DRC-2012-001974 - 0901a0688030cb2a

ORC-2012-001974
September 14,2012
VIA E-MAIL AND OVERNIGHT DELIVERY
Mr Rusty Lundberg
Director
Utah Division of Radiation Control
195 North 1950 West
PO Box 144850
Salt Lake City, UT 84114-4850
Re Radioactive Materials License UT 1900479, Response to Utah Division of Radiation Control
("DRC") Round 1 Interrogatory for the Revised Infiltration and Contaminant Transport Modeling
Report of March 2010
Dear Mr Lundberg*
Energy Fuels Resources (USA) Inc (EFR) transmitted on May 31, 2012 and August 15, 2012 our
responses to DRC's Round 1 Interrogatories for the White Mesa Mill Reclamation Plan Revision 5 0
This letter transmits EFR's responses to DRC's Round 1 Interrogatories on the March 2010 Revised
Infiltration and Contaminant Transport Modeling Report
We have included with this transmittal two hard copies and two copies in pdf format on CD ROM
Please contact me if you have any questions or require any further information
Yours very truly,
ENERGY FUELS RESOURCES (USA) INC.
Jo Ann Tischler
Director, Compliance
cc David C Frydenlund
DanL Hillsten
John Hultquist DRC
Harold R Roberts
David E Turk
Katherine A Weinel
Energy Fuels Resources (USA) Inc.
Lakewood, CO 80228
225 Union Boulevard, Suite 600
Phone. [phone]

ENERGY FUELS RESOURCES (USA) INC.

RESPONSES TO INTERROGATORIES –
ROUND 1 FOR THE REVISED
INFILTRATION AND CONTAMINANT
TRANSPORT MODELING REPORT, MARCH
2010;

SEPTEMBER 10, 2012

September 10, 2012

TABLE OF CONTENTS
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10CFR40 APPENDIX A, CRITERION
6(1); INT 01/1: INCONSISTENCIES BETWEEN REVISED ICTM REPORT AND RECLAMATION
PLAN REV 5.0 ............................................................................................................................................. 1 
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10CFR40 APPENDIX A, CRITERION
6(1); INT 02/1: COMPARISON OF COVER DESIGNS, SENSITIVITY ANALYSES, ‘BATHTUB’
ANALYSIS, AND RADON EMANATION MODELING .......................................................................... 7 
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10 CFR40 APPENDIX A,
CRITERION 6(1); INT 03/1: MOISTURE STORAGE CAPACITY OF COVER .................................. 31 
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10 CFR40 APPENDIX A,
CRITERION 1; INT 04/1: EVALUATION OF POTENTIAL FLOW THROUGH TAILINGS CELL
LINERS ...................................................................................................................................................... 35 
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4 -05/1: CONTAMINANT
TRANSPORT MODELING ....................................................................................................................... 46 

September 10, 2012
Interrogatory 01/1: R313-24-4; 10CFR40 Appendix A Criterion 6(1): Inconsistencies between Revised ICTM Report and Reclamation Plan Rev 5.0 Page 1 of 70
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10CFR40 APPENDIX A,
CRITERION 6(1); INT 01/1: INCONSISTENCIES BETWEEN REVISED ICTM REPORT AND
RECLAMATION PLAN REV 5.0
REGULATORY BASIS:
UAC R313-24-4 invokes the following requirement from 10CFR40, Appendix A, Criterion 6(1): In disposing
of waste byproduct material, licensees shall place an earthen cover (or approved alternative) over
tailings or wastes at the end of milling operations and shall close the waste disposal area in accordance
with a design which provides reasonable assurance of control of radiological hazards to (i) be effective
for 1,000 years, to the extent reasonably achievable, and, in any case, for at least 200 years, and (ii) limit
releases of radon-222 from uranium byproduct materials, and radon-220 from thorium byproduct
materials, to the atmosphere so as not to exceed an average release rate of 20 picocuries per square
meter per second (pCi/m2s) to the extent practicable throughout the effective design life determined
pursuant to (1)(i) of this Criterion. In computing required tailings cover thicknesses, moisture in soils in
excess of amounts found normally in similar soils in similar circumstances may not be considered. Direct
gamma exposure from the tailings or wastes should be reduced to background levels. The effects of any
thin synthetic layer may not be taken into account in determining the calculated radon exhalation level. If
non-soil materials are proposed as cover materials, it must be demonstrated that these materials will not
crack or degrade by differential settlement, weathering, or other mechanism, over long-term intervals.

INTERROGATORY STATEMENT:
Refer to Executive Summary, Section 2.1, Figures 2-2 and 3-1, Table 3-1, and Appendices D through
N of the ICTM Report Rev 2:

1. Revise the description of the proposed evapotranspiration (ET) cover, including revised cover
material characteristics (e.g., soil textures [percent clay content, etc…], expected in-place
saturated soil layer hydraulic conductivities, particle size distributions, porosities and bulk
densities) for each layer of the cover and revised thicknesses, where applicable, to be consistent
with the ET cover description that will be presented in the next revision of Reclamation Plan Rev.
5.0 reflecting the responses to comments contained in the Round 1 Interrogatories submitted on
the Reclamation Plan rev. 5.0 and these Round 1 interrogatories. Update Figures 2.2 and 3-1 to
reflect the ET cover thicknesses and materials and to be consistent with the descriptions to be
provided in the updated Reclamation Plan.
Response 1 (May 31, 2012):
The Revised Infiltration and Contaminant Transport Modeling (ICTM) Report was
submitted during March 2010 and was meant to introduce the conceptual design of an
evapotranspiration (ET) cover. Infiltration modeling contained within the 2010 Revised
ICTM Report indicated that the design and construction of a monolithic ET cover is the
preferred alternative for infiltration control. The construction of an ET cover as proposed
in the 2010 Revised ICTM Report was in contrast to previous iterations of the
Reclamation Plan that were based on the cover design from 1996. Therefore, the
Reclamation Plan Revision 5.0 submitted during September 2011 used the initial March
2010 conceptual design of the ET cover as a starting point, but modified some of the
material descriptions and thicknesses to provide an update to the analyses based on
additional information subsequently collected after the 2010 Revised ICTM Report was
submitted. The gap in time between publication of these two reports, combined with the
collection of additional soils data between March 2010 and September 2011, explains
September 10, 2012
Interrogatory 01/1: R313-24-4; 10CFR40 Appendix A Criterion 6(1): Inconsistencies between Revised ICTM Report and Reclamation Plan Rev 5.0 Page 2 of 70
the discrepancy between reports. Consequently, the material layering, thicknesses, and
physical characteristics presented in the next iteration of the ICTM Report including text,
figures, and tables will be consistent with next iteration of the Reclamation Plan.

2. Update analyses in the referenced Appendices to reflect ET cover characteristics that are
consistent with the descriptions to be given in the next revision of the Reclamation Plan Rev 5.0.
Response 2 (May 31, 2012):
In addition to the response to Comment One of this interrogatory, the analyses
presented in the next iteration of the ICTM Report including information and analyses
presented in the appendices will be consistent with the next iteration of the Reclamation
Plan. Applicable appendices that would be updated include Appendices E, F, G, H, and
N. However, as discussed later in this interrogatory response document, we propose
eliminating Appendix F from the next iteration of the Report to minimize confusion.

3. Provide an updated Appendix D (Vegetation Evaluation for the Evapotranspiration Cover) that
reflects information to be presented in the next revision of the Reclamation Plan Rev. 5.0 on
vegetation occurrence and the proposed revegetation plan and that addresses the additional
considerations and additional information described or requested in “INTERROGATORY
WHITE MESA RECPLAN REV 5.0 R313-24-4; 10CFR40 APPENDIX A; INT 11/1:
VEGETATION AND BIOINTRUSION EVUALATION AND REVEGETATION PLAN”.
Response 3 (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
The next iteration of the ICTM Report will include an updated Appendix D that reflects
the request for additional information (i.e., vegetation occurrence and proposed
revegetation plan). Revisions to Appendix D will be consistent with material presented in
Attachment G included with Denison’s August 15, 2012 Responses to the Reclamation
Plan, Revision 5.0 Interrogatories (Denison, 2012). Overall, supporting text was updated
to:
• Include the results of a plant and burrowing animal survey that was completed at
the mill site during June 2012.
o The results of the plant survey were used to support the range of percent
vegetative cover and root density/distribution for plant species that are
expected to occur on the cover during the design performance period.
o The plant survey, and the similarity in environmental conditions between
Monticello and White Mesa, suggests that a plant cover estimate of 40%
is a reasonable estimate for a long-term average, while a percent plant
cover of 30% is a reasonable estimate for a reduced performance
scenario. The root density/distribution for plants species expected to
occur on the cover is summarized in Table 01/1/3-1.
• Include a discussion regarding the sustainability of the cover system as it relates
to potential climate change and plant community succession and potential for
species colonization.
o From the review of climate change literature applicable to the southwest
United States and an analysis of the impact of various climate change
scenarios, the most likely plant community type that will be maintained
September 10, 2012
Interrogatory 01/1: R313-24-4; 10CFR40 Appendix A Criterion 6(1): Inconsistencies between Revised ICTM Report and Reclamation Plan Rev 5.0 Page 3 of 70
throughout the 200 to 1,000 year performance period is a community
dominated initially by cool season grasses, with a long-term transition to
dominance by warm season grasses as atmospheric CO2 and
temperature continues to increase and precipitation decreases and shifts
from winter storage to pulse dominated.
o While shrubs such as big sagebrush could establish through natural
succession during the short-term, it is unlikely that big sagebrush will be
sustainable on site, but it is likely to establish through natural succession
before the effects of climate change alter the environment.
Table for Response 3 (September 10, 2012):
Table 01/1/3-1. Root biomass for species expected to occur on the cover system
Depth
Root Biomass
Anticipated Performance
(g/cm3)
Root Biomass
Reduced Performance
(g/cm3)
0-15 0.11 0.04
15-30 0.17 0.12
30-45 0.035 0.02
45-60 0.023 0.015
60-75 0.021 0.014†
75-90 0.019 0.0
90-107 0.011 0.0
Note: † Maximum rooting depth under the reduced performance scenario would be 68 cm.

Reference for Response 3 (September 10, 2012):
Denison Mines (USA) Corp. 2012. Responses to Interrogatories – Round 1 for
Reclamation Plan, Revision 5.0, March 12. August 15, 2012.
4. For Appendix E (Comparison of Cover Designs Based on Infiltration Modeling), Appendix F
(Evaluation of the Effects of Storm Intensity on Infiltration through Evapotranspiration Cover),
Appendix G (Sensitivity Analysis Comparing Infiltration Rates through the Evapotranspiration
Cover Based on Vegetation, Biointrusion, and Precipitation), and Appendix H (Radon Emanation
Modeling for the Evapotranspiration Cover):
a. Provide revised discussion of the impacts of the results of an updated frost penetration
calculation and the maximum predicted frost penetration depth for the cover system
b. Provide revised discussion and revised infiltration analyses to:
i. Reflect the results of the updated frost penetration depth analysis requested in
“INTERROGATORY WHITEMESA RECPLAN 5.0 R313-24-4; 10CFR40,
APPENDIX A, CRITERION 6; INT 10/1: TECHNICAL ANALYSES - FROST
PENETRATION ANALYSIS”
ii. Address the additional considerations and additional information described or
requested in “INTERROGATORY WHITE MESA REV’D ICTM R313-24-4;
10CFR40 APPENDIX A, CRITERION 6(1); INT 02/1: COMPARISON OF
COVER DESIGNS, SENSITIVITY ANALYSES, ‘BATHTUB’ ANALYSIS, AND
RADON EMANATION MODELING”
Response 4a (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.

September 10, 2012
Interrogatory 01/1: R313-24-4; 10CFR40 Appendix A Criterion 6(1): Inconsistencies between Revised ICTM Report and Reclamation Plan Rev 5.0 Page 4 of 70
Frost Penetration
The frost penetration analysis for the tailings cover system was revised and presented in
Denison (2012a) to address interrogatories for the Reclamation Plan, Revision 5.0
(DRC, 2012). This analysis again will be updated after approval of the conceptual final
cover design is obtained. Revisions will be completed to be consistent with the revised
cover design presented in the August 15, 2012 responses to the interrogatories for the
Reclamation Plan, Revision 5.0 (Denison, 2012b). The frost penetration analysis
requires revision to incorporate additional data collected from a site investigation
conducted on April 19, 2012 to further evaluate cover borrow materials. It is anticipated
that the results of the updated analyses will be similar to the analyses presented in
Denison (2012a), with a frost penetration depth on the order of 81 cm (32 in).
Pedogenic Processes
Soil cover layers and their respective hydraulic and physical material properties
potentially could be affected from wet/dry, freeze/thaw, and other pedogenic processes
as suggested by Benson et al. (2011). However, as noted in Benson et al. (2011),
potential changes to the cover can be minimized by designing the cover system to be as
close as practical to the anticipated equilibrium state under long-term conditions;
furthermore, their study also noted that long-term changes are more prone to occur for
less permeable soils compared to more permeable soils. Because the frost penetration
depth is not anticipated to exceed the depth of the erosion protection and water storage
layers (combined depth of 107 cm), a minor increase or decrease in the frost penetration
depth will not affect the radon barrier and grading layers that are located beneath 107
cm. Therefore, any potential modifications to the hydraulic and physical properties of the
cover that could be influenced by freeze/thaw processes would be restricted to the
erosion protection and water storage layers.
As explained in more detail in interrogatory 02/1, response 1 of this document, the
hydraulic test results for the soils stockpiled at White Mesa are within the range of
parameter values anticipated to occur long-term as noted by Benson et al. (2011).
Based on this comparison, and the relatively permeable nature of the soils, corrections
to account for potential pedogenic processes are not warranted at this time because the
physical and hydraulic properties at the emplaced conditions are such that post-
construction changes should be minimal.
Response 4b (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
See Response 4a regarding the effect of pedogenic processes (including freeze/thaw)
on the infiltration modeling.
As explained in more detail in interrogatory 02/1, response 1 of this document, the cover
model has been updated from those presented in the 2010 Revised ICTM Report to
reflect additional laboratory test results and revisions to the cover design as presented in
Denison (2012b). In regard to the next iteration of the ICTM Report, applicable
appendices that would be updated include Appendices E, F, G, H, and N. However, we
propose eliminating Appendix F from the next iteration of the Report to minimize
confusion.
Revisions to the amount of percolation that may recharge the tailings affects the “bathtub
effect” calculations, and are discussed in interrogatory 02/1, response 1 in this
September 10, 2012
Interrogatory 01/1: R313-24-4; 10CFR40 Appendix A Criterion 6(1): Inconsistencies between Revised ICTM Report and Reclamation Plan Rev 5.0 Page 5 of 70
document. These potential effects and resultant calculations will be incorporated into
the next iteration of the ICTM Report.
References for Response 4 (September 10, 2012):
Benson, C.H. W.H. Albright, W.H., Fratta, D.O.,Tinjum, J.M., Kucukkirca, E., Lee, S.H.,
J. Scalia, J., Schlicht, P.D., and Wang, X. 2011. Engineered Covers for Waste
Containment: Changes in Engineering Properties and Implications for Long-Term
Performance Assessment(in 4 volumes). NUREG/CR-7028, Prepared for the
U.S. Nuclear Regulatory Commission, Washington, D.C., December 2011.
Denison Mines (USA) Corp. 2012a. Responses to Interrogatories – Round 1 for
Reclamation Plan, Revision 5.0, March 12. May 31, 2012.
Denison Mines (USA) Corp. 2012b. Responses to Interrogatories – Round 1 for
Reclamation Plan, Revision 5.0, March 12. August 15, 2012.
Utah Department of Environmental Quality, Division of Radiation Control (DRC), 2012.
Denison Mines (USA) Corp’s White Mesa Reclamation Plan, Rev. 5.0;
Interrogatories – Round 1. March 2012.
5. For Appendices K through N, provide updated/revised information and/or results to reflect
updated information and results provided as requested for Appendices E through H in Items 1
through 4 of this interrogatory.
Response 5 (May 31, 2012):
The updated frost penetration analysis will not affect the analyses presented in Appendix
K (tailings pore water source term chemistry). For the calculations presented in
Appendix L (potential water flux rates through the liners), the flux rates predicted at the
end of dewatering were assumed to equal the rate during post-closure conditions. This
approach was incorporated as an appropriate simplification based on discussions and
comments previously received by the Division (DRC, 2009) and responses provided by
MWH (2009). Additionally, the upper boundary condition assigned in Appendix M
(reactive flow and transport model through the bedrock vadose zone) used the flux
calculations presented in Appendix L. Therefore Appendices K, L, and M do not require
revision based on modifications to the frost penetration analysis. Appendix N (model
input/output files) will require revision once the infiltration modeling is updated and a
future iteration of the ICTM Report is submitted.
References for Response 5 (May 31, 2012):
Utah Division of Radiation Control (DRC), 2009. White Mesa Uranium Mill, Ground
Water Discharge Permit No. UGW370004, Infiltration and Contaminant Transport
Modeling Report: DRC Review Comments, Request for Additional Information,
Letter from Thomas Rushing of DRC to David C. Frydenlund of Denison Mines
dated February 2, 2009.
MWH, 2009. Revised responses to the Utah Division of Radiation Control’s February 2,
2009 comments on the Denison Mines (USA) Corp. November 2007 Infiltration
and Contaminant Transport Report, Letter from Ryan Jakubowski & Douglas
Oliver of MWH to David Frydenlund & Harold Roberts of Denison Mines dated
December 1, 2009.

September 10, 2012
Interrogatory 01/1: R313-24-4; 10CFR40 Appendix A Criterion 6(1): Inconsistencies between Revised ICTM Report and Reclamation Plan Rev 5.0 Page 6 of 70
BASIS FOR INTERROGATORY:
Section 3.3 and Figures 2.2 and 3-1 of the revised ICTM Report present the thickness of the ET cover as
9.3 feet extending from the cover surface to the top of the tailings. The Reclamation Plan, Rev. 5.0
(Section 3.2.2, Appendix G, and Figure 1-1 of Appendix G), describes the ET cover as being 9 feet thick
from the cover surface to the top of the tailings. Revisions need to be made to the ICTM Report to be
consistent with the ET cover details to be presented in the next revision of the Reclamation Plan Rev. 5.0.
Also, the description of the materials comprising the ET tailings cover design is different in the ICTM
Report than in the Reclamation Plan Rev. 5.0. The ICTM describes the ET tailings cover design from top
to bottom as follows:
• 0.5 ft (15 cm) Erosion Protection Layer (gravel-amended topsoil mixture)
• 3.5 ft (107 cm) Water Storage/Biointrusion/Frost Protection/Radon Attenuation Layer (random
fill soil [sandy clayey silt])
• 2.8 ft (75 cm) Radon Attenuation Layer (highly compacted loam to sandy clay
• 2.5 ft (75 cm) Radon Attenuation and Grading Layer (random fill soil [sandy clayey silt])loam to
sandy clay

However, Figure 1-1 of the Reclamation Plan Rev. 5.0 describes the water storage/biointrusion/frost
protection/radon attenuation layer as a loam to sandy clay with the radon attenuation layer being
comprised of highly compacted loam to sandy clay The intended proposed tailings cover design needs to
be made consistent for the ICTM Report and the next revision of the Reclamation Plan Rev. 5.0.
Finally, on page E-2, it is stated that "TITAN Environmental (1996) completed a freeze-thaw evaluation
based on site-specific conditions which indicated that that the anticipated maximum depth of frost
penetration was 6.8 inches (0.6 ft)." The frost penetration depth estimate presented by TITAN
Environmental (1996) is out of date and needs to be replaced with an updated frost penetration depth
calculation.
Refer to the Basis for Interrogatory sections in “INTERROGATORY WHITEMESA RECPLAN 5.0 R313-
24-4; 10CFR40, APPENDIX A, CRITERION 6; INT 10/1: TECHNICAL ANALYSES - FROST
PENETRATION ANALYSIS”, “INTERROGATORY WHITE MESA REV’D ICTM R313-24-4; 10CFR40
APPENDIX A, CRITERION 6(1); INT 02/1: COMPARISON OF COVER DESIGNS, SENSITIVITY
ANALYSES, ‘BATHTUB’ ANALYSIS, AND RADON EMANATION MODELING” and
“INTERROGATORY WHITE MESA RECPLAN REV 5.0; R313-24-4; 10CFR40 APPENDIX A; INT 11/1:
VEGETATION AND BIOINTRUSION EVUALATION AND REVEGETATION PLAN” for additional
information and bases for this interrogatory.

REFERENCES:
Denison Mines (USA) Corp. 2010. Revised Infiltration and Contaminant Transport Modeling Report,
White Mesa Mill Site, Blanding, Utah (Revision 2), March 2010.
Denison Mines (USA) Corp. 2011. Reclamation Plan, White Mesa Mill, Blanding, Utah, Radioactive
Materials License No. UT1900479, Revision 5.0, September 2011.

September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 7 of 70
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10CFR40 APPENDIX A,
CRITERION 6(1); INT 02/1: COMPARISON OF COVER DESIGNS, SENSITIVITY
ANALYSES, ‘BATHTUB’ ANALYSIS, AND RADON EMANATION MODELING
REGULATORY BASIS:
UAC R313-24-4 invokes the following requirement from 10CFR40, Appendix A, Criterion 6(1): In disposing
of waste byproduct material, licensees shall place an earthen cover (or approved alternative) over
tailings or wastes at the end of milling operations and shall close the waste disposal area in accordance
with a design which provides reasonable assurance of control of radiological hazards to (i) be effective
for 1,000 years, to the extent reasonably achievable, and, in any case, for at least 200 years, and (ii) limit
releases of radon-222 from uranium byproduct materials, and radon-220 from thorium byproduct
materials, to the atmosphere so as not to exceed an average release rate of 20 picocuries per square
meter per second (pCi/m2s) to the extent practicable throughout the effective design life determined
pursuant to (1)(i) of this Criterion. In computing required tailings cover thicknesses, moisture in soils in
excess of amounts found normally in similar soils in similar circumstances may not be considered. Direct
gamma exposure from the tailings or wastes should be reduced to background levels. The effects of any
thin synthetic layer may not be taken into account in determining the calculated radon exhalation level. If
non-soil materials are proposed as cover materials, it must be demonstrated that these materials will not
crack or degrade by differential settlement, weathering, or other mechanism, over long-term intervals.

INTERROGATORY STATEMENT:
1. Please refer to Sections 3-1, 4-1 and Appendix E, F, G, and H of the ICTM Report: Please
provide the following:
a. Provide additional information to justify the assumed cover soil layer properties, including the
value of porosity of 0.25 in Table H-3 for the Erosion Protection Layer. Demonstrate that the
values used in modeling appropriately reflect: (a) the composition and characteristics of the soil
and gravel components of the admixture layer and of other layers in the cover system; and (b) the
level of compaction proposed for each cover layer (see also “INTERROGATORY WHITEMESA
RECPLAN REV 5.0 R313-24-4; 10CFR40, APPENDIX A, CRITERION 6(4); INT 12/1: REPORT
RADON BARRIER EFFECTIVENESS).

Response 1a (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
ET Cover Material Properties
MWH has completed site-specific tests to evaluate the hydraulic and physical properties
of stockpiled materials that will be used to construct the ET cover; therefore the cover
soil layer material/hydraulic properties are updated from values previously used to
support the Revised ICTM Report. The testing of borrow source materials was
performed by the Wisconsin Geotechnics Laboratory at the University of Wisconsin-
Madison. The laboratory testing program consisted of two phases:
• Phase I testing consisted of standard material characterization tests, including
Atterberg limits, specific gravity, and full gradation. The test results were
reviewed, along with the existing soils data, to refine the Phase II laboratory
testing program.
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 8 of 70
• Phase II testing consisted of standard Proctor compaction, water retention
characteristic testing, and saturated hydraulic conductivity tests in the vertical
direction (Ks). The moisture retention and hydraulic conductivity tests were
compacted to 85% of maximum dry unit weight and optimum water content from
the standard Proctor compaction tests. Low relative compaction was used to
simulate a structured soil representing a longer term condition, as suggested in
Benson et al. (2011).
The results of the laboratory testing program established to evaluate the hydraulic and
physical properties of cover materials is summarized below. The results of the Phase I
laboratory testing program (standard material characterization tests) were reviewed,
along with the existing soils data, to define the Phase II laboratory testing program
(compaction, water retention, and hydraulic conductivity tests). This information, along
with the original laboratory data summary reports, is included in Attachment B to
Denison (2012). Evaluation of the index tests identified three soil groups, and the results
of the Phase I program were then used to select individual samples that would bracket
the range in material properties within the three soil groups:
• Group B: These materials are more broadly (B) graded with some to no plasticity,
and represent approximately 48% by volume of the existing stockpiles. On
average, these materials contain approximately 30% gravel. This group is
represented by samples W2-B1/2 and W5-B1/2 because they bracket the particle
size distribution for the plastic soil samples, and by sample W8-B1/2 to represent
the non-plastic soil samples.
• Group U: These materials are more uniformly (U) graded with some plasticity,
and represent approximately 47% by volume of the existing stockpiles. On
average, these materials contain approximately 2% gravel. This group is
conservatively represented by sample W9-B1/2 because it contains the highest
amount of gravel for the soil samples within Group U. The topsoil samples fall
within Soil Group U and are represented by sample E1-A1/2 which has a
measured plasticity index (PI) equal to the average of the PI values for all the
topsoil samples and the gradation of the sample represented the average
gradation of the topsoil samples.
• Group F: These materials are fine textured (F), plastic, and represent
approximately 5% by volume of the existing stockpiles. On average, these
materials contain approximately 0.5% gravel. This group is conservatively
represented by sample E3-A1/2 because it contains the lowest amount of fines
for the clay samples.
The results of the Phase II program were then categorized to bracket the range in
hydraulic properties. The hydraulic characterization tests identified three hydraulic
classification groups for the cover soils as summarized in Table 02/1/1-1, the results of
which were used to parameterize the cover model:
• High hydraulic conductivity and low storage. This group would be considered as
an upper bound scenario since water flow through the cover should be higher,
and storage should be low, compared to the other soil samples. This group is
represented by sample W2-B1/2.
• Intermediate hydraulic conductivity and storage. This group would be considered
as a base case (average) scenario since water flow through the cover should be
intermediate compared to the other soil samples. This group is represented by an
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 9 of 70
average of the soil sample test results (W2-B1/2, W5-B1/2, W8-A1/2, and W9-
B1/2).
• Low hydraulic conductivity and high storage. This group would be considered as
a lower bound scenario since water flow through the cover should be slower, and
storage should be high, compared to the other soil samples. This group is
represented by sample W9-B1/2.
The topsoil samples are represented by sample E1-A1/2.
The hydraulic classification groups were used to parameterize the cover model for the
proposed ET cover. The erosion protection layer was represented by sample E1-A1/2,
the water storage layer was represented by the samples identified above, and the radon
barrier layer and grading layer were represented by the samples identified above after
applying a correction factor to account for a decrease/increase in the saturated water
content. The correction factor was calculated from the anticipated change in porosity for
the different compaction efforts. In the model, the values for alpha and n were not
changed to maintain a conservative approach for simulation of the radon barrier layer.
While increasing compaction would lower alpha by reducing the largest pore size, and
lower n by making a more uniform pore size distribution, keeping these parameter
values constant will simulate a soil that releases water more readily than a soil at higher
compaction (Tinjum et al., 1997). Additionally, the Ks values for the radon layer were
assumed to equal that for the water storage layer to maintain a conservative approach.
The Ks of the radon barrier layer will be lower near term and perhaps long term, and
assuming a higher Ks is conservative. Parameter values for the grading layer were
assumed to equal those for the water storage layer. Effects on the grading layer, which
has a lower relative compaction compared to the water storage and radon barrier layers,
are not anticipated to affect the model results since this layer is located below the
terminal root zone and low permeability radon barrier layer. The three scenarios and
corresponding input values are summarized in Table 02/1/1-2 for the proposed ET cover
design. The material layering and thicknesses of the cover design are based on those
determined to be in compliance with radon emanation at the surface as evaluated by the
RADON model (Denison, 2012):
• The erosion protection layer will be placed at 85% standard Proctor at optimum
water content and will include 25% gravel as add-in to the topsoil materials.
• The water storage layer will be placed at 85% standard Proctor at optimum water
content and will include variable gravel contents based on the material
distribution for the soils.
• The radon barrier layer will be placed at 95% standard Proctor at optimum water
content and will include variable gravel contents based on the material
distribution for the soils.
• The grading layer will be placed at 80% standard Proctor at optimum water
content and will include variable gravel contents based on the material
distribution soils.
Soil cover layers and their respective hydraulic and physical material properties
potentially could be affected by wet/dry, freeze/thaw, and other pedogeneic processes
as suggested by Benson et al. (2011). However, as noted in Benson et al. (2011),
potential changes to the cover can be minimized by designing the cover system to be as
close as practical to the anticipated equilibrium state under long-term conditions;
furthermore, their study also noted that long-term changes are more prone to occur for
less permeable soils compared to more permeable soils. The hydraulic test results for
the soils stockpiled at White Mesa are within the range of parameter values anticipated
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 10 of 70
to occur long-term as noted by Benson et al. (2011), indicating that the hydraulic
properties used as input are likely to represent long-term conditions. For example, soil
that will be used to construct the water storage layer reported a range of values as
follows:
• The hydraulic conductivity values at White Mesa range between 35 and 130 cm/d
while those reported by Benson et al. range between 0.86 and 43 cm/d (with a
recommended value equal to 4.3 cm/d).
• The saturated volumetric water contents at White Mesa range between 0.23 and
0.40 while those reported by Benson et al. range between 0.35 and 0.45 (with a
recommended value equal to 0.40).
• The alpha values at White Mesa range between 0.0073 and 0.022 cm-1 while
those reported by Benson et al. range between 0.001 and 0.033 cm-1 (with a
recommended value equal to 0.02 cm-1).
• The n values at White Mesa range between 1.26 and 1.32 while those reported
by Benson et al. range between 1.1 and 1.5 (with a recommended value equal to
1.3).
Based on this comparison, and the relatively permeable nature of the soils, corrections
to account for potential pedogenic processes are not warranted at this time because the
physical and hydraulic properties at the emplaced conditions are such that post-
construction changes should be minimal. Furthermore, the soil properties for the water
storage layer are similar to data collected at the Monticello site for long-term conditions
that accounted for pedogenic processes (Ks of 13 cm/d; saturated volumetric water
content of 0.41; alpha of 0.0021 cm-1; and n of 1.30) as reported in Benson et al. (2008).
Possible Future Climate Conditions
The climatic dataset for the period of record for the Blanding weather station was
summarized to provide insight into future climate conditions that may occur at White
Mesa. Precipitation (P) and potential evapotranspiration (PET) data were tabulated to
define a range of possible future climate conditions, in the context of the historical
record, which may reasonably be expected to occur during the performance period of
the cover system. These contemporary data are then evaluated in the context of
inferred paleoclimate records and paleorecharge studies to justify a range of future
climate conditions that are reasonably anticipated to occur within 200 years and up to
1,000 years in the future. The use of paleoclimate in predicting future climate conditions
implicitly assumes that past climatic states can be used to infer future climatic states,
and that proxies used to infer paleoclimate or to evaluate paleorecharge provides an
accurate representation of past climatic processes.
The precipitation (P) and potential evapotranspiration (PET) dataset for the period of
record for the Blanding weather station (November 1, 1904 through March 31, 2011)
were obtained from the Utah Climate Center on April 12, 2012. The Revised ICTM
Report used data obtained from the Utah Climate Center on January 25, 2007 such that
summary data presented in the Revised ICTM Report should be fairly consistent with the
data presented below. Precipitation represents daily measurements while PET
represents a daily calculated value using Hargreaves’ Equation (Allen et al., 1998). The
Blanding weather station is located immediately west of Blanding at an elevation of
approximately 1,841 meters above mean sea level. As a comparison, the mill is located
slightly south of Blanding at an elevation of approximately 1,707 meters.
The daily data were summed for each month of the year and the corresponding monthly
tabulations were summed to generate annual (calendar) and seasonal (winter) values.
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 11 of 70
Months that had greater than five missing measurements were not included in the
summations, to be consistent with the approach used by the Western Regional Climate
Center. The period from November through February was selected to represent the
amount of winter precipitation, with nomenclature consistent with the definition of a water
year. These months were selected as the critical time period in which percolation may
occur since the difference between the long-term monthly averaged values for P and
PET was negative (December and January) or slightly positive (November and
February). Summary statistics are tabulated in Table 02/1/1-3 to illustrate the range in P
and PET throughout the period of record. No statistical relationship was found between
the annual amount of P and PET. A weak negative correlation was found between the
amount of winter P and winter PET such that winters with higher amounts of precipitation
occasionally had lower amounts of PET.
The following intervals exceeded the maximum allowable number of missing days for
monthly P summations: November 1904 through February 1905; October 1908;
December 1909; June 1910 through April 1912; March 1917 through April 1917; October
1925; April 1927; January 1931; January 1989 through August 1989; April 2005; and
December 2010.
The following intervals exceeded the maximum allowable number of missing days for
monthly PET summations: November 1904 through February 1905; April 1906;
December 1909; June 1910 through April 1912; April 1914 through May 1914; April 1915
through June 1915; November 1915; March 1917 through April 1917; November 1918
through December 1918; October 1922; October 1925; April 1927; May 1928; May 1929;
January 1931; March 1931 through April 1931; January 1989 through August 1989; April
2005; October 2010 through January 2011; and March 2011.
For the Blanding weather station, the following years had the highest amount of annual
precipitation (January through December):
• 1906 had 599 millimeters
• 1957 had 569 millimeters
• 1941 had 547 millimeters
• 1997 had 513 millimeters
• 1915 had 511 millimeters
For the Blanding weather station, the following years had the highest amount of winter
precipitation (November through February):
• 1909 had 333 millimeters
• 2005 had 321 millimeters
• 1979 had 307 millimeters
• 1993 had 296 millimeters
• 1907 had 288 millimeters
For the Blanding weather station, the following years had the lowest amount of annual
precipitation (January through December):
• 1956 had 128 millimeters
• 1950 had 138 millimeters
• 1976 had 160 millimeters
• 1934 had 182 millimeters
• 1964 had 201 millimeters
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 12 of 70
For the Blanding weather station, the following years had the lowest amount of winter
precipitation (November through February):
• 1964 had 23 millimeters
• 1981 had 29 millimeters
• 1970 had 33 millimeters
• 1918 had 33 millimeters
• 2006 had 35 millimeters
The deviations of annual and seasonal data from the long-term annual and seasonal
means are plotted in Figure 02/1/1-1.
The time-series evaluation illustrates the following observations:
• The following years were characterized by below average annual and below
average winter precipitation diagnostic of a “dry” period:
o 1934 through 1977 (excluding 1941-1942, 1949, 1957, and 1965 which
were wet).
o These observations are consistent with regional records that indicate that
a major drought period occurred during the early 1930s through 1940 and
from the early 1950s through the mid-1960s.
• The following years were characterized by average annual and average winter
precipitation:
o 1917 through 1932 (excluding 1931 which did not have a complete
dataset).
• The following years were characterized by variable annual and variable winter
precipitation diagnostic of a variable period punctuated by “wet” and “dry” years:
o 1977 through 2009/2010 (annual/winter).
o These observations are consistent with regional records that indicate the
last several decades were generally wetter than the preceding century.
• The following years were characterized by above average annual and above
average winter precipitation diagnostic of a “wet” period:
o 1906 through 1907 possibly extending through 1916 however data is
lacking to verify this at the Blanding weather station or the nearby
Monticello weather station.
o These observations are consistent with regional records that indicate that
the beginning of the 20th century was represented by wetter conditions
compared to past climates.
An analysis of the 106-year climate record for the Blanding weather station indicates that
the early 1900s were wet, which transitioned to a series of multi-decadal average and
below-average precipitation periods. These were followed by a series of multi-decadal
precipitation periods leading up to the present that were variable but overall wetter than
the long-term average. Overall, the latter part of the 20th century and beginning of the
21st century were wetter than conditions during most of the 20th century. While some
periods appear to be characteristic of “more wet” or “more dry” conditions all of the
periods are interspersed with variable amounts of precipitation.
The cover model presented in the Revised ICTM Report for the base case scenario used
the 57-year period between 1932 and 1988 to represent the base case climatic scenario.
This 57-year period was selected because it contained a nearly continuous time series
record, a mixture of the largest annual and daily precipitation events, and consecutive
wet years. Also, this timeframe included the period 1978 through 1987, which is a 10-
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 13 of 70
year timeframe characterized by above average amounts of annual (386 millimeters)
and winter precipitation (141 millimeters). The average annual precipitation between
1932 and 1988 is approximately 315 millimeters. The use of the timeframe from 1932
through 1988 for a base case scenario appears to be valid, and is consistent with
suggestions in the literature that suggest the use of a nearly-continuous time series
record.
Paleoclimate and Paleorecharge
A brief summary of the inferred paleoclimate and paleorecharge in the southwest United
States is presented below to facilitate comparison of measured contemporary data with
inferred paleoclimate and paleorecharge data for use in determining a wet precipitation
scenario in the cover model.
• At the Black Mesa Basin located in northeastern Arizona, Zhu (2000) used
radiocarbon age and hydraulic data to estimate recharge to groundwater during
the past 65,000 years. Chloride mass balance calculations give a range of
recharge rates between 5 to 20 millimeters per year (mm/yr) for most
groundwater samples whose age is younger than 6,000 years. Effective
calibration of the model was obtained when the climate, and recharge, was
assumed to be fairly stable for the last 6,000 years, with higher amounts of
recharge applied during the Pleistocene and lower amounts of recharge applied
during the transition from the Pleistocene to the Holocene (around 10,000 years
ago). These results suggest that climate at this location has not significantly
changed in the last 6,000 years. Current annual precipitation within this semiarid
basin is greater than 320 millimeters in areas higher than 1,800 meters elevation,
with most recharge occurring during the winter and spring months. In contrast,
during the late Pleistocene inferred temperatures were 5 to 6°C cooler than today
and recharge rates were two to three times higher than today, but during the mid-
Holocene inferred summer temperatures were 2 to 4°C warmer than today and
recharge rates were 50% lower than today. The highest recharge rates were
estimated to occur between 14,000 and 17,000 years ago.
• At the High Plains located in the Texas panhandle, Scanlon et al. (2003) used
chloride mass balance calculations and vadose zone data to estimate recharge
to the vadose zone and groundwater during the transition from wet/cool
conditions to dry/warm conditions that occurred 12,000 to 15,000 years ago.
Pluvial conditions during the Pleistocene were modeled to establish initial
conditions for long-term simulations during the Holocene coincident with the
transition to semiarid conditions and xeric vegetation. Water fluxes during the
Pleistocene were determined to equal 1.3 mm/yr based on chloride
concentrations. Simulations suggest that downward flow occurred during this
timeframe but that a drying front was initiated during the transition from the
Pleistocene to the Holocene (around 10,000 years ago). This drying front slowly
propagated downward through time, which eventually resulted in upward flow
conditions for at least the last 1,000 to 2,000 years. Recharge during the
Holocene were determined to be negligible (<0.1 mm/yr) in order to reproduce
the upward gradients in water potential and chloride profiles. Current annual
precipitation in this area is around 500 millimeters.
• In the Four Corners region of the southwest United States, Waugh and Peterson
(1995) assembled packrat midden, pollen, plant macrofossils, and
dendrochronology data to infer paleoclimate (temperature and precipitation)
during the Pleistocene and Holocene for conditions at Monticello, Utah. No
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 14 of 70
information regarding time-variant changes in recharge have been completed for
this location. Late Pleistocene conditions were estimated to have an inferred
mean annual temperature of 2°C and inferred precipitation around 800
millimeters. Between 13,000 to 10,500 years ago the inferred wet climate
continued but conditions were assumed to be slightly warmer. By the early
Holocene (around 10,000 years ago) the inferred climate was significantly
warmer and drier, although seasonal patterns of wet winters and dry summers
were thought to persist. The middle Holocene (around 8,000 to 4,000 years ago)
the inferred climate was a lot warmer and wetter, with inferred mean annual
temperature around 10°C and inferred annual precipitation around 600
millimeters. The vegetation record in packrat middens suggests that the winters
were cold and dry, and that precipitation occurred during the summer monsoon.
The continuance of inferred summer monsoon rains between years 700 to 1200
is inferred from population growth of the Anasazi. Summer rains were thought to
have abated during 1250 to 1850 based on forest fire records, but resumed to
present day conditions around 1840. The beginning of the 19th century was wet
based on climate records and dry land farming activities which began in 1905 but
were abandoned in 1932 because of drought. Dendrochronology within the
Upper Colorado River Basin suggests that the early 1900s was anomalously wet
relative to the preceding centuries, and there were numerous droughts more
sustained and/or severe than those of the 20th century, most notably during the
1500s. Overall, during the last 2,000 years the climate was variable and not well
constrained for the Monticello area.
Overall, the results of these paleoclimate/paleorecharge studies suggest that the climate
was warmer and drier during the transition from the Pleistocene to the Holocene (around
10,000 years ago) compared to current conditions, and that recharge was significantly
lower than present rates. During the last 1,000 to 6,000 years the climate is thought to
have been fairly comparable to current conditions with the exception of numerous
droughts more sustained and/or severe than those of the 20th century. Considering this
observation, climate was likely slightly warmer and drier during the last 1,000 to 6,000
years compared to current conditions, and recharge rates were likely less than current
rates. During the last 100 years the latter part of the 20th century and beginning of the
21st century were wetter than conditions during most of the 20th century. Moving
forward, there is considerable debate regarding climatic conditions during the next 200
to 1,000 years. However, the most consistent view regarding future climate in the
southwest United States is for warmer conditions and greater evaporative loss of water;
it also appears likely that winter precipitation may decrease and summer precipitation
may increase (Denison, 2012). Taken together, the next 200 to 1,000 years at White
Mesa likely will be comparable to conditions measured during the period of record (last
106 years) or slightly warmer and drier with less winter precipitation and greater
evaporative loss of water. And the assumption that a year with an abnormally wet winter
could occur for a decade or longer does not appear to be a valid. A more likely scenario
is that drought conditions will occur for a decade or longer rather than the extremely wet
conditions.
Cover Design Sensitivity Analyses
The purpose of Appendix E (cover design sensitivity analysis) in the Revised ICTM
Report was to compare the potential performance of different conceptual cover designs
based on model results assuming as-built material properties. MWH does not believe
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 15 of 70
that inclusion of additional sensitivity analyses is warranted for the four different
conceptual cover designs assuming weathered material properties.
ET Cover Sensitivity Analyses
For the proposed ET cover, the analysis presented above suggests that the soils will be
constructed in a manner that is close to the anticipated equilibrium state under long-term
conditions, and that correcting the material properties to account for pedogenic
processes is not warranted at this time. Rather, additional sensitivity analysis assuming
various scenarios for the soil material properties, and vegetative conditions, has been
completed for the proposed ET cover design. These updated model results are
discussed below.
Soil Properties:
The model-simulated water flux rate through the tailings cell cover during the
anticipated 57-year climate record (between 1932 and 1988) is shown on Figure
02/1/1-2 for the upper bound and lower bound hydraulic scenarios. These
climatic conditions are the same as those applied in the Revised ICTM Report.
The base case scenario is also plotted and compared to the departure from the
average amount of winter precipitation (November through February).
The average water flux rate is summarized below:
• The upper bound hydraulic scenario had an average water flux rate equal
to approximately 6.0 mm/yr or about 1.9% of the average annual amount
of precipitation.
• The base case scenario had an average water flux rate equal to
approximately 2.8 mm/yr or about 0.9% of the average annual amount of
precipitation.
• The lower bound hydraulic scenario had an average water flux rate equal
to approximately 2.4 mm/yr or about 0.8% of the average annual amount
of precipitation.
These rates are approximately 5 to 12 times higher than the value reported in the
Revised ICTM Report. The higher values are attributed to the laboratory Ks
results which were on the order of 80 cm/d (9x10-4 cm/s) while the value used in
the previous model was on the order of 8 cm/d (9x10-5 cm/s) for the water
storage layer. In actuality, the water flux rates could be slightly lower than
modeled depending on the Ks value assumed for the radon barrier layer. In the
model, the Ks value for the water storage and radon barrier layers were assumed
to be equal. Because the radon barrier layer will be compacted to 95% standard
Proctor compaction the material would be expected to have a lower permeability
than used in the model. Additional data may be collected to evaluate the Ks
values at 95% compaction. Overall, these simulated values are slightly higher
than measurements collected at the Monticello site for the last 12 years (average
percolation rate of 0.63 mm/yr with a minimum and maximum rate of 0 and 3.8
mm/yr).
Vegetation Properties:
The model-simulated water flux rate through the tailings cell cover during the
anticipated 57-year climate record (between 1932 and 1988) is shown on Figure
02/1/1-3 for the upper bound and lower bound vegetation scenarios assuming
30% and 40% cover (base case hydraulic scenario). The upper bound scenario
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 16 of 70
assumes a reduced root biomass distribution as presented in Table 01/1/3-1.
The lower bound vegetation scenario assumes a lower wilting point pressure
head equal to -30,000 cm, and a minimum surface pressure head equal to -
150,000 cm. The water stress response function for grass was selected from the
default database in HYDRUS. The database does not distinguish between
different species of grass, and transpiration is assumed to cease at soil water
pressures below the assumed wilting point of -8,000 cm. However, plants in
semiarid environments, many of which were selected for the ET cover, commonly
maintain transpiration at significantly lower (more negative) soil water pressures.
For example, crested wheatgrass can survive in soil water conditions where the
soil water pressure ranges between -20,000 and -40,000 cm (Chabot and
Mooney, 1985; Brown, 1995). Unless otherwise noted, all simulations assume
the default wilting point of -8,000 cm and a minimum surface pressure head of -
15,000 cm.
The average water flux rate is summarized below assuming 40% cover:
• The upper bound vegetation scenario had an average water flux rate
equal to approximately 4.9 mm/yr.
• The lower bound vegetation scenario had an average water flux rate
equal to approximately 0.7 mm/yr.

The average water flux rate is summarized below assuming 30% cover:
• The upper bound vegetation scenario had an average water flux rate
equal to approximately 5.1 mm/yr.
• The lower bound vegetation scenario had an average water flux rate
equal to approximately 0.9 mm/yr.
Climate:
A review of the literature indicates that there is no preferred methodology in
selecting a period to represent an increased precipitation scenario that may
occur in the future. The following quotations were taken from various
publications:
“The importance of site-specific climate and the multiyear effects
suggest that extended site-specific meteorological time series
should be used when designing [covers] to ensure that long-term
accumulation of water is not problematic and that the [cover] will
perform as intended. Methods currently employed in practice
include using the wettest year on record three or five years in a
row or the 10-year period with the highest average precipitation.”
Khire et al. (2000)
“The selected climatic periods or scenarios can have a significant
impact on [cover] predicted performance. Use of weather
generators is recommended for sites that have insufficient
historical data or where the goal is to evaluate cover performance
outside the bounds of the available historical data (e.g., to
evaluate the worst-case year in a 100-year period when only 40
years of historical data is available).” ITRC (2003)
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 17 of 70
“An adequate measurement of the climate at a site requires the
longest available record and should contain a minimum of 20
years of data. The importance of long records can be illustrated
by the annual precipitation from Coshocton, OH: while the 35-year
average annual precipitation is 37 inches, one 5-year period
averaged 88 percent of the overall average (32.6 inches) and
another averaged 115 percent (42.6 inches). Clearly, a short
record may not accurately describe the climate at a site and
should not be used for design.” Hauser and Gimon (2004)
The Monticello uranium tailings repository, located north of Blanding and at a
slightly higher elevation, used an end-member year to simulate percolation for a
high precipitation scenario (Abraham and Waugh, 1995). Abraham and Waugh
(1995) indicate that the 1993 data measured at the Monticello weather station
represented a record amount of winter precipitation. Monthly summations for
Monticello weather station were downloaded from the Western Regional Climate
Center on May 7, 2012. The cumulative precipitation measured during January
and February 1993 represents the third highest total behind 1980 and 2005. The
December and February 1993 measurements represent the sixth and second
highest measured values, respectively. The cover design modeling for the
Monticello repository assumed ten consecutive years of the 1993 precipitation
data to simulate a high precipitation scenario. The 1993 data is anticipated to be
similar to a Holocene wet climate scenario (up to about 13,000 years ago) based
on paleoclimatology studies (Waugh and Peterson, 1995). The winter
precipitation was selected as the most critical period at Monticello because
performance of the cover is most sensitive to precipitation that occurs while
evaporation rates are low and plants remain dormant. A similar assumption is
also incorporated into the analyses for White Mesa.
Based on the paleoclimate and paleorecharge studies summarized above, and
forecasted conditions that may occur in the future as a result of climate change,
the assumption of applying the maximum annual or winter precipitation value for
a ten year period does not appear to be justified. Rather, repetition at a lower
frequency for one of the wetter and drier winter seasons is a more practical
approach to determine if the ET cover will have sufficient storage capacity to
minimize percolation during a period of climatic stress:
• To simulate an increased precipitation scenario we used the Blanding
1993 winter precipitation (296 mm) and PET data repeated for a five year
period as part of the 57-year simulation incorporated in the Revised ICTM
Report. The January and February measurements recorded during the
1993 winter season correspond to the maximum and second highest
measured values recorded during the period of record, respectively. The
winter precipitation during 1993 corresponds to the 4th wettest year and
the 97th percentile, and is anticipated to be similar to a Holocene wet
climate scenario (up to about 13,000 years ago) based on information
presented by Waugh and Peterson (1995).
• To simulate a decreased precipitation scenario we propose using the
Blanding 1977 winter precipitation (39 mm) and PET data repeated for a
five year period as part of the 57-year simulation incorporated in the
Revised ICTM Report. The winter precipitation during 1977 corresponds
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 18 of 70
to the 7th driest year and the sixth percentile, and is anticipated to be
similar to a short-term drought.
This approach is functionally similar to the approach incorporated to design the
Monticello repository, and therefore is considered appropriate for this semiarid
site in the same region.
The model-simulated water flux rate through the tailings cell cover for the upper
bound and lower bound climate scenarios is shown on Figure 02/1/1-4 (base
case hydraulic scenario, 40% cover, anticipated root biomass, and default wilting
point). The average water flux rate is summarized below:
• The upper bound climate scenario had an average water flux rate equal
to approximately 6.4 mm/yr.
• The lower bound climate scenario had an average water flux rate equal to
approximately 2.6 mm/yr.
Summary
The cover model for the proposed ET cover was updated from the model
presented in the Revised ICTM Report. Updates to the model included
parameterization using the results of a site-specific physical/hydraulic
characterization testing program, and revised estimates for the range in
vegetation parameters and climatic conditions that may occur during the
performance period of the cover system within 200 years and up to 1,000 years
in the future. The range in hydraulic properties for the cover system, and the
anticipated frost penetration depth, suggest that soils used to construct the
erosion protection and water storage layers will be emplaced at conditions that
will minimize post-construction changes to the material layers.
The range in model-simulated water flux rate through the tailings cell cover
indicates that the hydraulic properties of the cover and the assumed wilting point
pressure for the vegetation exhibit the primary control in determining the amount
of water that may recharge the tailing cells. The minimum and maximum water
flux rates through the cover for sensitivity evaluated for this comment response
document were 0.7 mm/yr and 6.4 mm/yr, respectively. The modeling approach
incorporated into the Revised ICTM Report repeats the 57-year simulation
duration in a consecutive manner to assign initial conditions and evaluate long-
term infiltration rates such that dynamic equilibrium conditions will be achieved in
modeling the cover system performance.
The model-simulated water flux rate through the ET cover indicates that the
available storage capacity of the cover should be sufficient to significantly reduce
percolation, and the ET cover should function properly as designed. The
transport of water below the root zone, and radon barrier/grading layers, and into
the tailings material would occur when the storage capacity of the overlying soil
materials is exceeded; for example, during periods that receive above average
amounts of winter precipitation.

b. Provide additional sensitivity analyses projecting potential performance of the four different
conceptual cover designs where the cover materials are assumed to have experienced
degradation under postulated worst-case long-term conditions. Specifically, adjust parameters
(including at least, bulk density and porosity, in accordance with recommendations in NUREG-
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 19 of 70
1620, Section 5.1.3 [NRC 2003])) of soil and/or clayey materials within the maximum projected
frost-impacted zone for the 1,000-year recurrence interval (see also “INTERROGATORY
WHITEMESA RECPLAN 5.0; UAC R313-24-8; 10CFR40, APPENDIX A, CRITERION 6; INT
10/1: TECHNICAL ANALYSES - FROST PENETRATION ANALYSIS”). Consistent with
recommendations provided in Benson et al. 2011, adjust other cover soil properties (e.g.,
hydraulic conductivities and the α [or alpha] parameter in the mathematical expression for the
soil water characteristic curve [SWCC]) consistently for all alternative cover systems considered
(or justify why inconsistent parameter values are appropriate) in assessing long-term degraded
conditions.
Response 1b (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
See Response 1a.

c. Define and justify a range of possible future climate conditions that may reasonably be expected
to occur during the performance period of the closed tailings embankment system (up to 1,000
years), taking into account the projected variability of climate conditions over such time periods.
Provide infiltration modeling results that incorporate such peak/higher precipitation and/or
minimum evapotranspiration conditions. Alternatively, provide detailed justification why
consideration of such changed climatic conditions in the infiltration simulations is not justified or
would be otherwise inconsistent with relevant guidance and policy determinations and with
regulatory precedent established on other projects of a similar nature (Note: on similar projects,
formal future climate analysis techniques have been used to forecast possible future climate
states occurring during the next 1,000 years, and infiltration sensitivity analyses were performed
to assess long-term future cover system performance under these projected future climate
conditions). Incorporate worst-case meteorological conditions into the sensitivity analyses and
the “bathtub” analysis for the proposed evapotranspiration (ET) cover system.
Response 1c (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
See Response 1a and Response 2 to this interrogatory.

d. Extend the timeframe for calculations projecting the “bathtub effect” to a period of up to 1,000
years. Adjust soil properties in the proposed ET cover components to include initial and worst
case long-term degraded cover conditions as stated in Item 1 of this interrogatory. Incorporate
potential worst-case forecasted future climate conditions as stated in Item 2 of this interrogatory.
Response 1d (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
See Response 1a and Response 2 to this interrogatory.

e. Provide additional justification for selecting a three-consecutive-year period for the higher
precipitation regime in the infiltration sensitivity analysis provided in Appendix G. Discuss and
evaluate the appropriateness of results and/or recommendations from other published studies
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 20 of 70
(other than the Khire et al. 2000 study cited in Appendix G) for arid and semi-arid sites and
assumptions that were made for other similar projects (e.g., Monticello, Utah tailings repository
design, where a 10-consecutive-year wetter period was used in infiltration sensitivity analyses).
Demonstrate that the duration of the wetter period used in the sensitivity analyses ensures that
dynamic equilibrium conditions will be achieved in modeling the cover system performance.
Response 1e (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
See Response 1a.

References for Response 1 (September 10, 2012):
Abraham, J.D. and W.J. Waugh, 1995. Technical Task Cover Sheet, Re: Cover
Drainage and Leakage Rate Estimation Using the HELP 3.1 Code, Prepared
March 10, 1995.
Allen, R.G., L.S. Pereira, D. Raes, and M. Smith, 1998. FAO Irrigation and Drainage
Paper, No. 56, Crop Evapotranspiration (Guidelines for Computing Crop Water
Requirements), FAO - Food and Agriculture Organization of the United Nations,
Rome, Italy, pp. 300.
Benson, C.H., S.H. Lee, X. Wang, W.H. Albright, and W.J. Waugh, 2008. Hydraulic
Properties and Geomorphology of the Earthen Component of the Final Cover at
the Monticello Uranium Mill Tailings Repository, Geo Engineering Report No. 08-
04.
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 and Implications for Long-Term
Performance Assessment, Volume 1 and 2, NUREG/CR-7028, Report Prepared
for the U.S. Nuclear Regulatory Commission, December.
Brown, R.W. 1995. The Water Relations of Range Plants: Adaptations to Water Deficits,
In: Bedunah, D.J. and R.E. Sosebee (eds.), Wildland Plants: Physiological
Ecology and Developmental Morphology, Society for Range Management,
Denver, Colorado, pp. 291-413.
Chabot, B.F. and H.A Mooney 1985. Physiological Ecology of North American Plant
Communities. Chapman and Hall. New York, NY.
Denison Mines (USA) Corp. 2012. Responses to Interrogatories – Round 1 for
Reclamation Plan, Revision 5.0, March 12. August 15, 2012.
Hauser, V.L. and D.M. Gimon, 2004. Evaluating Evapotranspiration (ET) Landfill Cover
Performance Using Hydrologic Models, Report Prepared for the U.S. Air Force
Center for Environmental Excellence, January 2004.
Interstate Technology & Regulatory Council (ITRC), 2003. Technical and regulatory
guidance for design, installation, and monitoring of alternative final landfill covers,
Report Prepared by Prepared by the Interstate Technology & Regulatory Council
Alternative Landfill Technologies Team, December 2003.
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 21 of 70
Khire, M.V., C.H. Benson, and P.J. Bosscher, 2000. Capillary barriers: design variables
and water balance, Journal of Geotechnical and Geoenvironmental Engineering,
126(8):965-708.
Scanlon, B.R., K. Keese, R.C. Reedy, J. Simunek, and B.J. Andraski, 2003. Variations in
flow and transport in thick desert vadose zones in response to paleoclimatic
forcing (0–90 kyr): Field measurements, modeling, and uncertainties, Water
Resources Research, 39(7), 1179, doi:10.1029/2002WR001604.
Tinjum, J., C. Benson, and L. Blotz, 1997. Soil-Water Characteristic Curves for
Compacted Clays, Journal of Geotechnical and Geoenvironmental Engineering,
123(11):1060-1070.
Waugh, W.J. and K.L. Peterson, 1995. Paleoclimatic Data Application: Long-Term
Performance of Uranium Mill Tailings Repositories, Paper as Part of a
Conference Proceedings, Climate Change in the Four Corners and Adjacent
Regions: Implications for Environmental Restoration and Land-Use Planning,
September 12-14, 1994, Paper published September 1994.
Zhu, C., 2000. Estimate of recharge from radiocarbon dating of groundwater and
numerical flow and transport modeling, Water Resources Research, 36(9):2607–
2620.

September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 22 of 70
Tables and Figures for Response 1 (September 10, 2012):
Table 02/1/1-1. Identified hydraulic classification groups and scenarios for the soils used to parameterize the
cover model.

Scenario ID Soil Type Storage
(cm)
Ks(cm/d)
Gravel
(%)
Upper
Bound W2-B1/2 B 11.2 130
[1.5x10-3 cm/s] 41
Base Case
(Average) - B & U 18.1 62
[7.2x10-4 cm/s] 15
Lower
Bound W9-B1/2 U 21.7 35
[4.1x10-4 cm/s] 0
Note: Storage or available water content was computed at the difference between the volumetric water content
at field capacity and wilting point tensions multiplied by the thickness for the erosion protection and water
storage layers. Storage accounts for reduced capacity from the amount of gravel calculated using the
approach suggested by Bouwer and Rice (1984). The amount of gravel for the upper bound scenario was
taken as measured. The amount of gravel for the base case scenario was based on a weighted average for
the material volumes and percentage of gravel for the B, U, and F soil types. The amount of gravel for the lower bound scenario was assumed to equal zero based on previous geotechnical results.

Table 02/1/1-2. Parameter values used to parameterize the cover model for the three hydraulic scenarios
modeled using the van Genuchten-Mualem functions.

Cover
Layer Purpose Thickness
(cm)
θr(-)
θs(-)
α
(1/cm)
n
(-)
Ks
(cm/d)
l
(-)
ρb(g/cm3)
Upper Bound Soils
1 Erosion
Control 15 0.02 0.32 0.0080 1.35 11 0.5 1.70
2 Water
Storage 107 0 0.23 0.022 1.32 130 0.5 1.85
3 Radon
Barrier 110 0 0.16 0.022 1.32 130 0.5 2.07
4 Grading 76 0 0.26 0.022 1.32 130 0.5 1.74
Base Case Soils (Average)
1 Erosion
Control 15 0.02 0.32 0.0080 1.35 11 0.5 1.70
2 Water
Storage 107 0 0.34 0.011 1.30 62 0.5 1.67
3 Radon
Barrier 110 0 0.27 0.011 1.30 62 0.5 1.87
4 Grading 76 0 0.37 0.011 1.30 62 0.5 1.58
Lower Bound Soils
1 Erosion
Control 15 0.02 0.32 0.0080 1.35 11 0.5 1.70
2 Water
Storage 107 0 0.40 0.0073 1.26 35 0.5 1.56
3 Radon
Barrier 110 0 0.33 0.0073 1.26 35 0.5 1.75
4 Grading 76 0 0.43 0.0073 1.26 35 0.5 1.47
Note: The saturated and residual volumetric water contents for the erosion protection and water storage layers
were corrected for the amount of gravel calculated using the approach suggested by Bouwer and Rice (1984). The base case scenario was obtained by averaging the B and U soil samples: the saturated/residual
volumetric water contents, n, and ρb were arithmetically averaged while α and Ks were geometrically averaged.

September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 23 of 70

Table 02/1/1-3. Summary statistics of climatic data for the Blanding weather station period of record
November 1904 through March 2011
Statistic
Annual
Precipitation
(mm)
Winter
Precipitation
(mm)
Annual
PET
(mm)
Winter
PET
(mm)
Maximum 599 333 1,315 165
90th Percentile 467 211 1,287 145
Average 335 128 1,219 130
Median 320 119 1,227 130
10th Percentile 228 44 1,151 113
Minimum 128 23 1,025 103
Standard Deviation 94 69 54 13
Number of Years 91 100 86 98
Notes: Annual data represent the calendar year while the winter data represent November
through February. Data are rounded to the nearest whole number. PET = potential
evapotranspiration; mm = millimeters.

September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 24 of 70
Figure 02/1/1-1. Departure from the average long-term precipitation. The grey dashed lines indicate one
standard deviation from the annual average.

Figure 02/1/1-2. Model-simulated water flux rate exiting the bottom of the ET cover during the anticipated 57-
year climate record between 1932 and 1988 for the lower and upper bound hydraulic
scenarios (left) and for the base case scenario (right). The base case scenario is compared
to the departure from the average amount of winter precipitation.

Figure 02/1/1-3. Model-simulated water flux rate exiting the bottom of the ET cover during the anticipated 57-
year climate record between 1932 and 1988 for the lower and upper bound vegetation
scenarios assuming 40% cover (left) and 30% cover (right).

‐300
‐200
‐100
0
100
200
300
1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005
Pre
c
i
p
i
t
a
t
i
o
n
 (m
m
)
Time (Year)
Departure from Annual Mean Departure from Winter Mean
0.01
0.1
1
10
100
1000
0 102030405060
Wa
t
e
r
 Fl
u
x
 Ra
t
e
 (m
m
/
y
r
)
Time (years)
Lower Bound Soils Upper Bound Soils
‐200
0
200
0.01
0.1
1
10
100
1000
0 102030405060
De
p
a
r
t
u
r
e
 fr
o
m
 Wi
n
t
e
r
 Me
a
n
 (m
m
)
Wa
t
e
r
 Fl
u
x
 Ra
t
e
 (m
m
/
y
r
)
Time (years)
Base Case Soils Deviation from Winter Mean
0.01
0.1
1
10
100
1000
0 102030405060
Wa
t
e
r
 Flu
x
 Ra
t
e
 (m
m
/
y
r
)
Time (years)
Reduced Root Biomass Decrease in Wilting Point
0.01
0.1
1
10
100
1000
0 102030405060
Wa
t
e
r
 Fl
u
x
 Ra
t
e
 (m
m
/
y
r
)
Time (years)
30% Cover & Reduced Biomass Decrease in Wilting Point
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 25 of 70
Figure 02/1/1-4. Model-simulated water flux rate exiting the bottom of the ET cover during the lower and
upper bound climate scenarios (left) and cumulative drainage (right).

2. Refer to Revised ICTM Report, p. ES-6, Sections 4.1.2 and 5.1.2, and Appendix G: Please
justify assuming a tailings porosity of 57% in evaluating infiltration/potential for “bathtubbing”
of leachate on the liner systems. Perform and report results of sensitivity analyses that assess the
dependence of result on variations in the values of tailings porosity used in analyses.

Response 2 (September 10, 2012):
This response supersedes the response provided in the response document submitted
May 31, 2012.
The next iteration of the ICTM Report will evaluate any potential for a “bathtub effect”
using a tailings porosity of 47% to be consistent with the value used in the radon
emanation analyses presented in Denison (2012) for the revised cover design. A
sensitivity analysis regarding an upper and lower bound estimate for the tailings porosity
(44% and 51%), as based on the range in tailings compaction, for 200 years and
extending through 1,000 years, is discussed below.
The “bathtub effect” calculations are based on the shortest transport distance between
the upper surface of the tailings and the water level in the tailings after dewatering
ceases for Cells 2 & 3 and Cells 4A and 4B:
• For Cells 2 & 3, the average tailings thickness is approximately 6.4 meters and
the water level in the tailings after dewatering ceases is approximately 1.1
meters, resulting in a vertical distance of approximately 5.3 meters.
• For Cells 4A & 4B, the average tailings thickness is approximately 7.5 meters
and the water level in the tailings after dewatering ceases is approximately 0.3
meters, resulting in a vertical distance of approximately 7.2 meters.
Using the average water flux rate through the cover for the base case hydraulic scenario
(2.8 mm/yr), and assuming the materials are at 50% saturation (volumetric water content
of 22% to 26%), it would take on the order of 410 to 480 years for water from the cover
to reach the water level in the tailings for Cells 2 & 3 and on the order of 570 to 660
years for Cells 4A and 4B. As a result, there is no potential for a bathtub effect during
the 200 years after closure.
0.01
0.1
1
10
100
1000
0 102030405060
Wa
t
e
r
 Fl
u
x
 Ra
t
e
 (m
m
/
y
r
)
Time (years)
Dry Climate Wet Climate
0
0.1
0.2
0.3
0.4
0.5
0 102030405060
Cu
m
u
l
a
t
i
v
e
 Dr
a
i
n
a
g
e
 (m
)
Time (years)
Dry Climate Wet Climate
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 26 of 70
Commencing approximately 410 to 480 years after closure, head would start to build up
on the liner for Cells 2 & 3. However, using the same literature-based assumptions for
frequency of potential defects in the liners beneath the cells as was used for the Revised
ICTM Report, water levels in the cells will equilibrate when the rate of inflow is equal to
the assumed rate of outflow. The corresponding long-term water level in the cells would
occur when the modeled long-term flux of water through the cover is equal to the
calculated potential water flux rate through the liner. The water level in the tailings that
corresponds to these equilibrated conditions is approximately 2.9 meters for Cells 2 & 3,
which means that, during long-term conditions, water from the cover may increase water
levels in the cells by up to 1.8 meters, from 1.1 meters to 2.9 meters, at which time
equilibrium will be reached and further increases in water levels will not occur.
Therefore, since the equilibrium water level of 2.9 meters is lower than the average
tailings thickness of 6.4 meters, the calculations indicate that the accumulation of
leachate head within the tailings would not rise above or over-top the maximum liner
elevation internal to Cells 2 & 3 (i.e., create a “bathtub” effect) during the 1,000 years
after closure or at any time thereafter.
Commencing approximately 570 to 660 years after closure, head would start to build up
on the secondary liner for Cells 4A & 4B. However, using the same literature-based
assumptions for frequency of potential defects in the liners beneath the cells as was
used for the Revised ICTM Report, water levels in the cells will equilibrate when the rate
of inflow is equal to the assumed rate of outflow. The corresponding long-term water
level in the cells would occur when the modeled long-term flux of water through the
cover is equal to the calculated potential water flux rate through the liner. The water
level in the tailings that corresponds to these equilibrated conditions is approximately 4.5
meters for Cells 4A & 4B, which means that, during long-term conditions, water from the
cover may increase water levels in the cells by up to 4.2 meters, from 0.3 meters to 4.5
meters, at which time equilibrium will be reached and further increases in water levels
will not occur. Therefore, since the equilibrium water level of 4.5 meters is lower than
the average tailings thickness of 7.5 meters, the calculations indicate that the
accumulation of leachate head within the tailings would not rise above or over-top the
maximum liner elevation internal to Cells 4A & 4B (i.e., create a “bathtub” effect) during
the 1,000 years after closure or at any time thereafter.
Reference for Response 2 (September 10, 2012):
Denison Mines (USA) Corp. 2012. Responses to Interrogatories – Round 1 for
Reclamation Plan, Revision 5.0, March 12. August 15, 2012.
3. Refer to Appendix E, p. E-5, Paragraph 2 of the ICTM Report: Please clarify/provide the
information referenced as being included in Attachment E-1 (not apparently provided in the
report).
Response 3 (May 31, 2012 and September 10, 2012):
The information presented in Attachment E-1 was mistakenly left out of the report. The
information in Attachment E-1 was included as Attachment A in the May 31, 2012
response document (Denison, 2012).
Reference for Response 3 (May 31, 2012 and September 10, 2012):
Denison Mines (USA) Corp. 2012. Responses to Interrogatories – Round 1 for the
Revised Infiltration and Contaminant Transport Modeling Report, March 2010.
May 31, 2012.
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 27 of 70

BASIS FOR INTERROGATORY:
Various sets of assumptions were made when estimating parameter input values for various cover
materials for use in the infiltration model simulations and in the infiltration comparisons evaluating the
hydraulic performance of the four different cover designs. However, several simplified assumptions were
included, and additional justification/ rationale needs to be provided to support the representativeness
and appropriateness of these input values. Site-specific testing data should be better developed and
utilized and real correlations developed between field parameters and laboratory results, and between
soil properties and soil compaction levels for each of the different proposed ET soil cover layers.
Properties assumed for the various soil layers in the proposed ET cover system need to be fully justified.
For example, the porosity value of 0.25 listed in Table H-3 for the Erosion Protection Layer has not
adequately been justified and appears to be low. The value should be determined through calculation
(e.g., using the U.S. Bureau of Reclamation Earth Manual estimation formula for total density of a
soil/gravel admixture), information in Earth Manual or elsewhere on predicted percentages of Proctor
maximum dry densities obtainable using standard compactive effort in relation to percent of gravel
present, and correcting for the percentage of maximum density corresponding to the specified compaction
level), followed by calculations of the void ratio and porosity.
The meteorological and soil parameter values used in the sensitivity analysis should better reflect the
range of possible future meteorological and hydrological conditions that may occur at the site during the
long-term performance period of the closed tailings embankment cover system. Adjusted bulk density and
porosity values for the portion of the cover potentially affected by the maximum frost penetration depth
over a 1,000-year recurrence period should be employed in the radon emanation model as per NUREG-
1620 recommendations. Equivalent or consistent adjusted soil properties should be used in cover
infiltration simulations or adequate justification provided for assuming different material properties. The
estimates of the material parameters used in the infiltration sensitivity analyses performed to assess long-
term cover performance need to be reasonably conservative, considering the uncertainty associated with
these values.
Determination of soil properties should be based on testing of soils from the site and more precise
correlations of key soil properties (e.g., soil layer hydraulic conductivity vs. relative soil compaction
level) should be developed with supporting information describing the test method and its precision,
accuracy, and applicability provided. It needs to be demonstrated that the parameter values selected and
used in the performance analyses are conservative. For example, the code (HYDRUS) default-defined
hydraulic conductivity values (based on particle size gradation information – Table E-1 in Appendix E to
the Updated Tailings Design Report) may not always be conservative. The infiltration model should
result in a representative and a reasonably conservative (given the uncertainty in some values) long-term
infiltration estimate. Determination of variations in hydraulic conductivity with actual relative
compaction levels for on-site soil samples, and associated permeameter tests used to determine saturated
hydraulic conductivities of son-site soils with testing of on-site soils to determine the soil water retention
curves could likely result in considerably less uncertainty in soil parameter input values used in modeling
(e.g., see McCartney and Zornberg 2006).
An adequate range of climate data providing a conservative representation of recorded historical climate
conditions in the site area (e.g., Blanding, Utah climate data for the period 1904 through the most recent
year available), and a conservative estimate of the range of future climate conditions that might
reasonably be expected to occur during the performance period of the closed tailings embankment system
are required for evaluating the long-term performance of the embankment’s cover system. The
evaluation should consider projections of long-term extreme events and potential shifts in climate states
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 28 of 70
that could reasonably expected to occur over 100’s of years to up to 1,000 years, as well as annual and
decadal variability in meteorological parameters. To better capture and assess uncertainties in long-term
performance of the tailings embankment cover system resulting from possible future changes in climate
conditions, a projection (e.g., first approximation) of possible future climate states at the White Mesa site
should be developed using a future climate forecasting approach similar to or equivalent in approach
to the future climate analysis approach used in other recent studies completed for similar facilities in
Utah, such as the Monticello tailings repository (e.g., see Waugh et al. 1995; Sharpe 2004).
Identification of the potential climate conditions should be based on analysis of several facts and
considerations, including, but not limited to: (1) Annual total precipitation amounts that have occurred at
the Blanding Meteorological Station (e.g., 23.50 inches, and 24.42 inches, in 1906 and 1908,
respectively) that are higher than the range of annual precipitation values considered in the current
Infiltration and Contaminant Transport Model (ICTM) Report, which only considered Blanding climate
data acquired between 1932 and 1988;
(2) Subtotals of precipitation amounts that have occurred during any two, or any three consecutive
months at Blanding (e.g., 9.04 inches combined total precipitation for January and February 1993 and
11.33 inches combined total precipitation for December 1992 through February 1993; 7.98 inches
combined total precipitation for January and February 2005 and 10.46 inches combined total
precipitation for December 2004 through February 2005; 11.95 inches combined total precipitation for
December 1908 through February 1909; 5.75 inches total precipitation for April and May 2011
combined ; etc…) which are higher subtotal amounts than for any of the same consecutive sets of months
that were included in the 1932-1988 data set considered in the current ICTM Report and three-month
sub-total precipitation amounts recorded at Blanding that were higher than during the same three months
as the Summer 1987 summer monsoon period selected for use in the sensitivity analysis presented in the
current ICTM Report. Also, in 1908 and 1909, the months of December alone were the second highest,
and the highest of record, for any winter season months with 6.20 and 6.84 inches, respectively. This
further suggests that winter-season precipitation conditions may be expected to be the most critical (most
conservative) as a basis for extrapolating potential abnormal future wetter weather conditions for use in
assessing the effects (sensitivity)of such possible future conditions on modeled infiltration performance
(see also items (4) and (5) below);
(3) Site-specific monitoring data, if any, from measurements made within a cover test cell considered
representative of the proposed ET cover system, that might indicate one or more sets of consecutive
months of the year when infiltration rates in the cover would likely be the highest;
(4) Identification and justification for selecting a specific climatological data set such as choosing
precipitation data for the wettest consecutive months or sets of consecutive months recorded at Blanding
that may correspond to those months when the highest on-site infiltration rates would be expected to
occur through the ET cover system, for use in extrapolating (forecasting) potential long-term climate
conditions at the White Mesa site. In this regard, additional information should be provided to justify not
selecting the wettest consecutive winter months observed for the precipitation period of record for
Blanding, e.g., rather than selecting the 92-day-long 1987 summer monsoon season as was done in the
sensitivity analysis in Appendix F in the current ICTM Report, as the basis for extrapolating potentially
wetter future climate conditions, since doing the former could likely result in more moisture breakthrough
than that predicted by the current modeling;
(5) A description of the specific historical climate data set (e.g., wettest three consecutive winter
months, if selected), or other sub-annual or annual data set(s) selected, and a description of the
procedure used for extrapolating this data set or these data sets to simulate inferred future climate
conditions at the White Mesa site should be provided;
(6) A projection (e.g., first approximation) of possible future climate states at the White Mesa site
should be developed based on paleoecological evidence and/or a global/regional climate change model
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 29 of 70
using a future climate forecast approach, e.g., involving the use of analogue present-day climate sites,
similar in rigor to the future climate analysis approach used in other recent studies completed for other
similar facilities in Utah (e.g., see Waugh et al. 1995; Sharpe 2004); and
(7) A description of the correlation of the extrapolated climate conditions derived based on the
considerations listed in items (1) through (5) above to future climate conditions (climate states)
forecasted using the future climate analysis approach, as described in item (6) above, should also be
provided.
NUREG/CR-7028, a peer-reviewed report published for the NRC in December 2011, reports the findings
from investigations of several earthen and soil/geosynthetic cover systems to assess changes in properties
of cover materials in those cover systems 5 to 10 years following their construction. A key conclusion of
the report is that findings from these investigations demonstrate that changes in the engineering
properties of cover soils generally occur while in service and that long-term engineering properties
should be used as input to models employed for long-term performance assessments. The report indicates
that changes in hydraulic properties occurred in all cover soils evaluated due to the formation of soil
structure, regardless of climate, cover design, or service life. The report includes recommendations for
appropriate input based on the data that were collected. This document therefore contains information
important to the design of the final cover system for the White Mesa uranium tailings management cells
area. Additional sensitivity analyses should be performed that allow for and incorporate effects of
potential long-term degradation of the cover materials in a manner consistent with conclusions and
recommendations given in NUREG/CR-7028, i.e., that “engineering properties of cover soils change
while in service and…that long-term engineering properties for soils cover materials should be used as
input for performance assessments”.
Based on available information and data for other uranium mill tailings, a porosity value of 57% may be
considered more representative of the finer particle fraction of the tailings (slimes) than the tailings
materials on average (mixture of sands and clays/silt materials) in the saturated and unsaturated portions
of the tailings masses in the cells. Although a porosity of 57% may be considered conservative for
estimating radon flux through the cover (Appendix H of the ICTM Report), such an assumption may not
be appropriate for the infiltration and bathtub analyses, for which a lower average porosity value
appears to be warranted (e.g., approximately 39% to 40%, based on data for the Moab uranium tailings).
Additional justification should be provided supporting the use of a lower porosity value in the
infiltration/bathtubbing analyses and revised analyses and conclusions should be provided that
incorporate the lower porosity value.
Material referenced as being included in Attachment E-1 of Appendix E was not provided.
REFERENCES:
Benson, C.H. W.H. Albright, W.H., Fratta, D.O.,Tinjum, J.M., Kucukkirca, E., Lee, S.H., J. Scalia, J.,
Schlicht, P.D., and Wang, X. 2011. Engineered Covers for Waste Containment: Changes in Engineering
Properties and Implications for Long-Term Performance Assessment(in 4 volumes). NUREG/CR-7028,
Prepared for the U.S. Nuclear Regulatory Commission, Washington, D.C., December 2011.
Denison Mines (USA) Corp. 2010. Revised Infiltration and Contaminant Transport Modeling Report,
White Mesa Mill Site, Blanding, Utah (Revision 2), March 2010.
Denison Mines (USA) Corp., 2011. Reclamation Plan, White Mesa Mill, Blanding, Utah, Radioactive
Materials License No. UT1900479, Revision 5.0, September 2011.
September 10, 2012
Interrogatory 02/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Comparison of Cover Designs, Sensitivity Analyses, Bathtub Analysis, and
Radon Emanation Modeling Page 30 of 70
Khire, M.V., Benson, C.H., and Bosscher, P.J. 2000. Capillary Barriers: Design Variables and Water
Balance. Journal of Geotechnical and Geoenvironmental Engineering. August 2000.
McCartney, J.S, and Zornberg, J.G. 2006. Decision Analysis for Design of Evapotranspirative Landfill
Covers”, Proceedings UNSAT ’06, April 2-6, Carefree, AZ, ASCE, pp. 694-705.
NRC 2003. NUREG-1620: Standard Review Plan for the Review of a Reclamation Plan for Mill Tailings
Sites under Title II of the Uranium Mill Tailings Radiation Control Act of 1978. Washington DC, June
2003.
Sharpe, S. 2004. Future Climate States at Monticello, Utah. Desert Research Institute, February 25,
2004.
Waugh, W.J. and Petersen, K.L. 1995. “Paleoclimatic Data Application: Long-Term Performance of
Uranium Mill Tailings Repositories,” in: W.J. Waugh (ed.), Climate Change in the Four Corners and
Adjacent Regions: Implications for Environmental Restoration and Land-Use Planning, CONF9409325,
U.S. Department of Energy, Grand Junction, Colorado, USA, pp. 163185 (1995). Available at:
http://www.osti.gov/energycitations/product.biblio.jsp?osti_id=167170
September 10, 2012
Interrogatory 03/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Moisture Storage Capacity of Cover Page 31 of 70
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10 CFR40 APPENDIX A,
CRITERION 6(1); INT 03/1: MOISTURE STORAGE CAPACITY OF COVER
REGULATORY BASIS:
UAC R313-24-4 invokes the following requirement from 10CFR40, Appendix A, Criterion 6(1): In disposing
of waste byproduct material, licensees shall place an earthen cover (or approved alternative) over
tailings or wastes at the end of milling operations and shall close the waste disposal area in accordance
with a design which provides reasonable assurance of control of radiological hazards to (i) be effective
for 1,000 years, to the extent reasonably achievable, and, in any case, for at least 200 years, and (ii) limit
releases of radon-222 from uranium byproduct materials, and radon-220 from thorium byproduct
materials, to the atmosphere so as not to exceed an average release rate of 20 picocuries per square
meter per second (pCi/m2s) to the extent practicable throughout the effective design life determined
pursuant to (1)(i) of this Criterion. In computing required tailings cover thicknesses, moisture in soils in
excess of amounts found normally in similar soils in similar circumstances may not be considered. Direct
gamma exposure from the tailings or wastes should be reduced to background levels. The effects of any
thin synthetic layer may not be taken into account in determining the calculated radon exhalation level. If
non-soil materials are proposed as cover materials, it must be demonstrated that these materials will not
crack or degrade by differential settlement, weathering, or other mechanism, over long-term intervals.
INTERROGATORY STATEMENT:
Refer to Appendix F of the ICTM Report: Please provide the following:

1. Redefine and further justify the critical meteorological design event (or sequence of contiguous
events). State and justify the basis for the critical event conditions addressing the location of the
meteorological weather station for determining the wettest year on record; duration of the
critical event (i.e., single-day storm or multiple-day storm; number of consecutive days of rainfall
followed by a large, single-day rainfall event). Justify excluding recorded historical
monthly/daily precipitation data for Blanding, Utah from consideration in all infiltration
analyses conducted in the ICTM Report that indicate larger two-month-long and three-month-
long precipitation amounts than the 92-day-long 1987 summer monsoon season used in the
sensitivity analysis in Appendix F (see also INTERROGATORY WHITE MESA REV’D ICTM;
R313-24-4; 10CFR40 APPENDIX A, CRITERION 6(1); INT 02/1: COMPARISON OF COVER
DESIGNS, SENSITIVITY ANALYSES, ‘BATHTUB’ ANALYSIS , AND RADON EMANATION
MODELING above). Identify the month(s) of the year that would be expected to comprise the
most critical percolation period. Justify why consideration of summer monsoon conditions (when
plant cover would be more developed and ET rates more enhanced) has been considered to be
more conservative than assuming the most critical meteorological period as occurring during the
winter months.
Response 1 (May 31, 2012):
Appendix F was incorporated in the 2010 Revised ICTM Report to specifically address
comments previously received by the Division (DRC, 2009). The Division requested that
justification should be provided to confirm whether distributing the daily precipitation data
at a uniform rate throughout the day is representative of true field conditions and storm
intensity effects. A series of model simulations that used daily and hourly input data as
implemented in Appendix F concluded that the model simplification of using daily input
rather than hourly input does not affect the predictive results. Appendix F was not
intended to form the basis of a sensitivity analysis regarding an increased precipitation
scenario. The increased precipitation scenario was addressed in Appendix G. We
September 10, 2012
Interrogatory 03/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Moisture Storage Capacity of Cover Page 32 of 70
propose that Appendix F should be removed from the next iteration of the ICTM Report
because it may be a source of confusion.
References for Response 1 (May 31, 2012):
Utah Division of Radiation Control (DRC), 2009. White Mesa Uranium Mill, Ground
Water Discharge Permit No. UGW370004, Infiltration and Contaminant Transport
Modeling Report: DRC Review Comments, Request for Additional Information,
Letter from Thomas Rushing of DRC to David C. Frydenlund of Denison Mines
dated February 2.

2. Provide additional details regarding the assumed gradient at the soil cover/atmosphere interface
and include, as needed, an increase to an otherwise assumed gradient of unity to address the
potential for higher infiltration rates into the cover due to matric suction gradients greater than
unity (corresponding to low suction at the soil surface and a higher suction corresponding to the
initial moisture content) - see, e.g., McCartney and Zornberg 2006. Discuss how localized
surface ponding, if it were to occur, would or would not affect the assumptions about the gradient
at the soil cover interface;
Response 2 (May 31, 2012):
A gradient of unity was not assumed for the soil cover/atmosphere interface, and the
spreadsheet model implemented by McCartney and Zornberg (2006) is not equivalent to
the numerical model implemented in the 2010 Revised ICTM Report. In the 2010
Revised ICTM Report, the upper surface boundary condition representing the air-soil
interface follows a system-dependent boundary condition. The soil surface boundary
condition within the model may change from prescribed flux to prescribed head type
conditions and vice-versa. A switch to a prescribed head type boundary condition would
occur, for example, if the precipitation rate exceeds the infiltration capacity of the soil,
resulting in either surface runoff or accumulation of excess water on top of the soil
surface (ponding). The infiltration rate in this case is not controlled any more by the
precipitation rate, but instead by the infiltration capacity of the soil to transmit water after
removal through runoff or from ponding at the soil surface. In the model presented in the
2010 Revised ICTM Report, a maximum surface ponding depth of five centimeters was
assigned. Localized surface ponding, if it were to occur, would act to increase hydraulic
gradients along the air-soil interface resulting in greater amounts of water that could
infiltrate into the cover until the surface pond reservoir was depleted.
References for Response 2 (May 31, 2012):
McCartney, J.S. and J.G. Zornberg, 2006. Decision Analysis for Design of
Evapotranspirative Landfill Covers, Paper presented at the UNSAT ’06
Conference: 694-705.

3. Revise the water balance analysis to demonstrate that the cover system will provide sufficient
moisture storage capacity to retain precipitation resulting from a redefined, largest and most
critical meteorological event/set of conditions (most stressful hydraulic condition(s)) that the
cover might be exposed to during its required performance life (1,000 years, to the extent
practicable and technically and economically feasible, and in no case less than 200 years).

September 10, 2012
Interrogatory 03/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Moisture Storage Capacity of Cover Page 33 of 70
Response 3 (May 31, 2012):
As discussed previously in this interrogatory response document, the infiltration
modeling was updated to account for an increased precipitation scenario; and this
scenario was selected to correspond to the most critical time period for which percolation
through the cover could occur during its required performance life. However, such an
evaluation would be presented in another appendix (e.g., Appendix G) because it is not
applicable to the original intent of Appendix F.

4. Discuss, justify, and apply a recommended safety factor to the design of the cover to provide
additional assurance that the thickness of the cover system will be adequate to accommodate the
most stressful hydraulic conditions determined in Items 1 and 2 above , as required, and to also
address uncertainties relating to the following (e.g., Khire et al. 2000; Hauser et al. 2001;
Hauser and Gimon 2004):
a. The size of the soil water reservoir in the cover soil must be adequate to contain the
predicted extreme event/conditions (critical event or events) and potentially uncertain,
intense future storm events;
b. The potential variability of climate conditions over the required performance evaluation
period;
c. The time required to empty the soil-water reservoir; and
d. Other factors, such as the potential long-term degradation of the cover materials due to
desiccation cracking, water erosion, freeze-thaw damage, and other environmental
processes (see, e.g., Benson et al. 2011).
Response 4 (May 31, 2012):
The inclusion of a factor of safety (FOS) to the design of the cover is not appropriate
considering the conservative nature of the assumptions used to evaluate the cover
design and performance.

BASIS FOR INTERROGATORY:
Estimates of deep percolation through the cover are of particular concern for ET cover design and
evaluation. The performance of ET covers should be estimated for large and critical climatic events
expected to occur during the service life of the cover. Therefore, a major concern for ET cover
performance is the determination of the greatest storage capacity required for the ET cover during a
defined, most-critical meteorological event or set of consecutive (contiguous) meteorological events.
Critical events causing maximum soil-water storage may result from a single-day storm, a multiple-day
storm, or other events.
As a further check for ensuring that the proposed surface cover layer thickness is adequate, an evaluation
should be completed that uses suitable long-term simulations performed with the most stressful conditions
that the cover is likely to endure (Khire et al. 2000). The assessment should include any potentially
wetter future climate conditions that may reasonably be expected to occur during the performance period
of the embankment cover system spanning up to on the order of 1,000 years following the end of the
institutional control period, as described in INTERROGATORY WHITE MESA REV’D ICTM; R313-24-
4; 10CFR40 APPENDIX A, CRITERION 6(1); INT 02/1: COMPARISON OF COVER DESIGNS,
SENSITIVITY ANALYSES, ‘BATHTUB’ ANALYSIS, AND RADON EMANATION MODELING above.

September 10, 2012
Interrogatory 03/1: R313-24-4; 10CFR40; Appendix A Criterion 6(1): Moisture Storage Capacity of Cover Page 34 of 70
REFERENCES:
Benson, C.H. W.H. Albright, W.H., Fratta, D.O.,Tinjum, J.M., Kucukkirca, E., Lee, S.H., J. Scalia, J.,
Schlicht, P.D., and Wang, X. 2011. Engineered Covers for Waste Containment: Changes in Engineering
Properties and Implications for Long-Term Performance Assessment (in 4 volumes). NUREG/CR-7028,
Prepared for the U.S. Nuclear Regulatory Commission, Washington, D.C., December 2011.
Denison Mines (USA) Corp., 2011. Reclamation Plan, White Mesa Mill, Blanding, Utah, Radioactive
Materials License No. UT1900479, Revision 5.0, September 2011.
Hauser, V.L., Weand, B.L., and Gill, M.D. 2001. Alternative Landfill Covers. July 2001.
Hauser, V.L and D.M. Gimon 2004. Evaluating Evapotranspiration (ET) Landfill Cover Performance
Using Hydrologic Models. January 2004.
Khire, M.V., Benson, C.H., and Bosscher, P.J. 2000. Capillary Barriers: Design Variables and Water
Balance. Journal of Geotechnical and Geoenvironmental Engineering. August 2000.
McCartney, J.S, and Zornberg, J.G. 2006. Decision Analysis for Design of Evapotranspirative Landfill
Covers”, Proceedings UNSAT ’06, April 2-6, Carefree, AZ, ASCE, pp. 694-705.
U.S. Nuclear Regulatory Commission (NRC) 2003. Standard Review Plan (NUREG–1620) for Staff
Reviews of Reclamation Plans for Mill Tailings Sites under Title II of the Uranium Mill Tailings
Radiation Control Act”, NUREG-1620, June, 2003.
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 35 of 70
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10 CFR40 APPENDIX A,
CRITERION 1; INT 04/1: EVALUATION OF POTENTIAL FLOW THROUGH TAILINGS
CELL LINERS
REGULATORY BASIS:
Refer to UAC R313-24-4, which invokes the following requirement from 10CFR40, Appendix A, Criterion 1:
The general goal or broad objective in siting and design decisions is permanent isolation of tailings and
associated contaminants by minimizing disturbance and dispersion by natural forces, and to do so
without ongoing maintenance. For practical reasons, specific siting decisions and design standards must
involve finite times (e.g., the longevity design standard in Criterion 6). The following site features which
will contribute to such a goal or objective must be considered in selecting among alternative tailings
disposal sites or judging the adequacy of existing tailings sites:
• Remoteness from populated areas;
• Hydrologic and other natural conditions as they contribute to continued immobilization and
isolation of contaminants from ground-water sources; and
• Potential for minimizing erosion, disturbance, and dispersion by natural forces over the long
term.
• The site selection process must be an optimization to the maximum extent reasonably achievable
in terms of these features.
• In the selection of disposal sites, primary emphasis must be given to isolation of tailings or
wastes, a matter having long-term impacts, as opposed to consideration only of short-term
convenience or benefits, such as minimization of transportation or land acquisition costs. While
isolation of tailings will be a function of both site and engineering design, overriding
consideration must be given to siting features given the long-term nature of the tailings hazards.
Tailings should be disposed of in a manner such that no active maintenance is required to preserve
conditions of the site.
INTERROGATORY STATEMENT:
Refer to Appendix L (Evaluation of Potential Water Flow through Tailings Cell Liners) of the ICTM
Report: Please provide the following:

1. Revise and provide justification for the estimated saturated hydraulic conductivity of the
compacted foundation [liner bedding] layers underlying the geomembrane in Cells 2 and 3,
which are both comprised of a compacted gravel-sand mixture derived from crushing of loose
sandstone. possibly with washed concrete sand used in some areas);
Response 1 (May 31, 2012):
The material installed beneath the liners in Cells 2 and 3 consists of crushed Dakota
sandstone that was compacted with a smooth drum roller, but in some locations, in
which a smooth base grade was available, portions of the liner were placed over
sections of in situ Dakota sandstone (H. Roberts, 2012). The Second Phase Tailings
Management System Construction Report generally is consistent with this observation:
Energy Fuels Nuclear Inc. (1983) noted that a gravel-sand mixture derived from crushing
of loose [Dakota] sandstone, with some washed concrete sand in some areas, was used
to construct the compacted bedding layer immediately beneath the liner in Cell 3; and
that a similar process and materials were used for the liner bedding material in Cell 2.
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 36 of 70
No estimates of the hydraulic conductivity for the in-place liner bedding materials
beneath Cells 2 and 3 are available for incorporation into the analyses. However, to
support the 2010 Revised ICTM Report, saturated hydraulic conductivity tests in the
vertical direction (Ks) were measured on in-tact core-samples of the Dakota sandstone
using a flexible wall permeameter (Appendix B). While these hydraulic measurements
were used to support the solute transport model for the bedrock vadose zone the data
also can be used as a starting point to assess the potential Ks value for the liner bedding
materials.
The following in-tact core-sample intervals, measured in feet below ground surface (ft
bgs) for the various monitoring wells (MWs), and the corresponding Ks measurements in
meters per second (m/s) are as follows for the Dakota sandstone (Appendix B):
• MW-30 35.5-36.0 ft bgs was measured at 8.1x10-6 m/s.
• MW-30 44.0-44.5 ft bgs was measured at 8.2x10-8 m/s.
• MW-23 55.5-56.0 ft bgs was measured at 1.1x10-6 m/s.
For these three measurements, the corresponding geometric mean was equal to
9.0x10-7 m/s (9.0x10-5 cm/s), which is approximately eleven times higher than the
minimum value and nine times lower than the maximum value. While it may be difficult
to extrapolate these in-tact core-sample measurements to that of crushed and
compacted sandstone, the geometric mean for the core samples of Dakota sandstone
can be used as a starting point to support the Ks value assumed for the liner underlay
materials. In actuality, the Ks value for these materials could be higher or lower than the
measurements based on the reported variability and because the liner underlay material
consists of crushed Dakota sandstone which may have experienced some compaction
from hydraulic loading and tailings deposition. The Ks value used in the calculations can
be compared to the calculated potential water flux rates through the liners for Cells 2 and
3 to evaluate the credibility of the assumption that the geometric mean Ks value for the
core samples of Dakota sandstone can be used as input.
Other information needed to estimate potential water flux rates through the liners as
summarized in Appendix L includes the pressure head on the liner, the coefficient of
contact between the liner and the liner underlay materials, the thickness of the liner
underlay materials, and the defect sizes and frequencies. For Cells 2 and 3 the
following assumptions were incorporated into the calculations consistent with the
approach adopted in Appendix L:
• Contact between the geomembrane and the underlying soil bedding materials is
assumed to be good, thus the coefficient of contact (dimensionless empirical
coefficient) is assumed to be 0.21 in the calculations. Good contact assumes
that the liner was laid on a well-prepared, smooth soil surface with good wrinkle
control.
• The base case pressure head scenario assumed a maximum pressure head
equal to 5.82 meters.
• The thickness of the liner underlay materials is based on as-constructed records
(D’Appolonia Consulting Engineers, 1982) and was assumed to be 0.15 meters
(6 inches).
• The circular defect size for a small hole (SH) was assumed to equal 2 millimeters
in diameter, while a large hole (LH) was assumed to equal 10 millimeters in
diameter.
• Various defect frequencies for different scenarios were assumed as summarized
below:
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 37 of 70
o Upper bound scenario assumed 1 SH and 3 LH defects per acre.
o Base case scenario assumed 1 SH and 1 LH defects per acre.
o Lower bound scenario assumed 1 SH defect per acre.
• The footprint for each of Cells 2 and 3 was assumed to be 283,290 square
meters (70 acres).
The calculated potential water flux rates through the liners can be used to evaluate
which calculation methodology should be used based on criteria presented by Giroud et
al. (1997):
• Flow through a liner overlying a low permeability medium should use the Giroud
1997 Equation (Giroud, 1997). The applicability of using the Giroud 1997
Equation can be assessed by calculating the upper limit of the Ks value (kG) for
applying this equation (i.e., Ks < kG). The calculations in Appendix L used the
Giroud (1997) Equation.
• Flow through a liner overlying a permeable medium should use the Bernoulli
Equation (Giroud and Bonaparte, 1989). The applicability of using the Bernoulli
Equation can be assessed by calculating the lower limit of the Ks value under
practical conditions (kB’) for applying this equation (i.e., Ks > kB’).
• Flow through a liner overlying a semi-permeable medium, representative of in-
between conditions, should use the Giroud et al. 1997 Equation (Giroud et al.,
1997). This equation is applicable for conditions in which kG < Ks < kB’.
The value for kB’ for a LH defect is 1x10-1 m/s while the value for kG for a LH defect
assuming 1.0 meter of head is approximately 2x10-5 m/s. The value for kB’ for a SH
defect is 4x10-3 m/s while the value for kG for a SH defect assuming 1.0 meter of head is
approximately 3x10-7 m/s. Based on the Dakota sandstone geometric mean value for Ks
(9x10-7 m/s) the Giroud 1997 Equation is applicable for the LH defects and is close to the
limit for the SH defects (3x10-7 m/s). However, for the SH defects, values of Ks and KG
are close enough such that the use of the Giroud 1997 Equation is considered to be
acceptable for the calculations presented herein.
The maximum potential water flux rates through the liners using the geometric mean Ks
of the Dakota sandstone for the liner underlay materials are summarized below. These
values are compared to the maximum potential water flux rates through the liner as
presented in the 2010 Revised ICTM Report Appendix L using a lower Ks value. The Ks
value in Appendix L was based on the geometric mean Ks for the compacted platform fill
(2x10-9 m/s; Appendix E and Attachment E-1), because at the time the study was
completed it was believed that these materials comprised the liner underlay. The data
used to calculate the geometric mean Ks for platform fill were taken from measurements
reported by Chen and Associates (1978; 1979).
The maximum potential water flux rates, assuming the base case pressure head
scenario and two Ks values for the liner underlay materials, are as follows:
• Using the geometric mean Ks for the Dakota sandstone (9x10-7 m/s) the
maximum calculated potential water flux rates are approximately 1,650, 760, and
320 mm/yr for the various defect frequency scenarios.
• Using the geometric mean Ks for the compacted platform fill (2x10-9 m/s) the
maximum calculated potential water flux rates are approximately 18, 8, and 4
mm/yr for the various defect frequency scenarios.
• The calculated maximum potential water flux rates through the liners using the
geometric mean Ks for the Dakota sandstone are approximately 100 times higher
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 38 of 70
than the maximum values that were calculated using the geometric mean Ks for
the compacted platform fill.
Two order-of-magnitude calculations were completed using the estimated maximum
potential water flux rates through the liner to place these values into context. The first
calculation evaluated the time required to drain all of the water from the impoundment.
The second calculation evaluated the advective transport time to the upper surface of
the water table. The drainage time calculations assumed a 5.82 meter saturated
thickness, tailings porosity of 40 percent, a 283,290 square meter footprint, and no
additional input of process water. The travel time calculations assumed a unit gradient
and steady state conditions and was calculated according to the following equation:
Time = Distance / Velocity in which the Velocity = Water Flux / Volumetric Water
Content. The travel time calculations assumed a minimum vadose zone thickness of
12.8 meters from Appendix C and volumetric water contents (VWCs) from Appendices B
and C that correspond to the above referenced maximum potential water flux rates
through the liner. The calculated advective transport time will result in longer timeframes
than a modeled scenario since the maximum pressure head (and water flux rate) occur
after the start-up of operations and transient conditions would act to decrease modeled
travel times.
• Using the geometric mean Ks for the Dakota sandstone (9x10-7 m/s):
o The tailings impoundment was calculated to drain within approximately 1
year and 7 years for the maximum leakage rate and upper and lower
bound defect frequency scenarios, respectively.
o The calculated advective transport time to the upper surface of the water
table is approximately 1 year and 5 years for the upper and lower bound
defect frequency scenarios (VWCs of 16.0 percent and 13.6 percent),
respectively.
• Using the geometric mean Ks value for the compacted platform fill (2x10-9 m/s):
o The tailings impoundment was calculated to drain within approximately
185 years and 953 years for the maximum leakage rate and upper and
lower bound defect frequency scenarios, respectively.
o The calculated advective transport time to the upper surface of the water
table is approximately 69 years and 259 years for the upper and lower
bound defect frequency scenario (VWCs of 9.7 percent and 8.1 percent),
respectively.
If the Ks of the liner underlay is assumed to equal (2x10-8 m/s), ten times lower than the
value assigned in Appendix L of the 2010 Revised ICTM Report, the maximum
calculated potential water flux rate through the liner for the base case pressure head
scenario is 100 mm/yr for the upper bound defect frequency scenario.
• Using a Ks value that is 10 times lower than the compacted platform fill assigned
in Appendix L:
o The calculated timeframe for the impoundment to drain is approximately
24 years.
o The calculated advective transport time to the upper surface of the water
table is approximately 15 years for the upper bound defect frequency
scenario (VWC of 11.9 percent).
These order-of-magnitude calculations suggest that the Ks value assigned to the liner
bedding material is biased high when using the geometric mean of the Dakota
sandstone and is probably closer to the value assumed in Appendix L (2x10-9 m/s).
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 39 of 70
Additionally, there is strong evidence to suggest that no significant leakage has occurred
through the liner systems beneath Cells 2 & 3 over the past 30 years (e.g., Cell 2 went
into service during 1980 and Cell 3 went into service during 1983). Evidence that: (1)
Cells 2 and 3 are not significantly leaking, and (2) the Ks value for the liner bedding
material is lower than the geometric mean of the Dakota sandstone and closer to the
value assumed in Appendix L, is listed below:
• No significant leakage indicated by the leak detection systems.
• No leakage indicated by mounding of the perched aquifer water table surface.
• No observations of contamination (e.g., acid leaching, dissolution of carbonates,
gypsum precipitation, staining) in the bedrock core samples were recorded
during drilling of monitoring wells installed adjacent to the cells during spring
2005 as noted during inspection of the core by MWH (Appendix C).
• Total uranium was detected at background levels in bedrock core samples
collected while drilling monitoring wells adjacent to the cells as noted by analyses
presented in Appendix A.
• No contaminants detected in groundwater at levels above natural background
concentrations (INTERA, 2007a; 2007b; 2008). The lack of groundwater
contamination is corroborated by the following:
o The apparent groundwater age beneath the tailings cells is dominated by
water that is at least approximately 55 years old as determined from
measurements of tritium and helium in groundwater within the vicinity and
downgradient of the mill (Hurst and Solomon, 2008). In other words,
recharge at the land surface occurred prior to 1952 (Schwartz and Zhang,
2003) and takes at least 55 years to reach the perched aquifer.
o Groundwater beneath the tailings cells is not influenced by more modern
water that may have leaked from the tailings cells.
o No contaminants detected in groundwater as evaluated through
measurements of stable isotopes for oxygen and sulfate in groundwater
within the vicinity and downgradient of the mill (Hurst and Solomon, 2008)
indicative that significant leakage from the tailings cells have not
occurred.
Furthermore, the accuracy of the Giroud 1997 Equation was evaluated using numerical
simulations completed by Foose et al. (2001). Overall, leakage rates predicted with the
Giroud 1997 Equation were higher and therefore more conservative than those predicted
based on hydraulic theory for facilities constructed with a composite liner. Therefore,
Giroud’s 1997 Equation is anticipated to over-predict flow rates through the liners for
Cells 2 and 3. Additionally, one of the assumptions implicit with the Giroud 1997
Equation is that the material above the liner (e.g., slimes drain and overlying tailings)
readily transmits all available water, which may not be accurate for consolidated fine-
grained tailings that may effectively seal the defects through physical (compaction and
plugging by tailings) or chemical effects (precipitation of minerals such as gypsum or
amorphous silica in the slimes drains). In reality, the tailings may limit the transmission
of water, thus actual flow rates for any given hole size could be less than the calculated
potential flow rates through the liners.
Overall, the advective transport ranges, age dating of groundwater, geochemical
observations within groundwater, conservative approach of the Giroud 1997 Equation,
and potential limitations of the Giroud 1997 Equation suggest that the Ks value assigned
to the liner underlay materials using the value assumed in Appendix L is considered to
be a reasonable and appropriate assumption, and that an attempt to decrease this value
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 40 of 70
would result in potential leakage rates that do not appear to be realistic (i.e., too
conservative). Therefore, a higher Ks for the liner bedding materials does not seem to
be justified to represent potential in-place liner conditions beneath Cells 2 and 3 and the
calculations presented in the 2010 Revised ICTM Report do not require adjustment.
References for Response 1 (May 31, 2012):
Chen and Associates, Inc., 1978. Earth Lined Tailings Cells, White Mesa Uranium
Project, Blanding, Utah. Prepared for Energy Fuels Nuclear, Inc. 18 July.
Chen and Associates, Inc., 1979. Soil Property Study, Proposed Tailings Retention
Cells, White Mesa Uranium Project, Blanding, Utah. Prepared for Energy Fuels
Nuclear, Inc. January 23.
D’Appolonia Consulting Engineers, Inc., 1982. Construction Report, Initial Phase –
Tailings Management System, White Mesa Uranium Project, Blanding, Utah.
Prepared for Energy Fuels Nuclear, Inc.
Energy Fuels Nuclear, Inc., 1983. Construction Report, Second Phase Tailings
Management System, White Mesa Uranium Project, SUA-1358, Docket 40-8681.
Foose, G.J., C.H. Benson, and T.B. Edil, 2001. Predicting leakage through composite
landfill liners, Journal of Geotechnical and Geoenvironmental Engineering
127(6): 510-520.
Giroud, J.P., and R. Bonaparte, 1989. Leakage through Liners Constructed with
Geomembrane Liners, Parts I, II, and Technical Notes, Geotextiles and
Geomembranes 8:27-67, 71-111.
Giroud, J.P., 1997. Equations for calculating the rate of liquid migration through
composite liners due to geomembrane defects, Geosynthetics International 4(3-
4):335-348.
Giroud, J.P., King, T.D., Sanglerat, T.R., Hadj-Hamou, T., and Khire, M.V., 1997. Rate
of Liquid Migration Through Defects in a Geomembrane Placed on a Semi-
Permeable Medium, Geosynthetics International 4(3-4):349-372.
Hurst, T.G, and D.K. Solomon, 2008. Summary of Work Completed, Data Results,
Interpretations, and Recommendations for the July 2007 Sampling Event at the
Denison Mines, USA, White Mesa Uranium Mill, near Blanding, Utah. Prepared
for the Utah Division of Radiation Control. May.
INTERA, Inc., 2007a. Revised Background Groundwater Quality Report: Existing Wells
for Denison Mines (USA) Corp.’s White Mesa Uranium Mill Site, San Juan
County, Utah. Prepared for Denison Mines (USA) Corporation. October.
INTERA, Inc., 2007b. Revised Addendum Evaluation of Available Pre-Operational and
Regional Background Data Background Groundwater Quality Report: Existing
Wells for Denison Mines (USA) Corp.’s White Mesa Mill Site, San Juan County,
Utah. Prepared for Denison Mines (USA) Corporation. November.
INTERA, Inc., 2008. Revised Addendum Background Groundwater Quality Report: New
Wells for Denison Mines (USA) Corp.’s White Mesa Mill Site, San Juan County,
Utah. Prepared for Denison Mines (USA) Corporation. April.
Roberts, H., 2012. Email correspondence from Harold Roberts, Denison Mines (USA)
Corp., to Ryan Jakubowski, MWH Americas, Inc. April 10.

September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 41 of 70
Schwartz, F., and H. Zhang, 2003. Fundamental of groundwater, John Wiley & Sons,
New York, NY.

2. Provide additional justification to support the various assumed lower bound, base case, and
upper bound geomembrane defect frequencies for the liners in Cells 2, 3, 4A, and 4B. Justify the
upper bound assumption of 1 small hole and 3 large hole defects per acre for the geomembrane
defect frequency in the Cells 2 and 3 liners and the assumption of 1 small-hole defect per acre as
the base case assumption for the geomembrane defect frequency for Cells 4A and 4B, or
alternatively, provide revised assumed defect frequencies to ensure that the assumed defect
frequencies are adequately conservative and reasonably represent actual or potential in-place
liner conditions; and
Response 2 (May 31, 2012):
Cells 2 and 3
The defect frequencies assumed in Appendix L for Cells 2 and 3, as summarized in
response to Comment One of this interrogatory, were based on a range of values
reported in the literature. General recommendations follow that one SH defect is
anticipated per acre, while the number of LH defects will depend on the quality of the
installation. Good installation quality for the liner may have between one to four LH
defects per acre. Good installation quality was assigned for the liners in Cells 2 and 3,
and SH defects were also incorporated into the analysis completed in Appendix L. The
actual defect frequencies of the installed liners beneath Cells 2 and 3 are unknown. It is
conceivable that the defect frequencies may be higher or lower than the values assumed
in Appendix L. It is also conceivable that any defects that remained after construction
and quality control have subsequently been sealed through physical (compaction and
plugging by loaded tailings) or chemical effects (though the precipitation of minerals
such as gypsum or amorphous silica in the slimes drains).
If, for example, the following defect frequencies are assumed:
• Upper bound scenario assuming 6 SH and 6 LH defects per acre
• Base case scenario assuming 4 SH and 4 LH defects per acre
• Lower bound scenario assuming 2 SH and 2 LH defects per acre
This would represent a large increase and therefore overly conservative approach for
the defect frequencies compared to Appendix L. The corresponding calculated
maximum potential water flux rates through the liners are summarized below:
• Using the geometric mean Ks for the compacted platform fill, and the assumed
defect frequencies in the 2010 Revised ICTM Report (Appendix L), the maximum
calculated potential water flux rates through the liners are approximately 18, 8,
and 4 mm/yr for the various defect frequency scenarios.
• Using the same Ks value but the revised defect frequencies referenced above the
maximum calculated potential water flux rates through the liners are
approximately 50, 35, and 15 mm/yr for the various defect frequency scenarios.
The advective transport time to the upper surface of the water table for the latter
calculations is equal to approximately 28 years and 38 years for the upper bound and
base case defect frequencies (VWCs of 11.0 percent and 10.5 percent), respectively. A
flux rate of 15 mm/yr would take approximately 75 years to reach the upper surface of
the perched aquifer (VWC of 8.8 percent). Compared to the calculated potential water
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 42 of 70
flux rates through the liner presented in the response to Comment One of this
interrogatory, and using the same Ks for the liner underlay materials, approximately
equivalent flux rates were calculated for the scenarios that assumed 1 SH and 3 LH
(upper bound) and 2 SH and 2 LH (lower bound) defect frequencies.
The higher defect frequencies assigned above and the corresponding transport times
through the vadose zone are not supported by the geochemical observations that imply
that groundwater beneath and downgradient of the mill is at least approximately 55
years old as determined from measurements of tritium and helium in groundwater within
the vicinity of the mill (Hurst and Solomon, 2008). Therefore, a higher defect frequency
does not seem to be justified to represent potential in-place liner conditions beneath
Cells 2 and 3, and the calculations presented in the 2010 Revised ICTM Report do not
require adjustment.
Cells 4A and 4B
The defect frequencies assumed in Appendix L for Cells 4A and 4B were based on a
range of values reported in the literature for modern constructed liner systems with
excellent quality control and quality assurance. General recommendations follow that
one SH defects is anticipated per acre, while the number of LH defects will depend on
the quality of the installation. An excellent installation quality for the liner may have up to
one LH defect per acre, while good installation quality for the liner may have between
one to four LH defects per acre. Excellent installation quality was assigned for the liners
in Cells 4A and 4B, and only one scenario was considered assuming one SH defect per
acre for the secondary liner (Appendix L). The cell footprint was assumed to be 161,880
square meters (40 acres). The calculations presented in Appendix L only included
potential water flux rates through the secondary liner using the maximum amount of
head (0.004 meters) on the secondary liner for Cell 4A provided by Geosyntec
Consultants (2006). Cell 4B was assumed to have the same amount of head on the
secondary liner because of their similar designs.
Calculated potential flux rates through the primary liner were not provided in Appendix L.
However, the volume of solutions pumped from the leak detection system (LDS) beneath
Cells 4A and 4B can provide a check against the calculated potential water flux rates
through the liners based on the assumed defect frequency. Measurements of solution
volumes pumped from the LDS were provided by Denison Mines (H. Roberts, 2012).
• Cell 4A went into service on October 1, 2008. Cumulative solution volumes
pumped from the LDS are reported at:
o 253,955 gallons from the start-of-service through November 11, 2009.
This amounts to a measured water flux rate through the primary liner of
approximately 5 mm/yr from October 1, 2008 through November 11, 2009
(406 days of data).
o 264,010 gallons from the start-of-service through April 1, 2011. This
amounts to a measured water flux rate through the primary liner of
approximately 0.2 mm/yr from November 11, 2009 through April 1, 2011
(506 days of data).
o 269,486 gallons from the start-of-service through April 30, 2012. This
amounts to a measured water flux rate through the primary liner of
approximately 0.1 mm/yr from April 1, 2011 through April 30, 2012 (395
days of data) and approximately 2 mm/yr from October 1, 2008 through
April 30, 2012 (1,307 days).
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 43 of 70
• Cell 4B went into service on February 2, 2011. Cumulative solution volumes
pumped from the LDS are reported at:
o 361,013 gallons from the start-of-service through April 30, 2012. This
amounts to a measured water flux rate through the primary liner of
approximately 7 mm/yr from February 2, 2011 through April 30, 2012 (454
days of data). The measured leakage rates for Cell 4A and Cell 4B are
similar when nearly equivalent duration intervals were examined (e.g.,
approximately the first year of operations). However, it appears that the
flow rate totalizer started having problems during late November 2011
and may have been recording a higher pumping rate than actual
conditions.
The maximum calculated potential water flux rate through the primary liner in Cells 4A
and 4B can be calculated using the Bernoulli Equation. The maximum pressure head for
Cell 4A and Cell 4B that is currently being measured is approximately 7.6 and 4.6
meters of head, respectively. The maximum potential flux rate through the primary liner
integrated using the defect frequency scenario identified above (1 SH defect per acre) is
equal to approximately 179 mm/yr for Cell 4A and 140 mm/yr for Cell 4B. The calculated
value is approximately 35 to 1,500 times higher than the measured value for Cell 4A and
approximately 20 times higher than the measured value for Cell 4B. While a higher
defect frequency could be present for the secondary liner beneath the cells the
significant discordancy between the measured and calculated values indicates that the
assumed defect frequency for liner conditions beneath Cells 4A and 4B and the
calculations presented in the 2010 Revised ICTM Report do not require adjustment.
References for Response 2 (May 31, 2012):
Geosyntec Consultants, 2006. Cell 4A Lining System Design Report for the White Mesa
Mill, Blanding, Utah. Prepared for International Uranium (USA) Corporation.
2006.
Hurst, T.G, and D.K. Solomon, 2008. Summary of Work Completed, Data Results,
Interpretations, and Recommendations for the July 2007 Sampling Event at the
Denison Mines, USA, White Mesa Uranium Mill, near Blanding, Utah. Prepared
for the Utah Division of Radiation Control. May.
Roberts, H., 2012. Email correspondence from Harold Roberts, Denison Mines (USA)
Corp., to Melanie Davis, MWH Americas, Inc. May 1.

3. Revise the calculations of potential flow through the Cell 3 and Cell 2 liner systems using a more
suitable and appropriate methodology such as the modified methodology developed by Giroud
and others (Giroud et al. 1997a) for estimating the rate of liquid migration through defects in a
geomembrane placed on a semi-permeable medium. Utilize and incorporate information from
Giroud et al. 1997a as appropriate to interpolate between results obtained using the Giroud
equation (as it was used in Appendix L of the current ICTM Report) and results that would be
obtained using Bernouli’s equation.
Response 3 (May 31, 2012):
The methodology implemented to calculate potential water flux rates through the liners
beneath Cells 2 and 3 was provided in response to Comment One of this interrogatory.

September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 44 of 70
BASIS FOR INTERROGATORY:
The Construction Report, Second Phase Tailings Management System (Energy Fuels Nuclear, Inc. 1983)
indicates that a gravel-sand mixture derived from crushing of loose sandstone, with some washed
concrete sand in some areas, was used to construct the compacted bedding layer immediately underlying
the geomembrane in Cell 3. That report also indicates that a similar process and similar materials were
used for constructing the compacted bedding layer beneath the geomembrane liner in Cell 2. On page L-
7 of Appendix L, the saturated hydraulic conductivity for these compacted bedding layers is assumed to
be 2.0 x 10-7 cm/sec. This value is likely too low to be representative of these in-place compacted
materials. Giroud et al. 1997a developed a modified methodology for calculating the rate of liquid
migration through a defect in a geomembrane liner underlain by a semi-permeable medium. This
modified methodology appears to be more appropriate for calculating leakage rates through the
geomembrane liners in Cells 3 and 2 and should therefore be used instead of the method used in
Appendix L for estimating flow through defects in liners in Cells 2 and 3.

Additional justification should be provided to support the various assumed geomembrane defect
frequencies for the different geomembrane liners in Cells 2 and 3 vs. Cells 4A and 4B for the lower
bound, base case, and upper bound scenarios. Additional justification should be provided to demonstrate
why higher assumed base-case and/or upper bound defect frequencies would not be considered more
reasonably conservative assumptions and more reasonably representative of actual or potential in-place
liner conditions for some or all of the cell liners for the purpose of estimating potential leakage rates
through the various liner systems. Justify why a lower bound assumption of 1 small defect per acre for
Cells 2 and 3 (the same assumption as made for the base case for Cells 4A and 4B) is adequately
conservative for the Cell 2 and Cell 3 liners given that they were constructed 30 or more years ago when
construction quality assurance practices might have been somewhat less rigorous than those would have
been used during installation of high density polyethylene geomembranes in Cells 4A and 4B.
Additionally, the merit and applicability of assuming a geomembrane defect frequency (four defects per
hectare (10,000 m2) analogous to that discussed in Giroud et al. 1997b, which suggests an average of
approximately 1.62 defects per acre for a typical defect frequency for a modern constructed liner, should
be discussed for the Cells 4A and 4B liners, particularly given that this defect frequency was used in
previous leakage equations for calculating leakage rates to support the design of the liner system in Cell
4B. Further, for assessing a range of potential upper bound (worst –case) defect frequencies for the Cell
2 and Cell 3 liners, consideration should be given to other published data, such as Nosko and Touze-Folz
2000, which provide estimates of actual liner defect frequencies (the Nosko and Touze-Folz data suggest
a post-construction, pre repair average defect frequency of approximately 5 defects per acre of liner
installed - based on study of over 300 landfill liners before construction quality assurance measures were
undertaken to reduce the presence of defects but not eliminate them completely). Allowance should also
be made for additional defects to occur after liner construction is complete.

REFERENCES:
Denison Mines (USA) Corp. 2010. Revised Infiltration and Contaminant Transport Modeling Report,
White Mesa Mill Site, Blanding, Utah (Revision 2), March 2010.
Energy Fuels Nuclear, Inc. 1983. Construction Report, Second Phase Tailings Management System.
White Mesa Uranium Project. SUA-1358. Docket 40-8681.
Giroud, J.P., King, T.D., Sanglerat, T.R., Hadj-Hamou, T., and Khire, M.V. 1997a. “Rate of Liquid
Migration Through Defects in a Geomembrane Placed on a Semi-Permeable Medium”, Geosynthetics
International, Vol. 4, Nos. 3-4, pp. 349-372.
September 10, 2012
Interrogatory 04/1: R313-24-4; 10CFR40; Appendix A Criterion 1: Evaluation of Potential Flow through Tailings Cell Liners Page 45 of 70

Giroud, J.P., King, T.D., Sanglerat, T.R., Hadj-Hamou, T., and Khire, M.V. 1997b. “Leachate Flow in
Leakage Collection layers Due to Geomembrane Defects”, Geosynthetics International, Vol. 4, Nos. 3-4,
pp. 215-2922.

Nosko, V. and Touze-Foltz, N. 2000. Geomembrane Liner failure: Modeling of its influence on
Contaminant Transfer. Proc. 2nd European Conf. on Geosynthetics, Bologna, Italy, 2: 557-560.
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 46 of 70
INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4 -05/1: CONTAMINANT
TRANSPORT MODELING
PRELIMINARY FINDING:
Refer to UAC R313-24-4, which invokes the following requirement from 10CFR40, Appendix A, Criterion 1:
The general goal or broad objective in siting and design decisions is permanent isolation of tailings and
associated contaminants by minimizing disturbance and dispersion by natural forces, and to do so
without ongoing maintenance. For practical reasons, specific siting decisions and design standards must
involve finite times (e.g., the longevity design standard in Criterion 6). The following site features which
will contribute to such a goal or objective must be considered in selecting among alternative tailings
disposal sites or judging the adequacy of existing tailings sites:
• Remoteness from populated areas;
• Hydrologic and other natural conditions as they contribute to continued immobilization and
isolation of contaminants from ground-water sources; and
• Potential for minimizing erosion, disturbance, and dispersion by natural forces over the long
term.
• The site selection process must be an optimization to the maximum extent reasonably achievable
in terms of these features.
• In the selection of disposal sites, primary emphasis must be given to isolation of tailings or
wastes, a matter having long-term impacts, as opposed to consideration only of short-term
convenience or benefits, such as minimization of transportation or land acquisition costs. While
isolation of tailings will be a function of both site and engineering design, overriding
consideration must be given to siting features given the long-term nature of the tailings hazards.
Tailings should be disposed of in a manner that no active maintenance is required to preserve conditions
of the site.
INTERROGATORY STATEMENT:
1. Refer to Revised ICTM Report, Section 2.2 Site Characteristics and Section 4.3 Uncertainty
and Assumptions: Provide additional information on the potential presence and distribution of
fractures and/or joints, and uncemented/higher permeability intervals in the unsaturated zone
portions of the Dakota Sandstone and Burro Canyon geologic units underlying the site area,
including the footprint area of and downgradient vicinity of Cells 1, 2, 3, 4A, and 4B. Describe
the possible effects of such fractures and/or joints, and uncemented/higher permeability intervals,
on the flow and transport of potential contaminants through the vadose zone, including potential
effects on estimated contaminant travel times to the perched groundwater zone beneath the
tailing management cells.
Response 1 (May 31, 2012):
The potential occurrence of increased flow and transport from the presence and
distribution of fractures and/or joints in the unsaturated zone of the Dakota sandstone
and Burro Canyon Formation underlying the site area is not supported by geologic and
hydrogeologic observations as summarized below:

• The lack of faulting and lack of extensive jointing combined with the largely
structurally intact and sub-horizontal dip of the geologic units should act to limit
the downward movement of water within the bedrock vadose zone. Structural
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 47 of 70
control of water movement is likely limited due to the absence of faults and the
apparent low frequency of joints.
o The Dakota sandstone and Burro Canyon Formation are nearly flat-lying
with reported dips of less than 1 degree to the south (D’Appolonia, 1979).
o No faults have been mapped or identified within the Blanding Basin
(Kirby, 2008) or White Mesa area (D’Appolonia, 1979).
o The Dakota sandstone and Burro Canyon Formation are relatively devoid
of joints. Where present, primary joints are nearly vertical and north
striking, and commonly several meters in length; secondary joints are
oriented east-west and commonly terminate at the intersection with
primary joint surfaces (Kirby, 2008).
ƒ Observational data collected during a 1994 drilling program of
three perched monitoring wells and four angled borings beneath
Cell 3 and Cell 4A concluded that few fractures were present in
the cores or observed in video logs. And where such features
were present the fractures were closed and/or sealed with gypsum
(HGC, 2010a).
• The potentiometric surface map in the vicinity and downgradient of the tailings
cells resembles that of a perched aquifer system in that the projected and
inferred lines of equipotential are nearly parallel to one another (Figure 2-5 of the
2010 Revised ICTM).
• Age dating of groundwater in the vicinity and downgradient of the tailings cells
indicates that infiltration takes longer than 55 years to travel through the vadose
zone except in the vicinity of the wildlife ponds (Hurst and Solomon, 2008). The
recharge mound near the wildlife ponds, combined with the absence of tritium in
groundwater beneath the mesa, imply that the bedrock vadose zone can
generally be considered as recharge-limited rather than permeability-limited
(Hurst and Solomon, 2008), which corroborates the assumption that recharge
and flow through the bedrock vadose zone is predominately via matrix flow rather
than through fracture flow.
The potential occurrence of increased flow and transport from the presence of
uncemented and/or higher permeability intervals in the unsaturated zone of the Dakota
sandstone and Burro Canyon Formation underlying the site area is not supported by
geologic and hydrogeologic observations as summarized below:
• Sub-horizontal units of potentially more permeable lithologic units (e.g.,
conglomerate and sandstone with intermittent conglomeratic features) cannot be
correlated as broad continuous lenses beneath the tailings cells (HGC, 2010b).
However, these units can be correlated short distances between boreholes as
thin discontinuous lenses of limited thickness.
• A poor correlation between conglomeritic intervals and enhanced permeability
has been observed through interpretation of hydraulic testing (HGC, 2010a):
o Hydraulic conductivity tests in the horizontal direction for conglomeratic
lenses beneath Cell 4B (MW-16) were approximately 5.1x10-7 m/s which
is within the middle of the range of values (2.9x10-7 to 9.1x10-6 m/s)
reported for the more massive sandstone lithology (HGC, 2010a).
o Hydraulic conductivity tests in the horizontal direction for conglomeratic
lenses beneath Cell 3 (angled borings) were lower than 1x10-7 m/s for
three tests and higher than 1x10-7 m/s for one test (HGC, 2010a).
o The similar hydraulic behavior between the sandstone and conglomeritic
lenses can be explained because the conglomerate matrix is represented
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 48 of 70
by sandstone and the gravel-sized clasts within the conglomerate are
generally present in low percentages (less than 30 percent).
• The presence of unidentified high permeability discontinuous conglomeritic layers
of limited thickness within the vadose zone would likely result in more timely
detection of any seepage that may occur; because these units, if present, could
act to spread any seepage over a wider area, and make such fluids less likely to
pass undetected between monitoring wells (HGC, 2010a).
A discussion regarding the material presented above will be included in the next iteration
of the ICTM Report to some extent within Sections 2.2 and 4.3.
References for Response 1 (May 31, 2012):
D’Appolonia, 1979. Engineers Report, Tailings Management System, White Mesa
Uranium Project, Blanding, Utah. Prepared for Energy Fuels Nuclear, Inc. June.
Hurst, T.G, and D.K. Solomon, 2008. Summary of Work Completed, Data Results,
Interpretations, and Recommendations for the July 2007 Sampling Event at the
Denison Mines, USA, White Mesa Uranium Mill, near Blanding, Utah. Prepared
for the Utah Division of Radiation Control. May.
Hydro Geo Chem, Inc. (HGC), 2010a. Letter from Stewart J. Smith of Hydro Geo Chem,
Inc. to David Frydenlund, Esq. of Denison Mines. February 8.
Hydro Geo Chem, Inc. (HGC), 2010b. Letter from Stewart J. Smith of Hydro Geo Chem,
Inc. to David Frydenlund, Esq. of Denison Mines. February 12.
Kirby, S., 2008. Geologic and hydrologic characterization of the Dakota-Burry Canyon
Aquifer near Blanding, San Juan County, Utah, Special Study 123, Utah
Geological Survey.

2. Refer to Revised ICTM Report, Section 2.2.4: Please summarize the geochemical characteristics
of the perched groundwater and discuss in greater detail the potential relevance of perched zone
water geochemistry to the development of specific geochemical modeling input assumptions made
for the vadose zone in Appendix M (address, for example, the effects of dissolved oxygen
concentration, redox conditions).

Response 2 (May 31, 2012):
A brief summary of the geochemical characteristics of the perched groundwater as it
relates to the development of input assumptions for the vadose zone will be included in
the next iteration of the ICTM Report. The following discussion is anticipated to be
included in Appendix M, to support the reactive transport model for the next iteration of
the ICTM Report.
The geochemical modeling requires input assumptions regarding (i) the chemistry of the
tailings pore water, (ii) the chemistry of the bedrock vadose zone pore water, and (ii) the
mineralogy of the bedrock vadose zone. The geochemical characteristics of the vadose
zone pore water can be constrained by the solid phases present within the bedrock and
the pore water chemistry within the perched aquifer. The conceptual model used to
support the geochemical modeling input assumptions for the vadose zone is explained
below, while the conceptual model used to support the geochemical modeling input
assumptions for the tailings pore water seepage chemistry is explained in the response
to Comment Six below.
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 49 of 70
Recharge, especially from the unlined wildlife ponds east of the mill site, represents the
dominant source of the perched groundwater beneath the mill (Hurst and Solomon,
2008). Recharge from water (percolation) within shallow portions of the bedrock is likely
to be near atmospheric conditions with a concentration of dissolved oxygen (DO) on the
order of 8 milligrams per liter (mg/L). As percolation continues to migrate through the
bedrock, the oxidation of organic compounds is likely to consume a portion of the
dissolved and gaseous oxygen through aerobic respiration because this process is the
most energetically favorable redox reaction anticipated to occur within the oxic vadose
zone (Langmuir, 1997). Redox reactions within the bedrock and resultant pore water
chemistry will be controlled by the water to rock and microbial reactions that occur during
transport through the vadose zone. The presence of iron hydroxides and carbonate
minerals in the bedrock suggests that oxic conditions coincident with aerobic respiration
at near neutral pH are likely to dominate within the vadose zone. Therefore, the vadose
zone pore water is anticipated to contain DO at detectable concentrations that reflect
oxic conditions consistent with the presence of hydrous ferric oxide (HFO) within the
bedrock.
Additionally, as the percolation continues to migrate downward through the vadose zone,
the recharge water will eventually mix with the perched groundwater and equilibrate with
minerals present within the zone of saturation. While there are naturally-occurring
concentrations of chloride, sulfate, uranium, and other trace elements in the vadose
zone initially, the modeling assumed zero concentrations as a simplification. Initial
solution concentrations in the vadose zone pore water were estimated by assuming
equilibrium of calcite with HFO (Appendix M), consistent with minerals observed in the
bedrock, such that only calcium, carbonate, and dissolved oxygen were included as
aqueous species initially present within the vadose zone pore water.
Dissolved Oxygen
The concentration of DO is not being measured for wells screened within the perched
aquifer because this analyte is not required by the groundwater discharge permit.
However, DO was measured at a subset of the monitoring wells as part of the evaluation
completed by the University of Utah (G. Hurst, 2012). These measurements are
summarized below (see Table 05/1/2-1). The concentration of DO within groundwater
appears to be related to the vadose zone thickness such that a thinner vadose zone
correlates with a higher DO concentration in groundwater. Therefore, taking the
measurement reported for the minimum vadose zone thickness (MW-30) to be
consistent with the numerical model, the concentration of DO within the vadose zone
pore water is likely greater than the 5 mg/L concentration reported for groundwater at
this monitoring location. Within the bedrock vadose zone, the partial pressure of oxygen
was fixed in the geochemical model assuming a DO concentration of 2 mg/L in pore
water. Therefore, this assumption appears to be valid for the bedrock vadose zone,
based on measurements of DO in groundwater for the minimum vadose zone thickness,
at least as an initial condition.
Redox
The study completed by the University of Utah (Hurst and Solomon, 2008) did not
measure concentrations within groundwater for any complete pair of redox species.
However, in spite of observations of pyrite at some locations in the aquifer matrix, the
presence of DO within groundwater suggests that oxic conditions and aerobic respiration
are likely to dominate redox reactions in the vadose zone and in groundwater influenced
by seepage from the wildlife ponds. Therefore, the oxygen redox couple is anticipated to
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 50 of 70
control redox reactions within the bedrock vadose zone and in groundwater, at least as
an initial condition. Whether or not this assumption would hold true if potential tailings
seepage water was transported through the vadose zone will be discussed in the
response to Comment Six below.
References for Response 2 (May 31, 2012):
Langmuir, D., 1997. Aqueous Environmental Geochemistry, Prentice-Hall.
Hurst, T.G, and D.K. Solomon, 2008. Summary of Work Completed, Data Results,
Interpretations, and Recommendations for the July 2007 Sampling Event at the
Denison Mines, USA, White Mesa Uranium Mill, near Blanding, Utah. Prepared
for the Utah Division of Radiation Control. May.
Hurst, G. 2012. Email correspondence from Greg Hurst, University of Utah, to Ryan
Jakubowski, MWH Americas, Inc. April 22.

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 51 of 70
Table for Response 2 (May 31, 2012):
Table 05/1/2-1. Dissolved oxygen (DO) measurements for a subset of wells screened within the
perched aquifer.
Well Location
DO at Upper
Measurement Point
(mg/L)
DO at Lower
Measurement Point
(mg/L)
Vadose Zone
Thickness
(m)
MW-5
Between Cell 3
and 4B 0.02 0.01 22.9
MW-11
Between Cell 3
and 4A 0.25 0.01 19.5
MW-14 South of Cell 4A 0.14 0.04 22.9
MW-15
Between Cell 4A
and 4B 1.43 0.24 19.5
MW-29
Between Cell 2
and 3 0.41 0.03 22.6
MW-30
Between Cell 2
and 3 5.24 5.12 12.8
MW-31
Between Cell 2
and 3 8.33 9.10 13.1
Notes: Data provided electronically by G. Hurst. Samples were collected during July 2007. Dissolved oxygen
measurements were made at the depths at which the passive diffusion samplers (PDSs) were deployed. The PDSs were
deployed approximately 1 meter above the bottom of the screened interval and 1 meter below the top of the screened
interval. In wells that did not have a fully saturated screened interval (MW-5, 14, 15, 29, 30, 31), the top diffusion sampler
was placed approximately 1 meter below the top of the water level. Vadose zone thickness taken as difference between
cell depth and water table depth.

3. Refer to Revised ICTM Report, Section 3.4.4, Contaminants Modeled: Please provide the
rationale and justification for using aluminum, versus some other constituent, to obtain charge
balance in the HP1 (PHREEQC) simulations.
Response 3 (May 31, 2012):
Aluminum was selected to obtain charge balance because it was not measured in the
various solutions representing the input chemistry, and the solutions had a negative
charge imbalance which suggested cation deficiency. This additional text will be added
to Sections 3.4.4 and Appendix M of the next iteration of the ICTM Report.

4. Refer to Appendix C, Table C-4, p. C-15 in Appendix C to the ICTM Report: Please provide a
corrected maximum ANP value for MW-24 and corrected arithmetic and geometric means for
ANP in the TW4-22 boring. Please confirm the results used in calculating the statistics for all of
the borings and revise the summary statistics presented in Table C-4 as necessary. If the
statistical results in Table C-4 for the entire population change, please revise reactive transport
model as needed, to reflect these changes and report the results.
Response 4 (May 31, 2012):
The values used to compute the statistics were derived from arithmetic averaging of
primary and duplicate sample results (where applicable) after rounding to the nearest
whole number for the acid neutralization potential (ANP) data. Duplicate samples for
materials submitted May 2009 were labeled as being blind (i.e., collected from MW-100
with arbitrary depth intervals). The following sample locations corresponded to the
following sample labels:
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 52 of 70
• MW-24 80.0-80.3 Duplicate labeled as MW-100 11.1-11.3. The primary and
duplicate samples results for ANP were equal to 25 and 28 grams of CaCO3 per
kilogram of rock (g CaCO3/kg rock), respectively.
• MW-24 56.0-56.2 Duplicate labeled as MW-100 16.0-16.2. The primary and
duplicate samples had the same ANP results.
• TW4-22 69.8-70.0 Duplicate labeled as MW-100 19.8-20.0. The primary and
duplicate samples results for ANP were equal to 10 and 29 g CaCO3/kg rock,
respectively.
• MW-23 103.0-103.3 Duplicate was labeled as is to be consistent with the original
sample labeling criteria for samples submitted to the laboratory during February
2007.
Therefore the summary statistics presented in Appendix C are valid and do not require
adjustment. Clarifying text will be included in Appendix C for the next iteration of the
ICTM Report to avoid future confusion.

5. Refer to Appendix M, p. M-10, Paragraphs 2 and 3: Please provide and justify the bulk density
of the bedrock used to convert the ANP and HFO values from rock mass to rock unit volume.
Response 5 (May 31, 2012):
The dry bulk density of the bedrock was assigned to equal 2.0 grams per cubic
centimeter (g/cm3). This value was based on the measurement reported for MW-23
(55.5-56.0 ft) to be consistent with the sample interval test results used to parameterize
the bedrock hydraulic properties (Appendix C). The value assigned to the model is
approximately equal to the arithmetic average (2.1 g/cm3) of the samples tested
(Appendix B and C). Clarifying text will be included in Appendix M for the next iteration
of the ICTM Report to avoid future confusion.

6. Refer to Appendix M, p. M-11, Paragraph 1: Please justify the assumption that the redox
conditions in the tailing slimes drainage and the vadose zone are controlled by the oxygen
(O2/H2O) couple. Perform and report results of sensitivity analyses that assess the dependence of
result on variations in the values of redox value.
Response 6 (May 31, 2012):
The presence of measured DO in groundwater suggests that oxic conditions and aerobic
processes are likely to dominate redox reactions in groundwater and also in the vadose
zone as discussed in the response to Comment Two in this interrogatory. Whether or
not this assumption would hold true if potential tailings seepage water was transported
through the vadose zone is discussed below.
The conceptual model used to support the geochemical modeling input assumptions for
the tailings pore water seepage chemistry is based on the following observations and
hypothesized processes anticipated to occur in tailings pore water at depth.
Concentrations of DO in the tailings pore water at depth are not available because this
analyte is not required to be measured for the GWDP, license or tailings sampling plan.
Surface water or tailings pore water near the upper surface of the impoundment is likely
to be near atmospheric conditions with a concentration of DO on the order of 8 mg/L.
The concentration of DO at depth can only be estimated and bounded by the chemistry
of the tailings solutions combined with potential redox reactions that may occur at depth
within the impoundment. The tailings solutions are acidic and contain detectable
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 53 of 70
concentrations of some organic compounds and elevated concentrations of chloride,
nitrogen (nitrate plus nitrite and ammonium), sodium, and sulfate. The oxidation of
minerals within the ore during acid leaching is the likely cause of elevated concentrations
of iron and other ions in the tailings solutions.
Based on the chemistry of the tailings solutions discussed above the following redox
reactions are anticipated to occur within the tailing pore water at depth to some extent.
The DO within the tailings pore water at depth is likely to be consumed to some extent if
oxidation of organic compounds, ammonium, and ferrous iron occurs. If DO was
completely consumed anoxic conditions would develop. Considering the solution
chemistry of the tailings pore water, and the energetics of the reaction, nitrate reduction
and ammonium oxidation would be the next most favorable anaerobic reaction to occur
once DO is consumed. Therefore, considering the elevated concentrations of nitrogen
species within the tailings pore water at depth, and thermodynamic constraints,
measurements for the nitrogen species can be used to calculate redox (pe) conditions in
the tailings pore water at depth. Redox conditions calculated from the nitrogen species
can then be used to infer redox reactions and mineral stability within the iron system.
Using the chemistry for the tailings pore water at depth (Appendix K) but assuming a DO
concentration of zero:
• The redox value calculated by PHREEQC for the upper bound concentrations is
approximately 629 millivolts (mV). Using the nitrogen calculated pe value iron
would partition approximately 70% as ferrous iron. The saturation index for HFO
is calculated to be 0.07, implying that the solution is at equilibrium with HFO.
• The redox value calculated by PHREEQC for the base case concentrations is
approximately 631 millivolts (mV). Using the nitrogen calculated pe value iron
would partition approximately 67% as ferrous iron. The saturation index for HFO
is calculated to be 0.15, implying that the solution is slightly supersaturated with
HFO, and HFO could precipitate barring any sort of kinetic constraints.
• The redox value calculated by PHREEQC for the lower bound concentrations is
approximately 628 millivolts (mV). Using the nitrogen calculated pe value iron
would partition approximately 72% as ferrous iron. The saturation index for HFO
is calculated to be 0.02, implying that the solution is at equilibrium with HFO.
For these redox conditions, the calculated pe of the tailings pore water at depth is
approximately 10.6 and HFO is anticipated to be a stable phase within the tailings pore
water at depth. These calculations are used to support an alternative conceptual model
for conditions in which the iron redox couple is used to constrain redox conditions within
the vadose zone during reactive transport. This conceptualization differs from the
scenario presented in the 2010 Revised ICTM Report in that the concentration of DO
was fixed to equal 2 mg/L in the tailings pore water. Originally, this assumption was
incorporated into the model to maintain consistency with the DO concentration in the
vadose zone as a simplification. The sensitivity of this assumption is evaluated and
discussed below.
Sensitivity Analysis
The initial concentration of DO within the vadose zone was assigned to equal 2 mg/L, as
supported by data presented in response two of this interrogatory. The oxygen couple
was used to determine the redox of the initial solutions (a pe of 13.6) within the vadose
zone. This is equivalent to the parameterization for the 2010 Revised ICTM Report.
The initial concentration of DO within the tailings pore water was assigned to be infinitely
small (essentially 0 mg/L), and the iron couple was used to determine the redox of the
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 54 of 70
seepage water (a pe of 10.6) and during reactive transport through the vadose zone.
Furthermore, for this sensitivity analysis, concentrations of DO within the vadose zone
during reactive transport were not fixed but allowed to vary as a function of the
geochemical reactions. Subsequent calculations did not assume a fixed Eh, rather
redox conditions were allowed to change as a function of the geochemical reactions with
redox being controlled by the iron couple. This is generally the preferred approach
rather than fixing the Eh to a specific value that may have little to no quantitative
meaning during reactive transport. Additionally, the mass of HFO initially present within
the vadose zone was allowed to change based on changing redox conditions. To
maintain geochemical conditions consistent with this alternative conceptual model, the
mass of HFO available to participate in sorption was allowed to vary depending on the
simulated redox and geochemical conditions; the conceptual model incorporated into the
2010 Revised ICTM Report limited sorption to the mass of HFO initially present in the
vadose zone. The simulated pH, redox, aqueous concentration of uranium, and sorbed
concentration of uranium within the shallow vadose zone for these two conceptual
models are plotted below (see Figure 05/1/6-1 and 05/1/6-2).
The iron redox couple scenario has a slightly higher pH immediately beneath the liners
compared to the oxygen redox couple scenario. This slightly higher pH is related to the
mass of HFO that precipitates, which is slightly higher for the iron redox couple scenario:
the precipitation of HFO consumes acidity which results in a slightly higher pH. Both
scenarios show complete dissolution of calcite at about the same depth. The iron redox
couple scenario has a lower pe immediately beneath the liners compared to the oxygen
redox couple scenario. The slightly lower pe is based on the value input at the upper
boundary. The lower pe at greater depths for the iron redox couple scenario is lower
than the oxygen redox couple scenario; because DO that is originally present within the
vadose zone is oxidized and consumes electrons.
The concentration of uranium in the shallow vadose zone is significantly lower for the
iron redox couple scenario compared to the oxygen redox couple scenario. Decreased
transport of uranium is attributed to increased sorption onto HFO for the iron redox
couple scenario because more HFO is present within the vadose zone. The increased
mass of HFO is consistent with the conceptualization that the mass of HFO available to
participate in sorption was allowed to vary depending on the simulated redox and
geochemical conditions.

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 55 of 70
Figures for Response 6 (May 31, 2012):

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 56 of 70

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 57 of 70

7. Refer to Appendix M, p. M-11, Paragraph 2: Please provide justification for using a chloride
diffusion coefficient (1.75 cm2/day) for seawater in the model. Perform and report results of
sensitivity analyses that assess the dependence of results on variations in the values of the
diffusion coefficient used in analyses.
Response 7 (May 31, 2012):
The diffusion coefficient (Dw) for the solutes modeled was assumed to equal 1.75
centimeters squared per day (cm2/d). The value for Dw was based on measurements for
chloride at infinite dilution in water at 25°C, and do not represent measurements in
seawater. The value of Dw for some of the more common aqueous species are
summarized below (see Table 05/1/7-1). The maximum value of Dw corresponds to the
value for chloride while the minimum value of Dw corresponds to the value for uranium
(uranyl ion).
Diffusive transport was assumed to be species-independent as a necessary
simplification because the model is not set-up to simulate multicomponent diffusive
transport. Diffusive transport through the variably saturated porous media is simulated
as an effective diffusion coefficient (De) that is calculated as a function of space and time
as the product of Dw and tortuosity factor in the liquid phase (τw). While the value for Dw
is constant, the model simulated values for De vary through space and time as a function
of the simulated spatial variations in water content and resultant tortuosity.
In unsaturated porous media the value for De tends to increase as the degree of
saturation increases because of the increased connectivity of the porous media
(decreased tortuosity). For example, using the value for Dw referenced above (1.75
cm2/d), if the VWCs were equal to 8 percent and 15 percent the corresponding values for
De would equal approximately 0.14 and 0.62 cm2/d, respectively. The calculated values
for De are within the range of values presented in the literature:
• Hu and Wang (2003) report values for De equal to approximately 0.13 and 0.22
cm2/d for a sand soil at VWCs approximately equal to 8 percent and 15 percent.
• Badv and Mahmoundi (2009) report values for De from 0.30 to 0.60 cm2/d for a
silty sand soil at VWCs approximately equal to 19 percent and 37 percent.
In part, the value for De represents an intrinsic property of the porous media such that a
direct comparison with values reported in the literature should be completed with caution
because diffusive transport will vary depending on the retention curve which influences
the tortuosity factor.
Theoretically, a higher value for De will result in increased diffusion which will tend to
increase mass spreading resulting in faster transport times for the migrating diffusive
front during transport through the vadose zone. These postulated effects are confirmed
by the following plot which illustrates the chloride concentration profile within the bedrock
vadose zone using the maximum and minimum values for Dw reported in Table 05/1/7-1.
The concentration profiles using two different values for Dw are plotted after 240 years
for the base case flow and transport scenario described in the 2010 Revised ICTM
Report (see Figure 05/1/7-1).
The results presented above indicate that the higher value of Dw equal to 1.75 cm2/d
based on measurements for chloride as implemented in the 2010 Revised ICTM Report
result in increased diffusive transport (more conservative assumption) at the leading
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 58 of 70
edge of the plume. Therefore, a sensitivity analysis for the value assigned to the
diffusion coefficient at infinite dilution in open water (Dw) is not warranted.
References for Response 7 (May 31, 2012):
Badv, L. and M. Mahmoudi, 2009. The study of moisture and contaminant migration
through soil, Paper presented at the 33rd International Association of Hydraulic
Engineering & Research (IAHR) Congress: Water Engineering for a Sustainable
Environment.
Hu, Q. and J.S.Y. Wang, 2003. Aqueous-phase diffusion in unsaturated geologic media:
a review, Critical Reviews in Environmental Science and Technology, 33(3):275-297.
Li, Y-H. and S. Gregory, 1974. Diffusion of Ions in Sea Water and Deep-Sea Sediments,
Geochemica et Cosmochemica Acta 38:703-714.

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 59 of 70
Table and Figure for Response 7 (May 31, 2012):

Table 05/1/7-1. Diffusion coefficients of various ions at infinite dilution in water at 25°C.
Cations cm2/d Anions cm2/d
Potassium K+ 1.69 Bicarbonate HCO3- 1.02
Sodium Na+ 1.15 Chloride Cl- 1.75
Calcium Ca2+ 0.69 Carbonate CO32- 0.83
Magnesium Mg2+ 0.61 Sulfate SO42- 0.92
Cadmium Cd2+ 0.62 Arsenic H2AsO5- 0.78
Cobalt Co2+ 0.60 - - -
Copper Cu2+ 0.63 - - -
Iron Fe2+ 0.62 - - -
Nickel Ni2+ 0.58 - - -
Uranium UO22+ 0.37 - - -
Zinc Zn2+ 0.62 - - -
Aluminum Al3+ 0.48 - - -
Note: Diffusion coefficients reproduced after Li and Gregory (1974).

8. Refer to Appendix M, p. M-11, Paragraph 4: Please justify the assumption to establish the initial
soil water pressure heads within the bedrock vadose zone as that those resulting from percolation
at a rate equal to 1% of the average annual precipitation. Compare the resulting pressure head
distribution in the vadose zone with the water content distribution that could be expected to result
from potential leakage from the tailings cells area, especially the area of Cells 2 and 3 (see also
“INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10 CFR40 APPENDIX A,
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 60 of 70
CRITERION 1; INT 04/1: EVALUATION OF POTENTIAL FLOW THROUGH TAILINGS CELL
LINERS”).
Response 8 (May 31, 2012):
Initial Pressure Heads
The initial pressure heads within the bedrock vadose zone were assigned to represent
conditions prior to construction of the tailings cells. Initial values were based on an
assumed pre-tailing-cell recharge rate equal to 1 percent of the average annual amount
of precipitation between 1932 and 1988. The assumption that 1 percent of the annual
precipitation will occur as recharge generally is in agreement with recharge studies
completed at other semiarid sites. For example, a synopsis of recharge measurements
suggests that average recharge rates in arid and semiarid environments can vary
between 0.1 to 5 percent of the long-term average annual amount of precipitation
(Scanlon et al., 2006). The review paper authored by Scanlon et al. (2006) reported the
following recharge rates for relatively comparable environments in the southwestern
United States as compared to Blanding:
• 3 percent of precipitation for the Middle Rio Grande Basin in central New Mexico.
This regional recharge rate (8.5 mm/yr) was estimated using a steady state,
inverse groundwater model and carbon-14 age dating. Recharge occurs
primarily in the surrounding mountain block and mountain front settings through
ephemeral streams, with little or no recharge in inter-stream basin floor settings.
• 0.4 percent of precipitation on the Pajarito Plateau of northern New Mexico
beneath a pinon juniper cover within the shallow subsurface soils. This localized
recharge rate (0.2 mm/yr) was estimated using cumulative chloride water plots.
The major source of recharge at this location is derived from snowmelt and
spring rains.
• 0.02 to 2 percent of precipitation for the High Plains in the Texas panhandle.
Vadose zone modeling studies found that focused higher recharge rates (11
mm/yr) occur beneath ephemeral lakes and playas while little to no recharge
(<0.1 mm/yr) in inter-playa settings.
• 1 to 6 percent of precipitation for the Black Mesa basin in northeastern Arizona.
This regional recharge rate (5 to 20 mm/yr) was estimated using carbon-14 age
dating combined with a coupled carbon-14 flow and transport model. This
recharge rate was independently verified using chloride mass balance. The
aquifer is recharged seasonally from precipitation in the highlands principally
during the winter and spring, with less recharge at lower elevations.
Based on the measurements reported above a recharge rate assuming 1 percent of the
average annual amount of precipitation (approximately 3 mm/yr) is justified for use in
determining the initial conditions of the bedrock vadose zone. This assumption is further
supported by numerical modeling completed by Scanlon et al. (2003) for conditions of a
single deep borehole located in the High Plains of Texas. Their model simulations
suggest that water potential and chloride profiles at depth are out of equilibrium with
current climatic forcing, and reflect Pleistocene climate conditions. Current water fluxes
in the shallow subsurface, which developed over thousands of years, currently are
upward. Their simulations further suggest that the drying front was initiated during the
Pleistocene/Holocene climate shift, and that chloride concentrations at depth are low,
which suggests that water fluxes during the Pleistocene were quite high on the order of
1.3 mm/yr while Holocene recharge rates are negligible at <0.1 mm/yr.
Comparison of Water Content and Pressure Head Profiles
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 61 of 70
The volumetric water content throughout the vadose zone beneath Cells 2 and 3 during
the operational, dewatering, and post-closure steady-state timeframes is plotted in
Figure 4-4 of the 2010 Revised ICTM Report. A comparison between water contents
and pressure head is plotted below (see Figure 05/1/8-1). The synoptic timeframes
plotted correspond to maximum head conditions (13 and 23 years), the end of
dewatering (33 years), and post-closure steady-state (100 and 240 years). The change
in subsurface hydraulic conditions during the assumed maximum tailings seepage
duration from 0 through 23 years is evident by the increasing VWC and decreasing
pressure head within the shallow vadose zone (less than 5 meters depth). In response
to dewatering and reduced potential seepages rates, the VWC and pressure head
conditions within the vadose zone become drier and eventually reach steady state
conditions. The nearly identical profiles after 100 and 240 years indicate that steady
state flow conditions have developed in the bedrock vadose zone.
References for Response 8 (May 31, 2012):
Scanlon, B.R., K. Keese, R.C. Reedy, J. Simunek, and B.J. Andraski, 2003. Variations in
flow and transport in thick desert vadose zones in response to paleoclimatic
forcing (0–90 kyr): Field measurements, modeling, and uncertainties, Water
Resources Research, 39(7), 1179, doi:10.1029/2002WR001604.
Scanlon, B.R., K.E. Keese, A.L. Flint, L.E. Flint, C.B. Gaye, W.M. Edmunds, and I.
Simmers, 2006. Global synthesis of groundwater recharge in semiarid and arid
regions, Hydrological Processes, 20: 3335–3370.

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 62 of 70
Figure for Response 8 (May 31, 2012):

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 63 of 70

9. Refer to Appendix M, Figures M-3 and M-4: Please state and justify the value(s) of the effective
uranium retardation factor that would be consistent with the HP1 model output for the bedrock
vadose zone. Please see (summarized in Appendix M of the Revised ICTM Report, Figures M-3
and M-4,) which shows concentration profiles for sulfate and uranium, clearly indicating that
uranium is transported more slowly than sulfate. Please quantify the rate of uranium transport
relative to species, such as sulfates, that are not retarded.
Response 9 (May 31, 2012):
In the 2010 Revised ICTM Report sulfate is not transported as a conservative species.
Sulfate is allowed to sorb onto HFO and precipitate as gypsum. Sulfate concentrations
within the bedrock vadose zone are predominately dictated by mineral precipitation
reactions, as noted in Figure M-3. Even though the precipitation of gypsum results in a
faster transport rate compared to uranium, sulfate nonetheless, is retarded. Therefore,
the most applicable conservative species for comparison of uranium sorption/retardation
would be with chloride or fluoride.
The distribution coefficient (Kd) in units of milliliters per gram (mL/g) and the
corresponding retardation factor (Rf) in dimensionless units, as calculated from the
output of the reactive transport model after 240 years, are summarized below for the
conditions represented in Figure M-4 (see Table 05/1/9-1). Data below 2.3 meters
depth are not tabulated since the concentration of uranium in pore water was less than
0.005 mg/L.
The calculated Kd values ranged between approximately 0.002 to 1.5 mL/g
corresponding to calculated Rf values between approximately 1 and 41. The calculated
Rf values suggest that uranium is moving at about the same rate as groundwater for
acidic pH conditions and 41 times slower than groundwater for near-neutral pH
conditions.
A comparison between calculated Kd values listed above, and those presented in the
literature need to be made with some caution. This is because of the site-specific nature
of the values calculated using the results of the reactive transport model may not directly
correlate with other studies. Krupka et al. (1999) summarized measured Kd values for
uranium for the following pH values:
• pH 3 a minimum Kd of < 1 mL/g.
• pH 5 a minimum Kd of 25 mL/g.
• pH 7 a minimum Kd of 63 mL/g.
Significant research has investigated Kd values at the Naturita site in Colorado for a
variety of test conditions. For example, the geometric mean Kd value for measurements
in Curtis et al. (2006) is approximately 1.6 mL/g covering a range between 0.55 and 12.5
mL/g for sediments suspended in wells at the Naturita field site at near-neutral pH. For
comparison, the average Kd value for the reactive transport model at near-neutral pH is
approximately 0.43 mL/g. The minimum Kd values presented by Krupka et al. (1999) are
orders of magnitude higher than the Kd values calculated using the output of the reactive
transport model, while the Kd values presented by Curtis et al. (2006) are approximately
a factor of four higher. The model calculated Kd values for the iron redox couple
scenario (defined in the response to comment six of this interrogatory) varies between
approximately 0.5 and 18 mL/g within the upper 0.75 meters with the majority of the
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 64 of 70
calculated values hovering around 10 mL/g. Therefore, the sorption and attenuation of
uranium is reasonably represented by simulations using the reactive transport model.
References for Response 9 (May 31, 2012):
Curtis, G.P., J.A. Davis, and D.L. Naftz (2006) Simulation of reactive transport of
uranium(VI) in groundwater with variable chemical conditions, Water Resources
Research, 42, W04404, doi:10.1029/2005WR003979.
Krupka, K.M., D.I. Kaplan, G. Whelan, R.J. Serne, and S.V. Mattigod (1999)
Understanding variation in partition coefficient, Kd, values Volume II: Review of
Geochemistry and Available Kd Values for Cadmium, Cesium, Chromium, Lead,
Plutonium, Radon, Strontium, Thorium, Tritium (3H), and Uranium, U.S.
Environmental Protection Agency 402-R-99-004B.

September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 65 of 70
Table for Response 9 (May 31, 2012):

Table 05/1/9-1. Calculated distribution coefficients and retardation factors
as a function of depth after 240 years.
Depth
(m)
pH
(s.u.)
Aqueous U
Concentration
(mg/L)
Volumetric
Water
Content
(-)
Kd
(mL/g)
Rf
(-)
0 – 0.15 3.2 21 0.073 0.0015 1.04 
0.20 – 0.45 4.7 12 0.073 0.68 20 
0.50 – 0.70 4.7 3.1 0.074 0.87 25 
0.75 – 0.95 4.9 0.38 0.074 1.5 41 
1.0 – 2.15 7.3 0.073 0.075 0.14 4.6 
2.2 – 2.3 7.3 0.015 0.076 0.72 20 
Note: Values summarized above were calculated as the arithmetic average of simulated
values within the noted depth interval for conditions represented by Figure M-4.
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 66 of 70
10. Refer to Appendix M, Figures M-3 and M-4, pp. M-25 and M-26: Please clarify why the initial
concentrations for sulfate or uranium are not shown at a depth of 0 feet on Figures M-3 and M-4
and/or revise the figures as necessary.
Response 10 (May 31, 2012):
The results plotted in Figures M-3 and M-4 were meant to illustrate concentration profiles
after various times, and not the source term concentration. The concentration at the
upper boundary (0 centimeters depth) is a function of the applied boundary condition
(mass flux rate equal to time variable flux multiplied by the concentration); therefore,
plotting the source term concentration would be contradictory to the boundary condition
implemented at the upper surface and likely would be a source of confusion. The initial
source term concentrations are summarized in Table M-1, which the reader can easily
transfer to Figures M-3 and M-4 if necessary.

11. Refer to Appendix M, Figure M-4. Please explain why dissolved uranium concentration at the
top of the vadose zone appears to decrease from 50 years to 100 years but then to increase from
100 years to 240 years.
Response 11 (May 31, 2012):
The decrease in uranium concentration (approximately 0.7 mg/L) within the upper 15
centimeters from 50 years to 100 years is attributed to subtle differences in sorbed
concentrations. The increase in uranium concentration from 100 years to 240 years
(approximately 7 mg/L) is attributed to decreased sorption from increased surface site
loading resulting from increased total mass released into the vadose zone.

BASIS FOR INTERROGATORY:
The initial soil water pressure heads in the vadose zone beneath existing Cells 2 & 3 may be higher than
the initial soil water pressure heads derived from an assumption of 1% of the average annual
precipitation (1% of 13.3 in/yr or 3.4 mm/yr). Leakage from Cells 2 & 3 may have already occurred. In
Appendix L the estimated leakage rate through the liners in Cell 2 and 3 during the operational phase is
calculated as 8.3 mm/yr (Base Case scenario) with estimated lower and upper bound values of 3.5 and 18
mm/yr; these values area likely underpredicted as the methodology used in that calculation does not
appear to be conservative (see also “INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10
CFR40 APPENDIX A, CRITERION 1; INT 04/1: EVALUATION OF POTENTIAL FLOW THROUGH
TAILINGS CELL LINERS”). Please discuss the potential effects on vadose zone flow and transport if the
initial soil water pressure heads in the vadose zone were derived from the flux rate through the Cells 2
and 3 liners as calculated using the alternative flux rate calculation approach (Giroud et al. 1997)
recommended in INTERROGATORY WHITE MESA REV’D ICTM; R313-24-4; 10 CFR40 APPENDIX A,
CRITERION 1; INT 04/1: EVALUATION OF POTENTIAL FLOW THROUGH TAILINGS CELL
LINERS” .
The presence and distribution of fractures and/or joints and/or uncemented zones in the bedrock
materials beneath the tailing management cells area is not discussed in the Revised ICTM Report, and no
discussion is provided regarding the potential effects of such fractures and/or joints and/or uncemented
zones on subsurface contaminant flow and transport. The possible presence and distribution of such
fractures and/or joints in the bedrock materials should be discussed in the Revised ICTM Report, along
with an evaluation of the potential effects of such fractures and/or joints and/or uncemented zones on
subsurface contaminant flow and transport. For example, the 1978 Environmental Report (e.g., see
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 67 of 70
Dames & Moore 1978., p. 2-106) indicates the following: “…jointing is common in the exposed Dakota-
Burro Canyon sandstones along the mesa’s rim…more often than not, the primary joints are parallel to
the cliff faces and the secondary joints are almost perpendicular to the primary joints… two sets of joint
attitudes exist [in these sandstone units] ..to the west side of the project site…These sets range from N 10-
180 E and N 60-85 E0 and nearly parallel to the cliff faces”.

In addition, information provided by UMETCO (UMETCO 1993, p, 2-3) indicates that “during an
investigation of the White Mesa site, a number of fracture attitudes were measured (in the Dakota and
Burro Canyon sandstone units) along the rims of Corral and Cottonwood Canyons [in the general site
area], ..(with) analysis of the data indicating the presence of two joint sets… [and] distances between the
joints in each set varies from 5 to 20 feet, …the primary joints strike from north-south to N 200 E with a
vector mean of N 110 E and the secondary fractures have a strike ranging between N 400 W to N 600 W
with a vector mean of N 470 W… All joint sets observed were near vertical to vertical.”
The boring log for Borehole No. 19 (see Dames & Moore 1978, Plate A-9; International Uranium
Corporation [IUC] 2000, Figure 1.5.3-1), installed near the Cell 4B footprint, indicates horizontal
fracturing may be present at one or more depth zones (e.g., 45 ft, and 53-58 ft below ground surface)
within the Dakota Sandstone unit underlying and/or adjacent to the area of proposed Cell 4B. That
boring log also indicates the occurrence of some orange iron staining and considerable limonite staining
along bedding fractures (which suggest zones of localized movement of groundwater) as well as some
uncemented zones of rock within the Dakota Sandstone materials.
An injection test conducted within the Dakota unit in Boring 19 penetrating the Dakota and Burro
Canyon units yielded permeability values that differed by more two orders of magnitude, depending on
whether the tested interval spanned a zone (37.5 – 52.5 ft below ground surface) containing
“considerable near horizontal fracturing and some orange staining” (permeability of 9.12 x 10-4 cm/sec)
or had no reported fracturing (permeability 6.77 x 10-6 cm/sec).
The issue of the potential presence of fractures and/or joints and/or uncemented zones in the bedrock
materials beneath and in the vicinity of the Cell 4B tailing management cells area and the potential
effects of such features on vadose zone flow and transport was previously considered and evaluated in
responses provided by Denison Mines (USA) Corp (DUSA), with attached letters from Hydro Geo Chem,
Inc., to First Round Interrogatories submitted to DUSA by the Division on the Cell 4 B Design Report
(DUSA 2010a ) and Second Round Interrogatories submitted to DUSA by the Division on the License
Amendment Request and Environmental Report for Cell 4B (DUSA 2010b; 2010c). A similar
discussion/evaluation should be included in the ICTM Report to assesses the potential significance of
such features on the transport modeling assumptions and approach.
The maximum ANP value of 27 g CaCO3/kg rock listed for MW-24 (Table C-14, p. C-15 of Appendix C)
does not appear to be correct based on a review of the ACZ analytical data sheets provided in Appendix
A. The correct maximum value appears to be 25 g CaCO3/kg rock. It also appears that the arithmetic and
geometric means for ANP in the TW4-22 boring may also be incorrect. Data used to randomly check the
arithmetic and geometric means for boring TW4-22 were obtained from the ACZ analytical data sheets.
Statistical results for the entire population presented on Table C-4 are used as input to the Reactive
Transport Model described in Appendix M. If these results change, please modify the reaction transport
model as needed.
The discussion presented in Section 2.2.4 of the ICTM Report refers to a number of hydrogeologic and
background groundwater quality reports but does not summarize information on any pertinent
geochemical conditions that are relevant to the development of input parameters for use in the transport
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 68 of 70
modeling. The potential relevance of the perched zone geochemical data, if any, to the development of
geochemical modeling input assumptions made in Appendix M should be discussed and discussion should
be provided as to whether the vadose zone input and results are consistent with existing perched water
geochemical conditions at the site.

The fixed dissolved oxygen concentration (2 mg/L) arbitrarily chosen and used to define the (O2/H2O)
redox couple may be an overestimate of the likely redox potential conditions in the tailing slimes
drainage. With modeling conditions fixed in this way, all calculations in Eh-pH space will be confined to
a line just below the upper stability limit for water. Bass Becking et al. (1960) and Garrels and Christ
(1965) showed the inadequacy of this approach for all but a few rare surface geologic situations. Redox
equilibrium is typically not established in most waters because of the presence of living organisms, the
dependence of most redox reactions on biological catalysis, and the slow kinetics of many oxidation and
reduction reactions. The redox potential should therefore correspond to the potential range of the
predominant redox reaction under given conditions.
The tailing slimes drainage chemistry data presented in Table K-1 indicate that the tailing slimes contain
ammonia and dissolved iron which suggests that the redox conditions in the tailing slimes drainage may
be less than those defined by the chosen fixed dissolved oxygen concentration for the oxygen redox
couple. It is important to have a reasonable redox estimate for both the tailing slimes and the vadose zone
because it the redox potential value controls solubility and/or precipitation of some constituents/solids
such as Fe2+/HFO during reactive transport. For example, if more reducing tailing slimes drainage
percolates through the vadose zone, the assumed redox condition in the vadose zone may be less and
result in the dissolution of HFO which serves as a sorption site for uranium and other constituents. Thus,
less sorption would occur and uranium might be transported to the underlying perched zone. Because the
reactive transport model will likely be sensitive to redox and the uncertainty in redox, the redox value
should be included as a parameter in the sensitivity analyses.
A summary of the existing dissolved oxygen (DO) and/or oxidation-reduction potential (ORP) data for the
vadose or perched zones, as well as area groundwater seeps should be presented so that these data can
be compared to the dissolved oxygen concentration (2 mg/L) assumed for the vadose zone (pages M-10
thru M-12 in Appendix M) to determine if the assumed vadose zone oxygen content is consistent with
those found in the perched zone. Relevant information might be found in the INTERA hydrogeology
reports, background reports, etc. cited on pages 2-12 and 2-13 of the Revised ICTM Report.
The diffusion coefficient would be expected to affect transport of solutes through groundwater, including
the amount of time required for peak solute concentrations to arrive at a downgradient location.
Chloride diffusion coefficients reported in the literature (e.g., Barone et al 1990; Barone et al 1992;
Kincaid et al 1995; Rowe and Badv 1996; Badv and Faridfard 2005) suggest that a smaller chloride
diffusion coefficient may be more reasonable than the one selected because the salinity of water in the
vadose zone will be less than seawater. Because the reactive transport model will likely be sensitive to the
diffusion coefficient, the diffusion coefficient should be included as a parameter in the sensitivity
analyses.
The HP1 reactive transport model (HYDRUS-1D coupled with PHREEQC) does not use the traditional
concept of a distribution coefficient (Kd) from which a retardation factor can be calculated; rather it uses
a surface complexation modeling approach that is functionally similar to the methodology developed by
the U.S. Geological Survey for the U.S. Nuclear Regulatory Commission, as presented in NUREG/CR-
6820 (Davis and Curtis 2003). According to information presented in Appendix M, in this modeling
approach, uranium adsorption is allowed to compete with other metals, which would decrease the total
amount of uranium that could adsorb. The transport model shows the concentration front of uranium
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 69 of 70
proceeding more slowly than the concentration of species, such as sulfate, that are not retarded (see
Appendix M, Figures M-3 and M-4). Therefore, while the conceptual basis of the transport model is
different from a simple Kd and retardation factor approach, the predicted uranium transport could still
be described by an “effective” retardation factor, e.g., relative to the “effective” retardation factors for
other modeled species. An estimate should be made of effective retardation factor for uranium, that
would be consistent with the output of the reactive transport model, and the resulting predicted
“effective” attenuation behavior for uranium should be further discussed and compared to observations
or model predictions for other case studies/ similar sites, if available, and further discussion and
evaluation provided in the context of demonstrating the suitability/adequacy of the modeling approach
used.

The model results depicted on Figures M-3 and M-4 do not appear to show the initial concentrations for
sulfate (62,847 mg/L) or uranium (24.3 mg/L) introduced at depth 0 feet. The initial concentrations
should be shown.
REFERENCES:
Badv, K. and M. R. Faridfard. 2005. Laboratory determination of water retention and diffusion
coefficient in unsaturated sand. Water, Air, and Soil Pollution, v. 161, no. 1-4, pp. 25-38
Barone, F. S.; R. K. Rowe, R. M. Quigley. 1990. Laboratory determination of chloride diffusion
coefficient in an intact shale. Canadian Geotechnical Journal, v. 27, no. 2, (April), pp. 177-184.
Barone, F. S.; R. K. Rowe, R. M. Quigley. 1992. Estimation of chloride diffusion coefficient and tortuosity
factor for mudstone. Journal of Geological Engineering, v. 118, no. 7, (July), pp. 1031-1046.
Bass-Becking, L.G.M., I.R. Kaplan, and D. Moore. 1960. Limits of the natural environment in terms of pH
and oxidation-reduction potentials. Journal of Geology, v. 68, pp. 243 - 284.
Dames & Moore 1978. Environmental Report - White Mesa Uranium Project, San Juan County, Utah for
Energy Fuels Nuclear, Inc. January 30.
Davis, J.A. , and G.P. Curtis, 2003. Application of Surface Complexation Modeling to
Describe Uranium(VI) Adsorption and Retardation at the Uranium Mill Tailings Site
at Naturita, Colorado, Report NUREG CR-6820, U. S. Nuclear Regulatory
Commission, Rockville, MD., pp. 223.
Denison Mines (USA) Corp. 2010a. Round 1 – Interrogatory Response for the Cell 4B Design Report,
White Mesa Mill Site, Blanding, Utah. January 2010.
Denison Mines (USA) Corp. 2010b. Second Round of Interrogatories from Review of License Amendment
and Environmental Report for Cell 4B. DUSA Letter with attachment dated February 8, 2010.
Denison Mines (USA) Corp. 2010c. Second Round of Interrogatories from Review of License Amendment
and Environmental Report for Cell 4B. DUSA Letter with attachment dated February 12, 2010.
Denison Mines (USA) Corp. 2010d. Revised Infiltration and Contaminant Transport Modeling Report,
White Mesa Mill Site, Blanding, Utah (Revision 2), March 2010.
Garrels, R.M and C. L. Christ. 1965. Solutions, Minerals, and Equilibria. Freeman, Cooper & Company,
San Francisco, California. 450 pp.
September 10, 2012
Interrogatory 05/1: R313-24-4: Contaminant Transport Modeling Page 70 of 70
Giroud, J.P., King, T.D., Sanglerat, T.R., Hadj-Hamou, T., and Khire, M.V. 1997. Rate of Liquid
Migration Through Defects in a Geomembrane Placed on a Semi-Permeable Medium”, Geosynthetics
International, Vol. 4, Nos. 3-4, pp. 349-372.
International Uranium (USA) Corporation (IUC). 2000. Reclamation Plan – White Mesa Mill, Blanding,
Utah. Source Material Reference No. SUA-1358. Docket No. 40-8681. Rev. 3, July.
Kincaid, C. T., J. W. Shade, G. A. Whyatt, M. G. Piepho, K. Rhoads, J. A. Voogd, J. H. Westsik, Jr., M. D.
Freshley, K. A. Blanchard, B. G. Lauzon. 1995. Performance Assessment of Grouted Double-Shell Tank
Waste Disposal at Hanford, WHC-SD-WM-EE-004, Rev. 1, Westinghouse Hanford Company. Richland,
WA.
Rowe, R. K. and K. Badv. 1996. Chloride migration through clayey silt underlain by fine sand or silt.
Journal of Geotechnical Engineering, v. 122, no. 1 (January), pp. 60-68.
UMETCO Minerals Corporation 1993. Peel Environmental Services. Groundwater Study, White Mesa
Mill. January.