DSHW-2012-002052 - 0901a068802b5343

HAND DELIVERED
m 0 8 2012
7 March 2012 UTAH DiViSION OF
8200-FY12-070 SOLID & HAZARDOUS WASTE
Mr. Scott T. Anderson, Executive Secretary
State of Utah Department of Environmental Quality
Division of Solid and Hazardous Waste
195 N.1950 W.
P.O, Box 144880
Salt Lake City, Utah 84114-4880
Re: ATK Launch Systems-Promontory EPA ID number UTD009081357
Response to Comments on Addendum to Waste Characterization and Air
Dispersion Modeling Protocol For Use in the Human Health and Ecological Risk
Assessments
Dear Mr. Anderson:
ATK has completed responses to the comments provided by the Utah Division of Solid
and Hazardous Waste (UDSHW) regarding the Addendum to the Waste Characterization
and Air Dispersion Modeling Protocol. This information is necessary to conduct the
Human Health and Ecological Risk Assessments for our OB/OD operations.
Please contact me if you have any questions concerning this report. My telephone
number is [phone] or you can contact Blak Palmer at [phone].
Sincerely
Paul V. Hancock, Manager
Environmental Remediation
cc: Jeff Vandel
ATK LAUNCH SYSTEMS FINAL RESPONSES TO TECH LAW COMMENTS REGARDING
PRELIMINARY REVIEW OF ADDENDUM ATK LAUNCH SYSTEMS WASTE CHARACTERIZATION
AND AIR DISPERSION MODELING PROTOCOL FOR USE IN THE HUMAN HEALTH AND
ECOLOGICAL RISK ASSESSMENTS
MARCH, 2012
Below is a preliminary evaluation of the Addendum ATK Launch Systems Waste Characterization and Air
Dispersion Modeling Protocol for Use in the Human Health and Ecological Risk Assessments dated
January, 2012 (Addendum Protocol).
1 Section 4.4, OB/OD Treatment Scenarios, Page 4.4-4: Under the discussion entitled Discrete
Receptor Assessment, the text mentions that "new" meteorological data files will be developed.
Based on the text provided, it is not ciear what is meant by a "new" meteorological data file. It is
preferred that all air modeling use the exact meteorological data file used in previous modeling. It
was expected that the hours in the source strength file would correspond to the hours in the
previously used meteorological data file: the first hour in source strength file would correspond to
the first hour in the data file (12 midnight to 1 AM on January 1) and the last would correspond to
the iast hour in the data file (11 PM to Midnight December 31), Once the "worst case" hours for
the sources were identified, the amount of waste treated during those hours would be entered for
the appropriate source for the appropriate hour and zeros entered for the other hours of the year.
If the "new" data file is a file that re-orders the records of the previously used meteorological data
file, the air dispersion modeling report must include a full, detailed explanation of the re-ordering
process, the true identity (i.e., date, time) of each record in the new file, and a demonstration that
the re-ordered file produces the same result as would be obtained with the original meteorological
data file. Revise the Addendum Protocol to clarify the meaning of "new" meteorological data files
and address the additional documentation requirements discussed above,
ATK Response: The meteorological data to be used for the discrete and general grid modeling
will be composed of the same exact meteorological data files used in previous modeling. There
is no new meteorological data being used. We are basically reordering the meteorological data to
address the new methodology in the modeling analysis to focus on worst case events and ATK's
proposed treatment schedule. The new modeling analysis is being conducted using "worst case"
meteorological events that were identified in the short-term (1-hour) modeling results for discrete
and maximum impact general grid receptors using the Addendum treatment quantities. Once the
"worst case" hours for the sources were identified, these hours were placed at the beginning of
each meteorological data file for each receptor in sequential order for each source. For example,
the M136 treatment unit will conduct a total of 269 treatments events per year for all M-136
sources. The first 269 records of the meteorological data file for each receptor contains the worst
case meteorological events in sequential order (Source 1-156 events, Source 2-52 events.
Source 3 - 52 events. Source 4 - 6 events and Source 5 - 3 events). Because M-136 Source 1
will conduct 312 annual events, the worst case annual air dispersion factor for 156 events will be
also used for the impact associated with the second burn per day. The hours in the M-136
source strength file will correspond to the worst case event hours in the previously used
meteorological data file. As a result, no model calculations (all source strengths = 0) will be made
for meteorological data hours beyond record 269 to compute the annual average. The same
procedure is being used for M-225 which will have a total of 30 events per year (Source 1 - 24
events and Source 2 - 6 events). The hours in the M-225 source strength file will correspond to
the worst case event hours in the previously used meteorological data file. As a result, no model
calculations (all sources strength = 0) will be made for meteorological data hours beyond record
30 to compute the annual average. Using this procedure is easier than using the corresponding
hours in the previously used meteorological data file because the worst case event occurred at
different times ofthe year in each ofthe five years of onsite met data.
The air dispersion modeling report will include a detailed discussion of the re-ordering process,
the true identity (i.e., date, time) of each record in the new file. Only one demonstration case
Page 1 of 3
showing the re-ordered file produces the same result as would be obtained with the original
meteorological data file will be prepared and included in the report.
Section 4.4, OB/OD Treatment Scenarios, Page 4.4-5 and 4.4-6: Under the discussion
entitled General Grid Receptor Assessment, the text describes how hours that likely produce
worst-case dispersion will be identified and subsequently used in the proposed air modeling
analysis. The discussion is not clear and should be revised to promote clarity and transparency
in the Addendum Protocol. For example, the relationship between the maximum onsite and offsite
general grid receptors and the general grid sectors should be established at the beginning of the
last paragraph on page 4.4-5. It is recommended that a figure be provided to illustrate the
relationship. In addition, the first full paragraph on page 4.4-6 refers to the master list Please
provide an additional descriptor that assists in identifying this list (e.g., the master list of 250 worst
case events).
ATK Response: The Protocol Addendum will be revised to provide a clearer discussion
describing the relationship between the maximum onsite and offsite general grid receptors and
the general grid sectors as presented in Tables 4-1A and 4-2A for M-136 and M-225 respectively.
Figures will be used to show the relationship of general grid receptors and the associated master
list of general receptor grids.
Section 4.4.1.2, Other Modeling Assumptions for M-136, Page 4.4-8: It is not clear why the
release height specified in the previous protocol document for Burn Station 14 (2 meters) was
changed to 1 meter in the Addendum Protocol, Expand this discussion to explain why this
change was made.
ATK Response: The actual text stating that the release height for Station 14 is 2 meters
appeared on Page 2-10 of the October 2011 Preliminary Modeling Report. This statement
reflects a typographical error in the report. Burn Station 14 had a one meter release height in the
previous modeling and it will not be changed for the revised modeling analysis. The Burn Station
release height will remain at 1 meter A review of the original protocol approved by UDSHW
(Page 4-9) clearly states that all OB sources at M-136 have a release height of 1 meter Table 4-
1 of the Original Protocol also explains the release height for each source and indicates a
release height is 1 meter for Burn Station 14.
Section 4.4.1.2, Other Modeling Assumptions for M-136, Page 4.4-9: Based on the
discussion presented in Section 4,4.1.2, it is not ciear if particle-bound constituents will be
modeled. The text indicates that gas phase and particle phase air concentrations and particle
phase gravitational deposition will be modeled. UDSHW has commented on this issue in the past
but it is not clear that ATK Promontory plans to model gravitational settling using a surface area
weighted particle size distribution as outlined in Section 3.2.3 of EPA's HHRAP. Revised this
section to indicate if modeling will be performed for particle-bound constituents. If not, describe
how particle-bound constituents emitted from the OB and OD processes at M-136 will be
addressed in the air modeling analysis.
ATK Response: ATK does not plan to conduct model calculations for particle-bound analysis for
M-136 because OBODM does not contain an algorithm for particle-bound calculations. However,
ATK is proposing to assume that OBODM calculated particulate concentrations are equal to
particle-bounds concentrations.
Section 4.4.2.2, Other Modeling Assumptions for M-225, Page 4.4-11: Based on the
discussion presented in Section 4.4.2,2, it appears that gas phase and particle phase air
concentrations and particle phase gravitational deposition will be modeled. It is not clear if
particle-bound constituents will be modeled. UDSHW has commented on this issue in the past
but it is not clear that ATK Promontory plans to model gravitational settling using a surface area
weighted particle size distribution as outlined in Section 3.2.3 of EPA's HHRAP. Revised this
section to indicate if modeling will be performed for particle-bound constituents. If not, describe
Page 2 of 3
how particle-bound constituents emitted from the OB and OD processes at M-225 will be
addressed in the air modeling analysis,
ATK Response: ATK does not plan to conduct model calculations for particle-bound analysis for
M-225 because OBODM does not contain an algorithm for particle-bound calculations. However,
ATK is proposing to assume that OBODM calculated particulate concentrations are equal to
particle-bounds concentrations.
Section 4.5.4, Flat and Complex Terrain Modeling, Page 4.5.4.-1: ATK Promontory has
provided a new discussion on the need to include flat terrain and complex terrain modeling in the
air modeling analysis. While the added text does a credible job in establishing the reason for
performing both flat and complex terrain modeling, it does not describe how the limitations
introduced by OBODM will be circumvented in the air modeling analysis. The discussion should
be expanded to address the following issues:
• Because gravitational settling cannot be modeled in complex terrain using OBODM, the
Addendum Protocol should describe how the deposition of particle-phase and particle
bound constituents wiil be modeled or calculated at locations in complex terrain;
• The discussion should be accompanied by a statement on the conservatism inherent in
the approach proposed by ATK Promontory for modeling deposition in complex terrain;
and
• Indicate that areas of flat terrain modeling and complex terrain modeling will be depicted
on figures in the air dispersion modeling report.
ATK Response: OBODM does not calculate particulate deposition in complex terrain. ATK is
planning to use OBODM model annual air concentration (pg/m^) results for the "gas" phase at
complex terrain receptors to determine complex terrain deposition rates (e.g., micrograms of
deposition per square meter of soil surface area) in flat. This approach is consistent with
OBODM guidance and HHRAP (USEPA, September, 2005) and is considered to be conservative
because the gas air concentrations calculated are nondepleted (e.g., removal of mass from the
cloud as a function of downwind distance is not accounted for).
A conservative deposition velocity of 0.03 meters/second (m/s) will be used for this assessment,
which is the default value specified in the HHRAP (USEPA, September, 2005) guidance. As a
comparison, the gravitational settling velocity for particles of 2 g/cm^ and 10 pm diameters is
reported as approximately 0.01 m/s (DOE 1984, page 755). Dry deposition in flat and complex
terrain will be calculated as follows:
Complex Terrain Annual Non-gravitational Dry Deposition (pg/m^) = Annual Air Concentration
(pg/m^) X Deposition Velocity (m/s).
Page 3 of 3