United States
Environmental Protection
Agency
Office of Solid Waste and
Emergency Response
(5102G)
EPA-542-R-00-003
August 2000
www.epa.gov/tio
cluin.org
c/EPA
Innovations in Site
Characterization:
Geophysical Investigation at
Hazardous Waste Sites
-------
-------
Notice
This material has been funded wholly by the United States Environmental Protection Agency
under Contract Number 68-W7-0051. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
Copies of this report are available free of charge from the National Service Center for
Environmental Publications (NSCEP), PO Box 42419, Cincinnati, Ohio 45242-2419; telephone
(800) 490-9198 or (513) 489-8190 (voice) or (513) 489-8695 (facsimile). Refer to document
EPA-542-R-00-003, Innovations in Site Characterization: Geophysical Investigation at
Hazardous Waste Sites. This document can also be obtained through EPA's Clean Up
Information (CLUIN) System on the World Wide Web at http://cluin.org. For assistance, call
(301) 589-8368.
Comments or questions about this report may be directed to the United States Environmental
Protection Agency, Technology Innovation Office (5102G), 401 M Street, SW, Washington,
D.C. 20460; telephone (703) 603-9910.
-------
-------
Foreword
This document contains eleven case studies designed to provide performance information for
innovative uses of geophysical technologies that support less costly screening for site
characterization. The case studies and the technologies that they highlight are included in-the
following table:
Site Name
Baker Woods Creosoting
Ciba-Geigy Hamblet and Hayes
Crystal Oil Refinery
Kansas Underground Storage Tank
Kelly Air Force Base, Zone 4
Marshalltown Former Manufactured Gas Plant
New Mexico State Highway and Transportation
(NMSHTD) Underground Storage Tank
Investigation
New Hampshire Plating.
Tinker Air Force Base
Trail Road Landfill
Wurtsmith Air Force Base
Geophysical Technology
Ground Penetrating Radar,
Electromagnetometry
Ground Penetrating Radar
Ground Penetrating Radar, Electrical
Resistivity
Electrical Conductivity
Vertical Seismic Profiling
Electrical Conductivity
Magnetometry, Electromagnetometry, Natural
Gamma Logging
Seismic Reflection Surveys, Ground
Penetrating Radar, Natural Gamma,
Electromagnetic Induction Logging
Seismic Reflection
Natural Gamma Ray, Magnetometry, Electrical •
Conductivity, Temperature, Density
Ground Penetrating Radar, Electromagnetic
Induction, Magnetometry
These case studies are part of a larger series of case studies that will include reports on new
technologies as well as novel applications of familiar tools or processes. They are prepared to
offer operational experience and to further disseminate information about ways to improve the
efficiency of data collection at hazardous waste sites. The ultimate goal is enhancing the cost-
effectiveness and providing flexible tools for characterizing hazardous waste sites.
in
-------
-------
Acknowledgments
This document was prepared for the United States Environmental Protection Agency's (EPA)
Technology Innovation Office. Special acknowledgment is given to all of the Principal
Investigators for their thoughtful suggestions and support in preparing these case studies.
IV
-------
-------
TABLE OF CONTENTS
Page
Notice ii
Foreword iii
Acknowledgments iv
INTRODUCTION 1
1.0 Methodology for Site Selection 1
2.0 Overview of Case Studies 3
3.0 Geophysical Technology Descriptions 3
Ground Penetrating Radar 3
Electromagnetometry 4
Seismic Reflection/Refraction 4
Electrical Conductivity/Resistivity 4
Natural Gamma 5
Magnetometry 5
4.0 Summary of Performance 5
5.0 Summary of Lessons Learned 6
Table ES-1: Geophysical Investigation Sites and Technologies 8
Table ES-2: Summary of Geophysical Investigations 9
Table ES-3: Performance of Geophysical Technologies 11
Bibliography 13
6.0 Case Studies
Baker Wood Creosoting Company 19
Ciba-Geigy Hamblet & Hayes Site 37
Crystal Refinery 51
Kansas Underground Storage Tank Site 65
Kelly Air Force Base 79
Marshalltown Former Manufactured Gas Plant Site 91
New Hampshire Plating Co 107
New Mexico Highway and Transportation Department 123
Tinker Air Force Base 143
Trail Road Landfill 159
Wurtzsmith Air Force Base 173
-------
-------
INTRODUCTION
Throughout the 1990s, the methods used to characterize hazardous waste sites have
changed considerably. Site managers have found that the collection of a limited number
of high-quality, high-cost, analytical data points that dominated site characterization in
the early part of the decade resulted in a lack of broader understanding of site conditions.
The earlier characterization practices also often required long time horizons for the
compilation of sufficient data to support remedial decisions. The high costs, long time
frames, and limited nature of the information associated with earlier practices have led to
the emergence of a number of innovative techniques designed to speed the data collection
process, increase the amount of information collected, and lower the overall cost of data
collection. The various agencies and departments of the Federal government with
responsibility for the characterization and cleanup of hazardous waste sites had all
adopted some form of expedited site characterization process by the end of the decade.
One set of technologies that has found a natural application in the context of expedited
site characterization has been geophysical characterization technologies.
Increasingly traditional geophysical technologies have found new and innovative uses at
hazardous waste sites. Geophysical technologies have been used for decades in other
industries, principally the petroleum and mining industries, for their ability to describe
geological structures deep within the earth's crust. This proven track record that has been
easily transferred to the characterization of hazardous waste sites. In fact, geophysical
technologies, such as ground penetrating radar, electromagnetometry, and magnetometry,
have been in wide use already at hazardous waste sites to locate buried drums and
structures that often constitute source areas. The use of geophysical technologies is
rapidly expanding to other applications in hazardous site characterization, including the
direct detection of aqueous and nonaqueous phase contamination. In several of the
investigations discussed in this volume, geophysical technologies were able to detect the
presence of either dense or light, nonaqueous phase liquids (D/LNAPLs).
One of the principal missions of the U.S. Environmental Protection Agency's Technology
Innovation Office (EPA/TIO) is to disseminate information on the cost and performance
of innovative technologies and approaches applicable to the characterization and
remediation of hazardous waste sites. The dissemination of this information can
stimulate the adoption and use of innovative technologies and approaches on an ever
widening scale. This report contains case studies of the innovative application of
different geophysical technologies and methods at 11 hazardous waste sites. The
technologies described in these case studies do not represent the entire range of
geophysical technologies, but do represent innovative applications of the better-known
technologies.
1.0 Methodology for Site Selection
In order to prepare a set of 11 case studies that explored the use of geophysical methods
at hazardous waste sites, EPA decided that case studies would be prepared for sites only
when:
• the investigation took place within the previous five years, to ensure that
knowledgeable information sources could still be readily identified;
• the investigation sought to identify site contamination;
-------
• site contamination problems were similar to those encountered under the Resource
Conservation and Recovery Act (RCRA), Superfund, and Underground Storage
Tank programs; and
• the technology was used in a full-scale application.
EPA initially set out to identify as many geophysical investigations that met the above
criteria as possible. Through contacts in EPA's Office of Research and Development
research laboratories, requests for information posted in relevant Internet discussion
forums, and requests to members of EPA's Groundwater Forum, a group of technical
experts distributed across the ten EPA Regional Offices, EPA identified more than 40
individuals with relevant experience. EPA contacted most of these individuals to collect
some basic information and compare it to the above criteria.
Over the same period, EPA performed a review of the technical literature published
within the previous three years to identify applications that might be used. The literature
review was confined to the following sources:
• Proceedings from the Symposium of Applications of Geophysics to Engineering
and Environmental Problems (SAGEEP);
• Groundwater;
• Soil and Groundwater Cleanup; and
• Environmental Science and Technology.
From the individuals contacted and the articles identified, EPA identified 27 applications
of geophysical technologies that were recent and relevant to the goals of the project. To
reduce this number toll, EPA applied the following criteria to each geophysical
investigation:
• Geophysical investigations demonstrated the technology's capability to directly
detect, or facilitate the detection of, contamination;
• Adequate documentation was available for the site and the performance of the
technology. Also, EPA could ensure reasonable access to the principal
investigator, or another person with knowledge of the application; and
• The site was a Superfund site.
Sixteen of the 27 applications were eliminated using this criteria. In some cases,
geophysical investigations were selected that did not meet all of these criteria. The
geophysical investigation at the Marshalltown Former Gas Manufacturing Plant (FMGP)
and the Crystal Refinery sites were conducted as demonstrations, not as a full-scale
applications. The investigation at the Trail Road Landfill did not evaluate a hazardous
waste site, but a municipal landfill. In each of these cases, the investigations represented
techniques for which full-scale applications could not be identified, yet whose results
provided relevant and useful information.
-------
2.0 Overview of Case Studies
The case studies describe a number of geophysical technologies and methods that were
used at sites with significantly different geological settings and a wide range of types of
subsurface contamination. Table ES-1 presents a summary of the geology, contaminants,
and geophysical methods used at the 11 case study sites.
The geological settings in which these technologies have been used ranged from simple
geological settings with relatively homogeneous stratigraphy, such as those found at
Wurtzsmith Air Force Base to the highly heterogeneous mix of sand and clay layers
found at the New Mexico Highway Safety and Transportation Department (NMHSTD)
site. Overall, simpler geological settings provided fewer challenges to the collection of
high quality geophysical data.
The types of contamination that were being characterized fell primarily into three broad
groups, chlorinated solvents, petroleum-related compounds, and polyaromatic
hydrocarbons. Inorganic contaminants were investigated at two of the sites. At seven of
the sites, contaminants were found in a nonaqueous phase liquid (NAPL), either as a
dense (DNAPL) or light (LNAPL) compound.
The types of geophysical technologies represented in the eleven case studies include:
Ground Penetrating Radar17 (GPR);
• Electromagnetometry (EM);
• Electrical Conductivity or Resistivity;
• Seismic Reflection or Refraction;
• Magnetometry; and
• Natural Gamma Logging.
The purposes for which geophysical investigations were undertaken varied from the more
traditional characterization of site stratigraphy to directly monitoring contaminants in the
media. Some geophysical technologies, such as electrical conductivity,
electromagnetometry, and to a lesser extent, ground penetrating radar, were able to
directly detect the presence of contaminants by measuring the change in soil
conductivities caused by the chemical compounds. Other technologies, such as seismic
reflection and refraction, magnetometry, and gamma logging, cannot directly detect the
presence of contaminants but are powerful tools in identifying subsurface lithologies that
provide preferential pathways for the migration of contaminants.
3.0 Geophysical Technology Descriptions
Ground Penetrating Radar
Ground penetrating radar (GPR) uses high-frequency radio waves to determine the
presence of subsurface objects and structures. A GPR system radiates short pulses of
II Ground penetrating radar is a technique that belongs in the larger set of electromagnetometry, but is
treated here as a separate technology due to its widespread use.
-------
high-frequency electromagnetic energy into the ground from a transmitting antenna. This
wave propagates into the ground at a velocity that is related to the electrical properties of
subsurface materials. When this wave encounters the interface of two materials having
different dielectric properties, a portion of the energy is reflected back to the surface,
where it is detected by a receiver antenna and transmitted to a control unit for processing
and display.
Electromagnetometry
The EM method is based on measuring the response of an electromagnetic field induced
into the earth. A small coil transmits low frequency signals, one to ten kilohertz. The
low frequency, very long wavelength electromagnetic fields produced by the transmitter,
induce current flow in electrically-conductive media in the earth. This induced current
flow produces secondary electromagnetic fields that radiate back to the surface. A
receiving coil detects the secondary field and measures its strength and phase relative to
the transmitted signal. The data are presented as the relative amplitude of the secondary
signal, in parts per million (ppm).
The depth of penetration of the transmitted field is a function of the frequency of
operation. Lower frequencies penetrate deeper, while higher frequencies are attenuated
more rapidly. This frequency dependent penetration depth provides the opportunity to
interpret multifrequency EM data to evaluate the depth and size of targets.
Seismic Reflection/Refraction
Seismic methods use an artificial seismic source to create direct compressional waves that
travel into the ground where they are reflected back to the surface when the waves
encounter boundaries between soil layers with different electrical properties. Some
waves are refracted along the interface of such layers by traveling along the contact
between geologic boundaries. The signals continue until they reach the surface.
Subsurface stratigraphy is mapped by measuring the travel time necessary for a wave to
pass through one layer to another, refract along the interface, and return to the geophones
at the surface.
Reflection energy is received by the geophone and recorded as a trace. Each trace
represents a station and each subsurface reflector or event should be visually identifiable
on the trace, and connected to other traces within the survey. The ability to visually
connect traces with an identifiable reflector, such as the bedrock surface, across ,many
such traces can be an indicator of the seismic survey accuracy within localized areas.
Electrical Conductivity/Resistivity
Electrical conductivity is an inherent property of a material to conduct an electrical
current and the electrical properties of soils can be measured using conductivity probes.
Current is injected into the earth through a pair of electrodes, and the potential difference
is measured between the pair of potential electrodes. The current and potential electrodes
are usually arranged in a linear array. Common arrays include the dipole-dipole array,
pole-pole array, Schlumberger array, and the Wenner array.
Variations in shallow soil conductivity (resistivity is the inverse of conductivity) are
caused by changes in soil moisture content, conductivity of groundwater, and properties
-------
that can be related to lithology. Soil conductivity is a function of grain size, with finer
grains producing higher values and coarser grains resulting in lower values.
Natural Gamma
Natural gamma logging is the continuous physical measurement of the release of natural
gamma radiation from the soil and rocks surrounding the length of the borehole. Natural
gamma measurement begins by lowering the detector to the bottom of a hole, allowing it
to equilibrate to the different subsurface temperature, then reeling the detector up the hole
at a steady rate of between five and 10 feet per minute. The gamma log measures the
total gamma radiation emitted by a particular stratum in counts per second as the detector
is raised in the well column. Interpretation of the gamma log depends as much on the
absolute value of the gamma counts as it does on the rate of change in gamma counts as
the detector passes from one material to the next. Statistical variations in gamma
emissions, significant at low counting rates, are smoothed out by integration over a short
time interval. If the hole is logged too quickly, however, the smoothing effect leads to
erroneous results by shifting the peaks in the direction of logging.
Magnetometry
Magnetometers measure variations in the magnetic field of the earth, and local
disruptions to the earth's field, including the presence of naturally occurring ore bodies
and man-made iron or steel objects. Whether on the surface or in the subsurface, iron
objects or minerals cause local distortions or anomalies in this field. When used together,
the use of both total field magnetic and magnetic susceptibility logs allows for the
detection of ferromagnetic minerals. A magnetometer's response is proportional to the
mass of iron in the target. The effectiveness of magnetometry results can be reduced or
inhibited by interference (noise) from time-variable changes in the earth' s field and
spatial variations caused by magnetic minerals in the soil, or iron debris, pipes, fences.
buildings, and vehicles.
4.0 Summary of Performance
In each investigation, the geophysical technologies performed as expected and the results
were used by site managers to support a variety of site decisions. Those decisions ranged
from remedy selection and design to optimizing the performance of existing remedies.
Table ES-2 provides a summary of the results obtained in the 11 geophysical
investigations. In five, the geophysical technology was able to directly detect
contaminants, greatly aiding the delineation of contamination at those sites. In the
remaining investigations, the results described critical geological structures that
influenced the migration of contaminants. This information aided site managers in
identifying appropriate sampling locations for better delineation of contamination.
Table ES-3 provides a summary of the amount of data that was collected during the
geophysical investigation, the approximate cost2', and the difficulties encountered. The
information in this Table helps to underscore one of the greatest advantages that
geophysical technologies offer: their cost-effectiveness. For all but two of the
investigations, the approximate cost was less than $10 thousand. In one of the other two
cases, the higher costs represented an investigation with a substantially larger scope,
21 Investigation costs were estimated based on information supplied by the geophysical investigator.
5
-------
while the cost for the remaining investigation could not be separated from the overall cost
of the soil sampling investigation.
Although no difficulties were reported for five of the investigations, those reported in the
other investigations reflect the limitations of some of the technologies used. Ground
penetrating radar surveys found that dense clays and silts limited the depth to which
measurements could be taken. Standing water and cultural noises, such as airports and
railroads, posed difficulties for the collection of seismic data, in some cases. The
electrical conductivity probe used at one site often broke when large cobbles or boulders
were encountered.
Several of the technologies discussed in this report have been used successfully in
investigations of nonaqueous phase liquids (NAPLs). Electromagnetics, including
Ground Penetrating Radar (GPR), were used at two creosote site and were able to detect
discernable differences in the electrical conductivity of NAPL-saturated soils. At another
site, a seismic reflection survey was able to identify deep paleochannels in the bedrock
face in which DNAPLs had pooled and through which DNAPLs were migrating.
5.0 Summary of Lessons Learned
Overall, these technologies performed well and provided valuable information. Some of
the lessons learned during their application were related to improving data collection,
while others were related to the value of the technology, itself. Some of the principal
lessons learned included:
• Electromagnetometry can be useful in detecting certain types of contamination in
both soils and groundwater. The presence of organic compounds in soil and
inorganic compounds in groundwater change the electrical conductivity of the
medium from background levels.
• The performance of ground penetrating radar surveys can be improved with the
use of different antenna configurations. The typical co-pole configuration, while
sensitive to subsurface geometries that are flat and oriented parallel to the ground
surface, is less sensitive to angular features in the subsurface. At the Baker
Woods site, buried pits and vaults were more clearly identified with a cross-pole
configuration, while they were less visible in GPR data collected with a co-pole
configuration.
• Electrical conductivity probes are able to detect the presence of NAPLs directly.
NAPLs change the electrical properties of the soils in which they reside, and the
probes were able to measure these changes. When they occur in soils that should
not otherwise be providing such readings, it may be an indication of the presence
ofaNAPL.
• Geophysical technologies can be used in concert with each other to produce
complementary results that increase both perspective and confidence. At the Trail
Road site, several different borehole methods were used to gain a comprehensive
understanding of the stratigraphy and to identify zones of groundwater
contamination. Some of the methods were sensitive to dense, impermeable layers
while others were sensitive to coarser, sandy materials. Electrical conductivity
and water temperature were used in concert to detect the presence of contaminants
in the groundwater.
-------
Mature petroleum-related LNAPL plumes can be detected as electrically-
conductive soils located at the capillary fringe. Natural biodegradation of the
compounds works to mobilize inorganic materials in the soils whose presence
increases the soil conductivity. This approach was successfully pursued at two of
the sites: Crystal Refinery and Wurtzsmith Air Force Base.
Seismic data collection can be sensitive to cultural sources of noise which can
interfere with reflected acoustical energy from bedrock structures. Several
statistical procedures were employed during the Tinker Air Force Base
investigation to improve the quality of data collected.
-------
Table ES-1: Geophysical Investigation Sites and Technologies
Site Name and Location
Baker Wood Creosoting Company
Marion, OH
Ciba-Geigy Hamblet & Hayes Site
Lewiston, ME
Crystal Refinery
Carson City, Ml
Kelly Air Force Base
San Antonio, TX
Kansas UST
Salina, KS
Marshalltown FMGP
Marshalltown, IA
New Hampshire Plating Company
Merrimack, NH
NMHSTD UST
Deming, NM
Tinker Air Force Base
Tinker, OK
Trail Road Landfill
Nepean, Ontario, Canada
Wurtsmith Air Force Base
Oscoda, Ml
Geology
Silty loam over clay
Sandy fill over clay
Sandy loam, sand, clay
over limestone
Sand, gravel, clay mix
over limestone
Clay over sand
Glacial till over
limestone
Silty clay over granite
Sandy clay with clay
layers over shale
Mix of clay, sand layers
over sandstone
Sand, gravel over clay
and limestone
Sand, gravel over clay
and sandstone
Contaminants
Polyaromatic hydrocarbons
Chlorinated solvents;
Petroleum hydrocarbons
Petroleum hydrocarbons
Chlorinated solvents
Petroleum hydrocarbons
Polyaromatic hydrocarbons
Chromium
Chlorinated solvents
Chromium
Chlorinated solvents
Dissolved inorganic and
orga'nic compounds
Petroleum hydrocarbons
Geophysical Method Used
Ground Penetrating Radar
Electomagnetometry
Ground Penetrating Radar
Ground Penetrating Radar
Electrical Resistivity
Seismic Reflection
Electrical Conductivity
Electrical Conductivity
Seismic Reflection
Ground Penetrating Radar
Natural Gamma
Electomagnetometry
Magnetometry (R)
Electomagnetometry (R)
Natural Gamma
Seismic Reflection
Electomagnetometry
Natural Gamma
Magnetometry
Electrical Conductivity
Temperature
Ground Penetrating Radar
Electomagnetometry (R)
Magnetometry (R)
Note: (R) indicates that the method was used in a reconnaissance survey for buried materials that might interfere with primary
technology.
-------
Table ES-2: Summary of Geophysical Investigations
* ' ' v ,, * ,'"•',/" „ , >.
, '' \v **: \ -.*,•'<*
" :'> u« >j" ' * ,
ivv , ' , -< ^ ^ v,
.-Site Nartie and Location
Baker Wood Creosoting Company
Marion, OH
Ciba-Geigy Hamblet & Hayes Site
Lewiston, ME
Crystal Refinery
Carson City, Ml
Kelly Air Force Base
San Antonio, TX
Kansas LIST
Salina, KS
Marshalltown FMGP
Marshalltown, IA
New Hampshire Plating Company
Merrimack, NH
NMHSTD UST
Deming, NM
- ' rf t * v ' « "* '
, ,1 ' ,? ^ f* '< ' J "
tg'l \t f ~ t Y -l , \
Geophysicjal Method
\i "( ,v Used *: - ^
Ground Penetrating Radar
Electomagnetometry
Ground Penetrating Radar
Ground Penetrating Radar
Electrical Resistivity
Seismic Reflection
Electrical Conductivity
Electrical Conductivity
Seismic Reflection
Ground Penetrating Radar
Natural gamma
Electomagnetometry
Magnetometry (R)
Electromagnetometry (R)
Natural gamma
* " 7 ',, . ^ v \,N?"
V ^ ^ v •'- *
> ^ •&• « 4 "•
'- / Purpose ^ *
Delineate source areas
and soil contamination
Characterize stratigraphy
Monitor groundwater
contamination
Map bedrock topology
Characterize stratigraphy
Characterize stratigraphy
Characterize stratigraphy
Monitor groundwater
contamination
Characterize stratigraphy
for sampling point location
4 .
T •< s
„ _' $
I <• ^"^ \
' v/^x" . •;
" ' *. Results', >
~, 5 \ i
GPR identified buried structures
that later investigation found to be
contaminated. EM delineated
near-surface soil contamination
Found topographic low where
later sampling found pooled
DNAPL
Identified LNAPL mass located at
water table
Identified channels in bedrock
where later sampling found
pooled DNAPL
Found saddle-like formation in
confining layer that acted as
preferential migration pathway for
LNAPL
Clearly identified lithology,
including layers not yet identified;
probe able to directly detect
DNAPLs
Delineated stratigraphy; identified
zones of groundwater
contamination
Gamma logs identified clay layers
that influenced vapor migration in
vadose zone; logs were used to
position in situ soil gas samplers
-------
Table ES-2: Summary of Geophysical Investigations
Site Name and Location
Tinker Air Force Base
Tinker, OK
Trail Road Landfill
Nepean, Ontario, Canada
Wurtsmith Air Force Base
Oscoda, Ml
Geophysical Method
Used
Seismic Reflection
Electromagnetometry
Natural Gamma
Magnetometry
Electrical Conductivity
Density
Temperature
Ground Penetrating Radar
Electromagnetometry (R)
Magnetometry (R)
Purpose
Characterize stratigraphy
for new well installation
Monitor groundwater
contamination
Monitor groundwater
contamination
Results
Characterized stratigraphy;
identified permeable layers
Developed continuous lithologic
logs; conductivity and
temperature logs identified zones
of groundwater contamination
Identified unknown LNAPL plume
Note: (R) indicates that the method was used in a reconnaissance survey for buried materials that might interfere with primary
technology.
10
-------
Table ES-3: Performance of Geophysical Technologies
," " ", '''"• ' \ T> ' ' Ilf '
^* *., ' , "H f s", !* '
, < • , '"X"v" iU *' "> ," ,,. '
' ' Sjte^amf ajicl tdfcalion! r ;
1 \ «x ? . ' »!»•,' f
Baker Wood Creosoting Company
Marion, OH
Ciba-Geigy Hamblet & Hayes Site
Lewiston, ME
Crystal Refinery
Carson City, Ml
Kelly Air Force Base
San Antonio, TX
Kansas LIST
Salina, KS
Marshalltown FMGP
Marshalltown, IA
New Hampshire Plating Company
Merrimack, NH
NMHSTD UST
Deming, NM
.•'^v.^%;
* v t v >¥>••>*
' ^Geophysical, r
u'; Method Msed *'.
Ground Penetrating
Radar
Electromagnetometry
Ground Penetrating
Radar
Ground Penetrating
Radar
Electrical Resistivity
Seismic Reflection
Electrical Conductivity
Electrical Conductivity
Seismic Reflection
Ground Penetrating
Radar
Natural gamma
Electromagnetometry
Magnetometry (R)
Electromagnetometry (R)
Natural Gamma
^ % i; ,, •;**' * * /«!,/"* w ,. * ' . . ** <
' ^\^ ''' y^ft:.x'vr ^\ •' j--
,V»V.' ';•,;"-% Bineflts/Diffl^ultles • \ • !"
Benefits: 100 traverses over 0.7 acres
Difficulties: GPR depth was limited by shallow dense clay soils;
EM depth was limited by nearby structures
Benefits: 85 traverses over 0.1 acres in two days for $4 thousand
Difficulties: Dense clay limited depth of penetration; swampy
areas limited access
Benefits: 2 traverses over 2.3 acres for $5.8 thousand
Difficulties: No significant problems reported
Benefits: 317 station measurements for $15.9 thousand
Difficulties: Railroad noise interfered with data collection
Benefits: 10 logs over 3.7 acres for $3.6 thousand
Difficulties: No significant problems reported
Benefits: 27 logs in 5 days for $7.9 thousand
Difficulties: Probes broke when encountered cobbles and
boulders; weathered bedrock was not distinguishable in logs
Benefits: 33 station measurements/7 logs/5,800 ft. of profiles
over 13.1 acres for $43.1 thousand
Difficulties: Dense clay and sediments limited depth of
penetration for GPR and seismic signals
Benefits: 33 profiles over 15 acres for less than $70 thousand3'
Difficulties: No significant problems reported
3/ Price includes cost of soil gas survey
11
-------
Table ES-3: Performance of Geophysical Technologies
Site Name and Location
Tinker Air Force Base
Tinker, OK
Trail Road Landfill
Nepean, Ontario, Canada
Wurtsmith Air Force Base
Oscoda, Ml
Geophysical
Method Used
Seismic Reflection
Electromagnetometry
Natural Gamma
Magnetometry
Electrical Conductivity
Density
Temperature
Ground Penetrating
Radar
Electromagnetometry (R)
Magnetometry (R)
Benefits/Difficulties
Benefits: 17,510 feet of profiles over 100 acres
Difficulties: Muddy surface conditions interfered with data
collection
Benefits: 5 measurements in 8 logs for $4.2 thousand
Difficulties: No significant problems reported
Benefits: 2,700 feet of profiles for $7.7 thousand
Difficulties: No significant problems reported
Note: (R) indicates that the method was used in a reconnaissance survey for buried materials that might interfere with primary
technology.
12
-------
TECHNICAL REFERENCE BIBLIOGRAPHY
Abbey, D.G., Mwenifumbo, C.J., and Killeen, P.O. The Application of Borehole
Geophysics to the Delineation ofLeachate Contamination at the Trail Road Landfill Site:
Nepean, Ontario. Symposium on the Application of Geophysics to Environmental &
Engineering Problems (SAGEEP) (1997). Pg. 163-171.
Adams, Mary-Linda, Herridge, Brian, Sinclair, Nate, Fox, Tad, and Perry, Chris. 3-D
Seismic Reflection Surveys for Direct Detection ofDNAPL. 1st International Conference
on Remediation of Chlorinated & Recalcitrant Compounds (1998).
Ahrens, T. J., ed. A Handbook of Physical Constants. Am. Geophys. Union. (1995)
Allen, P. A. Earth Surface Processes. Blackwell, Oxford. (1997)' -
Annan, A.P., et al. Geophysical Monitoring ofDNAPL Migration in a Sandy Aquifer.
Society of Exploration Geophysicists Technical Program, 62nd Annual International
Meeting and Exposition, 25-29 October 1992.
ASTM (American Society for Testing & Materials). Special Procedures for Testing Soil
and Rock for Engineering Purposes. ASTM. (1970)
ASTM. Sampling, Standards and Homogeneity. ASTM, Spec. Tech. Publ. 540. (1973)
Bachrach, R., Nur, A., and Rickett, J. Seismic Detection of Viscous Contaminant Using
Shallow Seismic Reflection. Symposium on the Application of Geophysics to
Environmental & Engineering Problems (SAGEEP) (1998). Pg. 685-694.
Bainer, Robert W., Milligan, Paul A., Rector, James W., Carr, Bradley, and Doll,
William. Preliminary Report on the Use of Geophysical Vertical Seismic Profiling
Techniques for Site Characterization of Subsurface Structures at the Oak Ridge National
Laboratory, Oak Ridge, Tennessee. Lawrence Livermore National Laboratory (1998).
Balanis, C.A. Antenna Theory: Analysis and Design. Wiley Press, NY. (1996)
Bauman, P.D., Lockhard, M., Sharma, A., and Kellett, R. Case Studies of2D Resistivity
Surveying for Soils, Waste Management, Geotechnical, and Ground-water Contaminant
Investigations. Symposium on the Application of Geophysics to Environmental &
Engineering Problems (SAGEEP) (1997). Pg. 261-269.
Borns D.J., Newman G., Stolarczyk L.S., and Mondt W. Cross Borehole
Electromagnetic Imaging of Chemical and Mixed Waste Landfills. Symposium on the
Application of Geophysics to Environmental & Engineering Problems (SAGEEP) (1993).
Pg. 91-105.
Berryman, J. G. Mixture Theories for Rock Properties, in Rock Physics & Phase
Relations. American Geophysists Union. (1995)
Brewster, M. L., Annan, A. P., Greenhouse, J. P., Kueper, B. H., Olhoeft, G. R., Redman,
J. D., and Sander, K. A. Observed Migration of a Controlled DNAPL Release by
Geophysical Methods. Ground Water, volume 33 #5 pg. 977-987, (1995).
Buderi, R. The Invention That Changed the World. Simon & Schuster. (1996)
Butler, D.K., et al. Comprehensive Geophysics Investigation of an Existing Dam
Foundation - Engineering Geophysics Research and Development. The Leading Edge, v.
9, no. 9. (1990)
Carr, M. H. Water on Mars. Oxford University Press, NY. (1996)
13
-------
Chapelle, F. H. Ground-water Microbiology and Geochemistry. John Wiley, NY. (1993)
Clement, William P., Cardimona, Steve, and Kadinsky-Cade, Katharine. Geophysical
and Geotechnical Site Characterization Data at the Groundwater Remediation Field
Laboratory, Dover Air Force Base, Dover, Delaware. Symposium on the Application of
Geophysics to Environmental & Engineering Problems (SAGEEP) (1997). Pg. 665-673.
Conyers, L.B. and Goodman, D. Ground-penetrating Radar: an Introduction for
Archaeologists: Altimira. (1997)
Daily, William, Ramirez, Abelardo, and Johnson, Richard. Electrical Impedance
Tomography of a Perchloroethelyne Release. Journal of Environmental & Engineering
Geophysics. January 1998 Volume 2 Issue 3 pg. 189-201.
Delaney, Allan J., Strasser, Jeffrey C., Lawson, Daniel E., Arcone, Steven A., and
Evenson, Edward B. Geophysical Investigations at a Buried Disposal Site on Fort
Richardson, Alaska. U.S. Army Corps of Engineers® Cold Regions Research and
Engineering Laboratory Report 97-4 (1997).
Dhont, Jeffrey, and Singh, Udai. Addressing Non-Aqueous Phase Liquids and Dissolved
Plumes at Two Adjacent Superfund Sites with Commingled Groundwater Contamination.
Non-Aqueous Phase Liquids (NAPLS) in Subsurface Environment: Assessment and
Remediation Proceedings (1996).
Dubey, A.C. and Cindrich, I., eds. Detection Technologies for Mines andMinelike
Targets. SPJJB, Bellingham, WA. (1995)
Duke, S.K. Calibration of Ground Penetrating Radar and Calculation of Attenuation and
Dielectric Permittivity Versus Depth. MSc Thesis, Dept. of Geophysics, Colorado School
of Mines, Golden, CO. (1990)
Feddes, R. A., ed. Space and Time Scale Variability and Interdependences in
Hydrological Processes. Cambridge Univ. Press. (1995)
Gardiner, C. W. Handbook of stochastic methods for physics, chemistry and the natural
sciences. 2nd corrected ed. Springer-Verlag, Berlin. (1990)
Gelb, S., and Wonder, J.D. ESC Demonstration: D-Area Oil Seepage Basin - Savannah
River Site: A Case Study. Symposium on the Application of Geophysics to
Environmental & Engineering Problems (SAGEEP) (1998). Pg. 551-560.
Greenhouse, J.P., et al. Geophysics and Solvents: The Borden Experiment. The Leading
Edge, April, v. 12, no. 4. (1993)
Gueguen, Y. and Palciauskas, V. Introduction to the Physics of Rocks. Princeton Univ
Press, Princeton, NJ. (1994)
Hess, K. M. and Wolf, S. H. Techniques to Determine Spatial Variations in Hydraulic
Conductivity of Sand and Gravel. U.S. Environmental Protection Agency. EPA/600/2-
91/006. (1991)
Hillel, D. and Elrick, D. E. Scaling in Soil Physics. Principles and Applications. Soil
Science Society of America, Madison, Wisconsin, SSSA Spec. Publ. 25. (1990)
Hubert, A. and Schafer, R. Magnetic Domains. Springer, Berlin. (1998)
Huffman JJI, A.C. Characterization of Three-dimensional Geological Heterogeneities
Using Ground Penetrating Radar. MSc Thesis, Dept. Geophysics, Colorado School of
Mines, Golden, CO. (1992)
14
-------
Hunt, C. P., et al. Magnetic Properties of Rocks and Minerals, in Rock Physics and Phase
Relations. American Geophysists Union, Washington DC. (1995)
Kalma, J. D. and Sivapalan, M., eds. Scale Issues in Hydrological Modeling. Wiley, NY.
(1995)
Kaufinann, R.D., Yuhr, L.B., and Wonder, J.D. ESC Phase 1: Locating and Mapping
Drilling Mud Pits at the Central Nevada Test Site. Symposium on the Application of
Geophysics to Environmental & Engineering Problems (SAGEEP) (1998). Pg. 541-550.
Keithley, J. F. The Story of Electrical and Magnetic Measurements: From Early Days to
the Beginnings of the 20th Century, 50 B.C. to about 1920 A.D. IEEE Press, NY. (1998)
Kirkendall, B. A Rapid Limited 3-dimensional Near-field Modeling Program for Ground
Penetrating Radar. Dept of Geophysics, Colorado School of Mines, Golden, CO, MSc
thesis. (1998)
Kraus, J. D. Electromagnetics, 4th Ed. McGraw-Hill, NY. (1991)
Larson, T.H., Krapac, I.G., Dey, W.S., and Suchomski, C.J. Electromagnetic Terrain
Conductivity Surveys Used to Screen Swine Confinement Facilities for Groundwater
Contamination. Symposium on the Application of Geophysics to Environmental &
Engineering Problems (SAGEEP) (1997). Pg. 271-279.
Lemke, Seth R., and Young, Charles T. Leachate Plume Investigation Using Mise-A-La-
Masse Resistivity. Symposium on the Application of Geophysics to Environmental &
Engineering Problems (SAGEEP) (1998). Pg. 839-847.
Lindsley, D. H., ed. Oxide Minerals. Their Petrologic and Magnetic Significance.
Mineral. Soc. Am., Washington, DC. (1991)
Liu, L., and Quan, Y. GPR Attenuation Tomography for Detecting DNAPLS.
Symposium on the Application of Geophysics to Environmental & Engineering Problems
(SAGEEP) (1997). Pg. 241-259.
Liu, Zhi-Ming, and Doll, William E. Seismic Reflection Processing for Characterization
of a Hazardous Waste Site. Symposium on the Application of Geophysics to
Environmental & Engineering Problems (SAGEEP) (1997). Pg. 291-299.
Lucius, I.E., Olhoeft, G.R., and Duke, S.K., eds. Third International Conference on
Ground Penetrating Radar, Abstracts of the Technical Meeting. U.S. Geological Survey
Open-File Report 90-414. (1990)
Lucius, I.E., et al. Properties and Hazards of 108 Selected Substances - 1992 Edition.
U.S. Geological Survey Open-File Report 92-527. (1992)
Lucius, I.E., and Olhoeft, G.R. Statistical Analysis of Field-scale Subsurface
Heterogeneity at the Princeton, Minn., Management Systems Evaluation Area Using
Ground Penetrating Radar, in Agricultural Research to Protect Water Quality
Conference Proceedings, Feb. 21-24, 1993
Lucius, J.E., and Olhoeft, G.R. Geophysical Investigations of Heterogeneity and Scale at
Princeton, Minnesota, Management Systems Evaluation Area. in. Proceedings ofU. S.
Geological Survey Toxic Substances Hydrology Program Review. USGS WRI Report 94-
4015.(1996)
Martin, D. H. Magnetism in Solids. Iliffe Books, London .(1967)
Mitchell, J. K. Fundamentals of Soil Behavior, 2nd ed. Wiley, NY, 437p. (1993)
15
-------
Morey, R.M. Ground Penetrating Radar for Evaluating Subsurface Conditions for
Transportation Facilities. NAS/NRC/TRB NCHRP Synthesis Report 255. (1998)
Moskowitz, B. M. Fundamental Physical Constants and Conversion Factors, in Global
Earth Physics, AGU Reference Shelf Vol. 1. Am. Geophysical Union, Washington, DC.
(1995)
Murray, Craig, and Keiswetter, Dean. Application of Magnetic and Multi-frequency EM
Techniques for Landfill Investigations: Case Histories. Symposium on the Application
of Geophysics to Environmental & Engineering Problems (SAGEEP) (1998). Pg. 445-
452.
Nobes, D.C. How Important is the Orientation of a Horizontal Loop EM System?
Examples From a Leachate Plume and a Fault Zone. Symposium on-ther Application of
Geophysics to Environmental & Engineering Problems (SAGEEP) (1998). Pg. 453-458.
Olhoeft, G.R. Quantitative Statistical Description of Subsurface Heterogeneities with
Ground Penetrating Radar at Bemidji, Minnesota in Proceedings of U.S. Geological
Survey Toxic Substance Hydrology Program Technical Meeting. USGS WRI91-4034.
(1991)
Olhoeft, G.R. Spatial Variability, in Proceedings ofNSF/EPRI Workshop on Dynamic
Soil Properties and Site Characterization, v. 1. Palo Alto, Electric Power Research
Institute Report NP-7337. (1991)
Olhoeft, Gary R. Geophysics Advisor Expert System Version 2.0. U.S. Geological
Survey Open-File Report 92-526 (1992).
Olhoeft, G.R. Site Characterization Tools, in. Proceedings of Third Int'l. Conf. on Ground
Water Quality Research, June 21-24, 1992
Olhoeft, G.R. Geophysical Detection of Hydrocarbon and Organic Chemical
Contamination in Bell, R.S., ed, Proceedings on Application of Geophysics to
Engineering, and Environmental Problems. Society of Engineering and Mining
Exploration Geophysics. (1992)
Olhoeft, G.R. Velocity, Attenuation, Dispersion and Diffraction Hole-to-hole Radar
Processing, in Proceedings of the Fourth Tunnel Detection Symposium on Subsurface
Exploration Technology, Colorado School of Mines, Golden, CO, 26-29 April 1993.
Olhoeft, G.R., and Capron, D.e. Laboratory Measurements of the Radiofrequency
Electrical and Magnetic Properties of Soils from near Yuma, Arizona. U.S. Geological
Survey Open-File Report 93-701. (1993)
Olhoeft, G. R. Geophysical Observations of Geological, Hydrological and Geochemical
Heterogeneity, in Symposium on the Application of Geophysics to Engineering and
Environmental Problems., Mar 27-31, 1994
Olhoeft, G. R. Modeling Out-of-plane Scattering Effects, in Proceedings of the Fifth Int'l.
Conf. on Ground Penetrating Radar, Kitchener, Ontario, 12-16 June 1994.
Olhoeft, G. R. and Capron, D. E. Petrophysical Causes of Electromagnetic Dispersion, in
Proceedings of the Fifth Int'l. Conf. on Ground Penetrating Radar, Kitchener, Ontario,
12-16 June 1994.
Olhoeft, G. R., Lucius, J. E., and Phillips, S. J. Geophysical Tracking of the Injection of
Trench Stabilization Material at U. S. Doe Hanford Site, Richland, Washington. USGS
Open File Report 94-146. (1994)
16
-------
Olhoeft, G. R., et al. Buried Object Detection with Ground Penetrating Radar, in
Proceedings ofUnexploded Ordnance (UXO) Detection and Range Remediation
Conference, Golden, CO. May 17-19, 1994
Olhoeft, G. R. Electrical, Magnetic and Geometric Properties That Determine Ground
Penetrating Radar Performance, in Proceedings of 7th Int 7. Conf. On Ground
Penetrating Radar. May 27-30, 1998
Opdyke, N. D. and Channell, J. E. T. Magnetic Stratigraphy. Academic Press, NY.
(1996)
Peck, Timothy J., Lige, Joy E., MacFarlane, Ian D., and Barranco, Frank T.
Characterizing In Situ DNAPL Distribution, Mobility State, and Dissolution. Non-
Aqueous Phase Liquids (NAPLs) in Subsurface Environment: Assessmentrand
Remediation Proceedings (1996).
Pellerin, L., Alumbaugh, D.L., and Pfeifer, M.C. The Electromagnetic Integrated
Demonstration at the Idaho National Engineering Laboratory Cold Test Pit. Symposium
on the Application of Geophysics to Environmental & Engineering Problems (SAGEEP)
(1997). Pg. 725-734.
Powers, M.H., et al. Gprmodel: One-dimensional Full Waveform Forward Modeling of
Ground Penetrating Radar Data. U.S. Geological Survey Open-File Report 92-532.
(1992)
Powers, M.H. and Olhoeft, G. R. Modeling Dispersive Ground Penetrating Radar Data,
in Proceedings of the Fifth Int'1. Conf. on Ground Penetrating Radar. Kitchener, Ontario,
12-16 June 1994.
Powers, M.H. and Olhoeft, G. R. GPRMODV2: One-dimensional Full Waveform
Forward Modeling of Dispersive Ground Penetrating Radar Data, Version 2.0. U. S.
Geological Survey Open File Report 95-58 (1995)
Powers, M.H. and Olhoeft, G. R. Modeling the Gpr Response of Leaking, Buried Pipes,
in Proceedings of SAGEEP. Keystone, Colorado. (1996)
Powers, M.H. and Olhoeft, G. R. Computer Modeling to Transfer GPR UXO
Detectability Knowledge Between Sites, in UXO Forum Conference Proceedings. March
26-29, 1996.
Sandberg, S.K., Rogers, N.T., Karp, K.E., Goodknight, C.S., and Spencer, L.F. IP and
TEMfor Discrimination and Resolution in Mapping Ground-water Contamination at
Monument Valley. Symposium on the Application of Geophysics to Environmental &
Engineering Problems (SAGEEP) (1998). Pg. 795-804.
Sander, K.A., Olhoeft, G.R., and Lucius, J.E. Surface and Borehole Radar Monitoring of
a Dnapl Spill in 3d Versus Frequency, Look Angle and Time, in Bell, R.S., ed,
Proceedings of the Symposium on the Application of Geophysics to Engineering and
Environmental Problems. Society of Engineering and Mineral Exploration Geophysics.
(1992)
Schmidbauer,E. and Mirwald, P. W. Electric Conductivity of Cordierite, in Mineralogy
Petrology, v. 48. (1993)
Schon, J. H. Physical Properties of Rocks. Fundamentals and Principles ofPetrophysics.
2nd. ed. Pergamon Press. (1998,)
Smith, G. S. an Introduction to Classical Electromagnetic Radiation. Cambridge Univ.
Press. (1997)
17
-------
Stumm, W. Chemistry of the Solid-water Interface — Processes at the Mineral-water and
Particle-water Interface in Natural Systems. Wiley-Interscience. (1992)
Tailing, D. H. and Hrouda, F. The Magnetic Anisotropy of Rocks. Chapman & Hall,
London. (1993)
Tiab, D. and Donaldson, E. C. Petrophysics. Theory and Practice of Measuring Reservoir
Rock and Fluid Transport Properties. Gulf Publishing Co. (1996)
Torquato, S. Random Heterogeneous Media. Microstructure and Improved Bounds on
Effective Properties. Appl. Mech. Rev., v. 44. (1991)
Tyburczy, J. A. and Fisler, D. K. Electrical Properties of Minerals and Melts, in. Mineral
Physics and Crystallography. AGU, Wash.DC. (1995)
Wright, D.L., Olhoeft, G.R., and Grover, T.P. Velocity, Attenuation and Dispersion
Electromagnetic Tomography in Fractured Rock. USGS WRI Report 94-4015. (1996)
Wright, D.L., et al. High-speed Digital Radar Systems and Applications to Subsurface
Exploration, in Proceedings of the Fourth Tunnel Detection Symposium on Subsurface
Exploration Technology. Colorado School of Mines, Golden, CO, 26-29 April 1993.
Wright, D.L., et al. Electromagnetic and Seismic Tomography Compared to Borehole
Acoustic Televiewer and Flowmeter Logs for Subsurface Fracture in Mapping at the
Mirror Lake Site, New Hampshire. USGS WRI Report 94-4015. (1996)
Wyatt, D.E. and Temples, T. J. Ground-penetrating Radar Detection of Small-scale
Channels, Joints and Faults in the Unconsolidated Sediments of the Atlantic Coastal
Plain. Environmental Geology - Abstract Volume 27 Issue 3 (1996)
18
-------
6.0 CASE STUDIES
-------
-------
Baker Wood Creosoting Company
Case Study Abstract
Baker Wood Creosoting Company
Marion, Ohio
Site Name and Location:
Baker Wood Creosoting Company
Marion, Ohio
Period of Site Operation:
1890'sto 1960's
Operable Unit: N/A
Points of Contact:
Mark Durno
U.S. EPA
25089 Center Ridge Road
Westlake, OH44145
216-522-7260
Mr. Mark Vendl
Mail Code SRT-4J
USEPA Region 5
77 West Jackson Boulevard
Chicago, IL 60604-3507
Geophysical Technologies:
Ground penetrating radar
Electromagnetic induction
Geological Setting:
Two to three feet of silt loam underlain
by a firm calcareous clay
Date of Investigation:
January and February, 1999
Current Site Activities:
Assessment of sediments in the Little
Scioto River is being performed. Future
plans include installing five or six shallow
water wells to determine if the
groundwater is contaminated.
Technology Demonstrator:
U.S. EPA and
.Ohio State University
Purpose of Investigation: To locate possible buried waste pits or other contaminant-filled structures and to delineate the
extent of contamination within the surficial soils.
Number of Images/Profiles Generated During Investigation: 1 00 GPR traverses
Results: Lateral extent of contamination determined in the shallow subsurface by EM and GPR. GPR was operated in a
cross- and co-pole antenna configuration which clearly identified a series of buried vaults containing highly contaminated
material.
19
-------
Baker Wood Creosoting Company
555 EXECUTIVE SUMMARY ••••••••••^MMM^Mi^MBHMMH
The fonner Baker Wood Creosoting Company is located on 60 acres in Marion. The site is located
approximately one-half mile northwest of downtown Marion. The Little Scioto River is located one mile
to the west of the site. The property was used from the 1890s to the 1960s as a wood treating facility,
and the preservatives used were most likely creosote, petroleum, and other solvents. All buildings have
been removed from the site, but the concrete pads that supported the creosote storage tanks and a former
pump house remain. The geophysical study was conducted within the area that encompasses the former
tank area and pump house.
The surficial soils consist of a two- to three-foot surface layer of silt loam, underlain by a firm calcerous
clay. Glacial till containing occasional thin interbedded sand layers extends from beneath the surface
soil to Silurian limestone/dolomite bedrock, which is present at depths of approximately 13 to 25 feet
below ground surface in the area. The limestone/dolomite bedrock appears to contain a shallow and deep
aquifer. Regional groundwater flow direction of the deep aquifer is believed to be influenced by the
quarry located northeast of the Baker Wood site and by the municipal well field situated west of the site.
Typically, the generalized groundwater flow is westward towards the Little Scioto River.
A geophysical investigation was conducted in 1999 to delineate the extent of contamination prior to
conducting a time critical removal action. The information in this report was derived from the
interpretive report for the geophysical investigation. Two geophysical methods were used during this
investigation. A ground penetrating radar (GPR) survey was conducted first, followed by a frequency
domain electromagnetic induction (EM) survey. The GPR was used to locate subsurface structures that
might contain contamination while EM was used to detect anomalous soil conductivities that might
indicate the presence of contamination in the surface and near-surface soils.
The GPR survey identified nine areas with significant subsurface anomalies in the study area. Five of the
areas included vaults buried underneath each of four tank pads, a creosote-filled pit, and a trench. The
EM survey found areas of low conductivity soils that indicate the potential location of contaminated soil.
Areas of low conductivity were less prominent in the lower frequency data than in the higher frequency
data indicating that contamination was predominantly present in the near surface. Subsequent
exploratory trenching and screening analysis of soils was conducted in the nine areas identified in the
geophysical investigation, and significant contamination was found in five of them.
Although soil sampling from 1996 showed contamination in the same area as the GPR survey showed,
the lateral extent of contamination was unknown prior to the GPR survey. The GPR survey provided
information on lateral extent. Based on the GPR survey, it was estimated that 1800 cubic yards of
contamination existed at the Baker Woods site. Because contamination was found to a depth of five and
six feet in some locations, and the GPR was only able to see to four feet, an additional 400 cubic yards of
contamination was found and removed.
20
-------
Baker Wood Creosoting Company
SITE INFORMATION
Identifying Information
Baker Wood Creosoting Company
Holland Road and Kenton Street
Marion, Ohio
Background [1, 2, 3, 4, 5, 7]
Physical Description: The former Baker Wood Creosotmg Company is located on 60 acres in
Marion, Ohio, in the north-central part of the state, as shown in Figure 1. The Baker Wood
Creosotmg Company is located at the northwest comer of Holland Road and Kenton Street (State
Route 309), and is approximately one-half mile northwest of downtown Marion. The Little
Scioto River is located one mile to the west of the site. The topography of the site is flat with a
shallow westward gradient.
All buildings have been removed from the site, but the concrete pads that supported the creosote
storage tanks and a former pump house remain.
The pads and former pump house are located
within an area of approximately 130 by 50 feet,
just south of a gravel access road. The geophysical
study was conducted on a 300- by 100-foot area
that encompasses the former tank area and pump
house. This part of the site is located in the
southeast section of the 60 acres (See Figure 2).
Site Use: The property was used from the 1890s to'
the 1960s as a wood treating facility, and was Figure 1: Site Location
owned by the Baker Wood Creosoting Company. Historical information indicates that the
process used pressure vessels to treat railroad ties and other wood products. The preservatives
used were most likely creosote, petroleum, and other solvents. It is currently owned by Baker
Wood Limited Partnership and is an inactive site.
It was believed that chemical wastes were discharged to the combined sanitary/storm sewer that
is located adjacent to the site, along the southern border. The sewer flows west and discharges
directly into North Rockswale Ditch. Drawings indicate that the old sewer tie-ins from the
facility may still be in use. This combined sanitary/storm sewer is thought to be a direct link to
the surface water contaminant migration pathway leading to the North Rockswale Ditch.
21
-------
Baker Wood Creosoting Company
SITE INFORMATION
Process Building
Creosote Storage
Tanks
Pump House
Holland Road
LEGEND
1996 Soil sample locations
Pre-existing buildings
or structures
Figure 2: Baker Wood Site Map [1]
Release/Investigation History:
Numerous sampling events have been conducted in and around the Baker Wood site since the
1940s. In 1988 and 1991, the Ohio Environmental Protection Agency (EPA) collected sediment
samples from the Little Scioto and Scioto Rivers. Analysis of the samples showed high
concentrations of polycyclic aromatic hydrocarbons (PAHs). Investigators observed on both
occasions that the banks and bottom sediments of the Little Scioto River were heavily saturated
with a black material with a creosote odor. When disturbed, the bottom sediments released an
substance that left an oily sheen on the water's surface. The U.S. EPA and Ohio EPA collected
soil samples in 1996 around the former creosote storage tanks and pump house. Analytical
results from the soil samples revealed some of the highest concentrations of PAHs ever recorded
in the published literature.
22
-------
Baker Wood Creosoting Company
SITE INFORMATION
Regulatory Context:
The U.S. EPA and Ohio EPA have conducted response actions at the Baker Woods Creosbtmg
Company site under a time critical removal authority provided under the Comprehensive
Environmental Response, Compensation, and Liability Act (CERCLA), as amended by the
Superfund Amendments Reform Act (SARA).
Site Logistics/Contacts
State Lead Agency: Ohio EPA
Federal Oversight Agency: U.S. EPA
Remedial Project Manager:
Mark Dumo
U.S. EPA
25089 Center Ridge Road
Westlake, OH 44145 ... .
216-522-7260
Site Contact:
Mr. Mark Vendl
Mail Code SRT-4J
USEPA Region 5
77 West Jackson Boulevard
Chicago, EL 60604-3507
312-886-0405
Geophysical Subcontractors:
Dr. Jeffery Daniels
Department of Geological Sciences
Ohio State University
125 South Oval Mall
Columbus, OH 43210-1398
614-292-4295
23
-------
Baker Wood Creosoting Company
I MEDIA AND CONTAMINANTS
Matrix Identification [3]
Type of Matrix Sampled and Analyzed:
Subsurface soil consisting of silt loam.
Site Geology/Stratigraphy [3, 5]
The surficial soil profile at the Baker Wood site consists of a two- to three'-foot- surface layer of
silt loam, underlain by a firm calcerous clay. Glacial till containing occasional thin interbedded
sand layers extends from beneath the surface soil to Silurian limestone/dolomite bedrock, which
is present at depths of approximately 13 to 25 feet below ground surface (bgs) in the area. The
limestone/dolomite bedrock appears to contain a shallow and deep aquifer. The shallow aquifer
is encountered at approximately 40 feet bgs, and the deep aquifer is encountered at about 250
feet bgs.
Regional groundwater flow direction of the deep aquifer is believed to be influenced by the
quarry located northeast of the Baker Wood site and by the municipal well field situated west of
the site. Typically, the generalized groundwater flow is westward towards the Little Scioto River.
Contaminant Characterization [1]
Primary Contaminant Groups: The primary contaminants of concern at this site are volatile
organic compounds (VOCs) and PAHs.
Matrix Characteristics Affecting Characterization Cost or Performance [1]
Clays in the soils and high soil moisture content posed a significant challenge for GPR data
collection during the investigation by limiting the depth to which measurements could be taken.
Both caused excessive signal attenuation resulting in late signal arrival times. Investigators tried
to correct for this interference by using a 300 MegaHertz (MHZ) antenna, but the radio tower,
located on the adjacent property, caused interfering noise at that frequency. As a result, a 500
MHZ antenna was used for the investigation, but at this frequency, the investigation depth was
limited to three to four feet bgs. The investigation team believed this to be a sufficient depth
based on prior knowledge of site conditions.
Standing water, which ranged in depth from 10 to 15 inches, on the site resulted in late signal
arrival times, but the standing water was mapped so that the data interpretation would not be
affected. The late arrival times were due to the water having a relatively lower velocity than the
surrounding areas that did not have water present.
24
-------
Baker Wood Creosoting Company
i GEOPHYSICAL INVESTIGATION PROCESS mmmmummm^mmum^mm
Investigation Goals [1]
The goals for this project were to locate possible buried waste pits or other contaminant-filled
structures, and to delineate the extent of contamination within the surficial soils.
Geophysical Methods [1,6]
Two geophysical methods were used for this investigation. A ground penetrating radar (GPR)
survey was conducted first, followed by a frequency domain electromagnetic induction (EM)
survey. The GPR was used to locate subsurface structures that might contain contamination
while EM was used to detect anomalous soil conductivities that might indicate the presence of
contamination in the surface and near-surface soils.
GPR employs an extremely short electromagnetic pulse that penetrates into the earth. A portion
of the energy is reflected back to the surface, where it is detected by the receiving antenna. The
amplitude of the reflected pulse depends primarily on the soil's dielectric constant, or the
measure of electrical conductivity of soils. GPR anomalies result when there is a contrast in the
bulk dielectric property between materials, marking a boundary between geologic structures.
The time lapse between transmission and receipt of the EM signal it is measured in nanoseconds
(ns) and is transmitted to a control unit for processing and display.
The GPR study design for the Baker Wood Site called for the collection of two complementary
sets of data. The data were collected using the GPR in a co-pole and then in a cross-pole antenna
configuration (See Figure 3). The collection and comparison of the two types of data added an
analytical dimension to the GPR data that improved the GPR data, thus improving the
interpretation of the results. Each antenna configuration is sensitive to different types of objects
in the subsurface.
The polarization of reflected electromagnetic energy depends on the geometry of the reflecting
surface. Relatively flat subsurface targets or ones with small curvature reflect relatively large
currents of linearly polarized signals. Targets that are not planar, or have irregular surfaces,
scatter or depolarize the EM waves. A co-pole antenna configuration is primarily sensitive to
linearly polarized reflections. The cross-pole configuration is most sensitive to depolarized
reflections, while being less sensitive to energy that is scattered parallel to the transmit antenna.
Thus, the use of both antenna configurations allowed investigators to identify anomalies
representing a wider variety of subsurface geometries.
25
-------
i GEOPHYSICAL INVESTIGATION PROCESS
Receive Antennas
Transmit Antennas
Multi-component GPR antenna arrangements
Traverse direction
Receive Antenna
Transmit Antenna
Traverse direction
Receive Antenna
1
Transmit Antenna
Co-pole perpendicular to traverse direction Cross-pole with transmit antenna
perpendicular to traverse direction
Figure 3: Antenna Configurations for Co-Pole and Cross-Pole GPR
Measurements
26
-------
GEOPHYSICAL INVESTIGATION
o
o
CO
o
CM
I
O
OJ
O
CO
I
O
.1
....
N
w -< >~ E : ; Creosote Tank Foundations • : Pump House Foundation
s •
Figure 4: Geophysical Survey Area Showing North-South Traverse Lines [1]
Geophysical surveys were conducted on a 100- by 300-foot grid with a 3-foot spacing between
the north-south traverse lines. This area included the foundations of the creosote storage tanks
and the former pump house (see Figure 4). Two separate surveys were conducted during January
and February 1999. The data from the two surveys was compared to identify variations in the
results due to changes in soil moisture. No significant variation was detected in the results of the
two surveys.
The GPR survey was conducted using a Geophysical Survey Systems, Inc. (GSSI) 500 MHZ
multi-component antenna with a Subsurface Interface Radar (SIR)-10 recording system with a
fixed number of traces recorded per distance traveled. The GPR system was towed using a
survey wheel to accurately position the data spatially.
The EM survey was conducted to identify the spatial extent of soil contamination, identified b y
the survey as areas of anomalous low conductivity resulting from creosote contamination within
surficial soils. It has been postulated that when organic contamination interacts with and
displaces soil moisture in the vadose zone, a decrease in conductivity can result. In the areas
where the highest levels of creosote contamination were found, the EM survey showed the lowest
conductivity values in the entire area.
The EM method is based on measuring the response of an electromagnetic field induced into the
earth. Low frequency signals, one to ten kilohertz, are transmitted by a small coil. The low
frequency, very long wavelength, electromagnetic fields produced by the transmitter induce
current flow in electrically conductive media in the earth. This induced current flow produces
27
-------
i GEOPHYSICAL INVESTIGATION PROCESS
secondary electromagnetic fields which will radiate back to the surface. A receiving coil detects
the secondary field and measures the strength and phase relative to the transmitted signal.
This EM survey was conducted using a GSSI GEM-300. For the GEM-300 system, the
secondary field that is measured is split into in-phase and quadrature components that are
expressed in parts per million (ppm) against the primary induced field strength. The in-phase
response is sensitive to metal conducting targets and is referred to as the metal detector mode,
while the quadrature phase response is sensitive to non-metallic conductors and is referred to as
the terrain conductivity mode.
EM measurements were taken every two feet along the same traverse lines on the 100- by 300-
foot survey grid used in the GPR survey. Measurements were taken at three different
frequencies: 2010 Hz (2kHz), 4410 Hz (4 kHz), and 9810 Hz (9kHz), with the long axis of the
instrument oriented parallel to the survey lines and the dipole axis oriented vertical to the plane
of the ground. The variation in frequencies provided investigations to different depths. The
depth of penetration of the transmitted field is a function of the frequency of operation or
frequency of the EM signal. Lower frequencies penetrate deeper, while higher frequencies are
attenuated more rapidly.
I GEOPHYSICAL FINDINGS I
Technology Calibration
No independent calibration information was required for the GPR and EM instruments used in
this investigation.
Investigation Results [1]
The GPR survey identified nine areas with significant subsurface anomalies in the study area.
Subsequent exploratory trenching and screening analysis of soils at the nine areas found
significant contamination in five of them. The five areas included vaults underneath each of four
tank pads, a creosote-filled pit, and a trench. The GPR findings discussed in this case study are
limited to those that focus on the creosote-filled pit and one of the tank pad vaults as they are
representative of the data collected around the other significant anomalies.
Figure 5 shows both two-dimensional (2-D) and three-dimensional (3-D) displays of the cross-
pole data collected southeast of the former pump house. Three-dimensional displays were
generated by stacking multiple 2-D profiles and provide an enhanced visualization of the GPR
anomaly. The anomaly in this profile was determined to be a creosote-filled pit during
subsequent sampling and analysis of the soils in the area. Co-pole data collected along the same
set of traverses contained more clutter, making identification of the anomaly difficult.
28
-------
Baker Wood Creosoting Cnmnanv
I GEOPHYSICAL FINDINGSl
Figure 6 presents 2-D and 3-D views of both co- and cross-pole data collected near the two
easternmost storage tank pads. A backfilled-trench, which later was discovered to contain
creosote-contaminated drainage tile, can be seen in the profiles, located between the two storage
pads. A comparison of the views generated using co- and cross-pole data shows that the cross-
pole data contains less clutter and has better resolution. The clutter present in the co-pole data
nearly obscures the trenched area between the two pads.
The EM survey in-phase data showed anomalous regions of relative high conductivity in the
vicinity of the tank and pump house foundations as a result of rebar within these structures.
These regions of relatively high conductivity were also evident in the quadrature responses at the
4 kHz and 9 kHz frequencies measured. Areas of low conductivity, shown as light areas in
Figure 7, indicate the potential location of contaminated soil. Research has shown that as the soil
moisture becomes contaminated with organic compounds, including those found at this site, the
electrical conductivity of those soils decreases. Areas of low conductivity were less prominent
in the 4 kHz data than in the 9 kHz data indicating that contamination is predominantly present in
the near surface. The 4 kHz data showed areas of low conductivity in the vicinity of the tank
foundations, which correlated with soil contamination that was found at greater depths.
Figures 8 shows where the geophysical survey found anomalies and where trenching was to be
conducted based on the anomalies. Figure 9 shows where the creosote-filled pits and vaults were
located. Comparing the two figures it is apparent that the accuracy with which the GPR survey
identified the location of the vaults and pit was within a few feet.
Although soil sampling from 1996 showed contamination in the same area as the GPR survey,
the lateral extent of contamination was unknown prior to the GPR survey. Based on the GPR
survey, it was estimated that 1,800 cubic yards of contamination existed at the Baker Woods site.
During excavation, contamination in some locations was found to a depth of 5 or 6 feet, but
primarily, the contamination was excavated from the same depths as those indicated by the EM
survey. By the end of the cleanup project, the total soil removed was 2,200 cubic yards. Thus,
estimations the results of the geophysical surveys were within 20 percent of the volume of
contaminated material excavated at the site.
29
-------
Baker Wood Creosotine ConriDanv
I GEOPHYSICAL FINDINGSl
0
0
Distance (ft)
25 50
75
100
10~
H 15-
20-
Creosote-fiHed pit
North
a) 2-D profile line
*
Creosote-filled pit
o North
b) 3-D block view
Figure 5: Cross-Pole Data Collected Near Pump House Foundation [1]
loeft
-------
25
15 -
20 -
Tank pad
75 J, 100
Tank pad AmrtngrelnT "• '• ' X^ '[ •' ;£*!?•"
.Trenched area -
North-
Crwtote-fflletl
vault
185.,.
«T
Crewwte-tUlcd"""^-}-,'
v»oH | /
WOfl
10 -
15 -
20 -
North-
100ft
Figure 6: GPR Data Collected Near Eastern Storage Pads [1] [Poor Quality Original]
II
o
w
o
W
HH
O
O
I— I
2!
O
OS
63
77
o
a.
i
en
O
3'
rQ
n
O
3
to
3
-------
Baker Wood Creosoting Company
I GEOPHYSICAL FINDINGSl
Surface water
300ft
ppm
''—IlKJO
t—J177S
I7S>
1725
1700
1675
I860
MBS
1600
1575
1530
1S2S
1500
1475
1450
142S
1400
137S
13SD
132S
1300
'300ft
Oft
ppm
:37TO
3730
MOD
3«0
36TO
3570
3530
3480
3450
3410
3370
3330
3290
3250
3210
3170
3130
2360
Figure 7: Electromagnetic Conductivity of Soils in Study Area [1]
32
-------
Baker Wood Creosotine Comnanv
I GEOPHYSICAL FINDINGSl
as
S
a
S.
S
Scale: 1 inch = 70 feet
t
N
I I Pump House Foundation
Creosote Tank Foundations
• Areas Marked for Excavation Based on Anomalies"
El Areas Marked to Test Extent of Contamination
Figure 8: Locations Selected For Screening by Geophysical Surveys [1]
0
x>
n
t
Scale: 1 inch = 70 feet
| [ Pump House Foundation
O Creosote Tank Foundations
— Creosote-filled Drainage Tile
• Creosote-filled Pit j
! N Bi Creosote-filled Vaults i
* - '
i I—1 Area Excavated to Remove Creosote in Soils j
Figure 9: Locations of Significant Soil Contamination [1]
33
-------
Baker Wood Creosottng Company
[GEOPHYSICAL FINDINGSl
Results Validation [1, 7]
Using the GPR and EM results together, investigators were able to identify nine areas for further,
invasive investigation. Exploratory trenches were excavated in each of the nine areas and soil
samples were taken from the trenches. The soil samples were analyzed in the field using a field
portable flame ionization detector. Based on these analytical results, significant contamination
was found in five of the nine areas.
I LESSONS LEARNED ^•••^•••••^^^•••^^^••^••S
Some of the lessons learned from this investigation include the following:
• The effectiveness of this GPR survey was improved with the collection and analyses of
both co-pole and cross pole data.
• Standing surface water and layered clay soils attenuated the GPR signal in certain
portions of the study area, interfering with the interpretation. These areas were mapped.
It is anticipated that results would have been clearer in the absence of standing water.
• The GPR survey was successful in identifying subsurface structures that held
contaminated material, including a vault hidden beneath a pit. The EM survey was
successful in identifying areas of suspected soil contamination. Information from both
surveys were used to identify nine areas for investigation. Trenches were excavated and
the soils analyzed in each area. Significant contamination was found in five of the areas.
• Although soil sampling from 1996 showed contamination in the same area as the GPR
survey showed, the lateral extent of contamination was unknown prior to the GPR
survey. Based on the GPR survey, it was estimated that 1800 cubic yards of
contamination existed at the Baker Woods site. Because contamination was found to a
depth of 5 and 6 feet in some locations, and the GPR was only able to see to 4 feet bgs,
an additional 400 cubic yards of contamination was found and removed.
• The low conductivity areas identified in the EM survey correlated with areas of high
concentrations of creosote contamination and were verified through soil sample analysis
and exploratory trenching.
34
-------
Baker Wood Creosoting Company
I REFERENCES
1.
Guy, E.; Daniels, J.; Holt, J.; Radzevicius, S. Geophysical Investigations at the Former
Baker Wood Site, Marion, Ohio. Ohio State University. August 1999.
2. U.S. Environmental Protection Agency. Environmental News Release No. 99-OPA117.
April 26, 1999.
3. Ohio Environmental Protection Agency. Bottom Sediment Evaluation of Little Scioto
River, Marion, Ohio. 1992.
4. U.S. Environmental Protection Agency. Fact Sheet: U.S. EPA Continues Cleanup At
Baker Wood Creosoting Site. July 1999.
5. Ohio Environmental Protection Agency. Integrated Assessment (IA) Report: Baker
Wood Creosoting Site, Marion, Marion County. 1998.
6. Personal Communication with Mark Vendl of U.S. EPA. September 30, 1999.
7. Personal Communication.with Mark Durno of U.S. EPA. September 30, 1999. ._..._
35
-------
THIS PAGE LEFT BLANK INTENTIONALLY
36
-------
Ciba-Geigy H&H
Case Study Abstract
Ciba-Geigy Hamblet & Hayes (H&H) Site
Lewiston, ME
Site Name and Location:
Ciba-Geigy Hamblet & Hayes Site
Period of Site Operation:
1963 to 1995 solvent
repackaging/chemical redistribution
Operable Unit:
Not applicable
Geophysical Technologies:
Ground Penetrating Radar (GPR)
CERCLIS #
Not applicable
Current Site Activities:
A ground-water pump and treat system and
an air sparging/soil vapor extraction
system were installed at the site in 1996,
and have been operating since early 1997.
Additional investigation work on areas
where dense nonaqueous phase liquids
(DNAPLs) have been found are also
ongoing.
Point of Contact:
Stephen Walbridge
Harding Lawson Associates
511 Congress Street
P.O. Box 7050
Portland, ME 04112-7050
(207) 828-3482
swalbrid@harding.com
Geological Setting:
The surficial unit is the Presumpscot
Formation, a marine deposit consisting
of varying amounts of clay, silt, and
fine sand. Overlying this formation is
"a unit primarily composed of sandy fill.
Below the Presumpscot Formation is a
sand and gravel unit.
Technology Vendor:
Geophysical Survey Systems, Inc.
13 Klein Dr.
North Salem, NH 03073-0097
(603)893-1109
Fax (603) 889-3984
sales@geophysical.com
Purpose of Investigation:
The purpose of the GPR survey was to provide information on the continuity and topographic relief of clay layers in near-
surface soils beneath the site. Identifying these high and low points of the clay layer would help identify where DNAPL
might accumulate.
Number of Images/Profiles Generated During Investigation: 85 traverses
Results:
The GPR survey successfully identified continuous reflectors that represent silty clay layers in the shallow subsurface soils
beneath the site. There was an observed parallel relationship of the various sand, silt, or clay layers that are present in the
shallow subsurface soils that suggest the topography of the interpretive layer mimics the topography of the massive silty clay
known to exist 19 to 22 feet below the ground surface (bgs) in the area of the GPR survey. This would provide a downward
sloping pathway for DNAPL to move along until accumulating in topographically low areas identified, such as beneath the
southwest corner of the leachfield.
37
-------
Ciba-Geigy H&H
•••EXECUTIVE
The Ciba-Geigy Corporation (Ciba-Geigy) Hamblet and Hayes (H&H) site is a complex of buildings
located off of Crowley Road in Lewiston, Maine. The facility was primarily known for its solvent
repackaging activities. Suspected site contamination was associated partly with an incident which
occurred in 1983 when a valve was inadvertently left open and approximately 1,000 gallons of xylenes
were spilled onto the ground. An environmental assessment was conducted and revealed that a
contaminated groundwater plume and contaminated soil, primarily consisting of chlorinated solvents and
xylenes, existed at the site.
The surficial geologic unit is the Presumpscot Formation, a marine deposit consisting of varying amounts
of clay, silt, and fine sand. Overlying the Presumpscot Formation at the site is a unit that is composed of
sandy fill to an approximate depth of 7 feet below ground surface (bgs). Underneath the Presumpscot
Formation is a sand and gravel unit that extends to depths of 45 feet bgs.
As part of the third Phase of the site investigation process, a ground penetrating radar (GPR) survey was
conducted to map the top of the clay surface. The information presented in this report was derived from
the interpretive report of the geophysical investigation. At least four reflectors were identified. The four
reflectors were interpreted to represent the top of the silty clay layers that comprise the upper portion of
the marine clay formation found at the site. The uppermost reflector was interpreted to be a silty clay
layer and was chosen for further interpretation. The silty clay layer was present on most profiles and
determined to be continuous throughout the study area. A topographic low for potential dense non-
aqueous phase liquid (DNAPL) pooling was identified near the western corner of the site.
The GPR data were accurate to site conditions and the confidence level of the decisionmakers in the
results was high. Their confidence in the level of accuracy of the GPR data was validated through later
investigations and comparisons to soil boring data. Overall the GPR survey was an effective tool for
identifying continuous reflectors that represented silty clay layers in the shallow subsurface soils beneath
the site.
As a result of the GPR survey, the topographic low point of the upper surface of the underlying aquitard
was determined. This low point was chosen as a location to install an extraction well, since this would
be a potential area where DNAPL might pool. DNAPL was encountered during the installation of the
extraction well, confirming the results of the GPR survey. Later comparisons to soil boring data also
verified the accuracy of the GPR data.
38
-------
Ciba-Geigy H&H
I SITE INFORMATION I
Identifying Information
Ciba-Geigy Hamblet & Hayes (H&H) Site
Lewiston, ME 55952
Resource Conservation and Recovery Act (RCRA) Site
Background [1]
Physical Description: The H&H site is a complex of buildings located to the southeastern side
of Crowley Road in Lewiston, Maine (Figure 1) which occupies an area approximately 450 feet
(ft) wide by 600 ft long on a 5.5 acre parcel of land at approximately 190 ft above mean sea level.
The site slopes gently from northeast to southwest, toward No Name Brook. Surface drainage
from around the buildings collects and flows into a drainage ditch that encompasses the site.
Overall, surface drainage primarily flows southwest from the site into No Name Brook. Swampy
conditions exist in the area of monitoring wells MW-205A and MW-205B, which is primarily to
the south of the study area (Figure 2). A mounded leachfield was built in 1979 to replace the
former leachfield and was used for treating sanitary wastes at the site. The previous leachfield
was located beneath what is currently the northeastern portion of the truck loading
warehouse/office building.
The study area for the Ground Penetrating Radar (GPR) survey was between Crowley Road and
the truck loading warehouse/office building (350 ft by 400 ft in area) and in the immediate area
along the southeast side of the building (100 ft by 250 ft in area). This covered an area from the
former underground storage tanks (USTs), which is the source area, to the railroad tracks (Figure
2)-
Site Use: The facility began operations as a solvent repackaging facility in 1963 as the Polar
Chemical Division of Hamblet & Hayes Co. (H&H), which was then was purchased by Ciba-
Geigy in 1978. The facility ceased solvent repackaging operations in 1985 and changed to a
chemical redistribution facility. While operating as a repackaging facility, bulk chemicals were
received by tank truck and railroad freight car and then stored in the warehouse and in a series of
eight USTs and two aboveground storage tanks. The USTs were located on the northwestern
side of the flammable materials storage building, which is located in the north end of the site.
One of the aboveground storage tanks was located adjacent to the flammable materials storage
building and the other on the southeastern side of the truck loading warehouse/office building.
Chlorinated and non-chlorinated solvents were stored in the storage tank areas. Solvents were
pumped from the tanks and repackaged into drums and other containers for distribution. The
containers were then loaded onto trucks and shipped for delivery. In 1989, the site was
39
-------
SITE INFORMATION!
Ciba-Geigy H&H
> /w !TS2?
i^v^IF
: 1IST7S. 7J M»«JTE SEFBSS OOADRANCa^. U5W1STOM. *«6 (1379)
Figure 1: Site Location Map for Ciba-Geigy Hamblet & Hayes [1]
40
-------
SITE INFORMATION
Ciba-Geigy H&H
£ I
i g
t t
— HI
£ S
is S
3 1
i
|
I
Ik
Is
I
o
1
i I
I *\ j j \\Q a s s s •*• *• ® I
41
-------
i SITE INFORMATION j
Ciba-Geigy H&H
purchased by Van Waters & Rogers, Inc., and operations ceased in 1995. La 1997 Ciba-Geigy
reacquired the property. Currently, limited truck parking and trailer transferring are the only
activities at the site.
A groundwater pump and treat system and an air sparging/soil vapor extraction (SVE) system
were installed at the site in 1996, and have been operating since early 1997. The pump and
treatment system includes four extraction wells screened in the shallow silty sand aquifer,
including EW-401 at the location of dense nonaqueous phase liquids (DNAPLs), and one
extraction well (EW-501) screened in the underlying sand and gravel aquifer. The sparging/SVE
system is located in the former UST area where light nonaqueous phase liquids (LNAPLs) and
very high volatile organic compound (VOC) concentrations have been found in both the
saturated and unsaturated zones.
Release/Investigation History: Suspected site contamination occurred in 1983 when a tank
valve was inadvertently left open and approximately 1,000 gallons of xylenes were spilled onto
the ground. H&H employees reported that xylenes ran along the asphalt driveway surface and
ponded in a low area off the asphalt directly across from the front of the truck loading
warehouse/office building. The spill was promptly reported to the Maine Department of
Environmental Protection (MEDEP) and emergency response crews responded to the spill by
excavating the ponded, free product xylenes and contaminated soils. A recovery sump was
installed at the corner of the flammable materials storage building in order to recover the portion
of xylenes that infiltrated the ground and was floating on top of the groundwater. H&H
employees reported that xylenes were skimmed from this sump for approximately four years after
the incident. The pumping was discontinued in 1987 due to low or nonexistent levels of
recoverable product.
In 1985, the USTs were removed under the supervision of MEDEP personnel. The excavation
was backfilled with soils excavated from around the tanks, along with clean, off-site backfill
material. The tanks appeared to be in excellent condition, but a solvent odor was noticeable. No
soil or water samples were collected as a part of the tank removal process.
The investigation that documented the suspected contamination at the site was a result of the
1989 property transfer Phase I investigation program. ABB Environmental Services, Lac. was
contracted by Ciba-Geigy Corporation (Ciba-Geigy) to conduct an environmental assessment and
develop a plan for any necessary cleanup of suspected soil and groundwater contamination at the
H&H site. This assessment generated enough data to determine that a contaminated groundwater
plume, primarily consisting of chlorinated solvents and xylenes, existed at the site. Soil
contamination was also identified as being present in the vicinity of the former UST area.
Contamination has also been identified in the sediments and surface water of No Name Brook.
Recent investigations have been conducted in areas where some DNAPL was identified during
the installation of extraction well EW-401. The presence of DNAPL was first confirmed in EW-
401 in November of 1994. EW-401 is located adjacent to piezometer PZ-4 (Figure 2).
42
-------
Ciba-Geigv H&H
SITE INFORMATION
Regulatory Context: This is a RCRA site where the MEDEP is providing oversight on all
aspects of work done at the site, including work plan reviews and approval and field site visits.
On March 27, 1997 the MEDEP entered into a compliance order by consent with Ciba-Geigy.
This order detailed the requirements and remedial objectives of the groundwater pump and
treatment system that was installed in 1996 and has been operational since early 1997 [5].
Site Logistics/Contacts
Federal Lead Agency: None
State Oversight Agency:
Maine Department of Environmental
Protection
Project Manager:
Peter Blanchard
Maine Department of Environmental
Protection
17 State House Station
Augusta, ME 04333-0017
207-287-7880
Peter J.Blanchard@state.me.us
Geophysical Subcontractor:
ABB Environmental Services Inc. (Now
Harding Lawson Associates)
511 Congress Street
P.O. Box 7050
Portland, ME 04112-7050
(207) 828-3482
Ciba-Geigy (Now Ciba Specialty
Chemicals) Manager:
Tom Smith
Ciba Specialty Chemicals Company
Remediation Services
P.O. Box 71
Oak Ridge Parkway
Toms River, NJ 08754
(732) 914-2867
I MEDIA AND CONTAMINANTS I
Matrix Identification
Type of Matrix Sampled and Analyzed: Subsurface soil and clay
Site Geology/Stratigraphy [1]
Native subsurface soils consist of a stratified sequence of outwash sands, peat, marine clay, and
sand and gravel layers. The upper layer encountered is a sandy fill layer, which consists of both
natural and man-made fill materials that overlay natural organic materials (peat). Debris such as
bricks, cinders, and spent coal can also be found in this layer. Beneath the sandy fill layer is a
silty sand layer, which consists primarily of fine sands and silts that varies in thickness from 1
foot to 19.5 ft. The silty sand is underlain by a marine clay layer known as the Presumpscot
Formation, which was deposited during the recession of the late Wisconsinan glacier. The
marine clay primarily consists of a blue-gray silty clay with a trace of fine sand, and with various
thickness of fine gray sand lenses with a weathered brown silty clay layer typically overlying the
blue-gray material. On average the clay was encountered at 10 ft below the ground surface (bgs)
43
-------
Ciba-Geigy H&H
I MEDIA AND CONTAMINANTS •••••i^HHMmBBHHMBMH
and ranged in thickness from 14 to 61 ft. The layer encountered beneath the clay is the sand and
gravel stratum, which consists of a wide range of soils types and gradations, ranging from clean
poorly graded sands to well-graded till with cobbles and boulders throughout. This layer was
encountered at depths ranging from 15 ft bgs to 73.5 ft bgs. Depth of bedrock was first
determined by borings and a seismic refraction survey conducted in 1993. This investigation
indicated that depth to bedrock is believed to be at an average of 55 ft bgs [2]. The silty clay, if
present in layers or as a massive deposit, is characterized by very low hydraulic conductivities.
Groundwater was encountered in all the soil borings taken during the site investigation, at depths
ranging from the ground surface to 6 ft bgs. Direction of groundwater flow is generally
southwest across the site toward No Name Brook and into the surface drainage ditches. This
holds true for both normal conditions and after heavy precipitation events have occurred.
Hydraulic conductivity values for the various subsurface strata are as follows: silty sand layer
has a mean conductivity value of S.VxlO"4 centimeters per second (cm/sec); silty sand and marine
clay interface zone has a value of l.lxlO"4 cm/sec; the marine clay has a value of 4.2xlO~8 cm/sec;
and the sand and gravel stratum has a value of 2.0xlO"3 cm/sec.
Contaminant Characterization [1]
Primary Contaminant Groups: Primary contaminants of concern found at the H&H site
include: chlorinated solvents (1,1,1-trichloroethane, tetrachloroethene (PCE), trans-1,2-
dichloroethene, and methylene chloride), aromatic hydrocarbons (benzene, ethylbenzene, toluene
and xylenes), naphthalene, and ketones. Floating free product organic solvents exist at the
former UST area. PCE has been found as a DNAPL at the site.
Matrix Characteristics Affecting Characterization Cost or Performance [1]
Parameters affecting performance of the GPR include a shallow water table and the nature of the
soils encountered at the site. To the south side of the truck loading warehouse/office building
penetration of GPR was limited by the dense fill that lies above the massive silty clay near the
ground surface. Swampy conditions existed in the area of MW-205A and MW-205B to the west
of the GPR study area (Figure 2), which limited the use of the GPR system and the extent of the
GPR survey, due to limited access to this area. No other factors were reported to impede the
effectiveness of the GPR survey or results.
The average depth to be surveyed was between 0 and 15 ft bgs. This is the area where the clay
layer is almost always encountered, since the average depth to clay is 10 ft bgs. This was an
ideal depth for the GPR survey to be effective in detecting the clay layer and whether or not it
was continuous across the study area.
A waterline, which was installed by the City of Lewiston, exists approximately 8 to 10 ft bgs on
the southeastern side of Crowley Road (Figure 2). The H&H facility's water line connection
between the City of Lewiston water line and the truck loading warehouse/office building
parallels the northeastern side of the mounded leachfield and is at a depth of 11 ft bgs. These
44
-------
Ciba-Geigy H&H
I GEOPHYSICAL INVESTIGATION PROCESS ^•••^•••••••H
water lines were not reported to interfere with the effectiveness or the results of the GPR survey.
Investigation Goals [1]
The goal of the geophysical investigation was to provide information on the continuity and
topographic relief of the clay layer beneath the site. This would help identify low areas in the
clay layer where DNAPL could potentially accumulate.
Geophysical Methods [1]
The GPR technique uses high-frequency radio waves to determine the presence of subsurface
objects and structures. A GPR system radiates short pulses of high-frequency electromagnetic
(EM) energy into the ground from a transmitting antenna. This EM wave meanders into the
ground at a velocity that is related to the electrical properties of subsurface materials
(specifically, the relative dielectric permitivity of the materials). When this wave encounters the
interface of two materials having different dielectric properties (i.e., soil and water), a portion of
the energy is reflected back to the surface, where it is detected by a receiver antenna and
transmitted to a control unit for processing and display. The major principles involved for GPR
are similar to reflection seismology, except that electromagnetic energy is used instead of
acoustic energy.
For this investigation a Geophysical Survey Systems, Inc. (GSSI) Subsurface Interface Radar
(SIR) System-3 GPR system equipped with 100 and 500 MHz antennae were used. Two-way
travel times of 50 to 75 nanoseconds were used. The GPR system was towed by hand across the
study area at a speed of 0.25 miles per hour (mph).
As part of the site investigation process, a GPR survey was conducted to map the top of the clay
surface. However, before the GPR study was conducted at the H&H site, a pilot study was
conducted by ABB Environmental Services, Inc. on July 2, 1991. This study indicated that GPR
would be effective in profiling the shallow subsurface strata. This led to the more
comprehensive survey that was conducted on October 23 and 24, 1991. The survey area was
between Crowley Road and the truck loading warehouse/office building and in the immediate
area along the southeast side of the building. This covered an area from the former USTs (the
source area) to the railroad tracks (Figure 2). An 18-inch steel culvert approximately 50 ft south
of the main GPR study area was also examined as part of this investigation. This steel culvert
was used for depth of penetration calibration.
The survey consisted of two separate grid areas (Figure 2). One grid area was established in the
front yard of the H&H site as a 50 foot by 50 foot grid oriented N30°W (magnetic). For this grid
the GPR traverses were conducted in northeasterly and southeasterly directions with 50 foot and
10 foot spacing, respectively. The second grid area was located to the rear of the facility and had
GPR traverses spaced 20 ft apart with perpendicular orientation to the southeast wall of the truck
loading warehouse/office building. For the 18-inch steel culvert, two short traverses were
oriented perpendicular to the culvert.
45
-------
Ciba-Geigy H&H
I GEOPHYSICAL INVESTIGATION PROCESS I
Technology Justification
The objective of the geophysical investigation was to determine the topographic relief of the clay
layer and whether it was continuous across the site. GPR was considered to be an ideal method
for being effective in detecting the clay since the average depth to be surveyed was between 0
and 15 ft bgs.
• GEOPHYSICAL FINDINGS I
Technology Calibration [11
Boring log information and an onsite steel culvert were used for depth of penetration calibration.
Investigation Results [1]
Close examination of the 85 GPR profiles revealed a minimum of four reflectors that could be
identified on most of the traverse profiles that were generated. These four reflectors were
interpreted to represent the top of the silty clay layer surfaces within the-transitional zone of the
upper portion of the Presumpscot Formation.
The uppermost reflector was found at about 6 ft bgs at the location of MW-106. This was
identified on most of the data profiles performed at the site. This uppermost reflector is
interpreted to be a silty clay layer within the upper part of the Presumpscot Formation. The
surface of this uppermost silty clay reflector generally slopes westward below the survey area at
a rate of 5.5 ft per 100 ft. A local topographic low near the western corner of the leachfield
exists at approximately 19 ft bgs. This topographic low exists approximately 10 ft below the
elevation of the clay surface at the eastern edge of the survey area. Interpretation of data
suggested that the topographic low could be part of a trough that trends in a southerly direction.
Unfortunately the survey did not extend far enough westward to define the shape of the clay
surface in the vicinity of monitoring wells MW-205A, and MW-205B. This was due to the
swampy conditions in this area that would not allow access with the GPR equipment.
The depths to the top of the uppermost reflector were tabulated in nanoseconds of two-way travel
time. Travel times were then converted to depth using a conversion factor of 5.75 nanoseconds
per foot of depth. Depth in feet was then converted to elevation and an interpretive map of
elevation contours of the uppermost reflector was created (Figure 3).
Examination of the data collected from the two traverses above the steel culvert indicated that the
culvert is not surrounded by transmissive sands and gravels that would act as pathway for
groundwater migration. It is thought that the culvert is most likely surrounded by compacted fill.
The general conclusion after examining the GPR results was that there was an observed parallel
relationship of the various layers of sand, silt, or clay that are present within the shallow
subsurface soils of the Presumpscot Formation. This suggested that the topography of the
46
-------
I GEOPHYSICAL FINDINGSl
Ciha-Geiov
interpreted layer mimics the topography of the top surface of the massive silty clay found at
approximately 19 to 22 ft bgs in the area of the GPR survey. Based on this survey and
subsurface explorations, the marine clay formation appears to be continuous across the study
area. If present, DNAPL could move downslope along the top surface of the silty clay layers and
accumulate in the topographical low areas, such as the low area identified near the southwest
corner of the leachfield.
Results Validation
Confidence in the level of accuracy of the GPR data was verified through later investigations and
comparisons to soil boring data. A digital model created from the soil boring data would produce
the same shape when compared with the GPR data [6]. This confirms the results of the GPR
survey that the clay formation appears to be continuous across the study area.
An extraction well (EW-401) was installed in November, 1994 at a location previously
determined from the GPR survey to be a topographic low point on the upper surface of the clay
formation [4]. PCE was found as a DNAPL along with other VOCs at this well; this confirmed
that the location identifiedby the -GPR survey was a topographic -low-point-where-DNAPL might
accumulate.
47
-------
[GEOPHYSICAL FINDINGSl
48
-------
Ciba-Geigy H&H
I LESSONS LEARNED
Some of the lessons learned during this investigation include:
• Overall, the GPR survey was an effective tool for identifying continuous reflectors that
represented silty clay layers in the shallow subsurface soils beneath the site [6]. The
confidence level of the decisionmakers in the results was high. Confidence in the level
of accuracy of the GPR data was verified through later investigations and comparisons to
soil boring data.
Penetration of the GPR survey was limited on the southeast side of the truck loading
warehouse/office building, since it did not identify subsurface reflectors. This is
interpreted as a result of the presence of dense fill that lies above the massive silty clay
near the ground surface [1].
As a result of the GPR survey, the surface of the clay layer underneath the site was
mapped, and the topographic low point of the upper surface was determined. This low
point was chosen as a location to install an extraction well, since this would be a
potential area where DNAPL might pool. DNAPL was encountered during the
ifistallatioh'of the ex1racTibri'well,~as susp~ected [4J. . "
49
-------
Ciba-Geigy H&H
IREFERENCES
ABB Environmental Services, Inc. Site Investigation Report. Prepared for Ciba-Geigy
Corporation, Hamblet & Hayes Site, Lewiston, Maine. February 1992.
ABB Environmental Services, Inc. Site Investigation Report Addendum. Prepared for
Ciba-Geigy Corporation, Hamblet & Hayes Site, Lewiston, Maine. March 1993.
ABB Environmental Services, Inc. Work Plan: Well Installation and Decommissioning.
Prepared for Ciba-Geigy Corporation, Hamblet & Hayes Site, Lewiston, Maine. August
1994.
ABB Environmental Services, Inc. Work Plan: WellMW-408A Investigation and Dense
Non-Aqueous Phase Liquid Assessment. Prepared for Ciba-Geigy Corporation, Hamblet
& Hayes Site, Lewiston, Maine. February 1995.
Personnel Communication with Mr. Peter J. Blanchard of the Maine Department of
Environmental Protection. Septembers, 1998.
Personnel Communication with Mr. Tom Smith of Ciba-Geigy.- September 3, 1998.
Personnel Communication with Mr. Stephen Walbridge of Harding Lawson Associates.
September 18, 1998.
50
-------
Case Study Abstract
Crystal Refinery
Crystal Refinery-
Carson City, MI
Site Name and Location:
Crystal Refinery
Carson City, MI
Period of Site Operation:
1935 to early 1990s
Operable Unit:
Not Applicable
Point of Contact:
David Monet, Geologist
Environmental Response Division
MIDEQ
245 Colrain, SW
Wyoming, MI 49548-1013
(616)246-1739
Geophysical Technologies:
Ground Penetrating Radar (GPR)
Electrical resistivity
Geological Setting:
Alluvial sand and loam soils overlay a
shallow, fine to coarse sand aquifer
separated from deeper aquifers by a
clay aquitard. Bedrock occurs at 350 ft
bgs and is composed of sandstone,
shale, limestone, siltstone, and clay.
CERCLIS #
Not Applicable
Current Site Activities:
Groundwater pump and treatment system
via French drains/capture trenches
Technology Demonstrator:
William A. Sauck, PhD
Department of Geosciences
Western Michigan University
Kalamazoo, MI 49008
(616)387-4991
sauck@wmich.edu
Purpose of Investigation:
Investigation of the hypothesis that electrical properties of the zone impacted by a hydrocarbon plume in a natural
environment change over time from electrically resistive to electrically conductive due to biodegradation, and that this shift in
conductivity can be measured using geophysical methods.
Number of Images/Profiles Generated During Investigation:
1 GPR profile, 1 dipole-dipole resistivity profile, 1 vertical resistivity probe profile and associated soil boring
Results:
The investigation confirmed the hypothesis that an older light non-aqueous phase liquid hydrocarbon plume in the natural
environment will shift the bulk resistivity of the impacted zone from high resistivity to low resistivity over time.
Project Cost:
Estimated total cost for the investigation of this type was approximately 55,795.
51
-------
Crystal Refinery
w^m EXECUTIVE
The Crystal Refinery is located in northwest Carson City in central Michigan. To the north and
northwest there is a pine forest, and to the east and southeast, there are agricultural lands, residences, and
commercial businesses. Site topography is characterized by rolling hills and uniformly western sloping
plains. The geology at this site was created by two types of moraines created during the Wisconsin
Glacial Period; a ground moraine and two end morainic ridges. The soil in the ridges is composed of
clay, while soils deposited by glacial outwash in the ground moraine are sands and loams. The alluvial
soils overlay a shallow, fine to coarse sand aquifer which is separated from deeper aquifers by a clay
aquitard.
For this geophysical investigation two methods were used. The first was ground penetrating radar
(GPR), which uses high-frequency radio waves to determine the presence of subsurface ^objects and
structures. The second method was electrical resistivity, which injects electric currents into the earth
through a pair of current electrodes, and the potential difference is measured between a pair of potential
electrodes. The investigator chose GPR and electrical resistivity as the best methods to recognize the
geoelectric properties of the volume of earth containing a hydrocarbon contaminant plume.
The GPR profile revealed a strong and continuous reflector which was interpreted as the water table at a
depth of"l 0-18 ieet—ThisTS-in-agreementwith known water-table-measurements..- A-second reflector was
visible in the profile just above the water table from the west to approximately 100 meters (m). Depths
to this reflector are computed as 2.7 m in the west down to 5.5 m at its lowest point. Soil boring data
indicate that this reflector is coincident with the top of a layer containing residual product and exhibiting
oil staining and a strong gasoline odor. The dipole-dipole resistivity profile demonstrated high resistivity
in the vadose zone and a gradient to low resistivity. The areas of low resistivity were interpreted as the
saturated zone and the clay aquitard-beneath it. The .vertical resistivity probe was placed in a known area
of free product. The vertical resistivity probe revealed high resistivity in most of the vadose zone, in
correlation with the dipole-dipole profile. However, near the base of the vadose zone and in the
uppermost part of the saturated zone, a pronounced resistivity minimum was encountered.
This investigation was done as a field demonstration by Western Michigan University and all the
equipment was owned by the University. Therefore, there were no direct labor and equipment costs
associated with this geophysical investigation. However, the estimated cost of initiating such an
investigation using three different geophysical methods would cost approximately $6,000.
Biodegradation of mature LNAPL plumes can produce geochemical changes in the materials at the
capillary fringe that mobilize inorganic compounds from the subsurface materials. The change in pH and
ion charge of the materials increases the conductivity of the subsurface materials. This increase in
conductivity can be detected using electromagnetic methods, such as ground penetrating radar [2]. Light
hydrocarbon free-product and associated dissolved plumes are dynamic systems. Therefore the
application of geophysical techniques to investigations such as this should be conducted in conjunction
with geochemical investigations [2].
52
-------
Crystal Refinery
ISITE INFORMATION I
Identifying Information
Crystal Refinery
Carson City, MI 48811
Background [21
Physical Description: Crystal Refinery is located on North Williams Street in northwest Carson
City in central Michigan. Carson City is a small rural town and the site is located in a residential
and commercial area. The Carson City Park is located on the southern border of the site, and
Fish Creek forms the western border. To the north and northwest there is a pine forest, and to the
east and southeast, there are agricultural lands, residences, and commercial businesses (Figure
1). The center of Carson City is located approximately 3,000 feet to the southeast of the site.
The site is approximately 32.5 acres and consists of two separate parcels. The larger southern
parcel contains the petroleum refinery, storage tanks, lagoons, loading docks, and several
buildings. The northern parcel contains storage tanks, a valve station, and a disposal area. Both
parcels are partially fenced and gated. Site topography is characterized by rolling hills and
uniformly western sloping plains. Between the two parcels is a cemetery. South of the southern
parcel is a city park.
Site Use: Crystal Refinery began processing crude oil in 1935 and operated until the early 1990s.
The site received crude oil from both an underground pipeline and railroad cars. The average
production of the refinery was approximately 84,000 gallons of oil per day. Total tank storage
capacity, including above-ground storage tanks (ASTs) and underground storage tanks (USTs),
was an estimated 10,000,000 gallons. Between 1957 and 1962, two additional ASTs were
constructed on the northern parcel, adding 2,000,000 gallons to the total tank storage capacity.
Eight cooling lagoons on the southern parcel received waste sludges from site operations. These
sludges, copper chloride, Fuller's Earth, and styrene materials were transferred to and disposed
of on the northern parcel until the mid-1970s. In 1970, four french drains were installed.
Recovered oil and water from these drains was pumped to an oil/water separator. Remaining
liquids were then pumped to a second separator and discharged into one of the lagoons. The site
is currently inactive.
Release/Investigation History: Shortly after site operations began, an oil seep was discovered
in the north recovery area. The Michigan Department of Natural Resources (MDNR)
investigated this seep in 1945, however there is no information regarding the results of this
investigation. In 1968, MDNR performed an evaluation of Crystal Refinery to determine the
extent of oil contamination in groundwater. Twenty-two wells were installed, both on- and off-
site, to establish the lateral extent of contamination and groundwater flow direction. A heavy oil
slick was noted on the backwaters of Fish Creek at this time.
53
-------
ISITE INFORMATION!
C!rvsfal Refinery
STORAGE:. -T\!
rANK
GRAND
TRUNK
'HRAiLROAD
DISSOLVED FPCX3UCT BOUMDARV
FREE PRODUCT 1H|CKNE55
APPnOX. S7E
•HJI^^- G.
-------
(~!rvstal T? efi n erv
•SITE INFORMATION mm
A large release of crude oil occurred in January, 1973 when a fractured check valve on a
receiving line burst. Eighty-eight thousand gallons of crude oil flowed over frozen ground
toward Fish Creek. Although some was intercepted on site, oil did enter Fish Creek. The
ensuing cleanup was performed by Crystal Refinery.
hi 1982, Crystal Refinery conducted a hydrogeological investigation required by MDNR. More
wells were installed to evaluate both the lateral and vertical extent of contamination. This
investigation estimated that as much as 4,000,000 gallons of oil in might be present in the
groundwater, 117,000 cubic yards of soil might be seriously impacted by oil contamination, and
86,000 cubic yards of soil may have been marginally impacted. In 1983, Crystal Refinery
installed purge wells in an attempt to address the groundwater contamination. Water pumped
from the wells was skimmed, sent to a separator, and then to one of the lagoons.
A 1989 EPA visit to the site documented degradation of containment measures, such as erosion
around the lagoons and degradation of insulation in storage tanks. EPA instructed Crystal
Refinery to address these conditions, and in 1992, MDNR required the development of a
remedial action plan (RAP) addressing both groundwater and soil contamination. A RAP was
completed in 1992 addressing only groundwater issues. MDNR accepted the RAP as an interim
response, but stated it was inadequate until soil concerns were addressed. Since 1993, Crystal
Refinery has continued to address groundwater, but has not performed any remedial measures
addressing soil contamination. The EPA razed (removed) all above-ground facilities in Fall,
1998.
Regulatory Context: MDNR has been the lead agency in overseeing and approving the Crystal
Refinery activities and decisions. However, in April 1997, MDEQ (formerly MDNR) referred
the site to the US EPA Region 5 Emergency Response Branch.
Site Logistics/Contacts
Federal Lead Agency:
US EPA Region 5 Emergency Response
Branch
Project Manager:
David Monet, Geologist
Environmental Response Division
MIDEQ
245 Colrain, SW
Wyoming, MI 49548-1013
(616) 246-1739
State Lead Agency:
Michigan Department of Environmental
Quality
Geophysical Subcontractor:
William A. Sauck, PhD
Department of Geosciences
Western Michigan University
Kalamazoo, MI 49008
(616) 387-4991
sauck@wmich.edu
55
-------
Crystal Refinery
I MEDIA AND CONTAMINANTS I
Matrix Identification
Type of Matrix Sampled and Analyzed: Subsurface soils and groundwater
Site Geology/Stratigraphy
The geology at this site was created by two types of moraines created during the Wisconsin ^
Glacial Period; a ground moraine and two end morainic ridges. The soil in the ridges is
composed of clay, while soils deposited by glacial outwash in the ground moraine are sands and
loams. The alluvial soils overlay a shallow aquifer composed of fine to coarse sand. The
shallow aquifer is separated from deeper aquifers by a clay aquitard. The thickness of the
shallow aquifer ranges from 15 feet in the western portion of the site to 30 feet in the eastern
portion. Bedrock occurs at 350 feet below ground surface and is formed by the Jurassic and
Saginaw Formations. The bedrock is composed of primarily sandstone, shale, limestone, clay,
and siltstone.
Contaminant Characterization
Primary Contaminant Groups: The contaminants of concern are residual oil and hydrocarbons
both as dissolved phase and free product. Hydrocarbons are present as light, non-aqueous phase
liquids (LNAPLs). Both crude oil (north) and refined products (south).
Matrix Characteristics Affecting Characterization Cost or Performance [21
Some fading of the ground penetrating radar (GPR) reflections occurred and may have been
related to enhanced soil conductivities which limit the effective depth of penetration of the radio
waves. Some of the problems affecting the performance of the electrical resistivity data
interpretations were related to equivalence and suppression. Lower resistivities at the fringes of
survey lines may have been the result of a shallower water table to the west, and, to the east,
lower resistivities may have been caused by the presence of road salt. Other factors such as the
surface conditions, and subsurface distribution of conductive zones may also play an important
role in controlling the electrical signature of surface geophysical measurements at hydrocarbon
impacted sites.
56
-------
Crystal Refinery
I GEOPHYSICAL INVESTIGATION PROCESS
Investigation Goals
The geophysical investigation was undertaken in November 1997 as part of academic research
for Western Michigan University's Department of Geology. The purpose of the investigation
was to test the proposition that the electrical properties of the soil moisture intermingled with a
hydrocarbon plume change over time from electrically resistive to electrically conductive. The
conventional model, based on controlled spill and lab experiments, is that groundwater and soils
contaminated with hydrocarbons exhibit lower electrical conductivity arid'lower relative
permittivity than the surrounding uncontaminated media. The hypothesis tested in this study is
that hydrocarbon spills in the natural environment will change the bulk properties of the
impacted zone from electrically resistive to electrically conductive over time due to
biodegradation of the hydrocarbons. Conductivity is enhanced by the leaching of inorganics
from the soil and aquifer materials by organic acids produced by microbial activity during
degradation of the hydrocarbons [1].
Geophysical Methods [2]
For this geophysical investigation two methods were used. The first was GPR, which uses high-
frequency radio waves to determine the presence of subsurface objects and structures. A GPR
system radiates short pulses of high-frequency electromagnetic (EM) energy into the ground
from a transmitting antenna. This EM wave propagates into the ground at a velocity that is
related to the electrical properties of subsurface materials (specifically, the relative dielectric
permittivity of the materials). When this wave encounters the interface of two materials having
different dielectric properties (i.e., soil and water), a portion of the energy is reflected back to the
surface, where it is detected by a receiver antenna and transmitted to a control unit for processing
and display. The major principles involved for GPR are similar to reflection seismology, except
that EM energy is used instead of acoustic energy and the propagation times are much shorter.
The GPR survey was conducted using the Geophysical Survey Systems Inc. (GSSI) Subsurface
Interface Radar-10A+ (SIR-10A+) with 300 MHZ bistatic antennae. The modulation frequency
was set at 300 MHZ with a recording time of 160 nanoseconds (ns). The survey used a constant
gain setting and a 3-scan moving average horizontal filter. The GPR system was towed for 230
meters along two lines 20 meters apart at 15 and 35 meters south of the refinery boundary
(Figure 1).
The second geophysical method used was electrical resistivity. During resistivity surveys,
current is injected into the earth through a pair of current electrodes, and the potential difference
is measured between a pair of potential electrodes. The current and potential electrodes are
generally arranged in a linear array. Common arrays include the dipole-dipole array, pole-pole
array, Schlumberger array, and the Wenner array. The apparent resistivity is the bulk average
resistivity of all pore fluids, soils and rock influencing the flow of current. Resistivity is the
inverse of conductivity. It is calculated by dividing the measured potential difference by the
input current, and multiplying by a geometric factor (specific to the array being used and
electrode spacing). Models of the variation of resistivity with depth can be obtained using model
curves or forward and inverse modeling computer programs.
57
-------
Crystal Refinery
[GEOPHYSICAL INVESTIGATION PROCESS
Electrical resistivity was measured with the Iris Syscal R2 Deep Resistivity-IP System using the
axial dipole-dipole array configuration with dipole separations between 1 and 5 and the Wenner
array configuration with a 2-inch electrode spacing for vertical resistivity measurements in the
vertical probes. Both the dipole-dipole and vertical profiling (not along a line, but a single point)
were conducted along a line 20 meters south of the refinery. The Iris Syscal R2 Deep
Resistivity-IP System is menu-driven and has internal storage memory and weighs approximately
6 kilograms.
i GEOPHYSICAL FINDINGS
Technology Calibration
No calibration was reported as being necessary for this investigation. However, for the
resistivity system calibration is usually done digitally by the microprocessor based on correction
values stored in memory. The correction values are found in final production testing and are also
established during later periodical recommended yearly checks at authorized service centers.
Vertical probes have been calibrated in a water tank to determine the correction factor for the
body of the probe (2" OD PVC cylinder-perfect insulator).
Investigation Results [2,3]
The GPR profile revealed a strong and continuous reflector occurring at 40 ns near the west end
and 70 ns further east along the profiles (Figure 2). This reflector is interpreted as the water
table and a depth of 3.5 to 5.5 m was computed. This is in agreement with known water table
measurements. Another, parallel, reflector is visible in the profile just above the water table
from the west to approximately 100 meters. At 100 meters, this reflector dips to the east and
then merges with the W.T. reflector at 140 meters. After 160 meters, this reflector is visible as a
separate event again and rises to 50 ns. Depths to this reflector are computed as 2.7 m in the
west down to 5.5 m at its lowest point at 140 meters east. Soil boring data indicate that this
reflector is coincident with the top of the layer containing residual product and exhibiting oil
staining and a strong gasoline odor. The appearance of this reflector on the GPR profile may be
due to viscous residual product in the vadose zone blocking sediment pore space and altering the
permeability.
The dipole-dipole resistivity profile demonstrates high resistivity in the vadose zone and a
downward gradient to low resistivity (shown as lighter shades in Figure 3). The areas of low
resistivity are interpreted as the saturated zone and the clay aquitard beneath it. The fact that no
anomalous features attributable to the free product plume were observed is possibly due to
problems of equivalence and suppression associated with resistivity interpretations.
The vertical resistivity probe is located in a known area of free product. The vertical resistivity
probe revealed high resistivity in the vadose zone (staying around 1000 Ohm-meters on a log
scale), in correlation with the dipole-dipole profile. However, at approximately 3.8 m bgs,
58
-------
Crystal Refinerv
I GEOPHYSICAL FINDINGS
electrical resistivity suddenly decreases to 15 Ohm-meters. This is lower than background water
resistivities of 30 Ohm-meters. According to soil boring data, this zone of low resistivity begins
just above the water table and is coincident with the layer containing free product observed
between the upper reflector and the W.T. reflector in the GPR survey. This is interpreted to
confirm that a natural environment zone which has been saturated with hydrocarbon for a period
of time (in this case, 50 years) exhibits an increased conductivity (decreased resistivity), contrary
to the conventional model that it will display conductivities less than the uncontaminated areas.
Results Validation [2]
Geochemical data, including dissolved oxygen, pH, and specific conductance was collected from
five on-site wells (Table 1). The locations of these wells can be seen in Figure 1. The high
measurements of dissolved oxygen and low conductivity in OW-10 and OW-21 indicate minimal
impact by hydrocarbon contamination. Similar measurements in OW-43 showed low dissolved
oxygen and the highest conductivity. This well is located at the margin of the dissolved phase
plume, as is OW-44. Measurements in OW-31 revealed the lowest dissolved oxygen and a
corresponding high conductivity. The variations in measurements are attributed to varying rates
of biodegradation. Waters from below the impacted zone were 3-5 times more conductive than
background. The low dissolved oxygen rates correlate with high conductivities and indicate
microbial activity is breaking down the hydrocarbons. The use of ambient dissolved oxygen
ultimately results in (involves bacterial process, then chemical leaching process) elevated
conductivities.
59
-------
(~!rvstal Rpfinprv
I GEOPHYSICAL FINDINGS
W
Time
10 30
70 90 110 130 150 170 190 210 m
E
WT
WT
fine 35
Shadow Zone?
Shadow Zone
Figure 2: Ground Penetrating Radar Profiles of Line 15 and Line 35 [2]
w
0.0
40.0
80.0
120
160
E
meters
1
2
3 .
4
5 J
0.0
a) Measured Apparent Resistivity Pseudosection
40.0 80.0 120
160
Depth 0.0
b) Calculated Apparent Resistivity Pseudosection from Inversion Result Below
40.0
80.0
120
160
(m)
c) Inversion Result, Shown as Subsurface Resistivity Pattern
Figure 3: Resistivity Pseudosections [2],
60
-------
Crystal Refinery
I GEOPHYSICAL FINDINGS
CD
JC.
o
CL.
CD
Q
-180 —
-240 —
-300-
10
Vadose Zone
Approx. Water Table
Anomalous
Conductivity
Zone
Carson City,VRP1A
[200 m W,20 m S]
2" Wenner Array
Oct. 19, 1997 WAS
i i i i i i i
100
Resistivity, Ohm-m
1000
Figure 4: Vertical resistivity probe located at 0.0 mE, 20 mS, 2" Wenner Array,
semi-log scale [2].
61
-------
I GEOPHYSICAL FINDINGS
C!rvstal T?efinerv
Table 1: Geochemical Data Used in Results Validation
Well ID
OW-10
OW-21
OW-31
OW-42
OW-43
OW-44
Dissolved
Oxygen (mg/L)
7.2
7.2
0.3
0.7
0.6
1.2
pH
6.4
7.1
6.37
7
6.76
6.5
Specific
Conductance
(mS/m)
32
30
93
90
158
101
I LESSONS LEARNED
The lessons learned during this investigation are the following:
• Biodegradation of mature LNAPL plumes can produce geochemical changes in the
materials at the capillary fringe or zone of mixing that mobilize inorganic compounds
from the subsurface materials. The change in pH and ion charge of the materials
increases the conductivity of the subsurface materials. This increase in conductivity can
be detected using electromagnetic methods, such as ground penetrating radar. This
phenomenon will be limited to "mature" plumes, and depending on the specific chemical
nature of the plume and the viability of the indigenous microbial population, may not be
observed at all sites [2].
• Ground penetrating radar was able to clearly identify the water table and the top of the
impacted zone [2].
• No anomalous regions which can be attributed to the free product plume could be
observed along the horizontal resistivity profile. This was likely due to the problems of
equivalence and suppression, which often plague resistivity interpretations [2].
However, electrical resistivity data from fixed vertical resistivity probes showed
resistivity minima which coincide with GPR shadow zones with relation to the depth of
the water table.
• Light hydrocarbon free-product and associated dissolved plumes are dynamic systems.
Therefore the application of geophysical techniques to investigations such as this should
be conducted in conjunction with geochemical investigations [2]. This will result in a
better understanding of site conditions.
62
-------
Crystal Refinery
REFERENCES
Baedecker, M.J., Cozzarelli, I.M., Eganhouse, R.P., Siegel, D.I, and Bennett, P.C. Crude
Oil in a Shallow Sand and Gravel Aquifer - III. Biogeochemical Reactions and Mass
Balance Modeling in Anoxic Groundwater; Applied Geochemistry, vol. 8, pp. 569-586.
1993.
Atekwana, E., W.A. Sauck, and D.D. Werkema, Jr. Characterization of a
Complex Refinery Groundwater Contamination Plume Using Multiple Geoelectric
Methods. Proceedings of the Symposium on the Application of "Geophysics to
Engineering and Environmental Problems (SAGEEP 1998), pp. 427-436.
Sauck, William A. A Conceptual Model for the Geoelectrical Response ofLNAPL
Plumes in Granular Sediments. Proceedings of the Symposium on the Application of
Geophysics to Engineering and Environmental Problems (SAGEEP 1998) pp. 805-817.
Personal communication with Phil Sirles of Microgeophysics. Wheat Ridge, CO.
. DecemberJ.O, 1998- . .
63
-------
THIS PAGE LEFT BLANK INTENTIONALLY
64
-------
Kansas Underground Storage Tank (UST) Site
Case Study Abstract
Kansas Underground Storage Tank (UST) Site
Salina, KS
Site Name and Location:
Kansas UST Site
Salina, KS
Period of Site Operation:
Unknown
Operable Unit:
Not applicable
Point of Contact:
Wesley McCall, 913-825-1842
Geophysical Technologies:
Electrical conductivity
Geological Setting:
Low conductivity clays overlying
~sandy water-bearing units —
CERCLIS #
Not applicable
Current Site Activities:
Long-term Monitoring
Technology Demonstrator:
Geoprobe Systems
Salina, KS -• - --
Purpose of Investigation:
Characterize the subsurface stratigraphy and identify structures that could influence groundwater flow patterns
Number of Images/Erofiles-GeneratedDuring Investigation: HLconductivity logs to depthsxanging from 50 to 60 feet
below ground surface
Results: The survey identified a continuous confining clay layer overlying and upper and lower aquifer. A contour of the
contact surface between the upper aquifer and the confining clay layer was generated and a topographic high was identified
that might create a migration pathway for LNAPLs to the northwest, a. direction that is opposite from the generalized
groundwater flow to the east.
65
-------
Kansas Underground Storage Tank (UST) Site
I EXECUTIVE SUMMARY I
The Kansas Underground Storage Tank (KS UST) site is located within the city limits of Salinas, KS
adjacent to an exit off Interstate 70. The site is situated in a light commercial area, with service stations,
motels and restaurants on the adjacent lots. The study area was 160,000 square feet in size and is
situated behind a former service station where the KS Department of Environment and Health suspected
than an unidentified source of groundwater contamination existed.
Groundwater contamination was discovered at the site during a Phase II investigation, conducted in
support of a real estate transaction on a nearby property. Groundwater monitoring in the area established
a plume of petroleum hydrocarbons moving toward the east. However, contamination was detected in
several wells to the north of the suspected source area, as well. The geophysical investigation was
conducted in 1995 as a cost-effective method for characterizing the subsurface stratigraphy.
The site geology consists of a surficial layer of low hydraulic conductivity clays and sands to a depth of
approximately 46 feet below ground surface (bgs). Below the clay and sand layer, subsurface materials
grade into alluvial sands and gravels. Below the alluvial sands lies another clay layer that separates the
upper sand layer from another, deeper, sand layer. The Wellington formation forms the bedrock at the
site and consists of gray and green shales. Groundwater is encountered at approximately 20 to 40 feet
bgs.
The geophysical investigation was carried out using the Geoprobe* Direct Image5 Soil Conductivity
System. This direct push technology does not require a pre-existing borehole to perform the logging
process as the conductivity probe is driven directly into virgin unconsolidated formations. Additionally,
no drill cuttings are generated during the logging process, which significantly reduces waste generation
and potential exposure hazards. Electrical conductivity logs were calibrated by comparing them with
lithologic logs from continuous core samples taken in two locations. Conductivity logs were taken from
10 borings across the study area to depths of 40 to 60 feet below ground surface. The logs indicated the
consistent presence of the surficial clay layer to a depth of approximately eight to 10 feet bgs.
Furthermore, a comparison of the logs indicated that the surficial clay layer had a saddle-like structure
with a ridge trending northward. The investigation concluded that the surficial clay layer acted as a
confining layer, and that petroleum contamination floating on the water table was being forced northward
beneath this ridge by artesian pressure.
The Soil Conductivity System was found to be a cost-effective approach for characterizing the subsurface
stratigraphy. Conductivity logging can provide consistent information on stratigraphy and when accurate
surface elevations are obtained from each boring, a contour map can be developed for any of the
lithologic units that are identified in the survey. This information can be used to identify subsurface
structures that might provide migration pathways for non-aqueous phase liquids, either light or dense.
66
-------
Kansas Underground Storage Tank (UST) Site
SITE INFORMATION I
Identifying Information
Kansas Underground Storage Tank (UST) Site
Salina, Kansas
Investigation Date: August 1995
Background [1,2]
Physical Description: The Kansas Underground Storage Tank (KS UST) site is located in a
commercial area within the city limits of Salinas, Kansas. Located on a small parcel of land
adjacent to the 1-70 exit ramp, the site is surrounded by service stations, motels, and a fast food
restaurant (see Figure 1). The 160,000 square foot study area for the geophysical investigation
was located to the north of Diamond Drive where the Kansas Department of Health and
Environment (KDHE) suspected that an unidentified source of groundwater contamination was
located. The study area lies in flat terrain with little topographical relief. The Saline River lies
to the north at a distance of one mile.
Site Use: The site is the location of a former Amoco service station where past spills of
petroleum products had contaminated the soils and the groundwater. There are other potential
sources, however, located to the north of the site, such as a former truck stop and several ditches
that may have been used to dispose of petroleum products. A motel is presently located on the
site.
Release/Investigation History: Groundwater contamination was discovered during a Phase II
investigation conducted to support the real estate transaction that led to the construction of a
motel on the site of the former service station. The geophysical investigation was conducted in
1995.
Regulatory Context: The KS UST site is managed under the Kansas Underground Storage
Tank Fund, and all compliance requirements are set by that program.
67
-------
Kansas TTndprgrnnnH Storage Tank Site
SITE INFORMATION
Site Logistics/Contacts
State Lead Agency: Kansas Department of
Health and Environment (KDHE)
Federal Oversight Agency: None
KDEH Project Manager:
Scott Lang
Kansas Department of Health and
Environment
2501 Marketplace, Suite D
Salina, KS 67401-7699
785-827-9639
Geophysical Subcontractor:
Wesley McCall
Geoprobe Systems, Inc.
601 N. Broadway
Salina, Kansas 67401
913-825-1842
O iw INCH PVC NoumvG wai
© SCffiEN FOOT 15 SAMPlfR
2\ CONDUCTMir LOG
SOU.COE _mm
sn£
Figure 1: KS UST Study Area with Potentiometric Surface [1]
68
-------
Kansas Underground Storage Tank (UST) Site
i MEDIA AND CONTAMINANTS m^^fma^^mmmmmmmmi^mmm
Matrix Identification [1]
Type of Matrix Sampled and Analyzed: Subsurface clays, sands, and gravels
Site Geology/Stratigraphy [1, 2]
Surface and near-surface soils in the study area consist primarily of low hydraulic conductivity
clays and sands to a depth of 46 feet below ground surface (bgs). Below this level the clays
grade into alluvial sands and gravels overlying another clay layer. This clay layer is continuous
throughout the study area and separates the upper sand layer from another sand layer. Bedrock is
present beneath the lowest sand layer as the Wellington Formation, consisting of grey and green
shales.
The formation from a depth of about 20 to 40 feet bgs grades from clayey-silts to fine sandy silts
to medium-grained sands with depth. There is evidence that this upper aquifer exists under
confined conditions. When wells are screened at depths of 22 to 24 feet bgs, they do not yield
water, yet when screened at, or below, 28 feet bgs the static groundwater level rises to 18 feet
bgs. Generalized groundwater flow is to the east and toward the Saline^River at an average rate
of less than one-half foot per day. A second aquifer exists in the deepest sandy layer overlying
the Wellington Formation and is separated from the upper aquifer by a clay-silt layer over five
feet thick.
Contaminant Characterization [2]
Primary Contaminant Groups: The principal contaminants of concern include benzene,
toluene, ethylbenzene, and xylenes (BTEX). Free product contamination found in some wells
indicated that the contaminants were present as light, non-aqueous phase liquids (LNAPLs).
Matrix Characteristics Affecting Characterization Cost or Performance [1,3]
The ground surface in the study area included grass- and gravel-covered areas which posed no
problems for the probe. In two areas, however, the surface was concrete and holes were bored
through the concrete before beginning the push in these areas.
Consistency in site lithology facilitated the conductivity survey. The site has several
geologically distinct layers whose conductivity values are markedly different. Furthermore, the
layers are relatively continuous throughout the study area.
Excessive soil moisture that might have interfered with conductivity readings was not a problem
at any location in the study area.
69
-------
Kansas Underground Storage Tank (UST) Site
GEOPHYSICAL INVESTIGATION PROCESS •••^^^••^•••S
Investigation Goals [3,4]
The goal of the investigation was to identify any subsurface structures that might influence
groundwater flow directions. The initial investigation found BTEX contamination on the
southern side of Diamond Drive, and generalized groundwater flow to the east. When
contamination was found in monitoring wells to the northwest of the suspected source area, the
question was raised: were there other undiscovered source
areas to the northwest, or were there groundwater flow
dynamics that could result in the northerly migration of
contaminants?
Geophysical Methods f 1,3,5]
The geophysical survey was carried out using the
Geoprobe® Direct Image® Soil Conductivity System,
operated in a Wenner array configuration. In this
configuration, an electrical current is passed through the soil
and the soil conductivity is measured by four electrical
contacts. The conductivity value is a function of grain size,
with finer grains producing higher values and coarser grains
resulting in lower values. The units of measurement for
conductivity are milliSeimens per meter (mS/m). The
Seimen is the inverse of the Ohm, the standard measure for
electrical resistivity.
The Direct Image® system consists of a steel probe running
through four stainless steel rings, as is shown in Figure 2.
The SC200 probe is eight inches long and varies in diameter
from one inch at the tip to 1.125 inches at the base. An
electrical grade plastic insulates the rings from the steel
shaft. A shielded data transmission cable is attached to the
probe by a waterproof seal.
The probe is advanced using a percussion probing machine
which weighs 1,680 pounds and is mounted on the back of a
truck. The percussion machine delivers as much as 18,000
pounds of downward force to the drive end of the probe.
The depth and rate of advancement are measured using a
string pot system. When the probe is retracted, the
percussion machine can exert as much as 25,000 pounds of
retraction force.
.1"(25mm)O.D.
STEEL PROBE ROD
SHIELDED CABLE FOR
SIGNAL TRANSMISSION
ENGINEERING TYPE .
PLASTIC PROViDEa
• MECHANICAL-STRENGTH
AlflJ ELECTiFHliAt""
ISOtSTiON ,•' '"
CONTACT RING
. TAPERED GEOMETRY TO
ENHANCE SOIL CONTACT
Figure 2: SC200 Electrical
Conductivity Probe
70
-------
i Kansas Unden
I GEOPHYSICAL INVESTIGATION PROC1
round Storage Tank fTTST't
The Direct Image® software, running on a PC laptop connected to the instrumentation box,
provides a "real-time" display of the conductivity signal, probe depth, and rate of advancement.
Individual logs can be printed in the field. Data from the logs can be easily output to a
spreadsheet software, or a modeling software, such as SURFER. The Direct Image® software
also includes a calibration routine for the probe that should be run before each push to ensure
that the probe is operating correctly.
Technology Justification •"•• - - •-•• •
The presence of the surficial clay layer overlying the surficial aquifer indicated the need for a
geophysical technology that would not be impeded by this layer. It was necessary to penetrate
this near surface clay layer and delineate the stratigraphy below the clay. The effectiveness of
ground penetrating radar would have been limited by this layer and it was too close to the surface
for a seismic method to be effective. Therefore, Geoprobe's® Direct Image® Soil Conductivity
System was a more effective and less expensive option to identify any subsurface structures that
might influence groundwater flow directions.
I GEOPHYSICAL FINDINGS I
Technology Calibration [1]
A lithology log was developed from a continuous core sample taken to a depth of 40 feet below
ground surface (bgs) at the northwest corner of the study area. Discrete interval samples were
collected in a boring located at the northeast corner of the study area at depths of 46 to 48 feet
bgs and 54 to 56 feet bgs. The deeper samples provided information on the distinct lithologic
units found at those depths. Conductivity logs were taken in the same two locations. The
conductivity values were correlated with the visual identification of discrete lithologic units in
the core samples. The results of the correlation are shown in Table 1.
There is a unique range of conductivity values for each of the lithologic units with the exception
of the units encountered between two and 32 feet bgs. In this interval, there is a silty clay layer
that is somewhat coarser than the clay layers above and below it. The electrical conductivity of
the silty-sandy clay layer, 70 to 140 mS/m, overlaps that of the over- and underlying clay layers
which ranges from 125 to 240 mS/m. Thus, conductivity measures falling between 125 and 140
mS/m are indicative of transition intervals between the low permeability clay layers and the more
permeable silty-sandy clay layer.
Below the second clay layer is a sandy layer, known as the upper aquifer whose upper surface
was encountered at depths ranging from 36 to 46 feet bgs. The upper aquifer ranges in thickness
from 5 to 15 feet. Conductivity values within the upper aquifer ranged from 20 to 40 mS/rn. At
the base of the upper aquifer is a clay layer that acts as an aquitard between the upper and lower
aquifers. The aquitard was found at depths ranging from 46 to 52 feet bgs, and was found to be
approximately 5 to 8 feet thick. Conductivity values in the aquitard range from 80 to 100 mS/m,
making these materials similar in composition to those encountered between 13 and 32 feet bgs.
71
-------
Kansas Underground .Storage Tank fTJST> Site
GEOPHYSICAL FINDINGS!
Table 1: Correlation of Conductivity Values With Discrete Lithologic Units
Conductivity
Range
(mS/m)
Oto75
125 to >240
70 to 140
125 to > 240
20 to 40
80 to 100
20 to 40
Depth
Range
(feet bgs)
Oto2
2 to 13
13 to 32
36 to 46
46 to 52
52 to 60
Location of
Samples
CND02
CND02
CND02
CND02
CND02
CND10
CND10
Generalized Lithologic Description
Organic rich topsoil and gravel fill
Clays, brown with some caliche
development
Silty to fine sandy brown clays
Clays, brown with some caliche
development
Medium to coarse grained granitic sands
with sparse fine gravels, water-saturated
Gray clay-silt
Medium to coarse grained granitic sands
with sparse fine gravels, water-saturated
Source: [1]
Investigation Results
Conductivity logs were collected from 10 borings across the study area to depths ranging from 40
to 60 feet bgs. The locations of the borings can be seen shown in Figure 1, and the logs
themselves are shown in Figures 3a and 3b. The uppermost clay layer is visible in each of the 10
logs as a rise in conductivity to values at or greater then 200 mS/m. The densest portion of this
clay layer, and consequently the highest electrical conductivity value, was found consistently at a
depth of approximately eight to 10 feet bgs. The consistency with which this unit is found in the
logs suggests that it is continuous throughout the study area and has a low surface gradient.
The decline in conductivity values that can be seen in each of the 10 logs between 13 and 32 feet
bgs can be interpreted as the gradual grading of subsurface materials from the upper clay layer
into a coarser silty clay layer. At the base of this silty clay layer, the subsurface materials seem
to grade once again into a tighter clay formation. This graduation can be seen in most of the logs
as conductivity values rise again due to the finer grain size within the clay layer. In the logs
CND01 and CND06, the clay layer is either missing, or its composition is less dense than in other
portions of the study area. In the logs for CND08 and CND09, the clay layer appears to be
present at shallower depths of approximately 20 to 25 feet bgs. The consistency with which this
clay layer seems to be present suggests that it may act as a confining layer to the upper aquifer.
This conclusion seems to be supported by the fact that when wells are screened at a depth of 22
to 24 feet bgs, they yield no water, but when screened at or below 28 feet, the static groundwater
level in the casing rises to 18 feet bgs.
72
-------
GEOPHYSICAL FINDINGS
The sandy gravel layer that constitutes the more transmissive zone of the upper aquifer can be
seen clearly in each of the 10 logs at depths ranging from 36 to 46 feet bgs. As the probe passed
through these materials, there was a marked decrease in conductivity values. At the base of the
upper aquifer, there is a dense layer of clay materials which can be seen in the logs as a sharp
increase in conductivity values at depths ranging from 45 to 55 feet bgs. Most logs were
terminated at this depth to prevent penetration of the aquitard in highly contaminated zones. The
rise in conductivity values as the probe entered this layer is clearly visible, even in the shallower
logs. The consistency with which this unit was encountered in the logs suggests that is laterally
continuous and may act as an aquitard separating the upper and lower aquifers. The sandy gravel
layer that constitutes the lower aquifer can be seen in the logs for CND01, CND02, and CND10.
These borings were the only three pushed to the full 60-foot depth because they were outside of
contaminated areas and approximately bound three corners of the site. Because only three
borings were advanced to this depth, the continuity of the lower aquifer cannot be determined.
A contour map was generated of the contact surface between the sandy layer of the upper aquifer
and the overlying clay layer, and is shown in Figure 4. Surface elevations were taken at each
boring location and the distance from the surface to the top of the sandy layer was calculated.
ECLogCND02
OniS/ia
300mS/m
ECLogCNDOl
OmS/m
10 20 30 40 50
Depth (feet below ground surface)
60
Figure 3a: Electrical Conductivity Log for CND1 - CND05 [1]
73
-------
GEOPHYSICAL FINDINGS!
Kansas Underground Storage Tank HTSTt Site
300 fflS/m
ECLogCNDIO
OmS/m
300mS/m
ECLogCND09
OmS/m
300mS/m
ECLogCNDOS
ECLogCNDO?
ECLogCNDOfi
OmS/m
10 20 30 40 50
Depth (feet below ground surface)
60
Figure 3b: Electrical Conductivity Logs for CND06 - CND10 [1]
Where the transition from the clay to the sand layer was more gradual and less defined, as in the
logs for CND02 and CND05, the depth at which the conductivity value reached 75 mS/m was
taken as the top of the sandy layer. The calculated depths to the top of the sandy layer were
plotted on a site map and the elevations were contoured by hand. In Figure 4, the resulting
contours show that the upper contact surface of the sandy layer has a complex saddle-like
structure, with a topographic high running to the northwest through CND04 and CND06 and
branching northeast toward CND03. Monitoring wells 5, 10, and 11 were highly contaminated
and sporadically contained free product. Combining the information provided by this map and
the finding that the overlying clay layer is continuous throughout the study area, the investigator
concluded that LNAPLs within the groundwater flowing under confined conditions could
migrate along the topographic high in a northwesterly direction. This would make the direction
of the LNAPL migration contrary to the generalized groundwater flow to the east. The
topographic low in this surface at CND09 east of the source also prevented contaminants from
moving east with groundwater flow to well MW08, in which no contamination had been
detected.
74
-------
Kansas Underground Storage Tank (TTST"> Site
GEOPHYSICAL FINDINGS I
Results Validation [4]
No further efforts were undertaken to validate the findings of the geophysical investigation.
NAPL FLOW PATH
166- TOP OF SAND CONTOURS
TWO INCH PVC MONITORING WELL
SCREEN PO!Mr 15 SAMPLER
CONDUCTIVHY LOR
I I SOIL CORE
Figure 4: Contour Map of Upper Sand Layer [1]
mmm LESSONS LEARNED
Some of the lessons learned during this investigation include:
• Electrical conductivity logging is a cost-effective approach for characterizing subsurface
stratigraphy in unconsolidated materials. Alternative approaches to gathering the same
information would use traditional well drilling methods, such as a hollow stem auger.
The daily cost of an auger can be as high as $5,000.
« Conductivity logging can provide consistent information on stratigraphy. A review of
the 10 logs produced at this site shows that the same lithologic units were identified at
consistent depths across the study area. In part, the consistency was due to the similarity
of the materials in the various units across the study area.
• When accurate surface elevations are obtained for each boring, a contour map can be
developed for any of the lithologic units that were identified in the survey. This
information can be used to identify subsurface structure that might provide migration
pathways for non-aqueous phase liquids, either light or dense.
75
-------
Kansas Underground Storage Tank fUSTl Site
GEOPHYSICAL FINDINGS ••••^•••••^•••••••••••Bi
• Two soil boring logs typically are sufficient to produce an unambiguous correlation
between the conductivity results and the observed lithologies. In very heterogeneous
formations, additional location/depth targeted samples may be needed.
76
-------
Kansas Underground Storage Tank (UST) Site
[REFERENCES
1. McCall, Wesley. Electrical Conductivity Logging to Determine Control of Hydrocarbon
Flow Paths in Alluvial Sediments. Geoprobe Systems. December 1995.
2. Geocore Services, Inc. Quarterly Monitoring Report for KDHE UST Trust Fund Site.
March 11, 1998.
3. Personal Communications with Scott Lange, KDHE. August 18, 1998.
4. Personal Communications with Wesley McCall, Geoprobe Systems. September 15,
1998.
5. Fax Communication with Geoprobe Systems. September 15, 1998.
77
-------
PAGE LEFT BLANK INTENTIONALLY
78
-------
Case Study Abstract
Kelly Air Force Base
Kelly Air Force Base
San Antonio, Texas
Site -Name and Location:
Kelly Air Force Base
San Antonio, Texas 78241-5842
Period of Site Operation:
From the 1940s-1980s, Site MP
housed metal plating and degreasing
operations for aircraft
Operable Unit:
Zone 3 Groundwater Operable Unit
SS037
Point of Contact:
Rhonda Hampton
Zone 3 Groundwater Project
Manager
307 Tinker Drive
Kelly Air Force Base, Texas 78241
(210) 925-3100 Extension 226
Geophysical Technologies:
Vertical seismic profile
Geological Setting:
Quaternary alluvium overlying several
hundred feet of Cretaceous limestones,
shales, and clays
CERCLIS #
Not Applicable
Current Site Activities:
Soil organic vapor survey complete,
second phase began in April 1998 which
consisted of (1) a seismic reflection survey
of 318 points, (2) soil samples from 83
geoprobe locations and nine soil borings,
(3) installation of one recovery and three
monitoring wells, and (4) the extraction of
approximately 1,000 gallons of dense
nonaqueous phase liquid (DNAPL)
Technology Demonstrator:
SeisPulse Development Corporation
93 Northridge Terrace #27
Medford, Oregon 97501
541-535-4641
Purpose of Investigation:
To map the bedrock surface and determine the locations of structural highs and lows where DNAPL could collect
Number of Images/Profiles Generated During Investigation: 317 shotpoints along 16 seismic reflection lines
Results: The surface of the Navarro Formation was mapped utilizing data acquired from SeisPulse Development Corporation
"Vertical Reflection profile" (VRP) method of seismic survey and the SeisPulse seismic source. Depressions on the Navarro
Formation were found to contain pooled DNAPL.
79
-------
Kelly Air Force Base
555 EXECUTIVE SUMMARY mm^^^mmmmmm^mammmmmmm^^^
Kelly Air Force Base is situated in metropolitan San Antonio within the encircling route 410 freeway.
The site is located in the eastern portion of Kelly AFB, north of the Tinker Drive and Berman Road
intersection. Surrounded by offices, industrial buildings, and an adjacent Union Pacific Rail yard, the
site is approximately 90,000 square feet and consists of two former buildings (258 and 259) and an
adjacent container storage area. Surface topography can be described as flat with maximum changes in
elevation of 4.17 feet northward across the site.
From the 1940s to the 1980s, Building 258 and 259 housed metal plating and mechanical degreasing
operations. Both buildings were leveled in 1981 and only their foundations remain. Solvents leaked
from the degreasers, especially in the propeller line, which ran the width of Building 258. In December
1997, during an initial phase of well monitoring, a dense nonaqueous phase liquid (DNAPL) was
discovered beneath former Building 258. Based on this discovery, a second phase of field activity began
in April 1998, which included a seismic reflection survey, soil sampling, installation of a recovery well
and monitoring wells, and the removal of approximately 1,000 gallons of DNAPL.
The Kelly AFlBTtudy area is underlain by a thin layer of fill material averaging several feet in thickness.
Fill material may be absent in some surface locations, exposing the lower strata. Beneath the fill material
are Quaternary alluvial deposits ranging in thickness from 30 to 45 feet, and consist of clay, gravel, and
sand.
A seismic reflection survey was conducted in 1998 as part of a larger site investigation. The information
presented in this report was derived from the interpretive report of the geophysical investigation. The
Vertical Reflection Profile (VRP) (seismic) method was used to determine the structural highs and lows
of the confining layer of the shallow aquifer that could provide migration pathways for dense non-
aqueous phase liquids (DNAPLs). The seismic survey included the acquisition of 16 seismic reflection
lines which radiated in a general northwest to northeast pattern across the site. An estimation of each
seismic reflection peak associated with the top of the bedrock was made and the bedrock surface was
identified and mapped. Multiple depressions were identified in cross sections produced with the seismic
reflection data. The DNAPL was located in one of the depressions.
The "near offset" method of acquisition results in a series of discrete vertical reflection data points.
However, obstacles such as buried building foundations act as reflective surfaces and can hinder
reflection interpretation associated with the bedrock. Strong acoustical vibrations, such as rail yard
activity, act as an additional seismic source and can also interfere with reflections. The bedrock surface
was mapped and larger depressions were identified. However, the survey was unable to find small or
narrow channels that could also facilitate DNAPL transport. Seismic surveying should be utilized in
conjunction with other geophysical methods in understanding the subsurface and possibly contributing to
the discovery of DNAPLs.
80
-------
SITE INFORMATION
Identifying Information
Kelly Air Force Base
Kelly Air Force Base (AFB), Site MP
San Antonio, Texas 78241-5842
Building 258 SWMU
CERCLIS #: Not applicable
Investigation Date: 16 April to 19 April, 1998
Background [1, 2]
Physical Description: Kelly Air Force Base (AFB) is located in central Bexar County, Texas
(Figure 1). The base is situated in metropolitan San Antonio within the encircling Route 410
freeway. The site is located in the eastern portion of Kelly AFB, north of the Tinker Drive and
Berman Road intersection, shown in Figure 2. The site is surrounded by offices, industrial
buildings, and an adjacent Union Pacific Rail yard, to the southeast. The site is approximately
90,000 square feet of the approximately 371 acres of Zone 3 and consists of two former buildings
(258 and 259) and an adjacent container storage area. Surface topography can be described as
flat with maximum changes in elevation of 4.17 feet northward across the site.
Site Use: Aircraft operations and maintenance were performed at Kelly AFB from the 1940s to
the 1980s. During this period, Building 258 and 259 housed metal plating and mechanical
degreasing operations. Both buildings were leveled in 1981 and only their foundations remain.
The entire area was then converted to an asphalt parking lot. Currently Kelly AFB is host to
several tenant organizations representing Air Force, Army, and other government organizations.
Kelly AFB is in transition from an Air Force Base to an industrial park. Transfer of ownership to
the city is scheduled for completion July 13, 2001.
Release/Investigation History: Former Buildings 258
and 259 were used for aircraft maintenance and metal
plating. Solvents leaked from the degreasers, especially
in the propeller line, which ran the width of Building 258.
In December 1997, during an initial phase of well
monitoring, a dense nonaqueous phase liquid (DNAPL)
was discovered beneath former Building 258. Based on
this discovery, a second phase of field activity began in
April 1998, which included a seismic reflection survey,
soil sampling, installation of a recovery well and
monitoring wells, and the removal of approximately 1,000
gallons of DNAPL. The asphalt parking surface was
removed from the foundations of former buildings 258
and 259 for inspection, sampling, and, cleaning. At the
conclusion of the investigation, the area was paved again. ,.,. , „.. T ,.
0 ' r s> Figure!: Site Location
81
-------
Kelly Air Force Base
SITE INFORMATION
Figure 2: Site Map [1]
Regulatory Context: RCRA and Texas National Resource Conservation (TNRCC) regulations.
Site Logistics/Contacts
Federal Lead Agency: U.S. Air Force
Federal Oversight Agency: EPA
Remedial Project Manager:
Rhonda Hampton
Zone 3 Groundwater Project Manager
SA-ALC/EMRI
307 Tinker Drive
Kelly Air Force Base, Texas 78241
(210) 925-3100 Extension 226
Geophysical Subcontractor:
SeisPulse Development Corporation
93 Northridge Terrace #27
Medford, Oregon 97501
541-535-4641
Site Contact:
Dr. Yuequn Jin Ph.D.
Senior Hydrogeologist
SAIC
4242 Woodcock Drive
Suite 150
San Antonio, Texas 78228
(210)731-2200
yuequn.jin.@cpmx.saic.com
82
-------
Kelly Air Force Base
I MEDIA AND CONTAMINANTS
Matrix Identification
Type of Geology Investigated: Bedrock surface
Site Geology/Stratigraphy [1]
The Kelly AFB study area is underlain by a thin layer of fill material averaging several feet in
thickness. Fill material may be absent in some surface locations, exposing the lower strata.
Beneath the fill material are Quaternary alluvial deposits ranging in thickness from 30 to 45 feet.
consisting of clay, gravel, and sand (see Figures 3 and 4). The upper sequence of the alluvial
deposits are composed of a black to brown clay ranging in thickness from 15 feet to 28 feet. The
lower sequence of the alluvial deposits are composed of a permeable clayey gravel unit ranging
in thickness from eight to 25 feet. Interbedded within the clayey gravel unit are sand and silt
layers of various thicknesses which tend to laterally grade into gravelly clay and gravel with little
clay.
A shallow groundwater aquifer lies within this permeable zone of clayey gravel at approximately
20 to 25 feet below ground surface (bgs). Groundwater generally flows-eastward, off base, at
0.07 to 3.2 feet/day with a hydraulic gradient of approximately 0.001. Flow is influenced by
compositional variation within the alluvium and channel-like features of interbedded gravel and
clayey gravel formed on the bedrock surface. Beneath the alluvium lies several hundred feet of
Cretaceous limestones, shales, and clays that compose the Navarro Formation, which acts as a
bedrock aquitard. The bedrock separates the near-surface soil from the Edwards Aquifer, which
is San Antonio's sole drinking water source.
Contaminant Characterization [1]
Primary Contaminant Groups: The primary groundwater contaminants at the site are
halogenated aliphatics. Groundwater maximum contaminant levels (MCL) were exceeded for
chlorinated solvent concentrations including tetrachloroethene (PCE), trichloroethene (TCE), cis-
1,2-dichloroethene (cis-l,2-DCE), and vinyl chloride (VC). Inorganic contaminants, including
chromium, nickel, manganese, and arsenic were detected below their MCLs.
Matrix Characteristics Affecting Characterization Cost or Performance [1]
The seismic survey was affected by rail yard seismic noise originating northeast of the site. The
passing trains of the active rail yard caused earth-penetrating acoustical vibrations that acted as
reflective sources, obscuring the actual bedrock reflections. Seismic energy reflects from any
change in density, which can be the bedrock surface or a buried foundation. In addition to the
rail yard seismic noise, buried structures affected the seismic survey by acting as another
reflective boundary. The shallow depth of bedrock did not hinder the survey. The overall
homogeneity of the alluvium and lack of topographical relief facilitated the seismic survey.
83
-------
Kelly Air Force Base
I GEOPHYSICAL INVESTIGATION PROCESS
Investigation Goals [4]
The goal of the geophysical investigation was to map the bedrock surface and determine the
locations of structural highs and lows that could provide migration pathways for DNAPL.
Geophysical Methods [2]
The SeisPulse Development Corporation patented the SeisPulse Seismic Source, which acquires
seismic reflection data utilizing the Vertical Reflection Profile (VRP) method. The VRP method
utilizes near-vertical ray reflection paths to acquire data. Since the 1920s, types of seismic
sources have included vibrating masses, combustible gas mixtures, weight drops, dynamite,
shotguns, and modified rifles. All of these sources produce destructive ground roll and airwave
interference. The SeisPulse seismic source inhibits the destructive interference by producing an
elastic (deformable) wave.
The SeisPulse system is a propane-combustion lightweight portable seismic source. A high-
velocity pressure ridge resulting from the propane-air combustion is directed down an attached
funnel-shaped wave guide. This source wave impacts the earth's surface, resulting in a strong
seismic wave. Energy travels along the surface and into the earth, where it reflects from
subsurface horizons such as limological boundaries or erosional surfaces. The differences
between interval velocities indicate changes in soil or rock properties.
Vertical resultant reflections at the site were received by a Mark Products (Mark 40A) geophone
placed at a one-foot distance from the base of the seismic source. Shot point spacing of 10 feet
between each station was used with the exception of two stations where seismic data was
acquired at five-foot intervals, for a total of 1,606 shot points. The seismic reflection profile was
not affected by the shot point station spacing. Reflection data was recorded on a 12-channel
Geometric S-12 seismograph. Vertical reflection data from the initial source firing is recorded
on channel one, and the next shot is recorded on channel 2 and summed together with the first
shot on channel 12. This process is repeated until data acquisition for that specific station is
complete. The station data is saved to an individual data file, and the source and geophone are
moved to the next station. The data was recorded at a 1/4 millisecond sample interval and a
record length of 512 millisecond.
The seismic source and geophone were operated by one technician while another technician
acquired the data. At each shot point the records where collected, edited, and summed together
to increase the signal to noise ratio. This additive process assumes that all coherent reflective
energy arrives at the same time while noise is random and will not arrive at the same time.
Minimal separation of the source and geophone allows the ground roll and air wave to pass over
the receiving geophone before the arrival of target data. The "near offset" method of placing the
geophone within one foot of the source eliminates the need for large field crews and long cable
layouts, which allows interpretation of survey material within 24 hours.
Technology Justification
84
-------
Kellv Air Force Rase
I GEOPHYSICAL INVESTIGATION PROCESS
The choice of seismic reflection for investigating the bedrock topography at the Kelly AFB was
based on the need for a cost-effective method for mapping depressions in the bedrock which may
represent migration pathways for contaminants. The depth to bedrock was well within the
effective range for seismic reflection, yet not so shallow that the return wave would conflict with
the ground roll wave.
[ GEOPHYSICAL FINDINGS
Technology Calibration [2]
A "walk-away" noise test was conducted to assure that data acquired is a reflection and not
another source of wave energy such as source signatures or ground roll. The test involves
placing a single geophone (receiver), at a one-foot distance from the source, igniting the source,
and recording one trace. The geophone is moved in one-foot increments away from the
stationary source and the test is repeated. After a typical distance of nine to 11 feet, the returning
signals are studied to determine the reflector from the noise background. Ground roll energy will
move at a constant velocity whereas reflector energy does not; this velocity difference is used to
identify the reflector data from background noise. The ground roll or direct surface wave was
identified from the reflector reflection data. The time necessary to complete a walk-away noise
test using the SeisPulse system is approximately 30 minutes. Two walk-away tests were
conducted at the site.
A geologic description from fifteen borings provided the depth to bedrock data, allowing proper
identification and timing of reflection events. Two of the borings were used to complete down-
hole velocity check shot surveys. The velocity surveys were initiated by suspending a two-inch
diameter down hole geophone tool by means of a cable to the deepest accessible depth. A
hammer and plate source were used near the boring and the time required for energy to travel to
the geophone was recorded. The geophone was raised three feet and the procedure was repeated.
The result was a set of one-way travel times, from the surface to various depth, which were used
to determine the interval and average velocity of the overlying alluvium to make a general time-
depth calculation required for the upcoming seismic survey.
The ability to visually connect or "time tie" line intersections with an identifiable reflector such
as the bedrock surface throughout the survey can be an indicator of the seismic survey accuracy
within localized areas. Reflection energy is received by the geophone and recorded as a trace.
Each trace represents a station and each subsurface reflector or event should be visually
identifiable on the trace, and connected to other traces within the survey. Since the bedrock is a
continuous surface, each trace should have a event that marks its boundary and that event can be
time tied to the next trace reflection event. This connection of events makes up the seismic
profile. The bedrock surface reflectors were "time tied" at all line intersections.
85
-------
TCellv Air Force Rase
I GEOPHYSICAL FINDINGS
Investigation Results [1]
The purpose of this investigation was to map the bedrock surface and determine the locations of
structural highs and lows in which DNAPLs could collect. The seismic survey included the
acquisition of 16 seismic reflection lines which radiated in a general northwest to northeast
pattern across the site. An estimation of each seismic reflection peak associated with the top of
the bedrock was made and the bedrock surface was identified and mapped.
Figure 3 shows a geologic cross-section of the typical stratigraphy from northwest to southeast.
The bedrock surface is shown at approximately 35 feet bgs and relatively horizontal. In the
northwest section, the depression which collected DNAPL was shown. The approximate
dimension of the DNAPL pool is approximately 100 feet by 50 feet with a maximum thickness of
seven feet.
Figure 4 shows a southwest to northeast cross-section of the bedrock surface approximately 40 to
45 feet bgs. The bedrock surface varies in nature and areas of highs and lows are evident. Two
depressions are shown separated by a relatively high area. The southwest depression is
approximately 20 feet long and five feet deep. The northeast depression is approximately 30 feet
long and five feet deep. DNAPLs were not discovered in either of the low areas.
Results Validation [1]
On May 30, 1998, three hollow-stem auger soil borings were taken to verify the seismic survey
data used to map the bedrock surface. Two borings reached the bedrock surface at depths which
closely matched the seismic survey reported depths. The third boring encountered a DN.APL
pool at approximately 37 feet bgs and was terminated for health and safety reasons. This boring
verified the existence of depressions in the clay surface as indicated by the seismic survey.
Comparison of lithologic units encountered at 15 wells or soil borings depths to calculated
seismic depths indicate the seismic findings are consistent with actual depths to bedrock. The
seismic depth errors ranged between -2.5 to +2.4 feet, and resulted from the use of a constant
wave velocity. Interval velocities may have varied from station to station due to near-surface
differences within the alluvium. The wells and soil boring locations were chosen based on their
close proximity to the seismic lines.
86
-------
Kelly Air Force Base
l LESSONS LEARNED
IBB ft
3 FIU.
3 ClAY
= SANO. CLAYEY GRAVEC
J AND GRAVEL
i SOIL BORING
i SOUD WALL CASING
H ^HIGH WATER LEVEL
L ^-LOWWATER LEVEL
E PRODUCT AND ELEVATION
Figure 3: Northwest to Southeast Cross-Section, in Feet Above Mean Sea Level [1]
Figure 4: Southwest to Northeast Cross-Section, in Feet Above Mean Sea Level [1]
Lessons learned at the Kelly AFB include the following:
87
-------
I GEOPHYSICAL FINDINGS
Kellv Air Fnrrp Base
The "near offset" method of acquisition results in a series of discrete vertical reflection
data points. However, obstacles such as buried building foundations act as reflective
surfaces and can hinder reflection interpretation associated with the target bedrock.
Strong acoustical vibrations such as rail yard activity act as an additional seismic source
and can also interfere with bedrock reflections.
The varying nature of DNAPLs suggest that they can follow cracks, offsets, and smaller
scale features found on the boundary surface. At the site, the seismioreflection survey
was implemented to define areas on the bedrock surface in which DNAPLs could collect
and potentially migrate. The bedrock surface was mapped and larger depressions were
identified. However, because of the error rates, the survey was unable to find small or
narrow channels that would facilitate DNAPL transport. Seismic surveying should be
utilized in conjunction with other geophysical methods in understanding the subsurface
when investigating to find DNAPLs.
-------
Kelly Air Force Base
[REFERENCES]
San Antonio Air Logistics Center, Environmental Management Restoration Operations.
Current Conditions Report for the Building 258 Solid Waste Management Unit, Kelly
AFB, Texas. Science Applications International Corporation. July 1998.
Science Applications International Corporation. Seismic Reflection Geophysical Survey
Report, Kelly AJFB, Texas. SeisPulse Development Corporation. May, 1998.
Personal communication with Mike King, President of SeisPulse Development
Corporation, Vancouver, Washington. September 4, 1998.
Personal communication with Rhonda Hampton. Zone 3 Groundwater Project Manager.
Kelly AFB, Texas. September 17, 1998.
89
-------
PAGE LEFT INTENTIONALLY BLANK
90
-------
Marshalltown FMGP
Case Study Abstract
MarshaUtown Former Manufactured Gas Plant (FMGP) Site
Marshalltown, IA
Site Name and Location:
Marshalltown FMGP
Period of Site Operation:
1870's to 1946-gas manufacturing
and electric generation
Operable Unit:
Not Applicable
Geophysical Technologies:
Electrical conductivity
CERCLIS #
Not Applicable
Current Site Activities:
Currently 100 employee.? still work on-
site. Groundwater monitoring is being
performed.
Point of Contact:
Albert Bevolo Ph.D.
Ames Laboratory
Iowa State University
125SpeddingHall
Ames, IA 50011
(515)294-5414
Geological Setting:
Pleistocene glacial till, glaciolacustrine
deposits, fluvial deposits, and loess lie
unconformably over discontinuous
layers of limestone and dolomite.
Technology Demonstrator:
Geoprobe Systems
Corporate Headquarters
601 N. Broadway
Salina,KS 67401
1-800-GEOPROBE
Purpose of Investigation:
The geophysical investigation was undertaken as part of a demonstration under the Department of Energy's Expedited Site
Characterization program. The goal of the demonstration was to compare the capability of electrical conductivity surveys
with more traditional methods for characterizing subsurface stratigraphy, such as borehole geophysical logging and cone
penetrometer testing (CPT).
Goals related to the soil conductivity probe (SCP) were to confirm and further refine the site geologic and contamination
conceptual models as defined through Phase I activities. Another goal was to define the topography of the lower cohesive
unit (LCU) in order to find low-points where dense non-aqueous phase liquids (DNAPLs) might collect.
Number of Images/Profiles Generated During Investigation:
27 conductivity profiles produced from 700 feet of log in 27 holes
Results:
SCP provided high vertical resolution data from which transitions between high conductivity clay and low conductivity sands
could be readily identified. Calibration with soil borings and CPT showed that the main stratigraphic units were readily
distinguishable. The SCPs provided clear information on stratigraphic transition depths which were readily integrated with
data from CPT and boring logs.
A secondary and unexpected result was the apparent response of the SCPs to DNAPL-saturated soils by distinct decreases in
conductivities.
91
-------
Marshalltown FMGP
wSSm EXECUTIVE SUMMARY «^^MHHBHH^^MMHHHHBH^^HBI
The Marshalltown Former Manufactured Gas Plant (FMGP) site is located in an old industrial area in
Marshalltown, Iowa. Gasification by-products of the manufacturing process, including coal tar, coke,
and other materials, were stored on-site in unlined pits. A site investigation found polycyclic aromatic
hydrocarbons (PAHs) in soil samples at levels substantially above background levels. Another
investigation, conducted during an underground storage tank (UST) removal, showed the presence of
petroleum hydrocarbons in excess of applicable action levels in soil and groundwater.
The site is situated on the edge of the flood plain of Linn Creek where the ground surface is flat to gently
sloping. Near surface soils consist of a wide range of fill materials of low plasticity and varying in
thickness from 0.5 to 14 feet. This is underlain by fine-grained cohesive soils consisting of low plasticity
silty clay with interbedded sandy and gravelly clays, ranging in thickness from 6 to 14 feet. Limestone
bedrock is approximately 50 feet below the ground surface. A steep ridge in the bedrock surface, with
about 25 feet of relief, trends northwest-southeast across the site.
The Marshalltown site was used a demonstration site for the comparison of various technologies used in
the site characterization process. The focus of this case study is an assessment of the performance of a
soil conductivity probe (SCP) used to delineate soil stratigraphy and its utility in the Expedited Site
Characterization Process. The information in this report was derived from the interpretive report of the
geophysical investigation. A site-specific goal for the SCP was to define the topography of the lower
cohesive unit (LCU) to identify low-points where dense non-aqueous phase liquids (DNAPLs) might
accumulate. The probe determines soil conductivity by measuring the electric potential across electrodes
in direct contact with the soil.
Results of the investigation revealed that the SCP provided very useful and reliable stratigraphic data.
The upper cohesive unit (UCU) and the LCU contact was inferred by a distinct rise in the conductivity
values. Both of these contacts could easily be identified on most of the soil conductivity logs. A
secondary and unexpected result was the apparent response of the SCP to DNAPL-saturated soils,
measured by decreases in conductivities.
Since this was a demonstration of the SCP when the product was first developed, no direct costs were
associated with this investigation. However, based on current models of equipment and prices associated
with them an estimated cost of $7,875 can be associated with an investigation of this type.
The capabilities of the SCP were proven. The probe was very versatile in that it could maneuver into
small spaces and could penetrate most soil subsurface materials. The SCP was not able to clearly
identify weathered bedrock and probes would break when unexpected bedrock or larger sized gravel and
cobbles were encountered. The probe was also operationally efficient and could be operated by a single
person if necessary. The information collected by the probe can also be used to enhance the site
contamination model. The probe has the ability to provide much more detailed stratigraphic information
than conventional auger borings. This is very important when considering the fate and transport of
contaminants. The SCP detected DNAPL-saturated soil as a distinct decrease in conductivity.
92
-------
Marshalltown FMGP
I SITE INFORMATION
Identifying Information
Marshalltown FMGP Site
Marshalltown, Iowa 50158
Operable Unit: --N/A
CERCLIS#: --N/A
Background [1]
Physical Description: The Marshalltown Former Manufactured Gas Plant (FMGP) is located in
in an old industrial area of Marshalltown, Iowa (Figure 1). The site contains several buildings
from the FMGP and former electric plant and is approximately 2.5 acres in size (Figure 2). The
nearest residential properties are located several hundred feet to the north. The site is located on
the edge of the floodplain of Linn Creek which flows west to east approximately 800 feet south
of the site and discharges into the Iowa River roughly 2.5 miles northeast of the site. Site
topography is flat to gently sloping, with approximately 10 feet of relief across the site.
Site Use: A manufactured gas plant operated at the site from the 1870's-until 1946. When the
site first opened in the 1870's, under the name of Marshalltown Gas Light Company, gas
manufacturing was accomplished by coal carbonization. Electrical generation began at the site
between 1888 and 1892. In 1892 the Marshalltown Gas Company, the Marshalltown Electric
Company, and the Marshalltown Street Railway Company consolidated and became the
Marshalltown Light, Power, and Railway Company, which brought the electrical and gas
operations under common ownership. Between 1910 and 1921, the gas manufacturing process
was converted from coal carbonization to carbureted water gas and the ownership was transferred
to the Iowa Railway and Light Corporation. Plant operations continued until 1946.
The FMGP site is currently used as the service and materials distribution center for Alliant
Utilities gas and electric operations. The site is currently owned by IES Utilities who merged
with other mid-western power generation/distribution
companies and is now known as Alliant Utilities.
There are 100 employees that work on-site, and the
plant is scheduled to close within the next two years so
remediation can begin [2]. ._*"""- "v v
% Marshalltown «
Figure 1: Site Location
93
-------
Marshalltown FMGP
X.
SITE INFORMATION
» a i-» * * * - •-
isn i!'
• ea&.
i
" "il 1 : |Mg .„ '-' j
gl-fl-rr^, -/
i *-«-r :
K j
= J± f !? i«
IS I IJ; >
ISJ1^ IB!
:K
jLgl 1| iffjVX .'"v... ../g? \M i|] !
i: ,-^» ssg'rf—^nru. igj
* j-! ^..«4C-^-- .Bi ,^ --JSU?i!
aso'oc— •—i - 4 - •* I ; . * i
«•—»•"•«» i^a *— —_jCjl: '. - - * :
II- |g: i i I I
^\ — \\ \\i
-, .
\\\ f Jiy--?<3^-
\\\ I /Ts-kT.yi '-:•-.
L1_^
H i lii
: J4«
i • " : 1J
M k-n|
i.' • \! v
/ J , • U;^X
/ \i .- i hrir -
\ "r^r^rr^N
es
O
J?
"es
3
a
o
o
S
08
o
C3
S
-------
Marshalltown FMGP
SITE INFORMATION
Release/Investigation History: Major by-products of the gas manufacturing process are coke,
tar, ash, and spent purifiers. The final disposition of these by-products on the site is unknown,
but a substantial portion of the tar may have been disposed on site. The investigation of waste
product disposal at the site began with a preliminary study conducted in 1986. In 1987, soil
samples collected in a follow-up site investigation revealed the presence of polycyclic aromatic
hydrocarbons (PAHs) compounds at levels substantially above background levels. In November,
1988, an underground storage tank (UST) was removed from an area near the west end of the
east wall of the former spray pond (Figure 2). Petroleum hydrocarbons in excess of applicable
action levels were detected during this removal action. In 1990, a detailed remedial investigation
was begun to address requirements under a 1989 Consent Order between the Iowa Department of
Natural Resources (DDNR) and Alliant Utilities. The investigation identified a tar pit, two
different tar separators, and a tar well as potential contaminant sources. A comprehensive soil
and groundwater sampling program was included as part of this investigation. Visible coal tar or
fuel contamination was found in the soil sampled from several borings at the site. The site
activities that are the focus of this case study include the second phase of the site investigation
process and the evaluation of remedial alternatives.
Regulatory Context: The EDNR is the lead agency coordinating all the activities of the
Marshalltown FMGP site. The site is being addressed as a result of a 1989 Consent Order
entered into between the DDNR and Alliant Utilities. The EDNR is utilizing CERCLA
requirements and guidances [3].
Site Logistics/Contacts
State Lead Agency:
Iowa Department of Natural Resources
Project Manager:
Dr. Johanshir Golchin
Iowa Department of Natural Resources
Wallace Building-502 E. 9th Street
Des Moines, IA 50319-0034
(515)281-8925
Site Contact:
Albert Bevolo P.h.D
Ames Laboratory
Iowa State University
125 Spedding Hall
Ames, IA 50011
(515)294-5414
Geophysical Subcontractor:
Geoprobe Systems
601 N.Broadway
Salina,KS 67401
1-800-GEOPROBE
PRP Contact:
Dean Hargens
Alliant Utilities
P.O. Box 351
Cedar Rapids, IA 52406
1-800-822-4348
95
-------
Marshalltown FMGP
I MEDIA AND CONTAMINANTS mm
Matrix Identification [1]
Type of Matrix Sampled and Analyzed: Subsurface soil and occasional weathered rock
Site Geology/Stratigraphy [1]
The overall stratigraphy of the glacial drift in the area of the site consists of loess over Kansan
till, which overlies Mississippian-age limestone and Pennsylvanian shale bedrock. Near-surface
soils include a wide range of fill materials (clay, gravel, sand, cinder, and other debris) of low
plasticity ranging in thickness from 0.5 to 14 feet. This is underlain by fine-grained cohesive
soils consisting of low plasticity silty clay with interbedded sandy and gravelly clays, ranging in
thickness from 6 to 14 feet. This layer is also known as the upper cohesive unit (UCU). This is
followed by a granular unit comprised mostly of various types of sand. A layer of low plasticity
clayey lacustrine soil and low to high plasticity glacial till separates the granular unit from
bedrock in most areas of the site. This layer is commonly referred to as the lower cohesive unit
(LCU). Depth to bedrock ranges from 20 feet below the ground surface (bgs) in the northern part
of the site to 40 feet bgs in the southwestern portion. According to drilling information, a steep
ridge in the bedrock surface with about 25 feet of relief trends northwest-southeast across the
site.
Depth to groundwater at the site averages between 18 to 20 feet bgs. Hydraulic conductivity
measurements indicate values that range from 0.0029 to 0.00076 cm/sec for the granular soils.
Groundwater in the alluvial sediments tends to flow in a southern direction toward Linn Creek.
Bedrock groundwater flow characteristics are not well established and appear to be strongly
influenced by the activity of production wells in the area that tap the Mississippian aquifer. The
limestone bedrock is part of the Mississippian Burlington and Gilmore City Formations and are
part of the regional Mississippian aquifer.
Contaminant Characterization [1]
Primary Contaminant Groups: The primary contaminants of concern at the site include the
following: benzene, toluene, ethylbenzene, xylene, phenols, and PAHs, such as naphthalene and
phenanthrene. Some contaminants are known to be present as dense non-aqueous phase liquids
(DNAPLs).
Matrix Characteristics Affecting Characterization Cost or Performance [1,4]
Due to surface obstructions encountered on-site, such as buildings, sheds, etc., explorations could
not be conducted at some locations. Unexpected cobbles, larger sized rocks, boulders, or
bedrock were encountered and posed a problem for the probes. Several probes were broken
when these units were encountered.
Factors such as non-uniform infiltration of highly saline solutions from winter road salting
operations, poor ground-to-probe contacts at shallow depths, and the diverse nature of the
surficial fill contributed to the erratic conductivity data gathered within the top six feet from the
96
-------
Marshalltown FMGP
[MEDIA AND CONTAMINANTS)
surface. Irregularity in the conductivity trace can also be attributed to thinly interbedded seams
of silts, sands, clays, and gravels up to depths of approximately 15 feet bgs. Weathered rock did
not have a distinct conductivity signature, and could not be distinguished from soils on the basis
of its conductivity.
I GEOPHYSICAL INVESTIGATION PROCESS
Investigation Goals [1,4]
The geophysical investigation was undertaken as part of a demonstration under the Department
of Energy's Expedited Site Characterization program. The goal of the demonstration was to
compare the capability of electrical conductivity surveys with more traditional methods for
characterizing subsurface stratigraphy, such as borehole geophysical logging and cone
penetrometer testing (CPT).
Goals related to the soil conductivity probe (SCP) were to confirm and further refine the sue
geologic and contamination conceptual models as defined through Phase I activities. Another
goal was to define the topography of the LCU in order to find low points where DNAPLs might
collect.
Geophysical Methods [1, 5,6]
The focus of this case study is an assessment of the performance of the Geoprobe® SC100
conductivity probe (SCP) in delineating soil stratigraphy. The determines soil conductivity by
measuring the electric potential across electrodes, which are in direct contact with the soil.
Electrical conductivity varies with soil type, with clays exhibiting higher
conductivities than silts, and sands and gravels having the lowest
conductivity. The probe has a vertical resolution of 0.05 feet with a data
rate of 20 samples per second, and a maximum depth range of 80 feet bgs.
The SCP system is small and very maneuverable.
//'
The conductivity probe, shown in Figure 3, consists of a steel shaft
running through the center of four stainless steel contact rings. An
engineering grade plastic electrically isolates the rings and the shaft from
each other. The probe was operated in the Wenner array configuration
which reacts linearly to variations in formation conductivity and yields
good vertical resolution by using all four electrodes. The probe is
approximately eight inches long and has a diameter that tapers from 1-1/8
inch at the top to 1 inch at the point. The taper assures a firm
ground-to-probe contact. The probe assembly threads directly to standard
Geoprobe® probe rods. A signal cable is threaded through the inside of
the rod string and into a PC-based data acquisition system housed in a
ruggedized case. Depth measurements are obtained by a stringpot system
_i
.? .._ _L
Figure 3:
Conductivity
Probe [3]
97
-------
Marshalltown FMGP
I GEOPHYSICAL INVESTIGATION PROCESS
configured to measure the distance from the driving mechanism to the ground surface. The
stringpot signal is used both to determine probe depth and speed at which the probe is
advancing.
Technology Justification
The reasons that soil conductivity was selected for the investigation at the Marshalltown site
were not site-specific, but were related to the investigation as a demonstration for the expedited
site characterization process. However, due to the many cultural interferences at the site such as
chain-link fences, stacks of steel piping, steel storage sheds, and vehicle traffic, other
geophysical methods such as ground penetrating radar and electromagnetic offset logging could
not provide useful results [4].
I GEOPHYSICAL FINDINGS
Technology Calibration [1]
In order to calibrate the SCP with local soils at the site, 123 sample soil cores (which resulted in
127 soil samples) were collected from 27 geoprobe locations at specific depths for standard core
logging and visual soil classification (Figure 4). These core samples were two feet long and
were collected from six subsurface zones at specified depths using another Geoprobe® system.
The units of measurement for conductivity are milliSeimens per meter (mS/m). The Seimen is
the inverse of the Ohm, the standard measure for electrical resistivity. Direct calibration of the
SCP with soils collected from the site revealed the following comparisons in conductivity values:
Clayey soils: 60 - 140 mS/m; sandy soils: 30 - 40 mS/m; gravels: 20 - 35 mS/m (Table 1 and
Table 2).
The SCP was advanced adjacent to two existing borings for calibration: MW-3A and B-8.
Distances between the borings and probes were 16 and 13 feet respectively. The SCP was also
calibrated against CPTs. CPT is a reliable direct-push geotechnical method of characterizing
soils on the basis of its physical resistance to penetration. The conductivity profiles were
compared directly to the stratigraphic logs for these borings and with the CPT results.
98
-------
Marshalltown FMGP
I GEOPHYSICAL FINDINGS
640 h
620
600
580
560
540
520
500
480
460
440
420
400
380
360
M146 M-69
ML68
ML66
MW3
«W2
ML58
Ml?
ML55
ML35
ML36
ML34
MI. ML59
M140
HL53
ML63
ML2S'
ML29
ML31
400 450 500 550 600 650 700 750 800
Figure 4: Soil Conductivity Probe Push Locations [1]
Investigation Results [1, 4]
850 900 950 1000 1050
In the Spring of 1994, Geoprobe obtained 27 logs at the Marshalltown FMGP site, using the
Geoprobe® 4200 to push the SCI00 conductivity probe into the subsurface. Each push was
halted upon a confident identification of the lower cohesive unit (i.e., rapid rise in soil
conductivity), in locations where the lower cohesive unit was absent, or upon probe refusal.
Push depths ranged from 14 feet bgs to 40 feet bgs. The logs were then used to determine the
confining bed of clay where contaminants might be found. These logs were merged into 3D
models and used to determine soil sampling depths and locations.
The conductivity logging system produced 700 feet of log in 27 holes over a period of 5 working
days. The system was operated by a two-man crew. Operation by a one-man crew is possible,
although productivity would be significantly lower. The data required minimal post-processing
(deletion of negative or repeat values). Digital conductivity and probing speed data and field
printouts were provided at the end of each work day for integration into the existing site model.
99
-------
Marshalltown FMGP
I GEOPHYSICAL FINDINGS
Twenty-seven SCP penetrations were combined with data from soil borings, Geoprobe® core
samples, and CPT penetrations. Locations of the various penetrations are shown in Figure 4. In
most conductivity logs, readings from three to five feet bgs exhibited erratic conductivity values.
Explanations offered for this phenomenon included infiltration of extremely saline solutions
resulting from salting of roads during winter weather and poor ground-to-probe contact at
shallow depths. The UCU interface and the LCU interface were easily identified in most
locations of the site. The UCU contact with the granular unit could be inferred by the distinct
drop in soil conductivity between 14 and 17 feet bgs. The LCU contact with the granular unit
above it was identified by an increase in soil conductivity between 30 and 32 feet bgs. Figure 5
shows cross section A-A' showing SCP conductivity profiles and stratigraphic zones as
determined by soil boring logs and CPT data. The conductivities of the units and conductivity
changes across the transitions in ML-28 and ML-45 are shown in Table 1 and Table 2.
A secondary and unexpected result was the apparent response of the SCP to DNAPL-saturated
soils by decreases in conductivities. At the base of the granular unit above the LCU in ML-45,
soil conductivity shows a 20 mS/m drop from about 50 mS/m to about 30 mS/m. When depths
are adjusted for small stratigraphic variations between sites, the depth of conductivity decrease
corresponds closely to a known zone of DNAPL, as seen in the boring logs for B-8.
100
-------
Marshalltown FMGP
I GEOPHYSICAL FINDINGS
eft
CC
LJJ
CO
NQliVA313
o
o>
cc
en
ce
O
U
C3
.Sf
*3
cs
I
6JO
E
101
-------
Marshalltown FMGP
I GEOPHYSICAL FINDINGS
Results Validation [1, 2]
In addition to comparison with the calibration boring logs and to the discrete soil sample cores,
the SCP was also compared to laser-induced fluorescence (LIF) logs that were run in conjunction
with the CPT logs. LIF is an innovative method of near-continuous screening for contaminants,
in this case PAHs.
The 20 mS/m drop in conductivity at the base of the granular unit above LCU in ML-45
corresponds closely with a high LEF reading indicative of the presence of DNAPLs. There is also
a good correlation between high LIF and low conductivities at the contact between the UCU and
the top of the granular unit at around 20 feet bgs. The conductivity shows a decrease of about 20
mS/m compared to the more stable value of about 50 mS/m through the relatively
uncontaminated section of the granular unit.
Conductivity dips recorded within the granular zone of all profiles at the site were compared with
the Geoprobe® soil core logs, with LIF profiles, and with boring logs. The results of this
analysis showed that conductivity dips within the granular zone could be attributed to DNAPL
contamination 75 percent of the time and to uncontaminated gravel 25 percent of the time.
The SCP filled in absent stratigraphic information between two locations with known
stratigraphic profiles. The major stratigraphic unit contacts were very noticeable on the soil
conductivity logs and the data was used to generate a database for a three-dimensional site
stratigraphic model. When sections from this model were compared to the nearby soil boring
logs, a correlation of stratigraphic units within one to two feet was revealed. This was attributed
to the difference in stratigraphy over the relative distance between the SCP locations and the soil
boring locations, and to the use of different sampling technologies. The investigators believed
that the SCP depths were more reliable for sampling because it involved same technology, i.e.
Geoprobe direct push depth indicators. Another demonstration at this site used a more invasive
Geoprobe® unit to collect large bore soil samples at the identified locations and also confirmed
the information in the SCP conductivity logs.
The CPT and conductivity soil profiles show significantly more detail than the boring logs and
reveal that the CPT and conductivity soil profiles were in close agreement with the boring logs
(Table 1 and Table 2). Depths to stratigraphic contacts between ML-28 CPT and ML-28 SCP
are similar, but differ from the soil boring log MW-3A by about two feet (possibly due to
differences in stratigraphy as a result of distance between locations). The ML-45 SCP
corresponds well with the CPT and soil boring log. Overall, the side-by-side comparisons of the
SCPs and the soil boring logs indicated stratigraphic correspondence of the unit contacts to
within about one to two feet.
102
-------
Marshalltown FMGP
[ GEOPHYSICAL FINDINGS
Table 1: Comparison of Soil Stratigraphy Results for ML-28 SCP and CPT, and MW-3A
Stratigraphy
ucu
Transition depth for UCU
Granular Unit
Transition depth for
Granular Unit
Conductivity
mS/m
130-190
140->70
20-80
50->90
ML-28 SCP
Depth
(Feet bgs)
16-17
30
ML-28 CPT
Depth
(Feet bgs)
17
29-30
MW-3A Log
Depth
(Feet bgs)
17
32
Source: [1]
Table 2: Comparison of Soil Stratigraphy Results for ML-45, ML-60, and B-8
Stratigraphy
UCU
Transition depth for UCU
Granular Unit
Transition depth for
Granular Unit
LCU
Conductivity
mS/m
90-140
90->30
20-80
25->85
70-90
ML-45 SCP
Depth
(Feet bgs)
17-18
35-36
ML-60 CPT
Depth
(Feet bgs)
nd
34
B-8 Log
Depth
(Feet bgs)
19
36
nd=no data
Source: [1]
103
-------
Marshalltown FMGP
I LESSONS LEARNED
Lessons learned for the Marshalltown site include the following:
• Geophysical survey techniques are an important part of the expedited site
characterization process, however not all techniques are appropriate for all sites.
Marshalltown had complex stratigraphic conditions that led to significant error and
uncertainty in the some of the geophysical survey results. Therefore, potential
limitations of each geophysical method must be carefully considered on a site specific
basis.
• The stratigraphic correlations between the push technologies (SCP and CPT) and the
borehole log data demonstrated that the contacts between soil units can generally be
interpreted from the CPT and soil conductivity logs with confidence. Correlations were
generally within one to two feet, but this variance was attributed to the distance between
SCP locations and the soil boring locations.
• The SCP was more maneuverable and more versatile than the CPT and could penetrate
most soil subsurface materials. The probe was also operationally much more efficient
than the CPT and could be operated by a single person if necessary. SCP provides
reliable high-resolution demarcation between high conductivity clays and silts and low
conductivity sands and gravels. With proper calibration from soil borings, the SCP can
provide reliable infill information between boring logs and can be used to enhance the
site conceptual model. The probe provides more detailed stratigraphic information than
conventional auger borings.
The SCP appears to respond to DNAPL-saturated soil by exhibiting a distinct
conductivity dip. Since other factors, most notably sand or gravel lenses, can cause
conductivity dips, the SCP cannot by itself detect DNAPLs. However, dips in
conductivity in generally low conductivity (sandy permeable) layers immediately above
high-conductivity (clayey low-permeability) zones would certainly be a target to
investigate pooling of DNAPLs. This would be especially true if the stratigraphic
transition occurred at lower elevations compared to surrounding logs. Close inspection
of SCP results can provide a good screening tool for the location of accumulated
DNAPLs.
• Twenty-seven SCP penetrations were combined with data from soil borings, Geoprobe®
core samples, and CPT penetrations. Locations of the various penetrations are shown in
Figure 4. The SCP provided high vertical resolution data from which transitions
between high conductivity clay and low conductivity sands could be readily identified.
Comparison with soil borings and CPT logs showed that the main stratigraphic units
were readily distinguishable and that the transition depths agreed among the three
methods.
104
-------
Marshalltown FMGP
I LESSONS LEARNED
The SCP was not able to clearly identify weathered bedrock. The probes would break
when unexpected bedrock or large gravel/cobbles were encountered. This could be
remedied by using more indestructible equipment at developed sites where
heterogeneous fill layers would be expected or are already known to exist.
105
-------
Marshalltown FMGP
•REFERENCES
1.
Ames Laboratory, Ames Expedited Site Characterization Demonstration at the Former
Manufactured Gas Plant Site, Marshalltown, Iowa. Ames Laboratory, Ames Iowa.
1996.
2. Personal Communications with Dean Hargens, Alliant Utilities. September 10, 1998.
3. Personal Communications with Johanshir Golchin, Iowa Department of Environmental
Resources. September 17, 1998.
4. Bevolo, A., Kjartanson, B., Stenback, G., and Wonder, J., Site Characterization,
Expedited. Ames Laboratory, Ames Iowa. 1997
5. Christy, C., Christy, T., and Wittig, V. A Percussion Probing Tool for the Direct
Sensing of Soil Conductivity, Technical Paper. Geoprobe Systems, Salina Kansas.
March 1994.
6. Personal Communications with Albert Bevolo, Ames Laboratory. August 24, 1998.
7. Fax Communication with Geoprobe Systems. September 15,1998.
106
-------
New Hampshire Plating Company
Case Study Abstract
New Hampshire Plating Co.
Merrimack, NH
Site Name and Location:
New Hampshire Plating Co.
Merrimack, NH
Period of Site Operation:
1962 to 1985
Operable Unit:
N/A
Geophysical Technologies:
Marine seismic-reflection surveys
Ground-penetrating radar
Natural gamma
EM borehole logs
CERCLIS*
NHD0010091453
Current Site Activities:
EPA issued a proposed cleanup plan to
the public in January 1998. The EPA
plans to conduct a treatability study in
Summer or Fall, 1999.
In addition, on-going groundwater
monitoring is being conducted.
Point of Contact-
Thomas Mack
U.S. Geological Survey
603-226-7805
Geological Setting:
Alluvial terrace underlain by
glaciolacustrine sediments and till. The
underlying bedrock consists of schists
and phyllites with minor amounts of
granite and gneiss.
Technology Demonstrator:
United States Geological Survey
Purpose of Investigation:
To use geophysical methods to identify contamination from the New Hampshire Plating Company and to determine
underlying lithology. The geophysical methods used were marine-seismic reflection, ground-penetrating radar, and natural-
gamma radiation and electromagnetic-induction borehole logging.
Number of Images/Profiles Generated During Investigation:
Seismic-Reflection Profiles: 3 lines
Ground-penetrating Radar Profiles: 4 lines
Borehole Geophysical Logs: 7
Results: The natural gamma used in combination with the EM logs identified a probable plume of groundwater
contamination from the electroplating facility. The contamination is moving toward Horseshoe Pond and Merrimack River.
107
-------
New Hampshire Plating Company
EXECUTIVE SUMMARY i
The New Hampshire Plating Company Site is located on 13.1 acres in Merrimack, New Hampshire. On
the property are a former plating facility, a large pond, and four lagoons. The Merrimack River is located
about 600 feet (ft) east of the plating facility and the northern end of Horseshoe Pond is located about
500 ft south of the plating facility. From 1962 until 1965 the property was used for electroplating
activities. The contamination was a result of the facility discharging 35,000 to 60,000 gallons per day of
electroplating wastes into the four unlined lagoons. The wastes consisted of cyanide plating baths and
sludges, acids and chlorinated solvents.
The site geology is comprised of an alluvial terrace consisting of sand, silt and some gravel. Under the
alluvial terrace is glaciolacustrine sediments and till which consists primarily of sand, silt, clay and some
gravel. Under the glaciolacustrine sediments and till is bedrock consisting of schist and phyllites with
minor amounts of granite and gneiss. The water table is encountered at depths ranging from five to 20 ft
in the study area.
A geophysical investigation was conducted at the site as part of a larger effort to delineate site conditions
and the scope of contamination. The information presented in this report was derived from the
interpretive report of the geophysical investigation. The geophysical investigation used four different
technologies. Continuous seismic reflection and ground penetrating radar surveys were performed to
characterize the site geology and locate bedrock structures. The seismic surveys were conducted near the
shore of Horseshoe Pond and along the eastern side and the middle of the Merrimack River. The profiles
conducted in Horseshoe Pond indicated that the bedrock ranged from the water surface to less than 20 ft
below (the survey could only identify to that depth at this location-this may not be true elsewhere). The
profiles conducted in the Merrimack River indicated that bedrock ranged from 10 to 50 ft below the
water surface. The"ground penetrating radar (GPR) method was used to produce geophysical profiles of
the land area around the facility and the lagoons. The GPR profiles successfully determined soil types to
a depth of 30-35 ft.
Natural gamma logs were used to delineate the stratigraphy of sub-surface materials in eight monitoring
wells, and electromagnetic induction (EM) logs were used to identify zones of conductive groundwater
that would indicate the presence of contaminated groundwater. The gamma logs correlated well with
existing lithologic logs for the wells. The EM logs showed significant spikes, indicating possible zones
of contamination. The depths at which the spikes occurred correlated well with measures of specific
conductance of groundwater taken as part of the on-going monitoring program in those wells.
The seismic and radar surveys were moderately successful, but some difficulties were encountered due to
the presence of fine-grained sediments in the bottom of the pond and in the soils around the facility.
Fine-grained sediments limited the penetration of the radar signals, resulting in blank areas on the
profiles. The GPR profiles conducted on water bodies were inconclusive because the signal was
attenuated by the water column and was unable to penetrate beneath the bottom sediments. The gamma
and EM logs were very successful in characterizing the stratigraphy and identifying zones of highly
conductive groundwater that may indicate contaminated groundwater.
108
-------
New Hampshire Plating Company
SITE INFORMATION
Identifying Information
New Hampshire Plating Company
Wright Avenue
Merrimack, New Hampshire
Background [1]
Physical Description: The New Hampshire Plating Company is located on Wright Avenue in
Merrimack, Hillsborough County, New Hampshire. The surrounding area is primarily used for
light industrial and commercial purposes, with some residential areas nearby. The site covers
13.1 acres of leased property and includes the former plating facility and four lagoons (see Figure
1). The plating facility is located approximately 600 ft west of the Merrimack River. The study
area consisted of the land around the former facility and the lagoons, and extended east to the
Merrimack River and south to Horseshoe Pond, to determine if the contamination was moving in
those directions. The site lies in the 100-year floodplain of the Merrimack River, and the
topography of the site has little relief.
Site Use: The property was used from 1962 until 1985 for electroplating activities. The four
lagoons on site were used for disposal of wastes and wastewaters resulting from the
electroplating operations. These lagoons were unlined and had no leachate detection or
collection.
Release/Investigation History: From 1962 to 1985, the facility discharged on-site 35.000 to
60,000 gallons per day (gpd) of electroplating wastes into a series of four unlined lagoons. The
wastes included cyanide plating baths and sludges, acids, and chlorinated solvents. Discharge of
degreasing solvents into lagoon was discontinued in the late 1970s. In 1980, the New Hampshire
Plating Company (NHPC) notified the EPA that it was a hazardous waste disposal facility under
Subtitle C of the Resource Conservation and Recovery Act (RCRA). An inspection by the New
Hampshire Department of Environmental Services (NHDES) and the Environmental Protection
Agency (EPA) in April 1982 noted several RCRA violations. As a result, the New Hampshire
Division of Public Health Services issued a Notice of Violations and Order of Abatement to
NHPC. In February 1983, the State of New Hampshire filed a civil suit against NHPC. NHPC
halted operations in 1985 because it lacked the financial resources necessary to meet compliance
standards and continue hydrogeologic investigations at the property.
In June 1987, a contractor for New Hampshire Division of Environmental Services treated the
lagoon system with lime and a sodium hypochlorite solution, removed debris, drums, and plating
tank liquids to a regulated disposal facility, and conducted a superficial cleaning of the
manufacturing building. In 1990, EPA used emergency funds to solidify the contaminated sludge
and soil at the property.
109
-------
New Hampshire Plating Company
SITE INFORMATION
Regulatory Context: This investigation was conducted to support the characterization of waste
under the Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLA). Manufacturing operations at the site were regulated under Subtitle C of the
Resource Conservation and Recovery Act (RCRA), and the New Hampshire Division of
Environmental Services.
71°29'30-
4^51-30'
42-51'
FORMER PLATING FAC1UTY
,\
MW-8r • \_7 ?
MERRIMACK
RIVER
MW-10S MW-105
B-10S • B-ll
B-lod
HORSESHOE
POND
MW-106 MONITORING WELL LOGGED
AND NUMBER
B-10S ADJACENT WSLL(S)
B-100 AND NUMBER
200 400 600 FEET
Base modifed tram U.S. Amy Coips of Engineers. 1:1.200. 1992.
Figure 1: Site Map of NH Plating Company [1]
110
-------
SITE INFORMATION
New Hampshire Plating Company
Site Logistics/Contacts
Federal Lead Agency: EPA
Federal Oversight Agency: None
Remedial Project Manager:
Jim Dilorenzo
EPA Region 1
617-918-1247
Geophysical Subcontractor:
Thomas Mack
United States Geologic Survey
361 Commerce Way
Pembroke, NH 03275
603-226-7805
! MEDIA AND CONTAMINANTSl
Matrix Identification
Type of Matrix Sampled and Analyzed: Sand, silt, fine-grained lake bottom sediments, and
coarse-grained stratified drift.
Site Geology/Stratigraphy
Site geology is composed of an alluvial terrace consisting of sand, silt, and some gravel, ranging
in thickness from less than five to 25 feet (ft). The total thickness of surficial sediments
overlying bedrock in the study area ranges from zero at the northwestern bank of Horseshoe Pond
to greater than 120 ft between the center of the Horseshoe Pond and the plating facility, and
approximately 20 ft along the northeastern bank of Horseshoe Pond.
The alluvial terrace is underlain by glaciolacustrine sediments and till. The glaciolacustrine
sediments consist primarily of sand, silt, and clay with some gravel. Coarse-grained sediments
are interspersed with fine-grained lake bottom sediments within this unit. The till is a poorly
sorted mixture of silt, sand, and gravel with some boulders and clay. The total thickness of these
units could be as much as 100 ft.
Underlying the glaciolacustrine sediments and till is a bedrock unit consisting of schists and
phyllites with minor amounts of granite and gneiss. The bedrock surface forms a north-south
trending trough and outcrops at the northwestern bank of Horseshoe Pond.
The depth to the water-table is approximately 5 to 20 ft in the study area. Groundwater flows
beneath the site to the east toward the Merrimack River and to the south toward Horseshoe Pond.
Ill
-------
New Hampshire Plating Company
I MEDIA AND CONTAMINANTSi
Contaminant Characterization
Primary Contaminant Groups: The primary contaminants of concern include cadmium,
chromium, copper, cyanide, iron, nickel, zinc, tin, arsenic, lead, manganese, sodium, and
trichloroethylene.
Matrix Characteristics Affecting Characterization Cost or Performance
The effectiveness of the ground penetrating radar (GPR) survey was limited by the fine-grained
bottom sediments in the water bodies, as well as by the depth of the water. The fine-grained
sediments found along the bottom of Horseshoe Pond limited the penetration of the seismic
signal, which had already been attenuated by the water column. Although such sediments were
less prevalent in the Merrimack River, similar problems occurred along some of the survey lines
there. The fine-grained soils present across the site had a similar effect on the land GPR surveys
in some locations, resulting in blank records in the images along certain survey lines. These
blank areas on the profiles occurred under fine-grained sediments and also coincided with areas
known to contain highly conductive groundwater.
The effectiveness of the seismic survey was limited by the presence of organic sediments on the
bottom of Horseshoe Pond, and, to a lesser degree, along the bottom of the Merrimack River.
The seismic survey was, however, able to clearly identify the fine-grained sediments that
impeded the performance of the GPR survey.
No site characteristics impeded the performance of the natural gamma or EM borehole logs, such
as complex lithology, e.g. clay content, organic matter, magnetic minerals content, etc.
I GEOPHYSICAL INVESTIGATION PROCESS.
Investigation Goals
The goal of the investigation was to confirm stratigraphic information collected during earlier
investigations and to characterize zones of highly conductive groundwater that may indicate the
presence of a groundwater contaminant plume. Continuous seismic reflection and ground
penetrating radar surveys were used to confirm the stratigraphy across the site. Natural gamma
logs were used to develop vertical profiles of stratigraphy in existing monitoring wells.
Electromagnetic induction logs were used to identify the zones of conductive groundwater.
Geophysical Methods
The continuous seismic reflection method uses an acoustic source to emit a signal downward into
the subsurface and measures the travel time of a seismic signal from the surface to subsurface
reflectors, i.e. bedrock, and back. Travel time, or velocity, is typically measured in feet per
second (ft/s). In New England geology, the velocity of sound through saturated glacial sediments
ranges from 4,000 to 6,000 ft/s [2]. This study used a seismic sound velocity of 5,000 ft/s to
112
-------
New Hampshire Plating Company
i GEOPHYSICAL INVESTIGATION PROCESS BHIMM^^M^MM
calculate penetration depth of the seismic signal. The continuous seismic reflection profile was
carried out using Geoacoustic's Geopulse© equipment operated at a frequency of 700 to 1500
hertz.
The ground penetrating radar (GPR) method uses an antenna to radiate short pulses of high-
frequency radio waves into the subsurface and a receiving antenna to record variations in the
reflected return signal. Interpretations of GPR profiles depend on the sediments penetrated and
the scale adjustments as those sediments change. For example, the electromagnetic velocity in
saturated unconsolidated sediments is 0.2 feet per nanosecond (ft/ns) as compared to 0.4 ft/ns in
unsaturated unconsolidated sediments. The primary factor limiting depth of penetration is the
electrical conductivity of the sediments [1]. The type of equipment used was a GSSI System 10
operated at a frequency of 80 megahertz.
Natural gamma logging is the continuous physical measurement of the release of natural gamma
radiation from the soil and rocks surrounding the length of a borehole. The gamma log measures
the total gamma radiation, in counts per second, as the detector is raised in the well column. In
the glaciated sediments of the northeast, fine-grained sediments rich in clay are generally more
radioactive than quartz sand or carbonate rocks [1]. The natural gamma-data were gathered using
Century Equipment Natural Gamma detector which is capable of logging at a rate of 30 ft/min.
The electromagnetic induction method uses a transmitter coil that generates an electromagnetic
field that induces currents in the earth. A receiver coil intercepts the electromagnetic fields
generated induced current as a voltage that is linearly related to subsurface conductivities.
Subsurface conductivities are measured in terms of millisiemens/centimeter (uS/cm). The EM
conductivity of unconsolidated glacial sediments is primarily affected by the presence of clay
minerals and the conductivity of ground water. The presence of ions in water, such as dissolved
metals, increases the electrical conductivity of that water [1]. The EM data were acquired using
a Century Equipment EM Flow Meter, Model 9721.
I GEOPHYSICAL FINDINGS
Technology Calibration
The only calibration conducted was to focus the EM probe so that the maximum response was
obtained from soils about one foot from the probe, or the center of the borehole. This measure
was taken to avoid any interference with the well casing materials. The seismic, ground
penetrating radar and natural gamma instruments needed no calibration.
113
-------
New Hampshire Plating Company
I GEOPHYSICAL FINDINGS
Investigation Results
Continuous seismic reflection profiles were generated along 16 lines around Horseshoe Pond.
Along some lines, organic sediments at the bottom of the Pond impeded the penetration of the
seismic wave, resulting in ambiguous results. The profiles generated for Horseshoe Pond
indicated that bedrock, overlain at times by coarse-grained sediments, was identified at depths
ranging from the surface of the pond to approximately 20 ft below the surface. At the
northernmost end of the pond, the method was unable to detect bedrock known to be present at a
depth of 100 ft. The investigator believed that the inability of the method along this line was due
to the presence of organic bottom sediments that prevented the penetration of the seismic signal.
The seismic survey of the Merrimack River included eight lines along the western shore and five
lines taken along the midline of the river. Bedrock was found underneath the river at depths
ranging from 10 to 50 ft below the river surface, and to depths of 20 ft along the shoreline. The
results from the river survey were markedly better than those for the pond survey. The lack of
organic sediments in the river resulted in a better profile of the bedrock surface in all lines.
Ground penetrating radar profiles were generated along 12 lines around the plating facility and
extending north and south from the facility. Along most of the lines, the survey found that the
subsurface materials consisted mainly of silt, sand, and clay. The subsurface materials graded to
a fine sand along the southernmost survey line, close to Horseshoe Pond. Bedrock outcrops can
be seen in a few of the profiles. The fine materials tended to limit the penetration of the radar
signal to depths no greater than 35 ft, and obscured the water table along most lines. In several
of the profiles, blank areas appear below a depths of 10 to 12 ft. These blank areas were
interpreted to represent the failure of the radar signal to penetrate fine-grained soils, or the
presence of highly conductive groundwater. Radar surveys taken on the pond and the river
yielded ambiguous results, as the radar signal was attenuated by the water and failed to penetrate
the fine-grained sediments along the pond bottom.
Natural gamma logs were developed for seven existing monitoring wells, one upgradient of the
plating facility and its lagoons, and six downgradient. The gamma logs were useful in
delineating the stratigraphy of the subsurface materials and identifying permeable and
impermeable zones. The EM log was used to vertically delineate zones of increased electrical
conductivity to identify potential contaminant plumes.
Three of these logs have been reproduced in Figures 2 to 4. The EM and gamma logs in these
figures are shown along with the lithologic logs developed at the time the wells were installed
(presented to the right of the geophysical logs). The lithologic logs indicate significant
heterogeneity in the distribution of layers of coarse to fine materials. The only consistent stratum
found in each of the logs was the near-surface very fine sand layer. The gamma readings in each
of these borings correlated closely with the lithologic logs. Gamma readings were clearly higher
in strata composed of finer sands and silts and lower in sandy strata. Gamma counts of less than
100 counts per second (cps) consistently were measured in strata that were identified in the
lithologic logs as medium to coarse sands. Higher counts, in the range of 100 to 150 cps, were
registered in the layers of fine sand and silts.
114
-------
New Hampshire Plating Company
[ GEOPHYSICAL FINDINGS
The EM logs were taken to identify zones of conductive groundwater that may indicate the
presence of chromium-contaminated groundwater. In each of the logs shown in Figures 2 to 4,
there are significant spikes in the EM readings, indicating possible zones of contamination.
MW-8r (Figure 2), which is immediately downgradient of the lagoons and directly west of the
plating facility, showed three distinct zones of conductive ground water, centered at altitudes of
approximately 15, 58, and 92 ft above mean sea level (msl). In MW-104d which is located to the
south of the lagoons, only one such zone can be seen, centered at 75 ft msl (Figure 3). In MW-
108d, located to the east of the plating facility, a single zone is seen at 54 ft msl (Figure 4). It is
interesting to note that in each of the wells, the spikes in the EM readings occur in strata
composed of finer materials.
Results Validation
The results of the EM measurements were compared with water samples that were being
collected as part of the ongoing monitoring. Specific conductance was measured in each well
and the result printed on the geophysical logs in Figures 2 to 4 (shown to the right of the
lithologic log). The altitude at which the conductance log is printed and indicates the depth of
the well screen. The measured conductance correlated well with the locations with high
conductivity from EM logging in two of the three logs. In MW-8r and MW-108d, the specific
conductance was higher at the depths at which the EM measurements were also high. In MW-
104d, however, high specific conductance was measured near the bottom of the well, but the EM
readings at that altitude were not high relative to a higher location in the well. Measures of
specific conductance can be sensitive to naturally occurring ions, as well as ions associated with
chromium contamination.
115
-------
New Hampshire Plating Company
I GEOPHYSICAL FINDINGS
MW-8r
GAMMA, IN COUNTS PER SECOND
CONDUCTIVITY, IN MILLISIEMENS PER METER
118
108
LU
UJ
LU
CO
LU
I
<
LU
LU
LU
Q
150 200 250
Screened intervai
Well number
Specific conductance
of water sample
B-8s
100 microsiemens per
centimeter at 25
degrees Celcius
EXPLANATION
MEDIUM TO COARSE
SAND
VERY FINE TO FINE
SAND
VERY FINE SAND, SILT.
OR CLAY
WATER TABLE
EM ELECTROMAGNETIC
CONDUCTIVITY
Figure 4. Borehole geophysical logs, lithologic section, screened interval, and associated specific
conductance of ground water in Merrimack, New Hampshire at well MW-8r.
Figure 2: Geophysical Log for MW-8r [3]
116
-------
New Hampshire Plating Company
I GEOPHYSICAL FINDINGS
MW-104d
GAMMA, IN COUNTS PER SECOND
CONDUCTIVITY, IN MILLISIEMENS PER METER
50 100 150 200 250
112?
LU
LU
102
92
82
72
<
UJ
CO
LU
O 62
03
<
LLJ
LU
LL
52
LU 42
Ci
Z3
H;
v- 32
22
12
asp-/
1
•?M
HM:
;B;
'*^=,X;
'•K';r!
Screened interval
Well number
Specific conductance
or water sample
MW-104S
660micrpsiemens per
centimeter at 25
degrees Celsius
M W-104d
800 micrpsiemens per
centimeter at 25
degrees Celsius
FINE TO MEDIUM !§
SAND &
' FINE TO FINE
EXPLANATION
VERY FINE SAND. SILT, EM ELECTROMAGNETIC
OR CLAY CONDUCTIVITY
WATER TABLE
Figure 5, Borehole geophysical logs, lithologic section, screened intervals, and associated
specific conductance of ground water in Merrimack, New Hampshire at well MW-104d.
Figure 3: Geophysical Log for MW-104d
117
-------
New Hampshire Plating Company
I GEOPHYSICAL FINDINGS
MW-108d
GAMMA, IN COUNTS PER SECOND
CONDUCTIVITY, IN MILLISIEMENS PER METER
0 50 100 150 200 250
120
Screened interval
Weil number
Specific conductance
of water sample
MW-108s
420microsiemens per
centimeter at 25
degrees Celsius
MW-108d
81 Omicrosiemens per
centimeter at 25
degrees Celsius
FINE TO MEDIUM B
SAND ^
VERY FINE SAND. SILT,
OR CLAY
EXPLANATION
VERY FINE TO FINE EM ELECTROMAGNETIC
SAND CONDUCTIVITY
WATER TABLE
*• WA' fcH 'ABLt
Figure 8. Borehole geophysical logs, lithologic section, screened intervals, and associated
specific conductance of ground water :n Mcrrimack. New Hampshire at well MW-108d.
Figure 4: Geophysical Log for MW-180d
118
-------
New Hampshire Plating Company
I LESSONS LEARNED
The lessons learned during this investigation are the following:
• Downhole technologies provide more information on the stratigraphy and potential
plumes of contaminated groundwater beyond shallow depths than surface geophysical
techniques.
• Downhole technologies are effective in developing contour maps of lithologic units
including subsurface structures that might promote or inhibit contaminant pathways.
• The two borehole technologies may be especially effective for large sites with deep
contamination and complex stratigraphy. Borehole technologies may be useful along
with boring and monitoring well placement during initial site evaluations. For sites with
limited areal extent, shallow contamination, and/or simple stratigraphy, borehole
technologies may be less cost-effective.
• Caution should be taken not to mistake natural conductivity in EM surveys with
contamination.
• Both EM and natural gamma were much more effective in delineating deep subsurface
features than seismic-reflection and GPR.
• Interpretation of the results of the investigation is useful in identifying the utility of the
borehole technologies and comparison with utility of seismic-reflection and GPR. The
following paragraphs discuss the technology usefulness and limitations as they relate to
this investigation.
• The wells that had been installed for on-going monitoring missed the most contaminated
sections of the aquifer. These wells were unable to identify the contaminated zones with
typical monitoring well techniques. EM was useful in identifying likely elevations
where plumes of contaminated groundwater exist. However, the levels of peak
conductivity in both the measurement of groundwater samples and EM are strong
indicators of contamination, for this site, because the background well had much lower
conductivity. Further, the pattern of downgradient conductivity was consistent with a
groundwater contamination plume pattern.
• Natural gamma logging can provide consistent information on stratigraphy. When
accurate surface elevations are obtained for each boring, a contour map can be developed
for any of the lithologic units that were identified in the survey. This information can be
used to identify subsurface structures that might provide migration pathways for
contaminants.
The two borehole technologies may be especially effective for sites with the following
features: unconsolidated sediments, large areal extent, deep contamination that may have
119
-------
New Hampshire Plating Company
[ LESSONS LEARNED
traveled far, and for more complex stratigraphy. In these situations, the borehole
technologies may be most effective in combination with examination of boring logs and
limited initial monitoring well placement. The monitoring wells are cost-effectively
placed in combination with the borehole technologies investigation and are useful for
correlating results. Following the combination of initial efforts of boring log
examination, groundwater monitoring, and borehole technologies evaluations, additional
borings and monitoring wells may be placed more cost-effectively, than if only borings
and monitoring wells were placed.
However, for sites with limited areal extent, shallow contamination, and/or simple
stratigraphy, borehole technologies (EM and natural gamma) are less cost-effective.
120
-------
New Hampshire Plating Company
I REFERENCES
1. Mack, Thomas J. Geophysical Investigations in the Vicinity of a Former Electroplating
Facility in Merrimack, New Hampshire: U.S. Environmental Protection Agency, Region
1, 1994.
2. Haeni, P.P. Application of continuous seismic-reflection methods to hydrologic studies.
Ground Water, 1996, v. 24, no. 1, p. 23-31.
3. Beres, Milan Jr., and Haeni, F.P. Application of ground-penetrating-radar methods in
hydrogeologic studies. Ground Water, 1991, v. 29, no. 3, p. 375-386.
121
-------
PAGE LEFT BLANK INTENTIONALLY
122
-------
NMSHTD District 1 Headquarters
Case Study Abstract
New Mexico State Highway and Transportation Department (NMSHTD)
Underground Storage Tank (UST) Investigation
Deming, New Mexico
Site Name and Location:
NMSHTD UST Investigation
Deming, NM
Period of Site Operation:
1955 to present
Operable Unit: N/A
Points of Contact:
Phil Ramos, 505-827-5528
Richard Meixner, 505-822-9400
Jim Viellenave, 303-278-191 1
Geophysical Technologies:
Magnetometry
Electromagnetics
Natural gamma logging
Soil gas analysis
Geological Setting: Quaternary bolson
alluvium underlain by Cretaceous
Mesa Verde and Mancos shale
Date of Investigation: July 1997
Current Site Activities:
Planning for Soil Vapor Extraction
Remedy
Technology Demonstrator:
Sunbelt Geophysics
P.O. Box 36404
Albuquerque, NM 87176
505-266-8717
TEG Rocky Mountain
400 Corporate Circle, Suite R
Golden, CO 80401
303-278-0104
Purpose of Investigation: The overall goal of this environmental investigation was to identify and characterize the source of
chlorinated VOC contamination in groundwater beneath the NMSHTD site. The goal of the magnetic and electromagnetic
survey was to locate buried materials that might be potential sources of contamination. The goal of the gamma log survey was
to guide the placement of the soil vapor sampling points.
Number of Images/Profiles Generated During Investigation: 33 natural gamma profiles.
Results: Magnetometry and electromagnetics identified two areas of concern. Natural gamma logs of direct-push boreholes
identified stratigraphic units that influenced the migration of contaminant vapors in the vadose zone. Permanent soil gas
sampling points were installed in the units identified by the gamma logs. The soil gas survey provided a representative
distribution of the contamination in the vadose zone.
123
-------
NMSHTD District 1 Headquarters
EXECUTIVE SUMMARY I
The New Mexico State Highway and Transportation Department (NMSHTD) District 1 Headquarters
site is located in Deming, New Mexico in the southwestern part of the state. The site covers
approximately 15 acres. The site lies within 10 miles of the U.S./Mexico border. The topography is
generally flat with little or no relief across the site. The Mimbres River is located north of the site and is
the only nearby water body. The property has been used since 1955 by the NMSHTD for vehicle
maintenance, steam cleaning, and other activities. During years of heavy roadbuilding, a materials
testing laboratory used 1,1,1-tetrachloroethane (1,1,1-TCA) on a regular basis for asphalt analyses.
Spent solvent was either disposed off site or recycled in an on-site still. The aggregate used in testing
was apparently rinsed with water, and the contaminated water was regularly rinsed down the drain and
into the septic system.
Site geology consists of deposits typical of an arid zone basin that has been filled in by erosion of
materials from the surrounding uplands. Locally interlayered sandy clay and clayey sand are present
with some gravel. A thick gravel layer was present at depths between eight and 15 feet below ground
surface (bgs). Depth to groundwater is 100 to 150 feet bgs.
A geophysical investigation was completed as part of a soil gas survey conducted at the site in 1997. The
information presented in this report was derived from the interpretive report of the geophysical
investigation. Geophysical methods were used to identify buried materials and to find optimal locations
for the placement of soil gas sampling points. A reconnaissance survey was performed over the study
area using magnetometry and electromagnetics (EM) to identify buried materials that might be sources of
contamination. The reconnaissance survey was conducted over a 25-acre area from July 7 to July 22,
1997. The survey identified numerous areas of buried materials, but only two were of interest. A septic
tank was identified to the southeast of the building that had housed the materials testing laboratory. Also
identified was another area located approximately 75 feet to the north of the septic tank on the east side
of the building. Natural gamma logs taken in direct push boreholes were used to identify clay lenses that
might impede the migration of soil gas vapors. Soil gas sampling points were installed just below these
lenses.
The gamma logs were successful in locating the clay lenses that were controlling vapor migration in the
vadose zone. The resulting soil gas survey identified areas of groundwater contamination related to the
septic field that was acting as a source area.
124
-------
NMSHTD District 1 Headquarters
SITE INFORMATION
Identifying Information
Figure 1: Site Location
New Mexico State Highway and Transportation Department (NMSHTD)
District 1 Headquarters Underground Storage Tank (UST) Site
2912 East Highway 80
Deming, NM
Background
Physical Description: The NMSHTD District 1 Headquarters Underground Storage Tank (UST)
site is located in Deming, New Mexico in the southwestern part of the state, as shown in Figure
1. The site covers approximately 15 acres. This investigation was
extended off site to investigate various properties that may have
contributed to the on-site contamination [1].
The site lies within 10 miles of the U.S./Mexico border. The
topography is generally flat with little or no relief across the site. The
Mimbres River is located north of the site and is the only nearby water
body [2].
Site Use: The property has been used since 1955 by the NMSHTD for
vehicle maintenance, steam cleaning, and other activities. Figure 2
shows a map of the site and the immediate area. During years of heavy
road building, a materials testing laboratory used 1,1,1-tetrachloroethane (1,1,1-TCA) on a
regular basis for asphalt analyses. Spent solvent was either disposed off site or recycled in an
on-site still. The aggregate used in testing was apparently rinsed with water, and the
contaminated water was regularly rinsed down the drain and into the septic system. This
improper disposal of 1,1,1-TCA-contaminated water contributed to the contamination of the site
[1, 2, 3, 4].
Release/Investigation History: During a tightness test of underground storage tanks at the
Deming site in July and August 1989, NMSHTD found leaks in underground storage tanks and
petroleum hydrocarbon contamination of subsurface soil. Subsequent investigations confirmed
the presence of gasoline-derived and chlorinated volatile organic compounds (VOCs) in
groundwater at concentrations in excess of New Mexico Water Quality Control Commission
health standards [1,3,4].
In June 1996, Daniel B. Stephens & Associates (DBS&A) conducted a shallow soil gas survey
using passive soil gas samplers to detect contamination in the vadose zone. The survey was
conducted over an area of known VOC-contaminated groundwater, but the only VOC detected
was perchloroethene (PCE). Furthermore, the distribution of PCE was not representative of the
distribution of chlorinated VOCs known to be present in the groundwater beneath the site. The
investigators from DBS&A believed that the local geology affected the movement of
contaminant vapors beneath the site, possibly preventing the shallow samplers from registering
125
-------
NMSHTD District 1 Headmiarters
SITE INFORMATION
chemicals known to be present in groundwater approximately 100 feet below ground surface
(bgs)[l,3j.
Regulatory Context: New Mexico UST Regulations and New Mexico Water Quality Control
Commission standards [1,3].
MatKfebond
T«Sng laboratory-
***> ' isrw
L «=rnd
5 flawing
1 «••"» aws
1 °
T— v ijn ^
"^ "?* ! Lrr"
^
S«pUe «y»»m— jpiV^26
im-rii^^'
SO-2S *
v
sc-u
a
SC-Jl S^10
£- . .
Sfi-l!
,HW-6 0
(UW-1M)
0
SO-22
NMSHTD wNcit tndroad
oqjptnor* atorag*
SG-9 ^t^"^
-^"^ _ MV«112
1 X V«»-ii«u .
\UK-102S
SG-2S ' UW-1W N.
O K *L
: . "?. ^
T • /-£ievo:«i toftk
J '— ^
1 ^SC-J
Former USTs— v. ,— ^ • Office end vesci?
^st l moiflKncnc* TOOtr
. V.'' *»
0 0
SG-*
ASTs— . ^ iiw-iso '
^--Q •
1*1-1 OSS
SIW09«
eonooy 1
0 SI-1<> u»'-!MO "
SO 19
/
jr
so- is
vv
!
f
State
PoBce
-------
NMSHTD District 1 Headmiarters
SITE INFORMATION
Site Logistics/Contacts
State Lead Agency: NMSHTD
Federal Oversight Agency: Not applicable
Remedial Project Manager:
Richard Meixner
Daniel B. Stephens and Associates, Inc.
Albuquerque, NM 87176
505-822-9400
Site Contact:
Phil Ramos
New Mexico Highway and Transportation
Department
Albuquerque, NM 87176
505-827-5528
Geophysical Subcontractors:
David Hyndman
Sunbelt Geophysics
P.O. Box 36404
Albuquerque, NM 87176
505-266-8717
James Viellenave
TEG Rocky Mountain
400 Corporate Circle, Suite R
Golden, CO 80401
303-278-0104
I MEDIA AND CONTAMINANTS I
Matrix Identification [3]
Type of Matrix Sampled and Analyzed: Subsurface clays, sands, and gravels.
Site Geology/Stratigraphy [3,4,5]
Regional geology in the area of the site consists of Quaternary alluvium underlain by Cretaceous
Mesa Verde and Mancos shale. In some areas the eroded materials have been reworked by local
streams. The local stratigraphy consists of deposits typical of an arid zone basin that has been
filled in by erosion of materials from the surrounding uplands. Locally interlayered sandy clay
and clayey sand are present with some gravel. A thick gravel layer was present at depths
between eight and 15 feet bgs. Some confining layers are present that may influence the
migration of contaminant vapors. Depth to groundwater is 100 to 150 feet bgs.
Contaminant Characterization [1,3]
Primary Contaminant Groups: The primary contaminants of concern include trichloroethene
(TCE), perchloroethene (PCE), 1,1-dichloroemene (1,1-DCE), and 1,1,1-dichloroethane (1,1,1-
DCA). A combination of benzene, toluene, ethylbenzene, and xylene (BTEX) was also present
but was not addressed in the investigation described here. The most frequently detected
chlorinated VOC in soil gas was 1,1-DCE.
127
-------
NMSHTD District 1 HeaHmiarters
I MEDIA AND CONTAMINANTS I
Matrix Characteristics Affecting Characterization Cost or Performance
The magnetometry survey was affected in three areas by the presence of sources of magnetic
interference, such as fences, buildings, etc, to the degree that the survey in those areas was
replaced by an electromagnetic survey [3].
The natural gamma detector used at the Deming site was found to be sensitive to temperature
change. After it was lowered into the hole, the field team allowed it to equilibrate to the lower
subsurface temperature before recording the counts of natural gamma radiation. Later models of
the detector are designed to be impervious to temperature differences and has been used in 100°F
temperatures [3, 6]. The detector is impervious to humidity and water. It functions in
groundwater and has been used in the rain and snow [5].
Certain geologic materials, such as granite-derived cobbles and gravel in conglomeratic deposits,
organic rich deposits, and phosphate and potash (K2CO3) deposits, have low natural gamma
radiation levels, and natural gamma logging may be insufficient to distinguish layers composed
of these materials. However, these materials were not present in the alluvial deposits examined
at this site [3, 5, 6].
The presence of a gravel layer between eight and 15 feet bgs and a tendency of the deeper
materials to collapse when the probe was advanced led the investigators to conduct the gamma
logging inside of the drive rods [3].
I GEOPHYSICAL INVESTIGATION PROCESS I
Investigation Goals [1,3]"
The overall goal of this environmental investigation was to identify and characterize the source
of chlorinated VOC contamination in groundwater beneath the NMSHTD site. The goal of the
magnetic and electromagnetic survey was to locate buried materials that might be potential
sources of contamination. The goal of the gamma log survey was to guide the vertical placement
of the soil vapor sampling points.
Geophysical Methods
A reconnaissance survey was performed over the study area using magnetometry and
electromagnetics (EM). The magnetometry survey was carried out using a Geometries G-858
cesium magnetometer. Magnetic data were acquired every two feet along parallel traverses
separated by 10-foot intervals. The EM data were acquired using a Geonics EM-61 high
precision metal locator every 0.6 feet along parallel traverses separated by five-foot intervals.
Natural gamma data were gathered using a Mt. Sopris Slim Line prototype instrument with a
sodium iodide detector to measure the impinging natural gamma radiation. (A commercial
version of this instrument has been developed since the date of this investigation [9]). The
128
-------
GEOPHYSICAL INVESTIGATION PROCESS
Figure 3 - Typical Gamma Log [7]
detector was 0.75 inches by 24 inches in size, and
was attached to a 200 foot cable. An MGX data
logger was connected to the cable. Gamma logging
is useful in borings ranging from one to six inches
in diameter [1, 3, 6].
Natural gamma logging is the physical measurement
of the release of natural gamma radiation from the
soil and rocks surrounding a borehole. Natural
gamma logging is based on the principle that more
intense natural gamma radiation is emitted from
clay-rich formations, which are usually higher in
naturally radioactive elements, than clay-poor
formations. Most natural gamma radiation occurs in
clays containing thorium, uranium, or potassium 40.
Figure 3 shows a typical natural gamma log for
some consolidated sedimentary deposits. Note the
higher counts for clay-rich units like shale,
particularly marine shale, and bentonite.
Natural gamma measurement begins with the
lowering of the detector to the bottom of a hole, allowing it to equilibrate to the different
subsurface temperature, and then reeling the detector up the hole at a steady rate of between five
and 10 feet per minute (allowing the logging of a 50-foot hole in five to 10 minutes). The level
of gamma radiation being emitted by a particular stratum is measured in counts per second (cps).
Interpretation of the gamma log depends as much on the absolute value of the gamma counts as it
does on the rate of change in gamma counts as the -detector passes from one material to the next.
Statistical variations in gamma emissions, significant at low counting rates, are smoothed out by
integration over a short time interval. If the hole is logged too quickly, however, the smoothing
effect leads to erroneous results by shifting the peaks in the direction of logging. The lower left-
hand portion of Figure 3 illustrates the result of logging too fast [8].
Multiport soil gas wells were installed at depths between 20 and 60 feet bgs using direct
push/hammer (Strataprobe®) technology. The Strataprobe® unit consisted of a dual ram with a
hydraulic hammer vibrating component capable of producing a high-frequency impact with an
8,000 pound static reaction weight and more than 35,000 pounds of pullback capacity. The
truck-mounted hydraulic percussion hammer unit was used to advance 1.75-inch outer diameter
rods with an expendable 2-inch diameter tip into the subsurface until downward progress ceased
due to refusal [1, 3].
After refusal, the rods were disconnected at the surface in order to conduct a subsurface natural
gamma profile in the borehole to depths of approximately 50 feet bgs. The probe was pushed to
129
-------
, NMSTTTD District 1
I GEOPHYSICAL INVESTIGATION PROCESS ^^^^
total depth (averaging 50 feet), and the gamma logging was conducted from inside the pipe.
Permanent vapor sampling points were installed as the pipe was withdrawn [1, 3, 6].
I GEOPHYSICAL FINDINGS
Technology Calibration fl, 3]
No calibration of the magnetometer or the EM detector were performed, as this is not general
practice.
To calibrate the natural gamma log readings for the Deming site, gamma logs were taken in three
existing monitoring wells: MW-102, MW-109, and MW-111. The gamma readings were
correlated with the lithology and stratigraphy that had been previously described for these wells.
Figures 4 through 6 show the lithologic logs for these wells on the left and the corresponding
gamma log on the right.
An examination of the logs revealed an acceptable level of correlation. In each of the gamma
logs, the presence of a near-surface layer of silty clay was indicated by an increase in the gamma
counts as the detector passed through that material. The individual logs, however, did show a
difference in the absolute values of gamma counts for this layer, as gamma counts rose to levels
of 115 to 135 cps in MW-109 and MW-111, and to levels of approximately 275 cps in M W-102.
The difference in the absolute level of gamma counts probably indicated that the silty layer in
MW-109 and MW-111 contained a smaller proportion of coarser materials than it did in MW-
102. The near-surface silty layer was present in each of the lithologic logs.
The layer of coarse gravel that was identified in the monitoring well logs at depths ranging from
eight to 15 feet bgs can be seen in each of the calibration logs as both a small decline in absolute
gamma counts to levels of less than 100 cps and as an increase in the distance from peak to peak
in the log. Again, there appears to be a higher proportion of silty materials mixed with the gravel
in the log for MW-102, as evidenced by the higher gamma counts for similar materials shown in
that log than in the other two gamma logs.
The layer of silty materials present in the lithologic log for MW-109 and MW-111 at
approximately 25 feet bgs is identifiable in the gamma logs for those wells. The grading from
coarser sand materials into a silty clay can be seen as the gamma count rises above 100 cps. That
silty layer was not present in the lithologic log for MW-102, and no indication of such a layer
can be seen in the gamma log for that well.
The next interval in the lithologic logs is composed of predominantly sandy materials, although
the size of the interval varies across the three wells. In MW-102, this interval extends to an
approximate depth of 40 feet bgs, and the gamma log for that well shows small variations in
gamma counts which remain in the 150 to 225 cps range. In MW-109, the sandy layer extends
deeper to an approximate depth of 48 to 50 feet bgs. This layer can be seen in the gamma log for
MW-109 as an interval over which the gamma counts remain largely within a narrow range
130
-------
NMSHTT) District 1 Hearinnarters
I GEOPHYSICAL FINDINGS
between 63 and 75 cps. The interval between 25 and 55 feet bgs in MW-111 is largely composed
of sand with gamma counts varying between 50 and 100 cps. This interval is interrupted at a
depth of approximately 38 feet bgs by a sandy clay material that can be seen in the gamma log as
gamma counts rise to approximately 125 cps.
The lithology of these wells over the remaining interval is significantly different. The lithology
of MW-102 below 40 feet bgs grades from sandy silt to sandy gravel and back to sandy silt at a
depth of 84 feet bgs. The short interval of sandy gravel, shown on the lithologic log from 65 to
75 feet bgs, can be seen in the gamma log as a lower set of gamma counts beginning at
approximately 68 feet bgs and extending to 78 feet bgs. The lithologic log shows a gradual
coarsening of materials below a depth of 50 feet bgs in MW-109 to a depth of 78 feet bgs. The
gamma log for this well, instead, shows a similar coarsening over the interval from 50 feet bgs to
60 feet bgs as the gamma counts gradually decline. From 60 feet bgs to a depth of approximately
75 feet bgs there appears to be a gradual increase in gamma counts. Such an increase may only
indicate the presence of silty materials mixed in with the sand that are not readily evident in a
visual inspection of the same materials.
Overall, there appears to be an adequate correlation between the lithologic logs and the gamma
logs for the three monitoring wells for successful calibration. In addition, the gamma logs reveal
the presence of fine-grain materials when the same materials are not noted in the lithologic logs.
This may be due to the fact that the lithologic logs were developed for another use and were used
here as a matter of convenience, or that gamma readings provide a more sensitive measure of
subtle changes in stratigraphic units than can be achieved with a visual inspection.
131
-------
I GEOPHYSICAL FINDINGS
NMSHTD District 1 TTpaHnnarters
f
CM
O
§
25
l
50
l
75
I
'm
o
CO O '
>o t*
o
U
tt \o eo o e« a
t- t> t* oo " =
132
-------
NMSfTTD District 1 Hfeari quarters
(GEOPHYSICAL FINDINGS
m
VI
I
o
o
—]-
25
i
50
I
CT>
O
cr>
c
o
•-6
P
I !
Cs
o
!C
O
i.
O
i!
0 Q
I
H
Q
J
J«!«/«^
« * O 06 O
N «t « 09 O
Cfl if \D CO p
133
Gaza*
c* 5 vo oo o
W ^ 1O CO O
e* *
-------
I GEOPHYSICAL FINDINGS
NMSHTD District 1 Hpadniiarfprs
f
cr>
S
25
l
<
.-'
"oi
P
d
^
=r?
'
=2
O
J
1 1 1 1^
MM!
MM!
i 1 IJ^
1 1 J J J
.n
"a
U
M M 1. M i M M M M &
a a a a a «
-------
NMSHTD District 1 Headnnartprs
i GEOPHYSICAL FINDINGS m^mmmf^mmmmmfmmmmmmmmm^^m
Investigation Results [3]
The reconnaissance survey was conducted over a 25-acre area from July 7 to July 22, 1997. The
survey identified numerous areas of buried materials, but only two were of interest. A septic
tank was identified to the southeast of the building that had housed the materials testing
laboratory. Also identified was another area located approximately 75 feet to the north of the
septic tank on the east side of the building. The latter area was described as a concentration of
buried metal materials of unknown origin.
The gamma logging and the installation of the soil gas wells occurred between July 8 and July
25, 1997. The installation of the soil gas wells began in the southern part of the Deming site near
the septic tank and the area of buried materials to the north in order to determine whether
chlorinated VOCs were present in the vadose zone. The area over which vapor sampling points
would be placed was based on the results of the reconnaissance survey. The vertical placement
of vapor sampling points was based on the field reading of geologic units as indicated by natural
gamma logs. Specifically, the gamma logs were used to find permeable layers positioned below
impermeable layers. This geologic setting forms a migration pathway for contaminant vapors,
and the characterization of migration pathways is an important step in contaminant detection.
An examination of the gamma logs from several holes revealed the presence of a series of fairly
consistent layers of sandy clay or similar material beginning at about 15 to 22 feet bgs (just
below the gravel), 27 to 32 feet bgs, and finally at 38 to 50 feet bgs, particularly in those wells
near the septic system. Beneath each of these layers the subsurface materials tended to grade
into coarser, sandy materials. The gamma log signature was reviewed in the field, and from this
information certain intervals were selected for the installation of gas points. A more permeable
sampling interval (identified by lower gamma counts) was selected for each gas point location.
Examples of three gamma logs indicating the presence of clay layers and the location of vapor
sampling points are shown in Figures 7 through 9.
In Figure 7, the first vapor sampling point was placed in a screened interval between 25 and 30
feet bgs. At this level, the sampling point was located below the silty material that was present
from 20 to 25 feet bgs, shown in the gamma log where the gamma count rises through 124 cps.
The silty materials would impede the upward movement of contaminant vapors. The screen was
placed in the relatively coarser material (located just below the silty material) through which
vapors would be likely to migrate. The second and third screen intervals were similarly located
in coarser materials located just below a less permeable layer, indicated in the gamma log by
sharp increases in gamma counts.
In Figure 8, the first sampling point is located just below the thin silty layer encountered at a
depth of approximately 21 feet bgs. At this depth, it appears that there is an interval of ,
approximately 19 feet of coarse sandy material, and the screen was positioned at the top of this
interval. The second sampling point was located in a screened interval between 38 and 43 feet
bgs. At this depth, the screen is located toward the bottom of a layer of coarse materials that
extends from 31 to 47 feet bgs. The silty layer that would be expected to impede the migration
of contaminant vapors is at least eight feet above the top of the screened interval. The third
135
-------
NMSHTD District 1 Headnnarters
I GEOPHYSICAL FINDINGS
sampling point is located in a screened interval between 55 and 60 feet bgs. In this position, the
sampling point is located directly below the silty material that can be see at a depth of 53 feet
bgs, where the gamma counts rise sharply to nearly 90 cps.
Figure 9 shows the gamma log for SG09, in which only two vapor sampling points were placed.
The first point was located in a screened interval between 21 and 24 feet bgs. At this depth, the
sampling point was located directly below the silty layer seen at 21 feet where the gamma counts
peak at approximately 70 cps. The lower sampling point was placed in a screened interval
between 46 and 55 feet bgs. This screen was placed more to take advantage of the coarse
materials located in that interval than to use a distinct impermeable layer located directly above.
The coarsing of the materials in this interval can provide a migration pathway for contaminant
vapors.
136
-------
NMSTTTT) District 1 TTparimiarters
I GEOPHYSICAL FINDINGS
ndicates vapor i
sampling point
J Ft/Mi n
(A:\SG02\SG02.GB1)
Figure 7: Vapor Sampling Points in SG02 [3]
137
-------
I GEOPHYSICAL FINDINGS
NMSHTD District 1 ffpaHnnarfers
Indicates vapor
sampling point
3 Ft/Mln
(A:\SG08\SG08.GB1)
Figure 8: Vapor Sampling Points in SG08 [3]
138
-------
NMStTTH Distrirf 1
I GEOPHYSICAL FINDINGS
Indicates vapor
sampling point
.soil gas point 09
} Ft/Win
(A:\SGQ9\SGQ9.Ge2)
NGorrma
soil gas point 09
Figure 9: Vapor Sampling Points in SG09 [3]
139
-------
I GEOPHYSICAL FINDINGS
Results Validation [1,3]
NMSHTD District 1
No additional activities were conducted to validate the findings of the natural gamma logging.
The results of the soil gas sampling conducted after the installation of the permanent sampling
points served as an indirect validation. The distribution of VOC contamination that was revealed
was representative of previous sampling, and was centered around the two source areas that had
been identified by the reconnaissance survey.
I LESSONS LEARNED
There were several important lessons learned during the Deming investigation. These are
discussed below.
• Natural gamma readings should be calibrated to the site stratigraphy using site-specific
knowledge of local geology. The use of lithologic logs from existing wells can save the
time and effort that would be expended if the logs had to be generated during the same
investigation.
• Natural gamma logs appear to be more sensitive to subtle changes in stratigraphy than is
the visual inspection of lithologic logs, which often is based on the personal
interpretation of the geologist.
• The natural gamma logs were used successfully to make well point placement decisions
in the field at the time when the sampling points were being installed. In less dynamic
investigations, well point placement decisions might be delayed rather than made in the
field, potentially resulting in delays in the investigation.
• Because the natural gamma signature does not degrade or decay over time, this
information is representative for present and future investigations as well. The same
information that was used to guide sampling point locations can be used at a later date to
guide the installation of screening intervals for a soil vapor extraction system.
• Gamma logs are a useful tool for identifying interbedded impermeable layers that may be
thin and difficult to locate. This tool can be used to guide the placement of subsurface
sampling points, or screening intervals for soil vapor extraction or pump and treat
systems in geologically heterogenous materials.
140
-------
NMSHTP District 1 Headquarters
1 REFERENCES BHBBMBBBBHBHMMHBnBBHBBMBBOHMHH
1. Meixner, Richard, et al. Combination ofDownhole Gamma Logging and Soil Vapor
Measurements to Track Low Level Chlorinated Solvent Contamination in the Deep
Vadose Zone. Proceedings of the Symposium on the Application of Geophysics to
Engineering and Environmental Problems (SAGEEP 1998) pp. 369-378.
2. Personal communication with Phil Ramos, New Mexico State Highway Transportation
Department. September 24, 1998.
3. Daniel B. Stephens & Associates, Inc. Deming Surface Geophysical Survey and
Multipart Soil Gas Well Installation and Sampling. September 12, 1997.
4. Personal communication with Richard Meixner, Daniel B. Stephens & Associates, Inc.
Septembers, 1998.
5. L.W. LeRoy, editor. Subsurface Geology. Colorado School of Mines, Golden, CO.
1977.
6. Personal communication with Jim Viellenave, TEG Rocky Mountain, Inc. September
17, 1998.
7. Telford, W.M., et al. Applied Geophysics. Cambridge University Press. Cambridge,
GB. 1984.
8. Personal communication with Jim Viellenave, TEG Rocky Mountain, Inc. September
11,1998.
9. Personal communication with Jim Viellenave, TEG Rocky Mountain, Inc. September
15, 1998.
141
-------
PAGE LEFT BLANK INTENTIONALLY
142
-------
Tinker Air Force Base
Case Study Abstract
Tinker Air Force Base
Oklahoma City, OK
Site Name and Location:
Tinker Air Force Base
Period of Site Operation:
1942-present
Operable Unit:
Not applicable
Geophysical Technologies:
Electromagnetics
Seismic reflection
Seismic Modeling
RCRA Permit #
1571724391
Current Site Activities:
Portions of the base are undergoing
investigation and remediation under the
Installation Restoration Program
Point of Contact:
Sara Sayler
OC-ALC/EMR
7701 Arnold Street, Suite 221
Tinker Air Force Base
Tinker, OK 73145-9100
405-734-3058
sara.sayler@tinker.af.mil
Geological Setting:
Permian-age sedimentary rocks
overlain by Quaternary alluvium, sand
dunes, and terrace deposits
Technology Demonstrator:
IT Corporation
312 Directors Drive
Knoxville,TN 37923-4799
Phone - 423-690-3211
Fax - 423-690-3626
Purpose of Investigation:
To help identify and map possible conduits for preferential groundwater flow in the shallow subsurface. The site-specific
decision being supported was to obtain additional information in planning the optimal placement of the installation of
groundwater recovery wells.
Number of Images/Profiles Generated During Investigation:
17,510 linear feet of seismic profiles collected along 8 survey lines. This case study focuses on the modeled and interpreted
compressional (p-wave) results from a portion of Line 4 and an intersecting portion of Line 5, These lines were selected to
represent typical site data and anomalies. This intersection is also the location of an interpreted sand channel near the
intersection of the two lines, making it easier to demonstrate the continuity of the interpreted sand channel.
Results:
Interpretation of the seismic data indicates several places where sand channels cut into the Upper Saturated Zone/Lower
Saturated Zone (USZ/LSZ) aquitard. The seismic reflection survey results will be used to recommend the placement of Phase
II recovery well drilling locations. The results of Phase I groundwater recovery well yield tests indicated good correlation
with several of the seismic anomalies identified in the target zones by the seismic survey. Seismic modeling was conducted to
provide support for the interpretation of the seismic results. A EM-31 terrain conductivity survey was conducted along the
eight seismic lines to screen for large scale anomalies caused by metallic objects that might interfere with the seismic survey.
Most of the anomalies were due to surface metal, such as the chain-link fences, nearby structures, and monitoring well
monuments.
143
-------
Tinker Air Force Base
EXECUTIVE SUMMARY I
Tinker AFB covers 4,277 acres and is located on the southeast edge of the Oklahoma City metropolitan
area. The base is situated within the North Canadian River drainage basin and drains into the Crutcho
and Soldier Creeks and overlies a complex aquifer system that includes the Garber-Wellington
Formation. The Southwest Quadrant Stabilization System (SQSS) Area is the location of two landfills
that were used sporadically for disposal over a forty-year span from the 1940s to the late 1960s for
disposal of sanitary and industrial wastes, including paints and solvents.
Near-surface geology at Tinker AFB consists of clays and clayey silts that are interbedded with thin,
clayey sand layers, reaching a maximum thickness of approximately 60 feet in the western and
southwestern parts of the base. The deeper geology is comprised of mostly unconsolidated materials,
which are composed of predominantly fine-grained sandstone, with lesser amounts of siltstone and shale.
Bedrock formations dip to the southwest by approximately 0.5 degrees, or by 40 to 50 feet per mile.
Groundwater occurs at the site in four water-bearing units, but only the surficial unit was the target for
this study. Groundwater can occur at depths as shallow as 20 feet, but public water supplies are drawn
from depths of greater than 400 feet.
The purpose of the seismic survey was to locate permeable layers in the subsurface that might indicate
preferential pathways for groundwater flow. This information is being used to site new extraction wells
for the groundwater pump and treat system. Seismic methods were chosen as a cost-effective method for
gathering information on the subsurface stratigraphy. The geology of the area is highly complex and
other investigative methods, such as soil borings, would have yielded less information at a higher cost.
Two geophysical methods were used during this investigation: electromagnetic (EM) reconnaissance
survey, and a seismic survey. The EM survey was conducted to screen for subsurface conditions that
might cause interference in the seismic data collection. The seismic survey was conducted to identify
conductive layers in the subsurface that might be paths for groundwater migration. Seismic modeling
was conducted to provide analytical support for the interpretation of the seismic results.
The seismic survey revealed the presence of sand channels that were incised into the uppermost aquitard
and sand lenses located within that aquitard. Seismic modeling significantly improved the investigator's
understanding of the seismic anomalies that were found by providing an analytical benchmark against
which to compare the seismic results. Strong correlation was found between the location of significant
seismic anomalies and known groundwater flow pathways.
The target structures were relatively shallow and groundroll effects were not a significant source of
interference. Poor surface conditions and cultural sources did, however, posed difficulties to data
collection. Several seismic data processing techniques, such as refraction statics, spectral whitening, and
mute analysis, significantly reduced the level of interference in the data and allowed for increased frequency and
resolution of the seismic results.
144
-------
Tinker Air Force Base
SITE INFORMATION!
Identifying Information
Tinker Air Force Base (AFB)
Oklahoma City, OK
RCRA Permit* 1571724391
Southwest Quadrant SMU
Background [1,2]
Physical Description: Tinker AFB covers 4,277 acres and is located on the southeast edge of the
Oklahoma City metropolitan area (see Figure 1). The base is bordered by Sooner Road to the
west, Douglas Boulevard to the east, interstate highway 1-40 to the north, and SE 79th street to the
south The base is situated within the North Canadian River drainage basin and drains into the
Crutcho and Soldier Creeks. The base overlies a complex aquifer system that includes the
Garber-Wellington Formation .
* Tinker A.F.B.*
The topology of the seismic study area, located in ,^^mm^™_«»,
the southwest portion of the base, is characterized
by nearly level plains to gently rolling hills, with
surface elevations ranging from 1,240 to 1,270 feet
above mean sea level. The surface consists of
alluvial soils near streams and flood plains, and *>„
residual soils resulting from weathered bedrock [1].
Site Use: The Southwest Quadrant Stabilization Fi§ure 1: Site Location
System (SQSS) Area is the location of two landfills
that were used sporadically for disposal over a forty-year span from the 1940s to the late 1960s.
Landfill #2 was used during the 1940s and 1950s for disposal of sanitary and industrial wastes,
including paints and solvents. Landfill #4 was used from 1961 to 1968 for the disposal of
drummed solvents, and sludges from petroleum and solvent storage tanks.
Release/Investigation History: On-site disposal of industrial wastes occurred from 1942 until
1979 when off-site disposal became the standard disposal practice. Organic solvents, including
trichloroethylene (TCE), tetrachloroethylene, and 1,2-dichloroethylene, were used for degreasing
and aircraft maintenance. In the past, waste oils, solvents, paint sludges, and plating waste
generated from maintenance activities were disposed in Industrial Waste Pits Numbers 1 and 2,
located about 1 mile south of Soldier Creek and Building 3001. hi 1997, a groundwater treatment
system was installed to treat contaminated groundwater.
Regulatory Context: Actions at this site are being undertaken in compliance with Federal and
State regulations under the Resource Conservation and Recovery Act (RCRA).
145
-------
SITE INFORMATION!
t" Ail* f?nrr*t
Area of Case Study Interest
Scale
=^
0 500 1000 Feet
Figure 2: Site location map with seismic survey lines [1].
146
-------
Tinker Air Force
SITE INFORMATION I
Site Logistics/Contacts
Federal Lead Agency:
United States Air Force
Federal Oversight Agency:
Environmental Protection Agency
Site Contact:
Sara Sayler
OC-ALC/EMR
7701 Arnold Street, Suite 221
Tinker Air Force Base
Tinker, OK 73145-9100
405-734-3058
sara.sayler@tinker.af.mil
I MEDIA AND CONTAMINANTS!
Matrix Identification
Remedial Project Manager:
Ruby Williams
EPA Region 6
214-665-6733
1445 Ross Avenue Suite # 1200
Dallas, TX 75202-2733
williams .ruby@epa.gov
Geophysical Subcontractor:
IT Corporation
312 Directors Drive
Knoxville, TN 37923-4799
423-690-3211
Type of Matrix Sampled and Analyzed: Subsurface soil/Bedrock
Site Geology/Stratigraphy
Near-surface geology at Tinker AFB consists of the Permian-age Hennessey Group and the
Garber-Wellington Formation. The Hennessey Group is composed of clays and clayey silts that
are interbedded with thin, clayey sand layers, reaching a maximum thickness of approximately 60
feet in the western and southwestern parts of the base. The Hennessey Group is underlain by the
mostly unconsolidated materials of the Garber Formation, which are composed of predominantly
fine-grained sandstone, with lesser amounts of siltstone and shale. The deeper Wellington
Formation has a similar lithology to the Garber Formation, and together the two comprise the
1,000-foot thick Garber-Wellington Formation. Bedrock formations dip to the southwest by
approximately 0.5 degrees, or by 40 to 50 feet per mile.
Groundwater occurs at the site in four water-bearing units, that include the Hennessey Water
Bearing Zone (HWBZ), the Upper Saturated Zone (USZ), the Lower Saturated Zone (LSZ), and
the Producing Zone (PZ). The HWBZ consists of fine-grained sediments with very low
transmissivity and large vertical hydraulic gradients. Beneath the HWBZ, the USZ is the
uppermost waterbearing zone of the Garber-Wellington aquifer. The USZ is made up of
permeable sand channels and lenses. It is generally believed that the HWBZ and the USZ are not
hydraulically connected. The USZ/LSZ aquitard is comprised of overlapping clay layers with
interbedded thin sand lenses and is not of uniform thickness. The aquitard ranges in thickness
from less than 10 feet to more than 25 feet, with the base of the aquitard occurring at a depth of
147
-------
Tinker Air Force Rase
I MEDIA AND CONTAMINANTS*
approximately 110 feet below ground surface (bgs) in the southwest portion of the study area.
The LSZ ranges between 140 and 200 feet in thickness, and is separated from the underlying PZ
by the 30- to 100-foot-thick LSZ-PZ aquitard. The PZ extends from between 210 and 280 feet
bgs to more than 1,000 feet bgs, and is used as the primary source of groundwater on the base
and elsewhere in the Oklahoma City area. Groundwater can occur at depths as shallow as 20
feet, but public water supplies are drawn from depths of greater than 400 feet.
Contaminant Characterization ^, ,
Primary Contaminant Groups: Volatile organics such as TCE.
Matrix Characteristics Affecting Characterization Cost or Performance [1,2]
Certain subsurface structures, such as utilities and buried metallic objects, can cause interference,
or noise, in the seismic data. To screen for such structures, frequency-domain electromagnetic
(EM) data were acquired along the seismic lines. No significant sources of subsurface
interference were found.
Significant interference with the seismic data collection was caused by muddy surface conditions
and standing water along portions of the survey lines, particularly along Line 8. The seismic
energy was significantly attenuated as it passed through the saturated soils, resulting in poor
seismic data quality along those lines. Other surficial sources of interference included cultural
noises, such as pumps, vehicles, wind and aircraft.
Groundroll can cause interference in data collection and interpretation. Groundroll is caused by
seismic waves that travel horizontally toward the geophone sometimes obscuring the collection
of seismic waves originating from deeper structures. During the investigation at Tinker AFB, the
the high velocity surface materials reduced interference from the slow groundroll and the targets
of principal interest were shallow and mostly located outside of the groundroll noise cone.
I GEOPHYSICAL INVESTIGATION PROCESSI
Investigation Goals \2,3]
To help identify and map permeable materials in the USZ and the USZ/LSZ aquitard that might
indicate preferential groundwater flow in the shallow subsurface. This information was used to
place extraction wells to optimize the groundwater extraction system. The primary target of
interest was near-surface sand channels and lenses within the USZ aquifer and the USZ/LSZ
aquitard which may form preferential flow channels in the subsurface [2].
148
-------
Tinker Air Force Rase
(GEOPHYSICAL INVESTIGATION PROCESSl
Geophysical Methods [1, 2]
Two geophysical methods were used during this investigation: electromagnetic (EM)
reconnaissance survey, and a seismic survey. The EM survey was conducted to screen for
subsurface conditions that might cause interference in the seismic data collection. The seismic
survey was conducted to identify conductive layers in the subsurface that might be paths for
groundwater migration.
The EM survey was performed to detect sources of potential interference to seismic data
collection. The EM survey sought to identify variations in the electrical conductivity of
subsurface materials that might be caused by buried objects, conductive fluids, or geologic
discontinuities. By artificially applying a known electromagnetic field to the ground surface by
means of a transmitter, investigators measure the presence of disruptions to the known
electromagnetic field with a receiver. These disruptions, termed EM anomalies, can result from
either geological changes or the presence of metallic objects, such as pipes, drums, cables, tanks,
etc., in the subsurface. The EM survey conducted at Tinker AFB was used to identify buried
materials that might interfere with seismic survey by scattering or attenuating seismic waves. A
Geonics Limited EM-31 terrain conductivity meter coupled to an Omnidata DL720 digital data
logger was used to collect quadrature-phase and in-phase component data along the length of
each seismic line in the survey area.
The seismic reflection method was used to collect seismic data in the subsurface with which
permeable layers in the subsurface can be identified. These permeable layers may act as
groundwater migration pathways, and may be good locations for future extraction wells. In a
seismic reflection survey, an artificial seismic source is used to create an acoustic wave that
propagates downward through the soil layers. When the wave reaches a soil layer whose seismic
conductivity is significantly different from that of the overlying soils, a portion of the wave is re-
directed to the surface. A geophone, or electromechanical transducer, is used at the surface to
receive the reflected wave energy. Subsurface stratigraphy is then mapped by measuring the
travel time necessary for a wave to pass through one layer to another, refract along the interface,
and return to the geophones at the surface. Seismic field equipment used to conduct the survey
consisted of three 48-channel Geometries Strataview® seismographs in a master-slave
configuration, totaling 144 channels. Single, 40-Hz vertical geophones were used for collection
of p-wave data.
Seismic modeling was conducted to provide support for the interpretation of the seismic results.
Seismic models were developed to depict the anticipated seismic response of various types of
subsurface stratigraphy that might be encountered in the study area, i.e. sand channels cutting
into the aquitard or sand lenses embedded in a clay layer. Well lithologies and sonic logs
acquired in wells located within the survey area were used to develop estimates of the seismic
velocities of the various soil types found within the study area. These estimates were used to
construct hypothetical seismic models of subsurface structures of varying thickness and
composition. When seismic anomalies were encountered in the survey data, the actual seismic
response was compared to the modeled response of different stratigraphic features to help
identify the type of subsurface stratigraphy that might create such an anomalous response. The
149
-------
Tinker Air Force
• GEOPHYSICAL INVESTIGATION PROCESSl
use of seismic models greatly aided investigators in their interpretation of the seismic results by
providing a set of benchmarks against which actual results could be compared [4].
Seismic lines were chosen to satisfy three criteria:
• The survey area should include parallel and perpendicular coverage of a known geologic
strike. The lines within the area were located so that velocity data could be acquired in
at least one well on at least one line of the survey, and each of the lines were to have at
least one tie with another line in the survey.
• The survey area should include areas in which known contaminant plumes were present;
and
• The survey lines should be placed in areas whose groundwater is under hydraulic control
from the groundwater pump and treat system.
Prior to collecting geophysical data, each seismic survey station was geospatially surveyed using
a Global Positioning System (GPS). The ability to geospatially reference the seismic profiles
allowed investigators to understand the relationships between the individual seismic profiles and
the larger site geology. Horizontal and vertical geospatial accuracies were kept within 0.5 feet
and 0.1 feet, respectively.
A field test of the seismic parameters was used to evaluate the relative merits of collecting
different types of seismic wave during the survey, and to determine the optimal distance between
the seismic source and the geophones. Data on two types of seismic waves: compressional (p-
wave) and shear(s-wave) wave, were acquired along a short test section of Line 5 to evaluate and
compare the results such surveys would produce at the site. While p-wave data, the seismic wave
that is projected downward, were less complex to collect and interpret, s-wave data often provide
a higher resolution. S-wave data are collected as the seismic energy is transmitted horizontally
from the source to the receiver. Along this test section, p-wave data were recorded using vertical
source impacts and vertical geophones, and s-wave data were recorded using horizontal impacts
and horizontal, s-wave, geophones. Based on the noise tests conducted in the field and the depth
of the target, a 5-foot station and 10-foot shotpoint (energy source) spacing were used for the
survey. The investigators decided that the p-wave data would provide sufficient resolution to
identify the targeted subsurface structures, and, therefore, no additional s-wave data were
collected.
150
-------
Tinker Air Force Base
[GEOPHYSICAL FINDINGS I
Technology Calibration [1, 2]
The calibration needed for a successful seismic survey is to establish the relationship between the
depth of an anomaly in the subsurface and the time it takes a seismic wave to propagate to that
anomaly and return back to the surface. In other words, the seismic time must be "tied" to depth.
To establish this link at Tinker AFB, investigators collected vertical seismic profiles (VSPs) and
sonic logs. The sonic logs were used to construct synthetic seismograms. A synthetic
seismogram is a statistical comparison of seismic velocity, soil density, and depth values used to
convert seismic velocity data into depth.
The VSP data were acquired in two monitoring wells with maximum depths of approximately
150 feet in order to better understand the subsurface velocities. For each profile, a geophone is
locked in a well at regular depth intervals and used to record the energy from a surface source at
each interval. The time lapse recorded between source and receiver is a measure of the time
necessary to go from the surface to the geophone in the well is displayed as a time versus depth
graphic.
Data from an existing sonic log in a nearby well was also used to link the seismic time data to
depth. Together, the sonic log and the VSP data, were used to generate a synthetic seismogram.
The seismogram provides a correlated display of seismic velocities, time and known depths to
reflectors. These correlations establish the link between the seismic velocity of certain
subsurface materials and the depths at which those materials were encountered.
Investigation Results [1,2]
The EM-31 reconnaissance survey was conducted along the eight seismic lines and revealed
large-scale anomalies caused by metallic objects. Most of the anomalies were caused by surface
metal, such as the chain-link fences, nearby structures, and monitoring well monuments.
Subsurface anomalies were also identified as subsurface pipelines that cross the area, such as a
north-to-south trending pipe that exists in the western part of the site.
More than 17,000 linear feet of seismic data were collected along eight survey lines and the
resulting profiles identified four reliable locations for future extraction wells. For the purposes of
this case study, however, the discussion presented focuses only on the interpreted and modeled results from
the intersecting portions of Lines 4 and 5, as shown in Figure 2 on page 3. The seismic results along both
lines showed a sand channel near the intersection of the lines, increasing the reliability the interpreted sand
channel and its continuity. None of the EM anomalies, discussed above, were located within this
area [3].
Muddy surface conditions and standing water along portions of the survey lines caused
significant variation in and interference with the quality of the seismic data collected. Two
statistical solutions were applied to improve the quality of the data. Refraction statics, proved
most effective in minimizing the noise in the data. The adjusted data had less variation and
improved resolution.
151
-------
I GEOPHYSICAL FINDINGS!
Tinkfi* Air TTnrpp
The seismic reflection survey was conducted along each line using a 5-foot interval between
geophones and a 10-foot shotpoint interval. Data were acquired using a 0.5-millisecond (msec)
sampling rate; the record length was 1,024 msec. As data quality warranted, source impacts per
shotpoint were adjusted along each line.
The seismic profile generated along Line 4 at its intersection with Line 5 is shown in Figure 3. In
this profile, the top of the USZ/LSZ aquitard was interpreted to,be. at approximately 70-75 feet
bgs (25 msec), and the bottom of the aquitard was interpreted to be at approximately 110-115
feet bgs (35 to 40 msec). Although only the USZ was targeted for this study, other structures can
be seen in Figure 3, such as the base of the LSZ, where the LSZ/PZ aquitard occurs, at
approximately 295 feet bgs (75 msec). One significant anomaly can be seen in this Figure,
centered on Station 295, and is outlined by hash marks. This feature was interpreted to be a large
sand channel within the upper portion of the USZ/LSZ aquitard which is laterally continuous and
was considered to be part of a larger structure that can be seen nearby on the Line 5 section. The
channel was presumed to trend north to northeast, roughly in line with the high-yield B6 and B7
recovery wells. The seismic model data indicate that this anomaly could be caused by the
presence of a low conductivity materials (i.e. sand) embedded in higher conductivity materials
(i.e. silts and clays). Above this channel, the seismic data indicate the presence of a low velocity
medium, likely a sand within the USZ aquifer. This channel was suggested as a good location for
a future extraction well.
The seismic profile generated along Line 5 at its intersection with Line 4 is shown in Figure 4.
The bottom of the USZ/LSZ aquitard can be seen at a depth of 110-115 feet bgs and a noticeable
low in this area (represented by the dashed line)-indicated the presence of materials with similar
seismic velocity incised into the base of the aquifer, and/or the accumulation of slower velocity
materials locally, such as would be expected from a sand channel. Two anomalies appear in the
Line 5 data shown in Figure 4. A broad and subtle anomaly extending between stations 680 and
693 at a depth of approximately 75-80 feet bgs was interpreted as a small incised sand lens at the
bottom of the USZ aquifer. The feature extending from station 650 to 667 at a depth of 80-90
feet bgs was interpreted to be part of the same large sand channel that occurs along Line 4.
A seismic interpretation map of the entire survey area with the locations of interpreted channels
and lenses above and within the upper portion of the USZ/LSZ aquitard is shown in Figure 5 A.
Several channels were interpreted near the bottom of the USZ/LSZ aquitard, and several deep
channel systems were interpreted within the LSZ (not discussed here). There is a substantial
concentration of interpreted lower aquitard channels in the southern and southeastern portion of
the survey area, and deep channel systems in the central and southeastern portions of the survey
area.
152
-------
Tinker Air Force
I GEOPHYSICAL FINDINGS^
Results Validation [1]
Seismic data indicating areas of high hydraulic conductivity were compared to well yield test
results from the existing Phase I recovery wells which were drilled to an average depth of 80 to
90 feet. All of the Phase I wells were drilled in locations based on engineering factors, plume
location, and groundwater flow direction. Wells A8, B3, B6, and B7 were drilled near anomalies
identified by the seismic data discussed in this case study. Figures 5A and 5B show the
correlation between higher yield zones, as determined with pump tests on.the.Phase I wells, and
the locations of the sand channel identified at the intersection of Lines 4 and 5. On Line 5.
anomalies centered on Station 658 correspond to high yields on the B6 and B7 recovery wells.
Another anomaly found along Line 5 and interpreted as a sand channel, centered on Station 205.
corresponds to the high yield recovery well B3. On Line 6, the anomaly centered near Station
390 corresponds to the high yield A8 recovery well. Several other interpreted sand channels
have not been verified at this time. These areas present target locations for possible future Phase
El recovery wells.
Further validation was provided by sonic logs taken in five of the extraction wells that were
located on or near the seismic survey lines. These helped to confirm reflector identification and
also demonstrated good correlation between the seismic findings and well tests [3].
153
-------
Legend
TopoFUSZ/LSZAquitard
Channel / Lens at Hie Top of or Within USZ/LSZ Aquilaul
BollomofUSZ/LSZAquilard
Interpreted Channel Extending into LSZ
Inteq>reted Location of LSZ/PZ Aquitard
SW
SP.-2GOO
- Line 4 - Interpretatcd Reflection Section (Stations 260 - 330)
2650
2?0.0
276.0
260,0
2850
2SOO 2D5.0
3000
3050 310.0
315.0
.3200
325.0
NE
3300
«w«^^^
^'•'M'hKlU.;tt^*^^
^ •M&Mhlhk&tkVktfcikfck'liVv^ vk1kk'h^'vVk^.kVCwk^VwS^kk>.\^^kii^
101)
i^l}),^)^^^^^^)^^^
IU50- ft)l!J|^Ji)^^
Figure 3: Interpreted Reflection Section on Line 4 [2].
II
oo
H-<
n
O
Gfl
-------
I GEOPHYSICAL FINDINGS I
Tinker Air Force Base
•If
•< •&
5 .5
5 -o SP
CO
•O
cu
"S
a.
cu
Tt
CU
(aas)
155
-------
Tinker Air Force Rase
I GEOPHYSICAL FINDINGS I
FIGURE 5ft
SEISMIC INTERPRETATION MAP
LEGEND».
1—^>.
88 V
NOTES*
SE1SMIC LINES SWJVING STATIONS
(NOT SHOWN ON
TINKER AIR FORCE BASE PROPERTY LINE
RECOHENOEO PHASE 1J RECOVERY *£LL LOCATION
PHASE ! RECOVERY WELL LOCATIONS
FIGURE 5B
CONTOUR MAP OF PHASE I RECOVERY WELL YIELD TEST RESULTS
SEISMIC L[NcS SHOUINC STATIONS
PHAS£ 1 RECOVERY WELL LOCATIONS
TEST
TEST
TEST
^lELD > S GPM
?IELD 4 - 5 CPU
riELD 3 - 4 GPM
1£ST YIELD 2 - 3 CPM
TEST YIELD 1 - Z 0PM
TEST YIELD < 1 GPM
SCALE
-'-u""
soo i6oc FH:ET
Figure 5: Location of Significant Seismic Anomalies and Correlation with Well Yields [2]
156
-------
Tinker Air Force Base
I LESSONS LEARNED
Lessons learned at the Tinker APB study site include the following:
• The seismic survey results identified several areas in which permeable zones in the
subsurface are located and which may be favorable locations for future well installation.
Four of these sites represent locations where the highest potential for drilling into
significant sand channels is thought to exist [1].
The EM survey successfully identified large scale anomalies caused by metallic objects,
such as the chain-link fences, nearby structures, and monitoring well monuments. The
EM anomalies found did not present a problem for the seismic data quality.
• Existing well yield data correlates well with several of the anomalies that are interpreted
as channels. Incorporation of recently acquired sonic log data and lithologic logs from
extraction wells drilled near any of the seismic lines with the seismic data will be
particularly useful for refining stratigraphic and depth correlation [1].
• The relatively high seismic velocities in the unconsolidated sediments at this site reduce
the spatial resolution that can be attained from the data. S-wave data should increase
resolution compared to the p-wave data. However, for this site, the s-wave data proved
inferior when compared to p-wave data, especially when the additional cost for acquiring
and processing the s-wave data is considered[2].
• The results developed for this site are only valid for two-dimensional cross sections of
the subsurface beneath each seismic line. If delineation of the spatial distribution of
features between lines is required, the acquisition of three-dimensional data should be
considered at this site [2]. Three-dimensional seismic techniques were the preferred
method, but due to the large areal extent of the survey area and associated data
acquisition and processing costs, the two-dimensional method was used [1].
The application of seismic data processing algorithms, such as refraction statics and
spectral-whitening, reduced the level of interference in the data. This, combined with
thorough velocity and mute analysis along the seismic lines, allowed for increased
frequency and resolution of the seismic results [1].
Seismic results often reveal a number of anomalous results attributable to a large variety
of conditions, such as poor surface conditions, interference from cultural sources, or
variation in seismic wave generation. These anomalies may, on the other hand, represent
the target structures. The use of seismic models in this survey aided the investigators by
helping them quickly identify whether the anomalous results were due to difficulties in
data acquisition or target structures. Moreover, by comparing seismic anomalies to
model results, investigators were able to refine their interpretation of the anomalous
responses by helping them to distinguish between different lithologic changes, such as a
discontinuity in a clay layer and a sand channel incised into the clay layer [4].
157
-------
i REFERENCES!
Tinker Air Force Base
Hackworth, Jeffrey B., Sayler, Sara, Steensma, Gilein J., and Marcum, David W.
Seismic Reflection Survey For Mapping Ground-water Migration Pathways at Tinker Air
Force Base, Oklahoma. Proceedings of the Symposium on the Application of
Geophysics to Engineering and Environmental Problems (SAGEEP) 1998.
IT Corporation and Steensma, G. J. 2-D Seismic Reflection Survey Southwest Quadrant
Stabilization System Area Tinker Air Force Base Oklahoma City, Oklahoma. Draft
report. June 1997.
Personal communication with Sara Sayler of Tinker Air Force Base. August 25, 1999.
Personal communication with Sara Sayler of Tinker Air Force Base. December 17
1999.
158
-------
Trail Road Landfill
Case Study Abstract
Trail Road Landfill
Nepean, Ontario, Canada
Site Name and Location:
Trail Road Landfill
Nepean, Ontario, Canada
Period of Site Operation:
Early 1980's to the present
Operable Unit:
Not applicable
Geophysical Technologies:
Natural gamma
Maghetometry
Electrical conductivity
Density
Temperature
CERCLIS #
Not applicable
Current Site Activities:
The Nepean Landfill is capped and closed.
Stages land 2 of the Trail Road Landfill
are capped and closed. Stage 3 is currently
being filled and stage 4 is ready to be
opened. Stages 3 and four have leachate
collection systems. In general,
groundwater is monitored 3-4 times a
year for chemicai contamination.
Point of Contact:
Keith Watson, 613-838-2799
Darin Abbey, 604-291-5429
C. Jonathan Mwenifumbo, 613-996-
2312
Geological Setting:
A complex mixture of sand, gravels,
and silt overlying a lacustrine clay
plain. Limestone bedrock underlies a
glacial till deposit of sand and gravel
which lies under a silty clay layer.
There is a shallow aquifer which
discharges into a deep aquifer.
Technology Demonstrator:
Darin Abbey, Carleton University,
Ottawa, Canada,
Purpose of Investigation:
The overall goal of this investigation was to show that leachate plume delineation could be accomplished through interpreting
data from a full suite of geophysical logs.
Number of Images/Profiles Generated During Investigation:
Eight composite profiles illustrating the results of the logs from each of the above mentioned technologies.
Results:
The use of geophysical measurements from boreholes can provide a continuous vertical profile of the geology and
hydrogeology. This information can be used to understand the factors controlling the groundwater composition, and
ultimately leachate movement in the subsurface. The geophysical techniques overcome the traditional monitoring limitation
of fixed vertical sampling positions for chemical analytes.^
159
-------
Trail Road Landfill
[EXECUTIVE SUMMARY I
The Trail Road and Nepean Landfill sites are located within the Region of Ottawa-Carleton, Canada,
with a population of 750,000. The site, approximately 500 acres, is surrounded by light industry, and
farmland. The Nepean Landfill began operation in the early 1960s and accepted waste until the early
1980s when it was deemed nearly full and the Trail Road Landfill was opened. The Trail Road Landfill
is currently serving as a municipal sanitary landfill accepting non-hazardous waste including residential
garbage, construction, commercial, institutional, and light industrial waste.
Leachate, believed to originate from the unlined Nepean Landfill and the stages 1 and 2 of the Trail Road
Landfill, has been detected in the groundwater below the site. The leachate consists of a complex
mixture of organic and inorganic constituents as well as elevated levels of calcium, magnesium, chloride.
sulphate, potassium, ammonia, other nitrogen compounds, other dissolved organic carbons, phenols, and
iron.
The landfill site is positioned on a glacial outwash plain which has a complex mixture of sands, gravels,
cobbles, clays, and silt. The surface soil consists of a discontinuous dense layer of silt and clay
(approximately two meters) beneath which is a layer of sand and gravel which overlies a limestone
bedrock forming a deep aquifer, present at a depth of 10 to 30 meters. A clay layer is present beneath
part of the Trail Road Landfill site. The clay layer separates the sand and gravel ridge into an upper and
lower aquifer.
A geophysical investigation was conducted at the landfill to demonstrate an innovative method for
monitoring a landfill leachate plume. The information contained in this report was extracted from the
interpretive report of the investigation. Six different geophysical methods were combined in borehole
applications to collect the geophysical data for this investigation. The six methods were: natural gamma.
gamma-gamma, total magnetic and magnetic susceptibility, electrical conductivity and temperature.
Geophysical logs were developed in eight existing monitoring wells.
The geophysical logs correlated well with existing lithologic logs, and identified the presence of a
surficial clay layer and a perched aquifer on that layer where leachate may collect. Logs of the deeper
aquifer generally showed little evidence of contamination, with the exception of one well, in which a
significant anomaly was detected. The conductivity and temperature logs were interpreted to sho\v the
presence of leachate contamination in this one well.
Lessons learned at the Trail Road Landfill site were that the major advantage of geophysical logs over
traditional sampling techniques is that they provide a continuous representation of the subsurface
conditions. The logs can provide a measurement of total dissolved solids as a proxy for ions in water. A
major failing of the traditional sampling approach is the fixed vertical screen position. Although actual
chemical identification cannot be done by geophysical methods, groundwater with anomalously high
conductivities would indicate the need for chemical analyses. These examples show the need for
conductivity data to be interpreted in conjunction with other geophysical measurements to illustrate the
anomaly in conductivity at a certain depth within a well.
160
_
-------
Trail Road Landfill
I SITE INFORMATION
Identifying Information
Trail Road Landfill
Environment and Transportation Department
Solid Waste Division
Region of Ottawa-Carleton
4475 Trail Road, R.R. #2
Richmond, Ontario, KOA 220
CANADA
Background
Physical Description: The Trail Road Sanitary Landfill site, which includes the Nepean and
Trail Road landfills, is located within the Region of Ottawa-Carleton, Canada, with a population
of 750,000. The site, approximately 500 acres, is surrounded by light industry and farmland. The
terrain consists of grasslands and light forests. Running tangent to the eastern side of Trail Road
Landfill is Highway 416. Likewise the southern side is bordered by a lesser road, Trail Road,
which also borders the northeastern side of the Nepean Landfill (which is located southwest of
the Trail Road Landfill). Moodie Drive runs along the western boundary of the Nepean Landfill.
The south end of the entire site is bordered by Barnsdale Road and Cambrian Road runs
northeast through the northern boundary of the site, but is not immediately adjacent to the
landfills (see Figure 1)[1, 2, 3]. South of the Trail Road Landfill, there is a sand and gravel ridge
which serves as a divide for surface water runoff. Surface water flows from this ridge to either
the north or the south. For the Trail Road Landfill, the general site surface water flow is in a
north to northeasterly direction but is interrupted by site excavations.
The Nepean Landfill began operation in the early 1960s and accepted all landfill waste until the
early 1980s when it was deemed nearly full and the Trail Road Landfill was opened. Thereafter,
until is was capped in 1993, only construction waste was disposed of in the Nepean Landfill.
This landfill is not lined but it is capped with a polyethylene liner and soil [1].
Site Use: The Trail Road Landfill is currently serving as a municipal sanitary landfill accepting
solid non-hazardous waste including residential garbage, construction, commercial, institutional,
and light industrial waste. The Trail Road Landfill was opened in 1980 and has been
continuously operated in stages (see Figure 1). The first two stages are closed and capped with
polyethylene and soil but are not lined and do not have leachate collection systems. Stage 3 was
constructed with a 60 centimeter (cm)- thick competent clay and a high density polyethylene
liner. The third stage, which opened in 1991, is nearly full, and will be capped with a
polyethylene liner and soil. Stages 3 and 4 have leachate collection systems. Stage 4 is not yet
operational [1,2].
161
-------
Trail Road Landfill
SITE INFORMATION
Release/Investigation History: Leachate, believed to originate from the unlined Nepean
Landfill and the stages 1 and 2 of the Trail Road Landfill, has been detected in the groundwater
below the site. The leachate consists of a complex mixture of organic and inorganic constituents
as well as elevated levels of calcium, magnesium, chloride, sulphate, potassium, ammonia, other
nitrogen compounds, other dissolved organic carbons, phenols, and iron [3].
The groundwater is monitored on a variable basis. All wells are monitored up to 3 times a year
for indicators including chloride, boron, bromide, BOD, DOC, and iron [1,2].
Regulatory Context: Not Applicable
Figure 1: Site Map [3] [Poor Quality Original]
162
-------
Trail Road Landfill
SITE INFORMATION
Site Logistics/Contacts
Site Contact:
Keith Watson
Regional Municipality of Ottawa-Carleton
4475 Trail Road, R.R. #2
Richmond, Ontario, KOA 220
(613) 838-2799
I MEDIA AND CONTAMINANTS!
Matrix Identification [3,5] _^_
Geophysical Investigator:
Geological Survey of Canada
Mineral Resources Division
601 Booth Street
Ottawa, Ontario, Kl A OE8
CANADA
Type of Matrix Sampled and Analyzed: Subsurface clays, cobbles, sands, and gravels
Site Geology/Stratigraphy [3, 5]
The landfill site is positioned on a glacial outwash plain which has a complex mixture of sands,
gravels, cobbles, clays, and silt (Figure 2). A discontinuous dense layer of silt and clay
(approximately two meters in thickness) separates two aquifers. The silt and clay layer is
complete under the Nepean Landfill but not under all of the Trail Road Landfill and acts as an
aquitard to a perched aquifer.
NORTH
120-
110-
100-
90-
80-1
70 J
Oewaterinq
Pond -
h«— Trail Road Landfill -H
Cambrian
Road
ih
! I
Nepean Landfill
SOUTH
i
Legend
Refuse
I I Fin« to medium sand
smandclay
Coarse sand and gravel
I — 4 Limestone bedrock
Figure 2: North to South Cross Section of Site [3]
Approximately 500 meters from the northern boundary of Trail Road Landfill on the north side
163
-------
Trail Road Landfill
I MEDIA AND CONTAMINANTS!
of Cambrian Road is a large de-watering pond used to catch the local groundwater discharge.
The pond water eventually discharges into the Jock River which is located approximately 1 km to
the north. Southwest of Trail Road is the Nepean Landfill. Surface water runoff flows in a south
to southwesterly directly from Trail Road [2, 3].
There are two aquifers, separated by clay, underlying the entire site. A shallow sand aquifer
flows in a north to northeasterly direction under the Trail Road Landfill. Surface water
penetration creates a shallow groundwater flow in a south to southwesterly direction under the
Nepean Landfill. The deep aquifer, located in a layer of bedrock at a depth ranging from 10-30
meters flows in a south to north direction[2].
Contaminant Characterization
Primary Contaminant Groups: The contaminants consists of chemicals within groundwater
from landfill leachate. The leachate consists of a complex mixture of organic and inorganic
constituents, and is produced by the percolation of water through the waste, which dissolves and
suspends some of the chemicals by chemical reaction. The leachate has elevated concentrations
of calcium, magnesium, chloride, sulphate, potassium, ammonia, other nitrogen compounds,
other dissolved organic carbons, phenols, and iron [3]. The leachate has characteristically high
conductivity, hardness, alkalinity, and total dissolved solids (TDS). Exothermic reactions within
the landfill can cause elevated groundwater temperatures.
Matrix Characteristics Affecting Characterization Cost or Performance [3, 5]
The density readings taken for sediments above the water table contained a low bias because the
density calibration logs assumed a water-filled well. No other characteristics of this site affected
the performance of the geophysical technologies.
I GEOPHYSICAL INVESTIGATION PROCESS
Investigation Goals [3,5]
The purpose of this study was to show that by measuring the physical properties of the
subsurface, borehole geophysics can refine the hydrogeological interpretation of the landfill site.
The interpretation of gamma ray, density, magnetic susceptibility, total magnetic field, electrical
conductivity and temperature logs can serve to refine the understanding of the underlying
geology and the existence of a leachate plume. Borehole geophysics can also be used to
delineate areas of leachate contamination with greater efficiency than sampling and chemical
analysis of analytes.
164
-------
Trail Road Landfill
I GEOPHYSICAL INVESTIGATION PROCESS!
Geophysical Methods [3, 5]
Six different geophysical methods were combined in eight borehole applications to collect the
geophysical data for this investigation. The six methods were: natural gamma, gamma-gamma.
total magnetic and magnetic susceptibility, electrical conductivity, and temperature.
Gamma Ray and Density
The natural gamma probe detects variation in natural radioactivity of the material surrounding
the well. In sediments, 40K is the principal source of natural gamma radiation which is present in
clay minerals such as illite and montimorillonmte. The presence of clay layers can be detected
by an increase in gamma emissions. Clays, with their low permeability can have the effect of
precluding the vertical flow of groundwater and leachate. The technique can be used to
determine accurate boundaries between sediment layers, sequences in grain size fining or
coarsening which are generally much more accurate than lithologic logs developed by hand.
A gamma-gamma method was used to estimate the density of the geologic units. The density is
determined by reading the "scatter back" of a gamma ray emitted from a source crystal
containing Cobalt 60 on the probe. The application of density measurements to hydrogeology
relies upon the assumption that the lower the density of the formation the greater the porosity and
therefore potential for groundwater flow. It can be predicted that the areas within the sands.
gravels and cobbles with lower densities will likely have the most water flow, while the
limestone bedrock and clays having the least water flow. The gamma data were collected using
the IFG Corporation Logging System, utilizing a dual-purpose 512 channel spectral natural
gamma and gamma-gamma density probe
Magnetometry (Total and Susceptibility)
Magnetic susceptibility provides a direct measure of the presence of magnetic sediments. Most
unconsolidated sediments have little or no magnetic signature. Thus a higher magnetic
susceptibility indicates the presence of iron rich mafic sand, gravel or cobbles. The use of both
total field magnetic and magnetic susceptibility logs allow for the detection of ferromagnetic
minerals such as pyrite (FeS2). The measurement of the three orthogonal magnetic field
components, which represent the local value of the normal ambient field of the Earth as modified
by the remnant magnetization of adjacent sediments. The identification of such magnetic zones
indicates layers that may have higher permeabilities, and, therefore, may be potential flow paths
for groundwater. Total magnetic field, magnetic susceptibility and temperature were measured
using the BMP-04 multi-parameter probe containing a 3 orthogonal fluxgate magnetometer.
Electrical Conductivity
Perhaps the most useful geophysical measurement for detecting groundwater contamination is
electrical conductivity. This geophysical method measures the conductivity of subsurface media
by generating a current between two electrodes and measuring the potential difference. Electrical
conductivity is measured in units of milliSeimens per meter (mS/m). Because soil is a poor
165
-------
Trail Road Landfill
(GEOPHYSICAL FINDINGSi
conductor, most electric current flow occurs in the soil water when ions such calcium,
magnesium, potassium, sodium, dissolved iron, chloride and sulphates are present. Leachate
from a landfill typically contains large amounts of these type of ions. Since natural waters can
contain many different ions, both ionic and uncharged, electrical conductivity cannot be used to
make accurate estimates of specific ion concentrations. A linear relationship between total
dissolved solids (TDS) and the electrical conductivity of groundwater exists.
Conductivity measurements were taken using the Geonics EM-39 system consisting of one
transmitter coil and one receiver coil operating at 39.2 kHz.
Temperature
Temperature readings can indicate at what depths there is flowing groundwater as well as aid in
determining location of exothermic chemical reactions from contamination. This information
can be used to characterize the extent of leachate plumes and potential areas of groundwater
contamination. The temperature is measured by a thermistor is cable of detecting temperature
variations of+/-0.001°C: Characteristic water temperature profiles can be amplified using -
calculated temperature gradient logs to compare with measured temperatures. The temperature-
depth profile can be modified by water flow, or exothermic chemical reactions in the leachate.
[GEOPHYSICAL FINDINGSi
Technology Calibration [3]
Geophysical methods often include calibration of the measurement instrument to a quantitative/
semi-quantitative standard. Natural gamma probes are calibrated to models of known 40K
radioactivity. Density is calibrated to models with a known density. Conductivity was calibrated
to the ambient conductivity of monitoring wells in which chemical sampling had found no
contamination. A background conductivity level of approximately 11 mS/m was established.
Temperature was calibrated to the ambient temperature of an upgradient background well. Total
magnetic field and magnetic susceptibility readings were zeroed by holding the probe at least 1.5
m above the ground and away from any metal objects.
Investigation Results [3,5]
Each of eight existing monitoring wells were used in the geophysical investigation. Four of the
wells, M66, M83, M76, and M77, are located downgradient from the landfills. The wells are
arranged along a line perpendicular to the groundwater flow between the landfills and the quarry
(see Figure 1). Three others are located, at the downgradient edges of Trail Road Landfill. The
two sets of monitoring wells are well placed to monitor contaminant migration from the landfill
toward the quarry.
The geophysical logs taken in the four wells located midway between the landfills and the quarry
are shown in Figures 3 to 7. These four wells were selected for use in this analysis because of
166
-------
Trail Road Landfill
I GEOPHYSICAL FINDINGSl
their location across the groundwater migration pathway. If leachate contamination were
migrating to the quarry, it would be detected in one or more of these four wells.
The use of multiple geophysical methods allows the results of one method to be used to validate
the findings of another. For example, in each of the four lithologic logs shown, a clay layer is
present at shallow depths, i.e. less than 5 meters. In each case, the results of the gamma and the
spectral gamma-gamma logs confirm this finding, as indicated by the sharp peak in counts per
second at similar depths. The results of the density logs taken at depths"abfove the water table are
not valid, as the instrument calibration assumed a water-filled well. The magnetic logs, both
total and susceptibility, are used to detect a coarsening in the subsurface materials, resulting from
the presence of gravels and cobbles. Such coarse layers may be potential migration pathways for
groundwater. In Figures 3 to 7, the magnetic logs do not indicate such layers at the depth at
L1THO
GAMMA
(CPS)
J Fme-MedSand
3 —. SiltyClay
Fine Sand
-I Sand.Gravel
Figure 3: Geophysical Log for Well M66
TempGrad
(mK/m)
167,
-------
I GEOPHYSICAL FINDINGSi
Trail Road Landfill
GAMMA DENSITY SGG MT MS MONITOR COND TMP
UTHO (CPS) (g/cm3) RATIO (nT) (mSI) DETAILS (mS/m) (°C)
Fine-Mad Sand |
Fine-MedSandi
Sand.Gravel K
o o
j I.
o. -i
j I.
Figure 4: Geophysical Log for Well M83 [5]
TempGrad
(mK/m)
GAMMA DENSITY SGG MT MS MONITOR COND TMP
UTHO (CPS) (g/cm3) RATIO (nT) (mSI) DETAILS (mS/m) (°C)
i
•3
Topsoil I
Clay|
Sand.Clay.Soams j
Siity, fine Sand
8-
-i
-i
Till
Figure 5: Geophysical Log for Well M76 [5]
168
TempGrad
(mK/m)
-------
Trail Road Landfill
I GEOPHYSICAL FINDINGSl
LITHO
GAMMA
(CPS)
Density
1 1
j Sand.Gravel.Cobbles
CJ ;
TempGrad
(mlC'm)
Figure 6: Geophysical Log for Well M77
which the clay layer is encountered. By inference, this result can be taken as a validation of the
lithologic findings. Electrical conductivity logs also confirm the presence of the clay layer at the
depths shown in the lithologic logs. The peak in conductivity measurements shows a distinct
peak at depths at which clay is present in the lithologic log and lower values where sandy soils
predominate. Conductivity and temperature logs .were taken to identify the presence of leachate
contamination. An examination of these logs in the four wells does suggest that in only one well,
M77, may such contamination be present. The conductivity log for M77 clearly shows two
anomalous spikes at depths of approximately eight and 20 meters. The first peak occurs at the
water table. While some conductivity increase can be expected as the probe comes into contact
with water, the reading in this well may also indicate the presence of contaminated groundwater.
The second peak in conductivity occurs at approximately 20 meters. At this depth, there is no
indication in the gamma or lithologic logs of clay lenses that might cause such a peak in
conductivity measurements. Density and magnetic logs, indicators of porosity, both show the
presence of a porous layer which may be controlling groundwater flow at this depth. The
temperature log at this depth shows a marked increase in temperature, rising to a maximum of 7.7
°C, that may be due to the presence of exothermic reactions occurring in the groundwater. These
findings, taken together, suggest the presence of leachate contamination at this depth.
169
-------
Trail Road Landfill
I GEOPHYSICAL FINDINGS!
Results Validation [3,5]
Chemical sampling at the landfill, as part of the on-going monitoring effort, confirmed the
findings of the geophysical investigation.
I LESSONS LEARNED
Lessons learned at the Trail Road Landfill site were the following:
• Geophysical logs provided a continuous representation of the subsurface conditions
which was a major advantage over the fixed depth readings obtained with traditional
monitoring methods. The information obtained using fixed-depth sampling was relevant
only at the depth the readings were taken. Geophysical logs provided continuous
readings for the full depth of the borehole.
• The geophysical logs successfully delineated the leachate plume migrating from the
landfill as regions of groundwater with anomalously high electrical conductivity.
Chemical analyses conducted as part of the on-going monitoring-program at the landfill
confirmed the presence of leachate contamination moving from the landfill.
• The use of several, complementary, geophysical methods provided a cross-validation
between the results of various methods. This cross-validation increases the confidence
with which the geophysical data are interpreted.
170
-------
Trail Road Landfill
(REFERENCES MMMBMUMM^HM«HMM^M^HHMMHI
1. Personal Communication with Dave Ryan, Trail Road Landfill Engineer, Region of
Ottawa-Carleton, Ontario, Canada. December 1, 1998.
2. Personal Communication with Keith Watson, Site Manager, Region of Ottawa-Carleton,
Ontario, Canada. November 25, 1998.
3. Abbey, Daron G., et al. The Application of Borehole Geophysics to the Delineation of
Leachate Contamination at the Trail Road Landfill site: Nepean, Ontario. Proceedings
of the Symposium on the Application of Geophysics to Engineering and Environmental
Problems (SAGEEP) 1998.
4. Personal Communication with Barbara Elliot, Borehole Geophysical Scientist,
Geological Survey of Canada, Mineral Resources Division, Borehole Geophysics
Section, Ottawa, Ontario, Canada. December 14, 1998.
5. Abbey, Daron G. The Application of Borehole Geophysics to the Delineation of
Leachate Contamination at the Trail Road Landfill site: Nepean, Ontario. Carleton
University, Ottawa, Ontario. December 1995.
6. Personal Communication with John Stowell of Mt. Sopris Instruments. Golden, CO.
December 1, 1998.
171
-------
PAGE LEFT BLANK INTENTIONALLY
172
-------
Wurtsmith Air Force Base
Case Study Abstract
Wurtsmith Air Force Base
Oscoda, Michigan
Site Name and Location:
Wurtsmith Air Force Base
Oscoda, MI 48750
Period of Site Operation:
1923 -1993
Operable Unit:
OT-16b
Geophysical Technologies:
Ground penetrating radar
Electromagnetic induction
Magnetometry
CERCLIS #
MI5570024278
Current Site Activities:
None
Point of Contact:
Paul Rekowski
BRAC Environmental Coordinator
AFB Conversion Agency/DD
Wurtsmith
3950 East Arrow Street
Oscoda, MI 48750
(517)739-5161
prekowski@afbdal .hq.af.mil
Geological Setting:
Coastal sand plain consisting of 60 feet
of sand and gravel overlying glacial-
lacustrine silty clays
Technology Demonstrator:
William A. Sauck, PhD
Department of Geosciences
Western Michigan University
Kalamazoo, MI 49008
(616)387-4991
sauck@wmich.edu
Purpose of Investigation:
To better explain/define a GPR shadow zone discovered during an earlier geophysical investigation of a well-established
dissolved hydrocarbon plume to the west. This GPR shadow zone was suspected to be a light non-aqueous phase liquid
(LNAPL) plume.
Number of Images/Profiles Generated During Investigation:
2700 feet of GPR lines/profiles
Results:
The investigation was a complete success and verified the accidental discovery of the newly named OT-16b LNAPL plume
found during a previous GPR investigation of the neighboring FT-02 plume site that was conducted during December 1994.
Overall, results indicate that biodegradation of a residual light hydrocarbon product plume and subsequent chemical processes
led to changes of the conductivity of soils and groundwater in the capillary fringe. In general, the GPR shadow zone is
coincident with the dissolved residual product plume.
173
-------
Wurtsmith Air Force Base
•EXECUTIVE SUMMARY l
Wurtsmith Air Force Base is located in northeastern losco County and covers a 5,223-acre area located
on the northeastern-part of Michigan's lower peninsula, approximately 2 miles west of Lake Huron. The
land surface is a five-mile wide plain bounded on the west by 80-foot high bluffs. Several small streams
flow from the bluffs and discharge into a swampy area west of the base. The shallow subsurface
stratigraphy is known to consist of uniform and well sorted fine to medium sands that coarsen with depth.
A sand and gravel aquifer of glacial origin underlies the base. The water table is about 10 to 12 feet
below land surface at the OT-16b site.
Three non-intrusive geophysical techniques were used in the characterization of a newly discovered
plume. These included electromagnetic (EM) induction, ground penetrating radar (GPR), and
magnetometry. An EM survey was chosen to search for any buried metal objects. Magnetometry was
used to determine the presence and location of buried magnetic materials that may have been missed by
the EM survey. Due to the uniform geologic conditions present at the site, GPR was used to further
investigate the newly discovered plume.
The EM survey identified an unmarked utility line and areas where caution should be exercised when
drilling wells at the site. The magnetometer survey revealed that no unknown buried steel objects existed
at the site. The GPR data identified that the conductive plume is located in the upper portion of the
aquifer. Overall, results indicate that biodegradation of a residual light hydrocarbon product plume and
subsequent chemical processes led to the generation of a secondary conductive plume in the aquifer.
Generally the anomalous GPR zone is coincident with the dissolved product plume.
One of the goals of this investigation was to challenge the conventional model of the geophysical
properties of hydrocarbon plumes. The conventional model, based on controlled spill and lab
experiments, is that groundwater and soils contaminated with hydrocarbons exhibit lower electrical
conductivity and lower relative permittivity than the surrounding uncontaminated media. The alternative
model tested in this study is that hydrocarbon spills in the natural environment will change the impacted
zone from electrically resistive to electrically conductive over time due to biodegradation of the
hydrocarbons.
Geophysical methods at the newly-discovered OT-16b site provided coverage of a large area in a short
period of time. The geophysical methods were non-intrusive and were less expensive than drilling wells
randomly or on a grid for plume delineation downgradient from the possible source. The results obtained
from the three different techniques were complimentary in making conclusions. The exceptional
geologic uniformity of this site provided a uniform background resistivity environment for a geophysical
investigation where even a subtle shadow effect could be observed. The conductive nature of this plume,
totally derived from insulating hydrocarbon fuels, fits the chemical and electrical model for mature
plumes undergoing natural attenuation.
174
-------
Wurtsmith Air Force Base
SITE INFORMATION
Identifying Information
Wurtsmith Air Force Base
Oscoda, MI 48750
Operable Unit: OT-16b
CERCLIS No.: MI5570024278
Background [2,5, 6]
Physical Description: Wurtsmith Air Force Base (AFB) is located in northeastern losco County
and covers a 5,223-acre area located on the northeastern part of Michigan's lower peninsula,
approximately 2 miles west of Lake Huron. The site is bordered to the north and northeast by
Van Etten Lake; to the southeast and east by the Village of Oscoda; to the northwest by State
Forest woodlands, and to the southwest by Allen Lake and wooded marshlands. Approximately
1,943 acres of the base are owned by the Air Force, 2,466 acres are leased, and 814 acres are
registered as easement tracts.
The land surface is a five-mile wide plain bounded on the west by 80-foot high bluffs. Several
small streams flow from the bluffs and discharge into a swampy area west of the base. The Au
Sable River, which flows eastward and discharges into Lake Huron, is located less than one mile
south of the base. The land between the base and the river is swampy. The altitude of the land
surface drops from 750 to 580 feet as it slopes toward the river.
The newly discovered OT-16b plume study area where this geophysical investigation took place
is located 450 feet to the east of a former fire training area site known as FT-02, shown in Figure
1.
Site Use: The FT-02 site was used by the Air Force for 24 years as a bi-weekly fire training
facility. Typical exercises involved the combustion of several thousand gallons of JP-4 jet fuel
and other hydrocarbon fuels. Most but not all of the fuel would bum, which would leave the rest
to percolate into the ground along with fire retardant chemicals used to extinguish the fires. In
1982, a concrete fire-containment basin with an oil-water separator was constructed to help
reduce the amounts of fuel entering into the subsurface. Until this point, an unknown quantity of
fuel had already infiltrated into the subsurface. It was reported that overflows persisted after the
separator was installed in 1982.
Release/Investigation History: Fuels used in the fire training exercises at FT-02 were stored
nearby in a vaulted underground storage tank (UST) at the OT-16b site. This underground
collection and supply tank was removed in 1993, but a concrete pad and steel perimeter posts still
mark its location. The protective vault was free of any signs of hydrocarbon spillage. Therefore,
the tank was removed and the vault was backfilled. An underground pipeline had been used to
transport the waste fuels and solvents from the collection tank to the burning pad at the center of
the fire training area (Figure 1). This pipeline passed leak tests at the time it was
175
-------
Wurtsmith Air Force Base
SITE INFORMATION I
decommissioned [6]. Contamination may have occurred in the past from spillage during
refilling activities of the UST.
In December of 1994 an integrated geophysical investigation was undertaken at the FT-02 study
site 450 feet to the west of OT-16b. This investigation consisted of ground penetrating radar
(GPR), electrical resistivity using dipole-dipole profiling and Schlumberger vertical electrical
sounding, and self potential methods [6]. The results of several reconnaissance GPR survey lines
conducted to examine the background response of FT-02 revealed several strong reflectors. One
zone of attenuated GPR reflections was spatially correlated with the area of known hydrocarbon
contamination, as determined from soil borings and hydrochemical studies [6]. When the
positions of the GPR 'shadow' zones were plotted on a map, the resulting pattern was spatially
coincident with the mapped position of the plume from hydrochemical studies [6]. Some of the
'shadow' zones were not coincident with the area of the known FT-02 plume (Figure 1) and
caused speculation as to what they represented. The investigator came back in the Spring of
1996 to initiate this geophysical investigation in the area of the OT-16b site to determine what
these other 'shadow' zones represented.
Former Tank Location
CPR Survey Lines
DPR/IP Survey Line
VRP Location
Survey Grid Notes
LNAPL Plume
Roads
Figure 1: Study area location with GPR profile lines shown. The boxed area to the east of the FT-02 plume
site is the OT-16b site. Source: [4,5].
176
-------
Wurtsmith Air Force Base
SITE INFORMATION
Regulatory Context: The site is being addressed through Federal actions. Wurtsmith AFB was
proposed to the National Priorities List (NPL) on January 18, 1994, but its addition to the NPL
has not yet been finalized. In July 1991, the Base Realignment and Closure (BRAC) Commission
recommended the closure of Wurtsmith AFB. On June 30, 1993, the installation closed as
scheduled. The BRAC Cleanup Team (BCT) was formed in fiscal year 1994. The BCT consists
of representatives of the Air Force, U.S. Environmental Protection Agency (EPA) Region 5, and
the Michigan Department of Environmental Quality (MDEQ). The BCT works with a number of
other agencies and organizations to complete environmental actions necessary before property at
the base can be transferred to the private sector.
Site Logistics/Contacts
Federal Lead Agency:
United States Air Force
Federal Oversight Agency:
EPA Region 5
State Oversight Agency:
Michigan Department of Environmental
Quality Response Division
Robert Delaney
P.O. Box 30473
Lansing, MI 48909-7973
(517)373-7406
delaneyr@state.mi.us
EPA Remedial Project Manager:
Diana Mally
U.S. EPA
77 West Jackson Boulevard
Chicago, IL 60604
(312)886-7275
mally.diana@epamail.epa.gov
Geophysical Subcontractor:
William A. Sauck, PhD
Department of Geosciences
Western Michigan University
Kalamazoo, MI 49008
(616) 387-4991 -
sauck@wmich.edu
Site Contact:
Paul Rekowski
BRAC Environmental Coordinator
AFB Conversion Agency/DD Wurtsmith
3950 East Arrow Street
Oscoda, MI 48750
(517)739-5161
prekowski@afbdal .hq.af.mil
177
-------
Wurtsmith Air Force Base
I MEDIA AND CONTAMINANTS I
Matrix Identification
Type of Matrix Sampled and Analyzed: Groundwater and subsurface soil
Site Geology/Stratigraphy [4]
Based on previous borings completed by the United States Geological Survey (USGS) at the
neighboring FT-02 site, the shallow subsurface stratigraphy is known to consist of uniform and
well-sorted fine to medium sands that coarsen with depth. A sand and gravel aquifer of glacial
origin underlies the base and is comprised of a brown to gray-brown medium coarse sand
containing some gravel. The water table is about 10 to 12 feet below land surface at the OT-16b
site. The aquifer overlies a thick clay layer found at an average depth of 65 feet. The clay layer
is mostly brown to gray, relatively impermeable, and cohesive. Its thickness at the base is not
accurately known because no lithologic logs exist that extend to the maximum depth of the clay
layer. However, the clay is known to be at least 13 feet thick in one location. At Oscoda and at
places east and north of Van Etten Lake, the clay unit is at least 125 feet thick and may be as
thick as 250 feet. It slopes downward to the east at 10 to 30 feet per mile. In general, the surface
of the unit dips inward to low points in the northeast part of the base and in an area just northeast
of Van Etten Lake. Mississippian sandstone, shale, and limestone formations dipping southwest
into the Michigan Basin constitute the bedrock beneath the base.
A groundwater divide cuts diagonally across the base from northwest to southeast. South of the
divide, groundwater flows to the Au Sable River; north of the divide, it flows to Van Etten Creek
and Van Etten Lake. Groundwater flow ranges from about 0.8 feet per day in the eastern part of
the base to about 0.3 feet per day in the western part.
Contaminant Characterization [4]
Primary Contaminant Groups: The primary contaminants of concern at the OT-16b site
include fuel related contaminants such as benzene, toluene, ethylbenzene, and xylene (BTEX).
Matrix Characteristics Affecting Characterization Cost or Performance [2,5]
For the electromagnetic (EM-31) method in the vertical dipole mode, 18 feet is the maximum
depth of detection of a highly conductive. Since the contaminant plume found at the site is only
moderately conductive, approximately 3.3 times the conductivity of the background aquifer, it is
not likely that the EM-31 can effectively discriminate between the weak signature of the
contaminant plume below the water table and the conductivity of the uncontaminated
groundwater. The depth of the contaminant plume is below the water table (15 feet), which is
close to the limit of penetration for the EM-31 instrument. However, results from the EM-31
survey are still useful for other aspects of site characterization, since they clearly indicate where
subsurface objects or utilities may exist and caution should be used in drilling future wells at the
site.
There were no reported characteristics of the site that affected the magnetometer survey results.
178
-------
Wurtsmith Air Force Base
I MEDIA AND CONTAMINANTS I
However, there are some limitations when using a magnetometer generally. In a relatively
"clean" area, a single drum may be theoretically detected to a depth of 20 feet from the surface.
In practice, however, numerous smaller, near-surface iron objects will obscure the weaker deep
target. A more realistic maximum depth of detection is 5 to 10 feet. Large masses of drums may
be detected easily to depths of 10 to 40 feet.
The clarity of Ground Penetrating Radar (GPR) results can be affected by heterogeneous
conditions in the subsurface. However, the study site has been noted to have exceptional
geologic uniformity. The results of the GPR survey were enhanced by these uniform conditions.
iGEOPHYSICAL INVESTIGATION PROCESS!
Investigation Goals
Overall, the goal of this geophysical investigation was to use three different geophysical
techniques (GPR, magnetometry, and EM) to explore and better define a suspected light non-
aqueous phase liquid (LNAPL) plume that was encountered during a GPR investigation
approximately 450 feet to the west of the FT-02 plume site [8]. A specific goal of the
magnetometer survey was to search for any buried steel objects that might have been missed by
the EM induction survey [2, 5]. GPR was then used to delineate the boundaries of the newly
discovered plume.
One of the goals of this investigation was to challenge the conventional model of geophysical
properties of hydrocarbon plumes. The conventional model, based on controlled spill and lab
experiments, is that groundwater and soils contaminated with hydrocarbons exhibit lower
electrical conductivity and lower relative permittivity than the surrounding uncontaminated
media. The alternative model tested in this study is that hydrocarbon spills in the natural
environment will change the impacted volume from electrically resistive to electrically
conductive over time due to biodegradation of the hydrocarbons. Conductivity is enhanced by
the leaching of inorganics from the soil and aquifer materials by organic acids and carbonic acid
produced by microbial activity during degradation of the hydrocarbons. This model suggests that
the conventional model can not be applied uniformly to all hydrocarbon plume sites and the
geoelectrical signature of a plume will vary with time and position [7].
Geophysical Methods [2, 5]
The investigation took place over several days in May 1996. The EM induction method is often
used to explore for metal objects based on the principle of EM induction. This induction
technique uses two coils: a transmitter and a receiver. EM surveys detect variations in the
conductivity of subsurface materials. Buried objects, conductive fluids, and geologic
discontinuities can be detected by artificially applying known electric fields to the ground surface
by means of the transmitter, and the receiver records the presence of disruptions to the known
field. These disruptions, termed EM anomalies, can result from geological changes or the
presence of metallic objects, such as pipes, drums, cables, tanks, etc., in the subsurface.
179
-------
[ GEOPHYSICAL INVESTIGATION PROCESS!
Wurtsmith Air Force Base
For the EM survey, a Geonics EM-31 was carried at waist level using the vertical dipole mode.
A grid of 25 feet by 50 feet was established and results from the survey were plotted using
Geosoft™ software.
The second method used in this investigation was the magnetometry survey. Magnetometers
measure variations in the magnetic field of the earth, and local disruptions to the earth's field, the
presence of naturally occurring ore bodies, and man-made iron or steel objects such as buried
drums, tanks, or ordinance. Whether on the surface or below, iron objects or minerals cause
local distortions or anomalies in this field. Originally designed for mineral exploration,
magnetometers are now used in the environmental field for locating buried steel drums, tanks,
pipes, and iron debris in trenches and landfills. A magnetometer's response is proportional to the
mass of iron in the target. The magnetometer can only sense ferrous materials such as iron and
steel; other metals like copper, tin; aluminum, and brass are not ferromagnetic and cannot be
located with a magnetometer. The effectiveness of magnetometry results can be reduced or
inhibited by interference (noise) from time-variable changes in the earth's field and spatial
variations caused by magnetic minerals in the soil or iron debris, pipes, fences, buildings, and
vehicles. Many of these problems can be minimized by careful selection of the type of
instrument and field procedures used for the survey.
Magnetometry was used in this investigation to determine the presence and location of buried
magnetic materials using a 50 feet by 50 foot grid, which had already been established for the
EM survey, magnetic data were collected using a Geometries G-858 cesium vapor magnetometer.
Using this data, a magnetic field intensity map of the area was produced for interpretation.
The third geophysical method used in the OT-16b site geophysical investigation was ground
penetrating radar (GPR). GPR uses high-frequency radio waves to determine the presence of
subsurface objects and structures. A GPR system radiates short pulses of high-frequency EM
energy into the ground from a transmitting antenna. This EM wave penetrates into the ground at
a velocity that is related to the electrical properties of subsurface materials. When this wave
encounters the interface of two materials having different electromagnetic properties (i.e., soil
and water), a portion of the energy is reflected back to the surface, where it is detected by a
receiver antenna and transmitted to a control unit for processing and display. The major
principles involved for GPR are similar to reflection seismology, except that EM energy is used
instead of acoustic energy, and the time scale for GPR is a million times shorter than that of
seismic phenomena.
For this investigation a Geophysical Survey Systems, Inc. (GSSI) Subsurface Interface Radar
(SIR) System-10 GPR system along with 100 MHZ antennae recording for a scan time of 400
nanoseconds (ns) was used. The 100 MHZ Transmitter-Receiver pair were operated with a
separation of 1.45 meters between mid-points. The site was traversed in the west to east and
south to north directions along lines spaced 50 feet apart, using a van to tow the antennae. No
post-processing was done other than horizontal scales normalization. This GPR system used
180
-------
Wurtsmith Air Force Base
fixed gain vs. depth function. No gain equalization or automatic gain control processing were
used. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
i GEOPHYSICAL FINDINGS •^••••••••••^^^••MBMMHM
Technology Calibration [8]
For the EM-31 the only calibration necessary is setting the zero on the instrument. A region of
very resistive ground was identified and its conductivity was accurately measured using
conventional techniques. GPR readings were taken in the same location and the instrument gains
were set at this point. No further calibration was reported to be necessary for the GPR or
magnetometer used in this investigation.
Investigation Results [2,5]
The EM survey revealed a linear alignment of anomalies extending from a manhole located at
coordinate S18 to coordinate P12 (Figure 1), and at least three other anomalies parallel to this
alignment. These anomalies were attributed to communication or electric cables buried in the
ground. No anomalous regions associated with the suspected conductive groundwater plume
were visible on any of the interpretive maps produced from the EM survey. The lack of EM-31
response from the conductive plume just below the water table at 15 feet was attributed to the
plume being nearly at the limit of depth penetration for the instrument. -One unmarked utility line
was discovered. The results indicated where caution should be taken when drilling wells at the
site. A strong anomaly beneath the old taxiway in the northeast comer of the map was detected
but the source is unknown.
The magnetic survey revealed that there was a strong low in the magnetic field in the vicinity of
the old UST vault. The UST vault (still in place, but now filled) had been surrounded at the
surface by 20 steel posts filled with concrete. The posts were attached to a flat slab of concrete.
The strong low was attributed to the potential for the steel having a strong reversed remnant
magnetization. The magnetic survey revealed no other buried steel objects at the site. This was
an indicator that the buried cables found by the EM-31 survey are nonmagnetic but electrically
conductive. A strong magnetic low found beneath the asphalt taxiway in the northeast corner of
the map remains unexplained.
The GPR data revealed that there is a particularly strong reflector representing the water table at
approximately 10 to 12 feet (five meters) shown on Figure 2 as an inverted triangle. This was
caused by the sharp change in the relative permittivity in the transition from unsaturated to
saturated sand. The central areas of the pair of two-dimensional profiles show pronounced signal
attenuation, creating an amplitude shadow zone (between "R" and "T" on line 14 in Figure 3, and
between "S" and "U" on line 16 in Figure 4). The conductive zone causing the attenuation is at
the tops of these shadows. The shadow begins at or just below the water table, so the conductive
plume is located in the upper part of the aquifer. This same phenomenon was seen along all
other lines crossing the plume area. This conductive plume in the groundwater below the highly
resistive hydrocarbon liquids fits the alternative geoelectrical model proposed for mature plumes [7].
181
-------
Wurtsmith Air Force Base
I GEOPHYSICAL FINDINGS I
The investigator's recognition and understanding of the significance of the GPR shadow zone
below the FT-02 plume led to the discovery of the new contaminant plume. From the shadow
zones on the GPR profiles, it was possible to create a map that showed the extent of the
conductive plume at the OT-16b site (Figure 2). The broad proximal end of the plume is possibly
due to the spillage of fuel on the asphalt taxiway, as well as possible surface spillage during
refilling operations at the former underground collection tank location. Another result of the
Figure 2: Ground penetrating radar profile of Line 14 showing the strong amplitude shadow caused by the
proximal end of the neighboring FT-02 plume, and the somewhat weaker shadow at the right end caused by thj
OT-16b plume. Source: [8]
GPR investigation was the observation of some paleo-dune morphologies that underlie the area at
a depth of approximately 40 feet (Figure 4 at location "Q").
Overall, results indicate that biodegradation of a residual light hydrocarbon product plume and
subsequent chemical processes led to the generation of a secondary conductive plume in the
aquifer that is coincident with the dissolved product plume. This coincides with the newly
developed hypothesis that hydrocarbon spills in the natural environment cause changes from
electrically resistive to electrically conductive over time due to biodegradation of the
hydrocarbon impacted zone. Conductivity is enhanced by the leaching of inorganics from the
soil and aquifer materials by organic acids produced by microbial activity during degradation of
the hydrocarbons [7]. Generally the GPR shadow zone is coincident with the dissolved product
plume (Figure 2).
182
-------
Wurtsmith Air Force Base
I GEOPHYSICAL FINDINGS]
- ~ WAFB OT.16&" Un«14 WtoE,
Figure 3: 100 MHZ GPR profile for line 14 (at 150 feet N coordinate on Figure 1) oriented with west to the left.
Scan length is 400 ns, showing amplitude shadows starting below the water table (about 70 ns or 12 feet);
horizontal scale is 50 feet between marks. Source: [2,5]
0 —
» Amplitude Shadow
,p*wC1!sai<*&&''CS*&*. * ft ** **
/«^r*' •S**SSS;*07r
WAFB; PlumeOT-16b; May9,3996 w ^^'
Figure 4:100 MHZ GPR profile for line 16 oriented with west to the left. Scan length is 400 ns, showing
amplitude shadows starting below the water table (about 70 ns or 12 feet); horizontal scale is 50 feet
between marks. Source: [2,5]
183
-------
Wurtsmith Air Force Base
I GEOPHYSICAL FINDINGS I
Results Validation [2, 5]
Several months after the initial geophysical investigation took place in May of 1996,
borings were taken at three locations on the newly discovered OT-16b plume. Soil and
groundwater samples were taken at various depths. One soil sample revealed
approximately 16 inches of a dark, viscous residual hydrocarbon product near the water
table. The conductivities of the aquifer water were at a maximum at the top of the
saturated zone and then diminished to background levels at depths of 10 feet below the
water table. This indicated that the anomalous conductive zone was less than 10 feet
thick and the water samples had a conductivity contrast of 2.5 to 3.3 above background
levels.
In addition, after the geophysical investigation was completed, a review of Wurtsmith
AFB airphoto archives led to the discovery that a maintenance building occupied the site
area until the 1970's. The UST was installed later, after the building was removed. When
the UST was removed there was no evidence of soil contamination. This indicates that
the source of the newly discovered contaminant plume was probably as a result of the
drainage of solvents and fuels from the floor of the maintenance building.
184
-------
Wurtsmith Air Force Base
i LESSONS LEARNED
Lessons learned at the Wurtsmith site include the following:
Geophysical methods at the newly discovered OT-16b site provided coverage of a large
area in a short period of time. The geophysical methods were non-intrusive and were
less expensive than drilling wells randomly or on a grid for plume delineation
downgradient from the possible source. The investigation was considered a complete
success and, using purely surface geophysical methods, verified the 1994 initial "blind"
discovery of a new groundwater contaminant plume [2, 5].
• The use of more than one geophysical method provided synergy, as each technique was
responsive to a different property. Therefore, the results obtained using the different
techniques were complimentary. The GPR outlined the conductive groundwater plume
and also revealed the details of the sand stratigraphy. The shallow EM discovered a
complex of buried electrical utility lines where only one line had been previously known.
Finally the magnetic survey revealed no buried steel objects, which was helpful in
characterizing the site as "tank-free [2, 5]."
• The conductive nature of this plume, totally derived from insulating hydrocarbon fuels,
fits the chemical and electrical model for mature plumes undergoing natural attenuation
[7]. The anomalous geophysical response is due to the electrically conductive ionic
nature of the plume, not due to any direct response to residual or dissolved hydrocarbons.
The investigators would not extrapolate these results to investigations of dense non-
aqueous phase liquid spills [2, 5].
It is clear that at this site biodegradation of a residual light hydrocarbon product plume
and subsequent chemical processes led to changes of the conductivity of soils and
groundwater in the capillary fringe and underlying aquifer. The broad proximal end of
the plume is potentially due to fuel spillage on the asphalt taxiway, as well as possible
surface spillage during refilling operations at the former underground collection tank
location and the floor of the maintenance building [2].
• The exceptional geologic uniformity of this site provided a uniform background
environment for a geophysical investigation where the shadow effect could be observed
[8]. The amplitude shadow is not visible if the GPR scan length or range can only reach
the water table. The shadow will also be destroyed if automatic gain control or other
gain equalization is applied during either acquisition or post-processing of data.
Therefore the appropriate setting of field acquisition parameters and careful post-
processing are necessary to record and preserve the GPR amplitude shadows [2, 5].
185
-------
Wurtsmith Air Force Base
I REFERENCES
1. Baedecker, M.J., Cozzarelli, I.M., Eganhouse, R.P., Siegel, D.I, and Bennett, P.C. Crude
Oil in a Shallow Sand and Gravel Aquifer - III. Biogeochemical Reactions and Mass
Balance Modeling in Anoxic Groundwater; Applied Geochemistry, vol. 8, pp. 569-586.
1993.
2. Bermejo, Jose L., Sauck, William A., and Atekwana, Estella A. Geophysical Discovoy
of a New LNAPL Plume at the Former Wurtsmith AFB, Oscoda, Michigan. Fall 1997
GWMR Vol XVII, No. 4, pp. 131 to 137.
3. Radian Corporation. Installation Restoration Program—Phase I-Records Search
Wurtsmith AFB, Michigan. April 1985.
4. Stark, J.R., Cummings, T.R., and Twenter, F.R. Groundwater Contamination at
Wurtsmith AFB, Michigan. USGS Water-Resources Investigations Report 83-4002.
Prepared in Cooperation with the USAF. 1983.
5. Sauck, William A., Atekwana, Estella A., and Bermejo, Jose L. Characterization of a
Newly Discovered LNAPL Plume at Wurtsmith AFB, Oscoda, Michigan. Proceedings of
the Symposium on the Application of Geophysics to Engineering and Environmental
Problems (SAGEEP) 1998 pp. 389-408.
6. Sauck, William A., Atekwana, Estella A., and Nash, Mike S. High Conductivities
Associated with an LNAPL Plume Imaged by Integrated Geophysical Techniques.
Journal of Environmental and Engineering Geophysics. Volume 2, issue 3, January
1998, p. 203-212.
7. Sauck, William A. A Conceptual Model for the Geoelectrical Response of LNAPL
Plumes in Granular Sediments. Proceedings of the Symposium on the Application of
Geophysics to Engineering and Environmental Problems (SAGEEP) 1998 pp. 805-817.
8. Personal Communication with Dr. William A. Sauck of Western Michigan University.
November 20, 1998.
9. Personal Communication with Dr. Michael J. Barcelona of the University of Michigan.
November 20, 1998.
10. Personal Communication with Phil Sirles of Microgeophysics Corporation. Wheat
Ridge, CO. December 10, 1998.
186
------- |