DSHW-2010-052472 - 0901a068801e2d22 (6)

June 14, 2010

David P. Gosen, P.E.
Director, Environmental Services
ATK Launch Systems – Promontory
P.O. Box 707
Brigham City, UT 84302-0707

RE: Waste Characterization and Air Dispersion Modeling Protocol
ATK Launch Systems - Promontory Facility
UTD009081357

Dear Mr. Gosen:

The Division of Solid and Hazardous Waste has completed its review of ATK’s Waste Characterization and Air Dispersion Modeling Protocol documents. The waste characterization information
and air dispersion model will be used to conduct the Human Health and Ecological Risk Assessments for open burning and open detonation. Our comments and questions on the documents are
enclosed with this letter.

As you are aware, the Division is committed to completing permitting work for the Subpart X units by September 30, 2010. Due to the short amount of time left to accomplish this goal,
we request that ATK submit its responses to the attached comments by June 28, 2010. If you have any questions, please contact Jeff Vandel at [phone].

Sincerely,

ORIGINAL DOCUMENT SIGNED BY SCOTT T. ANDERSON ON 6/14/2010

Scott T. Anderson, Director

STA\JV\tjm

Enclosure

c: George Gooch, ATK Launch Systems
Blair Palmer, ATK Launch Systems
Lloyd Berentzen, Deputy Director, Bear River Health Department

ATK LAUNCH SYSTEMS, INC., PROMONTORY FACILITY
RCRA SUBPART X EVALUATION AND PERMIT MODIFICATION
PROMONTORY, UTAH

EVALUATION OF ATK’S WASTE CHARACTERIZATION AND AIR DISPERSION MODELING WORK PLAN

GENERAL COMMENTS

1. ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1, Section 2.0, ATK Promontory Facility Process Description

Section 2.3.1, M-136 Treatment Activities, and Section 2.3.2, M-225 Treatment Unit, indicate trays at M-136 and M-225, respectively, may be lined with soil but most trays do not contain
soil. No discussion of the potential impact of the soil on the open burning process is offered in either section. In addition, the criteria used to determine if a burn tray should
be lined is not provided. Please revise the ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1 dated November 4, 2009 (the Air Dispersion Modeling
Work Plan) to discuss any impacts of lining the burn trays with soil on process emissions. Please include the criteria used to determine if soil lining is necessary. Please also indicate
if soil lined burn trays will be addressed in the air modeling for M-136 and M-225. If not, please propose an approach for addressing the use of soil lined trays in the uncertainty
analysis of the risk assessment or demonstrate that impacts from the soil lining are insignificant.

2. ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1, Section 4.0, Air Quality Modeling Methodology

Section 4.0 does not clearly identify the types of air quality impacts to be modeled. Table 4-3, Summary of Deposition Modeling Parameters, indicates air concentration will be modeled
for gas phase runs and deposition rate will be modeled for particulate phase runs. However, information provided in Section 3.0 indicates gas phase, particle phase, and particle-bound
phase constituents will be emitted from all three types of wastes considered in the air modeling analysis. It is expected that, at a minimum, gas phase air concentrations, particle-phase
air concentrations, particle-bound phase air concentrations, particle phase gravitational deposition, and particle-bound phase gravitation deposition would be modeled using OBODM. Please
revise Section 4.0 to clearly indicate the types of impacts to be modeled by OBODM.

3. ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1, Section 4.0, Air Quality Modeling Methodology

Section 4.4, OB/OD Treatment Scenarios, provides general information on the treatment scenarios to be addressed in the air modeling analysis. Stakeholders are referred to Sections 4.4.1
and 4.4.2 for a summary of source parameters and other assumptions to be used in modeling the M-136 and M-225 treatment units. However, Sections 4.4.1 and 4.4.2, and referenced Tables
4-1 and 4-2, do not provide the dimensions of the merged sources, Source 1 at M-136 and Source 1 at M-225, to be addressed in the air modeling analysis. In addition, the second bulleted
item in Section 4.4.1.2, Other Modeling Assumptions for M-136, indicates modeled sources within M-136 will be assigned to source groups so individual contributions from different sources
and different types of wastes can be delineated. There is no mention of source groups in the discussion related to M-225; thus, it is unclear if source groups will be used in modeling
those sources. Please revise the Air Dispersion Modeling Work Plan to provide the dimensions (i.e., length, width, and depth) that will be used to characterize merged Source 1 at M-136
and merged Source 1 at M-255 in the air modeling analysis. Also, please indicate if source groups will be used in modeling the sources at M-225. Please provide a table that lists the
modeled sources to be included in each source group used in the air modeling analysis.

4. ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1, Section 4.0, Air Quality Modeling Methodology

Section 4.4 of the Air Dispersion Modeling Work Plan indicates the air modeling results will be subjected to post processing to account for the treatment quantities proposed in Tables
2-1 and 2-2. Stakeholders are referred to Section 4.9, Post-Processing Activities, for discussion of the proposed post processing steps. Step 1 is described as a review of all air
modeling results to determine the maximum one-hour and annual air dispersion and deposition factors. No information on manipulation or modification of the air modeling results, as implied
by the discussion in Section 4.4, is described. Please revise Section 4.9 to indicate how (or if) the air modeling results will be modified (i.e., post processing) to reflect the treatment
quantities proposed in Tables 2-1 and 2-2.

5. ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1, Section 4.0, Air Quality Modeling Methodology

Based on the information provided in Section 4.4.1.2, Other Modeling Assumptions for M-136, and the first paragraph of Section 4.4.2, M-25 Treatment Unit Scenarios, it appears 1.5 meters
will be used as the source diameter for detonation processes. Table 4-1, M-136 Source Parameters, and Table 4-2, M-225 Source Parameters, also list ground level as the effective release
height for OD sources. Section 3.4, Source Data, of Volume I of the OBODM User’s Guide (Bjorklund et al., 1998a) states OBODM can be directed to compute the effective release height
and diameter of the initial cloud from a detonation. Alternately, the user can enter the effective release height and initial diameter parameters. Section 3.4 continues that, for an
instantaneous source (i.e., the type of source assumed for a detonation), OBODM requires the diameter of the resulting fireball. Based on the information provided in the Air Dispersion
Modeling Work Plan, it is not clear how OBODM will be configured in modeling OD sources. Please revise the Air Dispersion Modeling Work Plan to indicate if OBODM will calculate the
effective release height for OD sources or if the user will supply the required effective release height and initial diameter. Regardless of which option is indicated, please ensure
that the text and/or Tables 4-1 and 4-2 reflect the information that will be supplied as input to OBODM when modeling the OD sources at ATK.

6. ATK Launch Systems Waste Characterization and Air Dispersion Modeling Protocol, Revision 1, Section 4.0, Air Quality Modeling Methodology

Details regarding how the air modeling was performed will be available in the air modeling files. These files must be provided to the Division in order to convey a clear understanding
of how the air modeling analysis was conducted and to allow for the re-creation of the analysis, if necessary. Please revise the Air Dispersion Modeling Work Plan to indicate electronic

copies of all OBODM input and output files will be submitted to the Division. Further, please state that copies of the model-ready meteorological data file, as well as any other files
(e.g., hourly source strength files) used in generating the modeled results will be provided.

SPECIFIC COMMENTS

7. Section 2.3.1, M-136 Treatment Activities, Page 2-4

Based on the description provided in Section 2.3.1, it is not clear that Table 2.1, M-136 Risk Assessment Treatment Unit Wastes Treated, Treatment Limits, and Model Quantity, reflects
the appropriate established daily quantity limits. Table 2.1 indicates one of the options for achieving the 50,000 pounds per day established daily quantity limit at Source 2 is to
burn 1.1 pure propellant and contaminated material at one burn station. However, the discussion in Section 2.3.1 indicates 1.1 neat propellant and contaminated materials would be limited
to 20,000 pounds per day at Burn Station 13, the only station associated with Source 2. Please review the discussion in the text and the information in Table 2.1 and revise them as
appropriate to clarify the type, nature and quantity of potential burn materials allowed to ensure consistency within the text and table.

8. Section 2.3.2, M-225 Treatment Unit, Page 2-6

Based on the description provided in Section 2.3.2, it is not clear that Table 2.2, M-225 Risk Assessment Treatment Unit Wastes Treated, Treatment Limits, and Model Quantity, reflects
the correct description of the established daily quantity limits. Table 2.2 indicates one of the options for achieving the 2,000 pounds per week established daily quantity limit at
Source 1 is to burn 1,000 pounds per tray of 1.1 or 1.3 pure propellant and contaminated material. However, the discussion in Section 2.3.2 states open burning treatment quantities
are limited to 500 pounds per tray of propellants and 1,000 pounds per tray of propellant contaminated material. Please review the discussion in the text and the information in Table
2.2 and revise them for clarity and consistency. Further, please change the header on the column entitled Established Daily Quantity Limits to Established Weekly Quantity Limits.

9. Section 3.2.1, Class 1.3 Waste Emission Factors, Page 3-5

In order to characterize the class 1.3 wastes that are treated at the facility by open burning or detonation, ATK made the effort to conduct emissions testing on three different compositions
of waste materials (PW100, PW85-15 and PW65-35). Apparently, ATK is planning to use an average value, calculated from the three sets of emission factors, to evaluate the risk associated
with open burning/detonating these waste materials. The option of comparing the risk associated with burning wastes with a higher percentage of trash with wastes containing a lower
percentage of trash will be lost using this approach. Please explain why ATK has decided to use an average value as opposed to calculating the risk associated with each composition.

10. Section 3.2.3, Category E Emission Factors, Page 3-6

The second paragraph of Section 3.2.3 states that constituent data from the Munitions Items Disposition Action system (MIDAS) database was reviewed to determine that the M816, 81-mm
Infrared (IR) Illumination Cartridge was a suitable surrogate for the Category E wastes treated at ATK. However, the components of the M816, 81-mm IR Illumination Cartridge are not
provided. Please revise Section 3.2.3 to include a comparison of the constituents in the Category E wastes and those in the M816, 81-mm IR Illumination Cartridge to fully demonstrate
that the Cartridge is an appropriate surrogate.

11. Section 4.1, Air Quality Dispersion Model Selection, Page 4-3

The next to last bulleted item in Section 4.1 indicates the modeling domain includes areas of complex terrain. As stated in the discussion, OBODM contains a screening-level algorithm
for estimating air quality impacts in complex terrain. However, this algorithm is restricted to predicting air concentrations. As stated in Section 2.3 of Volume I, User’s Instructions,
and Section 2.7, Complex Terrain Screening Procedures, of Volume II, Technical Description, of the Open Burn/Open Detonation Dispersion Model (OBODM) User’s Guide (Bjorklund et al.,
1998a, 1998b), the complex terrain option cannot be used when calculating concentration with gravitational deposition occurring or gravitational deposition for particulates with appreciable
settling velocities. The Air Dispersion Modeling Work Plan does not address how gravitational settling will be addressed or calculated in complex terrain. Please revise the Air Dispersion
Modeling Work Plan to propose an approach for calculating particulate deposition (via gravitational settling) in areas of complex terrain.

12. Section 4.2, Land Use Analysis, Page 4-4

The last paragraph of Section 4.2 indicates a count analysis was conducted as part of determining the land use classification for the modeling domain. It appears that the Administration
and Manufacturing area falls within the 3 km radius that was included in the analysis. What land use classification was assigned to this area? The details of the analysis are not provided
in the Air Dispersion Modeling Work Plan. For completeness and transparency, please revise the Air Dispersion Modeling Work Plan to include the count analysis or provide a statement
that the count analysis will be included in the air dispersion modeling report.

13. Section 4.4, OB/OD Treatment Scenarios, Page 4-5

The discussion at the bottom of page 4-5 indicates a heat content of 1,471 calories per gram (cal/g) was determined for 1.3 class materials using the NASA-Lewis Thermochemical Model.
Further, the discussion indicates the facility will be conducting tests to determine the heat content for 1.1 and Category E class materials. Some general information related to the
determination of the heat content for 1.3 class materials is provided; however, this information is not sufficient to demonstrate to stakeholders how the value was derived. Please revise
Section 4.4 to describe how a value of 1,471 cal/g was determined for 1.3 class materials. Please summarize all input information considered and illustrate how the NASA-Lewis Thermochemical
Model was used in obtaining the reported value. Please provide electronic copies of the input and output information for any model runs. Also, please describe how the heat contents
for 1.1 and Category E class materials will be provided to the Division and provide details regarding how the values will be determined.

14. Section 4.4.1.1, M-136 Source Parameters, Page 4-6

It is unclear what treatment quantities will be used in the air dispersion model analysis for the different waste classifications (e.g. 1.1 or 1.3) that are treated at M-136 and M-225.
Table 4-1 (referenced in this section of the text) shows the modeled treatment quantity per event for Source 1 is 106,500 pounds. Will this quantity be used for all waste classifications?
Table 2-1 shows different daily quantity limits for the waste classifications shown. How will these limits be represented in the air dispersion model analysis? This question also
applies to the other sources at the two burn grounds. In addition, what is the daily quantity limit for the Category E waste category?

15. Section 4.4.1.2, Other Modeling Assumptions for M-136, Page 4-7

The third bulleted item on page 4-7 indicates each OB source will be based on an average pan size of six feet by 17 feet. It is not clear how this average size pan was determined.
Please revise the Air Dispersion Modeling Work Plan to describe how the average pan size of six feet by 17 feet was determined.

16. Section 4.4.1.2, Other Modeling Assumptions for M-136, Page 4-7

The fourth bulleted item on page 4-7 states a release height of 2.0 meters will be assumed for Burn Station 14 at M-136. Table 4-1, M-136 Source Parameters, indicates an effective
release height of one meter for the OB pans at burn station 14 (i.e., source 3 at M-136). Please revise the text and table to consistently reflect the value to be used in the air modeling
analysis.

17. Section 4.4.2, M-225 Treatment Unit Sources, Page 4-8

Based on the information provided in the second paragraph of Section 4.4.2 it appears the OB and OD sources at M-225 “…will be modeled as a single emission source…” in a single model
run. No information regarding how this will be accomplished is provided. Please revise the Air Dispersion Modeling Work Plan to describe how the OB and OD operations at M-225 will
be modeled as a single emission source. As part of the description, please confirm the OBODM can model two operations with different source characteristics at the same time and at the
same location. In addition, please provide the source inputs that will be used in OBODM to model both processes in the Air Dispersion Modeling Work Plan.

18. Section 4.4.2.1, M-225 Source Parameters, Page 4-8

In regard to the modeled treatment quantity and daily quantity limits for the M-225 Burn Grounds, Table 4-2 (referenced in this section of the text) shows a modeled treatment quantity
of 2,000 pounds per event for Source 1. Draft permit condition VI.C.5.a. states that the maximum treatment quantity shall not exceed 4,500 pounds at M-225. Please clarify what the
maximum quantities are for the M-225 Burn Grounds. Does ATK intend to use the maximum quantity for the air dispersion modeling analysis?

19. Section 4.5.1, Particle Phase Dry Deposition, Pages 4-10 and 4-11

ATK has proposed the same particle size distribution for modeling both OB and OD operations. No information supporting or justifying this approach is provided. Please revise the Air
Dispersion Modeling Work Plan to include information supporting the use of identical particle size distributions in modeling OB and OD operations. If necessary, please include information
establishing the need to use one PSD for both operations. In addition, please indicate that the use of a single PSD in the air modeling analysis will be addressed as a source of uncertainty
in the air modeling and risk assessment reports.

20. Section 4.5.1, Particle Phase Dry Deposition, Pages 4-10 and 4-11

As discussed in Section 3 of US EPA’s Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (USEPA, 2005), particle deposition should be addressed in two phases,
the particle phase and the particle-bound phase. Constituents present in the particle phase are modeled using a mass weighting of the assumed particle size distribution while particle
bound phase constituents are modeled using a surface area weighting of the assumed distribution. Section 3.2.3 of the HHRAP outlines a technique for calculating the surface area weighting
factors from the mass fractions for each particle size category. The discussion in Section 4.5.1 of the Air Dispersion Modeling Work Plan does not address the difference between modeling
gravitational deposition for particle phase and particle-bound phase constituents. Please revise Section 4.5.1 to describe how the difference between gravitation deposition for particle
phase and particle-bound phase constituents will be addressed in the air modeling analysis.

21. Section 4.5.[2], Gas Phase Dry Deposition, Page 4-11

Section 4.5.[2] indicates ATK will estimate gas phase dry deposition using the equation presented on page 4-11 (gas phase dry deposition = annual gas concentration x deposition velocity
x conversion factor). Section 2.5.2, Dry Deposition, of Volume II of the OBODM User’s Guide (Bjorklund et al., 1998b) explains that OBODM models gravitational deposition but not dry
deposition (defined as the product of total ground-level dosage and an empirical or theoretical dry deposition velocity). Equation 2-49 is provided as the preferred means of estimating
dry deposition from OBODM air modeling results:

Dry (x,y) = vd x D(x,y,0)

Where:

Dry (x,y) = dry deposition at downwind distance x and crosswind distance y;

vd = dry deposition velocity; and
D(x,y,0) = total ground level dosage (predicted by OBODM) at (x,y).

Please revise Section 4.5.[2] of the Air Dispersion Modeling Work Plan to demonstrate the methodology used to estimate gas phase dry deposition at ATK is equivalent or more conservative
than the approach outlined in Section 2.5.2 of Volume II of the OBODM User’s Guide.

22. Section 4.6, Receptor Networks, Page 4-11

This section discusses the applicable receptor grids that would be used and references the use of the United States Geological Service (USGS) Digital Elevation Map (DEM) to support
the receptor grid development. However, it is not clear from what source (e.g., vendor, web site) this information will be obtained. Please revise the Air Modeling Work Plan to address
this issue.

23. Section 4.6.1, Discrete Receptor Grid, Page 4-12

It is understood that OSHA exposure concentration values will be used to evaluate ATK worker exposure at each treatment unit. However, it appears that the Administration and Manufacturing
Area, and other similar areas at the facility, should be included as discrete receptors since the workers in these areas are not directly involved with the activities at the treatment
units. Please provide a discussion on the assessment of the risk posed to these workers associated with open burning at the facility.

24. Section 4.7.1, Surface Data, Page 4-16

In addition to the information already provided, please indicate if surface data from Hill Air Force Base (AFB) has been used as a surrogate (i.e., “substitute”) for missing surface
data in previous air dispersion modeling analyses at ATK. In addition, the Air Dispersion Modeling Work Plan should indicate the use of surface data from Hill AFB will be addressed
as a source of uncertainty in the air modeling and risk assessment reports.

25. Section 4.7.2, Upper Air Observations (Mixing Height Data), Page 4-17

The first full paragraph on Page 4-17 concludes with “ATK will use a combination of upper air data from Salt Lake City and surface temperature observations from Hill AFB to produce
twice-daily mixing heights.” The second full paragraph on Page 4-17 indicates cloud cover and ceiling height observations from Hill AFB will be used “…in the calculation of stability
class…” Based on the information furnished in Section 4.7, Meteorological Data, it was expected that surface temperature observations, cloud cover, and ceiling height observations from
the M-245 meteorological monitoring station would be used for these purposes. No information supporting the use of Hill AFB data in these analyses has been provided. Please review
the information presented in Section 4.7.2 to ensure it accurately reflects the information sources ATK will use in preparing the model-ready meteorological data file for the air modeling
analysis. If ATK intends to use Hill AFB surface temperature observations, cloud cover data, and ceiling height observations in lieu of data from M-245, Section 4.7.2 should be revised
to explain why the Hill AFB data must be used and demonstrate that the Hill AFB observations are the best values to use in determining mixing heights stability class for the ATK air
modeling analysis.

26. Section 4.8, Comparison to Air Quality Standards and Exposure Criteria, Page 4-18

The contribution from background and the contribution from ATK OB and OD operations should be clearly identified in the comparisons to air quality standards described in Section 4.8.
Please revise Section 4.8 to indicate the contributions from background and OB and OD operations will be clearly delineated in the comparative analyses.

REFERENCES

Bjorklund, J. R., J. F. Bowers, G.C. Dodd. and J.M. White, 1998a. Open Burn/Open Detonation Model (OBODM) User’s Guide, Volume I, User’s Instructions, DPG Document No. DPG-TR-96-008a,
February.
Bjorklund, J. R., J. F. Bowers, G.C. Dodd. and J.M. White, 1998b. Open Burn/Open Detonation Model (OBODM) User’s Guide, Volume II, Technical Description, DPG Document No. DPG-TR-96-008b,
April.
USEPA, 2005. Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities, Office of Solid Waste and Emergency Response, EPA530-R-05-006, September.

1

195 North 1950 West • Salt Lake City, UT
Mailing Address: P.O. Box 144880 • Salt Lake City, UT 84114-4880
Telephone [phone] • Fax [phone] • T.D.D. [phone]
www.deq.utah.gov
TN201000481 Printed on 100% recycled paper

State of Utah

GARY HERBERT
Governor

GREG BELL
Lieutenant Governor

Department of Environmental Quality

Amanda Smith
Executive Director

DIVISION OF SOLID AND HAZARDOUS WASTE
Scott T. Anderson
Director