DSHW-2013-003436 - 0901a0688037e1d0

Launch Systems Group
P.O. Box 707
Brigham City, UT 84302 Division of
and Hazardous Waste
www.atk.com MAY 1 5 2013
14 March 2013
8200-FY14-022
Scott T. Anderson, Director
Utah Department of Environmental Quality
Division of Solid and Hazardous Waste
P.O. Box 144880
195 North 1950 West
Salt Lake City, Utah 84114-4880
Subject: ATK Launch Systems Promontory Facility, Response to Utah Division of Solid and
Hazardous Waste Additional Comments Regarding the New SWMU Assessment Report,
Promontory EPA ID #UTD009081357
Dear Mr. Anderson:
On March 14, 2013, your office submitted a letter with additional comments in response to the ATK
New SWMU Assessment Report for SWMU #681. ATK's responses to these comments are
included with this letter.
If you have questions regarding these comments, please contact Paul Hancock at [phone].
Sincerely,
George Gooch, Manager
Environmental Services
ATK Launch Systems Promontory Facility
Response to Division of Solid and Hazardous Waste Additional Comments on New SWMli
Notification and Assessment Report for SWMU #681
Original DSHW Comment:
Please provide the complete output, including the transient time steps, of the Hydrus Model that
was run to investigate the potential for the release from M-705 to reach groundwater.
A TK Response:
The transient time steps of the Hydrus Model run were a minimum of 26 minutes and a maximum
of 5 days. The complete output files from the Hydrus Model run used in this simulation are
included with this submittal as Attachment 1.
Additional DSHW Comment:
The Division is concerned that the answer to the question of whether groundwater may be
impacted by the release from building M-705 is based solely on the Hydrus Model that has been
run without any specific subsurface data. We believe that a more rigorous assessment of the
potential for an impact to groundwater from the release is warranted. The installation of a new
monitoring well down-gradient of M-705 could provide the most definitive answer to this
question.
ATK Response:
Prior to the construction of building M-705, engineering exploratory soil borings were collected to
determine if the soil met the requirements to support the building load. The depths of these
borings, on average, were over 25 feet. Through the boring logs and soil sieve analysis, the soils
were classified as a clay. The sections of the pertinent soil report are included in Attachment 1.
With the conclusive evidence that the first 25 feet below M-705 are clay, this further reinforces
the well log used to support the Hydrus Model. Additionally, there are several other well logs in
the same general area that also show similar subsurface data.
More details of the operating conditions at building M-705 were obtained for the trench where the
crack was found. This trench is associated with the reject water from the sand filter process and
has a much lower volume than the regular process water; this flow is estimated at around 2 gpm.
The process runs about 4 hours per day, 5 days a week. A conservative estimate is that a tenth of a
gallon of water per minute could seep through the crack. The trench is inspected once per month
so it is assumed that the crack was present for one month. When it was first found, the depth of
the crack was determined using a narrow probe and found to be about 2 to 3 inches. The building
construction drawings show that the reinforced concrete pad under the trench is 10 inches thick so
the crack may not have extended to the ground.
In order to provide a more rigorous and conservative assessment for the soil-to-groundwater
pathway, an additional Hydrus Model run was conducted using the known 25 feet of clay and then
assuming that the remaining formation to groundwater was sand using both a 1 and 2 gpm release
Page 2
(24 hours a day) for 30 days and also including precipitation . Results of this modeling run show
that the released would not reach groundwater. These results are included in Attachment 1.
There is an existing perchlorate plume in the surrounding wells, therefore a new well in the area
of M-705 will likely show the existing contamination and it would be difficult to differentiate this
from any possible new contamination.
Based on the previous and current HYDRUS modeling results and with support ofthe pre-
construction soil testing at M-705 along with simmilar lithology in existing area wells, the
potential for the perchlorate to reach groundwater is unlikely.
Original DSHW Comment:
Please provide a blown-up version of the particle trace map that was submitted so the path from
the M-705 building to wells J-7 and J-8 may be seen in more detail.
A TK Response:
A larger particle trace map is included with this submittal as Attachment 2.
Additional DSHW Comment:
Thankyou for providing the blown-up version of the particle trace map. Based on the map, it
appears that the particle trace from M-705 is to the west beneath Blue Creek and then pretty
much straight south until near well H-8 where it turns to the southwest. It appears that if
contamination from the release at M-705 were to reach groundwater, a contaminant plume may
not be seen at wells J-7 or J-8.
In addition, at approximately well H-l, the particle trace continues due south to the mudflats as
opposed to turning toward the east and Shotgun and Pipe Springs. Is this consistent with the
solute transport model?
ATK Response
Hydraulic conductivity was discretized into zones in the model. Since the actual parameter
distributions are far more complex than the model has the power to replicate, and probably far
more complex than can be measured, zones were used to group similar values together. Because
of the zoning of hydraulic properties, when viewed closely, groundwater obviously appears to
flow counter intuitively to the direction that would be expected. Locally, groundwater may indeed
flow north to south at M-705 rather than west then south. With three wells representing
approximately 100 acres in the area of M-705, conductivities were averaged and grouped. Each
grid in the model has a uniform length and width of 200 feet. This results in flow that looks to be
travelling at right angles as it follows hydraulic zones near M-705.
Typically, a groundwater model's ability to accurately predict groundwater flow in real-world
situations is poor. At best groundwater models, despite their high degree of precision, are
qualitative predictors of future behavior. A major cause of the lack of accuracy is the severe
Page 3
discrepancy between the scale of measurement necessary to understand aquifer parameters for
accurate modeling and the scale of measurement generally made under the restraints of time and
budgets.
Contaminants that are detected in the springs may come primarily from fractures/faults rather than
direct groundwater flow as indicated in the comment. Shotgun and Pipe Springs occur because
groundwater flow in fractured bedrock is forced to the surface when it encounters lower-
permeability valley sediments and/or groundwater discharges to the surface from deeper high-
pressure zones at the edges of the block-faulted mountains. A major task of the modeling effort at
the Promontory facility was to determine the pathway that contaminated groundwater follows to
reach the springs. There is evidence that adjacent springs are highly influenced by different
primary sources. For example, Pipe Spring and Shotgun Spring are only 400 feet apart but they
discharge at different elevations and have substantially different dissolved solids and contaminant
concentrations.
Historic and current potentiometric surface maps show that groundwater flow is to the south-
southeast from well H-l. The particle trace shows more of a south-only flow path controlled by
defined hydraulic zones in the model. The calibrated steady-state potentiometric surface map from
the transport model shows that water in the model travels toward Shotgun and Pipe Springs.
Original DSHW Comment:
What is the status of the old wells TCC8 and TCC8a? Are well casings still in place? Based on
their apparent location, it would be very useful if potentiometric and/or analytical data could be
collected from one of these wells.
ATK Response:
As noted above, A TK has information on wells TCC8 and TCC8a. A well log indicates well TCC8
was drilled with a 4 3A - inch casing to a depth of458feet. However, there is no indication it was
completed...
Additional DSHW Comment:
ATK has indicated that well TCC8a was sampled last fall and that no contaminants were detected.
Is the potentiometric surface data for the well consistent with the groundwater flow model? Is
there any data available that may be used to calculate a hydraulic conductivity for the well? It
was also indicated in the meeting held at our office on March 6, 2013 that ATK may recalibrate
the groundwater flow model. Please ensure that applicable data from well TCC8a (i.e., head data
and concentration) is used when the model is recalibrated.
ATK Response:
The calibrated steady-state potentiometric surface from the groundwater flow model indicates a
water surface elevation of 4320 feet near TCC8A. The actual water surface elevation from the
recent measurement is 4311 feet, a difference of 9 feet. This is fairly consistent with the flow
Page 4
model and can be tightened up when the model is recalibrated. Bear in mind that water levels
fluctuate by several feet over a few years due to drought or rainfall and thus, fluctuating recharge.
Until we can determine the depth of casing and perforated or screened interval of the well, we will
not have the necessary hydraulic data to conduct a slug test. ATK will try to determine the depth
and perforated interval sometime this summer and then run a slug test to calculate a hydraulic
conductivity. ATK will again sample TCC8a this spring to verify that the well does not contain
contamination. All of this information will be included in the groundwater model recalibration
which is currently planned for 2014.
Attachment 1
Additional HYDRUS Modeling
Including the M-705 Preconstruction Soil Study
Evaluation of Perchlorate Infiltration at Building M-705
ATK Launch Systems, Promontory Utah
The finite-element model HYDRUS v4.14 was used to evaluate the potential for
perchlorate to percolate from the surface to the groundwater table at building M-705. For
this model, it was assumed that perchlorate contaminated water leaked through a floor
drain in the building at rates of 2 gpm and 1 gpm for a period of 31 days. The flow was
stopped after 31 days and the model was allowed to run. Scenarios evaluated included
average precipitation for a 1 year period as well as no precipitation to represent
conditions that may exist at M-705 since discharge occurred under the building which is
not exposed to rainfall. Additionally, the upper 25 feet of soil at or near the building has
been identified as sandy clay (Chen & Associates). Soils below 25 feet were modeled
as sand to provide a conservative estimate of infiltration.
This model, which is recommended by the U.S. Environmental Protection Agency1 for
use in these situations, is based on standard soil physics concepts and is capable of
modeling both flow and contaminant transport in the unsaturated zone. Version 4.14 of
this public domain model was used, as downloaded from the Internet at http://www.pc-
proqress.com/en/Default.aspx?h1d-downloads .
According to recorded depth to water from nearby wells, groundwater occurs at a depth
of about 140 feet. To simulate the unsaturated zone completely, a 140-feet thick
unsaturated zone was simulated in Hydrus. The initial moisture profile was taken to be in
equilibrium with the initial ground water level at 140 feet. The soil was modeled using a
van Genuchten-Mualem single porosity model. As a conservative measure, hysteresis
in the moisture response curve was ignored. A bottom boundary condition of deep
drainage was assumed as the vertical drainage flux depended on the position of
groundwater level.
In summary, assumed conditions were as follows:
(1) 140 ft unsaturated soil profile at M-705;
(2) van Genuchten-Mualem single porosity model without hysteresis;
(3) Soil type is clay to 25 feet then sand to depth;
(4) Atmospheric boundary condition allowing for overland flow applied for upper
boundary condition;
(5) Deep drainage at bottom boundary condition;
(6) Initial moisture profile to be in equilibrium with the groundwater level.
(7) 1-year Rainfall infiltration model
(8) Typical annual rainfall of 14 inches
Daily precipitation and reference evapotranspiration data from 1962 to 2010 at THIOKOL
PROPULSION F S station were download from Utah weather center
http://climate.usurf.usu.edu/products/download.php. The chosen hydraulic station
' U.S. Environmental Protection Agency. 1996. Soil Screening Guidance: Technical Background Document.
EPA/5407R-95/128. Office of Solid Waste and Emergency Response. Washington, D.C.
THIOKOL PROPULSION F S station is located closest to our study area with the same
weather variance. Closed to the 14-inch average annual precipitation, 1964 was
modeled as a typical hydraulic year that consisted of 117 rain days and a total rainfall
depth of 14.13 inches during the whole year. Ignoring the variance of daily precipitation,
the simulated rainfall series was simplified into 12 rainfall events, each of them occurring
only at the beginning of each month and lasting the same period of raining time
happened in the reality in the whole month. The precipitation rate remained evenly with
the average monthly precipitation rate during the raining time.
Monthly average potential evaporation was download from Western Regional Climate
Center http://www.wrcc.dri.edU/htmlfiles/westevap.final.html#UTAH . In a typical year
(1964), the evaporation amount after each rainfall event was calculated. Then the
evaporation amounts in each month are added to represent monthly evaporations for the
typical study year.
A one-year rainfall infiltration model was simulated with the above calculated monthly
precipitation and monthly evaporation data.
The result of the modeling effort is presented in the attached figures. Each figure for 1-
year simulation contains the predicted depth of the wetting or concentration front for
each month.
With one year of rainfall infiltration, the maximum wetting front in M-705 will only reach to
75 feet. Compared to the groundwater depth of 140 feet at M-705, the model result
shows that the contaminant at the ground surface could not reach the groundwater
within a one year period.
Figures
HYDRUS GRAPHS
Figure 1. Water Content Change with Depth Over Time at M-705
2 gpm Discharge to Drain/Average Annual Precipitation
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
-•—initial time
After 1 month
After 2 month
After 3 month
-*— After 4 month
-•—After 5 month
-f After 6 month
After 7 month
After 8 month
After 9 month
After 10 month
After 11 month
After 12 month
Water content (theta)
Figure 2. Water Content Change with Depth Over Time at M-705
2 gpm Discharge to Drain/No Precipitation
0.05 0.1 0.15 0.2 0.25 0.3
—i—
0.35 0.4 0.45
-•—initial time
-•—After 1 month
After 2 month
-*— After 3 month
-*— After 4 month
-•—After 5 month
—I— After 6 month
After 7 month
After 8 month
After 9 month
After 10 month
After 11 month
After 12 month
Water content (theta)
Figure 3. Water Content Change with Depth Over Time at M-705
1 gpm Discharge to Drain/Average Annual Precipitation
0.05 0.1 0.15 0.3 0.2 0.25 0.35 0.4
••— initial time
After 1 month
After 2 month
After 3 month
«— After 4 month
•—After 5 month
H— After 6 month
— After 7 month
After 8 month
After 9 month
After 10 month
After 11 month
After 12 month
0.45
Water content (theta)
Figure 4. Water Content Change with Depth Over Time at M-705
1 gpm Discharge to Drain/No Annual Precipitation
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
-•—initial time
-•—After 1 month
After 2 month
-*— After 3 month
-*— After 4 month
-•—After 5 month
H—After 6 month
After 7 month
After 8 month
After 9 month
After 10 month
After 11 month
After 12 month
Water content (theta)
OPEN FIELD
M aw tr* DM - E x Ii ling floor l».»l
Ell*. 100' •••u*i«d
• as • a <
> 5
BO AO ING
CHAIN *,Gt
-- I
Eltv. 96.0'
TRAILER
-7
V . OPEN FIELD
NORTH
0 (0* 120'
APPROXIMATE SCALE
BU1L DING M-70S
MORTON THIOKOL
Chen & Associate* LOCATION Of EXPLORATORY BORINGS Ftgur* I
chen and associates, inc.
Moisture Content • 13.1 percent
Qty Unit Weight * 95,3 pel
S*«npieo< Slightly Sandy Clay
B-l at 2' From
Cxnpr >ss .011 HE 9 1 ecting
too 0 t 1 0 to
APPLIED PRESSURE — ksf
Moisture Content * 10.4 p«rc«nf
Ory Unit Wetgnt « 87.7 pet
sempteot: Slightly Sandy Clay
From B-2 at It'
Additional c jn prsssion u ion *et in 5
Note: Differ •er : cal scale
01 1.0 to
APPLIED PRESSURE — ksi
100
Job No. 535787 SWELL-CONSOLIOATION TEST RESULTS Ftg..
chen and associates, inc.
OT CO CU
ll
I «
o e*.
Moisture Conient • 19.2 P«rctnt
Ory Un« We<gM • 100.0 PCf
SMipwet Slightly Sandy Clay
From: B-2 at 19'
Iddi io ia < oan ession i pan «»et ting
Oi io 10
APPLIED PRESSURE — ksf
too
Moisture Content * 10.2 percent
Ory Unit WetQ.nl « 88.8
sempieof: Slightly Sandy Clay
From B-3 at 2'
AdditLon il II: sion unon in »
c 5
u 6
too 0 1
Job Mo. 535787
1.0 10
APPLIED PRESSURE — ksf
SWELL-CONSOLIDATION TEST RESULTS Fig..
CA-l-79
e o
01
s
u
CL e o
°. 10
M 12
Note:
0.1
Dif
chen and associates, inc.
Moisture Content t 23. S
Oy UVW WtirjM • 80.2
Sam** aay
percent
pel
•Vom: B-3 at 9*
A Idit: .on ll tssion ujon tett
s 6
100 0.1 1.0 10
APPLIED PRESSURE - ksf
er tnt i e
idle .on il
i i cil seals
1
Moisture Content i 8.9
Ory Umt wetgnt * 86.3
**»••• <* Clay
From: B-4 at 4'
percent
pet
Iff 3 sion u(on vett in
1.0 10
APPLIED PRESSURE — ksf
too
SWELL-CONSOLIDATION TEST RESULTS Fig.
CHEN AND ASSOCIATES
Job Mo. 535787
TABLE I
SUMMARY OF LABORATORY TEST RESULTS
SAWfiC LOCATION
Of MH
(recti
NATUftAL
MOlStURC
content
IV.I
NAtuAAL
D(MS>tT
CRAOATION
J»NO
•CACCNT KISSING
HO. too
iitvt
ATTCHOfnS LIMITS
LIOUIO
LIMIT
(V.I
fLASTIOTV
IhCCl
UKCON>IN(0
COM'MCSSIVC
SMCttCtn
Utter
Soluble
aces
son. on
• tONtXll t»r>£
B-1 13.1
19.6
95.3
90.5
92
91 2690 <0.1
SI. Sandy Clay
SI. Sandy Clay
B-2 7.5 92.3 25 90 1800 SI. Sandy Clay
10.4 87.7 91 SI. Sandy Clay
19 19.2 100.0 98 SI. Sandy Clay
B-3 10.2 88.8 88 SI. Sandy Clay
15.3 84.4 42 95 1220 Clay
23.5 80.2 95 Clay.
B-4 11.6 84.0 22 94 590 SI. Sandy Clay
8.9 86.3 95 Clay
B-5
B-6
8.5
8.3 84.7
89
96
26
24
SI. Sandy Clay-Silt
Clay-Silt
••I
tl... 102
• I
EU.. 10) )
B-3
tit*. 10/
••4
CU*. 104
as
EU*. 101
B-6
EU*. 102
no
10}
100
>
_95
.90
-IS
11
_ 70
136/12 ac. D.I •* tn • «5.)
114/12 "m ' W
M r « - i*.*
\ DO • 90.} •100 • tl
I UC - 2690
\WS <0 I
] 10/12
]j*/!2
]24/12
] 25/12 « • I.S
00 . 92.)
] 11/12 -MO . 90
** V DC . 1*00
\IC - 10.4
I DO . 67.7
1-200 • 91
' ] 10/12
] 11/12
] 11/12 wc • 19.2
DD • 100.0
-200 • 96
'] 16/12
_£) 17/12
|)I/I2 UC - 10.2
DD - u.a
||3/,2.-200 • 64
\ WC • I).I
' DD • 64.4
| -200 • 9)
MIC • 1220
llt/12 WC • 2).)
DD • 60.2
-200 - 9)
]20/12
]24/l2
'/j"l»/12
52/12 WC • 11.6
DO • 64.0
21/12 -200 . 9*
\J1C . 190
1 WC • 6.9
! DO • 66.3
.-200 . 91
14/12
19/12
' ' ] 23/13
»
| ] 26/12
]))/t2 VC . ».)
-200 > 69
S.c Fliur. ) for Lt|tnd «nd Holt*
119/12 k-C . 6.)
DO • 64.7
18/12 -200 - 96
LL . 24
PI - 1
no
101_
ioo_
9J.
90_
70-1
z
Q < >
»I1HII Chen at Associate* LOGS or EXPLORATORY BORINGS FtgiM* 2