United States
Environmental Protection
Agency
Office of
Solid Waste and
Emergency Response
Publication 9355.4-13
EPA 540R-93-073
PB93-963343
September 1993
Superfund
&EPA
Evaluation of the Likelihood of
DNAPL Presence at NPL Sites
National Results
Word-searchable Version Not a true copy
-------
9355.4-13
EPA 540-R-93-073
PB93-963343
September 1993
EVALUATION OF THE LIKELIHOOD OF DNAPL PRESENCE
AT NPL SITES
National Results
FINAL REPORT
Office of Emergency and Remedial Response
Hazardous Site Control Division
401 M Street S. W.
Washington, B.C. 20460
Word-searchable Version Not a true copy
-------
NOTICE
The conclusions outlined in this document are intended solely for technical support to EPA personnel.
They are not intended, not can they be relied upon, to create any rights, substantive or procedural,
enforceable by any party in litigation with the United States. The Agency reserves the right to act at
variance with these policies and procedures and to change them at any time without public notice.
For additional copies of this report please contact:
National Technical Information Service (NTIS)
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
(703) 487-4650
Word-searchable Version Not a true copy
-------
CONTENTS
Page
EXECUTIVE SUMMARY viii
Chapter 1: INTRODUCTION 1
1.1 Background 2
1.2 Potential Scope of the DNAPL Problem in Superfund 6
1.3 Overall Study Strategy 8
Chapter 2: DATA COLLECTION AND MANAGEMENT 9
2.1 Data Needs 9
2.2 DNAPL Survey Response 11
Chapter 3: ANALYSIS OF DNAPL OCCURRENCE 13
3.1 Site History Ranking 13
3.2 Ground Water Contamination Ranking 24
3.3 Composite Site Ranking 3 7
3.4 Effect of Hydrogeologic Setting on DNAPL Occurrence 43
3.5 Relationship of Site Use to DNAPL Occurrence 47
3.6 Site Contaminant Type and DNAPL Occurrence 53
References
Appendix A DNAPL Site Assessment Survey Form
Appendix B Estimating the Potential for DNAPL Occurrence at
Superfund Sites
iii
Word-searchable Version Not a true copy
-------
LIST OF FIGURES
Figure 1 -1. Schematic cross section of a site at which TCE has been discharged to
a waste lagoon over a 20 year period as part of an aqueous solution 4
Figure 1-2. Schematic cross section of a site at which TCE has been discharged to
a waste lagoon over a 20 year period as a separate organic liquid 5
Figure 3-1. Distribution of site history rankings for the 40 known DNAPL sites and the
270 sites at which DNAPL probability must be estimated 22
Figure 3-2. Distribution of the contaminants found most frequently at the highest
concentrations (as a percentage of their pure-phase solubility) in ground water 34
Figure 3 -3. Distribution of ground water contamination rankings for the 40 known
DNAPL sites and for the 270 sites at which DNAPL probability must be
estimated (see Table 3-8 for key to classes) 36
Figure 3-4. Comparison of Site History Ranking and Ground Water Contamination
Ranking for the 270 sites at which the potential for DNAPL occurrence must
be inferred 39
Figure 3-5. Potential for DNAPL occurrence at 270 sites evaluated. Rankings defined in
Table 3-9 41
Figure 3-6. Distribution of the 310 sites of this study according to Hydrogeological
Setting. Refer to Table 3-11 for explanation of settings 44
Figure 3-7. Results of the Site Flistory and Ground Water Contamination Rankings
as Related to Hydrogeologic Setting 45
Figure 3-8. Site use distribution for the 310 sites 47
Figure 3-9. Site History Ranking and Ground Water Contamination Ranking by
Site Use Type 51
Figure 3-10. Distribution of the 310 sites evaluated according to site contaminant
type 53
Figure 3-11. Site Use Distribution for the 98 Chlorinated Solvent Sites 54
IV
Word-searchable Version Not a true copy
-------
Figure 3-12. Site use distribution for the 155 sites in the mixed industrial solvents
category. Refer to Figure 3-11 for key 56
Figure 3-13. Relationship of Contaminant Type to Likelihood of Subsurface
DNAPL 65
Figure 3-14. Extrapolation of the Study Results to the Universe of NPL Sites 66
Word-searchable Version Not a true copy
-------
LIST OF TABLES
Table 1 -2. Organic contaminants detected most frequently in ground water at
Superfund sites 7
Table 2-1. Basic Information on Each Site Collected for the DNAPL Site
Assessment Study 10
Table 2-2. Number of sites evaluated in each region during the study period
(November 91 - December 92) 11
Table 3 -1. Site History Ranking Characteristics and the Number of Sites Fitting
Each Category 18
Table 3-2. Site History Ranking Assignments from Combinations of DNAPL
Indicators 20
Table 3-3. Number of Sites Reporting DNAPL Indications from Site History
Information 21
Table 3-4. Relationship of Degree of Site History Understanding to Site
History Ranking 23
Table 3-5. Summary of factors that contribute to less-than-saturation concentrations
of DNAPL compounds in ground water at sites with single-component
DNAPL source 27
Table 3-6. Concentrations of Tetrachloroethylene, Trichloroethane, Trichloroethylene,
and Methylene Chloride expressed as percentages of their pure-phase
solubilities, and the number of Superfund sites in this study (out of 310)
reporting each level of contamination 29
Table 3-7. Summary of factors that contribute to less-than-saturation concentrations
of dissolved-phase chemicals emanating from a multi-component
DNAPL source, in addition to those listed in Table 3-5 30
Table 3 -8. Contaminant Ranking Assignment, ranking of sites based on maximum
percentage solubilities of DNAPL Compounds 32
Table 3-9. Definitions of the Four Composite Rankings 37
Table 3-10. Matrix for combining the site history ranking and ground water contamination
rankings at sites for which the potential for DNAPL
VI
Word-searchable Version - Not a true copy
-------
occurrence must be estimated 40
Table 3-11. Descriptions of the Hydrogeological Settings of Heath (1984) 43
Table 3-12. Major Categories of Site Uses 48
Table 3-13. Compounds reported at > = 0.01% solubility in ground water at the
ninety-eight chlorinated solvent sites 55
Table 3-14. Main Compounds reported at >0.01% Solubility in Ground Water at
Mixed Industrial Solvent Sites 57
Table 3-15. Compounds Found at > 0.01% Solubility in Ground Water at
Creosote Sites 59
Table 3-16. Compounds Found at > 0.01% Solubility in Ground Water at Coal Tar
Sites 60
Table 3-17. Compounds found in ground water at PCB/Solvent sites 64
vu
Word-searchable Version Not a true copy
-------
EXECUTIVE SUMMARY
This document presents the results of a survey undertaken by the U.S. Environmental Protection
Agency's (EPA's) Office of Emergency and Remedial Response (Superfund). The survey was designed
to estimate the proportion of National Priorities List (NPL) sites where Dense Nonaqueous Phase Liquids
(DNAPLs) may be present. Earlier studies by OERR suggested that DNAPLs may be more common at
hazardous waste sites than previously thought, and may act as a continuing source of contamination thus
reducing the ability of pump-and-treat systems to attain cleanup goals within expected timeframes
(Evaluation of Ground Water Extraction Remedies, Phase II, EPA 9355.4-05). This study represents the
first systematic nation-wide review of NPL sites designed to estimate the extent of subsurface DNAPL
contamination.
Superfund sites with DNAPL contaminants pose special problems and challenges with respect to
site investigation and remediation because DNAPLs comprise a separate liquid phase whose behavior
differs significantly from that of the dissolved phase. Unlike the transport of dissolved contaminants,
DNAPL migration is gravity driven and relatively unaffected by ground water flow. DNAPL transport is
strongly influenced by small-scale geological heterogeneities, and the resulting subsurface distribution of
DNAPLs can be extremely complex. Further, DNAPLs can migrate vertically through fractures in rock
or clay layers, and thus, can contaminate deep aquifer systems. Since many DNAPLs are clear liquids in
their pure product form, they are difficult to recognize, even when directly encountered in the subsurface.
As a result of these characteristics, conventional site investigation methods which are used successfully at
non-DNAPL sites may produce misleading data when used at DNAPL sites, and in some cases may cause
site conditions to worsen. Once they reach the saturated zone, DNAPLs constitute a major long-term
source of dissolved-phase contamination that is difficult or, in some cases, impossible to remove with
current technology. Indeed, because of their unique characteristics and behavior in the subsurface,
DNAPLs pose a serious challenge to conventional site investigation and remediation techniques.
In summary, this study developed an estimate of the likelihood of DNAPL in ground water by
re-evaluating existing site data at a large sample of NPL sites. The results of the study are intended to aid
policy makers by serving as a basis for assessing ground water remediation policy and guidance in the
Superfund program. The results of this study also suggest that the emphasis of future research efforts should
be placed on chlorinated solvents and mixed solvents sites, as these represent the majority of sites having
DNAPL-related compounds.
An additional goal of the project was to assess the usefulness of various indirect indicators of
DNAPL presence associated with site historical activities and ground water contaminant information. The
results of this study indicate that certain indirect indicators correlate well with DNAPL presence. This can
benefit site managers by helping focus data
Vlll
Word-searchable Version Not a true copy
-------
gathering efforts early in the site assessment and investigation phases.
This report does not detail all of the procedures that a site interpreter would use to determine the
presence of DNAPLs at a specific site. There is no viable substitute for careful and thorough evaluation of
all site data by an experienced site interpreter. However, the methodology used in this study to estimate
the likelihood of DNAPL presence is quite similar to the method outlined in the Fact Sheet "Estimating the
Potential for DNAPL Occurrence at Superfund Sites" (Appendix B, EPA/9355.4-07FS) and as such
could be applied to any existing site where appropriate information exists. However, it is not intended to
serve as a substitute for a complete and thorough site evaluation by an experienced site interpreter. The
methodology used in this study will be used to revise and update the above-referenced Fact Sheet. For a
detailed discussion of the scientific and technical issues associated with DNAPL compounds and their
behavior in the subsurface environment please refer to the recently published technical guidance document
"DNAPL Site Evaluation" (EPA 600/R-93-022).
The study included a screening level evaluation of 712 NPL sites (roughly 55% of all NPL sites,
as of 1991) in Regions 1, 3, 5, 6, and 9. At forty-four of the 712 sites, DNAPLs were observed directly
in the subsurface. The likelihood of DNAPL occurrence at the remainder of these sites was estimated
based on more detailed analysis of a subset of 310 sites (25% of the NPL sites), including 40 of the sites
where DNAPLs were observed directly. Finally, these results were then extrapolated to all NPL sites.
Detailed information for each site studied was obtained from Remedial Investigation and other site
characterization reports, direct discussions with Remedial Project Managers, and regional hydrogeologists.
Nearly all major physiographic regions in the U.S. and virtually all categories of Superfund site types were
covered by the study. The conclusions drawn in this report are based solely on the site historical information
and site characterization information provided for review.
Two separate ranking systems were developed that, when applied to site information, would yield
a relative ranking of low, medium, or high for the likelihood of subsurface DNAPL. The two ranking
systems were based on site historical information and site contaminant information, respectively. These
separate rankings for each site were then combined via a matrix table into a single estimate of the likely
presence of DNAPL at that site. The sites where DNAPLs were observed directly were used to measure
the applicability and effectiveness of the two ranking systems.
Three additional factors were evaluated in order to determine what influence they had on
determining the likelihood of DNAPL presence. The three factors were: 1) hydrogeologic setting; 2) site
use type; and 3) site contaminant type.
The results of this study provide the backdrop for a number of other important technical guidance
documents and Fact Sheets. These include: "Evaluation of the Technical Impracticability of Ground Water
Restoration"; "Presumptive Remedies: Strategies and
ix
Word-searchable Version Not a true copy
-------
Treatment Technologies for CERCLA Sites with Ground Water Contamination" , "Site Characterization
for DNAPLs"; and "Methods for DNAPL Extraction."
CONCLUSIONS
Extrapolation of the survey results to the current universe of NPL sites indicates that approximately
60% of all NPL sites exhibit a medium to high likelihood of having DNAPLs present as a source of
subsurface contamination. A further breakdown yields the following: approximately 5% of the sites fall
within the category where DNAPL presence is "definite or known"; 32% of sites have a "high potential"
for DNAPL presence; 20% have a "medium potential"; 27% fall within the "low potential" category; and
16% are "unlikely" to have DNAPLs present. In some instances the lower likelihood of DNAPL presence
may be the result of inadequate knowledge of past site activities and/or inadequate site characterization.
Thus the results of this study suggest that the presence of DNAPLs should be considered carefully in
planning site investigation and cleanup strategies for most Superfund sites.
The analysis of hydrogeologic setting on DNAPL occurrence indicated that there was no
identifiable hydrogeologic setting that had a greater likelihood of exhibiting subsurface DNAPL than
another. In addition, dissolved-phase DNAPL contamination was just as likely to be present in aquifers
with a deep vadose zone as those with a shallow water table.
The relationship of site use to DNAPL occurrence was evaluated in order to determine if certain
site uses (site types) exhibited a greater likelihood for subsurface DNAPL than others. The results indicated
that indeed, certain site types continuously ranked "high" in likelihood of DNAPL presence. Site categories
with the highest likelihood of having DNAPL include: wood-treating sites, general manufacturing sites,
organic chemical productions sites, and industrial waste landfills. Sites within these categories should be
assumed to have a medium to high likelihood of DNAPL presence and site managers should design site
investigation and remediation activities accordingly. A more detailed list of site types falling under these four
general categories is included in the main body of the report.
The relationship between site contaminants and DNAPL occurrence was evaluated in order to
determine if there were certain suites of compounds present at concentration levels above their theoretical
maximum solubilities that would exhibit a higher likelihood of subsurface DNAPLs than at sites where that
situation does not exist. The results correlate well with the types of DNAPL compounds associated with
specific site types. The contaminants most directly associated with DNAPL presence included: creosote
compounds, coal tar compounds, Polychlorinated Biphenyls (PCBs), chlorinated solvents, and mixed
solvents. However, even though creosote, coal tar, and PCB sites were easily linked with specific site uses,
and have a relatively high likelihood of subsurface DNAPL, they represent only a very small proportion of
the universe of NPL sites. The majority of NPL sites where
Word-searchable Version Not a true copy
-------
likelihood of DNAPLs is high exhibit chlorinated and mixed solvent contaminants in ground water.
The results of this study also suggest that the emphasis of future research efforts for ground water
remediation should be placed on chlorinated solvents and mixed solvents sites, as these represent the
majority of sites having DNAPL-related compounds present as a separate phase and as a source of
dissolved-phase ground water contamination.
The site historical information ranking system correlated well with the information from the sites
known to have DNAPLs present. The historical information focused on site use, past disposal practices
and release of DNAPL compounds throughout the period of site operation. This information can yield
important direct and indirect evidence that DNAPLs have been released. However, the lack of such
information does not constitute evidence that DNAPLs were absent at a site.
The ground water contaminant ranking system (expressed as a per cent of maximum solubility) also
correlated well with information from the sites known to have DNAPLs present. While the presence of a
DNAPL-related compound dissolved in ground water is one of the best indirect indicators of the likelihood
of DNAPL presence, the presence of dissolved-phase DNAPL does not confirm the presence of a
pure-phase DNAPL source in the subsurface. However, certain concentrations are now generally accepted
by the research community as indicating a high likelihood of a subsurface source of DNAPL across a wide
range of site types (i.e. 1% or more of a compound's solubility). However, concentrations representing less
than 1% of a compound's solubility does not indicate the absence of a subsurface DNAPL source.
XI
Word-searchable Version Not a true copy
-------
CHAPTER 1
INTRODUCTION
Dense nonaqueous phase liquids (DNAPLs) are chemical compounds that are heavier than water
in their pure form. Examples of such compounds are chlorinated solvents, which were associated with many
site operations common at Superfund sites. By far, the largest group of DNAPL compounds encountered
at Superfund sites consists of chlorinated organic solvents. Because of their unique characteristics and
behavior in the subsurface, DNAPLs may pose a serious challenge to conventional site characterization and
remediation techniques.
Superfund sites with DNAPL contaminants pose special problems and challenges with respect to
site investigation and remediation (Cohen and Mercer, 1993; Ruling and Weaver, 1991; Mackay and
Cherry, 1989). DNAPLs comprise a separate liquid phase whose behavior differs significantly from that
of the dissolved phase. DNAPL migration is gravity driven and relatively unaffected by ground water flow
and often moves in a manner that is independent of ground water flow. DNAPL transport is strongly
influenced by small-scale geological heterogeneities, and the ultimate subsurface distribution of DNAPLs
can be extremely complex. DNAPLs can migrate vertically through fractures in rock or clay formations and
thus, can contaminate deep aquifer systems. Once DNAPLs have entered the subsurface environment, they
can act as a source of contamination for an extremely long period of time by releasing gas phase and
aqueous phase chemicals to soil and ground water. Many DNAPLs are clear liquids in their pure product
form and are therefore difficult to recognize, even when directly encountered in the subsurface.
Conventional investigation methods which are used successfully at non-DNAPL sites may produce
misleading data when used at DNAPL sites, or in some cases cause site conditions to worsen. Once they
reach the saturated zone, DNAPLs constitute a major long-term source of dissolved-phase contamination
that can be difficult or, in some cases, impossible to remove with current technology.
Scientific knowledge concerning the occurrence and behavior of DNAPLs in ground water was
nearly non-existent in 1980 when the Comprehensive Environmental Response, Compensation and Liability
Act (CERCLA) was enacted. As a result, many Superfund site investigations in the 1980s were carried
out without regard for possible DNAPL presence. However, field data collected during these investigations
include both direct observations of DNAPLs and indirect evidence of DNAPL sources. These data provide
a valuable resource for understanding the impact of DNAPLs at Superfund sites and for guiding future
efforts to define DNAPL contamination.
The primary goal of this study is to estimate the likelihood of occurrence of DNAPLs in ground
water at Superfund sites nation-wide through the re-evaluation of existing site data. The results of this study
are intended to provide a technical and scientific basis for refining ground water remediation policy and
guidance in the Superfund program. Secondary goals of this project are to:
Word-searchable Version Not a true copy
-------
N assess the usefulness of various indirect indicators of DNAPL occurrence
from existing Superfund site documents;
N raise awareness of DNAPL issues in the EPA regional offices and highlight
specific sites at which DNAPLs occurrence is likely; and
N identify groups of Superfund sites that have similar DNAPL-related characteristics
in order to provide a framework for long-term program planning and research.
This report is not a guidance document, however, the methodology used to estimate the likelihood of
DNAPL presence is based slight modifications of the method outlined in the Fact Sheet "Estimating the
Potential for DNAPL Occurrence at Superfund Sites" and could be applied to any existing site where
appropriate information exist. The methodology used in this study will aid in refining and revising the above
referenced Fact Sheet. Please refer to the guidance document "DNAPL Site Evaluation" (EPA
600/R-93-022) for a detailed discussion of the scientific and technical issues associated with DNAPL
compounds and their behavior in the subsurface environment.
1.1 Background
The Superfund program specifically addresses sites where past, rather than current, activities have
led to the contamination of soil and water resources. Contamination at many Superfund sites has been
occurring over many years, or in some instances, several decades. Typically the contamination results from
waste handling and disposal practices no longer allowed, and frequently involves contaminants that are
resistant to rapid breakdown. Common among these contaminants are synthetic organic compounds, a
category of compounds manufactured in large quantities since the second World War. Many of these
synthetic organics, particularly the chlorinated solvents, are denser than water in their pure form.
The environmental media most commonly affected by contamination at Superfund sites are soil and
ground water. A review of data collected from the current sites on the NPL indicates that 85% of the sites
have ground water contamination and 72% have soil contamination (USEPA, 1991). An EPA report of
Superfund Records of Decision (RODs) indicates that, of the 591 sites for which Records of Decision
(RODs) have been signed address ground water contamination, 90% (535 sites) report ground water was
contaminated with organic compounds. A central task of the Superfund program, then, is to address the
contamination of ground water resources by organic compounds.
Ground water investigations differ from other kinds of environmental studies in that they involve a
significant amount of inference. An understanding of potential sources and avenues for contaminant release
generally id reconstructed from historical information on site practices. A three-dimensional site conceptual
model of subsurface contamination
Word-searchable Version Not a true copy
-------
generally is developed using data from relatively sparsely placed soil and ground water sample locations,
and contaminant behavior must be inferred from an understanding of the interactions of chemical properties
and site hydrogeologic conditions.
Historically, for many Superfund investigations, the site conceptual model has assumed that the
sources of ground water contaminants lie primarily in the unsaturated zone, near the ground surface. Also,
contaminants are generally considered to have been released to the environment as part of an aqueous
solution rather than in their pure liquid form. Under this conceptual model, rainwater infiltration through the
unsaturated zone is considered to be the major mechanism of contaminant transport from the surface to
ground water; and all contaminants in the saturated zone are either dissolved in ground water or sorbed to
aquifer material.
This site conceptual model has driven nearly all Superfund site investigations through the 1980s and
early 1990s. Perceiving a site in this manner affects the kinds of data collected at a site as well as the
remedial actions selected. A revised conceptual model, where subsurface DNAPL is a source of
dissolved-phase contamination should influence both the site investigation techniques and the options for
ground water remediation.
Potential differences between a non-DNAPL site and a DNAPL site are illustrated in Figures 1-1
and 1-2. Figure 1-1 shows a plan view and cross sectional diagram of a site at which the source of
contamination is an unlined hazardous waste lagoon that received trichloroethylene (TCE) waste over a
twenty year period, all in the form of an aqueous solution. In this case, the TCE is transported through the
unsaturated zone in dissolved form, reaching the water table and forming a plume of dissolved TCE in the
upper aquifer. The plume migrates in the direction of ground water flow. In contrast, Figure 1-2 shows a
site at which TCE was discharged into the lagoon over a similar active period as a separate immiscible
liquid. For this second case, the pure TCE has migrated into the subsurface, where it acts as a source of
dissolved contamination. The TCE DNAPL has traveled out the base of the lagoon through the upper sand,
leaving behind immobile blobs (residual) trapped in the pore spaces of the sand. The first clay layer has not
acted as a barrier to contaminant migration. Since DNAPL transport is gravity driven, the TCE pooled on
depressions on the clay surface and penetrated fractures or rootholes in the clay, where it then entered the
second aquifer. Both diagrams depict the same hydrogeological setting, with massive sand units interbedded
with clay layers. These figures represent simplifications of actual subsurface conditions which may be
encountered at Superfund sites.
Word-searchable Version Not a true copy
-------
Figure 1-1. Schematic cross section of a site at which TCE has been discharged to a waste
lagoon over a 20 year period as part of an aqueous solution.
Residua! Saturation of
DNAPL in So« From Spill
Infiltration and
Leaching
Vadosa
Zone
Plume of-Dissolved
; Contaminants; ; \
Groundwater
Flow
Former Waste Pond
Dissolved Phase Plume
Ground Water Flow Direction
Word-searchable Version - Not a true copy
-------
Figure 1-2. Schematic cross section of a site at which TCE has been discharged to a waste
lagoon over a 20 year period as a separate organic liquid.
DNAPL Zone
contains free-phase DNAPL in
or lenses and/or residual DNAPL
Former Waste Pond
DNAPL Entry Location
Dissolved Plume
Ground Water Flow Direction
Word-searchable Version - Not a true copy
-------
This study reviewed Superfund site data collected as if a site looks like Figure 1-1, and
re-examined it to determine whether site data are actually more consistent with Figure 1-2. The relative
likelihood of DNAPL occurrence is inferred by asking the following types of questions: Are site operators
certain that no TCE was released in a nonaqueous form? Were high concentrations of dissolved TCE
unexpectedly found in the lower zones? Were nonaqueous liquids observed in soil cores from the base of
the lagoon? Has a pump and treat system removed a contaminant mass that is many orders of magnitude
larger than that which is present in the dissolved plume? Together these types of information can help to
indirectly assess whether DNAPL sources might be present below the water table. This study did not
attempt to estimate the mass of DNAPL in the subsurface at any site, and does not describe the methods
for doing so.
1.2 Potential Scope of the DNAPL Problem
Information from recent studies suggests there is a potential for DNAPL contamination at many
Superfund sites. As previously noted, approximately 85% of the sites on the NPL reported ground water
contamination during the Hazard Ranking System scoring process. Approximately 800 NPL sites, or 66%
of the sites listed on the NPL, report solvents as waste materials (NPL Characterization Project, 1991).
Forty-nine sites used creosote, and eight were coal gasification plants, which routinely disposed of coal tar.
Of the twenty organic contaminants detected most frequently in ground water at Superfund sites (Table
1-2), thirteen are DNAPLs or DNAPL-related compounds. Of these thirteen, most are chlorinated
solvents.
In the late 1980s, EPA conducted a study of the efficacy of pump and treat systems at 24
Superfund sites (Evaluation of Ground-Water Extraction Remedies, EPA Directive 9355.4-05). One of
the conclusions of this study was that, a key factor preventing efficient site clean-up within a reasonable
timeframe was the failure of remedial designs to account for the possibility of subsurface DNAPL. A more
recent study of pump and treat remediation at 11 chlorinated solvent sites (Harman et al, 1993) found that
the two major limits to aquifer restoration were inadequate site characterization and presence of unidentified
reservoirs of subsurface DNAPL sources.
Despite the widespread use of DNAPL compounds, and the common detection of these
contaminants dissolved in ground water, very few Superfund sites report direct observations of DNAPLs
in the subsurface. An informal poll of the EPA Regions conducted as part of this study found 44 sites (less
than 5% of the NPL sites) at which DNAPLs had been directly observed. Further, most of the encounters
have been accidental. Therefore, in order to assess the pervasiveness of DNAPL at Superfund sites, this
study used indirect indicators of DNAPL sources to assess the potential for DNAPL occurrence in the
absence of direct observation.
Word-searchable Version Not a true copy
-------
Table 1-2. Organic contaminants detected most frequently in ground water at Superfund sites.
Ranking Chemical
1 Acetone
2 Bis (2-ethylhexl) phthalate
3 Toluene
4* Trichloroethylene
5* Chloroform
6* Methylene Chloride
7* Dichloroethylene, 1,2-
8* Trichloroethylene, 1,1,1-
9 Benzene
10* Tetrachloroethylene
11 Xylenes
12* Dichloroethane, 1,1
13 Ethylbenzene
14* Di-n-butyl phthalate
15* Naphthalene
16 Methyl Ethyl Ketone
17* Chloroebenzene
18* Dichloroethylene, 1,1
19* Phenol
20* Carbon Bisulfide
SOURCE: Superfund Chemical Analysis Results (SCAR), downloaded from the CLP Analytical Results
Database (CARD). The CARD database was published in 1988 and contains results from the Contract
Laboratory Program (CLP) analyses of samples taken from Superfund sites.
* = DNAPLs or DNAPL-related compounds.
Word-searchable Version - Not a true copy
-------
1.3 Overall Study Strategy
The approach used for this study was to screen a large sample of sites to identify good candidates
for detailed analysis, evaluate a subset of sites in detail for indirect indications of DNAPL occurrence, and
then extrapolate the findings to all NPL sites across the country. 712 sites were surveyed. The detailed
analysis included 310 Superfund sites in five EPA Regions and was performed using a standardized data
collection form. The form requested information pertaining to DNAPL indicators that were most uniformly
available in site documents. The subgroup of sites studied constitutes one quarter (25%) of the sites listed
on the NPL.
In selecting the subset of sites, only those were considered that were far enough along in the site
investigation process that sufficient data could be obtained for an evaluation of DNAPL occurrence, and
to obtain a range of hydrogeological settings and site use types.
At forty sites, DNAPL had been directly encountered in the subsurface. Although this subgroup is
not entirely representative of DNAPL sites addressed by the Superfund program as a whole, it represents
the only available standard for measuring the relative importance of indirect indicators of DNAPL
occurrence. For the other 270 sites studied in detail, dissolved organic contaminants had been detected
in ground water but there were no direct observations of DNAPL in the saturated zone. For this subset,
a ranking system was developed that assigned a high, medium, or low potential for DNAPL occurrence.
The system separately analyzed the site use history and ground water data, and then combined the
information into a single estimate, using a modified version of that outlined in the Fact Sheet "Estimating the
Potential for Occurrence of DNAPL at Superfund Sites (Appendix B).
Once the potential for DNAPL presence had been estimated for a site, it was grouped with similar
sites to see if there were other factors that would influence DNAPL occurrence. Sites were grouped by
hydrogeological setting, prior use, and ground water contaminant type. The final task was to extrapolate,
from the results of this study to the remaining NPL sites. This provided an indication of the pervasiveness
of DNAPLs at all NPL sites. These results would then allow the Superfund program to evaluate, and refine
as appropriate, the policies associated with ground water remediation.
Word-searchable Version Not a true copy
-------
CHAPTER 2
DATA COLLECTION AND MANAGEMENT
In developing a data collection strategy for this study, two factors were balanced: the number of
sites evaluated and the amount of information collected from each site. To incorporate a representative
sample of site uses and hydrogeologic settings, as many sites as possible were evaluated given the time
frame of the study. Specific goals of the data collection effort were to:
N collect site information in a consistent manner for
comparative analysis;
N obtain enough detailed information on the sites
known to have DNAPLs to test the assumptions
regarding indirect indicators of DNAPL
occurrence;
N obtain information from a broad spectrum of sites,
those with both high and low DNAPL probability;
and
N collect information encompassing a range of site
uses and hydrogeological settings.
In all, detailed information on 310 Superfund sites in five EPA Regions, including 40 sites at which DNAPL
had been directly observed in the saturated zone, were collected and evaluated.
2.1 Data Needs
Site information from Remedial Investigation (RI) and other site characterization reports, and other
site documents provided the bulk of the information used to evaluate the potential presence of DNAPL.
The tool used for recording this information was a site survey form (Appendix A). The site survey form was
very detailed, and included information that would enable evaluation of the indirect DNAPL indicators listed
in the DNAPL Fact Sheet (Appendix B). One form was completed for each site. A list of general
categories of site information collected is provided in Table 2-1.
Word-searchable Version Not a true copy
-------
Table 2-1. Basic Information on Each Site Collected for the DNAPL Site Assessment Study.
Site History
N Site use and years of use
N Historical industrial and waste disposal practices
N Hazardous substances and chemicals on-site
N Information on known releases of hazardous substances and chemicals
Site Investigation
N Observation of LNAPLs and DNAPLs
N Maximum observed concentrations of organic chemicals in ground water
N Main contaminant sources
N Presence of DNAPL-related spatial and temporal patterns in ground water
Extent of Field Program
N Stage in the Superfund process
N Number of monitoring wells and ground-water samples analyzed for organics
N General understanding of hydrogeology, contaminant sources, and ground-water contamination
Hydrogeological Information
N Unconsolidated and bedrock materials
N Depth to bedrock and to ground water
N Dimensions of ground-water plume
Survey Response
N General comments on survey content
N Comments on DNAPL information and research needs
10
Word-searchable Version Not a true copy
-------
2.2 DNAPL Survey Response
The number of sites for which detailed information was obtained from each region is listed in Table
2-2.
Table 2-2. Number of sites evaluated in each region during the study period (November 91 -
December 92).
EPA
Region
Region 1
Region 3
Region 5
Region 6
Region 9
Total
Number of
Superfund sites in
Region*
84
162
267
74
125
712
Number of
Superfund sites
evaluated
in detail
79
92
74
23
42
310
Percent of
Superfund sites
evaluated
in detail
94%
57%
28%
31%
33%
44%
This number represents the number of active sites on the NPL in the region at the time of the study
(FY 92).
11
Word-searchable Version Not a true copy
-------
CHAPTER 3
ANALYSIS OF DNAPL OCCURRENCE
The methodology used for establishing the potential for DNAPL occurrence in ground water at
Superfund sites is based on the approach outlined in the Fact Sheet, "Estimating the Potential for
Occurrence of DNAPLs at Superfund Sites" (Appendix B). In keeping with the Fact Sheet, two broad
categories of Superfund site data were considered:
(1) information from the site use history, and
(2) data obtained during the site investigation of
ground water contamination
These data were evaluated independently and then combined into a single estimate of the relative
probability of the presence of subsurface DNAPL. In order to apply the method consistently across a wide
variety of site types, specific means of answering the questions posed by the DNAPL Fact Sheet were
defined. Based on experience evaluating a large number of sites, modifications and refinements were made
to the Fact Sheet approach. This chapter outlines the method of ranking sites for DNAPL probability and
discusses the findings regarding the potential for DNAPL occurrence at the 270 NPL sites. The ranking
system was also applied to the 40 sites where DNAPLs were observed present, as a measure of the
effectiveness of the methodology.
The ranking system uses a baseline of information that was easily obtainable for the majority of
sites, and by its nature cannot consider all of the complexities of each site. DNAPL potential is not a
parameter that is easily quantified, and the best estimates of DNAPL occurrence result from careful
weighing of many lines of evidence. The site rankings may be suitable for long-term program planning, for
targeting sites for further study, and for establishing broad trends. A site ranking should not be taken as the
definitive word on the occurrence of DNAPL, or if present, it's mass at any given site. For individual sites,
there is no viable substitute for careful and thorough evaluation of all site data by an experienced site
interpreter.
3.1 Site History Ranking
Investigation of site uses over the active period of operation can yield important indirect evidence
that DNAPLs have been released. This section describes the method of analyzing site history information
and applies a site history ranking system to the 40 known DNAPL sites and to the 270 sites for which the
potential for DNAPL occurrence was to be estimated.
Method for Evaluating Site History Information
Currently, the DNAPL Fact Sheet poses three questions regarding the site use
13
Word-searchable Version Not a true copy
-------
history:
(1) Does the industry (site) type suggest a high probability of historical DNAPL release?
(2) Does a process or waste practice employed at the site suggest a high probability of
DNAPL release?
(3) Were there any DNAPL-related compounds used in appreciable quantities atlhe site?
Modifications to these questions were made in order to focus on actual knowledge of onsite use,
disposal, and release of DNAPLs throughout the site history. The last question was expanded to include
general types of substances (e.g., solvents, oils, pesticides, etc.) that may have been present at the site. The
term, "appreciable quantities" was defined as at least 5 drums per year. Although much smaller quantities
can easily migrate to ground water (Poulsen and Kueper, 1991) and cause substantial dissolved-phase
contamination, quantities of fewer than five drums per year are unlikely to have been documented for most
Superfund sites. Finally, information was gathered on known releases of DNAPL substances to the
environment, specifically the form (non-aqueous vs. dissolved in water) of these releases.
Considering these modifications, five aspects of the site history must be answered in order to obtain a site
history ranking using this system:
(1) Does the site type suggest a high probability of historical
DNAPL release?
(2) Did site operations include industrial processes or waste
management practices that suggest a high probability of DNAPL
release?
(3) Were any DNAPL-related compounds or substances used in
appreciable quantities (>5 drums/yr) at the site?
(4) If DNAPL-related substances were present on site, were there
known releases of them?
(5) If there were known releases, were the materials released
primarily in nonaqueous form, or as components of an aqueous
solution?
The method of ranking site history information is based on "yes" answers to the above questions,
or positive indicators of DNAPL presence (contained on the survey form,
14
Word-searchable Version Not a true copy
-------
Appendix A). This approach was taken because there is no site historical information that canbe used to
entirely rule out the possibility of DNAPL releases. Careful logs of daily site operations and chemical
inventories were rarely kept at Superfund sites. DNAPL compounds, particularly the chlorinated solvents,
are so widely used that their presence, at least in small quantities, is possible at virtually any site.
Consequently, the direct knowledge of a DNAPL-related practice, substance, or release can be used in
establishing DNAPL probability for a site, but a lack of such knowledge does not constitute evidence that
DNAPLs were absent at the site.
In order to answer the five site history questions, lists of site types, hazardous substances,
and site operations that are associated with DNAPL contamination were developed. These lists,
presented as Table 3-1, expand upon those found in the DNAPL Fact Sheet and in unpublished
work by Cherry and Feenstra (1991). Table 3-1 is used to determine whether or not the answers
to the first three questions related to the site history ranking are "yes" or "no". A site receives
a "yes" answer if it falls within the categories of site types listed in section A, has handled
hazardous substances listed in section B, or has site operations listed in section C. If a site does
not fit under one of these categories it receives a "no". The "yes" answers are then recorded
in the first three columns of the Site History Ranking Assignment (Table 3-2).
Two questions remain to be answered in order to determine the final site history ranking
from Table 3-2. They both refer to the form in which a release of DNAPL-related compound
occurred. The DNAPL compound may have been released in a nonaqueous form, (e.g. pure
solvent discharged to an unlined lagoon) or an aqueous form (e.g. solvent washed from a floor
with water and discharged to a dry well). If either of these conditions occurred, then a "yes"
answer is recorded in the appropriate column in Table 3-2. The final site history ranking is then
read from the far right hand column. The history ranking can then be applied to the matrix table
combining the site history ranking and the ground water contamination ranking (Table 3-10).
Table 3-2 is the "Site History Ranking Assignment" table. It shows the possible combinations of
"yes" answers to the five site history questions, and the assignment of the history ranking based upon the
answers. The site history ranking ranges from 1 (low DNAPL likelihood) to 6 (high DNAPL likelihood).
For instance, the eighth line of Table 3-2 describes a site which reports a facility type and waste disposal
practice that have a high probability of DNAPL release, but DNAPL-related substances were not present
in appreciable quantities at the site. This combination of answers is assigned a rank of 2.
Table 3-1 also lists the number of sites that reported each of the DNAPL-related facilities,
substances, or practices for the 40 known DNAPL sites and the 270 sites for which DNAPL probability
was to be estimated. More than half of the facility types designated as "DNAPL-related" are reported for
the known DNAPL sites. All of the listed DNAPL-related substances are reported at the known DNAPL
sites except for asphalt. Nearly 90% of the DNAPL-related site operations are reported at the known
DNAPL sites. These data
15
Word-searchable Version Not a true copy
-------
suggest that the site history features targeted by the study for their potential association with DNAPL
contamination do in fact characterize sites with known DNAPL contamination.
Many of the DNAPL-associated site history characteristics are also reported at the 270 sites for
whichDNAPL probability must be estimated. Of the DNAPL-related site types, electronics and electrical
equipment manufacturing and fabricated metal production were the most frequently reported manufacturing
activities. Other site types frequently encountered included organic chemicals production, liquid hazardous
waste disposal, storage and transport facilities, and solvent recycling.
Solvents were by far the most pervasive DNAPL substances, and they were reported at nearly
three quarters (75%) of the sites. Metal cleaning and degreasing, solvent loading and unloading, storage
of drummed solvents, and storage of solvents in underground tanks were commonly reported industrial
practices. Two waste disposal practices, dumping of liquid wastes onto the ground and discharge of liquids
to lagoons and surface impoundments were practiced at a majority of the sites. Spills and leaks were
reported at nearly half of the sites. These findings indicate that use and disposal of DNAPLs, particularly
solvents, occurred relatively routinely at the subgroup of Superfund sites included in this study.
It is clear from Table 3-2 that the controlling factor in the assignment of a higher ranking is the
reported presence of DNAPL substances on-site. Site operations and practices are given lesser weight
because they merely imply the use or disposal of DNAPL-related compounds, rather than absolutely
confirming them. Known releases of DNAPL substances, particularly in a non-aqueous form, significantly
increase the likelihood of subsurface DNAPL. All sites at which there was a known release of a DNAPL
substance in nonaqueous form receive the highest ranking (6).
One potential source of bias in the estimate of DNAPL probability from site history information is
a noted tendency for those who provided site data to infer historical practices from site characterization
information. For example, it would be natural to infer a historical release of DNAPL substances for a site
at which DNAPL had been directly observed in the subsurface, even if the release was not actually
reported. Such an inference would bias the site history evaluation for the known DNAPL sites in favor of
a higher site history ranking. The survey form specifically requests that no site history information be inferred
from site investigations.
Results of Site History Evaluation
Table 3-3 shows the number of sites reporting at least one of the five DNAPL indicators from site
history information. For the known DNAPL sites, 85% or more reported each of the indicators. For the
remaining 270 sites, more than 61% of the sites had at least one DNAPL-related site type, while nearly
three quarters of the sites reported DNAPL-related substances onsite, indicating there are site types other
than those targeted at which DNAPL-related substances are present in appreciable quantities. More than
90%
16
Word-searchable Version Not a true copy
-------
reported site operations for which one would expect some use or disposal of DNAPL substances. This
finding implies that the potential number of sites using DNAPL compounds is actually higher than the
number reporting known chemical use. For example, at some sites, metal cleaning and degreasing may have
been a routine practice, but the solvents used may not have been documented as part of site activities.
Alternatively, the quantities may have been too small to report. Almost two thirds of the sites (56%)
reported releases of DNAPL substances and chemicals, either dissolved in water, as a separate,
nonaqueous phase, or in unknown form. At one third of the sites (33%), releases of DNAPLs in a
nonaqueous form are known to have occurred.
Figure 3-1 presents the distribution of the historical ranking for the 310 sites of this study. The same
ranking method was applied to the 40 known DNAPL sites and to the 270 sites at which DNAPL
occurrence was to be estimated. The distributions are presented separately for comparative purposes. The
distribution of site history ranking for the known DNAPL sites represents a standard against which the site
evaluation techniques can be measured.
Ideally, the known DNAPL sites should all receive a site history ranking of 6, the highest DNAPL
probability. As shown in Figure 3-la, the site history ranking for 85% of the known DNAPL sites is in fact
6. This distribution indicates that, for 34 of the 40 known DNAPL sites, there was a documented release
of a DNAPL to the environment. However, for six of the sites, no releases were reported over the history
of site use. One example of such a site is a dry cleaning plant where no spills or leaks of dry cleaning fluids
were ever documented, even anecdotally, yet subsequent site investigations revealed a loading area draining
to a drywell that had clearly received DNAPL releases. Other releases that occur beneath the ground
surface, such as leaks from pipelines, are also rarely discovered in advance of site investigations.
Of the remaining 270 sites there is a wider range in assigned rankings (Figure 3-lb), but the
majority of sites are clustered in the higher probability range. Fully 80 percent of these sites receive rankings
greater than or equal to three, signifying a medium to high likelihood of subsurface DNAPL. This finding
confirms the initial expectation that the use and disposal of DNAPL compounds was common at Superfund
sites, and that site practices permitted either deliberate or accidental release of these substances to the
environment. Using site history information alone, there are very few sites at which the possibility of
subsurface DNAPL can be ruled out.
Based on the large proportion of the known DNAPL sites that received the highest site history
ranking, we are confident that the combination of DNAPL indicators targeted from site history information
is in fact highly associated with subsurface DNAPL. However, it is also clear that a medium or low site
history ranking cannot be used to discount the possibility of subsurface DNAPL. For some portion of sites,
lower rankings may instead reflect a lack of knowledge of actual site activities.
17
Word-searchable Version Not a true copy
-------
Table 3-1. Site History Ranking Characteristics and the Number of Study Sites Fitting Each Category.
A. Facility Types Sites with
Observed Remaining
DNAPLs (40) sites (270)
General Manufacturing
Aircraft maintenance and repair 0 3
Aircraft manufacturing 0 4
Automobile and motorcycle manufacturing 0 1
Capacitors and transformers manufacturing 2 1
Electronics and electrical equipment manufacturing 2 39
Engine manufacturing 0 3
Fabricated metal product manufacturing 1 30
Tool and die manufacturing 0 0
Weapons and explosives manufacturing 0 6
Waste Management
Liquid hazardous waste disposal 6 37
Liquid hazardous waste incineration 0 8
Liquid hazardous waste storage and transport 1 14
Liquid hazardous waste treatment 3 2
Solvent recycling 3 13
Transformer reprocessing and/or recycling 0 1
Organic Chemical Production
Coal gasification 3 0
Coking operations (steel industry, etc) 4 2
Organic chemical manufacturing 2 10
Organic chemical packaging, distribution, and storage 0 6
Pesticide distribution, packaging, and transport 0 1
Pesticide and herbicide production 1 8
Solvent manufacturing 1 3
Solvent packaging, distribution, transport and recycling 0 1
Transformer oil production 0 0
Miscellaneous
Wood preservation 13 5
Dry cleaning plant 2 2
Fire-fighter training area 1 2
18
Word-searchable Version - Not a true copy
-------
Table 3-1. Site History Ranking Characteristics and the Number of Study Sites Fitting Each
Category, (continued)
Sites with
Observed Remaining
B. Hazardous Substances DNAPLs (40) sites (270)
Asphalt 0 3
Capacitor and transformer debris 3 5
Coal tar 85
Creosote 14 5
PCB-laden waste oils 3 12
PCBs 3 12
Pesticides 2 29
Solvents, chlorinated 13 115
Solvents, undifferentiated 9 89
Transformer oil 3 1
C. Site Operations
Industrial Practices
Electronic parts and electronics cleaning 2 34
Metal cleaning and degreasing 1 59
Metal machining 0 16
Paint and lacquer stripping 0 5
Solvent loading and unloading 5 60
Storage of drummed solvents in uncontained areas 7 60
Storage of solvents in underground tanks 4 47
Storage of solvents in above-ground tanks 3 28
Tool and die operations 1 3
Transformer salvage or recycling 1 3
Wood treatment 13 5
Waste Management Practices
Drum disposal/burial 5 78
Lagoon/liquid waste surface impoundment 22 91
Leaks from above-ground tanks 5 38
Leaks from underground tanks and pipelines 7 59
Liquid wastes discharged to septic systems 2 21
Liquid wastes dumped from tank trucks 1 24
Liquid wastes dumped onto open ground 12 112
Liquid wastes released to drains and sumps 5 33
Releases during chemical loading and unloading 11 34
Releases during fires or explosions 6 12
Spills 17 55
Underground injection wells 2 3
19
Word-searchable Version Not a true copy
-------
Table 3-2.
Site History Ranking Assignments from Combinations of DNAPL Indicators.
DNAPL-Related
Facility
Type
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Site
Operations
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Hazardous
Substances
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Known Release
Aqueous or
Unknown Form
Y
Y
Y
Y
Y
Y
Y
Y
Non-aqueous
Form
Y
Y
Y
Y
Y
Y
Y
Y
Hist
Ranki
ng
1
1
1
1
1
1
2
2
2
2
3
3
3
3
3
3
4
4
5
5
6
6
6
6
20
Word-searchable Version - Not a true copy
-------
Table 3-3. Number of Sites Reporting DNAPL Indications from Site History Information.
DNAPL Indications from Site History Information
At least one industrial facility associated with
the use or disposal of DNAPLs
More than 5 drums per year of DNAPL-related
compounds on site
At least one industrial or waste management practice
with a likelihood of DNAPL release
Known release of DNAPL compounds (dissolved in
water, as a separate phase,
or in unknown form)
Known release of DNAPL compounds in nonaqueous
form
Observed
DNAPL
Sites (40)
34
(85%)
36
(90%)
39
(98%)
37
(93%)
34
(85%)
Remaining
Sites (270)
164
(61%)
197
(73%)
245
(91%)
152
(56%)
90
(33%)
21
Word-searchable Version Not a true copy
-------
Site History Rank
B.
§
2345
Site History Rank
Figure 3-1. Distribution of site history rankings for the 40 known DNAPL sites and the 270 sites at
which DNAPL probability must be estimated.
22
Word-searchable Version - Not a true copy
-------
Certainty of Estimates from Site History Information
Since the site history ranking system relies on positive indicators of DNAPL presence, it may not
accurately represent DNAPL probability for sites at which the history of site operations is not well known.
In evaluating the certainty of the site history rankings, two factors were considered: (1) the general amount
of site history knowledge; and (2) the relative availability of the specific indicators targeted.
To provide some measure of the amount of site history knowledge, the survey form asked the site
manager or hydrogeologist to provide their opinions on the general degree of site history understanding.
Table 3-4 shows the number of sites for which the site history is considered very well, well, generally, or
poorly understood. At least a general knowledge of the site history was available for 94% of the sites.
Managers of the known DNAPL sites generally reported a greater degree of site history understanding than
the other sites studied. This greater site history knowledge may have contributed to the fact that DNAPL
was encountered at these sites.
Table 3-4. Relationship of Degree of Site History Understanding to Site History Ranking.
How well understood
is the site history?
Very Well
Well
Generally
Poorly
DNAPL
Sites (40)
17
14
9
0
Remaining
Sites (270)
48
121
86
15
Percent
of 270
Sites
17%
45%
32%
6%
Average Site
History
Ranking
(270 sites)
5
4
4
1-3
For the remaining 270 sites, Table 3-4 also shows the average site history ranking for the various
categories. Sites that are very well understood have a significantly higher ranking, on average, than those
that are poorly understood, so there is some potential for an underestimation of DNAPL potential for sites
where historical practices are not well documented. The implication of these results is that careful
documentation and research of historical site practices will increase both the certainty of DNAPL site
diagnosis and the likelihood that DNAPL-related substances or practices will be discovered.
The specific knowledge of the individual indicators evaluated has bearing on the
23
Word-searchable Version Not a true copy
-------
certainty of the site history evaluation as well. Some amount of research and inference was often required
to answer relatively obvious and straight-forward site history questions posed by the data collection form.
The information that was easiest to extract from site investigation reports was the facility type and a
description of the general site operations, two features that carry less weight in the ranking system. The
names of specific chemicals used or disposed of onsite were less commonly known than general categories
of substances. Liquid organics were commonly reported at the sites studied, but the form of these liquids
(aqueous solutions or pure-phase compounds) was not clearly identified in site documents. The mechanisms
of release of organic liquids were usually documented, but their form upon release was not often reported,
even when known. For example, of the sites with known releases of DNAPL chemicals, 30% had no
information on the form of the release. As site investigators become more knowledgeable about techniques
of investigating potential DNAPL sites, documentation of the form will improve. Since the form of the
compound upon release is a key factor in the site history ranking, any improvements in the reporting of this
particular aspect of the site history will also increase the reliability of DNAPL site diagnosis. For this study,
the majority of sites evaluated included sufficient site history knowledge and documentation.
3.2 Ground Water Contamination Ranking
Data from site investigations provide information on the possible routes of transport of DNAPL to
the subsurface, and can assist in evaluating the likelihood that DNAPL has reached the saturated zone. The
DNAPL Fact Sheet poses three questions concerning data collected during site investigations:
(1) Has DNAPL been found in monitoring wells, observed in soil
cores, or physically observed in the aquifer?
(2) Do chemical analyses of ground water or soil indicate the
possible presence of DNAPL at the site?
(3) Is it likely that the existing field program could miss DNAPL at
the site?
This study separated the Fact Sheet's methodology into two parts. First, the potential for subsurface
DNAPL was established based on direct observations of DNAPLs and chemical analysis of ground water
(questions 1 & 2) and each site has assigned a ground water contamination ranking. The extent of the
field program (question 3) is then evaluated to provide an indication of the certainty of the ground water
contamination ranking.
24
Word-searchable Version Not a true copy
-------
Method of Evaluating Ground Water Contamination Data
In this study the analysis focused on parameters that were available at a broad spectrum of site
types and were of comparable quality from site to site. Unfortunately some data that are very useful for
establishing DNAPL probability for individual sites could not be applied to the full range of site types
encountered. For example, this study's method of evaluating site contamination differs from that of the Fact
Sheet in that it did not consider chemical analyses of soils. For individual sites, properly collected soil data
can be crucial in establishing DNAPL likelihood. Most soils data were from the unsaturated zone, and the
companion data necessary for evaluating the significance of the concentrations, such as the soil moisture
content and fraction of organic carbon, were rarely provided. The method of collecting soil samples was
not uniform, and site-to-site comparisons would not necessarily have been valid. Thus, soils data was not
used as a factor in the ground water contamination ranking.
One indicator that does not appear in the Fact Sheet was added; the presence of temporal trends
in concentrations of DNAPL compounds that suggest the possibility of a subsurface, nonaqueous, source.
As site data were reviewed, it was determined that many sites had removed major near-surface sources
of contamination, yet plume generation continued, and the zones of maximum dissolved-phase
contamination did not appear to move with time. At these sites, the potential for a subsurface DNAPL
source is higher than at sites showing a rapid decline in concentrations in near-source wells.
In order to address questions 1 and 2 of the Fact Sheet, this study evaluated site contamination
information to answer these four questions:
(1) Have there been direct DNAPL observations in ground water
samples, monitoring wells, soil cores, or test pits?
(2) Do maximum concentrations of DNAPL-related compounds
in ground water (as a percentage of their pure-phase solubilities)
indicate the possible presence of DNAPL in ground water?
(3) Do spatial patterns of dissolved-phase contamination include
concentrations of DNAPL compounds that are inexplicably high
at depth beneath source areas?
(4) Do temporal trends in concentrations of DNAPL compounds in
ground water indicate the possible presence of a subsurface,
nonaqueous source?
As previously noted, there are 40 sites at which DNAPL presence is certain. These sites were used
to test the assumptions regarding the data that indirectly indicate DNAPL occurrence, by ignoring the
DNAPL find and evaluating ground water information in a
25
Word-searchable Version Not a true copy
-------
manner comparable to the evaluation of the 270 sites for which DNAPL probability must be estimated.
In analyzing ground water data, each site is classified according to the magnitude of the
concentrations of DNAPL in ground water expressed as a percentage of their pure-phase solubilities. The
maximum concentrations of compounds detected in ground water was used rather than examining the entire
body of chemical data for each site. Based on the theoretical understanding of the distributions of
dissolved-phase contamination emanating from a site, these maximum concentrations are most indicative
of the presence of a nonaqueous source below the water table.
For the 40 known DNAPL sites, the maximum concentrations that would have been detected if
the DNAPL had not been directly encountered were evaluated by reviewing the information on the
maximum concentrations from wells other than those in which the DNAPLs were found. This approach
ensured that ground water data from these sites would be comparable to ground water data from the 270
sites at which DNAPL was to be estimated.
For a single-component DNAPL, the concentration of the compound in ground water that is in
equilibrium with the DNAPL should theoretically equal the pure-phase solubility of the compound. For
example, the concentration of dissolved TCE in ground water contacting a TCE DNAPL should be 1,000
mg/L, or 100% of TCE's solubility. As the dissolved contaminant is carried away from the DNAPL source,
concentrations will reduce to lesser and lesser percentages of the compound's pure-phase solubility.
Factors that produce dissolved-phase concentrations that are significantly lower than the pure-phase
solubility, even in samples obtained quite near a single-component DNAPL source, are summarized in
Table 3-5. It is clear from this table that the concentrations observed will depend greatly on individual site
conditions and investigation techniques.
As noted by Cherry and Feenstra (1991), site conditions are so variable that it is not possible to
accurately prescribe the dissolved chemical concentration that reflects the presence of subsurface DNAPL.
However, computer modelling has shown that, in a hypothetical aquifer of horizontally layered sands with
a tetrachloroethylene (PCE) DNAPL source, ground water samples taken from wells 50 m down gradient
from the source will yield dissolved concentrations of only 0.1% to 5% of PCE's solubility (Anderson et
al., 1991). Case studies of known DNAPL sites also point to the remarkably low concentrations that can
be observed in routine monitoring prior to a DNAPL encounter (Kueper and McWhorter, 1991). The
concentrations that are now generally accepted by the research community as indicating subsurface
DNAPL across a wide range of site types are on the order of 1% or more of a compound's solubility
(Cherry and Feenstra 1991, EPA Fact Sheet, Cohen and Mercer, 1993).
As a reference point for understanding the magnitude of concentrations represented by various
percentage solubilities, Table 3-6 lists these two parameters for four DNAPLs:
26
Word-searchable Version Not a true copy
-------
Table 3-5. Summary of factors that contribute to less-than-saturation concentrations of DNAPL
compounds in ground water at sites with a single-component DNAPL source.
Factor
Explanation
Borehole Dilution
DNAPL will be heterogeneously distributed over vertical intervals
tapped by monitoring wells. The 10-50 foot well screens typical of the
Superfund program are likely to draw water from both DNAPL-
contaminated and relatively uncontaminated strata or fracture systems.
The sample obtained from such a well would be diluted relative to that of
a well screened over a shorter interval tapping a DNAPL zone.
Well Placement
Regardless of the screened interval, wells equidistant from a DNAPL
source in the downgradient flow direction can have widely varying
dissolved concentrations depending on whether they are tapping the
transport route of dissolved contaminants emanating from DNAPL pools
or residual. The DNAPL zones can also be very small relative to the
spacing of wells. These conditions especially hold true in fractured rock
systems and in very heterogeneous overburden.
Ground Water
Sample Collection
Method
Excessive purging can dilute water samples. Some known DNAPL sites
have reported that higher dissolved concentrations are obtained when
kemmerer bottles or bottom-loading bailers are used to extract water
from the base of wells than when standard sampling techniques are used.
Dispersion
Dissolved contaminants emanating from a DNAPL source will be subject
to dispersion, particularly in the direction of ground water flow. Their
concentrations will reduce with time and with distance from the DNAPL
source.
DNAPL
Dissolution
kinetics
Dissolution of contaminants from the DNAPL may occur too slowly in
relation to diffusion or advection of the dissolved phase away from the
DNAPL-water interface to attain the theoretical dissolved concentration
expected under equilibrium conditions. This factor would especially hold
true in settings with naturally high ground water velocities or near
pumping wells.
27
Word-searchable Version Not a true copy
-------
TCE, TCA, PCE, and methylene chloride. For example, 1% of the pure-phase solubility is a much lower
concentration for the least soluble compound, PCE, than for the other three. Table 3-6 also shows the
number of sites in the study reporting each of the contamination levels.
For a multi-component DNAPL, the solubilities of each of the constituents in ground water will
generally be depressed in proportion to the mole fraction of the compound in the DNAPL. These
depressed solubilities are called effective solubilities. For example, in a DNAPL composed of half TCE
and half PCE, the effective solubility of TCE will be 500 mg/L (half of TCEs pure-phase solubility of 1,000
mg/L) and the effective solubility of PCE will be 75 mg/L (half of PCE's pure-phase solubility of 150 mg/L).
Ground water directly in contact with a multi-component DNAPL, then, could contain dissolved
concentrations that are 100% of the effective solubilities of its constituents, but lesser percentages of the
pure-phase solubilities.
28
Word-searchable Version Not a true copy
-------
Table 3-6. Concentrations of Tetrachloroethylene, Trichloroethane, Trichloroethylene, and Methylene Chloride expressed as percentages
of their pure-phase solubilities, and the number of Superfund sites in this study (out of 310) reporting each level of contamination.
Percent of
Pure - Phase
Solubility
100%
50%
10%
3%
1%
0.1%
0.01%
Tetrachloroethylene
(PCE)
ug/L
150,000
75,000
15,000
4,500
1,500
150
15
# sites*
___
9
23
41
52
89
120
1,1,1 Trichloroethane
(TCA)
ug/L
950,000
475,000
95,000
28,500
9,500
950
95
# sites*
___
2
13
25
39
78
101
Trichloroethylene
(TCE)
ug/L
1,000,000
500,000
100,000
30,000
10,000
1,000
100
# sites*
___
9
29
49
78
131
165
Methylene Chloride
(Dichloromethane)
ug/L
13,200,01
6,600,000
1,320,000
396,000
132,000
13,200
1,320
# sites*
)0
0
1
o
11
28
47
* Number of sites reporting this concentration or higher.
Word-searchable Version Not a true copy
-------
All of the factors that contribute to lowering these saturation concentrations in actual ground water
samples for a single-component DNAPL (Table 3-5) also apply to multi-component DNAPLs. Additional
factors that affect ground water concentrations for multi-component DNAPLs are listed in Table 3-7. For
individual sites with several DNAPL compounds present at high concentrations in ground water, the most
suitable method for evaluating site data is to calculate effective solubilities and then express contaminant
concentrations as a percentage of these solubilities. The contaminant concentrations that are generally
accepted to be indicative of multi-source DNAPLs are 1% or more of the compound's effective solubility.
In some cases approximations of the effective solubilities of compounds can be back-calculated from
ground water concentration data obtained from a single sample with high hits of DNAPL chemicals
(Feenstra, 1990).
Table 3-7. Summary of factors that contribute to less-than-saturation concentrations of
dissolved-phase chemicals emanating from a multi-component DNAPL source, in
addition to those listed in Table 3-5.
Factor
Explanation
Initial
DNAPL
Composition
The aqueous solubility of each DNAPL constituent will be depressed in
proportion to its mole fraction in the DNAPL.
DNAPL
Weathering
Over time, a greater mass of the more soluble constituents of the DNAPL will
dissolve into the ground water, leaving behind a DNAPL composed of a lesser
and lesser proportion of the most soluble constituent. These changes in DNAPL
composition will lower the solubilities of the soluble DNAPL constituents, and the
ground water concentrations will reflect these changes.
In this study, a broader representation of site conditions was sought, and therefore the maximum
concentrations of contaminants site-wide and over the entire period of investigation rather than
concentrating on a single sample from a single well was collected. The site analytical data therefore often
(although not exclusively) come from many locations, and many different sample events. By taking this
approach, we have accounted for the likelihood of heterogeneously distributed sources and sample
locations at individual sites (to the extent possible given the constraints of this study). This approach renders
the data unsuitable for the back-calculation of effective solubilities. Instead, all concentration data
30
Word-searchable Version Not a true copy
-------
are expressed as a percentage of the pure-phase solubilities of the compounds.
Multi-component DNAPL sources were accommodated by considering the three DNAPL
compounds found at the highest concentrations in ground water, rather than considering only the highest
contaminant concentration. Three compounds were chosen to keep the method simple and applicable to
the greatest number of sites.
The system for assigning ground water rankings is outlined in Table 3-8. Note that a site with only
one DNAPL compound at 1% of its pure phase solubility is classified with a lower probability of subsurface
DNAPL than a site with three DNAPL compounds at 1% of their solubilities. This is because one would
expect lower contaminant concentrations at sites with DNAPL sources containing three compounds than
at sites with DNAPL sources containing one compound.
Applying the methodology used in this study for developing a ground water contaminant
ranking is straight forward. Effective solubilities are not calculated for this method, one simply
calculates the maximum per cent solubilities for the three DNAPL compounds present at the
highest concentration in the dissolved phase and then applies that information to the Contaminant
Ranking Assignment (Table 3-8). The numerical ranking is then read from the far left-hand
column. The contaminant ranking can then be applied to the matrix table combining the site
history ranking and the ground water contaminant ranking (Table 3-10) for obtaining the overall
likelihood of DNAPL presence at a site.
Word-searchable Version Not a true copy
-------
Table 3-8. Contaminant Ranking Assignment, ranking of sites based on maximum percentage
solubilities of DNAPL Compounds.
Ranking by
Magnitude of
Ground Water
Contamination
Maximum Percentage Solubilities for DNAPL-related
Compounds in Ground Water
Likelihood of
Subsurface
DNAPL
No DNAPL compounds or
One DNAPL compound at < 0.1% or
Two at < 0.03% or
Three at < 0.01%
Low
One DNAPL compound at 0.1% to 1% or
Two at 0.03% to 0.1%
Three at 0.01% to 0.03%
Low
One DNAPL compound at 1% to 3% or
Two at 0.3% to l%or
Three 0.01% to 0.3%
Medium
One DNAPL compound at 3% to 10% or
Two at 1% to 3% solubility or
Three at 0.3 to 1% solubility
High
One DNAPL compound at 10% to 50% or
Two at 3% to 15% solubility or
Three at 1% to 5% solubility
High
One DNAPL Compound at > 50% or
Two at > 25% or
Three at > 15%
Very High
32
Word-searchable Version - Not a true copy
-------
To establish the final ground water contamination ranking, direct DNAPL observations in the
unsaturated zone were noted and DNAPL-related spatial patterns and temporal trends in ground water
contamination were examined. It was assumed that sites with DNAPL observations in the unsaturated zone
have a higher likelihood of subsurface DNAPL than sites without direct observations, so these sites are
raised by one point in the classification system.
In the ranking system, spatial patterns and temporal trends in ground-water contamination are
considered significant only if DNAPL-related compounds are among the major contaminants at the site.
A site is considered to have major DNAPL-related compounds contamination if the ground water ranking
(Table 3-8) is two or greater. At such sites, half a point is added to the site's ground-water contamination
ranking if high concentrations of DNAPL compounds found at depth beneath source areas cannot be
adequately explained by dissolved-phase transport.
"Significanf temporal trends are defined as sharp decreases in concentrations for 3 or more years,
slight increases for 5 or more years, sharp increases for 3 or more years, and steady concentrations for 5
or more years. If any of the last three conditions hold true, half a point is added to the ranking. The
observation of sharp decreases in concentrations over time leads to a half point decrease in the ranking.
In total, these adjustments do not change the ranking of a site by more than one point, and they are
most important for sites with a ground water contaminant ranking of two and three, where the
concentrations are not high enough to place a site definitively in a "high" category. The lesser reliance on
spatial patterns and temporal trends to establish site ranking in part reflects the difficulty in interpreting data
from site investigations that were not specifically designed to characterize these aspects of site
contamination.
Results of Ground Water Contamination Evaluation
As with the site history information, this section separately examines the various indicators that
factor into the ground water contamination ranking and then presents the composite ranking in barchart
form. To give a general feeling for the major DNAPL-related contaminants observed in ground water at
significant concentrations, Figure 3-2 shows the distribution of the DNAPL-related compounds found most
frequently at the maximum concentrations in ground water. The three most prevalent contaminants, TCE,
PCE, and 1,1,1 TCA, are all chlorinated hydrocarbons that are used ubiquitously as industrial solvents. At
the subgroup of Superfund sites evaluated in this study, use of chlorinated solvents and site operations
associated with their use were commonly reported. In terms of the compounds found in ground water, then,
the data match the expectations from site history information.
33
Word-searchable Version Not a true copy
-------
O
<«»
d
i-
0)
40
30
/ft A .
20
10-
0*
x x x x H|\ xxx xn
>>X\HVX\S1
,V/vJKv,VP
/
>VxW
WxV
-Vxv/
VxxxV
\VxV
>VxV/
VxV/
X N % % '
' f t f f
\ \ X \ '
\ S S \ '
Pv/VxV
/
'\'x'x'\'
t f f t t
\ \ \ N
S f f S S
\ \ \ \
VsVx'
V%Vx'
f f f f f
\ \ \ s
V V V \
V%Vx'
'xVxV
Vs'xV
'x'x'x'x'
Vx'xV
Vx'x'x"
xVxVx
'x'x'x'x'
f f f f f
/w/v
vXX'
XvX
'x'x'x'x'
'x'x'x'x'
;%%
X S \ \
f f f f f
\ \ \ \
X X X X
/VxVx
X' -~"1"--^" ' jM
x^'vB
'xxxsxxxxfi
xsxVxvI
xxx xH
Vx'x'xB
xxxNxSxNH
'X-xI
'*'' H .-^__. _
^"j" -"H/j . J
xVx^JKvxvJ / _
x x x xH9ix >.>.>. ^^ t \ \ x \ HI
" " " ^Hh v x " 'Of r j. ».-,A " v " "B
x \ x xBI^ \xx <«B v x x x IK. xxx x |H_
MTCL DCE TCA TCE PCE CLBZ DCB PCB
Contaminant
Figure 3-2. Distribution of the contaminants found most frequently at the highest concentrations (as a
percentage of their pure-phase solubility) in ground water.
34
Word-searchable Version - Not a true copy
-------
The key ground water contamination indicators used to establish DNAPL probability were the
three maximum concentrations of DNAPL compounds as a percentage of their pure phase solubilities.
Figure 3-3 shows the results of applying the ranking system outlined in Table 3-8 to the 40 known DNAPL
sites and to the 270 sites at which DNAPL probability must be estimated.
Most of the known DNAPL sites received the three highest ground water contamination ranking.
For these sites, the system would have predicted a high likelihood of subsurface DNAPL prior to the actual
identification of DNAPL at the site. However, some of the known DNAPL sites receive lower rankings.
At this latter group of sites, site monitoring outside of the DNAPL find did not detect the expected high
concentrations of DNAPL compounds in ground water. This result emphasizes the fact that the ability to
establish DNAPL likelihood based on ground water data is limited by the scope of field investigations. One
group of DNAPL sites, the four that received a ranking of 1, are somewhat atypical in that they are all
creosote/coal tar sites at which DNAPL was found so early in the investigation that all efforts were aimed
at locating and characterizing the free phase and very few ground water samples were taken.
Of the 270 sites at which DNAPL has not been directly observed, 60% received a ranking of three
or greater, signifying a medium to very high likelihood of subsurface DNAPL. Sites in this group with
rankings of 5 or 6 can be considered very likely candidates for subsurface DNAPL. The status of sites with
rankings of 2, 3, and 4 is less clear, and for these sites, the other indicators such as spatial and temporal
patterns of dissolved-phase contamination, can help to estimate DNAPL presence.
35
Word-searchable Version Not a true copy
-------
801
Ground Water Contamination Rank
30-
123456
Ground Water Contamination Rank
Figure 3-3. Distribution of ground water contamination rankings for the 40 known DNAPL sites and
for the 270 sites at which DNAPL probability must be estimated (see Table 3-8 for key
to classes).
36
Word-searchable Version - Not a true copy
-------
3.3 Composite Site Ranking
Once the site history ranking and ground water contamination ranking was established, these two
rankings were combined into a single estimate of DNAPL probability. For purposes of this study, there are
four categories, defined in Table 3-9. This section outlines the method of combining the rankings and
discusses the results of the composite site ranking.
Table 3-9. Definitions of the Four Composite Rankings.
DEFINITE
DNAPL directly encountered below the water table in soil cores and/or ground
water samples.
HIGH
DNAPL strongly suspected based on ground water data and site history
information. Proceed with site investigation and remediation plans assuming
subsurface DNAPL source is present.
MEDIUM
Information from site history and ground water investigation indicate moderate
potential for subsurface DNAPL. Important to gather additional site information
regarding possible DNAPL presence. Best to proceed as if site is a DNAPL site
until further investigations indicate otherwise.
LOW
Based on available site history and ground water information, DNAPL sources are
unlikely. DNAPL potential at some sites in this category may be underestimated due
to lack of information. Modify expectations if further investigation show evidence of
DNAPL sources.
Combining the Site History and Ground Water Rankings
The 40 known DNAPL sites receive a composite ranking of DEFINITE. For the 270 sites at
which DNAPL probability must be estimated, a matrix was developed (Table 3-10) for assigning a high,
medium, or low potential for subsurface DNAPL based on the independent rankings each site received
from the Site History Ranking and the Ground Water Contamination Ranking. Figure 3-4 shows the
distribution of the site history and
37
Word-searchable Version Not a true copy
-------
ground water contamination rankings for the 270 sites. Based on site history information alone, most sites
have a medium to high potential for DNAPL occurrence. The history ranking system has a tendency to rank
sites at the high end of the scale. Based on ground water data, there is a wider range in the potential for
subsurface DNAPL. The ground water contamination ranking system has a tendency to rank sites at the
lower end of the scale. This could, in part, be due to the limited amount of ground water characterization
data available for a site.
In combining the two ranking factors, the greatest emphasis was placed on information carrying the
greatest certainty. Both ranking systems were based on positive indicators of DNAPL occurrence, so the
higher rankings carry greater certainty than the lower rankings. For sites receiving a high ranking based on
ground water data but a low ranking based on site history information, the ground water data prevails
because it more accurately reflects the status of contamination in ground water. For the opposite case, a
high site history ranking and a low ground water contamination ranking, the site history information carries
more weight, particularly when the extent of site characterization is low. For sites with a low ranking on
both counts, there is some potential that a lack of site knowledge is contributing to the low rankings, but
a low match in the rankings can add to the reliability of site information for well characterized sites.
In order to apply the combined ranking system to a site one must first determine the site
history ranking from Table 3-2 and the ground water contaminant ranking from Table 3-8. Using
Table 3-10, locate the intersection point of the site history ranking and the ground water
contamination ranking. Refer to Table 3-9 for the explanation of the combined ranking.
38
Word-searchable Version Not a true copy
-------
40i
2345
Site History Rank
30-1
1 23456
Ground Water Contamination Rank
Figure 3-4. Comparison of Site History Ranking and Ground Water Contamination Ranking for the
270 sites at which the potential for DNAPL occurrence must be inferred.
39
Word-searchable Version - Not a true copy
-------
Site
Hist
Rank
6
5
4
3
2
1
Ground Water Contam. Rank
6
Hi
Hi
Hi
Hi
Hi
Hi
5
Hi
Hi
Hi
Hi
Hi
Hi
4
Hi
Hi
Med
Med
Med
Med
3
Med
Med
Med
Med
Med
Med
2
Med
Med
Lo
Lo
Lo
Lo
1
Med
Lo
Lo
Lo
Lo
Lo
Table 3-10. Matrix for combining the site history ranking and ground water contamination rankings at
sites for which the potential for DNAPL occurrence must be estimated.
40
Word-searchable Version Not a true copy
-------
Results of Composite Ranking
Figure 3-5 shows the results of applying the combined ranking system to the 270 sites for which
we are estimating the potential for DNAPL occurrence. Sixty five (65%) percent received a medium to high
ranking, while 35% have a low potential for DNAPL occurrence. Table 3-9 provides an explanation of
the implications of these rankings.
LOW
95 sites (35%)
xVxVxx-
\x\ss-
XXX/XXXXXXX..^
xVxVxVxVxVxV>
VxVxVxVxWxVxX
XXXSXXXXXXXNXXJ
,'xxVx'x'Vx'\f>f\f\.J'\'\'\f\ .
xxxxxxxxxxxxxxx
HIGH
109 sites (40%)
f*ffrfff*fff*f* r^ f
v'A'X'x'x'x'x'x'x%'x
-------
3.4 Effect of Hydrogeological Setting on DNAPL Occurrence
Transport of DNAPL in the subsurface has been shown to be very sensitive to the geological media
through which it passes (Poulsen and Kueper, 1992). Site geology has the potential to affect many aspects
of DNAPL contamination, including the likelihood that DNAPL will reach the saturated zone, the ultimate
depth of DNAPL transport, the extent of lateral spreading of the DNAPL zone, the likelihood that DNAPL
pools will form, and the spatial distribution of the dissolved-phase plume emanating from a DNAPL source.
These factors in turn affect the ease of site characterization for DNAPLs and the overall potential for site
remediation.
Hydrogeological Categories
Geological information was collected as part of this study so that the relationship of the
hydrogeological setting of sites to the likelihood of subsurface DNAPL could be assessed. For example,
a thick unsaturated zone could offer some protection against migration of the nonaqueous phase to the
water table. As a starting point in the analysis, each site was assigned a hydrogeological setting category
according to those defined by Heath (1984). Table 3-11 names and describes the twelve broadly defined
hydrogeological regions in the United States.
The maj ority of Superfund sites are located in six settings: the Northeast and Superior Uplands, the
Glaciated Central Region, the Non-Glaciated Central Region, the Piedmont-Blue Ridge Region, the
Atlantic and Gulf Coastal Plain, and the Western Alluvial Basins. All of these settings share the common
characteristic of a flat to gently rolling topography. In addition, most U.S. population centers are located
in these six hydrogeological regions. The Non-glaciated Central and Piedmont regions have poor ground
water yields, while the remaining four have relatively abundant ground water resources. Regions such as
the Western Mountain Ranges and Columbia and Colorado Plateaus are more rugged and less populated,
and contain far fewer industries and hazardous waste sites, and were not considered in this study.
Each of the 310 sites evaluated was assigned a hydrogeological setting category based on detailed
geological information. In collecting data on the hydrogeology, concentration was focused on the geological
character of deposits directly beneath source areas, so the category would reflect the nature of the material
through which a DNAPL might have passed. Figure 3-6 shows the distribution of the sites studied
according to hydrogeological setting.
Figure 3-7 shows the results of the site history ranking and ground water contamination ranking as
they relate to sites located in the various hydrogeologic regions. These results indicate that no single
hydrogeologic setting has a significantly greater likelihood of subsurface DNAPL than another.
42
Word-searchable Version Not a true copy
-------
Table 3-11. Descriptions of the Hydrogeological Settings for sites studied, Heath (1984).
Hydrogeological Setting
for Sites Studied
Western Alluvial Basins
Non-glaciated Central
Region
Glaciated Central Region
Piedmont and Blue Ridge
Region
Northeast and Superior
Uplands
Atlantic and Gulf Coastal
Plain
Description
Thick alluvial (locally glacial) deposits in basins and valleys
bordered by mountains
Thin regolith over fractured sedimentary rocks
Thick glacial deposits over fractured sedimentary rocks
Thick regolith over fractured crystalline and metamorphosed
sedimentary rocks
Thick glacial deposits over fractured crystalline rocks
Complex interbedded sands, silts, and clays
Notes: Superfund sites are generally concentrated in the six highlighted regions Health's cutoff for
"thick" vs "thin" deposits is 5 meters
43
Word-searchable Version Not a true copy
-------
100
» X X X X
f f f
. N S N X
f t f f
L \ \ X \
f f f t
> N N N N
f f f t
. V S \ N
f f f f
. X \
f f f S
, X X XX
'x'x'x'x
'x'x'x'x
'x'x'x'x
X X X X
f
X X X X
f t f
X X XX
f t f f
X X X X
f f
X X X X
t t t
X X X X
f f t f
X X X X
NESU GLCR NGCR PMBR AGCP WALB
Figure 3-6. Distribution of the 310 sites of this study according to Hydrogeological Setting.
Refer to Table 3-11 for explanation of settings.
44
Word-searchable Version Not a true copy
-------
Site History Rank
^t
S3
S
00
6"
5'
4'
3-
2'
i
*777*-,
.. \ \ \
s s / s
k X X X
s / s s
b X \ \
S S S f
f t t f
b \ \ X
t / s s
<. X \ X
i, X X X
S S / S
V\Vx
VVVx
k X X X
k X X X
S f / S
k X X X
t S f /
X X X X
f S f t
\ \ X \
/ S f f
S / f /
X X X X
/ S f f
X X X X
xVxV
,v,v
X S \ \
f f f /
\ \ N S
S X N \
- S X \ \
X XX
X XX
XXX
X XX
XXX
X XX
f f f
X XX
X X X X
- X X X X
XXX
XXX1
XXX
XXX
X X X *
XXX
X X X <
, X X X '
XXX
, X X X
XXX
XXX'
XXX
XXX
X X X
- XXX'
XXX
X XX
f f f
XXX
X XX
f f f
X X
' Vx
f f f
X X
XXX
' 'X
X. S \ \
- \ \ \ S
\ s \ s H
' ' ' 'B
x / x xH
x x x x H
f f f xra
x x x x a
xxx n
f f f XHJ
s'v' '1
x" ' '1
xxx n
/ t / s&
S S f XH
x X x H
x's'x' "1
- Xv 'I
NESU GLCR NGCR PMBR AGCP WALB
Hydrogeological Setting
Ground Water Contamination Rank
I
6-
s-
4-
3-
2-
i
&%'
/VVV
xvxv
-///:
X XX X
X X X >
x x x *
'XXX.
X X X X
'XXX,
X X X X
'XXX,
'x'x'x'x
x/VV
'XXX
X X X X
x'x'x'x
'x'x'x'x
f t f t
xxx
X XX
xxx
X XX
X XX
k X X X
X XX
- V X X X
t f e t
XNXXXXXX
X X X «i
/ t t f
X X X >
X X X >
t t f f
XXXV
'x'x'v^
xVxV
X X X >
f f f f
. X X X N
X X X X
x x'x'x
x x'x'x
X
X X X X
X X X X
/x'/x'
x'x'x'x
x'x'x'x
s. x x x
/x'x'x'
/x'x'x'
f f f f
< X X X
S. X X X
«b X X X
/x'x'x'
f f f f
VSxVx
V X X X
f f t f
-V X X X
NESU
GLCR NGCR PMBR AGCP WALB
Hydrogeological Setting
Figure 3-7. Results of the Site History and Ground Water Contamination Ranks as Related
to Hydrogeologoic Setting
45
Word-searchable Version-Not a true copy
-------
3.5 Relationship of Site Use to DNAPL Occurrence
The former use of a Superfund site has bearing on the ability to predict DNAPL contamination as
well as on the likelihood of DNAPL occurrence. Site use also affects the nature of the site contamination,
the remedial options, and the degree of difficulty of site remediation. To explore the association between
site use and DNAPL occurrence, site uses were divided into nine major categories:
o Organic Chemical Production
o General Manufacturing
o Industrial Waste Management
o Combination Landfill
o Federal Facility
o Wood Treatment
o Inorganic Chemical Production
o Metal Industry/Mining
o Miscellaneous
Figure 3-8 shows the site use distribution for the 310 sites evaluated, and the specific site uses associated
with these categories are listed in Table 3-12. The proportion of sites with direct observations of DNAPL
below the water table are shown in a darker shade. The site uses where DNAPL observations were most
commonly reported are wood treaters, organic chemical producers, and industrial waste managers.
Figure 3-9 shows the average site history rankings and ground water contamination rankings for
each site use. Wood treatment operations received high marks for both rankings, partly due to the large
percentage of known DNAPL sites in this group of sites. After wood treatment, organic chemical
producers, industrial waste sites, and general manufacturing sites have the highest observed percentage
solubilities and number of known DNAPL sites, and thus the highest ground-water rankings. In summary,
the site findings indicate that some site uses will have a greater likelihood of subsurface DNAPL that others.
Those with the highest probability are: wood treatment sites, organic chemical production sites, general
manufacturing sites, and industrial waste disposal sites.
46
Word-searchable Version Not a true copy
-------
O
i.
E
3
270 Sites
40 DNAPL sites
Org Man IndW LF Fed Wood Inorg Met Oth
Site Use
KEY:
ORG - organic chemical production: MAN - general manufacturing: INDW - industrial waste
management: LF - combination landfill: FED - federal facility: INORG - inorganic chemical production:
MET - metals industry/mining: OTH - miscellaneous
Figure 3-8. Site use distribution for the 310 sites.
47
Word-searchable Version Not a true copy
-------
Table 3-12. Major Categories of Site Uses.
Organic Chemical Production
Asphalt production or distribution plant
Coal gasification facility
Coal mining
Coking operations (steel industry)
Oil and gas mining
Oil storage (fuel oil, etc)
Organic chemical manufacturing
Organic chemical packaging, distribution and storage
Paint and dye production
Pesticide distribution, packaging, and transport
Pesticides and herbicide production
Petroleum refining and related industries
Pharmaceutical manufacturing
Resin and glue manufacturing
Solvent manufacturing
Solvent packaging distribution, transport and recycling
Synthetic fiber production
Transformer oil production
General Manufacturing
Agricultural equipment manufacturing
Air craft manufacturing
Air craft maintenance and repair
Automobile and motorcycle manufacturing
Automobile body repair or paint shop
Battery manufacturing
Capacitors and transformers manufacturing
Ceramics manufacturing
Construction company
Electronics and electrical equipment manufacturing
Engine manufacturing
Engine repair
Fabricated metal product manufacturing
Food manufacturing, packaging, and distribution
Lumber and wood products manufacturing
Other manufacturing
Paper and allied products manufacturing
48
Word-searchable Version Not a true copy
-------
Table 3-12. Major Categories of Site Uses (continued)
General Manufacturing (continued)
Plastics manufacturing
Printing or publishing facility
Rubber products manufacturing
Tannery
Textile mill
Textile printing and processing
Tool and die manufacturing
Weapons and explosives manufacturing
Weapons maintenance and repair
Industrial Waste Management
Drum reconditioning facility
Industrial landfill
Liquid industrial/hazardous waste disposal
Liquid industrial hazardous waste incinerator
Liquid industrial hazardous waste storage and transport
Liquid industrial hazardous waste treatment
Midnight dumping
Petroleum-related waste disposal
Solvent recycling
Transformer reprocessing and/or recycling
Waste oil processing, storage, transport
Landfill
Combination municipal and industrial landfill
Other Waste Facilities
Municipal landfill
Publicly owned sewage treatment works
Recyclers of solid waste
Septic services
Solid waste incineration facility
Tire disposal facility
Waste storage and transfer facility
Waste transportation
49
Word-searchable Version Not a true copy
-------
Table 3-12. Major Categories of Site Uses (continued)
Federal Facilities
Department of Energy facility
Military Base
Wood Treatment
Wood preservation plant
Inorganic Chemical Production
Asbestos manufacturing
Chemical manufacturing (unspecified)
Chemical mixing and batching (unspecified)
Fertilizer manufacturing
Inorganic chemical manufacturing
Inorganic chemical packaging, distribution, and storage
Inorganic waste processing
Non-metallic mineral mining
Metal Industry/ Mining
Battery recycling
Electroplating facility
Metals mining
Metal recycling
Ore mill
Primary metals industry
Salvage/scrap yard
Miscellaneous
Airport
Dry cleaning plant
Fire-fighter training area
Nuclear power plant, radiation lab, etc
Power plants (non-nuclear) and associated facilities
Railroad yard and rail car maintenance facility
Research laboratory, agricultural station, or similar facility
Unknown
50
Word-searchable Version - Not a true copy
-------
Site History Rank
*N
£
o>
-------
3.6 Site Contaminant Type and DNAPL Occurrence
The hazardous substances that have been used, stored, or disposed of at Superfund sites vary
widely in their compositions and physical and chemical properties. In evaluating the potential for subsurface
contamination by DNAPLs, it is useful to group sites with similar contaminants. Based on the site types
encountered during this study, and on DNAPL site groupings suggested by Cherry and Feenstra (1991),
eight categories of contaminant types were established for which one would expect distinctive types of
subsurface contamination.
N inorganic chemicals
N light petroleum products
N chlorinated solvents
N mixed industrial solvents
N creosote
N coal tar
N PCB oil/solvent
N other organic compounds (including pesticides)
Site history information (type of chemicals used or stored at the site over its history of operation)
and site characterization information (key ground water contaminants) were used to assign contaminant type
categories. The first two categories have a relatively low DNAPL likelihood. The remaining six site types
all have a significant potential for subsurface contamination by DNAPL chemicals. Figure 3-10 shows the
distribution of site contaminant types for the 310 sites of this study.
Inorganic Chemical Sites
Inorganic element sites are those at which no organic contamination of ground water has been found
and for which the key site uses are thought to have generated only inorganic chemicals. This study included
two sites that indicated minor organic contamination in soils, but in general these were excluded.
Light Petroleum Product Sites
Light petroleum product sites are those at which the only hazardous substances used on site were
lighter than water, and for which little or no DNAPL-related compounds have been found in ground water.
One example of a Superfund site use in this category was a rubber manufacturing plant, which had other
DNAPL compounds present.
Chlorinated Solvent Sites
These are sites at which the main contaminants are chlorinated solvents. Product
52
Word-searchable Version Not a true copy
-------
200
LPET MIXSOLVCLSOLVPCBSOLV CREO CTAR OTHER
Contaminant Type
Key:
LPET - Light petroleum products
CLSOLV - Chlorinated solvents
MIXSOLV - Mixed industrial solvents
PCBSOLV - PCB oil/ solvents
CREO - Creosote
CTAR - Coal tar
OTH - Other organic compounds
Figure 3-10. Distribution of the 310 sites evaluated according to site contaminant type.
53
Word-searchable Version - Not a true copy
-------
or waste streams, or ground water contamination composed entirely of one or more of these compounds
place a site in the chlorinated solvents category. Ninety-eight sites, or nearly a third of the sites received
this designation.
Since solvent use is common among a wide range of industries, there are numerous site uses
associated with this site contaminant category. Figure 3-11 shows the site use distribution of the 98
chlorinated solvent sites evaluated. Table 3-13 lists the frequency of detection of chlorinated solvents at
greater than 0.01% of solubility at these sites.
Org Man IndW LF Fed Inorg Met Oth
Site Use Category
ORG - organic chemical production: MAN - general manufacturing: INDW - industrial waste management:
LF - combination landfill: FED - federal facility: INORG - inorganic chemical production: MET - metals
industry/mining: OTH - miscellaneous
Figure 3-11. Site Use Distribution for the 98 Chlorinated Solvent Sites.
54
Word-searchable Version - Not a true copy
-------
Table 3-13. Compounds reported at > =0.01% solubility in ground water at the ninety-eight
chlorinated solvent sites.
Compound
Carbon Tetrachloride
Chlorobenzene
Chloroform
Dichlorobenzene, 1,2-
Dichloroethane, 1,1-
Dichloroethane, 1,2-
Dichloroethylene, 1,1-
Dichloroethylene, cis 1,2-
Dichloroethylene, trans 1,2-
Methylene Chloride
Tetrachloroethylene
Trichloroethane, 1,1,1-
Trichloroethane, 1,1,2-
Trichlorethylene
Trichlorofluoromethane, 1,1,2
Form and color of
Pure Product
Colorless liquid
Colorless liquid
Clear liquid
Colorless liquid
Colorless oily liquid
Colorless oily liquid
Colorless liquid
Colorless liquid
Colorless liquid
Colorless liquid
Colorless liquid
Colorless liquid
Colorless liquid
Clear or blue liquid
Colorless liquid
No. of
Sites
2
4
1
5
14
5
30
9
31
11
42
40
5
76
1
% Sites
2
4
1
5
14
5
31
9
31
11
43
41
5
78
1
Mixed Industrial Solvent Sites
These are sites at which a great range of DNAPL compounds have been used, stored, or disposed
of, but site contaminants are generally dominated by a few chlorinated solvents. They may also contain
BTEX compounds, pesticides, and poly-nuclear aromatics, and phenols. One hundred fifty-five sites in this
study received this designation. Figure 3-12 shows the distribution of site uses associated with mixed
solvent sites. In general, industrial waste management sites and landfills receive the widest range in waste
materials and account for the greatest number of these sites.
Table 3-14 shows the main chemicals observed at the 155 mixed industrial solvent sites and their
frequency of detection at >0.01% of solubility. The compounds seen most frequently were the
monoaromatic hydrocarbons and chlorinated solvents.
55
Word-searchable Version - Not a true copy
-------
en
o>
C
0)
M
"o
CO
o
O)
o
i-
Org ManlndW LF Fed Inorg Met Oth
Site Use Category
Figure 3-12. Site use distribution for the 155 sites in the mixed industrial solvents category.
Refer to Figure 3-11 for key.
56
Word-searchable Version - Not a true copy
-------
Table 3-14. Main Compounds reported at >0.01% Solubility in Ground Water at Mixed Industrial
Solvent Sites.
Compound
Light Petroleum Products:
Benzene
Ethylbenzene
Styrene
Toluene
Xylenes
Chlorinated Solvents:
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
1,2 Dichlorobenzene
1,4 Dichlorobenzene
1,1 Dichloroethane
1,2 Dichloroethane
1,1 Dichloroethylene
cis-1,2 Dichloroethylene
trans- 1,2 Dichloroethylene
Methylene Chloride
1,1,2,2 Tetrachloroethane
Tetrachloroethylene
1,1,1 Trichloroethane
l,l,2Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Vinyl Chloride
Pesticides:
1,2 Dichloropropane
Ethylene Dibromide
# sites
60
80
7
105
78
7
31
12
13
18
18
40
21
33
10
61
34
6
73
59
8
82
4
52
8
3
% sites
39
52
5
68
51
5
20
8
8
12
12
26
14
21
6
39
22
4
47
38
5
53
3
34
5
2
57
Word-searchable Version Not a true copy
-------
Table 3-14. Main Compounds reported at >0.01% Solubility in Ground Water at Mixed Industrial Solvent
Sites (continued).
Compound
Poly Nuclear Aromatics and Phenols:
Acenapthene
Benzo(a)anthracene
Chrysene
Dimethylphenol, 2,4-
Fluoranthene
Fluorene
Methyl naphthalene, 2-
Naphthalene
Pentachlorophenol
Phenanthrene
Phenol
Pyrene
Other:
Acetone
Benzoic acid
Dibenzofuran
Isophorne
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Tetrahydrofuran
# sites
6
5
5
6
5
6
11
28
4
6
3
8
3
4
1
4
20
17
2
% sites
4
3
O
4
3
4
7
18
3
4
2
5
2
3
1
3
13
11
1
Creosote/Coal Tar Sites
Creosote sites are a relatively small and distinctive group in the Superfund program. They are
related to only two site uses: wood preservation, and creosote production. Of the 15 sites in this category,
10 have had direct observations of creosote DNAPL in the saturated zone, and six reported LNAPLs
floating on the water table. Contaminants typically found in ground water at creosote sites are listed in table
3-15.
Coal tar sites are generally associated with coal gasification or coal tar production operations.
Contaminants typically found in ground water at coal tar sites are listed in table 3-16. Creosote and coal
tar sites should be considered as definite DNAPL sites.
58
Word-searchable Version - Not a true copy
-------
Table 3-15. Compounds Found at > 0.01% Solubility in Ground Water at Creosote Sites
Compound
Benzene
Ethylebenzene
Toluene
Xylenes
Acenapthene
Anthracene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Chrysene
o-cresol
p-cresol
Dibenzo(a,h)anthracene
Dibenzofuran
Dimethylphenol, 2,4-
Fluoranthene
Fluorene
2-Methyl Napthalene
Napthalene
Pentachlorophenol
Phenanthrene
Phenol
Pyrene
# sites
3
5
5
6
12
6
1
1
3
7
1
2
1
3
2
8
11
9
12
9
12
1
9
% sites
20
33
33
40
80
40
7
7
20
47
7
13
53
20
13
53
73
60
80
60
80
7
60
59
Word-searchable Version Not a true copy
-------
Table 3-16. Compounds Found at > 0.01% Solubility in Ground Water at Coal Tar Sites
Compound
Benzene
Ethylbenzene
Styrene
Toluene
Xylenes
Acenapthene
Anthracene
Chrysene
Dimethylphenol, 2,4-
Fluoranthene
Fluorene
2-Methyl Napthalene
Napthalene
Phenanthrene
# sites
6
5
3
5
5
5
3
3
3
4
4
5
5
3
% sites
75
63
38
63
63
63
38
38
38
50
50
63
63
38
PCB/Solvent Sites
PCB contamination usually encompasses a class of chlorinated compounds that includes up to 209
variations or congeners with different physical and chemical characteristics. They were commonly used as
mixtures called Aroclors. The most common are Aroclor-1254, Aroclor-1260, and Aroclor 1242. PCBs
alone are not usually mobile. However, they are often found with oils, which may carry the PCBs as a
separate phase. PCBs are most commonly associated with electrical transformer manufacturing, salvage,
and recycling site uses. Table 3-17 shows the DNAPL compounds found at PCB sites in this study.
Relationship Between Contaminant Type to the Likelihood of Subsurface DNAPL
Figure 3-13 shows the relationship of contaminant type to the likelihood of DNAPL presence for
the site history ranking and ground water contamination ranking. The results of this study indicate that
certain contaminant types can be directly associated with a medium to high probability of subsurface
DNAPLs. Those that continuously received a high ranking include creosote, coal tar, and PCBs. However,
these sites tend to represent a
60
Word-searchable Version Not a true copy
-------
small proportion of Superfund sites, are easily linked to specific site uses and tend to have a relatively small
impact in terms of volume of subsurface contamination (when compared to the solvent sites).
In addition, the chlorinated solvent and mixed solvent sites that represent the maj ority of Superfund
sites, are associated with a wide range of site uses, and cover the entire range of likelihood of subsurface
DNAPL. However, current research indicates that they have a relatively large impact in terms of volume
of subsurface contamination.
Conclusions
Figure 3-14 shows the extrapolation of the results of this study to current universe of sites on the
NPL. Approximately 60% of NPL sites either have, or could be expected to have a medium to high
potential of having DNAPLs present, providing a source of ground water contamination in the subsurface.
The remainder of sites could be expected to fall within the category of "low to unlikely." This means that
the potential for subsurface DNAPL should be considered at the majority of Superfund sites. Site
characterization efforts should focus on determining the potential of DNAPL presence early in the site
investigation process.
In order to extrapolate the results to the entire universe of NPL sites, four categories of site
conditions were established (listed below). Ground water contamination information for each NPL site was
evaluated and each site was assigned to one of the four categories. The sources of ground water
contaminant information included the NPL Site Characterization Database, the NPL Summary Booklets,
RPMs, remedial investigation reports, and other site documents. For the five regions that were visited for
this study, all site contaminant information was verified. The sites in the remaining five regions were assigned
to a category based on the information obtained from the sources just mentioned. The results were as
follows:
N observation of DNAPLs below the water table (5%)
N organic contaminants in ground water, but no DNAPL observation (80%)
N only inorganic contaminants in ground water (10%)
N no contaminants in ground water (5%)
At the top of the spectrum are 5% of sites for which DNAPL contamination has been established
with certainty. At the bottom of the spectrum, the 10% of sites with inorganics only and the 5% with no
ground water contamination, are those at which DNAPLs can be ruled out. The remaining 80% are those
at which no DNAPL has been observed, but organic contaminants are present in the dissolved phase and
thus, have some potential for DNAPL contamination.
61
Word-searchable Version Not a true copy
-------
The 270 sites which were ranked via the methods discussed in this study were drawn from the pool
of NPL sites with organics in the ground water. They make up a little over 25% of that group. This subset
of 270 sites was representative of the distribution of site use types of the overall 80% of the sites exhibiting
organic contamination. Therefore, since the distributions were the same, a high degree of confidence is
placed on the direct extrapolation of the proportions from the subset of 270 sites to the sites with organics
present in ground water. The final percentages of sites falling within the high, medium, and low probabilities
were calculated by adding those sites with DNAPLs observed, only inorganics present, and no ground
water contamination back into the set of all sites. Therefore, the percentage of sites in the high, medium,
or low categories are lower for the set of all NPL sites that for the subset of sites. The results of this study
suggest that it is important for any future refinements of policies for investigating and addressing
contaminated ground water at Superfund sites to consider carefully the likely presence of DNAPLs.
The site historical information ranking system correlated well with the information from the sites
known to have DNAPLs present. The historical information focused on site use, past disposal practices
and release of DNAPL compounds throughout the period of site operation. This type of information can
yield important direct and indirect evidence that DNAPL have been released. However, the lack of such
information does not constitute evidence that DNAPL were absent at a site.
The ground water contaminant ranking system (expressed as a per cent of maximum solubility) also
correlated well with information from the sites known to have DNAPLs present. The presence of a
DNAPL compound in ground water is one of the best indirect indicators of the likelihood of DNAPL
presence. The presence of dissolved-phase DNAPL in ground water does not confirm the presence of a
pure-phase DNAPL source in the subsurface. However, the concentrations that are now generally
accepted by the research community as indicating a high likelihood of a subsurface source of DNAPL,
across a wide range of site types, are on the order of 1% or more of a compound's solubility.
The analysis of hydrogeologic setting on DNAPL occurrence indicated that there was no
identifiable hydrogeologic setting that had a greater likelihood of exhibiting subsurface DNAPL than
another. In addition, dissolved-phase DNAPL contamination was just as likely to be present in aquifers
with a deep vadose zone as those with a shallow water table.
The relationship of site use to DNAPL occurrence was evaluated in order to determine if certain
site uses (site types) exhibited a greater likelihood for subsurface DNAPL than others. The results indicated
that indeed, certain site types continuously ranked as having a high likelihood of DNAPLs present. Those
with the highest likelihood of having DNAPLs include: wood-treating sites, general manufacturing sites,
organic chemical productions sites, and industrial waste landfills.
The relationship between site contaminants and DNAPL occurrence was evaluated
62
Word-searchable Version Not a true copy
-------
in order to determine if there were certain suites of compounds present at concentration levels above their
theoretical maximum solubilities that would exhibit a higher likelihood of subsurface DNAPLs than at sites
where that situation does not exist. The results correlate well with the types of DNAPL compounds
associated with specific site types. The contaminants most directly associated with DNAPL presence
included: creosote compounds, coal tar compounds, Polychlorinated Biphenyls (PCBs), chlorinated
solvents, and mixed solvents. However, even though creosote, coal tar, and PCB sites were easily linked
with specific site uses, and have a relatively high likelihood of subsurface DNAPL, they represent only a
very small proportion of the universe of NPL sites. The majority of NPL sites exhibit chlorinated and mixed
solvent contaminants present in ground water. These sites are more difficult to assess because they are
associated with a wide range of uses.
The results of this study also suggest that the emphasis of future research efforts for ground water
remediation, emphasis should be placed on chlorinated solvents and mixed solvents sites, as these represent
the majority of sites having DNAPL-related compounds present as a separate phase and as a source of
dissolved-phase ground water contamination.
63
Word-searchable Version Not a true copy
-------
Table 3-17. Compound found in ground water at PCB/Solvent sites
Compound
Benzene
Ethylbenzene
Toluene
Xylenes
Dichlorobenzene, 1,2-
Dichlorobenzene, 1,3-
Dichlorobenzene, 1,4-
Trichlorobenzene, 1,2,3-
Trichlorobenzene, 1,2,4-
Trichlorobenzene, 1,3,5-
PCB 1242
PCB 1248
PCB 1254
PCB 1260
PCBs (total)
Tetrachloroethylene
Trichloroethylene
Color and Form of Pure
Product at Room
Temperature
Colorless liquid
Colorless liquid
White volatile crystals
Platelets
Colorless liquid
Crystals
Clear, colorless oil
Colorless oil
Light yellow viscous liquid
Yellow soft sticky resin
Yellowish oily liquid
Number
of Sites
2
3
4
3
2
3
3
3
5
1
2
1
5
6
3
3
5
Percent
of Sites
18
27
36
27
18
27
27
27
45
9
18
9
45
55
27
27
45
64
Word-searchable Version Not a true copy
-------
Site History Rank
CO
DJD
61
5'
4'
3
2"
1-
xxx -B
x x x >
VxV.
xxx
xxx
VxV
f f f f
xxx
f f f f
xxx
VxV
VxV
xxx
VxV
f f f /
.XXX
k'xxx'x'
f f f 1
kxV*x\
k. X X X
f f f *
f S S S
y .
^ X X X
f / f .
^ X X X
t f t <
/ / s .
v.'x'x'x
f f f
\ X X X
f S S
x'xVx
S S f ,
s s s
t f f
xVx'x
x'xVx
s s s
VxV
\ X X *
' S S S
\ X X *
' S S t
-VxV
r * f f
X X X ^
-VxV
LPET MIXSOLV CLSOLV PCBSOLV CREO
CTAR OTHER
Contaminant Type
Ground Water Contamination Rank
1
£
o
4
6
5'
4'
3-
2-
WT*
X \ X
xS'Vx'
'X*
f f f f
XXX
VxV
'x'x'x'
VxV
VxV
f f f f
f f f /
/Xv
A-;-;-
S / * j
'xVA
'/Vx\
k X X X
f f f 4
'xVx\
b X X X
f f f f
f f f
X \ X X
f f f
X X X X
f f f
X X X X
*x*
X X X X
f f f
m
XX X
f f f
:* -;
S / f
% *
VxV.
VxV.
xVx'x
^
X X X X
X X X X
f f f
X X X X
f f f
VxV
Xvx
f f f
VxV
^77^
VxV.
fxVx",
K x / x
1 X X X ^
K X X /
LxxW
[N X X ^
K x'x'-
K x"x'<
I x x
1 x x
^ x x x
r x x x >
XXX
xxxVx
;v>:
xxxVx
X \ X
XXX
';
xV,Nx
VxV
LPET MIXSOLV CLSOLV PCBSOLV CREO CTAR OTHER
Contaminant Type
Figure 3-13. Relationship of Contaminant Type to Likelihood of Subsurface DNAPL.
65
Word-searchable Version - Not a true copy
-------
NPL Sites
310 Sites
Studied
DNAPL
Observed
(5%)
High Potential
for DNAPL
Occurrence
(32%)
Medium Potential
for DNAPL
Occurrence
(20%)
Low Potential
for DNAPL
Occurrence
(27%)
Definite
40 Sites
High
(40%)
Medium
(25%)
270 Sites
Unlikely
Only Inorganics
in ground water
or No ground
water contamination
(16%)
Figure 3-14. Extrapolation of the Study Results to the Universe of NPL Sites.
66
Word-searchable Version - Not a true copy
-------
REFERENCES
Anderson, M.R., R.L. Johnson, and J.F. Pankow. Dissolution of Dense Immiscible Solvents into
Ground Water: Laboratory Experiments Involving a Weil-Defined Residual Source. Submitted
to Ground Water, in press.
Cherry, J.A., S.Feenstra, 1991. Identification of DNAPL Sites: Eleven Point Approach, Draft
Document, Waterloo Center for Ground Water Research, University of Waterloo, Kitchner,
Ontario.
Cohen, R.M., J.W. Mercer, 1993. DNAPL Site Evaluation: CRC Press, Inc. Boca Ralton, FL
Feenstra, S., D.M. Mackay, and J.A. Cherry, 1991. A Method for Assessing Residual NAPL Based
On Organic Chemical Concentrations in Soil Samples. Ground Water Monitoring Review,
Spring 1991, 128-136.
Feenstra, S., 1990. Evaluation of Multi-Component DNAPL Sources by Monitoring of
Dissolved-Phase Concentrations. Presented at the International Association of Hydrogeologists
Conference on Subsurface Contamination by Immiscible Fluids, Calgary, Alberta, April 18-20.
Harman, J., D.M. Mackay, and J.A. Cherry, 1993. Goals and Effectiveness of Pump and Treat
Remediation Final Draft Pre-print.
Heath, R.C., 1984. Ground Water Regions of the United States. USGS Water Supply Paper 2242.
Ruling, S.G and J.W. Weaver, 1991. Dense Nonaqueous Phase Liquids. U.S. EPA/540/4-91/002,
21pp.
Kueper, B.H. and D.B. McWhorter, 1991, The Behavior of Dense Nonaqueous Liquids in Fractured
Clav and Rock. Ground Water, 29(5): 716-728.
Mackay, D.M. and J.A. Cherry, 1989. Ground Water Contamination: Pump-and-Treat Remediation.
Environmental Science and Technology, 23(6): 620-636, ACS.
Poulson, M. and B.H. Kueper, 1992, A Field Experiment to Study the Behavior of Perchloroethylene
in Unsaturated Porous Medium. Environmental Science and Technology, 26(5): 889-895,
ACS.
U.S. EPA, Estimating the Potential for Occurrence of DNAPL at Superfund Sites. EPA Publication
9355.4-07FS, December 1991.
Word-searchable Version Not a true copy
-------
U.S. EPA, Evaluation of Ground Water Extraction Remedies: Phase H EPA Publication 9355.4-05,
February 1992.
U.S. EPA, NPL Site Characterization Database. Computer Database, EPA, Hazardous Site
Evaluation Division.
U.S. EPA, CLP Analytical Results Database rCARDY
Word-searchable Version Not a true copy
-------
APPENDIX
Word-searchable Version Not a true copy
-------
DNAPL SITE ASSESSMENT STUDY
********** Please use blue or red ink when filling out this form thank you! **********
Site name:
EPA ID:
Site RPM:
RPM telephone number:
Your name, position, and
telephone number
(if not RPM)
For how many years have you been
involved with this site?
State:
At what state in the Superfund process is this site
(RI in progress, RD, RA, etc.), especially with respect
to ground water contamination?
Does this site have organic chemical contamination?
Sites without organic chemical contamination:
Yes
Maybe No Unknown
Does the site have groundwater contamination
with inorganic chemicals? Yes
Please fill out only section 1 A.
of this form (pg. 1).
Sites with known or possible organic chemical contamination:
Maybe No Unknown
Specifically, is ground water at the site
contaminated with organic chemicals?
Please fill out the rest of this form.
Yes
Maybe No Unknown
Word-searchable Version Not a true copy
-------
DNAPL SITE ASSESSMENT STUDY
REGION 9
Organization of the Data Collection Form
1. Site History Page
A. Site Use 1
B. Hazardous Substances and Chemicals 2
C. Releases of Hazardous Substances and Chemicals 3
D. Additional Comments 6
2. Site Investigation
A. Observation of Subsurface NAPLs 7
B. Contamination of Ground Water 9
C. Extent of Field Investigation 15
D. Additional Comments 16
3. Background Site Information
(This section to be filled out by project hydrogeologist)
A. Geologic and Hydrogeologic Setting 17
B. Plume Information 19
4. References and Final Comments
A. Reference Documents 20
B. Respondent Opinion on Possibility of DNAPLs 21
C. Comments on Survey 22
Word-searchable Version Not a true copy
-------
1. Site History Section
LA. Site Use
If the activity associated with the contamination at this site is
completely unknown, please check here and proceed to Section 2
(Site Investigation Section, p. 7).
What were the major uses of this site? Please record the key site activity(ies) (e.g. combination
municipal and industrial landfill, computer chip manufacturing, wood preservation, solvent recycling)
and the period during which the activity occurred (i.e. 1952 - 1975), to the best of your knowledge.
Activity Period
(years^
Start Stop
Is this site a multi-source site (that is, does it have a
number of distinct source facilities, such as an industrial
park, or is it a very large facility with multiple source
areas, such as a military base)? Yes No Unknown
If yes, you may choose to answer the questions on this form with respect to only one or a few source
areas that are most likely to have DNAPL on (for instance, areas with chlorinated solvent disposal). See
the project hydrogeologist for more explanation.
Please add any comments you would like to make on historical site uses.
HISTCMT1
Word-searchable Version Not a true copy
-------
l.B. Hazardous Substances and Chemicals
Please circle the abbreviations for those substances which are known to have been used, produced, stored,
or disposed of at the site in significant quantities (> 5 drums/year). The substances and chemicals listed
below are related to NAPLs to varying degrees (those marked with a " *" are strongly related to DNAPLs).
AUFL
BTRY
CPTF *
CTAR *
CREO *
CUTO
FOIL
GASO
ISEW
KERO
LABC
OCHM
OCWS *
PNTS
PCBO *
TPCB *
PCP *
PEST *
PRPL
SOLC *
SOLN
SOLV *
STEM
TRNF *
UMOL
WOIL
Automotive fluids (transmission, brake, etc.)
Batteries and/or associated wastes
Capacitor and/or transformer debris (PCB-related)
Coal tar
Creosote
Cutting oils
Fuel oils
Gasoline
Industrial sewage
Kerosene
Laboratory chemicals and/or wastes
Organic chemicals (besides PCBs and solvents)
Organic chemical waste drums and/or containers
Paints, lacquers and/or pigments
PCB-laden oils
PCBs
Pentachlorophenol
Pesticides and/or herbicides
Propellants, jet fuel
Solvents:
Chlorinated
Non-chlorinated
Undifferentiated
Still and/or tank bottoms
Transformer oil
Used motor oil
Waste oils
Please list any other hazardous substances or chemicals that are known to have been used, produced,
stored, or disposed of in significant quantities at the site. If you list chemicals, please indicate only those
chemicals for which records or other knowledge of historical site use exist, not chemicals whose historical
presence is inferred from their current presence as site contaminants:
OTHERSUBST
Word-searchable Version Not a true copy
-------
l.C. Releases of Hazardous Substances and Chemicals
This section documents the potential mechanisms for release of hazardous chemicals and substances at the
site.
I.C.I Industrial Practices
Please circle the abbreviations for any industrial practices which have occurred at this site. These practices
typically use DNAPL chemicals and have a moderate to high probability of historical DNAPL release.
ELCL Electrical parts and electronics cleaning
FFTA Fire fighter training
MTCL Metal cleaning and degreasing
MTMC Metal machining
PTST Paint and lacquer stripping (of furniture, etc.)
S AST Storage of solvents in aboveground tanks
SUST Storage of solvents in underground storage tanks
SDRM Storage of drummed solvents in uncontained areas
SLUL Solvent loading and unloading
TLDI Tool-and-die operations
TRNF Transformer breaking or recycling
WDPR Wood treatment
Please list below any other industrial practices which may have used DNAPLs (chlorinated solvents, coal
tar, creosote, PCB-laden oils) and possibly caused their release at this site:
OTHERINDP
Word-searchable Version Not a true copy
-------
l.C.2. Waste Disposal Practices and Unintentional Releases
Please circle the abbreviations for those waste disposal practices and unintentional releases which occurred
at this site. The practices which indicate a high probability of DNAPL release (assuming DNAPL
substances were present on-site) are marked with a "*".
Waste Disposal Practices
DRUM
LAGO
LWIN
LWSW
LWSS
LWOG
LWTT
LWDS
SLDG
SWIN
SWLF
SWSW
UGIW
Drum disposal/burial
Lagoon/liquid waste surface impoundment
Liquid waste incineration
Liquid wastes discharged to surface water bodies
Liquid (non-sanitary) wastes discharged to septic system or dry well
Liquid wastes dumped onto open ground or into unlined trenches
Liquid wastes dumped from tank trucks
Liquid wastes released from drains and sumps
Non-sewage sludge disposal
Solid waste incineration
Solid waste landfill
Solid waste discharged to surface water bodies
Underground injection well
Unintentional Releases
LAST
LDRM
LUTP *
CLUL
EXFR
SPIL *
Leaks from aboveground tanks
Leaks from drum storage areas
Leaks from underground tanks and pipelines
Releases during chemical loading and unloading
Releases during explosions or fires
Spills
Please list below any other means by which hazardous substances and chemicals were released to the
environment at the site:
OTHERWASTP
Word-searchable Version Not a true copy
-------
l.C.3. Known Releases of DNAPL Related Substances
Specifically, were there known releases to the environment,
such as spills, leaks, or disposal, of the DNAPL-related
substances or chemicals present at the site?
If DNAPL-related substances or chemicals
were released, were they released primarily
as a separate non-aqueous phase or
dissolved in water?
Yes Maybe No
No DNAPL substances present
Sep. phase Dissolved Both Unknown
No known releases
RELCMT
Considering the substances and chemicals present, please estimate the total volume of organic chemicals
released to the environment at this site, to the best of your knowledge. (Record a range, if necessary).
Units: gallons drums
Roughly, what is the uncertainty associated with this answer?
Low Medium High Very High
Check here if the volume released cannot be estimated:
Word-searchable Version Not a true copy
-------
I.D. Additional comments
In your opinion, how well understood is the site history Very Well
of this site, especially concerning the activities Well
and substances that caused contamination? Generally
Poorly
Please discuss below any additional information about site history that may be relevant to the probability
of DNAPL occurrence:
HISTCMT2
Word-searchable Version Not a true copy
-------
2. Site Investigation Section
2.A. Observation of Subsurface Non-aqueous Phase Liquids (NAPLs)
In this section, we are specifically interested in observations of separate organic liquids in the subsurface.
Subsequent sections will address the dissolved phase.
Was the possible presence of subsurface NAPLs
investigated directly during any site investigations?
Yes
No
Unknown
Specifically, was the possible presence of DNAPLs
investigated?
Yes
No
Unknown
If yes, what techniques were used to look for DNAPLS?
LOOKCMT
Were any non-aqueous phase liquids (NAPLs)
observed in the subsurface at this site?
(if you are uncertain, check boring logs
for observations of oily liquids)
Yes
Maybe
No
If yes, what was their nature?
Lighter than water (LNAPL)
Denser than water (DNAPL)
Unknown
If a NAPL was observed, had it
reached the water table?
Yes
Maybe
No
If NAPLs have been or may have been observed, please describe how they were encountered (in a test
pit, soil boring, ground water sample, etc.). Also note whether the NAPLs were found within contained
waste zones (for example, within a lined lagoon or landfill, or outside the boundaries of waste areas. Please
be as specific as possible.
NAPLENC
Word-searchable Version Not a true copy
-------
IfNAPLs have not been observed or only LNAPLs have been observed, please proceed to Section 2.B.
(Contamination of ground water, p. 9). If known or suspected DNAPLs were observed at this site, please
fill out the questions on this page:
Where were DNAPLs observed with respect to
the water table? Above Below Both Unclear
In which material was DNAPL observed? Unconsolidated material
(Circle all that apply) Bedrock
Ground Water
Surface Water
What is the maximum depth below ground surface
at which DNAPLs have been observed? (feet)
If the DNAPL was analyzed, please describe its composition below or attach a copy of the analytical
results. (We are interested in a sample of the free-phase DNAPL itself, not an associated ground-water
sample).
Chemical % in DNAPL
If measured, what was the density
of the DNAPL mixture? (g/cm3)
How much, if any, DNAPL has been removed
from the subsurface? (Please include units.)
If you have any additional comments on the DNAPL observation (e.g. what was its color and texture?),
please record them below:
DNAPLCMT
Word-searchable Version Not a true copy
-------
2.B. Contamination of Ground Water
This section records indirect indicators of DNAPL presence using levels and patterns of dissolved-phase
ground-water contamination.
2.B. 1 Sources of Ground Water Contamination
Please circle the abbreviations for the main sources for ground water contamination at the site. Those
sources most associated with DNAPL contamination (assuming that DNAPL substances were present
on-site) are marked with an "*".
AGST Aboveground tanks
LWOG * Area(s) of liquid waste dumping
BDRM * Buried drums
DRSP Drains and/or sumps
DRMS Drum storage areas
IPRA Industrial processing areas
FFTA * Fire fighter training area
LAGO * Lagoons/trenches for liquid waste disposal
LWIN Liquid waste incinerator
LULA Loading and unloading areas
SSYS Septic systems
SWLF Solid waste landfill
SOLU * Solvent use area
SPIL * Spill area
UGIW * Underground injection well
UGST * Underground tanks and pipelines
Other major sources of ground water contamination:
OTHSOURC
Please estimate the horizontal area of the source(s) at the site. (Record a range, if necessary. Record the
original source area if the source has since been removed.):
Units (circle one): acres ft2
Check here if the source area cannot be estimated:
What is the typical depth to ground
water at the site (feet)? Mn.: Max.
Word-searchable Version Not a true copy
-------
10
2.B.2 Major Ground Water Contaminants
Please list the compounds that you consider to be the major ground water contaminants at the site. Include
inorganic chemicals if these are driving the site investigation.
Are organic chemicals present in bedrock at
concentrations greater than MCLs?
2.B.3. Maximum Contaminant Concentrations
Yes
Maybe
No Unknown
Please use the following table (Table 1: Maximum Contaminant Concentrations) to record the maximum
concentrations of organic chemicals found in ground water at this site. The table includes the organic
contaminants most commonly found at Superfund sites and gives their densities and water solubilities, as
listed in the Subsurface Remediation Guidance Table 3 (EPA/540/2-90/01 Ib).
N We want to define the maximum ground-water concentrations observed over the entire site history,
not just in the latest sampling rounds, so please try to provide those to the best of your knowledge.
N Unless the site has significant semi-volatile contamination in the ground water (as at creosote or
coal tar sites), you may confine your answers to the volatile organic compounds.
N If there are major organic site contaminants which are not listed on the table, please include them
on the lines at the bottom of the table.
Theoretically, ground water in direct contact with DNAPL should exhibit concentrations of the DNAPL
chemicals that equal the chemicals' effective solubilities (i.e., if the DNAPL contains 50% TCE, the
ground-water concentration should be 500 mg/1 which is 50% of TCE's solubility limit). However, due to
sampling procedures and heterogenous DNAPL distribution in the subsurface, the maximum observed
concentrations of DNAPL-forming chemicals even at sites at which DNAPLs have been directly observed
are often much lower than the chemicals' effective solubilities. Depending on site conditions, concentrations
as low as a few percent of a chemical's solubility can represent an indication of subsurface DNAPL.
Word-searchable Version Not a true copy
-------
11
Table 1: Maximum Concentration Table
See Section 2.B.3. for instructions in filling out this table. Chemicals marked with a "*" have densities
greater than water and will behave as DNAPLs in the subsurface if present as a nonaqueous liquid.
Chemical
Density
(g/cm3)
Water Sol.
(Ppb)
Max.GW
(Ppb)
Volatiles
Benzene 0.877 1,780,000
2-Butanone
(Methyl ethyl ketone) 0.805 268,000,000
* Carbon Tetrachloride
(Tetrachloromethane) 1.595 800,000
* Chlorobenzene 1.106 490,000
Chloroethane
(Ethyl Chloride) 0.941 5,700,000
* Chloroform
(Trichloromethane) 1.485 8,220,000
* 1,1-Dichloroethane 1.175 5,500,000
* 1,2-Dichloroethane 1.253 8,690,000
* 1,1-Dichloroethylene 1.214 400,000
* Cis-l,2-Dichloroethylene 1.284 3,500,000
* Trans-1,2-Dichloroethylene 1.257 6,300,000
* Total-U-Dichloroethylene1 1.27 9,800,000
BENZ
MEK
CTET
CLBZ
CLEA
CLFM
IDC A
2DCA
1DCE
C2DC
T2DC
2DCE
1,2-Dichloropropane 1.158 2,700,000
Ethyl Benzene 0.867 152,000
Ethylene Dibromide
(1,1-Dibromoethylene) 2.172 3,400,000
4-Methyl-2-Pentanone
(Methyl isobutyl ketone) 0.802 19,000,000
Methylene Chloride
(Dichloromethane) 1.325 13,200,000
Styrene (Vinyl Benzene) 0.906 300,000
Key:
2DCP
EBNZ
EDB
MIBK
MTCL
STYR
Density: Density, g/cm3, generally at 20° C.
Water sol.: Solubility in water, generally at 20 ° C.
Max. GW: Maximum concentration of chemical observed in ground water at site, reported in ug/1 or ppb.
1 The densities and solubilities for these totals vary depending upon the exact mix of constituents.
Note: To calculate the percentage of aqueous solubility for a compound, divide the maximum concentration (Max. GW) by chemical's solubility in water (Water
sol.) and multiply by 100.
Word-searchable Version-Not a true copy
-------
12
Table 1: Maximum Concentration Table (cont.)
Chemical
Density
(g/cm3)
Water Sol.
Max.GW
Volatilesfcont.)
1,1,2,2-Tetrachloroethane
Tetrachloroethylene
Tetrahydrofuran
Toluene
1,1,1 -Trichloroethane
1,1,2-Trichloroethane
Trichloroethylene
Vinyl Chloride
(Chloroethylene)
Xylenes (total)1
Semi-Volatiles
Aroclor 1242
Aroclor 1254
Aroclor 1260
Acenaphthene
Anthracene
Benzo(a)anthracene
Bis-(2-ethylhexyl)phthalate
Chrysene
o-Cresol (2-Methylphenol)
p-Cresol (4-Methylphenol)
m-Cresol (3-Methylphenol)
Total cresols
(Methylphenols)1
1 The densities and solubilities for these totals vary depending upon the exact mix of constituents.
1.600
1.625
0.889
0.867
1.325
1.444
1.462
0.912
0.87
1.385
1.538
1.440
1.225
1.250
1.174
0.981
1.274
1.027
1.035
1.038
1.03
2,900,000
150,000
300,000,000
515,000
950,000
4,500,000
1,000,000
1,100,000
568,000
450
12
3
3,900
75
14
400
6
31,000,000
24,000,000
23,500,000
78,500,000
PCA
PCE
THF
TOLU
1TCA
2TCA
TCE
VNCL
TXYL
PC42
PC54
PC60
ACNP
ATHR
BATR
BEHP
CRYS
OCRS
PCRS
MCRS
TCRS
Word-searchable Version Not a true copy
-------
Table 1: Maximum Concentration Table (cont.)
13
Chemical
Density
(g/cm3)
Water Sol.
Max.GW
Semi-Volatiles (cont.)
1,2-Dichlorobenzene 1.306 100,000
l-4,Dichlorobenzene 1.248 80,000
2,4-Dimethylphenol 1.036 6,200,000
2,4-Dinitrophenol 1.680 6,000,000
Fluoranthene 1.252 265
Fluorene 1.203 1,900
2-Methyl Naphthalene 1.006 25,400
2DCB
4DCB
4DMP
4DNP
FLRA
FLRE
2MNP
Naphthalene
Pentachlorophenol
Phenol
Phenanthrene
Pyrene
1.162
1.980
1.058
0.980
1.271
31,000
14,000
84,000,000
1,180
148
NAPH
PCP
PHNL
PHNT
PYPJsT
1,2,4 Trichlorobenzene
1.574
30,000
124T
Other site contaminants:
Please add any comments you would like to make on the information in this table.
MAXCONTTBL
In what geologic unit were the maximum
concentrations found?
Unconsolidated material
Bedrock
Unclear
Word-searchable Version Not a true copy
-------
14
2.B.4. Patterns of Ground Water Contamination
Because the movement of DNAPLs is driven primarily by gravity and capillary forces, DNAPLs may
migrate into unexpected areas in the subsurface, such as the deeper portions of aquifers, below aquitards,
or areas which are hydrologically upgradient or across-gradient. This behavior gives rise to patterns of
ground-water contamination which differ from those generated by completely dissolved-phase
contamination. Subsurface DNAPL will also act as a long-term source of dissolved contamination, so
concentrations of DNAPL chemicals in the most contaminated monitoring wells are likely to remain steady
or increase over long time periods. Erratic concentration data, both spatial and temporal, are also expected
at DNAPL sites.
Spatial patterns:
Are ground-water concentrations of DNAPL-
related organic contaminants expectionally high at
depth below any source areas?
Yes Maybe No
Unknown
If yes or maybe, can these high concentrations be
explained by ground-water flow patterns in these
locations, such as downward vertical gradients?
Temporal patterns:
Yes
Partially
No
In general, how have the concentrations of DNAPL-related ground water contaminants in the most
contaminated wells changed over time at the site?
Increased sharply (>1 order of magnitude)
Increased slightly ( <1 order of magnitude)
Remained steady (no consistent increase or decrease)
Decreased slight (<1 order of magnitude)
Decreased sharply (>1 order of magnitude)
Insufficient data to observe pattern
On how long a period of years is this observation based?
In how many of these years were samples analyzed?
Word-searchable Version Not a true copy
-------
15
If you would like to explain any of your answers further or mention other interesting contamination patterns
at the site, please do so below. We are particularly interested in any ground water contaminant patterns
that are not easily explained by dissolved-phase transport. In addition, note any soils data that may be
indicative of subsurface DNAPL. Please see the DNAPL project hydrogeologist if you would like
assistance in interpreting soil or ground water contamination patterns at your site.
CONTCMT
2.C. Extent of Field Investigation
2.C.I. Sampling activities
Approximately how many ground-water monitoring points (wells,
well-points, etc.) are associated with the site?
(Please count each sampling point separately. For example, nested
monitoring wells whose screens are located at three discrete depths
count as 3 points. Please do not include nearby residential wells.)
Approximatley how many pairs, nests or clusters of wells
installed at multiple depths exist at the site?
Please estimate the number of ground-water samples that were analyzed for the major site contaminants
over all stages of site investigation and circle the appropriate range. (Please do not include samples from
residential wells)
None 161-10 11-25 26-50 51-75 76-100 101-150 151-200 >200
Is there a pump-and-treat system for ground-water
clean-up operating at the site? Yes No
If yes, how long has the system been in operation?
Word-searchable Version Not a true copy
-------
16
2.C.3. Site Investigation Summary
In your opinion, how well understood are the following aspects of the site?
(please use the key as a general guide to answer this question)
Key:
Very Well
No further information necessary for RD/RA.
Well
A good general understanding, but questions
remain in specific areas
Generally
Some understanding, but substantial
characterization effort still needed
Poorly
Only preliminary information available
Contaminant Sources:
Very Well
Well Generally Poorly
Site Hydrogeology:
Very Well
Well Generally Poorly
Ground Water Contamination:
Very Well
Well Generally Poorly
2.D. Additional comments
Please describe any additional information from site investigations that may be relevant to the probability
of DNAPL occurrence (for example, do pump and treat results match expectations?):
CHARCMT
Word-searchable Version Not a true copy
-------
17
3. Background Site Information
This section is designed to help us understand the site and how DNAPLS might behave in the site setting.
The proj ect hydrogeologists will be extracting information from documentation that you supply. Please copy
the following information from site investigation reports and attach to your completed survey form:
Check here if you have attached this information, or write NA if the information is not available for your
site:
1. Site Map
2. Geological Cross Section of Site
3. Description of Site Geology (such as from RI)
4. Plume Map and Cross Section (if available)
*******
TO BE FILLED OUT BY DNAPL PROJECT HYDROGEOLOGISTS
************
3.A. Geologic and Hydrogeologic Setting
We would like a general understanding of the geologic and hydrogeologic setting of the site. The movement
of DNAPLs in the subsurface is very sensitive to geologic heterogeneity and will tend to flow along areas
of increased permeability and downward through fractures. In addition, the question of whether DNAPLs
will reach ground water is influenced by both the thickness and composition of the unsaturated zone.
Typical depth to bedrock
at the site (feet):
Geologic description
Mia:
Max.
Unconsolidated sediments:
Were some or all of the unconsolidated sediments
or soils deposited by glaciers or glacial-related
water bodies?
All
Some None
Unknown
Word-searchable Version Not a true copy
-------
18
Please describe the composition and texture (such as silt, sand with clay layers, etc.) of the unconsolidated
materials. If roughly horizontal layers exist, please describe them from top to bottom. Include an estimate
of the thickness of each layer, if possible.
UNCONSOL
Bedrock:
Please indicate the bedrock types which comprise the upper 150 feet of bedrock below the site (check all
that apply). If necessary, explain your choices or add any additional information on the lines below.
Sedimentary
M Metamorphic I
Igneous
CONG
LIME
SAND
SILT
SHAL
Conglomerate
Limestone-
dolomite
Sandstone
Siltstone
Shale
NICE Gneiss
QTZT Quartzite
SHST Schist
SLAT Slate
MRBL Marble
GRNT Granite
BSLT Basalt
Other:
BEDROCK
Does water move in the bedrock primarily through pore spaces,
through fractures, or through solution channels?
Pore Spaces Fractures Solution channels Unknown
Is the site located in karst terrain?
Yes
No Possible, but unknown
Word-searchable Version Not a true copy
-------
19
3.B. Plume information
Has a ground-water plume been mapped at the site? Yes No
If yes, please describe the dimensions of the plume below.
Estimated length of plume (ft):
Estimated average width of plume (ft):
Estimated average thickness of plume (ft):
Boundary used to define plume:
Chemical (or TVOC):
Concentration (ppb):
Please indicate the approximate
date of this information
If the volume or mass of contaminants in plume
has been calculated, please record that
amount here (including units).
Comments on information in sections A. and B.:
BKGDCMT
END OF SECTION TO BE FILLED OUT BY PROJECT HYDROGEOLOGISTS
Word-searchable Version Not a true copy
-------
20
4. References and Final Comments
4.A. Reference Documents
Which documents do you recommend that we consult if we want to find more information concerning site
history and ground-water contamination at this site? (Which do you refer to most often?) If you use the
standard Superfund documents listed below, please check those that you use and indicate their dates and
authors (typically consulting firms). If you use other documents, please describe them on the blank lines
below:
HRS Scoring Package Date:
Remedial Investigation/Feasibility Study:
Title:
Date: Author:
Additional or Supplemental Remedial Investigation:
Title:
Date: Author:
Record(s) of Decision Date(s):
DOCUMENTS
Word-searchable Version Not a true copy
-------
21
4.B. Respondent opinion on possibility of DNAPLs
B ased on the information that you have provided here as well as your broader knowledge of site conditions,
what is your estimate of the probability of DNAPLs at this site?
Definite
High
Medium
Low
Please explain your estimate briefly:
OPNEXPL
While the exact measures to be taken in the case of possible DNAPL contamination vary from site to site,
we recommend that at the least, the impact of potentially present DNAPLs be considered when planning
further site investigations and remedial actions. The Quick Reference Fact Sheet on DNAPLs contains a
list of the implications for site investigations if there is a moderate to high probability of DNAPLs.
Word-searchable Version Not a true copy
-------
22
4.C. Comments on Survey
For our information, how many hours did you
spend filling out this form?
This survey will also be conducted in other regions. We would appreciate any suggested improvements or
comments that you would like to make about this form:
SURVCMT
Please note below any particular information concerning DNAPLs that would be helpful to you in your job
or particular topics concerning DNAPLs that you think deserve more research:
DNAPLINFO
Word-searchable Version Not a true copy
-------
APPENDIX B
Word-searchable Version Not a true copy
-------
United States R.S. Kerr Office of Solid Waste Publication: 9355.4-07FS
Environmental Environmental and Emergency December 1991
Protection Agency Research Laboratory Response
V>EPA Estimating Potential for
Occurrence of DNAPL
at Superfund Sites
Office of Emergency and Remedial Response
Hazardous Site Control Division (OS-220W) Quick Reference Fact Sheet
GOALS
The presence of Dense Non-Aqueous Phase Liquids (DNAPL) in soils and aquifers can control the ultimate success or failure of
remediation at a hazardous waste site. Because of the complex nature of DNAPL transport and fate, however, DNAPL may often be
undetected by direct methods, leading to incomplete site assessments and inadequate remedial designs. Sites affected by DNAPL may
require a different "paradigm," or conceptual framework, to develop effective characterization and remedial actions (2).
To help site personnel determine if DNAPL-based characterization strategies should be employed at a particular site, a guide for
estimating the potential for DNAPL occurrence was developed. The approach, described in this fact sheet, requires application of two
types of existing site information:
Historical Site Use Information Site Characterization Data
By using available data, site decision makers can enter a system of two flowcharts and a classification matrix for estimating the potential
for DNAPL occurrence at a site. If the potential for DNAPL occurrence is low, then conventional site assessment and remedial actions
may be sufficient. If the potential for DNAPL is moderate or high, however, a different conceptual approach may be required to account
for problems associated with DNAPL in the subsurface.
BACKGROUND
DNAPLs are separate-phase hydrocarbon liquids that are denser than water, such as chlorinated solvents (either as a single component
or as mixtures of solvents), wood preservative wastes, coal tar wastes, and pesticides. Until recently, standard operating practice in a
variety of industries resulted in the release of large quantities of DNAPL to the subsurface. Most DNAPLs undergo only limited
degradation in the subsurface, and persist for long periods while slowly releasing soluble organic constituents to groundwater through
dissolution. Even with a moderate DNAPL release, dissolution may continue for hundreds of years or longer under natural conditions
before all the DNAPL is dissipated and concentrations of soluble organics in groundwater return to background levels.
DNAPL exists in the soil/aquifer matrix as free-phase DNAPL and residual DNAPL. When released at the surface, free-phase DNAPL
moves downward through the soil matrix under the force of gravity or laterally along the surface of sloping fine-grained stratigraphic
units. As the free-phase DNAPL moves, blobs or ganglia are trapped in pores and/or fractures by capillary forces (7). The amount of the
trapped DNAPL, known as residual saturation, is a function of the physical properties of the DNAPL and the hydrogeologic
characteristics of the soil/aquifer medium and typically ranges from 5% to 50% of total pore volume. At many sites, however, DNAPL
migrates preferentially through small-scale fractures and heterogeneities in the soil, permitting the DNAPL to penetrate much deeper than
would be predicted from application of typical residual saturation values (16).
Once in the subsurface, it is difficult or impossible to recover all of the trapped residual DNAPL. The conventional aquifer remediation
approach, groundwater pump-and-treat, usually removes only a small fraction of trapped residual DNAPL (21, 26). Although many
DNAPL removal technologies are currently being tested, to date there have been no field demonstrations where sufficient DNAPL has
been successfully recovered from the subsurface to return the aquifer to drinking water quality. The DNAPL that remains trapped in the
soil/aquifermatrix acts as a continuing source of dissolved contaminants to groundwater, preventing the restoration of DNAPL-affected
aquifers for many years.
Word-searchable Version Not a true copy
-------
DNAPL TRANSPORT AND FATE - CONCEPTUAL MODELS
The major factors controlling DNAPL migration in the subsurface include the following (5):
the volume of DNAPL released;
the area of infiltration at the DNAPL entry point to the subsurface;
the duration of the release;
properties of the DNAPL, such as density, viscosity, and interfacial tension;
properties of the soil/aquifer media, such as pore size and permeability;
general stratigraphy, such as the location and topography of low-permeability units;
micro-stratigraphic features, such as root holes, small fractures, and slickensides found in silt and/or clay layers.
To describe the general transport and fate properties of DNAPL in the subsurface, a series of conceptual models (24) are presented
in the following figures.
Case 1: DNAPL Release to Vadose Zone Only
After release on the surface, DNAPL moves vertically downward
under the force of gravity and soil capillarity. Because only a small
amount of DNAPL was released, all of the mobile DNAPL is
eventually trapped in pores and fractures in the unsaturated zone.
Infiltration through the DNAPL zone dissolves some of the soluble
organic constituents in the DNAPL, carrying organics to the water
table and forming a dissolved organic plume in the aquifer. Migration
of gaseous vapors can also act as a source of dissolved organics to
groundwater (16).
DNAPL Gaseous
Vapors
Residual
Saturation of
DNAPL in Vadose
Zone
Infiltration, Leaching and
Mobile DNAPL Vapors
Dissolved Contaminant
Plume From DNAPL Soil Vapor
Groundwater
Row
Dissolved Contaminant
Plume From DNAPL
Residual Saturation
After W«l»rtoo Centra tor Groundwaier Research. 1969
Case 2: DNAPL Release to Unsaturated and Saturated Zones
If enough DNAPL is released at the surface, it can migrate all the way
through the unsaturated zone and reach a water-bearing unit. Because
the specific gravity of DNAPL is greater than water, it continues
downward until the mobile DNAPL is exhausted and is trapped as a
residual hydrocarbon in the porous media. Groundwater flowing past
the trapped residual DNAPL dissolves soluble components of the
DNAPL, forming a dissolved plume downgradient of the DNAPL
zone. As with Case 1, water infiltrating down from the source zone
also carries dissolved constituents to the aquifer and contributes
further to the dissolved plume.
Residual Saturation
of DNAPL in Soil
From Spill
Groundwater
Flow
Dissolved
Contaminant Plume
Residual
Saturation in Saturated Zone
After Waterloo Centre tor Groundwater Research 1989
Word-searchable Version Not a true copy
-------
CONCEPTUAL MODELS - Continued
Case 3: DNAPL Pools and Effect of Low-
Permeability Units
Mobile DNAPL will continue vertical migration until it is trapped as
a residual hydrocarbon (Case 1 and Case 2) or until low-permeability
stratigraphic units are encountered which create DNAPL "pools" in
the soil/aquifer matrix. In this figure, a perched DNAPL pool fills up
and then spills over the lip of the low-permeability stratigraphic unit.
The spill-over point (or points) can be some distance away from the
original source, greatly complicating the process of tracking DNAPL
migration.
Dissolved
Contaminant
Plume
Low Permeability
Stratigraphic Unit
After Waterloo Centre tor Groundwater Research. 1989
Case 4: Composite Site
In this case, mobile DNAPL migrates vertically downward through the
unsaturated zone and the first saturated zone, producing a dissolved
constituent plume in the upper aquifer. Although a DNAPL pool is
formed on the fractured clay or rock unit, the fractures are large enough
to permit vertical migration downward to the deeper aquifer (see Case
5, below). DNAPL pools in a topographic low in the underlying
impermeable unit and a second dissolved constituent plume is formed.
Dissolved
Contaminant
Plumes
frier Waterloo Centre for Ground Water Research, 1989.
Case 5: Fractured Rock or Fractured Clay System
DNAPL introduced into a fractured rock or fractured clay system
follows a complex pathway based on the distribution of fractures in
the original matrix. The number, density, size, and direction of the
fractures usually cannot be determined due the extreme heterogeneity
of a fractured system and the lack of economical aquifer
characterization technologies. Relatively small volumes of DNAPL can
penetrate deeply into fractured systems due to the low retention
capacity of the fractures and the ability of some DNAPLs to migrate
through very small (<20 microns) fractures. Many clay units, once
considered to be relatively impermeable to DNAPL migration, often
act as a fractured media with preferential pathways for vertical and
horizontal DNAPL migration.
After Waterloo Centre tor Ground Water Research. 1989
Word-searchable Version Not a true copy
-------
Does Historical Site Use Information Indicate Presence of DNAPL?
U
C
o
f->4
CO
1-4
&
g
o»
U
§
fc
YES
Does the
industry type suggest a
high probability of historical
DNAPL release?
(see Table 1)
process or waste
practice employed at the site
suggests a high probability of
historical DNAPL release?
(see Table 2)
Were any
DNAPL-related chemicals
used in appreciable quantities at the site?
(> 10-50 drums/year)
(see Table 3)
Go To Next Page
C MAYBEJ (
INSTRUCTIONS
1. Answer questions in Flowchart I
(historical site use info. - page 3).
2. Answer questions in Flowchart 2
(site characterization data - page 4).
3. Use "Yes,' "No,"and "Maybe"
answers from both flowcharts and
enter Occurrence of DNAPL matrix
(page 5).
TABLE 1
Industries with high probability
of historical DNAPL release:
Wood preservation (creosote)
Old coal gas plants
(mid-1800stomid-1900s)
Electronics manufacturing
Solvent production
Pesticide manufacturing
Herbicide manufacturing
Airplane maintenance
Commercial dry cleaning
Instrument manufacturing
Transformer oil production
Transformer reprocessing
Steel industry coking
operations (coal tar)
Pipeline compressor stations
TABLE 2
Industrial processes or waste
disposal practices with high
probability of historical DNAPL
release:
Metal cleaning/degreasing
Metal machining
Tool-and-die operations
Paint removing/stripping
Storage of solvents in
underground storage tanks
Storage of drummed solvents in
uncontained storage areas
Solvent loading and unloading
Disposal of mixed chemical
wastes in landfills
Treatment of mixed chemical
wastes in lagoons or ponds
TABLE 3 DNAPL-Related Chemicals (20):
Note:
The potential for DNAPL release increases -with the size
and active period of operation for a facility, industrial
process, or -waste disposal practice.
Halogenated Volatiles
Chlorobenzene
1,2-Dichloropropane
1,1-Dichloroethane
1,1 -Dichloroethylene
1,2-Dichloroethane
Trans-1,2-Dichloroethylene
Cis-1,2-Dichloroethylene
1,1,1 -Trichloroethane
Methylene Chloride
1,1,2-Trichloroethane
Trichloroethylene
Chloroform
Carbon Tetrachloride
1,1,2,2-Tetrachloroethane
Tetrachloroethylene
Ethylene Dibromide
Halogenated
Semi-Volatiles
1,4-Dichlorobenzene
1,2-Dichlorobenzene
Aroclor 1242, 1254, 1260
Chlordane
Dieldrin
2,3,4,6-Tetrachlorophenol
Pentachlorophenol
Non-Halogenated
Semi-Volatiles
2-Methyl Napthalene
o-Cresol
p-Cresol
2,4-Dimethyl phenol
m-Cresol
Phenol
Naphthalene
Benzo(a)Anthracene
Flourene
Acenaphthene
Anthracene
Dibenzo(a,h)Anthracene
Flouranthene
Pyrene
Chrysene
2,4-Dinitrophenol
Miscellaneous
Coal Tar
Creosote
Note: Many of these
chemicals are found mixed
with other chemicals or
carrier oils
Word-searchable Version - Not a true copy
-------
Do Site Characterization Data Indicate Presence of DNAPL? J
CN
Has DNAPL
been found in monitoring wells,
observed in soil cores, or physically
observed in the aquifer?
(see Table 4)
Do chemical
analyses of groundwater
or soil indicate the possible presence of
DNAPL at the site?
(see Table 5)
(Standard
Field
Program)
Is it likely that
the existing field program could
miss DNAPL at the site?
(see Table 6)
(Extensive Field Program)
INSTRUCTIONS
I. Answer questions in Flowchart 1
(historical site use info. - page 3).
2. Answer questions in Flowchart 2
(site characterization data - page 4).
3. Use "Yes," "No," and "Maybe"
answers from both flowcharts and
enter Occurrence of DNAPL matrix
(page 5).
Co To Next Page
(
NO
( MAYBE J ( YES
TABLE 4
Methods to confirm DNAPL in wells:
NAPL/water interface probes that signal a
change in conductivity of the borehole fluid
Weighted cotton string lowered down well
Pumping and inspecting recovered fluid
Transparent bottom-loading bailers
Mechanical discrete-depth samplers.
In general, the depth of DNAPL accumulation
does not provide quantitative information
regarding the amount of DNAPL present (24).
Methods to confirm DNAPL in soil samples:
Visual examination of cores or cutting may not be
effective for confirming the presence of DNAPL
except in cases of gross DNAPL contamination.
Methods for enhancing visual inspectionof soil
samples for DNAPL include:
Shaking soil samples in a jar with water to
separate the DNAPL from the soil (14).
Performing a paint filter test, in which soil is
placed in a filter funnel, water is added, and the
filter is examined for separate phases (20).
TABLES
Conditions that indicate potential for
DNAPL at site based on laboratory data:
Condition 1:
Concentrations of DNAPL-related chemicals
(see page. 3) in groundwater are > l%of
pure phase solubility or effective solubility,
(defined in Worksheet 1, pg. 6) (25).
Condition 2:
Concentrations of DNAPL-related chemicals
on soils are > 10,000 mg/kg (equal to 1% of
soil mass) (8).
Condition 3:
Concentrations of DNAPL-related chemicals
in groundwater calculated from water/soil
partitioning relationships and soil samples
are>pure phase solubility(see Worksheet 2,
Pg- 6).
Condition 4:
Concentrations of DNAPL-related chemicals
in groundwater increase with depth or
appear in anomalous upgradien/across
gradient locations (25).
Note: This procedure is designed primarily for hydrogeologic settings comprised of gravel, sand, silt, or clay
and may not be applicable to karst or fractured rock settings.
Word-searchable Version - Not a true copy
TABLE 6
Characteristics of extensive field
programs that can help indicate the
presence or absence of DNAPL (if
several are present, select "NO"):
Numerous monitoring wells, with wells
screened in topographic lows on the
surface of fine-grained, relatively
impermeable units.
Multi-level sampling capability.
Numerous organic chemical analuses
N of soil samples at different depths
using GC or GC/MS methods.
Well-defined site stratigraphy, using
numerous soil borings, a cone
penetrometer survey, or geophysics.
Data from pilot tests or "early
action"projectsthat indicate if the site
either:
1) responds as predicted by solute
transport relationships
(Suggest no DNAPL)
or
2) responds as if additional sources of
dissolved contaminants are present in
the aquifer
(Suggests DNAPL is present) (11).
-------
Potential for Occurrence of DNAPL at Superfund Sites J
DNAPL Category
P^MMJE"^*^ ^ W~ 'tf&ySF^
^s
l«
5 w
Ifi Yes
Ss
Is 2
'g o Maybe
.20
a 8
£ No
:*^ Do C
&^ Pre
Yes
I
I
I
haracterization Dal
sence of DNAPL? (
Maybe
I-II
II
II
a Indicate
Chart 2)
No
II
II - III
III
Category
I Confirmed or high
potential for DNAPL at
site.
Moderate potential
for DNAPL at site.
IE Low potential for
DNAPL at site.
Implications for Site Assessment
The risk of spreading contaminants increases with the proximity to a potential DNAPL zone. Special precautions
should be taken to ensure that drilling does not create pathways for continued vertical migration of free-phase DNAPLs.
In DNAPL zones, drilling should be suspended when a low-permeability unit or DNAPL is first encountered. Wells
should be installed with short screens (< 5 feet). If required, deeper drilling through known DNAPL zones should be
conducted only by using double or triple-cased wells to prevent downward migration of DNAPL. As some DNAPLs
can penetrate fractures as narrow as 10 microns, special care must be taken during all grouting, cementing, and well
sealing activities conducted in DNAPL zones.
In some hydrogeologic settings, such as fractured crystalline rock, it is impossible to drill through DNAPL with existing
technology without causing vertical migration of the DNAPL down the borehole, even when double or triple casing is
employed (4).
The subsurface DNAPL distribution is difficult to delineate accurately at some sites. DNAPL migrates preferentially
through selected pathways (fractures, sand layers, etc.) And is affected by small-scale changes in the stratigraphy of
an aquifer. Therefore, the ultimate path taken by DNAPL can be very difficult to characterize and predict.
In most cases, fine-grained aquitards (such as clay or silt units) should be assumed to permit downward migration of
DNAPL through fractures unless proven otherwise in the field. At some sites it can be exceptionally difficult to prove
otherwise even with intensive site investigations (4).
Drilling in areas known to be DNAPL-free should be performed before drilling in DNAPL zones in order to form a
reliable conceptual model of site hydrogeology, stratigraphy, and potential DNAPL pathways. In areas where it is
difficult to form a reliable conceptual model, an "outside-in" strategy may be appropriate: drilling in DNAPL zones
is avoided or minimized in favor of delineating the outside dissolved-phase plume (4). Many fractured rock settings may
require this approach to avoid opening further pathways for DNAPL migration during site assessment.
Due to the potential risk for exacerbating groundwater contamination problems during drilling through
DNAPL zones, the precautions described for Category I should be considered during site assessment.
Further work should focus on determining if the site is a "DNAPL site."
DNAPL is not likely to be problem during site characterization, and special DNAPL precautions are probably
not needed. Floating free-phase organics organics (LNAPLs), sorption, and other factors can complicate
site assessment and remediation activities, however.
Word-searchable Version - Not a true copy
-------
Worksheet 1: Calculation Of Effective Solubility (from Shm, 1988; Feenstra, Mackay, & Cherry, 1991)
For a single-component DNAPL, the pure-phase solubility of the organic constituent can be used to estimate the theoretical upper-level
concentration of organics in aquifers or for performing dissolution calculations. For DNAPLs comprised of a mixture of chemicals, however,
the effective solubility concept should be employed:
S° = the effective solubility (the theoretical upper-level dissolved-
phase concentration of a constituent in groundwater in equilibrium with a mixed
DNAPL; in mg/1)
c e = xi S-
Where S ; = the mole fraction component i in the DNAPL mixture (obtained from a lab
analysis of a DNAPL sample or estimated from waste characterization data)
X; = the pure-phase solubility of compound i in mg/1 (usually obtained
from literature sources)
For example, if a laboratory analysis indicates that the mole fraction of trichloroethylene (TCE) in DNAPL is 0. 1 0, then the effective solubility
would be 110 mg/1 [pure phase solubility of TCE tomes mole fraction TVE: (1100 mgl) * (0.10) = 110 mg/1]. Effective solubilities can be
calculated for all components in a DNAPL mixture. Insoluble organics in the mixture (such as long-chained alkanes) will reduce the mole
fraction and effective solubility of more soluble organics but will not comtribute dissolved-phase organics to groundwater. Please note that
this relationship is approximate and does not account for non-ideal behavior of mixtures, such as co-solvency, etc.
Worksheet 2: Method for Assessing Residual NAPL Based on Organic Chemical
Concentrations in Soil Samples (from Feenstra, MacKay, and Cherry, 1991)
To estimate if NAPLs are present, a partitioning calculation based on chemical and physical analyses of soil samples from the saturated
zone (from cores, excavations, etc.) Can be applied. This method tests the assumption that all of the organics in the subsurface are either
dissolved in groundwater or adsorbed to soil (assuming dissolved-phase sorption, not the presence of NAPL). By using the concentration
of organics on the soil and the partitioning calculation, a theoretical pore-water concentration of organics in groundwater is determined. If
the theoretical pore -water concentration is greater than the estimated solubility of the organic constituent of interest, then NAPL may be
present at the site. See Feenstra, MacKay, and Cherry (1991) for a description of the complete methodology.
Step 1: Calculate S j , the effective solubility of organic constituent of interest. [See Worksheet 1, above. |
Step 2: Determine Koc, the organic carbon-water partition coefficient from one of the following:
A) Literature sources (such as 22) or
B) From empirical relationships based on Kow, the octanol-water partition coefficient, which is also found in the
literature (22). For example, Koc can be estimated from Kow using the following expression developed for
polyaromatic hydrocarbons (8):
Log Koc = 1.0* Log Kow-0.21 | Other tmfmaA relationships between Koc
n^M^^MMMMHMMMHiMMJ and Kow are presented in refs. 4 and 15.
Step 3: Determine foe, the fraction of organic carbon on the soD, from a laboratory analysis of dean soils from the site.
Values for foe typically range from 0.03 to 0.00017 mg/mg (4). Convert values reported in percent to mg/mg.
Step 4: Determine or estimate pb, the dry bulk density of the soil, from a soils analysis. Typical values range from 1.8 to 2.1
gms/cc (kg/1). Determine or estimate Sj suggests possible presence of DNAPL
e
Cw< Sj suggests possible absence of DNAPL
Word-searchable Version Not a true copy
-------
GLOSSARY(adapted from Cherry, 1991):
DNAPL: A Dense Non-Aqueous Phase Liquid. A DNAPL can be either a single-component DNAPL (comprised of only one chemical) or a mixed
DNAPL (comprised of several chemicals). DNAPL exists in the subsurface as free-phase DNAPL or as residual DNAPL (see following definitions).
DNAPL does not refer to chemicals that are dissolved in groundwater.
DNAPL ENTRY LOCATION: The area where DNAPL has entered the subsurface, such as a spill location or waste pond.
DNAPL SITE: A site where DNAPL has been released and is now present in the subsurface as an immiscible phase.
DNAPLZONE: The portion of a site affected by free-phase or residual DNAPL in the subsurface (either the unsaturated zone or saturated zone).
The DNAPL zone has organics in the vapor phase (unsaturated zone), dissolved phase (both unsaturated and saturated zone), and DNAPL phase
(both unsaturated and saturated zone).
DISSOLUTION: The process by which soluble organic components from DNAPL dissolve in groundwater or dissolve in infiltration water and
forma groundwater contaminant plume. The duration of remediation measures (either clean-up or long-term containment) is determined by 1)
the rate of dissolution that can be achieved in the field, and 2) the mass of soluble components in the residual DNAPL trapped in the aquifer.
EFFECTIVE SOLUBILITY: The theoretical aqueous solubility of an organic constituent in groundwater that is in chemical equilibrium with a
mixed DNAPL (a DNAPL containing several organic chemicals). The effective solubility of a particular organic chemical can be estimated by
multiplying its mole fraction in the DNAPL mixture by its pure phase solubility (see Worksheet 1, page 6).
FREE-PHASE DNAPL: Immiscible liquid existing in the subsurface with a positive pressure such that it can flow into a well. If not trapped in
a pool, free-phase DNAPL will flow vertically through an aquifer or laterally down sloping fine-grained stratigraphic units. Also called mobile
DNAPL or continuous-phase DNAPL.
PLUME: The zone of contamination containing organics in the dissolved phase. The plume usually will originate from the DNAPL zone and
extend downgradient for some distance depending on site hydrogeologic and chemical conditions. To avoid confusion, the term "DNAPL plume"
should not be used to describe a DNAPL pool; "plume" should be used only to refer to dissolved-phase organics.
POOL and LENS: A pool is a zone of free-phase DNAPL at the bottom of an aquifer. A lens is a pool that rests on a fine-grained stratigraphic
unit of limited areal extent. DNAPL can be recovered from a pool or lens if a well is placed in the right location.
RESIDUAL DNAPL: DNAPL held in soil pore spaces or fractures by capillary forces (negative pressure on DNAPL). Residual will remain trapped
within the pores of the porous media unless the viscous forces (caused by the dynamic force of water against the DNAPL) are greater than the
capillary forces holding the DNAPL in the pore. At most sites the hydraulic gradient required to mobilize all of the residual trapped in an aquifer
is usually many times greater than the gradient that can be produced by wells or trenches (27).
RESIDUAL SATURATION: The saturation (the fraction of total pore space containing DNAPL) at which DNAPL becomes discontinuous and
is immobilized by capillary forces (14). In unsaturated soils, residual saturation typically ranges from 5% to 20% of total pore volume, while in
the saturated zone the residual saturation is higher, with typical values ranging from 15% to 50% of total pore volume (14,17). At many sites,
however, DNAPL migrates preferentially through small-scale fractures and heterogeneities in the soil, permitting the DNAPL to penetrate much
deeper than would be predicted from application of typical residual saturation values (16).
Defined Areas at a DNAPL Site , f. ,DN.APL*°N£ . Dissoived-PhasePLUME
(contains free-phase DNAPL in pools or
lenses and/or residual DNAPL)
DNAPL ENTRY LOCATION^
(such as a former waste pond) *
Groundieater Flow Direction
Word-searchable Version - Not a true copy
-------
References
1. Anderson, M.R., R.L. Johnson, and J.F. Pankow, The Dissolution of Residual Dense Non-Aqueous Phase Liquid (DNAPL) from a Saturated Porous
Medium, Proc.: Petrol. Hcarb. and Ore. Chemicals in Ground Water. NWWA, Houston, TX, Nov., 1987.
2. Cherry, J. A., written communication to EPA DNAPL Workshop, Dallas, TX, R. S. Kerr Environmental Research Laboratory, U.S. EPA, Ada, OK,
Apr. 1991.
3. Connor, J.A., C.J. Newell, and D.K. Wilson, Assessment, Field Testing, and Conceptual Design for Managing Dense Nonaqueous Phase Liquids
(DNAPL) at a Superfund Site, Proc.: Petrol. Hcarb. Org. Chemicals in Ground Water.NWWA. Houston, TX, 1989.
4. Domenico, P.A. and F. W. Schwartz, Physical and Chemical Hvdrogeology. Wiley, New York, NY, 1990.
5. Feenstra, S. and J.A. Cherry, Subsurface Contamination by Dense Non-Aqueous Phase Liquids (DNAPL) Chemicals, International Groundwater
Symposium. International Assoc. of Hydrogeologists, Halifax, N.S., May 1-4, 1998.
6. Feenstra, S., D. M. MacKay, and J.A. Cherry, A Method for Assessing Residual NAPL Based on Organic Chemical Concentrations in Soil Samples,
Groundwater Monitoring Review. Vol. 11, No. 2, 1991.
7. Hunt, J.R., N. Sitar, and K.D. Udell, Nonaqueous Phase Liquid Transport and Cleanup, Water Res. Research. Vol. 24 No. 8, 1991.
8. Karickhoff, S.W., D.S. Brown, and T.A. Scott, Water Res. Research. Vol. 3, 1979.
9. Keller, C.K., G. van der Kamp, and J.A. Cherry, Hydrogeology of Two Saskatchewan Tills, J. of Hydrology, pp. 97-121, 1988.
10. Kueper, B.H. and E.O. Frind, An Overview of Immiscible Fingering in Porous Media, J. of Cont. Hydrology. Vol. 2, 1988.
11. Mackay, D.M. and J.A. Cherry, Ground-Water Contamination: Pump and Treat Remediation, ES&T Vol. 23, No. 6, 1989.
12. Mackay, D.M. P.V. Roberts, and J.A. Cherry, Transport of Organic Contaminants in Ground Water. ES&T. Vol. 19, No. 5, 1985.
13. Mendoza, C.A. and T. A. McAlary, Modeling of Ground-Water Contamination Caused by Organic Solvent Vapors, Ground Water. Vol. 28, No. 2,
1990.
14. Mercer, J.W. and R.M. Cohen, A Review of Immiscible Fluids in the Subsurface: Properties, Models, Characterization and Remediation, J. of Cont.
Hydrology. Vol. 6, 1990.
15. Olsen, R.L. and A. Davis, Predicting the Fate and Transport of Organic Compounds in Groundwater, HMC. May/June 1990.
16. Poulson, M. and B.H. Kueper, A Field Experiment to Study the Behavior of Perchloroethylene in Unsaturated Porous Medium. Submitted to
ES&T. 1991.
17. Schwille, F., Dense Chlorinated Solvents in Porous and Fractured Media: Model Experiments (English Translation), Lewis Publishers, Ann Arbor,
MI, 1988.
18. Shiu, W.Y., A. Maijanen, A.L.Y. Ng, and D. Mackay, Preparation of Aqueous Solutions of Sparingly Soluble Organic Substances: II.
Multicomponent System - Hydrocarbon Mixtures and Petroleum Products, Environ. Toxicology & Chemistry. Vol. 7, 1988.
19. Sitar, N., J.R. Hunt, and J.T. Geller, Practical Aspects of Multiphase Equilibria in Evaluating the Degree of Contamination, Proc. of the Int. Asso.
of Hvdrog. Conf. on Subsurface Cont. by Immiscible Fluids. April 18-20, Calgary, Alb., 1990.
20. U.S. EPA, Dense Nonaqueous Phase Liquids. EPA Ground Water Issue Paper, EPA/540/4-91-002, 1991.
21. U.S. EPA, Evaluation of Ground-Water Extraction Remedies. Volume 1 (Summary Report). EPA/540/2-89/054, 1989.
22. Verschueren, K, Handbook of Environmental Data on Organic Chemicals. Van Nostrand Reinhold, New York, NY, 1983.
23. Villaume, J.F., Investigations at Sites Contaminated with Dense, Non-Aqueous Phase Liquids (NAPLs), Ground Water Monitoring Review. Vol. 5,
No. 2, 1985.
24. Waterloo Centre for Ground Water Research, University of Waterloo Short Course, Dense Immiscible Phase Liquid Contaminants in Porous and
Fractured Media. Kitchener, Ont., Nov., 1991.
25. Waterloo Centre for Ground Water Research, University of Waterloo Short Course, Identification of DNAPL Sites: An Eleven Point Approach.
Kitchener, Ont., Oct., 1991.
26. Wilson, J.L. and S.H. Conrad, Is Physical Displacement of Residual Hydrocarbons a Realistic Possibility in Aquifer Restoration?, Proc.: Petrol.
Hcarb. and Org. Chemicals in Ground Water. NWWA, Houston, TX, NWWA, Nov. 5-7, 1984.
NOTICE: The policies and procedures set out in this document are intended solely as guidance. They are not intended, nor can they be relied upon, to create any
rights enforceable by and party in litigation with the United States. EPA officials may decide to follow the guidance provided in this memorandum, or to act at
variance with the guidance, based on an analysis of specific site circumstances. The Agency also reserves the right to change this guidance at any time without
public notice.
For more information, contact: Randall R. Ross
R.S. Kerr Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Ada, Oklahoma 74820
Authors: Charles J. Newell, Groundwater Services, Inc., Houston, Texas
Randall R. Ross, R. S. Kerr Environmental Research Laboratory
Word-searchable Version - Not a true copy
------- |