to
the record supporting these conclusions; but has not caused the quantitative model demonstrating
compliance with UAC R313-25-9(5)(a) to require revision. Accordingly, DU PA v1.4 remains
the basis for demonstrating compliance of the disposal facility.
The Deep Time model has been revised to v1.5 in response to concerns identified in
Interrogatory CR R313-25-8(5)(A)-18/3. The revised model continues to demonstrate radon
fluxes that are comparable with those observed in Deep Time model v1.4.
Compliance with UAC R313-25-9(5)(a) is affirmed by DU PA v1.4 and Deep Time model v1.5,
together with their supporting documentation as supplemented by the interrogatory/response
cycle.
5.0 References
Benson, C.H., and T. Gurdal, 2013. Hydrologic Properties of Final Cover Soils. In Foundation
Engineering in the Face of Uncertainty, Honoring Fred H. Kulhawy, Geotechnical
Special Publication No. 229, edited by J.L. Withiam, et al., pp. 283–297, American
Society of Civil Engineers (ASCE), Reston VA
EnergySolutions, 2015. Radioactive Material License UT2300249: Safety Evaluation Report for
Condition 35.B Performance Assessment; Response to Issues Raised in the April 2015
Draft Safety Evaluation Report, EnergySolutions LLC, Salt Lake City UT, November
2015
EnergySolutions, 2018. Radioactive Material License UT2300249: Responses to the Amended
and New Interrogatories Related to Clive DU PA Modeling Report Version 1.4 Dated
November 2015, EnergySolutions LLC, Salt Lake City UT, April 2018
EnergySolutions, 2020. LLRW and 11e.(2) Construction Quality Assurance/Quality Control
(CQA/QC) Manual, Revision 28c, EnergySolutions LLC, Salt Lake City UT, February
2020
IAEA. The Safety Case and Safety Assessment for the Disposal of Radioactive Waste. Vienna,
Austria: International Atomic Energy Agency; Specific Safety Guide No. SSG-23; 2012.
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 49
ICRP. Radiation protection recommendations as applied to the disposal of long-lived solid
radioactive waste: ICRP Publication 81. Annals of the ICRP 28:13-22, 1998.
Nelson, J.D., et al., 1986. Methodologies for Evaluating Long-Term Stabilization Designs of
Uranium Mill Tailings Impoundments, NUREG/CR-4620, ORNL/TM-10067, United
States Nuclear Regulatory Commission (NRC), Washington DC, June 1986
Neptune, 2011. Final Report for the Clive DU PA Model version 1.0, Neptune and Company
Inc., Los Alamos NM, June 2011
Neptune, 2014. Final Report for the Clive DU PA Model, Clive DU PA Model v1.2, NAC-
0024_R2, Neptune and Company, Inc., Los Alamos NM, August 2014
Neptune, 2015. Final Report for the Clive DU PA Model, Clive DU PA Model v1.4, NAC-
0024_R4, Neptune and Company Inc., Los Alamos NM, November 2015
NRC, 2000. A Performance Assessment Methodology for Low-Level Radioactive Waste Disposal
Facilities: Recommendations of NRC's Performance Assessment Working Group,
NUREG-1573, United States Nuclear Regulatory Commission, Washington DC, June
2000
NRC, 2011. Technical Analysis Supporting Definition of Period of Performance for Low-Level
Waste Disposal, United States Nuclear Regulatory Commission, Washington DC, 2011
NRCS, 2001. Part 650 Engineering Field Handbook, National Engineering Handbook, Chapter
14, Water Management (Drainage), Natural Resources Conservation Service, United
States Department of Agriculture, Washington DC, April 2001
NRCS, 2007. Part 650 Engineering Field Handbook, Chapter 7, Grassed Waterways, Natural
Resources Conservation Service, United States Department of Agriculture, Washington
DC, December 2007
SC&A, 2015a. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
2, prepared for Utah Department of Environmental Quality, SC&A, Vienna VA, April
2015
SC&A, 2015b. Utah Division of Radiation Control, EnergySolutions Clive LLRW Disposal
Facility, License No: UT2300249; RML #UT 2300249, Condition 35 Compliance Report;
Appendix A: Final Report for the Clive DU PA Model, Safety Evaluation Report, Volume
1, prepared for Utah Department of Environmental Quality, SC&A Inc., Vienna VA,
April 2015
Clive DU PA Model—Response to Model Version 1.4 Amended Interrogatories
24 April 2020 50
SCS, 1984. SCS Engineering Field Manual, Chapter 7, Grassed Waterways, Soil Conservation
Service, 1984
Stantec, 2020. Phase 1 Basal-Depth Aquifer Study Report, Clive Disposal Facility, prepared for
EnergySolutions LLC, Stantec Consulting Services Inc., Salt Lake City UT, March 2020
Tauxe, J., 2015. Radioactive Waste Disposal and Protection of the Future Public, International
Journal of Technoethics 6 (2) 86–95 doi: 10.4018/IJT.2015070107
Utah DEQ, 2017. Division of Waste Management and Radiation Control, EnergySolutions Clive
LLRW Disposal Facility License No: UT2300249; RML #UT 2300249, Amended and
New Interrogatories Related to Clive DU PA Modeling Report Version 1.4 Dated
November 2015, Utah Department of Environmental Quality (DEQ), Salt Lake City UT,
May 2017
Utah DEQ, 2019a. Depleted Uranium Performance Assessment (DU PA); Clive Facility; Model
Version 1.4 Amended Interrogatories; Radioactive Materials License #2300249, Utah
Department of Environmental Quality, Salt Lake Cuty UT, July 2019
Utah DEQ, 2019b. Clarification Letter for DU PA Compliance Period of 10,000 Years RML #
2300249 Utah Department of Environmental Quality, Salt Lake City UT, August 29,
2019

- NAC-0032_R5

Deep Time Assessment for the Clive
DU PA

Deep Time Assessment for the Clive DU PA
Model v1.5
30 March 2020

Prepared by
NEPTUNE AND COMPANY, INC.
1505 15th St, Suite B, Los Alamos, NM 87544
Deep Time Assessment for the Clive DU PA
30 March 2020 ii
1. Title: Deep Time Assessment for the Clive DU PA
2. Filename: Deep Time Assessment v1.5.docx
3. Description: This report describes details of the “deep time” component of the Clive DU PA Model. The “deep
time” model addresses long term effects (beyond 10,000 years post-closure) of disposal of DU at the Clive
facility.
Name Date
4. Originator Bruce Crowe, Aharon Fleury 30 March 2020
5. Reviewer Sean McCandless, Paul Black, Dan Levitt 30 March 2020
6. Remarks

19 Feb 2020: Edited multiple sections on eolian and lake sedimentation for clarification and justification of
distribution of sediment thickness per event for intermediate lakes; added Table to Appendix B from data
developed by Jack Oviatt: Sediment Thickness Glacial Cycles – B. Crowe

R1-R4 prepared by Bruce Crowe and Robert Lee. Reviewed by Paul Black, Katie Catlett, and Dan Levitt.

4 Nov 2015: Added information to section 7, especially regarding dose calcs. K.Catlett and R. Perona

1 Nov 2015: Added to Table 1 dose parameters. Revised and added text relevant to latest model (v1.4)
consolidation and further CSM clarification.

15 Oct 2015: Thorough edits and revisions to add latest GoldSim modeling and model simplification justification.
– K. Catlett

09 Sep 2015: Edits – B. Crowe and J. Oviatt

03 Sep 2015: Edits- B. Crowe and R. Lee.

27 Aug 2015: Merged “Deep Time Supplemental Analysis. . .” white paper with the Deep Time white paper – R.
Lee

30 Jul 2014: Updates and corrections for v1.2. White Paper now at rev 3. — R. Lee and J. Tauxe

3 Jul 2014; R2: Accepted track changes from R1 and added “a” and “b” to identify two Oviatt et al. (1994)
references – D. Levitt

Deep Time Assessment for the Clive DU PA
30 March 2020 iii

This page is intentionally blank, aside from this statement.
Deep Time Assessment for the Clive DU PA
30 March 2020 iv
CONTENTS
TABLES ........................................................................................................................................ vi
1.0 Deep Time Model Distribution Summary ............................................................................. 1
2.0 Introduction ............................................................................................................................ 3
3.0 Deep Time Model Overview ................................................................................................. 3
4.0 Background on Pluvial Lake Formation in the Bonneville Basin ......................................... 7
4.1 Long-term Climate ........................................................................................................... 7
4.2 Prehistorical Deep Lake Cycles ..................................................................................... 11
4.3 Shallow and Intermediate Lake Cycles .......................................................................... 15
4.4 Sedimentation ................................................................................................................. 18
4.5 Eolian Deposition ........................................................................................................... 19
5.0 Conceptual Overview of Modeling Future Lake Cycles ..................................................... 20
5.1 Introduction .................................................................................................................... 20
5.2 Future Scenarios ............................................................................................................. 20
6.0 A Heuristic Model for Relating Deep Lakes to Climate Cycles from Ice Core
Temperature ......................................................................................................................... 22
6.1 Introduction .................................................................................................................... 22
6.2 Glaciation ....................................................................................................................... 23
6.3 Precipitation ................................................................................................................... 25
6.4 Evaporation .................................................................................................................... 25
6.5 Simulations ..................................................................................................................... 27
7.0 Deep Time Modeling Approach .......................................................................................... 29
7.1 Introduction .................................................................................................................... 29
7.2 Deep Lake Characteristics .............................................................................................. 29
7.3 Intermediate Lake Characteristics .................................................................................. 31
7.4 Sedimentation Rates ....................................................................................................... 32
7.5 Eolian Depositional Parameters ..................................................................................... 39
7.5.1 Field Studies ............................................................................................................. 39
7.5.2 Probability Distributions for the Depth and Age of Eolian Deposition ................... 39
7.6 Destruction of the Federal DU Cell ................................................................................ 42
7.7 Radionuclide Concentration in DU Waste ..................................................................... 47
7.8 Radionuclide Concentration in Sediment ....................................................................... 47
7.9 Radioactivity in Lake Water .......................................................................................... 48
7.10 Modeling of 222Rn Flux .................................................................................................. 50
7.10.1 Waste and Sediment Water Content ......................................................................... 51
7.11 Human Health Exposure and Dose Assessment ............................................................ 52
8.0 References ............................................................................................................................ 52
Appendix A ................................................................................................................................... 58
Appendix B ................................................................................................................................... 60

Deep Time Assessment for the Clive DU PA
30 March 2020 v
FIGURES

Figure 1. Comparison of delta deuterium (black line) from the European Project for Ice
Coring in Antarctica (EPICA) Dome C ice core and benthic (marine) oxygen-18
record (blue line) for the past 900 ky [from Jouzel et al. (2007)] ................................. 5
Figure 2. Benthic oxygen isotope record for 700 ka (from Lisiecki and Raymo, 2005) .............. 14
Figure 3. Temperature deviations for the last 810 k (from Jouzel et al., 2007) ............................ 23
Figure 4. Glacial change as a function of temperature for the coarse conceptual model ............. 26
Figure 5. Two example simulated lake elevations as a function of time, with Clive facility
elevation represented by green line ............................................................................. 28
Figure 6. Probability density functions for the start and end times for a deep lake, in yr prior
to the 100-ky mark and yr after the 100-ky mark, respectively. ................................. 31
Figure 7. Probability density function for sedimentation rate for the deep-water phase of a
deep lake ..................................................................................................................... 34
Figure 8. Historical elevations of the Great Salt Lake .................................................................. 35
Figure 9. Simulated transgressions of a deep lake including short-term variations in lake
elevations .................................................................................................................... 37
Figure 10. Probability density function for the total sediment thickness associated with an
intermediate lake (or the transgressive of regressive phase of a deep lake) ............... 38
Figure 11. Eolian deposition rate results for 1,000 realizations (m/yr). ....................................... 44
Figure 12. Probability density function for the area over which the waste embankment is
dispersed upon destruction .......................................................................................... 46

Deep Time Assessment for the Clive DU PA
30 March 2020 vi
TABLES
Table 1. Summary of distributions for the Deep Time Model container ........................................ 1
Table 2. Lake cycles in the Bonneville basin during the last 700 ky1 .......................................... 13
Table 3. Lake cycles and sediment thickness from Clive pit wall interpretation (C. G. Oviatt,
personal communication) 1 ......................................................................................... 18
Table 4. Thickness measurements from field studies of eolian silt near Clive ............................. 40
Deep Time Assessment for the Clive DU PA
30 March 2020 1
1.0 Deep Time Model Distribution Summary
A summary of parameter values used in the Deep Time Model component of the Clive DU PA
Model is provided in Table 1. For the purpose of this white paper, deep time refers to the period
between 10 thousand yr to 2.1 million yr; approximately when the progeny of 238U reach secular
equilibrium with 238U and peak activity.
For distributions, the following notation is used:
• N( μ, σ, [min, max] ) represents a normal distribution with mean μ and standard deviation
σ, and optional min and max if truncation is needed,
• LN( GM, GSD, [min, max] ) represents a log-normal distribution with geometric mean
GM and geometric standard deviation GSD, and optional min and max if truncation is
needed,
• U( [min, max] ) represents a uniform distribution with minimum min, and maximum max,
• Beta( μ, σ, [min, max] ) represents a generalized beta distribution with mean μ, standard
deviation σ, minimum min, and maximum max, and
• Gamma( μ, σ ) represents a gamma distribution with mean μ and standard deviation σ.

Table 1. Summary of distributions for the Deep Time Model container
Model Parameter Value or Distribution Units Reference
DepthEolianDeposition
long-term eolian deposition
depths
N(μ=72.7, σ=5 min=Small,
max=Porosity_Unit4)
cm Section 7.5
AgeEolianDeposition
long-term eolian deposition
ages
Beta(μ=13614,
σ=263.3,min=13000,max=15000)
yr Section 7.5
EolianCorrelationFactor
correlation between eolian
deposition depth and Eolian
deposition age
U(0.5,1.0) — Section 7.5
LakeDelayTime
time at which the first
intermediate lake occurs
50,000 yr Section 4.1

IntermediateLakeDuration
length of time that Clive is
covered by an intermediate
lake event
LN(GM=500, GSD=1.5,min=0,
max=2500)
yr Section 7.3
Deep Time Assessment for the Clive DU PA
30 March 2020 2
Model Parameter Value or Distribution Units Reference
IntermediateLakeSedimentA
mount
sediment thickness for each
intermediate lake event
LN(GM=2.82, GSD=1.71) m Section 7.4
DeepLakeStart
Lake start time before the
end of the 100,000-year
glacial cycle
LN(GM=14000, GSD=1.2,min=0,
max=50000 )
yr Section 7.2
DeepLakeEnd
Lake end time after the most
recent cold peak within the
100,000-year climate cycle
LN(GM=6000, GSD=1.2,min=0,
max=50000)
yr Section 7.2
DeepLakeSedimentationRate
rate of the sedimentation
during the open water phase
of a deep lake
LN(GM=1.2E-4, GSD=1.2) m/yr Section 7.4
SiteDispersalArea
the area across which the
destroyed site is spread
Gamma(mean=24.2332,
stdev=11.43731)
Km2 Section 7.6

IntermediateLakeDepth
depth of an intermediate lake
at Clive
Beta(μ=30, σ=18,min=0, max=100) m Section 7.9
DeepLakeDepth
depth of a deep lake at Clive
Beta(μ=150, σ=20,min=100,
max=200)
m Section 7.9
TotalEmbankmentVolume
original total volume of the
embankment
3,231,556 m3 Section 7.8
DiffusionLength
Diffusion length for the deep
time sediments
N(μ=0.5, σ=0.16 min=0.0,
max=Large)
m Section 7.9
external_DCF_modifiers See table in ES external DCF
modifiers.xlsx Excel file
— Section 7.11
Deep Time Assessment for the Clive DU PA
30 March 2020 3
Model Parameter Value or Distribution Units Reference
DCFs and parameters within
the DCFs container
See Dose Assessment white paper
for parameter values and reference
— See Dose
Assessment
white paper
Rn_flux_ratio
ratio of Rn-222 flux at
different sediment thickness
to flux with no overlaying
cover
Thickness
0.001
0.5
1.0
1.5
2.0
3.0
4.0
5.0
6.0
7.5
Rn-222 flux
1.00000
4.392E-1
1.972E-1
8.750E-2
4.000E-2
8.140E-3
1.656E-3
3.371E-4
6.881E-5
1.00E-30
— Section 7.10
* “Large” is a very large number, and “Small” is a very small number, as defined by GoldSim.
2.0 Introduction
This white paper provides documentation of the development of parameter values and
distributions used for modeling scenarios of the fate of Federal DU Cell waste for the Clive DU
PA model in deep time. Data sources are identified and the rationale applied for developing
distributions is described. The intent of this white paper is to describe the characteristics and
potential sedimentation processes occurring during the 2.1 million yr interval of the Deep Time
Model and the subsequent effects of these processes on waste disposed at the Clive site.
3.0 Deep Time Model Overview
There are two major components of the Clive DU PA Model. The first component addresses
quantitative contaminant fate and transport and subsequent dose assessment for 10,000 yr
(10 ky). That modeling is based upon projections of current societal conditions into the future
and assumes no substantial change in climatic conditions. The second component addresses
“deep time” scenario calculations from 10 ky until the time of peak radioactivity. For this PA,
peak radioactivity associated with the ingrowth of progeny from 238U occurs at about
2.1 million yr in the future (2.1 My).
The initial Deep Time Models for this site, the Deep Time container of the Clive DU PA v1.0 and
v1.2 Models and the Deep Time Supplemental Analysis (DTSA) Model (Clive DU PA Model
vDTSA.gsm), addressed DU waste stored above and below the surrounding grade in an
embankment. The DTSA model is a standalone model, not directly linked to the PA model. The
models assume destruction of the embankment via wave action from a possible return of a lake
to the Clive area under future glacial period conditions, and subsequent dispersal of waste. With
a review of this modeling, a decision was made by the State of Utah to require EnergySolutions
to dispose of all DU waste below the surrounding grade, and thus no waste per se would be
exposed or dispersed upon return of a lake (SC&A, 2015). The only possible mechanisms for
dissolution and dispersal of radionuclides would then be associated with radon emanation into
Deep Time Assessment for the Clive DU PA
30 March 2020 4
the embankment materials and diffusion of dissolved radionuclides upwards. The current PA
model (v1.5) retains this assumption, and the 10 ky model and the revised Deep Time Model are
now integrated. Additional factors such as eolian (i.e., wind-borne) deposition are also now
included. Below is a brief summary of the current conceptual site model (CSM) for the Deep
Time Model. These terms and details are explained and discussed further in this report.
• Time scale of interest: 10 ky to 2.1 My post-closure.
• Waste placement: All DU waste is buried below grade in a covered embankment.
• Pluvial (i.e., caused by increased precipitation) lake occurrence: This is driven largely by
glacial cycles of cooler and wetter climate conditions. “Deep” lakes occur no more than
once per 100-ky cycle. “Intermediate” lakes can occur independent of a deep lake, or as
transitory events during the transgressive (rising lake) or regressive (falling lake) phases
of a deep lake. While intermediate lakes can occur during interpluvial cycles, they are
unlikely to occur during the current interpluvial cycle because of the unusually high
atmospheric C02 content. An assumption of the Deep Time Model is that an intermediate
lake will not rise to the elevation of the Clive site without a return to pluvial conditions.
• Destruction of embankment: The embankment will be eroded to the level of the former
Lake Bonneville surface (current grade at the time of the first lake return) by wave action
and sediment churning during the first return of a deep or intermediate lake.
Radionuclides present in the above-grade part of the embankment (as a result of transport
processes) will be dispersed and mixed with sediments during active lake erosion across
the area of the lake. The waste itself will not be exposed.
• Release of radionuclides: Radionuclides in the dispersed sediments will be released to
lake water upon destruction of the embankment via diffusion. Radon is allowed to diffuse
upward through the sediment when a lake is not present.
• Fate of radionuclides: Radionuclides will partition between water and sediments
according to their solubility and sorption properties. Insoluble DU will be buried by lake
sediments. Radionuclides settle out in sediments after lakes recede.
• Sedimentation: Eolian deposition occurs while lake levels are below the Clive site and
these deposits are incorporated within (reworked) lake sediments after the first lake
returns. The Deep Time model tracks eolian sedimentation rates only for the first future
glacial cycle (see section 4.5). Clastic sedimentation will dominate during times of active
intermediate lakes with transitions to carbonate precipitation (marl, laminated marl) when
the lake levels increase in elevation and surface waves no longer affect lake-bottom
sediments (Clive elevation). Short-term changes in lake levels may result in transient
deposition of eolian and marl sediments. These deposits are partly to completely
reworked by wave action during intermediate lake activity and are aggregated in the
Deep Time model as intermediate lake deposition. The basic Deep Time Model
scenario involves projecting the future environment conditions upon the Pleistocene and
Holocene epochs (Quaternary period) record of climate variations and lake formation in
the Bonneville Basin. The conceptual model of the past environment is based upon
scientific records (sediment borehole logs, ice cores, deep ocean cores) of the past eight
glacial/climate cycles that have lasted approximately 100 ky each. The model considers
Deep Time Assessment for the Clive DU PA
30 March 2020 5
cycles from the beginning of an interglacial period onwards. In the past 100-ky cycles,
after an interglacial period, the average temperature drops and average precipitation
increases throughout the glacial cycle, until the relatively cold period (typically an ”ice
age”) ends and the next interglacial period begins (Figure 1). The Earth is currently in an
interglacial period. The first 10 ky of the Clive DU PA Model is projected under
interglacial conditions, and the Deep Time Model calculations include an evaluation of
the effect on the Federal DU Cell of future 100-ky glacial cycles for the next 2.1 My. The
critical aspect of a future glacial period is the potential return of a pluvial lake to the
elevation of the Clive site with accompanying lakeshore wave activity that would destroy
the Federal DU embankment. Thus, the objective of the Deep Time Model is to assess the
potential impact of glacial period pluvial lake events upon and associated with
radionuclide release/dispersal from the Federal DU Cell from 10 ky through 2.1 My post-
closure.
Figure 1. Comparison of delta deuterium (black line) from the European Project for Ice
Coring in Antarctica (EPICA) Dome C ice core and benthic (marine) oxygen-18
record (blue line) for the past 900 ky [from Jouzel et al. (2007)]
The approximate historical 100-ky glacial cycles are depicted in Figure 1. The current
interglacial period is shown on the left edge of the figure. The last ice age finished between
12,000 and 20,000 yr ago (12 ka and 20 ka, indicated as “ky B.P.” in the figure). In the last
glacial maximum (represented as a trough on the far-left side of Figure 1), the major Western
United States water body Lake Bonneville, which covered much of Utah, reached its maximum
extent. Antarctic ice core data as well as benthic marine isotope data (described below) show
similar patterns for the past 800 ky. These 100-ky cycles are used as the basis for modeling the
return and recurrence of lake events in the Clive area.
The Deep Time Model should be regarded as conceptual and stylized and is not intended as a
prediction of expected future conditions at the Clive site. The intent is to estimate potential future
Deep Time Assessment for the Clive DU PA
30 March 2020 6
radionuclide releases from the remains of the Federal DU Cell, rather than to provide a
quantitative, temporally-specific prediction of future conditions, or an assessment of exposure or
doses to possible humans. Doses to potential human receptors and the presence and
characteristics of human populations in the Clive area during this time period are entirely
speculative. When a lake inundates the waste site, there will be no receptors at that location.
Additionally, calculation of radiological dose to human at times beyond 10 ky is not required by
Utah state regulations (Utah 2015). Instead, these regulations specify a “qualitative” assessment
with radionuclide release simulations for this period. Organizations such as the International
Atomic Energy Agency (IAEA 2012) have indicated that calculating doses beyond a few hundred
yr is not defensible; thus, quantitative dose assessment, particularly subsequent to lake events
related to the interglacial cycle, is insupportable from scientific and technical perspectives.
However, if the Deep Time Model results such as radon flux are considered in the context of
gauging system performance, such results may provide limited insight into the behavior of the
disposal system in deep time. Based on potential future radon fluxes, a rancher dose was
calculated in the Deep Time Model to provide a context for the radon flux results.
A “deep” lake is defined here as a large glacial-period lake on the scale of the prehistoric Lake
Bonneville (present in the area from about 32 ka to 14 ka). Such lakes have occurred in several
of the past 100-ky climate cycles. An “intermediate” lake is defined as a lake that reaches the
elevation of Clive (described further below). These lakes are assumed to occur in the
transgressive and regressive phases of a deep lake, but evidence of such lakes is difficult to
identify and interpret because lake deposits are reworked during their transgressive and
regressive lake phases. It is assumed that the first deep or intermediate lake that reaches the
elevation of Clive will destroy and disperse the Federal DU Cell embankment via wave action.
This dispersal mixes radionuclides with lake sediments. The characteristics of these mixed
sediments are dependent upon the duration and intensity of the lakeshore processes (e.g., wave
sediment churning, and formation of spits and bars from longshore drift).
Wave action associated with transgressing and regressing intermediate lakes will rework the
lake-sediment interface to a depth that is controlled by the dynamics of the wave action.
Evidence of wave action and sedimentary processes for past levels of Lake Bonneville is
preserved in the area’s sedimentary and geomorphic features. This evidence includes
paleoshorelines, fan and river deltas, wave-cut cliffs, bayhead barriers and spits (Sack, 1999;
Schofield et al., 2004; Nelson, 2012). The most relevant lake features from the geologic record
are paleoshorelines. Schofield et al. (2004) divide Lake Bonneville shorelines into erosion-
dominated and deposition-dominated. The elevation difference between shoreline bench deposits
and shoreline fronts from these studies provides a time-integrated analog for the dynamics of
wave action during shoreline transgressions and regressions (see Figure 3 in Schofield et al.
2004). These elevation differences are about 90 cm for erosion-dominated shorelines and 40 to
65 cm for deposition-dominated shorelines. Thus, the process of wave action is assumed to
remove approximately the same thickness of sediment (0.5 to 1 m) as the residual embankment
thickness (<1 m).
Any periods in which a lake does not exist are assumed to experience eolian (i.e., wind-borne)
deposition. Although some removal of embankment materials and sediment via wave action is
expected, this is not modeled explicitly. Instead, these effects are assumed to be relatively small
compared to eolian and lake deposition effects, and are assumed to have roughly a net zero effect
on overall sedimentation before and after the return of an intermediate or deep lake (remaining
Deep Time Assessment for the Clive DU PA
30 March 2020 7
embankment thickness is about 0.5 m and removal depth is about 0.5 m). The current model thus
explicitly considers eolian and lake deposition only as contributors to sedimentation thickness.
Other major geologic or climatic events could also occur in the next 2.1 My. Events such as
major meteorite impacts, and volcanic activity such as eruptions associated with the Yellowstone
Caldera could also be considered. Such future catastrophic events are often screened from
consideration in PAs on the basis of a low probability of occurrence and/or limited
consequences. In this case, a major meteorite impact and a future volcanic eruption at
Yellowstone were not screened. Instead, the impacts of these events are considered to be so
catastrophic on a global scale that their effects would far outweigh any potential radionuclide
releases from the Federal DU Cell. The same applies to major climate changes outside of those
associated with glacial cycles, although impacts of anthropogenic climate change on future lake
events are partially considered here.
4.0 Background on Pluvial Lake Formation in the Bonneville
Basin
4.1 Long-term Climate
Large-scale climatic fluctuations over the last 2.6 My (the Quaternary Period, the current and
most recent of the three periods of the Cenozoic as defined in the geologic time scale;
http://www.geosociety.org/science/timescale/) have been studied extensively in order to
understand the mechanisms underlying those changes (Hays et al., 1976, Berger, 1988, Paillard,
2001, Berger and Loutre, 2002). These climatic signals have been observed in marine sediments
(Lisiekcki and Raymo, 2005), land records (Oviatt et al., 1999), and ice cores (Jouzel et al.,
2007). These large-scale fluctuations in climate have resulted in glacial and interglacial cycles,
which have waxed and waned throughout the Quaternary Period. The causes of the onset of the
last major northern hemisphere glacial cycles 2.6 million yr ago (Ma) remain uncertain, but
several studies suggest that the closing of the Isthmus of Panama caused a marked reorganization
of ocean circulation patterns that resulted in continental glaciation (Haug and Tiedemann, 1998,
Driscoll and Haug, 1998). Future glacial events are likely to be caused by a combination of the
Earth’s orbital parameters as well as increases in freshwater inputs to the world’s oceans
resulting in a disruption to oceanic thermohaline circulation (Driscoll and Haug, 1998).
Changes in the periodicity of glacial cycles have been linked to variations in Earth’s orbit around
the Sun. These variations were described by the Serbian scientist Milutin Milankovitch in the
early 1900s, and are based upon changes that occur due to the eccentricity (i.e., orbital shape) of
Earth’s orbit every 100-ky, the obliquity (i.e., axial tilt) of Earth’s axis every 41 ky, and the
precession of the equinoxes (or solstices) (i.e., wobbling of the Earth on its axis) every 21 ky
(Berger, 1988). For the first 2 My of the Pleistocene (the first major Epoch of the Quaternary
Period), Northern Hemispheric glacial cycles occurred every 41 ky, while the last million yr have
indicated glacial cycles occurring every 100-ky, with strong cyclicity in solar radiation every 23
ky (Berger and Loutre, 2002; Paillard, 2006). The shift from shorter to longer cycles is one of the
greatest uncertainties associated with utilizing the Milankovitch orbital theory alone to explain
the onset of glacial cycles (Paillard, 2006).
Deep Time Assessment for the Clive DU PA
30 March 2020 8
Hays et al. (1976), who analyzed changes in the isotopic oxygen (δ18O) composition of deep-sea
sediment cores, suggest that major climatic changes have followed both the variations in
obliquity and precession through their impact on planetary insolation (i.e., the measure of solar
radiation energy received on a given surface area in a given time). In its most common form,
oxygen is composed of eight protons and eight neutrons (giving it an atomic weight of 16). This
is known as ”light” oxygen because a small fraction of oxygen atoms have two extra neutrons
and a resulting atomic weight of 18 (18O), which is then known as ”heavy” oxygen. 18O is a rare
form and is found in only about 1 in 500 atoms of oxygen. The ratio of these two oxygen
isotopes has changed over the ages and these changes are a proxy to changing climate that have
been used in both ice cores from glaciers and ice caps, and cores of deep sea sediments. Thus,
variations in δ18O reflect changes in oceanic isotopic composition caused by the waxing and
waning of Northern Hemispheric ice sheets, and are thus used as a proxy for previous changes in
climate (cf. Figure 1).
Slightly different external forcing and internal feedback mechanisms can lead to a wide range of
responses in terms of the causes of glacial-interglacial cycles. The collection of longer ice core
records, such as the European Project for Ice Coring in Antarctica (EPICA) Dome C core located
in Antarctica, has highlighted the clear distinctions between different interglacial-glacial cycles
(Jouzel et al., 2007). Variation in climatic conditions appears to be sufficient that large
differences have occurred in each of the past several 100-ky cycles. At the present time, the
EPICA Dome C core is the longest (in duration) Antarctic ice core record available, covering the
last 800 ky (Jouzel et al., 2007).
There is considerable uncertainty associated with the number, timing, and recurrence interval of
glacially-influenced pluvial lakes in the Bonneville Basin. The 100-ky glacial cycle is roughly
correlated with the occurrence of deep lakes (Balch et al. 2005, Davis 1998), and there appear to
be smaller, millennial scale (“Dansgaard-Oeschger”) cycles within this larger cycle that are not
necessarily uniform (Madsen, 2000). For example, the Little Valley lake cycle peaked in
elevation at about 135 ka, the Cutler Dam lake cycle peaked about 65 ka, and the Bonneville lake
cycle peaked about 18 ka (Machette et al., 1992).
Many studies highlight the importance of past atmospheric composition in the dynamics of
glaciations across the Northern Hemisphere, in addition to orbital influences (Masson-Delmotte
et al., 2010; Clark et al., 2009; Paillard, 2006). Carbon dioxide (CO2) is a well-known influence
on the atmospheric “greenhouse effect” (i.e. warming due to trapping of solar heat), and is a
globally well-mixed gas in the atmosphere due to its long lifetime. Therefore, measurements of
this gas in Antarctic ice are globally representative and provide long-term data important for
understanding past climatic changes. Direct measurement of CO2 trapped in the EPICA Dome C
core indicates that atmospheric CO2 concentrations decreased during glacial periods due to
greater storage in the deep ocean, thereby causing cooler temperatures from a reduction of the
atmosphere’s greenhouse effect (EPICA, 2004). Warmer temperatures resulting from elevated
concentrations of CO2 released from the ocean contribute to further warming and could support
hypotheses of rapid wasting at the end of glacial events (Hays et al., 1976). Earlier interglacial
events (prior to 420 ka), however, are thought to have been cooler than the most recent
interglacial events (since 420 ka) (Masson-Delmotte et al., 2010).
The predicted effect of anthropogenic CO2 on glacial cycles has evolved over time. For example,
Berger and Loutre (2002) conducted simulations including orbital forcing (i.e., cycles largely
Deep Time Assessment for the Clive DU PA
30 March 2020 9
driven by orbital variables) coupled with insolation and CO2 variations over the next 100-ky.
Their results indicated that the current interglacial period could last another 50 ky with the next
glacial maximum occurring about 100 ky from now. The scientific record (cf. Figure 1) supports
this pattern of variability across the 100-ky glacial cycles. Berger and Loutre (2002) effectively
indicate that the current 100-ky cycle will not be as glacially intense as some of the previous
cycles. They also quote J. Murray Mitchell (Kukla et al, 1972, p. 436) who predicts that “the net
impact of human activities on climate of the future decades and centuries is quite likely to be one
of warming and therefore favorable to the perpetuation of the present interglacial.” Archer and
Ganopolski (2005) conducted simulations suggesting that the combination of relatively weak
orbital forcing and the long atmospheric lifetime of carbon released from fossil fuel and methane
hydrate deposits could prevent glaciation for the next 500 ky over two glacial cycle eccentricity
minima. Cochelin et al. (2006) used a paleoclimate model to simulate the next glacial inception
under orbital and atmospheric CO2 forcings. Three scenarios were modeled: an impending
glacial inception under low CO2 levels; a glacial inception in 50 ky for CO2 levels of 280 to
290 ppm; and no glacial inception for the next 100-ky for CO2 levels of 300 ppm or higher.
Tzedakis et al. (2012a) defined interglacial periods as episodes where global climate is
incompatible with the wide global extent of glaciers, and examined differences in such
interglacial durations over the last 800 ky. They noted that the onset of interglacials occurs within
2 ky of the boreal summer insolation maximum consistent with Milankovitch forcing, whereas
the end of interglacials does not occur consistently on a similar part of the insolation curve.
Reduction in summer insolation is identified as a primary trigger for glacial inception, but
multiple other feedbacks including atmospheric CO2 concentrations combine to modify the
timing of glacial inception. They further recognized two main groups for mean duration of
interglacials: 13±3 ky and 28±2 ky. In a related paper, Tzedakis et al. (2012b) suggest that the
end of the current interglacial could occur within the next 1,500 yr if atmospheric CO2
concentrations were reduced to about 240 ppm, but no glacial inception is projected to occur at
current atmospheric CO2 concentrations of 400 ppm, consistent with the conclusions of Archer
and Granopolski (2005).
Jansen et al. (2007) in Chapter 6 of the fourth assessment report of the Intergovernmental Panel
on Climate Change (IPCC) concluded that “it is very unlikely that the Earth would naturally
enter another ice age for at least 30 ky.” These conclusions were updated and strengthened in
Chapter 5 of the fifth IPCC assessment report (Masson-Delmotte et al., 2013).
“Since orbital forcing can be accurately calculated for the future…, efforts can be
made to predict the onset of the next glacial period. However, the glaciation
threshold depends not only on insolation but also on the atmospheric CO2
concentration… Models of different complexity have been used to investigate the
response to orbital forcing in the future for a range of atmospheric CO2 levels.
These results consistently show that a glacial inception is not expected to happen
within the next approximate 50 ky if either atmospheric CO2 concentration
remains above 300 ppm or cumulative carbon emissions exceed 1000 PgC
[petagrams of carbon—one petagram is 1015 g]. Only if atmospheric CO2 content
was [sic] below the pre-industrial level would a glaciation be possible under
present orbital configuration… Simulations with climate–carbon cycle models
show multi-millennial lifetime of the anthropogenic CO2 in the atmosphere…
Even for the lowest [emissions] scenario, atmospheric CO2 concentrations will
Deep Time Assessment for the Clive DU PA
30 March 2020 10
exceed 300 ppm until the year 3000. It is therefore virtually certain [i.e., a
greater-than 99% probability] that orbital forcing will not trigger a glacial
inception before the end of the next millennium.”
Current CO2 levels are approximately 400 ppm (http://co2now.org/images/stories/data/co2-mlo-
monthly-noaa-esrl.pdf), and have been steadily rising over the past 150 yr due to anthropogenic
sources. Preindustrial levels of CO2 were about 280 ppm, and CO2 levels associated with glacial
periods tend to be about 240 ppm (Tzedakis et al., 2012b). It would require major reductions in
CO2 emissions worldwide in order to return to preindustrial levels, and/or engineering solutions
(e.g., “scrubbing” on a massive scale) to remove CO2 from the atmosphere so that pre-industrial
levels are attained. However, the Clive DU PA Model projects current knowledge as a
fundamental assumption, therefore it is assumed here that no major anthropogenic CO2-reduction
interventions will occur, and that CO2 levels will continue to rise, or at least will not attain
preindustrial levels within the next 50 ky or longer.
The Bonneville basin watershed is large and integrates runoff from the eastern Great Basin and
transition region of the Colorado plateau. Long-term changes in evaporation and precipitation
over a large region are required to sustain rising of a lake to the Clive elevation. These conditions
may be expected to occur only with a return to glacial conditions given climate model forecasts
of increased aridity for the southwest United States. Climate change risks to municipal water
supplies in Utah have been modeled using watershed hydrology models that explore temporal
changes in average conditions (temperature, precipitation, runoff), and severe drought and water
supply scenarios (e.g., the Salt Lake City Department of Public Utilities, Bardsley et al., 2013).
These types of studies are both prudent and timely, but future projections of decade scale data
are highly uncertain. Indeed, projection of the global climate change model results to regional
models has been a developing topic in the succession of IPCC reports. Warming temperatures
associated with anthropogenic climate effects will likely have appreciable impacts on the
Southwestern United States, but current drought projections do not exceed paleoclimate records
of droughts over the last two millennia (Woodhouse et al., 2010; Morgan and Pomerleau, 2012).
Multi-model ensemble studies of future climate projections from 16 global climate models show
both decreases and increases in streamflow projections for the upper Colorado River Basin
(Harding et al., 2012). Cook et al. (2010) suggest caution in projecting climate model projections
for the arid Southwest.
Regardless, the weight of evidence reviewed and summarized in the sequence of IPCC reports is
considered to be substantive and persuasive, and this information supports the current modeling.
It is assumed that CO2 levels will continue to rise for the foreseeable future, or will not decrease
below pre-industrial levels. It is also assumed that the IPCC and associated climate projection
studies are valid, with a high degree of confidence, including their conclusion that the inception
of the next glacial period will probably not occur for at least 50 ky.
The following sub-sections present an overall background on past events in the Bonneville basin
that are driven by major shifts in climate, and that are presumed to occur in the distant future as
well.
Deep Time Assessment for the Clive DU PA
30 March 2020 11
4.2 Prehistorical Deep Lake Cycles
The Bonneville basin is the largest drainage basin in the Great Basin of the Western United
States. It is a hydrologically closed basin of over 134,000 km2, and has previously been occupied
by deep pluvial lakes. Pluvial lakes typically form when warm air from arid regions meets
chilled air from glaciers, creating cloudy, cool, rainy weather beyond the terminus of the glacier.
The increase in rainfall and moisture can fill the drainage basin, forming a lake. This kind of
climate was evident during the last glacial period in North America, and resulted in more
precipitation than evaporation, hence the rise of Lake Bonneville.
Numerous studies have investigated previous lake cycles in the Bonneville Basin. These include
studies of Lake Bonneville shoreline geomorphology (Currey et al., 1984), palynological (i.e.,
pollen) studies of deep boreholes (Davis, 1998), and studies of the geochemistry of deep-water
lacustrine depositional sequences (Eardley et al., 1973; Oviatt et al., 1999, Balch et al., 2005).
Analysis of these sediment cores can be used to help understand previous lake levels and
characteristics as well as establish the approximate age of previous lake cycles (e.g., Oviatt et al.,
1999).
Oviatt et al. (1999) analyzed hydrolysate amino acid enantiomers for aspartic acid, which is
abundant in ostracode protein. Ostracodes are small crustaceans that are useful indicators of
paleo-environments because of their widespread occurrence and because they are easily
preserved. Ostracodes are highly sensitive to water salinity and other limnologic changes.
Therefore, portions of sediment cores that contain ostracodes indicate fresher, and hence
probably deeper, lake conditions than the modern Great Salt Lake (Oviatt et al., 1999). An
important exception to the deep lake interpretation inferred from the presence of ostracodes is
wetland/spring discharge areas. While wetland sites contain abundant ostracodes, the sites can
generally be discriminated from deep lake carbonates by their lithology and stratigraphic position
of the former within transgressive and regressive lake cycles. To establish the approximate
timing of previous lake cycles, Oviatt et al. (1999) examined sediments from the Burmester
sediment core originally collected in the early 1970s near Burmester, Utah (Eardley et al., 1973).
Burmester is approximately 65 km east of Clive on the southern edge of the Great Salt Lake, at
an elevation of 1286 m above mean sea level (amsl). The Clive area has an elevation of
1307 m amsl. Oviatt has also collected sediment data from Knolls (to the west of Clive) and at
Clive itself (described further below). These data are largely consistent with the more recent
layers from Burmester, indicating similar sedimentation processes at work at least during these
time periods.
Data from the 307-m Burmester core suggest that a total of four deep-lake cycles occurred
during the past 780 ky (Table 2. ). Oviatt et al. (1999) found that the four lake cycles correlated
with marine δ18O stages 2 (Bonneville lake cycle: ~24 to 12 ky), 6 (Little Valley lake cycle:
~186 to 128 ky), 12 (Pokes Point lake cycle: ~478 to 423 ky), and 16 (Lava Creek lake cycle:
~659 to 620 ky).
Oxygen isotope stages are alternating warm and cool periods in the Earth’s paleoclimate which
are deduced from oxygen isotope data (Figure 2). These stages suggest that deep pluvial lake
formation in the Bonneville basin occurred in the past only during the most extensive Northern
Hemisphere glaciations. There are many interacting mechanisms that could control or ‘force’
glaciation and deglaciation. For example, Oviatt (1997) and Asmerom et al. (2010) suggested
Deep Time Assessment for the Clive DU PA
30 March 2020 12
that these extensive glaciations were controlled by the mean position of storm tracks throughout
the Pleistocene, which were in turn controlled by the size and shape of the ice sheets. Other
glaciation forcing mechanisms have been suggested. The review by Ruddiman (2006) suggests
that insolation changes due to orbital tilt and precession, greenhouse gas concentrations, changes
in Pacific Ocean circulation, and possibly other interacting mechanisms could contribute to
glaciation and deglaciation cycles in North America, and thus pluvial lake existence and size.
Lyle et al. (2012) suggests that lake levels in the Pleistocene western US were influenced by
stronger spring/summer precipitation fed by tropical Pacific air masses, rather than higher
numbers of westerly winter storms.
Balch et al (2005) conducted a more recent detailed study on ostracode fossils in Great Salt Lake
sediment (i.e., under the lake). Other fossil invertebrates were also used as paleoecological
indicators in this study. Both brine shrimp and brine fly fossils are indicators of hypersaline
environments because they have a much higher salinity tolerance than most other invertebrates.
This study’s findings were consistent with Oviatt et al.’s (1999) later cycles, but as the core was
not as deep the findings are not as useful for the present purpose as the Burmester data. The
Burmester core data are more germane to the present modeling effort because they represent a
relatively long time period in which to establish the occurrence of pluvial lakes in the region.
Deep Time Assessment for the Clive DU PA
30 March 2020 13
Table 2. Lake cycles in the Bonneville basin during the last 700 ky1
Lake Cycle Approximate
Age2
Maximum
Elevation
Lake Level Influences
Great Salt
Lake (current
level)
present 1284 m (4212 ft) in
1873
Interglacial climate; human intervention
Bonneville
(Gilbert
Episode)
11.6 ka 1295 m (4250 ft) Beginning of interglacial climate;
Bonneville
(Provo
Shoreline)
17.4 to 15.0 ka 1445 m (4740 ft) Glacial climate; new threshold at Red Rock
Pass, Idaho (natural dam collapse)
Bonneville
(Bonneville
Shoreline)
18.0 ka 1552 m (5090 ft) Glacial climate; threshold at Zenda near Red
Rock Pass, Idaho
Bonneville
Transgression
~30 to 18.0 ka Glacial climate
Bonneville
(Stansbury
Shoreline)
26 to 24 ka 1372 m (4500 ft) Glacial climate; transgressive phase of Lake
Bonneville
Cutler Dam ~80 to 40 ka <1380 m (<4525 ft) Glacial climate
Little Valley ~128 to 186 ka 1490 m (4887 ft) Glacial climate
Pokes Point 417 to 478 ka 1428 m (4684 ft) Glacial climate
Lava Creek ~620 to 659 ka 1420 m (4658 ft) Glacial climate
Elevations are not corrected for isostatic variations.
1 Note the various levels of the last major lake cycle, Lake Bonneville.
2 Approximate ages derived from Currey, et al. (1984) Link et al. (1999) and Oviatt et al. (1999). Bonneville cycle
approximate age presented as calibrated yr.
However, note that there is considerable uncertainty associated with the number, timing, and
recurrence interval of lakes in the Bonneville Basin. The 100-ky glacial cycle is roughly
correlated with the occurrence of deep lakes (Balch et al., 2005; Davis, 1998), and there appear
to be smaller, millennial-scale cycles within this larger cycle that are not necessarily uniform
(Machette et al., 1992; Madsen, 2000). It is likely that intermediate lakes have also occurred in
each glacial period, but their deposits have been reworked by wave action during subsequent
lake cycles and intermixed with local sediments.s. Sediment mixing that occurs during lake
formation can also mask the existence of previous intermediate lakes. Thus, it is impossible to
have complete confidence in historical lake formation characteristics and formation.
Lake Bonneville is the last major deep lake cycle that took place in the Bonneville basin and is
widely described in the literature (Hart et al., 2004; Oviatt and Nash, 1989; Oviatt et al., 1994a,
1999). Lake Bonneville was a pluvial lake that began forming approximately 28 to 30 ka,
forming various shorelines throughout its existence and covering over 51,000 km2 at its highest
level (Matsubara and Howard, 2009).
Deep Time Assessment for the Clive DU PA
30 March 2020 14
Figure 2. Benthic oxygen isotope record for 700 ka (from Lisiecki and Raymo, 2005)1
Most studies indicate that the high-stand (i.e., the highest level reached) of the lake at the Zenda
threshold (1,552 m amsl), located north of Red Rock Pass, occurred approximately 18.0 ka. The
high-stand of the lake was followed by an abrupt drop in lake level due to the catastrophic failure
(landslide) of a natural dam composed of unconsolidated material at approximately 17.4 ka. As a
result of this flood, the lake dropped to a level of 1,430 m amsl, called the Provo level (Miller et
al. 2013). The Provo level is the maximum level that any future deep lake is likely to reach
without major regional tectonic changes (Currey et al., 1984; Oviatt et al., 1999).
A more recent study (Miller et al., 2013), using radiocarbon dating for Provo shoreline gastropod
deposits, estimates that the dam collapse and Bonneville flood event occurred between 18.0 and
18.5 ka, and therefore the high-stand may have occurred earlier. However, Miller et al. (2013)
indicate that “uncertainties in [gastropod] shell ages may be as large as thousands of yr, and the
major shorelines of Lake Bonneville and the Bonneville flood require more work to establish a
reliable chronology.” The lake regressed rapidly during the last deglaciation, then increased
again to form the Gilbert episode ~ 11.6 ka, which remained below the elevation of Clive
(Oviatt, 2014). The lake then receded to levels of the current Great Salt Lake at approximately
10 ka for the remainder of the Holocene Epoch.

1Red (warm periods) and blue (cool periods) values correspond to marine isotope stages based upon Lisiecki and
Raymo (2005). Lake stages identified by Oviatt et al. (1999) are also included in blue text.
Deep Time Assessment for the Clive DU PA
30 March 2020 15
Glacial cycles can be discerned in Figure 2 by considering each cycle from the beginning of the
interglacial period and ending each cycle at the peaks that correspond to deep lake occurrence.
Using this approach, the current glacial cycle started around 12 ka, Lake Bonneville occurred at
the end of the last complete cycle, and Cutler Dam occurred in the middle of the last 100-ky
cycle. The previous 100-ky cycle resulted in the Little Valley lake. The Pokes Point lake
occurred five cycles ago, and The Lava Creek lake seven cycles ago. These deep lakes have been
identified in sediment cores and in shorelines around the Bonneville Basin. However, it is likely
that many more shallow lakes have also occurred in each glacial period, but the shorelines have
been destroyed by subsequent deeper lakes.
The types of sediment resulting from the formation and long-term presence of lakes in the
Bonneville basin are diverse and can be divided into three components (Schnurrenberger et al.,
2003): 1) chemical sediment (inorganic materials formed within the lake), 2) biogenic sediment
(fossil remains of former living organisms), and 3) terrigenous or clastic sediments (grains and
clasts that are mechanically and chemically fragmented from existing material, transported and
deposited by sedimentary processes). A fourth type of associated sediment, not formed by lakes,
includes eolian deposits consisting of windblown grains of sand, silt or dust (i.e., loess). These
deposits can locally be interbedded with lake sediments and may be affected by soil-forming
processes (i.e., pedogenesis) during prolonged periods of subaerial exposure. All four types of
sediments can be intermixed by lake-wave action or bioturbation, and deposited as clastic
sediments during transgressive and regressive lake cycles.
There is considerable uncertainty in the number of lakes, particularly lakes of intermediate size
that might have existed in the Bonneville basin. However, the main focus of the Deep Time
Model is to evaluate the presence of lakes that inundate Clive in future glacial cycles, and to
approximately match the net sedimentation patterns of the past glacial cycles. In order to inform
the potential for radionuclide releases, the high-level, conceptual modeling of lake cycles that
was conducted here did not assume any particular mechanism of glaciation and deglaciation. For
example, the modeling simply assumed a 100-ky cycle, regardless of the mechanism. The model
addresses deep lakes by allowing them to return in some glacial cycles, and intermediate lakes by
allowing them to occur both independently and as part of the transgressive and regressive phases
of deep lake cycles.
4.3 Shallow and Intermediate Lake Cycles
The current Great Salt Lake is an example of a shallow lake, as is the reinterpreted Gilbert
episode lake that has been shown to have not reached the elevation of the Clive site (Oviatt,
2014, contrasted with the map of Curry, (1982). The specific depths of lakes are not important in
the Deep Time Model, aside from calculations with regard to lake chemistry and dominant
processes of sedimentation. Under current climate conditions, only shallow lakes will occur.
Under future climate conditions, some glacial cycles will produce deep lakes in the Bonneville
basin, and intermediate lakes will occur during the transgressive and regressive phases of deep
lakes, or during glacial cycles that do not produce deep lakes. The approximate timing of the
return of the first intermediate lake is important in the Deep Time Model, because it is assumed
that the Federal DU Cell embankment is destroyed upon the occurrence of the first intermediate
lake.
Deep Time Assessment for the Clive DU PA
30 March 2020 16
A key assumption of the Deep Time Model, based upon core sediment studies and estimated
deep lake sedimentation rates, is that the net depositional rate of deep lakes is lower than the
sediment depositional rate for intermediate lakes. The conceptual basis for this assumption is that
sedimentation rates are dependent on basin location, presence or absence of fluvial deposition,
wave dynamics, availability of local sediment sources, slope, water chemistry and biological
activity. Biogenic carbonate deposition is likely to occur under a wide range of lake conditions,
but the ratio of carbonate deposition to clastic sedimentation will increase as the lake deepens
because of the reduction in sedimentary influx with increased distance from shoreline processes
and decreased wave activity.
There are recognized trends in carbonate mineralogy that can be correlated with lake volume and
indirectly lake depth (cf., Oviatt, 2002; Oviatt et al., 1994b; Benson et al., 2011). The transitions
from low-magnesium calcite to high-magnesium calcite to aragonite generally reflect increasing
lake salinity and increasing magnesium concentration, which occurs with decreasing lake
volume. Similarly, for a hydrologically closed pluvial lake system, the relative concentration of
total inorganic carbon should typically decrease as lake size increases. The δ18O of deposited
carbonate can be correlated with rising lake levels because of the interplay between the δ18O
value of river discharge entering a lake and the δ18O value of water vapor exiting the system via
evaporation (Benson, et al., 2011). The mineralogy and isotopic composition of carbonate
composition can be obtained from sediment cores. Interpretation of the data is complicated by
multiple processes, including: local groundwater discharge; introduction of glacial rock flour;
and, reworking of lake sediments during transgressive and regressive lake cycles.
Intermediate lake events are known to have occurred in the Clive area. These are documented in
Table 3 (C.G. Oviatt, Professor of Geology, Kansas State University, personal communication
December 2010, January 2011, and various email communication referred to as “C.G. Oviatt,
personal communication.”). These events are evident from a pit wall interpretation at the Clive
site (Appendix A; C.G. Oviatt, unpublished data) as well as at the ostracode and snail record
present in the Knolls sediment core (12 km west of Clive near the Bonneville Salt Flats;
Appendix B; C.G. Oviatt, unpublished data). In 1985 Lake Bonneville sediments were described
and measured in a pit wall during early development of the Clive disposal facility (Oviatt, 1985).
Lake sediments of intermediate and deep lakes were briefly studied during field studies at Clive
in the winter of 2014 (Neptune, 2015a). These studies confirmed:
1. The pit walls described by C.G. Oviatt in 1985 have been removed during quarrying
and/or disposal operations at the Clive site.
2. Soil-modified eolian silt (mean thickness 73 cm) was observed in the upper part of quarry
walls throughout the Clive site.
3. The stratigraphy of sediments of Lake Bonneville in modern quarry-wall exposures are
consistent with the 1985 pit wall interpretations (Appendix A).
4. Quarry-wall deposits of gravel and sand at the Clive site contain distinctive volcanic
clasts of black andesite derived from the Grayback Hills north of Clive. These deposits
are part of the transgressive Lake Bonneville sedimentary sequence.
5. Pre-Lake Bonneville lake sediments with interbedded-soils and eolian sands were
observed in one deep quarry wall at the north end of the Clive site. These sediments are
Deep Time Assessment for the Clive DU PA
30 March 2020 17
consistent with the 1985 pit-wall interpretations but the new exposures were
insufficiently studied to established sediment correlations and the deposition chronology.
Stratigraphic correlations between 1985 studies and the new field studies (Neptune, 2015a) are
shown in Appendix A.
From the Clive pit wall interpretation, it is presumed that at least three intermediate lake cycles
occurred prior to the Bonneville cycle, although there is uncertainty associated with the age and
duration of the intermediate lake deposits. For example, these intermediate cycles could be part
of the transgressive phase (i.e., rising lake level) of the Lake Bonneville cycle (C.G. Oviatt,
personal communication). By analyzing the Knolls Core interpretation, the Little Valley cycle is
present at approximately 16.8 m from the top of the core. Given that the pit wall at Clive was
6.1 m high and does not capture the Little Valley cycle, it is possible that other smaller lake
cycles occurred in the Clive region in addition to the three intermediate lake events noted in
Table 3 (labeled as Pre-Bonneville Lacustrine Cycles).
There are few data to support the specific number of lakes that might have reached Clive or the
rate of sedimentation. There is also uncertainty associated with the particular times that these
cycles occur, as age dating (e.g., via radiocarbon dating) has not been performed in the Great Salt
Lake area. Most studies examine the degree of lake salinity using fossil records, and are
associated with cores that are in or near the Great Salt Lake. For example, Balch et al. (2005;
Fig. 6) estimated that there were six “saline/hypersaline” (i.e., shallow to intermediate) lake
cycles that occurred between the Lake Bonneville and Little Valley cycles, and approximately
that same number between the Little Valley cycle and the maximum age evaluated (300 ky).
However, this work does not inform the question of whether these lakes may have reached the
elevation of Clive, nor does similar work such as Davis (1998).
It is also possible that intermediate lakes could reach the elevation of Clive under unusual
conditions not necessarily associated with a return to a glacial cycle. The areal extent of lakes is
not only determined by elevation, but also by local topography, precipitation, temperature,
characteristics of inflow and outflow sources, and other factors. For instance, the Great Salt Lake
‘spilled’ over a 1285-m (4217-ft) amsl topographic barrier to the west of the present lake into the
area of the present Great Salt Desert as recently as the 1700s (Currey et al., 1984). This expanded
lake was about 15 m lower than the Clive site, and slightly higher than the current surface
elevation of the Great Salt Lake.
Precise dating of shorelines for the Great Salt Lake and variants is unfortunately lacking.
Radiocarbon dating for the Pyramid Lake area in Nevada indicates that this lake’s levels have
lowered approximately 35 m from the late Holocene Epoch (3.5 to 2.0 ky) to today (Briggs et al.,
2005). Radiocarbon and tree-ring dating to determine lake levels in the Carson Sink area in
Nevada indicates that lake elevations have risen approximately 20 m twice in the last 2000 yr
(Adams, 2003). It is not possible at this time to interpolate from these studies to the Great Salt
Lake area. Observational evidence shows that lake levels have not reached the elevation of the
Clive site during post-Bonneville time. The current high atmospheric C02 contents are expected
to continue for tens of thousands of years and will surpress a return to glacial conditions.
Therefor a lake is not expected to return to the Clive elevation under current climate conditions.

Deep Time Assessment for the Clive DU PA
30 March 2020 18
Table 3. Lake cycles and sediment thickness from Clive pit wall interpretation (C. G.
Oviatt, personal communication) 1
Lake Cycle Thickness of
Sediment Layer (m)
Depth Below
Ground Surface (m)
Soil-modified eolian silt1 1.05 1.05
Lake Bonneville Regressive Phase (reworked marl) 0.43 1.48
Lake Bonneville Open Water (white marl) 1.29 2.77
Lake Bonneville Transgressive (littoral facies) 0.76 3.53
Pre-Bonneville Lacustrine Cycle 3 (possible shallow lake) 0.71 4.24
Pre-Bonneville Lacustrine Cycle 2 (possible shallow lake) 0.62 4.86
Pre-Bonneville Lacustrine Cycle 1 (possible shallow lake) 1.14 6.00
1 The upper sedimentary sequence is no longer interpreted as a Gilbert lake phase (Oviatt, 2014). It is surficial
eolian deposits and soils based on recent field studies (Neptune, 2015a). The pit wall described in the 1985 studies
has been removed during quarrying and/or disposal operations.
4.4 Sedimentation
During all pluvial lake cycles, evaporites are deposited, as well as carbonates in the form of
tufas, marls, and mudstones. These sediments may contain varying components of shells (e.g. of
mollusks), and ostracodes (Hart et al., 2004). Terrigenous sedimentation however, accounts for
the major thickness of sediment observed throughout the Clive area sediment core record (C.G.
Oviatt, personal communication). The geomorphological evidence in the form of depositional
and erosional landforms produced at lake shorelines are carved into the landscape in the
Bonneville basin and provide examples of the erosional capacity of lake systems over long time
periods. Given the difficulty in separating chemical, biogenic, and terrigenous sediment deposits
in cores and natural exposures, the estimates reported below are assumed to be representative of
cumulative sedimentation from all causes during a lake event.
Brimhall and Merritt (1981) reviewed previous studies that analyzed sediment cores of Utah
Lake, a freshwater remnant of Lake Bonneville that formed at approximately 10 ka. They
suggest that up to 8.5 m of sediment has accumulated since the genesis of Utah Lake, implying
an average sedimentation rate of 0.85 mm/y or 850 mm/ky. Within the Bonneville basin as a
whole the major lake cycles resulted in substantial accumulations of sediment based upon the
depth of the cores analyzed (e.g., a 110 m core that corresponds to the past 780 ky, or four deep
lake cycles [Oviatt et al., 1999]). This accumulation averages about 140 mm/ky. Einsele and
Hinderer (1997) indicate that sediment accumulation in the Bonneville basin occurred at a rate of
120 mm/ky during the past 800 ky. The Knolls Core suggests that there has been 16.8 m of
sediment formed in the last glacial cycle, or nearly 170 mm/ky.
Interpretations of the Clive pit wall (C.G. Oviatt, unpublished data) indicate that the
sedimentation rate at the Clive site for the Lake Bonneville cycle is on the order of 2.75 m over a
17 to 19 ky time period (140 to 160 mm/ky). By contrast, shallow lacustrine cycles that occurred
prior to Lake Bonneville (but after the Little Valley cycle) indicate that the amount of sediment
deposited during each cycle is approximately 1/3 that of the Bonneville sediment deposited. The
timing of these shallow lake cycles is uncertain, however it can be approximated when
Deep Time Assessment for the Clive DU PA
30 March 2020 19
comparing the Clive pit wall interpretation to the Knolls Core (C.G. Oviatt, personal
communication). The Little Valley lake cycle is exhibited in the Knolls Core at a depth of
approximately 17 m, which is roughly 14 m deeper than the beginning of the transgressive phase
of the Bonneville lake cycle event noted on the Clive pit wall interpretation. Given the Little
Valley event occurred 150 ka, a sedimentation rate can be approximated for the depth between
this event and the transgressive phase of the Bonneville cycle of 110 mm/ky.
4.5 Eolian Deposition
Post-Lake Bonneville eolian deposition has occurred and will continue to occur at the Clive site
under current conditions. The expected primary mode of eolian deposition at the Clive site is
deposition of fine-grained silt from suspension fallout during episodic wind storms.
Exceptionally strong surface winds could potentially transport sand-sized material by saltation.
Evidence supporting these conclusions include (Neptune, 2015a):
• The presence of soil-modified eolian silt in the upper part of quarry-wall exposures at
multiple locations in the Clive site. The presence of these deposits requires continuing
eolian activity in the region and long-term maintenance of stable surfaces that promotes
preservation of the eolian deposits (suspension fallout) and soil-forming processes.
• Holocene dune deposits of eolian sand and silt in road cut exposures within 0.5 km of the
Clive site.
• Active gypsum sand dunes located approximately 13.5 km west of the Clive site.
• Active dune fields in the Lake Bonneville basin west and southwest of the Clive site
(Jewell and Nicoll, 2011).
Replicate measurements of the thickness of eolian deposits located in quarry wall exposures in
the Clive site are presented in Neptune (2015a), and are used below to develop input probability
distributions for the Deep Time Model. These deposits are relevant to expected future eolian
sedimentation before the first return of an intermediate or deep lake; with the rise of a future lake
to the elevation of the Clive site, wave activity will rework the eolian sediments and intermix
them with clastic lakeshore sediments.
The Deep Time model simulates eolian deposition only for the first future glacial cycle. The
expected long duration (> 50,000 yrs) prior to the first return of a lake to the Clive elevation will
result in eolian deposition exceeding 2 m, which exceeds the expected depth of sediment
reworking during lake activity. Subsequent glacial cycles will have atmospheric C02 contents
consistent with past Quaternary glacial cycles and the duration of interglacial cycles will revert
to past patterns (13,000 to 28,000 years; Tzedakis et al. 2012a). Eolian deposition during these
future interglacials will only partly exceed the depths of sediment reworking. Eolian
sedimentation of these interglaicials are represented in the sediment event thickness of
intermediate lakes.
Deep Time Assessment for the Clive DU PA
30 March 2020 20
5.0 Conceptual Overview of Modeling Future Lake Cycles
5.1 Introduction
There is a lack of data and peer-reviewed literature that would allow accurate and precise
prediction of the direct effects of future climate change on intermediate and deep lake formation
in the Bonneville basin. However, assuming no major changes from prehistorical climate cycles,
there is a possibility of another major lake cycle occurring in the Bonneville basin within the
next few million yr. Variations in the Earth’s orbital parameters in combination with increases in
inputs of freshwater into the oceans could lead to another major ice age and could alter long-term
climatic patterns in the Bonneville basin, resulting in deep lake formation. The Clive site might
be subjected to deep lake formation in the future, unless anthropogenic effects on atmospheric
CO2 concentrations cause major long-term changes in glacial cycles and climatic patterns.
An overview of the Deep-Time CSM was presented at the beginning of this report. The basic
intent of the Deep Time Model is to allow a lake to exist that is sufficiently large that the above-
ground embankment of the Federal DU Cell will be destroyed. It assumes that the sedimentation
rates for each glacial cycle are similar. The exact timing of the recurring lakes is not important,
the current 100-ky cycle excepted. The Deep Time Model allows the possibility of a deep lake to
return in each 100-ky cycle. It also allows intermediate lakes to recur at a frequency that allows
the assumed 100-ky sedimentation rate to be satisfied. The current 100-ky cycle is not modeled
explicitly. It is possible that the current interglacial period will last for at least another 50 ky due
to anthropogenic influences, which is unusually long compared to the interglacial period for
recent 100-ky ice age cycles.
5.2 Future Scenarios
Representative lake occurrence scenarios for deep time are described below. Note that there are
two components of the models used to represent these scenarios. The first is modeling lake
formation and dynamics, based upon the scientific record, literature, and expert opinion. The
second is modeling the fate of the Federal DU Cell.
The Great Salt Lake represents the current condition of a shallow lake in the Bonneville Basin.
Lakes such as this are likely to exist in all future climatic cycles, but will not reach the elevation
of the DU waste embankment at Clive and thus will not affect the waste embankment. For the PA
model, it is assumed that destruction of the waste embankment will result from the effects of
wave action from an intermediate or deep lake. This assumption separates intermediate and
shallow lakes. In this destruction scenario, the embankment material above grade is assumed to
disperse through a combination of wave action/churning and dissolution into the water column
above the waste dispersal area. Radionuclides present in the embankment dissolve into the lake
and eventually return to the lakebed via precipitation or evaporation as the lake regresses.
Some radionuclides in the water column will bind with carbonate ions and precipitate as
chemical and biogenic sediment, while radionculides bound to embankment materials will
remain within the clastic sediment as the lake eventually recedes. Wave action during the lake
recession is expected to rework and mix the chemical, biogenic and clastic lake deposits. The
combined complexity of processes affecting the compositional and sedimentary features of
lacustrine deposits (Fritz, 1996) and the mixing of lake sediments during regressive and
Deep Time Assessment for the Clive DU PA
30 March 2020 21
transgressive lake cycles makes it difficult to develop quantitative models of chemical and
physical processes affecting the distribution of waste radionuclides in lake waters and sediments.
In reality, waste radionuclides dissolved in lake waters will mix and be diluted by lake
circulation driven by prevailing winds and geostrophic balances (Jewel, 2010). Waste-sediment
mixes will be dispersed by wave action and longshore drift. Sediment concentrations will
decrease over time because the amount of waste does not change other than through decay and
ingrowth, whereas more sediment is added progressively over time. The model makes two
simplifying assumptions. First, sediments are thoroughly mixed throughout the total sediment
depth. In the Deep Time Model the sediment layers are considered to be a single mixing cell.
Second, diffusion can occur into the lake through this mixing cell, throughout the total sediment
depth. The mixing cell allows for radionuclides to diffuse through a short diffusion length,
relative to the depth of the mixing cell (sediment depth). Although sediment concentrations will
decrease over time and lake concentrations would be expected to do so concurrently, lake
concentrations do not necessarily decrease over time in the Deep Time Model because of the
single mixing cell.
The Deep Time Model assumes that changes in climate will continue to cycle in a similar fashion
to the climate cycles that have occurred since the onset of the Pleistocene Epoch. These changes
follow those observed in the marine oxygen isotope record (Figure 2). The record captures major
climate regime shifts on a global scale and is used in this scenario in conjunction with expert
opinion (C.G. Oviatt, personal communication) plus site-specific sediment core and Clive pit
wall information to determine the approximate periodicity of lake events. However, uncertainties
exist due to the limitations related to the quality of the sediment core data.
It is assumed that during the 100-ky climatic cycles intermediate or deep lakes will reach the
elevation of Clive. Although a definitive distinction is not made, lakes that reach the elevation of
Clive but do not develop into a deep lake are considered intermediate lakes. These intermediate
lakes are also assumed to be large enough (depth and geographic area)to support wave action that
will destroy the embankment. Intermediate lakes occur during the transgression and regression
of a deep lake, or may be the primary lake-form during a glacial cycle that does not produce a
deep lake, perhaps in conjunction with glacial cycles that are shorter and less severe than the
100-ky glacial cycles previously discussed.
In general, variation in lake elevation is assumed to be associated with all types of lakes. The
variation is due to local temporal changes in temperature, evaporation and precipitation. For
example, the Great Salt Lake has seen elevation changes of several meters in the past 30 to 40 yr.
The Great Salt Lake has also seen greater elevation changes in the past 10 ky, but in no cases
since post-Bonneville time has the Great Salt Lake reached the elevation of Clive (Oviatt, 2015).
Sedimentation is assumed to occur during these intermediate lake events at higher annual rates
than is assumed to occur for the open-water phase of deep lakes. This is based upon the pre-
Bonneville lacustrine cycles that are documented in Table 3 (Clive pit wall interpretation, see
Appendix A). The lake is assumed to recede after some period of time, at which point a shallow
lake (e.g., similar to the Great Salt Lake) will occupy Bonneville basin until the next
intermediate or deep lake cycle.
In the deep lake scenario, a deep lake forms throughout the Lake Bonneville basin in response to
major glaciation in North America and the Northern Hemisphere, following the ongoing 100-ky
Deep Time Assessment for the Clive DU PA
30 March 2020 22
glacial cycle. Increases in precipitation and decreases in evaporation over the long term, and
subsequent increases in discharge to the Bonneville basin via rivers that drain high mountains
along the eastern side of the basin have resulted in lakes that are more than 30 m deep and cover
an area similar to that of the most recent deep lake episode (e.g., Lake Bonneville, Provo
Shoreline). A similar extent of lake formation (geographic area, lake depth) is assumed to occur
in the future. Under such a scenario, the depth of a lake at the location of the Clive facility could
range over depths of many tens of meters. Resulting lake sedimentation at the Clive site will be
high rates of deposition of clastic sediments during intermediate lake events and much slower
rates of carbonate deposition during deep lake events. A key difference between the deep lake
scenario and the intermediate lake scenario is that both the transgressive and regressive phases of
lake formation are included with intermediate lakes. Transgressive and regressive phases of lake
formation can lead to brief periods of rising and falling water levels in both phases. These phases
of transgression and regression are also assumed to have higher sedimentation rates than the
deep-water phase. Upon the complete regression of a deep lake to levels below the elevation of
the Clive site, it is assumed that eolian deposition will dominate and only intermediate lakes will
form until the deep lake associated with the next climate cycle occurs.
6.0 A Heuristic Model for Relating Deep Lakes to Climate
Cycles from Ice Core Temperature
6.1 Introduction
In this section, a model is presented for estimating lake elevation that uses surface temperature
deviations from the EPICA Dome C ice core data (Jouzel et al., 2007), which is used to support
the modeling of future intermediate and deep lakes in the Deep Time Model. The model of lake
elevation is not intended to be highly accurate, but rather is aimed at capturing the major lake-
cycle features as shown in the studies conducted by Oviatt et al. (1999), Link et al. (1999), and
the sediment core and pit wall interpretations (C.G. Oviatt, personal communication). This model
is not used as a predictive model but rather to form a basis for the character and dynamics of lake
events in the Deep Time Model.
The deep-sea benthic δ18O record is in excellent agreement with the EPICA Dome C deuterium
measurements for the last ~810 ky (Jouzel et al., 2007). Temperature anomaly data for the past
810 ky were obtained from the World Data Center for Paleoclimatology, National Oceanic and
Atmospheric Administration/National Climate Data Center. These data are made available based
on calculations described in Jouzel et al. (2007), and are plotted in Figure 3. From the 810 ky of
data, the temperature deviations range from Tmin = –10ºC to Tmax = +5ºC. This range is used to
bound extreme events.
Water balance in the Bonneville basin is affected by many complex processes (temperature,
storm tracks, patterns of precipitation), so modeling water balance simply as a function of
temperature alone is not expected to produce precise results, but instead provides a coarse
representation. The conceptual model is based upon a water balance reservoir model of
precipitation versus evaporation. If precipitation outpaces evaporation, the lake elevation
increases. If evaporation outpaces precipitation, then the lake elevation decreases. Precipitation
and evaporation are affected directly by temperature, but long-term patterns of precipitation are
Deep Time Assessment for the Clive DU PA
30 March 2020 23
affected more greatly by the presence or absence of continental glaciation in North America.
Thus, glaciation is modeled first using a simple reservoir model depending on temperature.
6.2 Glaciation
The water balance model begins by constructing a “continental glacier”; an artificial construct
that represents a glacier large enough to affect precipitation levels in the Bonneville Basin. The
extent of glaciation in proximity to the Bonneville basin is assumed to be zero initially, which is
a reasonable approximation for the start time of 785 ka, a start time chosen because it
corresponds to a warmer climate phase (data from Jouzel, et al., 2007; see Figure 3). For each
time step of 500 yr, an increase in glacial magnitude is dependent on temperature deviation (ΔT)
as scaled in Jouzel (see Figure 3):

Figure 3. Temperature deviations for the last 810 k (from Jouzel et al., 2007)
Deep Time Assessment for the Clive DU PA
30 March 2020 24
𝐺𝑙𝑎𝑐𝑖𝑎𝑙!""#$#%&(Δ𝑇)=,
0 if Δ𝑇≥Δ𝑇'(!)1
𝑁'*
2(𝑒+!"⋅(./!#$%–./)–1)5 if Δ𝑇<Δ𝑇'(!) (1)
where
NGA is a normalizing constant:
𝑁'*=𝑒+!"⋅(./!#$%2./&'( )
RGA is a rate parameter (yr-1), and
TGMax is a threshold temperature (degrees Celsius).
As glaciation here is an artificial construct for modeling purposes, the units and scale of the
glacial “magnitude” are arbitrary. The parameters of the precipitation model described below
must be calibrated appropriately to the scale of the glaciation model.
For each time step, the decrease in glacial magnitude is also modeled as a function of
temperature:
𝐺𝑙𝑎𝑐𝑖𝑎𝑙345$6!7$#%&(Δ𝑇)=,
0 Δ𝑇≤Δ𝑇'(#&𝑆'8
𝑁'8
2𝑒+!*⋅(./2./!#+,)–15 if Δ𝑇>Δ𝑇'(#&
where
NGS is a normalizing constant:
𝑁'8 =𝑒+!*⋅(./-$%2./!#$%) (4)
RGS is a rate parameter (yr-1), and
TGMin is a threshold temperature (degrees Celsius).
The change in glacial magnitude for a time step is thus:
𝐺𝑙𝑎𝑐𝑖𝑒𝑟$=𝑚𝑎𝑥[0,𝐺𝑙𝑎𝑐𝑖𝑒𝑟$29 +𝐺𝑙𝑎𝑐𝑖𝑎𝑙!""#$#%&(Δ𝑇$)−𝐺𝑙𝑎𝑐𝑖𝑎𝑙345$6!7$#%&(Δ𝑇$)]
where the t subscript is a time step index. The time step used for the model is 500 yr.
The parameters of the model were calibrated heuristically to compute parameters that produced a
glacial cycle that appeared reasonable for this coarse model. The set of parameters computed
was:
Δ𝑇'(!)=−6
𝑅'*=0.25
Δ𝑇'(#&=−6.0
𝑅'8 =0.2
𝑆'8 =5
Deep Time Assessment for the Clive DU PA
30 March 2020 25
The change in the glacial magnitude for a particular time step as a function of temperature is
shown in Figure 4. These values lead to slow growth during the very cold phases (Jouzel
temperature deviations of less than –6°C) of the glacial cycle, and rapid recession during warm
phases (temperature deviations of greater than –6°C).
6.3 Precipitation
A coarse model for precipitation in the Bonneville basin was developed dependent on global
temperature (as precipitation generally increases with global temperature), lake surface area
(which affects recharged evaporation), and an additional effect that depends of the magnitude of
the continental glacier. The precipitation in meters of annual rainfall is modeled as:
𝑃$(Δ𝑇$,𝐿$29,𝐺$29)=𝐵:+𝑅:/⋅Δ𝑇+𝑅:;8*⋅𝑆𝐴(𝐿$29)+𝑆:'⋅𝑒+.!⋅'/01
where
BP is a baseline precipitation,
RPT is a coefficient of linear effect of global temperature,
RPLSA is a coefficient of linear effect of the surface area of the lake, and
SA(L) is the surface area in km2 associated with lake elevation L.
The effect of temperature and lake surface area are modeled as linear, while the glacial effect is
exponential with respect to glacier size. The set of parameters calibrated to the glacial magnitude
model are:
𝐵:=0.30
𝑅:/=0.005
𝑅:;8*=2 × 102<
𝑆:'=0.06
𝑅:'=0.03
The precipitation is then converted to a volume by multiplying by the area of Bonneville basin
(approximately 47,500 km2).
6.4 Evaporation
Evaporation rate in the region is modeled as a function of temperature:
𝐸$(Δ𝑇$)=𝐵=+𝑆=
𝑁=
⋅𝑒+2⋅(./2./-+,)
where 𝑁=is a normalizing constant:
𝑁==𝑒+2⋅(./-$%2./&'( )
The evaporation is then converted to a volume by multiplying by the area of the basin.

Deep Time Assessment for the Clive DU PA
30 March 2020 26
The calibrated parameters are:
𝐵==0.32;
𝑆==0.3
𝑅==0.05
Δ𝑇>#&=−10
Δ𝑇>!)=5
If the precipitation volume exceeds the evaporation volume, then the difference is added to the
lake volume, and the lake elevation is calculated from the total lake volume.

Figure 4. Glacial change as a function of temperature for the coarse conceptual model

Deep Time Assessment for the Clive DU PA
30 March 2020 27
If the evaporation volume is greater than the precipitation volume, then the total evaporation is
adjusted downward to adjust for the actual surface area exposed (rather than the full surface area
of the basin as used in the initial calculation). The difference between the adjusted evaporation
and the precipitation is then subtracted from the lake volume, and the lake surface elevation is
calculated from the total lake volume.
Δ𝑉𝑜𝑙𝑢𝑚𝑒$=,
[𝑃$(Δ𝑇$)−𝐸$(Δ𝑇$)]⋅𝑆𝐴5!3#& 𝐸$(Δ𝑇$)<𝑃$(Δ𝑇$)
[𝑃$(Δ𝑇$)−𝐸$(Δ𝑇$)]⋅𝑆𝐴(𝐿$29)
𝑆𝐴5!3#&
𝐸$(Δ𝑇$)≥𝑃$(Δ𝑇$)
6.5 Simulations
For simplicity, lake volume and glacial magnitude are assumed to be zero at the first time step
(785 ka), as that time step corresponds to a warm climate phase. The values for the parameters
given above are calibrated graphically to produce reasonable precipitation versus evaporation
values. Several lake elevation histories were simulated by simulating the parameter values of the
model probabilistically. The distributions for the parameters were lognormal with medians equal
to the parameter values listed in Equations (6), (8), and (11). The simulations provide a variety of
behaviors depending on the combination of parameters simulated.
A few common features are apparent in the simulated results. The largest lakes tend to occur at
the times of Lake Bonneville, Little Valley, and Lava Creek, and the smallest 100-ky cycle lake
occurs in δO18 cycle 14 (~533 ka), which matches the scientific record. When the simulated
glaciation effects are small (RGA and RGS), precipitation change in the model is due primarily to
temperature change. In this case, deep lakes form with few intermediate lakes, as the lake
elevation history in the top graph in Figure 5 shows. When glaciation effects are larger, then deep
lakes tend to last longer, and intermediate lakes form, as the lake elevation history in the lower
graph of Figure 5 shows.
The simulation models were then calibrated further by combining the simulated lake histories
with sedimentation rates seen in sediment cores. Based upon the results of this coarse model
calibration, some assumptions are carried forward to the Deep Time Model.
1. The 100-ky cycle in global temperature is a strong indicator of the return of a deep lake.
While not all simulations showed a lake returning to the Clive elevation in every 100-ky cycle
(particularly δO18 cycle 14), the results were consistent enough to treat as systematic behavior for
a heuristic model.
2. Intermediate lakes should be a part of the Deep Time Model, because sedimentation rates
did not calibrate well with simulations that produce only deep lakes.
3. Intermediate lakes are more likely to occur in the later stages of the 100-ky cycle than in
the early stages, primarily in conjunction with the slowly decreasing temperatures across the
cycle (as opposed to the relatively rapid warming period that occurs at the end of a 100-ky
cycle).
Deep Time Assessment for the Clive DU PA
30 March 2020 28

Figure 5. Two example simulated lake elevations as a function of time, with Clive facility
elevation represented by green line
Deep Time Assessment for the Clive DU PA
30 March 2020 29
7.0 Deep Time Modeling Approach
7.1 Introduction
The GoldSim systems analysis software (GTG, 2011) is used to construct the Clive DU PA
Model v1.5. The same Species list of contaminants, material properties, and site geometry are
retained from the Clive DU PA Model v1.2. The standalone DTSA Model is combined with the
deep time container of the Clive DU PA Model v1.2 in the Clive DU PA Model v1.5 deep time
container.
The DU waste inventory for the start of deep time is taken from the Clive DU PA Model v1.5
Federal DU Cell Disposal container at the time the first lake returns, which changes for each
realization. The DU waste is disposed below current grade. Contaminant fate and transport are
captured in the Federal DU Cell until the first lake returns. Radionuclides that have migrated
above grade when the first lake returns are dispersed across the lake area and assumed to be
available to diffuse into any lake that appears. The dispersal is limited to at least 2m above
current grade because eolian processes deposit at least 2 m of material in the 50,000 years or
more before a lake returns. Remaining radioactivity in the lowest six waste layers and the 2m of
contaminated material that is above grade that at the time the first lake appears are used as the Rn
flux inventory for the Deep Time Model.
The Deep Time Model is largely a heuristic representation of deep time. The underlying concepts
are that a lake will return to the elevation of Clive at some point in the future, and new lake
sediments will be sufficiently thick after the first lake recedes that radon flux will meet present-
day regulatory guidelines2. Contaminant fate and transport after the first lake returns are not
evaluated in the Deep Time Model, excepting radioactive decay and the ingrowth of progeny.
As previously discussed, the depth of lake and eolian sediments removed at the Clive location
due to wave action and the residual material from the destroyed embankment are expected to be
approximately equal, and their effects essentially cancel. Therefore, the thickness of residual
embankment material and sediment overlying the disposed DU waste at the time when the first
intermediate lake recedes will be effectively equivalent to the thickness of eolian sediments
deposited up until that point in time and the sediment thickness deposited by the first
intermediate lake event.Both deposits will raise the elevation of the surrounding grade. The Deep
Time Model calculates radon ground surface flux from radionuclides in the disposed DU waste
buried beneath this layer. Dose to a rancher from this radon flux is calculation to provide a
reference point to interpret the significance of the radon flux.
7.2 Deep Lake Characteristics
The 100-ky climate cycle is treated as a sufficiently robust effect to create a hypothetical lake
that will reach and exceed the elevation of the Clive site during each glacial cycle. The exact
time of occurrence is not a crucial parameter, due to the slowly-changing concentrations during

2 As discussed in Section 3.0 above, there is not a regulatory point of compliance for deep time. Present-day radon
flux guidelines are considered in the context of gauging system performance, as such results may provide limited
insight into the behavior of the disposal system in deep time.
Deep Time Assessment for the Clive DU PA
30 March 2020 30
deep time. Thus, the lake is set to be present during each 100-ky interval, with time beginning at
10 ky (the end of the performance period for the quantitative dose assessment component of the
PA).

There is limited information from the Quaternary geologic record for the duration of time that the
Clive location has been under water. Lake Bonneville has been estimated to have been present at
the elevation of Clive for an interval of approximately 17 ky (Oviatt et al., 1999). Durations of
pre-Lake Bonneville deep lakes are uncertain. Thus, a conservative choice was made to allow
deep lakes to remain an average of about 20 ky (conservative in the sense that more
radionuclides will migrate into the water column). The occurrence time for each deep lake is set
by choosing a start time some number of yr prior to the 100-ky mark. The start time is
represented by a lognormal distribution with geometric mean of 14 ky prior to the 100-ky mark,
and a geometric standard deviation of 1.2. The end time is represented by a lognormal
distribution with geometric mean of 6 ky after the 100-ky mark, and a geometric standard
deviation of 1.2. These distributions are depicted in Figure 6.
Deep Time Assessment for the Clive DU PA
30 March 2020 31

Figure 6. Probability density functions for the start and end times for a deep lake, in yr
prior to the 100-ky mark and yr after the 100-ky mark, respectively.
7.3 Intermediate Lake Characteristics
Intermediate lakes are modeled as potentially occurring during the transgressive and regressive
phases of deep lakes and at any time between deep lake events. In order to reflect the slow
decrease in temperature over the 100-ky cycle, the occurrence time for intermediate lakes is
modeled as a Poisson process with a rate that increases linearly over the cycle time, from a rate
of 1 to 12 lakes per 100 ky. The lower number of lakes is established by the assumption of at
least 1 deep lake per 100 ky cycle. The upper number of lakes is required for consistency with
the heuristic lake-elevation model. This process produces an average of about 4 intermediate
lakes per 100 ky.

Deep Time Assessment for the Clive DU PA
30 March 2020 32
There is virtually no information for the duration of intermediate lakes, due to the high mixing
rate of lake sediments, which prohibits establishing the chronology of individual stratigraphic
layers from studies of cores of intermediate lake sediments. Thus, a distribution was chosen to
roughly calibrate with the heuristic model: lognormal with geometric mean of 500 y and
geometric standard deviation of 1.5.
7.4 Sedimentation Rates
As previously mentioned, the Deep Time Model makes a distinction between deep and
intermediate lakes with regard to sedimentation.
• The sedimentation patterns of deep lakes are assumed to be similar to observed intervals
of carbonate marl from Lake Bonneville or Lake Provo, and are assumed to occur no
more than once per 100-ky glacial cycle. The depth of deep lakes is significantly greater
than the depth of wave action and slow precipitation of carbonate is assumed to be the
dominant sedimentation process.
• Intermediate lakes are defined as lakes that reach and exceed the altitude of the Clive site
but are not large (or deep) enough that carbonate sedimentation is the dominant mode of
lake sedimentation. The transgressive and regressive phases of the Bonneville and Provo
shoreline lakes represent intermediate lakes formed during transient lake cycles where
the lake levels exceeded the elevation of Clive and lake sedimentation was dominated by
clastic deposits associated with wave activity and reworking of pre-existing lake and
eolian sediments (see Table 2 for the chronology of the lake cycles and descriptions of
the pre-Lake Bonneville transgressive deposits in Neptune 2015a).
• Shallow lakes, similar to the modern Great Salt Lake, are assumed to exist at all other
times, but these are irrelevant to the geomorphology of the Clive site and thus are not
explicitly modeled.
Deposition of eolian and lake sediments in the area of the Clive facility is a continuous process
that occurs during shallow, intermediate and deep lake periods. During shallow lake periods, as
observed in present-day conditions, eolian deposition of silt/loess from suspension fallout is the
primary sedimentary mechanism. However, eolian deposits are rarely observed in sediment
cores, presumably because of reworking of the depositions by wave action during lake
transgressions and mixing with lake-derived sediments. Note however that the upper part of the
Clive quarry exposure is now known to be of eolian origin (Neptune, 2015a) and paleosoils and
eolian deposits have been observed in the pre-Lake Bonneville sedimentary deposits at Clive and
described in the Burmester core indicating prolonged periods of subaerial exposure.
Intermediate lake sediments include chemical, biogenic, and terrigenous sediments, with their
proportions dependent on the size and duration of the lake and the interplay between deep lake
deposition and near-shore sedimentary processes. Schofield et al. (2004) note that the large fetch
of Lake Bonneville (distance of wave forming winds over the water) produced a variety of wave-
dominated erosional and depositional sedimentary and geomorphic features. They identified
cross-sections of erosion-dominated and deposition-dominated shorelines and the composite
sedimentation rates of shoreline profiles will be dependent on local process of wind/wave
erosion and deposition and supply of sediments from alluvial fans flanking pluvial lakes
(Schofield et al., 2004). Moreover, eolian depositional layers are not commonly identified in the
Deep Time Assessment for the Clive DU PA
30 March 2020 33
sediment cores, so the model effectively combines eolian deposits with lake sediments. The
mixing probably occurs during intermediate lake cycles, which are likely to be the first lakes
after interglacial periods. These assumptions require that there is a mixing depth associated with
each lake recurrence. However, the mixing process itself makes it difficult to assign mixing
depths for the different layers in the sediment cores. Mixing depths are probably determined by
the dynamics of wave activity and resulting erosion/deposition and fluctuations in lake depths
during transgressive and regressive lake cycles.
Neptune (2015a; Section 6.0) described the conceptual model of current and future eolian
deposition at the Clive site. The following events are expected as lake elevations increase during
glacial cycles and exceed the elevation of the Clive site:
1. Increased water table elevation and eastward expansion of playa areas now located west
of the Clive site. These changes will increase the rate of supply of eolian source material
and eolian depositional rates.
2. Development and eastward migration of dune fields toward the Clive site similar to the
modern dunes at Knolls, 14 km west of the site. Saltation deposition associated with
sand dune development would increase and become the dominant mode of deposition at
the Clive site. Sand deposition associated with transitory dune formation will locally
exceed 2 to 3 meters in thickness.
3. When lake waters reach the Clive elevation, wave action will erode local surface
topography projecting above the aggrading surface, rework eolian deposits, and intermix
eolian sand and silt with shoreline sands/oolitic sands and local accumulations of
gravels deposited by longshore drift. The low surface gradient will result in widespread
dispersal of these complex reworked eolian and shoreline lake deposits.
Deep lakes, in contrast, will have slower rates of clastic sedimentation as the lake waters deepen
and the active lake shoreline moves eastward away from the Clive site. The dominant process of
sedimentation is precipitation of chemical and biogenic material from the lake waters. Studies of
the sediment cores from the Clive region are able to distinguish between layers associated with
intermediate lakes with predominant sediment mixing, and sedimentary layers associated with
deep lakes that are dominated by carbonate layers (marl).
For deep lakes, a sedimentation rate is modeled as a lognormal distribution with geometric mean
of 120 mm/ky and geometric standard deviation of 1.2, a distribution that covers the range of
observed depositional rates for deep lakes. This distribution is represented in Figure 7. The
sedimentation rate is applied for the simulated duration of the deep lake.
For intermediate lakes (and shallow phases of deep lakes), there is high likelihood of multiple
short-term transgressions and regressions in lake elevations with respect to the elevation of
Clive. For example, the Clive pit wall (Appendix A) shows three distinct lakes after the deep-
water phase of Lake Bonneville and three distinct lakes prior to the deep-water phase of Lake
Bonneville. Without further systematic study of sediment cores and trench sections in and around
the Clive site, including chronology studies, it is impossible to determine if these distinct lakes
were separated by a few years or a few hundred years; i.e., whether they are distinct lake events
or simply part of the transgression and regression of Lake Bonneville. However, based upon
current behavior of the lake, some year-to-year variation in the lake elevation occurs, in addition
to the longer-term trends in lake elevation.
Deep Time Assessment for the Clive DU PA
30 March 2020 34

Figure 7. Probability density function for sedimentation rate for the deep-water phase of a
deep lake
Another heuristic model was constructed to evaluate the effect of the short-term variation. The
lake elevation for the years 1848 through 2009 is available from the Saltair Boat Harbor
monitoring site (USGS, 2001), as shown in Figure 8. The year-to-year variation can be modeled
as a second-order autoregressive process AR(2) (Brockwell and Davis, 1991), a model that
accounts for year-to-year temporal correlations in the variation. An AR(2) process was simulated
and added to a transgressive or regressive curve based upon the simplified model previously
presented. Examples of these simulations are given in Figure 9. As can be seen in the figure, the
short-term variation can result in lakes covering the Clive elevation for a short time, receding for
a short time, then rising again, often multiple times in a single transgression cycle. A similar
simulation was performed for simulated intermediate duration lakes as well. The transgressive
and regressive phases of a deep lake are assumed to behave similarly to the intermediate lakes in

Deep Time Assessment for the Clive DU PA
30 March 2020 35
that they averaged about four total occurrences of “mini-lakes;” i.e., occurrences of a rise above
the elevation of Clive followed by a drop below for at least one year.

Figure 8. Historical elevations of the Great Salt Lake
The distribution for the sediment thickness from intermediate lake events at the Clive site (Figure
10) was developed from three sources of information:
1. Composite sediment thicknesses associated with the 100 ky glacial cycles observed in
cores studies for the Knolls, Burmester and Saltair sites (13 to 26 meters net deposition;
Appendix B, Table B1). These deposits include intervals of eolian, intermediate lake and
deep lake deposition. The thickness of individual deposits can be estimated assuming
typical durations of eolian and deep lake depositional states during glacial cycles and the
estimated depositional rates for the respective depositional modes (see Figure 7 for deep
lake deposition rates and Section 7.5 for eolian deposition rates). These two sediment

Deep Time Assessment for the Clive DU PA
30 March 2020 36
components are subtracted from the glacial cycle sediment thickness. The remaining
sediment thickness is the sediment thickness attributed to the net activity of intermediate
lakes. The thickness of individual intermediate lake events is estimated from the Poisson
process used to simulate the number of intermediate lakes during a glacial cycle (see
section 7.3). Uncertainty in the sediment thickness distribution using this approach is
from the variations in estimates of glacial thickness intervals for the 100 ky cycles,
uncertainty in the eolian and deep lake depositional intervals and rates, and the
extrapolation of sediment thickness from the sites of core studies to the Clive site. For the
latter, the Clive site is near local topographic highs that are sources of local clastic
sediments. This probably results in underestimation of intermediate lake sediment
deposition at the Clive site.
2. Interpretations of quarry wall sections in the Clive site (see Section 4.3, Table 3 and
Appendix A). Uncertainty in these sediment thicknesses are primarily from limited
chronology data and assumptions concerning lake sequence boundaries.
3. Observations of the thickness of the pre-Lake Bonneville transgressive lake deposits in
the Clive site (Neptune 2015a). These sediments vary from about 0.5 to greater than 3 m
in thickness and can be attributed to the composite depositional processes of a
transgressive lake associated with the evolution of Lake Bonneville. The uncertainty in
this approach is the thickness of this sedimentary sequence has not been studied
systematically along and across the Clive site.

These three sources of sediment thickness information were incorporated into a model for
multiple mini-lake behavior. First, the number of mini-lakes associated with an intermediate lake
was simulated as 1 plus a Poisson random variable with rate 3 (the “plus 1” being necessary to
ensure at least one event in order to match the definition of a lake event). The sedimentation for
each mini-lake was simulated using a distribution based upon the sedimentary deposits of mini-
lakes exposed in the Clive pit wall, using the six distinct “mini-lakes” in Table 3 (all layers
except the one that corresponds to the deep-water phase of Lake Bonneville). These data are
represented in a lognormal distribution of sediment thickness with geometric mean 0.75 m and
geometric standard deviation 1.4.
Deep Time Assessment for the Clive DU PA
30 March 2020 37

Figure 9. Simulated transgressions of a deep lake including short-term variations in lake
elevations
The total sedimentation for all mini-lakes associated with a simulated intermediate lake cycle
was then added together to produce a total sedimentation for the intermediate lake. A distribution
was then based upon all simulated intermediate lake sedimentations, a lognormal distribution
with geometric mean 2.82 m and geometric standard deviation 1.71, as presented in Figure 10.
Note that the sedimentation distribution for intermediate lakes is represented as a distribution of
composite sediment thickness per lake event and contrasts with a distribution of sedimentation
rates assumed for deep lakes. The net effect is that the sedimentation rates are on the order of 13
to 26 m per glacial cycle (100-ky). For the duration of the model (2.1 My), this implies
sedimentation of more than 300 m. The Basin and Range system accommodates this rate of
sedimentation because it is an extensional system; i.e., sedimentation continues as the basins
expand and subside, maintaining similar elevation in each cycle.
Deep Time Assessment for the Clive DU PA
30 March 2020 38

Figure 10. Probability density function for the total sediment thickness associated with an
intermediate lake event
An additional consideration is the unique sedimentation history for the first lake to reach the
Clive elevation during the next glacial period, the critical scenario for establishing the maximum
radon release rates for the Clive site. Two factors suggest the sediment thickness distribution may
underestimate the actual intermediate lake deposition for the first lake event at the Clive site.
1. The first lake event associated with the next glacial cycle is expected to occur after an
exceptionally long interglacial interval ( > 50 ky). This interval will result in increased
local eolian deposition and accumulation of clastic sediments on topographic surfaces
within and surrounding the Clive site. These factors will increase the amount/volume of
clastic sediments available for reworking by wave activity during the first lake event.

Deep Time Assessment for the Clive DU PA
30 March 2020 39
2. The sediment thickness distribution (Figure 10) includes both transgressive and
regressive lake events. Field studies at the Clive site (2015a) show that intermediate lakes
of regressive lake sequences tend to be thinner because of the combined effects of the low
topographic gradient near the Clive site, and the mantling of the topography by fine-
grained lake deposits which decreases availability of local clastic sediments (see Neptune
2015; their Figure 12).

7.5 Eolian Depositional Parameters
Studies of eolian deposits in multiple quarry exposures at the Clive site and in surface exposures
west and southwest of the site show that deposition of eolian sand and silt is now occurring and
will continue to occur in the future as long as the at grade site elevation is exposed at the surface
(above the elevation of lake levels; Neptune, 2015a).
7.5.1 Field Studies
Field studies of the eolian depositional history at the Clive Disposal Site were conducted in
December 2014 to provide information for characterizing eolian deposits and establishing eolian
depositional rates for the original DTSA Model (Neptune, 2015a). The primary goals of the field
studies were to evaluate the modern geological and depositional setting of the Clive site, and to
assess the stratigraphy of the Holocene and Pleistocene lake sedimentation of Lake Bonneville
and post-lake depositional processes within the Clive site including the following:
1. Re-evaluating the stratigraphic section previously described by Oviatt (1985, cited in
Neptune, 2015b).
2. Describing the eolian sediments and processes affecting the sediments.
3. Measuring variations in thickness of the deposits across the site.
4. Providing sufficient replicate measurement at multiple sites to estimate eolian sediment
thicknesses and the variation in eolian sediment thicknesses at the Clive site.
The field studies achieved these primary goals, and the replicate measurements of the thickness
of eolian deposits located in the upper part of the stratigraphic section were made at multiple
locations on and in the vicinity of the Clive Disposal Site. The data are presented in Neptune
(2015) and are used below to develop input probability distributions for the Deep Time Model.
7.5.2 Probability Distributions for the Depth and Age of Eolian Deposition
The Deep Time Model requires specification of input probability distributions for the depth of
eolian deposition and the age of the eolian deposits. Together, these two variables provide the
information needed to estimate the rate of eolian deposition. The distribution for the depth of
eolian deposition is based on the field data described above (Neptune, 2015a), whereas the
distribution for the age of the eolian deposits are derived from a summary paper by Oviatt
(2015).
An assumption is made that the described eolian deposits at the Clive site represent an integrated
time interval of eolian sediment accumulation, modification by processes of soil formation and
minor modifications by processes of surface erosion. These deposits approximate a steady-state
Deep Time Assessment for the Clive DU PA
30 March 2020 40
representation of eolian processes since the regression of Lake Bonneville and these processes
should continue into the future until conditions at the site change considerably (e.g., natural
climate change). The distributions are based on the depth of eolian deposition since Lake
Bonneville regressed below the elevation of Clive and estimations of the age at which regression
below the Clive elevation occurred (Neptune, 2015a) These distributions are used to model
future eolian deposition until the return of a lake at the elevation of Clive.
The data presented in Table 4 from Neptune (2015a) are the measured thicknesses of eolian silt
in quarry walls and excavated surfaces for the Clive Disposal Site. The mean of the deposits is
72.7 cm, and the standard deviation is 16.6 cm. There are 11 data points, and the data are
reasonably symmetric about the mean. Consequently, a normal distribution is specified for the
Deep Time Model with a mean of 72.7 cm and a standard error of 5.0 cm. A reasonable
simulation range considering ± 3 standard errors would be 57.5 to 87.5 cm. The minimum of the
normal distribution was set to a very small number and the maximum was set to a very large
number so that the distribution was not unnecessarily restricted.
This distribution represents spatio-temporal scaling, so that the distribution is of the average
depth of eolian deposition at the Clive site since Lake Bonneville regressed below the site. This
provides the best representation of the future eolian depositional rates over the long timeframes
and spatial scales of the Deep Time Model.
Table 4. Thickness measurements from field studies of eolian silt near Clive
Neptune Field Studies December 2014
Site GPS Coord GPS Coord Silt Thick Date
UTM E UTM N (cm) (mm/dd/yy)
Clive 29-1 321354 4508262 90.0 12/16/14
Clive 29-2 321390 4508256 80.0 12/16/14
Clive 29-3 321423 4508248 80.0 12/16/14
Clive 29-4 321502 4508236 60.0 12/16/14
Clive 29-5 321239 4508283 110.0 12/16/14
Clive 5-1 320813 4504729 55.0 12/16/14
Clive 5-2 320869 4504730 70.0 12/16/14
Clive 5-3 320914 4504731 60.0 12/16/14
Clive 5-4 321041 4504732 70.0 12/16/14
Clive Hand-Dug-1 322093 4507482 70.0 12/17/14
Clilve hand-Dug-2 320445 4507035 55.0 12/17/14
Mean 72.7
Std Error 5.0

Note that several replicate measurements were taken at each location (usually three or four), and
the results represent the average thickness at each location. These data are also supported by
previous data collected from shallow core studies at Clive, which also are presented in Neptune
(2015a). The documentation and uncertainty in the measurements of the eolian sediment
thickness from the core studies data is not as precise as those made in the Neptune field study;
however, the data are supportive of the results of the field study, indicating very similar patterns
Deep Time Assessment for the Clive DU PA
30 March 2020 41
of eolian thickness data. These data provide another 21 data points that have an average of 71 cm
depth of eolian deposits, with a standard error of 4 cm. Because of the uncertain pedigree and
lesser precision of the data from the core studies, they were not used in the distribution
development. Their use would have resulted in a much tighter distribution because of the scaling
effects of spatio-temporal averaging.
Ages of the deposits were determined from radiocarbon dating. The summary paper by Oviatt
(2015) provides the most recent compilation and interpretation of radiocarbon ages for the
chronology of Lake Bonneville. Based on information summarized in Figure 2 of Oviatt (2015)
and supported by the supplemental radiocarbon data referenced in the paper, the preferred
estimate for the age of the final regression of Lake Bonneville below the altitude of the Clive site
is about 13.5 ka. (Clive elevation 1304 m). A reasonable lower bound on the youngest or
minimum age for this event is 13.3 ka based on radiocarbon ages determined from organic
material collected in post-Bonneville wetland deposits (Oviatt, 2015). The reasonable oldest or
maximum age of lake regression at the Clive site is constrained by the age of the Provo shoreline
and reliable radiocarbon ages for sites above the altitude of the Clive site and below the Provo
shoreline. This reasonable maximum age is estimated to be about 14.5 ka. A distribution was
developed based on these values from Oviatt (2015) and on expert elicitation of Oviatt. Oviatt
suggested that values around 13.5 ka were more likely. Based on this information a beta
distribution was fit to approximate elicited quantiles. The following quantile inputs were used:
• Absolute minimum possible age – 13,000 yr
• Reasonable minimum age – 13,300 yr
• Most likely age – 13,500 yr
• Reasonable maximum age – 14,500 yr
• Absolute maximum possible age – 15,000 yr
After considering possible quantiles for the middle three terms, a beta distribution fit was agreed
upon with the following parameters:
• Minimum – 13,000 yr
• Maximum – 15,000 yr
• α (shape 1 parameter) – 3.318
• β (shape 2 parameter) – 7.498
This beta distribution has a mean of approximately 13,600 yr and a standard deviation of
approximately 270 yr. The mean is reasonably close to the specified most likely age of 13,500
yr. Quantiles of this beta distribution are provided below:
• 2.5% – 13,174 yr
• 10% – 13,284 yr
• 20% – 13,378 yr
• 50% –13,592 yr
• 80% – 13,846 yr
Deep Time Assessment for the Clive DU PA
30 March 2020 42
• 90% – 13,988 yr
• 97.5% – 14,207 yr
The distribution is slightly positively skewed, hence the median is slightly less than the mean,
and the difference between the maximum and the median is greater than the difference between
the minimum and the median. Note that averaging is not employed for this distribution. The
distribution simply reflects the age over which eolian deposition has occurred. The rate of eolian
deposition is averaged for spatio-temporal scaling by dividing the depth of deposition by the age
over which deposition has occurred as described in the next section.
In principle, the rate of eolian deposition is the deposition thickness divided by the age over
which deposition occurs. However, an assumption is made that greater ages imply greater depths,
in which case there is a correlation between depth and age of eolian deposition. There are no data
to inform a correlation between these two variables. Although elicitation could be performed to
develop a correlation, the approach taken is to specify the correlation as uncertain across a range
of 0.5 to 1. In a sense, this distribution is chosen to indicate that the “data are more likely to be
correlated than not-correlated.” A uniform distribution is used across this range, but this input
will be tracked specifically in sensitivity analysis to determine if it is an important predictor of
the Deep Time Model output.
Using the input distributions and the correlation described above, the resulting distribution of
rate of eolian deposition in the model has a mean of approximately 5.3 × 10-5 m/yr, (roughly 53
cm every 10 ky) with a standard deviation of approximately 3.0 × 10-6 m/yr. A histogram of the
eolian deposition rate for 1,000 realizations is depicted in Figure 11. Quantiles from these
simulated data include:
• 5% – 4.84E-05 m/yr
• 10% – 4.96E-05 m/yr
• 20% – 5.10E-05 m/yr
• 50% – 5.34E-05 m/yr
• 80% – 5.58E-05 m/yr
• 90% – 5.71E-05 m/yr
• 95% – 5.81E-05 m/yr
The distribution is symmetric, as evidenced by the normal distribution fit that is laid over the
histogram. The normal distribution has the mean and standard deviation as specified above, and
the quantiles, which show similar differences between the 95% quantile and the median and the
5th quantile and the median.
Overall, this intermediate product of the Deep Time Model suggests eolian deposition rates of
slightly more than 0.5 m every 10,000 yr.
7.6 Destruction of the Federal DU Cell
Destruction of the Federal DU Cell embankment was modeled assuming future lakes have
sufficient wave energy to destroy the above-ground portions of the cell. The precise lake
Deep Time Assessment for the Clive DU PA
30 March 2020 43
elevation needed for this to happen is not considered for the model, but the intermediate lakes
that occur in the model are intended to match this definition.
The first lake in the time period assessed is more likely to be an intermediate lake but can be
either an intermediate lake or a transgressive deep lake. The destructive energy is equivalent in
either case, as the conceptual model treats the transgressive phase of a deep lake as behaving
similarly to an intermediate lake.
The mass of material that is within the embankment above the grade of the surrounding land is
assumed to be eroded to grade and dispersed by wave action. This volume of above grade
material in the embankment, including fill material and cap material, is assumed to be mixed
with the sediment associated with the intermediate lake, and subsequently spread across a
dispersal area determined by the dynamics of wave activity.
The dispersal area parameter used in the original Deep Time model was estimated for a projected
area where the above grade embankment material could be spread by wave action using different
assumptions for the final dispersal thickness of the volume of embankment material. The
dispersal area was designed to be conservative (small sediment dispersal areas) giving higher
waste concentrations in sediment allowing increased dissolution of waste in lake water.
Deep Time Assessment for the Clive DU PA
30 March 2020 44

Figure 11. Eolian deposition rate results for 1,000 realizations (m/yr).
With below grade disposal of DU, the approach to estimating the dispersal area is revised and
based on a conceptual model for processes affecting the Clive disposal site with the return of a
lake. The following assumptions are used for the revised lake return scenario:
1. The Clive site will be affected by the return of a lake at some time in the future. The
lake event will be either an intermediate lake or the transgressive phase of a deep lake
with the lake processes the same for either event (degradation of the site by near-
shoreline wave action).
Histogram with Normal Curve
Eolian Deposition Rate
Fr
e
q
u
e
n
c
y
4.5e-05 5.0e-05 5.5e-05 6.0e-05 6.5e-05
0
20
40
60
80
10
0
12
0
Deep Time Assessment for the Clive DU PA
30 March 2020 45
2. Eolian deposition will occur during the interval after waste emplacement and before the
first return of a lake to the elevation of the Clive site.
3. Wave action associated with the lake return is assumed to completely remove the
above-grade embankment material above the DU waste.
4. Wave action will churn (rework) the eolian deposits and lake sediments. The maximum
depth of reworking of the eolian deposits is assumed to be about 1 meter based on the
geometry of shoreline deposits for Lake Bonneville.
5. Radionuclides within the above grade embankment will be dispersed by wave action
and mixed with eolian deposits and lake sediments.
6. The alternative models used for estimating sediment dispersal areas include:
a. Analogue sites of modern sedimentary processes dispersing sediments at
shorelines of the Great Salt Lake;
b. Field assessments of sediment dispersal during the transgressive phase of Lake
Bonneville at and around the Clive Disposal Area (Neptune, 2015a);
c. Assessment of wind directions from dune forms west and southwest of Clive
(Jewell and Nicoll, 2011)
Google Earth© imagery was used to identify and determine the areas of active shoreline
sedimentation for the Great Salt Lake assuming these patterns provide analogues for wave action
and sediment dispersal for the lake return scenario at Clive. Dispersal area estimations assumed
no longshore drift (minimum areas) and one dominant direction of longshore drift (maximum
areas). The Great Salt Lake analogue may be somewhat conservative (underestimate sediment
dispersal) for two reasons. First, the fetch length for a lake return at the Clive elevation would be
longer than the fetch length for the Great Salt Lake. Second, the observed sedimentation patterns
of the Great Salt Lake represent relatively short term dynamics of lakeshore processes – the
dispersal area of sediments for the return of a lake at the Clive site and erosion of the
embankment would likely develop over a timescale of multiple decades.
Google Earth© imagery was used to estimate alternative sediment dispersal areas using
constraints from field observations of the distribution of conglomerate and sand deposits of the
transgressive phase of Lake Bonneville. These estimations combined data from surface
landforms and quarry-wall exposures of lake sediments at Clive.
Finally, alternative sediment dispersal patterns were estimated using Google Earth© imagery for
Clive by centering the sediment dispersal at the Clive embankment and adjusting the dispersal
patterns for the topographic features of the Clive area.
The following percentiles were assigned to the composite data to establish a distribution for the
sediment dispersal parameter:
• 1%: 4 km2 from smallest measured dispersal area
• 5%: 10 km2 assuming only west-east wind directions
• 15%: 15 km2 averaging dispersal areas for no longshore drift
Deep Time Assessment for the Clive DU PA
30 March 2020 46
• 30%: 16 km2 averaging dispersal areas for N-S and SW-NE longshore drift
• 50% 24 km2 assuming multidirectional winds and longshore drift
• 75% 36 km2 averaging all single direction longshore drift dispersal areas
95% 55 km2 from maximum measured dispersal area
A gamma distribution was used to fit the percentages above, with mean of 24.2332 and standard
deviation of 11.43731. A typical probability density function of this distribution is shown in
Figure 12.

Figure 12. Probability density function for the area over which the waste embankment is
dispersed upon destruction
Deep Time Assessment for the Clive DU PA
30 March 2020 47
7.7 Radionuclide Concentration in DU Waste
After a lake recedes, radionuclides in the original DU waste disposal volume are not likely to
move to the surface in any significant amounts via diffusion or other processes. This section
discusses processes that are likely to occur in deep time relating to the original DU waste source.
Infiltration rates will increase over time, moving material downward via advection, counteracting
potential upward diffusion of radionuclides. The climate will become cooler and wetter, entering
a glacial period, resulting in the lake return. Estimates of future net infiltration at Clive are
supported by work for the Yucca Mountain Project. Faybishenko (2006) developed models
predicting infiltration rates for future climate states based on factors including predicted
precipitation, evapotranspiration, and temperature. The meteorological stations at Simpson and
Spokane in Faybishenko (2006, Table 3) provide a reasonable range of infiltration rates for Clive
of 40 mm/yr to 73 mm/yr, for a glacial phase.
An external, finely-discretized GoldSim model was used to test diffusion behavior at Clive,
along with higher infiltration rates. The model results showed that if infiltration increased even
to 10 mm/yr, downward advection would dominate upward diffusion in the model. Infiltration of
40 mm/yr to 73 mm/yr would move radionuclides that had diffused above the original grade
back below grade.
Dry periods during the inner-glacial timeframes would be expected to behave like current
conditions. Because of the uncertain nature of the deep time future conditions and timing and
because it is important to keep the Deep Time Model simple, it was assumed that until the first
lake returns, radionuclides migrate upwards via the processes of diffusion and plant and animal
transport and that the associated material and radionuclides above grade is spread across the site
dispersal area and is available to diffuse into an intermediate or deep lake. These simplifying
assumptions ignore increases in infiltration during wetter periods in the climate cycle, which is a
conservative approach.
7.8 Radionuclide Concentration in Sediment
The radioactivity per unit volume of sediment following the dispersal of the waste is estimated
using Equation 13 below.
The model calculates radioactivity by volume in the sediment layers, after the embankment has
been destroyed. The current implementation always mixes sediment with the full amount of
waste, and does not consider a mixing depth; i.e., the waste is always fully mixed and not
covered by sediment. Thus, radioactivity concentration in sediment is initially calculated under
the assumption that all of the waste in the waste embankment is mixed evenly with the sediment
that forms as a result of the lake destroying the embankment.
Concentration in sediment is initially calculated under the assumption that all of the waste that
was above grade in the waste embankment is mixed evenly with the sediment that forms with the
lake that destroys the embankment.
𝐶sediment =!embankment
"material above grade#"sediment
.

Deep Time Assessment for the Clive DU PA
30 March 2020 48
where
Rembankment is all remaining radioactivity in the embankment,
Vmaterial above grade is the volume of material in the above grade portion of the embankment
(estimated as 3,231,556 m3), and
Vsediment is calculated as the depth of sediment due to lake processes multiplied by
the area over which the waste is dispersed.
This calculation assumes that there is no loss of waste from the initial dispersal region. While
this calculation is counter to the modeling of dissolution into the water column of the lake, a
simplifying assumption is that all waste that dissolves into the lake precipitates back into the
sediment upon recession of the lake.
The concentrations in sediment are modeled as constant, except for decay and ingrowth, until a
new lake occurs. When a new lake occurs, the sedimentation associated with that lake is likely to
mix with some portion of the top layer of existing sediment and leave the lower layers of the
sediment buried beneath. However, for simplicity, a conservative approach is to mix all sediment
that contains waste, effectively keeping some portion of the waste near-surface. The
concentration is again the total radioactivity divided by the volume containing waste, but the
volume that contains waste now has the additional volume of sediment associated with the
current lake.
7.9 Radioactivity in Lake Water
When lake water is present, radionuclides will partition between the water phase and the solid
phase depending on element-specific solubility and sorption properties. Radionuclides remaining
in the pore water will then diffuse into the lake. The waste is likely to mix over a wide area of the
lake, and many forms of the waste are likely to bind with carbonate ions in the water, ultimately
precipitating into carbonate sediments. As a conservative assumption, upon recession of the lake,
all waste is assumed to precipitate back into the local sediments, meaning that all radionuclides
in the sediments are returned to the sediments when the lake regresses.
When a lake returns, the sediments are assumed to be fully saturated, and radionuclides are
partitioned from the sediment to the pore water within the sediment using the same partitioning
coefficients (Kd) used for other sedimentary soils in the model. An important difference between
the assumptions for this model and the model for transport from the embankment in the 10-ky
model is that the lake water is assigned a different solubility for uranium for the Deep Time
Model. While solubilities for all other radionuclides remain the same, the solubility for uranium
is reduced to that of U3O8 which is appreciably lower than other forms of uranium originally
present in the waste. This change in solubility for uranium is adopted because it is expected that
by the time the first lake returns, soluble uranium forms (UO3) either will have been leached
from the embankment into the shallow aquifer or will have been converted into U(IV), which is
also very insoluble.
As radionuclides associated with the sediments dissolve into the pore water, they diffuse into the
lake water using a constant flux model based upon Fick’s first law, with the following
assumptions:
Deep Time Assessment for the Clive DU PA
30 March 2020 49
• The concentration in sediment remains constant over the deep time period. The sediment
concentration should in fact diminish over time if enough mass is migrated into the water,
but for simplicity, the sediment concentrations are kept constant across time steps.
• The diffusion length from the radionuclides in the sediment diffusing into the lake is
about 0.5 m. This diffusive length value assumes the mixing depths of the sediment
correspond to diffusive processes from the sediment into the lake.
• Mixing depths are expected to be between 0 and 1 m, with 0.5 m being most likely. The
distribution was set up as a normal distribution with mean of 0.5 m and standard
deviation of 0.16 m so that 99% of the distribution will be between 0 and 1 m. The
distribution is truncated at 0 m so that no negative diffusion lengths are chosen.
Fick’s law for this case estimates the mass diffusing from a given volume of sediment into the
lake with time. The mass (or activity) per area per time is the flux. Fick’s law states that this flux
is given by the difference in mass concentration over a distance (the concentration gradient)
multiplied by a free-water diffusion coefficient, across a diffusive area. The calculation assumes
that there is a stagnant interface boundary layer of water between the sediment and the open
water that is the thickness of the diffusion length (~0.5 m). The assumption is also made that the
mass concentration is zero in the open water. The difference in concentration across the stagnant
layer is then the concentration in the sediment Cv minus the concentration in the open water or
Cv – 0 g/mL. Fick’s law applied to diffusion is used to define the mass (or activity) flux J:
𝐽= +
∆$ *= 𝐷> A3
5456
where
R is the mass (M) activity (T-1),
ΔT is the length of the time period (T),
A is the area of the sediment that contains the waste (L2),
Dm is the diffusion coefficient for the radionuclide in water (L2/T), and
bbdy is the thickness of the boundary layer.

Multiplying both sides of the equation by ΔT·A gives
𝑅=Δ𝑇⋅𝐷>⋅A7
B.9 D ⋅𝐴
Concentration in lake water is calculated based upon the conservative assumption that the
radioactive material does not dilute in a large basin of the lake but rather remains in the water
column immediately above the dispersed area. The activity concentration in the lake water is
then calculated by dividing the total activity, R, by the volume of lake water. The volume of lake
water is the product of the lake depth and the dispersal area:
𝐶E =𝑅
𝐷⋅𝐴
Deep Time Assessment for the Clive DU PA
30 March 2020 50
where
Cv is concentration (M/L3 or T-1/ L3),
R is the mass (M) or radioactivity (T-1),
A is the area of the sediment that contains waste (the dispersed area, as L2), and
D is the depth of the lake (L).
There is an insufficient record of lake elevations to construct a data-based distribution for lake
depth. Thus, the distributions for lake depth are chosen based upon the conceptual model. Depths
for intermediate lakes have a Beta distribution with mean of 30 m, standard deviation 18 m,
minimum of 0 m, and maximum of 100 m. Depths for deep lakes have a Beta distribution with
mean 150 m, standard deviation 20 m, minimum of 100 m, and maximum of 200 m.
For intermediate lakes, the time step is about the duration of the intermediate lake. For deep
lakes, the lake may exist for several time steps in the GoldSim model, in which case the time step
is the portion of the time step for which the lake is present. When deep lakes cross multiple time
steps, the concentration in sediment is allowed to change between time steps (only due to decay
and ingrowth) and the activity in the lake water is accumulated over those time steps.
7.10 Modeling of 222Rn Flux
Radon-222 flux through the overlying sediment is calculated using the approach described in the
Nuclear Regulatory Commission (NRC) Regulatory Guide 3.64 Calculation of Radon Flux
Attenuation by Earthen Uranium Mill Tailings Covers (NRC, 1989). These equations were
developed for estimating radon flux from uranium mill tailings buried under a monofill cover.
For the Deep Time Model, an assumption is made that the material above the below-grade DU
waste and the additional lake sedimentation is homogenous material with properties similar to
those of the surrounding Unit 3 sediments. The use of an analytical model such as that described
in NRC (1989) allows radon flux to be estimated through a homogeneous cover of varying
thickness with minimal complexity.
The increasing depth of material covering the disposed DU waste over time will result in
attenuation of radon flux. However, this rate of attenuation will be partly offset by the slowly
increasing activity of the radioactive progeny of 238U. Previous modeling results, such as those
from the Clive DU PA Model v1.2, indicated that sediment accumulation overwhelms the
influence of progeny ingrowth.
Although the median and mean sediment thickness track closely, the mean radon ground surface
flux is much larger than the median. This strongly skewed result for radon flux is a consequence
of the non-linearities inherent in the NRC radon ground surface flux calculation. These are
equations (9) through (12) in NRC (1989):
Deep Time Assessment for the Clive DU PA
30 March 2020 51

(17)
The definitions of variables are available in the NRC Regulatory Guide (1989), but the salient
point is that these equations will produce a highly non-linear result, Jc, which is the ground
surface flux of radon. Although all of the inputs to the calculation are essentially normal
distributions, the division calculations, exponents, etc. in the equations produce non-linear
results.
Modeling of radon transport to the surface of the intact Federal DU Cell in the Clive DU PA
Model v1.2 does not lend itself to such simplified analytical solutions, because the cover is
constructed of layers with widely-varying properties. Radon diffusive flux is therefore integrated
with other transport processes employing a column of well-mixed cells, allowing for the vertical
redistribution of radionuclides over time throughout the disposal system by diffusive, advective,
and biotic processes. Because the above-ground part of the Federal DU Cell is assumed to be
dispersed by wave action from the first intermediate lake, these processes are not relevant to the
Deep Time Model except insofar as they affect radionuclide concentrations in the below-grade
waste cells.
7.10.1 Waste and Sediment Water Content
Volumetric water contents are defined for the DU waste, and for sediments overlying the waste,
in order to support radon diffusive flux calculations through these sediments.
In the 10,000-year model, the waste material is assumed to be Unit 3 material. In the Deep Time
Model it is also assumed to be Unit 3, for both the mound material that is directly above the
waste but below grade when the first lake returns and for the sediment material that is deposited
from deep and intermediate lakes in deep time. Sediment porosity is assumed to be the same as
Unit 3 porosity.
The Deep Time Model water contents for the cover materials after the first lake recedes are
based on concentrations of waste materials just above the DU waste, in Waste Cells 17 – 21 and
the upper-most waste cell containing DU waste, Waste Cell 22. This cell may not be completely
full of DU waste, because the discretization of the model may not match exactly the
discretization of the disposed wastes, so Cell 22 was included in these calculations for moisture
content. Because the Deep Time Model is now fully integrated in the v1.5 model, these values
are taken directly from the waste properties and align with those directly for each model
realization.
Deep Time Assessment for the Clive DU PA
30 March 2020 52
7.11 Human Health Exposure and Dose Assessment
In the Deep Time component of the GoldSim model, external radiation dose and radon inhalation
dose are evaluated for the time period after a lake returns. Specifically, this special analysis
evaluates dose at a time immediately after the first intermediate lake has formed and
subsequently receded. The wave action of the lake is assumed to have destroyed the
embankment. The DU wastes at this point in time when the intermediate lake has receded are
covered by a thickness of material equal to the thickness of the eolian sediments that have been
continually deposited at a constant rate over time, plus the deposition of lake sediments while the
intermediate lake exists. The lower stratum of material of thickness equivalent to the eolian
sediments is comprised of the waste layers that existed above the DU waste in the embankment.
Although these wastes at one time contained radionuclides that had migrated upwards from the
DU, these radionuclides are assumed to have been dissolved and dispersed during the time when
the intermediate lake was present. Therefore, both these materials as well as the lacustrine
sediments are assumed to be practically free of DU-related radionuclides in this modeling.
The purpose of the dose calculations the Deep Time component of the GoldSim model is to
determine whether hypothetical doses in Deep Time may be higher or lower than doses
calculated for the 10,000-year performance period. The Deep Time dose calculation results are
not considered to have independent validity. Rather, they are a tool for evaluating the relative
radiation dose during these two time periods. For the dose assessment during the first 10,000
years of the Clive DU PA v1.5 Model, two future use exposure scenarios are identified for the
Clive site: ranching and recreation. However, only ranching receptors are evaluated for the Deep
Time component of the model because their utilization of the area including the Clive site is far
greater than that of recreational users and their doses are therefore higher.
The radiological assessment method for the Deep Time component of the GoldSim model
calculates total effective dose equivalent (TEDE) as the product of exposure (behavioral)
parameters, dose conversion factors (DCFs), and the concentrations of radium and gamma-
emitting radionuclides in the DU waste. The calculations are analogous to those described for
the Ranching scenario during the 10,000-year performance period with two exceptions:
1. Radon flux is calculated using the approach described in the Nuclear Regulatory
Commission (NRC) Regulatory Guide 3.64 Calculation of Radon Flux Attenuation by
Earthen Uranium Mill Tailings Covers (NRC, 1989). These equations were developed
for estimating radon flux from uranium mill tailings buried under a monofill cover, and
the properties of the overlying materials are homogenous material with properties similar
to those of the surrounding Unit 4 sediments.
2. The external DCFs are multiplied by radionuclide-specific modifying factors to account
for the attenuation of external gamma radiation due to the material that overlies the DU
waste. The modifying factors were calculated using the RESRAD computer code by
evaluating the ratio of external dose at different cover thicknesses to external dose with
no overlying material.
8.0 References
Adams, K.D., 2003, Age and paleoclimatic significance of late Holocene lakes in the Carson
Sink, NV, USA, Quaternary Research, Vol. 60, pp. 294–306, 2003.
Deep Time Assessment for the Clive DU PA
30 March 2020 53
Archer, D. and A. Ganopolski, 2005. A movable trigger: fossil fuel CO2 and the onset of the next
glaciation. Geochemistry, Geophysics, Geosystems, 6(5), doi:10.1029/2004GC000891.
Asmerom, Y., Polyak, V. J., and S. J. Burns, 2010. Variable winter moisture in the southwestern
United States linked to rapid glacial climate shifts. Nature Geoscience, 3, 114-117.
Balch D.P., A.S. Cohen, D.W. Schnurrenberger, B.J. Haskell, B.L.V. Garces, J.W. Beck, H.
Cheng, and R.L. Edwards, 2005. Ecosystem and paleohydrological response to Quaternary
climate change in the Bonneville basin, Utah, Palaeogeography, Palaeoclimatology,
Palaeoecology, Vol. 221, pp. 99–121.
Berger, A., 1988. Milankovitch theory and climate. Reviews of Geophysics, 26(4): 624-657.
Berger, A. and M. F. Loutre, 2002. An exceptionally long interglacial ahead? Science, 297: 1287-
1288.
Benson, L.V., S.P. Lund, J.P. Smoot, D.E. Rhode, R.J. Spencer, K.L. Verosub, L.A. Louderback,
C.A. Johnson, R.O Rye, R.M. Negrini, 2011, The Rise and Fall of Lake Bonneville Between
45 and 10.5 ka, Quaternary International, Vol. 235, p. 57-69.
Briggs, R.W., S.G. Wesnousky, and K.D. Adams, 2005, Late Pleistocene and late Holocene lake
highstands in the Pyramid Lake subbasin of Lake Lahontan, Nevada, USA, Quaternary
Research, Vol. 64, pp. 257–263, 2005.
Brimhall W. H. and L. B. Merritt, 1981. The geology of Utah Lake – implications for resource
management. Great Basin Naturalist Memoirs, 5: 24-42.
Brockwell, P.J. and R.A. Davis, 1991. Time Series: Theory and Methods. Springer-Verlag, New
York, NY.
Clark, P. U., Dyke, A. S., Shakun, J. D., Carlson, A. E., Clark, J., Wohlfarth, B., Mitrovica, J. X.,
Hostetler, S. W., and A. M. McCabe, 2009. The Last Glacial Maximum. Science, 325: 710-
714.
Cochelin, Anne-Sophie B., L.A. Mysak, Z. Wang, 2006, Simulation of long-term future climate
changes with the green McGill paleoclimate model: The next glacial inception. Climate
Change, 79, 381-401.
Cook, E.R., R. Seager, R.R. Helm Jr, R. S. Vose, C. Herweijer, and C. Woodhouse, 2010.
Megadroughts in North America: Placing IPCC Projection of Hydroclimatic Change in a
Long-Term Palaeoclimate Context, Journal of Quaternary Science, 25, Issue 1, 48-
61.Currey, D.R., G. Atwood, and D.R. Mabey, Map 73 Major Levels of Great Salt Lake and
Lake Bonneville, Utah Geological and Mineral Survey, Salt Lake City, UT, May 1984
Curry, D.R., 1982. Lake Bonneville: Selected Features of Relevance to Neotectonic analysis.
U.S. Geological Survey Open-File Report 82-1070.
Davis, O.K., 1998. Palynological evidence for vegetation cycles in a 1.5 million yr pollen record
from the Great Salt Lake, Utah, U.S.A., Palaeogeography, Palaeoclimatology,
Palaeoecology, Vol. 138, pp. 175–185.
Driscoll, N. W. and G. H. Haug, 1998. A short circuit in thermohaline circulation: a cause for
Northern Hemisphere glaciation. Science, 282: 436-438.
Deep Time Assessment for the Clive DU PA
30 March 2020 54
Einselle, G. and M. Hinderer, 1997. Terrestrial sediment yield and the lifetimes of reservoirs,
lakes, and larger basins. Geologische Rundschau, 86: 288-310.
EPICA community members, 2004. Eight glacial cycles from an Antarctic ice core. Nature, 429:
623-628.
Faybishenko, B. 2006. Climatic Forecasting of Net Infiltration at Yucca Mountain Using
Analouge Meteorological Data. Lawrence Berkeley National Laboratory. DT#46762
QA:NA. MOL.20061204.0226.
Fritz, S.C., 1996, Paleolimnological records of climatic change in North America, Limnol.
Oceanogr. 45, 882-889.
GTG (GoldSim Technology Group), 2011. GoldSim: Monte Carlo Simulation Software for
Decision and Risk Analysis, http://www.goldsim.com
Harding, B.L., A.W. Wood, and J.R. Prarie, 2012. The Implications of Climate Change Scenario
Selection for Future Streamflow Projection in the Upper Colorado River Basin, Hydrol.
Earth Syst. Sci, 16, 3989-4007.
Hart, W. S., Quade, J., Madsen, D. B., Kaufmann, D. S., and C. G. Oviatt, 2004. The 87Sr/86Sr
Ratios of Lacustrine Carbonates and Lake-level History of the Bonneville Paleolake System.
GSA Bulletin, 116: 1107-1119.
Haug, G. H. and R. Tiedemann, 1998. Effect of the Formation of the Isthmus of Panama on
Atlantic Ocean Thermohaline Circulation. Nature, 393: 673-676.
Hays, J. D., Imbire, J., and N. J. Shackleton, 1976. Variations in the Earth's Orbit: Pacemaker of
the Ice Ages. Science, 194: 1121-1132.
IAEA. The Safety Case and Safety Assessment for the Disposal of Radioactive Waste. Vienna,
Austria: International Atomic Energy Agency; Specific Safety Guide No. SSG-23;
2012.Jansen, E. J. Overpeck, K.R. Briffa, J.-C. Duplessy, F. Joosv, V. Masson-Delmotte, D.
Olago, B. Otto-Bliesner, W.R. Peltier, S. Rahmstorf, R. Ramesh, D.. Raynaud, D. Rind, O.
Solomina, R. Villalba, and D. Zang, 2007, Palaeoclimate. In: Climate Change 2007: The
Physical Science Basis. Contribution of Working Group I to the fourth Assessment Report of
the Intergovernmental Panel on Climate Change [Solomon, s. D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tigor and H.L Miller (eds.)] Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
Jewell, P.W. 2010, River incision, circulation, and wind regime of Pleistocene Lake Bonneville,
USA, Palaeogeography, Palaeoclimatology, Palaeoecology, 293, 41-50.
Jewell, P.W., and K. Nicoll, Wind Regimes and Eolian Transport in the Great Basin, U.S.A.,
Geomorphology 129, 1-13 (2011).
Jouzel, J., Masson-Delmotte, V., Cattani, O., Dreyfus, G., et al., 2007. Orbital and millennial
Antarctic climate variability over the past 800,000 yr. Science, 317: 793-796.
Kukla, G. J., R. K. Matthews, J. M. Mitchell Jr., Quat. Res. 2, 261 (1972)
Deep Time Assessment for the Clive DU PA
30 March 2020 55
Lisiecki, L. E. and M. E. Raymo, 2005. A Pliocene-Pleistocene stack of 57 globally distributed
benthic δ18O records. Paleoceanography, 20, PA1003, doi:10.1029/2004PA001071.
Link, P. K., Kaufmann, D. S., and Thackray, G. D., 1999. Field guide to Pleistocene lakes
Thatcher and Bonneville and the Bonneville flood, southeastern Idaho, In: Hughes S. S. and
G. D Thackray (eds.), Guidebook to the Geology of Eastern Idaho, Idaho Museum of Natural
History, p. 251-266.
Lyle, M., L. Heusser, C. Rovelo, M. Yamamoto, J. Barron, N.S. Diffenbaugh, T. Herbert,
D. Andreasen, 2012, Out of the Tropics: The Pacific, Great Basin Lakes, and Late
Pleistocene Water Cycle in the Western United States, Science, Vol. 337, p. 1629–1633.
Machette, M.N., S.F. Personius, and A.R. Nelson, 1992. Paleoseismology of the Wasatch fault
zone: A summary of recent investigations, interpretations, and conclusions, in Gori, P.L., and
W.W. Hays, eds., Assessment of regional earthquake hazards and risk along the Wasatch
Front, Utah, U.S. Geological Survey Professional Paper 1500-A-J, pp. A1–A71, 1992.
Madsen, D.B., 2000. Late Quaternary Paleoecology in the Bonneville Basin, Utah Geological
Survey Bulletin 130, 2000.
Masson-Delmotte, V., Stenni, B., Pol, K., Braconnot, P., et al., 2010. EPICA Dome C record of
glacial and interglacial intensities. Quaternary Science Reviews, 29: 113-128.
Masson-Delmotte, V., M. Schulz, A. Abe-Ouchi, J. Beer, A. Ganopolski, J.F. Gonzalez Rouco, E.
Jansen, K. Lambeck, J. Luterbacher, T. Naish, T. Osborn, B. Otto-Bliesner, T. Quinn, R..
Ramesh, M. Rojas, X. Shao, and A. Timmermann, 2013, Information from Paleoclimate
Archive. In: Climate Change 2013: The Physical Science Basis. Contribution of Working
group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change,
Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia,
V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA.
Matsubara, Y. and A. D. Howard, 2009. A spatially explicit model of runoff, evaporation, and
lake extent: application to modern and late Pleistocene lakes in the Great Basin region,
western United States. Water Resources Research, 45, W06425,
doi:10.1029/2007WR005953.
Miller, D.M., C.G. Oviatt, and J.P. Mcgeehin, 2013. Stratigraphy and chronology of Provo
shoreline deposits and lake-level implications, Late Pleistocene Lake Bonneville, eastern
Great Basin, U.S.A., Boreas, Vol. 42, pp. 342–361, 2013.
Nash, W.P., 1990, Black Rock Desert, Utah, in C.A Woods and J. Kienle, eds. Volcanoes of
North America, Cambridge University Press, Cambridge p. 271-273.
Nelson, D.T., 2012. Geomorphic and Stratigraphic Development of Lake Bonneville’s
Intermediate Paleoshorelines during the Late Pleistocene, PhD Dissertation, University of
Utah, 237 p.
Neptune and Company, Inc., 2015a, Neptune field studies, December, 2014, Eolian depositional
history Clive Disposal Site.
Deep Time Assessment for the Clive DU PA
30 March 2020 56
Neptune and Company, Inc., 2015b, Final report the Clive DU PA Model, Clive DU PA Model
v1.5, Neptune and Company In., (November, 2015).
NRC (Nuclear Regulatory Commission), 1989. Calculation of Radon Flux Attenuation by
Earthen Uranium Mill Tailings Covers, U.S. Nuclear Regulatory Commission, Office of
Nuclear Regulatory Research, Task WM 503-4, June 1989.
Oviatt, C. G. and W. P. Nash, 1989. Late Pleistocene basaltic ash and volcanic eruptions in the
Bonneville basin, Utah. Geological Society of America Bulletin, 101: 292-303.
Oviatt, C. G., McCoy, W. D., and Nash, W. P., 1994a. Sequence stratigraphy of lacustrine
deposits: a Quaternary example from the Bonneville basin, Utah. Geological Society of
America Bulletin, 106: 133-144.
Oviatt, C.G., G.D. Habiger and J.E. Hay, 1994b. Variation in the Composition of Lake Bonneville
Marl: A Potential Key to Lake-Level Fluctuations and Paleoclimate, Journal of
Paleolimnology, Vol 11, p. 19-30.
Oviatt, C. G., 1997. Lake Bonneville fluctuations and global climate change. Geology, 25(2):
155-158.
Oviatt, C. G., Thompson, R. S., Kauffman, D. S., Bright, J., and R. M. Forester, 1999.
Reinterpretation of the Burmester core, Bonneville Basin, Utah. Quaternary Research, 52:
180-184.
Oviatt, C.G., 2002, Comparing the Shoreline and Offshore Sedimentary Records of Lake
Bonneville. Abstracts with Programs - Geological Society of America 34(6): 292.
Oviatt, C. G., D. M. Miller, J. P. McGeehin, C. Zachary, and S. Mahan, 2005. The Younger Dryas
phase of Great Salt Lake, Utah, USA, Palaeogeography, Palaeoclimatology, Palaeoecology,
219: 263-284.
Oviatt, C.G., and B. P. Nash, 2014, The Pony Express Basaltic Ash: A Stratigraphic Marker in
Lake Bonneville Sediments, Utah, Miscellaneous Publication 14-1, Utah Geological Survey
10 p.
Oviatt, C.G., Chronology of Lake Bonneville, 30,000 to 10,000 yr B.P. Quaternary Science
Reviews 110, 166-171, (2015).
Paillard, D., 2001. Glacial cycles: toward a new paradigm. Reviews of Geophysics, 39(3): 325-
346.
Paillard, D., 2006. What drives the Ice Age cycle? Science, 313: 455-456.
Ruddiman, W.F., 2006. Orbital Changes and Climate, Quaternary Science Reviews, Vol. 25,
pp. 3092–3112.Sack, D., 1999.The composite nature of the Provo Level of Lake Bonneville,
Great Basin, Western North America, Quaternary Research, v. 52, 316-327.
SC&A, Inc. 2015. Safety Evaluation Report: Volume I. for Utah Department of Environmental
Quality. April 2015.
Deep Time Assessment for the Clive DU PA
30 March 2020 57
Schnurrenber, D. J. Russell, and K Kelts, 2003, Classification of lacustrine sediments based on
sedimentary components, Journal of Paleolimnology 29, p. 141-154.
Schofield, I., P. Jewell, M. Chan, D. Currey, and M. Gregory, 2004, Shoreline development,
longshore transport, and surface water dynamics, Pleistocene Lake Bonneville, Utah, Earth
Surf. Process. Landforms, 29, 167-1690.
Tzedakis, P.C., E.W. Wolff, L.C. Skinner, V. Brovkin, d.A. Hodell, J.F. McManus, and D.
Raynaud, 2012a, Can we predict the duration of an interglacial? Climate Past 8, 1473-1485.
Tzedakis, P.C., J.E.T. Channel, D.A. Hodell, H.F. Kleiven, and L.C. Skinner, 2012b,
Determining the natural length of the current interglacial, Nature Geoscience, 5, 138-141.
United States Geological Survey (USGS), 2001. National Water Information System data (Water
Data for the Nation), accessed December, 2010 URL: http://waterdata.usgs.gov
Utah, State of, 2015, Utah Administrative Code Rule R313-25. License Requirements for Land
Disposal of Radioactive Waste - General Provisions. As in effect on September 1, 2015.
(http://www.rules.utah.gov/publicat/code/r313/r313-025.htm, accessed 5 Nov 2015).
Woodhouse, C.A., D. M. Meko, G. M. MacDonald, D.W. Stahle, and E. R. Cook, 2010. A
1,200-year Perspective of 21st Century Drought in Southwestern North America, Proc. Natl.
Acad. Of Sciences, 107, 21283-21288.

Deep Time Assessment for the Clive DU PA
30 March 2020 58
Appendix A
A.1 Clive Pit Wall Interpretation (C. G. Oviatt, unpublished data) and
stratigraphic comparison with quarry wall studies from Neptune
(2015a).

Deep Time Assessment for the Clive DU PA
30 March 2020 59

Deep Time Assessment for the Clive DU PA
30 March 2020 60
Appendix B
B.1 Knolls Core Interpretation (C. G. Oviatt, unpublished data)
Deep Time Assessment for the Clive DU PA
30 March 2020 61

Table B.1 Sediment Thickness Estimates for 100 ky Glacial Cycles (data from
C. G. Oviatt, personal communication, 2020)

Core Location Age Control Depth (m) Sediment
Thickness
Per 100 ky (m)
Burmester Lava Creek B Ash 86 13
Knolls Lava Creek B Ash 107 16
Saltair Lava Creek B Ash 167 26

Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 1 of 13 April 24, 2020

6.5 Deep Time Results
The deep time model addresses in a heuristic fashion the fate of the Federal DU Cell from 10 ky
to 2.1 My, the time at which DU reaches secular equilibrium. The model addresses the needs
identified in the Section 5(a) of R313-25-9 of the UAC to perform additional simulations for the
period where peak dose occurs, for which the results are to be analyzed qualitatively. Even
though the deep-time model runs to 2.1 My and there is huge uncertainty in predicting human
society and evolution that far into the future, rancher doses are calculated to provide a context for
radon fluxes which are calculated when no lake is present. The output of the deep time model is
also presented in terms of concentrations of radionuclides in relevant environmental media.
The deep-time model considers the return of lakes in the Bonneville Basin that reach or exceed
the elevation of Clive. Two classes of lakes are considered. The first is a deep lake similar to
Lake Bonneville that inundates the Clive facility. It is deep and adds to materials that are
currently on Bonneville Basin floor. This type of lake is assumed to occur once every 100 ky in
line with the 100-ky climate cycles that have occurred for the past 1 My or so. The second type
of lake is shallower and is termed an intermediate lake. It is also assumed to inundate the Clive
facility and adds sediment materials but is not a deep lake like Lake Bonneville. It is more
similar to the Gilbert Lake that occurred at the end of the last ice age. This type of lake is
assumed to occur several times in each climate cycle in response to colder, wetter conditions.
Return of a lake at or above the elevation of Clive is assumed to result in the destruction of the
Federal DU Cell. The above-grade embankment material and radionuclides are assumed to be
dispersed through wave action. The dispersal area forms the basis for the lake volume in which
radionuclides are dissolved and ultimately settle back to the basin floor through precipitation or
through evaporation as the lake recedes. The lake cycle involves movement of the radionuclides,
subject to continuing decay and ingrowth, from the sediment into lake water and back to
sediment as the lake forms and recedes. The dispersed radionuclides are assumed to be fully
mixed with the accumulated sediment. Sediment accumulates on average at the rate of about
17 m per 100-ky climate cycle. The current Unit 3 layer of sediment at Clive, which is derived
from Lake Bonneville, is assumed to be a confining layer.
The lake cycle effects on transport processes are complex. Sediment core records show
significant mixing of sediment, but also can be used to identify significant lake events in the past
several hundred thousand years. The extent of sediment mixing is not well understood. The
mechanisms for dispersal of a relatively soft pile of material in the middle of a desert flat are not
well understood. The extent of mixing of dissolved materials in a deep lake is also not well
understood. The Model, consequently, is simplified to the point of acknowledging lake return,
destruction of the Federal DU Cell, and cycling of radionuclides between periodic lakes and
basin sediments.
In particular, the model overly simplifies the lake cycle processes and the effect of those
processes on the transport of radionuclides. It limits the dispersal of radionuclides through time.
Destruction of the Federal DU Cell is assumed to occur with a lake that at least reaches the
elevation of Clive. This means that even a very shallow lake is assumed to destroy the
embankment. With the sediment acting as one large mixing cell, lake water diffusion can occur
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 2 of 13 April 24, 2020

across the entire depth of the sediment, no matter how deep. The simplified model ignores
increased precipitation and cooler conditions as the time of lake return approaches, which would
move radionuclides downwards in the sediment. With these simplifying assumptions, some
(perhaps unreasonably) high lake water and sediment concentrations are predicted by the Model.
The area of dispersal of the Federal DU Cell is captured with a simple distribution that reflects
the area of an intermediate lake. This fixes a dispersal area. Dissolution into the lake is assumed
to occur and to be mixed in the entire lake. The same dispersal area is used for both intermediate
and deep lakes, limiting both the volume of water within which dissolved materials might mix
and the area in which precipitates and evaporates can return.
Although the embankment material is dispersed within a specified dispersal area, isolation of any
part of the sediment profile is assumed not to occur. That is, the sediment is assumed to
completely mix with previous sediment for every lake event. Lake sedimentation does not allow
burial or isolation of previously formed sediment layers. Since different lakes can be identified in
sediment cores, this again limits the dispersal of the radionuclides.
The model, therefore, represents a closed system that cycles radionuclides from lake water to
sediment and back again. Decreased concentrations in sediment are obtained because of the
increased sediment load, but the mass of radionuclides available to diffuse into each lake is not
different in time, except from decay and ingrowth.
Deep Time Model results such as radon flux are considered in the context of gauging system
performance and may provide limited insight into the behavior of the disposal system in deep
time. Based on potential future radon fluxes, a rancher dose was calculated in deep time to
provide a context for the radon flux results, consistent with the rancher scenario from the first
10,000 years of the model.
Conceptually, deep time will result in a combination of repeated isolation of sediment layers and
more dispersal than modeled. This will cause mixing over ever increasing areas and volumes,
rather than mixing within a closed system. Consequently, concentrations of radionuclides will
decrease with each lake cycle and with each climate cycle. However, the constraints of the model
do not allow lake water concentrations to decrease with each cycle, and sediment concentrations
decrease only because of the additional mass of sediment within which the DU waste is mixed.
In light of the simplifications in the model, the results for the deep time scenario are presented
primarily within the first 100-ky cycle, in which the first intermediate or deep lake will return
and the Federal DU Cell will be obliterated. Consideration of model assumptions should be used
when interpreting results beyond the first 100-ky cycle. Summary statistics lake water
concentrations are presented at 82,500 years, which is the timestep at which the greatest
percentage of lakes is present in the first 100-ky cycle. The focus of the deep-time results is,
consequently, the effects of dispersal on concentrations of 238U and its progeny in lake water and
sediments within the first 100-ky climate cycle, as well as 222Rn flux and rancher dose after the
first lake recedes. Progeny of 238U presented include 230Th and 226Ra. Unless otherwise noted,
deep time results are presented for 1000 realizations in order to capture the temporal changes in
these results most clearly.
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 3 of 13 April 24, 2020

6.5.1 Sedimentation and Lake Timing Results
Thickness of the sediment above the DU waste is shown in Figure 1. The next lake to reach the
elevation of Clive is assumed to occur no sooner than 50 ky into the future, so only eolian
deposition, at a constant (uncertain) rate, contributes to accumulation of sediments in the vicinity
of Clive. Note that the embankment exists until the advent of the first lake, so the eolian
deposition thickness up to 50 ky is the only sediment accumulation in the vicinity of Clive.
When a lake reaches the Clive elevation, eolian deposition is augmented by the deposition of
lake-derived sediments. Because the number and timing of such lakes and the depth of deposited
sediment are uncertain, the variability in sediment thickness after 50 ky is considerably greater
than in the initial 50-ky modeling period. The change in the slope of the sediment thickness
curve at approximately 75 ky reflects the deposition of sediment from deep lakes that often
appear at this time within the 100-ky climate cycle.

Figure 1. Evolution of sediment thickness in deep time.
The increasing depth of material covering the disposed DU waste over time will result in
attenuation of radon flux. However, this rate of attenuation will be partly offset by the slowly
increasing activity of the radioactive progeny of 238U. Modeling results indicate that sediment
accumulation overwhelms the influence of progeny ingrowth. This is revealed by inspection of
the results of individual model realizations, where radon flux is always highest at the model time
just before the second intermediate lake appears and then decreases over time to the end of the
modeling period.
The time when the first intermediate lake returns after 50 ky is modeled as a Poisson process and
varies with each model realization. 100% of intermediate lakes occur within the first 85 ky of
the simulation. Deep lake start times are modeled as a log-normal distribution which occur
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 4 of 13 April 24, 2020

before 100 ky. As shown in Figure 2, the likelihood that the first lake to reach Clive has appeared
increases with time from 50 ky such that there is approximately a 80% probability that a lake
will have appeared by approximately 66 ky. Figure 3 displays the distribution for time of
appearance of second lake (intermediate or deep) to reach the Clive elevation.

Figure 2. Time of appearance of first intermediate lake to reach the Clive elevation.

Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 5 of 13 April 24, 2020

Figure 3. Time of appearance of second lake (intermediate or deep) to reach the Clive
elevation.

6.5.2 Lake Sediment Concentrations
Results are presented similarly in
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 6 of 13 April 24, 2020

Table 1 for concentrations of 238U and its progeny in sediment derived from successive lakes.
These results are statistical summaries of lake concentrations at 82.5 ky. The peak occurrence of
a lake across 10,000 realizations, the time at which a lake is most likely to be present, is at
82.5 ky. By that point in time, 230Th and 226Ra have ingrown sufficiently to be present in
concentrations greater than those of 238U.
Sediment thickness increases with time at a mean rate of about 16.5m per 100 ky climate cycle.
In the model the first lake is expected to appear in the second half of the current climate cycle;
i.e., no earlier than 50ky.

Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 7 of 13 April 24, 2020

Table 1. Statistical summary of deep time sediment concentrations at model year 82,500.
Based on 1000 realizations.
25th Percentile Median Mean 95th Percentile
U-238 sediment
concentration
(pCi/g)
6.4E-04 3.8E-03 2.5E-02 1.2E-01
Ra-226 sediment
concentration
(pCi/g)
4.4E-04 1.9E-03 6.6E-03 2.5E-02
Th-230 sediment
concentration
(pCi/g)
4.3E-04 1.8E-3 6.2E-02 2.5E-02

Time history plots of radionuclide concentrations in future lake sediments for 238U and its
progeny 230throium and 226radium are presented in Figure 4, Figure 5, and Figure 6, respectively,
over 2.1 My. These plots show a large increase in concentrations as a consequence of the first
lake event, with subsequent decreases as the sediment load increases.

Figure 4. Time history of concentrations of uranium-238 in sediments
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 8 of 13 April 24, 2020

Figure 5. Time history of concentrations of thorium-230 in sediments

Figure 6. Time history of concentrations of radium-226 in sediments
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 9 of 13 April 24, 2020

6.5.3 Lake Water Concentrations
A summary of lake water concentrations of 238U and some of its progeny are presented in Table
2. These results are statistical summaries of lake concentrations at 82.5 ky, the time at which a
lake is most likely to be present. By that point in time, 230Th and 226Ra have ingrown sufficiently
to be present in computable concentrations.
Table 2. Statistical summary of deep time lake concentrations at model year 82,500. Based
on 1000 realizations.
25th Percentile Median Mean 95th Percentile
U-238 lake
concentration
(pCi/L)
1.2E-06 5.6E-05 1.1E-02 5.9E-02
Ra-226 lake
concentration
(pCi/L)
5.7E-02 2.5E-01 6.8E-01 2.5E+00
Th-230 lake
concentration
(pCi/L)
5.6E-02 2.4E-01 6.5E-01 2.5E+00

Time history plots of lake water concentration statistics for 238U and its progeny, 230Th and 226Ra,
are presented in Figure 7, Figure 8, and Figure 9, respectively, across 2.1 My. These are
presented on a log scale to capture the full concentration range. The jagged nature of the plots is
due to the fact that lake water concentrations are zero when there is no lake present, and
intermediate lakes only occur on average 4 times per 100 ky. Peak lake water concentrations
tend to occur near the end of the period of the deep lake, which provides time for the
radionuclides to dissolve into the lake.
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 10 of 13 April 24, 2020

Figure 7. Time history of concentrations of uranium-238 in lake water

Figure 8. Time history of concentrations of thorium-230 in lake water
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 11 of 13 April 24, 2020

Figure 9. Time history of concentrations of radium-226 in lake water

6.5.4 Radon flux results (10,000 realizations)
A statistical summary of radon flux results after the first lake recedes are presented in Table 3.
The time when the first intermediate lake returns after 50 ky is modeled as a Poisson process and
varies with each model realization. Therefore, the time of peak radon flux also varies with each
realization. Median values are below the 10,000-yr timeframe regulatory limit of 20 pCi/m2s.
Table 3. Statistical summary of radon-222 flux concentrations.
Radon-222 flux (pCi/m2-s)
result mean median
(50th %ile) 95th %ile
After 1st lake 24.3 8.9 100.8
Before 2nd Lake 29.7 10.6 123.2
Results are based on 10,000 realizations, seed 1

Radon flux over time is shown in Figure 10. Although radon flux will be highest at times closest
to 50 ky, in most realizations a lake will not have occurred until closer to 60 ky. The change in
the slope of radon flux curve before 100 ky in Figure 10 reflects the deposition of sediment from
a deep lake that appears by this time within the 100-ky climate cycle.
Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 12 of 13 April 24, 2020

The peak of the mean radon flux shown in Figure 10 is approximately 13 pCi/m2-s. The peak
occurs in the Model at about 59,500 yr. This value is lower than the mean radon flux after the
first lake recedes (above) since that value occurs at various points in time and the mean flux in
Figure 10 is calculated for each point in time.
Although the median and mean sediment thickness track closely (Figure 1), the mean radon
ground surface flux is much larger than the median. This strongly skewed result for radon flux is
a consequence of the nonlinearities inherent in the NRC radon ground surface flux calculation.
These are equations (9) through (12) in NRC (1989), here reproduced without detailed
explanation:

The definitions of variables are available in the NRC Regulatory Guide (1989), but it is clear that
these equations will produce a highly nonlinear result, Jc, which is the ground surface flux of
radon. So, even though all the inputs to the calculation are essentially normal distributions, the
complexity of dividing one by another and involving powers (e.g. ex) and hyperbolic tangent,
produces a nonlinear result.

Attachment 2 to Clive DU PA Model – Response to Model Version 1.4 Amended Interrogatories
EnergySolutions DU PA Deep Time Results v1.5 Page 13 of 14 April 24, 2020

Figure 10. 222Rn ground surface flux in deep time.

6.5.5 Rancher radon results (10,000 realizations)
Doses to a rancher receptor are calculated to provide a context for the radon flux calculations,
using radon flux after the first lake recedes. The rancher exposure scenario provided the greatest
dose to a receptor in the Model from 10,000 years, so it was used here for comparison. Rancher
dose is less than 1 mrem/yr even at the 95th percentile of the results.
Table 4. Statistical summary of doses to ranchers.
Rancher dose after first lake recedes (mrem/yr)
simulation scenario mean median
(50th %ile) 95th %ile
After 1st lake 0.18 0.060 0.72
Before 2nd Lake 0.22 0.071 0.90
Results are based on 10,000 realizations, seed 1

One of the objectives of a PA, as defined in the UAC R313-25-9 is site stability. The
performance standard for stability requires the facility must be sited, designed, and closed to
achieve long-term stability to eliminate to the extent practicable the need for ongoing active
maintenance of the site following closure. If the intent is to minimize the need for ongoing active
maintenance, as stated, then obliteration of the Federal DU Cell in deep time achieves this goal,
since concentrations are low and the need to maintain the site disappears completely. In addition,
continued deposition through eolian processes in inter-glacial periods, or through lake deposition