PB82-103896
USE OF REMOTE SENSING TECHNIQUES IN A SYSTEMATIC
INVESTIGATION OF AN UNCONTROLLED HAZARDOUS WASTE SITE
N. L. Cichowicz, et al
The MITRE Corporation
Bedford, Massachusetts
September 1981
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
KITS
-------
EPA-600/2-81- 187
September 1981
PB52-10
USE OF REMOTE SENSING TECHNIQUES IN A SYSTEMATIC
INVESTIGATION OF AN UNCONTROLLED HAZARDOUS WASTE SITE
by
Nancy L. Cichowicz
Robert W. Pease, Jr.
Paul J. Stoller
Harold J. Yaffe
The MITRE Corporation
Metrek Division
Bedford, Massachusetts 01730
Contract No. 68-01-5051
Project Officer
Stephen C. James
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
NOTICE
THIS DOCUMENT HAS BEEN REPRODUCED
FROM THE BEST COPY FURNISHED US BY
THE SPONSORING AGENCY. ALTHOUGH IT
IS RECOGNIZED THAT CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
IN THE INTEREST OF MAKING AVAILABLE
AS MUCH INFORMATION AS POSSIBLE.
-------
TECHNICAL REPORT DATA
(Pleaze read I,iwvcrzons on the revene before corn plenng)
1 REPORT NO. 2.
EPA—600/2—81— 187 ORD Repor.t
3. RECIPIENTS ACCESSIOf’NO.
PB82 10389 6
4. TITLE AND SUBTITLE
Use of Remote Sensing Techniques in a Systematic
.
Investigation of an Uncontrolled Hazardous Waste Site
5. REPORT DATE
September 1981
8. PERFORMING ORGANIZATION CODE
7 AuT). OR(S)
Nancy L. Cichowicz, Robert W. Pease,Jr., Paul J.
Stoller, Harold J. Yaffe
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
BRD1A
11. CONTRACT/GRANT NO.
68—01—5051 -
9. PERFORMING ORGANIZATION NAME AND ADDRESS
The Mitre Corp.
Metrek Division
Bedford, Mass. 01730
12. SPONSORING AGENCY NAME AND ADDRESS 13. TYPE OF REPORT AND PERIOO COVERED
Municipal Environmental Research Laboratory — Cm., OH Final
Office of Research and Development 14.SPONSORINGAGENCYCOOE
U.S. Environmental Protection Agency EPAI600I14
Cincinnati, Ohio 45268
16. SUPPLEMENTARY NOTES - - - -
Project Officer: Stephen C. James (513) 684—7871 -
See also EPA—600/2-81- iR
16. A8S RACT
This report describes the use and evaluation of several remote sensing techniques
in conjunction with direct sample collection in order to develop a systematic approach
for subsurface investigations at uncontrolled hazardous waste sites. Remote sensing
techniques (electrical resistivity, seismic refraction, ground—penetrating radar, and
metal detection) were employed to determine the extent (and sequence) to which they
may be integrated with the more conventional methods of test drilling, installation of
monitoring wells, and excavation for determining information such as the following:
—— nature and extent of ground water contamination
presence and number of buried drums
— topography and condition of bedrock
—— costs and effectiveness of several abatement methods.
Both the remote sensing and conventional sampling methods were used at an abandoned
hazardous waste dump in Coventry, Rhode Island.
T7. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IOENTIP!SRS/OPEN ENDED TERMS
C. COSATI Field/Group
Remote sensing
Uncontrolled hazardous waste site
Electrical resistivity
Seismic refraction
Ground—penetrating radar
Metal detection
l3B
21. —
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Reporrj
Unclassified
20. SECURITY CLASS (This page;
Unclassified
22. PRICE
EPA Førm 2220-I (9-13)
I
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U. S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solu-
tion and it involves defining the problem, reassuring its impact, and search-
ing for solutions. The Municipal Environmental Research Laboratory develops
new and improved technology and systems for the prevention, treatment, and
management of wastewater and solid and hazardous waste pollutant discharges
from municipal and community sources, for the preservation and treatment of
public drinking water supplies, and to minimize the adverse economic, social,
health, and aesthetic effects of pollution. This publication is one of the
products of that research; a most vital communications link between the
researcher and the user community.
This report describes the use and evaluation of several remote sensing
techniques in conjunction with direct sample collection in order to develop
a systematic approach for subsurface investigations at uncontrolled hazardous
waste sites.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
iii
-------
ABSTRACT
This report describes the use and evaluation of several remote sensing
techniques in conjunction with direct sample collection in order to develop a
systematic approach for subsurface investigations at uncontrolled hazardous
waste sites. Remote sensing techniques (electrical resistivity, seismic re-
fraction, ground—penetrating radar, and metal detection) were employed to de—
teruu.ne the extent (and sequence) to which they may be integrated with the more
conventional methods of test drilling, installation of monitoring wells, and
excavation for determining information such as the following
• nature and extent of ground water contamination
• presence and number of buried drums
• topography and condition of bedrock
o costs and effectiveness of several abatement methods.
Both the remote sensing and conventional sampling methods were used at an aban-
doned hazardous waste dump in Coventry, Rhode Island.
iv
-------
ACKNOWLEDGEMENTS
The project team is appreciative of the support given by the following
MITRE personnel toward the completion of this investigation: Lynne S. Arden,
Donna T. Howarth, and Milton V. Wilson for report preparation and coordination;
Irwin Frankel, Ronald N. Hoffer, and John L. Menke for their critical review;
Joan S. Garber and Marilyn L. Pyne for assistance in project management; and
Kern E. Sails and Barbara J. Trinklein for support in field activities.
The assistance of Stephen C. James, Project Officer, and Donald E. Sanning
of the U.S. EPA Solid and Hazardous Waste Research Division is greatly appre-
ciated as is the support given by personnel of the project subcontractors:
• Caputo and Wick, Ltd. (surveying and map preparation)
Rumford, Rhode Island
• Energy Resources Co., Inc. (chemical analysis)
Cambridge, Massachusetts
• Fred C. Hart and Associates, Inc. (electrical resistivity and metal
New York City, New York detection surveys)
a Geophysical Survey Systems, Inc. (ground—penetrating radar survey)
Hudson, New Hampshire
• Geotechnical Engineers Inc. (consultants for bedrock coring and
Winchester, Massachusetts sampling)
• Guild Drilling Co. (well installation and rock coring)
East Providence, Rhode Island
• Stephen A. Alsup and Associates, Inc. (seismic refraction survey,
Newton, Massachusetts vertical electrical resistivity
survey)
V
-------
1
1
• 2
• 3
• 6
• 9
• 9
• 10
• 12
12
19
• 19
A. Summary of Conclusions, Recommended Actions, and Comparison
of Abatement Alternatives: Phase I 61
B. Summary of Evaluation of Long—Term Abatement Options: Phase II 67
Page
• iv
• V
vii
ix
CONTENTS
Abstract
Acknowledgements
Figures
Tables
1. Introduction
Purpose and Scope
Summary of the Remote Sensing Methods
History of the Site
Site Investigation: Phase I and Phase II.
2. Remote Sensing Techniques
Electrical Resistivity
Seismic Refraction
Metal Detection
Ground—Penetrating Radar
3. Results of Field Studies
Plume Delineation
Determination of Bedrock Topography and
Depth of Buried Drums
Determination of Trench Location and Geometry.
4. Evaluation of the Remote Sensing Techniques
Detection of Subsurface Contamination
Elucidation of Bedrock Topography and Condition.
Determination of Subsurface Trench Limits. .
Detection of Buried Drums
5. Recommendations
Systematic Approach for Abandoned Site Investigations.
Research Needs
References
Appendices
25
31
41
41
43
44
45
47
47
58
60
vi
-------
FIGURES
Number Page
1 Location of the Hazardous Waste Dump Site in Coventry, Rhode
Island 4
2 Outline of Trench Locations at the Coventry Site as Determined
by Metal Detection Survey 5
3 Approximate Location of Seismic Refraction Profiles
4 Radar Profile Taken Outside Trench Boundary 14
5 Radar Profile Taken Within Trench Boundary Showing “Signatures”
of Buried Drums 15
6 Radar Profile Taken Within Trench Boundary Showing Buried Drums
and Suspected Chemical Contamination 16
7 Contour Map of Apparent Resistivity Values 21
8 Apparent Resistivity Depth Profiles for Several Locations Near
the West Trench 22
9 Cumulative Resistivity Depth Profiles for Several Locations
Near the West Trench 24
10 Results of Seismic Refraction Survey: Line 1 26
11 Results of Seismic Refraction Survey: Line 2 27
12 Results of Seismic Refraction Survey: Line 3 28
13 Results of Seismic Refraction Survey: Line 4 29
14 Subsurface Profile of the West Trench as Determined by Seismic
Refraction 32
15 Illustrative Trench Geometry 33
16 Comparison of Northeast and Northwest Trench Locations as
Detected by Ground—Penetrating Radar and Metal Detection 34
vii
-------
FIGURES (concluded)
Number Page
17 Location of South and West Trenches as Determined by Ground—
Penetrating Radar and Metal Detection 35
18 Recommended Sequence of Activities at Coventry Site 54
viii
-------
TABLES
Number Page
1 Techniques Used in Phase II to Provide Information Needed to
Select an Abatement Alternative 7
2 Field Activities, Picillo Property, Coventry, Rhode Island. . . . 8
3 Subsurface Conditions as Inferred from Compressional Wave
Velocities 30
4 Estimated Rectangularized Dimensions of Surface of Trenches . . . . 37
5 Estimated Number of Buried Drums Based on Extrapolation of Best
Available Data 38
6 Comparison of Remote Sensing Techniques 42
7 Systematic Approach to Determine Nature and Extent of Problem at
Coventry Site 50
8 Major Informational Needs for Implementation of Certain Abatement
Activities at Coventry, Rhode Island 57
9 Summary of Recommended Research Needs for Remote Sensing Methods
for Hazardous Waste Site Investigations 59
i x
-------
SECTION 1
INTRODUCT ION
Section 1 covers the purpose and scope of the work documented in this
report, briefly introduces the remote sensing techniques used in the study,
presents a brief history of the uncontrolled hazardous waste site which was
investigated, and describes the results of the investigations undertaken.
PURPOSE AND SCOPE
This report describes the use and evaluation of several remote sensing
techniques in conjunction with direct sample collection in order to develop a
systematic approach for subsurface investigations at uncontrolled hazardous
waste sites. It is one of two reports concerning the Coventry site which
were prepared by The MITRE Corporation for the U.S. Environmental Protection
Agency, Solid and Hazardous Waste Research Division (EPA/SHWRD). The second
report, entitled “Evaluation of Abatement Alternatives: Picillo Property,
Coventry, Rhode Island”, (MITRE Technical Report 80WOO253), represents the
continuation of a site subsurface investigation initially funded by the Rhode
Island Department of Environmental Management (DEM) and contains specific re-
sults, conclusions, and recommendations to the DEM for abatement actions.
Remote sensing techniques were employed to determine the extent (and se-
quence) to which they may be integrated with the more conventional methods of
test drilling, installation of monitoring wells, and excavations for deter-
mining information such as the following:
• nature and extent of ground water contamination
• presence and number of buried drums
• topography and condition of bedrock
• costs and effectiveness of several abatement methods.
The following remote sensing techniques were used during an Investigation
of an abandoned hazardous waste dump site in Coventry, Rhode Island:
• electrical resistivity
• ground—penetrating radar
• seismic refraction
• metal detection.
1
-------
Direct investigative techniques included:
• collection of soil and ground water samples through installation of
shallow and deep monitoring wells
o collection of bedrock samples
• collection of surface water samples
• chemical analysis of water and soil samples for priority pollutants
and total volatile organics
o excavation of buried drums of c1 iemical waste.*
Neither list is intended to be an exhaustive summary of the available
techniques, especially the remote sensing methods, which could be used in an
investigation of this type. The investigative techniques are not, in them-
selves, unusual, each having been employed previously in a similar applica-
tion (see next subsection). The important aspect of this study is that the
remote sensing and direct data collection techniques were applied in combina-
tion, which enabled a broad yet thorough understanding of the situation prior
to the recommendation of the most suitable alternative for abatement of site
pollution.
Section 2 of this report describes how the previously listed remote sens-
ing techniques were applied at the Coventry site and Section 3 presents the
results obtained. Advantages and limitations of each method are discussed
and evaluated in Section 4 and recommendations concerning investigations at
other sites and research needs are presented in Section 5.
SUMMARY OF THE REMOTE SENSING METHODS
Electrical resistivity surveying is a remote sensing technique that, when
employed as part of an investigation of an abandoned waste site, may be used
to locate the lateral and vertical extent of ground water contaminated with
ionic species and to monitor the movement of the contaminant plume with time.
The technique has often been used in ground water contamination studies. Re-
sistivity measurements are taken from the ground surface using portable equip-
ment and are used, for example, for determining the most advantageous place-
ment of monitoring wells. The method cannot be universally employed however,
as both natural conditions and man—made obstacles may affect its success at a
particular location.
Seismic refraction techniques have been applied to the exploration of new
ground water supplies and seismic refraction surveying has three potential
uses at an uncontrolled hazardous waste site:
• determination of bedrock topography in order to locate the position
of deep wells to bedrock and cost of interceptor trenches
*Conducted and funded by the Rhode Island Department of Environmental Manage-
ment.
2
-------
• determination of boundaries of large numbers of buried drums for the
purpose of estimating their number
• deter ’ination of soundness (extent of fracturing) of bedrock to eval-
uate ics effectiveness as an impermeable base for leachate interceptor
trenches or constructed physical barriers.
Metal detection and ground—penetrating radar may be utilized remotely to
locate buried drums or other underground objects of interest. The two tech-
niques effectively complement each other because they differ in information
obtained, ease of use, interpretation of data, and cost. Radar also has the
potential to provide limited data of a qualitative nature related to the con-
dition and approximate density of buried drums, to the construction of trenches
in the earth, and to the location of soil heavily contaminated with leaking
chemicals. It has been applied in various situations, such as to locate buried
sewer lines and cables or to determine the thickness of sea—ice. Ground—
penetrating radar was used during the early stages of the investigation at
Love Canal, and its application at abandoned hazardous waste sites appears to
be increasing.
HISTORY OF THE SITE
The abandoned hazardous waste dump site which was investigated is located
in Coventry, Rhode Island, approximately 20 miles southwest of Providence.
The site encompasses approximately 7.5 acres of cleared ground surrounded by
woods and wetland in a relatively rural area of the state (see Figure 1). An
undetermined quantity of chemicals had been placed into the ground both by the
burial of 55 gallon drums in five separate locations and by direct discharge
into trenches (see Figure 2). There are approximately 30 to 40 dwellings
within a one—mile radius of the dump site, but none in the downgradient area
of discharge. A swamp, 1,200 ft to the northwest of the site, is the surface
discharge area of chemicals leaching from the dump. This swamp discharges to
a body of water called Whitford Pond which is a source of irrigation water for
a cranberry bog located approximately one mile from the swamp’s outlet. To
date, no evidence of chemical contamination in Whitford Pond has been found,
based on sampling conducted by the DEN and the U.S. Environmental Protection
Agency (EPA), Region I.
State of Rhode Island officials were alerted to the dumping activities
by a fire and explosion on September 30, 1977. A court order issued on Novem-
ber 18, 1977 prohibited the property owner (Warren Picillo) from continuing
dumping activities or otherwise altering the site. From the end of 1977 to
mid—1979, the DEN conducted field investigations to quantify the seriousness
of the situation.
In October 1979, the DEN contracted with The MITRE Corporation to conduct
a systematic site assessment, and in April 1980 the investigation continued
under funding by the EPA/SHWRD. Although the investigation of the Coventry
site was conducted in two discrete phases with separate project reports, over-
all project continuity was maintained. Phase I funding was shared by the DEN
and EPA/SHWRD* and Phase II was completely undertaken by EPA/SHWRD.
*EPA/SHWRD funded all chemical analysis and the preliminary evaluation of
abatement methods. 3
-------
TO PROVIDENCE
PERRY 11111 ROAD
BRIDGE
ROAD
Figure 1.
Location of the Hazardous Waste Dump Site in Coventry, Rhode Island
-------
—..—S • %
5 S 55 •‘ ‘S
.5 ‘S
‘S
S
‘S S /
. 5
S ..’
i0
LEGEND.
-500--
--- 500 ---
— — LAND SURFACE CONTOURS (FEET ABOVE MSL
STONE FENCE
SWAMP
AREASOF HIGH METAL CONTENT DETECTED NEAR GITOUNI) S))RFUC I
0 VISIBLE METAL DRUMS
‘5
5’
S_.S •‘ ‘S..
___
S . ’
S
“S
\
‘S
‘S
. 5
.5
___ 616— —.. ..
‘a,
Ba,
LAND SURFACE CONTOURS BASED ON USGS CONTOURS AND ARE INACCURATE
WHERE DUMPING ACTIVIIIES HAVE DISTURBED GROUND
U i
-- 630--...
SCALE
-
160
O 60 (00 lEO
FEET
‘ . 5
NOR1 EST
\ TRENCI4 ’
\ ‘
\ \
Figure 2. Outline of Trench Locations at the Coventry Site as Determined by Metal Detection Survey
-------
SITE INVESTIGATION: PHASE I AND PHASE II
The overall purpose of the Phase I and Phase II investigations was to:
a) determine the nature and extent of a problem consisting of buried drums and
subsurface chemical contamination; and b) evaluate the potential costs and ef-
fectiveness of the following abatement methods:
• site encapsulation
• leachate collection and treatment
• drum removal and disposal
as well as of the “no action” alternative.
The techniques employed for data collection during the Phase I effort
were: electrical resistivity; metal detection; installation of monitoring
wells; and chemical analysis of soil, ground water, and surface water. The
field methods employed, data collected, conclusions drawn, and recommendations
made to the DEM are documented in the Phase I project report: “Hazardous
Waste Investigation: Picillo Property, Coventry, Rhode Island”, April 1980
(MITRE Technical Report 80W00032). Although the extent of the problem was de-
fined and abatement options were preliminarily evaluated, certain key pieces
of information (concerning the presence of fracturing or contamination of the
bedrock and the condition and number of the buried drums) needed to be ascer-
tained before a permanent solution could be selected. The Phase I report pre-
sented the abatement options, identified the necessary additional information,
and made recommendations for immediate and near—term actions to protect the
public health and to collect additional data. Summaries of the principal con-
clusions, recommended actions, and comparison of abatement methods are pre-
sented In Appendix A.
The relationships among the abatement methods, the additional information
needs at the conclusion of the Phase I study, and the techniques employed in
Phase II to obtain that information are shown in Table 1. Phase II was under-
taken by MITRE with EPA/SHWRD funding (ground—penetrating radar, seismic re-
fraction, bedrock sampling, and chemical analysis) and by the DEN under inter-
nal state funding (exploratory excavation of Northeast Trenches). These
activities and their purposes are summarized in Table 2.
The recommendations concerning long—term abatement methods determined
from the Phase II field activities are contained in Appendix B.
6
-------
Table 1
Techniques Used in Phase II to Provide Information
Needed to Select an Abatement Alternative
Alternative Action
A
dditional Information Required
at end of Phase I to
Select Alternative
Phase II Technique to
Obtain Information*
1.
No Action
•
•
•
condition of source (drums)
state of nearby pond
contaminant underfiow at
swamp
• radar, exploratory excava-
tion
• additional wells, chemical
analysis of soils and water
samples
•
ultimate disposition of all
pollutants
2.
Drum Removal and Disposal
•
condition of source (drums)
• radar, e’ loratorv excava—
(excavation, testing, and
proper disposal of drums
and contents, and contam—
mated soils)
•
tion
condition of soil
• exploratory excavation,
chemical analysis of soil
samples
3.
Site Encapsulation (con—
•
condition of source (drums) • radar, exploratory excava—
struction of impermeable
barriers around source
of pollutants)
•
tion
condition of bedrock
• seismic refraction, core
drilling, deep wells
4.
Leachate Collection and
Treatment
a. Limited Option (in—
•
condition of source (drums) • radar, exploratory excava—
terceptor trenches
constructed adjacent
to site walls)
•
condition of bedrock tion
• seismic refraction, core
drilling, deep wells
b. More Complete Option
•
same as above • same as above
(interceptor trenches
constructed 600 feet
downgradient of site
walls)
*
Metal detection had previously been used to locate trenches; electrical resistivity to delineate
leachate plume. Radar could have been ennloyed in lieu of or in conjunction with metal detec-
tion, as recommended for other sites; potential radar effectiveness was unknown at the time of
the initial survey.
7
-------
Table 2
Field Activities
Picillo Property, Coventry, Rhode Island
Phase Activity Purpose
1 a o lateral electrical resistiv— o location of plume
ity profile
• placement of monitoring wells
• metal detection survey o location of buried drums
• soil, ground, and surface • determination of ground water
water sampling elevations
• determination of occurrence
and type of contamination
1 1 b • ground—penetrating radar • elaboration of number and
location of buried drums
• limited excavation of buried • determination of number and
drumsC condition of drums
o seismic refraction survey o determination of bedrock pro-
file
o evaluation of potential to
determine lower boundary of
burled drums
• estimation of soundness of
bedrock
• bedrock sampling • determination of soundness of
bedrock
• determination of contamination
of bedrock
o surface and ground water • elaboration of subsurface
sampling contamination
• vertical electrical re— • evaluation of potential to
sistivity profile determine lower boundary of
burled drums
a. Funded by the Rhode Island Department of Environmental Management.
b. Funded by the EPA/SHWRD.
c. Funded and conducted by Rhode Island Department of Environmental
Management.
8
-------
SECTION 2
REMOTE SENSING TECHNIQUES
Section 2 gives a brief description of each of the remote sensing tech-
niques employed during Phases I and II at the Coventry site, including dis-
cussions of the equipment, other previous applications, and the specific pro-
cedures used in the field.
ELECTRICAL RESISTIVITY
The electrical resistivity of a geological formation depends upon the
conduction of electric current through the particular subsurface materials.
Since most of the geologic formations that contain water have high resistivi—
ties, the electrical resistivity of a saturated rock or soil is primarily a
function of the density and porosity of the material and the concentration of
the conducting ions within the saturating fluid. In a resistivity survey, an
electric current is passed into the ground through a pair of current elec-
trodes and the potential drop is measured across an inner pair of potential
electrodes. The “apparent resistivity” is determined by the equation, Ra =
2rA(V/I), where A is the electrode spacing, V is the potential difference,
and I is the applied current. The depth of penetration is controlled by the
distance between the electrodes (called the A—spacing) and is approximately
equal to half of this distance.
Resistivity measurements can be taken in the form of either lateral or
depth profiling. In lateral profiling, the Wenner electrode configuration is
used, where the A—spacing is fixed and the electrodes are moved ahead in a
straight line. This method allows lateral converage at a more—or—less con-
stant depth, and can be used to define aquifer limits or to delineate varia-
tions in ground water quality. To obtain a depth profile, a series of mea-
surements are taken at different electrode spacings to the left or right from
a centralized electrode. This technique is called the Lee modification of
the Wenner electrode array. Plotting the apparent resistivity against the
electrode spacing gives an indication of the resistivity of various layers at
successively greater depths. Interpretation is most successful in areas hav-
ing a simple layered structure since the apparent resistivity is a measure of
the effects of all the layers between the maximum depth of penetration and
the surface. Depth profiling has often been used to determine the thickness
of glacial aquifers overlying bedrock and to locate the saltwater/freshwater
interf ace in coastal aquifers. Additional information regarding the elec-
trical resistivity technique may be found in U.S. EPA (1978) or Freeze and
Cherry (1979).
9
-------
Both lateral arid depth profiling surveys were conducted at the hazardous
waste site in Coventry using a Bison Instruments Model 2350B Earth Resistivity
meter powered by a 90—volt battery. An electrode spacing of 20 feet was used
for the lateral profiles in the areas of the trenches and the swamp where the
depth of gromd water contamination was suspected at 10 to 15 feet. Two lat-
eral profiles using an A—spacing of 50 feet were also conducted approximately
2000 feet west and north of the immediate site walls, where it was suspected
that the contamination might be detected at depths of 25 to 35 feet.
Additionally, seven depth profiles using the Lee modification were con-
ducted in the vicinity of a large trench containing buried drums located by
the metal detection survey near the western boundary of the site. Electrodes
were set at intervals of 1, 2, 4, 8, 16, and 32 feet at each profile location,
permitting a maximum depth of investigation of approximately 15 feet. Resis-
tivity readings were taken at each spacing interval for the left, right, and
central spacings. This particular spacing interval was chosen based upon the
intents of the investigation, which were to identify changes in subsurface
contamination and to locate the boundary defining the bottom of the trench.
The lateral surveys were conducted by Fred C. Hart arid Associates, Inc.
and the vertical surveys by Stephen A. Alsup and Associates, Inc.
SEISMIC REFRACTION
The seisnuc refraction method is based on the principle that elastic
waves (mechanical rather than electromagnetic) travel through different sub-
surface strata at different velocities. Elastic waves are introduced to the
ground surface by an energy source, usually a small explosion or a hammer blow
on a steel plate for shallow investigations. The refracted waves are detected
by small seismometers (geophones) located on the surface at various distances
from the energy source. A seismograph records the travel time between the vi-
bration and the arrival of the elastic wave at the geophones. Plotting arriv-
al time versus distance from the energy source to geophone from a series of
seismograph records enables the determination of strata depths and their seis-
mic velocities through the use of simple refraction thenry. Greater detail
concerning geophysical surveys may be found in Dobrin (1960).
Seismic surveys have been used in hydrogeologic investigations to provide
subsurface geologic information, such as depth to bedrock and presence of
buried bedrock valleys, more rapidly and at lower cost that could be deter-
mined through actual test drilling. Interpretation of seismic data, however,
is difficult in areas having complex stratigraphy or stratigraphy exhibiting
little contrast in propagational velocities. In addition, propagational ve-
locities within the subsurface layers must increase with depth because low—
velocity layers that underlie high—velocity layers are completely obscured by
the refraction of the high—velocity layer. For these reasons, limited test
drilling should be conducted in conjunction with the seismic method to confirm
the subsurface geologic interpretation.
Seismic refraction profiling of approximately 2,850 linear feet was per—
formed at the Coventry site in two days of field work. Figure 3 shows the
10
-------
Figure 3.
Approximate Location of Seismic Refraction Profiles
(lines not surveyed)
11
S
LAND SURFACE CONTOURS
(FEET ABOVE MSI)
STONE FENCE
UNIMPROVED DIRT ROAD
MONITORIONG YE1L
SURFACE WATER STATION
(APPROX LOCATION)
-------
location of seismic profiles. A Geometrics/Nimbus Model ES121OF Multichannel.
Seismograph was used to record and collect the voltage outputs from 12 Mark
Products L—15 vertical geophones spaced at 20—foot intervals for each refrac-
tion spread. The energy source used to initiate each record and shock wave
was a 30—pound weight drop or 10—pound sledge hammer blow on a steel plate
with an attached impact start switch. The Model ES121OF Seismograph, which
includes a digital memory of waveform from each data channel, allows repeti-
tions of the elastic waves from a series of hammer blows thereby enhancing the
ability to detect signals and pick arrival times. The use of the hammer drop
as the energy source (which is preferable for safety reasons to using small
explosive changes in this type of investigation) would be more difficult with-
out the digital memory of the seismograph. Impact points for this survey were
at the end of, and quarterly along, the refraction spread, providing a locus
for depth calculations at 80—foot intervals along each spread. Data continu-
ity and repetition were achieved by repeating end shots where refraction lines
were longer than one spread length.
The seismic refraction survey was conducted by Stephen A. Alsup and As-
sociates, Inc.
METAL DETECTION
The entire 7.5 acre site in Coventry was surveyed by personnel from Fred
C. Hart and Associates, Inc., with a Fisher M—Scope (Model TW—5) metal detec-
tor. This equipment is designed for locating buried metal objects by inducing
an electromagnetic field around the object in response to radiation from a
transmitter. The average depth of detection for metal objects is dependent on
the amount of background t?noiseu Thus in areas free of buried metal, the
probable depth of detection for metal objects was approximately six to eight
feet. In areas of buried drums, the sensitivity setting of the instrument had
to be cut back, resulting in a potential depth of metal detection of approxi-
mately four to five feet. In areas where buried drums were suspected, based
on disturbed ground or the initial gross scan of the overall site, the survey
was conducted by traversing closely spaced grid lines.
GROUND-PENETRATING RADAR
The technique of ground—penetrating, or impulse radar involves the repet-
itive propagation of short—time duration (on the order of a few nanoseconds,
nsec) pulses of electromagnetic energy in the radar frequency range downward
into the ground from a broad bandwidth antenna on (within a few inches of)
the surface. Reflections from subsurface interfaces are received by the an-
tenna during the off—period of the pulsed transmission, processed electronic-
ally, and recorded to yield a continuous profile of subsurface conditions as
the antenna/transmitter—receiver unit is moved across the ground surface.
The depth to an interface, or the surface of a “target’ t such as a metal
drum, is determined by measuring the time for a radar pulse to travel to the
interface and reflect back to the surface. In air, radar travels at the speed
of light or about one nsec/ft. Typical speeds in soil range from six nsec/ft
for drier soils to eight nsec/ft for wetter soil conditions, hence depth cal-
ibration would be based on a round trip velocity in the range of 12 to
12
-------
16 nsec/ft of depth. A real—time display of the depth profiles can be obtained
for field interpretation using a strip—chart recorder, and the data can be
recorded more completely and permanently on magnetic tape for subsequent play-
back and interpretation.
Impulse radar has been used as a geophysical technique in such applica-
tions as surveying archeological sites, locating sewer lines and buried cables
prior to construction activities, and profiling lake and river bottoms (see
Morey and Harrington (1972) and Campbell and Orange (1974)). The application
to locating buried drums of chemical wastes is relatively new, and further
refinements both in the technology and in data interpretation are anticipated.
The field survey was conducted by the equipment manufacturer, Geophysical
Survey Systems, Inc. (GSSI). The equipment used was GSSI’s SIR System 7 ground
radar. The survey of the trench areas, representing approximately two acres,
took two days. Following experimentation with two alternative antennas and
center frequencies, GSSI Model 3105AP operating at a center frequency of 300
MHz and GSSI Model 3102 operating at 600 MHz, the latter was chosen for most
of the survey due to its improved spatial resolution at shallower depths. The
operating depth varies approximately as the inverse square of the frequency,
all else being equal.
One large trench (labelled the West Trench on Figure 2) located by the
metal detection survey was surveyed with the 300 MHz antenna set at a nominal
depth of 25 feet, later calibrated at 24.4 feet, based on average soil condi-
tions. The other trenches (labelled Northwest, Northeast, and South) were
subsequently surveyed using the 600 MHz antenna set at a nominal depth of 12.5
feet. The survey was conducted according to a rectangular grid. All trenches
were surveyed longitudinally by using parallel radar transects at spacings of
ten feet. Transverse transects, or cross—cuts, were made at intervals of 20
feet for the Northeast Trench and 40 feet for the West and Northwest Trenches.
The antenna unit was pulled along each transect manually, and the data re-
corded by wire connection with equipment located in a stationary van on the
site, which also served as the power source. The major equipment components
were a control unit with cathode ray tube display, a tape recorder, a graphic
(chart) recorder, and a solid state inverter.
The radar beam has a spread of ±450 in the fore and aft directions, and
±200 laterally. Any target detected within this beam will be recorded as
being directly below the point of the surface where the signal is transmitted
and received, and signals are reflected only from surfaces perpendicular to
the direction of the signal. The use of a 10—foot grid spacing thus resulted
in a sampling approach, as opposed to full coverage of the subsurface volume.
However, even with a very fine grid, a fraction of the buried drums would be
missed by the radar due to (a) their orientation or (b) their being “shielded”
by metal drums closer to the surface, since metal is a near—perfect reflector
of radar energy.
Illustrative data from the survey are shown in Figures 4, 5, and 6. The
ground surface is at the top of each figure whereas the bottom of the figure
corresponds to a depth of approximately 12 feet. The vertical dashed lines
13
-------
Figure 4. Radar Profile Taken Outside Trench Boundary
14
-------
Figure 5. Radar Profile Taken Within Trench Boundary
Showing “Signatures” of Buried Drums
15
-------
Figure 6. Radar Profile Taken Within Trench Boundary Showing Buried Drums and
Suspected Chemical Contamination (Blurriness in Left—Center of Photo)
16
-------
are markers produced electronically in the field, which correspond to 10—foot
intervals. Figure 4 i1lustr ces a subsurface profile where there are no
buried drums. This profile :as taken over an area of undisturbed soil.
In Figure 5, there are a number of individual targets identifiable by
the characteristic hyperbolic “signature.” This signature results from the
increased travel tlrne bet ’eert the target and the antenna when the beam ap-
proaches or moves away from the target versus when it is di ectly over the
target. Each reflecting target will produce three characteristic hyperbolas.
It is possible for a skilled interpreter to distinguish between the signature
caused by a drum or boulder either in the field or from the recorded data. In
the field, a metal object can be determined instrumentally by comparing the
polarity of the target signal to the background signal. A metal object which
is essentially a perfect reflector, will produce a signal that is “in phase”
with background. An object such as a boulder will produce a signal reversed
from background. Close examination of the recorded data will also show another
reflecting signal produced by a metal object in addition to the three charac-
teristic hyperbolas. This fourth image found above the characteristic three
is also caused b the reflection from metal and would not be present on the
data if the target were a boulder.
Figure 6 shows a blurred effect which is interpreted as being caused by
a concentration of contaminants, as from a leaking drum. Chemical analysis
by the DEM has shown that some of the chemicals being released from the drums
were ionic, which is a characteristic that would increase the attenuation of
the radar signal strength. Therefore a noticeable blurred contrast relative
to the average signal strength is cbserved on data taken over a trench con-
taining a high concentration of contamination.
17
-------
SECTION 3
RESULTS OF FIELD STUDIES
Section 3 describes the results of field studies at the hazardous waste
dump site in Coventry.
PLUME DELINEATION
Having determined the information needed to evaluate the long—term abate-
ment alternatives, a Phase I hydrogeologic investigation was planned to achieve
certain general objectives. Those objectives were:
• to identify the source of contamination and type of contaminants in
soil and water
o to define the areal extent of ground water contamination and the di-
rection of flow
• to determine the quantities of contaminated ground water flowing away
from the site.
A principal component of the site investigation was the installation of shal-
low monitoring wells in order to collect soil and water samples and to deter-
mine ground water elevations.
Because natural conditions at the site were such that measurement of
electrical resistivity was expected to be successful (see Section 4), a later-
al profiling survey was performed to facilitate the placement of monitoring
wells and a depth profiling survey was conducted to determine vertical contam-
ination patterns. The surficial geology map of the particular quadrangle on
which the site was located showed that the area was characterized by a rela-
tively thin mantle of till or outwash overlying crystalline bedrock. Nine
monitoring wells previously installed on top of the site by the property owner
indicated a shallow unconfined aquifer within the glacial deposits, with the
direction of flow toward the swamp. A surface water sample analyzed by the
State of Rhode Island in the summer of 1979 showed above average concentra-
tions of iron and chloride, both of which would increase the conductivity of
the ground water. All of these factors: shallow water table, generally uni-
form unconsolidated material overlying crystalline bedrock, and sharp con-
trasts between contaminated and natural water, indicated that the resistivity
method could be used to establish the lateral and vertical extent of the
ground water contamination. The information obtained could then be used to
aid the selection of locations for additional monitoring wells.
Preceding page blank 19
-------
Figure 7 shows the apparent resistivity values obtained during the later-
al profiling survey, plotted on a contour map. The measurements taken using
both the 20— and 50—foot A—spacing are given. Rather than showing that the
plume was moving directly toward the swamp in a northwesterly direction as
suspected from measurement of water levels in existing monitoring wells, the
apparent resistivity values gave an indication of two distinct plumes (a west-
ern and a northern plume) at different depths with each having a separate
source. With increasing distance from the site, however, the two plumes joined
finally to discharge in the swamp.
The western plume, which was defined primarily using the 20—foot A—spac-
ing, appears to be generally within 10 feet of the surface. Most of the plume
moving toward the north was defined using the 50—foot A—spacing and appears
generally deeper than 20 feet below the surface. However, some shallow con-
tamination is also apparent along the northern border of the site near the
trenches. Shallow bedrock off the northwest corner of the site was considered
the most likely explanation for the high apparent resistivity values between
the two plumes, although this explanation was later proven to be false (see
next subsection). Hence, the results of the resistivity survey suggest that
additional monitoring wells be located to determine the existence of the shal-
low bedrock and to substantiate the presence of two separate plumes.
Additionally, a discovery of a contaminant source along the partly grass—
covered western edge of the site was important information, since it served
to indicate how far south and west the monitoring well program ideally should
extend. As discussed in Section 2, locating this additional source of contam-
ination may also have been possible using the radar technique based upon com-
parison of signal strength.
As mentioned earlier, several depth profiling resistivity surveys were
conducted in the vicinity of the West Trench to determine vertical changes in
contamination. Figure 8 is a plot of the apparent resistivity versus approx-
imate depth below the surface (computed as half of the A—spacing) for several
profiles taken over and outside of the trench boundary. Presentation of the
data in this manner allows comparison of the “normal” or background resistiv-
ity patterns observed outside the West Trench where no contamination is ex-
pected, to the patterns observed over and downgradient from the trench where
the occurrence of contamination is anticipated.
Profiles 1 and 6, taken outside the major plume boundary, as indicated
by the lateral resistivity survey, are indicative of background patterns.
Profile 2 was taken in the open, unfilled trench where no drums were burled,
but bulk chemicals were suspected to have been discharged. This profile shows
the effect of free—standing water in the trench and perhaps some slight con-
tamination within the first few feet of the surface, and then an increase in
resistivity to near background conditions. This indicates that either the open
trench was not used for bulk dumping of significant amounts of chemicals with
high electrical conductivity (since conductive effects decrease rapidly below
several feet of depth), or that significant portions of the chemicals have
been removed by leaching. Profiles 3, 4, 5, and 7, taken over or downgradient
from the West Trench, show the varying degrees of contamination within the
plume. A very high concentration of drums may explain the extremely low
20
-------
I- ’
—— — — — LAND SIMFACE CONTOURS FIST ASOVI MILl
STONE FENCE
2k. SWAMP
CONTOUR OF APPARENT RESISTIVITY VALUES O*4M-FTl AT
20 FOOT A WACSNO
_ _ CONTOUR OF APPARENT RESISTIVITY VALUES OHMFTJ AT
00 FOOT A ACINO
AREAE OF LOW APPARENT RESISTIVITY VALUES
Figure 7. Contour Map of Apparent Resistivity Values
-------
APPARENT RESISTIVITY (OHM — FEET)
-a
0
o 0
o 0
o 0
0
-I
rn
D
rn
-V
-4
I
m
0
C ,)
C
‘1
‘ i i
-I’
m
m
-4
Figure 8. Apparent Resistivity Depth Profiles for Several Locations Near the West Trench
-------
resistivity measured at the location of profile 4. The volume of highly con-
ductive chemicals at this location is inferred as being very large in order
to have such a strong effect on the resistivity measurements. A very local-
ized concentration of conductive chemicals would not be expected to have such
a strong effect with this particularly wide electrode spacing (32 feet).
Figure 9 shows the same seven profiles, but on a plot of “cumulative” re-
sistivity versus approximate depth below the surface. Slope changes or breaks
in the cumulative curves generally indicate the depth to the underlying unit,
and the direction of slope change indicates the relative resistivity values
of the subsurface materials. Thus, an increasing slope signifies that the
underlying unit has a higher relative resistivity, whereby a decreasing slope
denotes a lower relative resistivity in the underlying unit. In accordance
with the interpretation of Figure 8, Figure 9 shows higher relative resistivi—
ties with depth (no contamination) for profiles 1 and 6. Interpretation of
profile 2 is more straightforward using the cumulative resistivity plot.
Slight decreases in slope, denoting lower resistivity or presence of soil
moisture or contamination, are evident between two and five feet and below
ten feet. Profiles 3, 4, 5, and 7 show a continued decrease in slope with
depth, which is expected because of their locations within the buried drum
area or downgradient from the trench.
Determination of the depth of the bottom of the West Trench was not pos-
sible using the vertical resistivity plots. It is not clear, however, whether
the particular A—spacing used for these profiles (32 feet) was wide enough to
allow adequate penetration necessary to detect the bottom of the trench. A
second possibility is that there is no detectable change in contamination at
the bottom of the trench thereby causing a continuation of the low resistivity
readings beyond the region of buried drums.
Fifteen monitoring wells were installed following the lateral resistivity
survey. Refusal depths, tentatively assumed to reflect the approximate top of
bedrock, did indicate a mound in the bedrock surface off the northwest corner
of the site. Four wells in this vicinity were dry, which also gave credence
to the results obtained from the resistivity survey, namely the existence and
location of two plumes. In addition, soil samples taken from these seine lo-
cations were much less contaminated than soil samples taken from borings lo-
cated within the plume boundaries. Consideration of these factors seemed to
indicate that ground water flow was being diverted around a bedrock mound and
this had resulted in the detection of high apparent resistivity values in this
area. It was found by the Phase II bedrock drilling, seismic refraction sur-
vey, and chemical analysis of soil and ground water that the bedrock mound did
not exist and that contaminated ground water was indeed traveling in this lo-
cation. In general, the ground water is at a greater depth below the surface
in this region than the other surveyed areas, thus resulting in higher rela-
tive resistivity values, and subsequent incorrect interpretation.
Although a wider A—spacing (50 feet) was used in this northwest region,
the subcontractor failed to calibrate the readings by surveying in uncontami-
nated areas. It is additionally possible that the 50—foot A—spacing was in-
sufficient to obtain interpretive results due to the depth of the saturated
zone (up to 25 feet) in this area and that 70 to 100—foot electrode widths
23
-------
CUMULATIVE RESISTIVITY (OHM — FEET)
WEST )Z
— 4 .—.—--—-
TRENCH OPEN
TRENCH
SCALE
I I 1
0 50100
FEET
-I
0
0
0
1
2.5
5
>
0
-v
0
-l
rn
D
m
-u
-I
I
m
I-
0
‘1
0
rn
‘1
m
m
-I
10
15
LOCATION OF PROFILES
7 6
N
Figure 9. Cumulative Resistivity Depth Profiles for Several Locations Near the West Trench
-------
should have been employed. Lines of current from one electrode to another
travel in semi—elliptical paths; because of the path slope, the Influence of
the saturated zone on resistivity readings decreases in a non—linear relation-
ship with depth. The water table will effect a minor change on resistivity
values if it is at or near the maximum extent of the current lines. There-
fore, it is recommended for other surveys that the nominal depth of the signal
be twice the expected depth of the water table.
DETER1 1INATI0N OF BEDROCK TOPOGRAPHY AND DEPTH OF BURIED DRUMS
Complete verification of the shallow bedrock off the northwest corner of
the site was not possible until the bedrock coring and the seismic survey were
performed. The drilling showed that the refusal depths of the previous bor-
ings had actually been due to boulders and/or very dense till. At each boring
location, the bedrock (a granite gneiss) was discovered to be 10 to 30 feet
deeper than anticipated. The seismic survey indicated that the bedrock sur-
face was gently rolling, varying from approximately 10 feet below ground sur-
face near the swamp to approximately 70 feet below ground surface on top of
the site. Figure 10 shows the seismic profile from W21 through the northwest
region of the site (see Figure 3 for approximate location). Subsurface condi-
tions inferred from the compressional wave velocities are given in Table 3.
Interpretation of the velocity units is given where correlation with the test
borings is possible.
The deep boring drilled in the area between the two plumes (W23) showed
that the bedrock was highly weathered and fractured. A piezometer installed
in the fractured bedrock indicates that the granite gneiss is hydraulically
connected to the unconsolidated glacial deposits. Therefore, ground water is
not being diverted around a shallow bedrock mound as had been inferred from
the resistivity survey and Phase I drilling, but is actually moving over this
area toward the swamp at depths greater than 20 feet. A ground water sample
taken from W23 was found to contain a diverse assortment of volatile organic
pollutants similar in concentration to samples taken from wells within the two
plumes.
The seismic profiling proved to be most useful for the determination of
the depth to dense or competent bedrock, which is the information needed when
considering certain on—site abatement alternatives such as source encapsula-
tion or leachate collection trenches. Typically, compressional wave veloci-
ties above 11,000 ft/sec are generally indicative of very dense bedrock. Fig-
ures 11, 12, and 13 show the remaining seismic profiles taken at the Coventry
site to gain information about depth to bedrock. Interpretation of velocity
units from correlation with test borings is given where appropriate.
The method of seismic refraction was used over the West Trench in an ex-
perimental attempt to determine the depth of the base of the buried drums.
Neither ground—penetrating radar nor metal detection was able to show the
lower boundary of drums and vertical electrical resistivity measurements re-
vealed no readily—interpretable trends. The importance of knowing the depth
of trenches is most clearly shown by the fact that the number of drums esti-
mated in the Northeast Trenches during the Phase I study were significantly
lower than what was found during the exploratory excavation. A remote sensing
25
-------
SE
NW
W21
+ 100
560
U i 540
520
VERTICAL 10
SCALE 20
30
FT
HORIZONTAL SCALE
20 60 100FT
. p i p •
200
300
400
W23
600
SOURCE S A ALSUP AND ASSOCIATES, INC.
I ’ , )
0 ’
700
FRACTURED 12000-13000
DIABASE (?)
INTRUSIVE
560
540
520
500
480
GRANITE
GNE ISS
16000
LEGEND
• SHOT POINTS
— INFERRED SUBSURFACE
VELOCITY BOUNDARIES
5200 SEISMIC WAVE VELOCITIES
IN FT/SEC
TEST BORING LOCATION
Figure 10. Results of Seismic Refraction Survey: Line 1
-------
$
N
460
HORIZONTAL SCALE
20 60 100 FT
I I I I I
VERTICAL 10
SCALE 20
30
FT
SHOT POINTS
— INFERRED SURSURFACE
VELOCITY BOUNDARIES
5200 SEISMIC WAVE VELOCITIES
IN FT/SEC
4 TEST BORING LOCATION
SOURCE S A ALSUP AND ASSOCIATES. INC
100
200
300
400
500 600
I I
GNEISS
440
800
16400
W22
+
520
500
LEGEND
460
:RACTURED
GRANITE
GNE ISS
440
Figure 11. Results of Seismic Refraction Survey: Line 2
-------
w
E
100
200
300
+
W24
400
5O0
480
440
uJ
500
600
20000
700
GRANITE
GN E ISS
500
18500-20000
480
HORIZONTAL SCALE
20 60 100 FT
I I I I I
460
LEGEND
VERTICAL 10
SCALE 20
30
0
FT
5200
SHOT POINTS
INFERRED SURSURFACE
VELOCITY BOUNDARIES
SEISMIC WAVE VELOCITIES
IN FT/SEC
+
TEST BORING LOCATION
SOURCE S A ALSUP AND ASSOCIATES, INC
Figure 12. Results of Seismic Refraction Survey: Linc 3
-------
S N
100 200 300 400
I I I I 4W23
520
1200 160
-J ___________________________
100 1600 ‘—FINE-MEDIUM
C l)
SAND
LU
> 500
0
‘— GRAVELLY SAND
::: 00 144004600 FRACTUREDGRANITE
4
GN E ISS
0
460
460
LU
-J
LU
440 44
LEGEND:
• SHOT POINTS
HORIZONTAL SCALE
20 60 100FT — INFERREDSURSURFACE
I I I I I
VERTICAL 10 I VELOCITY BOUNDARIES
SCALE 20 1 5200 SEISMIC WAVE VELOCITIES
30 IN FT/SEC
FT + TEST BORING LOCATION
SOURCE S.A. ALSUP AND ASSOCIATES, INC.
Figure 13. Results of Seismic Refraction Survey: Line 4
-------
Table 3
Subsurface Conditions as Inferred from
Compressional Wave Velocities
Wave Velocity
Subsurface Conditions
(Feet/Second)
600 — 2,200 Very loose and unconsolidated soil or fill, not
saturated with ground water or other fluid, may
include ablation tills, very recent sand/gravel
deposits.
2,400 — 3,600 More compact deposits than above, but of interme-
diate density. Often Includes non—saturated coarse
sands and gravels, some ablation tills, and some
compacted fill materials.
4,000 — 5,600 Materials of either type above, with ground water
saturation. Degree of saturation and permeability
generally increases with increasing velocity to
the mid—range values, then may decrease because of
finer grain sizes in the deposits.
6,000 — 8,600 Typically dense glacial tills, either with or with-
out ground water saturation. May include deeply
weathered or fractured bedrock, with possible marine
clays in the lower part of the velocity range.
9,000 — 11,000 Moderately to weakly weathered or fractured bedrock,
may include very dense lodgement tills.
above 11,000 Typically very dense to dense sound and competent
bedrock units.
Note: There is overlap among the ranges above with regard to the
particular type of deposit represented by the compressional
wave velocities. Geological interpretation is commonly re-
quired for identification of deposit type.
Source: S. A. Alsup and Associates, Inc.
30
-------
method that can effectively deteririne depth of drums would greatly aid other
similar investigations for the determination of drum number and cost estimates.
The results of the seismic profile of the West Trench are shown by Figure
14. Three distinct velocity units are shown on the profile and it is believed
that the upper one, varying from 7 to 14 feet, represents the disturbed soil
surrounding the buried drums.
According to Table 3, the intermediate unit wave velocities (3200 — 3600
ft/sec) shown on Figure 14 do not imply saturated conditions. However, the
water table is known 1 o be midway within this intermediate velocity unit (ap-
proximately 25 feet below the ground surface). Therefore it is evident that
Table 3 presents limited approximations of subsurface conditions and that both
the interpretation of the base of the West Trench and the actual characteris-
tics of the 3200 — 3600 ft/sec velocity unit need to be confirmed through test
drilling and/or excavation.
DETERMINATION OF TRENCH LOCATION AND GEOMETRY
The location and dimensions of the trenches used to estimate the number
of drums in each trench were based on a combination of data from metal detec-
tion, ground—penetrating radar, and the exploratory excavation. The results
from the seismic profiling of the West Trench were used to estimate the lower
limit for the bottom of the trenches, even though these data have not been
confirmed. Figure 15 illustrates trench geometry with a typical trench cross—
section and longitudinal section. For the purposes of estimating the number
of drums contained in the trenches, the angle of the vertical side walls was
assumed to be 600, the angle of the declining surface of drums 450, and the
angle of descent at the trench ends 450 The angle of repose for disturbed
site soil is approximately 450, but excavated side walls were shown to main-
tain a much steeper slope. Since the radar probed to a depth of 12 feet in
contrast with the four to six feet in the vicinity of the trenches for metal
detection, the radar would be expected to present a somewhat more accurate in-
dication of trench boundaries. In the case at hand, trench boundaries from
the two techniques were compared for each trench. Figures 16 and 17 show this
comparison of the areal outlines of the trenches as reported by the subcon-
tractors who conducted the field surveys. Although there is not complete
overlap between the outlines determined by the two methods, it is suspected
that the deviation is due mainly to inaccuracies in the reporting of the metal
detection results. The radar found two trenches in the “Northeast Trench”,
versus the single trench identified previously with metal detection (Figure
16); the explanation for this is not known. On the other hand, the radar data
for the West Trench had to be supplemented by data from the metal detection
(Figure 17).
The radar provided, in addition, some useful qualitative information on
the way drums were placed and on the trench construction. For example, al-
though there were isolated instances where several drums appeared to be neatly
stacked, this was the exception rather than the rule; the drums for the most
part appeared to be randomly stacked based on the radar data, and at least in
the top eight or so feet below the surface (where individual drums most clear-
ly could be identified) the drums appeared to be present in clusters as opposed
31
-------
S N
100 200
I I
-J
0 550 —
w 1100
530 530
3200-3600
510 510
490 490
I . . )
N.)
LEGEND:
HORIZONTAL SCALE 0 SHOT POINTS
20 60 100 FT — INFERRED SURSURFACE
VELOCITY BOUNDARIES
VERTICAL to-i
SCALE 20 .J 5200 SEISMIC WAVE VELOCITIES
IN FT/SEC
30 J
FT
SOURCE: S.A. ALSUP AND ASSOCIATES, INC.
Figure 14. Subsurface Profile of the West Trench as Determined by Seismic Refraction
-------
TRENCH DEPTH
TRENCH DEPTH
1
Figure 15.
Illustrative Trench Geometry
BOUNDARY SEEN
BY RADAR AND
METAL
DETECTION
GROUND
a TRENCH CROSS-SECTION
BOUNDARY SEEN BY RADAR AND METAL DETECTION
GROUND SURFACE
b. TRENCH LONGITUDINAL VIEW
33
-------
Os
20$
40S
60$
40E 120E
RADAR
GRID NORTHWEST
TRENCH
160E 200E 240E
NORTHEAST
TRENCHES
360E
LEGEND
STONE FENCE
+ MONITORING WELL
_.— TRENCH BOUNDARY BY RADAR
TRENCH BOUNDARY BY
—— METAL DETECTION
SCALE
I I I I I J
O 20 40 60 80 100
FEET
Figure 16.
Comparison of Northeast and Northwest Trench Locations as Detected by
Ground—Penetrating Radar and Metal Detection
-------
75N 115N 155N 195N 215N 255N
15E
35E
55E
85E
145E
+
W4
---
Lri
WEST TRENCH
W-
GRID
55N
LEGEND.
STONE FENCE
± MONITORING WELL
TRENCH BOUNDARY BY
RADAR
TRENCH BOUNDARY BY
METAL DETECTION
SCALE
I I I I I
O 20 40 60 80
FEET
Figure 17. Location of South and West Trenches as Determined by Ground—Penetrating Radar and Metal Detection
-------
to being uniformly dense throughout a trench. Also, the top surface of the
druns displayed an “angle of repose” from the sides to the center of the
trench cross—section.
The radar was not able to detect the bottom of the trenches, partly be-
cause the upper drums masked what was beneath. Even in the West Trench, where
a 25—foot nominal depth was probed, the trench bottom could not be located
from the data. The radar data can often be used, however, to determine the
interface between the sides of the trenches and the undisturbed soil. Radar
signals from within the trench are generally stronger than signals from out-
side the trench. This contrast is attributed to the fact that the disturbed
soil within the trench has a higher dielectric constant because it is more
porous and has a greater moisture content than undisturbed soil. For future
work at other sites, it is suggested that deep radar probing at and just out-
side a trench boundary may be successful in determining the maximum depth of
drums, depending on the steepness of the side of the trench relative to the
radar beam, the clarity of the radar signal at this depth, and the subsurface
material at the given site.
In order to produce estimates for the number of drums remaining buried,
a theoretical trench geometry shown by Figure 15 was employed. It was also as-
sumed for the purpose of the drum estimates that a two—foot layer of soil
covered the top of the burial area and two nominal trench depths of 14 and 22
feet were used in order to bracket the range determined from remote sensing
and direct excavation.* The bottom of the trenches are assumed to be level
with no irregularities. Straight sides for the horizontal widths and lengths
have also been assumed; the dimensions used for determining the volumes of
each trench are shown in Table 4.
Two densities of drums (percent of volume of drums within trench volume
below the cover layer of soil) were used for the drum number estimates: 90
percent and 50 percent. A drum density of 90 percent represents the closest
packing arrangement possible for cylinders without regard to interferences im-
posed by the actual geometry of the trench boundaries.
An exploratory excavation of the Northeast Trenches was conducted by the
DEM as a result of Phase I recommendations by MITRE. An actual drum density
of 54 percent was calculated for the Northeast Trenches using the results**
from the excavation combined with the theoretical geometry shown by Figure 15.
The calculated 54 percent density was rounded off to 50 percent for the lower
limit calculations of the drum number estimates.
The estimated range of the number of drums remaining buried at the
Coventry site is found in Table 5. The drum estimate was performed by cal-
culating the volume of each trench and multiplying the volume by the assumed
*14 ft — seismic survey of West Trench
22 ft — average depth of deeper Northeast Trenches
**2300 drums removed from two trenches with the following dimensions: Trench
A = 120 ft long, 20 ft wide, and 22 ft deep (average), and Trench B = 60 ft
long, 15 ft wide, and 17 ft deep (average).
36
-------
Table 4
Estimated Rectangularized Dimensions of Surface of Trenches
Trench Location
Width (feet)
Length (feet)
Northwest
50
235
West
45
240
South
30
60
Source: Ground—penetrating radar and metal detection
survey.
37
-------
Table 5
Estimated Number of Buried Drums
Based on Extrapolation of Best Available Data
Trench
Location
Maximum
Drum Density
Drums Randomly
Stacked
d
= 14
ft
d = 22
ft
d
= 14
ft
d = 22
ft
Northwest
14,800
22,400
8,200
12,400
West
13,500
20,200
7,500
11,200
South
1,700
2,100
1,000
1,200
Total
30,000
44,700
16,700
25,000
Notes: d = nominal trench depth
Random stacking indicated by results of excavation of
Northeast Trenches, approximated by 50 percent drums,
50 percent earth by volume in trench below two—foot
cover and assumed trench geometry, as shown by Figure
10.
Drums are assumed to be uncrushed, 55—gallon drums.
38
-------
drum density to yield the total volume of drums. The estimate for the number
of whole drums is provided by dividing the total volume by the volume of a
single drum (7.35 cu ft). As Table 5 shows, the overall range varies by a
factor of two and a half, from 16,700 to 44,700, while the more likely range
based upon the observed depth of the Northeast Trenches is less than a factor
of two, from 25,000 to 44,700.
The above estimates are for whole, uncrushed 55 gallon drums. The numbers
will necessarily increase if some are crushed, enabling closer packing. Crushed
drums, however, are more likely not to contain chemicals in liquid form. The
drum number estimates can be corrected for the presence of crushed drums by
multiplying by g/(f + g — gf), in which f represents the fraction of crushed
drums and g is equal to the ratio of the volume of a whole drum to the volume
of a crushed drum. If g = 2 and f = 0.3, for example, as indicated by the ex-
ploratory excavation of the Northeast Trenches, there would be 18 percent more
drums (whole plus crushed); however, there would be 17 percent fewer whole
drums.
Prior to the radar survey, an estimated range of drums was made which was
substantially lower than the estimates presented here. The earlier analysis
plausibly assumed that the trenches with buried drums were of similar construc-
tion to that of an unfilled trench on the site: nine feet in depth and with
sides of slope 450 As a lesson for other similar sites, it is wise to keep in
mind that without the benefit of more accurate information, the “worst case”
corresponds to a steep sided trench (with angle of repose depending on local
soils as well as the method of trench construction) with depth approximately to
the water table, to bedrock, or to the maximum feasible excavation depth.
39
-------
SECTION 4
EVALUATION OF THE REMOTE SENSING TECHNIQUES
Section 4 presents an evaluation of the remote sensing techniques employed
at the Coventry site, including other potential uses, and the limitations in
each case. The above information is presented in summary form in Table 6.
DETECTION OF SUBSURFACE CONTAMINATION
The use of electrical resistivity as a remote sensing technique to delin—
eate the limits of the contamination plume was extremely appropriate at the
Coventry site. As described in Section 3, natural conditions at the site were
such that the resistivity survey was valuable in locating most of the contami-
nated ground water for the positioning of monitoring wells. A limitation of
the technique, however, is exemplified by the incorrect interpretation of a
relatively deep plume off the northwest corner as being an area of ground
water diversion.
The following conditions are necessary for using resistivity as a remote
sensing technique in a site investigation:
• contrast between the qualities of the contaminated and background
ground water, such that the conductivity of one is very different
from the other
• relatively shallow depth to the contaminated water body below ground
surface, such that conductivity differences within the water body are
not masked by overlying strata
• few changes in geology, such that there are no great variations in
either the thickness or physical characteristics of the unsaturated
zone over the site.
Knowledge of certain man—made interferences, such as buried electrical conduc-
tors or paved areas which might mask resistivity contrasts in an otherwise
suitable area, is also required. Therefore, a basic understanding of the geo-
logic and hydrologic conditions at the site, as well as adverse man—made con-
ditions, is essential to determine whether or not the resistivity method is
applicable.
In addition to determining the extent of contaminated ground water, lat-
eral electricial resistivity profiling may be used to monitor changes in both
the quality and position of the contamination with time. This application is
most useful in situations where it is necessary to monitor the effectiveness
of a particular method chosen to decrease the impact of the contaminant
Preceding page blank 41
-------
Table 6
Comparison of Remote Sensing Techniques
Depth Profiling
ietsric Retraction (Non—
explosive Method)
Purpose
• determine lateral extent of
contaminated ground water
• facilitate placement of mon-
itoring wells end optimize
their number
• monitor changes in plume
position and direction
• indicate change in contamina-
tion with depth
• establish vertical control in
areas of complex stratigraphy
• determine depth and topogra—
phv of bedrock
• determine depth of trench
containing buried drums
• locate areas of high metal
content (e g , buried drums)
• locate buried objects (e g
buried drums)
• provide qualitative infor—
nation regarding drum density
• detect interfaces between
disturbed and uodisturbed
soil (e.g., bottom of tren-
ches)
• detect plumes of high chemi-
cal concentration
Advantages
• procedure less expensive than
drilling
• procedure more rspid than
drilling
• equipment light—weight, able
to be hand tarried
• survey may be conducted in
vegetated areas
• procedure less expensive and
safer than coring or excavation
• procedure more rapid than
coring or excavation
• survey may be conducted in
vegetated areas
• procedure less expensive and
safer then excavation or rsdsr
• procedure more rapid then ex—
csvstion or radar
• equipment light—weight, able
to be hand—carried
• survey may be conducted in
vegetated areas
• procedure less expensive and
safer than excavation
• procedure more rapid than ex-
cavation
• procedure deeper—penetrating
then metal detection
• procedure yields more infor-
mation thsn metal detection
• procedure may be used over
paved sress
• limited ability to detect
non—conductive pollutants
• technique unsuitable if no
sharp contrast between con—
tsminsted and nsturai ground
water
• interpretation difficult if
water table is deep
• interpretation difficult if
lateral variations in stra—
tigrsphy exist
• interpretation difficult if
rsdicsl changes in topogra-
phy are not accounted for in
choice of A—spacing
• technique unsuitable in paved
sress or areas of buried con-
ductive objects
same as above
• technique Lnsuitable if no
sharp velocity contrast se—
tween units of interest
(e p , trench containing
buried drums and surrounding
soil)
• survey requires access road
for vehicle
• depth of penetration varies
with strength of energy
source
• low velocity unit obscured
by overlying high velocity
io U ts
• interpretation difficult in
regions of complex ntracigrs—
phy
• technique unsuitable for the
detection of non—metallic
obj ects
• technique unsuitable for ob-
jects below five feet
• technique unsuitable for de-
termination of number or ar-
rangement of buried objects
• technique unsuitable for
vegetated areas
• data requires sophisticated
interpretation
• underlying objects obscured
by those above
• survey requires access road
for vehicle
Technique
ilecrricai Resistivity
Lateral Profiling
Limitations
name as above
Metal Detection
Ground—Penetrating Padar
42
-------
source(s). In this case, it is not being suggested that resistivity be em-
ployed in place of monitoring wells to detect changes in ground water quality,
but rather it should be used in conjunction with wells. Consequently, the
number of wells required as well as the total cost for the monitoring program
may be reduced by this dual system of detection.
Vertical electrical resistivity profiling is used primarily, as it was at
the Coventry site, to detect changes in contamination with depth below ground
surface at a fixed point. Interpretation of the data is easiest when both the
apparent and cumulative resistivity values are plotted, as shown in Section 3.
The two curves may be used to locate areas of comparatively higher or lower
electrical conductivity, such as where bulk dumping occurred, or to obtain in-
formation regarding vertical changes in stratigraphy at sites having lateral
variations in subsurface structure. In the latter situation, the depth pro-
filing provides the most valuable information when conducted near a test bor-
ing location, so that the resistivity measurements can be interpreted using
the geologic log.
It is possible that ground—penetrating radar may also be employed to de-
tect areas of subsurface contamination. The exploratory excavation of the
Northeast Trenches showed that plume areas previously identified by the radar
survey (blurred areas of Figure 6) were due to nested pockets of liquids.
Contaminated areas can be distinguished by radar from both soil and ground
water because of the greater signal attenuation caused by their higher (in
most cases) dielectric constant.
The use of both of these remote sensing techniques to detect subsurface
contamination is limited to areas where pollutants having a high electrical
conductivity have been dumped, buried, or improperly stored, such that leak-
age causes a distinct contrast with background water quality. Future research
concerning remote methods of detecting subsurface contamination should be di-
rected toward the identification of a means of detecting non—conductive pollu-
tants, such as organic chemicals which tend to be prevalent at abandoned waste
sites.
ELUCIDATION OF BEDROCK TOPOGRAPHY AND CONDITION
The use of seismic refraction as a remote sensing technique applied in
ground water investigations is not new. At the Coventry hazardous waste site,
seismic refraction was employed to determine the depth to bedrock, to identify
channels in the bedrock surface, and to determine the thickness of surficial
fracture zones in the crystalline bedrock. Each of these applications has
been attempted previously in other hydrogeologic studies. At Coventry, the
interpretation of the seismic wave velocities generally showed good correla-
tion with the depth to bedrock as determined by actual rock coring. In addi-
tion, the seismic results showed that the bedrock surface undulated, varying
from approximately 10 to 70 feet below ground surface. However, no indication
of the thickness of surficial fractures in the bedrock surface was possible by
the seismic technique, although this is not considered unusual since fractures
seen in the core samples were numerous but narrow. Only wide fractures would
dissipate the seismic energy sufficiently to be interpreted as “non—competent”
bedrock.
43
-------
There are several advantages, as well as limitations to the use of seis-
mic refraction in determining the bedrock topography. The main advantage,
which relates to the simple non—explosive seismic method employed at Coventry,
is the relatively low cost as compared to coring as a means of locating and
identifying bedrock. Although coring will still be required in order to vali-
date the interpretation of the seismic data, it is probable that the amount of
drilling will be reduced since a seismic survey will allow the selection of
appropriate and possibly less expensive (shallower) drilling locations.
Seismic refraction surveys are generally limited to areas which have a
simple two or three layer structural configuration. Simple stratigraphy aids
in interpretation, as well as reduces the amount of direct subsurface sampling
necessary. Large contrasts in the seismic velocity of each layer are also es-
sential. Additionally, it should be recognized that the seismic refraction
method will not detect a low—velocity layer beneath a high—velocity layer.
This is due to the fact that waves refracted downward by the low—velocity lay-
er will never by detected by the surface receiver before waves are refracted
upward by the high—velocity layer. The depth of penetration of the seismic
method is also limited by the strength of the energy source. However, for
shallow (up to 100 feet) investigations similar to the Coventry site the ham-
mer drop method is adequate.
Ground—penetrating radar can also be used to determine the location of
the soil—bedrock interface, although this application was not attempted at the
Coventry site. In areas of low soil conductivity, the depth range of the
electromagnetic impulse can be as great as 30 meters (approximately 100 feet).
However, soils having high electrical conductivity inhibit the transmission
of the signal, thereby resulting in a decrease in the depth of penetration.
Therefore, radar is preferable to seismic refraction only in situations where
the depth to bedrock is very shallow (less than 30 feet) and a large survey
area must be covered in a short period of time.
DETERMINATION OF SUBSURFACE TRENCH LIMITS
The feasibility of using seismic refraction and ground—penetrating radar
to locate the side and bottom limits of the trenches was tested at the Coven-
try site. Results showed the following:
• radar could be used to locate the side boundary between disturbed and
undisturbed soil
o radar was not successful in locating the base of the trenches, which
tended to be obscured by the buried drums and by the relatively high
moisture content of the overlying material (resulting in a high di-
electric medium)
• seismic refraction was only moderately successful in locating the
side boundaries of buried drum areas. Differences in velocities were
noted between what may be interpreted as inside versus outside the
trench, but actual limits to the units cannot be determined without
direct subsurface sampling
44
-------
• an interface that may be interpreted as the base of the trench was
located by the seismic refraction method, although additional test
borings or confirmation by excavation is necessary for validating the
results.
Seismic results from this study or from a controlled experiment must be
calibrated with test boring logs to determine if a trench/undisturbed soil in-
terface can be located by the refraction method. The radar technique appears
useful in locating side boundaries, but not in the definition of the bottom
primarily due to the high attenuation of signal by the trench material (drums,
moisture and contaminants). Both techniques are limited by the fact that the
results from each require skilled interpretation.
DETECTION OF BURIED DRUMS
Metal detection and ground—penetrating radar were used at the Coventry
site to determine the outlines of trenches containing buried drums, and to the
extent feasible, provide additional information as to the distribution of the
drums. Because the radar has the ability to penetrate deeper than the metal
detector, it was assumed that the radar would define the trench outline more
accurately. However, a metal detector can easily and inexpensively locate the
approximate trench outline using a widely—spaced grid, and then other tech-
niques can be used to focus on the outlined areas. Additionally, the infor-
mation desired from the metal detection survey is obtained while in the field
and does not require extensive interpretation.
Ground—penetrating radar is a feasible method to use to refine boundaries
of the trench by providing information on the distribution of buried drums.
The technique is not without limitations, however, nor can it be considered
the only means available to provide this information. Magnetometry has also
been considered for this same purpose, but radar was the technique selected
for the Coventry study. Once rough trench outlines have been determined by
the metal detector, a magnetometer survey of these approximate areas could be
conducted with regular grid spacing (on the order of five feet) to identify
the changing signal character at the trench boundaries. Bensen and Glaccum
(1980) have shown magnetometry to be successful in identifying the approximate
spatial distribution, depth to the “center of mass,” and quantity of drums
present, in a survey over a linear pile of buried drums about two meters deep.
Other remote sensing methods that have been considered to obtain informa-
tion on the distribution of drums include various acoustic mapping techniques
or “geotomography” (geophysical tomographic techniques) (Eddy and Guttrich,
1979). For maximum sensitivity using the acoustic technique, an acoustic wave-
length on the order of 1 to 2 feet (approximately 10 Khz) is required, which is
unfortunately within the audible range. The maximum depth of penetration is
normally on the order of 50 wavelengths (in this case 50 to 100 feet), which
would usually be well in excess of what is required at an abandoned dump site.
Depending on drum content (high loss if empty, moderate loss for liquids and
low—density powders, least loss for compacted soils), acoustic signals would
penetrate through drums at shallow depths, thereby permitting detection of drums
at lower depths than with radar. However, multiple reflections could generate
display artifacts that might be misinterpreted as drums at excessively great
45
-------
depths. It is also possible that large boulders could be confused with drums,
but as with radar, there may be several methods for distinguishing between
the two either in the field or from the data.
Geotomography was also considered by MITRE for defining drum distribution
by using cross—borehole probing with either or both acoustic or electromagnetic
signals. These procedures would yield substantially higher assurance that all
drums had been located, but would involve much higher cost than either radar
or acoustic methods for drilling and subsequent computer processing.
Despite the fact that the acoustic technique may be able to view drums
buried beneath other drums (if drums and soil were tightly packed) better than
the radar method, the latter was selected due to the relative ease of perform-
ing the survey. Acoustic mapping techniques require considerably more site
preparation since the transducer must be physically coupled to the ground.
Although the survey area also should be relatively free of vegetation for a
radar survey, the operational problems are much less than for acoustic tech-
niques. Geotomography was considered prohibitively expensive and a magnetom-
eter survey ineffective in determining the information necessary.
As shown previously in Figure 5, the radar technique was able to identify
individual buried drums in the trenches. The radar data showed, to an approx-
imate depth of 12 feet, that most of the drums were in a random distribution
rather than arranged in orderly stacks. It is suggested that information ob-
tained about drum distribution at the Coventry site be validated by conducting
trench excavation in such a manne: that data are collected in a rigorous and
systematic manner.
The main limitation regarding the use of radar for the detection of bur-
ied drums is the requirement for skilled interpretation of the resulting data.
As can be seen from Figures 4, 5, and 6 the radar—produced signals are complex
and it is difficult to distinguish between reflective objects, such as boul-
ders or drums. Blurriness of the signal is produced from high dielectric con-
stant media, such as ground water or leaking chemicals, thus possibly causing
their obliteration (which may be enhanced by computer processing). A signif-
icant obstacle to data interpretation arises from the relative inexperience
inherent in the use of ground—penetrating radar for buried drum determination.
There presently does not exist an adequate data base from various sites to
provide signatures of drums in various configurations and arrangements which
can be said to be known or proven. Another drawback is that the data inter-
pretation is usually performed out of the field resulting in a delay between
the actual survey work and the use to which it is put. Because interpretation
is so difficult, research on the “signatures t ’ from known buried drums would
be very beneficial. This type of scientific experiment was not feasible at
the Coventry site because this was not the intent of the DEN when it conducted
the exploratory excavation.
46
-------
SECTION 5
RECOMMENDATIONS
Section 5 presents recommendations regarding the use of remote ensing
techniques in abandoned site investigations and recommendations for research
needs.
SYSTEMATIC APPROACH FOR ABANDONED SITE INVESTIGATIONS
Abandoned hazardous waste sites present varying degrees of difficulty to
investigators. For example, abandoned sites which are areally extensive, ru-
ral (with hindering vegetation), or in areas of complex geology and hydrology,
represent troublesome environments for investigation. Therefore it is impor-
tant to develop approaches for thorough, but rapid and cost—effective assess-
ments of these difficult situations. In most cases, a well designed and exe-
cuted investigative program will include remote sensing techniques in addition
to direct measurement. Premature action to drill wells, collect and analyze
various air, water, and soil samples, or perform excavations without careful
planning and proper integration of available techniques may result in unneces-
sary exposure to hazardous conditions and in an inaccurate or incomplete
understanding of the total problem. - - - -
Remote sensing techniques may be used to provide reasonably accurate as-
sessments of subsurface contamination, the location and extent of buried
drums, and other data needs for determining appropriate methods of abatement.
It must be stressed, however, that not all critical information can be ob-
tained remotely, since each of the techniques has limitations, both theoreti-
cal and site—specific, and consequently, direct sampling should be under-
taken at every uncontrolled hazardous waste site.
To accomplish site investigations in the most efficient manner, a system-
atic approach is necessary to take advantage of the information that can be
extracted from remote sensing methods. In addition, a systematic approach
allows a reduction in the time and cost, and an increase in the effectiveness
of direct sampling.
In general, the following two objectives must be addressed by all inves-
tigations at uncontrolled hazardous waste sites:
• determination of the nature and extent of the problem and the result-
ing effects on public health (both actual and potential)
o determination of environmentally sound and cost—effective methods to
effectively abate the problem (if abatement is deemed necessary).
47
-------
The first activity of any investigation should be the identification of the
specific data needed to meet each objective After this has been accomplished,
the various techniques available for data acquisition, both remote and direct.
can be evaluated with regard to the type of inforrr’ation that can be obtained
from each in relation to the specific conci cions at the site. Although not
always the case, it may be reasonaoly assu-red that remote sensing techniques
should be used in advance of the n ore d:caL data acquisition methods of bor-
ings, or excavations. This is not intenOc’i to imply, however, that all direct
sampling should be held in abeyance. There have been numerous instances in
which emergency action is dependent upon ir mediate results from air, water,
and soil sampling, and for such cases remote sensing techniques should be used
secondarily.
Table 6 of Section 4 summarizes the purpose, advantages, and limitations
of each of the four remote sensing methods used at the Coventry site. It is
important that this type of information be consulted prior to development of
an investigatory program. Even though there are disadvantages inherent to
each technique, proper sequencing and phased studies can potentially result in
an overall optimized approach. It must be emphasized that as the study pro-
gresses, preliminary conclusions will necessarily be modifiec and the nature
of direct sampling activities will need to be evaluated continuously. It is
recommended that final conclusions not be drawn solely from the results of re—
mote sensing methods. Direct sample collection should be undertaken for all
studies.
Because no single procedure would be appropriate for all abandoned haz-
ardous waste site investigations, the conditions at the Coventry site will be
used for the basis of development of systematic procedures with the expecta-
tion that some of the concepts can be applied elsewhere, as appropriate. The
sequence of activities shown below represent the idealized case and may not
have necessarily been followed in actual practice.* This is due to the fact
that the exact capabilities and limitations of some of the remote sensing
methods (in particular ground—penetrating radar and seismic refraction for
determining the lower boundary of buried drums) were unknown at the outset.
The Coventry site indeed provided a valuable proving ground for the mix of
the remote sensing techniques.
Objective: Determination of Nature and Extent of Problem
The following conditions were posed by the nature of the Coventry site
at the outset of the investigation:
• an unknown number of drums buried on a cleared seven—acre site: loca-
tion of drum burial areas incompletely known
• an unknown quantity of unknown chemicals bulk—discharged into trenches
on the site
*The exact investigatory procedures are presented in the Phase I and Phase II
reports referenced in Section 1.
48
-------
• surface discharge of chemicals at a swamp, 1200 feet away and 50 feet
lower than the site
o potential for ground water to flow away from the site over an arc of
approximately 2700, two thirds of which is moderately vegetated, not
including the swamp
o subsurface material consisting of hard, boulder—studded glacial till
amid numerous bedrock outcrops.
The following items had to be determined for a comprehensive understand—
ing of the nature and extent of the problem:
• direction, rate, and extent of subsurface migration of contaminants
o location of surface discharge areas of contaminated ground water and
the subsequent fate of the contaminants
o identification of most harmful contaminants
o location of areas of contaminated soil and buried drums and determina-
tion of the potential of this source of pollutants for long—term re-
lease.
An example of a systematic approach to achieve the above objectives is
presented in Table 7 and the phasing of site activities is shown by Figure 18.
It should be noted that Figure 18 shows a two—phase monitoring well installa-
tion and sampling program. The purpose of a phased investigation is to obtain
a preliminary understanding of the problem prior to final planning of all di-
rect sampling in order to more effectively guide subsequent activities. Addi-
tionally, this figure shows that, in general, remote sensing preceeds direct
sampling in order to reduce the time and cost of the latter and to help ensure
that the full extent of the situation is identified.
In the case of buried drums, it is suggested that metal detection, rather
than ground—penetrating radar, be used to locate the burial areas because of
the relatively lower cost and greater portability of the former technique.
However, radar has greater penetration and should be used in all areas where
drums are suspected, but not found by metal detection. A limited excavation
may then be required to gain information about the depth, condition, and con-
tents of the drums so that the concept of total drum excavation and chemical
disposal can be evaluated with the other abatement alternatives. Limited ex-
cavation was considered feasible for the Coventry site, since total drum ex-
cavation did not appear significantly more expensive than other abatement al-
ternatives.
Surface water, ground water, and soil sampling are necessary for any
hazardous waste site investigation. Monitoring wells are best located once
the extent and direction of the plume have been determined. It is recommended
that monitoring wells be placed at the following locations:
• upgradient from the source of contamination (to monitor background
conditions)
49
-------
Table 7
Systematic Approach to Determine Nature and Extent of Problem at Coventry Site
Direction, Rate,
and Extent of
Subsurface Con-
taminants
S
determine steal extent
of contaminated ground
water
• assess hydrogeologic and geo-
logic settings of the site by
reconnaissance and study of
topographic maps, serial pho-
tographs, aiid all existing
data
• conduct vertical resistivity
profiles in various locations
to determine approximate
depths of wuitamination
• conduct lateral resistivity
survey choosing A—spacing
based on result’, of vertical
resistivity profile
• install monitoring wells in-
side and outside of contauuu—
nated zones as defined by
resistivity survey for con—
fi nnat toni I purposes
S
establish permanent field
grid for usc hy all remotc
sensing surveys, grid loca-
tions can be trnnsfcrred to
site map by aerial pliotogra—
phy or land survey
• choose wide enough A-spacing
for lateral survey to mini-
mize influence of varintions
of anticipated depth to
ground water from ground
surface
• determine vertical ex-
tent of contaminated
ground water
• conduct scisinuc refraction
survey ovcr contaminated
area to dLteumioc depths to
bedrock and vcrtical subsur-
face profuics
• install cluster wells con-
sisting of horings screened
in soil and u i bedrock both
inside and ooi side of con-
taminated ?OiiLS
• use established field grid
far seismic survey
• use results of seismic re-
fraction survey to select
most economic locations for
bedrock borings
• determine whether vertical
gradients exist and whether
contaminants are present in
bedrock fractures
• install bedrock wclia as a
seond—phase drilling effort
and install any additionally—
needed monitoriuug wells based
upon f i rs t—piuise resii Its
Objective Data Needs luuvestul,-itory Methods Comments
La
0
-------
Table 7 (continued)
2. Location of Sur-
face Discharge
Areas of Ground
Water Contamina-
tion and Deter-
mination of Fate
of Pollutants
3. Chemical Identi—
ficatiori of all
Principal Contam—
I nants
Data Needs
• determine direction and
rate of subsurface mi-
gration
• locate surface discharge
areas
• determine fate of pollu-
tants
• determine principal con-
taminants
• determine dispersion of
contaminants
invcsLlgatory Methods
• instal I minimum number of
monitoring wells to define
ground water flow, well lo-
cations should be hased on
results of resistivity survey
and chemical analytical needs
(see third objective in this
table)
• perform in—situ permeability
tests In selected monitoring
and hedroik wells
• reconnoiter site, locate dis-
charge irtas by sight and air
quality liLasuring devices
• use results of electrical re-
sistivity and water table map
to identify surface receptors
of subso , face discharge
• conduct downstiesm sampling,
including rate of flow of
surface water
• sample sediment and air around
discharge area
• analyze cuspos ite samp I us
from selected wells and sur-
face waters for priority
pollutants
• analyze icctud wells and
surface waters for selecred
compounds based upon priori iv
1)011 utant ,rnalvsis
• select Indirator analyses
based rr on hr ion ty pu I I rrt,riit
screening include general
water qii ii i ry tests (gIl, ton—
duct iv I ty , iron, cli br nit
lO t)
Comments
• because the ground surface
contours at Coventry site
indicate a potentially large
arc of subsurface travel,
wells at edge of contaminated
zones were necessary to de-
tail direction of outer fringe
of pollutants
• install cluster well uppradi—
cot of pollutant source to
determine background condi-
i tins
• use air quality measuring
devices to determine areas of
poor quality, indicating sur-
face discharge points also
indicates areas requiring
breathing protection devices
• mass balances should be cal-
culated for principal pollu-
tants to determine ultimate
dispersion mechanism(s)
• upstream samplinp necessary
for background conditions
• all water wells and surface
waters used for notable water
suoplies within a mile radius
sliotil d be precaritionari ly
sampled
• composites made op of wells
close to source of poi lutants
and surface discharge areas
should limit each composite
to unIv two adlaccnt wells
• m u i cator analysis provides
relatively low cost method
for riori—speci fic monitoring
ol pul I utant levels In moni
turing will network
Objective
Ln
I - .
-------
Table 7 (concluded)
Objective
Data Needs
Investigatory Methods
Comments
La
f.J
4. Location of Areas
of Buried Drums
and Contaminated
Soil and Deter-
mination of Life-
time of Future
Release of Chemi—
ca is
• locate areas of buried
drums and contaminated
soil
• reconnoiter site, search for
areas of disturbed soil or
vegetation, or areas of dis-
colored soil
• interview persons involved
with dumpiny activities for
information toncerning loca-
tion of trenches and method
of operations
• study prcvioiisiy taken aerial
photographs to obtain histor-
ic information
• use re m us of resistivity
survey to locate source of
contaminattd ground water
• conduct metal detection stir—
vey over all cleared or dim
turbed aruas of site
• conduct ground—penetrating
radar murvcy in burial areas
located by octal detection
survey
• use established field grid
for metal detection and ground—
penetrating radar survey
• metal detection used in ad-
vance of ground—penetrating
radar because of lower cost
and ease of use, however,
radar has greater penetration
and should be used in all
areas where drums are sus-
pected, but not found by
metal detection
• determine lifetime of
future release of chemi-
cals
• conduct stisniic refraction
survey ovur drum burial area
to determine depth of drums
• perform list tud excavation
of buried drums to determine
condition and contents of
drums, density of drums, lower
boundary of drums
• sample soil to drum burial
areas anti bulk chemical dis-
charge at ci analyze est racted
leacl iate or spec! ftc cliemicat
compounds durermined by pri-
ority polltaunt screening
• effectiveness of seismic re-
fraction method to determine
lower boundary of buried
drums remains subject to
yen ficatiun
• drum excavation limited to
data gathering only and shoutd
be tcrminated when sufficient
information obtained, refer to
Phase I I project report (ref-
erenced in Section 1) for ret—
onmicndations regarding excava-
tion procedures
• drum escavat Ion should be con-
ducted after all other site
activities completed ia ordur
to mininize personoel on site
Note’ Coventry site procedures used for illustrative purposes only, investigatory procedures and sentience may not
necessarily be directly applicable to other sites
-------
tal Detection J Groond—Penetrating
Survey “ Radar Survey
— location of buriad
drums
- determination of
horizontal boundaries
of buried drums
— location of deep
buried drums
‘iote Coventry site procedures used for illustrative purposes only, investigatory procedures and sequence nan not
necessarily be directly applicable to ocher sites
Preceding page blank
Figure 18. Recommended Sequence
— preliminary determination
of water table
— determination of aouifer
charscterlstics
— site reconnaissance
— review of prior
reports and chemical
an nlvs is
- intervier. or persona
associated with dump-
ing
— study of topographic
naps and aerial photo-
graphs
— establishment of
permanent field
grid
- production of
photogrammet nc
nap or site and
surrounding re-
gion
Source of
cOota’ninat ion
lower houndar or
drums and prelini—
nan estimate of
number of drums
i i ’
54
-------
of Activities at Coventry Site
- determination of contents.
depth, and density of
buried drans
— determination of soil
Contoninat ion
— determination of final
eotinate or nanber of
buried drurm
of bedrock
contamination and trana—
miss ieitv
— deterhnation of vertical
eradients
— determination or specific
contaminants and appropri-
ate indicator ten-s
— preliminary determination
of contaminated zones
— final determination of
site hydrogeology
— final determination of
site contamination and
effect on pubiic health
55
-------
o outside of the plume downgradient from the source of contamination (to
verify the extent of the plume and to mon or its movement)
• within the plume close to the source of contamination (to obtain sam-
ples before extensive dispersion, dilution, or attenuation)
• within the plume at the outer extent of contdmanation (to observe dis
persion, dilution, and attenuation).
It is important to define the death and topography of the bedrock not
only to evaluate certain abatement methods, but also to economically locate
the deep borings that may be needed for bedrock sa-ipling, permeability
testing, and water sampling. Seismic refraction is a very effective tool for
providing remote information on the configuration of subsurface strata. Addi-
tionally, the results can be used to locate bedrock wells, the position of
which will depend upon the objectives of the investigation. For example, some
investigators may wish to install bedrock wells only to determine formation
integrity as an acceptable base for certain abatement alternatives. Other in-
vestigators may be interested in locating bedrock wells to determine the pres-
ence of contaminants in specific regions or channels.
Objective: Determination of Methods to Abate the Problem
As shown in Table 8, the data needs for selecting and determining the
cost of abatement alternatives are similar to those for understanding the na-
ture and extent of the problem at the Coventry site. However, the two objec-
tives are best addressed separately, since the locations for direct sampling
and remote sensing may differ between the two, as may the use of the informa-
tion obtained.
It is possible to use remote sensing techniques as a “negative screening”
step in the evaluation of certain abatement options. An example of this con-
cept is given in the case of evaluating source encapsulation as an abatement
technique for the problem at Coventry. For encapsulation to be feasible, it
is necessary to have a low permeability base within a relatively shallow depth
from the ground surface. Seismic refraction has the potential to determine
whether the bedrock underlying the source of contamination should be ruled out
(negative screen) of consideration as an acceptable base. If, after a seismic
refraction survey, it is found that the bedrock is either too deep or too
fractured to function as an effective base, then rock coring is unnecessary.
The information obtained by this method is not sufficient, however, to prove
that the bedrock is sufficiently sound for encapsulation without actual test
borings and field—permeability tests.
There is one investigational method, estimation of the number of buried
drums, listed on Table 7 which is not included in the discussion of the pre-
ceding subsection. The remote sensing technique of seismic refraction has the
potential to determine the lower boundary of buried drums. The results of an
exploratory excavation can be compared to the seismic profile obtained at a
particular location and, if valid, the seismic profiles for all other buried
drum areas may be used for the drum number estimates. If there is no correla-
tion found between the seismic profile and the excavation (or if an exploratory
56
-------
Table 8
Major Informational Needs for Implementation of
Certain Abatement Activities at Coventry, Rhode Island
Alternative Informational Needs
Removal of Buried Drums and • Drum Condition
Disposal of Chemicals
• Drum Number
• Drum Contents
• Trench Location and Geometry
Encapsulation of Source • Drum Contents
o Imperviousness of Underlying
Strata
• Level(s) of Contamination
(soil and/or ground water and/
or bedrock)
o Trench Location and Geometry
Collection and Treatment of • Areal Extent of Contamination
Leachate (trenches and/or
• Type of Contamination
wells)
• Concentration of Contaminants
• Imperviousness of Underlying
Strata
• Aquifer Characteristics
No Action Alternative • Drum Contents
• Drum Condition
• Level(s) of Contamination
(soil and/or ground water and/
or bedrock)
• Type of Contamination
57
-------
excavation is not desired) it is recommended that the maximum feasible excava-
tion depth or depth to bedrock be used for the lower boundary of the buried
drums.
RESEARCH NEEDS
As described in the evaluation of the remote sensing techniques in Sec-
tion 4, each technique has a number of limitations regarding its application
at an abandoned hazardous waste site. Several of these limitations define
directions for future research, although many of the following recommendations
simply involve use of the technique in controlled, research—oriented experi-
mental situations. Recommendations regarding future research into direct
sampling methods is beyond the scope of this study. Table 9 presents a summary
of the following recommended research needs.
Ground—Penetrating Radar
It is recommended that future research into the use of ground—penetrating
radar at abandoned hazardous waste sites be primarily directed toward compiling
radar signatures of known buried drum configurations at known depths. It is
suggested that this research involve a combination of computer simulation and
actual field testing of drums (empty of hazardous chemicals) which were inten-
tionally buried. Additionally, radar signatures should be determined for the
following:
• drums buried in average mixed municipal refuse landfills
• drums in good condition versus drums in poor condition
• drums dumped underwater in marshy areas
• drums buried in rubble.
Changes in antenna design would be required in the latter two investigations,
as use of the present equipment depends upon relatively even and stable
vegetation—free surfaces. It is well known that landfills and wetlands are
common locations for the illegal burial of drums, and it would be extremely
useful to accumulate, as a data base, the signatures arising from these situa-
tions.
In addition to signature research, it is recommended that the usefulness
of radar be investigated at secure landfills, in order to determine its ef-
fectiveness in determining subsurface conditions; this would be useful to
regulatory agencies to verify that proper operations are taking place. It may
also be found that the landfill subsurface conditions could be monitored over
time using radar techniques. It is also suggested that the applicability of
radar to monitoring the stability of physical remedial structures at abandoned
hazardous waste sites be investigated.
Research into the area of signal enhancement would be useful for hazard-
ous waste site investigations because of the attenuation of the signal caused
by soil moisture. This would enable deeper penetration in moist soils (like
58
-------
Table 9
Summary of Recommended Research Needs for Remote
Sensing Methods for Hazardous Waste Site Investigations
Research Need
Ground—Penetrating Radar
Use
1. Determine signatures of drums buried
under known conditions:
• in soil
• in municipal refuse landfills
• in rubble
• underwater
• in poor vs. good condition
2. Determine the effectiveness of radar to
monitor subsurface conditions at aban-
doned sites at which remedial measures
have been undertaken and at secure chemi-
cal waste landfills.
3. Develop (or improve) methods of enhance-
ment of reflected signal.
Seismic Refraction
4. Investigate under controlled experimental
conditions whether the technique is ef-
fective for determining the lower boundary
of buried durms.
Compile data base to aid
in future interpretation.
Observe changes in physical
integrity of subsurface
structures at secured
sites.
Deeper penetration in
moist soils and easier in-
terpretation of complex
subsurface problems.
Determine the lower bound-
ary of buried drums to
estimate total number of
buried drums.
5. Determine the
refraction to
structures of
cure chemical
effectiveness of seismic
determine the subsurface
refuse landfills and Se—
waste landfills.
Observe changes in physical
integrity of subsurface
structures at secured
sites.
Electrical Resistivity
6. Determine the effectiveness of electrical
resistivity to monitor subsurface condi-
tions at abandoned sites at which re-
medial measures have been undertaken and
at secure chemical waste landfills.
7. Determine the effectiveness of electrical
resistivity to locate and monitor sub-
surface spills of nonconductive chemi-
cals.
Observe changes in physical
integrity of subsurface
structures at secured
sites.
Register the presence of
petroleum leaks into an
aquifer.
59
-------
those in New England) and may have added benefits related to interpretation
of complex situations.
Seismic Refraction
A comprehensive investigation into the usefulness of seismic refraction
as a tool to locate subsurface interfaces, such as trench bottoms, at aban-
doned hazardous waste sites should be made. In addition to expanding the use
at hazardous waste sites, experiments with the technique should be conducted
at municipal refuse landfills or secure landfills to determine the feasibility
of locating waste depths or cell structures or of monitoring the integrity of
engineered remedial actions.
Electrical Resistivity and Metal Detection
The usefulness of electrical resistivity and metal detection in specific
applications is widely recognized, and increased use in defining subsurface
contamination and locating buried metal objects at abandoned hazardous waste
sites can be expected. Additional applications for electrical resistivity
should be investigated, such as determining the ability of the technique to
monitor pollution from secure landfills or the effectiveness of in—site abate-
ment measures. It is also recommended that improvements be made in the detect-
ability of each technique, specifically in the depth of penetration for metal
detection and the use of electrical resistivity where the resistivity of the
polluted water is less than that of the background water.
REFERENCES
1. Bensen, R. C., and R. A. Glaccum. 1980. Site Assessment: Improving
Confidence Levels with Surface Remote Sensing. Proc. Mgmt. of Uncon-
trolled Hazardous Waste Sites , pp. 59—65.
2. Campbell, K. J., and A. S. Orange. 1974. A Continuous Profile of Sea
Ice and Freshwater Ice Thickness by Impulse Radar. Polar Record , Vol. 17,
No. 106, pp. 31—41.
3. Dobrin, M. B. 1960. Introduction to Geophysical Prospecting . McGraw—
Hill, New York, 446 pp.
4. Eddy, F. N., and C. L. Guttrich. 1979. Personal Communication.
5. Freeze, R. A., and J. A. Cherry. 1979. Groundwater . Prentice—Hall, Inc.,
Englewood Cliffs, New Jersey, 604 pp.
6. Morey, R. M., and W. S. Harrington, Jr. 1972. Feasibility Study of
Electromagnetic Subsurface Profiling . EPA—R2—72—082.
7. U.S. Environmental Protection Agency. 1978. Electrical Resistivity
Evaluations at Solid Waste Disposal Facilities , SW—729.
60
-------
APPENDIX A
SUMMARY OF CONCLUSIONS, RECOMMENDED ACTIONS, AND
COMPARISON OF ABATEMENT ALTERNATIVES: PHASE I
61
-------
Phase I Investigation: Summary of Conclusions
Chemical Contamination
• Ground water and surface water are contaminated with
chlorinated and non—chlorinated organic chemicals (total
concentration less than 100 ppm).
• Air quality near the swamp is degraded due to release of
chlorinated and non—chlorinated volatile organic che-nicals.
• Soil around the site is contaminated with phthalate esters
(total concentration less than 20 ppm).
Health Effects
o The chemicals detected are potentially hazardous, but the
principal potential threat to health appears to result
from poor air quality in certain sections of the swamp.
• The rural nature of the affected area reduces the threat
to public health, unless Whitford Pond is contaminated.
Hydrology and Buried Drums
• A bedrock mound located off the northwest corner of the
site diverts leachate into two primary plumes; however,
both plumes discharge to the swamp.
• The quantity of contaminated ground water flowing away
from the site is less than 260,000 gal/day.
• Drums are buried in two major trenches, one along the
western and one along the northern boundary of the site.
• The best estimated range of the total number of drums
buried is 3,500 to 9,000.
Abatement Options
• The estimated costs of long—term abatement alternatives
range from $750,000 to $2,970,000 (February 1980 dollars),
but may go significantly higher if specific chemicals such
as PCBs are found in sizable quantities.
• Information available at the present time is not sufficient
to accurately determine the most cost—effective solution.
63 Preceding page blank
-------
Phase I Investigation: Recommended Actions
- Activity Purpose Time Frame (1980 )
I. Post contaminated areas of Alert trespassers to threat to April
dunp and swamp public health
2. Analyze quantity and qual— Determine potential threat to April
ity of influent to Whitford public health
Pond
3. Analyze residential wells Determine potential threat to April
(within 1 mile radius of public health
site) for volatile organ—
ics
4. Sample air quality around Determine nature of hazard Periodically
swamp
5. Evaluate need to restrict Determine if nature of potential Periodically
access to contaminated hazard justifies cost of fencing
area of swamp contaminated regions of swamp
6. Excavate and dispose of Confirm continued existence of April — August
drums in northeast trench source of chemicals (including pro—
(backfilling with aerated curement)
soil)
7. Install absorbent booms Limit potential of surface April — May (in—
and sheets at several lo— pollutant flow to t’Thitford Pond stallation only)
cations and evaluate their
effectiveness
8. Examine bedrock for pres— Assist in the design of long— April — June
ence of fractures and term abatement measures
contamination
9. Install additional wells Define plume boundaries and April — June
investigate swamp underf low
of contaminants
10. Sample existing wells Monitor changes in water quality Periodically
11. Analyze condition of Determine potential threat to April — June
Whitford Pond (aquatic public health
life, surface water, and
sediment)
12. Determine all uses of Determine potential threat to April — May
Pond water in public health
addition to cranberry ir-
rigation
13. Conduct detailed evalua— Abate pollutants in a cost— July — September
tion of long—term abate— effective manner
ment approach and imple-
mentation plan for pre—
— —ferred approach
64
-------
Comparison of Abatement Methods at the Conclusion of Phase I
• effective if source of
contamination is ex—
boos ted
• effective if total mass
of contaminants volatil-
izes in swamp, the swas,
remains isolated, and
Wbitford Pond is unaf-
fected
• does not resooc source
of pollutants
• potential for future
release of pollutants
still exists
• uncontrolled release of
pollutants may cause
public health problems
• condition of source
(drums)
• state of nearby pond
• contoainsnt uoderf low
at swamp
• ultimate disposition of
all pollutants
• radar, esploracory cx—
cavat too
• additional wells, clam-
ical analy’ ls of soils
and water samples
2. Drum Removal and DIs-
posal (excavation,
testing, and proper
disposal of drums sod
contents, and contam-
inated soils)
• remove source of pollu-
tants
• ineffective if drums are
ruptured and chemicals
dispersed
• potential for Injury to
workers exists
• condition of source
(drums)
• condition of soil
• radar, exploratory es—
cavation
• eaploretory escovation,
chemical analysis of
soil samples
a’
U I
3 Site Encapsulation
(construction of im-
permeable barriers
around source of
pol tutanta)
• steps/controls pollution
at source
• woikitig conditions baler
than for drum removal
• does not renuvc source
of pollutants
• potential tor loture re-
lease of pollutants
still exists
• success of coetalitment
requires absence of
fractures in bedrock
surface
• condition of source
(drums)
• condition of bedrock
• radar, emploratory ex-
cavation
• seismic refraction,
core drilling, deep
wells
• requires periodic and
perpetual monitoring and
maintenance
4 Leachnte Collection
and Treatment
a LImited Option (in-
terceptor trenches
constructed adja-
cent to site walls)
• controls pollution at
source
• working conditions safer
than for drimi removal
• does not remove source
of pollutants
• success of coilsctlnn
depends on condition of
bedrock
• treotment system does
not removc all contami-
nants from leach te
• condition of source
(drums)
• condition of bedrock
• radar, exploratory cx—
cavat ion
• seismic refraction,
core drilling, dcep
wells
• unknown and poteotially
large life—cyclc tost
b More Complete Op-
tion (intereepter
trenches conat ructed
600 ft downgradient
of site walls)
• controls pollution at
source including addi-
tional downgradient
contaminated soil
• working conditions
aafer than for drum
removal
• same as obovc
• sane ax above
• Same ax above
1. No Action
Additiooai Information
Aicernative Key Advantsgea Key Diaodvatitigc Requircd to implement
Alternative
Technique to
Obtain lnfonnatioo
-------
APPENDIX B
SUMMARY EVALUATION OF LONG—TERN ABATEMENT OPTIONS: PHASE II
Preceding page bbnk
67
-------
SUMMARY EVALUATION OF LONG—TERM ABATEMENT OPTIONS
Long—Term Optiona Summary Evaluation
Option 1: Encapsulation Not Recommended
• significant source of chemicals in
liquid state (perpetual threat for
environmental release)
• deep bedrock (high cost)
• f-ractured bedrock (too permeable
for secure base)
Option 2: Interceptor Trenches Not Recommended
• deep bedrock (high cost)
• irregular bedrock surface (high)
cost)
o fractured bedrock (too permeable
for secure base)
Option 3: Drum and Chemical Removal Recommended
• source of contaminants removed
— with continued monitoring of
plume and swamp • dispersion of contaminants in
ground water monitored
No Action Alternative Not Recommended
• significant source of contamina-
tion (potential for long—term
continuous release)
• swamp not proved to be treatment
mechanism (potential for spread
of contaminants resulting in
human contact)
a. Cost of each option is different than shown in Phase I report although
titles of options are the same.
Preceding page blank 69
------- |