United States
Environmental Protection
Agency
National Exposure
Research Laboratory
Research Triangle Park, NC 5711
Research and Development
EPA/600/SR-98/020
July 1998
Project Summary
Monitoring and Assessment of
In-Situ Biocontainment of
Petroleum Contaminated
Ground-Water Plumes
R. Ryan Dupont, Darwin L. Sorensen, Marian Kemblowski, Mark Bertleson,
Dietrick McGinnis, Idris Kamil, and Yang Ma
Abstract
This two-year field research project
was conducted to assess the potential
for natural attenuation of gasoline-con-
taminated ground-water plumes at two
underground storage tank (LIST) sites in
northern Utah. An evaluation of rapid
site-assessment techniques for plume
delineation and subsurface site charac-
terization was carried out using cone
penetrometer and ambient temperature
headspace (ATM) analysis techniques.
An approach was developed for the col-
lection and evaluation of soil and
ground-water monitoring data for de-
termining the efficacy of in-situ
biocontainment and to identify "stabi-
lized" fuel-impacted ground-water
plumes. A decision framework was de-
veloped to guide the user in data collec-
tion, data reduction, data interpretation,
and decision making efforts to evaluate
the nature and potential extent of intrin-
sic plume bioattenuation taking place
under a given set of site conditions. Data
collected from the two field sites are pre-
sented and evaluated in detail, the rate
and extent of the natural attenuation of
ground-water plumes are quantified, and
recommendations regarding source re-
moval action are made for these two field
sites.
The information in this document has
been funded wholly or in part by the
United States Environmental Protection
Agency under CR 818835-01 to the Utah
Water Research Laboratory, Utah State
University. It has been subjected to the
Agency's peer and administrative re-
view, and it has been approved for pub-
lication as an EPA document. Mention
of trade names or commercial products
does not constitute endorsement or rec-
ommendation for use.
This Project Summary was developed
by the National Exposure Research
Laboratory's Characterization Research
Division, Las Vegas, NV, to announce
key findings of the research project that
is fully documented in a separate report
of the same title (see Project Report or-
dering information at back).
Introduction
Ground-water quality impacts resulting
from leaking underground petroleum stor-
age tanks are a significant environmental
concern due to the sheer number of such
tanks and the extent of possible environ-
mental contamination when they leak. Many
of these tanks are being removed and re-
placed or upgraded to eliminate the source
of possible petroleum contamination to un-
derlying aquifers. However, there remains
a large number of sites with ground-water
contamination above existing water quality
limits (300,000 confirmed releases as of
June 1995, with an additional 100,000 re-
leases expected by 2000 (e.g., personal
communication, Hal White, U.S. EPA,
OUST, 1995). While contamination from pe-
troleum storage tank releases can have a
significant impact on public health and the
environment, active remediation of these
contaminated sites can be difficult and ex-
pensive. Expectations of having to apply
-------
active remediation at all sites has generally
not allowed the focusing of resources on
those sites representing the greatest threat
to public health and the environment.
When gasoline is released into the sub-
surface, it distributes among the soil, gas,
and aqueous phases that make up this en-
vironment. As it moves through the unsat-
urated zone, it leaves behind vapors con-
taining volatile gasoline components and
residual liquid hydrocarbons retained within
the soil matrix. A large fraction of gasoline
is water soluble and migrates to underlying
aquifer with infiltrating water. Plumes of
ground-water hydrocarbon contamination
spread within the soil environment by
ground-water advection and diffusion. Many
gasoline components are biodegradable,
and a primary mechanism for their trans-
formation in the subsurface is via biodeg-
radation. Biotransformation takes place
under a variety of soil pH and oxidation re-
duction potential (redox) conditions, and in-
volves various terminal electron acceptors
(oxygen, nitrate, manganese, iron, sulfate,
carbon dioxide, carbon). If biodegradation
of a contaminant proceeds at a rate greater
than or equal to the rate of advance of the
contaminant front, the plume will be effec-
tively contained, and can eventually be com-
pletely remediated if the source of contami-
nation is removed.
The only costs associated with site clean-
up under these intrinsic biocontainment
conditions are related to the costs for source
removal and for the necessary, on-going
monitoring of contaminant distribution and
movement overthe lifetime of the dissolved
contaminant plume. Without reliable con-
tainment information, contamination may
continue to spread, worsening existing con-
ditions, posing increased public health risk,
and increasing final clean-up costs. To ob-
tain intrinsic degradation process rates for
contaminants at fuel release sites, chemi-
cal/biological parameters indicative of in-situ
biodegradation processes taking place in
these environments must be well under-
stood and documented. Once this is accom-
plished, routine measurements of these
parameters, or a subset of these param-
eters, would allow identification of condi-
tions under which in-situ biodegradation
could be relied upon to provide natural at-
tenuation of a contaminant plume.
The goal of this project was to identify
and validate field monitoring techniques,
and data reduction and reporting methods
that can be utilized to rapidly and conclu-
sively demonstrate the existence of intrin-
sic biodegradation reactions at leaking LIST
sites. With this demonstration of the natu-
ral containment of a hydrocarbon plume, ra-
tional decisions can be made regarding the
need to actively remediate it to ensure the
protection of public health and the environ-
ment. Based on an evaluation of field and
laboratory water quality determinations, soil
core measurements, and companion mod-
eling results, recommendations have been
made regarding the selection of process
variables, monitoring procedures, and data
reduction and reporting methods that should
be implemented at hydrocarbon contami-
nated sites to document intrinsic bioreme-
diation of ground-water plumes.
Procedure
The research was conducted in four
phases. The first was a site assessment/
characterization phase in which contami-
nant distribution and site hydraulic charac-
teristics were determined using rapid field
characterization methods. The second
phase involved process monitoring, in which
field techniques were used to monitor
ground-water and soil gas characteristics
that reflect in-situ biodegradation reactions
taking place throughout the field sites. The
third phase was the reduction of field data
to yield estimates of total dissolved plume
contaminant mass, center of mass, mass
center trajectory, contaminant degradation
rates, and estimated source lifetime. The
final phase involved the use of a three-di-
mensional analytical model to provide in-
situ biodegradation process verification and
long-term predictions of fate of the plume
as attenuated by natural biodegradation
reactions.
Two field sites with known contamination
from gasoline storage tanks were selected
for this study: the Hill AFB site and the
Layton, Utah site. Both sites were "aged"
with initially "constant" source terms where
it was reasonable to hypothesize that natu-
ral biodegradation reactions, and subse-
quent plume containment, had developed
and were quantifiable. Single and multi-level
ground-water monitoring probes were
placed within and around the area of the
contaminant plume at both sites. These
wells included upgradient "background"
wells, and wells within the area of contami-
nation, allowing the definition of the bound-
aries of the plumes with some certainty. A
gradient of chemical and biological condi-
tions was observed throughout each plume
so that transformation/degradation rates,
mass transfer rates, etc., could be esti-
mated. Plume characterization was carried
out using cone penetrometry and 5/8-in di-
ameter piezometer ground-water sampling
wells to rapidly and inexpensively collect soil
textural information and ground-water data
from sampling locations throughout each
plume. During long-term monitoring, site
characterization information was collected
using all of the ground-water sampling
points.
The following data were collected
throughout the site during the initial site
characterization phase and six times dur-
ing the process monitoring phase of the
project: in ground water-pH, O2, Fe2+, Mn2+,
NO3-, SO42', CI-; in soil gas—O2, CO2; and
in both soil gas and ground water- aromatic
hydrocarbons and total petroleum hydrocar-
bons (TPH).
Field data were reduced to generate es-
timates of total dissolved mass of contami-
nant using the Thiessen area method
(Chow et al., 1988) to assign a specific
plume area to each ground-water monitor-
ing point. Total dissolved plume mass was
estimated for each sampling event, while
the center of the total dissolved plume mass
was estimated by taking the first moment
about a defined axis of the masses repre-
sented at each sampling location within the
contaminant plumes. The movement of the
plume centroid over time was described
based on the changes in its absolute posi-
tion, and contaminant plume velocities were
calculated between each sampling interval.
Contaminant degradation rates were esti-
mated based on the change in total dis-
solved plume mass between each sampling
interval. Aquifer assimilation capacity was
estimated based on the change in terminal
electron acceptor concentration and mass
inside versus outside of the plume.
Finally, an analytical, three-dimensional
ground-water fate-and-transport model,
which accounts for advection, dispersion,
contaminant sorption and contaminant deg-
radation, was applied to both field sites to
validate intrinsic biodegradation reactions
observed at these sites. Hydraulic proper-
ties were selected based on measured field
data and information regarding the nature
of soil below each site. A source configura-
tion was established for each site. Model
input variables for measured source con-
centrations, contaminant properties, and
time since the release were varied to as-
sess the sensitivity of the model to these
parameters and to determine those combi-
nations of parameters producing the best
model fit of centerline contaminant concen-
trations for a tester data set. The effects of
source removal on the lifetime of the plume
and the maximum plume travel distance
were assessed using the site-specific, field-
data calibrated model.
Results and Discussion
Site Assessment/Monitoring
Techniques
Cone Petrometer Testing (CPT)
Methods
The application of cone penetrometer
testing (CPT) for soil textural data collec-
-------
tion coupled to piezometer well placement
is appropriate for initial site investigation
activities at sites where no prior data regard-
ing ground-water plume characterization
are available. These techniques can also
be effectively applied at sites where exist-
ing data are limited in scope and detail, as
was exemplified from the results obtained
from CRT and ground-water probe data
collected at the Hill AFB site.
The Hill AFB site was the location of a
former 18,000-gallon LIST that was exca-
vated in 1989. At the time of excavation,
holes were observed in the tank and petro-
leum odors were detected in the tank pit.
Five conventional ground-water monitoring
wells were installed at the site and soil gas.
Soil boring and limited ground-water moni-
toring data collected from the site from 1989
to 1991 led to the conceptual site model
suggesting a plume migrating to the south-
west.
Soon after site investigation activities
began, it became apparent that the initial
conceptual model of contaminant distribu-
tion and plume migration at the Hill AFB site
was significantly flawed. Ground probe
samples that were collected near the source
of contamination within the plume bound-
aries showed hydrocarbon levels at back-
ground concentrations. CRT analysis and
ground-water piezometer sampling were
moved into the north and northwest regions
of the site in an attempt to locate the con-
taminant plume. Field hydrocarbon screen-
ing data collected from these piezometers
were successful in identifying the true north-
westerly direction of the hydrocarbon plume
as indicated by the ground-water concen-
tration data shown in Figure 1. CRT mea-
surements collected in this study also pro-
vided insight into ground-water flow condi-
tions existing below the Hill site. Figure 2
clearly indicates a distribution of high and
low permeability deposits that correspond
to what appears to be an old stream bed
resulting in channeling of the regional west-
erly flow to the northwest direction beneath
the Hill site.
It is not known how widespread the find-
ings for the Hill AFB site are in terms of the
development of a less than complete pic-
ture of actual plume distribution and trans-
port using limited, conventional site investi-
gation data. It was found, however, that the
use of CRT methods facilitated the rapid
collection of soil textural information and the
placement and sampling of ground-water
probes that allowed a much more compre-
hensive picture of both structural and chemi-
cal features of the subsurface than was
possible using soil core sampling and labo-
ratory analysis procedures routinely applied
at LIST sites. This added information is com-
bined with the complete elimination of soil
• Single Level Piezometers
* Multilevel Piezometers
Q Monitoring Wells
-20 0 20 40 60 80 100 120 140 160 180
North Coordinates (ft)
Figure 1. Initial ground-water plume hydrocarbon data developed from field screening headspace
analyses conducted at the Hill AFB site, July 1992.
180
160-
140-
120-
100-
• Single Level Piezometers
* Multilevel Piezometers
O Monitoring Wells
-20
-20
20 40 60 80 100 120 140 160 180
North Coordinates (ft)
Figure 2. Textural map for soils at the ground-water table generated from CRT data collected at the Hill
AFB site in July 1992.
-------
cuttings and large volumes of contaminated
ground water that are often costly and
regulatorily challenging to manage, making
the screening techniques used in this study
ideal for application at many sites. The im-
portance of additional insights into local
ground-water flow conditions that can be
provided by these CRT techniques is also
unquestionable, especially when consider-
ing an intrinsic remediation management
option at a site.
Ambient Temperature Headspace
(ATM) Results
Two data sets (July and December 1992)
were collected from each field site to evalu-
ate the representativeness of ATM field
screening techniques compared to results
generated from standard laboratory-gener-
ated purge-and-trap hydrocarbon measure-
ments. This comparison between field- and
laboratory-generated results was based on
total hydrocarbon determinations using se-
rial dilutions of a gasoline-saturated water
standard in the field and hexane-equivalent
concentrations for laboratory results.
The ratio of field- and laboratory-deter-
mined ground-water hydrocarbon concen-
trations was quite variable, ranging from a
low of 0.0 to a high of 446. This ratio was
generally consistent for a given sampling
location between the two sampling times;
however, some ratios varied by one to two
orders of magnitude. The average ratio of
field- to laboratory-determined ground-wa-
ter hydrocarbon concentrations was greater
than 1 for both sampling events at both sites,
and ranged from 4.9 to 32.1. This result
suggests that the field ATM procedures used
in this study provided a conservative esti-
mate of contaminant concentration from
most sampling locations by a factor of 5 to
30.
A statistically significant relationship was
found between laboratory and field data for
both sites. A background field concentration
of 279 |ig/L was found when laboratory re-
sults indicated azeroTPH concentration at
the Hill AFB site, while at the Layton site, a
ground-water concentration of 3,680 |ig/L
was indicated from ATM readings before
laboratory purge-and-trap methods de-
tected TPH contamination. The slope of the
field versus laboratory hydrocarbon concen-
tration relationship was less than one for the
Hill AFB site, indicating that when true
ground-water concentrations (as defined by
laboratory purge-and-trap results) exceed
842|ig/L, field ATM values under-predict ac-
tual ground-water TPH concentrations. The
slope of the Layton relationship was greater
than one, however, indicating that unlike the
Hill AFB data, field ATH measurements con-
sistently over-predicted laboratory purge-
and-trap ground-water concentrations by
nearly a factor of two.
These findings suggest that there is a
general relationship between laboratory and
field ATH determined hydrocarbon concen-
trations, but this relationship is very much
site-specific. The utility of these field ATH
measurements appears then to be in the
initial site assessment phase where rapid,
semi-quantitative results are used for de-
tailed plume delineation efforts. Field ATH
measurements can be used to effectively
guide initial ground-water quality investiga-
tions and to optimize ground-water moni-
toring probe and monitoring well placement
for long-term site monitoring. Once this ini-
tial screening is completed, however, labo-
ratory analyses are necessary to provide
accurate ground-water quality data for fur-
ther site fate-and-transport and intrinsic re-
mediation evaluation.
Hill AFB Site Intrinsic
Remediation Results
Site Background Data
This site is located in the west area of Hill
AFB south of the city of Ogden, Utah. The
site is immediately north of Building 1141
that was used by the Air Force for small
vehicle maintenance. Generally, the regional
shallow and deep ground-water flow direc-
tion through the fluvial and lacustrine de-
posits underlying the base is from the moun-
tains on the east toward the Great Salt Lake
on the west. Hydraulic conductivity of the
shallow soils at the Building 1141 site, as
determined by slug tests performed on se-
lected wells by Engineering Science, were
found to range between 1.0x10-5 and 7.7
x 10-5 cm/s.
In December 1989, an 18,000-gallon bare
steel gasoline UST was excavated and no
free-phase hydrocarbons were observed at
this time, although observations of odors
and holes in the UST indicated that a re-
lease may have occurred. Two soil samples
and a ground-water sample were collected
and analyzed for benzene, toluene,
ethylbenzene, xylenes (BTEX), and total
petroleum hydrocarbons (TPH). Data from
this sampling event indicated that residual
and dissolved BTEX and TPH contaminants
were present in the soil and shallow ground
water. Field investigative activities con-
ducted in October and November 1990 in-
cluded a soil-gas survey, soil borings, instal-
lation of ground-water monitoring wells, col-
lection of soil and ground-water samples for
laboratory analyses, slug tests, and a site
survey. From these analyses, limited re-
sidual-phase and dissolved hydrocarbons
were documented in the soil and the upper
portion of the shallow aquifer immediately
downgradient and adjacent to the former
UST.
In March 1991, levels of dissolved con-
taminants in the ground water were similar
to those documented in November 1990. No
free-phase product was observed, indicat-
ing that the bulkofthe released product was
bound in capillary pore spaces of the shal-
low soils. In July and August 1992, CPTdata
were collected at 44 locations throughout
the Hill AFB site to augment the existing five-
well monitoring network. CPT data success-
fully identified a subsurface stream chan-
nel (Figure 2) at the ground-water table,
producing a significant northerly flow com-
ponent to the regional, westerly ground-
water flow.
Seven sampling events were used at the
Hill AFB site to describe the distribution and
movement of contaminants and electron
acceptors occurring at the site between April
1992 and January 1994. These data were
used to determine steady-state plume con-
ditions and to estimate total mass and mass
centervaluesforvariousanalyt.es at the site
overtime.
Intrinsic Remediation
Assessment
Determination of Steady-State
Plume Conditions
Based on centerline concentration data
observed at the Hill AFB site overtime, the
concentration profile for BTEX and TPH
constituents remained at pseudo-steady-
state after December 1992. To further evalu-
ate plume stability conditions, total dissolved
plume mass was estimated overtime.
To determine dissolved plume mass,
Thiessen areas were generated for each
sampling event using a fixed outer plume
boundary and individual areas determined
based on the actual sampling locations used
in a given sampling event. Specific com-
pound mass data shown in Figure 3 sug-
gest that steady-state conditions existed for
some of the compounds (benzene, toluene,
naphthalene) during a portion of the study
period, while a continuously declining mass
was seen for ethylbenzene and p-xylene
during the entire study period. All of the spe-
cific compounds of interest did show a sig-
nificant decline at the end of the study, how-
ever. Dissolved plume TPH mass confirmed
the non-steady-state conditions existing at
the Hill AFB site, as its mass was found to
exponentially decay over time. Center of
mass analyses showed migration of the
mass center downgradient from its location
in August 1992, although the total
downgradient distance traveled was small,
ranging from a maximum of 106 ft for TPH,
to a low of only 17 ft for naphthalene. This
-------
o Benzene
o Toluene
A Ethylbenzene
* p-Xylene
+ Naphthalene
Figure 3. Time course of total dissolved plume
mass estimates for BTEX and naphthalene con-
taminants at the Hill AFB site over the course of
the study.
decreasing contaminant mass and center
of mass moving downgradient suggest that
the source was finite, and despite plume mi-
gration, the reductions in contaminant mass
indicated assimilation was taking place.
Estimation of Parent Compound
Degradation Rate
Because the mass of contaminants in the
dissolved plume at the Hill site were found
to be decreasing over time, the plume ap-
peared to originate from a pulse source.
With a pulse source, contaminant degrada-
tion rates can be estimated by linear regres-
sion of total dissolved contaminant mass
data versus time. The results of these re-
gressions are shown in Table 1.
Estimation of Source Mass/
Lifetime
The projected lifetimes of the BTEX
ground-water plumes using this mass ver-
sus time regression approach were quite
short, being less than 2 weeks from the last
sampling event. This was due to the low
mass of contaminant remaining in the plume
as of January 1994. The naphthalene and
TPH plumes could persist for a much longer
time but were estimated to last only slightly
longer than 2 years at the current rate of
degradation. The TPH plume mass center
was projected to move the farthest of the
compounds analyzed in the study, a total of
156 ft downgradient from its position in
January 1994. This puts the mass center
approximately 50 ft downgradient of the
existing monitoring network, still within the
boundaries of Hill AFB, and without impact
to a downgradient receptor.
Predicting Long-Term Behavior
Consideration of a source removal sce-
nario for this site is not relevant as data in-
dicate that the initial contaminant mass has
been depleted from the original source area
through intrinsic processes of contaminant
dissolution, dispersion, and degradation.
The long-term behavior of the plume is
modeled using the degradation rates of the
contaminants within the "detached" dis-
solved plume and the rate at which the con-
taminants are migrating within the aquifer.
The results of this analysis are shown in
Table!
Decision Making Regarding
Intrinsic Remediation
Impacted Receptors
The long-term behavior of the contami-
nant plumes existing at the Hill AFB site pro-
jected from contaminant degradation and
transport data suggests that the maximum
extent of any plume of interest will only be
50 ft downgradient of the existing ground-
water monitoring network. This limited ex-
tent of potential contaminant migration en-
sures that no downgradient receptors will
be impacted by contamination at the Hill
AFB site over the projected lifetime of the
ground-water plumes that exists there.
Potential Aquifer Assimilative
Capacity
When the mass of electron acceptor mov-
ing onto a site equals or exceeds the sto-
ichiometric equivalent of dissolved contami-
nant in the plume, then it can be assumed
that the availability of the electron acceptor
will not limit future contaminant degradation
and plume attenuation. The electron accep-
tor concentrations quantified throughout the
contaminated portion of the Hill AFB site
were used along with the electron acceptor
concentrations measured in background
ground water to estimate the potential as-
similative capacity existing within the aqui-
fer. The hydrocarbon (HC) assimilative ca-
pacities related to dissolved oxygen (0.66
mg HC/L), nitrate (5.4 mg HC/L) and sul-
fate (10.7 mg HC/L) utilization were based
on the lowest observed concentration in the
background well. For iron (0.06 mg HC/L)
and manganese (0.02 mg HC/L), their as-
similative capacities were estimated based
on the smallest increase in the soluble con-
centrations of these solid phase electron
acceptors observed within the center of the
plume during the study.
Sulfate was the most significant electron
acceptor, accounting for more than 60 per-
cent of the potential assimilative capacity
at the site. The potential assimilative capac-
ity projected for the site was more than 90
times greater than that required to assimi-
late the TPH remaining there. This result
provides additional evidence that an intrin-
sic remediation management option for the
Hill site would be protective of public health
and the environment.
Layton Utah Site Intrinsic
Remediation Results
Site Background Data
The Elaine Jensen R.V. facility, Layton,
UT, was used as a recreational vehicle (RV)
sales and service facility during the study
and consisted of a service shop, an RV sales
building, and associated sales and display
Table 1. Contaminant center of mass velocities and degradation rates based on ground-water data
collected at the Hill AFB site from March 1992 to January 1994
Benzene
Distance Traveled (ft)
Contaminant Velocity (ft/d)
Contaminant Velocity (ft/yr)
Zero Order
Degradation Rate (g/d)
Zero Order
Degradation Rate (g/yr)
First Order
Degradation Rate (1/d)
First Order
Degradation Rate (1/yr)
Remaining Mass in 1/94 (g)
Time to Degrade
Remaining Mass -
Zero Order (d)
Time to Degrade 99.9% Mass -
First Order (d)
Travel Distance in
Degradation Time (ft)
25.7
0.05
18
=0.10
=36.9
0.41
=4.1
=0.20
Toluene
53.3
0.1
37.4
=0.099
=28.5
0.66
=6.7
=0.67
Ethyl-
benzene
42.9
0.08
30.1
0.063
23.4
0.53
8.4
0.67
P-
Xylene
55.6
0.11
39
0.058
22.3
0.8
13.8
1.5
Naphthalene
17.2
0.03
12.1
=0.026
=9.5
0.13
=266
Q
~O
TPH
106
0.2
74.2
0.009
3.285
28.8
768
156
-------
lots. The majority of the site is covered by
asphalt, and several underground utilities
are present. The ground surface slopes
gently toward the southwest and toward
U.S. Interstate Highway 15 which borders
the site directly to the west. The property
was in agricultural production until approxi-
mately 1958, at which time it was leased to
Sinclair Oil for retail gasoline sales from a
newly constructed gasoline station building
(the present service shop). The fuel stor-
age and dispensing system consisted at that
time of four USTs (2,000 gallons each) with
suction pumps located to the west of the
tanks. The property was subsequently uti-
lized for camper sales between 1968 and
1974, followed by use for retail gasoline
sales until 1984. The site was upgraded in
1974 with three additional 6,000-, 8,000-,
and 10,000-gallon USTs equipped with
pressurized dispenser systems. The tanks
were removed in 1984, and the property
was subsequently used forthe recreational
vehicle sales and service facility that occu-
pied the site during the study. The area sur-
rounding the site is characterized as mixed
residential and agricultural use.
Soils encountered at the site are prima-
rily sands in the southern portion of the site
and silts and clays in the northern half of
the site. Ground water is encountered at
approximately 8 to 10 ft below grade. The
direction of shallow ground-water flow is
predominantly southwest, although westerly
and northwesterly flows were observed at
the site during the study. Hydraulic conduc-
tivity values of the shallow aquifer at the
Layton site, determined by slug tests per-
formed during this study, indicated values
ranging from 0.78 to 3.1 ft/d, with an aver-
age of 1.5 ft/day.
A soil-gas survey was performed in 1990
at a depth of approximately 4 ft below
ground surface using a portable organic
vapor monitor (OVM) equipped with a pho-
toionization detector (PID). A total of 36 lo-
cations were surveyed for volatile organic
vapors. The highest vapor concentrations
(250 to 1,975 ppmv) were found in the two
tank excavations and in the dispenser is-
land area between the former tanks. An
outer area of lower vapor concentrations (0
to 250 ppmv) extended to the west and
southwest. Overall, an area of elevated
vapor concentrations of approximately
18,400 ft2 was identified. In January 1991,
soil borings were collected and three
ground-water monitoring wells were in-
stalled and screened across the water table
from 7 to 17 ft below ground surface.
In July and August 1992, CRT data were
collected at 27 locations to augment the
existing monitoring network. This detailed
site investigation effort provided analysis of
soil textural features from the surface to a
depth of 22 ft. This CRT information sug-
gested a finer grained material than was in-
dicated from soil boring data, and identified
a clay to silty clay lens 2 to 6 ft thick cover-
ing most of the site 12to 14 ft below ground
surface.
Seven sampling events were conducted
at the Layton site to describe the distribu-
tion and movement of contaminants and
electron acceptors taking place there be-
tween April 1992 and January 1994. These
data were used to determine steady-state
plume conditions and to estimate total mass
and mass center values for these various
analytes, and to model contaminant trans-
port and degradation occurring there over
time.
Intrinsic Remediation
Assessment
Determination of Steady-State
Plume Conditions
A westerly centerline transect was used
to make a determination regarding steady-
state plume conditions. Figure 4 shows
plume centerline data for combined BTEX
concentrations at the Layton site from July
1992 to February 1995. A pseudo-steady-
state was found to occur for both BTEX
components and TPH. Both BTEX and TPH
concentrations increased and decreased
along the plume centerline transect over
time, but no consistent pattern
downgradient of the source was evident.
Thiessen areas generated for each sam-
pling event using a fixed outer plume bound-
ary forthe Layton site are shown in Figure
5. Based on the Thiessen areas associated
with each monitoring point at each sampling
time, estimates of the total dissolved plume
mass and center of mass of BTEX, naph-
thalene, and TPH were made.
Despite a decline in dissolved contami-
nant mass observed at the Layton site be-
tween July 1992 and September 1993, fur-
ther sampling in January 1994 and Febru-
ary 1995 indicated that the dissolved mass
of the contaminants has remained essen-
tially constant over the two and one-half
year project period (Figure 6). The mobility
of the center of mass of all of BTEX, naph-
thalene and TPH was limited, with a maxi-
mum of only a 10-ft movement of TPH mass
over the two and one-half year field study.
The pseudo-steady-state contaminant
mass levels and limited center of mass
movement downgradient suggest that a
continuous source exists at the Layton site
which reflects a plume stabilized by con-
tinuing intrinsic attenuation mechanisms.
20 40 60 80 100 120
Distance Downgradient of CPT-19
Figure 4. Combined BTEX plume centerline con-
centration data collected at the Layton, UT, site
from July 1992 to February 1995.
CPT-17
CPT-09
s jCPT-Hx --X^ ^
A SCPT-15 - .
,.MW-03_^—-\ CPT-08 [ CPT-13
MLP.06 CPT-07 >/CP\.14
* VMLP-OSC -
CPT-19
0 MW-04
Figure 5. Outer plume boundary used for Layton
site plume total mass and mass center calcula-
tions. Thiessen areas forthe July 1993 sampling
event are shown.
12000
3 10000
200 400 600 800
Time (days)
Figure 6. Time course of total dissolved plume
mass estimates for BTEX and naphthalene con-
taminants at the Layton site during the study, July
1992 through February 1995.
-------
Estimation of Contaminant
Degradation Rate
Using the intrinsic remediation protocol
developed in this study, the estimation of
contaminant degradation rates with a
steady-state plume mass is carried out ei-
ther from an analysis of plume centerline
concentration data or through the calibra-
tion of a fate-and-transport model using field
data.
Plume centerline concentration
data
A steady-state, time-averaged concentra-
tion profile was used to estimate first order
degradation rates, the results of which are
summarized in Table 2. Naphthalene had
the slowest degradation rate, while benzene
was found to be the most degradable of the
compounds investigated. These degrada-
tion rates do not explicitly account for re-
ductions in concentration due to dispersion
and dilution (i.e., non-degradative) pro-
cesses taking place within the Layton aqui-
fer. The more preferred approach to esti-
mate contaminant degradation involves the
use of a ground-water fate-and-transport
model that is calibrated to field-generated
data.
Ground-Water Model Calibration
An analytical solution for the advection-
dispersion equation with degradation (as-
suming a continuous source with one-di-
mensional ground-water flow) was applied,
along with site-specific physical/chemical
input parameters, to model the fate and
transport of contaminants under actual field
conditions. Aquifer pore water velocities
were calculated based on measured values
of hydraulic gradient (0.01 ft/ft) and hydrau-
lic conductivity (1.5 ft/d) and estimated val-
ues of total aquifer porosity (0.38) using
Darcy's Law. An average pore water veloc-
ity of approximately 0.037 ft/d resulted.
Transverse dispersivity was assumed to be
1/20th of the longitudinal dispersivity while
vertical dispersivity was assumed to be
negligible (0.001 m). Based on current prac-
tice, a longitudinal dispersivity of 0.1 times
the plume length was used, resulting in lon-
gitudinal and transverse dispersivity values
of 14 and 0.7 ft, respectively. Contaminant
retardation factors were estimated from
compound-specific organic carbon normal-
ized soil/water partition coefficients (Kocs).
The soil organic carbon content, bulk den-
sity, and porosity values assumed for the
Layton aquiferwere 0.3%, 1.15 g/cm3, and
0.38, respectively.
The analytical solution used in this intrin-
sic remediation protocol assumes a con-
stant plane source perpendicular to the di-
rection of ground-water flow. The Layton site
vertical source dimension was set equal to
Table 2. Contaminant degradation rates estimated from time-averaged centerline concentrations
measured at the Layton site from July 1992 to February 1995
Compound
Benzene
Toluene
Ethylbenzene
p-Xylene
Naphthalene
First Order
Rate (1/d)
0.00087
0.00044
0.00026
0.00024
0.00014
95 Percent
Confidence Interval
(1/d)
0.00057
0.00010
0.00010
0.00013
0.00008
r2
0.8204
0.9742
0.9263
0.8737
0.8243
p Value
0.0129
0.0003
0.0021
0.0063
0.0123
10 ft, the approximate maximum thickness
of the contaminated ground-water column
observed in monitoring wells. The lateral
source dimension was based on an inspec-
tion of contaminant concentration profiles
perpendicularto ground-water flow nearthe
source area, approximately 100 ft.
From the site history, a range of simula-
tion times from 10 to 25 years was used for
model calibration. Contaminant degradation
rate and simulation time were varied to
evaluate the sensitivity of model output to
these parameters, and to determine those
combinations of parameters producing the
best model fit. From model calibration, a
source simulation lifetime of 25 years re-
sulted in the best overall model fit. Benzene
results are shown in Figure 7, while Table 3
summarizes model-calibrated degradation
rates for all contaminants of interest.
Estimation of Source Mass/
Lifetime
When a continuous source is identified
at a site, the estimation of source mass and
source lifetime is based on the total mass
of contaminant existing both above and
below the ground-water table. This mass is
generally estimated based on soil core con-
centration data collected from within the
source area. The estimated source area
mass at the Layton site appeared to be
greatly underpredicted by soil core results
based on the dissolved plume masses mea-
sured in February 1995. Source area mass
was then estimated based on typical re-
sidual hydrocarbon saturation within a
smear zone at and below the ground-water
table for a given soil texture. It was assumed
that approximately 10 percent of the sub-
surface porosity could be expected to be
occupied by product material in the clay to
silty clay soils below the Layton site, result-
ing in an estimated 241,000 Ib of TPH re-
maining in the source area.
Because of the large mass of contami-
nant remaining within the source area, with-
out source removal, site management re-
quirements could be in excess of 100 years
based on an ethylbenzene maximum con-
taminant level (MCL) of 700 |ig/L and a
source area ethylbenzene ground-water
concentration of» 1,900 ug/L measured in
February 1995. With source removal, the
time to reach the ethylbenzene MCL is re-
duced to slightly more than 10 years, with
benzene becoming the contaminant taking
the longest to assimilate, requiring approxi-
mately 20 years before reaching its MCL.
Source removal greatly reduces the length
of time for assimilation of the mass of all
contaminants at the Layton site, strongly
suggesting the need for source removal so
that the duration of site management be-
comes acceptable.
Predicting Long-Term Behavior
The time to reach the MCL following
source depletion can be predicted for each
component using the calibrated fate-and-
transport model described in this report by
the superposition of a continuous source
plume, with a source concentration equal
to the negative of the original source con-
centration, -Co, on top of the steady-state
plume concentration profile, at a point in
time corresponding to the time of source
removal. A benzene plume of greater than
the MCL of 5 ug/L is projected to persist for
18 years, while all other contaminants of
concern are projected to reach their MCL
values everywhere within the plume in less
than 7.5 years following source removal.
Decision Making Regarding
Intrinsic Remediation
Impacted Receptors
The projected long-term behavior of the
contaminant plumes existing at the Layton
site suggests that they have stabilized and
are not expected to migrate from the site
as long as existing site hydraulic conditions
persist. This limited extent of potential con-
taminant migration ensures that no
Potential Aquifer Assimilative
Capacity
The aquifer assimilative capacities re-
lated to dissolved oxygen (0.76 mg HC/L),
nitrate (19.6 mg HC/L), sulfate (30 mg HC/
L), iron (0.06 mg HC/L), and manganese
-------
Ground-Water Contaminant
Concentration (^ig/L)
7,000<
6,000-
5,000-
4,000-
3,000-
2,000-
1,000-
0 -
I
'1 Simulation Parameters:
' Degradation rate = 0.00086/c
•i R = 2.7, t = 25 yr
\
* » Field Measured Data
^ - Predicted Data
\
v_
*""^*«fc
0 20 40 60 80 100 120 140 160 180 200
Downgradient Distance from Source Area (
Figure 7. Results of benzene plume centerline calibration at the Layton site using data collected in
March 1993.
(0.03 mg HC/L) amounted to approximately
50.5 mg HC/L. With a maximum TPH con-
centration of nearly 100 mg/L and a maxi-
mum BTEX concentration of nearly 25 mg/
L observed during the study, this potential
assimilative capacity is marginal. This re-
sult provides additional evidence that some
form of source removal to reduce contami-
nant mass flux into the impacted ground
water is the preferred approach for the
Layton site. Without additional contaminant
source removal, intrinsic remediation ap-
pears marginally protective of public health
and the environment.
CONCLUSIONS AND
RECOMMENDATIONS
The research was conducted in four
phases: 1) site assessment and character-
ization; 2) process monitoring; 3) field data
reduction; and 4) three-dimensional analyti-
cal modeling. Based on the results of the
study, the following conclusions can be
made.
1. Conventional site assessment tech-
niques using a small number of large
diameter ground-water monitoring
wells and limited soil core and shal-
low soil gas survey data can be se-
verely limited in their ability to provide
a detailed understanding of subsur-
face soil conditions at a site. CRT
techniques can be used to improve
delineation of subsurface conditions
that may greatly impact local ground-
water flow and contaminant transport.
Coupled with placement of small di-
ameter ground-water monitoring
probes and field ATM measurements,
CRT measurements enabled the col-
lection of cost-effective data for accu-
rate plume delineation on nearly a
real-time basis in this study.
2. The field ATM method used in this
study appears more sensitive in situ-
ations with free product or high levels
of residual saturation, than at those
sites where weathered fuel contami-
nation exists. Field ATM measure-
ments can be used to effectively guide
initial ground-water quality investiga-
tions and to optimize ground-water
monitoring probe and monitoring well
placement for long-term site monitor-
ing. However, field screening data
should not be used as a substitute for
laboratory determined, long-term
ground-water monitoring data.
3. Centerline concentration profile and
dissolved contaminant mass analyses
indicated that a significant decline in
all contaminant mass levels took place
at the Hill AFB site over the course of
the study. Center of mass calculations
suggested that the plume was re-
sponding to a pulse source with con-
taminant attenuation. Dissolved plume
mass changes over time were used
to estimate zero and first order deg-
radation rates for BTEX, naphthalene,
and TPH contaminants.
The lifetime of the BTEX compounds
within the Hill AFB plume were short.
Approximately 270 days was pre-
dicted for 99.9 percent naphthalene
removal. More than 2 years was esti-
mated for the same removal efficiency
of TPH components under existing
site conditions. The decision to apply
intrinsic remediation at the Hill AFB
site is warranted based on: a) the con-
taminant degradation rates and plume
attenuation observed; b)the absence
of an impacted downgradient recep-
tor; and c) the potential aquifer assimi-
lative capacity that is more than 90
times greaterthan that required forthe
assimilation of the TPH and BTEX re-
maining in the dissolved plume. A
long-term monitoring scheme for both
compliance and process monitoring
purposes can be carried out using the
existing monitoring network. Annual
monitoring until 1997 should provide
adequate data to validate complete
plume assimilation to permit site clo-
sure.
Centerline concentration profile, dis-
solved contaminant mass and center
of mass analyses for the Layton site
indicated that the plume appeared to
be at steady-state through February
1995. Results suggest that the Layton
plume was acting as a continuous
source which was stabilized by on-go-
ing intrinsic attenuation mechanisms.
Changes in contaminant plume
centerline concentrations with dis-
tance from the source area and a con-
taminant fate-and-transport model de-
scribed in this study were also used
to estimate contaminant degradation
rates using measured March 1993
field data. Comparison of results from
the centerline concentration and
model calibration methods for degra-
dation rate estimates suggest that the
Table 3. Contaminant degradation rates estimated from model calibration of the Layton site data
collected March 1993
Compound
Calibrated Degradation
Rate(1/d)
Retardation
Factor, R
Benzene
Toluene
Ethylbenzene
p-Xylene
Naphthalene
0.00086
0.00062
0.00020
0.00037
0.00012
2.7
4.4
7.2
7.5
12.8
-------
simpler degradation rate approach
can be used to provide representa-
tive contaminant degradation rates
when a plume has reached steady-
state conditions.
6. The soil core data available from the
Layton site were limited, and residual
product volume estimates based on
gasoline residual saturation values re-
ported in the literature for silty clay
soils were used to estimate the mass
remaining in the Layton site source
area. From these residual mass esti-
mates, the lifetime of the total mass
of BTEX and naphthalene at the
Layton site, based on reaching their
MCL values in the plume, ranged from
approximately 30 years for toluene to
over 1 00 years for ethylbenzene. With
100 percent source removal, the re-
quired site management timeframe,
based on the ethylbenzene MCL, was
reduced to approximately 8 to 14
years. Under these site conditions,
however, benzene becomes the con-
taminant with the greatest duration of
concern, requiring approximately 18
to 22 years to reach its MCL value of
5
7. The decision to apply only an intrin-
sic remediation plume management
approach at the Layton site should be
made with caution. Despite apparent
plume stabilization and the absence
of an impacted downgradient recep-
tor, the low contaminant degradation
rates (approximately one order of
magnitude lower than at the Hill AFB
site), large residual masses of con-
taminant within the source area and
marginal potential aquifer assimilative
capacity observed at the site will re-
quire long-term monitoring and site
management (>30 years). Active
source removal and residual mass
remediation are warranted at the
Layton site to accelerate the rate of
contaminant removal from the site,
and shorten the length of time re-
quired for plume management.
Based on the findings in this study, the
following recommendations can be made:
1. The combined CPT/ATH procedures
used for initial plume delineation
proved to be rapid and cost-effective,
leading to a significantly improved un-
derstanding of subsurface conditions
at both of the field sites. This approach
should be refined and evaluated for
general application for site assess-
ment activities at UST sites.
2. The intrinsic remediation protocol de-
veloped in this study provides a logi-
cal, quantitative approach for evalu-
ating the presence and rates of con-
taminant assimilation within an aqui-
fer system. The protocol provides im-
provements over conventional plume
containment assessment methods.
Plume centerline concentration analy-
sis is used, along with plume mass
and center of mass determinations, to
incorporate the aerial aspects of
plume containment that have not nor-
mally been included in other field
evaluations of intrinsic remediation
processes. Its application should be
evaluated for routine use at UST sites.
3. The modeling approach used to de-
scribe intrinsic remediation processes
occurring at the two UST field sites al-
lowed the quantitative assessment of
contaminant migration and degrada-
tion using actual field data. The model
was easy to implement in a spread-
sheet environment and appeared to
provide a quantitative description of
contaminant plume profiles that were
observed at the two distinctly differ-
ent field sites evaluated in this study.
The model provided independent veri-
fication of plume steady-state condi-
tions, and allowed the rapid assess-
ment of the impact of various source
removal options on the duration of
contaminant plumes produced from
hydrocarbon releases at these sites.
Its application should be evaluated for
routine use at UST sites.
REFERENCES
Chow, V.T, D.R. Maidment, and L.W. Mays.
Applied Hydrology. New York: McGraw
Hill Book Co., 1988.
-------
R.Ryan Dupont, Darwin L. Sorensen, Marian Kemblowski, MarkBertleson, Dietrick
McGinnis, Idris Kamil and Yang Ma are with Utah Water Research Laboratory,
Utah State University, Logan, UT 84322-8200
Charlita Rosa/ ;'s the EPA Project Officer (see below).
The complete report, entitled "Monitoring and Assessment of In-Situ
Biocontainment of Petroleum Contaminated Ground-Water Plumes," (Order No.
PB98-145329; Cost: $47.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-605-6000
The EPA Project Officer can be contacted at:
Characterization Research Division
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Las Vegas, NV 89193-3478
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
Official Business
Penalty for Private Use $300
EPA/600/SR-98/020
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
10
------- |