United States Office of Research and EPA/600/R-01/025
Environmental Protection Development March 2001
Agency Washington DC 20460
&EPA Evaluation of the Protocol
for Natural Attenuation of
Chlorinated Solvents:
Case Study at the Twin Cities
Army Ammunition Plant
-------
EPA/600/R-01/025
March 2001
Evaluation of the Protocol for Natural
Attenuation of Chlorinated Solvents:
Case Study at the Twin Cities Army Ammunition Plant
John T. Wilson
Don H. Kampbell
Subsurface Protection and Remediation Division
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Ada, OK 74820
Mark Ferrey
Paul Estuesta
Site Remediation Section
Minnesota Pollution Control Agency
St. Paul, MN 55155-4194
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
Notice
The U.S. Environmental Protection Agency, through its Office of
Research and Development, produced this document through in-
house research conducted by staff of the Subsurface Protection and
Remediation Division of the National Risk Management Research
Laboratory under Task 3674, entitled Development, Evaluation, and
Revision of the Protocol for the Natural Attenuation of Chlorinated
Solvents in Ground Water. It has been subjected to the Agency's
peer and administrative review and has been approved for publica-
tion as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for
use.
All research projects making conclusions or recommendations
based on environmentally related measurements and funded by the
U.S. Environmental Protection Agency are required to participate in
the Agency Quality Assurance Program. This project was conducted
under an approved Quality Assurance Project Plan. The procedures
specified in this plan were used without exception. Information on the
plan and documentation of the quality assurance activities and
results are available from the Principal Investigator.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the
Nation's land, air, and water resources. Under a mandate of national environmental laws, the
Agency strives to formulate and implement actions leading to a compatible balance between
human activities and the ability of natural systems to support and nurture life. To meet this
mandate, EPA's research program is providing data and technical support for solving environ-
mental problems today and building a science knowledge base necessary to manage our
ecological resources wisely, understand how pollutants affect our health, and prevent or reduce
environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for investigation
of technological and management approaches for preventing and reducing risks from pollution
that threatens human health and the environment. The focus of the Laboratory's research
program is on methods and their cost-effectiveness for prevention and control of pollution to air,
land, water, and subsurface resources; protection of water quality in public water systems;
remediation of contaminated sites, sediments and ground water; prevention and control of
indoor air pollution; and restoration of ecosystems. NRMRL collaborates with both public and
private sector partners to foster technologies that reduce the cost of compliance and to
anticipate emerging problems. NRMRL's research provides solutions to environmental prob-
lems by: developing and promoting technologies that protect and improve the environment;
advancing scientific and engineering information to support regulatory and policy decisions; and
providing the technical support and information transfer to ensure implementation of environ-
mental regulations and strategies at the national, state, and community levels.
The Technical Protocol for Evaluating Natural Attenuation of Chlorinated Solvents in Ground
Water, EPA/600/R-98/128, was developed from experiences with relatively small plumes in
unconsolidated aquifers. Many EPA enforcement actions involve large plumes in fractured bed
rock aquifers. This report is an evaluation of the performance of the Technical Protocol for
Evaluating Natural Attenuation of Chlorinated Solvents in Ground Water on a very large plume
of chlorinated solvents in a landscape where much of the plume is contained in fractured
consolidated rock. This report identifies the successes and some of the shortcomings of the
Technical Protocol for Evaluating Natural Attenuation of Chlorinated Solvents in Ground Water
when it was applied to a large plume in a fractured consolidated rock aquifer.
Stephen G. Schmelling, Acting Director
Subsurface Protection and Remediation Division
National Risk Management Research Laboratory
-------
Abstract
At the request of staff in the EPA Regions, EPA's Office of Research and Development
carried out an independent evaluation of the Technical Protocol for Evaluating Natural Attenua-
tion of Chlorinated Solvents in Ground Water (EPA/600/R-98/128). The Protocol was devel-
oped around case studies on relatively small plumes in sand aquifers. Staff in the Regions
wished to know if the Protocol could be usefully applied to large plumes, or to plumes in aquifers
in fractured consolidated rock.
Disposal of waste solvents containing trichloroethylene and 1,1,1-trichloroethane at the Twin
Cities Army Ammunition Plant (TCAAP) in Minnesota resulted in the formation of a ground-water
contaminant plume over five miles long. As part of the remedial response, the U.S. Army
installed an extensive pump-and-treat containment system around the source areas intended to
prevent additional loading of contaminants to the downgradient portion of the aquifer.
The site was selected for an independent evaluation of the Protocol for two reasons. First,
the ground-water plume was very well characterized and an extensive historical sampling
database was available to complement the study. Second, although long-term monitoring
indicated that the concentrations of trichloroethylene and 1,1,1-trichloroethane attenuated with
distance from the source, natural attenuation did not contain the ground water plume within the
site boundary. Thus, the evaluation could determine if an analysis done according to the
Protocol would predict a large ground-water plume under existing site conditions.
The results show that the Protocol was successful in predicting the development of the
ground-water plume at TCAAP. The screening analysis in the Protocol predicts that the
geochemical environment at TCAAP is not favorable to rapid reductive dechlorination. The
modeling portion of the study indicated that the current ground-water plume should be expected
when the rate of reductive dechlorination is slow.
The study also shows that natural biodegradation complements the on-going efforts to
extract contaminated ground water at the source, and should greatly reduce the time required to
reduce the concentration of contaminants to U.S. EPA drinking water standards. If the rate of
natural biodegradation exhibited in the last ten years continues for the next twenty years, the
portion of the aquifer downgradient from TCAAP will be reclaimed.
IV
-------
Contents
Foreword iii
Abstract iv
Tables vi
Figures vii
SI Conversion Factors ix
Background and Goals of the Study 1
Geological and Geographical Context 4
Geochemical Context 6
Distribution of Volatile Organic Contaminants 7
Calibration of BIOPLUME III 8
Calibration of Water Flow 9
Calibration of the Rate of Natural Biodegradation 11
Role of Hydrogen in Natural Biodegradation at TCAAP 13
Evaluation of the Protocol 15
Conclusions Applied to TCAAP 16
Extension to Other Sites 17
References 17
-------
Tables
Table 1.
Table 2.
Table 3.
Table 4.
Table 5.
Monitoring Wells Selected for an Evaluation of the Natural Attenuation
of Chlorinated Solvents in Ground Water at TCAAP
Geochemical Indicators of Anaerobic Biotransformation of
Chlorinated Solvents in Ground Water at TCAAP
Apparent Attenuation of Concentrations of TCE and 1,1,1-TCA in
Selected Monitoring Wells with Distance Downgradient of Source
Area D on TCAAP
Apparent Rate of Attenuation in Concentration of TCE in Selected
Monitoring Wells at TCAAP
Concentration of Hydrogen, Oxidation/Reduction Electrode Potential
Against an Ag/AgCI Reference Eectrode, and Temperature of
Ground Water Sampled from TCAAP
20
21
22
22
23
VI
-------
Figures
Figure 1. Geological cross section at the Twin Cities Army Ammunition Plant,
showing the origin of the chlorinated solvent plume in ground water 24
Figure 2. Location of the Twin Cities Army Ammunition Plant (TCAAP) and
the associated plume of chlorinated solvents in ground water with respect
to the Mississippi River, to Interstate 35W, and to Interstate 694 25
Figure 3. Panel A. Locations of monitoring wells or monitoring well clusters on
the TCAAP that were involved in the geochemical study to evaluate
monitored natural attenuation 26
Figure 3. Panel B. Locations of monitoring wells or monitoring well clusters that are
downgradient of the TCAAP that were involved in the geochemical study
to evaluate monitored natural attenuation 26
Figure 4. Relative concentration of trichloroetheylene and its biological reduction
product cis-dichloroethylene in ground water from TCAAP 27
Figure 5. Relative concentration of 1,1,1-trichloroethane and its abiotic transformation
product 1,1-dichloroethylene in ground water from TCAAP 27
Figure 6. Relative concentration of 1,1,1-trichloroethane and its biological
reduction product 1,1-dichloroethane in ground water from TCAAP 28
Figure 7. The triangles connected by lines are the water table elevations used to
calibrate the initial conditions in BIOPLUME III 28
Figure 8. Kriged values for horizontal hydraulic conductivity used to calibrate
BIOPLUME III 29
Figure 9. Cells used to calibrate BIOPLUME III 29
Figure 10. The grid used to calibrate BIOPLUME III superimposed on surface
features at TCAAP 30
Figure 11. Heads modeled by BIOPLUME III just prior to beginning pump-and-treat
activities in 1988 31
Figure 12. Steady-state heads modeled by BIOPLUME III for the year 2020, after
initiation of pump-and-treat starting in 1988 31
VII
-------
Figure 13. Set-up screen for BIOSCREEN when calibrated to conditions in the
Hillside Sand Aquifer, hydraulic conductivity at 280 feet per day, calibrated
rate of bioattenuation 0.17 per year 32
Figure 14. Centerline output of BIOSCREEN under conditions in the Hillside
Sand Aquifer 32
Figure 15. Set-up screen for BIOSCREEN under conditions near New Brighton 33
Figure 16. Centerline output of BIOSCREEN calibrated to the conditions at New Brighton,
hydaulic conductivity is 480 feet per day, and the calibrated rate of natural
biodegradation is 0.28 per year 33
Figure 17. Projections of concentrations of TCE at TCAAP with and without
bioattenuation in 1969 34
Figure 18. Projections of concentrations of TCE at TCAAP with and without
bioattenuation in 1988 35
Figure 19. Projections of concentrations of TCE at TCAAP with and without
bioattenuation in 1998 36
Figure 20. Projections of concentrations of TCE at TCAAP with and without
bioattenuation in the year 2008 37
Figure 21. Projection of concentrations of TCE at TCAAP with and without
bioattenation in the year 2018 38
VIII
-------
SI Conversion Factors
Multiply
Area:
Flow rate:
Length:
Mass:
Volume:
Temperature:
Concentration:
Pressure:
Heating value:
English (US)
Units by
1 ft2
1 in2
1 gal/min
1 gal/min
1 MGD
1 ft
1 in
1 Ib
1 Ib
1ft3
1 ft3
1 gal
1 gal
°F-32
1 gr/ft3
1 gr/gal
1 Ib/ft3
1 Ib/in2
1 Ib/in2
Btu/lb
Btu/scf
Factor
0.0929
6.452
6.31 x 10-5
0.0631
43.81
0.3048
2.54
453.59
0.45359
28.316
0.028317
3.785
0.003785
0.55556
2.2884
0.0171
16.03
0.07031
6894.8
2326
37260
Metric (SI)
to get Units
m2
cm2
m3/s
L/s
L/s
m
cm
g
kg
L
m3
L
m3
°C
g/m3
9/L
g/L
kg/cm2
Newton/m2
Joules/kg
Joules/scm
IX
-------
-------
Background and Goals of the Study
At the request of staff in the EPA Regions, EPA's Office of Research and Development carried
out an independent evaluation of the Technical Protocol for Evaluating Natural Attenuation of
Chlorinated Solvents in Ground Water, EPA/600/R-98/128 (the Protocol). The Protocol was
developed around case studies on relatively small plumes in sand aquifers. Staff in the Regions
wished to know if the Protocol could be usefully applied to large plumes, or to plumes in aquifers
in fractured consolidated rock.
The U.S. Army, their contractors, and their former contractors at the Twin Cities Army
Ammunition Plant (TCAAP) are carrying out a pump-and-treat remedy for ground water
contaminated with chlorinated solvents. The ground-water plume at the TCAAP site is over five
miles long, and much of the plume is in fractured dolomite and fractured sandstone. The plume
was recommended to the Office of Research and Development as a site to conduct a test on the
procedures outlined in the Protocol.
A primary purpose of this study was to determine whether the Protocol would accurately predict
the behavior of a large plume where natural attenuation is not appropriate as the sole remedy.
The ground-water plume at the TCAAP site is well characterized. This offered the opportunity
to use the Protocol to predict the effect of natural attenuation on the ground-water plume and
compare those predictions with the actual site data.
The second goal was to evaluate the effect of natural attenuation on the time required to restore
ground water through the combination of active pumping and natural attenuation. The Protocol
was used to evaluate the contribution of biological reductive dechlorination and natural abiotic
transformation reactions to the natural attenuation of chlorinated solvents in the ground-water
plume at TCAAP. The time required to restore the aquifer through pumping alone was
compared to the time expected with a combination of pumping and natural attenuation.
The Protocol is organized around three lines of evidence as identified in the OSWER Directive
on Monitored Natural Attenuation (U.S. EPA, 1997). The lines of evidence are (1) historical
ground-water data that demonstrate a clear and meaningful trend of decreasing mass or
concentration over time; (2) hydrogeologic and geochemical data that can be used to demon-
strate indirectly the type(s) of natural attenuation processes active at the site, and the rate at
which such processes will reduce contaminant concentrations to required levels; and (3) data
from field or microcosm studies which demonstrate the occurrence of a particular natural
attenuation process at the site and its ability to degrade the contaminants of concern.
At the TCAAP, the concentrations of trichloroethylene and 1,1,1-trichloroethane at most
monitoring wells that are downgradient of the source area have decreased by at least a factor of
ten since monitoring began in 1988. A major contribution to the reduction in the size of the
plume and concentration of contaminants is the benefit of a pump-and-treat system that has
been in operation since 1988. In this evaluation, the first line of evidence was provided by
-------
comparing the disposition of the plume of trichoroethylene before pumping began in 1988 to the
disposition that would be expected if there were no natural biodegradation.
The initial step for natural biodegradation of trichloroethylene was presumed to be reductive
dechorination. To evaluate the second line of evidence for trichloroethylene, the geochemistry
of the ground water at the site was surveyed and compared to the geochemistry that is known to
be associated with reductive dechlorination. The rate of natural biodegradation of trichloroeth-
ylene was estimated by calibrating models to fit the distribution of contamination in the aquifer
before pump-and-treat began in 1988. The second line of evidence for natural attenuation of
1,1,1-trichloroethane was evaluated by comparing the observed distribution of
1,1,1-trichloroethane to the distribution that would be produced by the expected rates on abiotic
reactions.
To evaluate a third line of evidence for trichloroethylene, the concentration of hydrogen in
ground water was determined and compared to the concentration of hydrogen that is known to
support reductive dechlorination.
This Technical Brief is not intended to present a full application of the Technical Protocol
sufficient for remedy decision-making. At the TCAAP, that decision was already made and the
remedies are in place. The remedy for ground-water contamination on the TCAAP is pump-and-
treat. This system is designed to extract the contaminant mass and exhaust the plume on the
facility. The pumping system has been in continuous operation since 1988. The remedy for
contamination in the aquifer downgradient of TCAAP is also pump-and-treat. This system is
designed to extract water for water supply.
This Technical Brief does not present information on the identification of receptors, the risk
analysis, a discussion of the cleanup goals, or other information that would be required by
decision-making authorities to select natural attenuation as a remedy. It focuses on the
contribution of natural attenuation to the cleanup of contamination in ground water after the
source of contamination has been controlled by an active remedy. It applies those sections of
the Protocol that are used to extract rate constants from field data to estimate a rate of natural
biodegradation. It then uses those extracted rate constants to forecast the future behavior of the
plume of contamination in ground water after source control has been achieved.
At the TCAAP, most of the plume is deep below the water table. Volatilization is not expected
to reduce the concentrations of contaminants. The plume has reached its maximum extent, and
has started to recede. Sorption is not expected to reduce concentrations. The distribution of
chlorinated solvents in the plume is controlled by those processes that degrade or dilute the
contaminants.
Under these conditions, an apparent reduction in the concentration of a contaminant as ground
water moves downgradient can be caused by at least three phenomena. First, the downgradient
concentrations may be lower simply because the contaminant plume has not broken through at
that location. Second, the downgradient concentration may be lower as a result of dilution and
dispersion along the flow path. Third, the reduction in concentration may be due to biodegrada-
tion or abiotic transformation of the contaminant along the flow path.
A major goal was to determine which combination of phenomena best accounted for the
observed reductions in contaminant concentrations downgradient. This was done by calibrating
two mathematical models, BIOSCREEN and BIOPLUME III, to the site. BIOSCREEN and
-------
BIOPLUME III were developed to facilitate the modeling of natural attenuation. Both models can
be downloaded from the Subsurface Protection and Remediation Division's web site
(http://www.epa.gov/ada/csmos.html). BIOCREEN is a simple and easy to use model designed
to evaluate data collected along a plume centerline. It is essentially a one-dimensional model.
It allows estimates of advection, dispersion, sorption, and biodegradation. BIOPLUME III is a
two-dimensional model that is better suited to forecasting the future behavior of plumes. It
allows separate calibration of the rate of attenuation of the source of contamination, and the rate
of attenuation of contamination along the flow path.
Projections of the models based on sorption, dilution and dispersion alone, as well as
projections including natural transformations, were compared with the observed concentrations
of trichloroethylene along the flow path. Mathematical modeling indicated that sufficient time
has passed for the leading edge of the plume to complete the entire flow path from TCAAP to the
ultimate receptor of ground-water flow, the Mississippi River. The modeling also indicated that
simple dilution and dispersion alone could not account for the existing distribution of trichloroet-
hylene. The actual concentrations were much lower than would be expected, indicating that
some process was actively removing the trichloroethylene.
The attenuation of trichloroethylene and 1,1,1-trichloroethane in ground water is often a result of
biological transformations carried out under anaerobic conditions in the aquifer. The most
important biological process that is capable of transforming trichloroethylene and
1,1,1-trichloroethane is reductive dechlorination. In addition to reductive dechlorination,
1,1,1-trichloroethane undergoes an elimination reaction to form 1,1-dichloroethylene and a
hydrolysis reaction to form acetate (McCarty, 1996; Vogel and McCarty, 1987; Vogel et al.,
1987; Klecka et al., 1990). The rate of this abiotic reaction is equivalent to the rate of
biodegradation.
During reductive dechlorination a chlorine atom in the molecule is replaced with a hydrogen
atom. The chlorine atom is released to the environment as a chloride ion. Trichloroethylene is
reduced to dichloroethylene (primarily cis-dichloroethylene), which can be further reduced to
vinyl chloride, which in turn can be reduced to ethene and ethane (Barrio-Lage et al., 1990;
Vogel and McCarty, 1985). Similarly, 1,1,1-trichloroethane is reduced to 1,1-dichloroethane and
then to choroethane. The reduced organic carbon in the dechlorination sequence is conserved.
The sum of the molar concentrations of trichloroethylene, the dichloroethylenes, vinyl chloride,
ethylene, and ethane does not change as one chemical is dechlorinated to form another.
The plume of trichloroethylene and 1,1,1-trichloroethane has reached its maximum extent and
maximum concentrations; sorption should not act to remove contaminant mass. The plume is
deep below the water table; volatilization should not be an important removal mechanism. As a
consequence, it is reasonable to assume that the difference in concentration between an
upgradient and downgradient location is the missing mass that has been destroyed by
biodegradation or abiotic reactions. The measured concentration of transformation products in
water can be used to calculate the concentration of trichloroethylene or 1,1,1-trichloroethane
that must be degraded through reductive dechlorination or removed through abiotic reactions to
produce that concentration of transformation products.
There is very little vinyl chloride, ethylene, or ethane in ground water at TCAAP. The amount of
trichloroethylene or 1,1,1-trichloroethane that would be destroyed to produce the observed
concentration of daughter products is a small fraction of the missing trichloroethylene or
-------
1,1,1-trichloroethane. If vinyl chloride is formed, it does not accumulate, and it is not further
reduced to ethylene and ethane.
The fact that the daughter products of degradation are not present does not necessarily
preclude reductive dechlorination. It may be that the rate of degradation of daughter products
may be rapid compared to that of the parent compound. It is also possible that the missing
trichloroethylene in the plume at TCAAP was initially transformed to cis-dichloroethylene
through reductive dechlorination, and that the cis-dichloroethylene was further degraded. In
recent years it has been established that bacteria in aquifers can degrade cis-dichloroethylene
to carbon dioxide using Iron III, Manganese IV or oxygen as the electron acceptor (Bradley and
Chapelle, 1997; Bradley et al., 1998a, 1998b; Klier et al., 1999). They can also degrade any
vinyl chloride that might be formed using Iron III, Manganese IV, or oxygen (Bradley and
Chapelle, 1996, 1997; Davis and Carpenter, 1990; Hartmans and de Bont, 1992). This would
explain why the accumulation of reduced transformation products did not account for the
missing trichloroethylene. If this is the situation, then conditions in the ground water at TCAAP
must be favorable for reductive dechlorination of trichloroethylene to cis-dichloroethylene. The
most direct indicator of conditions favorable for reductive dechlorination of trichloroethylene is
the presence of hydrogen in the ground water at concentrations greater than 1 nanomolar (U.S.
EPA, 1998).
It is also possible that reductive dechlorination in ground water at TCAAP was weak, and only a
minor fraction of the trichloroethylene was transformed to cis-dichloroethylene, which then
accumulated only to a limited extent. Under this pattern, conditions would not favor the further
reduction of cis-dichloroethylene. Vinyl chloride would never form in the first place. In this
situation the apparent attenuation in concentration of the trichloroethylene and
1,1,1-trichloroethane would be caused by sampling monitoring wells that are askew of the
centerline of the plume of contamination, or miss the proper depth interval of the plume.
The first pattern that can explain the data is extensive biological degradation of the parent
compounds with a combination of reductive dechlorination and anaerobic oxidation of interme-
diate compounds. The second pattern is limited biological degradation of the parent compounds
with minor production of first intermediates, which accumulate. A major goal of this study was to
determine which of these two possible patterns of biotransformation best explain the behavior of
the chlorinated solvents at TCAAP.
Geological and Geographical Context
The Twin Cities Army Ammunition Plant (TCAAP) is located six miles north of the city of St. Paul,
Minnesota. Ground-water flow is toward the southwest to the Mississippi River. Figure 1
presents a geological cross section for the area on and just downgradient of TCAAP. The water
table aquifer at TCAAP is an unconsolidated sandy aquifer, the Hillside Sand. Below the Hillside
Sand is an aquifer in fractured dolomite, the Prairie du Chien Group. Below the Prairie du Chien
is an aquifer in fractured sandstone, the Jordan Sandstone. To the southwest and down the
hydraulic gradient from TCAAP, the Hillside Sand and the aquifers below it become confined by
non-permeable materials of the Twin Cities Formation.
-------
A variety of wastes containing chlorinated solvents have been disposed at TCAAP. Near
TCAAP, the highest concentrations of solvents are in the Hillside Sand, as pictured in Figure 1.
As the plume moves farther downgradient, the plume moves into the Prairie du Chien Group
(not shown in Figure 1).
A mathematical model (BIOPLUME III) was used to evaluate the role of natural biodegradation
of chlorinated solvents in ground water at TCAAP. Figure 2 presents the areas on TCAAP and
the neighborhoods downgradient that were included in the grid of the model. Depicted in
Figure 2 are TCAAP, Interstate Highway 694, Interstate Highway 35W, and U.S. Highway 10.
To provide a reference, the highways are depicted in subsequent figures of the area considered
in this study.
Wells from six locations were selected for this study. It is important to point out that using only
six monitoring locations to evaluate a plume that is over five miles long and over 1000 feet wide
would be unacceptable to select natural attenuation as a remedy in nearly all cases. The
distribution of contamination in three dimensions had been determined from data from over 250
monitoring wells in the site characterization phase of the Superfund RI/FS. These locations
were selected from wells that were installed during the site characterization phase of the
Superfund RI/FS, and from wells that are installed to monitor the progress of the active pump-
and-treat system. To describe the background conditions in the aquifer, a well cluster was
selected that was upgradient of the known releases of chlorinated solvents. Two well clusters
were selected in known "hot spots" of ground-water contamination on the TCAAP. Pump-and-
treat wells in the "hot spots" and at the facility boundary were included. All of the available wells
that were downgradient of the TCAAP were evaluated. All of these wells that had shown
consistently high concentrations of trichloroethene and 1,1,1-trichlorethane during previous
sampling were included in the study.
The locations of monitoring wells on TCAAP are depicted in the Panel A of Figure 3. The
locations of wells downgradient of TCAAP are depicted in the Panel B of Figure 3. Table 1 lists
construction details on the wells sampled in this study. Wells in the study were sampled in June
1996, December 1996, June 1997, November 1997 and June 1998.
Well 03U113, a shallow well that is not impacted by contamination (see top panel of Figure 3),
was sampled in June 1997. Well 03L113, a deep well not impacted by contamination, was
sampled in November 1996 and June 1997. Well 03L113 is located in a cluster with Well
03U113.
Monitoring Wells 03U020, 03M020, 03L020, and 04U020 are a cluster of wells. Their location
is labeled 03U020 in the top panel of Figure 3. These wells were sampled in November 1996,
June 1997, November 1997 and June 1998.
Monitoring Wells 03U002, 03M002, 03L002, and 04U002 are also a cluster of wells. Their
location is labeled 03U002 in the top panel of Figure 3. These wells were sampled in December
1996, June 1997 and June 1998.
Monitoring Wells 04U821, 191942, and 04U872 were sampled in June 1997, November 1997
and June 1998 (see bottom panel of Figure 3).
Three of the extraction wells were also sampled. Wells 03U314 (also labeled SC-2) and 03U317
(also labeled SC-5) are located at a hot-spot of contamination near the location of Monitoring
Wells 03U020 in Figure 3. Well 03F306 is located at the perimeter of TCAAP near the location
-------
of Well 03U002 in Figure 3. These wells were sampled in November 1997 and June 1998. The
relationship between the extraction wells and the monitoring wells is depicted at a larger scale in
Figure 10.
Geochemical Context
Table 2 presents data on the geochemistry of the waters sampled by the monitoring wells in this
study. Data are only presented for sampling conducted in June 1998. There was little variation
in the geochemical parameters in a particular well from one sampling date to another. The
chemical parameters will be evaluated based on the criteria presented in Table 2.3 of the
Technical Protocol for Evaluating Natural Attenuation of Chlorinated Solvents in Ground Water
(U.S. EPA, 1998).
With the exception of one sampling date in one well, the oxygen concentrations in the three
wells downgradient of TCAAP (19142, 04U821, and 04U872) were below 0.5 mg/liter. The
oxygen concentrations in the extraction wells and the deeper wells in the monitoring well
clusters on TCAAP were also below 0.5 mg/liter (Extraction Wells 03U314, 03U317, and
03F306, and Monitoring Wells 03M020, 03L020, 04U020, 03M002, 03L002, and 04U002). The
most shallow monitoring well in the two clusters on TCAAP has oxygen concentrations above
the concentration that is tolerated by dechlorinating microorganisms (Wells 03U020 and
03U002). The oxygen concentrations in the background wells followed the same pattern of
oxygen concentrations above 0.5 mg/liter near the water table, and less than 0.5 mg/liter deep
below the water table. Oxygen was above 0.5 mg/liter in the shallow well and below 0.5 mg/liter
in the deeper well (03U113 compared to 03L113). The concentration of oxygen in the plume of
chlorinated solvents does not preclude a contribution of reductive dechlorination to natural
attenuation.
The concentrations of Iron II are low in all the wells sampled for this study. The highest
concentration was 2.0 mg/liter, in Well 04U020, the deepest well in a cluster of wells on TCAAP.
Iron II tends to sorb to or chemically react with aquifer solids. The presence of any measurable
concentration of Iron II is good evidence for ongoing Iron III reduction. The highest concentra-
tions in the wells downgradient of TCAAP were 0.7 mg/liter in Well 191942 when sampled in
June 1998, and 1.0 mg/liter in Wells 04U821 and 04U872 when sampled in June 1997. Many of
the concentrations are below the criterion of 1 mg/liter published in Table 2.3 of the Protocol.
The concentrations of Iron II in the monitoring wells do not indicate that there is a highly reduced
geochemical environment that is generally associated with reductive dechlorination.
The concentrations of manganese in the plume at TCAAP are generally higher than the
concentrations of Iron II, indicating that manganese reduction may be more prevalent than iron
reduction in this aquifer. Manganese was measured in the field using a HACH™ Company DR
100 field colorimeter and the Periodate Oxidation Method (APHA, 1975). Because Mn IV is
insoluble at neutral pH, the measured concentrations of manganese are probably Manganese II.
In a ground water that is buffered by carbonate minerals in the aquifer matrix, an increase in
alkalinity may result from carbon dioxide produced by microbial metabolism. There is very little
change in the alkalinity of the ground water in the wells sampled for this study. The wells
-------
downgradient do not meet the criterion of twice background that is published in Table 2.3 of the
Protocol. In fact, there is no significant difference in the alkalinity in the contaminated wells and
in the background wells. Because the ground water is hard, substantial increases in alkalinity
would be necessary to meet the criterion. In any case, changes in alkalinity in ground water
from the wells in this study offer no indication that microbial metabolism has produced carbon
dioxide.
The concentrations of sulfate in wells from this study are low, and most are below the criterion of
20 mg/liter published in the Protocol. The same situation applies to concentrations of nitrate.
High concentrations of nitrate can inhibit reductive dechlorination. However, the concentrations
of nitrate in ground water from the wells in this study are low, and most are below the criterion of
1 mg/liter published in the Protocol. High concentrations of nitrate should not inhibit reductive
dechlorination at TCAAP.
Methane can be produced by biological processes under conditions that are strongly associated
with reductive dechlorination. None of the ground water sampled for this study had methane
concentrations that approached the criterion of 0.5 mg/liter published in the Protocol. Concen-
trations of methane offer no evidence of the anaerobic metabolic activity that might be
associated with reductive dechlorination.
As described in the Protocol, chloride should accumulate from reductive dechlorination, sulfide
should be produced from sulfate reduction, and dissolved organic materials may act as a
substrate that supports biological reductive dechlorination. The concentrations of chloride,
sulfide and total organic carbon (TOC) were determined in water from the monitoring wells.
Concentrations of chloride varied from 13.3 to 4.5 mg/liter in the background wells and from 14.9
to 30.7 mg/liter in the pump-and-treat wells on the TCAAP. The concentration of chloride in the
wells that were downgradient of TCAAP varied from 10.4 to 13.9 mg/liter. There was no clear
evidence of accumulation of chloride over background, and the accumulation did not meet the
criterion in the Protocol. The concentration of TOC in all the wells was uniformly low. The
maximum concentration determined was 3.8 mg/liter. Most analyses fell in the range of 1.9 to
0.8 mg/liter. There was no evidence that TOC was accumulating in the water. Sulfide was
never detected at TCAAP at concentrations exceeding the detection limit of 1.0 mg/liter.
In summary, the ground-water geochemistry at the TCAAP is inconclusive in assessing the role
of reductive dechlorination.
Distribution of Volatile Organic Contaminants
Reductive dechlorination requires an electron donor to serve as a source of reducing power and
metabolic energy to allow the process to proceed. The petroleum-derived aromatic hydrocar-
bons (benzene, toluene, ethylbenzene, the xylenes, and the trimethylbenzenes) serve as
electron donors to support reductive dechlorination in many contaminated ground waters
(Sewell and Gibson, 1991; Clarke, 2000). In the water samples collected for this study, the
concentrations of BTEX compounds were very low, less than 5 ug/liter in every case. Their total
concentration never approached the criterion of 0.1 mg/liter published in the Protocol. The
BTEX compounds are not available to support reductive dechlorination at TCAAP.
-------
Despite the weak indications of reductive dechlorination at TCAAP, cis-dichloroethylene is
present in the ground water from Extraction Wells 03U314, 03U317, and 03F306, TCAAP
Monitoring Wells 03U020, 03M020, 03M002, and downgradient Monitoring Well 191942 . This
compound is most likely produced as the reductive transformation product of trichloroethylene.
The distribution of trichloroethylene and cis-dichloroethylene with distance downgradient from
the source is presented in Figure 4. The concentration of cis-dichloroethylene was a small
fraction of the concentration of trichloroethylene. In general, the concentration of
cis-dichloroethylene was approximately 1% of the concentration of trichloroethylene.
The transformation product that is present in highest concentration relative to the potential
parent compounds is 1,1-dichloroethylene, which is produced through a spontaneous abiotic
reaction from 1,1,1-trichloroethane. Relatively high concentrations of 1,1-dichloroethylene are
found in the ground water from Extraction Wells 03U314, 03U317, and 03F306, TCAAP
Monitoring Wells 03U020, 03M020, 03U002, 03M002, and 03L002, and downgradient Monitor-
ing Wells 191942, 04U821, and 04U827. The distribution of 1,1,1-trichloroethane and
1,1-dichloroethylene with distance downgradient of the source is presented in Figure 5. In
general, the concentration of 1,1 -dichoroethylene was a factor of two to a factor of ten lower than
the concentration of 1,1,1-trichloroethane.
Another major potential reduction product of trichloroethylene is trans-dichloroethylene. The
potential reduction product of the dichloroethylenes is vinyl chloride. In all the ground water
sampled for this study, the concentration of trans-dichloroethylene and vinyl chloride was less
than 5 ug/liter.
A potential biological reduction product present in ground water from TCAAP is 1,1-dichloroethane,
produced from the reductive dechlorination of 1,1,1-trichloroethane (Galli and McCarty, 1989).
Figure 6 presents the distribution of 1,1,1-trichloroethane and 1,1-dichloroethane with distance
downgradient of the source. The concentration of 1,1-dichlorethane in the downgradient water
samples is approximately one-half of the concentration of 1,1,1-trichloroethane.
Although the concentration of the parent compounds was reduced by several orders of
magnitude, the combined concentrations of their transformation products never equaled the
concentration of the original contaminants that remained after natural biodegradation, much less
approached the original concentration of the contaminants.
Calibration of BIOPLUME III
Trichloroethylene was the most persistent contaminant in the plume (compare Figures 4, 5, and
6). The modeling of transport and fate of contaminants emphasized trichloroethylene.
The hydrological and geological parameters in the model used to describe transport and fate
were not independently selected from a review of available data. Values for hydraulic
conductivity, porosity, retardation due to sorption, and the coefficient of dispersion were taken
from a previous model developed for the plume as part of the Superfund Remedial Investigation
process (Camp Dresser & McKee Inc., 1991). This was done because the model calibrated for
the remedial investigation had been reviewed and approved by the appropriate regulatory
authority. The model calibrated for the remedial investigation did not account for biotransforma-
-------
tion of chlorinated solvents. Because the calibration of BIOPLUME III used the same values for
hydraulic conductivity, porosity, retardation due to sorption, and the coefficient of dispersion, the
differences between the forecasts of the remedial investigation model and the projections of
BIOPLUME III reflect the further contribution of biodegradation. Differences in the models were
not an incidental result of differences in the calibration of the flow of water.
Calibration of Water Flow
The natural biodegradation of trichloroethylene and its dechlorination products was described
with BIOPLUME III, a two-dimensional transport-and-fate model that is designed to facilitate an
understanding of natural biodegradation of organic compounds in ground water. BIOPLUME III
is available from the Center for Subsurface Modeling Support (CSMoS) a technical support
service located at the Subsurface Protection and Remediation Division, National Risk Manage-
ment Research Laboratory, Office of Research and Development, U.S. EPA. The web address
to download BIOPLUME III and to download instructions to install the model is
http://www.epa.gov/ada/models.html.
There is a potential for error in using a two-dimensional model to simulate three-dimensional
fate and transport processes. Because dispersion is restricted in the two-dimensional model to
longitudinal and transverse dispersion, the contribution of vertical dispersion to attenuation is
ignored. The two-dimensional model does not account for variations in contaminant concentration
with depth, thus the two-dimensional model underestimates natural attenuation through dispersion.
The model was calibrated assuming that the entire aquifer at a particular location experienced
the highest concentration measured at any depth interval at that location, rather than some
average of the actual distribution of contaminant concentrations. The two-dimensional model
assumes more contamination is present in the aquifer than is truly present.
The use of trichloroethylene and 1,1,1 -trichloroethane at TCAAP began after World War II. The
BIOPLUME III simulation of the trichloroethylene plume at TCAAP started with year 0 of the
simulation in 1950 and ended with year 70 of the simulation in the year 2020. The first
parameters used to calibrate BIOPLUME III were the parameters controlling the ground-water
flow properties. The initial conditions for hydraulic heads were imposed by tracing interpreted
ground-water contours from a map. BIOPLUME III inputs data on water table elevations by
importing a map showing isopleths of the elevation of the water table, then tracing the ground-
water contours with a computer mouse. The contours on the map used to calibrate BIOPLUME
III (Figure 7) were fit using professional judgment, not a contouring software application. On
TCAAP, most of the contamination is in the unconsolidated sand aquifer. Based on pumping
tests in the Hillside Sand Aquifer, the horizontal hydraulic conductivity along the southwestern
edge of TCAAP was set at 200 feet per day. Downgradient of TCAAP the plume enters the
fractured consolidated rock aquifers. Based on pumping tests in the Prairie du Chien group, the
horizontal hydraulic conductivity farther downgradient in the New Brighton area was set at
380 feet per day.
The distributed values for horizontal hydraulic conductivity are presented in Figure 8. Each red
dot on the figure is a log point where an input value for hydraulic conductivity of either 200 feet
per day or 380 feet per day was provided to the model, based on professional judgment and
maps of the distribution of contamination in the Hillside Sand aquifer and the Prairie du Chien
group. Simple kriging was used to distribute an estimated hydraulic conductivity to the other
cells in the grid.
-------
To be consistent with the Remedial Investigation, the coefficient of longitudinal dispersion was
set uniformly at 200 feet. The coefficient of transverse dispersion was set at 40 feet. The
effective porosity was set at 0.2; this relatively high value reflects the presence of solution
channels in the Prairie du Chien group.
The plume of trichloroethylene in ground water was simulated in BIOPLUME III by calibrating an
injection well in Cell 5 and in Cell 8 as depicted in Figure 9, injecting at the rate of 5 gpm. This
low rate of injection was selected to avoid errors in the simulation of the flow field. The
concentration of trichloroethylene in the injection wells was adjusted empirically until the
modeled concentrations matched the real distribution of trichloroethylene in 1988. A modeled
injected concentration of 2,500,000 ug/liter produced a simulated concentration in the plume of
27,000 ug/liter. The concentration in the injection wells was held constant until 1988. After 1988
the concentration in the injection wells was lowered according to a schedule that reflected actual
reductions in concentrations in monitoring wells in the source.
The plume of chlorinated solvents in ground water at TCAAP is presently being captured by a
series of extraction wells. Their effect was simulated by assigning a uniform rate of pumping to
the cell in the model grid that contained the well. The allocation of pumping wells in the
BIOPLUME III model is depicted in Figure 9. The following two paragraphs describe the
calibration of the pumping wells in the BIOPLUME III model.
Cells 1 through 8 are on TCAAP. They were modeled as pumping at a uniform constant rate
starting in 1988. Wells in Cell 1 were pumped at 500 gpm, wells in Cell 2 at 610 gpm, wells in
Cell 3 at 550 gpm, wells in Cell 4 at 510 gpm, wells in Cell 6 at 155 gpm, and wells in Cell 7 at
115 gpm. The total rate of pumping modeled for the TCAAP Ground Water Recovery System
(TGRS) system was 2,440 gpm. Cell 9 contained the pumpout system at New Brighton, also
labeled OU1. In the model, wells were pumped at Cell 9 at 1,000 gpm starting in 1950 extending
to 1996, then at 2,000 gpm extending to the year 2000, then at 2,700 gpm after the year 2000.
Cell 10 contained a pumping system for the OU3 remedy. Wells in Cell 10 were modeled at
1,000 gpm starting in 1992.
The relationship between the actual location of the pumping wells and the cells in the model grid
for BIOPLUME III is pictured in Figure 10. Data on the gallons pumped for each well since 1989
was divided by ten to estimate gallons per year, then by 525,600 to convert to gpm. The wells
in Cell 1 are B-12 pumping at 240 gpm, B-7 pumping at 248 gpm, and B-10 pumping at 210 gpm
for a total of 698 gpm. The wells in or just upgradient of Cell 2 are B-6 pumping at 229 gpm, B-9
pumping at 133 gpm, B-5 pumping at 185 gpm, B-8 pumping at 116 gpm, and B-4 pumping
at170 gpm fora total of 833 gpm. The wells in Cell 3 are B-3 pumping at 209 gpm, B-2 pumping
at 109 gpm, and B-1 pumping at 240 gpm for a total of 558 gpm. The wells in Cell 4 are B-11
pumping at 99 gpm, and SC-1 pumping at 29 gpm for a total of 128 gpm. The wells in Cell 6 are
SC-3 pumping at 97 gpm and SC-2 pumping at 44 gpm for a total of 141 gpm. The well in Cell
7 is SC-5 pumping at 103 gpm.
Figure 11 presents the simulated heads in 1988, just prior to the initiation of pump-and-treat
activities at TCAAP. Figure 12 presents the simulated heads in 2020. The model predicted that
the heads came to a steady state within a few years after the initiation of pumping.
The simulated effect of pumping was to lower the water table elevation at the downgradient
boundary of TCAAP by a little more than four feet. The steady-state head in Cell 1 was
829.5 feet in 1988 and 825.8 feet in 1990. The steady-state head in Cell 2 was 825.4 feet in
10
-------
1988 and 821.2 feet in 1990. The steady-state head in Cell 3 as 825.4 feet in 1988 and
821.1 feet in 1990. The steady-state head in Cell 4 was 825.3 feet in 1988 and 821.5 feet in
1990.
Calibration of the Rate of Natural Biodegradation
At certain times since 1950, large quantities (railroad tankcar loads) of used trichloroethylene
and 1,1,1-trichloroethane were released to disposal lagoons on the TCAAP. It is likely that
some of this material remains in the aquifer as DNAPL. A source of trichloroethylene and
1,1,1-trichloroethane in ground water near these disposal sites has persisted to the present
time.
The rate of natural biodegradation along the flow path at TCAAP was estimated assuming that
the plume had come to a steady state before the pump-and-treat activity was initiated in 1988.
The screening model BIOSCREEN was used to fit rate constants by curve matching to a simple
one-dimensional representation of the flow path. BIOSCREEN is available from the Center for
Subsurface Modeling Support (CSMoS), a technical support service located at the Subsurface
Protection and Remediation Division, National Risk Management Research Laboratory, Office
of Research and Development, U.S. EPA. The web address to download BIOSCREEN and to
download instructions to install the model is http://www.epa.gov/ada/models.html. The values
of hydraulic conductivity selected to calibrate BIOSCREEN were selected after a previous
review by a hydrogeologist with the Minnesota Pollution Control Agency.
Figure 13 presents the input screen for BIOSCREEN when the entire flow path was calibrated to
conditions appropriate to the Hillside Sand Aquifer. A horizontal hydraulic conductivity of
280 feet per day was input as 9.9E-02 cm/sec in Area 1. The hydraulic gradient of 0.002 was
extracted from Figure 13. A porosity of 0.22 was assumed. In Area 2 of the input screen, the
coefficient of longitudinal dispersion was set to 200 feet, and the coefficient of transverse
dispersion was set to 40 feet to be consistent with the assumptions in the models in the remedial
investigation. In Area 3, the retardation factor was assumed to be 1.3, to be consistent with the
assumptions in models in the remedial investigation. In Area 5, the modeled length was set to
30,000 feet, and the width of the model was set to 6,000 feet. The plume was modeled for forty
years, from 1950 to1990. Area 6 represents the width of the model as a transect perpendicular
to ground-water flow along the Southwest margin of TCAAP. The source for the BIOSCREEN
simulation is approximately equivalent to Cell 2 and Cell 3 in the BIOPLUME III simulation in
Figures 9 and 10.
Field data for comparison to BIOSCREEN are taken from Table 3, which lists the concentration
of trichloroethylene and the distance downgradient from the source for selected wells along the
centerline of the flow path from TCAAP. These concentrations are the highest that were
sampled in each well during routine monitoring over several years in the period prior to the
initiation of pumping in 1988. A rate of natural biodegradation was selected that provided the
best match of the projections of the model to the actual concentrations in the monitoring wells.
The best calibration of BIOSCREEN is presented in Figure 14. The red line is the expected
concentrations based on dilution, dispersion, and retardation, but no biodegradation. There was
adequate time for the trichloroethylene to move along the flow path, and bring the plume to
steady state. This has been confirmed by ground-water contaminant concentration trends
observed throughout the TCAAP monitoring network. A first-order rate of natural biodegrada-
tion of 0.17 per year (Area 4 in Figure 13) provided adequate calibration to the field data. A first-
order rate of 0.17 per year is equivalent to a half life of 4.1 years.
11
-------
Figure 15 presents a second calibration of BIOSCREEN to conditions that are more typical of
the area near New Brighton, where the plume has moved from the Hillside Sand into aquifers in
fractured consolidated rock. The hydraulic conductivity was raised to 480 feet per day
(1.7E-01 cm/sec). Under these conditions, an adequate calibration required a first-order rate
constant for natural biodegradation of 0.28 per year (compare Figure 16). A first-order rate of
0.28 per year is equivalent to a half-life of 2.8 years.
To determine the uncertainty in the rate constant as determined by curve matching, the data in
Table 3 were subjected to log-linear regression. The distance of each monitoring well from the
source was divided by the seepage velocity predicted by BIOSCREEN under conditions typical
of the area near New Brighton to estimate the travel time for water along the flow path to each
well. A linear regression was performed of the natural logarithm of concentration of trichloroet-
hylene against travel time along the flow path. The rate fitted by the regression was equivalent
to a rate of natural biodegradation of 0.28 per year. The slower 5% confidence interval
corresponds to a rate of at least 0.17 per year. The F statistic for the relationship is very high.
The probability that chance alone would give the appearance of degradation, if in fact there was
no degradation, is 0.21%.
As an independent evaluation of the rate of natural biodegradation of trichloroethylene that was
extracted by this approach, it was applied to the observed attenuation of 1,1,1 -trichloroethane in
the same wells. The well-established rate of abiotic transformation of 1,1,1-trichloroethane was
used as a benchmark to evaluate the overall rate of apparent attenuation. The concentrations
of 1,1,1-trichloroethane along the flow path are presented in Table 3. The ground-water
temperature at the TCAAP is near 10 °C to 12 °C. McCarty (1996) evaluated the effect of
temperature on the rate of abiotic transformation of 1,1,1-trichloroethane. He predicted an
average rate of abiotic transformation of 0.14 per year at 15°C. The observed rate of natural
attenuation extracted from the data in Table 3, assuming a seepage velocity of 1,600 feet per
year, was 0.4 +/- 0.19 per year at 95% confidence. The rate extracted assuming a seepage
velocity of 931 feet per year was 0.24 +/- 0.11 per year. This rate would include natural
biodegradation as well as the rate of abiotic transformation predicted by the relationships of
McCarty (1996) and Klecka et al. (1990). If there were no reductive dechlorination of
1,1,1 -trichloroethane and all the attenuation was caused by the abiotic processes, the rate fitted
to the field data at the higher estimate of seepage velocity would exceed the published rate by
no more than a factor of three. There is no practical or statistical difference between the rate
fitted to the field data at the lower estimate of seepage velocity and the published rate of abiotic
degradation.
For trichloroethylene, the simple one-dimensional model was best fit with a rate of 0.28 per year
when it was calibrated to conditions that predicted a high seepage velocity, and 0.17 per year
with a low seepage velocity. A regression through the data under conditions with a high
seepage velocity also predicted a rate of 0.28 per year, with a 95% confidence interval of at least
0.17 per year. The BIOPLUME III model was calibrated twice, using both rates to describe
biodegradation in the ground water. The "radioactive decay" option under decay parameters
was used in BIOPLUME III to simulate reductive dechlorination.
Monitoring data from 1988 to present indicate that the concentrations of trichloroethylene
entering the ground water from the residual sources are also attenuating. Most likely, this
results from the success of the extraction system. The reduction in the strength of the source
12
-------
was simulated by reducing the concentration of trichloroethylene in the hypothetical injection
wells in BIOPLUME III that simulate the source area. The concentrations were reduced
according to a schedule that produced a first-order reduction in concentration in the simulated
plume. Table 4 presents the results of log-linear regressions performed on the concentration of
trichloroethylene in twelve monitoring wells and on the TCAAP Ground Water Recovery System
(TGRS), over time. The arithmetic average rate of natural biodegradation was near 0.4 per year,
with a rate of at least 0.3 per year at 95% confidence. Not all the wells at TCAAP show this
marked, downward trend in concentration. To be conservative, BIOPLUME III was calibrated
with a simulated rate of attenuation of the source areas of 0.25 per year. The concentrations of
the source areas were held constant in the model until 1988 (the year the pump-and-treat
system began operation), then declined at a rate of 0.25 per year until the end of the simulation
in the year 2020. Under these conditions for attenuation of the source, BIOPLUME III was
calibrated for three conditions: for no natural attenuation due to biodegradation, for a
degradation rate in the plume of 0.28 per year, and for a degradation rate in the plume of
0.17 per year.
Figure 17 compares the three simulations for 1969. There is effectively little difference in the
simulations. Figure 18 compares the simulations in 1988, the year the extraction activities
began. Under conditions with no natural bioattenuation, the plume would have reached the
Mississippi River at high concentrations. Even with bioattenuation, the plumes would be
expected to expand (compare Figures 17 and 18). Figure 19 compares the three simulations in
1998, after ten years of pump-and-treat. The plumes predicted by the simulations with natural
attenuation are effectively at steady state (compare Figures 18 and 19). Figure 20 compares
the three simulations for the year 2008. The simulation with no natural bioattenuation shows
little change from the earlier simulations. In the simulations with bioattenuation, the concentra-
tions in the remaining hot spot are greatly reduced compared to the simulations of the plume
before pump-and-treat (compare Figures 20 and 18). In addition, the hot spot in the simulations
with bioattenuation has detached from TCAAP and is moving downgradient with the flow of
ground water. This illustrates the role of source control in achieving goals for remediation
through natural biodegradation. Figure 21 compares the three simulations for the year 2018. In
the simulation with no bioattenuation, most of the plume remains in the aquifer. Only the leading
edge has been replaced with uncontaminated ground water from upgradient. The simulation
with bioattenuation at a rate of 0.28 per year suggests that the plume will have essentially
disappeared by 2018. The simulation with bioattenuation at a rate of 0.17 per year predicts that
concentrations of trichloroethylene will be very low and approaching MCLs by the year 2018.
Role of Hydrogen in Natural Biodegradation at TCAAP
Conventional geochemical characterization of the ground water at TCAAP indicates that
biodegradation through reductive dechlorination is possible, but offers no evidence that it should
be expected. Mathematical models that incorporate biodegradation are the best fit to the long-
term monitoring data at the site. However, these lines of evidence by themselves do not present
compelling evidence that natural bioattenuation is important at TCAAP.
Molecular hydrogen is produced in ground water by fermentation reactions. This hydrogen is an
excellent substrate to support reductive dechlorination (Maymo-Gatell et al., 1995; Jakobsen et
13
-------
al., 1998). If adequate concentrations of hydrogen are available, reductive dechlorination
should occur. The concentration of dissolved hydrogen in ground water is dictated by the
dominant microbial processes in the aquifer.
Lovley, Chapelle, and their associates (Lovley et al., 1994) have developed a technique to
determine or predict the dominant electron accepting process in ground water by measuring the
concentration of dissolved hydrogen. They found that sulfate reduction required a minimum
hydrogen concentration of 1 nanomolar, while methanogenesis required a hydrogen concentra-
tion of 4 nanomolar. Reductive dechlorination occurs in ground water that is undergoing sulfate
reduction or methanogenesis. Following their approach, the U.S. EPA Protocol (U.S. EPA,
1998, see page 41) predicts that reductive dechlorination of solvents will occur if the concentra-
tion of hydrogen exceeds 1 nanomolar. Recent experimental evidence has supported this
prediction. Yang and McCarty (1998) found that a mixed culture growing on benzoate with
cis-dichloroethylene available as an electron acceptor (at 28 °C) poised the hydrogen concentra-
tion at 2 nanomolar. They interpreted this concentration as the minimum concentration of
hydrogen that would support utilization by organisms carrying out reductive dechlorination.
Fennell and Gossett (1998) reported that the lowest hydrogen concentration that would support
dechlorination (at 35 °C) was as low as 1.5 nanomolar.
Ground-water samples for the analysis of hydrogen were taken using a pneumatically driven
pump according to current EPA guidance (U.S. EPA, 1998). A glass sampling bulb, fitted with
a gas-tight septum, was nearly filled with ground water, allowing approximately 50 ml of air to
remain in the bulb. Ground water was introduced to the bulb at approximately 400 ml/minute
and allowed to escape from the bulb through a second port. Sequential sampling indicated that
equilibration between the aqueous phase and the gaseous phase was complete at 30 minutes.
Following the equilibration, a 10 ml aliquot of gas was immediately withdrawn from the bulb
through the septum using a gas tight syringe. In June 1997, hydrogen was sampled in the field
and analyzed in the field within minutes of collection. In June 1998, hydrogen was sampled in
the field and sealed in a glass bottle and shipped back to the Subsurface Protection and
Remediation Division, Robert S. Kerr Environmental Research Center, Ada, Oklahoma, for
analysis within a week of collection. The June 1998 samples are corrected for background
concentrations of hydrogen in trip blanks (equivalent to 0.89 +/- 0.18 nanomolar).
The screening process in the Protocol predicts that under the geochemical conditions shown in
Table 2, reductive dehalogenation processes are "occurring too slowly to contribute in a
meaningful way" to the attenuation of the plume (U.S. EPA, 1998). Hydrogen concentrations
are usually in the range of 0.2 to 0.8 nM under iron reducing conditions (Lovley etal., 1994; U.S.
EPA, 1998). Much of the ground water in the TCAAP plume contained hydrogen in this
concentration range. However, at many locations, hydrogen concentrations were found along
the axis of the plume at concentrations that were near or in excess of 1 nanomolar (Table 5).
In their paper on practical considerations for measuring hydrogen in ground water, Chapelle et
al. (1997) warn against measuring hydrogen in wells with metal screens (compare Table 1 for
construction materials of wells in this study). The background wells, Wells 03U113 and 03L113
are constructed entirely of PVC plastic. Bjerg et al. (1997) demonstrated that PVC, Teflon™,
and "probably" stainless steel are suitable materials for wells where samples will be analyzed for
dissolved hydrogen. The hydrogen concentrations in water from the deeper background Well
(03L113) indicated that the background conditions will support reductive dechlorination. Well
14
-------
04U821 is an open borehole. The hydrogen concentration in this well in 1997 indicates that
conditions were favorable for reductive transformation of chlorinated solvents as far downgradient
as Well04U821.
Interestingly, high concentrations of hydrogen were also observed in Monitoring Wells 03U020
and 03U002, where ground water also contained oxygen and nitrate. However, ground water at
these wells also contained cis-dichloroethylene and concentrations of chloride over five times
background level, indicating that degradation of trichloroethylene was occurring in this ground
water. It is possible that sampling these wells results in the mixing of ground water that is
stratified with respect to redox condition in the upper zones of this aquifer, thus yielding
conflicting geochemistry.
Much of the experience base for the range of hydrogen concentrations was built in relatively
warm ground water in the southeastern United States, or using laboratory systems maintained
at room temperature or above. Ground-water temperature at the TCAAP varied from 9.6 °C to
12.4 °C (see Table 5). Recently the typical range of hydrogen concentrations has been
examined for colder ground water more representative of TCAAP. The work of Jakobsen et al.
(1998) predicts that the characteristic hydrogen concentration of sulfate reduction at 10 °C
would be 0.3 nanomolar, well below the range predicted by Lovley et al. (1994). If temperature
has the same effect on the threshold for reductive dechlorination, the criterion of 1 nanomolar in
the U.S. EPA Protocol (U.S. EPA, 1998) may be conservative.
Evaluation of the Protocol
One objective of this study was to assess the accuracy of the EPA natural attenuation protocol
in predicting the fate of a ground-water plume at a site where natural attenuation was clearly
ineffective at containing ground-water contamination. TCAAP was chosen for this test because
the plume extends over five miles from the on-site source area and is very well characterized.
The ten years of sampling over 225 monitoring wells, in addition to numerous private and
municipal wells, have established not only the extent of the plume but also concentration trends
at many locations. Therefore, the results of this natural attenuation study could be verified with
an extensive contaminant database and subsurface investigation.
The major conclusions of this evaluation of the EPA Natural Attenuation Protocol are:
The screening analysis indicated that the geochemistry of the ground water did
not favor the reductive dechlorination of solvents and that a natural attenuation
remedy at this site was unlikely. The screening evaluation showed that the ground
water was basically manganese and iron reducing, with no evidence of sulfate
reducing or methanogenic activity. Under these conditions, the Protocol predicts that
the potential for reductive dehalogenation is low. This screening prediction was
verified by the magnitude of the ground-water plume. Reductive dehalogenation did
not prevent formation of a large plume.
Application of the Protocol extracted a rate of biodegradation that was slow.
The biodegradation rate at this site was estimated at 0.17 to 0.28 per year for TCE.
This rate is slow compared to other sites where natural attenuation appears to be an
15
-------
effective remedy. This low rate, coupled with the high ground-water velocity, is
consistent with the observed development of the very large solvent plume at the
TCAAP site. This biodegradation rate was derived from four sampling events at a
limited number of wells along what is considered the longitudinal axis of the plume,
consistent with the approach for extracting rate constants that is illustrated in the
Protocol.
The BIOPLUME III modeling accurately predicted the extent and magnitude of
the solvent plume at this site. The model was calibrated with the ground-water data
available from previous site investigations consistent with the Protocol's emphasis on
natural attenuation verification efforts. Even with a rate for biodegradation rate
included, the modeling indicated that a large ground-water plume would develop.
Actual ground-water contaminant data from the extensive network of monitoring wells
at the site confirm the accuracy of these modeling predictions.
The relatively high hydrogen concentrations detected in the iron/manganese-
reducing ground water environment were not predicted by the Protocol. More
research is needed in manganese reducing ground-water environments to correlate
hydrogen concentration to the dominant electron accepting processes. However, it
also demonstrated that hydrogen concentrations can, under some circumstances, be
a better predictor of the dehalogenation of trichloroethylene to cis-dichloroethylene at
a site than the geochemistry alone, underscoring the importance of hydrogen sam-
pling in natural attenuation studies as detailed in the Protocol and supporting
literature.
Overall, the study accurately predicted that natural attenuation processes in the ground water at
TCAAP would be insufficient to contain the ground-water contaminants within acceptable limits
at the site. These results confirm the ability of the Protocol to evaluate and successfully predict
the fate of ground-water plumes at similar sites that are not contained through natural
attenuation processes.
Conclusions Applied to TCAAP
Based on the concentration of hydrogen in the ground water, there is strong evidence that
biological degradation is destroying chlorinated solvents currently present in ground water at
TCAAP. Natural attenuation through biodegradation is the most plausible explanation for the
apparent reduction in the concentration of chlorinated organic compounds as the plume moves
downgradient from the source. Natural biodegradation complements the on-going efforts to
extract contaminated ground water at the source, and should greatly reduce the time required to
reduce the concentration of contaminants to U.S. EPA drinking water standards. If the rate of
natural biodegradation exhibited in the last ten years continues for the next twenty years, the
portion of the aquifer downgradient from TCAAP will be reclaimed.
16
-------
Extension to Other Sites
The Protocol for the Natural Attenuation of Chlorinated Solvents in Ground l/1/aterwas originally
developed from experiences with chlorinated solvents in unconsolidated sandy aquifers. At
these sites, the cost of monitoring wells was relatively cheap, and additional wells could be
installed to fill gaps in the monitoring data. In this study, the methods and approaches of the
Protocol for the Natural Attenuation of Chlorinated Solvents in Ground Water were applied to
wells that were available from previous site characterization efforts. The existing wells (several
hundred) were originally designed to delineate a plume, and to design a pump-and-treat
remedy. The number of wells that were installed to meet these goals were severalfold fewer
than would be necessary to justify a selection of natural attenuation as a sole remedy.
The number of wells available for study at the TCAAP was equal to or greater than the number
typically available at large and complex sites where plumes of contamination extend several
hundred feet deep into consolidated rock. It is unlikely that other large and complex sites will
have better infrastructure for monitoring than does the TCAAP. Therefore it is unlikely that the
Protocol can be used with existing monitoring wells to justify the selection of natural attenuation
as the sole remedy at large and complex sites.
The study shows that the Protocol can be used to evaluate the contribution of natural
biodegradation to a cleanup that is achieved through active control of the source of contamina-
tion. The information provided can be used to estimate the time required to reach cleanup
goals, and to design or optimize strategies for long-term monitoring.
References
American Public Health Association (APHA). 1975. Standard Methods, 14th edition, p. 277.
Barrio-Lage, G.A., F.Z. Parsons, R.M. Narbaitz, and P.A. Lorenzo. 1990. Enhanced anaerobic
biodegradation of vinyl chloride in ground water. Environ. Toxicol. Chem., 9:403-415.
Bjerg P.L., R. Jakobsen, H. Bay, M. Rasmussen, H.-J. Albrechtsen, and T.H. Christensen.
1997. Effects of sampling well construction on H2 measurements made for characterization
of redox conditions in a contaminated aquifer. Environ. Sci. Technol., 31(10):3029-3031.
Bradley, P.M., and F.H. Chapelle. 1996. Anaerobic mineralization of vinyl chloride in Fe(lll)-
reducing aquifer sediments. Environ. Sci. Technol., 30:2084-2092.
Bradley, P.M., and F.H. Chapelle. 1997. Kinetics of Dichloroethylene and VC mineralization
under methanogenic and Fe(lll)-reducing conditions. Environ. Sci. Technol., 31:2692-2696.
Bradley, P.M., F.H. Chapelle, and J.T. Wilson. 1998a. Field and laboratory evidence for
intrinsic biodegradation of vinyl chloride contamination in a Fe(lll)-reducing aquifer. J.
Contam. Hydrol., 31:111-127.
Bradley, P.M., J.E. Landmeyer, and R.S. Dinicola. 1998b. Anaerobic oxidation of [1,2-
14C]dichloroethene under Mn(IV)-reducing conditions. Appl. Environ. Microbiol., 64:1560-
1562.
17
-------
Camp Dresser & McKee, Inc. 1991. Phase 1A Multi-Point Source Ground Water Remedial
Investigation, New Brighton/Arden Hills, Minnesota, Final Report, Volume 1 of 2 prepared
for the Minnesota Pollution Control Agency, February 1991.
Chapelle, F.H., D.A. Vroblesky, J.C. Woodward, and D.R. Lovely. 1997. Practical considerations
for measuring hydrogen concentrations in groundwater. Environ. Sci. Technol., 31(10):2873-
2877.
Clarke, J.N. 2000. Reductive dechorination of chlorinated aliphatic compounds as a result of a
petroleum fuel release: case study of the former Phoenix bus terminal, Phoenix, Arizona. ]n:
Natural Attenuation Considerations and Case Studies, Remediation of Chlorinated and
Recalcitrant Compounds. G.B. Wickramanayake and A.R. Gavaskar, editors. The Second
International Conference on Remediation of Chlorinate and Recalcitrant Compounds,
Monterey, California, May 22-25, 2000. pages 105-112.
Davis, J.W. and C.L. Carpenter. 1990. Aerobic biodegradation of vinyl chloride in ground water
samples. Appl. Environ. Microbiol., 56:3878-3880.
Fennell, D.E. and J.M. Gossett. 1998. Modeling the production of and competition for hydrogen
in a dechlorinating culture. Environ. Sci. Technol., 32:2450-2460.
Galli, R. and P.L. McCarty. 1989. Kinetics of biotransformation of 1,1,1-trichloroethane by
Clostridium sp. Strain TCAIIB. Appl. Environ. Microbiol., 55:845-851.
Hartmans, S., and J.A.M. de Bont. 1992. Aerobic vinyl chloride metabolism in Mycobacterium
aurum Li. Appl. Environ. Microbiol., 58:1220-1226.
Jakobsen, R., H-J. Albrechtsen, M. Rasmussen, H. Bay, P.L. Bjerg, and T.H. Christensen.
1998. H2 concentrations in a landfill leachate plume (Grindsted, Denmark): In situ energetics
of terminal electron acceptor processes. Environ. Sci. Technol., 32(14):2142-2148.
Klecka, G.M., S.J. Gonsior, and D.A. Markham. 1990. Biological transformations of 1,1,1-
trichloroethane in subsurface soils and ground water. Environ. Toxicol. Chem., 9:1437-
1451.
Klier, N.J., West, R.J. and P.A. Donberg. 1999. Aerobic biodegradation of dichloroethylenes in
surface and subsurface soils. Chemosphere, 38(5): 1175-1188.
Lovley, D.R., F.H. Chapelle, and J.C. Woodward. 1994, Use of dissolved H2 concentrations to
determine distribution of microbially catalyzed redox reactions in ground water. Environ. Sci.
Technol., 28:1255.
Maymo-Gatell, X., V. Tandoi, J.M. Gossett, and S.H. Zinder. 1995. Characterization of an H2-
utilizing enrichment culture that reductively dechlorinates tetrachloroethylene to vinyl chloride
and ethene in the absence of methanogenesis and acetogenesis. Appl. Environ. Microbiol.,
61:3928-3933.
McCarty, P.L. 1996. Biotic and abiotic transformations of chlorinated solvents in ground water.
In the Proceedings of the EPA Symposium on Natural Attenuation of Chlorinated Organics in
Ground Water, EPA/540/R-96/5098.
Sewell, G.W. and S.A. Gibson. 1991. Stimulation of the reductive dechlorination of
tetrachloroethylene in anaerobic aquifer microcosms by the addition of toluene. Environ.
Sci. Technol., 25(5):982-984.
18
-------
U.S. EPA. 1997. Use of Monitored Natural Attenuation at Superfund, RCRA Corrective Action,
and Underground Storage Tank Sites. Office of Solid Waste and Emergency Response
Directive 9200.4-17.
U.S. EPA. 1998. The Technical Protocol for Evaluating Natural Attenuation of Chlorinated
Solvents in Ground Water, EPA/600/R-98/128.
Vogel, T.M. and P.L. McCarty. 1985. Biotransformation of tetrachloroethylene to trichloroethylene,
dichloroethylene, vinyl chloride, and carbon dioxide under methanogenic conditions. Appl.
Environ. Microbiol., 49:1080-1083.
Vogel, T.M. and P.L. McCarty. 1987. Abiotic and biotic transformations of 1,1,1-trichloroethane
under methanogenic conditions. Environ. Sci. Technol., 21:(12):1208-1213.
Vogel, T.M., C.S. Griddle, and P.L. McCarty. 1987. Transformations of halogenated aliphatic
compounds. Environ. Sci. Technol., 21:(8):722-736.
Yang, Y. and P.L. McCarty. 1998. Competition for hydrogen within a chlorinated solvent
dehalogenating anaerobic mixed culture. Environ. Sci. Technol., 32:3591-3597.
19
-------
Table 1. Monitoring Wells Selected for an Evaluation of the Natural Attenuation of Chlorinated
Solvents in Ground Water at TCAAP
Well Aquifer Screened Well Screen Depth Screen Riser
Depth Length Ground
(feet) (feet) Water
(feet)
Background
03U113
03L113
Lower Hillside Sand
Deep Sands
158
430
20
20
118
118
PVC
PVC
PVC
PVC
Source Area Pump-out Wells
03U314
03U317
(SC-2) Upper Hillside Sand
(SC-5) Lower Hillside Sand
186.5
430
20
10
132
114
ss
ss
s
s
Source Area Monitoring Wells
03U020
03M020
03L020
04U020
03U002
03M002
03L002
04U002
Upper Hillside Sand
Middle Hillside Sand
Lower Hillside Sand
Jordan Sandstone
Upper Hillside Sand
Middle Hillside Sand
Lower Hillside Sand
Prairie du Chien
135
185
229
260
89
167
238
280
20
20
20
10
20
20
20
10.5
106
106
106
103
70
73
73
75
ss
ss
ss
ss
ss
ss
ss
ss
PVC
PVC
PVC
PVC
PVC
PVC
PVC
PVC
Perimeter Pump-out Well
03F306
(B-5) Prairie du Chien
232
100
82.9
ss
s
Downgradientfrom TCAAP 4,600 feet
191942
04U821
Prairie du Chien
Prairie du Chien
183
147
18
5
47
43.6
open
hole
SS
s
s
Downgradientfrom TCAAP 17,000 feet
04U872
Prairie du Chien
382
30
137.9
open
hole
s
SS = Stainless Steel
S = Steel
20
-------
Table 2. Geochemical Indicators of Anaerobic Biotransformation of Chlorinated Solvents in Ground
Water at TCAAP (Reported are Values for the June 1998 Round of Sampling)
Well
Oxyge Mangane Iron (II) Alkalinity Sulfate Methane Nitrate
mg/liter mg/liter mg/liter mg/liter mg/liter mg/liter mg/liter
Background
03U113
03L113
1.8
0.1
0.02
0.33
0.0
0.0
247
224
13.3
4.5
0.001
0.018
0.59
Source Area Pump-out Wells
03U314(SC-2) 0.4
03U317(SC-5) 1.0
Source Area Monitoring Wells
03U020 6.7
03M020 0.1
03L020 0.5
04U020 0.1
0.65
0.37
0.25
0.0
236
234
17.2
17.2
0.011
0.001
0.44
0.59
0.70
0.80
6.0
0.8
0.0
0.0
0.25
0.65
291
231
227
231
32.6
2.9
4.2
5.9
0.001
0.035
0.015
0.007
2.35
0.11
<0. 1
<0. 1
03U002
03M002
03L002
04U002
3.3
0.2
0.0
0.0
0.1
1.1
0.34
0.0
0.0
0.25
0.15
274
284
230
237
14.2
10.7
6.2
2.8
0.001
0.001
0.008
0.011
0.62
0.57
<0. 1
<0. 1
Perimeter Pump-out Wells
03F306 (B-5) 0.5
0.66
Downgradientfrom TCAAP 4,600 feet
191942 0.1 <0.1
04U821 0.2 <0.1
0.0
0.7
0.1
242
7.4 0.004 0.17
260 12.4 0.005 <0.1
220 10.4 0.001 0.25
Downgradientfrom TCAAP 17,000feet
04U872 0.3 <0.5 0.25
269 25.1 0.017 <0.1
The ground water is depleted of oxygen, but there is little evidence of intense anaerobic microbial
activity, such as the accumulation of methane. There is accumulation of Iron (II) and Manganese (II),
but their concentrations are low.
21
-------
Table 3. Apparent Attenuation of Concentrations of TCE and 1,1,1-TCA in Selected Monitoring Wells
with Distance Downgradient of Source Area D on TCAAP (The Concentrations are the
Highest Detected in the Years Prior to Initiation of Pump-and-treat)
Well
03U020
03L002
04U821
04U850
04U872
04U882
Distance
Downgradient
(feet)
2,000
3,500
7,800
13,200
22,800
27,000
Highest TCE
Concentration
(|jg/liter)
11,000
2,700
950
910
168
67
Highest 1,1,1- Estimated Travel
TCA
Concentration
(|jg/liter)
6,000
4,200
170
71
19
7.7
Time
(years)
1.25
2.20
4.9
8.30
14.3
17.0
Table 4. Apparent Rate of Attenuation in Concentration of TCE in Selected Monitoring Wells at
TCAAP
Well
03U096
03U094
03U003
03U079
03U093
03U094
03L002
03L020
03L883
04U002
04U020
04U848
TGRS
MEAN
Time
Interval in
Regression
(years)
10
9
21
16
27
9
14
4
13
13
15
18
8
Number of
Samples in
Regression
10
9
21
16
27
9
14
4
13
13
15
18
8
First-Order
Rate of
Attenuation
(per year)
0.365
0.359
0.278
0.353
0.691
0.359
0.383
0.430
0.376
0.348
0.556
0.574
0.229
0.41
95% Confidence
Interval on the Rate of
Attenuation (per year)
0.202
0.245
0.206
0.277
0.574
0.245
0.296
0.218
0.239
0.283
0.457
0.486
0.185
0.30
22
-------
Table 5. Concentration of Hydrogen, Oxidation/Reduction Electrode Potential Against an Ag/AgCI
Reference Eectrode, and Temperature of Ground Water Sampled from TCAAP
Well Order Hydrogen
Shallow nanomolar
to Deep
Date of
Background
03U113
03L113
Source Area Moni
03U020
03M020
03L020
04U020
Source Area farthi
03U002
03M002
03L002
04U002
Downgradient 4,6
191942
04U821
Downgradient 17,000 feet from TCAAP
04U872 0.90 0.44
Redox
millivolts
Temp.
6/1997 6/1998 6/1997 6/1998 6/1997 6/1998
1.0
2.0
Ing Wells
1.0
2.0
3.0
4.0
/H/™\\ A/K^/"l K*O/H I
uowngraoi
1.0
2.0
3.0
4.0
i feet from
0.84
12.9
3.05
1.41
0.79
0.38
IQK^t
lent
0.76
0.36
1.07
1.5
TCAAP
0.97
1.16
0.8
1.16
3.2
7.5
13.6
5.5
1.4
1.3
0.2
36
0.86
0.18
-113
-123
126
104
-133
-171
94
97
45
-74
-117
-78
93.8
-52.0
97.0
-175.1
-115.0
-166.6
106.0
2.0
-5.1
-79.6
-162.0
-122.0
11.3
10.6
12.4
9.57
10.8
10.8
11.6
11.0
11.3
11
11.3
10.9
10.8
10.5
11.0
11.8
11.0
10.6
11.4
11.7
10.6
10.5
11.2
10.7
-140
-114.0 10.9
10.0
23
-------
lite D
=. 850
Sand -',-•
Jwm Cities
Formation
'
550
950
900
Lawrenee Formslior
800
aip: .ML
Cross Section A-A1 Srte D
Figure 1. Geological cross section at the Twin Cities Army Ammunition Plant, showing the origin of the
chlorinated solvent plume in ground water. The section A' to A extends from Northeast to
Southwest in the direction of ground-water flow (from right to left in the figure). The
distance along the section from A' to A is 4,500 feet. Site D is the most important source of
chlorinated solvents in ground water. Presented are the concentrations of trichloroethylene
in mg/liter. The vertical distribution of chlorinated solvents is determined by the distribution of
solvents in vertical clusters of monitoring wells.
24
-------
1649
Mississippi River
TCAAP
Chlorinated Solvent
Figure 2. Location of the Twin Cities Army Ammunition Plant (TCAAP) and the associated plume of
chlorinated solvents in ground water with respect to the Mississippi River, to Interstate 35W,
and to Interstate 694. The heavy dotted line at the head of the plume is the location of the
vertical cross section presented in Figure 1.
25
-------
VV-' Y A\ /N-mr^ X <^ \
B, y\ %^=^ \
IM
-------
£=
o
£=
CD
o
o
o
10000
1000
100
s
• i
* Trichloroethylene
a cis-Dichloroethylene
-5000
5000 10000 15000
Distance Downgradient (feet)
20000
25000
Figure 4. Relative concentration of trichloroethylene and its biological reduction product
cis-dichloroethylene in ground water from TCAAP. Wells in the study were sampled in June
1996, December 1996, June 1997, November 1997 and June 1998.
10000
-c- 1000
£=
°
CD
o
o
O
100
-5000
og B
* 1,1,1-Trichloroethane
D 1,1-Dichloroethylene
n
9
a
5000 10000 15000
Distance Downgradient (feet)
20000
25000
Figure 5. Relative concentration of 1,1,1-trichloroethane and its abiotic transformation product 1,1-
dichloroethylene in ground water from TCAAP. Wells in the study were sampled in June
1996, December 1996, June 1997, November 1997 and June 1998.
27
-------
g
"ro
"c
CD
o
£=
O
O
10000
1000
100
10
Figure 6.
-5000
» 1,1,1-Trichloroethane
n 1,1-Dichloroethane
n
n
D
°
5000 10000 15000 20000 25000
Distance Downgradient (feet)
Relative concentration of 1,1,1-trichloroethane and its biological reduction product 1,1-
dichloroethane in ground water from TCAAP. Wells in the study were sampled in June
1996, December 1996, June 1997, November 1997 and June 1998.
- m il
Figure 7.
•''
-
It
M
The triangles connected by lines are the water table elevations used to calibrate the initial
conditions in BIOPLUME III. Ground-water flow is from right to left; the interpreted contours
provided on the map used to calibrate water elevations have a difference in elevation of ten
feet.
28
-------
- .
*::-*
-IM,
II.'
I ,;.V«S=. 1,'UO.
Figure 8. Kriged values for horizontal hydraulic conductivity used to calibrate BIOPLUME III. Values
are in feet per day.
10
2' ;
3 ." > ;8
4 5 '6
_ : : : : : : : : :: : : ua—A«,HW,«
Figure 9. Cells used to calibrate BIOPLUME III. The cells labeled with a number are modeled with
either an injection well or an extraction well as described in the text. The cells with an * are
log points used to calibrate the distributed parameters in BIOPLUME III.
29
-------
.5
1 Mile
Figure 10. The grid used to calibrate BIOPLUME III superimposed on surface features at TCAAP. The
locations labeled BP or SC are pump-and-treat wells at TCAAP. The model only allows one
extraction or injection well per cell in the grid. The cells labeled 1 through 7 are modeled with
an extraction or injection well as described in the text and Figure 9. Well B-5 is also labeled
03F306, Well SC-2 is also labeled 03U317 and Well SC-5 is also labeled 03U314.
30
-------
- ill SI. I
Figure 11. Heads modeled by BIOPLUME III just prior to beginning pump-and-treat activities in 1988.
. «i ».
*-.-.*
IHJUU..
Figure 12. Steady-state heads modeled by BIOPLUME III for the year 2020, after initiation of pump-and-
treat starting in 1988.
31
-------
JmdMmlHt I**'
" ' =" : - f f *!wff »-WiW f>V
1.
lMfj«>* Vifeflif1'
ft*
H %-ii
-------
1.
1,
l,::a-:ij h, -t-K *. Et ii»ft
T r srT-y*f ? 5 ? "s *s p* t
1,
v#!f£:T. : "•
5,
a (12
f s i KAB'Swis": !ii* Jft'-ift te
I,
1st .'irds-f iit i'? CfislT ;•».«•&&
ar
S:
-------
1969 - NATURAL ATTENUATION RATE OF .17 PER YEAR
20000-
10000-
0 5000 10OOO 1SOOO 20000 2(000 30000 85000 40000 45000 500OO
1969 - NATURAL ATTENUATION RATE OF .18 PER YEAR
0 SOOO 10000 1SMXJ 20000 28000 30000 36000 40000 45000 50000
1969 - NO NATURAL ATTENUATION
20000-
15000-
10000-
SOOO 10000 15000 20000 25000 30000 35000 40000 45000 50000
-r
Figure 17. Projections of concentrations of TCE at TCAAP with and without bioattenuation in 1969.
Distances are in feet, concentrations are in |ig/liter.
34
-------
1988 - NATURAL ATTENUATION RATE OF .17 PER YEAR
20000-
10000-
0 5000 10000 15000 20000 26000 30000 35000 40000 45000 60000
1988 - NATURAL ATTENUATION RATE OF .28 PER YEAR
15000-
5OOO 10000 1SOOO 20000 25000 30000 35000 40000 4SOOO SOOOQ
1988 - NO NATURAL ATTENUATION
20000-
10000-
0 MOO 10000 15000 20000 26000 30000 35000 40000 45000 60000
Figure 18. Projections of concentrations of TCE at TCAAP with and without bioattenuation in 1988.
Distances are in feet, concentrations are in |ig/liter. Pump-and-treat began in 1988.
35
-------
1998 - NATURAL ATTENUATION RATE OP .17 PER YEAR
20000-
10000-
5000 10000 15000 20000 25000 30000 35000 4OOOO 45000 50000
1998 - NATURAL ATTENUATION RATE OF .28 PER YEAR
5000 10000 15000 20000 25000 30000 38000 40000 45000 fflJOOO
1998 - NO NATURAL, ATTENUATION
20000-
15000-
5000 10000 15000 20000 25000 30000 35000 40000 45000 50000
10000- =
Figure 19. Projections of concentrations of TCE at TCAAP with and without bioattenuation in 1998.
Distances are in feet, concentrations are in |ig/liter. A pump-and-treat remedy in the source
began in 1988.
36
-------
2OO8 - NATURAL ATTENUATION RATE OF .17 PER YEAR
20000-
10000-
HJOO 10000 15OOO 20000 2GOOO 30000 39000 40000 450OO 50000
2008 - NATURAL ATTENUATION RATE OF .28 PER YEAR
20000-
15000-
1000 10000 16000 20000 2SOOQ 30000 3SQOO 40000 41000 SOOOO
2OO8 - NO NATURAL ATTENUATION
15000-
0 5000 10000 1SOOD 20000 26OOQ 30000 35OOO 40000 45000 50000
Figure 20. Projections of concentrations of TCE at TCAAP with and without bioattenuation in the year
2008. Distances are in feet, concentrations are in |ig/liter.
37
-------
2018 - NATURAL ATTENUATION RATE OF .17 PER YEAR
20000-
10000-
5000 10000 15000 20000 25000 30000 3SOOO 40000 45000 50000
201 i - NATURAL ATTENUATION RATE OF .28 PER YEAR
20000-
10000-
SOOO 10000 15000 20000 26000 30000 35000 40000 45000 50000
2018 - NO NATURAL ATTENUATION
5000 10000 15000 20000 25000 3OOOO 35000 40000 45000 50000
Figure 21. Projection of concentrations of TCE at TCAAP with and without bioattenuation in the year
2018. Distances are in feet, concentrations are in |ig/liter.
38
------- |