Natural Attenuation of Fuel Hydrocarbons at Multiple
Air Force Base Demonstration Sites
Don H. Kampbell, USEPA/ORD/NRMRL/SPRD, Ada, OK 74821
Jerry E. Hansen U.S. Air Force/AFCEE, Brooks AFB, TX 78235
Bruce M. Henry, Parsons Engineering Science Inc., Denver, CO 80290
John R. Hicks, Parsons Engineering Science Inc., Denver, CO 80290
Introduction
A major initiative to evaluate monitored natural attenuation (MNA) of ground water
contaminated with fuel hydrocarbons began in June 1993 and continued through September
1999. The main emphasis was to evaluate natural degradation mechanisms to reduce dissolved
fuel components of benzene, toluene, ethylbenzene, and xylenes (BTEX). The results are
summarized in this report for the six-year study. During this time site characterization studies
were conducted at 28 spill sites on Air Force bases within all 10 USEPA regions. All sites were
evaluated for natural attenuation trends according to the procedures outlined in the technical
protocol document (Wiedemeier, et al. 1995). This research report has not been subjected to
U.S. EPA review and official endorsement of conclusions made should not be inferred.
Monitored natural attenuation refers to the use of naturally occurring attenuation
processes to decrease ground-water contaminants by physical, chemical, and biological
processes.
Almost 50 years ago Zobel (1950) demonstrated that aliphatic hydrocarbons are
susceptible to microbial degradation processes. Then, during the mid-1980s it was demonstrated
-------
that petroleum hydrocarbons could be biodegraded under both aerobic and anaerobic incubation
conditions. During the early 1990s studies were reported that biodegradation processes were a
beneficial effect of natural attenuation. In 1997, the University of Texas (Mace, 1997) reported
evidence by statistical analysis that natural attenuation and low aquifer permeability is effectively
remediating the majority of petroleum generated ground-water plumes at 605 within-state sites.
Monitored natural attenuation is less costly than conventional engineered treatment
techniques. It is often equally protective of the environment and human health, however, long-
term monitoring (LTM) and land use control are required to establish continuous protection.
The objective of the study was to document the effectiveness of natural attenuation and to
promote the use of MNA to cost-effectively achieve closure at the 28 Air Force Base facilities.
Currently most of the 50 states and all 10 USEPA regions will consider use of MNA as a cleanup
remedy for fuel-contaminated ground water. MNA application at the air bases in the United
States has resulted in considerable cost savings for soil site remediation. A survey (Ritz, 1996)
indicated that 53 of the 59 state agencies allow natural attenuation as a stand-alone remedial
option at fuel hydrocarbon contaminated sites.
Spill Site Treatability Study Protocol
Six primary tasks were performed for each study site. Completion of the treatability
study was a multigroup effort. An Air Force contractor, Parsons Engineering Science, Denver,
-------
performed a majority of the work as listed in the following six primary tasks.
• Each study begins with a site meeting and a MX A briefing between base officials and
concerned regulatory agencies.
• A site-specific work plan was prepared describing the study methods and tools.
« Site activities were performed to fill past data on the nature and extent of soil and ground-
water contamination and subsurface geochemical conditions.
• Ground-water monitoring wells and well points were sampled for contaminant
concentrations, and physical and geochemical biodegradation indicator parameters.
Biodegradation rates and geochemical trends were evaluated for the impact of natural
attenuation on contamination fate and transport. Future plume migration trends were
predicted by ground-water models such as Bioscreen (Newell et al., 1997) and Bioplume
III (Rifai et al., 1987).
• The implementability of MNA and in combination with engineered remediation
technologies were evaluated. A long term monitoring plan was developed.
The USEPA/NRMRL-Ada, Oklahoma was a co-investigator in most of the studies with
primary duties of reviewing work plans, collecting field data, sample analyses, and editing final
-------
technical reports. The U.S. Army Corps of Engineers provided at some base locations cone
penetrometer testing (CPT) to locate contaminant source areas and install small-diameter ground
water monitoring points.
Study Results
The site characterization studies indicated that natural attenuation was decreasing the
mass of dissolved fuel components of benzene, toluene, ethylbenzene, and xylenes (BTEX) at all
demonstration sites.
Plume Behavior Results
Historical ground-water data showing stabilization or decline of dissolved contaminant
concentrations is the most direct and convincing evidence for natural attenuation. At least two
sets of ground water quality data were available for 30 of the MNA test sites. Of these sites 87%
of the ground-water plumes were either stable or receding. The other 12 test sites without
historical data had fate and transport model predictions that most of the plumes were stabilized.
Only one site exhibited an expanding ground-water plume. This was at a recent fuel release
location so steady-state conditions may not have been attained.
BTEX Assimilative Capacity and Geochemistry Results
Microbial metabolism (Bouwer, 1992) can utilize fuel hydrocarbons as a primary electron
-------
donor in ground-water systems. Dissolved oxygen, nitrate, ferric iron, sulfate, and carbon
dioxide are common electron acceptors used. These compounds can easily be detected in ground
water, and their depletion, coupled with the accumulation of reaction components ferrous iron
and methane provided evidence regarding the preferred microbial pathways for contaminant
biodcgradation. Dissolved oxygen at elevated levels upgradient from lower levels in a plume
interior indicates that aerobic biodegradation of fuel hydrocarbons has occurred at the site.
Assimilative capacity can be computed by converting the relative mass of individual electron
acceptors available for bacteria utilization at a site into mass of BTEX that could be consumed
during the biodegradation reaction. The assimilative capacity identified the contaminant mass
that could theoretically be oxidized as one pore-volume of ground water traveled through the
plume core.
Based on multiple study sites, analytical data showed that sulfate reduction was the most
prominent process at 74% of the sites. One reason was that five of the study sites had very high
sulfate concentrations (>200 mg.'L). Sulfate reduction was an important attenuation mechanism
at 28% of the test sites when five sites with very high dissolved sulfate were excluded (Figure
1.1). Methanogenesis also was important as the predominant attenuation mechanism at 45% of
the sites. The contribution of the remaining attenuation processes decreased in the order of iron
reduction, aerobic oxidation, and denitrification. Our data indicated that as much as 97% of the
assimilative capacity of the ground-water systems was attributed to anaerobic biodegradation
processes.
Total BTEX assimilative capacity computed for the study sites ranged in mg/L from 23 to
-------
892 and averaged 64. Two-thirds of all sites had assimilative capacities that exceeded the
maximum observed dissolved BTEX concentration. Three-fourths of the ground-water plumes
that had increasing BTEX concentrations also had below-average assimilative capacities. These
phenomena indicated the importance of electron acceptor availability in limiting plume
advancement.
Field Biodegradation Rates
The fate and transport of contaminants dissolved in ground water was predicted by
estimating field-scale biodegradation rate constants. First-order kinetics was used to
approximate field-scale biodegradation. The apparent degradation rate was normalized for the
influence of dilution, sorption, and volatilization. Several methods were available for estimating
first-order biodegradation rates. They are as follows:
Use of a biologically recalcitrant compound (Kampbell et al., 1996).
* Use of a one-dimensional, steady-state analytical solution to the advection-dispersion
equation (Bear, 1979) (Buscheck and Alcantor, 1995).
* Use of ground water models where the biodegradation rate was adjusted during model
calibration to accurately simulate a measured plume.
Biodegradation Rate Results
-------
Biodegradation rates for dissolved BTEX plumes ranged more than three orders of
magnitude which was from 0.0002 to 0.08 day"1 (Figure 1.2). This was equivalent to half lives of
9.5 to 0.02 years, respectively. The geometric mean of the field biodegradation rates was 0.0019
day"1 or about a half-life of one year. A Laboratory study (Atlas, 1988) reported that rates of
microbial production increased by a factor of two for every 10°C increases in temperature. This
trend was not observed at our test sites where a site-averaged ground water temperature ranged
from 5.5°C to 26.9°C (Figure 1.3). There was a definite lack of correlation between field
biodegradation rates and ground water temperature. Our conclusion was that each site had a
microbial community adapted to efficiently degrade fuels at the site-specific temperature range.
Biodegradation rates did not correlate well with assimilative capacity (Figure 1.4). This
indicated that total assimilative capacity was not a reliable indicator of the biodegradation rate.
Biodegradation rates equal to or greater than 0.003 day"1 were present for contaminated aquifers
where ground-water velocities exceeded 300 feet per year (Figure 1.5). These phenomena
indicated that natural attenuation can be an effective remediation alternative at sites characterized
by rapid ground-water velocities. One reason for this is that electron-acceptor-enriched ground
water sweeps through the source area at a relatively rapid rate, which contributes to the reduction
of the source. For example, at one site where the estimated ground water velocity was 1600
ft/year the dissolved BTEX maximum upgradient concentration was 26,600 jj.g/L and 1650 feet
downgradient to the plume toe was 10 |xg/L BTEX. Biodegradation rates were compared to
plume length with the assumption that higher rates would result in shorter plumes (Figure 1.6).
Correlation did not support this assumption which indicated that contaminant concentrations, soil
type, ground water velocity, and other factors influence plume length to a greater degree.
-------
Modeling Results
Ground-water modeling was performed at all but three of the study sites. At most of the
sites the two-dimensional ground water flow and transport model Bioplume II (Rifai et al., 1987)
was used to predict natural attenuation trends and support development of long term monitoring
plans. The remaining sites were evaluated using the one-dimensional fate and transport models
Bioscreen (Newell et al., 1997), ONFD3 (Beljin, 1991) and analytical solutions by van
Genuchten and Alves (1982). For one site that had a very diverse hydrogeology, the three-
dimensional model using MODFLOW, (McDonald and Harbough, 1988), and MT3D, (S.S.
Papadoulous and Associates. Inc., 1996) was used. Simulation results indicated that the models
were most sensitive to field biodegradation rates and hydraulic conductivity. Simple analytical
models generally are adequate to predict the future migration and persistence of BTEX
contamination at a site because of their widespread acceptance.
Proposal Remedial Alternatives
Natural attenuation processes at one-third of the sites were effective enough to warrant
use of monitored natural attenuation in combination with institutional controls as the sole
remedial alternative. Measurable free-phase fuel product was present in about 45% of the sites.
The remedial alternative recommended for these sites consisted of monitored natural attenuation
and institutional controls for plume remediation and low-cost source removal technologies such
-------
as bioslurping and/or bioventing. A combination of monitored natural attenuation and
engineered source reduction was recommended. Remedial costs were favorable for short-term
engineered reduction or when surface waters or ground water receptors were impacted or could
be in the future.
The average estimated length of time required for natural attenuation to achieve State or
Federal ground-water quality standards for BTEX without engineered remediation was
approximately 30 years, based on conservative modeling assumptions. The engineered source
reduction addition using free-product recovery and/or removal of residual soil contaminants
reduced the average estimate for long-term monitoring to 20 years. More aggressive remediation
such as source area soil removal reduced the average estimated long-term monitoring to 14 years,
but greatly increases the remedial cost.
Long-Term Monitoring
A network of long-term monitoring and point-of-compliance wells was recommended at
each site to monitor natural attenuation trends and to protect downgradient receptors. The
recommended number of long-term monitoring and point-of-compliance wells ranged from five
to 22 and had an average of 11. Sampling frequency recommended for the sites ranged from
quarterly to biennial. However, annual sampling was recommended most frequently. The
recommended duration of long-term monitoring for all test sites averaged 22 years. The cost
estimated was $ 192K for a monitoring period of 30 years. A reduction in average time to 15
years for site cleanup using more aggressive technologies such as excavation or bioslurping was
-------
estimated to cost $816K.
Conclusions
Meaningful site characterization studies for natural attenuation of fuel hydrocarbons in
the subsurface involved combining multiple lines of evidence that includes geochemical and
documented loss of contaminant mass at the field scale. Also, delineation of the magnitude and
extent of the dissolved contaminant plume and source area is essential. Use of power-push
techniques such as Geoprobe to collect soil samples, investigate subsurface stratigraphy, and
install small-diameter ground-water monitoring points proved advantageous for rapid, low-cost
collection of adequate field data except where the water table was deeper than 25 feet below
ground surface. The use of a continuous-flow apparatus to protect extracted ground water from
reoxygenation was essential for accurate water quality measurements especially for dissolved
water and oxidation-reduction potential. Other analyses such as temperature, pH, alkalinity,
ferrous iron, and sulfide can be performed quickly and inexpensively in the field.
At sites with relatively simple hydrogeology, one dimensional models such as Bioscreen
are sufficient to determine persistence and migration potential of the dissolved contaminant
plume. These models will have limited adaptability where aquifer spatial heterogeneities,
weathering effects, or engineered source reductions are present. For these instances more
sophisticated two or three dimensional numerical ground-water models are recommended.
The newly ubiquitous occurrence of natural BTEX biodegradation has been widely
-------
documented in the literature. Laboratory aquifer soil microcosm studies are not necessary to
document site-specific biodegradation potential and the presence of a native fuel-hydrocarbon-
degrading microbial population. However, historical data information on a stabilized or receding
plume is needed at each site. Also, evidence of altered geochemical trends that support
occurrence of biological attenuation must be shown.
The results of the multiple base characterization studies determined that the geometric
mean of the dissolved BTEX biodegradation rates was 0.0019 day1 which was equivalent to a
contaminant half-life of one year. Natural attenuation rates were rapid enough to stabilize
hydrocarbon plume migration at even relatively high ground-water velocities. Temperature did
not influence biodegradation rates even as low as 5.5°C. Dominant biological attenuation
mechanisms were anaerobic biodegradation, particularly sulfate reduction and methanogenesis.
The average size of the ground-water contaminant plume for all site characterization studies was
seven acres.
This research project combined with other state and USEPA natural attenuation studies
have provided information for increasing regulatory acceptance of natural attenuation for
dissolved BTEX plumes. Most states are now receptive to the use of monitored natural
attenuation for dissolved BTEX plumes. Some have published guidance or regulations regarding
the conduct of natural attenuation studies.
A Case Study - Elmendorf AFB, Alaska, Site ST-41
-------
Site
The fuel spill at the site was released from four 1,000,000-gallon underground storage
tanks and associated piping. The plume of contamination was a weathered mixture of JP-4 jet
fuel and some aviation gasoline. The mixture was present both as free-phase and residual
product. The tanks were installed in the 1940s and have been documented to be leaking since the
1960s. Several thousand gallons of product had been released by 1984 and the tanks were
decommissioned in 1991. The aquifer was semiconfined and consisted of sand bounded above
and below by clay layers. The estimated ground-water velocity was 280 ft/yr and depth to the
ground-water surface was one to 35 feet.
Extent of Contamination
The volume of mobile LNAPL was estimated at 8770 gallons in a 100x100 feet area. A
free product recovery system began operation in 1993. The area of dissolved BTEX plume was
near 4.9 acres (Figure 1.7). The highest measured ground-water BTEX concentration was 43,300
Hg/L. The plume extended 700 feet from the source area.
Natural Attenuation
A biodegradation rate of 0.005 day"1 was estimated for the dissolved BTEX with aerobic
oxidation, denitrification, iron reduction, sulfur reduction, and methanogenesis being
contributing processes. The cool ground-water temperature averaging 6.7°C did not have a
-------
negative impact on biodegradation rates. The major degradation processes appeared to be
denitrification and sulfate reduction. The estimated total BTEX assimilative capacity of the sites
ground water was 22,300 fJtg/L.
Modeling and Historical Trends
Historical trends in contaminant distribution could not be determined because of
insufficient past ground-water data. A conservative Bioplume II model that assumed a constant
source over time, that is - no source weathering, indicated that natural attenuation would contain
the dissolved BTEX plume within 1000 feet of the source area (Figures 1.8 and 1.9). Natural
attenuation of dissolved BTEX and natural source weathering were predicted to result in
reduction of dissolved BTEX concentrations to below 1996 USEPA maximum contaminant
levels (MCLs) by 2024. A third simulation with continued operation of the existing free product
recovery system would achieve a 70% source reduction in five years (Figure 1.8). This would
decrease remediation time by eight years.
Recommendation for Site
Since the product-skimming recovery system was installed before our natural attenuation
site characterization study in 1994, it was recommended to continue source removal activities
until 1999 and perform LTM until 2009, then reevaluate remediation options beyond 2009.
References
-------
Atlas, R.M., 1988, Microbiology - Fundamentals and Applications: Macmillan Publishing
Company, New York.
Bear, J., 1979, Hydraulics of Groundwater. McGraw-Hill, New York, 569p.
Beljin, Milovan S., 1991, Solute: A Program Package of Analytical Models for Solute
Transport in Groundwater, Version 2.02. International Groundwater Modeling Center,
Holcomb Research Institute, Butler University, Indianapolis, Indiana.
Bouwer, E.J., 1992, Bioremediation of Subsurface Contaminants. In: Mitchell, R., (ed.).
Environmental Microbiology, Wiley-Liss, New York.
Buscheck, T.E. and Alcantar, C.M., 1995, Regression techniques and analytical solutions
to demonstrate intrinsic bioremediation. In: Proceedings of the 1995 Battelle
International Conference on In-Situ and On Site Bioreclamation. April 1995.
Kampbell, D.H., T.H. Wiedemeier, J.E. Hansen. 1996, Intrinsic bioremediation of fuel
contamination in ground water at a field site, Journal of Hazardous Materials 49,197-204.
Mace, R.E., et al., 1997, Extent, Mass, and Duration of Hydrocarbon Plumes from
Leaking Petroleum Storage Tank Sites in Texas: Texas Bureau of Economic Geology,
Geological Circular 97:1.
-------
McDonald, G. and Harbaugh, A.W., 1998, A modular three-dimensional finite-difference
groundwater flow model. U.S. Geological Survey Techniques of Water Resources
Investigations, Book 6, Chapter Al.
Newell, Charles J., McLeod, Kevin R., and James R. Gonzales, 1997, Bioscreen: Natural
Attenuation Decision Support System, Version 1.4. Prepared for the Air Force Center for
Environmental Excellence, Brooks AFB, Texas.
Rifai, H.S., P.B. Bedicnt, R.C. Borden, and J.F. Haasbeek, 1987, Bioplume II: Computer
Model of Two-Dimensional Transport Under the Influence of Oxygen Limited
Biodegradation. Dept. of Environ. Sci. and Engr,, Rice University, Texas.
Ritz, S. 1996. States speak out on natural attenuation. Soil and Groundwater Cleanup,
Jan.-Feb., 18-26.
S.S. Papadoulous and Associates, Inc., 1996, MT3D96: A Modular Thrcc-Dimensional
Transport Model for Simulation of Advection, Dispersion, and Chemical Reactions of
Contaminants in Ground-Water Systems. Bethesda, Maryland.
van Genuchten, M.T., and Alves, W.J., 1982, Analytical Solutions of the One-
Dimensional Convective-Dispersive Solute Transport Equation: U.S. Dept. of
Agriculture, Tech. Bull., No. 1661, 115p.
-------
Wiedemeier, T.H., J.T. Wilson, D.H. Kampbell, and J.E, Hansen, 1995, Technical
Protocol for Implementing Intrinsic Remediation with Long-Term Monitoring for Natural
Attenuation of Fuel Contamination Dissolved in Groundwater, U.S. Air Force Center for
Environmental Excellence, San Antonio, Texas.
Zobell, C.E., 1950, Assimilation of Hydrocarbons by Microorganisms. Advances in
Enzymology 10:443-486.
-------
Figure 1.1 Average Relative Contributions of BTEX Biodegradation Processes
in Site Groundwater
(Excluding 5 Sites With > 200 mg/L Sulfate Reduction Capacity)
Aerobic Oxidation
Denitrificadon
Iron (ID) Reduction
12%
Methanogenesis
45%
Sulfate Reduction
28%
Figure 1.2 Estimated BTEX Biodegradation Rates
25-1
l.E-04 l.E-03 l.E-02
Total BTEX Biodegradation Rate (day-1)
-------
Figure 1.3 Average Biodegradation Rates versus Groundwater Temperature
ca
w
&
cs
d6
c
_o
T3
C8
w
M
1)
T3
O
5
-------
Figure 1.5. First-Order Biodegradation Rate versus Groundwater Velocity
l.E-01
>>
a
-o
£ l.E-02
c
.2
c<3
u.
OJQ
u
¦o
c
3
L-
*o
O
i
C/5
1—
£
l.E-03
l.E-04
~
~
~
***
~~~~ ~ * ~
~ ~
. ~ .
*** ~
~
4
~ ~
1
~
1 1
10 100 1000
Groundwater Velocity (feet/year)
10000
Figure 1.6 First-Order Biodegradation Rate versus BTEX Plume Length
1.0000-
3 0.1000-
3
83
OC
c
0.0100-
a
•o
cs
u.
00
-------
Figure 1.7 Total BTEX in Groundwater, 1994, Elmendorf AFB, Alaska
Legend
Monitoring Well With Total
BTEX Concentration (pg/L)
CLP°
Inferred Line of Equal Dissolved
BTEX Concentration (pg/L)
Inferred Direction of
Groundwater Flow
ND Not Detected
Figure 1.8 Predicted BTEX Plume after 20 years (No Source Reduction)
Elmendorf AFB, Alaska
z
100 200
Legend
$ Monitoring Well
Line of Equal Simulated
£10° Dissolved BTEX
Concentration (|jg/L)
Inferred Direction of
Groundwater Row
-------
Figure 1.9 Predicted BTEX Plume after 21 Years (Continued Free Product Recovery)
Elmendorf AFB, Alaska
Legend
9 Monitoring Well
Line of Equal Simulated
100 Dissolved BTEX Concentration
^ ftjg/L)
Inferred Direction of
Groundwater Flow
-------
NRMRL-ADA-01241
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-02/026
2.
3.
4. TITLE AND SUBTITLE
5. REPORT DATE
Natural Attenuation of Fuel Hydrocarbons at Multiple Air Force Base
Demonstration Sites
6. PERFORMING ORGANIZATION CODE
7. AUTHOR (S)
'Kampbell, Don H.
'Hansen, Jerry E.
'Henry, Bruce M. and Hicks, John R
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
'U.S. EPA, Office of Research and Development, Natl. Risk Management
Research Lab., SPED, 919 Kerr Research Drive, Ada, OK 74820
JU.S. Air Force, AFCEE, Brooks AFB, TX 78235
'Parsons Engineering Science Inc., Denver, CO 80290
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
IAG RW57938631 US Air Force
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. EPA
NATIONAL RISK MANAGEMENT RESEARCH LABORATORY
SUBSURFACE PROTECTION AND REMEDIATION DIVISION
P.O. BOX 1198; ADA, OK 74820
13. TYPE OF REPORT AND PERIOD COVERED
Symposium Paper
14. SPONSORING AGENCY CODE
EPA/600/15
15. SUPPLEMENTARY NOTES
PROJECT OFFICER!
Don H. Kampbell 580-436-8564
16. ABSTRACT
A major initiative to evaluate monitored natural attenuation (MNA) of ground-water contaminated with fuel
hydrocarbons began in June 1993 and continued through September 1999. The main emphasis was to evaluate
natural degradation mechanisms to reduce dissolved fuel components of benzene, toluene, ethylbenzene, and
xylenes (BTEX) . The results are summarized in this report for the six-year study. During this time, site
characterization studies were conducted at 28 spill sites on Air Force bases within all 10 USEFA regions. All
sites were evaluated for natural attenuation trends according to the procedures outlined in the technical
protocol document (Wiedemeier, et al. 1995). This research report has not been subjected to U. S. EPA review
and official endorsement of conclusions made should not be inferred.
Monitored natural attenuation refers to the use of naturally occurring attenuation processes to decrease
ground-water contaminants by physical, chemical, and biological processes.
17.
KEY WORDS AND DOCUMENT ANALYSIS
A. DESCRIPTORS
B. IDENTIFIERS/OPEN ENDED TERMS
C. COSATI FIELD, GROUP
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS(THIS REPORT)
UNCLASSIFIED
21. NO. OF PAGES 21
20. SECURITY CLASS{THIS PAGE)
UNCLASSIFIED
22. PRICE
EPA FORM 2220-1 (REV.4-77)
PREVIOUS EDITION IS OBSOLETE
------- |