REMEDIAL ACTION MODELING ASSESSMENT
WESTERN PROCESSING SITE, KENT, WASHINGTON
                  Draft
                   by

               F. W. Bond
               C.  M.  Smith

  Battelle Project Management Division
  Office of Hazardous Waste Management
       Richland,  Washington  99352
              January 1985
              Prepared for
  U.S. Environmental Protection Agency
                Region X
       Seattle, Washington  98101

-------
                prc
                Planning Research Corporation
      REMEDIAL ACTION MODELING
            ASSESSMENT
         WESTERN PROCESSING

            DRAFT FINAL
            Prepared for

U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Waste Programs  Enforcement
      Washington, D.C.  20460
           Work Assignment  No
           EPA Region
           Site No.
           Date Prepared
           Contract  No.
           PRC No.        ••
           Prepared  By

           Telephone No.
36
X
10T16
February 4. 1985
68-01-7037
15-0360-49
Battelle
(Chris M. Smith)
(509) 943-6001

-------
                                 DISCLAIMER

     The  information  in  this  document  has  been  funded   by   the   U.   S.
Environmental  Protection Agency under Contract No. 68-01-6692, Task  Order  No.
107  to  CHZM'HILL  and  Contract No.  68-01-7037  to  PRC  Engineering,  with
Battelle Project Management Division, Office of Hazardous Waste Management as
the principal subcontractor.   Mention of  trade  names or commercial products
does not constitute endorsement or recommendation  for  use.
                                     11

-------
                                   ABSTRACT

     The Western  Processing Hazardous  Waste Site  consists  of  13  acres  near
Kent, Washington,  which operated  as  an  industrial  waste  recycling facility
from about 1960 through  1982.   During  1982,  the U.S. Environmental Protection
Agency  (EPA)  conducted surface water  sampling  around  the  site  and  found  26
priority pollutants,  all  of  which were  subsequently  found  on  site.    As  a
result  of  these findings and subsequent  efforts,  the EPA  initiated  a series
of   studies   to   characterize  the   site   and   evaluate   remedial   action
alternatives.
     One of  the efforts  initiated by  the EPA was  to develop  a groundwater
flow and  contaminant transport model  of  the  site to  be  used  in  evaluating
proposed"remedial  actions.  The development  and calibration  of this model  and
its use in evaluating remedial  actions  is discussed in this report.
     A  conceptual   model  of  the  study  area  was  formulated   based  on   the
available  hydrogeologic  and contaminant  data.   The  conceptual  model  formed
the framework  for  developing the  groundwater flow and contaminant transport
model of the area around the Western Processing  Site.
     Once  calibrated   a  limited   sensitivity  analysis  was   performed  to
determine  the  sensitivity of  the  model  results  to  changes   in  hydraulic
conductivity, porosity,  recharge,  dispersivity,  and retardation  factor.   The
model was  used to  evaluate the effectiveness of remedial actions proposed by
CH2M  HILL  and  the  Potentially  Responsible  Parties,  as   well   as   minor
modifications.  The  remedial  actions  considered include:   1)   no-action; 2)
source removal; 3) source  removal  and  pump and  treat;  4)  capping and pump and
treat;   5)  source  removal  and  a slurry  wall;  and  6)  source removal,  slurry
wall, and  pump and  treat.   Pumping rates of both 100 gpm  and  200  gpm  were
simulated in the model.
     Of the remedial actions simulated,  source removal  combined with pump and
treat   was   found   to   provide   the   greatest    reduction    in   mass   of
                                     iii

-------
trichloroethylene (TCE) from the  groundwater  flow system.  Simulations  showed
that all TCE  was  removed  from the groundwater  flow system in  15  and 5 years
at  the  100 gpm  and  200  gpm  pumping  rates, respectively.   The  slurry wall
around  the  site  prevented  the removal  of TCE  that  was off  site,  but  was
effective in reducing the lateral flow  of  clean  water to wells during pumping
and containing contamination after pumping ceases.
                                     IV

-------
                                   CONTENTS
Disclaimer	     .]]
Abstract	     in.
Figures	      V1
Tables	     ..?
Acknowledgement	•	    xni

     1.   Introduction	       1
     2.   Conclusions 	       3
     3.   Description of the Study Area	       '
               Climate	       7
               Geology	       9
               Hydrology	      10
               Waste Disposal History 	      10
     4.   Model Development  	      17
               Model Selection	      17
               Groundwater Flow Model Development 	      18
                    Structure	      18
                    Boundary Conditions 	      17
                    Hydraulic Conductivity	      21
                    Groundwater Potential 	      24
                    Hydraulic Stress	      24
                    Porosity	      24
               Contaminant Transport Model Development	      24
                    Contaminant Selection 	      25
                    Source Location 	      25
                    Source Area Concentration 	      26
                    Source Duration/Leach Rate	      26
                    Sorpti on/Retardation	      27
     5.   Model Calibration  	      28
               Flow Model Calibration 	      28
               Transport Model Calibration	      32
               Base Case Model Results	      36
               Numerical Dispersion 	      35
     6.   Sensitivity Analysis
               Hydraulic Conductivity (Runs 1-4)	      44
               Porosity (Runs 5 and 6)	!  !  !  !      44-
               Recharge (Runs 7 and 8)	!  !  !  !      45
               Retardation (Runs 9 and 10)	'.!!!!      45
               Dispersivity  (Runs 11 and 12)	!".!!!!        46
               Combination Runs (Runs 13-16)	'.'.'....      46
               Sensitivity Analysis Summary	.'.'.....      47
     7.   Assessment of Remedial Action Alternatives. ...'...,.      50

-------
               CH2M HILL Proposed Remedial Actions	     50
                    No Action (Extension of Base Case)	     51
                    Source Removal	     51
                    Source Removal combined with Pump
                    and Treat (100 gpm)	     55
                    Source Removal Combined with Pump
                    and Treat (200 gpm)	     58
                    Cap Combined with Pump and Treat (100 gpm). ...     58
                    Cap Combined with Pump and Treat (200 gpm). ...     60
               PRPs Proposed Remedial Action	     63
                    No-Action (Extension of the Base Case)	     63
                    Source Removal and Slurry Wall	     65
                    Source Removal,  Slurry Wall, Pump and
                    Treat (100 gpm)	     67
                    Source Removal,  Slurry Wall, Pump and
                    Treat (200 gpm)	     69
                    Source Removal,  Pump and Treat (100  gpm),
                    No Slurry Wall	     69
                    Source Removal,  Pump and Treat (200  gpm),
                    No Slurry Wall	     73
     8    Summary	    75
               Base Case	    75
               Remedial Actions 	    76

References	    79
Appendices

     A.   Calculation of Leach Duration and Amount from
          Source Area	    81
     B.   Recharge Calculations 	    89
     C.   Stream and Leakance Boundary Conditions 	    91
     D.   Calculation of Recharge Factor	    96

-------
                                  FIGURES
Number
   la     Comparison  of  the Total Mass of TCE Remaining
          in the Groundwater Flow System for the CH2M HILL
                                                                         Page
Ib


2
3
4
5
6
7

8

9

10

11

12

13
14

15

Kemecnai Action oases 	
Comparison of the Total Mass of TCE Remaining
in the Groundwater Flow System for the PRPs
Remedial Action Cases 	 ...
Western Processing Site Before 1984 Excavation 	
Smoothed Kriged Potential Surface for April, 1984 	
Western Processing Well and Reaction Pond Locations ....
Western Processing Model Area 	
Finite Element Grid of the Western Processing Site 	
Location of Surveyed Values for Mill Creek and
Drainage Ditch 	 • 	
Model -Predicted 1983 Top of Layer 1 Potential Surface
for the Base Case Simulation 	
Model -Predicted 1983 Top of Layer 2 Potential Surface
for the Base Case Simulation 	
Model -Predicted 1983 Top of Layer 3 Potential Surface
for the Base Case Simulation 	
Model -Predicted 1983 Top of Layer 4 Potential Surface
for the Base Case Simulation 	
Model-Predicted 1983 Bottom of Layer 4 Potential Surface
for the Base Case Simulation 	
Smoothed Kriged TCE Concentration Contours Fall, 1982
Model -Predicted 1983 Top of Layer 1 TCE Concentration
Contours for the Base Case Simulation ... ...
Model -Predicted 1983 Top of Layer 2 TCE Concentration
Contours for the Base Case Simulation 	
V

5
8
11
14
19
20

22

29

29

30

30

31
33

33

15
                                    vn

-------
16      Model-Predicted 1983 Top of Layer 3 TCE Concentration
        Contours for the Base Case Simulation	      34

17      Model-Predicted 1983 Top of Layer 4 TCE Concentration
        Contours for the Base Case Simulation	      35

18      Model-Predicted 1983 Bottom of Layer 4 TCE Concentration
        Contours for the Base Case Simulation	      35

19      Model-Predicted Top of Layer 1 Potential Surface for the
        Sensitivity Run with One-Tenth the Hydraulic
        Conductivity	     48

20      Model-Predicted Top of Layer 1 Potential Surface for the
        Sensitivity Run with One-Half the Hydraulic Conductivity
        and the Sensitivity Run with Double the Recharge	     48

21      Model-Predicted 1993 Top of Layer 1 TCE Concentration
        Contours for the No-Action Simulation 	     52

22      Model-Predicted 2003 Top of Layer 1 TCE Concentration
        Contours for the No-Action Simulation 	     52

23     "Model-Predicted 2033 Top of Layer 1 TCE Concentration
        Contours for the No-Action Simulation 	     53

24      Model-Predicted Top of Layer 1 Potential Surface for
        CH2M HILL Source Removal and Pump and Treat (100 gpm)
        Remedial Action 	     57

25      Model-Predicted Top of Layer 1 Potential Surface for
        CH2M HILL Source Removal and Pump and Treat (200 gpm)
        Remedial Action 	     59

26      Model-Predicted Top of Layer 1 Potential Surface for
        CH2M HILL Cap and Pump and Treat (100 gpm)
        Remedial Action 	     61

27      Model-Predicted Top of Layer 1 Potential Surface for
        CH2M HILL Cap and Pump and Treat (200 gpm)
        Remedial Action 	     61

28      Model-Predicted Top of Layer 1 Potential Surface for the
        PRPs Slurry Wall and Pump and Treat (100 gpm)
        Remedial Action	     68

29      Model-Predicted Top of Layer 1 Potential Surface for the
        PRPs Slurry Wall and Pump and Treat (200 gpm)
        Remedial Action 	     70

30      Model-Predicted Top of Layer 1 Potential Surface for the
        PRPs Pump and Treat (100 gpm) Remedial Action  with no
        Slurry Uall	     72
                                   vm

-------
 31      Model-Predicted  Top of Layer 1  Potential  Surface for the
         PRPs Pump and Treat (200  gpm) Remedial  Action with no
         Slurry Wall	    74

B-l      Locations of Regional  Values Used to Determine
         Boundary Conditions .	    95

-------
                                   TABLES


Number

   1      Summary of the Model Results for the CH2M HILL
          and PRPs Remedial Action Simulations
   2      Average Monthly Precipitation, Potential
          Evapotranspi ration (PET), and Actual
          Evapotranspi ration (AET) for the Seattle Area .......      9

   3      Comparison of Shallow and Deep Well  Potentials .......     12

   4      Summary of Soil and Groundwater Concentrations of
          TCE and DCE (EPA, 1982) at the Three Suspected
          Source Areas ........................     15

   5     "Parameters Used in the Final Calibrated Groundwater
          Flow and Contaminant Transport Model ............     23

   6      Comparison of Observed to Model-Predicted Maximum TCE
          Concentrations in the Groundwater at the
          Three Source Locations ...................     36

   7      Distribution of TCE in the Model Base Case Simulation ...     37

   8      Summary of Sensitivity Runs ................     39

   9      Summary of Mass of TCE (Ib) Exiting  the System for
          the Sensitivity Runs ....................     40

  10      Summary of Total  Mass of TCE (Ib) in the Groundwater
          System for the Sensitivity Runs ..............     41

  11      TCE Concentration (ppb) at the Sources in 1983 for
          the Sensitivity Runs ....................     42

  12      Percent Change of Mass and Concentration of TCE for
          the Sensitivity Runs Based on 1983 Results .........     43

  13      Model -Predicted Distribution of TCE  for the
          CH2M HILL No-Action Simulation ...............     54

-------
 14      Model-Predicted Distribution of TCE for the CH2M HILL
         Source Removal Action 	     56

 15      Model-Predicted Distribution of TCE for the CH2M HILL
         Source Removal Combined with Pump and Treat (100 gpm)
         Remedial Action 	     58

 16      Model-Predicted Distribution of TCE for the CH2M HILL
         Source Removal Combined with Pump and Treat (200 gpm)
         Remedial Action 	     60

 17      Model-Predicted Distribution of TCE for the CH2M HILL
         Cap Combined with Pump and Treat (100 gpm)
         Remedial Action 	     62

 18      Model-Predicted Distribution of TCE for the CH2M HILL
         Cap Combined with Pump and Treat (200 gpm)
         Remedial Action 	     62

 19      Model-Predicted Distribution of TCE for the PRPs
         No-Action Simulation	     64

 20      Model-Predicted Distribution of TCE for the PRPs
         Source Removal with a Slurry Wall Remedial Action 	     66

 21      Model-Predicted Distribution of TCE for the PRPs Source
         Removal, Slurry Wall, and Pump and Treat (100 gpm)
         Remedial Action 	     69

 22      Model-Predicted Distribution of TCE for the PRPs Source
         Removal, Slurry Wall, and Pump and Treat (200 gpm)
         Remedial Action 	     71

 23      Model-Predicted Distribution of TCE for the PRPs Source
         Removal and Pump and Treat (100 gpm) Remedial  Action
         (No Slurry Wall)	     71

 24      Model-Predicted Distribution of TCE for the PRPs Source
         Removal and Pump and Treat (200 gpm) Remedial  Action
         (No Slurry Wall)	     73

A-l      TCE Soil Concentrations Near Reaction Pond 1	     82

A-2      TCE Soil Concentrations Near Reaction Pond III	     82

A-3      TCE Soil Concentrations Around Well  21	           83

A-4      Summary of Data Used to Make Total  Mass Calculations at
         all Three Source Locations	                35

A-5      Summary of Estimated Total Mass of TCE Present at  the
         Three Suspected Source Areas	     85 •
                                    XI

-------
A-6      Time and Amount of TCE Leaching from the Unsaturated
         Zone at Reaction Pond I	    87
B-l      Runoff Program Results,
                                                                        90
C-l      Stream Boundary Option Data Used to Simulate
         Flux to Mill  Creek	    92

C-2      Stream Boundary Option Data Used to Simulate
         Flux to the Ditch	    93

-------
                                ACKNOWLEDGMENT

     The authors wish  to thank the  EPA Region  X  staff, particularly Project
Manager Fred Wolf  for  supporting this  project.   The efforts  of  EPA Region  X,
CH2M  HILL  (Bellevue  Office,   particularly Neil  Geitner  and  Jeff  Randall),
Hart-Crowser and Associates, and Gaynor Dawson,  Joe English,  and Jim Doesburg
of  Battelle  in  helping  to  formulate  the  initial   conceptual  model   and
providing technical  assistance and review throughout  the  project  are greatly
appreciated.  Finally,  the authors  would  like  to  thank Peggy  Monter, Naney
Painter, Beth  Eddy,  and  Dick  Parkhurst of  the Battelle Office  of  Hazardous
Waste  Management  who   were  instrumental   in  preparing this  report.    Our
successful  completion  of  this  project  to  a great  extent  has  been  possible
because of  the efforts  of the people acknowledged above.
                                     XII 1

-------
                                  SECTION 1
                                INTRODUCTION

     The goal  of this  project  was to  evaluate remedial  action  alternatives
for the Western  Processing  Hazardous  Waste Site in Kent,  Washington, using a
calibrated groundwater  flow and  contaminant  transport  model.   The  specific
tasks of the study included:
•    review available data and identify deficiencies;
•    develop a groundwater  flow  and  contaminant transport model of  the study
     area;
•    calibrate the flow and transport model with existing data;
•    perform a  limited  sensitivity analysis with .the  final  calibrated model;
     and
•    evaluate remedial  action alternatives for  the  site with  the  calibrated
     model.
     A  conceptual  model   of  the  flow  system  was  developed  based  on  the
available hydrogeologic data.  This conceptual  model formed  the framework for
developing the flow and transport numerical model of  the site.
     The Finite Element Three-Dimensional  Groundwater  (FE3DGW)  code  (Gupta et
al., 1979) was initially used to  model  the groundwater flow  within an area at
and around the Western  Processing  Site.   A finite element grid was  developed
and the  necessary data  on geologic structure,  boundary conditions,  hydraulic
conductivities, and hydraulic stress  were input in the code.  Once  the flow
model was calibrated, these data  were input to  the  three-dimensional Coupled
Fluid,  Energy,  and Solute  Transport  (CFEST)  code  (Gupta et  al.,   1982)  to
simulate  the  groundwater   flow   and   contaminant  transport.    CFEST   is  an
extension of FE3DGW in  that  it  uses  the  same  hydrologic data  structure and
finite element grid.  In addition, CFEST  includes  the  necessary parameters to
couple contaminant transport with groundwater flow.

-------
     Once  developed,  the  flow  and  transport  portions  of  the  model   were
calibrated  to observed  1982  through  1984  potentiometric  and  contamination
data provided  by EPA Region X.   In this phase  of  the project,  the  transport
modeling was  based  strictly on observed concentrations  of  trichloroethylene
(TCE)  in  the groundwater,  surface water,  and soil  at  the  site.   A  limited
sensitivity  analysis was  performed  with  the final  calibrated  (base  case)
model  in order  to test the  sensitivity  of  the model results to  variations  in
the basic model input parameters.
     The final  calibrated  flow and transport  model was  used  to predict  the
effectiveness  of  remedial  action  alternatives  proposed   for   the   Western
Processing Site  by CH2M  HILL and  the Potentially Responsible Parties  (PRPs).
The CH2M HILL proposed actions  were:    1)  source removal  combined  with  pump
and treat; and 2)  cap  combined  with pump and treat.  The  PRPs proposed action
(Landau Associates  and Dames and  Moore,  1984) consisted  of  a  combination  of
source  removal,   a  slurry  wall  around  the  site,   and  pump and treat.    In
addition to  the  basic remedial  action  runs,  a no-action  run  (extending  the
base case"'run into  the future) was performed  to  provide a benchmark  against
which  remedial  action  results   could  be  compared.     Also,   a  few  minor
variations  (e.g.,   using  different  combinations   of  actions   and  variable
pumping rates) to  the  basic remedial action  runs were simulated in  order  to
better understand the model results.
     In all cases,  the  flow portion of  the model  was used to predict  changes
in the  groundwater potential (i.e.,  drawdown) and  volumes  of  water removed.
The contaminant  transport portion  of the model was  used  to  predict  the  mass
of TCE removed from  the system  and average  concentrations up to  50 years  into
the future.

-------
                                  SECTION 2
                                 CONCLUSIONS

     A  groundwater  flow   and  contaminant  transport  model  of  the  Western
Processing Site has been developed and calibrated.   An  acceptable calibration
was achieved both  in  terms of matching model-predicted  to  observed hydraulic
potentials  and TCE  concentrations,  as  well   as   accurately  predicting  the
groundwater gain of Mill  Creek over the  model  region,   and  the  concentration
of TCE  in  the  creek.   The model  was also  able  to  closely match  the estimated
total mass of TCE in the groundwater flow  system based  on monitoring data for
the period 1982 through 1984.
     A  4:imited   sensitivity  analysis   was   performed   to  determine   the
sensitivity  of "the  model  results  to  changes  in  hydraulic  conductivity,
porosity,   recharge,   dispersivity,   and   retardation  factor.    The  analysis
showed  that  all parameters  tested  have  some  impact  on the  results  of  the
final calibrated model, and  no parameter can be changed  without altering the
current calibration.
     The  final  calibrated model  was used  to  evaluate  the effectiveness  of
remedial  actions proposed  by  CH2M  HILL  and the  PRPs.    All  model  runs  have
simulated   the  transport  of  TCE.    Other  contaminants  on  site will  behave
differently because  they  have different  sorption  properties,  are  present  in
different   quantities  at the  site,  and/or  have different  source locations).
It is important to  keep this in mind when  extrapolating  the model results for
TCE to a more comprehensive remedial  action for all contaminants  on site.
     A  summary of  the  results for  all   remedial  action  simulation  runs  is
shown in  Table 1.   A  comparison of  the total mass  of TCE remaining  in the
groundwater flc1" system for the CH2M HILL and PRPs remedial  action cases are
shown in  Figure  la and Ib,  respectively.   The CH2M  HILL  PRPs  base case and
no-action   cases  show  essentially the same results.  The  only  difference  is
that  the   slurry wall   elements  in  the  PRPs  case  reduced the  size  of  the

-------
                     TABLE 1.   SUMMARY OF THE MODEL  RESULTS FOR THE CH2M  HILL  AND PRPS
                                  REMEDIAL  ACTION  SIMULATIONS.
                                                to Hill Creek
                     TCE In System   TCE In System     1983 - 1988
TCE Discharging   TCE Discharging.'
                                                                                X TCE
                                                                                             * TCE
Years to   Average
                to Hill Creek  '  Discharging to   Withdrawn   Years to    Remove    Drawdown
                 1908 - 1993      Hill Creek    by Pumping   Remove     90% of   Over Site
Cases in 1988 ( Ib) In 1993 ( Ib)
CH2H HILL Cases
No-Action 13.075 9.611
Source Removal 12,141 8,587
Source Removal,
Pump and Treat
(100 gpm) 1,511 593
Source Removal.
Pump and Treat
(200 gpm) 0 0
Cap, Pump and Treat
(100 gpm) 1,307 51M
Cap, Pump and Treat
(200 gpm) 0 0
PRPs Cases
No-Action 12.389 9.136
Source Removal 12.388 9,875
Source Removal,
Slurry Wall.
Pump and Treat
(100 gpm) 2.577 1.069
Source Removal,
Slurry Mall,
Pump and Treat
(?!)0 gpm) 0 0
Source Removal,
Pump and Treat
(100 gpm) 1.276 845
Source Removal,
Pi/mp dnd Treat
l?(X> gpm) 0 0
(Ib) ,

4,822
4,781


1.824


529

1.767

500

4.480
3.536



2,349



467


1.575


467
(Ib)

3.597
3.447


246


0

213

0

3.392
2.438



1.463



0


418


0
and Ditch

100
100


13


3

9

3

100
100



18



3


II


J
Wells

N/A
N/A


87


97

91

97

N/A
N/A



82



97


89


97
all TCE

•50
50


15


5

15

5

-50
>50



15



5


20


5
the TCE

30
30


5


• 5

5

• 5

30
40



10



5


5


5
(ft)

N/A
N/A


4.0


8.5

4.5

9.0

N/A
N/A



6.0



12.0


4.0


8.5
N/A  * Not Appl(cable

-------
    E •*•>
    QJ VI
    at >>
     13
    *
    ) 01
    — C
20,000  •]

18,000

16,000

14,000

12,000

10,000

 8,000

 6,000


 4,000

 2,000

    0
                                                   No Action
                                                          Source  Removal  and Pump Treat  (100 gpm)
                                                          'Cap and Pump and  Treat (100 gpm)
                     Source Removal  and Pump and  Treat (200 gpm)
                    'Cap and Pump and  Treat (200  gpm)
                    1963
               1S72
1983
                                             1993
                                         2003
2013
2023
                                                                                2033
                                                   Year
       Figure  la.   Comparison of  the Total  Mass  of  TCE Remaining  in  the Groundwater
                    Flow System for the  CH2M Hill  Remedial  Action  Cases
3
-0
           1963
                                           Source Removal,  Pump and Treat(200 gpm),
                                           with and without Slurry Wall
                                            Source  Removal  and Slurry Wall

                                                No Action

                                                     Source Reiroval, Pump and Treat (100 qpm)
                                                     Slurry Wall

                                                           Source Removal, Pump and Treat (100  gpm),
                                                           No Slurry Wall
                                                                       2033
      Figure Ib.   Comparison of  the Total Mass of  TCE  Remaining  in  the  Groundwate-
                    Flow System for the  PRPs  Remedial  Action Cases

-------
Reaction Pond III  and Well  21  source area which in turn, slightly reduced  the
mass loading to the system.
     The results  for  the pump and treat  where  the wells were pumping 200  gpm
(758 1/min)  showed that all of  the  TCE  would  be  removed from  the  system  in
the  first  5 years.   All  cases  where  the  wells  were  pumping  100  gpm  (379
1/min)  showed that 85%  to  90%  of the TCE is removed in  the  first  5  years  and
the  remainder is  removed within the next 10 to  15 years.   The slurry wall  in
the  PRPs cases  prevented the pumping wells  from removing TCE that  had moved
off  site, and therefore, the  simulations  with  the  wall were less effective  at
removing the TCE.
     There  do  exist  some  small  differences  in the  results for  the various
model simulations.  But, for the most  part,  all  remedial  actions  were about
equally effective in cleaning up the TCE  in a 5 to 15 year time period.

-------
                                  SECTION 3
                        DESCRIPTION OF THE STUDY AREA

     The  Western  Processing  Site  is  located  within  the  City  of  Kent,
approximately  four  miles (6  km)  north of  the business district  (Figure 2).
The facility occupies an area of about 13 acres  (5  ha),  and when in operation
consisted  of  a  small  laboratory,  a  solvent  recycling  plant,  a  fertilizer
plant,  bulk   storage  tanks,   drum  storage  areas,  piles   of  flue  dust,
construction debris,  and large cement-block above-ground  storage  lagoons for
liquid  wastes,  cooling water,  and  process  water  (EPA,  1983).   Mill  Creek,
also  known as King  County Drainage  Ditch  No.  1,  runs  across  the northwest
corner of-rthe site from south to north.  A  drainage  ditch,  bicycle  trail, and
railroad tracks run along the eastern boundary of the site.
     During the fall  and winter of  1984,  the site was partially excavated to
remove  surface  structures,  waste  piles,  and some  of the  contaminated  soil.
The  modeling  discussed  in  this   document   uses  the  pre-excavation  ground-
surface  elevation  (averaging  25   ft  above  mean sea  level)  as a  reference
point.

CLIMATE
     The  annual  average rainfall  at  the  Western  Processing  Site  is  39  in.
(99 cm).  There is a well defined  dry  season in  the  summer  and a rainy season
in the winter.  Table 2  shows  the  monthly average  of precipitation, potential
evapotranspiration,   and   actual    evapotranspiration.      The   amount   of
precipitation  that  recharges the  aquifer  was  estimated  to  range from  4 to
12 in./yr  (10  to  30  cm/yr).   Using a method  described  by Dunne and  Leopold
(1978),  a  recharge  of 8   in./yr   (20  cm/yr)  was  obtained.    A  detailed
description of recharge calculations is contained in Appendix B.

-------
                                               n   MISCELLANEOUS DEBRIS
S&ATTU
  BELLfVUE
 RCNTON

 KENT
                                                     PLASTIC COVERED
                                                     PILE  OF    |
                                                     CONTAMINATED SOIL
               LAGOONS  FILLED WITH
               CONTAMINATED WATER
                                                          FERTILIZER PLANT
                EMPTY DRUMS


                        FULL DRUMS
BATTERY CHIPS
             STEEL MILL FLUE DUST
           CONTAINING HEAVY METALS
  NOT TO SCALE
  V///A PONOED  WATER
  ^^ ASPHALT
                                        DRUMS OF ZINC CHLORIDE
Figure 2.   Western Processing  Site  Before  1984  Excavation

-------
    TABLE 2.  AVERAGE MONTHLY PRECIPITATION, POTENTIAL EVAPOTRANSPIRATION
              (PET), AND ACTUAL EVAPOTRANSPIRATION (AET) FOR THE  SEATTLE AREA
                                               PET,** in.     AET,**  in.
                                                   0.3            0.3
                                                   0.6            0.6
                                                   1.2            1.2
                                                   1.8            1.8
                                                   3.1            3.0
                                                   3.8            2.9
                                                   4.5            2.0
                                                   4.1            1.6
                                                   2.8            1.9
                                                   1.8            1.8
                                                   0.8            0.8
                                                   0.5            0.5
Month Precipitation,* in.
January
February
March
April
May
June
July
August
September
October
November
December
Annual
5.73
4.24
3.79
2.40
1.73
1.58
0.81
0.95
2.05
4.02
5.35
6.29
38.94
                                                  25.3           18.4
          *(NOAA, 1974)
          **(EH1s, 1984)
GEOLOGY
     The Western  Processing  Site lies  in  the broad flood plain  of  the Green
River.   Elevations  in this valley average 20  ft  (6 m) above mean  sea level.
The  sediments  include alluvial  fan  deposits  of  sand, silt, peaty  silt,  and
clay more  than 150 ft  (45 m)  thick,  primarily derived  from Mt.  Rainer  and
transported by the White River (Luzier, 1969).
     The top  8 ft  (2.4  m) underlying  the Western  Processing  Site  consists
primarily of unsaturated  artificial  fill  material.   Below the  fill,  the site
is  underlain   by  a mixture  of  sand,  silt,   clay,   and  peat  to   a  depth  of
approximately 40 ft (12 m) (Unit  1).   Intermittent  clay lenses  are present in
this layer  (Hart-Crowser,  1984).   From 40 to  150  ft  (12  to 46 m)  belov,  the
surface  (Unit  2)  the  material consists  predominantly  of fine  to  medium sand
with occasional  layers  of silty  and/or gravelly  sand  (Hart-Crowser,  1984).
Underlying  Unit 2  is  a  thick  (greater than 218 ft)  layer  of silt which forms
an aquitard (CH2M HILL,  1984).

-------
          Potential contours
          1n feet above mean
          sea level
                                              FEET

                                            >  FLOW  DIRECTION
Figure 3.   Smoothed Kriged Potential  Surface  for April,  1984
                          11

-------
 site.   As a result of  this  action,  disposal at  the  site ceased in  1982.   In
 1982  and  1983, the  EPA  installed  a series of monitor wells and  collected soil
 and  water  samples.    An  initial  remedial  measure  was  implemented  (limited
 capping,  excavation and removal,   drum  storage,  etc.) to  contain  some of  the
 waste   until   a   more  permanent  remedial   action   could  be  designed   and
 implemented.
      In  1984,  the  EPA  did  more  extensive  excavation,  removing  all  of  the
 structures  and materials  (drums,  waste  piles,  buildings, etc.)  on site  and
 removing  some  of the contaminated soil.   Waste  disposal  records for the  site
 are  sketchy;  however,  a  review  of the  records  that  are available  indicates
 that  TCE  was one of the most common wastes received, and  that it was received
 from   about  1960  to   1980.     Sampling   results   confirm   the   widespread
 distribution of  TCE  in  on-site wells,  although it has been  detected (at  low
 levels) in  only  two of  the off-site wells.
     There  is  no evidence that trans-l,2-dichloroethylene (DCE)  had ever been
 received  at the  site,  however, DCE  has been found  on site  in  both soil  and
 groundwater  samples.    In some cases the  level  of DCE approaches  or exceeds
 the level of TCE.
     As  a  general   trend, .  it  has  been  found  that  the  occurrence of  the
 dichlorinated  species   increases  downgradisnt  from  source  areas.    These
 findings  led  to  the  conclusion that  degradation of TCE  is  the  likely source
 of  DCE.    The  most  plausible   mechanism  for  such  a  transformation   is
 biodegradation (Wood  et al., 1981).
     As  a  result  of  these  determinations,  the contaminant  transport  model
 calibration  assumed  that this  TCE  to  DCE transformation  process  occurs.
 Because this transformation  cannot be simulated in the model,  the sum of  TCE
 and DCE concentrations  was used to calibrate  the model.
     The  high  TCE  concentrations   measured  in the soil and/or  groundwater  in
 Wells 11, 15,  17, 20, 21, and  27  (Figure 4)  indicates the existence of  three
 probable TCE source areas:   1) Reaction  Pond I;  2) Reaction  Pond III;  and  3)
 around Well- 21.  A  summary of  the measured TCE  and DCE concentrations  in  the
 soil  and  groundwater at  the  three  suspected  source  locations is  shown  in
Table 4.   The  sum  of the TCE  and  DCE  concentrations in  the groundwater  was
 used for model  calibration.
                                      13

-------
  South 196th ST.
             )pC,i
                                                 1A.B  2   3
                                 Reaction Pond
                                    II
                                    Western Processing
                                    [Office Building
                                            22A.B

                                         WESTERN
                                       PROCESSING
    66    132    I98   264    330
   •  MONITOR WELL
I-III  REACTION PONDS         !

 •	  PROPERTY BOUNDARY  AND '   24
      PROPOSED SLURRY  WASTE
      LOCATION               '	
2SA, B
oil
                       WELL 30
        Figure 4.  Western  Processing Well and Reaction  Pond  Locations
                                    14

-------
                  TABLE 4.  SUMMARY OF SOIL AND GROUNDWATER CONCENTRATIONS OF TCE AND DCE (EPA, 1982)
                            AT THE THREE SUSPECTED SOURCE AREAS
                             Associated   Concentration TCE  (ppb)   Concentration DCE (ppb)   Total TCE and DCE
Source
Reaction Pond I
Reaction Pond III
Around Well 21
Well Number
11
15
17
20
27
21
Soil
312
580,000
558,000
676
NS
1,520
Groundwater
80,000
210,000
42,000
1,100
140,000
170,000
.; Soil
0
0
0
0
NS
24
Groundwater
0
0
0
0
0
390,000
in Groundwater
80,000
210,000*
42,000
1,000
140,000*
560,000*
tn
NS = No soil samples taken.
Concentration matched in the model calibration process.
Soil samples may have been taken from either the saturated or unsaturated zone and
  were not necessarily in contact with the groundwater.

-------
     The maximum  observed TCE  concentration  in  groundwater is  210,000 ppb,
while  the  maximum  observed  soil  concentration  is  558,000  ppb.    The  EPA
priority pollutant  human  health criteria for  TCE  in water  is 27  ppb  at 10"'
cancer risk (EPA,  1980).
                                     16

-------
                                  SECTION 4
                           -   MODEL DEVELOPMENT

     A  numerical  model  was  developed  to  simulate  groundwater  flow  and
contaminant  transport  at  the  Western  Processing  Site.    The  model  was
developed in two steps:
1)   a flow model  was developed to describe the  groundwater  flow in the area
     around the Western Processing Site; and
2)   the  flow  model  was  used  to form  the basis  of  a transport  model  which
     simulated the movement of contaminants in the groundwater.
Although  the model  was developed in two stages,  the  final  result is a single
model  which  can  be  used  to  simulate  groundwater  flow  and  contaminant
transport at the site.  Because  the model  was  developed in a staged approach,
the flow  and transport portions will be discussed separately.

MODEL SELECTION
     A three-dimensional  model  was selected  for the Western  Processing Site
because it is able to  simulate:  1) variations  in permeability with depth; 2)
the vertical flow within  the study  area; 3)  localized discharge to Mill  Creek
and the drainage ditch; and 4) slurry wall  and  pumping  depths  in the proposed
remedial   actions.
     The  numerical  codes  selected  to  model the  Western  Processing  Site  are
the FE3D6W flow code  and -the CFEST transport code.  The FE3DGW code simulates
groundwater  flow  while  its   companion  code,  CFEST,  simulates  contaminant
transport  coupled   with   groundwater  flow.    The  two  codes   are  completely
compatible  such  that  the  simulation  of  transport   phenomena  using  CFEST
proceeds   directly from calibration  of  FE3DGW  based on  flow  properties.   Both
codes  have  been  benchmarked  against   other  numerical  codes  and   have  been
verified   by solution  of  standard  analytical  problems (Gupta  et al., 1979 and
1982).

                                     17

-------
GROUNDWATER FLOW MODEL DEVELOPMENT
     The flow model  of the local area  around  the Western Processing Site  was
developed  based on  the  available  hydrogeologic  data.    The  model  area  is
2,790 ft  (850  m)  wide  and  4,020 ft  (1,225  m)  long.  The Western  Processing
Site is located near the center of the model region  (Figure 5).
     A  finite  element  grid  was  developed  for   the  local  model  region  to
properly  represent  the   area!   extent,   boundary  conditions,  and  primary
features  of  the hydrologic system.   The  grid  consists  of 311  nodes and  283
elements. The  two-dimensional  surface representation of  the  grid is shown  in
Figure 6.
       Data  files   were  developed   for  the aquifer  thickness   and  extent,
vertical   and   horizontal  hydraulic   conductivity,  and   hydraulic   stress
(recharge  and  discharge)  using  data received  from EPA  Region X.   The  data
used in the final  calibrated flow model are  discussed below.

Structure...
     The top 100 ft (30 m) below the water  table was  simulated in  the  model.
The  top  30 ft  (9  m)  was  simulated  as a  silt and fine sand  material  with
intermittent clay  lenses  (Unit  1 as defined  in  the previous  section).   The
intermittent clay lenses  were  simulated by  reducing the  ratio  of vertical  to
horizontal permeability.   A coarser  sand material  (Unit  2 -as  defined   in  the
previous section)  was  simulated  between 30 and  100 ft  (9  and 30  m)  below  the
water table.
     Although the region  was  simulated with two  geologic units,  the vertical
dimension in the model was simulated as four layers.   Layers  1  and  2 composed
Unit 1, while Layers 3 and 4  composed Unit 2.  The  layer thicknesses from  top
to bottom were 13  ft (4 m), 17 ft (5 m),  10  ft  (3  m),  and 60  ft (18 m).    This
subdivision  of  geologic  units  allowed  for  more  vertical  resolution  in  the
model,  as well  as  allowing   an accurate  representation  of  pumping   depth,
source  removal  depth,   and   slurry  wall  depth   in   the  remedial   action
simulations.
                                      18

-------
• WELL LOCHT10N MI MLMBt
  Figure 5.   Western  Processing Model  Area
                     19

-------
Boundary Conditions
     The  model  boundaries  were  defined  with  the  FE3DGW  code's  "leakance
boundary condition" option,  which can be  defined  as a combination of flux  and
constant head  (held) boundary  conditions.   This option uses the distance from
the boundary  to a known  potential,  the potential  at that distance,  and  the
cross-sectional  area  associated  with  each  node  to  calculate  the flux  and
potential  at  the boundary nodes  based  on the  model  conditions.   In  effect,
this option calculates the potential  along  the  boundary based on regional  and
local data.  The regional data used  to  calculate  the boundary conditions were
obtained from  regional  wells  and Green  River  elevations.   Elevations   along
the Green River were assumed to be  hydraulically  connected to the groundwater
system.
     Mill Creek  and the  ditch  to the east of the site were simulated  using a
stream boundary option.   Rather than  holding  the  groundwater  potential  at  the
elevation  of  the  creek  and  the  ditch,  this  option  allows  the  model  to
calculate^ the  groundwater potential  based  on  the  surface  water  elevation;
stream  bottom  elevation, cross-sectional  area,  thickness, and  permeability;
and minimum stream depth.   Surface water elevations at each  node  along Mill
Creek and  the  ditch,  were interpolated or  extrapolated from surveyed  values
obtained in  April, 1984  (Figure  7).   A more  detailed  description  of  the
boundary conditions is provided in Appendix C.
Hydraulic Conductivity
     Initially,  horizontal  (Kh)  and vertical  (Kv)  hydraulic conductivities
were assigned  to the two material  types discussed above based on  data  in  the
Hart-Crowser  report  (1984).     These  values  were  then  adjusted  in   the
calibration process until  a  good match was achieved  between model-predicted
and observed potentials.
     The hydraulic conductivities used  in the final  calibrated model for each
material  (Units  1 and  2) are shown in  Table 5.    The  values  were  uniform
throughout each  layer.   For  both material types,  the final calibrated values
were within the range of values reported by Hart-Crowser (1984).
                                      21

-------
  '   •  Well Location
     A Survey Location and Elevation in   i
       Feet Above Mean Sea Level
                                                II...
Figure 7.   Location  of  Surveyed  Values  for Mill  Creek and Drainage Ditch

                                   22

-------
     TABLE 5.  PARAMETERS USED IN THE FINAL CALIBRATED GROUNDWATER
               FLOW AND CONTAMINANT TRANSPORT MODEL
                                             Model  Layer
                                                       3~
Structure
  Top (ft AMSL)                  15         2       -15       -25
  Bottom (ft AMSL)                2       -15       -25       -85
Thickness (ft)                   13        17        10       60
Horizontal Hydraulic
  Conductivity (ft/day)           3.5       3.5      50       50
Vertical Hydraulic
  Conductivity (ft/day)           0.175     0.175     5        5
Kv/Kh                             0.05      0.05      0.1      0.1
Effective Porosity                0.15      0.15      0.15     0.15
Longitudinal Dispersivity (ft)   25        25        25       25
Transverse Dispersivity (ft)      5555
Retardation Factor                4444
Lithology                        silty     silty     sand     sand
                                  sand     sand
Surface Recharge = 8.0 in./yr
                               23

-------
Groundwater Potential
     Due  to  a  lack  of transient  data,  it  was  assumed  that  the groundwater
system  around  the Western Processing  Site  is in  steady  state.   A  review  of
the  transient  potential   data  that  have  been   collected  showed  that,  the
potentials do not change  enough to  significantly  alter the flow field or flow
velocities.   Therefore,  the  steady  state   assumption  is  considered  to   be
acceptable for the modeling effort.
     A  contour  surface  of  the   April,  1984  potential  data was   used   to
represent  the  initial  potential   conditions  (Figure  3).    This  surface  was
prepared  by  kriging  potential  data from 36  wells  on  and  around the site, and
measurements  along   Mill  Creek  and  the  ditch.    Kriging  is   a  statistical
technique  used  to estimate  a  surface from  spatially-distributed data.   The
model-predicted  potential  surface  was  compared  to  the  kriged  potential
surface in the model calibration process.
Hydraulic Stress
     The  only  hydraulic  stress   considered  within  the  model  region  was
recharge  from precipitation.   Recharge was  assumed constant over  the area  at
8  in./yr  (20 cm/yr).   The  only exception  was in  the  asphalted (capped) area
on  the site  (Figure 2)  where recharge  was set  at  0  in./yr.   A  detailed
description of the recharge calculations is  contained in Appendix B.
Porosity
     Measured values of porosity  are  not available for the Western Processing
Site.  As a reasonable estimate,  an effective porosity of  15%  was  used in all
layers of the model.

CONTAMINANT TRANSPORT MODEL DEVELOPMENT
     The  contaminant  transport  model" was  developed using  the  calibrated flow
model,  observed  or  estimated  migration  parameters,  and  estimates  of source
loading on  the  groundwater system  as  a function  of  time.  Data  input  files
were  developed  to define  source  concentrations,   leaching  rates,  retardation
factors,  and  dispersivity.   In most  cases,  these  data were not specifically
known  for the Western  Processing Site.    As a  reasonable  estimate,  initial
values were selected from  the  literature and  final values were derived  in the-

                                      24

-------
model calibration  process.   The  data used in  the  final  calibrated transport
model are discussed below.

Contaminant Selection
     A review  of the list  of  wastes received  at the  Western Processing Site
showed  that TCE  was accepted  throughout  the operating  life  of  the  site.
Also, high  concentrations of TCE  have  been measured in many of the wells on
site  and it  is  one  of  the  ubiquitous   contaminants.    Therefore,  TCE was
selected  for  use in  calibrating  the transport model, and  for  comparison of
the various remedial action alternatives simulated.
     As  discussed earlier,  the  DCE  observed  on   site  is  believed to  be a
degradation product of the TCE.   Since  the model  is not capable of  simulating
this degradation process, the  initial mass (concentration)  of TCE was  assumed
to  be   the  total  mass   (concentration)   of  TCE  plus   DCE.     The   total
concentration of TCE  plus DCE  at each  of  the three suspected source areas is
shown in-Table  4.   These totals  are  the  concentrations that  were matched in
the model calibration process.  From  this  point on, the total of TCE plus DCE
will be referred to simply as TCE.
Source Location
     A review  of the  sampling results for  TCE  in the  on-site  wells  (EPA,
1983) reveals three  probable  source  areas:   1) Reaction Pond I; 2) Reaction
Pond  III;  and  3) around  Well  21.   The area  of  the finite  elements  used to
simulate  Reaction Ponds  I and  III  was  set to  represent  the  actual  areas of
these ponds.   Reaction  Pond I  was  represented in  the model  by  elements with
                     ?        7
an  area  of  7,800 ft   (750 m  ).   Reaction Pond  III  was  represented  in the
                                          ?       p
model by elements with an area of 7,050 ft^ (655 m).
     For an unknown number of years during  the  operation  of the  site,  TCE was
mixed with fly ash  in the area  around Well 21.  High concentrations of TCE in
both the  soil and groundwater  at  the  location of Well 21  indicate a probable
source.    The area  of this source  is  unknown,  therefore,  it was  determined in
the  model  calibration  process by  the area  which  resulted  in  the   desired
concentration and  mass  loading  of TCE.   In  the final  calibrated  model the
                                                                             2
source around Well 21 was represented by an element with  an area of 1,600 ft
(149 m2).

                                      25

-------
Source Area Concentrations
     Leaching  of  TCE into  the  groundwater,  rather  than  direct infiltration,
was the  primary mechanism for introducing TCE  to  the model.   Therefore, the
initial TCE concentration  was defined at the elements  representing the  three
source  areas,  and  the  loading  rate at  each  site  was   calculated   as the
infiltration rate times  the surface area of the source  times the  initial TCE
concentration  at  the source.   The  initial  TCE concentrations  at  the  three
sources were defined  as  follows:   Reaction Pond I  —  6.0  x 105 ppb, Reaction
Pond  III —  1.9  x  106  ppb, and  around Well  21  —1.15  x  107   ppb.   The
infiltration rate was  constant over  the  entire model region  at 8  in./yr (20
cm/yr).   Using the areas  for each source given above,  the  loading  rates  in
the model  at Reaction  Ponds  I and III, and around  Well  21  were  195 Ib/yr (88
Kg/yr),  559  Ib/yr  (254  Kg/yr),  and 769 Ib/yr  (349  Kg/yr),  respectively,  over
the 20-year active disposal  life  of  the  site.   The total model-predicted mass
of  TCE  disposal  at  the  site  was   1,523  Ib/yr  (689  Kg/yr)  or  30,460  Ib
(13,781 Kg) over the 20-year disposal period.
Source Duration/Leach Rate
     The  sources  were   assumed   to  be  actively   leaching  TCE   into   the
groundwater for 20 years,  from 1963  through 1983.    The  rate  of  leaching was
constant over the-20 years at the rates given  above.
     The mass  of  TCE currently  in  the unsaturated  zone  at the three  source
areas  was  estimated  to  determine how long  leaching into  the  saturated  flow
system  would  continue   (see  Appendix  A).    This   information  was  used  to
determine how  long  to  keep the sources active  in  the model.  The  results  of
these  calculations   indicate  that  virtually  all   the  mass   of  TCE   in  the
unsaturated  zone  beneath  Reation Pond  III  and  around Well  21  has  already
leached  into the  saturated zone.   Therefore,  these  two sources were  turned
off in  the  model  in 1983.   The  calculations  show  that Reaction Pond  I would
continue to  leach TCE  into the groundwater for  about  20 years, and  that the
rat"  decreases  exponentially.   Based on  the  calculations,  the  model   input
concentration at Reaction  Pond  I was  set  at  full   strengh  (6 x 105  ppb) for
the first five years  into the future  (1984  -  1988), and then  it  was  reduced
by 75% at each  of the next three time steps (1989 -  1993 =  50,000  ppb,  1994 -
                                     26

-------
1998 =  37,500 ppb, and  1999  - 2003  =  9,400 ppb).  After  the year 2003,  all
sources of TCE were turned off in the model.
SorptIon/Retardation
     The  CFEST  model  uses  a  single  retardation  factor   for   the  entire
groundwater  flow  system  (all  layers),  which  was  determined  in  the  model
calibration  process.   The  retardation  factor  used  in  the  final calibrated
model was 4.0.
     The retardation  factor  can be  calculated  from the bulk density, actual
and  effective  porosity,  and  the  distribution  coefficient  (Kd)  for   the
material through which flow occurs.   A  discussion of these parameters and  how
they relate to a retardation factor of 4.0 is contained in Appendix D.
                                     27

-------
                                  SECTION 5
                              MODEL CALIBRATION

     The flow and transport models  were  calibrated  by adjusting certain model
input parameters until  a good match was achieved between  model-predicted and
observed data.   A  brief description of  the  calibration process for  both the
flow and transport models is provided below.

FLOW MODEL CALIBRATION
     Once the data  were input into the  FE3DGW code,  the model  was run in the
steady-state  mode   to   predict   groundwater  potentials.     The   model   was
calibrated by comparing the model-predicted  flow field to  measured potential
data.
     The difference  between model-predicted  and  measured  hydraulic potentials
was minimized by adjusting the following flow  model  parameters:   the vertical
and horizontal hydraulic  conductivity, the parameters controlling  the flow to
Mill Creek and the  drainage ditch (stream  bottom permeability and  thickness),
and the boundary conditions.
     The final model-predicted potential surface for the water  table (top of
Layer 1) (Figure 8)  compares  well with  the  kriged  potential data (Figure 3)
and  the  conceptual   model  of the flow  regime  within  the  study  area (i.e.,
localized  flow   to   Mill  Creek   and  the  ditch,  and  regional  flow  to  the
northwest).    Potential  surfaces   for  the model-predicted  top of Layers  2, 3,
4,  and  the  bottom   of  Layer 4   are  shown  in  Figures  9,  10, 11,   and  12,
respectively.  The  model-predicted groundwater  flux  to Mill Creek  along the
reach within the study  area  is  0.45 cfs  (1,101 m3/day).   This  value compares
well with a  gain of 0.5 cfs  (1,223 nr/day)  along Mill  Creek within the study
area as measured in  May, 1982, by EPA Region  X.
     Regional groundwater flow as predicted  by the  model  is to the northwest,
while localized  flow is to Mill  Creek.   Discharge to  Mill  Creek dominates the

                                      28

-------
ro
     Figure 8.  Model-Predicted 1983 Top of Layer 1
                Potential  Surface for the Base Case
                Simulation
Figure 9.  Model-Predicted 1983 Top of Layer 2
           Potential Surface for the Base Case
           Simulation

-------
to
o
       Figure  10.   Model-Predicted 1983 Top of Layer 3
                   Potential  Surface for the Base Case
                   Simulation
Figure 11.  Model-Predicted 1983 Top of Layer 4
            Potential  Surface for the Base Case
            Simulation

-------
                                                                                     "71   f
UJ
                           Figure 12.   Model-Predicted 1983 Bottom of Layer 4
                                       Potential  Surface for the Base Case
                                       Slmulat Ion.

-------
flow patterns  to  a depth of about  30  ft (9 m) and  its  influence can  be  seen
at 100 ft  (30  m).   The model predicts that  regional  groundwater flow  becomes
dominant somewhere between 30 and 50 ft  (9 and 15 m) below the surface.

TRANSPORT MODEL CALIBRATION
     Once  the  ^data were input into  the  CFEST code, the model was  run in the
transient  mode with five-year time  steps  from 1963 to  1983.    The  model  was
calibrated  by  comparing model-predicted to  measured  TCE  cpncentrations  for
1983, TCE  mass  loading  to  Mill Creek,  and  total  mass of TCE in  the  system  in
1983.
     The  difference between model-predicted  and  measured  data  was  minimized
by adjusting the retardation factor  and  source strengths  in the  model.  Leach
time was  assumed to  be  the period  of active disposal (20 years).  Because the
unsaturated zone is  very thin, it  was assumed that TCE  entered  the  saturated
zone soon  after disposal.
     A  sm'bothed kriged  concentration  contour plot  of TCE  within  the model
study area is  shown  in  Figure  13.   The model-predicted  TCE concentrations for
the  top  of  Layer  1  (Figure  14)  compare  reasonably  well with the  kriged
values.   The model-predicted TCE concentrations  for the  top of  Layers 2,  3,
4,  and the  bottom  of  Layer  4  are  shown  in  Figures  15,  16,  17,  and  18,
respectively.
     In addition to  matching the observed  location  of  the  TCE  plumes from the
three  source  areas,  it  was  important  to  match  the  maximum   observed  TCE
concentration  at  these  areas.     This  match  was  achieved  through  source
calibration.   The measured  and  model-predicted   concentrations   at  the three
                                       i
source locations are shown in Table 6.
     The  concentration  of TCE. in  Mill  Creek was  calculated  based  on model
results  and  compared  to the  measured  concentration.    The  model-calculated
concentration  of  39 ppb  based on  a  creek  flow of  15 cfs  (0.4 rn^/sec)  is
comparable to the creek TCE  concentration of  15 ppb  measured in  January 1984,
by EPA Region X.
                                     32

-------
Figure 13.  Smoothed Kriged TCE Concentration
            Contours, Fall , 1982
                                                                                              Numerical
                                                                                              Dispersion
                                                           Numerical
                                                           Dispersion^1
                                                                                              Numerical
                                                                                              Dispersion
                                                                                     \
                                                                               Numerical
                                                                               Dispersion
Figure 14.   Model-Predicted 1983 Top of Layer  1
            TCE Concentration Contours for  the
            Base Case Simulation

-------
Figure 15.  Model-Predicted 1983 Top of Layer ?
            TCE Concentration Contours for the
            Base Case Simulation
Figure 16.   Model-Predicted 1983 Top of Layer 3
            TCE Concentration Contours for the
            Base Case Simulation

-------
CO
en
        Figure  17.   Model-Predicted  1983  Top  of  Layer  4
                    TCE  Concentration  Contours  for  the
                    Case Case Simulation.
Figure 18.   Model-Predicted 1983 Bottom of Layer 4
            TCE Concentration Contours for the
            Dase Case Simulation

-------
            TABLE 6.  COMPARISON OF OBSERVED TO MODEL-PREDICTED MAXIMUM
                      TCE CONCENTRATIONS IN THE GROUNDWATER AT THE  THREE
                      SOURCE LOCATIONS
                                     Measured        Model-Predicted
                                  Concentration      Concentration
             Source Location          (ppb)         	(ppb)	
            Reaction Pond I          210,000            212,000
            Reaction Pond III        140,000            139,000
            Around Well 21     •      560,000            557,000

     The  total  mass of TCE  (actually  TCE plus  DCE)  in  the  flow   system in
1983 as predicted  by  the  model  was  17,100 Ib (7,737 Kg).  This value compared
well with the 18,000  Ib (8,144 Kg)  of  TCE plus  DCE as estimated independently
by CH2M HILL  (1985) and Landau  Associates and  Dames and Moore (1984) based on
the 1982  through  1984  measured  concentration data.  A  list of the parameters
used in the final calibrated model is shown in Table 5.

BASE CASE MODEL RESULTS
     The  base  case  is defined  as  the  20-year  simulation  period  from  1963
through 1983.   Over this  20-year  period,  the model predicted  that  a total  of
30,400  Ib (13,756 Kg) of  TCE  entered  the groundwater  flow  system.   Of  this
total, 17,100 Ib (7,738 Kg) remained in the  flow  system  in 1983.   Of the  mass
exiting the  system 97%  (13,300  Ib  (6,018 Kg))  discharged  to Mill  Creek and
the remaining 3%  discharged  to  the drainage ditch  along  the  eastern boundary
of the  site.    No TCE entered  the deeper,  regional  flow system  which flows
northwest toward the Green River.   The distribution of  TCE  in the groundwater
flow system for the base case is shown  in Table 7.

NUMERICAL DISPERSION
     Figures  14 and  15  have  low  concentrations  of  TCE upgradient  from the
sources   where   it   is   not  possible   (from   the  conceptual   model)   for
contamination to  have  occurred  from Western Processing.  Occurrences of  this
phenomena  are evidenced  by  the closed  loop 100 ppb  contours  west  of  Mill
Creek (which  is  upgradient  from the creek in the top  three model  layers) and
                                      36

-------
south and  east  of Well  21;  these contour lines  are  identified in Figure  14.
In all  cases,  these  upgradient occurrences  of TCE  are caused  by  numerical
dispersion.   Numerical dispersion  is  inherent  to  the  numerical  solution of
the  convective   dispersion   equation   contained  in  the   model   code   and,
therefore,  cannot  be  avoided.    Numerical   dispersion can   be   reduced  by
adjusting  the  model   node 'spacing,   however,   it  can  never  be  completely
eliminated.   In  all  TCE  concentration contour plots (i.e.,  remedial action
simulation  plots),  upgradient  occurrences of  TCE  have  been  attributed  to
numerical dispersion'and should be disregarded.
       TABLE 7.  DISTRIBUTION OF TCE IN THE MODEL BASE CASE SIMULATION
           Year
           1968
           1973
           1978
           1983
           Total
    TCE
Inflow fib)
   7,601
   7,601
   7,601
   7,601
  30,404
     TCE
Outflow (lb)
    1,049
    2,906
    4,183
    5,195
   13,333
TCE Remaining
in Groundwater
  System  (lb)
    6,552
   11,247
   14,655
   17,071
                                     37

-------
                                  SECTION 6
                             SENSITIVITY  ANALYSIS

     After the model  calibration  was  completed,  several  additional model runs
were made  to  test the sensitivity of  the  model  results  to changes in various
model  input  parameters.   The  sensitivity  runs were  performed with  the base
case model (1963  -  1983)  and the  base case extended 20 years into the future.
Thus,  the total  simulation  period  was 40  years  (1963  -   2003),   and  the
sensitivity runs  were performed  for  the case where no  remedial  actions were
simulated  in the model (base case and no-action case).
     The  model  parameters varied  in  the sensitivity  analysis  were hydraulic
conductivity, porosity,  recharge, retardation, and  dispersivity.   A  summary
of the sensitivity  runs  performed is  shown  in Table 8.  For the most part the
sensitivity analysis  consisted of doubling  and  halving  each  parameter  while
holding  all  other parameters  constant.  A  summary of the  values  of  the base
case model parameters is shown in Table 5.
     The  results  of  all  the sensitivity runs are  shown in Tables  9  through
12.  In  all  sensitivity runs, the mass  of  TCE entering the  groundwater flow
system from the source  areas was  not  changed (the infiltration at the sources
was the  same  as  the base case).   Table  9  summarizes  the mass  of  TCE  exiting
the  system  at  each  time   step  (5-year  intervals)   to  Mill   Creek   and  the
drainage ditch east  of  the  site.   In all   cases,  virtually all  (97%)  of the
TCE exiting  the   system  goes to  Mill  Creek.  Table  10  summarizes  the  total
mass of TCE remaining in  the system  at each time  step.  Table  11  reports the
model-predicted TCE  concentrations for  1983 at   the  three source  areas,  the
model-predicted groundwater  flux  to Mill Creek over  the  model  region,  and the
concentration of  TCE  in Mill  Creek  based  on the creek  loadings  predicted by
the model.  Table 12  summarizes the percent  changes  of mass and concentration
of TCE  based  on 1983 results.
                                      38

-------
                    TABLE 8.  SUMMARY OF SENSITIVITY RUNS
  Run
 Number   	Parameter(s)  Adjusted
    1     For all layers, the K is reduced by 0.5.
          Layers 1 and 2:  Kh = 1.75 ft/day, Kv = 8.75 x  10-2  ft/day.
          Layers 3 and 4:  Kh = 25 ft/day, Kv = 2.5 ft/day.

    2     For all layers, the K is doubled.
          Layers 1 and 2:  Kh = 7.0 ft/day, Kv = 0.35 ft/day.
          Layers 3 and 4:  K = 100.0 ft/day, Kv = 10.0 ft/day.

    3     The K for all layers is reduced by one-tenth.
          Layers 1 and 2:  Kh = 0.35, Kv = 0.0175.
          Layers 3 and 4:  Kh = 5.0, Kv = 0.5.

    4     The K for all layers is increased ten times.
          Layers 1 and 2:  Kh = 35, Kv = 1.75.
          Layers 3 and 4:  Kh = 500, Kv = 50.

    5—   The effective porosity is reduced by 0.5 for all layers; porosity1
          7.5%.

    6     The effective porosity is doubled for all layers; porosity = 30%.

    1     The recharge is reduced to zero.

    8     The recharge is increased to 16 in./yr.

    9     The retardation is reduced to zero.

   10     The retardation is increased to 8.

   11     The dispersivity is reduced by a factor of 5:  D,  = 5 ft, DT =
          1 ft.                                           L          '
   12
The dispersivity is increased by a factor of 5:  D.  = 125 ft,
n  - -3c f+                                        L
          DT = 25 ft.
   13     The effective porosity is 30% for all layers and the retardation is 2.

   14     The effective porosity of Layers 1 and 2 is 45% and the porosity of
          Layers 3 and 4 is 40%.

   15     For Layers 3 and 4:   Kh = 25 ft/day, Kv = 2.5 ft/day; and porosity
          = 25%

   16     Effective porosity = 25%, retardation = 8.0.

K = hydraulic conductivity
                                   39

-------
                                    TABLE  9.   SUMMARY OF  MASS OF TCE (LB) EXITING

                                              THE  SYSTEM  FOR THE SENSITIVITY RUNS
-F*
o
Time
Ste
1
2
3
4
5
6
7
8
Time
Step
1
2
3
4
5
6
7
8
p Year
1968
1973
1978
1983
1988
1993
1998
2003

Year
1968
Base
Case Run 1 Run 2
1,
2,
4,
5,
4,
3,
2,
2,
Base
Case
1,049
1973 2,906
1978 4,183
1983 5,195
1988 4,971
1993 3,708
1998 2,839
2003 2,000
049
906 1,
183 2,
195 3,
971 3,
708 3,
839 2,
000 2,

Run 9
3,003
6,672
7,380
7,404
4,759
1,621
720
140
604 1,768
823 4,514
770 6,004
506 6,836
616 5,454
045 3,048
689 1,729
419 814

Run 10
542
1,655
2,542
3,238
3,375
2,878
2,560
2,336
Run 3
175
632
1,126
1,513
1,676
1,513
1,284
1,130

Run 11
1,386
3,731
5,003
5,848
5,245
3,501
2,681
2,029
Run ft
4,493
8,160
7,508
7,879
3,139
244
61
15

Run 12
810
2,259
3,455
4,402
4,281
3,407
2,596
1,860
Run 5
1,859
4,706
6,181
6,949
5,418
2,897
1,573
707

Run 13
1,049
2,097
4,183
5,194
4,971
3,709
2,839
1,999
Run 6
542
1,655
2,543
3,238
3,375
2,877
2,560
2,336

Run 14
350
1,144
1,857
2,412
2,602
2,293
2,059
1,946
Run 7 Run 8
944 1,
2,645 3,
3,847 4,
4,824 5,
4,706 5,
3,636 3,
2,910 2,
2,192 1,

Run 15
969
2,677
3,783
4,655
4,453
3,428
2,893
2,405
143
131
466
491
162
723
725
780










Run 16

1,
1,
2,
2,
2,
1,
1,
310
029
691
202
373
100
869
762

-------
TABLE 10.  SUMMARY OF TOTAL MASS OF TCE (LB) IN THE GROUNDWATER
           SYSTEM FOR THE SENSITIVITY RUNS
Time
Step
1
2
3
4
5
6
7
8
Time
Step
1
2
3
4
5
6
7
8
Year
1968
1973
1978
1983
1988
1993
1998
2003
Year
1968
1973
1978
1983
1988
1993
1998
2003
Base
Case
6,552
11,247
14,665
17,071
13,075
9,611
6,833
4,848
Base
Case
6,552
11,247
14,665
17,071
13,075
9,611
6,833
4,848
Run 1
6,997
12,775
17,606
21,701
19,070
16,269
13,641
11,237
Run 9
4,598
5,527
5,748
5,945
2,161
784
125
0
Run 2
5,833
8,920
10,517
11,282
6,803
3,999
2,331
1,532
Run 10
7,059
13,005
18,064
22,427
19,827
17,193
14,694
12,373
Run 3
7,426
14,395
20,870
26,958
26,057
24,788
23,565
22,450
Run 11
6,215
10,085
12,683
14,436
9,966
6,709
4,089
2,075
Run., 4
3,
2,
2,
2,




Run
6,
12,
16,
19,
15,
12,
10,
8,
108
549
642
364
0
0
0
0
12
791
133
279
478
972
809
274
429
Run 5
5,742
8,637
10,057
10,709
6,266
3,613
2,101
1,409
Run 13
6,552
11,246
14,664
17,071
12,875
9,410
6,632
4,648
Run 6
7,059
13,005
18,063
22,426
20,026
17,393
14,894
12,573
Run 14
7,251
13,708
19,452
24,641
22,814
20,765
18,767
16,836
Run 7
6,657
11,613
15,367
18,144
14,413
11,021
8,172
5,995
Run 15
6,632
11,556
-15,374
18,320
14,842
11,658
8,826
6,436
Run 8
6,458
10,928
14,063
16,173
11,986
8,507
5,843
4,078
Run 16
7,292
13,864
19,774
25,173
23,774
21,918
20,104
18,354

-------
ro
                             TABLE  11.   TCE  CONCENTRATION (PPB) AT THE SOURCES IN 1983
                                        FOR  THE SENSITIVITY RUNS


             Location          Base Case      Run 1        Run 2         Run  3         Run 4        Run 5
Reaction Pond I
Reaction Pond III
Well 21
Mill Creek
212,000
139,000
557,000
39
315,000
232,000
700,000
27
121,000 •;
72,000
368,000
51
441,000
403,000
822,000
11
21,000
10,000
80,000
60
218,000
132,000
620,000
52
       Groundwater  Flux  Into
         Mill  Creek (cfs)             0.45         0.23         0.90         0.05         4.42         0.45
             Location          Base Case      Run 6        Run  7         Run  8         Run  9        Run 10
Reaction Pond I
Reaction Pond III
Well 21
Mill Creek
212,000
139,000
557,000
39
184,000
133,000
448,000
24
207,000
161,000
718,000
36
213,000
120,000
416,000
41
207,000
117,000
622,000
56
184,000
133,000
448,000
24
       Groundwater  Flux  into
         Mill  Creek (cfs)            0.45         0.45         0.44         0.46         0.45         0.45
       Location          Base Case     Run  11       Run 12       Run 13       Run 14       Run 15       Run 16
Reaction Pond I
Reaction Pond III
Well 21
Mill Creek
212,000
139,000
557,000
39
485,000
222,000
911,000
43
53,000
50,000
207,000
33
212,000
139,000
557,000
38
160,000
122,000
373,000
18
212,000
144,000
538,000
34
154,000
119,000
354,000
16
Groundwater Flux  into
  Mill Creek  (cfs)            0.45         0.45         0.45         0.45         0.45         0.42         0.45

-------
         TABLE  12.   PERCENT  CHANGE  OF  MASS  AND  CONCENTRATION  OF  TCE
                    FOR  THE  SENSITIVITY  RUNS  BASED  ON  1983  RESULTS
Run Mass
No. Exiting
1
2
3
4
5
6
7
8
9
10
11 ":"
12
13
14
15
16
-33
+32
-71
+52
+34
-38
-7
+6
+43
-38
+13
-15
0
-54
-10
-58
Total Mass
in System
+27
-34
+58
-86
-38
+31
+6
-5
-65
+31
-15
+14
0
+44
+7
+47
Reaction
Pond I
+49
-43
+108
-90
+3
+13
-2
0
-2
-13
+129
-75
0
-25
0
-27
Reaction
Pond III
+67
-48
+190
-93
-5
-4
+16
-13
-16
-4
+60
-64
0
-12
+4
-14
Well
21
+26
-34
+48
-86
+11
-20
+29
-25
+12
-20
+64
-63
0
-33
-3
-36
Mill
Creek
-31
+31
-72
+54
+33
-38
-8
+5
+44
-38
+10
-15
0
-54
-13
-58
Mill
Creek
Flux
-49
+100
-89
+882
0
0
-2
+2
0
0
0
0
0
0
-7
0
+ = increase
- = decrease
                                  43

-------
     A discussion of  the  results  shown  in the tables and  a  summary  of  all  the
sensitivity analysis model runs is presented below.

HYDRAULIC CONDUCTIVITY (RUNS 1 - 4)
     Four  sensitivity runs  were  made  where  the  base  case  horizontal   and
vertical  hydraulic  conductivities (K)  were changed  by factors  of  0.1, 0.5,
2.0,  and  10.0.    Decreasing  K  decreased  the groundwater  flux  through   the
system which:
1)   decreased the TCE mass loading to the creek and ditch;
2)   increased the total  mass of TCE in the system;
3)   increased the concentration of TCE at the three source  areas; and
4)   decreased the groundwater flux to Mill Creek.
Increasing  K  increased the  flux  through  the  system which  had  the opposite
effect of decreasing K as discussed above.
     Decreasing  the  K by  factors  of 2.0  and  10.0 increased  the  water table
elevation'over  the model  region  by  about  1.0 and  5.0 ft  (0.3  and  1.5  m),
respectively.   Increasing  the  K by a factor  of  10.0  decreased the potentials
by  about 0.5 ft (0.15  m),  whereas  increasing   K  by  a  factor  of  2.0 only
slightly decreased the potentials.

POROSITY (RUNS 5 AND 6)
     Two  sensitivity  runs were made  where the  base  case  effective porosity
was  changed  by   factors  of  0.5 and  2.0.    Decreasing  the  effective porosity
increased the groundwater velocity and  decreased  the  amount  of dilution.  The
effect of this change was to:
1)   increase the mass loading to  the creek and ditch;
2)   decrease the total mass in the system;
3)   increase the concentration at 2  of  the  3  source  areas  (Reaction Pond  III
     decreased slightly); and
4)   maintain the same groundwater flux to the creek.
Increasing the  porosity  had the opposite  effect  of decreasing  the porosity.
Changes in porosity had no impact  on  the groundwater potentials.
                                     44

-------
DISPERSIVITY (RUNS 11 AND 12)
     Two  sensitivity runs  were made  where  the base  case  longitudinal   and
transverse dispersivity  were changed by factors  of 0.2  and  5.0.   Decreasing
the dispersivity:
1)   increased the mass loading to the creek and ditch;
2)   decreased the total mass in the system;
3)   increased the concentration at the three source areas; and
4)   did not impact the groundwater flux to the creek.
Increasing the dispersivity had the opposite effect of decreasing it.
     Changes in dispersivity had no impact on the groundwater potentials.

COMBINATION RUNS (RUNS 13 - 16)
     Four  runs  were  made  in  which  either  two  parameters  were  changed
simultaneously or a single parameter was varied between layers.
     In Run  13  the  effective porosity was  doubled  and the retardation factor
was halved.   These parameters  both  impact  the  groundwater  velocity  and   the
changes offset  each  other,  therefore,  the results  are  virtually identical to
the base-case results.
     In  Run  14  the  effective  porosity  in  the  upper  and  lower units   was
increased  to  45%  and 40%,  respectively.    The  impact  of  this  increase is
similar to the impact of increasing porosity as discussed above.
     In Run  15  the  hydraulic conductivity in  the  lower  unit  (model  Layers 3
and 4) was halved  and the effective porosity  in  the  lower unit was increased
to 25%.  The impact of these changes was to:
1)   decrease the mass loading to Mill  Creek;
2)   increase the total  mass in the system;  and
3)   decrease the groundwater flux to Mill  Creek.
The change in  concentration  and flux was small.   The  parameter changes made
only in Unit 2  (gravelly  sand)  changed  the  model results slightly (Table  12),
indicating that  some  contaminant is migrating through  the lower  unit  of   the
model  to Mill Creek.
     In Run  16  the  effective porosity in all  layers  was  increased to 25%  and
the retardation was increased to 8.0.  This  case was  intended to better match
the parameters  used  by CH2M HILL  in their  feasibility study.   The effect of
these changes was to:
                                     46

-------
1)   decrease the mass loading to the creek and ditch;
2)   increase the total mass in the system;
3)   decrease the concentrations at the three source areas; and
4)   maintain the base case flux rate to Mill Creek.
     The  groundwater  potential   surfaces  were  virtually  unchanged  by  the
changes made in the combination runs.

SENSITIVITY ANALYSIS SUMMARY
     Only  three  of  the  sensitivity  runs  showed  significant  changes  in  the
potential  surfaces:   those where  the K was decreased  by factors of  2.0  and
10.0, and  where the  recharge was  doubled.   The potential  surfaces for the top
of Layer 1 for the case where K was decreased  by a  factor of  10.0 is shown in
Figure 19.  The potential  surface  for the  case where the K decreased a factor
of 2.0 and for the case where recharge  was doubled  are  virtually the same and
are shown  in Figure 20.
     Chan-gjng  the  hydraulic  conductivities  changes  the  groundwater  flux
through the system as  well  as  the velocity,  whereas changing  the porosity or
retardation factor  only impacts  the  velocity of  the  contaminant.   Changing
recharge has a small  impact on  groundwater flux, whereas  dispersivity  has no
impact on  flux.    Therefore, the  groundwater flux  to Mill  Creek  is  almost
completely  controlled  by  the  hydraulic conductivity.    Since  the  base  case
model  accurately  predicts  the  measured  value  of  groundwater  flux to  Mill
Creek over  the model  region the hydraulic conductivity  used  in  the  model  is
probably reasonable.
     The mass  of  TCE  exiting  the system  is  about equally controlled  by  the
hydraulic conductivity, the porosity, and the  retardation  factor.   If any one
of these parameters  is changed  by  a  factor  of  two,  the results are altered by
a factor of about plus or minus 35%.  This degree  of  sensitivity supports  the
values currently used in the model.
     Dispersivity has  the  greatest  influence on  concentration  at  the  three
source areas.   Dispersivity  is  also  the parameter  in which we  have  the least
confidence.  The  other parameter which  influences  the  source  concentrations,
is  the  hydraulic  conductivity.     Changes  in  porosity,  retardation,  and
recharge have  little influence  on the  concentrations  at  the  sources.   Since
                                     47

-------
-
00
                    WP 8EN HUN 14 TOT CTOP LRYEB 1)
     Figure  19.   Model-Predicted Top of Layer 1 Potential
                  Surface  for the Sensitivity Run with One-
                  Tenth  the Hydraulic Conductivity
                                                                             VP 8EHHUH 2PCT
                                LflYER 1)
Figure 20.
Model-Predicted Top  of  Layer 1 Potential
Surface for the Sensitivity Run with One-
Half the Hydraulic Conductivity and the
Sensitivity Run with Double the Recharge

-------
dispersivity>  is  the  primary factor controlling  source  concentration,  and the
model-predicted  concentrations  match   the  measured  values,  the  base  case
dispersivity is probably reasonable.
     The  limited sensitivity  analysis  illustrated  how  every parameter  has
some influence on the groundwater  potential,  the flow to Mill Creek,  the TCE
flux to Mill Creek, and/or  the  total mass  of  TCE in the system.   Changing any
of the parameters will  affect the current calibration.
                                     49

-------
                                  SECTION 7
                 ASSESSMENT  OF  REMEDIAL ACTION ALTERNATIVES

     Remedial action  alternatives for  the  Western Processing  Site have  been
proposed by  CH2M HILL and the  PRPs  for site restoration.  The  purpose  of the
remedial  action  assessment  performed  in  this  study  was to  simulate  these
proposed alternatives  in  the final  calibrated model  in order to  evaluate and
compare their effectiveness.
     The  base  case  consisted  of running  the  final  calibrated  model  for 20
years,  from  1963  through  1983.   In  order to  simulate  the  remedial  action
alternatives, certain model  parameters  were adjusted  in 1983  and the model
was run for an additional 50 years to  the year  2033.   When the pump and treat
alternative was simulated, water  was  withdrawn  for the first 30 years  (1983 -
2013) in  the CH2M  HILL cases and for the  first 5 years  (1983  - 1988} in  the
PRPs  cases.   The  model  results  during the prediction  period were  used to
determine  which  action  would   be most  effective in  reducing  the  level of
contamination at the site.
     The following section describes  the remedial  action alternatives and  how
they  were  simulated   in  the  model,   and   provides   the  results  of   each
simulation.  The  alternatives   are discussed in two sections,  those proposed
by CH2M HILL and those  derived  from  the alternative proposed by the PRPs.  A
summary of the  results is also provided  in  Section 2.

CH2M HILL PROPOSED REMEDIAL ACTIONS
     CH2M HILL requested  that   three  basic  simulations be  performed with  the
model:
1)   no-action;
2)   source removal combined  with pump and  treat; and
3)   cap combined with pump and treat.
                                      50

-------
The pump and  treat  cases  were  run  with two different pumping rates.   In  order
to better  understand  the  results,  a  simulation  involving  source removal  only
was also performed.  These three cases and variations are- discussed below.
No Action  (Extension of the Base Case)
     The no-action  simulation  consisted  of running  the  base case'model (final
calibrated model presented earlier) 50 years  into  the future.   The purpose  of
this simulation was to  determine the  predicted extent of TCE contamination  if
no remedial measures  are  implemented  at  the site.   This  simulation served  as
a  benchmark  against which  the CH2M  HILL  proposed  remedial actions  could  be
compared.
     The potential  surfaces for the  no-action case  are  the same  as  for the
base case  (Figures  8  to 12).   TCE  concentration contours at the  top  of Layer
1  in  the years 1993,  2003,  and 2033  (10,  20, and  50 years  into  the future)
are shown  in  Figures  21 through 23,  respectively.   The total  mass of TCE  in
the flow system,  and  the  total mass  of  TCE discharging  to  Mill  Creek and the
drainage ditch over 5 year intervals are shown in Table  13.
     Table  13  shows  that  of   the  31,699  Ib  (14,378  Kg)  that entered  the
groundwater  flow  system  between  1963  and 2003,  15%  and  4%  remain  in  the
system  in  the  years  2003.  and   2033  (20   and  50  years  in  the  future),
respectively.   Of the  17,071  Ib  (7,743  Kg)  remaining  in  the flow  system  in
1983,  about  88% and 99% discharges to Mill  Creek  and the  ditch by the years
2003  and  2033, respectively.   As  for  the base  case,  about 97%  of  the  TCE
exiting  the  system discharges   to Mill  Creek  and the remaining  3%  discharges
to the ditch.
Source Removal
     The  source  removal   action  assumes  that  6  ft  (2   m)  of  unsaturated
material is excavated  and  removed  from  the site.   The  effect of  this action
would  be to  remove the source  of  TCE that leaches  from  the  unsaturated  zone
into the saturated flow system.
     Source remova" was simulated  in  the  model  by  eliminating  further input
of TCE  beyond  1983.    Because Reaction  Pond  I  is  the   only  source  that
contributes TCE to  the  system  after 1983 (see Appendix  A),  source removal was
accomplished in the model  by eliminating this source after 1983.
                                      51

-------
               UP CF RUN 20 CONC (TOP UTTER 1)
Figure 21.  Model-Predicted 1993 Top of Layer  1  TCE
            Concentration Contours for the No-Action
            Slmula11 on
              UP CF RUN 20 CONC OOP LAYER 1)
Figure 22.  Model-Predicted 2003 Top of  Layer 1 TCE
            Concentration Contours for the  No-Action
            S f milla t ion

-------
en
CO
                                            VP CF HUN 20 CONC CTOP LflYCT 1J
                             Figure 23.  Model-Predicted 2033 Top of Layer  1  TCE
                                         Concentration Contours for the No-Action
                                         Simulation

-------
   TABLE 13.  MODEL-PREDICTED  DISTRIBUTION  OF  TCE  FOR
              THE  CH2M  HILL  NO-ACTION  SIMULATION
Time
Step
1*
2*
3*
4*
5
6
7
8
9
10
11
12
13
14
Total
End
Year
1968
1973
1978
1983
1988
1993
1998
2003
2008
2013
2018
2023
2028
2033
TCE
Inflow (Ib)
7,601
7,601
7,601
7,601
975
244
61
15
0
0
0
0
0
0
31,699
TCE
Outflow (Ib)
1,049
2,906
4,183
5,195
4,971
3,708
2,839
2,000
1,288
824
531
379
290
240
30,403
Total TCE in
System (Ib)
6,552
11,247
14,665
17,071
13,075
9,611
6,833
4,848
3,560
2,736
2,205
1,826
1,536
1,296
*Base case

Inflow is mass entering the groundwater flow system and
outflow is the mass exiting the system.
                          54

-------
     The  groundwater  potential   surfaces  for  the  source  removal  case  are
identical to those for the base  case  (Figures  8  to 12) because the  soil  below
the water table  was  undisturbed.   The source removal  action removes  so  little
mass  of  TCE from  the system, that  the TCE  concentration contour  plots  are
essentially the  same as those for the no-action case (Figure 21 through  23).
     The  total  mass of  TCE  in  the  flow  system,  and  the  total  mass  of  TCE
discharging to Mill  Creek  and  the ditch over five-year intervals are shown  in
Table  14.   The source removal action  removes  about 1,300  lb  (588  Kg)  of  TCE
from  the  groundwater flow  system.  This action  results  in a slight  reduction
in  the mass  of  TCE  remaining  in  the  system  and exiting  to the  creek  and
ditch.
     The  source  removal   action  is  not  effective  in cleaning  up  the  site
because most of  the TCE has entered the saturated  flow system by 1983.
Source Removal Combined with Pump and Treat (100 gpm)
     This  remedial   action  assumes  that the  entire  site  is  excavated  to  a
depth  of  6  ft  (2  m)   (source  removal)  followed  by  the  withdrawal  of
contaminated groundwater (pump and treat).
     The  source   removal  portion of  this   alternative was  simulated  in  the
model  as  discussed  above.   Pumping and  treatment  was  simulated  in  the model
by withdrawing water from  a system of  38  wells  spaced across  the area of  the
site.   Nodes  of  the finite element  grid  (Figure  6)  were used  to represent
wells, and  the  wells were  situated in  3 rows  running  north -  south along the
east,  central, and west  portions  of  the site.   It was assumed  that the wells
were  drilled  to  a  depth  of 30  ft  below the  water  table and  that  the  total
withdrawal rate  was  100  gpm  (379  1/min)  (2.6 gpm  (10  1/min) from each well).
Pumping  was  simulated  for  the   first  30  years  of   the  50  year  prediction
period.
     The  potential  surface for the top of Layer  1 for this case  is shown in
Figure 24.  The  average  drawdown  over the  site due to pumping  was  about 4 ft
(1.2 m).
     The total mass  of TCE  in  the flow  system, and the total  mass discharging
to Mill  Creek  and   the  pumping  wells  over  5-year  intervals  are  shown in
Table 15.   The model  results  show that 91% of the  TCE   is removed  from  the
flow  system  in  the  first  5 years,  and that  all   the TCE  is  removed  in 15

                                      55

-------
TABLE 14.  MODEL-PREDICTED DISTRIBUTION OF TCE FOR
           THE CH2M HILL SOURCE REMOVAL ACTION
Time
Step
1 - 4
5
6
7
8
9
10
11
12
13
14-...
Total
Total
End
Year
(Same
1988
1993
1998
2003
2008
2013
2018
2023
2028
2033
(1983 - 2033)
(1963 - 2033)
TCE
Inflow (Ib)
as for Base
0
0
0
0
0
0
0
0
0
0
0

TCE
Outflow (Ib)
Case)
4,929
3,554
2,587
1,734
1,054
623
355
217
135
99
15,287
28,620
Total TCE in
System (Ib)

12,141
8,587
6,000
4,266
3,212
2,589
2,234
2,017
1,882
' 1,783

                     56

-------
en
                           Figure 24.   Predicted Top of Layer 1 Potential  Surface
                                       for CH2M  HILL  Source  Removal  and  Pump  and
                                       Treat  (100  gpm)  Remedial Action

-------
years.   After  1983,  87%  of the  TCE exiting  the  system  is  removed by  the
pumping wells; the remaining 13% discharges  to Mill  Creek and the  ditch.

             TABLE  15.   MODEL-PREDICTED DISTRIBUTION OF TCE FOR THE
                        CH2M HILL  SOURCE  REMOVAL COMBINED WITH
                        PUMP AND TREAT (100 GPM) REMEDIAL ACTION

                                                               Total  TCE in
                                                               System  (Ib)
                                                                  1,512
                                                                    593
                                                                      0
                                                                      0
                                                                      0
Time
Step
1 - 4
5
6
7
8
14
Total
End
TCE
. TCE Outflow Mb)
Year Inflow Mb) Mill Creek
(Same
1988
1993
1998
2003
2033
as Base Case)
0
0
0
0
0
0

1,880
254
59
0
0
2,193
Pumping Wells

13,679
663
534
0
0
14,876
Source Removal Combined with Pump and Treat  (200 gpm)
     This  remedial  action  is similar to  the previous  action  except that  the
38  wells  were pumped  at   a  total   rate  of  200  gpm   (758 1/min)  instead of
100 gpm  (379  1/min).
     The  potential  surface for the  top of  Layer 1 for this  case is shown in
Figure 25.   The average drawdown over  the  site due to pumping was about  8.5
ft (2.6 m).
     The  total mass  of  TCE in  the flow system, and the total mass  discharging
to  Mill   Creek  and  the   pumping  wells  at  5-year  intervals   are  shown  in
Table 16.  The model results  show that  all  the TCE is  removed  from  the  system
in  the  first  5  years.   After  1983,  97% of  the  TCE  exiting  the  system is
removed  by the pumping wells;  the  remaining 3% discharges  to  Mill Creek  and
the ditch.
Cap Combined  with Pump  and Treat  (100 gpm)
     The   capping  action  assumes  that  the   site  is covered  with   a   low
permeability  material in order  to eliminate the infiltration of  water through
the  unsaturated  zone  and  the  resulting  leaching  of   TCE  into the  saturated

                                      58

-------
en
                                                     POT hor una n
                           Figure  25.   Predicted  Top  of  Layer 1  Potential  Surface
                                       for CH2M HILL Source Removal and Pump and
                                       Treat (200 gpm) Remedial Action

-------
Time
Step
1 - 4
5
6
14
Total
End
TCE
Year Inflow (Ib) Mil
(Same
1988
1993
2033
as Base Case)
0
0
0
0
TCE
Outflow (Ib)
1 Creek Pumping Wells

545
0
0
545

16,526
0
0
16,526
flow  system.   The  capping  action  would be  followed by  the installation  of
wells to pump and treat contaminated groundwater.

             TABLE  16.   MODEL-PREDICTED DISTRIBUTION OF TCE FOR THE
                        CH2M HILL SOURCE REMOVAL COMBINED WITH
                        PUMP AND TREAT  (200  GPM) REMEDIAL ACTION

                                                               Total  TCE in
                                                               System (Ib)
                                                                    0
                                                                    0
                                                                    0
     The  cap  was simulated  in  the model by  reducing the  recharge rate  over
the  site "'to 0.8  in./yr  (2 cm/yr) or  one-tenth  the estimated annual  average.
The  pump  and  treat  action was  simulated as  discussed  in  the  previous  case
where the pumping rate was 100 gpm (379  1/min).
     The  potential  surface for the top  of  Layer 1 for th-is  case is shown  in
Figure  26.    The  average  drawdown  over the  site due  to  pumping  was  about
4.5 ft  (1.4 m).
     The  total  mass of  TCE in  the  flow system,  the mass  entering from  the
unsaturated  zone, and  the mass  discharging   to  Mill Creek and  the  pumping
wells at  5-year  intervals  are  shown  in Table 17.  The model  results  show that
92% of  the  TCE  is removed  from the system in  the  first 5  years, and that  all
of  the  TCE  is removed  in  15 years.   After 1983, 88% of  the TCE exiting  the
system  is removed by the pumping wells;  the  remaining  12% discharges  to Mill
Creek and the ditch.
Cap Combined with Pump and Treat  (200  gpm]
     This remedial  action  is  similar  to the  previous  action except that  the
38  wells  were  pumped   at  a  total rate of  200  gpm  (758 1/min)  instead  of
100 gpm (379 1/min).
     The  potential  surface for the top  of  Layer 1 for this case is shown  in
Figure  27.  The  average  drawdown over the site due to pumping  was about 9  ft
(2.7 m).
                                      60

-------
                VP CF NUN M TOT CTDf UHBI 11
   wป CF iua
Figure 26.  Model-Predicted Top of Layer  1  Potential
            Surface for the CH2M HILL Cap and  Pump
            and Treat (100 gpm) Remedial  Action
Figure 27.
Model-Predicted Top of Layer  1 Potential
Surface for the CH2M HILL Cap and  Pump
and Treat (200 gpm) Remedial  Action

-------
5
6
7
8
14
Total
1988
1993
1998
2003
2033

98
25
7
0
0
130
1,822
220
47
0
0
2,089
14,040
528
530
0
0
15,105
        TABLE 17.  MODEL-PREDICTED DISTRIBUTION OF TCE FOR THE
                   CH2M HILL CAP COMBINED WITH PUMP AND TREAT  (100 GPM)
                   REMEDIAL ACTION
            End        TCE            TCE Outflow fib)        Total TCE  in
            Year   Inflow (1b)   Mill CreekPumping WeTTs    System (1b)
    1-4      (Same as Base Case)
                                                                  1,307
                                                                    584
                                                                      0
                                                                      0
                                                                      0
     The total  mass  of TCE  in the  flow system,  the  mass entering  from the
unsaturated zone, the mass discharging  to Mill  Creek,  and the mass removed by
pumping wells  at  5-year  intervals are shown  in  Table  18.  The  model results
show that  virtually  all  of the TCE  is removed from  the  system in the first 5
years.   The  only TCE remaining  in  the system  after  5  years   is  the  small
amount  that  enters  from  the  unsaturated zone.   After  1983,  97%  of the TCE
exiting  the  system  is   removed   by the  pumping  wells;  the   remaining  3/ป
discharges to Mill Creek  and the ditch.

        TABLE 18.  MODEL-PREDICTED DISTRIBUTION OF TCE FOR THE
                   CH2M HILL CAP COMBINED WITH PUMP AND TREAT (200 GPM)
                   REMEDIAL ACTION
                                                              Total TCE in
                                                               System  (Ib)
                                                                    0
                                                                    0
                                                                    0
                                                                    0
                                                                    0
Time
Step
1 - 4
5
6
7
14
Total
End
TCE
Year Inflow (Ib) Mi
(Same
1988
1993
1998
2033
as Base Case)
97
25
0
0
0
TCE Outflow (Ib)
11 Creek

519
0
0
0
519
Pumping Wells

16,649
25
0
0
16,674
                                       62

-------
PRPS PROPOSED REMEDIAL ACTION
     The PRPs  proposed remedial action  is  discussed  in  detail  in the  report
by Landau Associates and Dames  and Moore  (1984).   In summary, they proposed  a
combination of  source  removal,  a  slurry wall  around the site, and pumping  and
treatment.  In  order  to  more thoroughly analyze  their  proposal  in the  model,
a series of six model runs were performed.
     The first  model   run  was a no-action  simulation  to  provide  a benchmark
against which  the  remedial  action runs could  be  compared.   The second model
run  simulated   a  slurry  wall  combined  with  source removal.   The  third   and
fourth model  runs  simulated  the  combination  of  slurry wall,  source removal,
and  pump   and  treat,   at  total  pumping rates  of  100  and  200  gpm  (379   and
758 1/min).  To evaluate the effectiveness  of the  slurry  wall,  the  fifth  and
sixth  model  runs  were similar  to  the  third  and  fourth runs  except  that   the
slurry wall was removed.
No Action  (Extension of the Base Case)
     The PRPs   no-action simulation  is  basically the  same  as  the  no-action
case for the  CH2M  HILL simulations, except that  the finite  element  grid   was
modified for  the  PRPs simulations  to  contain elements  which  represent   the
slurry wall.   In the  no-action case the slurry  wall   elements  were  assigned
the  same   permeability as  surrounding  elements.   For  cases  involving   the
slurry wall, these elements  were assigned a  permeability of  2.8 x  10   ft/day
(10    cm/sec).   The purpose  of this simulation  was  to verify  that  the same
model  solution  was  achieved  for  the  finite element  grid   with  the  normal
permeability  slurry  wall  elements  as  for  the grid  without  the  slurry wall
elements.
     The potential  surfaces  for the  PRPs  no-action case  are  the  same  as   for
the original base case (Figures 8 to  12).  The  TCE  concentration contours  are
nearly  identical   to   those  for  the CH2M  HILL  no-action  case  (Figures  21
through 23).
     The total  mass  of TCE in the flow system, the mass entering  the system,
and the mass  discharging to  Mill  Creek and  the  ditch over  5-year  intervals
are shown  in Table 19.   The  numbers  in this  table are  different than those in
Table  13  (CH2M  HILL no-action  case)  because  the areas of the  Reaction Pond
III  and  Well   21  sources  are slightly  smaller  (the  slurry  wall  elements

                                      63

-------
    TABLE  19.   MODEL-PREDICTED  DISTRIBUTION  OF  TCE  FOR
               THE  PRPS  NO-ACTION  SIMULATION
Time
Step
1*
2*
3*
4*
5
6
7
8
9
10
r:.. 11
12
13
14
Total
End
Year
1968
1973
1978
1983
1988
1993
1998
2003
2008
2013
2018
2023
2028
2033
TCE
Inflow (Ib)
7,034
7,034
7,034
7,034
975
244
61
15
0
0
0
0
0
0
29,431
TCE
Outflow (Ib)
919
2,612
3,815
4,757
4,619
3,497
2,688
1,909
1,231
788
505
358
270
223
28,191
Total TCE in
System (Ib)
6,115
10,537
13,756
16,033
12,389
9.136
6,509
4,615
3,384
2,596
2,091
1,733
1,463
1,240
*Base case.
                          64

-------
removed a  3  ft  (1  m)  strip along  western  edge).   After  the first four  time
steps,  when  sources  around Reaction  Pond  III  and  Well  21  are  no  longer
active,  the   amount of  TCE  entering  the  system from  Reaction  Pond  I  is
Identical  to the base  case, thus,  it  appears  that adding  the extra  elements
does not change  the model  results  significantly.   As  a result, all PRPs cases
were simulated with  the  slightly reduced  area of the  source around  Reaction
Pond III and Well 21.
     The  results of  the  PRPs  no-action  run  served  as  a  benchmark  against
which  the  PRPs remedial  action  simulations could be  compared.    The  results
showed  that  28,136  Ib  (12,730  Kg)  of  TCE  entered  the  groundwater flow system
between 1963  and 1983,  and  that  an additional 1,295  Ib  (586 Kg)  leached  in
from  the   unsaturated  zone  over the  next  20 years.    By  1983,   12,103  Ib
(5,476 Kg) were  discharged  to either Mill  Creek (97%)  or the ditch (3%),  and
16,033 Ib  (7,254 Kg) remained in the system.
Source Removal and Slurry Wall
     This "model   run  simulated  a  combined source  removal  and  slurry  wall
remedial  action.   The  source  removal  action assumes  that  the  unsaturated
material contaminated with  TCE  is excavated and removed from the  site.   This
action  was simulated in the model  by eliminating all  input of  TCE   to  the
groundwater flow system after 1983.
     The  slurry  wall  was  simulated in the  model  as  a series  of  3 ft  (1 m)
wide elements  around the perimeter  of the  site (Figure  4).  The  slurry wall
elements were  assigned a permeability  of 2.8  x 10  ft/day (10    cm/sec)  and
the wall  extended to  a  depth  of  50  ft  below the surface (40  ft  below  the
water  table).   The  intent of  the  wall  was to  contain the  contamination on
site,   reduce  the  lateral  flow  to  the  wells,  and  increase  upward   vertical
flow.
     The potential  surfaces for this  case  are essentially  the  same as those
for the  base case  (Figure 8 -  12).   The  potential  on  site averages  about
0.6 ft  (0.2  m)  higher  than the base case  potential  due to  the  impact of  the
slurry wall.
     The distribution of TCE  in  the flow system  at 5-year  intervals  is shown
in Table 20.   Comparing  these  results to  the  results  for  the  no-action  case
                                     65

-------
TABLE 20.  MODEL-PREDICTED DISTRIBUTION OF TCE FOR THE PRPS
           SOURCE REMOVAL WITH A SLURRY WALL REMEDIAL ACTION
Time
Step
1 - 4
5
6
7
8
9
10
11
12
13
14
Total
End
Year
(Same
1988
1993
1998
2003
2008
2013
2018
2023
2028
2033
TCE
Inflow (Ib)
as for Base
0
0
0
0
0
0
0
0
0
0
0
TCE
Outflow (Ib)
Case)
3,645
2,513
2,232
1,712
1,265
864
608
457
368
309
13,973
Total TCE in
System (Ib)

12,388
9,875
7,643
5,931
4,666
3,802
3,194
2,737
2,369
2,060
                           66

-------
shows that  the  slurry wall prevents the  TCE  from migrating off  site,  thereby
increasing  the  mass  of TCE  on site  and reducing  the  TCE  loading  to  Mill
Creek.   All TCE  exiting the  system  either  goes  to Mill  Creek, (97%)  or  the
ditch (3%).
Source Removal, Slurry Wall, Pump and Treat (100 gpm)
     This model  run  simulated the remedial action  proposed by the  PRPs.    It
consisted  of a  combination  of source  removal,  pump  and  treat at  100   gpm
(379 1/min), and a slurry wall  around the perimeter of the  site.
     The  source  removal and  slurry wall  actions  were simulated  in the  same
manner  as  discussed  in the  previous  case.    As  stated  earlier,  for   this
remedial  action  involving  pumping from  shallow  wells,  the  purpose  of   the
slurry  wall was  not  only  to  contain  the contamination,  but  to  reduce   the
lateral  flow of water  from  off  site  to the wells,  and  induce  upward   flow
through  the highly contaminated  near-surface materials  on  site.    Pump   and
treat was  simulated  in  the same manner as in  the  CH2M  HILL cases.   Water  was
withdrawn from a network of 38 wells evenly  distributed  over the site.    The
wells  were  drilled  to  30 ft  (9  m)   below  the  water  table  and  the  total
withdrawal  rate from all wells  was 100  gpm  (379  1/min).   The major difference
between  the CH2M HILL and PRPs cases  is  that in the PRPs  cases,  pumping   was
simulated, for the  first 5  years of the 50-year  simulation period  instead of
for the first 30 years as in the CH2M HILL simulations.
     The potential surface  for  the top of Layer  1  for  this case  is  shown in
Figure 28.   The  average drawdown over  the  site  due to pumping is  about  6 ft
(1.8 m).
     The distribution  of TCE  in  the flow  system for  this case  is  shown in
Table 21.   The results  show that  84%  of the TCE  presently  in the flow system
(1983)  will  be removed  after  5 years of  pumping.   All  of  the  TCE  is  removed
from the system  in 15 years  (after  the first 5  years  it  discharges  to  Mill
Creek because  pumping  is  stopped).    Of  the  TCE  exiting  the  system  in   the
first 5 years, 84% is withdrawn by the  pumping  wells,  and the  remaining   16%
goes to Mill Creek  or  the ditch.
                                     67

-------
en
oo
                                             UP CF HUN 32 POT (TOP UtYEH 11
                              Figure  20.   Model-Predicted Top of Layer  1  Potential

                                          Surface for the PRPs Slurry Wall  and

                                          Pump and Treat (100 gpm) Remedial  Action

-------
             TABLE  21.   MODEL-PREDICTED  DISTRIBUTION  OF  TCE FOR THE
                        PRPS  SOURCE  REMOVAL,  SLURRY WALL,  AND
                        PUMP  AND  TREAT  (100  GPM)  REMEDIAL  ACTION
Time
Step
1 - 4
5
6
7
8
14
Total
End
TCE
TCE Outflow (Ib)
Year Inflow (Ib) Mill Creek Pumping Wells
(Same
1988
1993
1998
2003
2033
as Base Case)
0
0
0
0
0
0

2,153
1,508
1,069
• 0 '
0
4,730

11,303
0
0
0
0
11,303
Total TCE in
System (Ib)

2,577
1,069
0
0
0
Source Removal, Slurry Wall, Pump and Treat (200 gpm)
     This remedial  action  is similar to  the  previous action except  that the
38 wells 'were pumped at a total rate of 200 gpm  (758  1/min)  instead  of 100
gpm (379  1/min).  The 200  gpm pumping rate is twice the rate proposed  by the
PRPs.
     The  potential  surface for the top of  Layer  1 for this case  is  shown in
Figure 29.  The  average  drawdown over the site due to pumping  is  about 12 ft
(3.6 m).
     The  distribution  of TCE  in the flow  system for this  case  is  shown in
Table 22.  The results show that all  of  the TCE exits the flow system  in the
first 5  years.   Of the  total  leaving the system, 97%  is  withdrawn  by the
pumping wells and 3% discharges to the creek or ditch.

Source Removal, Pump and  Treat (100 gpm),  No Slurry Wall
     In order  to  evaluate  the benefit of  a slurry wall, this  case simulated
source removal and  pump  and  treat  (total  of 100 gpm) without the slurry wall.
All parameters were identical  to the previous  case which simulated  the PRฐs
proposed  remedial  action,  except  that  this  case  did not include  the  slurry
wall.
                                      69

-------
                UP CF HUM 33 POT (TOP LflYEB 1)
Figure 29.
Model-Predicted Top  of  Layer 1 Potential
Surface for the PRPs Source  Removal, Pump
and Treat (200 gpm), and  Slurry Wall

-------
            TABLE  22.  MODEL-PREDICTED  DISTRIBUTION  OF  TCE  FOR THE
                       PRPS  SOURCE  REMOVAL,  SLURRY WALL,  AND
                       PUMP  AND TREAT  (200 GPM)  REMEDIAL  ACTION
End
TCE
Year Inflow (Ib) Mil
(Same
1988
1993
2033
as Base Case)
0
0
0
0
TCE
Outflow (Ib)
1 Creek Pumpinq Wells

481
0
0
481

15,552
0
0
15,552
                                                              Total TCE  in
                                                               System (1b)
                                                                    0
                                                                    0
                                                                    0
     The potential surface  for  the top of Layer  1  for this case  is  shown in
Figure 30.   The  average drawdown over the site due  to pumping  was about 4 ft
(1.2 m).   The drawdown was greater  in  the case  with  the slurry  wall  (6 ft)
because the wall  cut off much of the lateral  flow.
     The--;distribution  of  TCE in  the flow system for  this  case  is  shown in
Table 23.  The results  show that  92% of the  TCE in the system in 1983 will be
removed after 5  years  of pumping  and  all  the TCE  exits the system  after 20
years  (after  the first 5 years  it goes to  Mill  Creek because  the  pumps are
turned off).   Of  the  TCE  exiting  the  system  in the  first  5  years,  89% is
withdrawn by the pumping wells  and  the  remaining  11% discharged to Mill Creek
or the ditch.

          TABLE 23.  MODEL-PREDICTED DISTRIBUTION  OF TCE FOR THE
                     PRPS SOURCE REMOVAL AND  PUMP  AND TREAT (100 GPM)
                     REMEDIAL ACTION (NO SLURRY WALL)

                                                              Total TCE in
                                                              System  (1b)
                                                                   ,276
                                                                    845
                                                                    276
                                                                      0
                                                                      0
Time
Step
1 - 4
5
6
7
8
14
Total
End
TCE
TCE Outflow (Ib)
Year Inflow (Ib) Mill Creek
(Same
1988
1993
1998
2003
2033
as Base Case)
0
0
0
0
0
0

1,623
431
569
276
0
2,899
Pumpinq Wells

13,134
0
0
0
0
13,134
                                      71

-------
ro
                                              CF HLJNM POT flOP UWCT
                            Figure 30.  Predicted Top of  Layer  1  Potential  Surface
                                        for the PRPs Pump and Treat  (100 gpm)
                                        Remedial Action with no Slurry Wall

-------
Source Removal, Pump and Treat (200 qpm), No Slurry Wall
     A simulation was  made with source  removal  and  pump and  treat (total  of
200 gpm) without the slurry  wall.   All  model  parameters were  identical  to  the
PRPs proposed  remedial  action with the  200 gpm pumping rate  except that  the
slurry wall was eliminated.
     The potential  surface for the top  of  Layer 1 for this  case  is shown  in
Figure 31.   The average drawdown over  the  site due  to pumping  was about  8.5
ft  (2.4  m).   The drawdown was greater  in  the case with the  slurry wall  (12
ft) because the wall cut off much of the lateral flow.
     The distribution  of TCE  in  the flow  system for  this  case  is  shown  in
Table 24.   The results  show that all  the  TCE exits  the  flow  system  in  the
first 5  years.   Of the  total exiting  the system,  97%  is  withdrawn  by  the
pumping wells and 3% discharges to Mill  Creek  or the ditch.

          TABLE 24.   MODEL-PREDICTED DISTRIBUTION OF  TCE FOR  THE
         "'           PRPS SOURCE REMOVAL AND PUMP AND TREAT (200 GPM)
                     REMEDIAL ACTION (NO SLURRY WALL)

             End        TCE            TCE Outflow (Ib)         Total TCE  in
            Year   Inflow (Ib)   Mill  Creek   Pumping Wells    System  (Ib)
               (Same as Base Case)
            1988        0           481           15,552             0
            1993        00                00
            2033        0.             0          	0             0
                        0           481           15,552
                                     73

-------
               UP CF RUN aO POT (TOP UTTER 1)
Figure 31.  Model-Predicted Top of Layer  1  Potential
            Surface  for the PRPs Pump  and Treat (200
            gpin)  Remedial  Action with  no  Slurry Wall

-------
                                  SECTION 8

                                   SUMMARY

     A  summary of  the base  case  and .remedial  action simulation  results  is
discussed in this section.

BASE CASE
     TCE  was  introduced to  the groundwater flow  system  in the model  over a
period  of 20 years  (1963  -  1983)  at  three  source  locations:    1)  Reaction
Pond I;  2)  Reaction  Pond  III;  and  3)  around  Well  21.    Over  this  20  year
period  the model  predicted that 30,400  Ib  (13,789  Kg)  of TCE was  spilled  at
the  site.   Over  the same period,  about 13,300  Ib (6,033 Kg)  discharged  to
Mill  Creek  or  the  railroad  drainage  ditch,  with  17,100  Ib  (7,756  Kg)
remaining  in  the flow  system -in 1983.   This  number compares  well  with the
estimated mass  of TCE in the  flow system of  18,000 Ib (8,165  Kg)  calculated
independently  by  CH2M HILL .(1985) and  Landau Associates  and  Dames  and Moore
(1984)  based on the 1982  through 1984  chemistry data for the  site.   Of the
30,400  Ib  (13,789   Kg)  of  TCE  that  was  spilled at  the  site,  the  model
indicated that  13%,  37%,  and 50% originated  in  the area  of Reaction  Pond  I,
Reaction Pond III, and Well 21, respectively.
     A summary of the results for both  the  CH2M  HILL  and  PRPs remedial action
simulations is  shown  in  Table 1 (Section 2).  A comparison of the total  mass
of TCE  remaining  in the groundwater  flow system for  the CH2M  HILL  and  PRPs
remedial action cases is shown in Figure la and Ib, respectively (Section 2).
     For both the CH2M HILL  and the  PRPs simulations, the base  and no-action
case  results  in  Figures  la  and  Ib  are essentially  the same.    The results
would  be  identical  except  that the  slurry  wall elements  in the  PRPs  cases
reduced the  size of  the  Reaction Pond  III and  Well  21   source  areas in the
model, thereby slightly reducing the mass loading of TCE  at these areas.
                                      75

-------
     The model results  show  that Mill  Creek has been  and  will  continue to be,
if  no  remedial   actions   are  taken,  the  primary  discharge  point  for  TCE
migrating from  the Western  Processing  Site.   In  the  base case  (1963  - 1983)
and the no-action case  (1983 - 2033),  about 97% of the TCE exiting  the system
discharges  to Mill  Creek,  and  the remaining  3% discharges  to the  drainage
ditch along the eastern boundary  of  the  site.   According  to the  model,  no TCE
enters  the  regional  groundwater  flow  system  which  flows toward  the  Green
River; all  the TCE  remains in  the local flow  system controlled  by  Mill  Creek
and the  drainage  ditch.   By 1983,  a  little  less than  half  (44%)   of  the  TCE
that was estimated  to  have entered the  flow system  during site  operation  had
exited to Mill  Creek and  the ditch.   For  the  no-action simulation, the  model
predicted that 89%  and 96% of the  total mass  of  TCE  that entered   the  system
discharged  to Mill  Creek  or the  ditch  by  the  years 2008  and  2033   (25  and 50
years into the future), respectively.

REMEDIAL ACTIONS
     Two sets of  remedial action  runs  were simulated with the model:   those
based on the  remedial  actions  proposed  by  CH2M  HILL;  and  those  based on the
remedial actions proposed  by the  PRPs.   The model runs based on  the CH2M HILL
proposed actions were:
1)   no-action (extend the base case into the future);
2)   source removal;
3)   source removal and pump and treat (100 gpm);
4)   source removal and pump and treat (200 gpm);
5)   cap and pump and treat  (100 gpm);  and
6)   cap and pump and treat (200 gpm).
     The model runs based  on the PRPs proposed actions were:
1)   no-action;
2)   source removal combined with a slurry wall;
3)   source removal, slurry wall, and pump  and treat (100  gpm);
4)   source removal, slurry wall, and pump  and treat (200  gpm);
5)   source removal and pump and treat (100 gpm) (no slurry wall); and
6)   source removal and pump and treat (200 gpm) (no slurry wall).
                                      76

-------
     The  no-action  simulations  consisted  of  running the  base  case  (1963  -
1983) 50 years  into the  future in order to provide  a benchmark  against  which
remedial action cases could be compared.
     The source removal  action (excavation of contaminated unsaturated soils)
alone is  ineffective  in  cleaning up the TCE  on site  because apparently  very
little TCE  remains  in the  unsaturated  soils.   According  to the calculations
shown in Appendix A,  only  about  1,300  Ib (590 Kg) of  TCE  currently exists  in
the  unsaturated  zone  on   site.    Of  this  total,  99%  is  present   in  the
unsaturated  soils  in  the   area  around Reaction Pond  I.    Most  likely,  the
source removal  action would be effective for  constituents  with high affinity
for  adsorption  such  as   metals  and  other   highly  sorbed   contaminants.
Therefore,  this action  should  not be construed  to  be universally ineffective
at the Western Processing Site based on the predictions for TCE.
     In the  CH2M HILL cases,  the source removal  and capping actions combined
with  pumping  and  treatment   achieved  about  the  same  results;  the  action
involv.ing...source removal  is slightly more effective  in  reducing the TCE mass
loading to Mill Creek, but  a slightly greater  contaminant  mass remains  in  the
system  than for  the   action  involving capping.   These results  are  similar
because  both  actions essentially  eliminate  the  leaching  of  TCE  from  the
unsaturated  zone  to  the saturated groundwater  system.    The  cap  eliminates
only about   5  gpm  (19  1/min)  of  recharge  over  the  site,  a  small  number
compared to the 100 gpm and 200 gpm (379 and 758 1/min) pumping rates.
     In the PRPs  cases, the  remedial   actions  without  the  slurry  wall  were
more effective  in  reducing the total mass  of  TCE  in the system and the mass
of TCE exiting  to  Mill  Creek  and the  ditch,  than  the cases  with  the  slurry
wall.   The  reason  for  this  is  that  the  slurry  wall  prevented  the  pumping
wells from  removing  contamination  which  is  outside  the  slurry  wall.    The
slurry wall  is effective in reducing the  lateral flow of water from the creek
to the  puming  wells,  and  thereby reducing the  total  quantity  of  water  that
requires  treatment.    Also,   the  slurry  wall  is  effective   in  containing
contamination  once  the  pumping  ceases.    The  slurry  wall  would  be  more
effective if a  cap  were  also   installed over the site  to reduce the amount of
recharge.    The  cap  would  reduce  the leaching  and  transport  of contaminants
through  the  system  and  under  the wall.   Even  without the  cap,  however,  the
rate of migration of  contamination  through  the site and under the wall   could

                                     77

-------
be low  enough  that  the  concentrations would  be  di-luted to  acceptable levels
in the groundwater flow system.
     In  all  cases   involving  pumping,  withdrawal  of  water  at  the  200- gpm
(758 1/min)  rate  removed all  the  TCE from  the system in  the first  5 years.
About  97%  of  the TCE  was  removed  by the  pumping wells  with the  remainder
discharging to Mill Creek (and a trace to  the  drainage  ditch).
     In the  CH2M  HILL cases  and  the PRPs case with the slurry wall,  pumping
at the  100  gpm (379 1/min)  rate  removed  all  the TCE  in  the first  15 years.
In all  three cases,  most of the TCE  (about  90%)  exits the sytem  in  the first
5 years.  Over the  15-year  period, between 85% and  90% of the TCE is  removed
by  the pumping  wells for  the  CH2M HILL cases.   In the  CH2M  HILL  cases,
pumping was  simulated for  the  first 30 years  (1983 - 2013), whereas  in  the
PRPs cases,  pumping  was simulated  for  just  the  first  5  years (1983  - 1988).
Therefore, in  the PRPs cases,  all  TCE exiting the system after  the  first  5
years discharges to Mill Creek or the ditch.
     In the PRPs case of source removal  and  pump  and treat at 100 gpm  without
the  slurry wall,  20 years  were required  for all   the TCE  to  exit the  system.
Ninty-two percent  of the TCE  exited the  system  during the  first  5  years of
pumping.   Of  this  amount, 89% was removed  by   the  pumping wells,  and  the
remainder discharged  to  Mill  Creek and the  ditch.  After  the pumping  ceased
an additional 15 years was  required for  the  TCE to  discharge  to Mill  Creek or
the ditch.
     The  drawdown  over  the  site  for  all   cases  involving pumping  ranged
between 4 and  12  ft (1.2 and  4.6  m).   The  100 gpm (379  1/min)  pumping  rate
resulted in  average  drawdowns  between 4 and 6  ft (1.2  and 1.8 m) whereas  the
200 gpm (758 1/min)  pumping rate  resulted in  average  drawdowns  between 8  and
12 ft  (2.4  and  3.7 m).  For the same pumping  rate,  the cases with the  slurry
wall had  about 2  to  3  ft  (0.6  to  0.9  m)  more drawdown  because  the  wall
eliminated lateral flow to  the wells.
     In summary,  to  a  large degree  the  CH2M  HILL  and PRPs  remedial  action
cases  performed about equally.  As expected,   the modeling results  indicated
that the  200  gpm (758 1/min)  pumping rate performed better  than  the  100  gpm
(379 1/min) rate.  These  data  and  other differences which  are discernable in
the  model  results  should  be considered  in  the selection   of  the   remedial
action alternative to be implemented at the site.

                                      78

-------
                                  REFERENCES


CH2M  HILL.    1985.    Feasibility  Study Western  Processing  Kent,  Washington.
Draft.  Seattle, WA.

CH2M   HILL.     1984.     Remedial  Investigation  Western   Processing  Kent,
Washington.  Final Draft.  EPA WA 37-0116.1,  Seattle, WA.

Dunne, T., and  L. B.  Leopold.   1978.   Water  in Environmental Planning.  W. H.
Freeman & Co., San Francisco, CA.

Ellis,  R.    1984.     Personal  Communication.    Soil  Conservation  Service,
Spokane, WA.

Environmental  Protection  Agency.    1984.   Western Processing  Alternatives
Assessment Study, 1983  Data.   Environmental  Services Division,  EPA  Region X,
Seattle, -WA.

Environmental  Protection Agency.    1983.    Investigation  of  Soil   and  Water
Contamination at  Western Processing,  King County, Washington.   Environmental
Services Division, EPA Region X, Seattle,  WA.

Environmental Protection Agency.    1980.   "Water Quality  Criteria  Documents;
Availability."    Federal   Register,   pp.   45,   231,   79318-79379,   Friday,
November 18.

Gupta,  S.  K.,  C.  T.  Kincaid,  P.  R.  Meyer, C.  A.  Newbill  and C.   R.  Cole.
1982.   A Multi-Dimensional Finite  Element  Code for  the Analysis of Coupled
Fluid,  Energy  and  Solute  Transport  (CFEST).PNL-4260.Pacific  Northwest
Laboratory, Richland, WA.

Gupta,  S.  K.,  C. R.  Cole and  F. W.   Bond.   1979.   Methodology for Release
Consequence  Analysis  —  Part  III, Finite-Element   Three-Dimensional  Ground-
Water  (FE3DGW)  Flow Model.   Formulation  Program Listing  and  User's Manual.
PNL-2939, Pacific Northwest Laboratory, Richland,  WA.

Hart-Crowser and Associates.  1984.   Hydrologic Assessment  Western  Processing
Kent. Washington.  Seattle, WA.

Landau Associates and Dames and  Moore.    1984.   Weste.-n  Processing Remedial
Action Plan,  Phase II.  Volumes  I, II, and III,  Edmond,  WA.

Luzier,  J.  E.    1969.   "Geology  and  Ground-Water  Resources  of Southwestern
King County, Washington."   Water-Supply  Bulletin  No. 28,  Department of Water
Resources, Olympia,  WA.


                                      79

-------
National  Oceanic  and  Atmospheric Administration.    1974.    CTimates  of  the
United  States  - Volume  II  - Western  States.   U.S.  Department  of commerce,
Washington, D.C.


Richter,  R.  0.   1981.   Adsorption of Trichloroethylene  by  Soils from Dilute
Aqueous Systems.   Contract  No.  F49620-79-C-0038,  Air  Force  Engineering  and
Services Center, Environics Division, Environmental Engineering Branch.

Wood,  P.  R., F. Z.  Parsons, J.  Demarco,  H.  J.  Harween, R.  F.  Long,  I.  L.
Payan and M. C. Ruiz.   1981.   Introductory Study of the Biodegradation of the
Chlorinated Methane, Ethane, and Ethene Compounds.   Draft Report  prepared for
Water  Supply Research  Division,   EPA  MERL,  Cincinnati,  OH.   Summary  paper
presented at the Annual AWWA Conference,  St. Louis, MO,  1981.
                                     80

-------
                                  APPENDIX A
          CALCULATION  OF  LEACH  DURATION  AND AMOUNT FROM SOURCE AREAS

     Three  sources of  TCE  have  been  identified  at  the  Western Processing
Site,  based  on  the  concentration  levels measured  in  the  soil  and water.
These three  sources  are:  1)  Reaction  Pond  I;  2)  Reaction  Pond  III; and 3)
around Well  21.    Calculations were  made  to estimate  the  mass  of TCE present
in  the  unsaturated zone (both in soil  and water)  of these  three locations
based on  1982 soil  concentration measurements  (EPA,  1983 and  1984).  These
mass  estimates were  then used  to  estimate  the  time  required  for TCE  to
completely  leach the  unsaturated  zone into the saturated  zone.   Time  of leach
calculations  were  based  on  the  average  annual  recharge  at  the   site  (8
in./yr).  The  results of these calculations were used to  determine the number
of years  past  1982 to keep  the sources  active in the model, and  to   estimate
the source strength for  each time step simulated.

SOURCE AREA DATA
     The  data  used to  estimate  the mass  of TCE  in the   unsaturated  zone of
each source area are summarized below.
Reaction Pond I
     The total volume of contaminated unsaturated soil  at Reaction Pond I was
estimated  to  be  46,250  ft3.   This  number  is  based  on  a  surface   area  of
        2
9,250 ft  and an average depth of  unsaturated  contaminated soil  of 5   ft.  The
surface area is the actual area associated  with  Reaction  Pond I.  The average
depth of  contaminated soil  was  estimated  from  the soil   sampling depths and
concentrations reported  in Table A-l.   The TCE concentration in the   soil was
estimated to  be  600,000  ppb based  on  the measured  soil  concentrations from
the borings at Wells  15  and 17  (Table  A-l).   In order  to represent  a worst
                                     81

-------
case,  the  concentration   was   assumed  to  be  the  maximum   observed  soil
concentration and to be uniform over  the  area of  the  disposal  site.

           Table A-l.   TCE  Soil  Concentrations Near Reaction Pond I
                                   Soil Concentration  (ppb)
Depth (ft)
3
6
9
Well 15
U
580,000
180,000
Well 17
U
558,000
350,000
U = Not Detected

Reaction Pond" III
     The total  volume  of contaminated, unsaturated  soil  at Reaction Pond  III
was estimated to  be 35,650 ft .   This number  is  based on  a  surface area of
        p
7,125  ft   and  an  average depth  of  contaminated  soil  of 5 ft.   The surface
area  is "-"the actual   area  of  Reaction  Pond  III.     The  average   depth  of
contaminated  soil   was   estimated   from   the  soil  sampling   depths   and
concentrations  reported  in Table  A-2.   The TCE concentration  in  the soil  was
estimated  to be   700  ppb  based  on   the  measured  soil   concentrations from
Well 20 (Table  A-2).   In order to represent a worst  case,  the concentration
was assumed  to  be the  maximum observed soil  concentration and was assumed to
be uniform over the area of the disposal site.

         Table  A-2.  TCE Soil  Concentrations Near  Reaction Pond III
                                   Soil Concentration (ppb)
                      Depth (ft)            Well 20
                           3                   U
                           6                   M
                           9                  676
                          12                  544
U = Not Detected
M = Present but  below minimum quantifiable limit.
                                     82

-------
Area Around Well 21
     The  total  volume  of  contaminated  unsaturated soil  in the  area  around
Well 21 was difficult  to estimate  because the area of  disposal  is not  known.
Therefore,  the  volume was  assumed to  be the same  as  that  of  Reaction Pond
III, 35,650 ft3.  The TCE concentration  in  the  soil  was estimated to be  1,600
ppb based on the measured soil concentration from Well 21 (Table A-3).

                Table  A-3.   TCE Soil Concentrations  in Well  21
                                   Soil Concentration (ppb)
                      Depth (ft)             Well 21
                           3                   U
                           6                   U
                           9                   116
                          12                 1,520
U = Not Detected

CALCULATION OF SOURCE AREA MASS
     The  calculations  of  mass in  the  unsaturated zone for  each  of  the three
source locations are presented below.
     For  all   source  areas,   the  actual   and  effective  porosities  of  the
unsaturated material  were  assumed to  be 0.25  and 0.15,  respectively.   The
distribution coefficient (Kd)  of TCE  in  the unsaturated  soils was set  at 0.2
at all source locations.
     The  equation  used to  calculate the  mass  of TCE at  each source  location
is:

        =  {mass in soil} +  {mass  in water}
        - (Cs)(Vols)(ys) + O)(Cw)(Vols)(Yw)                         (A-l)
where
          total mass of TCE in the soil (Ib)
          actual porosity
                                     83

-------
      8 = moisture content-
     C$ = TCE concentration (ppb) in the soil
   Vols = total volume of unsaturated soil (ft3)
     Ys = density of the soil  (lb/ft3)
     Cw = TCE concentration (ppb) in the water  fcw = ^  (C$)}
     Yw = density of the water (lb/ft3)
      W
     Kd = distribution coefficient

     The density of the soil was calculated as:

     Ys = (1 - 6A)(Y) = 121.5  lb/ft3

where:

     SA = actual porosity (0.25)
      Y = dry density of sand  (162 lb/ft3)

     The TCE  soil  concentrations were  based  on measured  soil  concentrations
in the  unsaturated zone  at wells  around  the  source  areas.   The  TCE  water
concentrations  were  estimated  to  be  five  times  greater  than  the  soil
concentrations  based  on  a  distribution  coefficient  (Kd)  of  0.2.    The  ppb
                                   Q
notation is equivalent to Ib per 10  Ib in the total mass calculation.
Reaction Pond I
     Based on the data  in Table  A-4,  the  total  mass of TCE in the unsaturated
zone at Reaction Pond I was calculated to be 4,670  Ib.   The total  mass in the
soil  and water was calculated  to be 3,370 Ib and 1,300 Ib, respectively.
                                      84

-------
      Table A-4.  Summary of Data Used to Make Total Mass  Calculations
                  at all Three Source Locations
                       Reaction Pond I    Reaction Pond  III     Well 21
         8                      0.15              0.15             0.15
         0A                     0.25              0.25             0.25
         Cs (ppb)          600,000               700           1,600
         Cw (ppb)        3,000,000             3,500           8,000
         Vol$ (ft3)         46,250            35,650          35,650
         Ys (lb/ft3)            121.5             121.5           121.5
         \ (lb/ft3)            62.4              62.4            62.4
         Kd                     0.2               0.2             0.2
Reaction Pond III
     Based on the data in Table A-4,  the  total  mass  of TCE in the unsaturated
zone at Reaction Pond III was calculated to be 4.2 Ib.  The  total  mass  of TCE
in the sot! and water were calculated to be 3.0  Ib and 1.2 Ib, respectively.
Around Well 21
     Based on the data in Table A-4,  the  total  mass  of TCE in the unsaturated
zone around Well 21 was  calculated  to  be  9.6  Ib.   The total  mass  in  the soil
and water were calculated to be 6.9 Ib and 2.7 Ib, respectively.
     The estimated total  mass  of  TCE  present in  the unsaturated  zone  (based
on  October,  1982,  soil   analyses)  at  the three  suspected   source  areas  is
summarized in Table A-5.

         Table  A-5.   Summary of Estimated Total Mass  of TCE  Present
                      at  the Three Suspected Source Areas
                                         Mass  of TCE  (Ib)
                                   In Soil    In Water    Total
               Reaction Pond I     3,370     1,300      4,670
               Reaction Pond III       3.0       1.2        4.2
               Around Well  21          6.9       2.7        9.6
                                     85

-------
DURATION AND AMOUNT OF LEACHING FROM SOURCE LOCATIONS
     The preceding calculations  show that Reaction  Pond  I  is the  only source
location  which contains  a  significant  quantity  of  TCE  in  the  unsaturated
zone.    Calculations  of  the  time  required   for  TCE  to  leach  out  of  the
unsaturated  zone   (duration),  and  of   the  amount that  would  leach  out  each
year,  were made  for Reaction  Pond  I.   This  information  was  used  in  the
modeling to determine the  length  of time into the future that  Reaction Pond  I
would  remain   active,  and  the  quantity  of  TCE  that  would  leach  from  the
unsaturated to the saturated zone at each time step.
Reaction Pond  I
     The calculations were  based  on the  concentration  of TCE in the water of
the  unsaturated zone  and  the rate  of  water movement  through the  unsaturated
zone  from  annual   recharge.   Equation  A-l  can  be  used to  calculate the TCE
concentration  in  the  water (C )  based  on the  total  mass of  TCE  remaining in
                              w
the  system  at  any given point in time.   For  convenience, the equation can be
rewritten as follows:

                MassTCE
      w  Vol  (Kd Y. + 6 y )                                            (A-2)
            9      J      W

     The Kd in equation A-2 is equivalent to C./C .
                                              5  W
     The volume  of  water  (Vol )  moving through  the  unsaturated zone from the
                              W
average annual recharge was calculated as follows:

   Volw = Rhg x A = 6,167 ft3/yr

where

    Rhg = average annual  recharge (0.67 ft/yr)
      A = area of Reaction Pond I (9,250 ft2)

This number  was  used to calculate  the mass of  water  (Mass  )  passing through
                                                            W
the unsaturated system, as follows:
                                     86

-------
  Massw = Volw x yw = 384,800 Ib/yr                                   Y

     The  initial  mass of  TCE  in the  system was 4,670  Ib.   For  each year  a
mass of TCE removed  from  the system was calculated as Cw times  Massw-  A  new
mass of TCE  in the system was calculated  as  the  previous  mass  minus  the mass
removed,  and  the procedure  was  repeated.   This iterative  process continued
until  virtually  all  of the TCE was  leached  from the system.   The results of
this calculation are shown in Table A-6.

            Table A-6.  Time and Amount of TCE  Leaching  from  the
                        Unsaturated  Zone at  Reaction  Pond  I
                  Mass TCE (Ib)                        Mass TCE   (Ib)
Year
1
2
3-....
4
5
6
7
8
9
10
Remaining
4,670
3,516
2,647
1,992
1,500
1,129
850
640
482
363
Total Removed
1,154
2,023
2,678
3,178
3,541
3,820
4,030
4,188
4,307
4,397
Year
11
12
13
14
15
16
17
18
19
20
Remaining
273
206
155
116
88
66
50
38
29
22
Total Removed
4,464
4,515
4,554
4,582
4,604
4,620
4,632
4,641
4,648
4,653
     Table A-6  shows  that virtually  all  of the TCE  in  the  unsaturated  zone
above Reaction Pond I  is  leached  into the  saturated  zone in 20 yr.  Over each
5-year period (the time step  used  in  the  model)  about 75% of  the  mass  of TCE
remaining  is  flushed  from  the system.    Based  on  these calculations,  the
Reaction Pond I  source was allowed to  leach for  20  years into the  future in
the model, and the source  strength was reduced  by 75% at every time step.
Reaction Pond III and Area Around Well 21
     The total  mass  of TCE  in the unsaturated  zone above Reaction  Pond III
and in the area around Well 21 was so  small that  it  was  not necessary to make
calculations  of  the  leach duration  and amount.   As a  result,  these sources

                                      87

-------
were allowed  to leach up  to the present,  and then  turned  off for  all  model
predictions into the future.

-------
                                 APPENDIX B
                            RECHARGE CALCULATIONS

     Recharge due  to precipitation  was calculated  using  the  water  balance
formula:

     Recharge =  Precipitation - Actual  Evaporation - Runoff            (B-l)

Average annual precipitation and actual evapotranspiration  for the study area
are about 39  in./yr  (99  cm/yr) and 18  in./yr  (46 cm/yr),  respectively (NOAA,
1974).   Runoff   was  calculated  using   a  method  developed  by  the U.S.  Soil
Conservation Service and modified by Dunne  and  Leopold  (1978).  The technique
is based  on a  simplified  infiltration model  of runoff,  daily precipitation
events,  and empirical  approximations   which  consider  such  factors   as  soil
type,   land  use,   vegetative  cover,  and  storm separation  interval  to determine
the antecedent soil  moisture conditions.
     A  program   developed  at  Battelle and  based  on   the  Soil  Conservation
Service method "was used  to  calculate  runoff for  the Western  Processing study
area.   The  calculations were made using daily  precipitation data for 1982 and
1983.   The  output from the  program  is  a list  of runoff estimates  for  a range
of runoff curve  numbers.    A  runoff curve number  of  70 was  selected  for the
study  area  based on the  soil  type  (Group  B),  land use  (residential  area with
one acre  lots),  and  total  impervious   area  (20%) (Dunne and  Leopold,  1978).
The curve number 70 corresponds to  curve number  85  if  normally wet antecedent
moisture conditions prevail, as might be expected for the area around Kent.
     The  results  of  the model  for the  two runoff  curve  numbers  at  several
storm separation intervals  for the  year 1982 and 1983 are shown in Table B-l.
                                     89

-------
                         TABLE B-l.   RUNOFF PROGRAM RESULTS
Storm Separation Runoff (in,
Interval
(days)
0
1
2
3
1982
CN 70
0.3
6.8
7.9
11.4
CN 85
2.5
14.8
16.9
20.5
./yr)

1983
CN 70
0.3
7.6
16.0
19.2
CN 85
2.9
17.3
25.1
28.1
              CN = Curve Number

     Using equation  B-l,  storm separation  intervals of 0,  1,  2,  and 3 days,
and averaging the runoff over two years  yields  estimated recharges of about 5
in./yr  (13  cm/yr)  and 12  in./yr  (30 cm/yr)  for  runoff curve  numbers  70 and
85, respectively.  The  curve number that  applies  to the area  around Kent is
probably between 70  and  85.   Therefore, the  actual  recharge was estimated to
be about 8 in./yr (20 cm/yr).
     In the  final  calibrated model  a recharge value  of 8  in./yr  (20 cm/yr)
was applied  uniformly over the  local model region  except for  paved  areas on
the Western Processing Site  (Elements 128,  129,  130, 143,  144, 145,  146, 152,
153, 154,  163, 164,  and  175)  where the recharge was set to 0 in./yr.
                                        90

-------
                                  APPENDIX C
                    STREAM  AND  LEAKANCE  BOUNDARY  CONDITIONS

      The boundary  conditions  in the  FE3DGW  model   were  defined  using  the
 stream boundary options to describe flux to Mill Creek  and  the ditch,  and the
 leakance boundary  option  to describe  flux  across the  perimeter  boundaries.
 This  appendix  provides a more  detailed discussion  of  the data  used  in  the
 model to implement  these options.

 STREAM BOUNDARY OPTION
      Surface water  bodies are  often  expressions  of the water  table.   This
 phenomenon  is  often treated  in a groundwater model by holding  the  groundwater
 elevation at the level  of the surface water.  The  stream option  in  the FE3DGW
 code  allows  the  potential   to  fluctuate   above  or  below   a   stream,   and
 calculates  a flux  (to  or  from  the  stream)  based on  the potential  difference
 between the elevation  of  the stream and  that  of  the groundwater.   The  data
 required by  the model  to   make this  calculation  are  the  stream  surface
 elevation;  the stream  bottom  elevation,  cross-sectional  area,  thickness,  and
 permeability;  and  minimum stream depth.    These data were  entered into  the
•model for  each  node along  Mill  Creek  and  the drainage  ditch.    The  model
 calculates  the  flux to (gaining)  or  from  (losing)   each  node using  Darcy's
 Law.   The  data used to implement the  stream option  in  the final  calibrated
 model  for   Mill  Creek  and  the  ditch  east  of  the  site  are  provided  in
 Tables C-l  and C-2,  respectively.
      The surface water  elevation at  nodes along  Mill  Creek  and the  ditch were
 interpolated and extrapolated from  measurements at  five  locations  along  the
 Creek and  two  along  the  ditch (Figure  7).  The  measurements  were  made  on
 April 10, 1984 by EPA Region  X.
                                      91

-------
                  TARLE C-l.  Stream Boundary Option Data Used to Simulate Flux to Mill Creek
ro
NUUC
NUMBED
2
12
22
33
34
35
47
58
75
92
122
134
150
159
177
186
204
214
230
239
251
263
275
287
286
289
290
301
309
286
285
272
271
270
L w C r K
Fl t'VATtn J
1
1
I
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I . U6
! ,ซ0
i.75
I .70
1 .67
1.63
1 .61
.60
.59
.57
.565
.558
.55
.5?
.5
.49
.4M
.46
.27
.07
10.87
10.67
L. * r. c i\
LF.'iGTM
246.0
J 6 9 . 0
27'>.0
246.0
246.0
246.0
1 4 * . 0
96.0
140.0
131.0
100. 0
1 >) ft . 0
7'J.O
50.0
K2.0
9H.O
1 1 "I . 0
130.0
165.0
215.0
246.0
29-5.0
10.47 2^5.0
10.35
10.3
10.27
1 0 . 1 a
10.1
10.05
10.21
10.17
10.13
10.09
\ 't 0 . 0
277.0
29'i.i
J24.0
560.0
215.0
1 'to . 0
2 1 :i . o
<> h ? . 0
S^M.O
1 0 . ofc 3/>ft.O
? K P. 10.01
?9 S
ซ*.
.<**
360.0
t 9 ft . t)
U '1 1 1 K
wlOTH
5.0
5.0
5.0
5.0
5.0
5.0
5..)
5.0
5.n
5.0
5.0
5 . 0
•* . o
5.0
5.0
5.0
5.0
5.')
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5,0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
s.o
ELEVATION
10.86
10. ft
10.75
10.70
10.67
10.63
10.61
10.60
10.59
10.57
10.565
10.558
10.55
0.52
0.5
0.49
0.48
0.46
0.27
0.07
9.87
9.67
9.47
9.55
9.3
9.27
9.1H
9.1
9.05
9.21
9.1 7
9.13
9.09
9.04
9.01
ซ.9fl
THICKNESS PERME
o.
o.
o.
o.
o.
o.
0.
o.
o.
0.
o.
o.
o.
o.
o.
o.
o.
0.
0.
o.
o.
0.
0.
o.
o.
0.
0.
o.
0.
o.
o.
o.
o.
o.
o.
o.
o.
0.
o.
o.
o.
0.
0.1 0.
O.I 0.
0.1 0.
0.1 0.
0.1 0.
0.1 0.
0.1 0.
0.1 0.
0.1 0.
0.1 0.
0. 0.
0. 0.
0. 0.
o. o.
o, o.
ABILITY
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
142
1 12
run unccrv
DEPTH
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.2S
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25

-------
TARLE C-2.  Stream Boundary Option Data Used to Simulate Flux to the Ditch
N'mt
NUMBER
51
65
62
99
1 1 1
1 15
1 29
1 11
157
166
1 M
193
209
219
234
LMtt K
ELEVATION
13.
13.
3.
i.
3 .
3.
3.
13.
1 3.
13.
1 3.
13.
13.
12.
1 2.
57
52
19
<45
U2
1
39
37
31
32
3
I -5
0
t\
6
LHt tr,
LENGTH
66. 0
I 25.0
1 >) ft . 0
^2.0
57.0
57.o
60.0
7n.i
(> ti . 0
57.0
5 ' . 0
7 i . 0
b2.n
125.0
73."
C.K
Hi
2
2
2
2
2
2
?
2
2
2
2
2
2
2
2
OTH ELEVATION
.ti 12.57
.0
.0
.0
.0
.0
. 0
. o
. 0
. o
.0
.')
.0
.0
. >*
2.52
2.19
2.15
? i^ 2
2.1
2.39
2.37
2.31
2.32
2.3
2.15
2,o
t.e
1.6
C*EtK
THICK
0.
o.
0.
o.
o.
o.
0.
0.
o.
0.
o.
o.
o.
o.
o.
NESS PERMC
i
i
i
i
i
i
i
i
i
i
i
i
i
l
i
o.
o.
o.
o.
o.
o.
o.
o.
o.
o.
o.
o.
o.
o.
o.
ABILITY
112
112
112
112
112
112
112
112
112
112
112
112
112
112
112
11N tซtEK
DEPTH
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25

-------
LEAKANCE BOUNDARY OPTION
     The  "leakance  boundary  condition"   option  of  the  FE30GW  code  allows
flexibility in defining external  boundaries  of the model region.   Rather than
a constant flux or  held potential at the boundary, the  leakance  option  allows
the potential  and  flux to be  varied  depending on  the  conditions  which  exist
within the study area.
     The data  required  by the model to make this calculation arethe  distance
from the boundary  to a known  potential,  the potential  at  that  distance,  and
the cross-sectional  area  of  the boundary.   These  data are  entered into  the
model   for each node along  the boundary,  including  both surface nodes  and
nodes  at  depth.    The model  calculates  a boundary flux  at  each  node  using
Darcy's Law, which  is then used to calculate the potential at  the boundary.
     A  map   depicting  the   regional   potentials  used   to  calculate   the
groundwater potential  at  certain  distances  from the  boundaries is  shown in
Figure  C-l.   This  map  shows  the distances  to the extended  boundary  and  the
gain or lass in potential  elevation out to these distances.
     The  Green River  elevations  were  interpolated  from  three  measurements
taken by EPA Region X in April,  1984 :
     1.   east of benchmark 32 (southwest of site) - 9.8 ft (3.0 m)  AMSL;
     2.   east of benchmark 22 (west of site) - 8.8 ft (2.7 m) AMSL; and
     3.   Tukwilla  Gauge (north of site) - 7.9 ft (2.4 m) AMSL.
                                     94

-------
   Regional Potential Value in Feet
         LQCflL MODEL RESIGN
                                                  +3 ft

                                            VU1M4 P1VCEB8DC SITE
Figure C-l.  Locations of Regional Values Used to Determine
             Boundary Conditions
                             95

-------
                                  APPENDIX D
                      CALCULATION OF RETARDATION FACTOR

     The retardation factor, K, can-be calculated  by  the  formula

     K = 1 + B Kd                                                        f0'1)

where B is defined as the  bulk  density  divided by the effective porosity,  and
Kd is the  distribution  coefficient.   The dry density of  the  silty  sand  at  the
Western  Processing   Site  was   estimated   as  the  bulk   density  of  sand
(2.6 giti/cnr)  times  the quantity  1.0 minus  the  actual  porosity  (1.0  - 0.4 =
0.6); or"1.6  gm/cm .  Kd's for  TCE  have been reported in the range of 0.1  to
1.0 cnr/gm  (Richter,  1981), depending on  the soil type.  A  Kd of 0.3  cnr/gm
was used to represent the  silty sand material at the Western  Processing  Site.
     Using  the  bulk  density  for silty  sand of  1.6  gm/cm   and an effective
porosity of  15%,  yields a value  of  10.7 for B.   Substituting  B  and  Kd into
equation D-l  yields  a K of 4.2,  which  means that the TCE  travels about four
times slower than the groundwater.
     This  value   of  K  was used  as  a  guide  in   determining  the  retardation
factor to use in  the final calibrated CFEST model.
                                      96

-------
            ANALYSIS OF THE WESTERN  PROCESSING MODEL SENSITIVITY
                      USING LATIN HYPERCUBE  SAMPLING

          To better understand  Battelle's  groundwater flow  and  contaminant
transport model of  the Western  Processing Site, a sensitivity  analysis was
performed  using  Latin  Hypercube  Sampling  (LHS).    LHS   is  a  constrained
sampling  scheme   which   selects   values  within  a   specified   range  and
distribution (Iman  and Shortencarier, 1984).  These values  can  then be used
as model  input parameters  so  that  the correlation between  the  tested input
parameters and the  model  results can be determined.
          A  Latin   Hypercube   Sample  is  a  multiparameter  (multivariate)
sample  composed of  a  number of  replications of individual  realizations  of
the  multiparameter  set.    Each  individual  parameter  is  assigned  to  a
probability   distribution   (normal,    lognormal,    uniform,    loguniform,
triangular,  beta,   or  user-defined).     The   range   of   each   probability
distribution is partitioned into  a  number of equally  probable  intervals;  a
uniform distribution is divided into  a number of  intervals  of  equal length.
LHS involves selecting a  parameter  value  for each  of the  equally probable
intervals  (according  to  the  conditional  distribution of that  interval) and
then randomly  permuting  the  orders  of intervals  in  order  to  introduce the
proper  degree of correlation between  parameters.  The  main  advantage of LHS
is  that  the  entire range  of  the   parameter  is  sampled   in  an  efficient
manner.
          The  LHS  technique  was  applied  to analyze  the   sensitivities  c"
five parameters with  respect to  the conditions  at  the Western  ฐrocessing
Site.    These  parameters  are:    the hydraulic conductivity  of  Unit  1;  t^e
hydraulic conductivity of Unit  2; the  effective  porosity;  the  recharge; and
the retardation factor for trichloroethylene  (TCE).   The sampling range and
distribution for  each parameter  are shown  in Table 1.  These  distributions
were input  into  the LHS  program which  generated  25 realizations  of these
parameters  (Table  2).   These  parameter  realizations  were then  input into
CFEST  to  create 25  realizations of  the  model output.   Note that  the  source
strength and dispersivity were not analyzed in this  study.

-------
        Table  1.  Range  and Distribution of Parameters
                   Used in LHS Analysis
      Parameter
Hydraulic Conductivity
  (Unit 1)

Hydraulic Conductivity
  (Unit 2)

Porosity

Recharge

Retardation Factor
        Range
    1 to  10 ft/day


   10 to  100 ft/day


     10% to 35%

6 in./yr to 12 in./yr

       2 to 8
Distribution

  Uniform


  Uniform


  Uniform

  Uniform

  Uniform

-------
             Table 2.  Western Processing Sensitivity (Latin
                       Hypercube Sample Input Vectors)
Run No.
1
2
3
4
5
6
7
1.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
K*
Unit 1
x (1)
1.40
6.26
6.80
9.31
4.07
5.30
6.47
3.62
1.00
2.42
9.27
7.72
7.24
1.95
3.08
4.86
7.98
5.34
8.85
9.88
2.59
3.47
4.57
5.73
8.52
K*
Unit 2
x (2)
60.0
21. 9
47.5
96.8
55.1
75.5
19.3
84.5
90.1
16.6
93.9
70.4
28.5
37.4
73.2
12.8
64.7
52.6
41.0
27.9
45.3
86.5
62.4
79.0
33.9
Porosity
x (3)
0.318
0.147
0.169
0.227
0.306
0.261
0.320
0.280
0.123
0.237
0.172
0.246
0.135
0.218
0.186
0.159
0.252
0.194
0.3*1
0.107
0.206
0.119
0.298
0.286
0.339
Recharge
in/yr
x (4)
11.1
10.5
10.6
7.24
8.91
7.82
6.47
8.52
10.8
9.50
9.28
7.05
8.26
6.70
6.01
8.66
6.96
9.71
11.5
9.92
7.66
11.6
10.2
8.16
12.0
Retardation
x (5)
7.54
7.85
4.30
4.78
4.04
7.13
3.58
6.49
4.57
2.90
5.31
7.41
6.29
5.38
2.20
5.63
6.62
5.87
5. 03
3.70
3.21
2.99
6.94
2.34
2.66
* ft/day

-------
          The  MINITAB   statistical   package  was   used  to   analyze   the
relationship   between   parameter  input   and  model   results.     The   C
classifications  used  in MINITAB  are  shown in Table  3.   Histograms  (which
show  the  distribution  of the model  results)  and descriptive  statistics  of
the model  outputs  (1983 concentrations,  flux  to Mill Creek,  and  total  mass
of  TCE   in  the   groundwater   system)   are   shown   in  Tables   4   and  5,
respectively.  The histograms  and descriptive statistics  indicate  that the
overall changes  in input result in less than  an  order of magnitude  change
in model-predicted 1983  source  concentrations.  But these  changes do  result
in order  of  magnitude  changes  in  Mill  Creek  flux  and  predicted 1983  Mill
Creek concentrations.
          The  model   results  were  regressed  upon  values   of  the   input
parameters (Apppendix  A).   The  regressions were  analyzed by  comparing the
t-statistics  (t-ratio)  for  the regression  equations.    A  summary of  the
regression t-statistics  is  shown in Table 6.   A higher  absolute  value  of
the  t-statistic   indicates  a greater  correlation  between the  parameters.
For example, in Table 6 the  highest  t-statistic  for the flux  to  Mill  Creek
is for  the hydraulic conductivity of Unit 1.   Therefore, the  flux to  Mill
Creek  is  primarily  controlled  by the  conductivity  of Unit  1.    In  this
example,  the  conductivity of Unit  2 has  the  next  highest  t-statistic and
therefore  is the  secondary  controlling  factor.   T-statistics  with  absolute
values  less  than  two  are   considered  insignificant.   Examination  of  t-
statistics for the input parameters indicates that the parameter  with the
greatest  affect  on the  model  results is the  hydraulic  conductivity of  Unit
1.   The results   of  the regression  analysis  for each  of  the  model  output
variables' are discussed below.
          The  predicted  1983   concentration  at  Well   21   is   primarily
controlled  by  the  hydraulic   conductivity  of  Unit   1,  followed  by  the
retardation factor,  effective  porosity, and  recharge.   The  predicted  1983
concentration  at  Reaction  Pond  I   is  almost  entirely  controlled by the
hydraulic conductivity  of Unit  1 with little  or  no  influence from the other
parameters.  The  dominance  of  the Unit  1 hydraulic  conduct! vi'ty indicates
that   the   hydraulics  (the  conductivity and  gradient)  are   dominating the

-------
          Table  3.  Key  to C Classifications

 Cl = 1'983 TCE Concentration  at Well  21  (ppb)
 C2 = 1983 TCE Concentration  at Reaction  Pond  I  (ppb)
 C3 = 1983 TCE Concentration  at Reaction  Pond  III  (ppb)
 C4 = Flux to Mill  Creek  (cfs)
 C5 = 1983 TCE Concentration  in Mill  Creek  (ppb)
 C6 = Hydraulic Conductivity  of Unit  1  (ft/day)
 C7 = Hydraulic Conductivity  of Unit  2  (ft/day)
 C8 = Porosity
 C9 = Recharge (ft/day)
CIO = Retardation
Cll = 1968 Total  Mass  of  TCE  in Groundwater System
C12 = 1973 Total  Mass  of  TCE  in Groundwater System
C13 = 1978 Total  Mass  of  TCE  in Groundwater System
C14 = 1983 Total  Mass  of  TCE  in Groundwater System
C15 = 1988 Total  Mass  of  TCE  in Groundwater System
C16 = 1993 Total  Mass  of  TCE  in Groundwater System
C17 = 1998 Total  Mass  of  TCE  in Groundwater System
C18 = 2003 Total  Mass  of  TCE  in Groundwater System

-------
               Table 4.  Histograms  of Model Results
P1TB  >  HI37 Cl

Histogram of Ci    N  =  25
(1983 TCE Concentration  at Well 21)
Midnoint   Count
                 7  *******
                                       i17B  >  HI37  C4

                                       Histogram of  C4
                                       (Flux to  Mill  Creek 1
                                       Kirfooint    Count
                    N  = 25
   -w-t-c-i'*.'
                     ******
                 0
                 3   ***
 N7B > HIST C2

 Histo_prarn  of C2    N = 25
 (1983 TCE Concentration at Reaction
 Micaoint    Count         Pond I)
ฃ0888
80888
1 1710800
120088
148888
1&8088
1 80888
280888
223888
248888
2&088C
258000
5
ฃ
4
4
0
2
c!
1
i
0
ft
0
1
*****
******
****
****•

•**
**
*



*
8. 1
8.2
ฃi. 3
8. 4
e.s
8. ฃ
0.7
0. a
8.9
1. 0
I. i
"17B > HI
1
1
3
4
4
3
4
1
4
1
2
37 C5
Histogram of C5
(1983 TCE
PI idooi nt
20
48
ฃ8
88
180
120
148
*
*
***
*
*,***
***
*•***
*
****
*
**

N = 25
Concentration in Mill Creek)
Count
1
4
2
9
8
ฃ
2
*
****
**
*********

******
**
                                             180
(1983  TCE Concentration  at Reaction
Kicaoint    Co MI-IT:       Pond III)
    ฃ'Z!'2'0i3      '  4   **** '

   IB'!'8 8 8        ฃ•   ******
   288C88        0
   24|Z|iZ100        5   *****
   2E8888        8
Histogram of  Cll
(1968  Total Mass of TCE in System)
tfidooint    Count
4888
5288
5ฃ'?8
ฃ888
ฃ400
ฃ888
7208
7ฃ88
1
8
1
3
4
10
5
1
*

*
***
#***
*****
*****
*

-------
                     Table 4.  (Cont'dK
  S  1  n .* ;
Histogram of C12    N = 25
(1973 Total Mass of TCE in  System)
Xidsoint    Count
ฃ0.00
7000
3ฃ:20
5'Z'00
183133
11000
12000
1 3000
14000
15000
1 *
0
0
1 *
7 *******
0
8 ********
A ****
3 ***
1 *
MTB > HIST C13
Histocram
(1978 Total
^licooint
ฃ000
,_B000
10000
12000
1A000
16000
18000
ฃ0000
22000
MTB > HIST
Hist oar am
(1983 Total
M iCDOint
6000
8000
10000
-ฃ'200
'. -000
1 ฃ000
13000,
2'?ฃ30
2ฃ.ฃ'i'0
ฃA^0 1?
26000.
28000
of C13 N = 25
Mass of TCE in System)
Count
1 *
0
0
7 *******
1 *
7 *******
5 *****
3 ***
1 *
CIA
of CIA N = 25
Mass of TCE in System)
Count
1 *
0
•?
2 **
5 *****
1 *
0
g *******.**.
2 **
3 ***
1 *
1 *
:ซ.~3  >  nIST Cl5

Histogram of  CIS   N =  25
(1988 Total Mass of TCE in  System!
tficooint   Count
        0        1  *
                 0
                 6
    1208)0        2
    10001?        3
    ฃ0(2 01?        2
                                                        1
                                                           ******
                                                           **
                                                           *********
                                                           *+
                                                           ****
                                                           *
                                       tfTB > HIST C16
                                       Histogram of CIS    N = 25
                                       (1993 Total  Mass of TCE in System)
                                       ^.iciacint    Count
                                               0         1   *
                                           4^00         3   ***
                                           8000         5   *****
                                          12000         7   *******
                                          160C0         3   ***
                                          ฃ0000         5   *****
                                          ฃ4000         0
                                          ฃ8000         1   *

                                       MTB > HIST C17

                                       Histogram of C17    N = ฃ5
                                       (1998 Total Mass of TCE in System)
                                       Midooint    Count
                                               0         1   *
                                           4000         7   *******
                                           3000         6   ******
                                          12000         5   *****
                                          j C. 71 iTl '21         ',   *
-''.~~.B.... .t ,--~'.'L2-. Z lo

~.iฃ".'-• crarfi OT  ClS   .-  -  _-j
(2003 Total Mass of TCE in System)
l^iciDoirit   Count
                                                            f\ =  2
                                                        5  *****
                                            ••_ 'ฃ.' C

                                            ^AJi?
                                                        3  *-ป•*-
                                                           *

-------
                       [able 5.  Descriptive Statistics  for Model Results

                                       REACTION  REACTION  MILL CREEK   MILL CREEK
                            WELL 21     POND I   POND;  III     FLUX     CONCENTRATION

N
MEHN
MEDlfYN
T MLlflNJ
STDEV
GEMEON
KAX
MINI
Q3
Ql
Cl
ฃ5
331701
332803
374252
1 38385
27677
703421
225303
474029
272914
€2
ฃ5
111-454
94962
106443
54375
10875
ฃ83913
54238
138981
70451
C3
ฃ5
159548
132970
152643
71768
14354
389244
8Q675
199807
102434
TMEAN.= Mean of the trimmed
SEMEAN = STDEV/ /N~
Q3 = Third quartile (7SX.)
Ql = First quartile (2^)
                                                                CD
                                                                C'=-
                                                                i™^J
                                                            0. 638
                                                            0. ฃ40
                                                            0.637
                                                            0. ฃ83
                                                            0. 057
                                                            1. 140
                                                            0. 139
                                                            0.909
                                                            0. 440
   sample (less the upper and lower 5%)
                                                 C7
                                                 ••"ปtr
                                                 C.U
                                               88. 1
                                               81. 6
                                               86. 8
                                               38. 3
                                                7. 7
                                              188.4
                                              . 17.0
                                              120.8
                                               60. 6
                                          Total  Mass  in  the System
              1968
               Cll
1973
N
VEPN
MEDIflN
TKCfiN
STDEV
BEMEON
MAX
MIN
Q3
. ••_ , f
ฃ643
6831
6678
576
115
750ฃ
4976
7046
-."• f^
; 1650
12170
1 1760
1864
373
14722
6244
1 ฃ958
1978
C13
ฃ5
15555
16377
15701
3426
685
ฃ1595
6145
17973
1 ฃ633
1983
C14
ฃ5
18640
19664
18785
5083
1017
ฃ8164
5777
ฃฃฃ80
1 4LT'c'B
1988
CIS
ฃ5
15371
16233
15433
6201
1ฃ40
ฃ7936
1385
19829
-6 :!.•.;•
1993
C16
ฃ5
1ฃ434
1ฃ818
1S395
6584
1317
27005
0
17157
:>?." /•
1998
C17
•~tET
C. wJ

9706
9869
6647
1329
ฃ6002
0
. 1 4629
377?
2003
CIS
PS
8ฃ79
6998
7910
649E
1298
ฃ5039
0
1ฃฃ93
c3'
-------
Table 6.  Sumtrary of T-Statistics for Regression  Equations
K Unit 1
K Unit 2
Porosity
Recharge
Retardation
Factor
1983
Well 21
Concentration
r!2.72
1.16
-4.33
-3.35
-7.27
1983
Reaction
Pond I
Concentration
-8.98
0.08
-1.59
0.03
-1.13
1983
Reaction
Pond III
Concentration
-9.73
0.89
-2.90
0.37
-2.89
Mill
Creek
Flux
32.59
6.47
1.29
-1.25
0.39
1983
Mill
Creek
Concentration
11.64
0.56
-7.33
1.25
8.95
1983
Total TCE
Mass in
Svstem
-11.40
-0.76
7.34
-1.14
8.90

-------
transport  at  this source  and  the retardation has  very  little effect.  The
predicted  1983 concentration  of  Reaction  Pond III  is  also  dominated  by the
hydraulic  conductivity   of   Unit  1,   while   the  effective  porosity  and
retardation factor  affect the concentration to a  significant  but to  lesser
degree.
           The hydraulic  conductivity of Unit  1  dominates the predicted flux
to Mill  Creek.   To a  lesser  degree, Mill  Creek flux  is  also  influenced by
the  conductivity  of Unit  2.   The predicted 1983  TCE  concentration in Mill
Creek  is  about   equally  dependent  on  the  conductivity  of  Unit  1,  the
retardation   factor,   and  the   effective  porosity,   with  the  hydraulic
conductivity  having a  slightly greater influence.
           Regressions  were also  performed  on  the predicted  total  mass of
TCE -remaining in  the  groundwater  system  and  the  input  parameters.   (Note
that  the  mass  input  was  constant  but   the  mass  exiting  the  system  is
variable.)   The   total  mass  in  the  system is  controlled  by  the hydraulic
conductivity  of  Unit  1  followed  by  the  retardation  factor  and effective
porosity.   The  conductivity  has  more control  after  the  sources are  no
longer  active.    Comparison  of  the  predicted  1983 concentration  at  three
sources  and  the   total  mass  remaining  in  the groundwater  system  indicates
that  the  total  mass   of  TCE  is  controlled  primarily  by  Reaction  Pond I,
followed by Reaction Pond  III.   Well  21 has no  effect on  the  total mass of
TCE in the system.   The  differences  between the contributions of  tne  source
areas is   probably due to  the  fact  that the Reaction Pond  sources  discharge
to '-1'ill  Creek  while  the  Well  21  source  may  nave  li'tle e'~ect  :n '•'-:~"
Creek.
           In  summary,  the LHS  analysis  has  shown that,  of  the  parameters
tested,-  the model  is  most" sensitive" to'the hydraul ic' "conductivi'ty of Unit
1.    The  flux  and  contaminant  loading  to   Mill  Creek,  the   total mass
remaining  in the  system,  and  the  peak  concentration at the source  areas are
all  controlled   by  the   Unit   1   conductivity.     Because   the   hydraulic
conductivity  of Unit  1 is the major  controlling   factor,  confidence  in the
conductivity  values  used  as  input to the nodel will '•esult in  the  Creates:
reliability in model  results.   The  conductivities  used in  the mode"!  have

-------
been  verified  by field  testing  (at  least  in the  northern  portion  of  the
site), therefore much  confidence  can  be placed  in  the flow-portion  of  the
model.

-------
         APPENDIX  A




MINITAB Regression Equations

-------
               Cl
                    C6-C10
The renression  ecuation is
Cl ป 993017  - 39189 C6 i- 366  C7  - 486237 C8  -  15677 C9 - 3
Pred ictor
Constant
C6
C7
ca
C3
C i|?i
Coef
338017
-39133
3ฃฃ. A
-486287
-15ฃ77
-33334
Std. Coef
ฃ3531
3082
315. 8
11ฃ33ฃ
4573
4677
t-rat 10
16.47
-12. 72-
1. IB
—4, 33
-3. 35
-7=27
3 = 4ii'4c,        R-so = 33.37C

Ana lysis of  Variance
 ource
•=
-------
       MTB  >  REGRESS C3 5 C6-C10

       The  rearessiori eauation is
       C3 - 364534 - ฃฃ327 C6 * ฃ08 C7 - ฃ42655. C8 * 1275 C3 -  10048  C10
Predictor
Constant
C5
C7
C8
C3
C10
Coef
364534
-22327
208. 1
-242655
1275
-10048
Std. Coef
45095
2294
235. 1
63607
3478
3481
t-rat io
8. 38
-9.73
0.89
-2.90
0.37
-ฃ. 89
       S  =  23805        R-sq = 86.346    R-so(adj)  = 82.753

       Analysis of Variance

       Source        DF ~          S3            MS
       Regression    5   1.0674E+11   2.1347E+10
       Error         19   1.687SE+10    853303925
       Total         24   1.2362E+11
MTB >  REGRESS C4 5 C6-Cxtf

The regression eauation  is
C4 = - 0.0335 •*• 0.104  C6 * 0.00211 C7 + 0.149  C8   - -3. .?•:•ฃ'?- C3
^red ict or
Icr star,.;
C6
C7
H5
•ri0
:: = 0.^41,
Coef
-0. 03353
U' . 1 'ซ. ^. s & /
•2.2021131
C. 143Z
^1 t \ฃl lฃ. _ Q ^ Q
•* 3 — -r-_
Std. Coef
D. 05256
0. 003187
C. 0003255
0. 1152
Spoils?
58. 293 ?-sc
t-rat io
-0. ฃ3
32. 53
S. 47
:"t!
c'. 39 v

Analysis of'Variance

Source       DF            SS           PO
Regression.-- 5- .  ^1,63276..-    .-C. 37ฃ5ฃ
Error        13      ฃ..23253      D. 0^172
ToTai        24      1. 31537

-------
The regression eauation  is
C5 = 126 + 10.0 C6  -*- 0.0496  C7  -  ฃ30 2e  * 1.64 ซ-3  -  11.7

Predictor       Coef      Std.Coef      t-ratio
                                           7. 41
                                          11.64
                                           3.5E,
                                          -7- 33
                                           j. •
                                          -a.

S =. li.il       3-so = 33.ฃฃฃ     R-so(ad.i) = 91.411

Anaivsis of Variance

Source        DF           S3           MS
Renression     5      32753.3       6550.7
Error         19        ฃ388.9        125.7
Total         ฃ4      35142.3
Constant
C6
C7
CO
C9
C10
125. S0
10. 0436
0. 04957
-ฃZ2i. 41
1 . 64 1
-_ 1 . 723
16.37
0. 6630
0. 08844
31.45
1 . 309
1.303
MTB  >  REGRESS  Cll  5 C6-C10

The  regression ecmation is
Cll  =  5973  -  138 Cฃ - 0.49 C7 - 33S5 C3 - ฃ4.1 19  -  154
ฐredictor
Constant
C8
C7
28
u Z?
21 '2i
Coef
5973. 3
*"" ฑ ^ f • w>wf
-ป. 486
2334. 3
-24. 07
154. 45
Std. Coef
378, 3
13. ฃ4
1 . 37ฃ
701. 4
ฃ3. 18 •
ฃ3. ฃ0
t-rat io
15.75
-7. 15
~ฃ . ill!
5. ฃ3
— G. 33
cr '-IQ
_j . c. y
                      = 35.100    R-sa(adi) =  81.173
Analysis  of  Variance
Source        DF            SS "     '    'MS
Regression     5       6764025      1356805
Error         13      '1187763        62514
Total         ฃ4       7371733

-------
The renression ecuation  is
C12 = 3ฃ98 - 453 C6 - 3.51 C7  +  12133  C8 - 31.0 C9 + 553 C10
Predictor
Constant
Co
C7
C8 .
i™* •""
.""' ' ">
— * x1
Coef
3698
-459. 25
-3.513
12133
-91.123
=;^T "?~7
<_f -Jwซ j r
Std.Coef
1019
51. 85
5.31A
1890
78. 63
73. ฃ3
t-rat 10
5.51.
-8. So
-0.ฃ6
6. 42
-1. 1&
7.03
S = ฃ73.3       '-so = 33. ฃ51     Ft-sc(adi)  = Sฃ. 3ฃ7

     •5 is of Variance

     •a       DF           SE
Error
"oral
 j
13
      rcS
14343548
  45233:.
       6334333
 >-
               ci
 The reqression eouation is
 C13 = 12002 - 8ฃ7 CS - ฃ.ฃ0 C7 t-  21ฃ35  C8  - 156 C3 + 1043 C10
Pred ict or
Const ar.~
Cฃ
C7
^~ G
C3
C10
3 = ^13
S -, o E v 5 ' = " "

'E/jrce'""' ' '
;.5GrS = 5 j.0.''
ป 'r" *" O v"
" •;- ai -
Coef
12Q02
-3ฃฃ. 33
-ฃ. ฃ'23
21ฃ35
-155. 7
1043. 3
~,-50
var ia.r.c =

' r'r '
5 15
13 2
24 28
Stc. Coef
lฃS7
85. 73
8. 732
3127
130. 1
130. 2
	 ~> - ~ "••-ป> 3
— _/ - . -Jll-'t "


ฃS
B 1-3 11 04 .5
3312534
1773ฃ3ฃ
t— ra~ 10
7. 1;L
-10. 10
-0.75
ฃ. 32
-1.20
8. 01
"" "" 3 ""

a
rฃ '
1 ฃ32220
124ฃ7ฃS


-------
PITS >  REGRESS C14 5 C6-C10

The rsaression squat' ion is
C14 = 13407 - 1322 C6 - 3. 0 C7 + 31008 C3  -
                                                 C9  + 1566 C10
Predictor
Const ant
C6
C7
C8
C9
C10
Ccef
13407
-1322.4
-9.03
3100S
-200. 2
1565.6
Std.Coef
2280
116.0
11.88
4227
175.8
176.0
t-rat io
' 5.88
-11.40
-0.76
7.34
-1. 14
8.90
S = 1537        R-SQ = 33.042

Analysis of Variance

             DF
Source
Regression    5
Error        19
Total        ฃ4.-
                          SS
                   57ฃ30ฃ344
                    43144264'
                   ฃฃ0051200
                                 R-sq(adj)  =  91.211
                                       MS
                                 115381332
                                  2270751
MTB >  REGRESS CIS 5 C6-C10

The renression equation is
CIS = 8906 - 1667 C6 - 7.1,ฃ7
3red
/*•
(~O
Co
C7
CB
C3
ictor

ns-cant








~~ J.


-
Coef
89'Ziฃ
667.0
-7. 10
36106
131. 5
Std



. Co
ฃ5
1 ~? 1
13.
ef
Sฃ
• fcJ
48
t-rat

-

4794

Cl? %V 1307. 1
c
—
17,09
V51S Of
Source
^s
.pr
35 si on
Er<"or
"™ " .
: a
;

Var
SF
. ,,,5
-9
14
R-SQ =
iance'

s .. ,.as,7
crer
*-i -l
33. 387

SS
3.223.68
488452'
133
139



..4
. ฃ
- R-SQ




(adj

MS
ซj •
I ^!.
-0.
7.
-C.
9.
) =


i
ซ.
ฃ.
j—i
i.
6
53
c-
^J
g
"^1
4.
56



92. 435
.. ... :.. 	 	 s .-v. 	


.. .17.3464.480 ... . . , ,, - ,,-.
2920445
32231 aais. '

-------
MTB > REGRESS C16  5 Cฃ—Z10

The redression  eauation is
Clฃ = 5543'- 1826  Cฃ - 0. ฃ C7 + 35863 CS  -  105 C3 + 1373 ClS?
Prea ictor
Constant
Cฃ
•""7
Co
C3
ซ - I"*
Coef
5543
-182ฃ. 1
-0. ฃ1
356S3
-105. 0
1373.2
Std. COST
2344
144. ฃ
14.82
5272
213. 3
213.5
t-ra~ i.o
•3 c*
-12. ฃ2
-0. 04
ฃ.30
-ฃ:. 43
3. 02
S =1330         R-SQ  = 93.548    R-sa(adj)  =

Analysis of Variance                       "

Source       DF            SB     ,      Kฃ3
Rscre = sior:    5     3731'?24o4    13452:i?435
Error        13     ฃ711333ฃ      3532537
Total        24    1040221824
MTB > REGHES3 C17  5  C6-C10

The rearession equation is
C17 = 2325 - 1884  C6 + 6. 3 C7 + 33023 C8 + 4  C3  +  1320 C10
Prsd ictor
Constant
u- O t&
C7 " '
ca
C3
C10
S = 2182
Coef
2325
-1383. 7
8.8ฃ
33023
4. 5
-1323. 3
R-SQ: =
Std. Coef
3302
iฃ7. 3
17.21
6121
254. 7
254. 8
31,4ฃ8 R-so
t-rat io
0.89
-11.22
0. 51
5. 33
0. !?ฃ
7.54
(ad.i)- = 8-3. 223
Analysis of Variance

Source       DF            SS           MS
Regression    .5   ... 97.00i345.44 ... ,194000912,.
Error   "" ''" 13'   ' "9'047845ฃ      47ฃ2024
"j-al     „  24    10ฃ0453^C'3

-------
                    •C6-C10
 he rsrression sauation is
Ci8 = 783 - 1372 C6 + 21.0 C7 -t- 28375  CS + 143 C3 * 1766 C10
Predictor
Constant.
C6
C7
C8
C3
C10
S = 2335
Pinalvsis of
Source
Regression
Error
Total
V.~B > RE3RE
Coef Std.Coef t-rat
783 3624 . 0.
-1871.6 184.3 -10.
21.00 18.83 1.
28375 6713 4.
143.4 273.5 '0-
176.6.0 273.7 6.
y
R-sa = 89.225 R-sq(adj) =
Variance
DF SS MS
5 302565760 180513152
19 108937848 5736729
24 10115ฃ3584
35 C14 Z Ci-13
io
2ฃ
15
11
31
51
31
86. 389



fiie regression equation is
C14 = 16232 +0.0009 Cl + 0.308 C2 - 0.203 C3
Predictor
Constant
Cl
C2
C3
3 = 32M
"••:=.'-<=.'-•= '-.'-"
vr.,'~~5i,_r]
Tot al
Source
Cl
L. j
Coef Std.Coef t-rat
16292 1355 8.
0.00085 0.01201 ' 0.
0.30807 0.06225 4.
-0.20253 0.05336 -3.
R-sa = 65.323 R-sq(adj) =
v a r i a r. c s
D- SS MS
24 620051200 " 	 '"
DF Seq SS
1 23078470
T ฃ52764656 ' 	 -'"•'
1 1.1 31 3-217 1ฃ
io
33
07
35
41
: 60. 363


•,- •••.,. ..-.- . ....,-.--- •••• •- ..;/> ••<_;-., -| "•
••••-. A

-------
 •he regression  eauation is
 C14 = 30298 -  134  C5 * 22. B  C6

 --sdic'cov'        Coef      S~a. 3oef     t—ratio
 C :-•=•: an-     30293. 4           32.7      326.98
 -5          -133.786        .  1.221     -109.60
 '-=              22.79          17.54         1.30

 S  = 173.8        R-sa = 33.893     R-sa(adj)  =  99.883

 Analysis of Variance

 Source       DF            SS            MS
 Regression    2     619386752     309693376
 Error        22        664470     .    302123
DF        Sec SS
 1    619335744
           51012-7
               1
MTB > RE3RESS C14 ฃ C4  C5

The regression eouation is
C:4 = 312317 -•• ฃ7 C4 -  133 C5

^r-ecictor        Coef       S.td.Coef      t-ra-io
Constant      32i31ฃ.6          iiZlia. 0       3i?3. e-4
C4               ฃ=,. 7          1Sฃ. 9         I3.il
C5           -133. i35ฃ          l.ฃ0ฃ      -1HZI.G6

-3 = 1T(5. 7        P-SG =  99. ฃ23    R-SG(acj) = -rr. B~

Analysis of Variance

Source        J)~     '    '  ...S3 	:'. ........ flS ....,..-.".
Rfecre'ssiofi   ' :":ฃ " '  'ol934i 1ง4 ' :  ' 3ฎ'3&7C59ฃ'
                                       32275
Erป'or
To-tal
Source
C4
C3
c!ci!
24
•DF-
1
1
71 '3257
620I2512S4
SBQ- SS
224095344
395245792

-------
         WESTERN PROCESSING
     HYDROLOGIC CHARACTERIZATION
                 by

             C. M. Eddy
           J. M. Doesburg
Battelle Broject Management Division
Office of Hazardous Waste Management
    Richlandj Washington   99352
            March, 1985
            Prepared for
  Environmental Protection Agency
              Region X
     Seattle, Washington  98101

-------
               WESTERN  PROCESSING HYDROLOGIC CHARACTERIZATION

          Slug  tests •  and  borehole  dilution  tests  were performed  at the
Western Processing  Site (Figure 1)  to  better characterize the  groundwater
flow system.   The tests were performed  during the  period February  11-13,
1985, by  Jim Doesburg, Mary  Lilga,  and Chris  Eddy of Battelle's Office of
Hazardous  Waste   Management.     The   results   indicate   that  hydraulic
                                                       2
conductivities of the  zones  tested range  from 5 x 10"   cm/sec  (142 ft/day)
to  4 x  10   . cm/sec   (1  ft/day).    The  test  procedures  and  results are
discussed below.  The  field forms and plots used  to  calculate the hydraulic
conductivity are included in Appendices A  and  B, respectively.

                           SLUG  TEST PROCEDURES

          The  hydraulic conductivity  of  an  aquifer  can be  determined by
instantaneously  changing  the water   level  in  a  well   and  observing the
recovery.    For the Western  Processing  Site,  a  3.5-in.  diameter  slug was
used to  change the  water level.   The slug  was  designed  to  displace the
water 5,408  cm , which  should result in a 66-cm change in  water level in a
4-in. diameter  well.    The  recovery  was  measured  after  both  inserting and
removing the slug using an electrical tape.
          The change in water level  over  time  was then   plotted on semi-log
paper  and the  method   described in  Bouwer  and  Rice  (1976)  was  used to
analyze the results.

                             SLUG TEST RESULTS

          A  summary of the slug test  results  is   shown  in  Table 1.    Four
zones  of  permeabilities could  be discerned.    Conductivities  vary between
the  shallow  and  deep   portions  of  the aquifer,  and the  north  and   south
portions  of  the  site.   The shallow  portion of  the  aquifer  in  the north of
the site has a conductivity of about  10"   cm/sec (3 ft/day), while the  deep
                                              _2
portion has  a  conductivity  of  about 4  x 10   cm/sec   (113 ft/day).  The

-------
Boring Locations
                 N
              A
                                                         t 13
                             • South  196th  Street
                 ./••"






   Legend


 .17  Qround Wซlซf WปIU



     80' Monitoring Well



-•-  60' Dual Monitoring Wซll









0         150       300

         =i	=
     Scale in Feet
                                                   .4
                                       v:!
                                     //4-84-5
                                                     .11
                                                 .K

                                    * ~^84-4
                                      .24
.25


— x-
                                                                   4-1
                                                            .7   x
                                                          is
                                                              ^Fence
                                                             /  Line
                                                      ^84-3
                                                                 Darrves A Moore
                 Figure 1.  Western Processing  Well Locations

-------
opposite is  true of the  southern portion of  the site; the  deep  portion of

the aquifer has a conductivity  of about 3 x 10   cm/sec  (8.5 ft/day), while
                                                f)
the .shallow  portion  is  approximately  2  x  10"  cm/sec   (57  ft/day).   The

values  for  conductivity  can  be  quantitatively  verified  by comparing  the

calculated values with the  geology of the screened  interval  (Table  2).  The
                                                      2
fine-to-medium  sand  has  a  conductivity  of 3  x 10   cm/sec  (85  ft/day),

while the presents of peat, clay, and/or silt decreased the  conductivity to

6 x 10"4 cm/sec (1.7 ft/day).


                  Table  1.  Summary of  Slug Test Results

                                      Hydraulic
                                    Conductivity
                         Well  No.     (on/sec)*
1A
IB
17A
84-1B
84-2
84-3
84-4A
84-4B
84-5A
84-5B
2 x 10"?
8 x io;J
6 x 10 ;
2 x io;j
9 x 10 I
8 x 10",
2 x 10":
4 x 10";
4 x ID",
5 x 10"
                         *Average of the results for
                         dropping and removing the
                         slug (see Appendix B).
       Table  2.  Comparison  of  Hydraulic  Conductivity and Lithology

                           Conductivity
        Well No.   Depth     (cm/sec)     	Lithology	

           1A-      15 ft     2 x IQ'l     Sand with Silt/Clay Lenses
           17A      15 ft     6 x 10";     Silt and Clay
         84-4A     25 ft     2 x 10"f     Fine-to-Medium Sand
         84-5A     25 ft     4 x 10~T     Peat/Fine-to-Medium Sand
           IB      30 ft     8 x 10"Z     Peat/Fine Sand
         84-1B     50 ft     2 x ID");     Fine-to-Medium Sand
          84-2     50 ft     9 x 10".:     Fine Silty Sand
          84-3     50 ft     8 x 10"^     Fine-to-Coarse Sand
         84-4B     50 ft     4 x 10";     Sandy Silt
         84-5B     50 ft     5 x 10       Fine-to-Medium Sand

-------
          The values  for conductivity  used  in Battelle's  groundwater flow
and  contaminant  transport  model  of  the  Western  Processing  Site  were
1.2 x 10"3 cm/sec  (3.5  ft/day)  for the  upper  30  ft, and 1.8  x  10"2 cm/sec
(50 ft/day) for the portion of the aquifer below 30  ft.   The values closely
match the  values obtained  from  the slug  tests  in  the northern  portion  of
the  site,  while these  values  vary an  order of magnitude  in the  southern
portion of the  site.  Since the  majority of  the contaminants of interest in
the model  were  disposed of  in  the northern portion  (Reaction Ponds  I  and
III) of  the  site,  the  difference  between the model  predicted and  measured
hydraulic  conductivities  will  probably  have little  effect  on the  overall
transport of contaminants.

                     BOREHOLE DILUTION TEST PROCEDURES

          Borehole  dilution  testing   as   described  by  Freeze  and  Cherry
(1979)   has  been used  extensively  in  Europe  on  a means  of  determining
groundwater  velocity.    The  theory  is  that   the   horizontal  velocity  of
groundwater through  a  well-bore can  be  determined  by measuring  the change
in concentration over time  of  a specific  ion introduced  into  the borehole.
This measurement  is  performed by  packing  off  a known portion of  a screen,
introducing a  known  ion,  and  measuring the  changes in  concentration  over
time.    Figure  2  is a schematic  of  the  apparatus  used   at the  Western
Processing Site.

                      BOREHOLE DILUTION  TEST RESULTS

          The borehole  dilution  tests  met  with mixed   results.    Several
equipment  problems  and  difficult  field  conditions  resulted  in tests  being
performed on only two boreholes.   Both  Battelle pumps  refused to  work  (both
were lab tested  prior  to field work),  and the  conductivity meter would not
function properly.   The  conductivity meter gave anomolous measurements when
the probe  was moved  near the lower mixing tube.   When the  probe was  moved
near  the  upper  mixing  tube,   the   anomolous measurements  ceased   (see

-------
Strip Chart Recorder
  T_   Conductivity Meter
     =0-
         Screen  <
                      Injection  Port
                                     Pump
                                             Ground  Surface
                       -*-Well Casing
                       i-* Packer
Outlet


Conductivity Probe


Inlet
       Figure 2.   Borehole Dilution  Test  Apparatus

-------
Figure 2).  We  obtained  a third pump from  the EPA  Manchester  lab which did
work.   Further  modifications  have been  made  to   the  test  apparatus  to
eliminate the.probe problem in the future.
          The tests  that  were  performed did  show   the  expected  change  in
conductance with time.  Because  of  the  errors in the  initial  readings,  the
calculated  velocities  were  suspect.    The  borehole  tests  indicated  that
horizontal groundwater flow near Well  17A  was much  less  than  that  of  Well
IB.   Unfortunately,  the  length  of  time that was  required to complete  the
test at Well  17A was longer than the time available  on  site.

                                 REFERENCES

Bouwer, H., and R. C. Rice.   1976.   "A  Slug Test for  Determining Hydraulic
Conductivity   of   Unconfined   Aquifers  with   Completely   or    Partially
Penetrating Wells."  Water Resources Research, Vol.  12, No.  3,  pp. 423-428.

Freeze, R. A., and J. A.  Cherry.  1979.  Groundwater.   Prentice-Hall,  Inc.,
Englewood Cliffs,  NJ.

-------
      APPENDIX  A



SLUG TEST FIELD FORMS

-------
SITE Western Processing
                                                        Page  1  of
Date 2/11/85  10:30
                             SLUG TEST FIELD FORM
Investigator  J. Doesburg
              C. Eddy
              M. Lllga
Borehole I
Radius of borehole  4 in./lO cm

Radius of casing    2 in./5 cm

Depth of well 	443 cm	
Length of screen     3 ft./91 cm

Reference point for
water level  measurments  Top of  PVC
 Casing
Elevation Steel Casing  23.38  ft. AMSL 713 cm

Static water level  250  cm  (2/ll/85)-246.5 cm  (2/13/85)

Static water level elevation M66  cm

Ground surface elevation _NA	

Slug volume   330 cu.  in. /5408 cm3

Anticipated  displacement 26  in./66 cm
Remarks: * Stop.watch  had  to  be  restarted  at  approximately
          30 minutes.

          This well was installed with  a  backhoe  (F.  Wolf verbal
          communication).  There is a large  gravel  pack  around
          the screen.

          Water  level  in  well dropped 21  cm  overnight.   Reading
          8:30 a.m.-2/12/85.  Puddles which  had  been  6"  deep in
          vicinity  were gone  in morning.
T

-------
                              WELL '#  1A
SLUG TEST FIELD FORM
1



1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

TIME


DROPP
0:00
0:05
0:15
0:25
0:35
0:45
0:60
1:15
1:30
1:45
2:00
2:15
2:30
2:45
3:00
3:30
4:00
4:30
5:00

DEPTH
TO
WATER


NG SLUG
250
221
229
232
"234
237
237.5
238.5
239
240
240.5
241
241.5
241.5
242
242.5
242.5
242.5
242.75

STATIC
MINUS
TEST



0
29
21
18
16
13
12.5
11.5
11
10
9.5
9
8.5
8.5
8
7.5
7.5
7.5
7.25

























#
20
21
22
23
24
25
26
27
28
29

1
2
3
4
5
6
7
8
9
10
11

TIME
6:00
7:00
9:00
11:00
15:00
20:00
25:00
* SEE
35:00
45:00
2:52:00
REMOVI
0:15
0:30
0:40
0:50
1:00
1:10
1:20
1:30
1:40
1:50
2:05

DEPTH
TO
WATER
242.75
243
243.5
244
244
245
244.5
REMARK
244.5
244.5
250
G SLUI
267
257
252.5
251
249
148
246.5
245.5
244.5
244
242.5
STATIC
MINUS
TEST
7.25
7
6.5
6
6
5
5.5
5.5
5.5
0

17
7
2.5
1
-1
-2
-3.5
-4.5
-5.5
-6
-7.5
















.








-------
                               Paqe  _3_ of 5_



                               WELL  #  1A
SLUG TEST  FIELD  FORM
1
^^^^^M
12
13
••i^-——
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32

TIME
2:15
2:30
2:45
3:00
3:15
3:30
4:00
4:30
5:00
5:30
6:00
7:00
8:00
10:00
12:00
14:00
17:00
22:00
28:00.
47:30
57:00

DEPTH
TO
WATER
242.5
242
241
241
241
241
240
239
238
237.5
236.5
235
234.5
234
234
234
234
233.5
233.5
235
234

STATIC
MINUS
TEST
-7.5
-8
-9
-9
-9
-9
-10
-11
-12
-12.5
-13.5
-15
-15.5
-16
-16
-16
-16
-16.5
-16.5
-15
-16
























#
34
35
36
37


















TIME
1:10:0
l:29:/)l
1:46:01
.8:26:01


















DEPTH
TO
WATER
) 235
) 236
) 236
1 257


















STATIC
MINUS
TEST
-15
-14
-14
7










































-------
Page 4_ of J_
WELL #   1A

#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
. 22

RET
TIME
0:00
0:05
0:16
0:24
0:31
0:39
0:47
0:57
1:10
1:22
1:35
2:00
2:20
2:40
3:00
3:30
4:00
4:30
5:00
5:45
6:30
7:15

ST
^••^^^^••^•••^
DEPTH
TO
WATER
246.5
217:5
225
228
230
232
233.5
235
237
238
239
240
241
242
242
242.5
243
243.5
244
244
244.5
245


• i -••
STATIC
MINUS
TEST
0
29
21.5
18.5
16.5
14.5
13
11.5
9.5
8.5
7.5
6.5
5.5
4.5
4.5
4
3.5
3
2.5
2.5
2
1.8

SLUG TES





















.
— .
••
T FIELE
# .
23
24
25
26
27
28
29
30

1
2
3
4
5
6
7
8
9
10
	
11
—
12
— 	
13
"
) FORM
•i . i i
TIME
8:00
9:09
10:00
12:00
14:00
17:00
20:07
3:30:39
SLUG Rl
0:16
0:27
0:40
0:55
1:13
1:30
1:45
2:00
2:20
2:40
_
3:00
—
3-20
-
3' 40
M—

DEPTH
TO
WATER
245
245.2
245.3
245.5
245.8
245.9
245.9
246
MOVED
259.5
259.5
259.5
259
257.5
256.5
256.5
255.5
255
254
—
253.5
"
OC 0
ฃ30
•
DCO C
CJฃ . J
—

STATIC
MINUS
TEST
1.5
1.3
1.2
1.0
.7
.6
.6
0

13
13
13
12.5
11
10
10
9
8.5
7.5
7
6r
. 0
V^BMI^^^^MH^^M






















. 	 . 	
	 	 	
	
	 • —

-------
                             Page 5_ of 5	



                             WELL # 1A
SLUG TEST FIELD FORM
#
14
15
16
17
18
19
20
21
22
23
24
25
26









TIME
4:00
4:30
5:00
5:45
6:30
7:15
8:00
9:00
10:00
11:00
13:40
15:00
17:00









DEPTH
TO
WATER
25.2
251.5
251
250.5
249.5
249
248.5
248
248
247.5
247
247
246.7









STATIC
MINUS
TEST
5.5
5
4.5
4
3
2.5
2
1.5
1.5
1
1
1
.2
































#






















TIME






















DEPTH
TO
WATER






















STATIC
MINUS
TEST














































-------
                                                        page i or
SITE  Western Processing
Date  2/11/85   10:00
                             SLUG TEST FIELD  FORM
Investigator     J. Doesburq
                 C. Eddy
                 M. Lilga
Borehole #  IB
Radius of borehole 10 cm
/////////7
Radius of. casing 5 cm
Depth of well 914 cm 	 - -
Length of screen 91 cm

Reference point for
water level measurments Top of PVC
Casing
Elevation steel Casing 788 cm AMSL
Static water level 276 cm
Static water level elevation -^512 cm
Ground surface elevation NA
Slug volume 330 cu. in. 75408 cm3
Anticipated displacement 66 cm










rif 	 ป•
*- t.*i

— 2j?w —



1



r





•yj





/ ii n
snfnt







-J-
i
L
1
j
Remarks:  Cold, wet.   Intemittent  rain

-------
                              Page 2_ of 2



                              WELL #  IB
SLUG TEST FIELD FORM
#



1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
TIME


DROPPI
0:00
0:05
0:15
0:25
0:35
0:45
0:55
1:05
1:15
1:25
1:35
1:45
1:55
2:05
2:20
2:40
2:55
3:10
3:25
DEPTH
TO
WATER


NG SLUG
276
217
218
220
222
224
228
228
230
231
233
234
235
237
239
242
243
244
246
STATIC
MINUS
TEST



0
59
58
56
54
52
48
48
46
45
43
42
41
39
37
34
33
32
30























#
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38



TIME
3:40
4:00
4:30
5:00
5:30
6:00
7:00
8:00
9:00
10:00
12:00
14:00
16:00
18:00
20:00
22:00
24:00
26:00
28:00



DEPTH
TO
WATER
247.5
249
251
253
254.5
256
259
261.5
264
265
268
269
271
?72.5
273.5
275
275
?75.5
276



STATIC
MINUS
TEST
28.5
27
25
23
21.5
20
17
. 14.5
13
11
8
7
5
3.5
2.5
1
1
.5
0














-












-------
SITE  Weytena Processing
Date 2/11/85  09:00
                                                        Page 1 of  1
                             SLUG TEST FIELD  FORM
Investigator  J. Doesburq
              C. Eddv
              M. Lilqa
Borehole I    HA
Radius of borehole  10 cm

Radius of casing  5 cm
Depth of well  457 cm
Length of screen  91 cm
Reference point for
water level measurments

	Casing	
Top of PVC
Elevation   Steel  Casing 767  cm AMSL

Static water level   284 cm	
Static water level elevation -M83  cm

Ground surface elevation   NA	
Slug volume   330  cu.  in. /5408 c

Anticipated displacement   66  cm
                                                 rrrr,
                                                                           \snnc
Remarks:  Recovery in  about  5  seconds.
          Maximum displacement about 8 .cm.

          Using 8 cm as  yo and .1  and 5  sec.  at y0and t gives a K of
          2X10-1  cm/sec.
          Well  11B PVC  casing is  broken.

-------
SITE  Western Processing,
                                                        Page  1 of
Date  2/12/85  08:50
Investigator   J.  Doesburg
               C.  Eddy
               M.  Lilga
                             SLUG TEST FIELD  FORM
Borehole #
Radius of borehole  10 cm
Radius of casing   5 cm
Depth of well   524 cm
Length of screen  91 cm
Reference point for
water level measurments    Top
 Casing-East Side(PVC Cut on Slant)

Elevation   Steel  Casing 741 cm AMSL

Static water level  196 cm (2/11/85)  205 cm (2/12/85)

Static water level elevation ^536 cm

Ground surface elevation   NA	

Slug volume   330  cu. in. 75408 cm3

Anticipated displacement    66 cm	
Remarks: 2/11/85  Raining
         2/12/85  Sunny

         Well 17B has screwed joints.


15)

^L



t
*-*-JA
h
T
L
••

\
!
\

-------
                              WELL #  17A
SLUG TEST FIELD FORM
#



1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

TIME


DROPP
0:00
0:05
0:15
0:30
0:40
0:45
0:55
1:10
1:25
1:35
1:45
2:00
2:10
- 2:20
2:35
2:50
3:00
3:10
3:30

DEPTH
TO
WATER


NG SLUC
205
142
147
151.5
153.5
156
158
162
164.5
167
168
170
171
173.
176
178
178.5
180
181

STATIC
MINUS
TEST

.

0
63
58
53.5
51.5
49
47
43
41.5
38
37
35
34
32
29
27
26.5
25
24

























#
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

TIME
3:40
3:55
4:10
4:25
4:35
4:55
5:15
5:25
5:40
6:00
6:30
7:00
7:30
8:00
8:30
9:00
9:30
10:00
11:00
12:00
13:00
14:45

DEPTH
TO
WATER
182
184
185
186
187
190
191
191.5
192
193
195
196
197.5
198
198.5
200
200.5
201
202
202.5
203
203
STATIC
MINUS
TEST
23
21
20
19
18
15
14
13.5
13
12
10
9
7.5
7
6.5
5
4.5
4
3
2.5
2
2























!

-------
                              Page  3_ of 4	



                              WELL  #  17A
SLUG TEST FIELD FORM
#
42
43
44
45
46
47
48
49
50

1
2
3
4
5
6
7
8
9
10
11
12
TIME
16:00
17:00
18:00
20:00
22:00
25:00
30:00
35:00
51:00
REMOV
0:08
0:25
0:40
0:50
1:00
1:15
1:30
1:45
2:00
2:10
2:30
2:45
DEPTH
TO
WATER
203.5
204
204
204
204
204
204
"204
204
NG SLUG
274
269
267
267
266
266
265
263
262
261.5
260.5
259.5
STATIC
MINUS
TEST
1.5
1
1
1
1
1
1
1
1

69
64
62
62
61
61
60
58
57
56.5
55.5
54.5























#
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
TIME
3:00
3:20
3:50
4:10
4:30
5:00
5:30
6:00
6:30
7:15
8:00
9:00
10:00
11:00
12:00
13:00
14:00
15:30
17:00
18:30
20:00
22:00
DEPTH
TO
259
258.5
256.5
255.5
255
253
252
250.5
249
247
246
243.5
241
240
237
235.5
234
231
229
227.5
226
224
STATIC
MINUS
TEST
54
53.5
51.5
50.5
50
48
47
45.5
44
.42
41
38.5
36
35
32
30.5
29
26
24
22.5
21
19
























-------
                              Page _4_ of  4



                              WELL I  UA
SLUG TEST FIELD FORM
#
35
36
37
38
39
40
41















TIME
24:00
27:00
30:00
35:00
40:00
50:00
1:46:0
Stop '














DEPTH
TO
WATER
221.5
219
216.5
214
212
209
) 205
/atch re














STATIC
MINUS
TEST
16.5
14
11.5
9
7
4
at 11:3
set at t





















i actual tim
me 10:53, s














#






t
top wat














TIME







:h time














DEPTH
TO
WATER







1:04














STATIC
MINUS
TEST






























r















-------
oITc _U
                                                        Page 1 of
                      nn
                             SLUG TEST  FIELD  FORM
Date
Investigator     j. Doesburq
                 r. Eddv
                 M. Liloa
Borehole #  I?B
Radius of borehole IQ cm
Radius of casing 5 cm
Depth of well 914 cm
Length ;6f screen 91 cm
Reference point for
water level measurements  Top of  PVC
 Easing
Elevation Steel Casing 736  cm AMSL

Static water level   235  cm	
Static water level elevation -\, 451  cm

Ground surface elevation    NA	
Slug volume   330  cu.  in./54Q8 cm3

Anticipated displacement   66  cm
I
Remarks:  Joints  are  screwed  together so  slug  won't
          fit  in  hole.

-------
SITE   Upstern Processing,
                                                        Page  1 of
Date   y/13/85
Investigator    c. Eddv
                H. L11g?
Borehole #   84-1-A
                             SLUG TEST  FIELD FORM
Radius of borehole  10 cm
Radius of casing  5 cm
Depth of well  762 cm
Length of screen  152 cm
Reference point for
water level measurements  Top of PVC
Casing
Elevation  NA
Static water level  302
cm
Static water level elevation

Ground surface elevation   N
Slug volume    330 cu. in. /54DR cm3

Anticipated displacement  66  cm
                                                                           I/////
Remarks: Unable to  slug test  due  to  presence  of screws in casing

-------
       Western Processing
Date   2/13/85
Investigator     c. Eddy
                 M. Lilga
Borehole #   84-1-B
                                                        Page 1 of
                              SLUG  TEST  FIELD FORM
Radius of borehole   10 cm
Radius of casing   5 cm
Depth of well   1524 cm
Length of screen   152 cm	
Reference point for
water level measunnents    Top of PVC
Casino
Elevation   MA
Static water level   321.5 cm
Static water level elevation   NA
Ground surface elevation   NA
Slug volume   330 cu. in./54Q8 cm3
Anticipated displacement  66 cm

Remarks:
STATIC.  U

-------
                              WELL I  84-
SLUG TEST FIELD FORM
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

1
2
3
4
5
6

TIME
INilJ
0:.00
0:08
0:21
0:30
0:38
0:47
1:05
1:34
2:32
2:49
4:00
5:41
5:54
6:30
8:11
REMOVI
0:11
0:25
0:33
0:37
1:00
1:44

DEPTH
TO
WATER
AL WATE
321.5
347
324
322.5
322
322
322
322
322.5
322.5
322.5
323
323
322.5
322
NG SLUG
322.5
324
323.5
322.5
322.5
322.5

STATIC
MINUS
TEST
0
25.5
2.5
1
.5
.5
.5
1
1
1
1
1.5
1.5
1
.5

.5
2
1.5
.5
.5
.5




DEPTH TO
WATER
INCREASES



















#
7
8
9




















TIME
2:30
3:27
4:10
RETEST
0:10
0:17
0:24
0:36
1:07
2:07
REMOVI
0:00
0:12
0:24
0:34
0:40
0:53
1:22
2:09
3:57
4:45


DEPTH
TO
WATER
322.5
322.5
322.5

355.5
327
323.5
322
322
322
^G SLU
322
325
324
323
322.5
322.5
322.5
322.5
322.5
322.5

STATIC
MINUS
TEST
.5
.5
.5

13.5
5
1.5
0
0
0

0
3
2
1
.5
.5
.5
.5
.5
.5









-.







.








-------
SITE
                                                        Page 1 of
                       no
                              SLUG  TEST FIELD FORM
Investigator   j. Doesburo
                  Eddv_
               M. Liloa
Borehole #    84-2
              Dames and Moore wel>sir,ale well, middle  of  eastern  side
              of site adjacent to fence.
Radius of borehole  IQ cm
Radius of casing  5
cm.
Depth of well  1524 cm
Length of screen  152 cm
Reference point for
water level measurements    TOD of PVC
Casino
Elevation
Static water le-vel   266.5
                           cm.
Static water level elevation

Ground surface elevation    i\
Slug volume   330 cu. in. /5408

Anticipated displacement  66 cm


Remarks:
                                                         STATIC,
                                                                                  H

-------
                              WELL *  84-2
SLUG TEST FIELD FORM
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

TIJ1E
INITIA
0:00
0:05
0:15
0:25
0:35
0:45
0:55
1:05
1:15
1:25
1:35
1:45
1:55
2:10
2:20
2:30
2:40
2:50
3:00
3:15
3:30
3:45

DEPTH
TO
WATER
. WATER
266. 5
208
212.5
215.5
218.5
220.5
222.5
"225
227
229
231.5
232.5
235
237
239
240.5
242
242.5
244
245.5
246.5
248.5

STATIC
MINUS
TEST
LEVEL
n
58.5
54
51
48
46
44
41.5
39.5
37.5
35
34
31.5
29.5
27.5
26
24.5
24
22.5
21
20
18

























1
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42



TIME
4:00
4:15
4:30
4:45
5:00
5:20
5:45
6:05
6:30
7:00
7:30
8:00
9:00
10:00
11:00
12:00
13:00
14:00
16:00
18:00



DEPTH
TO
WATER
249.5
251
252.5
253.5
255.5
255.5
257
257.5
259
260
261
261.5
262.5
263.5
264.5
265
265.5
265.6
266
266.5



STATIC
MINUS
TEST
17
15.5
14
13
11
11
9.5
9
7.5
6.5
5.5
5
4
3
2
1.5
1
.9
.5
0

























1

-------
                              Page _2_ of



                              WELL #  84-2
SLUG TEST FIELD FORM
#

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Ifl
19
20

TIME
REMO
0:20
0:45
1:00
1:10
1:20
1:25
1:35
1:40
1:50
2:00
2:10
2:20
2:30
2:55
3:25
3:45
4:00
4:15
4:30
5:00

DEPTH
TO
WATER
flNG SLt
321.5
319
314
312
310
309
308
306
305
303
302
300
298.5
295
291
289.5
?R7.5
?Rfi
285
283

STATIC
MINUS
- TEST
G
55
52.5
47.5
45.5
43.5
42.5
41.5
39.5
38.5
36.5
35.5
33.5
32
28.5
24.5
23
?1.
1Q.R
18.5
16.5
























#
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35







TIME
. 5-: 30
6:00
6:30
7:00
7:30
8:00
9:00
10:00
11:00
12:00
13:00.
14:00
15:00
16:20
17:20







DEPTH
TO
HATER
281
278.5
277
276.5
275
274
271.5
270.5
269.5
268.5
268
267.5
267
267
267







STATIC
MINUS
TEST
14.5
12
10.5
10
8.5
7.5
5
4
3
2
1.5
1
.5
.5
.5































-------
SITE. uPS|0rn Processing
Date  p/17/85
Investigator    c. Eddy
                M- Liloa
Borehole #   84-3
                                                        page  1 of	3
                             SLUG TEST  FIELD  FORM
Radius of borehole  10  cm
Radius of casing   5  cm
                                                    ////////
Depth of well   1524 cm
Length o'f screen   152 cm	
Reference point for
water level measurments   Top of PVC
Casin
Elevation  MA
Static water level   270 cm
Static water level elevation 	NA
Ground surface elevation   NA	
Slug volume
               330 cu. in. 75408  cm3
Anticipated displacement    66 cm

Remarks:
                                                                          -T777/

-------
                             Paqe J_ of J_



                             WELL #   84-3
SLUG TEST FIELD FORM
#
. •—
1
2
3
4
5

1
2
3
4
5
6
7
8
9
10
11

1
2
3
4

TIME
"INITI
0:00
0:10
0:30
0:55
1:15
REMOV
0:13
0:25
0:35
0:45
1:00
1:15
1:30
1:45
2:00
2:20
2:40
DROPPI
0:20
0:30
0:40
0:50

DEPTH
TO
WATER
IL WATEf
270
289
270.5
270
270
NG SLUC
288
278
274
273
272
271.5
271
271
270.5
270.5
270
NG SLUG
273
270
-269
268

STATIC
MINUS
TEST
LEVEL
0
1 19
.5
0
0

18
8
4
3
2
1.5
1
1
.5
.5
0

3
0
-1
-2



DEPTH TO
WATER INCRE















DEPTH TO
WATER INCR



#
5
6
USES
7
8
9
10

1
2
3
4
5
6
7
8
9
10
11
12

1
2

TIME
1:00
1:15
1:30
1:45
2:00
2:20
REMOV I F
0:1?
0:24
0:30
0:40
0:55
1:05
1:20
1:40
2:00
2:20
2:40
3:00
DROPPI
0:08
0:21

DEPTH
TO
WATER
269
269
269
269.5
269.5-
270
G SLUC
285
277
275
273
272.5
272
272
271
271
270.5
270.5
270
NG SLU
281
277

STATIC
MINUS
TEST
-1
-1
-1
-.5
-.5
0

15
7
5
3
2.5
2
2
1
1
.5
.5
0
s*
b
11
7






















DEPTH TO
WATER INCR.


-------
                              WELL #
SLUG TEST FIELD FORM
#
3
4


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

TIME
0:29
0:39
REMOV
0:00
0:18
0:28
0:34
0:41
0:48
0:58
1:10
1:20
1:30
1:40
1:50
2:00
2:20
2:40
3:00
3:30
4:00

DEPTH
TO
WATER
271.5
270
[NG SLU(
270
276.5
274.5
274
273
273
272.5
271.5
271
271
272
271.5
271.5
271
271
271
270,5
270.5

STATIC
MINUS
TEST
1.5
0

0
6.5
4.5
4
3
3
2.5
1.5
1
1
2
1.5
1.5
1
1
1
.5
.5










-













1


•



















TIME






















DEPTH
TO
WATER






















STATIC
MINUS
TEST






































•







-------
SITE  Western Processing
                                                        Page 1 of
Date  2/12/85
Investigator    c-
                M.  Lilga
Borehole #   84~4A
                             SLUG TEST  FIELD  FORM
Radius of borehole  10 cm
Radius of casing  5 cm
Depth of well   762 cm
Length of screen   152 cm	

Reference point for
water level measurments    TฐP of pvc
Casing
Elevation
             NA
Static water level   268 cm
Static water level elevation

Ground surface elevation     "A
Slug volume   330 cu. in. /5408 cm3

Anticipated displacement    66 cm
Remarks:
                                                                    z*
                                                                 t
H

-------
                               Paqe _2_ of J_



                               WELL # 84-4A
SLUG TEST FIELD FORM
#
1
2
3
4
5
6
7
R
9
10
11
12
13
14

1
2
3
4
5
6
7
TIKE
INITIA
0:00
0:10
0:20
0:35
1:00
1:15
1:35
1 !45
1:55
2:05
2:15
2:30
2:45
3:30
REMOV
0:22
0:35
0:46
0:55
1:05
1:15
1:25
DEPTH
TO
WATER
. WATER
268
255
267
268
266
266
266.5
7fi7
267
267
267
267
267
267
NG SLUG
287
277.5
274
273
271.5
270.5
270
STATIC
MINUS
TEST
LEVEL
0
13
1
0
2
2
1.5
1
1
1
1
1
1
1

19
9.5
6
5
3.5
2.5
2























#

8
9
10
11
12
13
Id
15
16

1
2
3
4
5
6
7
8
9
10
11
TIME

1:35
1:50
2:00
2:15
2:30
3 .- nn
v^n
4:15
5:00
DROPPI1
0:30
0:40
0:50
1:00
1:10
1:20
1:40
1:50
2:05
2:25
3:00
DEPTH
TO
WATER

269.5
269.5
269
269
269
?sq
?fiQ
268.5
268
G SLU(
263.5
263
264.5
265
265.5
265.5
266
267
267.5
267.5
267
STATIC
MINUS
TEST

1.5
1.5
1
1
1
1
i
.5
0

4.5
5
3.5
3
2.5
2.5
2
1
.5
.5
1
























-------
                              Pa.qe J_ of 3	



                              WELL #   84-4A
SLUG TEST FIELD  FORM
1
12
.13


1
2
3
4
5
6











TIME
3:30
4:00
REMOVI
0:00
1:05
1:20
1:45
2:00
2:15
2:35











DEPTH
TO
WATER
267.5
268
NG SLUG
268
274
269
269.5
269
269
268











STATIC
MINUS
TEST
.5
0

0
6
1
1.5
1
1
0

































#







-













TIME





















DEPTH
TO
WATER





















STATIC
MINUS
TEST




























f















-------
 SITE Western Processing
                                                         Page 1 of _4_
 Date 2/12/85
                              SLUG TEST FIELD  FORM
 Investigator    c. Eddv
                 M  Lilaa
 Borehole #    84-4B
 Radius of borehole  10 cm
 Radius of casing  5 cm

 Depth of well     1524 cm
 Length "of screen   152 cm

 Reference point for
 water level  measurements
Top of PVC
Casing
 Elevation  MA
 Static water level    269 cm
 Static water level  elevation    NA

 Ground surface  elevation     NA
 Slug  volume   330 cu. in./54Q8 cm3

 Anticipated  displacement     66  cm
                                                                           17777
 Remarks:    "Fizzing"  noise  noted  in  well  during first part
             of  slug test.

-------
                              Page _ฃ_ of  4



                              WELL #  84-4B
SLUG TEST FIELD FORM
1
.^^— ^—
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
, Ifl
19
20
2J
22
TIME
TMTIA
0:00
0:08
0:20
0:35
0:45
0:55
.1:10
1:25
1:40
1:50
2:00
2:15
2:30
2:45
3:00
3:30
3:50
4:15
4:40
5:00
5:20
5:40
DEPTH
TO
WATER
. WATER
269
205
208
209
210
211
2,12
212
213
213.5
213.5
214.5
215
215
216.5
217.5
218.5
220
221
221.5
222
223
STATIC
MINUS
TEST
LEVEL
0
64
61
60
59
58
57
57
56
55.5
55.5
54.5
54
54
52.5
51.8
50.5
49
48
47.5
47
46























#
23
24
25
26
27
28
29
30
31
32
33
34
35
36
.37
38
39
40
41
42
43
44
TIME
6:00
6:20
6:40
7:00
7-: 30
8:00
8:35
9:00
9:45
10:30
11:15
12:00
13:00
14:10
15:15
16:30
18:00
19:30
21:00
23:00
25:00
27:45
DEPTH
TO
WATER
224
225
225.5
226
227
228.5
228.5
230
231
232
233
234.5
236
237.5
239
240.5
242.5
245
246
248
249
251
STATIC
MINUS
TEST
45
44
43.5
43
42
40.5
40,5
39
38
37
36
34.5
33
31.5
30
28.5
26.5
24
23
21
.20
18
























-------
                              Paqe JL of A



                              WELL # 84.aa
SLUG TEST FIELD FORM
#
45
46
47
48
49
50
51
52
53

54
55
-55
57
58
59
60
61
62
63
64
65
66

TII1E
30:05
33:00
36:00
40:00
45:00
50:00
1:00:0
1:10:0
1:20:0
REMOVI
0:00
0:10
0:15
0:30
0:40
0:50
1:00
1:10
1:20
1:35
.1:45
2:00
2:20

DEPTH
TO
WATER
252.5
255
256
258
260
261.5
] 264
) 266.5
3 268
JG SLUG
269
3?7
328
326
324
323
321.5
320
319.5
317
316
314.5
310.5

STATIC
MINUS
TEST
16.5
14
13
11
9
7.5
5
2.5
1

0
58
59
57
55
54
52.5
51
5C.5
48
47
45.5
41.5

























#
67
68
69
70
71
72
73
74
75
7&
77
78
79
80
81
82
83
84
85
86
87
88

TIME
2:35
2:50
3:00
3:15
3:30
3:45
4:00
4:20
4:40
5:00
5:30
6:00
6:30
7:00
8:00
9:00
10:00
11:00
12:00
14:00
16:00
18:00

DEPTH
TO
WATER
310
308.5
307.5
306
304
303
301.5
300
298
296
294
292
290
289
285.5
282
280
278
277
274.5
272.5
271.5
•
STATIC
MINUS
TEST
41
39.5
38.5
37
35
34
32.5
31
29
27
25
23
21
20
16.5
13
11
9
8
5.5
3.5
2.5








•

















-------
                              Page  4_ of   4



                              WELL  #  84-4B
SLUG TEST FIELD FORM
#
89
90
91
92
93

94?
















TIME
20:00
22:00
25:20
28:00
32:00
33 mir
1:10:0(
















DEPTH
TO
WATER
270
270
270
270
270
utes is
271,









1






STATIC
MINUS
TEST
1
1
1
1
1
3:49 p.n
at 4:26























p.m.
















#























TIME























DEPTH
TO
WATER






















STATIC
MINUS
TEST














































	 . — ___

-------
                                                         Page 1 of  3
SITE   Upstprn Processing
 Date   ?/i?/85
 Investigator
                              SLUG TEST FIELD FORM
      C. Eddy
                   M. Lilga
 Borehole  #   84-5A
 Radius  of borehole   10  cm


 Radius  of casing    5  cm
 Depth  of well    762 cm
 Length  of  screen   152 cm
 Reference  point  for
 water  level  measurements   Top of PVC
Casing
 Elevation
NA
 Static  water level    241 cm
 Static  water level  elevation    NA


 Ground  surface  elevation   NA
 Slug  volume    330 c'u. in. /54Q8 cm3
                         /

 Anticipated  displacement   66 cm




 Remarks:

-------
                              Page _ฃ_ of J_
                              WELL #  84-5A
SLUG TEST FIELD FORM
#
1
2
3
4
5
6
7
8
9
10
11
12
1-3
14
15
16
17
18
19
20
21
22
~-
TIME
INIT:
0:00
0:12
0:27
0:39
0:51
1:05
1:15
1:25
1:35
1:45
2:00
2:15
2:30
2:45
3:00
3:20
3:40
4:00
4:30
5:00
5:30
6:00
DEPTH
TO
WATER
AL WATE
241
201
197
202
202.5
204
204.5
205
206
206.5
207
208
209
210
211
212.5
213
214
215
216
217
218
STATIC
MINUS
TEST
R LEVEL
0
40
44
39
38.5
37
36.5
36
35
34.5
34
33
32
31
30
.28.5
28
27
26
25
24
23























#
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

1
2
3
4
TIME
6:45
7:30.
8:15
9:00
10:00
11:00
12:30
14:00
15:30
17:00
18:30
20:00
22:00
24:00
27:00
30:00
35:00
REMOVI
0:05
0:14
0:22
033
DEPTH
TO
WATER
220.5
221
222.5
223
225
227
228
229.5
231.5
232
233
234
236
237.5
238
239
241
NG SLU
303
301
300
297
STATIC
MINUS
TEST
20.5
20
18.5
18
16
14
13
11.5
9.5
9
8
7
5
3.5
3
2
0
3
62
60
59
54
























-------
                              Page _1 of f_



                              WELL #  84-5A
SLUG TEST FIELD FORM
#
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

TIKE
0:44
0:50
1:10
1:20
1:40
1:50
2:05
2:15
2:30
2:45
3:00
3:20
3:40
4:00
4:30
5:00
5:30
6:00
6:45
7:30
8:15
9:02

DEPTH
TO
WATER
295
294
292
290
288
286
285
284
282
280.5
279
277
275
273.5
271.5
269.5
267.5
265.5
262.5
260
259
257

STATIC
MINUS
TEST
54
53
51
49
47
45
44
43
41
39.5
38
36
34
32.5
30.5
28.5
26.5
24.5
21.5
19
18
16

























#
27
28
29
30
31
32
33
34
35
37
38












TIME
10:00
11:00
12:00
13:30
15:00
17:00
19:00
22:00
25:00
29:00
33:07












DEPTH
TO
WATER
254
253
251.5
249.5
247
246
245
244
243
242
241












STATIC
MINUS
TEST
13
12
10.5
8.5
6
5
4
3
2
1
0





































-------
SITE   intern  Processing
                                                        Page  1  of _2_
Date   2/13/85
                             SLUG TEST FIELD  FORM
Investigator    C.  Eddy
                M.  Lilga
Borehole #   84-5B
Radius of borehole   10 cm

Radius of casing  5 cm
Depth of well  1524 cm
Length of screen  152 cm
Reference point for
water level measurments   Top of PVC
Casing
Elevation  NA
Static water level   259.5 cm
Static water level elevation   NA
Ground surface elevation   NA
Slug volume  330 cu. in. /54Q8 cm

Anticipated displacement   66 cm
Remarks:
I
                                                                    It
                                                                                 H

-------
                              Page _ฃ of  2




                              WELL #   84-5B
SLUG TEST FIELD FORM
#
1
2
3
4
5
6

1
2
3

1
2
3
4
5

1
2
3
4
5
TIME
WATER
0:00
0:16
0:25
0:35
0:49
l:-5
REMOV
0:OS
0:20
0:50
DROPP
0:10
0:18
0:30
0:45
1:01
REMOV If
0:12
0:19
0:25
0:35
1:10
DEPTH
TO
WATER
LEVEL ]
259.5
263.5
259.5
259
259
259.5
NG SLUC
" 259
259.5
259.5
NG SLUG
306
262
259.5
259.5
259.5
!G SLUG
261
260
259.5
259.5
259.5
STATIC
MINUS
TEST
NCREASES
0
4

-.5
-.5
0

.5
0
0

464.5
3.5
0
0
0

1.5
.5
0
0
0













#

1
2
3
4
5 .
6

1
2
3
4
WATER LEVEL
INCREASES 5









6
7







TIME
DROPPII
0:02
0:12
0:19
0:27
0:33
0:41
REMOV If
0:12
0:22
0:30
0:36
0:45
0:53
1:02







DEPTH
TO
WATER
G SLU<
256
277
261
260
259.5
259.5
G SLU(
260
259.5
259.5
259.5
259.5
259.5
259.5






•
STATIC
MINUS
TEST

3.5
17.5
1.5
.5
0
0

.5
0
0
0
0
0
0





















"









-------
           APPENDIX B

PLOTS USED TO CALCULATE HYDRAULIC
   CONDUCTIVITY  FROM SLUG TESTS

-------
For the EPA Wells:

     Length of Screen = 91 cm
     Borehole Radius = 10 cm
     Well Radius = 5 cm
     A* = 1.8
     B* = 0.2
For the Dames and Moore Wells:

     Length of Screen = 152 cm
     Borehole Radius = 10 cm
    -Well Radius = 5 cm
     A* = 2.0
     B* = 0.2
From Bouwer and Rice (1976).

-------
                 WELL  Ifl —  DROPPING  SLUG  (1)
                                Hydraulic Conductivity
                                Height of Water Column
                                                  YO
                                                   Y
                                                   t
                                               3 x 10
                                               181 cm
                                               29 cm
                                               1 cm
                                               205 sec
0.00
150.88
301.75
  452.63
TIME IN
  603.50
SECONDS
754.38
905.25
1055.13

-------
         WELL Ifl  — REMOVING SLUG  (1)
                                                  10"4 cm/sec
             Hydraulic Conductivity
             Height  of Water Column
                                YO
                                 Y
                                 t
                             8  x
                             181 cm
                             15 cm
                             5  cm
                             255 sec
127.5
255.0
  382.5     510.0
TIME IN  SECONDS
637.5
755.0
                                                             822.5

-------
         WELL  Ifi — DROPPING  SLUG  (2)
Hydraulic
Height of
Conductivity
Water Column
YO
Y
t
= 1 x 10" 3
= 181 cm
= 21 cm
= 1 cm
= 469 sec
cm/ sec
337.5
675.0
  1012.5
TIME IN
  1350.0
SECONDS
1687.5
2025.0
                                                          2362.5

-------
         WELL  IB — DROPPING SLUG  (1)
                       Hydraulic Conductivity
                       Height of Water Column
                                          YO
                                           Y
                                           t
                8 x 10
                699 cm
                59 cm
                10 cm
                532 sec
217.5
435.0
652.5
870.0
1087.5
1305.0
1522.5
                  TIME IN  SECONDS

-------
                      WELL 17R —  REMOVING  SLUG  (1)
                                     Hydraulic Conductivity = 2 x  10
                                     Height of Water Column = 313  cm
                                                       YO = 65 cm
                                                        Y = 30 cm
                                                        t = 736  sec
                                                                 -4
                                                       cm/sec
  o_
CO
or:
LJ
CO
o
     0.0
375.0
750.0
  1125.0
TIME IN
  15CO.O
SECONDS
1875.0
zso.o

-------
         WELL 17R —  DROPPING  SLUG  (1)
                        Hydraulic Conductivity
                        Height of Water Column
                                          YO
                                           Y
                                           t
                                     1 x 10
                                     313 cm
                                     60 cm
                                     2 cm
                                     723 sec
382.5
765.0
  1147.5     1530.0
TIME IN  SECONDS
1912.5
2225.0
2S77.5

-------
                  WELL  84-IB —  DROPPING SLUG  (1)
                                 Hydraulic Conductivity
                                 Height of Water Column
                                                   YO
                                                    Y
                                                    t
                                                3 x
                                                1,263 cm
                                                26 cm
                                                1 cm
                                                24 cm
                                            _2
                                          10   cm/sec
0.000
61.375
122.750    184.125    245.500
         TIME  IN  SECONDS
306.875
368.250
                                                                      429.625

-------
                      WELL  84-IB —  DROPPING  SLUG  (2J
  o_
LJ
LJ
O
                                     Hydraulic  Conductivity
                                     Height of  Water Column
                                                        YO
                                                         Y
                                                         t
                                                2 x
                                                1,263 cm
                                                13.5 cm
                                                1.0 cm
                                                21 sec
             _2
           10  cm/sec
     0.000
15.875
31.750     47.625     63.500
        TIME  IN SECONDS
79.375
95.250
111.125

-------
                       WELL  84-IB —  REMOVING  SLUG (2)
                                     Hydraulic  Conductivity
                                     Height of  Water Column
                                                        YO
                                                         Y
                                                         t
                                               1 x
                                               1,263 cm
                                               3.0 cm
                                               0.5 cm
                                               35 sec
                                           _2
                                         10   cm/sec
  o_
CO
a:
DJ
cu
o
  o.
                                                                        -a
     0.000
35.625
71.250
108.875
142.500
178.125
                                                   213.750
249.375
                                TIME IN  SECONDS

-------
                      WELL 84-2 — DROPPING SLUG  (1)
                                     Hydraulic Conductivity
                                     Height of Water Column
                                                        YO
                                                         Y
                                                         t
  o_
CO
a:
LJ
t->
LJ
LJ
o
      1 X
      1,313 cm
      57 cm
      2 cm
      623 sec
10"3  cm/sec
     0.0      1:5.0      270.0      405.0      540.0
                                TIME  IN SECONDS
675.0
810.0
945.0

-------
        WELL 84-2 — REMOVING SLUG (1)
                                                    -4
                                                  10   cm/sec
Hydraulic Conductivity
Height  of Water Column
                   YO
                   Y
                   t
                          8 x
                          1,318 cm
                          55 cm
                          10 cm
                          386 sec
130.0
280.0
390.0
       520.0
650.0
780.0
910.0
                  TIME  IN SECONDS

-------
                      WELL 84-3 —  REMOVING  SLUG  (1)
                                     Hydraulic Conductivity
                                     Height  of Water Column
                                                        YO
                                                         Y
                                                         t
                                              1 x 10  cm/sec
                                              1,315 on
                                              18 cm
                                              1 cm
                                              53 cm
  o_
CO
LJ
LJ
O
     0.0
20.0
40.0
   60.0       80.0
TIME IN  SECONDS
100.0
120.0
110.0

-------
         WELL  84-3 — REMOVING SLUG (2)
                                                  10   cm/sec
Hydraulic Conductivity
Height of Water Column
                  YO
                   Y
                   t
                          9 x
                          1,315 cm
                          16 cm
                          1 cm
                          60 sec
22.5
45.0
67.5
       90.0
112.5
                                                  135.0
157.5
                 TIME IN SECONDS

-------
        WELL  84-3  — REMOVING SLUG (3)
                                                  10   cm/sec
             Hydraulic Conductivity
             Height of Water Column
                                YO
                                Y
                                t
                             4 x
                             1,315 cm
                             6.5 -cm
                             1.0 cm
                             93 sec
30.0
60.0
   90.0      120.0
TIME IN  SECONDS
150.0
180.0
210.0

-------
                 NELL 84-4R  — DROPPING SLUG  (1)
                                Hydraulic Conductivity  = 4 x 10
                                Height  of Water Column  = 555 cm
                                                   YO  = 13 cm
                                                    Y  = 0.1 cm
                                                    t  = 20 sec
                                                             -2
                                                       cm/sec
0.00
25.25
52.50      78.75      105.00
        TIME IN  SECONDS
                                        131.25
157.50
153.75

-------
         WELL  84-4R —  REMOVING  SLUG  (1)
                        Hydraulic Conductivity = 5 x
                        Height of Water Column = 555  cm
                                          YO = 19 cm
                                           Y = 1.0  cm
                                           t = 98 sec
                                         10"3 cm/sec
37.5
75.0
  112.5
TIME IN
   150.0
SECONDS
187.5
225.0
2S2.5

-------
         WELL 84-4B ~ DROPPING  SLUG (1)
                                                     -4
                                                   10   cm/sec
              Hydraulic Conductivity
              Height  of Water Column
                                 YO
                                  Y
                                  t
                             2 x
                             1,316 cm
                             60 cm
                             10 cm
                             2,200 sec
600.0
1230.0
  1800.0     2400.0
TIME IN  SECONDS
3000.0
3600.0
                                                             4200.0

-------
                       WELL 84-4B  — REMOVING SLUG  (1)
                                      Hydraulic  Conductivity
                                      Height of  Water Column
                                                         YO
                                                          Y
                                                          t
        5  x
        1,316 cm
        60 cm
        10 cm
        613 sec
   -4
 10   cm/sec
to
en
LJ
E-H
LJ
OJ
o
     0.0      525.0      1050.0     1575.0     21X.O
                                TIME IN  SECONDS
2625.0
3150.0
3675.0

-------
                       NELL 84-Sfi — DROPPING SLUG  (.11
Hydraulic Conductivity-  3  x  10
Height of Water Column =  582  cm
                   YO =  40 cm
                    Y =  10 cm
                    t =  802  sec
                                                                      cm/sec
CO
<ฃ
CJ
Hi
CJ
     "0.0      262.5      525.0      787.5     1050.0

                                TIME IN  SECONDS
                 1312.5
1575.0
1837.5

-------
         WELL 84-5R — REMOVING SLUG (1)
                                                     -4
                                                   10   cm/sec
              Hydraulic Conductivity
              Height of Water Column
                                 YO
                                 Y
                                 t
                              5 x
                              582 cm
                              60 cm
                              3 cm
                              1,086 sec
0.0
247.5
495.0
  742.5     890.0
TIME IN  SECONDS
1237.5
1485.0

-------
                       WELL  84-5B —  REMOVING SLUG  (1)
                                                                   -2
                                      Hydraulic Conductivity = 3  x  10" c cm/sec
                                      Height  of .Water Column = 1,325 cm
                                                         YO = 0.5  cm
                                                          Y = 0.1  cm
                                                          t = 9  sec
(O  J
a:
CJ
ฃ-•
CiJ
CiJ
O
             >
               \
                                     -B-
     0.00      6.25
12.50
  18.75

TIME IN
   25.00

SECONDS
31.25
37.50
43.75

-------
                       WELL  84-5B —  DROPPING SLUG (2)
                                      Hydraulic Conductivity
                                      Height of Water Column
                                                         YO
                                                          Y
                                                          t
                                  _2
                            6 x 10   cm/sec
                            1,325  cm
                            47 cm
                            0.1 cm
                            18 sec
  o_
CO
s
E-ซ

O
0.000
7.625
15.250
22.875
                                            30.500
38.125
45.750
53.375
                                TIME  IN SECONDS

-------
                                 APPENDIX  D
          CALCULATION OF  LEACH  DURATION AND AMOUNT FROM SOURCE AREAS

     Three sources of TCE have  been  identified  at the  Western  Processing Site
based  on  the  concentration  levels  measured  in the  soil   and  water.   These
three  sources  are:   Reaction Pond I; Reaction  Pond  III;  and the  area around
Well 21.   A known disposal  site is not  located  in the vacinity  of  Well  21,
but the  high  concentrations of  TCE  in  soil  and water  samples  from  Well  21
indicate  that  disposal  occurred  in  this area.   Calculations  were   made  to
estimate  the mass of TCE present  in the  unsaturated  zone  (both  in  soil  and
water)  of these three  locations  based on  1982  soil  concentration measurements
(EPA,  198.2).    These mass  estimates were then  used  to   estimate  the  time
required  for  TCE to  completely leach out of  the  unsaturated  zone  into  the
saturated zone.   Time of  leach calculations  were based on  the  average annual
recharge  at  the   site   (8   in./yr   -  Appendix  A).    The  results  of  these
caluclations were used to determine the number  of years past 1982  to  keep  the
sources  active  in the  model,  and to  estimate the  source  strength  for  each
time step simulated.

SOURCE  AREA DATA
     The  data  used  to  estimate  the mass  of  TCE in  the unsaturated  zone  of
each source area are summarized below.
Reaction Pond I
     The total  volume of  contaminated unsaturated soil  at  Reaction Pond I  was
estimated to be 46,250 ft .  This  number is  based  on a surface  area  of 9,250
  o
ft   and   an  average  depth  of  unsaturated contaminated soil  of  5  ft.   The
surface area is the actual area  associated with Reaction Pond  I.   The average
depth  of  contaminated  soil   was  estimated from the soil sampling-depths  and
concentrations reported in Table D-l.   The TCE concentration  in  the  soil  was
estimated  to  be  600,000  ppb based  on  the measured soil  concentrations from
Wells  15  and  17   (Table  D-l).   In  order to  represent  a worst case,  the

-------
concentration   was   estimated   to   be  about   the  maximum   observed   soil
concentration  and  was assumed  to  be  uniform  over the  area  of  the  disposal
site.

           Table D-l.  TCE  Soil  Concentrations Near  Reaction Pond  I
                                   Soil Concentration (ppb)
Depth (ft)
3
6
9
Well 15
U
580,000
180,000
Well 17
U
558,000
350,000
U = Not Detected

Reaction Pond III
     The total  volume of contaminated  unsaturated  soil  at Reaction  Pond  III
was estimated to  be 35,650 ft .    This  number is based  on a surface  area  of
        2
7,125 ft_.; and an  average depth of  contaminated soil of  5  ft.   The surface
area is the actual  area associated with Reaction  Pond  III.   The average depth
of  contaminated  soil  was  estimated  from  the  soil  sampling  depths  and
concentrations reported in Table D-2.   The TCE concentration in  the  soil  was
estimated  to be   700  ppb  based  on  the  measured   soil  concentrations  from
Well 20 (Table D-2).   In order to represent  a worst  case,  the concentration
was estimated  to  be  about  the  maximum observed  soil concentration  and  was
assumed to be uniform over the area of the disposal  site.
          Table  D-2.   TCE  Soil  Concentrations  Near Reaction  Pond  III
                                   Soil Concentration (ppb)
                      Depth (ft)            Well 20
                           3                   U
                           6                   M
                           9                  676
                          12                  544
U = Not Detected
M = Present but below minimum quantifiable limit.

-------
Area Around Well 21
     The  total  volume  of contaminated  unsaturated soil  in the  area around
Well 21  was difficult to estimate  because  this  area is  not  a  known  disposal
area.   Therefore,  the volume was  assumed  to be the same as that of  Reaction
Pond  III,  35,650 ft^.   This volume is based  on  the  same surface  area  and
average  depth  as  those used for Reaction  Pond  III.  The TCE concentration in
the  soil  was  estimated  to   be   1,600  ppb  based  on  the  measured  soil
concentration from Well 21 (Table D-3).

           Table D-3.  TCE Soil  Concentrations Near Well  21 Source
                                    Soil Concentration (ppb)
                      Depth (ft)            Well 21
                           3                   U
                           6                   U
                           9                  116
                          12                 1,520
U = Not Detected

CALCULATION OF SOURCE AREA MASS
     The calculations  of mass  in  the  unsaturated zone for  each  of  the three
source locations are presented below.
     For  all   source  areas,  the   actual   and  effective porosities  of  the
unsaturated  material   were   set   at  0.25   and   0.15,   respectively.    The
distribution coefficient  (Kd)  and  retardation factor  for sorption of TCE  on
the  unsaturated  soils  at  all  source  locations  were  set  at   0.2   and  4.0,
respectively.
     The equation  used to calculate the mass of TCE at  each  source   location
is:

MassT-E = {mass in soil} + {mass in water}
        = (1 - 9A)(Cs)(Vols)(Ys) +  (9E)(Cw)(Vols)(Yw)                (D-l)

Where:

-------
MassTCE = total mass of TCE in the soil (Ib)
QA = actual water porosity
% = effective porosity
Cs = TCE concentration (ppb) in the soil
Vols = total volume of unsaturated soil (ft )
Ys = density of the soil (lb/ft3)
Cw = TCE concentration (ppb) in the water  {Cw = -^ (C$)}
Yw = density of the water (lb/ft3)
     The density of the soil was calculated as:

Ys = (1 - GA)(Y) = 121.5 lb/ft3
where:

0A = actual porosity (0.25)
Y = dry density of sand (162 lb/ft3)

     The  TCE  soil  concentrations  were  determined  based  on  measured  soil
concentrations in the unsaturated zone  at  wells  around  the source areas.   The
TCE water  concentrations were  estimated  to  be five  times greater  than  the
soil  concentrations based  on an  distribution  coefficient  (Kd)  of 0.2.   The
                                   Q
ppb notation is equivalent to lb/10  Ib in the total mass calculation.
Reaction Pond I
     Based on the data  in Table  D-4,  the  total mass of TCE in the unsaturated
zone at Reaction Pond  I was calculated  to  be 3,830  Ib.   The total mass in the
soil  and water was calculated to be 2,530  Ib and 1,300 Ib, respectively.

-------
       Table  D-4.   Summary  of  Data  Used  to Make Total Mass  Calculations
                   at  all Three  Source Locations
                       Reaction Pond I   Reaction Pond  III    Well 21
         0E                      0.15              0.15             0.15
         0A                      0.25              0.25             0.25
         Cs (ppb)          600,000               700           1,600
         Cw (ppb)        3,000,000             3,500           8,000
         Vo1s (ft3)         46,250            35,650          35,650
         Ys (lb/ft3)           121.5             121.5           121.5
         Yw (lb/ft3)            62.4              62.5            121.5
         Kd                      0.2               0.2             0.2
Reaction Pond III
     Based on the data  in  Table  D-4,  the total  mass of TCE in the unsaturated
zone at Reaction Pond III was calculated to be 3.5  Ib.  The  total  mass  in the
soil and_water were calculated to be 2.3 Ib and 1.2 Ib, respectively.
Around Well 21
     Based on the data  in  Table  D-4,  the total  mass of TCE in the unsaturated
zone around Well 21 was  calculated  to be 7.9  Ib.   The total mass  in  the soil
and water were calculated to be 5.2 Ib and 2.7 Ib, respectively.
     The estimated total mass  of TCE  present  in  the  unsaturated  zone  (based
on  October  1982  soil   analyses)  at  the  three   suspected  source  areas  is
summarized in Table D-5.
         Table  D-5.   Summary  of  Estimated Total Mass  of TCE  Present
                      at  the Three  Suspected  Source  Areas
                                         Mass of TCE (Ib)
                                   In Soil   In Water    Total
               Reaction Pond I     2,530      1,300      3,830
               Reaction Pond III       2.3        1.2        3.5
               Around Well 21          5.2        2.7        7.J

-------
LEACH DURATION AND AMOUNT FROM SOURCE LOCATIONS
     The preceeding calculations  show that  Reaction  Pond I  is the only  source
location  which  contains  a  significant  quantity  of TCE  in  the  unsaturated
zone.   Therefore,  calculations   to  determine the  time required  for  TCE  to
leach out  of  the unsaturated zone (duration), and the amount that would leach
out each year were  only made for Reaction Pond I.   This information was used
in  the  modeling to  determine  the  length  of  time  into the future  to keep
Reaction Pond  I  active, and  the  quantity  of TCE to leach from the unsaturated
to the saturated zone at each time step.
Reaction Pond  I
     The   leach   duration   and   amount  calculations   were  based   on  the
concentration  of TCE  in  the water  of  the  unsaturated  zone  and the  rate  of
water movement through  the  unsaturated  zone  from annual recharge.   Equation
D-l can be used  to calculate  the  TCE concentration in  the water  (C  ) based  on
                                                                   w
the total  mass  of  TCE  remaining  in the system  at  any given point  in time.
For convenience, the equation can be rewritten as follows:

                 MassTCE
 w   Vols  {Kd (1 - QA) Ys + % Ywl                                     (D-2)

     The volume of water  (Vol  )  moving through the  unsaturated  zone  from the
                              W
average annual recharge was calculated as follows:

Vol  = Rhg x A = 6,167 ft3/yr
   W

where

Rhg = average annual  recharge  (0.67 ft/yr - see Appendix A)
A = area of Reaction Pond I (9,250 ft2)

This  number  was  converted  to  a mass  of water  (Massw)  passing  through the
unsaturated system so the units  are consistent.

Mass  = Vol  x v  = 384,800 Ib/yr
    "      W    *W

-------