United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
P.O. BOX 93478
Las Vegas NV 89193-3478
Pre-issuance Copy
December 1987
Research and Development
&EPA
Soil-Gas and
Geophysical Techniques
for Detection of
Subsurface Organic
Contamination
-------
United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
Las Vegas, NV 89193-3478
Research and Development
EPA/600/S4-88/019 July 1988
&EPA Project Summary
Soil-Gas and Geophysical
Techniques for Detection of
Subsurface Organic
Contamination
Ann M. Pitchford, Aldo T. Mazzella, and Ken R. Scarborough
From 1985 through 1987, the Air
Force Engineering and Services
Center (AFESC) funded research at
the U.S. Environmental Protection
Agency (EPA) Environmental Moni-
toring Systems Laboratory in Las
Vegas, Nevada (EMSL-LV) through
an interagency agreement. This
agreement provided for investi-
gations of subsurface contamination
at Air Force installation Restoration
Program sites. The purpose of these
investigations was to demonstrate
and evaluate inexpensive and rela-
tively rapid reconnaissance tech-
niques which can detect and map
subsurface organic contamination.
This information can reduce the
number and improve the placement
of wells required in an investigation,
resulting in significant savings in
terms of costs and time.
The methods chosen for demon-
strations included active and passive
soil-gas sampling and analysis, and
the geophysical techniques of
electromagnetic induction (EM), and
d.c. resistivity. Field studies were
performed at four Air Force Bases:
active soil-gas measurements were
performed at all sites; d.c. resistivity
and EM measurements were per-
formed at three sites; and passive
soil-gas sampling was performed at
two sites. The techniques of
ground-penetrating radar and com-
plex resistivity were included in the
evaluations using experiences at
other locations. Based on this limited
set of cases and information from
published literature, general guide-
lines on the application of these
techniques for detecting organic
contamination were developed.
The active soil-gas sampling
technique successfully mapped sol-
vents, gasoline, and JP-4 con-
tamination at the four bases where it
was used. The passive soil-gas
technique was successful in some
cases, but not as successful as the
active technique, and further re-
search on the performance of the
technique is recommended before
the method is used widely. The
geophysical methods were suc-
cessful for site characterization, but
the EM and d.c. resistivity techniques
did not detect gasoline and jet fuel
number 4 (JP-4) contamination
when it was present. The use of EM
and d.c. resistivity for direct
detection of hydrocarbons appears to
be a subtle technique which depends
on a thorough understanding of
background information at the site,
the skill of the instrument operator,
and may depend on the length of
time the spill has been present. The
ground-penetrating radar and com-
plex resistivity techniques were used
successfully at a number of locations
for detecting organic contamination.
This work was conducted from
January 1985 to October 1987.
This Project Summary was
developed by EPA's Environmental
Monitoring Systems Laboratory, Las
Vegas, NV, to announce key findings
of the research project that is fully
-------
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
In 1984, the U.S. Environmental
Protection Agency (EPA) Environmental
Monitoring Systems Laboratory in Las
Vegas, Nevada (EMSL-LV) and the Air
Force Engineering and Services Center
(AFESC) entered into an interagency
agreement concerning investigations of
subsurface contamination at Air Force
Installation Restoration Program (IRP)
sites. Organic contamination was em-
phasized in these studies. The traditional
approach to these site investigations
involves the installation of wells and
analysis of ground-water samples, This
approach provides a direct measurement
of the contamination at the locations
sampled. However, information about the
extent and degree of contamination may
be limited by the number, cost and
possible locations of the wells. If
inexpensive, and relatively rapid recon-
naissance techniques could be used as
an aid to selecting the well locations, the
number of wells could be reduced. This
would save money and time.
The interagency agreement initiated
studies at four IRP sites to demonstrate
indirect methods for detecting and
mapping organic contamination in
ground-water and soil. The methods
chosen for evaluation were soil-gas and
geophysical measurements. These
measurement results then were com-
pared to ground water data obtained
during the same study. This made it
possible to evaluate the performance of
the soil-gas and geophysical tech-
niques. However, because of the wide
variety in contaminants and geological
conditions, care must be used when
applying the conclusions developed from
these site-specific studies to other
locations. To help to extend the results
from these studies to other site
conditions, additional examples were
assembled from the literature. Using all
this information, general guidelines were
developed for the use of these
techniques in investigations of organic
contamination of soil and ground water.
Approach
The overall approach to the project
was divided into two parts with activities
in each proceeding concurrently. These
parts consisted of working with a panel of
experts to broaden the ideas, ap-
proaches and experiences being used as
a basis for developing the guidelines;
and performing site investigations to
demonstrate the soil-gas and geo-
physical techniques. The Air Force
Bases (AFBs) selected are listed in Table
1. Each AFB provides differing geology,
climate, depth to water table, and con-
taminants, thus representing a variety of
situations for performing the com-
parisons.
This series of studies was intended to
help formulate a hierarchy of techniques
which could be logically adapted and
applied to detect contamination for a
variety of site conditions. However, the
results from the field studies fit better
into a framework of broad guidelines
rather than into a detailed strategy which
ranks techniques,
Field Study Results
The methods chosen for dem-
onstrations included active and passive
soil-gas sampling and analysis, and the
geophysical techniques of EM and d.c.
resistivity. Active soil-gas measure-
ments were performed at all sites;
resistivity and EM measurements were
performed at three sites; and passive
soil-gas sampling was performed at two
sites. Key results from these in-
vestigations are summarized in Table 2.
The active soil-gas sampling
technique successfully mapped solvents,
gasoline, and JP-4 contamination at all
four bases where it was used. Results
from Robins AFB demonstrated that the
choice of sampling depth can influence
the measurements obtained. At this AFB,
initial sampling at 1 meter revealed very
little contamination as shown in Figure 1,
while additional sampling at 2 meters
located more contamination, which is
shown in Figure 2. Thus, it is important
to perform depth profiles at a number of
locations during the initial phase of a
study, preferably in regions of known
(quantified) ground-water contam-
ination, in order to select the sampling
depth. Sampling depth is particularly
important at sites where relatively old fuel
spills have occurred, because chemical
or biological oxidation of the petroleum
hydrocarbons can remove fuel con-
stituents from the aerobic soil horizons.
The real-time nature of this method also
represents a significant advantage over
more time-consuming techniques since
the choice and number of sampling
locations can be evaluated as data are
obtained.
Two of the sites investigated with
active soil-gas techniques were also
investigated by a passive technique
which used adsorbent charcoal badges,
At these sites, tests were performed to
determine the feasibility of mapping the
contamination at these sites by selecting
the best exposure times for the badges.
Performing feasibility tests with the
badges was demonstrated to be very
important; an insufficient exposure time
may indicate an area is uncontaminated
when contamination actually is present,
Alternately, overexposure of the badges
may result in saturation of the sorbent
which would mask any relative
differences in soil-gas contamination at
the various sampling locations. This
passive soil-gas technique was not as
successful as the active technique in
detecting contaminated ground water
However, contaminated areas were
identified successfully in some cases.
Further testing of the performance of this
technique for a variety of contaminants
and geologic conditions is recommended
before the method is used widely. If on-
site personnel are available to conduct
the sampling, the low analytical cost of
this method has potential for reducing
site investigation costs in some cases.
The geophysical methods were suc-
cessful for site characterization, but the
EM and d.c. resistivity techniques did not
detect gasoline and JP-4 contamination
when it was present. This was attributed
to the natural variations in background
Table 1,
Geology, Climate, and Contaminants at&r Force Base Study Sites
Base
Geology
Climate
Contaminant
Holloman AFB sand, inter bedded clay
Phelps Collins ANGTB karst
Robins AFB marine sand
Tinker AFB clay
arid gasoline, JP-4, solvents
humid solvent, JP-4, buried metallic objects
humid JP-4, solvents
humid JP-4
-------
Table 2. Key Results from the AFB Investigations
Site and contaminants Method
Comment
Hoffoman AFB,
BX Service Station,
Gasoline
Robins AFB.JP-4 Spill,
JP-4
Tinker AFB, Fuel Farm
290, JP-4
Active soil-gas
sampling
EM, d.c. resistivity
Active sod-gas
sampling
Passive soil-gas
sampling
EM, d.c. resistivity
Active soil-gas
sampling
Passive soil-gas
sampling
EM, d.c. resistivity,
complex resistivity
Compares favorably with ground-water data. Demonstrates movement of contaminants along
utility corridors.
Do not detect organics because of natural variability in soil resistivity. Culture limited extent of
survey.
Compares favorably with ground-water data in spite of 20-year age of spill. Demonstrates
importance of depth of sampling.
Preliminary test has mixed results compared to ground-water data.
Do not detect organics because of natural variability in soil resistivity due to rainfall effects and
culture. AFB radar interferes with EM-34 measurements..
Compares favorably with ground-water data; technique effective in clay soil
Preliminary test has mixed results compared to ground-water data. Technique may be
responding to surface contamination at times.
Were not attempted due to high density of buried pipes and tanks, and fences and pipes on
surface.
resistivity which masked any resistivity
anomaly due to the presence of
hydrocarbons. Based on these results,
the use of EM and d.c. resistivity for
direct detection of hydrocarbons appears
to be a subtle technique which depends
on a thorough understanding of
background information at the site, the
skill of the instrument operator, and may
depend on the length of time the spill
has been present. This does not
preclude the use of these techniques in
site characterization. The techniques of
GPR and complex resistivity were not
demonstrated at the AFBs, but their
successful performance in detecting
hydrocarbons has been documented in
the literature. Table 3 summarizes the
general recommendations for application
of the geophysical techniques.
Note that only two techniques, GPR
and complex resistivity, are recom-
mended for routine use in detecting
organic contamination. GPR is com-
mercially available. Complex resistivity,
however, is the subject of several
research efforts, and is not widely
available. The d.c. resistivity and EM
techniques may sometimes be useful at
a site for detection of hydrocarbons, but
the conditions for which this is true are
not now understood. Other techniques
with greater likelihood of success should
be considered first.
Fundamentals for Planning Site
Investigations
To place these results in context,
recommendations for planning a site in-
vestigation also are presented. These
recommendations were prepared in
conjunction with members of the panel of
experts assembled to provide advice to
the project. The recommendations ad-
dress general considerations in design-
ing an investigation, provide examples
and references to similar cases in the
literature, list the steps in planning a
soil-gas investigation, and list issues to
be considered in planning a geophysical
investigation. The issues which should
be considered are presented in series of
questions organized by topic area, in-
cluding hydrology, the use of isotopes,
and water chemistry.
Conclusions
Demonstrations of soil-gas and
geophysical techniques at four AFBs
provided the basis for the development
of broad guidelines for the application of
these methods. The active soil-gas
sampling technique successfully mapped
solvents, gasoline, and JP-4 contam-
ination at the bases. The passive soil-
gas technique was successful in some
cases, but not as successful as the
active technique, and further research on
the performance of the technique is
recommended before the method is used
widely. The geophysical methods were
successful for site characterization, but
the EM and d.c. resistivity techniques did
not detect gasoline and jet fuel number 4
(JP-4) contamination when it was
present The use of EM and d.c. re-
sistivity for direct detection of hydro-
carbons appears to be a subtle technique
which may sometimes be useful at a site
for the detection of hydrocarbons, but the
reasons for this are not well understood.
Other techniques with greater likelihood
of success should be considered first.
The ground-penetrating radar and
complex resistivity techniques have been
used successfully at a number of
locations for detecting organic
contamination.
-------
Legend
Total Hydrocarbon Concentration
ffjg/L) in Soil-Gas
LF-1-2 O—Well Sampling
SG-6 •—Soil-Gas Sampling Location
„, 10,000*- ~lsoconcentration Contour Line
' <0.06—Total Concentration Value (ug/L)
SG-7
180.000
SG-//
<0.06
N-
Figure 1.
10 0 10 20
Scale in Meters
Concentrations of total hydrocarbons in soil gas at JP-4 spill site. Robins AFB, Sampling depth: 1 meter.
-------
Legend
Total Hydrocarbon Concentration
(ng/L) in Soil-Gas
LF-1-2 O — Well Sampling Location
•—Soil-Gas Sampling Location
>0.06—Total Concentration
.Q QQO—lsoconcentration Contour Line
'
000
61
10 0 10 20
Sacle in Meters
Figure 2. Concentrations of total hydrocarbons in soil gas at JP-4 spill site, Robins AFB. Sampling depth: 2 meters.
-------
Table 3. Generalized Applications of Geophysical Techniques
Application
Technique
Ground Penetrating Radar
(GPR)
Electromagnetics (EM)
DC, Resvstivity
Complex Resistivity
Seismic Refraction
Metal Detector
Magnetometer
Site
Characterization
yes
yes
yes
yes**
yes
no
no
Conductwe
Leachate*
yes
yes
yes
yes**
no
no
no
Metal Objects*
yes
yes
yes
yes**
no
yes
yes***
Organic
Contamination
yes
possibly
possibly
yes
no
no
no
*ln some cases, the organic contamination will be associated with inorgamc contamination:
examples include organics in metal drums and mixed organic-inorgamc leachate plumes,
**But d.c. resistiviiy is equally good and much cheaper.
**Ferrous metals only.
-------
The EPA authors Ann M. Pitch ford, Aldo T. Mazzella and Ken R. Scarborough,
are with the Environmental Monitoring Systems Laboratory, Las Vegas, NV
89193-3478.
Aldo T. Mazzella is also the EPA Project Officer (see below).
The complete report, entitled "Soil-Gas and Geophysical Techniques for
Detection of Subsurface Organic Contamination, " (Order No. PB 88-208
194/AS; Cost: $14.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, NV 89193-3478
United States Center for Environmental Research
Environmental Protection Information
Agency Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S4-88/019
-------
SOIL-GAS AND GEOPHYSICAL TECHNIQUES
FOR
DETECTION OF SUBSURFACE ORGANIC CONTAMINATION
by
Ann M. Pitchford, Aldo T. Mazzella, and Ken R. Scarborough
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89193-3478
Prepared for
U.S. Air Force
Tyndall Air Force Base, Florida 32403-6001
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
-------
NOTICE
The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under contract number 68-03-3245
to Lockheed Engineering and Management Services Co., Inc. It has been subject
to the Agency's peer and administrative review, and it has been approved for
publication as an EPA document. Mention of trade names or commercial products
does not constitute either endorsement or recommendation for use.
-------
ABSTRACT
From 1985 through 1987, the Air Force Engineering and Services Center
(AFESC) funded research at the U.S. Environmental Protection Agency (EPA)
Environmental Monitoring Systems Laboratory in Las Vegas, Nevada (EMSL-LV)
through an interagency agreement. This agreement provided for investigations
of subsurface contamination at Air Force Installation Restoration Program sites.
The purpose of these investigations was to demonstrate and evaluate inexpensive
and relatively rapid reconnaissance techniques which can detect and map sub-
surface organic contamination. This information can reduce the number of wells
required in an investigation, resulting in significant savings in terms of
costs and time.
The methods chosen for demonstrations included active and passive soil-
gas sampling and analysis, and the geophysical techniques of electromagnetic
induction (EM), d.c. resistivity, seismics and magnetics. Field studies were
performed at four Air Force Bases where these techniques were used as appropri-
ate. Active soil-gas measurements were performed at all sites; resistivity
and electromagnetic induction measurements were performed at three sites, and
passive soil-gas sampling was performed at two sites. The other techniques were
applied to characterize one study site. The techniques of ground-penetrating
radar and complex resistivity were included in the evaluations using experiences
at other sites. Based on this limited set of cases and information from pub-
lished literature, general guidelines on the application of these techniques
for detecting organic contamination were developed.
The active soil-gas sampling technique successfully mapped solvents, gaso-
line, and JP-4 contamination at all four bases where it was used. The passive
soil-gas technique was successful in some cases, but not as successful as the
active technique, and further research on the performance of the technique is
recommended before the method is used widely. The geophysical methods were
successful for site characterization, but the EM and d.c. resistivity techniques
did not detect gasoline and jet fuel number 4 (JP-4) contamination when it was
present. The use of EM and d.c. resistivity for direct detection of hydrocar-
bons appears to be a subtle technique which depends on a thorough understanding
of background information at the site, the skill of the instrument operator,
and may depend on the length of time the spill has been present. The ground-
penetrating radar and complex resistivity techniques were used successfully at
a number of locations for detecting organic contamination. This work was
conducted from January 1985 to October, 1987.
i i i
-------
IV
-------
CONTENTS
Figures vi
Tables vii
1 Introduction 1
Background 1
Objectives 1
Approach 2
2 Conclusions and Summary 5
3 Methods 7
Introduction 7
Summary of Methods used at AFBS 7
Soil-Gas Techniques 7
Geophysical Techniques 13
4 Field Investigations 18
Introduction 18
Discussion of Soil-Gas Results 18
Geophysical Results 21
5 General Considerations for Site Investigations 26
Conceptualizing the Problem 27
Contaminant Source Size 31
Components of Site Investigations 31
Preliminary Information 32
Geological Techniques and Issues 36
Hydrological Techniques and Issues 37
Geochemical Techniques and Issues 38
Isotopic Techniques and Issues 39
Literature Review 40
6 Planning a Soil-Gas Investigation 53
Introduction 53
Is Soil-Gas Sampling Appropriate 53
Developing an Investigation Plan 60
7 Planning a Geophysical Investigation 62
Introduction 62
Geophysical Techniques and Issues 64
References 67
-------
FIGURES
Number Page
1 Features of an active soil-gas sampling system 11
2 Passive soil-gas sampling badge and manifold 12
3 Concentrations of total hydrocarbons in soil-gas at JP-4 spill
site, Robins AFB Sampling depth 1 meter 21
4 Concentrations of total hydrocarbons in soil-gas at JP-4 spill
site, Robins AFB Sampling depth 2 meters 22
5 Fuel leak over unconsolidated sand and gravel aquifer 41
6 Solvent leak over sand and gravel aquifer 42
7 Landfill over unconsolidated sand and gravel aquifer 44
8 Sewage leach field over unconsolidated sand and gravel aquifer 45
9 Leak over deep aquifer 47
10 Fuel and solvent leak over two interconnected aquifers 48
11 Fuel leak over crystalline fractured rock 49
12 Fuel leak over thick fissured clay 51
13 Fuel leak over karst terrain 52
VI
-------
TABLES
Number
1 Panel of Experts 3
2 Geology Climate and Contaminants at Air Force Base Study Sites 4
3 Investigation Techniques Used at Air Force Base Study Sites 8
4 Key Results from the AFB Investigations 19
5 Study Site and Contaminant Characteristics Comparison of Soil-
Gas and Ground-Water Data 20
6 Classifications of Common Organic Contaminants 28
7 Useful Data for Selected Organic Contaminants 55
8 Classes of Organic Compounds 57
9 Characteristics of the Seven Geophysical Methods 63
10 Generalized Applications of Geophysical Techniques 64
VI I
-------
SECTION 1
INTRODUCTION
BACKGROUND
In 1984, the U.S. Environmental Protection Agency (EPA) Environmental
Monitoring Systems Laboratory in Las Vegas, Nevada (EMSL-LV) and the Air Force
Engineering and Services Center (AFESC) entered into an interagency agreement
concerning investigations of subsurface contamination at Air Force Installation
Restoration Program (IRP) sites. Organic contamination was emphasized in these
studies. The traditional approach to these site investigations involves the
installation of wells and analysis of ground-water samples. This approach
provides a direct measurement of the contamination at the locations sampled.
However, information about the extent and degree of contamination may be limited
by the number, cost and possible locations of the wells. If inexpensive, and
relatively rapid reconnaissance techniques could be used as an aid to selecting
the well locations, the number of wells could be reduced. This would save
money and time.
The interagency agreement initiated studies at four IRP sites to demon-
strate indirect methods for detecting and mapping organic contamination in
ground-water and soil. The methods chosen for evaluation were soil-gas and
geophysical measurements. These measurement results then were compared to
ground-water data obtained during the same study. This made it possible to
evaluate the performance of the soil-gas and geophysical techniques for these
locations. However, because of the wide variety in contaminants and geological
conditions, care must be used when applying the conclusions developed from
these site-specific studies to other locations. To extend the results from
these studies to other site conditions, additional examples were assembled from
the literature. Using all this information, general guidelines were developed
for the use of these techniques in investigations of organic contamination of
soil and ground water.
OBJECTIVES
The objectives of the interagency agreement are listed below:
• to evaluate techniques, other than directly sampling ground water, for
detecting subsurface organic contamination under a variety of conditions;
• to recommend appropriate applications for the alternative techniques
based on the field experience; and
• to recommend research that is needed to further the use of these
techniques.
-------
The approach for meeting these objectives is described below.
APPROACH
The overall approach to the project was divided into two parts with activi-
ties in each proceeding concurrently. These parts consisted of working with a
panel of experts to broaden the ideas, approaches and experiences being used as
a basis for developing the recommendations; and performing site investigations
to demonstrate the soil-gas and geophysical techniques. The reasons the soil-
gas and geophysical techniques were chosen are described below.
Soil-gas surveying is an emerging technology for detection of subsurface
contamination through the use of near-surface techniques. The techniques of
soil-gas surveying are based on the measurement of volatile organic compounds
(VOCs) in soil gas to detect contamination in the ground water below. VOCs
dissolved in ground water vaporize into the soil atmosphere according to their
vapor pressures and aqueous solubilities. In many situations, detectable
concentrations of VOCs are present in soil gas above contaminated ground water.
Because of this, soil-gas surveying often can be used to map contaminated
ground water at a site. Since VOCs are the major components of gasoline, jet
fuel number 4 (JP-4), and many industrial solvents, this technology can be very
useful for locating commonly-occurring contamination. In these studies, active
and passive soil-gas techniques were evaluated. The results were compared
to ground-water analyses.
Geophysical techniques, developed for mineral, soil engineering, and oil
investigations, are now beinq applied to hazardous waste site investigations.
Techniques frequently used include d.c. resistivity, electromagnetic induction
(EM), ground-penetrating radar (GPR), magnetics, and seismic methods. These
methods, individually or in combination, can often provide information about
geohydrologic features, locations of buried metal objects, locations of buried
trenches, and mapping of conductive leachates and contaminant plumes. These
applications of geophysics are well understood and documented. Electrical
geophysical techniques such as EM (terrain) conductivity have been used success-
fully on a number of occasions to directly detect organic contamination
(Germeroth and Schmerl, 1987; Saunders and Cox, 1987; Saunders and Germeroth,
1985; Valentine and Kwader, 1985; and Saunders, et. al., 1983). However, these
methods generally have not been accepted for routine use because the physical
response occurring is not well understood. The use of GPR and complex resis-
tivity to directly detect organic contamination has been documented for a
number of locations (Olhoeft, 1986). Geophysical techniques are subject to
interferences from a variety of sources depending on the technique. These
interferences, which include the presence of metal objects, pipelines, power-
lines, radio transmissions, and ambient noise, may prevent the collection of
useful data at a particular location. In this study, all the methods mentioned
above except ground-penetrating radar were demonstrated at one or more loca-
tions. At most bases, the geophysical techniques were used to determine
physical characteristics such as depth to bedrock, or depth to the water table.
In some cases, geophysical methods were used to locate buried metal objects.
The direct detection of organic contamination was also attempted using EM and
d.c. resistivity techniques. In this report, the emphasis will be on using
soil-gas and geophysical techniques to directly detect organic contamination.
-------
Panel of Experts
As mentioned earlier, a panel of experts was chosen to provide advice on
site investigation approaches. These experts and their fields of expertise are
listed in Table 1. Each expert was assigned the task of describing the approach
they use for investigations at hazardous waste sites. These approaches were
compiled for the entire group to review and later discuss at a 2-day meeting.
The resulting information was of value to the field studies in progress, and
where appropriate, has been included in this report.
TABLE_1. PANEL_OF EXPERTS
Name
Affiliation
Area of expertise
Dr. John Cherry
Hydrologist
Dr. Gary Robbins*
Hydrologist
Geoflow, Limited
Waterloo, Ontario, Canada
Woodward-Clyde Consultants
Santa Ana, CA
Dr. Thomas Spittler U. S. EPA, Region 1
Chemist Lexington, MA
Dr. Dorm Marrin**
Hydrochemist
Dr. Gary Olhoeft
Geophysicist
Mr. Wayne Saundersj
Geophysicist
Dr. Aldo Mazzella
Geophysicist
Dr. Eric Waltherf
Environmental
Scientist
Tracer Research Corp.
Tucson, AZ
U. S. Geological Survey
Denver, CO
Camp, Dresser, and McKee,
Inc. Annandale, VA
U. S. EPA, EMSL-LV
Las Vegas, NV
Lockheed EMSCO
Las Vegas, NV
organics in aquifers, wells,
ground-water sampling
organics in aquifers, soil-
gas, soil-core head space
analysis
soil-gas analysis,
analytical chemistry
soil-gas analysis, vapor
extraction, aquatic chemistry
electrical geophysical
techniques
electrical geophysical
techniques
electrical geophysical
techniques
environmental monitoring
programs
*Present affiliation, University of Connecticut, Storrs, Connecticut
**Present affiliation, Consulting and Research Scientist, La Jolla, California
tPresent affiliation, Terrascan, Inc., Springfield, Virginia
^Present affiliation, Versar, Inc., Columbia, Maryland
Selection of Study Areas for Investigation
The other part of the project consisted of field demonstrations and quali-
tative comparisons of soil-gas and geophysical results to ground-water data.
Geophysical techniques also were used as part of the site characterization.
Four Air Force bases (AFBs) were chosen for these investigations after review of
-------
preliminary information from a total of eighteen. The selection criteria
included the following:
• presence of JP-4, gasoline, or solvent contamination from a relatively
recent spill (within the last 20 years);
• type of source of contamination; e.g., surface spill, pipeline leak,
fire training area;
• depth to aquifer less than 100 meters;
• type of geology, e.g., karst, alluvium, marine sand; and type of soil;
• conductivity of aquifer;
ease of access both to the study area, and the ground surface; sites
with contamination under large paved areas were excluded; and
number of wells already in place, delineating contamination.
The bases selected are listed in Table 2. Each AFB provides differing geology,
climate, depth to water table, and contaminants, thus representing a variety of
situations for performing the comparisons. The methods used in the studies are
listed and described in Section 3, "Methods." The studies and results are
described in detail in individual site reports (Baker et. al., 1987; Pitchford
and Scarborough, 1987; and Pitchford et. al., 1987). Results from these site
investigations are summarized in Section 4, "Field Investigations."
TABLE 2. GEOLOGY, CLIMATE, AND CONTAMINANTS AT AIR FORCE BASE STUDY SITES
Base Geology Cl i mate Contaminant
Holloman AFB sand, interbedded arid gasoline, JP-4,
clay solvents
Phelps Collins karst humid solvents, JP-4, buried
ANGTB metallic objects
Robins AFB marine sand humid JP-4, solvents
Tinker AFB clay humid JP-4
This series of studies was intended to help develop a hierarchy of tech-
niques which could be logically adapted and applied to detect contamination for
a variety of site conditions. However, the results from the field studies fit
better into a framework of broad guidelines rather than into a detailed strategy
which ranks techniques. These broad guidelines are provided in Sections 5, 6,
and 7, "General Considerations for Site Investigations," "Planning a Soil-Gas
Investigation," and "Planning a Geophysical Investigation," respectively.
4
-------
SECTION 2
CONCLUSIONS AND SUMMARY
From 1985 through 1987, the Air Force Engineering and Services Center
funded research at the U.S. Environmental Protection Agency Environmental Moni-
toring Systems Laboratory in Las Vegas, Nevada through an interagency agree-
ment. This agreement provided for investigations of subsurface contamination
at Air Force Installation Restoration Program sites. The purpose of these
investigations was to demonstrate and evaluate inexpensive and relatively rapid
reconnaissance techniques which can detect and map subsurface contamination.
Information from these techniques will reduce the number of wells required in
an investigation, resulting in significant savings in terms of costs and time.
The methods chosen for demonstrations included active and passive soil-gas
sampling and analysis, and the geophysical techniques of electromagnetic induc-
tion, d.c. resistivity, seismics, and magnetics. Field studies were performed
at four AFBs; these techniques were used as appropriate. Active soil-gas
measurements were performed at all sites; resistivity and EM measurements were
performed at three sites; and passive soil-gas sampling was performed at two
sites. The other techniques were performed at one site only. The general
conclusions about the techniques based on the field work are summarized in
Section 4. Briefly, the active soil-gas sampling technique successfully mapped
solvents, gasoline, and JP-4 contamination at all four bases where it was used.
Results from one site demonstrated that the choice of sampling depth can influ-
ence the measurements obtained. Thus, it is important to perform depth profiles
at the beginning of a soil-gas study. The real-time nature of this method also
represents a significant advantage since the choice and number of sampling
locations can be evaluated as data are obtained.
The passive soil-gas technique was not as successful as the active tech-
nique in detecting contaminated ground water. However, contaminated areas were
identified successfully in some cases. Further testing of the performance of
this technique for a variety of contaminants and geologic conditions is recom-
mended before the method is used widely. Because of its low cost, this method
has great potential for reducing site investigation costs in some cases.
The geophysical methods were successful for site characterization, but the
EM and d.c. resistivity techniques did not detect gasoline and JP-4 contamina-
tion when it was present. The natural variations in background resistivity
masked any resistivity anomaly due to the presence of hydrocarbons. Based on
these results from a limited group of geologic settings, the use of EM and d.c.
resistivity for direct detection of hydrocarbons appears to be a subtle tech-
nique which depends on a thorough understanding of background information at the
site, the skill of the instrument operator, and may depend on the length of time
-------
the spill has been present.This does not preclude the use of these techniques
in site characterization. The techniques of GPR and complex resistivity were
not demonstrated at the AFBs, but their successful performance in detecting
hydrocarbons has been documented in the literature.
To place these results in context, recommendations for planning a site
investigation were presented next. These recommendations were prepared in
conjunction with members of a panel of experts assembled to provide advice to
the project. The recommendations address general considerations in designing
an investigation, provide examples and references to similar cases in the
literature, list the steps in planning a soil-gas investigation, and list
issues to be considered in planning a geophysical investigation.
-------
SECTION 3
METHODS
INTRODUCTION
This section serves two purposes. It summarizes the techniques which were
applied during the field studies; and it also briefly reviews key character-
istics of these techniques, providing references to detailed descriptions in
the literature.
SUMMARY OF METHODS USED AT AFBS
A complete summary of the techniques considered for use in the field
investigations is provided in Table 3. The goal of applying these techniques
was to either characterize the hydrogeology or determine the distribution of
contaminants so results from each of the selected techniques could be compared.
Soil-gas and ground-water sampling were conducted at all AFBs; the EM and d.c.
resistivity measurements were performed at three of the four AFBs.
Some of the techniques listed were not used in the site investigations,
although they might have provided useful information. In certain cases, the
situations were not appropriate, while in other cases, the equipment could not
be obtained easily. For example, because of expense and because sites were
chosen with wells already available, no new wells were installed. Similarly,
no soil cores were obtained, although normally this technique would be part of
an investigation. Ground-penetrating radar and complex resistivity also were
not demonstrated in the field. Cost and scheduling problems precluded the use
of ground-penetrating radar at Robins AFB where it would have been appropriate.
It would be useful to apply this technique at Robins at a later date. Complex
resistivity was not appropriate for three of the four study sites. Thick clay
was present at the fourth site, Tinker AFB, but the presence of underground
pipelines and tanks precluded the use of the technique. Aerial photography
would have been used at all the sites if suitable maps and historical informa-
tion had not been available.
SOIL-GAS TECHNIQUES
Introduction
Soil-gas sampling and analysis services are available commercially. The
technique has several common variations, but the key elements include the
collection of a sample of vapor from the soil and the detailed analysis of the
sample for key compounds which indicate the presence of contaminants. The
soil-gas sample can be removed from the soil by inserting a probe and extracting
-------
TABLE 3. INVESTIGATION TECHNIQUES USED AT AIR FORCE BASE STUDY SITES
Technique Phelps-Collins Holloman Robins Tinker
Sampling ground water yes
from existing wells
Sampling ground water no
using soil-gas probe
Soil cores no
Active soil-gas sensing yes
Passive soil-gas sensing no
D.C. resistivity yes
Electromagnetic yes
induction
Seismic yes
Ground-penetrating radar no
Complex resistivity no
Aerial photography no
yes
yes
no
yes
no
yes
yes
no
no
no
no
yes
yes
no
yes
yes
yes
yes
no
no
no
no
yes
no
no
yes
yes
no
no
no
no
no
no
s —— — — =
a sample under vacuum with a syringe or by lowering a syringe or sorbent tube
into a borehole and collecting a sample in situ. Probes can be inserted into
the ground manually or with pneumatic hammers and hydraulic rams, reaching
depths of ten meters in some soil types (Lappala and Thompson, 1984; Marrin,
1985a; Eklund, 1985; Kerfoot et al., 1986; Marrin, 1985b; Marrin, 1984; and
Devitt et. al., 1987). Alternately, a sample can be collected by burying a
collector with an absorbent such as activated charcoal (Kerfoot and Mayer,
1986b and Voorhees, 1984). After allowing a time period for diffusion of VOCs
into the sampling manifold and sorption onto the charcoal (e.g. ranging from
hours to days), the collector is removed.
Once the sample is collected, it is analyzed using on-site gas chromatog-
raphy, or transported to the laboratory for VOC analysis. Laboratory analysis
is more time consuming because of the additional handling required and cannot
provide real-time results.
The detection methods used for analysis of soil-gas samples include:
• Flame ionization detector (FID) for the full range of organic compounds
(primarily petroleum hydrocarbons);
-------
• Photoionization detector (PID) for the aromatic hydrocarbons and sulfur
species;
• Electron capture detection (ECD) for selective detection of halogenated
hydrocarbons;
• Hall Electrolytic Conductivity detector (HECD) for the specific detec-
tion of halogenated species, nitrogen-, or sulfur-containing organic
species; and
• The flame photometric detector (FPD) for sulfur and phosphorus
compounds (Devitt et. al., 1987).
The real-time analysis allows for selection of additional sampling locations so
that contamination can be mapped with greater resolution than could be achieved
if the same number of sample points were sampled on a regular grid.
Some initial indication of contamination can be obtained using a
commercially-available organic vapor analyzer (OVA), but since these field
screening devices do not provide compound-specific identification, the results
can be confusing if several types of contaminants are present. In addition,
these devices are not as sensitive as field gas chromatography (GCs), so it is
possible that low-level contamination, which is often associated with plume
boundaries, may not be detected.
Soil-gas techniques may be influenced by airborne, surface, and subsurface
VOCs. Thus air blanks and vertical profiles should be obtained periodically.
Also, variations in the air permeability of soils resulting from utility corri-
dors, clay layers, and fractures will influence the soil-gas results, requiring
careful interpretation. Driving gas sampling probes into the ground to depths
of 1 to 3 meters (3 to 9 feet) may create a safety hazard if the probes puncture
underground utilities or buried drums which cannot be located prior to sampling.
Two soil-gas sampling techniques were used in the studies at the AFBs. Both
techniques have advantages and disadvantages which are described below.
Active Soil-Gas Sampling
In active (grab) sampling, a hollow pipe is driven into the ground to a
prescribed depth and soil gases are pulled up to the surface through it. Alter-
natively, a syringe or evacuated cylinder may be lowered down a borehole and a
gas sample collected in situ. The sample is then analyzed by gas chromatography
at or near the sampling location. This method offers the benefit of producing
immediate results as the survey progresses, a feature which is attractive
because it allows the sampling plan to be changed on the basis of results. In
addition, preliminary measurements can be performed which permit investigators
to optimize certain survey parameters such as sampling depth and sample volume.
An additional advantage of on-site soil-gas analysis is the use of analytical
instruments to screen soil and ground-water samples which are produced. The
drawback of this approach is that it requires the presence of sophisticated
analytical and sampling equipment on site. The presence of this equipment, a
specialist to operate and maintain it, and associated support systems such as
generators, calibration standards and carrier gases, make the technology
-------
somewhat expensive. This technique is most appropriate at sites where plume
mapping is the major objective or where relatively low concentrations of VOCs
are anticipated in soil gas.
Active soil-gas investigations were conducted at a variety of locations at
all four AFBs. In determining the extent of the contaminated areas, the samples
were analyzed for methane, benzene, ethyl benzene, toluene, xylene, total
nonmethane hydrocarbons, and halogenated organics. The contractor used an ana-
lytical field van equipped with two Tracer GCS with FIDs and two computing inte-
grators which permitted real-time sampling and analysis of the soil gas. This
van was also equipped with a specialized hydraulic ram mechanism used to drive
and withdraw the sampling probes. The probes consisted of 2.1-meter (7-feet)
lengths of 1.9-centimeter-diameter (3/4-inch) steel pipes fitted with detachable
drive points. A hydraulic hammer was used to assist in driving the probes
through hard soil.
Soil-gas samples were collected from depths ranging from 0.6 to 2.4 meters
(2 to 8 feet) in the ground. The key features are shown in Figure 1. The
above-ground ends of the sampling probes were fitted with a steel reducer and a
length of polyethylene tubing leading to a vacuum pump. Some 3 to 5 liters of
gas were evacuated with the vacuum pump. Samples were collected by inserting a
syringe needle through a silicone rubber segment, just above the reducer, in
the flowing evacuation line and down into the steel probe. Ten milliliters (ml_)
of soil gas were collected for immediate analysis with one of the GCs. The soil
gas was subsampled in volumes ranging from 1 microliter (|JL) to 2 ml_, depending
on the expected concentrations of volatiles. The hollow steel probes were used
once and discarded; the steel reducers were cleaned and baked after each use.
Using the experience at the four AFBs as a guide, the cost for soil-gas sampling
and analysis for fifty locations in one area is $7,500. Thus the cost per
sample location is approximately $150.
Passive Soil-Gas Sampling
There are at least two passive soil-gas sampling techniques in use. Each
technique uses the same sampling technique, i.e., shallow burial in soil, but
the collectors and analysis are different. One technique uses a thin ferromag-
netic wire coated with adsorbent charcoal. When the sample is returned to the
laboratory, the wire is heated in a vacuum and the desorbed compounds are
analyzed by Curie point mass spectrometry (Voorhees, 1984). The other technique
uses commercially-available, charcoal-adsorbent organic vapor monitor badges
(3M™) that were designed to be worn by personnel for occupational exposure
monitoring (Kerfoot and Mayer, 1986). These badges cost $7 to $10 each based
on the quantity purchased. This method was used at Robins AFB and Tinker AFB
to demonstrate the use of this passive sampling technique in sandy and clayey
soils, respectively.
To collect a soil-gas sample, the badges were installed in sampling mani-
folds constructed from clean, 0.95 liter (l-quart) aluminum cans. The sampler
and manifold are depicted in Figure 2. The manifold-sampler combination was
placed in a shallow hole dug 0.3 meters (1 foot) into the ground, covered with
soil, and left in place for a set period of time based on estimated ambient
concentrations. Passive samplers should generally be buried 0.6 meters (2 feet)
10
-------
10 CC GLASS.
SYRINGE
SYRINGE NEEDLE
SILICONS
RUBBER TUBE
HOSE CLAM
1/4 INCH TUBING
SILICONE RUBBER
•TUBE CONNECTION
TO VACUUM PUMP
ADAPTER FOR SAMPLING
SOIL-GAS PROBE
CLEAR TUBING SLEEVE
CONNECTOR (DISPOSABLE)
SOIL-GAS FLOW
DURING SAMPLING
.3/4 INCH
GALVANIZED PIPE
f
-DETACHABLE DRIVE POINT
Figure 1. Features of an active soil-gas sampling system
(courtesy of Tracer Research Corporation).
11
-------
DIFFUSION PATH-
DRAFT SHIELD
CHARCOAL
OLLECTION SURFACE
MARKER RIBBON
WIRE LOOP
SOIL
Figure 2. Passive soil-gas sampling badge and manifold
(after Kerfoot and Mayer, 1986).
12
-------
below grade to minimize the effects of atmospheric air dilution of soil gases.
Then the samplers were removed, sealed, and returned to the laboratory. At the
EMSL-LV laboratory, the VOCs were solvent-desorbed and the resulting solution
analyzed by gas chromatography. The procedures used are described in Kerfoot
and Mayer, 1986; Mazzella et. al., 1987; and Pitchford and Scarborough, 1987.
The estimated cost per analysis for the charcoal badges is $75 to $100, based on
this experience. Analytical services for these badges are offered as a service
by a number of companies at costs ranging from $48 to $58 depending on the type
of analysis requested. Analysis of one to three compounds can be requested.
To perform a full-scale investigation using the passive soil-gas technique,
it is desirable to determine an optimum exposure time and depth for the study
area. This is accomplished by performing field calibration tests, which are an
important part of the study process. Through this procedure, it is possible to
assess whether the technique can be used to detect and delineate subsurface
contamination at a site.
GEOPHYSICAL TECHNIQUES
Introduction
Geophysical techniques measure a variety of properties of the earth. For
example, ground penetrating radar is a reflection technique that measures
changes in electromagnetic propagation velocity. Electromagnetic induction
measures the electrical conductivity of the subsurface with lower frequency
electromagnetic waves. D.C. Resistivity measures subsurface electrical resis-
tivity which is the reciprocal of conductivity. Seismic refraction Involves
transmission of sound waves into the ground. Using measurements of the travel
time of the waves, the thicknesses and depths of geological layers can be
established. Magnetometry measures anomalies in the earth's magnetic field
caused by ferrous objects such as iron or steel. These techniques can be used
for defining natural geologic features; locating conductive leachates and
contaminant plumes; locating buried trenches and locating metal objects (Benson
et. al., 1983). This section briefly reviews the characteristics of these
methods.
Electromagnetic Induction
Electromagnetic induction is the most rapid and inexpensive of the geo-
physical techniques discussed in this report. It is readily available commer-
cially and acquires data for electrical conductivity over a large area. Depth
of electromagnetic penetration is a function of coil spacing, frequency, and
electrical conductivity. These depths are typically on the order of meters to
tens of meters with hand-held instruments. There are a variety of commercially-
available instruments that can be used to explore different depths, depending
on the conductivity of the surface. Data are acquired by measuring the pertur-
bation in the signal between two coils of wire which is due to the presence of
a nearby conducting material (the earth), and which is proportional to the
conductivity of the material. If the site relief is greater than one meter,
the data may require topographic correction from the surface of the earth to
the water table. Nearby utilities, gas pipelines, power and telephone lines,
and metal fences and debris can interfere with the measurements. An electrical
13
-------
conductivity map can reflect changes in porosity, water saturation level,
salinity of the ground water, or the presence of clay lenses. It can generally
illustrate the uniformity of a site subsurface. Basic sources of information
about EM include Keller and Frischknecht, 1966; McNeill, 1980; Benson et. al.,
1983; and Greenhouse and Harris, 1983.
D.C. and Complex Resistivity
D.C. resistivity is also readily available commercially. Instead of coils
of wire which do not touch the earth, the d.c. resistivity method makes physical
contact to the earth using shallow (<0.3 meter) electrodes. By injecting
current directly into the ground and by measuring the voltage response, the
apparent resistivity of the earth is measured. Interpretation of these data
can indicate various layers, possibly reflecting the depths of the water table,
aquitards, and bedrock. The geometry of the electrode arrays and spacings
determines the depth of investigation. The technique requires more time than
EM to cover a given area. Resistivity soundings, however, can give more
detailed depth profiles than commercially available EM methods. The technique
requires topographic correction and may also be subject to interference from
utilities. The d.c. resistivity method can be used as described above for the
EM method for profiling and mapping. Basic sources of information about d.c.
resistivity include Zohdy, 1974; Benson et. al., 1983; and Greenhouse and
Harris, 1983.
Complex resistivity is the technique of measuring resistivity in both
magnitude and phase as a function of frequency (also called induced polariza-
tion). The technique requires costly equipment and more time than conventional
resistivity and is thus more expensive. However, the frequency dependent
measurement gives information about active chemical processes in the earth as
well as the same information acquired by EM or conventional resistivity This
technique has shown the ability to detect and map organic materials in the
presence of clay by mapping clay-organic reactions. There are few available
commercial sources for this technique. Basic sources of information about this
technique include Sumner, 1976; and Olhoeft, 1984b, 1985, and 1986.
Ground-Penetrating Radar
Ground-penetrating radar (GPR) is readily available commercially, rapidly
provides very high spatial resolution over a large area, can work close to
utilities, but is more expensive than EM or resistivity It is cheaper than
complex resistivity or seismic techniques. GPR emits transient electromagnetic
pulses with energy centered at frequencies in the range of 80 to 1000 megahertz.
The wave fronts are reflected when they encounter contrasts in the dielectric
constant, such as the water table, bedrock, and clay layers. The reflected
waves are plotted as a function of depth, and topographic correction is required.
The depth of penetration is controlled by the intrinsic conductivity of the
earth, the amount of inhomogeneity in the earth, and the amount of clay and
water present (Olhoeft, 1984a and 1986). In clay-free sand with resistivity
above 30 ohm-meters, the ground-penetrating radar can map bedding and stratig-
raphy, water tables, bedrock interfaces, and other features with dielectric
contrasts at a resolution of a few centimeters to depths of 30 meters (Wright
et. al., 1984). Five to ten weight percent montmorillonite clay will reduce
14
-------
the depth of penetration to less than one meter. As dielectric contrasts do
exist between most earth materials and many organic substances, it is possible
to detect certain kinds of organics with ground-penetrating radar (Kutrubes,
1986 and Olhoeft, 1986). Basic information may be found in Ulriksen, 1982 and
Benson et. al., 1983.
Seismic Techniques
Seismic compressional and shear wave, reflection and refraction techniques
are readily available commercially and can be used to determine stratigraphic
and lithologic layer thicknesses and depths. This is the most expensive of the
geophysical tools discussed in this report. Commercially-available seismographs
can plot the arrival times of elastic waves refracted or reflected from these
subsurface features. Sources of seismic energy include a sledge hammer striking
a steel plate on the ground, specialized shotguns, or explosives. Subsurface
velocities are measured or estimated to allow calculation of the depths from
the travel times, and topographic correction is required. Seismic refraction
works if each successively deeper layer has a higher propagation velocity. Both
seismic techniques can provide information at great depths, but they do not
easily provide information on features shallower than 3 meters (10 feet). Any
nearby loud noise source such as a busy highway or construction may interfere
with the survey. Seismic techniques are not as rapid as EM and GPR. The
seismic techniques work best in solid materials with no fractures and perform
very poorly in loose materials. In clay-free sandy soils, GPR will work better
than seismic techniques and with higher resolution. In clay-bearing soils,
seismic techniques will work better than GPR. Marine seismic techniques are
useful in mapping stratigraphy below rivers and lakes. As there are no acoustic
contrasts between geological materials and organic contaminants, seismic tech-
niques cannot directly map organic contamination. Basic information on seismic
techniques is in Benson et. al., 1983; Miller et. al., 1986; and Romig, 1986.
Magnetometry
Magnetometry is an inexpensive,readily available technique which measures
the intensity of the earth's magnetic field. The presence of ferrous objects
such as iron drums creates a perturbation in the local intensity and direction
of the earth's magnetic field. The change in intensity is proportional to the
mass of the object. Detection of these ferrous objects depends on the mass,
magnetic properties, orientation, and depth of the object; the intensity and
direction of the earth's magnetic field; and the sensitivity of the magnetome-
ter. A large number of magnetometers are available commercially; two common
types are the fluxgate and proton magnetometers. The fluxgate measures a
component of the magnetic field and the proton magnetometer measures the total
magnetic field. Magnetic field measurements can be made in two ways; the
magnetic field can be measured, or a difference (gradient) can be determined
between two different points. Total field measurements are more sensitive but
are also more susceptible to noise than the gradient measurements (Benson et.
a I., 1983). Cultural features such as buried pipes; metal buildings; and
magnetic properties of the soil may interfere with the measurements. According
to Benson et. al., 1983, this technique can detect buried drums, define bounda-
ries of trenches filled with drums or other steel objects; and locate iron
15
-------
pipes or tanks. Basic information about magnetic techniques can be obtained
from Mabey, 1974 and Benson et. al., 1983.
Detection of Organic Contamination
A number of physical mechanisms can make the detection of organics by
geophysical techniques possible. For example, in most near-surface rocks, the
dominant electrical conduction mechanism of current is through the water in the
pore spaces of the formation. If the electrolyte is replaced by a high resis-
tivity fluid, such as a petroleum hydrocarbon, the resistivity of the formation
may increase. However, the presence of clay minerals and buried metallic
objects, such as pipes, can also significantly alter the electrical resistivity
of the subsurface. These provide competing mechanisms to the conduction through
the pore space. One of the objectives of the studies at Holloman and Robins
AFBs was to investigate whether any change in resistivity due to the presence
of gasoline or JP-4 was detectable over man-made or naturally-occuring condi-
tions, such as changes in the porosity, saturation level, the presence of clay
minerals, or buried metallic objects. For these studies, EM and d.c. resistivi-
ty measurements were performed. Other techniques which may be of use for loca-
ting subsurface organics are ground-penetrating radar and complex resistivity.
The remainder of this section on GPR and complex resistivity is adapted
from Olhoeft, 1986. Dr. Olhoeft is a member of the panel of experts. GPR uses
the propagation of electromagnetic energy; thus it is sensitive to relative
dielectric permittivity as well as the electrical conductivity. Whereas the
electrical conductivity is more sensitive to the presence of inorganic than
organics, the dielectric permittivity is more sensitive to organics than inor-
ganic. GPR has the advantage that depth resolution is controlled by the fre-
quency of measurement and is constant with depth, whereas EM for example, has
poorer resolution with increasing depth. Organic materials have relative
dielectric permittivities that range from 2 to over 40 according to Akadov,
1980. Adding organics to clay may produce no effect or a large effect depending
on whether or not the two react. As GPR cannot penetrate clay, it cannot see
any effect unless the organics coat the clay and destroy the clay-water interac-
tion without adding a new clay-organic reaction. In this latter case, GPR may
map clay-organic processes much like complex resistivity. In one example, a
hydrocarbon plume was detected directly by the GPR as the change in contrast
between the dielectric permittivity of sand and water compared to the lack of
contrast between sand and oil. The dielectric permittivities of sand and water
are 4 and 80, respectively; the dielectric permittivity of oil is near that of
water, with a value of 2. Dielectric permittivities are unitless ratios. In
another example, the plume was inferred by an indirect change due to a soil-
organic reaction. GPR is most sensitive to changes in dielectric properties in
the unsaturated zone down to and at the water table. Below the water table,
GPR cannot see changes caused by water-soluble organics directly, but may infer
their presence from changes caused by the organics. Of course, GPR is equally
effective in mapping geology above or below the water table.
Complex resistivity acquires the same information as the other methods of
measuring electrical conductivity, but also measures the frequency dependence
of the electrical properties in terms of magnitude and phase (Sumner, 1976).
The added information relates to the chemical activity in the earth, and
16
-------
directly measures the presence of active chemical processes (Olhoeft, 1985).
Generally, higher phase and nonlinearity (Olhoeft, 1979 and Olhoeft, 1985)
indicate greater chemical activity. Inorganic processes of oxidation-reduction
and of cation exchange may be quantitatively observed with complex resistivity.
Organic electrochemistry (Baizer and Lund, 1983) suggests a variety of organic
processes that may be observable with complex resistivity. To date, only those
processes involving reactions between organics and clay minerals have been
observed both in the laboratory and at hazardous waste sites. In one example
cited, the inhibition of the normal montmorillonite cation exchange process by
the organics allowed mapping of the organic plume by complex resistity (Olhoeft,
1984) .
Since the complex resistivity requires clay to map organics through clay
organic reactions, and clay severely restricts the penetration of the GPR, the
two techniques are complementary. Further, for hydrogeological information,
GPR and seismic methods are complementary because increased clay content in
loose and sandy soils improves seismic methods.
17
-------
SECTION 4
FIELD INVESTIGATIONS
INTRODUCTION
This section presents highlights of the field investigations performed at
Holloman AFB, Robins AFB, and Tinker AFB. The highlights focus on results of
the performance of soil-gas and geophysical measurements for detecting organic
contaminants. The results are summarized in Table 4. To construct the table,
the site of most interest was selected for each AFB. Methods used to investi-
gate the site were listed along with a brief summary of the result. These
results are discussed further below.
DISCUSSION OF SOIL-GAS RESULTS
Four sites were investigated using soil-gas techniques. The site condi-
tions and results of comparisons of soil-gas results to ground-water results
are summarized in Table 5. The sites represent a limited set of site-specific
and compound-specific parameters. Because of the wide variety of contaminant
and geological conditions which may exist at a given site, the conclusions
should be extended to other sites only insofar as stratigraphic and contaminant
conditions are similar. Some of the more generally applicable conclusions are
summarized below.
The comparison portion of Table 5 provides the results from active and
passive soil-gas and ground-water sampling conducted at three AFBs. For each
site, soil-gas measurements nearest to ground-water sampling locations were
selected. Distances between the points of comparison ranged from a maximum of
15 meters (50 feet) at Holloman AFB, to 6 meters (20 feet) at Robins AFB, and
1.2 meters (4 feet) at Tinker AFB. The values were compared qualitatively by
classifying the concentrations as background or above background. The cases
for which the classifications of the soil-gas and ground-water data agreed were
counted and presented as a ratio to the total number of cases. For example, at
Holloman AFB, active soil-gas sampling results agreed with ground-water sampling
results for 8 of the 12 comparisons. These results are discussed in more
detail below.
Active Soil-Gas Sampling
The active soil-gas sampling technique was generally successful at all
AFBs in delineating contamination over the areas where contaminants were present
in the ground-water. Paired soil-gas and ground-water samples showed agreement
at approximately 75 percent of the locations. This percentage might have been
higher if the soil-gas sampling locations had been closer to the ground-water
18
-------
TABLE 4. KEY RESULTS FROM THE AFB INVESTIGATIONS
Base, site, and
contaminants
Method
Comment
Holloman AFB,
BX Service
Station,
Gasoline
Robins AFB,
JP-4 Spill,
JP-4
Active soil-gas sampling
EM, d.c. resistivity
Active soil-gas sampling
Passive soil-gas sampling
Compares favorably with ground-
water data. Demonstrates move-
ment of contaminants along
utility corridors.
Do not detect organics because
of natural variability in soil
resistivity. Culture limited
extent of survey.
Compares favorably with ground-
water data in spite of 20-year
age of spill. Demonstrates
importance of depth of sampling.
Preliminary test has mixed
results compared to ground-
water data.
EM, d.c. resistivity
Tinker AFB,
Fuel Farm
290, JP-4
Active soil-gas sampling
Passive soil-gas sampling
EM, d.c. resistivity,
complex resistivity
Do not detect organics because
of natural variability in soil
resistivity due to rainfall
effects and culture. AFB radar
interferes with EM-34 measure-
ments.
Compares favorably with ground-
water data; technique effective
in clay soil.
Preliminary test has mixed
results compared to ground-
water data. Technique may be
responding to surface contami-
nation at times.
Were not attempted due to high
density of buried pipes and
tanks, and fences and pipes on
surface.
19
-------
TABLE 5. STUDY SITE AND CONTAMINANT CHARACTERISTICS; COMPARISON
OF SOIL-GAS AND GROUND-WATER DATA
AFB
Age
Depth to
ground water
Contaminant (years) Soil type feet meters
Comparison
Active Passive
agree/total agree/total
Holloman Gasoline
Robins
Tinker
JP-4
JP-4
20
unknown
sand/clay
sand
clay
2.0
2.0
2.6
8/12
6/8
9/12
not avail-
able
6/9a
5/12
"These results for Robins AFB use the data for exposure times of 3 days. The
3-day exposure times provided more consistent results than the shorter times
of 1 and 2 days.
sampling locations. For example, the boundary of the contamination at Holloman
AFB was very distinct. In one case, the soil-gas and ground-water data did not
agree. This discrepancy appeared to result from the relatively long distance
between the soil-gas probe and the monitoring well. However, there are other
anomalies which cannot be attributed to spatial differences. This pattern of
soil-gas anomalies also seemed to be true for the other AFBs. In these cases,
the differences between soil-gas and ground-water data have been attributed to
local heterogeneities in soil or to sampling difficulties. It is important to
rely on the overall pattern indicated by the active soil-gas data, rather than
on single values, in estimating the location of ground-water contamination.
The depth of sampling can be very important as shown by the results at
Robins AFB. At this AFB, initial sampling at 1 meter revealed very little
contamination as shown in Figure 3, while additional sampling at 2 meters
located significant contamination, which is shown in Figure 4. It is important
to perform depth profiles at a number of locations during the initial phase of
a study, preferably in regions of known (quantified) ground-water contamination,
in order to select the sampling depth. Sampling depth is particularly important
at sites where relatively old fuel spills have occurred, because oxidation
(chemical or biological) of the petroleum hydrocarbons can remove fuel consti-
tuents from the aerobic soil horizons.
The real-time nature of the active soil-gas sampling was a significant
factor in the success of the investigations at each of the AFBs. At Robins AFB,
real-time results allowed an immediate change in sampling depth when discrepan-
cies were discovered. The availability of results soon after samples were
collected offered the opportunity to choose sampling locations and depths based
on the best information available.
At Tinker AFB, the active soil-gas technique did not appear to be affected
by the presence of clay except at a few locations. It was possible to determine
when the probe was inserted into impermeable clay by observing the vacuum
20
-------
Legend
TOTAL HYDROCARBON CONCENTRATION
(ug/L) in SOIL-GAS
LF-1-2Q—Well Sampling
SG-6 • —Soil-Gas Sampling Location
000-*" Isoconcentration Contour Line
x ^ *
< 0.06—Total Concentration Value (ug/L)
10 0 10 20
SCALE IN meters
Figure 3. Concentrations of total hydrocarbons in soil gas at JP-4
spill site, Robins AFB. Sampling depth: 1 meter.
21
-------
Legend
TOTAL HYDROCARBON CONCENTRATION
(ug/L) IN SOIL-GAS
LF-1-2Q Well Sampling Location
• Soil-Gas Sampling Location
>0.06—Total Concentration
•"• 10,000 — Isoconcentration Contour Line (ug/L)
130.000
130.000
•61
-N-
10 0 10 20
SCALE IN meters
Figure 4. Concentrations of total hydrocarbons in soil gas at JP-4
spill site, Robins AFB. Sampling depth: 2 meters.
22
-------
pressure necessary to extract a sample. The sampling depth could then be
adjusted to avoid the impermeable strata. In some cases, it was not possible to
avoid the impermeable strata and the sampling location was changed. Originally,
it was anticipated that the soil-gas technique would not work well at this site
because of the clay. However, the active sampling technique performed well.
Either the permeability of the clay was adequate for soil-gas sampling due to the
presence of coarse-grained sediments, or the clay itself had been contaminated by
gas-phase VOCs or by infiltration of product.
The use of the soil-gas probe for collecting ground-water samples proved
effective in confirming the soil-gas results in a timely way. This approach is
not a substitute for the standard procedure of installing and sampling wells,
but is useful in field survey activities when the goal is estimating the extent
of the contamination. This ground-water sampling approach may not work when
the soil is very hard, or when the depth to ground water is greater than approx-
imately 3 meters (10 feet).
Passive Soil-Gas Sampling
Two of the sites investigated with active soil-gas techniques were also
investigated using passive techniques. At these sites, only feasibility tests
were performed. The purpose of these tests was to determine the feasibility of
mapping the contamination at these sites and to select the best exposure times
for the badges. Performing feasibility tests with the badges is very important;
an insufficient exposure time may indicate an area is uncontaminated when
contamination actually is present. For the exposure times used at Robins AFB,
the contaminated zone was successfully identified one out of two times for a
1-day exposure, two out of three times for a 2-day exposure, and three out of
three times for a 3-day exposure. This emphasizes the importance of carefully
selecting an exposure time. Over-exposure of the badges may result in satura-
tion of the sorbent which would mask any relative differences in soil-gas
contamination at the various sampling locations.
The passive soil-gas sampling data showed varying degrees of success in
qualitative comparisons with ground-water data. For the longest exposure time
of 3 days at Robins AFB, the badges successfully identified one area as contami-
nated and one area as uncontaminated. These results were consistent with the
ground-water data. A third area of intermediate contamination was not detected
by the badges. It is possible that the location assumed to have intermediate
contamination was actually outside the zone of ground-water contamination and
was influenced by VOCs diffusing laterally. At Tinker AFB, the badge data
matched the ground-water data for 5 out of 12 cases. In some of these cases,
the badges may have been responding to near-surface contamination, rather than
to the ground-water contamination. These data suggest that the depth of sam-
pling may be as important a consideration for passive sampling as it is for
active sampling. Overall, the data obtained from passive soil-gas sampling
with badges showed less agreement with ground-water data than did the active
soil-gas method.
The choice of which soil-gas technique to choose depends on the nature of
the investigation. At this time, there are a number of issues which need
further study before the passive soil-gas method should be used widely. Because
23
-------
of its low cost, this technique offers great promise for some sites. The issue
of greatest concern is understanding the conditions under which the passive
badge technique does not detect existing ground-water contamination. The use
of a feasibility test to establish exposure times, sampling depths, and agree-
ment with areas of known subsurface contamination at a specific site is strongly
recommended. Passive samplers should generally be buried at least 2 feet below
grade to minimize the effects of atmospheric air dilution of soil gases.
As was indicated by the results from the AFB various sites, the active
soil-gas results and ground-water data generally show good agreement. For this
reason, the active soil-gas method is recommended for routine use. Depth pro-
files should be used at the beginning of a study to determine optimum sampling
depth. In a clay environment, consideration should be given to using vacuum
values as a criterion for assessing the validity and representative nature of
samples. Moisture and organic carbon content of soils can also affect the
predictive capability of soil-gas techniques even if field sampling can be
performed without difficulty.
GEOPHYSICAL RESULTS
A number of limited geophysical studies were conducted for the direct
detection of subsurface hydrocarbon contamination. At Phelps-Collins ANGTB,
the suspected concentration of organic contamination was so low (<100 parts
per billion) (Baker et. al., 1987) that geophysical surveys were not attempted
for direct detection. At Tinker AFB, the amount of cultural interference from
tanks and pipelines was so great that geophysical surveys also were not
attempted.
D.C. resistivity and shallow EM surveys were conducted at Holloman and
Robins AFBs. These surveys were not successful at either AFB for the direct
detection of subsurface hydrocarbon contamination. At Holloman AFB, the resis-
tivity anomalies due to the presence of substantial gasoline contamination
could not be distinguished reliably from the naturally-occurring resistivity
variations in the soil of the area. At Robins AFB, where substantial JP-4
contamination was present, resistivity anomalies in the d.c. resistivity or
EM results could not be distinguished from the natural background resistivity
variations in the area. A change in the near-surface resistivity properties
due to recent rainfall further tended to complicate and mask any detection. In
Saunders and Cox, 1987, the resistivity anomalies attributed to the hydrocarbon
contamination are represented by decreases in conductivity on the order of 30
to 50 percent from background values. It may be difficult to separate a signal
of this magnitude from background variations in many circumstances. For
example, at Robins AFB, background conductivities were low, approximately 3
to 5 millimhos. The measurements performed did not have the sensitivity to
separate relative changes on the order of 1 to 2 millimhos from the background
variations.
Based on this experience and the results of two cases reported in the
literature (Saunders and Germeroth, 1985 and Saunders and Cox, 1987), the use
of d.c. resistivity or EM measurements for detection of subsurface hydrocarbons
appears to be subtle techniques which depend on a thorough understanding of
background information such as near-surface geology and potential interferences,
24
-------
the skill of the instrument operatorand may depend on the length of time the
spill has been present. This does not preclude the use of these techniques for
site characterization to obtain basic information on the electrical resistivity
properties of an area.
GPR was not used in the current studies. However, other reports and
studies described in Olhoeft, 1986 indicate direct detection of subsurface
hydrocarbon contamination by GPR surveys has been successful. The use of GPR
is limited to sites that are relatively clay free and have resistivities greater
than about 30 ohm meters. The use of initial reconnaissance EM surveys can help
define whether a GPR survey should be attempted at a site.
When clays are present at a site and the resistivities are less than 30
ohm meters, the emerging technology of complex resistivity appears to have some
potential for the direct detection of subsurface organic contamination (Olhoeft,
1986) . At the present time, further research is needed to fully evaluate the
complex resistivity method.
25
-------
SECTION 5
GENERAL CONSIDERATIONS FOR SITE INVESTIGATIONS
This section describes broad and general considerations for a site investi-
gation. The role of a conceptual model and the types of technical concerns
which should be considered are addressed. Following these general guidelines,
nine examples are presented. These examples serve to illustrate the capabil-
ities of the various techniques; references to actual cases are provided.
The type of organic compounds present at a site with subsurface contami-
nation will determine how an investigation is to be conducted and whether the
various techniques are likely to be successful. For convenience, the organic
contaminants have been categorized into groups which have similar physical
properties. These groups, listed below, were developed by Dr. Dorm Marrin, a
member of the panel of experts. The group designations will be used in discus-
sions in this section and in Section 6. More specific recommendations for the
application of soil-gas and geophysical techniques are provided in Sections 6
and 7, respectively.
Halogenated Methanes, Ethanes, and Ethenes
These compounds include chloroform, carbon tetrachloride, trichlorofluoro-
methane (Freon-n), 1,1,1-trichloroethane (TCA), 1,2-dibromoethane (EDB), vinyl
chloride, and trichloroethene (TCE).
Halogenated Propanes, Propenes, and Benzenes
These compounds include 1,2-dibromo-3-chloropropane (DBCP), 1,2-dichloro-
propane, 1,3-dichloro-l-propene, chlorobenzene, and trichlorobenzene.
Halogenated Polycyclic Aromatics
These compounds include polychlorinated biphenyls (PCBS) and organochlorine
pesticides such as aldrin, chlordane, heptachlor, and dichloro-diphenyl-
trichloroethane (DDT).
Ci-Cs Petroleum Hydrocarbons
These compounds include benzene, toluene, xylene isomers, methane, pentane,
cyclohexane, isooctane, and complex products such as gasoline and JP-4.
26
-------
C9-C15 Petroleum Hydrocarbons
These compounds include trimethylbenzene, tetramethylbenzene, napthalene,
dimethylnapthalene, nonane, decane, and complex products such as diesel and Jet
A fuels.
Polycyclic Aromatic Hydrocarbons
These compounds include anthracene, benzopyrene, fluoranthene, benzo-
fluorene, chrysene and complex products such as motor oils and coal tars.
Low Molecular Weight Oxygenated Compounds
These compounds include acetone, ethanol, formaldehyde, methyl ethVketone,
tetrahydrofuran, and phenol.
This section, except for Parts D, J, and Table 6, was adapted from
materials provided by Dr. John Cherry, a member of the panel of experts listed
in Section 1. Part D was adapted from materials provided by Dr. Gary Robbins,
another member of the panel. Part J was initially developed during discussions
by the panel, but is presented in an amplified form which was first documented
in Walther et. al., 1986.
CONCEPTUALIZING THE PROBLEM
The first step in any site investigation, and one of the most important
steps overall, is the conceptualization of the problem. To conceptualize the
problem to a useful degree, some information must be available on the nature of
the contaminant source and on the hydrogeology of the site. Useful information
on these topics is nearly always available before a site investigation begins
and therefore it is usually possible to develop a useful conceptualization
before drilling programs and monitoring networks are designed. For the concep-
tualization, it is desirable to know if the source of contamination has organic
floaters or organic sinkers (i.e., halogenated organic liquids) or simply
miscible contaminant source liquids. Table 6 lists the density and aqueous
volubility of common organic contaminants and classifies those which are insol-
uble in and less dense than water as "floaters" and those which are insoluble
in and more dense than water as "sinkers." Those which are soluble in water
are termed "mixers."
The presence, or possible presence, of organic sinkers is a particularly
important issue requiring attention in the conceptualization because the organic
liquids can sink deep into aquifers along pathways usually controlled by
geologic features. The sinking and final position are rarely influenced much
by the rate and direction of ground water flow at the site. If it is known or
suspected that halogenated solvents were at some time used on the site property
or disposed of on the site property, it is appropriate to evaluate the possi-
bilities for a significant mass of the solvent to exist in pockets or pools at
some depth beneath the property. The pockets or pools may be the long-term
cause of ground-water contamination. Depending on the depth at which they are
located, the pockets or pools may control the depth and extent of contaminant
plumes emanating from the site.
27
-------
TABLE 6. CLASSIFICATIONS OF COMMON ORGANIC CONTAMINANTS
Name
GROUP A
Chloroform
CHC13
Carbon
Aqueous
volubility
9/m3a
8,000
800
Density Classification
1.4832 sinker
1.5940 sinker
Tetrachloride
ecu
1,1,1 Trichloro-
ethane, TCA
Trichloroethene
TCE, C1CH:CCl2
Ethylene Dibromide
EDB
BrCH2CH2Br
Methylene Chloride
CH2CI2
720
1,100
4
20,000
1.4714
1.4642
2.1792
1.3266
sinker
sinker
sinker
sinker
GROUP B
Chlorobenzene
CeHsCI
1,2-Dichloro-
propane
CH3CHCIH2CI
1,2 Dibromo-3-
chloropropane
(DBCP)
1,2,4-Trichloro-
benzene
CeHsCIs
500
2,700
1,000
25
1.1058
1.1560
2.080
1.4542
sinker
sinker
sinker
sinker
(continued)
28
-------
TABLE 6. (Continued)
Name
Aqueous
volubility
g/m3a
Density
Classification
GROUPC
Polychlorinated
biphenyls, PCBS
Dichloro, diphenyl,
trichloroethane
DDT
Aldrin
Chlordane
0.04 - 0.2
0.003
0.01
0.056
sinker
sinker
sinker
sinker
GROUP D
Benzene
CeHe
Toluene
C/Hs
Xylene isomers
CsHio
Methane
ChU
n-Pentane
CH3(CH2)3CH3
n-Octane
CH3(CH2)eCH3
1780
515
162 - 185
24
38.5
0.66
0.87865
0.8669
0.8802 to 0.8611
0.5547
0°C
0.6262
0.7025
floater
floater
floater
(gas)
floater
floater
GROUP E
1,2,3-trimethyl
benzene
C9H12
75
0.8944
f I loater
(continued)
29
-------
TABLE 6. (Continued)
Aqueous
volubility
Name g/m3a
1,2,4,5-tetra- 3.5
methyl benzene
CioHu
Naphthalene 34
CioHs
1,4-dimethyl- 2.4
naphthalene
C12H12
Nonane 0.12
CH3(CH2)7CH3
Decane 0.052
CH3(CH2)sCH3
Tetradecane 0.0022
CH3(CH2)12CH3
GROUP F
Benzopyrene 0.003
GROUP G
Acetone infin
CHsCOCHs
Ethanol infin
C2H50H
Formaldehyde infin
HCHO
Tetrahydro- infin
furan
C4H80
Density
0.8875
1.4003
1.0166
0.7176
0.7300
0.7940
>1
0.7899
0.7893
0.815
0.888
Classification
floater
sinker
sinker
floater
floater
floater
sinker
mixer
mixer
mixer
mixer
"All numeric solubility data are from MacKay and Shiu, 1981; all other data are
from Weast, 1969-1970, or Verschueren, 1983. Density values are for 20°C
unless noted otherwise.
infin = infinitely soluble
na = not available
30
-------
CONTAMINANT SOURCE SIZE
Knowledge of the size of the contaminant source is important because with-
out it, there is not much basis for deciding on the spacing of monitoring loca-
tions. The plume of contamination emanating from a contaminant source which is
small in areal extent is normally narrow. Narrow plumes require close lateral
spacing of measurement locations for detection or delineation. Although the
literature on dispersion contains much controversial and problematic informa-
tion, there is sufficient data now to conclude that in many aquifers, dispersion
in the transverse lateral direction is weak and that plumes often do not spread
much laterally as they increase in length. In other words, long narrow plumes
should be viewed as the rule rather than the exception. The implications of
this generalization are great. It means that the lateral spacing of monitoring
wells or other measurement locations must be significantly less than the width
of the contaminant source. At some sites, such as those that have a local leak
in a liner or those that have had leaks from tanks or hazardous liquid supply
lines, the small dimensions at the source present a formidable difficulty.
The lack of detailed information on the location and size of contaminant
sources, as is the case for many sites, presents the greatest obstacle to the
efficient development of site investigation plans. To achieve a good probabil-
ity of detecting zones of ground-water contamination at these sites, it is wise
to consider soil-gas and geophysical techniques for mapping the contamination
rather than installing many more monitoring wells or soil sampling holes.
COMPONENTS OF SITE INVESTIGATIONS
The goal of investigations of sites that are known to be or suspected to be
contaminated by organic materials is to determine the extent and severity of
soil and ground-water contamination and to concurrently determine environmental
parameters useful for planning remedial action. Once the extent of the problem
is known, and pertinent environmental parameters are determined, plans for reme-
dial action and long-term ground-water monitoring can be developed concurrently.
Many investigative techniques are available from which to select those
appropriate for the particular site under consideration. Techniques can be
selected from the following categories.
• geological
• hydrological
• geochemical
• environmental isotopes
• mathematical models
• soil-gas sampling and analytical chemistry
• geophysical
31
-------
The challenge in any site study is to select the most appropriate combina-
tion of techniques for the specific site. This section will discuss many of
these approaches. However, the details of planning soil-gas and geophysical
investigations will be discussed separately in Sections 6 and 7, respectively.
Geological techniques such as drilling and sampling of borehole materials
will be included in nearly all site studies because in situ sampling is necessary
to confirm the degree and extent of contamination indicated by other techniques.
Hydrological techniques such as the use of monitoring wells for permeabil-
ity tests and for hydraulic-head monitoring are also an important component of
nearly all site studies where organics occur. Permeability measurements and
ground-water elevation monitoring determine the ground-water flow pattern if
the geologic framework of the site is also known. Without adequate knowledge
of the geology of the sites, determination of the hydraulic head distribution
will normally not provide for a good interpretation of the flow net. Most
types of monitoring wells can be used for three purposes: permeability tests,
head measurements, and acquisition of water samples. In recent years, however,
there have appeared several dedicated monitoring devices that provide ground-
water samples but that are not useful for permeability testing or head moni-
toring. Thus, hydrological studies of a site are not necessarily an integral
part of the monitoring phase of an investigation.
In this report, environmental isotopes refer to those isotopes in the
ground water that can be used to assist in the determination of ground-water
age or origin. The isotopes of primary interest are tritium, oxygen-18,
deuterium and carbon 14 and 13. Of these, tritium is by far the most useful in
studies of sites of organic contamination. Tritium can be used to identify
ground water that is less than about 30 years in age.
Mathematical models have a potential to aid in the development of hydro-
logical or transport interpretations. Such models offer a formal means of
displaying or assessing conceptualizations of the conditions at the site.
Mathematical models are rarely a means of reducing much of the need for site
data. Instead, they offer possibilities for making better use of the data that
are obtained. Ground-water flow models very commonly serve a useful and often
essential component in site studies. However, solute transport models that
include the combined effects of advection, dispersion and retardation rarely
serve an essential role in a site investigation.
In the development of a strategy for an investigation of organic contamina-
tion at a site, all of the various investigative techniques or tools should be
considered to better select those particular items with potential or expected
usefulness. The planned investigation should be formulated in several phases.
The phases should generally be sequential in the earliest stages. The approach
should allow for extensive feedback as phases are completed so the new informa-
tion can be applied to improve the investigation strategy.
PRELIMINARY INFORMATION
Before beginning a study, information in a number of categories is essen-
tial to aid in the choice of monitoring techniques, the design of survey grids,
32
-------
the procedures for using the instruments,and the interpretation of the data.
As mentioned earlier, this list was compiled with the assistance of Dr. Gary
Robbins, a member of the panel of experts listed in Section 1. The categories
listed have been divided into broad groups: hydrogeology; soil, surficial
geology, and bedrock; site layout; and contaminant source information. Sources
for this information include local consulting engineers, county offices, state
geological and water surveys, U.S. Geological Survey reports and maps (Handman,
1983), the National Climatic Center, the Soil Conservation Service, construction
and foundation reports for structures on site. The information described in
the list will be helpful in choosing techniques and planning survey grids.
Without this information, there will probably be a need for application of
additional techniques to provide confidence that the contamination has been
detected successfully and completely.
Hydrogeology
a. Existing wells.
(1) Locations
(2) Uses, past and present
(3) Quality of ground water, presence of contamination, for different
aquifers
(4) Well logs and driller's logs
(5) Construction specifications
(6) Typical pumping rates, hydrologic parameters such as specific
capacity, transmissivity, hydraulic conductivity, storage coefficients
or storativity and the extent of pumping influence.
b. Altitude of the water table.
(1) Regional and on-site (elevation and depth)
(2) Seasonal fluctuations, if available
c. Thickness and distribution of aquifers and aquitards; existence of perched
aquifers.
d. Ground-water flow velocity or gradient, both local and regional.
e. Soil porosity, moisture, and lithology.
f. Recharge and discharge areas.
9 Basic climatic information, including annual precipitation, and monthly
temperatures.
h. Nature of drainage conditions, and flooding.
Soil, Surficial Geology, and Bedrock
a. Types, thickness, and lateral distribution of strata.
33
-------
b. Properties of soil including color, density, porosity, infiltration rates,
hydraulic conductivity, soil suction relations, grain size distribution,
Unified Soil Classification System (USCS) classification, moisture content,
soil chemistry, and organic content
c. Type and extent of fill if present.
d. Boring logs for nearby construction.
e. Stratigraphy and lithology.
f. Location and Type of Bedrock.
(1) Mass properties (faulting, fracturing, layering, dips, and strikes)
(2) Geologic maps
(3) Regional geology
(4) Regional gravity and magnetic data
(5) Depth to bedrock
Site Layout
a. Historical and current aerial photographs.
b. Present and past use of site.
c. Topography and nature of surface, in terms of woods, vegetation, bare
soil, outcrops.
d. Location of buildings, other facilities such as runways, and survey
markers.
e. Nature and location of roads for access.
f. Nature and location of pipelines, utilities, and underground facilities
which may be conduits for contamination, obstacles to investigation
activities, or both.
9 Location of power, water, and lighting which may be needed in investigation
activities.
h. Nature of pavement including type, thickness, and reinforcement.
i . Nature of activities on site which may influence subsurface conditions,
such as irrigation, pumping wells, dewatering systems, septic fields, etc.
. Nature and location of safety hazards.
34
-------
Contaminant Source Information
a. Identity of organic contaminants.
b. Locations of spills and leaks (tanks, sumps, dumps, pipelines, impoundments,
etc.).
c. Amounts spilled or leaked; past problems of similar nature.
d. Time or duration of events.
e. How problem was discovered.
f . Depth of contamination.
9. Characteristics of problem such as odors, seepage, or a contaminated well.
h. Actions to clean up problem to date.
i . Contamination due to other sources, including chemicals, concentrations,
extent of problem, time frame of problem, remedial actions being performed.
Status of Early Knowledge
In any site investigation for contaminant migration, determination of the
geological conditions is a key task. A good monitoring strategy cannot be
developed until a considerable amount of information is obtained on the geology
of the site. If very little is known about the geology of the site before the
investigation begins, then an important early step in the investigation should
be a preliminary geological investigation.
The Zone of Relevance
In most site studies, it is usually determined that there is a depth
controlled primarily by geological conditions below which contaminants have not
penetrated. The entire zone above this depth can be referred to as the zone of
relevance. For example, if the site is situated on an unconfined sand aquifer
with an aquitard of fractured clay, the sand would constitute the zone of rele-
vance. It is obvious that the site study should focus primarily on the zone of
relevance. But to determine where the bottom of the zone of relevance is
located, a component of the site study must extend beneath the bottom of the
zone. If the bottom of the zone of relevance can be located early in the site
study, the remainder of the study can proceed with greater efficiency. Often,
drilling to determine the geological conditions provides appropriate information
to draw a tentative conclusion regarding the bottom of this zone. The presence
of tritium can be used to help define the zone of relevance. This conclusion
can then be assessed by other means such as ground-water monitoring.
35
-------
GEOLOGICAL TECHNIQUES AND ISSUES
When selecting the geological techniques of drilling and coring for a site
investigation, many questions should be addressed. The following is a list of
questions which may be useful to consider. The order of the questions is not
significant.
• What type(s) of drill rig(s) should be used?
• Can the use of drilling mud or other drilling additives be avoided and
if so, how?
• What type of soil or rock sampling methods should be used?
• What soil or rock sampling interval should be selected?
• How should the soil or rock samples be stored?
• Should an organic vapor analyser be used in the field to screen the soil
and rock samples?
• To what depth should the boreholes be drilled?
• Should the boreholes be used for installation of monitoring wells or
should they be plugged?
• What techniques should be used to plug the holes?
• If it is expected that fractures are the main route for contaminant
migration, should angle boreholes be drilled as well as vertical holes?
• Has a geologist with specific knowledge or experience pertaining to the
local geology been consulted in the development of the preliminary
geological interpretation of the site?
• To what depth is it reasonable to expect root holes, animal burrows and
desiccation cracks to penetrate from the ground surface?
• When boreholes are drilled and sampled, to what depth are weathering
features identifiable?
• If it is known or suspected that dense immiscible liquids have been
used/spilled/buried at the site, what geological contact or layer would
most likely have acted as a barrier to the sinking of the liquid?
• If possible geological barrier beds or contacts can be identified, what
is the dip of the surface along which dense immiscible liquids might
move?
• What procedures should be used to prevent further spread of contamina-
tion if this barrier is penetrated?
36
-------
HYDROLOGICAL TECHNIQUES AND ISSUES
When selecting hydrological techniques for a site investigation, many
questions should be addressed. The following list of questions can help to
better define the important decisions to be made for a given site investigation:
• Can the available geological knowledge of the site and the general
hydrological setting be used to develop a preliminary estimate of the
ground-water flow pattern?
• Is the depth to the water table relevant information? How should it be
determined?
• Should a network of monitoring wells be used to determine the ground-
water flow pattern?
• What techniques would be best for determination of ground-water
velocity?
• What type of aquifer test should be used?
• Is there a need to use pumping tests to establish the degree of
hydraulic connection between one part of the site and some other part?
• Are there aquitards at the site and if so, do they act as barriers to
contaminant migration?
• Should laboratory permeameter tests be done on core samples?
• If fractured clayey deposits occur at the site, has drilling caused
smearing of the fractures in the borehole, thereby changing the
hydraulic properties?
• Are the ground-water flow conditions observed now at the site the same
as those that existed when ground-water contamination began to occur?
• If contaminant migration is occurring at the site, does it occur by
porous media transport or fractured media transport?
• If it is expected that the fractured media transport of contaminants
occurs, how can estimates of the bulk fracture porosity be obtained for
velocity estimates?
• If the mode of flow and transport is via fractures, what is the porosity
of the blocks between the fractures?
• At what depths or in what zones do the critical solute-transport paths
occur?
• Should detailed vertical profiles of hydraulic head be obtained to
assist in the identification of depths of critical flow paths?
37
-------
How much annual infiltration is expected to occur at the site?
In the zone in which contaminant migration is most likely to occur,
what is the degree of heterogeneity and what dispersion tendencies are
expected?
What are the locations and yields of water supply wells in the area?
How does the potentiometric surface respond to precipitation and what
does this indicate with respect to the ground-water flow system?
Is the geology of the site suitable for use of multilevel monitoring
devices in single boreholes or is it necessary to drill many holes to
different depths at each location in order to monitor at many depths?
GEOCHEMICAL TECHNIQUES AND ISSUES
When selecting geochemical techniques for a site investigation, numerous
questions should be addressed.
• What are the redox conditions in the ground-water zone and is it likely
that these conditions will affect transformations or degradation of
organic contaminants?
• What is the weight percent of solid phase organic carbon in the geo-
logic materials and what degree of contaminant retardation would it be
expected to cause?
• Is there evidence of transformations (i.e., biodegradation, hydrolysis)?
• If the geological media allow contaminant transport via fractures, what
will be the influence of the matrix diffusion effect?
• Can inorganic parameters such as major ions or electrical conductance
be used as indicators of transport paths or contamination?
• Would it be useful to measure parameters such as pH, Eh, CH4 and
dissolved oxygen when sampling monitoring wells?
• Do samples of water from monitoring wells contain H2S and if so, what
does this mean?
• Are there clay-rich aquitards at the site and are they such that molecu-
lar diffusion is the dominant influence on solute transport?
• Do core samples from the geologic deposits show evidence of chemical
weathering and if so, what does this indicate regarding the development
of fractures?
• Should diffusion coefficient measurements be made on core samples?
38
-------
ISOTOPIC TECHNIQUES AND ISSUES
The presence of tritium can be used to help define the zone of relevance.
If the contamination has occurred more recently than 1953, then the presence of
tritium may be a useful indicator of the possible extent of the hydrocarbons.
However, if organic chemicals which are more dense than water have been used at
the site, then tritium is generally not a good indicator of the expected zone
of contamination. This is the case because contaminants which are more dense
than water can sink through the aquifer into zones much deeper than tritium
moves under the influence of ground-water flow alone. References for this tech-
nique include Payne, 1972; Freeze and Cherry, 1979; and Cherry, Farzolden, and
Frind, 1983. Tritium analyses are commercially available from the University
of Miami Tritium Laboratory.
When selecting isotopic techniques for site investigation, several ques-
tions should be addressed. These are listed below. The order of the questions
is not significant.
• Is it likely that the site became contaminated after 1953 and if so,
should tritium be used as an indicator of the zone of active ground-
water movement that is susceptible to post-1953 contamination?
Is there evidence that organic contamination exists in ground-water
samples that have no tritium?
If it appears appropriate to use tritium in the site investigation,
what detection limit and precision is appropriate to request in the
tritium analyses?
• Should tritium profiling be used to determine whether or not an aquitard
beneath the site is leaky?
• Should water samples for tritium analysis be acquired from monitoring
wells or by extraction of water from cores?
• Can the mapping of tritium in ground water at the site serve as a means
of delineating the zone of relevance for site monitoring?
• Is it likely that organics as dense immiscible liquids have travelled
in a manner that would mean that tritium is not a good travel-path or
travel-time indicator?
• Is it likely that isotopes in addition to or other than tritium can
play a useful role in the site study?
• If the contamination is known to have or suspected to have originated
from a lagoon or pond, is it likely that oxygen-18 and deuterium will
serve as an indicator of the source water?
39
-------
LITERATURE REVIEW
Numerous studies of subsurface organic contamination have been conducted
and reported in the literature. This discussion provides a series of examples
of contaminant and geology combinations along with references to studies of
that type in the literature. Each of the cases is meant to represent a broad
category of contaminated sites and to serve as an aid to visualizing the distri-
bution of the contamination. With some idea of the likely behavior and features
of the contamination, it is easier to plan the investigation. A brief discus-
sion of indirect monitoring techniques which may be appropriate for each case
also is provided. These conceptual drawings were developed and techniques
selected during the 2-day meeting of the panel of experts.
The characteristics of the cases include sand and gravel, clay, fractured
bedrock, and karst limestone; shallow and deep aquifers; and fuel, solvents and
landfill leachate. Actual site conditions usually would be more complex than
these hypothetical cases (Mackay and Roberts, 1985), but complexity alone would
not dictate a different choice of techniques. Instead, complexity will increase
the number of techniques necessary for complete understanding of the contaminant
location.
Nine hypothetical cases have been developed to represent common combina-
tions of sources, contaminants, and hydrogeology. The hydrogeological medium
for six of the nine cases is sand and gravel. Clay was assumed to be present
in the sand and gravel, but at low enough proportion to not affect the organic
contaminant migration. The remaining three cases deal with the complexities of
clay, crystalline fractured rock, and karst terrain.
Fuel Leak Over Unconsolidated Sand and Gravel Aquifer
In this example, shown in Figure 5, gasoline or other non-alcohol fuel has
leaked from some surface or near-surface source continuously or frequently for
several years. The fuel, which is of lower density than water, floats on top
of the water table. Some fuel is trapped in the unsaturated zone as coatings
on soil particles; and some constituents dissolve in the ground water, while
others volatilize and diffuse away from the fuel pool. The horizontal transport
distance will depend on the volume of the leak, ground-water velocity, and other
parameters. Soil-gas techniques are likely to be successful in this situation;
and if the resistivity of the soil is greater than 30 ohm-meters, GPR may be
useful . For a discussion of an actual example with attendant complexities,
Hult, 1984 describes an investigation of a crude oil leak at Bemidji, Minnesota.
Marrin, 1985 describes a gasoline leak over a sand and gravel aquifer.
Solvent Leak Over Sand and Gravel Aquifer
This case, shown in Figure 6, is similar to Case 1 except that trichloro-
ethylene and other common chlorinated solvents are denser than water and sink
through the unconfined aquifer until an aquitard is reached. The solvent may
pool in depressions on the aquitard. Each pool of solvent then acts as a
secondary source contaminating the ground water, possibly for many years after
the surface source is removed. When performing the investigation, the aquitard
should not be penetrated unless appropriate drilling precautions are applied.
40
-------
OC
UI
u.
o
-J
u
o
o
z
<
o
a
u
O
UI
-I
.1
UI
u.
s
Q
LL)
|_ <-N
< U
5 JSu
~> o z
(TOO
<
<0<
CO OC
m o
o
u
J^iu
§5
H N
<
CO
cr
as
O)
(O
t_
en
•O
•o
c
(O
•o
at
o
to
c
o
u
o
^
to
u.
t-
o
LJ
-------
K
u
UL
O
Ul
tr
o
o
z
CO
tt
u
u
H
111
>
-J
O
o-
10
1
VO
a>
01
42
-------
It is important that the well not provide a conduit for the solvent to contami-
nate lower aquifers. As in the previous case, soil-gas techniques are expected
to be successful and GPR may be useful, if the resistivity of the soil is
greater than 30 ohm-meters. For an extensive discussion of an actual case in
Pensacola, Florida, where the investigation of creosote contamination including
phenols was complicated by the presence of clay lenses, see Mattraw and Franks,
1984. Bradley, 1980 and Marrin and Thompson, 1987 discuss an investigation of
soil gas above TCE-contaminated ground water. Walther et. al., 1983 describes
a site with a soil consisting of a mixture of sand, gravel, and clay, and
chlorobenzene and benzene contamination in ground water at Pittman, Nevada.
Landfill Over Unconsolidated Sand and Gravel Aquifer
A landfill, shown in Figure 7, is a much larger area source than most fuel
and solvent spills, usually covering acres of land. The permeability of land-
fills is usually higher than the surrounding soil, and this leads to ground
water mounding beneath the landfill after rainfall. Such a mound is a dome
in the topography of the water table, whose height and permanence depends on
the frequency and intensity of rainfall. Landfills usually leach organic and
inorganic contaminants. The inorganic contaminants may be roughly colocated
with the organic contaminants, although the two contaminant types migrate at
different speeds. In this case, the inorganic plume may be useful for locating
the organic contamination. Landfills generally produce gases which can vary in
composition according to the age of the fill. These constituents include nitro-
gen and hydrogen which may be released for brief periods on the order of 2 to 3
months and carbon dioxide and methane which may be released for several years
after the placement of the fill. Soil-gas sampling for carbon dioxide, methane,
or VOCs may be useful for locating the contamination. If the depth to the
aquifer is less than 8 meters., and a conductive ground-water plume is present,
EM, resistivity, or GPR may be useful for detecting the inorganic contamination.
Measurements using EM, GPR, and magnetics on top of the landfill can locate the
the presence of metal trash. Cherry, 1983 describes an actual case at Borden
landfill in Ontario, Canada. Seitz et. al., 1971 describes the effect of a
landfill on the hydrogeologic environment.
Sewage Leach Field Over Unconsolidated Sand and Gravel Aquifer
A centralized sewage leach field, shown in Figure 8, typically covers a
large area, similar to a landfill. The sewage water causes a ground-water mound
beneath the sewage leach field. The concentrations and presence of the organic
contaminants will be variable because some of the sewage is biodegradable. If
the water table is less than 30 meters (98 feet) in depth, the EM and resistiv-
ity techniques may detect a conductivity increase caused by the presence of
inorganic constituents such as ammonia and nitrates. Soil-gas sampling for
methane may be ineffective if soil moisture beneath the leach field reduces the
effective porosity to less than five percent (Marrin, 1984). Seismic techniques
can be considered for determining the depth of the water table when it is
greater than 30 meters. LeBlanc, 1984 describes the investigation of such a
case at Cape Cod, Massachusetts.
43
-------
oc
E
o
ff
o
flC
u
o
u.
o
S-
(D
3
cr
•o
o
•o
to
(U
s_
CO
44
-------
SEWAGE LEACH FIELD OVER SAND AND GRAVEL AQUIFER
SAND
OR
GRAVEL
OR WATER
OR TABLE
SEWAGE
LEACH
FIELD
UNSATURATED
(VADOSE)
ZONE
NiJ*? *J®s *'*4*^' -4
f ' "' '
w$^™&: ;^\
vV ^',' S-^'*' "S' f'i* t^* ^ rf ^
SATURATED
ZONE
* ,\is %, 'f *s *, ,5%
? ^ % * V ,. J i
:t ^*^?;>^'u<;
. ' '™*AsA. . '. %.% %/ ••s
DIRECTION OF GROUND-WATER FLOW
Figure 8. Sewage leach field over unconsolidated sand and gravel aquifer.
45
-------
Leak Over Deep Aquifer
In this case, shown in Figure 9, there is reason to suspect that the fuel,
solvent or landfill leachate has not reached the water table. This may be due
to the volume of the leak or the depth to the aquifer. Other reasons this may
occur include degradation, volatilization, sorption onto clay and soil particle
surfaces, impermeable layers of clay, moisture barriers, or low infiltration
rates. The leak will descend in a narrow, vertical column unless clay lenses
or variations in permeability redirect the flow. Analysis of soil gas near the
source can determine the lateral diffusion of the volatile organic contaminants.
Core sampling can establish the vertical extent and actual concentration of the
contaminant. When performing the investigation, care should be used to avoid
breaching confining layers, thereby allowing new migration paths for the contam-
inant. This case is similar to a study of contamination in the unsaturated zone
from low-level radioactive hospital waste containing organic solvents and
carriers (Nichols, 1986 and Beers and Morey, 1981).
Fuel and Solvent Leak Over Two Interconnected Aquifers
This case, shown in Figure 10, provides more structural complexity. Fuel
will float on top of the unconfined aquifer, while solvent will form pools on
top of both aquitards. Some solvent will dissolve in the fuel, and some fuel
components will dissolve in the solvent. The ground water may flow in different
directions in the two aquifers. Therefore, detecting the plume in the uncon-
fined aquifer does not locate the plume in the confined aquifer. The silt/clay
aquitard between the aquifers acts as a barrier to the upward migration of
organic vapors and to the penetration below the unconfined aquifer of electri-
cal current as might be used in a geophysical survey. Resistivity can provide
information on the hydrogeology, possibly identifying the depth of the water
table and both aquitards. Resistivity is more likely to determine the complex
structure than is EM. Seismic techniques can also be used to help determine
these depths. Soil-gas techniques can map the contamination in the upper
aquifer. The drilling of monitoring wells must proceed carefully. The aqui-
tards should not be penetrated unless appropriate drilling precautions are
applied. It is important that the drilling not provide a conduit for the
contaminants to move between aquifers, yet it is the only method to obtain
information on the lower aquifer. An actual example of a case with multiple
aquifers is the St. Louis Park study (Ehrlich et. al., 1982).
Fuel Leak Over Crystalline Fractured Rock
This case, shown in Figure 11, illustrates the complexity of contaminant
migration where the presence and orientation of the fractures dictates the
pathways followed by the organic contaminants. Aerial image analysis can find
major fractures or fracture patterns; field mapping should be used to check
these results. This information may suggest where to install wells. GPR, with
its continuous profiling capability, may locate fractures if the resistivity of
the rock is greater than 30 ohm-meters. Otherwise, seismic or resistivity tech-
niques should be used. Resistivities for many rock types are given in Benson
et. al., 1983. Soil-gas sampling at locations of fractures indicated by the
earlier techniques can be used to develop further information regarding the
presence and type of contamination in the fractures. Monitoring wells may have
46
-------
LEAK OVER DEEP
FUEL,
SOLVENT
OR
LANDFILL
VOLATILES
CLAY LENS
\\ :n
jj "• A
SAND
OR TABLE
GRAVEL
WATER,
Figure 9. Leak over deep aquifer.
47
-------
CO
c
u
U.
O
u
o
UJ
O
O
o
I-
c
u
o
UJ
UJ
O
-I
UJ
o>
•o
(U
•«->
u
(U
c
c
o
u
L.
(U
OJ
>
o
(U
o
CO
3
01
o
UJ
48
-------
FUEL LEAK OVER CRYSTALLINE FRACTURED ROCK
\ FUEL (
\LEAK ;
\
I ' <""\
f : ''- "' \
UNSATURATED
CVADOSE)
ZONE
SAND OR
GRAVEL
WATER
TABLE
*r\\o
DIRECTION OF
GROUND-WATER
FLOW
CRYSTALLINE
FRACTURED
ROCK
Figure 11. Fuel leak over crystalline fractured rock.
49
-------
to be angled to intercept the fracture pathways. Taylor, 1984 evaluates methods
for measuring hydrological variables in fractured rock units. Examples of
geological characterization of waste sites located over fractured rock include
Davison et. al., 1982; Olsson et. al., 1984; and Jones et. al., 1985.
Fuel Leak Over Thick Fissured Clay
In this case, shown in Figure 12, the leaked fuel travels along unknown
pathways through the clay to the top of the water table. The fuel forms a pool
at the water table and the fuel constituents dissolve from that pool into the
ground water and travel downgradient. As in the case above, the pathway is
unknown. Clay has a high electrical conductivity and it is difficult for the
electromagnetic energy of a current field to penetrate. The clay is also a
barrier to the upward diffusion of VOCs. Clay layers at any depth above the
first aquitard have the same effect. Monitoring wells and soil cores have the
highest likelihood of success. It may be worthwhile to have an experienced
geophysicist apply the complex resistivity technique. Zehner, 1983 describes a
hydrogeological investigation for a site with fractured shale.
Fuel Leak Over Karst Terrain
Karst terrain, shown in Figure 13, usually contains a network of complex
fractures, channels, caves, and underground streams which are the migration
pathways for contaminants. As in the crystalline fractured rock example,
aerial image analysis can be used to find major features of the karst network
with field mapping to check the results. Tracers are one of the best methods
to determine flow paths. This information may suggest where to install wells.
Ground-penetrating radar can be used to locate fractures if the resistivity of
the rock is greater than 30 ohm meters. Its capability of continuous profiling
is particularly useful in this case. D.C. resistivity should be used instead
of GPR if the apparent resistivity of the soil is less than 30 ohm meters.
Seismic techniques should be used along with both of these electrical methods
to obtain independent information on the location of major geologic features of
the karst terrain. Sampling and analysis of soil gas at channel openings can
indicate the presence and type of contamination. Monitoring wells may be
placed using the results of the surveys. The monitoring wells may have to be
angled to find the channels. Quinlan and Ewers, 1985 discuss the complexities
of ground-water monitoring in karst terrain and recommend a strategy which
monitors existing springs. A description of a site investigation in karst
terrain is given by Franklin et. al., 1981.
50
-------
FUEL LEAK OVER THICK FISSURED CLAY
FUEL
LEAK
I
CLAY
UNSATURATED
CVADOSE)
ZONE
SAND
OR
GRAVEL
WATER TABLE
DIRECTION OF
GROUND-WATER
FLOW
VOLATILES
FUEL POOL
FLOATING FRACTION
SATURATED
ZONE
SMte-^W^^^* •"****
SINKING FRACTION
Figure 12. Fuel leak over thick fissured clay.
51
-------
FUEL LEAK OVER KARST TERRAIN
VOLATILES
KARST
SOLUTION
CAVITY
FRACTURES
\
DIRECTION OF
GROUND-WATER
FLOW
Figure 13. Fuel leak over karst terrain.
52
-------
SECTION 6
PLANNING A SOIL-GAS INVESTIGATION
INTRODUCTION
This section was adapted from materials provided by Dr. Dorm Marrin, a
member of the panel of experts listed in Section 1. The purpose of this section
is to identify the types of subsurface contaminant problems which are amenable
to soil-gas sampling. Furthermore, it is designed to illustrate environmental
conditions under which soil-gas sampling is either not appropriate or subject
to misleading interpretations. Soil-gas investigations must be designed and
interpreted according to the hydrologic/geologic setting and chemicals which
are present at each site. A variety of investigative and interpretive tech-
niques may be used at a single site if there are variable environmental
conditions or multiple objectives to fulfill.
The section is organized into two parts; the first assesses the applicabil-
ity of soil-gas sampling under generalized conditions; and the second develops
an investigation plan by presenting a hierarchy of questions addressing the
geology, hydrology, contaminant types, and objectives. The first part includes
four questions which assess the applicability of soil-gas sampling for common
contaminant groups under general subsurface conditions and another four ques-
tions which require more detailed site-specific information. These questions
provide a basis for estimating the effectiveness and interpretive limits of
soil-gas analyses. In the second part are two questions which provide insight
into the formulation of an on-site investigation plan.
Questions presented in this section are applicable to all active soil-gas
sampling methods. Several of the criteria emphasized here are not directly
applicable to passive sampling techniques. The guidance was designed specifi-
cally for active soil-gas sampling techniques combined with immediate on-site
analysis of the samples. With this approach, the real-time data are produced
within a time period which permits the results to be used in selecting
subsequent sampling locations.
Questions contained in the two parts are designed to provide users with
the necessary criteria to decide whether soil-gas sampling is appropriate for
specific subsurface problems.
IS SOIL-GAS SAMPLING APPROPRIATE?
The first question is "What are the contaminants of interest at the investi-
gation site?" This question is fundamental in determining the applicability of
soil-gas sampling to a contamination problem. Once specific compounds are
53
-------
identified, their physical properties can be obtained from a variety of chemical
references. Generally, soil-gas sampling is most effective for compounds with
upper pressures above 1.0 kpa (5.0 mmHg) and Henry's Law constants above 0.1 to
0.5 kpa mVmol (Marrin, 1984). If vapor pressures and Henry's Law Constants are
not available, then compounds with boiling points below 125°C can probably be
detected in soil gas. This information is summarized for common ground-water
contaminants in Table 7.
The next question is "Do the contaminants of interest partition adequately
into the vapor phase?" Once the compounds of interest are identified and cate-
gorized, a determination can be made regarding the likelihood of their presence
in soil gas under a variety of environmental conditions. Henry's Law constants
are a measure of air-water partitioning and can be calculated from a compound's
vapor pressure, aqueous volubility and molecular weight according to Equation 1.
VM m
H = — (1)
S
where:
kpa • m
H = Henry's Law Constants -
mol
v = vapor pressure (kpa);
g
M = gram molecular weight — ; and
mol
rag 9
S =aqueous volubility — or — .
L m3
Vapor pressure and aqueous volubility must be calculated at the appropriate
environmental temperatures.
Groups C and F represent high molecular weight compounds which do not
partition adequately into the gas phase to be detected in soil gas under normal
circumstances. These compounds are of considerable environmental concern,
however, they are not amenable to soil-gas detection as described here. All
other contaminant groups contain compounds with a significant vapor phase. The
success in mapping compounds within these other groups is dependent on a number
of site-specific factors.
The next question is "Is the major subsurface contamination present in the
soil or the ground water?" The answer to this question determines whether
compounds must partition from the aqueous to the gas phase or whether they only
have to diffuse in soil gas. This distinction is particularly important for
compounds in Group G which have high vapor pressures but which are also very
water-soluble. The result is that these compounds diffuse quite readily once
in soil gas but tend to remain dissolved in the ground water. Hence, Group G
contaminants are amenable to soil-gas detection if they result from a surface
or vadose zone spill, but may not be present in soil gas as a result of moderate
ground-water contamination.
54
-------
TABLE 7. USEFUL DATA FOR SELECTED ORGANIC CONTAMINANTS
=======================================================================
Boiling Vapor Molecular Aqueous Henry's Law
point pressure weight solubility constant
Name Group °C kpa g/mol g/m3 kpa m3/mo1
Chloroform
CHC13
Carbon Tetra-
chloride
CC14
1,1,1-Trichloro-
ethane
TCA
1,1,2-Trichloro-
ethene
1,3-Dichloro-
propane
Chlorobenzene
Dichloro,
diphenyl ,
trichloroethane
DDT
Benzene
Toluene
Methane
n-Pentane
n-Octane
1,2,3-trimethyl-
benzene
Naphthalene
Benzopyrene
Acetone
A
A
A
A
B
B
C
D
D
D
D
D
E
E
F
G
61.7
76.6
74.1
87.0
112.0
132.0
185.0
80.1
110.63
-164.0
36.1
125.7
176.1
218.0
311.0
56.2
25.60
15.06
16.53
9.87
4.53
1.58
1.34 x 10-8
12.70
3.80
27,260
(25°C)
68.4
(25°C)
1.88
(25°C)
0.202
(25°C)
1.09 x 10-2
6.67 x 10-13
(25°C)
24.227
119.4
153.8
133.4
(25°C)
131.4
110.97
112.56
354.5
78.11
92.13
16.04
72.15
114.23
120.2
128.19
252.3
58.08
8,000
800
720
1,100
2,800
500
0.003
1,780
515
24
38.5
0.66
75
34
0.003
23 x 105
0.38
2.3
2.8
0.90
0.18
0.35
5.3 x 10-3
0.55
0.67
67.4
128
325
0.323
0.043
1.4 x 10-7
na
Source: Mackay and Shiu, 1981, except for acetone, from Lucius, 1987.
All values are for 20°C unless another value is noted.
na = not available
55
-------
It is also appropriate to ask "What is the depth to contaminated soil or
ground water?" This question is applicable to Groups B and E which have very
low aqueous solubilities and readily partition out of the ground water. Once
in the gas, however, these compounds tend to remain near the water table or the
original zone of soil contamination. Low vapor pressures and gas diffusion
coefficients make Group B and E compounds amenable to soil-gas analysis only
where probes can be placed near contaminated soil or ground water. Since soil-
gas probes are normally driven to a depth of 1 to 3 meters below ground surface,
there are obvious limitations to the remote detection of Group B and E compounds
using conventional soil-gas techniques.
Group D compounds are also affected by the depth to subsurface contamination
sources because of their tendency to be oxidized in the shallow soil. These
compounds have high Henry's Law constants (indicating favorable partitioning
out of the aqueous phase) and diffuse rapidly when introduced to the gas phase.
Thus, Group D compounds should migrate into the shallow soil gas in any environ-
ment which permits subsurface diffusion of volatile organic chemicals (VOCs).
The residence time of Group D compounds in shallow soil depends on subsurface
redox potentials and microbial activity. Low molecular weight petroleum hydro-
carbons are most predictably detected in shallow aquifers or from leaking
underground tanks where probes can be driven near the source of contamination.
To summarize, the answers to the first four questions indicate the applica-
bility of soil-gas sampling to broad contaminant groups under general site
characteristics. Polycyclic aromatic hydrocarbons, PCBs and organochlorine
pesticides (represented by Groups C and F) are rarely, if ever, detected by
soil-gas sampling. Volatile oxygenated compounds (Group G) are relatively
water-soluble, and therefore are most easily detected as soil, rather than as
ground-water, contaminants in close proximity to the source.
Compounds represented by Groups B and E are most often detected as a result
of shallow soil or ground-water contamination because they diffuse minimally and
tend to partition into aqueous or organic phases in the soil. Low molecular
weight petroleum hydrocarbons (Group D) can either be detected in soil gas
overlying shallow aquifers (where probes can be driven near the contamination
source) or deep aquifers (where probes can be driven below the oxidative zone
in soils). Ci and C2 halogenated hydrocarbons (Group A) are good candidates
for soil-gas detection under a wide range of environmental conditions. These
compounds possess low aqueous solubilities, high vapor pressures, high diffusion
coefficients and are relatively resistent to degradation processes in most
soils. This information is summarized in Table 8.
The next four questions address more specific aspects of the site. The
first question is "What are the approximate concentrations of subsurface con-
taminants?" The concentrations of VOCs in ground water combined with the depth
to water can be used to estimate chemical concentration gradients in soil gas.
Both contaminant flux rates and soil-gas/ground-water correlations are a func-
tion of chemical concentration gradients. Gradients are routinely measured in
the field by sampling soil gas in a vertical profile.
Contaminant flux rates are of interest because they provide an estimate of
the migration time between contaminated ground water and shallow soil gas.
56
-------
TABLE 8. CLASSES OF ORGANIC COMPOUNDS
Group
Group volatility Cl ass
A High Halogenated
methanes,
ethanes,
ethenes
B Moderate Halogenated
propanes,
propenes,
benzenes
c Very Halogenated
low polycyclic
aromatics
D High Ci-Cs
petroleum
hydrocarbons
E Low C9-C15
petroleum
hydrocarbons
F Very Polycyclic
low aromatic
hydrocarbons
G High Low molecular
weight,
oxygenated
•sst = = = = = — = = —
Examples
CHCIs, CCU,
TCA, PCE
DBCP,
chloro-
benzene
PCBs,
aldrin ,
DDT,
chlordane
Benzene,
toluene,
methane,
pentane,
isooctane,
JP-4,
gasoline
Diesel,
Jet A,
decane,
trimethyl-
benzene
Motor oil,
coal tar,
benzopyrene
Acetone,
tetra-
hydrofuran,
MEK*
Comments
Diffuse
rapidly.
Resist
degradation.
Resist
degradation.
Moderate
diffusion.
Not good for
soil-gas
analysis
Easily
oxidized.
Diffuse
rapidly.
Usually
oxidized.
Low diffusion
Not good for
soil-gas
analysis.
Diffuse
rapidly.
Soluble in
soil, water.
*MEK = methyl ethyl ketone.
For the electrical properties of these materials, see Kutrubes, 1986 and
Akadov, 1980.
57
-------
Soil-gas/ground-water correlations are usually determined by placing soil-gas
probes near existing monitoring wells. Linear regression analysis can then be
used to calculate a correlation coefficient for log-log plots of soil-gas vs.
ground-water concentrations. Anomalous concentrations are often indicative of
contaminant sources or small-scale geologic/hydrologic heterogeneities in the
vadose zone.
Another question is "What is the physical state of subsurface contaminants?"
Compounds may exist in the subsurface in several physical states including:
(1) dissolved in water, (2) dissolved in another organic phase, (3) adsorbed on
soil materials, or (4) present only in the gas phase. The physical state is
rarely determined directly but can be estimated on the basis of compound proper-
ties, disposal practices, and subsurface conditions. The physical state of a
compound determines the degree of phase partitioning, if any, which must occur
to permit gas-phase analysis of subsurface contaminants.
Low molecular weight hydrocarbons (Group D) have a density less than that
of water and will float as a thin film on the water table. The aromatic compo-
nents of Group D (e.g. benzene, toluene) are moderately water-soluble and there-
fore occur as dissolved as well as immiscible contaminants. Many compounds in
Group G also have densities less than 1.0 gram/mL, but are seldom encountered
as floating product due to their high aqueous volubility. All other contaminant
groups have densities greater than 1.0 gram/mL and will sink as immiscible
liquids in aquifers. Floating hydrocarbon products can act as a solvent for
high-density compounds (e.g. halogenated solvents) and retain them near the
surface of the water table. In that case, partitioning between gaseous and
organic phases becomes more important than the partitioning between gaseous and
aqueous phases. Volatilization of VOCs from organic solvents is determined by
the volatility of the solvent and the mole fraction of the solute (VOC) in the
solvent. For solvents-other than complex hydrocarbon products, the volatility
of VOCs from an organic phase can be estimated by Raoult's Law.
VOCs can be introduced directly to the soil-gas phase without
contaminating either soil grains or ground water. Underground utility lines
often introduce volatile compounds to soil gas from cracks and/or joints in
natural gas and sewer lines. Ci through C4 aliphatic hydrocarbons (Group D)
can be released from natural gas lines while a variety of solvent and fuel
vapors can diffuse from sewers carrying industrial wastes.
The next question is "What are the major hydrologic/geologic features
of the vadose zone?" This question is best answered by reviewing detailed
lithologic logs prepared by a qualified hydrologist or geologist during the
installation of monitoring wells or borings in the investigation area. Of
primary interest are clay lenses, perched water, buried foundations, and other
potential barriers to the vertical diffusion of gaseous contaminants. Subsur-
face diffusion barriers often result in soil-gas VOC concentrations which are
uncharacteristic of the underlying ground water. Chemical concentration gradi-
ents are locally disrupted by diffusion barriers because gaseous contaminants
are either absent or present at very low concentrations in soil gas overlying
the barrier.
58
-------
The presence of diffusion barriers does not preclude a soil gas investiga-
tion as long as the areal extent of barriers are minimal compared to that of
the subsurface plume. However, soil-gas sampling is ineffective in a situation
where a clean aquifer overlies a contaminated aquifer because contaminants are
unable to diffuse through the unconfined aquifer. Mapping of subsurface plumes
via soil-gas sampling is usually not affected by a few anomalous points due to
the large number of samples which are collected over an investigation site.
However, the location of potential diffusion barriers should be identified
prior to the interpretive phase of a soil-gas study.
In addition to gas diffusion barriers, the presence of soil moisture and
highly permeable zones (e.g. backfill or utility trenches) locally affect soil-
gas samples. High moisture levels reduce the air porosity of soils and inhibit
both soil-gas collection and gaseous diffusion. As the number of continuous
air-filled pores is reduced (due to increasing water saturation), the mass of
VOCs in soil gas also decrease. Representative soil-gas samples are rarely
obtained from soils with an air porosity below five percent. Conversely, back-
fill and gravels have high air porosities which often result in anomalously
high VOC concentrations relative to the underlying ground water. These coarse
materials are more permeable than the undisturbed soil and can provide a conduit
for laterally diffusing gaseous contaminants.
The site may be more complex than anticipated so it is worthwhile to ask,
"Are there surface or shallow vadose zone contaminant sources overlying the
major subsurface plume?" This question refers to localized contamination
sources other than the major ground-water plume. Unknown surface or shallow
soil contaminant sources can adversely affect the interpretation of regional
soil-gas data. Soil-gas probes can intercept laterally diffusing VOCs from a
surface source as well as vertically diffusing VOCs from ground water. Thus,
contaminant concentrations in" shallow soil gas can be anomalously high relative
to concentrations in the underlying water. A radial distance equal to three
times the depth to water has been empirically determined at several sites to be
the extent of lateral contaminant diffusion surrounding a source. Soil-gas
samples collected beyond this distance are normally representative of the
underlying ground water.
If surface sources are not identified prior to a soil-gas investigation,
they can be located using several techniques. Contaminant sources within the
boundaries of a regional plume are indicated by (1) an abrupt increase in soil-
gas contamination compared to surrounding points, (2) a change in chemical
composition of the soil gas, and (3) a significant deviation from the soil-gas/
ground-water concentration ratio calculated for an overall site. The presence
of soil contamination can be confirmed by analyzing a vertical profile of soil
gas. Vertical soil-gas profiles completed near a surface spill typically show
increasing VOC concentrations down to the depth of maximum soil contamination
and then decreasing concentrations toward the water table. Contaminated ground
water results in increasing VOC concentrations with depth from the ground
surface to the water table.
To summarize, the last four questions identify site-specific factors which
determine the subsurface behavior of VOCs. In contrast to the first four ques-
tions, these questions require a detailed understanding of the investigation
59
-------
site which may not exist until after a remedial investigation has been com-
pleted. Both the planning and interpretation of soil-gas studies are enhanced
by the number of the site-specific questions which can be answered initially.
There are no absolute limits for VOC concentrations in soil or ground water
below which soil-gas sampling is ineffective. The lower limit is a function of
(1) compound properties, (2) analytical detection limits, (3) depth and physical
states of the compound, (4) soil properties, and (5) interference from degrada-
tion processes or secondary contaminant sources. Soil-gas investigations are
also commonly performed as screening or initial assessment procedures where few
of the site-specific questions can be answered. Soil-gas sampling can be
conducted with a minimal amount of background information, however, the results
will generally be more difficult to interpret.
DEVELOPING AN INVESTIGATION PLAN
Two additional questions need to be answered when planning an investiga-
tion. The first is "What is the major objective of the soil-gas investigation?"
Formulating an investigation plan which specifies the spacing and siting of
soil-gas probes requires a clear understanding of the objectives. Delineation
of plume edges is most efficiently achieved by establishing a transect parallel
to the hydraulic gradient and sampling outward from the suspected source. Once
an initial boundary point is identified, subsequent sampling locations are
selected on the basis of real-time results. By contrast, locating downgradient
contaminant sources is best achieved by sampling soil gas on a pre-determined
grid covering the investigation site. Locating primary source areas is accom-
plished by either grid or real-time sampling, depending on the initial informa-
tion which is available.
The distance between sampling points is a function of the plume resolution
required. Soil-gas samples are commonly collected on 303- to 606-meter (1,000-
to 2,000-feet) intervals over large geographic areas where the objective is to
identify potentially contaminated regions. Such widely spaced probes cannot
provide resolution of individual plume characteristics. Plume definition is
accomplished by sampling probes on more closely spaced centers, depending on
the specific site.
Soil-gas samples generally should not be collected less than approximately
15 meters (50 feet) apart where high resolution mapping is required. Differences
in VOC concentrations between closely-spaced points are as likely to result
from small-scale heterogeneities in the shallow soil as from significant changes
in the parameter of interest (e.g. contaminant levels in the underlying ground
water). Locations of soil-gas samples are also determined by the access to
sampling areas and the ability to successfully drive probes into the underlying
soil. Generally, the minimum spacing of soil-gas probes is proportional to the
depth to ground water.
The other question in this section is "What are the general topography and
surface conditions at the investigation site?" This final question is designed
to provide additional information for the selection of sampling locations.
Topography should be considered if variations in land elevation result in sig-
nificant differences in the depth to water over an investigation site. The
60
-------
thickness of the vadose zone overlying contaminated ground water affects
chemical concentration gradients and thus the comparison of VOC concentrations
in soil gas. Topography is a more critical factor for shallow aquifers (less
than 6 meters below the ground surface) than for deeper ground water. When
sampling above contaminated ground water less than 6 meters deep, probes are
often driven to a constant height above the water table rather than to a
constant depth below the ground surface.
Surface conditions also influence the location of soil-gas samples over an
investigation site. Extremely wet surface conditions caused by ponded water
should be avoided because of problems associated with low air porosities in
soil. Recently disturbed soils (e.g. plowed or graded) often do not yield
representative soil-gas results due to the dilution and mixing of soil gas with
atmospheric air. If probes can be driven several feet below the disturbed
soil, VOC concentrations in soil gas are usually representative of subsurface
contamination.
The presence of man-made pavements covering soil may also affect the results
of soil-gas sampling. Asphalt or concrete surfaces can act as a barrier to the
gaseous diffusion of VOCs and alter the chemical concentration gradient in
shallow soil. Two adjacent probes sampled under exposed soil and pavement,
respectively, can yield quite different soil-gas concentrations. Generally,
VOC concentrations sampled at the same depth are higher under pavement than
under bare soil. This difference can affect data interpretation if soil-gas
samples are collected under both surfaces at the same site. Pavement materials
vary widely in their ability to restrict the diffusion of VOCs.
To summarize, applicability of soil-gas sampling to a specific site should
be determined by answering the first eight questions. The answers to the last
two questions are designed to provide information which can be used to optimally
design a soil-gas investigation. Information required for the investigation
plan can be provided by on-site observations and a clear objective for the soil-
gas study. In addition, cost and time constraints usually affect the final
investigation plan.
61
-------
SECTION 7
PLANNING A GEOPHYSICAL INVESTIGATION
INTRODUCTION
The purpose of this section is to provide general guidance in the choice
of geophysical techniques to detect organic contamination. Most geophysical
techniques are useful in site characterization activities and have been
developed for that purpose. Some techniques have capabilities which are useful
for the detection of inorganic, while other techniques have capabilities which
are useful in detecting organic contaminants. Table 9 reviews the characteris-
tics of seven geophysical techniques, highlighting features of each method. It
is intended to provide a capsule summary of the primary technical characteris-
tics of each method, including parameter measured, mode of measurement, depth
of penetration, resolution, and raw data format.
Table 10 summarizes common applications of the techniques. The categories
listed are general: site characterization refers to detecting layers, depths
of soil and rock, and depths to the water table; conductive leachate plumes
refers to detecting the vertical and horizontal extent of inorganic leachates;
metal objects refers to detecting objects such as drums, trash, pipes, and
cables; and organic contamination refers to detecting the vertical and horizon-
tal extent of organics floating on the water table, or present in massive
quantities in the soil. This table presents generalizations which are appli-
cable in most cases. However, exceptions exist because of the wide range of
site conditions and project objectives.
It is important to realize that techniques will be useful at some types of
sites, but not at others. This is chiefly due to the geologic conditions
present, but also may be due to instrument capabilities or interferences which
affect the performance of the techniques. For example, GPR is not effective
in clay soils. All electrical methods are affected by nearby metal objects.
D.C. resistivity performs best in resistive soil. Magnetics will not detect
copper, stainless steel , or aluminum, but is excellent for ferrous iron.
Seismic methods are affected by wind and airport noise, truck and train traffic,
and working drill rigs. Seismic methods do not perform well in unconsolidated
soil. In addition, interpretation of the data from these techniques is an
important part of the process. Topographic corrections may be needed to present
the data at a standard distance from the water table. By using mathematical
models and supplemental field measurements, it may be feasible to remove the
effects of buildings or utilities from the data. However, in complicated
geological situations, this may not be possible.
62
-------
TABLE 9. CHARACTERISTICS OF THE SEVEN GEOPHYSICAL METHODS
(MODIFIED FROM BENSON ET. AL., 1983)
-------
TABLE 10.
GENERALIZED APPLICATIONS OF GEOPHYSICAL TECHNIQUES
Application
Technique
Site Charac-
terization
Conductive
Leachate
Metal
Objects
Organic
Contamination
yes
yes
yes
yes
yes
yes
yes
possibly
yes
no
yes
no
no
yes
no
no
no
no
yes
no
no
yes
yes
possibly
yes
no
no
no
Ground Penetrating
Radar (GPR)
Electromagnetic
(EM)
D.C. Resistivity
Complex Resistivity
Seismic Refraction
Metal Detector
Magnetometer
*ln some cases, the organic contamination will be associated with inorganic
contamination. Examples include organics in metal drums and mixed organic-
inorganic leachate plumes.
Note that only two techniques are recommended for routine use in detecting
organic contamination. The successful application of these techniques, GPR and
complex resistivity, is discussed in more detail below. The d.c. resistivity
and EM techniques may be useful at a site for detection of hydrocarbons, but
other techniques with greater likelihood of success should be considered first.
GEOPHYSICAL TECHNIQUES AND ISSUES
When selecting geophysical techniques for a site investigation, many ques-
tions should be addressed. The following list of questions can help to better
define the important decisions to be made for a given site investigation. Many
of the questions have answers which apply to more than one technique. Thus ,
the questions have been organized into a group of questions which is general in
nature, and additional groups of questions which are specific to certain tech-
niques. The questions listed have been selected from an EPA-U.S. Geological
Survey computerized expert system now in development.
General
Some of the questions in this category are asked in a different manner in
the discussion of preliminary information provided in Section 5. Because of
their importance and for logical continuity, they are also included here.
64
-------
o Was the source of contamination a single event, a continuous leak which
has been repaired, or a continuing leak?
o Did the contaminants originate from a surface spill, deep injection, a
leaking trench, a leaking landfill, a leaking underground storage tank,
a leaking underground pipeline, a land treatment facility, a surface
impoundment, or are the contaminants different, like an intact, lost
barrel of waste?
o Where are the contaminants now? They may be present on the surface, in
the unsaturated zone, in the saturated zone, or in all these areas.
o Is this an areal search, a depth search, or both? This will determine
whether profiles or soundings are performed.
o What types of contaminants are present? Are there inorganic contami-
nants present which may serve as indicators of the presence of organic
contamination? If so, geophysical methods for locating conductive
plumes may be useful.
o Are there natural organics such as from a farm, forest, or swamp
present?
o Are the soils at the site preferentially water-wet or organic wet? A
soil is not wet by a liquid if the liquid forms beads on the surface of
the soil.
o Are the organics mostly in the water phase, adsorbed on soil solids, or
in the gas phase?
o Do the organics and inorganic react in any way? Are the organics being
modified by degradation, catalysis, or adsorption?
o Are volatile organics present at the surface? If so, soil-gas tech-
niques should be considered.
o What is the environment at the site? Examples are rural, suburban,
urban, industrial, landfill, military base, service station.
o How much of the site is covered by buildings? What type of access is
possible? Is it difficult to walk around the site, or is it possible
to drive over most of the site with a vehicle such as a van? Is any of
the site inaccessible due to property ownership, security reasons,
safety hazards, or difficulties such as swampy conditions?
Resistivity, EM, GPR, and Magnetic Techniques
o Are there any metallic objects on or near the site? Metallic objects
such as fences, pipelines, and electrical or telephone wires above or
below ground may interfere with EM, d.c. or complex resistivity, GPR,
and magnetometry. These type surveys may not be possible depending on
the amount of the site surface which is covered.
65
-------
below ground may interfere with EM, d.c. or complex resistivity, GPR,
and magnetometry. These type surveys may not be possible depending on
the amount of the site surface which is covered.
o Are metallic well casings installed at the site? Casings may also affect
EM, d.c. or complex resistivity, GPR and magnetometry measurements.
o Are pipelines catholically protected? If possible, the cathodic protec-
tion should be turned off during surveys. Cathodic protection does not
affect GPR.
o How much of the area is covered by concrete or asphalt? D.C. and
complex resistivity techniques require contact with the ground. Rebar
present in the concrete may interfere with EM and magnetic measurements.
o What is the range in topographic relief across the site? If it is
greater than 1 meter, then electromagnetic, resistivity, and ground-
penetrating radar data should have a topographic correction applied.
o Are radio, television, or radar facilities nearby? Measurements may be
affected.
GPR and Complex Resistivity
o Is clay present at the site? How much clay is present? Is it present
as layers, lenses, evenly mixed with other soil components, or massive?
GPR cannot penetrate clay. However, if the clay is present as lenses,
the GPR may be useful between the lenses. The techniques of GPR and
complex resistivity are complementary in that complex resistivity
requires the presence of clay to be successful.
o Is the zone of relevance above the clay? If so, then GPR may be
feasible.
o What are the properties of the contaminants? Are they soluble or
insoluble in water? Are they miscible, immiscible, or a mixture of
both? What is the density of the contaminants? Are they nonpolar,
anionic, cationic or a mixture of these? GPR locates organics that
phase-separate, i.e., are immiscible or insoluble, and float.
o What is the average electrical resistivity of the site in ohm meters?
If the resistivity is greater than 30 ohm meters, and clay is not
present, then the site is a candidate for GPR.
Surface geophysical techniques have the ability to provide useful informa-
tion at hazardous waste sites. Maps of electrical conductivity variation from
EM measurements or resistivity soundings can provide three-dimensional boundary
locations for hydrogeological and cultural features as well as direct detection
of inorganic contaminants. Direct detection of organics using these techniques
may be possible in some cases. Complex resistivity and GPR require more time
and expense, but can provide more detailed hydrogeological information and some-
times direct detection of organic contamination. GPR has the highest resolution
66
-------
of any geophysical technique, allowing it tdlook" through the gaps in urban
and high density utility environments. The data may require modelling to
remove the effects of buildings. When most of the precipitation is seasonal,
GPR data quality can be improved by performing the measurements during the
driest time of the year or during the time when soils are frozen. GPR signals
cannot penetrate some types of asphalt or closely spaced rebar or chicken wire.
GPR is most useful at sites with no clay, on problems with water-insoluble
organics above or floating on the water table. Complex resistivity is most
useful at sites containing clay, and on problems with water-soluble organics
below the water table. It also may provide an effective noninvasive monitor of
the performance of clay barriers around waste sites.
67
-------
REFERENCES
Akhadov, Y. 1980. Dielectric properties of binary solutions, Pergamon,
New York. 475 pp.
Baker, A. T., D. J., LaBrecque, D. L, Marrin, and A. T. Mazzella. February
1987. Geophysical and Soil-Gas Investigations: Phelps Collins Air
National Guard Base, Alpena, Michigan, TS-AMD-85580, prepared by Lockheed
Engineering and Management Services Co., Inc. for U.S. EPA, EMSL-LV,
Las Vegas, Nevada.
Baizer, M. M., and H. Lund, eds. 1983. Organic electrochemistry, 2nd cd.,
Marcell Dekker, New York. 1166 pp.
Beers, R., and R. Morey. 1981. Subsurface radar profiling field tests at
low-level nuclear waste sites: Maxey Flats, Kentucky and Beatty, Nevada,
NUREG-CR-1272, GC-TR-79-1023, U.S. Nuclear Regulatory Commission. 44 pp.
Benson, R. C., R. A. Glaccum, and M. R. Noel. 1983. "Geophysical Techniques
for Sensing Buried Wastes and Waste Migration," National Water Well
Association.
Bradley, E. 1980. Trichloroethylene in the ground water supply of Pease Air
Force Base, Portsmouth, New Hampshire, OF-80-557, U.S. Geological Survey
Water Resources Investigation. 22 pp.
Cherry, J. A., ed. 1983. "Migration of contaminants in ground water at a land-
fill: a case study," Journal of Hydrology. Vol. 63. pp. 1-197.
Cherry, J. A., R. N. Farzolden, and E. O. Frind. 1983. "Migration of Contami-
nation in a Landfill: A Case Study 3 - Tritium as an Indicator of Disper-
sion and Recharge." Journal of Hydrol., Vol 63, pp 51-80.
Davison, C. C., W. S. Keys, and F. L. Paillet. 1982. Use of borehole-
geophysical logs and hydrogeologic tests to characterize crystalline rock
for nuclear waste storage, Whitshell Nuclear Laboratory, Ontario, Canada:
Battelle ONWIColumbus, Ohio. ONWI-418. 103 pp.
Devitt, D. A., R. B. Evans, W. A. Jury, T. H. Starks, B. Eklund, and A. Gholson.
June 1987. Soil-Gas Sensing for Detection and Mapping of Volatile Organics.
U.S. EPA, EMSL-LV, Las Vegas, Nevada.
Ehrlich, G. G., D. F. Goerlitz, E. M. Godsy, and M. F. Hult. 1982. "Degrada-
tion of phenolic contaminants in ground water by anaerobic bacteria,
St. Louis Park, Minnesota," Ground Water, Vol. 20. PP. 703-710.
68
-------
Eklund, B. May 1985. Detection of Hydrocarbons in Groundwater by Analysis of
Shallow Soil Gas/Vapor, API Publication 4394.
Franklin, A. G., D. M. Patrick, D. K. Butler, W. E. Strohm, Jr., and M. E. Hynes-
in karst terrains and other areas susceptible to collapse, NUREG/CR-206,
Us. Nuclear Regulatory Commission. 229 pp.
Freeze, A., and J. A. Cherry. 1979. Groundwater, Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey, pp. 136-137.
Germeroth, R. M., and H. Schmerl. February 1987. "Jet Fuel from the Ground
up." Civil Enginee ring. pp. 64-66.
Greenhouse, J., and R. Harris. 1983. "Migration of Contaminants in Ground
Water at a Landfill: a Case Study, DC, VLF, and Inductive Resistivity
Surveys." J . H y d rol. Vol. 3. pp. 177-197.
Handman, E. H. 1983. Hydrologic and geologic aspects of waste management
and disposal: a bibliography of publications by U.S. Geological Survey
authors, Circular 907. Geol Surv. pp. 1950-81.
Hult, M. F., ed. 1984. Ground-water contamination by crude oil at the Bemidji,
Minnesota research site: U.S. Geological Survey Toxic Waste-Ground Water
Contamination Study, 84-4188. U.S. Geol. Surv. Water Resources Investiga-
tions Report. 107 pp.
Jones, J. W., E. S. Simp§onP,. Neuman, and W. S. Keys. 1985. Field and
theoretical investigation of fractured crystalline rock near Oracle,
Arizona, NUREG/CR 3736. U.S. Nuclear Regulatory Commission Report.
104 pp.
Keller, G. V., and F. C. Frischknecht. 1966. Electrical Methods in Geophysical
Prospecting, Pergamon Press, New York. p. 517.
Kerfoot, H. B., J. A. Kohout, and E. N. Amick. March 4-6, 1986a. "Detection
and Measurement of Ground-Water Contamination by Soil-Gas Analysis," In:
Proceedings of 3rd National Hazardous Wastes and Hazardous Materials
Conference, Atlanta, Georgia.
Kerfoot, H., and C. Mayer. 1986b. "The Use of Industrial Hygiene Samplers for
Soil-Gas Surveying," Ground Water Monitoring Review. Vol. 6, no. 4.
pp. 74-78.
Kutrubes, D. 1986. Dielectric permittivity measurements of soils saturated
with hazardous fluids. MSC Thesis. Dept. of Geophysics, Colorado School
of Mines, Golden, Colorado.
Lappala, E., and G. Thompson. 1984. "Detection of Ground-Water Contamination
by Shallow Soil Gas Sampling in the Vadose Zone: Theory and Applications."
In: Proceedings of the 5th National Conference on Management of Uncon-
trolled Hazardous Waste Sites, Hazardous Materials Control Research
Institute, Silver Springs, Maryland, pp. 20-28.
69
-------
LeBlanc, D. R. 1984.Sewage plume in a sand and gravel aquifer, Cape Cod,
Massachusetts. U.S. Geological Survey Water Supply Paper 2218. 28 pp.
Lucius, J. E. August, 1987. Physical and Chemical Properties and Health
Effects of Thirty Three Toxic Organic Chemicals. U.S. Geological Survey
Open File report, 87-428.
Mabey, D. R., 1974. "Magnetic Methods", in Techniques of Water Resource Inves-
tigations of the U.S. Geological Survey. Chapter D1. Applications of
Surface Geophysics to Ground-Water Investigations. U.S. Government
Printing Office, Washington, D.C. pp. 107-115.
Mackay, D., and W. Y. Shiu. 1981. "A Critical Review of Henry's Law Constants
for Chemicals of Environmental Interest," J. Phys. Chem. Ref. Data. Vol.
10, No. 4. pp. 1175-1199.
Mackay, D. M., P. V. Roberts, and J. A. Cherry. 1985. "Transport of organic
contaminants in groundwater," Environ. Sci. Tech. vol. 19. pp. 384-392.
Marrin, D. L. November 5-7, 1984a. "Remote Detection of Volatile Organic
Contaminants in Ground Water via Shallow Soil-Gas Sampling." In:
Proceedings of Petroleum Hydrocarbons and Organic Chemicals in Ground
Water, Houston, Texas. National Water Well Association, Dublin, Ohio.
pp. 172-187.
Marrin, D. L. 1984b. Investigations of volatile contaminants in the unsatura-
ted zone above TCE polluted ground water. Final report to U.S. EPA. Ada,
Oklahoma,
Marrin, D. 1985a. "Delineation of Gasoline Hydrocarbons in Ground Water by
Soil Gas Analysis." In: Proceedings of the Hazardous Materials Manage-
ment Conference West '85. Tower Conference Management Company, Wheaton,
Illinois, pp. 112-119.
Marrin, D. L. December 3-5, 1985b. "Delineation of Gasoline Hydrocarbons in
Ground Water by Soil-Gas Analysi^roc^eidings of the 1985 Hazardous
Materials West Conference, Long Beach, California.
Marrin, D., and G. Thompson. Jan-Feb 1987. "Gaseous Behavior of TCE Overlying
a Contaminated Aquifer," Ground-Water. Vol 25:1. pp. 21-27.
Mattraw Jr., H. C., and B. J. Franks, eds. 1984. Movement and fate of creosote
waste in ground water, Pensacola, Florida: U.S. Geological Survey Toxic
Waste—Ground Water Contamination Program. Open File Report 84-466.
U.S. Geol. Surv. 93pp.
Mazzella, A. T., A. M. Pitchford, and K. R. Scarborough. November 1987.
Geophysical and Soil-Gas Investigations at Robins Air Force Base, Georgia.
U.S. EPA, EMSL-LV, Las Vegas, Nevada.
McNeill, J. D. 1980. Electromagnetic Terrain Conductive Measurement at Low
Induction Numbers. Technical Note TN-6. Geonics, Ltd.
70
-------
Miller, R. D., S. E. Pullan, J. S. Waldner, and F. P. Haeni. 1986. "Field
Comparison of Shallow Seismic Sources," Geophysics. Vol. 51. pp. 2067-
2092.
Nichols, W. D. 1986. Geohydrology of unsaturated zone at the burial site
for low level radioactive waste near Beatty, Nye County, Nevada. U.S.
Geological Survey Open File Report 85-198. 85 pp.
Olhoeft, G. R. 1979. "Nonlinear electrical properties," in Nonlinear Behavior
of molecules, atoms and ions in electric, magnetic, or electromagnetic
fields. L. Neel, cd., Elsevier, New York. pp. 395-410.
Olhoeft, G. R. 1984a. "Applications and Limitations of Ground-Penetrating
Radar." I nExpanded Abstracts,54th Annual International Meeting and
Exposition of the Society of Exploration Geophysicists, Atlanta, Georgia.
pp. 147-148.
Olhoeft, G. R. 1 984QJay-Organic Reactions Measured with Complex Resis-
tivity." In: Expanded Abstracts, 54th Annual International Meeting and
Exposition of the Society of Exploration Geophysicists, Atlanta, Georgia.
pp. 356-358.
Olhoeft, G. R. 1985. "Low-frequency electrical properties," Geophysics.
vol. 50. pp. 2492-2503.
Olhoeft, G. R. November 12-14, 1986. "Direct Detection of Hydrocarbon and
Organic Chemicals with Ground Penetrating Radar and Complex Resistivity."
In: Proceedings of the National Water Well Association Conference on
Hydrocarbons and Organic Chemicals in Ground Water, Houston, pp. 284-305.
Olsson, O., 0. Duran, A. Jamtlid, and L. Stenburg. 1984. "Geophysical investi-
gations in Sweden for the characterization of a site for radioactive waste
disposal-an overview," Geoexploration. Vol. 22. pp. 187-201.
Payne, B. R. 1972. "Isotope Hydrology," Adv. Hydrosci., 8, pp. 95-138.
Pitchford, A. M., A. T. Mazzella, and K. R. Scarborough. November 1987.
Geophysical and Soil-Gas Investigations at Holloman Air Force Base,
New Mexico. TS-AMD-87580. U.S. EPA, EMSL-LV, Las Vegas, Nevada.
Pitchford, A. M., and K. R. Scarborough. November 1987. Soil-Gas Investigations
Tinker Air Force Base, Oklahoma. U.S. EPA, EMSL-LV, Las Vegas, Nevada.
Quinlan, J. F., and R. O. Ewers. 1985. "Ground-Water Flow in Limestone
Terranes; Strategy, Rationale, and Procedures for Reliable, Efficient
Monitoring of Ground Water in Karst Areas." IN: National Symposium and
Exposition on Aquifer Restoration and Ground-Water Monitoring. Proceed-
ings, National Water Well Association, Worthington, Ohio, pp 197-234.
Romig, P. R., ed. 1986. "Special issue-Engineering and Ground Water,"
Geophysics. Vol. 51. pp. 221-323.
71
-------
Saunders, W. R., R. W. Castle, and C. R. Foget. 1983. "Delineation of Subsur-
face Petroleum Spills Using Terrain Conductivity Measurements." In:
Proceedings of the 1983 Oil Spill Conf.Am. Pet. Inst./U.S. EPA/U.S. Coast
Guard, San Antonio, Texas, pp. 415-417.
Saunders, W. R. and R. M. Germeroth. 1985. "Electromagnetic Measurements for
Subsurface Hydrocarbon Investigations," In: Proceedings of the petroleum
Hydrocarbon and Organic Chemicals in Ground Water: Prevention, Detection,
and Restoration Conference, National Water Well Association.
Saunders, W. R., and S. A. Cox. May, 1987. "Use of an Electromagnetic Induc-
tion Technique in Subsurface Hydrocarbon Investigations." IN: Proceedings
of First National Outdoor Action Conference on Aquifer Restoration, Ground-
Water Monitoring, and Geophysical Methods, National Water Well Association.
pp 585-599.
Seitz, H. R., A. T. Wallace, and R. E. Williams. 1971. "The effect of a land-
fill on a hydrogeologic environment." In: Engineering geology and soils
engineering symposium, 9th proc., Idaho Dept. Hwys., Boise, Idaho.
pp. 185-208.
Sumner, J. 1976. Principles of induced polarization for geophysical explora-
tion, Elsevier, New York. 7 pp.
Taylor, R. W. 1984. Evaluation of geophysical surface methods for measuring
" hydrological variables in fractured rock units, OFR-17-84, NTIS PB84-
158021. U.S. Bureau of Mines. 145 pp.
Ulriksen, C. P. 1982. Application of impulse radar to civil engineering.
Ph.D. Thesis. Dept. of Civil Engineering, Lund Univ. of Tech, Finland.
Published in USA by Geophysical Survey Systems Inc., Nashua, New Hampshire.
Valentine. R. M.. and T. Kwader. February 12-14, 1985. "Terrain Conductivity
as a-Tool for Delineating Hydrocarbon Plumes in a Shallow Aquifer - A Case
Study." IN: Proceedings of Surface and Borehole Geophysical Methods in
Ground Water Investigations, pp. 52-63.
Verschueren, K. 1983. Handbook of Environmental Data on Organic Chemicals.
Van Nostrand Reinhold Co., New York. 1310 pp.
Voorhees, K. J. 1984. Application of a New Technique for the Detection and
Analysis of Small Quantities of Contaminants in the Soil. Petrex Sales
Literature, Denver, Colorado. (Unpublished manuscript. )
Walther, E. G., A. M. Pitchford, and G. R. Olhoeft. November 18-20, 1986.
"Strategy for Detecting Subsurface Organic Contaminants." IN: Proceedings
of National Water Well Association/American Petroleum Institute Conference
on Petroleum Hydrocarbons and Organic Chemicals in Ground Water.
Weast, R. C., ed. 1969-1970. Handbook of Chemistry and Physics. The Chemical
Rubber Company, Cleveland, Ohio. 50th Edition.
72
-------
Wright, D., G. Olhoeft, and R. Watts. 1984. "Ground-Penetrating Radar Studies
in Cape Cod." Surface and Borehole Geophysical Methods in Ground Water
Investigations. D. M. Neilson, cd., Worthington, Ohio, Nat. Water Well
Assoc. pp. 666-680.
Zehner, H. H. 1983. Hydrogeologic Investigation of Maxey Flats Radioactive
Waste Burial Site, Fleming, Kentucky. U.S. Geological Survey Open File
Report 83-133. 149 pp.
Zohdy, A. 1974. "Electrical Methods." Techniques of Water Resource Investiga-
tions of the U.S. Geological Survey, Applications of Surface Geophysics to
Ground-Water Investigations. Chapter D1. U.S. Government Printing Office,
Washington, D.C. pp. 5-66.
73
------- |