FRAMEWORK FOR THE
DEVELOPMENT OF THE NATIONAL
SEDIMENT INVENTORY
March 18,1993
United States Environmental Protection Agency
Office of Science and Technology
Standards and Applied Science Division
-------
-------
EXECUTIVE SUMMARY
In 1992, the U.S. Environmental Protection Agency (EPA) issued its draft
Contaminated Sediment Management Strategy, which included a recommendation for
the development of a national inventory of contaminated sediment sites. Also in
'1992, Congress passed the Water Resources Development Act of 1992 (WRDA),
which required EPA, in consultation with the National Oceanic and Atmospheric
Administration (NOAA) and the U.S. Army Corps of Engineers (COE), to conduct
a comprehensive national survey of data regarding aquatic sediment quality in the
United States. In an effort to help meet the objectives of EPA's Contaminated
Sediment Management Strategy and to comply with the mandates of the WRDA, EPA
has initiated the development of the National Sediment Inventory (NSI). This
document presents a framework for the development of the NSI. Included are a
discussion of potential EPA program uses for the Inventory, a review of existing
background studies and pilot inventories, a discussion of options considered for the
development of the Inventory, a description of the option selected, and a summary of
existing sediment assessment techniques.
The NSI will provide EPA with the ability to conduct a near-term screening
assessment of the national extent and severity of sediment contamination across the
country. Such an assessment would include the identification of sites that should be
targets for future, more intensive study, either to justify and recommend regulatory
actions for those sites which pose an obvious risk to the environment or to gather
additional information for those sites which appear to be contaminated but for which
there are insufficient data to reach a definitive conclusion. In addition, the Inventory
will provide valuable information to assist EPA in achieving the other, more long-
range goals of its draft Contaminated Sediment Management Strategy concerning
pollution prevention, remediation, and dredged material management. Data in the
Inventory can help EPA to identify point and nonpoint source discharges contributing
to sediment contamination, identify chemicals of concern, and set priorities for their
control. The Inventory will also provide information to assist in identifying and
prioritizing sites for enforcement and remediation and in identifying technically and
economically feasible alternatives for remediation. For dredged material management,
data in the NSI can provide initial screening information to program managers
concerning the location of potential and probable contaminated sites and identification
of contaminants of concern.
EPA recently conducted a review of several existing studies of contaminated sediment
problems and contaminated sediment inventory pilot studies. The purpose of this
review was to assess the current understanding of problems associated with
contaminated sediments and to build on the experience of other programs during the
in
-------
development of the Inventory. The following sources of information were reviewed
as part of this effort:
• National Perspective on Sediment Quality (Bolton et al., 1985);
• An Overview of Sediment Quality in the United States (Lyman et al., 1987);
• Contaminated Marine Sediments—Assessment and Remediation (NAS
1989);
• Summary Report for Contaminated Sediments Assessments in U.S. EPA
Region IV Coastal Areas (USEPA, 199 Ib) and Draft Evaluation of the
Region 4 Inventory of Coastal Sediment Sites (USEPA, 1992a);
• EPA Region 5 Inventory of Contaminated Sediment Sites (USEPA, 1992b);
• The Gulf of Mexico Program's Toxic Release and Contaminated Sediment
Inventories (TRI, 1992; unpublished information); and
• Proceedings of EPA's Contaminated Sediment Management Strategy
Forums (USEPA, 1992d).
The review of these studies provided considerable insight into methods for evaluating
contaminated sediments and the potential extent of sediment contamination.
Examination of the pilot inventories also helped identify a number of concerns related
to the development of the NSI, including the capabilities of different systems and
software for performing data searches and. compilations and the possibilities for
storing detailed monitoring data or summary data in relational, searchable databases
that would be nationally accessible.
Several options for the design of the NSI were considered in an attempt to address
these and other concerns. Initially consideration was given to housing the NSI in an
existing database system such as ODES (the Ocean Data Evaluation System) or
STORET (the STOrage and RETrieval System). Because of data entry cost and data
accessibility concerns, these options were ruled out until a modernized STORET
system is available. The following options were given final consideration:
• Create an inventory of summary data only compiled by either EPA
Headquarters or the Regions or
• Create an inventory of detailed monitoring data compiled by either EPA
Headquarters or the Regions.
Based on an evaluation of the advantages and disadvantages of each of these options,
EPA decided to create an inventory using detailed monitoring data. This inventory
iv
-------
will be compiled initially by EPA Headquarters and will include data from several
potential sources, including:
• Select data sets from STORET, e.g.,
- COE
- U.S. Geological Survey (USGS)
- EPA
- States
- BIOACC
• NOAA's National Status and Trends (NS&T) Program database
• ODES
• Region 4 Sediment Inventory
• Region 5 Sediment Inventory
• Gulf of Mexico Program Sediment Inventory
• COE Seattle District Sediment Inventory
• Great Lakes Sediment Inventory
• Environmental Monitoring and Assessment Program (EMAP)
• National Estuary Program (NEP)
• Fish and Wildlife Service (FWS)
• MacDonaild Database
• USGS
• National Source Inventory
These and other sources of data will be evaluated for inclusion hi the NSI. The
determination as to whether a given database will be included in the Inventory will
be based in part on the difficulty in obtaining the data, difficulty in analyzing the data,
and compatibility with other data. Available resources will also be a consideration
when determining which data sets to include.
Four major categories of detailed monitoring data will be collected for the
development of the NSI (Table 1). Several minimum data parameters have been
identified under each of these major categories. Some must be included in a database
before the data will be added to the NSI; others would be desirable, but their absence
v
-------
Table 1. National Sediment Inventory Data
Category Summary
If Minimum Data Element
8 In Computerized Format
| Location
| Sampling Date
1 Lat/Long
U Reach Number
I SITE CHARACTERISTICS
| Land Use
I] Management Status of Site
| Location of Haz Waste/Superfund
| Site
| Spill Information
| Frequency of Dredging
| Point Source Information
| Presence of Endangered1 Species
Source of Information
Lab Methods
Reid Methods
SAMPLING PARAMETERS
j Sediment Chemistry
j Total Organic Carbon
Grain Size
] Acid Volitile Sulfides
| Biological Data
| Fish Advisories
| Benlhic Abundance
! Fish Pathology
Necessary
•
•
•
•
•
•
If Available
•
•
•
•
•
•
•
•
•
•
•
,•
•
•
•
•
O
Comments
With data dictionary specifying field
names, widths, delimiters, or file structure
Including waterbody name
Conforming to EPA standards
Urban, industrial, rural, etc.
Remedial actions, etc.
i.e., dredging history
Current/historical
Sponsor or client name and address,
name of analytical lab or principal
nvestigator and address
Quality of data to be coded, method
detection limits used in analyses to be
ncluded
Quality of data to be coded
Biotoxicity, bioaccumulation
VI
-------
would not necessarily preclude data from being entered.
It should be noted that although no data set will initially be excluded from the NSI
because of a lack of information on quality assurance/quality control (QA/QC)
procedures or concerns associated with the QA/QC procedures employed, EPA
believes that information regarding the quality of monitoring and analytical data
should be considered when identifying and evaluating potential and probable
contaminated sediment sites. Therefore, EPA is preparing to include with the data in
the NSI a basic screening assessment of the potential or probable quality of data (if
known) from a particular data set. The name and address of a database contact will
also be provided to allow the user of the data to acquire specific information
concerning QA/QC samples, methods, and results. The results of QA/QC analyses
will not be included in the NSI.
Once the NSI is in place, the data will be evaluated to identify those sediment
chemistry sample observations which exceed predetermined threshold limits for each
contaminant. (EPA is currently evaluating existing sediment assessment techniques
that could be used to establish these thresholds.) The results of this evaluation will
be a computer-generated detailed listing of all observations that exceeded the sediment
quality threshold limits. Potential areas of concern will include those sites with
sufficient information to be classified as contaminated as well as sites hi need of
further assessment. Additional data related to each river or coastal reach segment
(based on EPA's Reach File) in which a sediment sample that exceeded a given
threshold was taken will also be included in the NSI. Such information will include,
when available, bioassay, bioaccumulation, benthic abundance, fish pathology, and
fish advisory data. These data will be included to allow the users of the NSI to assess
the correlation between sediment chemistry data and biological and other data.
All EPA Regional Offices will be sent a copy of the preliminary assessment and data
for review. The Regions will then be asked to review and comment on the
information presented. Specifically, they will be asked to identify and provide
additional computerized databases not included in the NSI that can supplement the
information presented. They will also be asked to gather additional QA/QC
information for data taken from their Region that were included in the NSI but about
which little or nothing is known concerning the QA/QC samples and procedures used
when gathering and analyzing samples. Following Regional review, EPA
Headquarters will enter the appropriate new data sets into the NSI and update the
QA/QC and other information based on the input from the Regions. The data in the
NSI will then be evaluated a second time.
Each of the sites identified based on the second, more complete evaluation of the data
in the NSI will be categorized as either those for which sufficient data exist to
characterize them as causing high risks or severe effects (probable contaminated sites)
or those which may be contaminated but for which additional information and further
assessment are needed (potential contaminated sites). This final categorization will
be based on consideration of a number of factors, including the following:
vn
-------
• Number of chemicals exceeding threshold limits;
• Number of observations exceeding threshold limits;
• Severity of contamination (i.e., concentration of contaminants of concern);
• Biological evidence of contamination and impacts to support conclusions
based on sediment chemistry data;
• Fish advisory information; and
• Quality of data used to identify the site as contaminated.
The final listing of sites resulting from the assessment of data in the NSI will
represent a snapshot of sediment contamination problems across the country. Any site
listed would be a target for future, more intensive study, either to justify and
recommend regulatory actions for those sites that pose an obvious risk to the
environment or to gather additional information for those sites which appear to be
severely contaminated but for which there are insufficient data to reach a definitive
conclusion.
The data compiled as part of the NSI can help managers prioritize future remediation,
regulatory, or assessment activities; guide decisions regarding the appropriate type and
scale of regulatory action needed to reduce contaminant inputs; and evaluate the
effectiveness of existing technology-based effluent guidelines, water quality-based
controls, and nonpoint source controls. The NSI data can also be used to help
identify and prioritize on a local, state, regional, or national level those specific
chemicals in need of stricter regulation.
EPA recommends that the NSI be maintained and updated on a regular basis to allow
future assessments of sediment quality on a local and regional basis as well as
nationwide. EPA also recommends that efforts be made to ensure that future sediment
quality monitoring programs include additional information and parameter
measurements (which may currently be missing from many data sets), which can be
used to more accurately assess the potential environmental impacts of sediment
contamination during future assessments. For example, sediment sampling programs
should include the measurement of total or percent organic carbon content, sediment
particle size, sediment reductive capacity, and salinity. The data should also meet
certain minimum data quality objectives, and results of data quality evaluations should
be reported with the data or, at a minimum, the QA/QC samples and procedures used
should be identified.
vm
-------
CONTENTS
TABLES AND FIGURES . . . . xi
1. INTRODUCTION 1-1
Background 1-1
Purpose of the National Sediment Inventory 1-4
2. PROGRAM USES ., 2-1
Assessment , • • 2-1
Pollution Prevention 2-17
Remediation 2-20
Dredged Material Management , 2-22
3. REVIEW OF BACKGROUND STUDIES AND PILOT INVENTORIES . 3-1
National Perspective on Sediment Quality 3-1
An Overview of Sediment Quality in the United States 3-8
Contaminated Marine Sediments—Assessment and Remediation 3-11
Summary Report for Contaminated Sediment Assessments in U.S. EPA
Region IV Coastal Areas ; . . . 3-13
EPA Region 5 Inventory of Contaminated Sediment Sites 3-15
Progress Report on the Gulf of Mexico Program's Toxic Release
and Contaminated Sediment Inventories 3-17
Proceedings of the EPA's Contaminated Sediment Management
Strategy Forums . . . . 3-18
4. OPTIONS CONSIDERED FOR THE DEVELOPMENT OF THE
NATIONAL SEDIMENT INVENTORY 4-1
Options Considered . •. 4-2
Discussion of Option Selected 4-11
5. APPROACH FOR DEVELOPING THE NATIONAL SEDIMENT
INVENTORY • • • , 5-1
Development of the National Sediment Inventory 5-2
Initial Evaluation • • • • • • 5-11
Review of the National Sediment Inventory 5-14
Schedule . 5-16
IX
-------
6. SEDIMENT ASSESSMENT TECHNIQUES 6-1
Equilibrium Partitioning 6-2
Sediment Quality Triad 6-2
Bulk Sediment Toxicity 6-6
Interstitial Water Toxicity Identification Evaluation . 6-7
Apparent Effects Threshold 6-7
Spiked Sediment Toxicity 6-7
Tissue Residue 6-8
Screening-Level Concentration 6-8
Long and Morgan (1990) 6-9
MacDonald (1992) . 6-9
7. CONCLUSIONS AND RECOMMENDATIONS 7-1
REFERENCES CITED
-------
TABLES AND FIGURES
Table Page
2-1 EPA Program Uses of the National Sediment Inventory 2-2
3-1 STORET Sampling Stations for Freshwater Sediments by Compounds . 3-5
4-1 Comparison of Selected Attributes and Problems Associated with
Each Option 4-4
5-1 National Sediment Inventory Data Category Summary 5-3
5-2 QA/QC Components Used in the GLNPO Procedure 5-5
5-3 Sediment Chemistry Sample Parameters 5-7
5-4 Bioassay Sample Parameters 5-7
5-5 Bioaccumulation Sample Parameters 5-8
5-6 Benthic Abundance Sample Parameters 5-8
5-7 Fish Pathology Sample Parameters 5-9
5-8 Fish Advisory Parameters • 5-9
6-1 Data Requirements and Issues Related to Sediment Assessment
Techniques • 6-3
5-1 Milestones for Completion of the National Sediment Inventory 5-17
XI
-------
-------
CHAPTER 1
INTRODUCTION
In recent years, the contamination of sediments in waterbodies of the United States
has become a national ecological and human health issue of concern. In response
to this concern, the U.S. Environmental Protection Agency (EPA) has proposed its
draft Contaminated Sediment Management Strategy. One of the recommendations
of this draft strategy is the development of a national inventory of contaminated
sediment sites. In addition, recent legislation passed by Congress (the Water
Resources Development Act of 1992, or WRDA) requires EPA to develop by 1994
a national inventory of sites with contaminated sediments. The purpose of this
document is to present the proposed framework for the development of an
inventory of contaminated sediment sites that will fulfill both the objectives of
EPA's Contaminated Sediment Management Strategy and the mandates of the
WRDA.
Background
Sediments have been described as the ultimate sink for pollutants (Salomons et al.,
1987). If that were entirely true, however, there would be no need to be concerned
about potential impacts from sediment-associated compounds. In fact, sediments
can function as both a source of and a sink for contaminants in the aquatic
environment, and they are capable of releasing contaminants to the overlying water
and biota slowly over extended periods of time or very quickly due to natural or
human perturbations. Likewise, compounds in sediment deposits may build up
over time as a result of inputs from a combination of sources. The following
sections present a brief overview of the potential sources of sediment
contamination, the transport and fate of sediment-associated contaminants, and the
potential environmental effects of these contaminants.
Sources
The problem of in-place, sediment-associated pollutants is both widespread and
localized. All regions of the United States and all types of waterbodies are
affected. Waterbodies usually receive both point and nonpoint discharges of
pollutants as a result of the various human activities that take place there. For
instance, bays and harbors are associated with contaminant sources from shipping,
among other activities. Upper reaches of streams are usually polluted by local
sources. Harbors, streams, and estuaries bordered by industrialized or urbanized
areas tend to have elevated levels of metals, organics, and other compounds
associated with human activity (Lyman et al., 1987). Sometimes the Contamination
is localized beneath an outfall of industrial or municipal waste; in other cases,
natural mixing processes and dredging disperse the pollutants. Nonpoint sources
of pollution such as surface water runoff and atmospheric deposition can also
596201\RDOCS\NICSS\Chapt«f.t
-------
NSI Framework
contribute to widespread sediment contamination. In addition, pollutants from
upstream sources are discharged into waterbodies by rivers and contribute further
to the problem of sediment contamination.
The ubiquitous nature of trace organic and metal compounds in sediments near
urban and agricultural areas and the association of large inputs of these
contaminants with runoff events tend to support the importance of contributions
from nonpoint sources, such as atmospheric deposition and land drainage. For
example, mining is an significant source of sediment contamination in some
regions, as are runoff and seepage from landfills, Superfund sites, and urban and
agricultural runoff (Hoffman, 1985; Livingston and Cox, 1985; Ryan and Cox,
1985; Baudo and Muntau, 1990). Pollution from nonpoint sources is primarily
related to land use characteristics. Agricultural runoff may contribute selenium,
arsenic, and mercury and a wide variety of pesticides. Urban runoff is a frequently
mentioned source of heavy metals, polychlorinated biphenyls (PCBs), and
polycyclic aromatic hydrocarbons (PAHs). However, it is often difficult to
determine the fraction of these contaminants contributed by runoff versus point
source discharges because the same contaminants can come from both (Baudo and
Muntau, 1990).
Although nonpoint sources may contribute the largest quantities of contaminants
to the aquatic environment, .the combined effect of varied source locations,
hydrology, and sediment characteristics can lead to a large variability in the
concentrations of contaminants found in a waterbody as a result of nonpoint
sources (Lyman et al., 1987). Point source releases, including accidental or
deliberate discharges, may result hi elevated localized contamination. Purposeful
and accidental contaminant additions include effluent discharges, spills, dumping,
and the addition of herbicides to lakes and reservoirs. Both industrial and
municipal point sources contribute a wide variety of contaminants to sediments.
Municipal point sources include sewage treatment plants and combined sewer
overflows (CSOs). Industrial point sources include chemical plants, pulp and paper
mills, steel mills, metal-working plants, electroplating plants, tanneries, refineries
and other petroleum industries, engine and automotive industries, and many other
industrial categories.
Transport and Fate
Sediment contamination may be contributed in the form of solids (e.g., mine
tailings), or contaminants may be discharged in the aqueous phase and sorbed onto
sediments, which are then deposited. The residence time of contaminants in
sediments depends on a number of biological, chemical, and physical factors such
as the degree of binding to sediments and the degradation rate. Physical factors
are perhaps the most important, however, because compounds that find their way
to sediments tend to be those which are moderately to strongly sorbed, somewhat
1-2
-------
Chapter One—Introduction
volatile, and chemically stable (Gillett, 1983). Rivers with sufficient discharge
velocity to resuspend sediments may flush themselves clean once inputs of
contaminants cease. On the other hand, deep lakes and reservoirs act as giant
settling basins for contaminated sediments and provide long residence times due
to relatively limited resuspension, compared to rivers and near coastal
environments. Dissolved compounds entering lakes, reservoirs, and especially
estuaries and marine environments niay precipitate, may flocculate, or may be
scavenged by sorption onto other particles and thereby be incorporated into bottom
sediments.
Once compounds reach the sediment, they are hardly static. Sediments should be
viewed as dynamic systems. Not only are compounds transported with sediments
through various physical processes including settling, resuspension, and deposition,
but chemical reactions can also change the particle matrix and the chemical
characteristics of sorbed contaminants. Infaunal organisms also redistribute
sediments through their burrowing and home-building activities and are capable of
unearthing old deposits (Krezoski and Robbins, 1985). Some animals feed by
ventilating their burrows, which facilitates contaminant exchange with the water
column (McCaffrey et al.; 1980).
Biological and chemical processes affecting sediment contamination include
sorption and desorption, degradation of organic matter, transformation of iron and
manganese oxyhydrates to sulfides and vice versa, and biotransformation of
contaminants by in situ macro- and microorganisms. These processes depend
somewhat on sediment characteristics. Fine sediments tend to adsorb larger
quantities of pollutants per gram than do coarse sediments because of their
relatively higher surface area. Sorption of organic material to sediments is
controlled to a large degree by the organic carbon content of the sediment. The
higher the organic carbon content, the greater the ability of the sediment to. bind
and thereby accumulate organic pollutants, sulfur, and some metals. Enhanced
sorption may also reduce the bioavailability of the contaminant to aquatic life.
Bioaccumulation and Potential Environmental Effects
Contaminated sediments can affect aquatic life by making some areas
uninhabitable, by providing a source of contaminants to the aquatic food chain, and
by adversely affecting the health of organisms (Lyman et al., 1987). For example,
fin rot and a variety of neoplastic lesions have been found in fish living above
sediments contaminated by PAHs located near a creosote plant on the Elizabeth
River in Virginia, while liver tumors and skin lesions have occurred in brown
bullheads from the Black River in Ohio, contaminated by PAHs from a coke plant
(USEPA, 1992d).
1-3
-------
NSI Framework
Bioaccumulation of contaminants in fish tissue and contamination of the food chain
are also important human health concerns because the consumption of fish
represents the most significant route of aquatic exposure of humans to many metals
and organic compounds (USEPA, 1992d). Many surface waters have fish
consumption advisories or fishing bans in place because of the high concentrations
of PCBs, mercury, dioxin, kepone, and other contaminants. There are currently at
least 1,183 fish consumption advisories in place in the United States, affecting all
but five states (Allison Greene, USEPA, Risk Assessment and Management
Branch, Standards and Applied Science Division, Office of Science and
Technology, telephone conversation, 19 November 1992). Water supplies also
have been closed because of contaminated sediments, and in some places
swimming is no longer allowed. Most sediment-related human exposure to
contaminants is through indirect routes involving the transfer of pollutants out of
the sediments and into the water column or the biota.
Several recent assessments of existing data on the Nation's marine and freshwater
sediments indicate widespread and potentially serious contamination problems.
However, some researchers and resource managers believe that sediment
contamination problems are not widespread but limited to "hot spots" caused by
historical rather than ongoing pollutant discharges. Thus, an inventory and
evaluation of sediment quality data and associated information will yield greater
insight into the nature and extent, as well as the causes, of sediment contamination
in both freshwater and saltwater systems.
Purpose of the National Sediment Inventory
EPA proposed the development of a national inventory of contaminated sediment
sites as part of the draft outline of the Agency's Contaminated Sediment
Management Strategy, distributed on March 5, 1992 (USEPA, 1992c). The goals
of this proposed strategy are the following:
• Prevent ongoing contamination of sediments that may cause
unacceptable risks to human health or cause ecological harm, so that
beneficial uses of the Nation's surface waters are maintained;
• When practical, clean up existing sediment contamination that adversely
impacts the Nation's surface waters or their uses or that causes other
significant effects on human health or the environment; and
• Ensure that sediment dredging and the disposal of dredged materials
continue to be managed in an environmentally sound manner.
1-4
-------
Chapter One—Introduction
The Strategy is designed to help coordinate sediment quality assessment and
management activities of EPA program offices and Regions, as well as other
federal, state, and local agencies. The Strategy presents a plan of action for
assessing, preventing, and remediating sediment contamination and for supporting
ongoing Agency programs for the management of dredged material. EPA's
proposed Contaminated Sediment Management Strategy has been presented at a
series of public forums. The Agency is in the process of revising the Strategy,
taking into consideration comments and recommendations voiced during the
national forums as well as in formal written comments (USEPA, 1992d).
Individuals who commented on the Strategy were in general very supportive of a
national inventory of contaminated sediment sites but raised concerns regarding the
quality of data included in the inventory and how this information will be used in
the future management of contaminated sediments in the United States.
While EPA was developing the NSI as part of its Contaminated Sediment
Management Strategy, Congress passed the Water Resources Development Act
(WRDA) of 1992. This Act requires EPA, in consultation with the National
Oceanic and Atmospheric Administration (NOAA) and the U.S. Army Corps of
Engineers (COE), to conduct a comprehensive national survey of data regarding
aquatic sediment quality in the United States. As part of this requirement, EPA is
to compile all existing information on the quality, chemical and physical
composition, and geographic location of pollutants in aquatic sediment, including
the, probable source of such pollutants. The act requires EPA to report on the
findings of this survey within 2 years of enactment of the WRDA.
To fulfill the statutory requirements of the WRDA and to advance the objectives
of the Agency's Contaminated Sediment Management Strategy, EPA has begun the
development of the NSI based on existing information. The purposes of this
inventory are as follows:
• Obtain the best possible near-term assessment of the national extent and
severity of sediment contamination (i.e., determine whether contaminated
sediments are a localized, "hot spot" problem or a widespread, national
problem);
• Identify areas that may be contaminated and in need of further
assessment; and
• Identify areas with sufficient data to be characterized as causing high
risks or severe effects so that Agency programs can target these areas
for appropriate action.
In a parallel effort, EPA is also developing an inventory of potential sources of
sediment contamination. Together, these two inventories will contribute to EPA's
1-5
-------
NSI Framework
ability to identify areas in need of enforcement or remediation or in need of
reduced point and/or nonpoint source inputs of contaminants through increased
effluent or best management practice (BMP) controls, as well as those areas in
need of further assessment. These inventories will also be designed as a part of
a comprehensive and continuing program to assess aquatic sediment quality trends
over time and to assess the effectiveness of future sediment quality management
programs.
As previously mentioned, the purpose of this document is to provide the framework
for the development of the National Sediment Inventory. As part of this effort, an
attempt has been made to identify the potential uses of such an inventory by
various EPA program offices. A discussion of these potential uses, is presented in
Chapter 2 of this document. Chapter 3 presents a discussion of background studies
and pilot inventories that were reviewed and provided insight and guidance for the
development of the framework for the NSI. The options evaluated for the
development of the Inventory are presented in Chapter 4. Chapter 5 presents the
process selected for the development of the Inventory, its proposed structure, and
the procedure to be used in reviewing and updating the information presented in
the Inventory. Chapter 6 presents a summary of several sediment assessment
techniques that are being considered for use in evaluating contaminated sediment
data. The final chapter of this document presents the conclusions and
recommendations for the continued maintenance of the NSI, which can be used for
future assessments of national trends in sediment contamination.
1-6
-------
CHAPITER 2
PROGRAM USES
EPA's Contaminated Sediment Management Strategy presents a plan for assessing,
preventing, and remediating sediment contamination and supports ongoing Agency
programs for the management of dredged material. The National Sediment
Inventory (NSI) is an important tool that can contribute significantly toward
fulfilling EPA's objectives related to each of these activities. The development of
the NSI will represent a major step toward assessing the problem of contaminated
sediment on a nationwide basis. The NSI will be a repository for sediment data
that managers can use in conjunction with their own database systems to evaluate
the extent and severity of sediment contamination across the country and to target
chemicals for sediment criteria development. If the NSI is maintained and updated
in the future, it can also be used in conjunction with other tools to assess trends
in sediment quality and the effectiveness of existing regulatory programs at the
federal, state, and local levels.
The NSI will contain a minimum set of data elements that must be present before
data can be included m the database. These include information related to the
sampling location, date, latitude/longitude, sediment chemistry, and source of data.
Additional data that will be added if available include site characteristics such as
land use, management status of the site, location of Superfund sites, spill
information, frequency of dredging, point and nonpoint source information, and the
presence of endangered species. Other QA/QC and sampling parameters to be
added to the inventory if available include laboratory methods; field methods; total
organic carbon (TOC); grain size (and other geological parameters); acid volatile
sulfides (AVS); and biological, fish advisory, benthic abundance, and fish
pathology data.
The NSI can contribute valuable data to assist EPA in carrying out its plan of
action for preventing and remediating sediment contamination and for managing
dredged material disposal. The potential uses of the NSI by various program
offices related to these activities are summarized in Table 2-1. Examples of uses
of the data in the NSI by EPA program offices related to their assessment,
pollution prevention, remediation, and dredged material management activities are
discussed below.
Assessment
Because assessment is the first step in identifying and remediating environmental
pollution, EPA program managers could use data from the NSI as a screening tool
to describe the nature and spatial extent of potential sediment contamination due
to activities managed or regulated by their program area to determine whether the
2-1
-------
2-2
-------
I
f
I
1
1
S •
c
I
£
-------
I
2-4
-------
tJ
v
U
t
a
i
i
'•3
w
11
.1
5
z
i
•8
3
I
•a
'g -g
i!
0
ft
mediation
«
06
1
Pollution Prevent
1
•4
** rf
1 ^
8
E
i
I
(
i
e
1
2 ^
3 ° «
a . oo
" >* bn *O
•o a s 1
c a § "»
tt|£l
J»( ™ 5 tn £,
g S S c .52
^ -s •§ § -s
,
l^^g 1" Sg l a
"i, «2 -S 'i §-'g:§oS 1
111 ill Hli.
«is iii mi
111 nl mi
.00 If g 1 g '
S en ^" g S Joo
oo?B «» «" ^°Si
<: ^ -< •< ^ «s
&I u o uS
|_
|I
CA S3
•s|
s "i
I 1
w ^> O *"*
=32 r«2 §
S '1 «2 § 'S
as C5 jo 'S *
•§* -S S '§ ts
a S S. S P
« 0 SVS S
S«2 2 8 8
•s-s
<8 g
• VN ^
€ *35
< 1. 2
S = i
«fl
•< S 8
^
eoo
3
OS
2
1
£
CM
*£
Q ^^
£ Q
^5 ^3j
2-5
-------
1
1
2-6
-------
I
I
£
•4«*
Z
1
§
H
I
I
ft.
c
£
Hi I
o
en
O CO
en ' «
SOO \O
6O9 O
S 8
en en
2-7
-------
73
'u «
5 e
4)
W o
> i£>
•= « S
••-» a> ra
•2 2
ca
as 55
u
•3
a
u
,
o o .
2-8
-------
C/l
•a
lit i
™ "e/3
a> as
4= -O
C3
I I
1-1 I
« Q
2 e
s ou u —•
S c s g
I3g(_l «'CS-00
III Mill
|
V
>
I
.2
I
*
3
on
2
a.
.S
I
I
£
|
2
<
CU
1
8 '&
S w
a
O
«« 8
.o,
o
S '
•< S i
S
cos «
cos U
eft ^3 Q
i S S
2-9
-------
1*
nv
Z
W
I
i
c
I
Poll
1
OS
e
o
§•0
.— •- S
•3 .O ra
B -a
g 13 g
i i 11
4? - .2 '»
S S ^
a e> B
•i-Sl
> -3 «
12
CO9
•s
§•
§•
c
4>
§>
,§> 2 «
1||E|
£ .S N-4 CM *C
•C
11
)•§
g
• •«
lit
*c ??
P 8.
s a
1
2 e
•a .2,
-
§ g g
a §5
W> .rH C
a" s 1 a s 1
111 5=81
S J3
e 2 S >-• e -a u
g g .2 3 -a .2 S
^5 M *S. ™ P ^3 ^
Ilillli
la
a
73
I
2-/0
-------
f
§
u
i?
2
1
*••
e
V
1
1
.2
Z
V
*£•
*^
0
(A
1
en
a
la
a,
•3
•c -
•2 S
08 55
11
•o «
£ 2
O
S
1
•8
§
«
g
«
•s
S
1
e
{
"o
£
Assessment
a
| j
11
8
' E
O
2
£
a.
s
•o ,_
2g
M S
• v2 ^~
111
Is I
f*>1 M
O rtt
"u o, .
| 3,2
.H S TJ
^ bO v
O Q 5
Eel.
Supplement monitoring
programs to determine
compliance with existing
regulatory programs and
determine the. need for
remedial actions -
additional assessment
activities
X— S
X
GOO
*
O U .
CJ CJ
•s s
|0
O ^
11
^
^.2
a e
1 £
1 2
1 ^ «
!•* O C
i—« ST*
H 5. a.
Supplement ongoing
monitoring programs and
determine the need for
additional assessments
Provide background data
to develop
characterization reports
0
-------
NSI Framework
sediments pose a threat. Several EPA program offices, including the Office of Air
and Radiation (OAR), Office of Solid Waste (OSW), Office of Pesticide Programs
(OPP), Office of Toxic Substances (OTS), Office of Water (OW), and others, could
use the data in the NSI to consider the spatial scale over which sediments are
contaminated by ongoing as well as historical sources of chemicals in an attempt
to evaluate and improve the effectiveness of their existing regulatory programs.
Potential applications for evaluating the data in the NSI for assessment activities
by EPA managers include:
• Evaluating the extent and severity of sediment contamination;
• Evaluating whether concentrations of contaminants are increasing or
decreasing to determine whether current regulatory programs at the state,
regional, and national levels adequately protect the quality of sediments;
• Monitoring the concentrations and types of contaminants in sediments
surrounding discharges to assess compliance with current regulatory
programs; and
• Targeting chemicals for sediment criteria development based on their
geographic distribution and concentration in the sediments.
Office of Air and Radiation
The Office of Air and Radiation (OAR) is responsible for controlling the
atmospheric deposition, of .contaminants under the Clean Air Act (CAA). As
pointed out earlier, atmospheric deposition may be an important source of sediment
contamination. The atmospheric loading of pollutants to aquatic systems has been
demonstrated, and the potential for these contaminants to bind to sediments is
significant. OAR could use the data in the NSI to assist in evaluating the presence
of atmospherically-borne pollutants in contaminated sediments, using sediment
chemistry data to determine whether existing control programs are effective.
Office of Enforcement
The Office of Enforcement (OE) is primarily responsible for the management,
oversight, and direction of the Agency's enforcement program, including activity
to enforce the Clean Air Act; Clean Water Act; Safe Drinking Water Act; Resource
Conservation and Recovery Act; Comprehensive Environmental Response,
Compensation, and Liability Act; Toxic Substances Control Act; Federal
Insecticide, Fungicide, and Rodenticide Act; and Emergency Planning and
Community Right-To-Know Act. The NSI will facilitate enforcement decision-
making in a number of ways. First, the NSI will provide more reliable and
consistent information than is currently readily available concerning risks posed by
2-12
DRAFT
-------
Chapter Two—Program Uses
contaminated sediment areas. Second, the NSI will provide valuable information
useful in enforcement priority-setting. In certain individual enforcement actions,
the NSI may also provide data to assist the Agency in (1) demonstrating an
imminent and substantial endangerment (necessary in some causes of action),
(2) providing key information on which appropriate injunctive relief can be
fashioned, (3) demonstrating the gravity of the violations (relevant to the
calculation of penalties), and (4) proving that violations occurred.
Office of Emergency and Remedial Response
The Office of Emergency and Remedial Response (Superfund) identifies,
investigates, and remediates hazardous waste sites under the Comprehensive
Environmental Response, Compensation, and Liability Act (CERCLA). Samples
are collected to characterize releases of hazardous substances from a site and to
determine whether such releases present a threat to human health, welfare, and the
environment. In general, remediation program managers rely on states and Regions
to bring to their attention sites with known or suspected contamination (USEPA,
1992d). Sediment chemistry and sampling location data from the NSI could
supplement other data to assist in identifying areas where contamination is
suspected, and states could use these data when developing their lists of sites for
possible inclusion on the National Priority List (NPL) (USEPA, 1992d).
Where remediation activities have already been performed at selected Superfund
sites, data from the NSI could assist in evaluating the success of the remediation
by mapping sediment chemical concentrations over time and space at the
remediated Superfund site.
Office of Federal Activities
The Office of Federal Activities is responsible for EPA's compliance with the
National Environmental Policy Act (NEPA), as well as the NEPA environmental
review program. The objective of the environmental review program is to ensure
that EPA's general environmental expertise and regulatory experience are made
available to federal decision-makers. This is carried out through interagency
coordination early in relevant federal planning processes to identify significant
environmental issues of concern to EPA; in-depth review of federal environmental
impact statements and, as appropriate, environmental assessments; and follow-up
coordination on actions where EPA has identified significant environmental
impacts.
EPA could use the NSI for both the NEPA compliance and environmental review
programs to obtain additional data on the affected environment and to help identify
potential impacts of proposals.
DRAFT
2-13
-------
NSI framework
Office of Federal Facilities Enforcement
The Office of Federal Facilities Enforcement (6FFE) is responsible for the
enforcement of all applicable environmental statutes and regulations in which EPA
has jurisdiction. This multimedia enforcement office establishes the framework
that ensures that the federal government is accountable to the public for its
environmental record. Using this accountability to establish credibility, OFFE's
Ten-Point Strategic Plan calls for the office to enforce the law; increase citizen
involvement in decision-making; incorporate environmental equity concerns;
prevent pollution; accelerate the cleanup of CERCLA sites; accelerate the reuse of
closed bases; promote the development of innovative technologies to improve and
reduce the cost of environmental cleanup and management; address the
environmental issues at the national nuclear weapons complexes; develop and
implement multimedia enforcement; and ensure that federal agencies are receiving
pertinent information and training with regard to environmental issues. OFFE
could use the NSI in each of the 10 strategic points to ensure that the appropriate
program office activities accurately reflect the federal facility environmental
activities related to sediments. For assessment, this could include, but would not
be limited to, the use of the NSI for targeted inspections and enforcement.
Office of Pesticide Programs
The Office of Pesticide Programs (OPP) reviews the uses of new and existing
chemicals to be registered as pesticides in order to determine their effects on
nontarget organisms (USEPA, 1992d). OPP managers could use the data in the
NSI to help assess the extent or spatial scale of pesticides present in sediments to
guide decisions regarding appropriate registration actions. The presence of high
concentrations of a pesticide at numerous sediment sites nationwide may indicate
that the chemical has a high potential for transport away from the point of
application and thus represents a potential route of exposure for nontarget
organisms and may cause unreasonable adverse effects on the environment. The
NSI will include available data on site-specific lexicological and environmental
effects, which OPP can use as part of its assessment procedure, outlined in Hazard
Evaluation Division, Standard Evaluation Procedure: Ecological Risk Assessment
(USEPA, 1986). The procedure involves the review of existing laboratory and
field lexicological data for the registration of any pesticide.
Office of Science and Technology
The Office of Science and Technology (OST) is currently developing chemical-
specific sediment quality criteria that will be used in several EPA programs to set
point source limits, evaluate the quality of dredged material proposed for disposal,
and evaluate contaminated sites for remediation (USEPA, 1989). OST could use
the data in the NSI to assist in identifying target chemicals of concern for sediment
criteria development by listing those chemicals which are present in the highest
2-14
DRAFT
-------
Chapter Two—Program Uses
concentrations and those which have the greatest spatial coverage nationwide.
Once these chemicals are identified, they could be further prioritized for criteria
development based on characteristics such as bioavailabUity, persistence, and
bioaccumulation potential depending on the parameters included in the data set.
The data in the NSI could also be used to assist in evaluating the effectiveness of
the technology-based effluent guidelines and water quality-based limits developed
by OST. Sediment chemistry data could be overlaid with point source discharges
to determine whether sediment contamination may be a problem at any of the
discharge sites. The NSI could also be used as a tool to help identify contributions
of chemicals from nonpoint source discharges (urban and agricultural). Further
evaluation using biological parameters such as fish consumption advisories,
biotoxicity and bioaccumulation studies, fish pathology and benthic community
structure could help to determine the magnitude of the contaminated sediment
problem.
Office of Solid Waste
The Office of Solid Waste (OSW) is responsible for assessing whether releases
from hazardous waste treatment, storage, and disposal facilities have contaminated
sediments and determines corrective action, including possible remediation under
the Resource Conservation and Recovery Act (RCRA). EPA inspects facilities that
have applied for a RCRA permit, as well as facilities that ceased operations before
the deadline for submitting applications for a final RCRA permit. If the inspection
indicates that there is sediment contamination, a more extensive inspection can be
performed to determine the extent of contamination. Data in the NSI could
supplement other data and assist managers in determining whether hazardous waste
facilities are being properly managed by overlaying areas of high pollutant
concentrations in the sediments with hazardous waste facilities.
The Office of Toxic Substances
Under the Toxic Substances Control Act (TSCA), the Office of Toxic Substances
(OTS) is responsible for assessing the risks resulting from possible releases of
existing and new chemicals that are manufactured, distributed, or disposed of. The
NSI could be a useful tool to help identify chemicals that occur in areas of
contaminated sediment and that should be considered for further review. The
presence of a chemical at numerous sites nationwide may indicate that the chemical
poses an unreasonable risk to human health or the environment (USEPA, 1990).
The data in the NSI could help to identify aquatic life or human health problems
at contaminated sediment locations caused by known chemicals if the appropriate
biological parameters are included in the data set. OTS managers could use such
information to select chemicals for further assessment.
DRAFT
2-15
-------
NSI framework
Office of Wastewater Enforcement and Compliance
The Office of Wastewater Enforcement and Compliance (OWEC) is responsible
for issuing National Pollutant Discharge Elimination System (NPDES) permits to
control point source discharges into the Nation's waters. OWEC managers could
use the data in the NSI to determine whether the NPDES discharges are
contributing pollutants that lead to sediment contamination and to help prioritize
NPDES permit limits to protect sediment quality. By overlaying sediment "hot
spots" with NPDES permit locations, program managers could review the
overlapping data sets to determine which NPDES permitted facilities might be
contributing to environmental impairment. The NSI could also be used to help
evaluate the extent and severity of sediment contamination potentially caused by
combined sewer overflows (CSOs).
Office of Wetlands, Oceans and Watersheds
The Office of Wetlands, Oceans and Watersheds (OWOW) is responsible for
several major programs that potentially deal with contaminated sediments, such as
nonpoint source pollution control; watershed protection; and, with the Army Corps
of Engineers, dredged material disposal in the oceans and coastal waters of the
United States.
The Assessment and Watershed Protection Division (AWPD) could use the data
in the NSI as part of its assessment of nonpoint source control programs. The data
could help AWPD to identify areas of high concentrations of pollutants in the
sediment where point source controls are in place. Total Maximum Daily Loads
(TMDLs) are equivalent to the loading capacity for a waterbody. TMDLs are used
in watershed management to allow the water resource manager to determine the
most effective point and nonpoint source pollution controls for a watershed
(USEPA, 199 la). The NSI could be used to help target waterbodies in need of
TMDLs by identifying potential "hot spots" in the watershed. The data could also
be used to help evaluate the effectiveness of existing TMDLs, especially for
waterbodies where releases of contaminants from sediments may contribute to
violations of water quality standards (USEPA, 1990).
The Oceans and Coastal Protection Division (OCPD) is responsible for assessing
and preventing pollution in the marine environment. OCPD and its Regional
counterparts assess potential impacts of ocean discharges and monitor the effects
of ocean dumping into marine and coastal waters. OCPD managers could use the
data in the NSI to support these activities by overlaying ocean discharge and
disposal sites with sediment contamination data. The National Estuary Program
(NEP), which is managed by OCPD, targets selected estuaries for national
assessment and pollution prevention activities. The NEP's estuary programs could
use the data in the NSI as part of the mandatory characterization phase to identify
contaminated sites within the estuary and target these sites for further action.
2-16
DRAFT
-------
Chapter Two—Program Uses
Regional and State Programs
The NSI will identify areas with potential sediment contamination. The approach
for the NSI includes coordinating data review with EPA Regional offices' and
states' programs to ensure that all relevant data have been included and
appropriately evaluated.
Other regional projects, such as the Great Lakes National Program Office
(GLNPO), the Gulf of Mexico Program Office (GOMP), and the National Estuary
Program (NEP), could use the NSI to supplement monitoring programs and to
determine whether additional assessments are needed where data gaps exist. The
NEP could also use the NSI as a source of background information for the
preparation of estuary characterization reports.
Pollution Prevention
EPA program managers have several tools available to control and prevent the
release of contaminants into the environment. Potential sources of sediment
contamination can be controlled at the national level through the registration of
chemicals, the restricted use of specific chemicals, and the development of
chemical-specific sediment quality criteria. On a local or site-specific level,
managers can implement pretreatment technologies at discharge facilities, modify
existing NPDES permits, and implement best management practices and TMDLs
for watershed management.
Potential applications for the data in the NSI for pollution prevention activities by
EPA managers include assisting in the following activities:
• Identifying point and nonpoint source discharges, associated industries,.
and other factors contributing to sediment contamination;
• Identifying chemicals of concern to set priorities for further
regulatory/planning activities;
• Assessing the effectiveness of existing technology-based and water
quality-based controls and the need for stricter controls of discharges
with high chemical concentrations; and
• Identify ing areas in need of controls to reduce agricultural and urban
nonpoint source inputs.
DRAFT
2-17
-------
NSI Framework
Office of Air and Radiation
Through the issuance of National Ambient Air Quality Standards (NAAQSs), OAR
can control emissions that may contribute to atmospheric deposition, leading
eventually to sediment contamination. If atmospheric pollutants are suspected as
sources of sediment contamination, the standards could be reviewed and reissued
to restrict these atmospheric inputs. The 1990 amendments to the CAA include
new, more stringent requirements for controlling toxic air pollutants. These new
requirements will address stationary-source emissions that may be sources of
sediment contamination. The data in the NSI could be used to assist in evaluating
the effectiveness of these new controls, as well as in determining whether
additional controls are necessary for controlling atmospheric inputs.
Office of Enforcement
Several OE policies encourage the adoption of compliance projects in enforcement
settlements that permanently prevent pollutants from entering the environment.
The NSI could greatly facilitate the adoption of pollution prevention projects by
defendants in EPA enforcement actions. Enforcement decision-makers could use
the NSI to help identify target pollutants contributing to the sediment problem, and
to negotiate enforcement settlements that implement process changes, technologies,
and house-keeping practices that will prevent future contamination.
Office of Federal Activities
The OFA could use the NSI as a tool to identify opportunities for pollution
prevention and assist in coordinating EPA and interagency programs in this area.
Office of Federal Facilities Enforcement
The data in the NSI could provide information to identify supplemental
environmental projects that would prevent continuing sediment contamination.
Office of Pesticide Programs
The data in the NSI could be evaluated based on chemical concentrations and
scope of contamination to assist in prioritizing pesticides for possible additional
testing requirements, use restrictions, special reviews, or recommendations against
reregistration (Southerland et al., 1992). If it is determined that a pesticide is
causing an unreasonable risk to human health or unacceptable adverse effects on
the environment, OPP has several options to control and prevent further
contamination. If the problem appears to be national in scope, OPP can cancel the
pesticide's registration and ban its use. In site-specific or localized situations,
OPP can modify the label to control or restrict its use.
2-18
DRAFT
-------
Chapter Two—Program Uses
Office of Science and Technology
The NSI could be a useful tool to help assess the effectiveness of treatment
technologies on the bioaccumulation or bioavailability potential of chemicals
regulated under technology-based effluent guidelines. For example, if discrete
amounts of a contaminant meet applicable effluent guidelines but sediment
sampling around industrial discharges shows elevated levels of the contaminant, the
current guidelines may need to be revised to reduce the potential for environmental
effects.
OST managers could also use data in the NSI to assist in efforts to control
nonpoint source pollution. For example, chemical data can be downloaded and
used to evaluate the impacts to waterbodies of pesticide applications hi a
geographic area. If pesticide concentrations are found far away from the source,
then stricter measures may be needed.
Office of Toxic Substances
OTS managers can use several risk management tools to control the release of
chemicals into the environment, ranging from information gathering, to imposing
use restrictions, to banning the use of the chemical entirely. The data in the NSI
could provide useful information on the distribution of the chemical under
consideration—whether it is widespread or highly localized—and thereby provide
OTS managers with one means of evaluating the degree of potential human or
environmental exposure, the populations or ecosystems at risk, and the need for
regulatory action to reduce environmental effects.
Managers at OTS could also use the lexicological information hi the NSI in
conjunction with other data to score chemicals for their potential for environmental
effects, including acute or other toxicity to organisms in the environment,
bioaccumulation hi fish tissue levels resulting in fish consumption advisories, or
evidence of ecological effects such as alteration of the benthic community structure
(Davies et al., 1979; USEPA, 1990). Sediment bioassay data, coupled with
chemical concentration data, could be used to help assess the degree of
bioavailability of sediment-associated compounds.
Office of Wastewater Enforcement and Compliance
OWEC managers could use data hi the NSI to help target CSOs and stormwater
discharges requiring stricter permit requirements if sediment contamination is
shown to be significant at these sources. The NSI could be used to help identify
industrial and municipal dischargers that contribute to contaminated sediments hi
order to revise NPDES permit limits. The sediment chemistry data in the NSI
could also be used to help support enforcement actions against permit violators if
DRAFT
2-19
-------
NSI Framework
significant levels of sediment contamination are observed at an NPDES-permitted
facility.
Office of Wetlands, Oceans and Watersheds
Within OWOW, the AWPD and OCPD program offices could use the NSI to help
evaluate point and nonpoint sources of concern and to develop appropriate
programs of research and activities to control or prevent pollutant discharges.
Regional and State Programs
Regional and state program offices could use the data in the NSI to prioritize and
develop management plans for waterbodies with potential and probable sites of
concern. Information from the NSI could also be used to implement voluntary
pollution prevention programs within states and under the NEP.
Remediation
The remediation of contaminated sediments is expensive and time-consuming. The
NSI could be used as a tool to help prioritize sites requiring remediation based on
chemical concentrations and adverse environmental effects. The NSI could also
be used with other, more site-specific data to help identify responsible parties and
facilitate enforcement-based remediation by geographically linking sources of
contaminants to the concentrations of chemicals found in sediments.
Potential applications of .the NSI to support remediation activities include:
• Assisting in the identification of point and nonpoint source discharges
contributing to sediment contamination;
• Providing additional data for evaluating site-specific environmental and
human health threats resulting from sediment contamination;
• Providing additional data for identifying and prioritizing sites for
remediation based on the spatial extent and severity of contamination;
• Helping to identify technically and economically feasible alternatives for
remediation; and
• Providing additional data to help in prioritizing sites for enforcement
activities based on the spatial extent and severity of contamination.
2-20
DRAFT
-------
Chapter Two—Program Uses
Office of Enforcement
The restoration of ecosystems damaged by noncompliance with environmental
statutes is strongly encouraged by several statutes and OE policies. Ecosystem
restoration and sediment remediation projects have been successfully implemented
through EPA enforcement settlements with defendants. The NSI could assist
enforcement decision-makers in targeting enforcement activity in part based on the
potential for sediment remediation. The NSI could also, in certain individual cases,
help to establish the specific enforceable requirements for sediment remediation
projects implemented through enforcement settlements.
Office of Emergency and Remedial Response
Contaminated sites are evaluated on a case-by-case basis within the context of the
Hazard Ranking System (HRS) to determine whether they should be placed on the
NPL (Federal Register, December 14, 1990). Although the guidance provided
under Superfund is not specific to sediments, the HRS has been modified to
include an evaluation of both human health and ecological impacts due to
contaminated sediment exposure. Information contained in the NSI could be useful
during the first stage of the HRS. The level of danger to human health or the
environment could be assessed for in-place pollutants based on the severity of
problems at contaminated sediment sites, including harmful exposure of humans
through consumption of contaminated fish (as evidenced by fish consumption
advisories); severe alterations in benthic community structure in the presence of
elevated levels of pollutants; a high incidence of fin rot, tumors, or other
pathological indicators in fish inhabiting the area; and a high degree of toxicity in
the sediments at that site to benthic organisms.
The data in the NSI, together with other site-specific data, could also be used to
assist in identifying Potentially Responsible Parties (PRPs) by geographically
linking sources of contamination to chemical concentrations found in the
sediments.
Office of Federal Activities
The data in the NSI could be used in conjunction with other information to help
determine whether remediation is necessary at a site. Remediation could be
performed as a project feature or a mitigative measure.
Office of Federal Facilities Enforcement
OFFE could use the NSI data throughout the CERCLA process to ensure that
clean-up activities include the evaluation of contaminated sediment, where
appropriate, and to identify the need for supplemental projects and/or injunctive
relief at these CERCLA sites.
DRAFT
2-21
-------
NSI Framework
Office of Solid Waste
Once contaminated sediments are found at hazardous waste facilities, OSW
managers perform detailed assessments to determine the extent of contamination.
The NSI could assist in determining the geographic distribution of contamination
from historic or ongoing discharges of hazardous waste from a site.
Office of Toxic Substances
As mentioned for Superfund and OSW, OTS managers could use data in the NSI,
along with other site-specific data, to help identify possible violations of TSCA
regulations and responsible parties for enforcement-based remediation efforts by
geographically linking sources of contamination to the concentrations of chemicals
found in sediments. For example, under TSCA's PCB disposal rule, sediments
may be remediated based on site-specific risks (USEPA, 1992d).
Office of Wastewater Enforcement and Compliance
Once it has been confirmed that a discharger has violated its permit and that the
discharger has caused sediment contamination, OWEC managers could use the data
in the NSI together with other data to help prioritize these violations in the order
of severity of impacts, based on lexicological information, and initiate
enforcement-based remediation efforts.
Office of Wetlands, Oceans and Watersheds
Remediation activities for OWOW are limited to OCPD. This program office
could use the NSI to identify sites in marine and estuarine locations needing
remediation and to help plan programs for contaminated sediment removal or other
appropriate actions to be conducted through the Secretary of the Army.
Regional and State Programs
The NSI could assist various EPA Regional offices, the Great Lakes and Gulf of
Mexico Programs, and the states in development of contaminated sediment
remediation projects by providing background data to identify sites for remediation.
Dredged Material Management
Dredging of the Nation's waterways is necessary to maintain open shipping
channels for commercial and recreational navigation. The COE estimates that
approximately 3 percent of the 400 cubic yards of material dredged annually is
highly contaminated and that an additional 30 percent is moderately contaminated
(OTA, 1987). The data in the NSI could be used by dredged material management
2-22
DRAFT
-------
Chapter Two—Program Uses
programs to provide additional background information to help evaluate the need
for chemical and biological testing of proposed discharges.
Potential applications of the NSI to support dredged material management activities
include:
• Providing additional background data to help evaluate the potential for
sediment contamination of material proposed for dredging and disposal
(Tier I);
• Assisting in the identification of chemicals of concern that might be
targeted for more extensive bioeffects studies (Tier III); and
• Providing additional background information hi the design of
management and monitoring activities after disposal of dredged material
has occurred.
Office of Federal Activities
Because OFA is responsible for EPA's environmental review program, the data in
the NSI could be used by OFA and regional environmental review programs to
help assess the need for comprehensive, programmatic environmental impact
studies to address long-range planning for dredged material management.
Office of Federal Facilities Enforcement
When sites are subject to environmental regulation, the data in the NSI could be
used to assist in characterizing sediment in dredged material management activities.
Office of Science and Technology
OST could use the NSI data to evaluate whether the present tiered testing
requirements for dredged material disposal are adequate to determine the potential
for sediment contamination.
Office of Wetlands, Oceans, and Watersheds
OWOW's Oceans and Coastal Protection Division (OCPD) co-regulates with the
COE the disposal of dredged materials in ocean waters under section 103 of the
Marine Protection, Research and Sanctuaries Act (MPRSA). A tiered testing
protocol has been developed to determine the dredged material's suitability for
ocean disposal. The protocol consists of four tiers: evaluation of existing data on
potential sources of contamination, sediment chemical analyses, acute bioassays
and bioaccumulation tests, and biological community field studies (USEPA,
1992d). Managers could use the data in the NSI to provide additional background
DRAFT
2-23
-------
NSI Framework
information to assist in performing Tier I evaluations of the dredged material,
which use existing data such as sediment grain size, chemical concentrations,
evidence of fish tissue contamination, and records of spills or discharges to
evaluate the need for chemical and biological testing. The NSI could also be
useful in assisting managers in designing their monitoring programs at ocean
disposal sites, particularly hi situations where sediments have proven to be
problematic in harbors; however, no sediment chemistry or bioeffects data have
been collected from disposal sites (USEPA, 1992d).
OWOW's Wetlands Division (WD) and the COE regulate the discharge of dredged
materials into waters of the United States under section 404 of the Clean Water
Act (CWA). A testing manual for the evaluation of dredged material proposed for
discharge under section 404 of the CWA is currently under development. Modeled
after the manual developed for dredged material disposal in ocean waters, the
manual includes an evaluation of existing data on potential sources of
contaminants. Managers could use the data in the NSI to help determine the need
for chemical and biological testing, as required in performing Tier I evaluations of
dredged material.
Regional and State Programs
EPA Regional offices could use the NSI data to assist in developing appropriate
Tier I assessments of sediments targeted for dredging and disposal. The NSI could
also be used as a tool to help develop feasible management alternatives in cases
where dredging of contaminated sediments has been prohibited.
2-24
DRAFT
-------
CHAPTER 3
REVIEW OF BACKGROUND STUDIES AND
PILOT INVENTORIES ,
Prior to the development of the framework for the National Sediment Inventory
(NSI), EPA conducted a review of several existing studies of contaminated
sediment problems and contaminated sediment inventory pilot studies. The
purpose of this review was to assess the current understanding of problems
associated with contaminated sediments and to build on the experience of other
programs during the development of the NSI. The following sources of
information were reviewed as part of this effort:
• National Perspective on Sediment Quality (Bolton et al., 1985);
• An Overview of Sediment Quality in the United States (Lyman et al.,
1987);
• Contaminated Marine Sediments—Assessment and Remediation (NAS,
1989);
• Summary Report for Contaminated Sediment Assessments in U.S. EPA
Region IV Coastal Areas (USEPA, 1991b) and Evaluation of the Region
4 Inventory of Coastal Sediment Sites (USEPA, 1992a);
• EPA Region 5 Inventory of Contaminated Sediment Sites (USEPA,
1992b); and
• Progress Report on the Gulf of Mexico Program's Toxic Release and
Contaminated Sediment Inventories (TRI, 1992; unpublished infor-
mation);
• Proceedings of EPA's Contaminated Sediment Management Strategy
Forums (USEPA, 1992d).
The following presents a summary of the purpose, approach, and results of each
of these studies.
National Perspective on Sediment Quality (Bolton et al., 1985)
In November 1984, EPA sponsored a Sediment Criteria Development Workshop
to assist the Criteria and Standards Division (CSD) in focusing its efforts toward
sediment criteria development. The workshop presented the results of a prelimi-
nary national-scale inventory of existing sediment concentration data. The early
3-1
-------
NSJ Framework
nationwide inventory was develqped to determine the amount of data available to
assess the status of th? Nation's sediments and to perform a preliminary assessment
of the data. The review was not intended to be exhaustive, but rather to highlight
the strengths and weaknesses; of existing data. A specific issue addressed in the
document was whether many of the Nation's sediments would exceed sediment
thresholds, developed based on the Equilibrium Partitioning Approach, or whether
the majority of sites would be considered "clean" with the exception of a few
localized hot spots. Thus, the study attempted to identify the incidence and
geographic distribution of sediments with high sediment contaminant concentrations
to gain a perspective on the extent of the problem. In addition to existing sediment
concentration data, field studies that related concentration data to biological effects
were reviewed.
The study briefly discussed several approaches for formulating defensible sediment
criteria, including the development of criteria based on the following:
• Concentrations at a reference site (i.e., the Background Approach);
• Existing water quality criteria:
- Sediment-Water Equilibrium Partitioning Approach,
- Water Quality Criteria Approach;
• A set of new criteria developed from additional testing of benthic organ-
isms:
- Sediment-Biota Equilibrium Partitioning Approach,
- Bioassay Approach.
Because many of these approaches are yet to be fully developed and refined, only
limited details of the original proposed approaches were given, with the exception
of the Sediment-Water Equilibrium Partitioning Approach.
The preliminary national-scale inventory of sediment concentration data relied on
both marine and freshwater data housed in STORET (EPA's STOrage and
RETrieval System), as well as reports produced by state and federal agencies. The
initial assessment of sediment contamination was conducted by comparing sediment
concentrations to threshold values derived for this purpose. Where applicable,
threshold values used for ranking concentration data in STORET were based on the
Sediment-Water Equilibrium Partitioning Approach (JRB Associates, 1984a, b).
No effort was made to judge the adequacy of the Equilibrium Partitioning
Approach for establishing sediment criteria at that time. In the Equilibrium
Partitioning Approach, threshold concentrations are extrapolated from water quality
criteria Final Chronic Values by assuming that chemical equilibrium has been
3-2
-------
Chapter Three—National Inventory
established between the concentration of thp compound in the aqueous phase of the
sediment interstitial water and the concentration of the compound in the organic
carbon phase of the sediment. Because most of the sample locations in STORET
do not provide a value for sediment organic carbon content, for the purpose of the
initial study an organic carbon content of 4 percent was assumed. Threshold
values were also derived for metals using the Equilibrium Partitioning Approach,
even though partitioning of compounds to organic carbon has received limited
acceptance as a binding mechanism for metals.
For convenience, the concentrations reported in the monitoring data were divided
into four ranges, designated as Level I (less than the threshold value), Level 2 (1-3
times greater than the threshold value), Level 3 (3-10 times greater than the
threshold value), and Level 4 (greater than 10 times the threshold value). The
number of measurements made that were less than the detection limit was also
noted. As an exception, a background approach was used to establish the criterion
for PAHs. A sediment total PAH concentration of 1 ppm dry weight was
established as the cutoff between nonpolluted and slightly polluted sediments.
EPA initially identified 48 chemical contaminants for inclusion in the data review.
These contaminants fell into seven chemical categories:
• Polynuclear aromatic hydrocarbons,
• Pesticides,
• Chlorinated hydrocarbons,
• Mononuclear aromatic hydrocarbons,
• Phthalate esters,
• Metals, and
• Miscellaneous.
As a result of difficulties in accessing the data (possibly due to the way in which
the data were requested), of the 48 compounds identified, data were retrieved from
STORET for only 22. Notable exclusions were found among the PAHs, including
acenapthene, benzo(a)pyrene, naphthalene, fluoranthene, chrysene, and pyrene.
(Fluoranthene was not identified as a chemical for this study.) Notable exceptions
among the pesticides identified for the study included endrin and dieldrin. Over
255,000 data records were processed. No attempt was made to judge the quality
of the data or the sampling techniques. Marine and freshwater data were processed
3-3
-------
NSI Framework
separately because of the relatively large amount of STORE! data for sediments
in streams, rivers, lakes, and reservoirs compared to marine data.
Information in STORE! was augmented with biological information collected for
the same sites from the literature. Many sources of information in addition to
journal articles and publications in the open literature and federal agency reports
were identified; however, there was time to review only information from the open
literature. Other notable sources of information identified but not incorporated into
this analysis included the National Oceanic and Atmospheric Administration's
(NOAA's) Ocean Pollution Data and Information Network (OPDIN), the National
Oceanographic Data Center (NODC) Marine !oxic Substances and Pollutants Data
File, and the computerized inventory of long-term monitoring programs prepared
for NOAA's Ocean Assessment Division.
National maps and maps of each region of the United States were provided as part
of the completed national inventory to illustrate the geographic distribution of sites
with high contaminant concentrations. For each chemical, the 200 highest
concentrations, or average concentrations for sites with multiple measurements,
were identified and plotted. Symbols for various compounds were drawn on the
maps at approximately the latitude and longitude of the sampling site where that
compound had been detected. The open literature was then examined in an effort
to find information indicating a correlation between sediment concentrations and
benthic community structure. This effort was hampered by the large number of
parameters examined and the lack of case study data for many chemicals.
STORET contained extensive freshwater sediment chemistry data but generally
lacked biological data. . Some correlation was found between sediment
concentrations and benthic community impairment discussed in the literature for
metals, PCBs, and PAHs, but not for the other compounds. The number of
freshwater locations where data were collected for each compound is shown in
Table 3-1. Where more than one sample was collected at the same station, the
values reported were averaged.
The freshwater data were analyzed by plotting a cumulative frequency distribution
of the log of the concentration for each compound. Data points reported as zero
were noted to give an indication of the proximity of the threshold value to the
detection limit. To add some perspective, the median, 90th, and 95th percentile
concentrations were identified on the cumulative frequency distributions. Log
concentration versus cumulative frequency plots were developed for the PAHs
acenapthene, anthracene, benzo(a)anthracene, fluorene, and phenanthrene and the
pesticides aldrin, chlordane, DDT, heptachlor, lindane, and toxaphene. Of the
PCBs, only Aroclor 1016 (1221) data were evaluated. Aroclor 1016 (1221) is a
mixture of PCB congeners that contains a greater percentage of the lower
chlorinated analogues than do other Aroclor mixtures. The threshold criterion for
3-4
-------
Chapter Three—National Inventory
Table 3-1. STORET Sampling Stations for Freshwater Sediments by Compounds
Chemical
Copper
Lead
Mercury
Zinc
Nickel
Arsenic
Cadmium
Acenapthene
Anthracene
Benzo(a)anthracene
Fluorene
Phenanthrene
Diethylphthalate
Dimethylphthalene
Aldrin
Chlordane
DDT
Heptachlor
Lindane
Toxaphene
Aroclor 1016 (PCB)
Cyanide
Number of
Freshwater
Sampling Percent in
Locations Level 1
300
32,000
25,000
23,000
16,000
16,000
20,000
400
400
1,500
400
400
400
300
20,000
13,000
17,000
13,000
4,000
13,000
900
1,200
99.9
92.5
92
96
57
94
97.5
all
all
all
all
all
64
65
97
77
89
98
96
82
82
63
Percent in
Level 2
none
5.0
4.0
1.5
32
3.5
1.0
none
none
none
none
none
20
34
2
16
7
1
1
14
7
10
Percent in
Level 3
0.1
1.5
2.0
1.5
7.0
2.4
1.0
none
none
none
none
none
16
1
0.5
5
3
0.5
2
2
8
10
Percent in
Level 4
none
1.0
2.0
1.0
3.0
0.1
0.5
none
none
none
none
none
none
none
0.5
2
1
0.5
1.0
2
3
17
Aroclor 1016 had to be derived in a special manner for the mixture or an
assumption would have to be made regarding the most important component.
Pavlov's approach (JRB Associates, 1984a) was used to derive threshold values for
Aroclor 1016, which resulted in a threshold value believed to be lower than that
for PCBs as a group. A threshold value of 0.28 mg/kg was used for both
freshwater and marine areas, based on the water quality criterion for
hexachlorobiphenyl.
The cumulative frequency diagrams for chemicals having over 5,000 points plotted
as smooth s-shaped curves. The most useful information provided by the study of
freshwater sediments was an illustration of the general scope of potential problems
based on the percentage of sites above the threshold concentrations of the various
contaminants.
STORET data for metal concentrations in freshwater sediments were quite
extensive. For the metals lead, mercury, zinc, nickel, arsenic, and cadmium, the
3-5
-------
NSI Framework
database contained between 5,000 and 20,000 measurements for each metal. The
copper data set was restricted to wet weight determinations because of limitations
in the STORET data transfer. (There would be as much data for copper as for the
other metals if these difficulties could be overcome in future evaluations.) Only
0.1 to 3 percent of the sites had metal concentrations at Level 4 (more than 10
times the threshold value). Freshwater data were not available for chromium.
Each of the PAHs identified for the analysis had a STORET data set consisting of
approximately 400 freshwater measurements, with the exception of benzo(a)anthra-
cene, which contained approximately 1,500 measurements. Although the most
frequently measured PAH in freshwater sediments was benzo(a)pyrene, it was not
included in the assessment because of difficulties in obtaining the appropriate data
for this compound from STORET. The PAHs were ubiquitous, as reflected in the
low incidence of no-detects. The overwhelming majority of the data was at
Level 1 (i.e., no points on the cumulative frequency diagrams fell above the
threshold values). Ninety-five percent of the reported concentrations were below
4.3 to 5.6 mg/kg, depending on the compound. Ninety-five percent of the data'for
benzo(a)anthracene were below 0.014 mg/kg. The only region where site-specific
biological information was available for these compounds was the Great Lakes
region, where toxicity tests had been conducted on the amphipod Diporea sp.
The STORET data set contained approximately 300 freshwater measurements for
both phthalate esters. Cumulative frequency plots of log concentration for
diethylphthalate and dimethylphthalate revealed that these compounds were also
ubiquitous in the environment. In addition, about 35 percent of the sites had
concentrations above the threshold. There were a few localized hot spots, although
none of the data fell in the Level 4 range (i.e., more than 10 times the threshold).
STORET data on pesticides in freshwater sediments were relatively extensive;
between 4,000 and 20,000 measurements were available for each pesticide. The
median pesticide concentrations, in general, were well below the threshold
concentrations. There were a few hot spots with more than 10 times the threshold
value, but most sites fell within Level 1.
The STORET database contained 917 freshwater sediment measurements for
Aroclor 1016. Approximately 18 percent of the data for Aroclor 1016 fell above
Level 1, and 3 percent fell above Level 4. Insufficient STORET data were
available to analyze any monoaromatic hydrocarbons.
Contaminant concentration data for marine environments were gathered from
published literature, literature with limited distribution, and STORET. The data
presented hi the report were incomplete in terms of both incorporation of existing
data and geographic coverage, but were provided to present a preliminary
perspective on a national basis. A literature search for marine sediment data
3-6
-------
Chapter Three—National Inventory
uncovered a wealth of information on animal-sediment relations and contaminant
effects in addition to chemical concentration data. Marine/estuarine data were
placed on national and regional maps in the same manner as the freshwater data.
Because STORET contained limited marine/estuarine data, only median
concentrations of the various chemicals, rather than cumulative frequency
distributions, were reported.
The status of the marine sediments was evaluated based on a comparison with
threshold values developed using the Equilibrium Partitioning Approach, except in
the case of polynuclear aromatic hydrocarbons, where the Background Approach
was used. A sediment total PAH concentration of 1 ppm dry weight basis was
chosen as the cutoff between polluted and nonpolluted sediments based on a
publication by Kites et al. (1980).
The complete data for coastal sites were included as a tabular listing by site and
contaminant that either indicated that no data were available or specified the
magnitude of the concentration relative to the threshold value (i.e., Level 1, 2,
etc.). PCBs, PAHs, and metals affected the most sites.
The most useful information that the study of marine and freshwater sediments
provided was an illustration of the general scope and magnitude of the problem.
Most of the sites had concentrations below the thresholds established for chemicals
based on the Equilibrium Partitioning Approach. The median concentrations, in
general, were well below the threshold concentrations. This was especially the
case for metals and PAHs. Since the use of the Equilibrium Partitioning
Approach for metals is questionable, other threshold values were also considered
in the study. The lower threshold values for copper, lead, mercury, zinc, nickel,
and cadmium established by EPA Region 5 would have resulted hi between 6 and
31 percent of the data being reclassified above the Level 1 range. For arsenic, the
alternate threshold of 3 mg/L suggested by EPA Region 5 would have reclassified
62 percent of the sediments into Levels 2, 3, and 4.
The PAHs and phthalates were ubiquitous, as reflected hi the low incidence of no-
detects. Although the phthalates were a concern in terms of the percentage of sites
above the threshold, little toxicity testing was available for these compounds.
PCBs and cyanide had the highest percentage of sites falling into the Level 4
category. Historic lack of concern about cyanide in sediments has led to a paucity
of toxicity testing of cyanide in benthic organisms.
Marine contaminated sediments tended to be localized, with the vast majority of
marine sediments unpolluted or unstudied. However, severe biological effects have
been attributed to sediment contamination in the Puget Sound and New York Bight
regions despite low concentrations of individual pollutants. The low concentrations
3-7
-------
NSJ Framework
reported for these areas may reflect the lack of marine sediment chemistry
sampling data available at the time of this report.
An Overview of Sediment Quality in the United States (Lyman et al., 1987)
The purpose of this study was to provide an overview of sediment quality in the
waters of the United States with emphasis on contaminated sediment sites. The
study was undertaken as an initial step toward the goal of compiling a
comprehensive national assessment of the nature and extent of sediment
contamination problems. Specific objectives listed in the report were to:
• Document the extent to which various sources are associated with
sediment contamination problems;
• Document approaches to, and the effectiveness of, remediation of
sediment contamination;
• Provide documentation of regional and state approaches to sediment
contamination problem identification and response; and
• Provide support and perspective to the development and eventual
implementation of sediment quality criteria through an inventory and
description of known contaminated sediment problem areas.
The study attempted to provide a picture of the geographic distribution, areal
extent, and severity of the contaminated sediment problem and to provide a better
understanding of contaminant sources (both ongoing and historic), the sites
involved, and the types of pollutants and their impacts. It was believed that
existing data could be used to help establish sediment quality criteria.
This study primarily focused on identifying specific locations or problem areas
with contaminated sediments, rather than on obtaining estimates of the
concentration levels in sediments. Only an inventory of existing data was
undertaken because it was believed that existing data in their current state do not
lend themselves to in-depth analysis and review. Existing data, although extensive
in some regards, are associated with varying sampling and analytical methods and
are widely scattered in many state and federal offices. Often, data have not been
compiled in a computer database or they reside in incompatible systems.
Information was gathered for the study from the recently published literature on
sediment contamination and from a series of telephone and personal interviews
with representatives of various federal and state agencies that deal with
contaminated sediments. Agencies contacted included NOAA, U.S. Army Corps
5-8
-------
Chapter Three—National Inventory
of Engineers (COE), U.S. Army Toxic and Hazardous Materials Agency
(USATHAMA), U.S. Fish and Wildlife Service (FWS), U.S. Geological Survey
(USGS), U.S. EPA Environmental Research Laboratory-Narragansett, and the 10
EPA Regions. Contaminated sediment case studies with data documenting causes
and effects were collected along with descriptions of contaminated sediment
management approaches used by various state and federal agencies. Because of
its emphasis on contaminated sediment management, this study relied to a greater
degree on information presented in the literature and expert opinion than on
chemical concentrations housed in databases. The classification of sediments as
contaminated was somewhat arbitrary because it was based on the diverse
classification techniques used by various agencies. The data collected,
unfortunately, were somewhat anecdotal and could not be used to quantify the
extent of the problem.
The information collected was used to compile a list of sediment contamination
problem areas, with most attention given to sites where documentation was
available. The list was not prepared based on a ranking of the worst sites, nor was
it intended to be comprehensive. It represented an early attempt to link sites with
sources. 'Based on information provided by the contacts, data gathering produced
information on sites that were thought to contain in-place pollutants. The study did
not attempt to present a detailed and complete analysis of in-place pollutants and
was limited to providing subjective information. Studies reviewed of a more
general nature included the following:
• Identifying and Prioritizing Locations for the Removal of In-Place
Pollutants (Johanson and Johnson, 1976);
• National Perspective on Sediment Quality (Bolton et al., 1985);
• Removal and Mitigation of Contaminated Sediments (SAIC, 1985);
• Preliminary Survey of Contaminant Issues of Concern on National
Wildlife Refuges (USFWS, 1986); and
• National Status and Trends Program: Progress Report and Preliminary
Assessment of Findings of the Benthic Surveillance Project - 1984
(NOAA, 1987).
In addition, existing databases were searched in an effort to compile existing
sediment monitoring data for the purpose of determining how well different areas
of the United States were represented by sampling. These data were compiled by
the EPA Regions. As a result, some Regions were better represented than others,
reflecting the lack of uniformity in regional data-gathering efforts.
3-9
-------
NSI framework
The review of documents and discussions with experts resulted in the identification
of 184 separate sites with noted or perceived impacts from in-place pollutants.
Most of the sites were located in the Northeast, along the Atlantic and Gulf Coasts,
and in the Great Lakes region. These are also the regions where most of the data
are concentrated. Surface waterbodies of all types were affected. Heavy metals,
PCBs, pesticides, and PAHs were the most frequently mentioned contaminants!
Biological impacts included reproductive impacts, effects on the structure of the
community, and fish kills. Bioaccumulation of compounds from sediments was
severe enough in some cases to warrant fishing bans or occasionally to prevent the
use of water supplies for drinking water.
Few new data were collected in this study. Instead, the study relied heavily on the
results of Bolton and others (1985) for information on the number and level of
various pollutants at sites across the United States. The tally of the number of
sites was based on concentration data reported in published and unpublished
literature and data housed in STORET for marine and estuarine sites. Since the
compilation of data in the 1985 study was far from complete, it was difficult to
draw more than very general conclusions. The study by Lyman and others (1987)
presented the same data collected earlier in a different form, emphasizing the
identification of specific harbors, bays, rivers, estuaries, lakes, and waterways that
were potentially impacted and specifying the contaminants. Information from a
survey of national wildlife refuges was treated in the same manner. The
concentration frequency diagrams presented by Bolton et al. (1985) were reviewed.
The study by Lyman and others (1987) identified the sites, by chemical category,
for which one or more of the 48 chemicals identified by Bolton and others (1985)
had been measured. Thus, the representation of each of the 48 chemicals, or
groups of chemicals, at sites where chemicals had been detected was evaluated.
The number of the sites having concentrations above predetermined threshold
values was also indicated.
General observations and conclusions of the study were as follows:
• All major harbors, rivers, and estuaries bordered by industrialized or
urbanized areas contain elevated levels of metals, organics, and other
anthropogenic contaminants. Sometimes areas of contamination are
highly localized and related to a point source discharge of industrial or
municipal effluent. Pinpointing sources is not always an easy task
because sediments can be highly mobile and can be altered by dredging.
• Although field studies documenting the relationship between elevated
levels of contaminants in sediments and effects were limited at the time
of the study, impacts of in-place pollutants were believed to be
significant. In places where other sources of pollution have been
3-10
-------
Chapter Three—National Inventory
regulated, in-place contamination may be the primary source
contributing to the impacts.
• The historical record of pollutant concentrations in sediments preserved
by sediment cores shows that in-place contamination has increased
rapidly through this century.
Heavy metals and metalloids, e.g., arsenic, were the most frequently mentioned
contaminants; 69 percent of the sites contained at least one metal or metalloid.
PCBs were mentioned at 34 percent of the sites; PAHs at 19 percent of the sites;
pesticides at 26 percent of the sites; and other organics (including oil and grease,
hydrocarbons, volatile organics, phenols, base/neutrals, and dioxin) at 25 percent
of the sites. The pesticides most frequently found were DDT and its derivatives,
dieldrin, and chlordane. These results were similar to those obtained by Bolton et
al. (1985) for marine and estuarine sediments.
Specific locations and one or two pollutants of concern for each site were listed.
The results presented by Lyman et al. (1987), however, may be misleading because
the compounds identified reflect sampling at these sites that may have failed to
detect additional compounds simply because they were not tested. Organics, for
instance, are not as frequently monitored as metals. For example, the New
England Division of the COE monitors for a list of metals and total PCBs. The
limited number of compounds monitored are used by the COE as indicators of
sediment contamination. Typically, only a few chemicals are analyzed by the COE
and others. Nevertheless, the study clearly documented the existence of in-place
pollution problems.
Contaminated Marine Sediments—Assessment and Remediation (NAS, 1989)
This report was prepared by the Committee on Contaminated Sediments and the
Marine Board, Commission on Engineering and Technical Systems, of the National
Research Council, based on the outcome of a symposium and workshop conducted
by the Committee on Contaminated Sediments. The committee was convened in
response to the growing national awareness of problems resulting from
contaminated marine sediments and was composed of experts in aquatic toxicology,
dredging technology, resource economics, sediment dynamics and transport, benthic
ecology, environmental law, and public policy. At the symposium, invited papers
were presented on the extent of sediment contamination across the Nation, methods
for classification of sediment contamination, risks to human health and the
ecosystem, and sediment resuspension and contaminant, mobilization. Five case
studies were examined to illustrate the different ways in which sediment
contamination problems are being addressed: New Bedford Harbor, Massachusetts,
and the upper Hudson River, New York (PCBs); James River, Virginia (kepone);
3-11
-------
NSJ Framework
and Commencement Bay, Washington, and the Navy Homeport Project in Everett
Bay, Washington (variety of chemicals). Two consecutive work group meetings
were then held to discuss the extent, methods of classification, significance with
regard to biological communities and human health, and resuspension of
contaminated sediments, and the selection of appropriate, economically feasible
management strategies and remedial technologies for handling contaminated
sediments. Major findings and recommendations of the committee with regard to
the issues covered at the symposium are detailed in the report.
The committee defined contaminated sediments as those sediments which contain
chemical substances at concentrations that pose a known or suspected
environmental or human health threat. Although studies by the U.S. EPA and
NOAA's National Status and Trends Program have identified widespread
contamination and hot spots in coastal waters near major urban areas the
'Committee concluded that "adequate data do not currently exist' for
comprehensively pinpointing or prioritizing candidates for remedial action" (NAS,
1989, 1-2). In addition to collecting data, the committee noted that research,'
development, and the use of assessment methodologies must focus on identification
of biological impacts and favored a tiered testing approach. Human health risks
should be examined from an epidemiological perspective. Other recommendations
included research into sediment transport, dredged material management strategies,
contaminant source control, and well-focused monitoring efforts.
The NAS (1989) report noted that the location and extent of contaminated marine
sediments should be comprehensively assessed on a national basis, but that such
efforts should not duplicate the National Status and Trends (NS&T) program or
involve detailed mapping. This national assessment could delineate contaminated
sediments, while the search for new sites or rectification of known sites could
continue as remediation was under way.
The importance of having appropriate data for analysis of contamination was noted.
The committee found that different programs had collected data for different
purposes with varying approaches and stated that data should not be used beyond
the limits or intent of the original monitoring program. Furthermore, there have
not been any generally accepted and validated sampling and analysis techniques,
testing protocols, or classification methodologies that would allow data
comparisons. The Committee proposed setting national criteria, standards, or
guidelines to achieve this purpose. An interagency committee was also proposed
to evaluate existing and emerging data on sediment contamination to focus limited
resources where research and monitoring were needed, to reduce redundancy, and
to eliminate improper uses of data. Criteria review, laboratory bioassays,' and
infaunal surveys should be used to determine and evaluate the significance of
contamination.
3-12
-------
Chapter Three—National Inventory
With regard to the use of data from various programs, workshop participants noted
that EPA's STORET system contained data that had not been validated by
comparison with primary literature sources and different sample collection
techniques and analytical protocols had been used at different sites. One of the
work groups believed that most of the STORET and literature data on sediments
were best interpreted in a qualitative sense.
Summary Report for Contaminated Sediment Assessments in U.S. EPA
Region IV Coastal Areas (USEPA, 1991b)
In an effort to better manage coastal and marine waters within Region IV, existing
data from reports and state and federal agency databases were compiled into a
dB ASEIII+™ format. The resulting database of contaminated sediment sites was
designed to help provide an understanding of potential and actual contaminated
sediment problems and to assist hi coastal management decision-making.
Data were collected from 80 different references including universities, COE
dredging evaluations, .state reports, NOAA Sea Grant Program reports, laboratory
reports, journal articles, city reports, South Florida Management District reports,
and studies by FWS, USGS, and EPA. A significant portion of the sediment
quality information obtained came from Florida's Department of Environmental
Regulation sediments database. Most of the data available for liquid, suspended
solid, and solid bioassays and bioaccumulation and toxicity tests were obtained in
conjunction with dredged material evaluations by the COE using procedures
outlined in the EPA/COE dredged material testing manuals. Other data were
gleaned from various reports and in most cases represented summaries of
information rather than raw data.
Data quality objective considerations for including or excluding data focused
primarily on the availability of records of where samples were taken; data were not
included in the inventory if the ,sample location could not be determined. If the
sample data could not be located, the report data were used. Information on
sampling methods, analysis methods, and parameters analyzed was included in the
files if available. The parameter fields from ODES were used initially to describe
the data; however, efforts to enter data into ODES were abandoned and the data
were entered into dBASE™ files.
The data collected for each sample included sample identification and location
information. Each sample was given a unique identification number. (The latitude
and longitude descriptors were not available for the majority of sites; therefore,
locations were estimated using maps.) Only concentration data for sediments were
entered. The depth at which the sample was taken was included when it was
available. The 129 EPA priority pollutant numbers were used to identify the
3-13
-------
NSI Framework
chemicals. If a chemical was not on the priority pollutant list, an additional
number was used. Only information available for specimens identified in the EPA
EPA/COE's "Green Book" was included in the bioassay data. Although a
screening process was developed to determine the minimum level of acceptability
for data pnor to their inclusion in the inventory, the study mentioned a high
variability in the quality of the data collected.
The Florida Department of Environmental Regulation (FDER) represented the
major source of government data. A regional STORET retrieval was not
mentioned in the pilot inventory although some of the data reviewed had been
entered into STORET. (STORET contains primarily inland data, whereas the
Region 4 inventory focused primarily on coastal areas.) The FDER files contained
.data on metals, pesticides, PAHs, alkanes, and phenols. Data on the latitude and
longitude of stations and on sampling and analytical methods were also included
in the FDER files.
The pilot inventory represents data collected from all coastal states in the Region
with the majority of the sample sites located in Florida (571 of 817 sites)'
Analytical data were available for metals and for organic constituents for most of
the sites represented in the inventory. Analytical data for pesticides were much
more limited.
Preliminary review of the pilot inventory data was based on the comparison of
concentration data to NOAA guidance levels for lead and copper. Analysis
indicated that lead and copper concentrations exceeded guidance levels for
sediment contamination in several samples taken from sites located primarily in
Florida. Locations that exceeded NOAA guidance levels for lead included Miami
River, Florida; Perdido Bay, Florida; Pascagula Ship Channel, Mississippi; Lower
Hillsborough River, Florida; Indian River Lagoon, Florida; St. Lucie Estuary,
Flonda; St. Johns River Estuary, Florida; Choctawhatchee Bay, Florida; Manatee
Pocket, Florida; and Charleston, South Carolina. Locations that exceeded the
guidance levels for copper in Florida, included Miami River, Lower Hillsborough
River, Indian River Lagoon, St. Lucie Estuary, St. John's River Estuary,
Choctawhatchee Bay, and Manatee Pocket.
A more detailed evaluation of the Region 4 data has been conducted (Evaluation
of the Region 4 Inventory of Coastal Sediment Sites, USEPA, 1992a). This more
complete analysis indicated that the data were characterized by a lack of TOC and
grain size data for normalization of chemical concentrations of contaminants, as
well as limited biological toxicity data, and that a majority of samples were from
contaminated sites versus "background" samples. Therefore, the MacDonald
(1992) weight-of-evidence approach adopted for the proposed FDER sediment
quality guidelines was used to obtain the Threshold Effects Level (TEL) and
Probable Effects Level (PEL) values for 20 metal, organic, and pesticide
3-14
-------
Chapter Three—National Inventory
contaminants. Identified regional chemicals of concern included metals (arsenic,
lead, mercury, copper, silver, and zinc) and organics (chrysene, pyrene,
fluoranthene, phenanthrene, PCBs, and acenapthene). The calculated Effects Index
(the sum of the ratios of all contaminants to their TELs for the given site) was also
used to rank sites by levels within the categories of metals, organics, and
pesticides. Further work is under way to address limitations identified in the use
of this procedure and to evaluate sources of contaminants in relation to
contaminated sites.
EPA Region 5 Inventory of Contaminated Sediment Sites (USEPA, 1992b)
A pilot inventory of sites that are suspected or known to have contaminated
sediment problems is being conducted by EPA Region 5. The inventory is being
developed as part of EPA's national strategy for addressing contaminated sediment
issues. Some of the objectives of the inventory are to assist hi determining the
extent of contaminated sediments in the Region; to aid in locating problem sites;
to help in determining where additional studies are needed; and to aid in
determining where prevention, remediation, and enforcement actions are needed.
The pilot inventory has also been designed to serve as an initial framework for the
National Sediment Inventory.
Information is being collected from several sources for inclusion in the pilot
inventory. These sources include the EPA Region 5 Environmental Review
Branch, EPA Region 5 Superfund, EPA Region 5 Water Division, EPA Great
Lakes National Program Office, COE, USGS, STORET, and State agencies.
Several of these agencies use STORET to store their data. In these cases, data
from STORET were used only when the primary source of data was not available.
The information on a particular site was taken from only the two most recent
available reports for inclusion in the inventory.
The database is designed to include information on the site identification, site
characterization, sediment sampling, and biological sampling results. The site
identification specifies a location by the site name, county, state, latitude and
longitude, USGS hydrologic unit, and EPA reach number. The reach numbers for
the site locations are obtained from STORET. The characterization of the site
describes the area as a whole and includes data on the size of the sampling area,'
the reach description, the industries within the reach, receiving waters, land use,
and site status. The characterization also includes whether the sampling area is
within a Federal Navigation Channel and, if so, the dates of the last two dredgings,
known impacts to the site, and fish advisory information. The sediment sampling
data fields divide the sampling information on chemicals into chemical classes.
Additional information on the physical description of the sediments, the results of
grain size analysis, and benthic community information are included. The sediment
3-15
-------
NSI Framework
sample data also include the number and type of samples, the sampling equipment
used, and the depth of maximum concentration if a core sample was used. Data
fields are also provided for additional types of testing and comments. The
biological sampling data fields include a complete reference for each biota
sampling. The date of sampling, species sampled, and type of samples (including
tissue analysis, benthic community analysis, and sediment toxicity/bioassay testing)
are recorded. Sediment toxicity testing data include date of sampling, date of
testing, test duration, species, type of assay, number of samples, and results. A
data field for comments is also included.
The data fields in the Region 5 inventory are divided into key fields and abstract
fields. Key fields include information such as the site name, the state, sampling
dates, chemical parameters, and site characteristics. The abstract fields provide
additional information such as the descriptive variables and references. The
abstract fields cannot be searched. The Region 5 contaminated sediment inventory
data are being entered into a dBASE™ file. The Region V inventory currently
includes data for all locations in the Region for which data were available.
Prioritization of these sites is currently under way.
As of the date of the Region 5 draft report, sediment and fish sampling information
had been collected for most of the states within Region 5. The State of Wisconsin
identified approximately 190 sediment sampling sites. The Wisconsin Department
of Natural Resources (WDNR) is collecting and organizing its sediment data, and
much of the information is in the form of sampling results rather than in report
format. Fish sampling information was obtained from WDNR's Fish Sampling
database.
The State of Michigan identified approximately 99 sediment sites. The Michigan
Department of Natural Resources (MDNR) maintains a database of sampling
reports. The MDNR also publishes an annual fish sampling report that includes
fish sampling reports for most sediment sites.
The State of Minnesota identified approximately 45 sediment sampling sites.
STORET is the primary source of sediment information in the state. The
Minnesota Pollution Control Agency supplied fish sampling information for the last
2 years. Sampling information prior to the last 2 years was in STORET.
Sediment data have been collected from several departments within the Ohio
Environmental Protection Agency. Additional data obtained in a lakes sampling
study and a toxic metals sediment study, as well as fish sampling data, will be
provided by EPA. Approximately 14 sites have been identified thus far from the
Ohio data.
3-16
-------
Chapter Three—National Inventory
As indicated in the EPA Region 5 draft report, data collection for the inventory had
not been completed in some states. Only data for Lake Michigan Basin sites for
Indiana and Illinois have been entered into the database. Other data will need to
be compiled and collated in the contaminated sediment database in order to address
the purposes of the National Sediment Inventory.
Progress Report on the Gulf of Mexico Program's Toxic Release and
Contaminated Sediment Inventories (TRI, 1992; unpublished information)
The Toxic Substances and Pesticides Subcommittee of the Gulf of Mexico Program
and EPA Region 6 have jointly funded two projects under the Gulf of Mexico
Program to examine sediment contamination in the nearshore waters of the Gulf
of Mexico. Because sediments in the Gulf are heavily impacted by industrial
discharges, especially those related to the oil industry in Texas and Louisiana, a
Pollutant Source Inventory for sediment-associated chemicals was undertaken to
identify the amounts and kinds of chemicals discharges. A Contaminated Sediment
Inventory that identifies sites of contamination in the nearshore waters of the Gulf
of Mexico is also being developed.
The Pollutant Source Inventory was prepared under the direction of the Toxic
Substances and Pesticides Subcommittee of the Technical Steering Committee for
the Gulf of Mexico Program. The inventory was compiled from data in (1) the
Toxics Release Inventory of the Gulf of Mexico (an inventory and evaluation of
surface water discharges for industrial and municipal sites in the coastal zone as
reported hi EPA's Toxic Chemical Release Inventory (TRIS) and Permit
Compliance System (PCS)); (2) a separate evaluation of pesticides applied to
cultivated fields that could drain into the Gulf (Pait et al., 1992); and (3) a separate
evaluation of discharges from nearshore oil and gas platforms, including land oil
spills and fluids forced out of sediments during offshore drilling operations
(produced waters). The sites were limited to those known to be contaminated, such
as sites identified as not meeting water quality standards in accordance with section
304(1) of the Water Quality Act, those sites closed to fish and shellfish harvesting
because of contamination, and other contaminated sites known to the
Subcommittee. The potential impacts, based on toxicity, of chemicals and
pesticides from these sources were compared for 29 estuarine drainage basins
entering the Gulf. A report on the results of the Source Inventory, Impact of Toxic
Substances and Pesticides on Nearshore Gulf of Mexico: A Preliminary
Comparison (Toxicity Indices) of Twenty-five Estuarine Drainage Systems Based
on Release of Toxics from Industrial and Municipal Sites and Pesticide Run-off
from Agricultural Operations in 1989, is now under review.
The Sediment Inventory for the Gulf of Mexico is patterned after that of
Region 4, using information obtained primarily from databases maintained by
3-17
-------
NSI Framework
Region 4 and Region 6, and from monitoring efforts in the coastal counties of
Texas, Louisiana, Mississippi, Alabama, and Florida. The purpose of the inventory
is to determine the nature and extent of sediment contamination in the nearshore
waters of the Gulf of Mexico, identifying sites in terms of locations, types, and
potential impact of pollutants present. The project will ultimately prioritize
geographic areas of concern with respect to the potential toxicity of the sediment.
Data requested for retrieval and entry include detailed monitoring data collected
since 1980 on sediment chemistry, toxicity, bioaccumulation, and associated grain
size. Databases examined include the Region 4 inventory, STORET, EMAP
(Environmental Monitoring and Assessment Program), Mississippi Sound Study,
Mobile Bay Study, Houston Ship Channel Study, and Calcasieu Lake Study.
Common elements among the data sets, such as station identification, locations
(longitude/latitude), analyses, and concentrations of chemicals, are loaded into a
FOCUS database. Variable or unique elements that help to determine a data set's
utility are identified in a written abstract with the contact for the submitting
agency. These elements include QA/QC information, methodology, evaluations,
additional data, and STORET analysis. The chemical and biological information
being collected is concurrently assessed by a set of sediment quality guidelines
based on the ER-L (effects range-low) and ER-M (effects range-medium) values
and MacDonald's (1992) TELs (threshold effects levels) and PELs (probable
effects levels) for approximately 30 contaminants. Data gaps are also being
identified, including locations that lack biological testing where chemistry data
levels of contamination equal or exceed guidelines, locations where contaminants
of concern have not been tested, and locations that have been undersampled.
The Sediment Inventory is nearly completed. Site evaluations will determine the
scope of sediment contamination problems and identify toxic chemicals and
geographical areas of concern. Additional ranking procedures based on the
available and missing chemical and biological data will be used to prioritize sites
potentially in need of remediation and areas in need of additional monitoring.
Contaminated sites will ultimately be matched with sources from the Gulf of
Mexico's Pollutant Source Inventory and reported fish consumption advisories.
This information will be available in a Gulf of Mexico Program toxics and
pesticides characterization report later this summer (Catherine Fox, USEPA, Risk
Assessment and Management Branch, Standards and Applied Science Division,
Office of Science and Technology, personal communication, 11 March 1993).
Proceedings of the EPA's Contaminated Sediment Management Strategy Forums
(USEPA, 1992d)
During 1992, EPA sponsored a series of public forums for the purpose of
discussing the draft outline of the Agency's Contaminated Sediment Management
3-18
-------
Chapter Three—National Inventory
Strategy. Each forum addressed a different issue related to the Strategy. The
topics for each forum were as follows:
• The geographic extent and severity of contaminated sediments (April 21-
22, 1992, Chicago, Illinois);
• Building alliances among federal, state, and local agencies to address the
problem of contaminated sediments (May 27-28, 1992, Washington,
DC); and
• Outreach and public awareness (June 16, 1992, Washington, DC).
The following subsections describe the specific issues discussed at each of the
forums and present the conclusions reached and recommendations made as a result
of these discussions.
Extent and Severity of Contaminated Sediments
Three specific topics of concern were addressed at the first forum: (1) the extent
of sediment contamination, (2) the severity of contamination with respect to human
health effects, and (3) the severity of effects with respect to ecological effects. A
series of presentations were given addressing each of these topics.
During the first series of presentations, evidence was given illustrating the
widespread nature of the problem of sediment contamination, with toxic hot spots
occurring in many areas across the United States. For example, the COE estimates
that 12 million of the 400 million cubic yards of sediment dredged each year from
the Nation's waterways are contaminated. Data from NOAA's National Status and
Trends Program indicate that sediment contamination is most severe near densely
populated urban areas.
Data were also presented to suggest that direct or indirect exposure to contaminants
in sediments can adversely affect human health. Although no acute or observable
toxicity resulting from exposure to contaminated sediments is evident, effects on
human health are seen in potential increased incidence of cancer, reproductive or
developmental toxicity, or neurotoxicity. The consumption of fish tissue
contaminated through bioaccumulation from sediments is a major concern, although
the effects of chronic exposure to contaminants from fish tissue is poorly
documented.
Evidence exists to link elevated concentrations of metals and organic chemicals in
sediment and elevated tissue burdens in aquatic organisms. Such tissue burdens
can result in a variety of effects including neoplasms, cataracts, enzyme induction,
fin rot, other lesions, decrease in the abundance and variety of benthic species, and
3-19
-------
NSI Framework
others. However, assessing ecological effects is more difficult than delineating the
extent of sediment contamination or even estimating potential human health effects
because important effects manifest themselves in ways that are often difficult to
detect.
Two major conclusions were reached at the end of the first forum: (1) contamin-
ated sediments are a national problem and (2) human health problems and
ecological harm have been documented at a number of contaminated sediment
sites. In addition, participants agreed that existing data on the extent of sediment
contamination are decentralized, and they generally supported the development of
a national inventory of contaminated sites based on site chemistry, health effects,
and intended uses. Participants also agreed that integrated assessments
encompassing the following are necessary to appraise the status of an ecosystem:
• Toxicity assessments;
• Sediment chemistry analyses;
• Tissue chemical analyses;
• Pathobiological studies; and
• Community structure studies.
Building Alliances
The forum on building alliances among federal, state, and local agencies to address
the problem of contaminated sediments was conducted in three parts to address the
following activities: assessment, prevention, and remediation. Presentations were
made regarding cooperation among the various government sectors during each of
these activities.
Participants agreed that the assessment of contaminated sites is an area in which
EPA's Contaminated Sediment Management Strategy needs clearer direction. The
Strategy must define contaminated sediments more precisely and propose a
mechanism for the effective use of assessment data to support sediment
management programs. In addition, participants felt that the Strategy should
identify and promulgate consistent quality assurance/quality control protocols for
sediment sampling and bioeffects testing, focus more attention on nonpoint source
(NPS) contamination, and actively encourage coordination with state agencies.
Panelists were divided on the following two issues:
• Should the Strategy encourage an effects-based assessment approach or
the development of numerical sediment quality criteria?
3-20
-------
Chapter Three—National Inventory
• Should the Strategy specify uniform effects-based testing methods or
call for different but comparable effects-based testing methods?
In terms of pollution prevention, participants urged EPA to clarify several aspects
of the Strategy. Participants felt that EPA should state clearly how sediment
quality criteria will be used as part of the Strategy. They also felt that EPA should
include stronger provisions for prevention of NFS contamination and should
identify ways to improve coordination between state and federal agencies.
Participants also urged EPA not to rely too heavily on models and to recognize the
value of case studies in understanding the problems associated with contaminated
sediments.
Participants agreed that contaminated sediment remediation must be limited to
human health and ecological risk reduction, although some participants cautioned
that human health risk assessments that are too conservative can lead to higher
remedial costs with little marginal benefit. Participants suggested that the Strategy
also address liability to facilitate more timely remedial actions. Finally,
participants believed that EPA should provide guidance on specific issues related
to managing contaminated sediments, including the following:
• Remediation of oil spills;
• Disposal of contaminated dredged material;
• Aquatic construction and maintenance activities;
• Management of sediments contaminated by stormwater discharges and
other nonpoint sources; and
• Use of natural recovery options.
The following overall conclusions were agreed on following discussions of the
need and approach for building alliances to address the problem of contaminated
sediments:
• EPA should expedite implementation of the Strategy;
• Development of a contaminated sediment inventory is a high-priority
need;
• More attention should be focused on NPS contamination hi the Strategy;
; • Addition of sediment toxicity and bioaccumulation tests to chemical
registration under the Federal Insecticide, Fungicide, and Rodenticide
3-21
-------
NSJ Framework
Act (FIFRA) and the Toxic Substances Control Act (TSCA) is a high-
priority need to prevent point and .nonpoint source contamination of
sediments; and
• Consideration should be given to developing an integrated federal
agency strategy on contaminated sediments.
Outreach and Public Awareness
Recommendations for effective public outreach were made by representatives from
state government and private sector organizations. The private sectors represented
included the regulated community, environmental advocacy groups, and public
awareness groups. The following recommendations concerning outreach and public
awareness were made by representatives from each of these groups:
• State government
- EPA should use existing state networks for public involvement and
information dissemination and allow states flexibility in adapting the
Strategy to local situations.
• Regulated community
a
- Sediment contamination is a local, "hot spot" problem, not
national problem.
- EPA should subject all data and conclusions about sediment
contamination to rigorous review.
- Contaminated sediments should be defined with respect to human
health and ecological risk, not numerical chemical criteria.
Environmental advocacy groups
- Current EPA public outreach efforts are inadequate.
- The public lacks confidence that EPA has a rational, defensible
program to manage contaminated sediments.
- EPA should take advantage of existing communication networks to
present information on contaminated sediments; establish face-to-face
contact whenever possible through meetings, workshops, or
conferences; and develop more engaging written and graphic
information.
5-22
-------
Chapter Three—National Inventory
• Public awareness groups. EPA should engage in the following activities
to promote public awareness:
- Make sure outreach efforts address specific needs of various target
audiences.
- Design materials to foster participation in effective policy making.
- Build consensus among conflicting interests.
- Develop a framework of institutions that will be self-sustaining and
carry the work of sediment management into the future.
Participants in the forum on outreach and public awareness were in agreement on
several issues related to EPA's Contaminated Sediment Management Strategy.
Participants agreed that EPA should get the public involved as soon as possible,
clearly indicating how long cleanup will take, conveying complete information
without skimping on details, and communicating the health risks associated with
sediment contamination in terms analogous to comparable risks that the public can
understand. EPA should link the contaminated sediment issue to visible effects,
such as beach closures and fish tissue consumption advisories. EPA must
articulate and remain accountable for achieving short-term goals and celebrate
interim successes while working toward long-term restoration. Finally, participants
emphasized that EPA must engage in active dialogue with the public and must be
responsive to public concerns.
3-23
-------
-------
CHAPTER.4
OPTIONS CONSIDERED FOR THE
DEVELOPMENT OF THE NATIONAL
SEDIMENT INVENTORY
As discussed in the previous chapter, a number of studies and pilot inventories have
attempted to assess the extent of chemical contamination in the Nation's freshwater,
estuarine, and marine sediments. Much has been learned to enable federal, state,
and local agencies to begin developing appropriate methods of remediation and initiating
enforcement actions to prevent future pollution problems at known sites. Although
these efforts have been important in calling attention to the problem of habitat
degradation and human health risks related to contaminated sediments on a local
and regional level, a comprehensive national study will determine more accurately
and more uniformly the extent and severity of the problem so that managers can
more effectively focus scarce resources and management approaches on areas impacted
by contaminated sediments. As discussed previously, EPA's proposed draft
Contaminated Sediment Management Strategy called for (1) the identification of
a list of chemicals of concern based on toxicity, persistence, and propensity to bind
to sediment particles; (2) the identification of sources of chemicals of concern in
sediments; and (3) the identification of sites with contaminated sediments based on
existing information. The latter two tasks requke the development of two inventories,
the Inventory of Sediment Contaminant Sources (the Source Inventory) and the National
Sediment Inventory.
The Source Inventory, now being developed, will list chemicals that have been detected
in sediments and the facilities responsible for these pollutant discharges based on
information contained in existing databases such as STORET, ODES, EMAP files,
pilot inventories, and other databases; and chemical concentrations in sediments reported
in the literature. The Toxic Chemical Release Inventory System (TRIS) and Permit
Compliance System (PCS) databases will be used to determine important point source
dischargers of sediment-associated chemicals. Pollutants of concern will be ranked
by criteria based on chemical adsorption/persistence and ecotoxicity. The Source
Inventory will also attempt to identify nonpoint source (agricultural and urban) inputs.
The proposed National Sediment Inventory (NSI) that is described in this document
will be a summary of locations known or suspected to have contaminated sediments
based on detailed monitoring data from national, regional, and state sampling programs.
The NSI will include concentrations of chemicals of concern measured in sediments
at each site, as well as other physical and chemical parameters when available. The
Inventory will also include available information on environmental effects such as
fish tissue contaminant concentrations, fish consumption advisory information, sediment
toxicity data, benthic community impairments, and other information. Possible
techniques to be used for determining whether sediments are contaminated include
available sediment quality guidelines for conventional, metal, and organic pollutants
4-1
-------
NSI Framework
such as those used by Regions 4 and 5 and other programs for their inventories of
contaminated sites.
This chapter presents a discussion of options considered for the development of the
NSI. The benefits and disadvantages of each option and the relative costs involved
are provided.
Options Considered
Several options have been considered for the development of the NSI to compare
feasibility, level of effort required, time, costs, and other factors. Early considerations
by EPA for the development of the NSI focused on obtaining all detailed sediment
and related monitoring data from various databases and entering them into a single
existing repository to be used for analysis and identification of contaminated sites.
This approach seemed to be the simplest in overall design and utility. Existing national
repositories available to house sediment data include STORET and ODES. Figures
on available data compiled by ERG (1991) indicated that there were approximately
26,600 sites for which sediment chemistry and related monitoring data were available.
Data from 85 percent of these sites had been entered into STORET; data from 4
percent had been entered into ODES; and data from 11 percent were in hard copy
only or had been entered into other computer formats. Another advantage of using
STORET as a national repository was the large community of skilled users (1,100
persons across the country) who regularly enter their data on a voluntary basis.
STORET also provides potential users with access to other information that could
be used to analyze sediment data, including water quality and fish tissue data, NPDES
permit data, watershed information, and population data. STORET, with the Reach
File, provides many opportunities to link and interface with these data sources for
streams, lakes, and coastlines. Also,, the system is immediately accessible at EPA
workstations on local area networks (LANs) hi each Regional Office, most state
offices, and many federal agencies. Moreover, STORET has an extensive capacity
to house additional data that is far beyond the capabilities of ODES and dBASE™.
For several reasons, however, both STORET and ODES were dropped from considera-
tion as the repository for the NSI. The costs associated with entering data into ODES
was the reason most often cited for not using it to house the Inventory. Although
many believe most of the Nation's sediment chemistry data, particularly freshwater
data, currently reside hi STORET, it was dropped from consideration mainly because
of the difficulties often cited in entering and accessing STORET data, the lack of
QA/QC data, and the lack of fields to hold the ancillary information necessary to
evalute sediment quality. As an alternative, this work will be coordinated with
"STORET Modernization" to facilitate the incorporation of data into a modernized
STORET system which is currently under design.
4-2
-------
Chapter Four—Development of the NSI
Examination of the Region 4 and Region 5 database formats, as well as other options,
identified a number of concerns for the design of the NSI including the capabilities
of different systems and software for performing data searches and compilations
and the possibilities for storing detailed monitoring data or summary data in relational,
searchable databases that would be nationally accessible. The operation of the Inventory
will require consideration of who would evaluate existing data and at what level
(Headquarters, Regional, or state level); what kind of assessments would be needed
to fully understand the problem at a particular site; and which program uses would
require what information. Planning these operational details would ensure inclusion
of the most essential features and aid in identifying an existing database system into
which the inventory could be integrated. At a minimum, the NSI must be capable
of maintaining biological, QA/QC, and other forms of data as well as chemical data;
it must be relatively easy for EPA Headquarters, the Regions, and states, in addition
to other federal agencies and researchers, to access, evaluate, and update data; the
Inventory must be relatively inexpensive to maintain and operate; and it must be
flexible enough to be modified as our scientific understanding of contaminated
sediments develops.
Following numerous discussions with EPA personnel and others, two primary options
for the design of the NSI were considered: (1) the development of a summary
inventory based on a statistical evaluation of individual databases and (2) the
development of an inventory containing detailed monitoring data from which
assessments would be conducted to identify potentially contaminated sites. The
variations on each of these options are discussed below. A summary of selected
attributes and problems associated with each option is presented in Table 4-1.
Option 1. Inventory of Summary Data Only
The option of developing an inventory of summary data was based on the approach
used by Region 5 (see Chapter 3). Under this approach data from individual databases
are summarized before the data are compiled into a single database. Sediment
chemistry data, as well as biological and other forms of data, would be included
in the summary inventory. The inventory produced by this approach would contain
only certain data parameters for each site; abstracts of QA/QC procedures and observed
impacts; and calculated mean, maximum, and minimum concentrations of chemicals
of concern.
Compiled bv EPA HeadQuarters. Under this option, EPA Headquarters would evaluate
and summarize the data from individual databases and create the summary inventory.
The actual detailed data would not be compiled into a single database. EPA would
access and summarize data from STORET, ODES, and NOAA's National Status
and Trends program, as well as from other EPA program offices. The summary
inventory would then be sent to the EPA Regions for review. The Regions would
4-3
-------
!
£
f
B
•o
3
important
iy
li
5? £9
S-3
1
1
&
1
^
1
•a
1
cd
"3
u
s
4-1
c
li
g s
2-1
1
aluation
t§3
t .
o
B
2 GO
o «
* 2
09 3
_
1
•a
O
e
id criteria
§
CO
!
i
+•>
ns difficul
valuatio
g
•g
2
M M eo 3
ills if I•§ -
32^5 e"o«S 2'3rs
5Ǥ-8Sg^SII
wj . O
I I
st a
II
S,!?
<§"s
4-4
-------
!
1
« s
•5 ^ 3 "5
<
-------
S
o
"g a e
5 - 3 -q * S
S &51 ^ §
.2 S <= 3 "° .S
j 11 -3 I b
^ 1111"~ **
«
ij
r
§ i ? i
w 3 a r
ex'
eo
.5 e
> a
« "O
2
u
•§
§
:S
!
111
Q SQ
-------
frl
e a
.2 "eS
60 ^
1
f
I
2 -o 5 -2
5 a cs a o.^e
•^ re a ii x 5
2 > cr 3
-------
NSI Framework
correct any errors noted in the summary inventory and would supplement the inventory
with data from individual database not already summarized by Headquarters.
Since EPA Headquarters would have primary responsibility for the development
of the summary inventory, procedures for identifying appropriate data, performing
summary statistics calculations, and compiling the data could be more easily controlled
than if each Region were compiling a separate summary inventory. Having a single
entity preparing the inventory would also cost less in terms of providing guidance
and training by EPA Headquarters since there would be less need for extensive review
of regional inventories to ensure their compatibility. Summary statistical analyses
could be conducted using the software compatible with the original data source, and
then the summary statistics and other data could be downloaded into the summary
database. A single summary inventory would allow quicker review of pertinent
nationwide information and take up less computer space than one containing individual
data points, allowing more flexibility in file size and hardware requirements.
One major disadvantage of the summary inventory is the difficulty in reevaluating
the original data as criteria for sediment contamination change. Each database would
have to the reanalyzed and the summary statistics run again. Then the summary
inventory would have to be updated. Another disadvantage of EPA Headquarters
preparing a summary inventory is that important regional data, available in a local
but not national inventory, could be overlooked initially. Although the Regions would
review and supplement the summary inventory the following year, summary statistics
would need to be recalculated each time sediment chemistry data were found for
a site or each time new developments in sediment quality criteria assessments
established new chemical contaminant thresholds. This would require a complete
reanalysis of the detailed monitoring data from each database used in the summary
inventory.
Compiled by Each Region. Because of the great diversity in sediment research and
data collection, it may be more appropriate to establish inventory programs by EPA
Region, based on the pilot inventories done by Regions 4 and 5, and to collect only
summary information into a national inventory. Each Region would be responsible
for identifying data from the above list of data sources and obtaining all pertinent
data from all of the categories of available data. The data would be carefully
scrutinized for minimum quality control requirements (for example, each Region
could track down original sources of data and validate the STORET records).
Headquarters would provide guidance on minimum data requirements; exactly what
types of information are needed for the inventories; the scope and extent of analysis
and discussion; the types of analyses to be performed (e.g., mean, maximum, minimum
for a particular chemical at a particular site derived from all data or only those meeting
certain quality control requirements); and the site summarization format. The summary
inventory thus completed would be sent to Headquarters, consolidated, evaluated,
4-8
-------
Chapter Four—Development of the NSI
and entered into the summary inventory on one system for each Region, and/or all
summaries would be placed on EPA's mainframe.
The NSI created in this manner would be sent back to the Regions so that they could
review the information and supplement it as necessary. Once gathered into the
inventory, the data could be used by the Regions to perform regional assessments
of sediment contamination.
Performing the assessment by Region would have the advantage that data could be
more rigorously examined, including going back to the original reports to determine
the methods used in biological and chemical analyses and to assess data quality.
Furthermore, sediment conditions differ throughout the country and are influenced
by local environmental changes that may be recognized only by local expertise.
The effort would allow an opportunity for coordination of Regional organizations
involved in sediment data collection.
A disadvantage of each Region setting up its own inventory is that Regional inventories
would have to be maintained on a continuing basis, with provisions made for uploading
them periodically into the national inventory. Also, additional costs and time would
be required in terms of development of guidance and training by Headquarters,
identification of appropriate data sets and their summarization by each Region, and
compilation of summary data into a database for each Region.
Option 2. Inventory of Detailed Monitoring Data Only
The option of developing an inventory of detailed monitoring data grew from the
above concerns that summary data could not be easily reevaluated whenever additions
or deletions were made in the database following Regional reviews or following
uploading of data from each Region, or if there were changes in sediment quality
criteria. Also, a detailed monitoring inventory could hold more information and
be more useful for other types of evaluations than the summary database. For this
approach, all categories of detailed monitoring data that exist in database formats
are provided by STORET, ODES, COE, NOAA, and other EPA and COE programs.
These databases would be compiled into an inventory containing sediment chemistry
data, as well as biological and other forms of data. The inventory produced by this
approach would contain all pertinent data parameters for each site (with standardized
parameter names and values), summary information on QA/QC procedures and observed
impacts, concentrations of chemicals of concern measured during different studies,
and pollutant source information. The detailed monitoring inventory would then
be evaluated according to predefined criteria, With results presented in report format.
Compiled bv EPA Headquarters. In this option, EPA Headquarters would be
responsible for obtaining all categories of detailed monitoring data available from
the above agencies and consolidating these data. The quality of the detailed monitoring
4-9
-------
NSI Framework
data would be assessed to the extent possible and coded by a screening assessment
of QA/QC information. All detailed sediment chemistry data would be evaluated
to select sites of concern based on established sediment quality criteria. (This approach
is similar to that used in the Gulf of Mexico Program's Contaminated Sediment Site
Inventory.) A preliminary summary report identifying chemicals and sites of concern
and other information would then be prepared. The summary report and detailed
monitoring data for each Region would be sent to the Regions for review. A final
summary report and inventory would then be prepared after reevaluating all of the
detailed monitoring data. The detailed monitoring data would be available to the
Regions.
A single entity compiling the detailed monitoring data would have more control over
the identification of appropriate data, standardization of parameter names and values,
programming that may be required to consolidate the data and convert units,'
manipulation of the data to ensure compatibility with statistical software and database
formats, and preparation of reports than if each Region were compiling a separate
detailed monitoring database. Having a single entity consolidating the data would
also cost less, since each database would be reviewed only once to ensure compatibility.
Although a detailed inventory containing individual data points would take up more
computer space than one containing only summary data, the preparation of the database
by EPA Headquarters should minimize redundant data points during the initial
consolidation process.
The disadvantage of EPA Headquarters consolidating the detailed monitoring data
is that important Regional data could be overlooked initially, perhaps necessitating
extensive changes and additional evaluations depending on the strength of data collected
in the above agencies' sediment quality databases.
Compiled by Each Region. For this option, EPA Headquarters would provide detailed
guidance on minimum data requirements, parameter names and values, data quality
information, data formatting, and other factors so that all categories of detailed
monitoring data could be gathered by each Region and consolidated into a single
database. Each Regional database would be consolidated by EPA Headquarters into
a single database and evaluated using established sediment quality criteria. The detailed
monitoring database would then be evaluated to identify areas of concern.
This system would allow each Region to identify the most useful and accurate data
sets. By allowing the work to be divided by Regions, databases could be scrutinized
carefully by workers within each area who would be able to assess the quality of
the data and their significance for priority contaminated sediment consideration based
on localized variables that could differ around the country. Regions would gain
further expertise in the sediment data and in using the database and inventory, thus
strengthening then: information base and their understanding of local contaminated
sediment problems. As with a Regionally prepared summary database, more effort
4-10
-------
Chapter Four—Development of the NSI
on the part of the Headquarters personnel would be required in the first year to provide
guidance to each Region and to develop the database. Stringent review of each database
would also be required before it could be consolidated into the national detailed
monitoring database to ensure compatibility and reduce problems during subsequent
evaluations.
Discussion of Option Selected
The final design of the NSI was determined on the basis of comparisons of the benefits
and disadvantages, and relative costs, of each option (Table 4-1). This evaluation
indicated that an analysis of detailed monitoring data (Option 2) compiled by EPA
Headquarters would be the best approach for the Inventory. The approach for
developing this inventory is described in the following chapter.
While summary information would take up less computer disk space and probably
allow quick retrievals of site information, evaluations of data summaries to identify
and prioritize chemicals and sites of concern would not be as accurate as using detailed
monitoring data. A major disadvantage of the summary approach is that once EPA
Headquarters has prepared the summary inventory, changes in the inclusion of sites
would require extensive reanalysis of the original data if sediment quality thresholds
for chemicals were changed. Statistical procedures for different inventories could
lack pertinent analytical software or have different calculations. If summary statistics
had to be performed by hand, as done by the Region 5 inventory, additional personnel
would be required to examine each data set, perform the calculations, and then recheck
the calculations for errors. '
Each change in the summary inventory required by limited or ongoing Regional
review and supplementation would provide opportunities for further mistakes to be
made and possibly entered into the database, necessitating extensive quality control.
Also, as noted by Manheim (1991), the disadvantage of summarizing information
is that the goals of synthesis will invariably change over time. If only the synthesis
information is stored, the basic data may ultimately be lost, making reanalysis very
costly. Furthermore, different managers may be using the inventory for different
purposes at the same time to examine different sediment problems. Summary
information might not provide all the parameters required for such different analyses.
Extensive programming, parameter identification, summary statistics calculations,
development of summarized abstracts for certain types of information (QA/QC,
environmental impacts, site descriptions), and data entry would require more tune,
personnel, and funds than simply working with detailed monitoring data. Thus, these
concerns, especially recent developments and changing procedures in establishing
criteria for evaluating toxicities of chemically contaminated sediments, suggest that
the summary database approach would not be appropriate on a national scale.
4-11
-------
NSI Framework
By consolidating all detailed monitoring data into a single inventory, evaluations
of the data could be performed at any time as sediment contamination thresholds
are updated. Assessment and coding of data quality would allow certain subsets
to be used for different types of evaluations. Such coding could also indicate whether
sufficient high-quality data had been collected at a particular site or whether further
analyses were necessary. Additional data, identified during Regional reviews, could
be more easily tailored for loading to the national database. Furthermore,'future
collections of detailed monitoring data could be easily added and evaluated. While
there is great interest, in general, in the collection and interpretation of new data
rather than historical data, existing data are more significant for sediments than for
measurements of the water column and biota because changes in sediments take
place more slowly (Manheim, 1991). Thus, having all current detailed monitoring
data together in one inventory with the capability of adding data as they are collected
would allow timely comparisons to assess the impact of various management
approaches, such as pollution prevention, remediation, and/or dredged material
management programs, on sediment quality in the United States.
The development of the inventory by a single entity, EPA Headquarters, would permit
control of data compilation and more uniform quality assessments. This would also
eliminate the need to check each Regional inventory for compatibility and extra
programming that may be required if the guidelines that were provided are not strictly
followed. Problems encountered during inventory consolidation, such as variable
names for the same parameters or concentration units that must be converted to the
units used by sediment quality criteria for evaluations, could be corrected for the
entire database at the outset. Quality assurance procedures established prior to database
consolidation and manipulation could also be more easily monitored for a single
entity than for multiple Regions. EPA is anticipating that available resources will
be less in FY 1993 than in FY 1994. Headquarters will be in a better position to
develop the detailed monitoring database at the start, with more funds available for
distribution to each of the Regions the following year for data review and
supplementation. The total funding allocated to the development of the NSI and
evaluation of the data in the NSI to identify potential and probable contaminated
sediment sites is estimated to be $750,000 to $1 million (FY 1993 and FY 1994).
The development of a single inventory by EPA Headquarters with EPA Regional
review of this database is the most cost-effective approach to developing the NSI.
Since the assessment of sediment data is still an evolving science and the criteria
used to evaluate the extent of contamination may be modified, it is believed that
a reevaluation/recreation of a summary inventory would be required and may result
in an overall increased cost of 30-50 percent. Similarly, if Regions were to develop
independent detailed inventories, each Region would be charged with investigating
the availability of additional data and compiling readily available data. By centralizing
readily available data compilation and eliminating the cost to EPA Headquarters
of aggregating Regional databases, the chosen method should result in a decreased
4-12
-------
Chapter Four—Development of the NSI
cost of 40-60 percent. In addition to a cost increase, the overall 2-year schedule
may be jeopardized if either of the other options is selected.
4-13
-------
-------
CHAPTER 5
APPROACH FOR DEVELOPING
THE NATIONAL SEDIMENT
INVENTORY
The development of a detailed monitoring database that provides end-user
computing or "ready-to-go" applications software is beyond the regulatory time
schedule and resource constraints of the current effort. In addition, no resources
have been identified for continued training, operation and maintenance, or support.
As a result, the first phase of developing the National Sediment Inventory is to
aggregate data from diverse sources, evaluate selected data, and disseminate the
preliminary evaluation and data to EPA Regions for review. The form of the
transmission of data to EPA Regions has not been finalized; however, it is
expected that XBASE-cOmpatible files would be the likely format along with a
hard-copy report summarizing the preliminary evaluation. The types of sample
data will include sediment chemistry, bioassay, bioaccumulation, pollutant source,
and fish advisory data. Sediment chemistry, biological effects data, and QA/QC
will be initially evaluated by EPA Headquarters in order to identify potential areas
and chemicals of concern.
By providing the data in a generic form along with a report summarizing the
analysis, Regions will be able to adopt portions of the data into their existing
systems or will have the necessary skill levels to use EPA-standard software such
as dBASE™. By having both the hard-copy report and data, more in-depth
reviews are anticipated during the second phase of the National Sediment Inventory
development. The Regional Offices will be able to review and evaluate all the
detailed data. The Regions may then provide EPA Headquarters with additional
data to be included with a revised analysis. EPA Headquarters will conduct a
second evaluation of the revised NSI and create a final report.
In addition, this project will be coordinated with concurrent projects such as
"STORET Modernization" and other Office of Information Resource Management
(OIRM) activities. For example, a necessary portion of this task is to develop
"cross-walks" between the naming conventions for sediment data used by existing
data systems. As a result, the lessons learned from this effort will be helpful for
the ongoing efforts under STORET Modernization. This cooperative effort will
also facilitate the incorporation of the data into a modernized STORET system at
a later time.
The following sections describe the process to be used in developing the NSI. It
should be noted that several critical issues—for example, data structure, data
prioritization, quality assurance/quality control (QA/QC) evaluation procedures, and
methods for defining thresholds above which a site will be considered
contaminated—have yet to be resolved. These issues and others are currently
-------
NSI Framework
being addressed. In addition, decisions on other issues may change over the life
of the project. Note that in the discussion that follows, reference to the National
Sediment Inventory refers to the data to be included for analysis, not an
information system to be developed.
Development of the National Sediment Inventory
Categories of Data to Be Evaluated
Four major categories of detailed monitoring data will be considered for the NSI:
data record, site characteristics, QA/QC, and sampling parameters (Table 5-1). The
organization of the information presented in Table 5-1 (and throughout the
remainder of this chapter) is for convenience of presentation and subsequent
discussion and does not necessarily reflect the computerized data structure and
format that will be implemented for the NSI. Several minimum data parameters
have been identified under each of these major categories. Some of these
parameters must be available before the data will be included in the NSI; others
would be desirable, but their absence would not preclude data from being included.
The general tendency of this effort is to include rather than exclude data for this
screening-level analysis. Invariably, the minimum data requirements for inclusion
in this inventory may preclude the use of certain portions of the inventory for other
program objectives described in Chapter 2. With proper identification, it is
believed that other programs will be able to selectively choose data for their
requirements. The major data categories and minimum data elements are described
further below.
Data Record. The data record must be in computerized format and must include
a data dictionary specifying field names, widths, delimiters, or file structure. Other
data that must be included in the data record are sampling location (including
waterbody name), sampling date, and latitude/longitude. If available, the reach
number (based on EPA's Reach File) should also be included.
Site Characteristics. There are several pieces of information related to site
characteristics that, if available, would be considered during the development of the
NSI, although none are considered critical pieces of information without which data
would be excluded. These include land use (e.g., agricultural, rural, urban,
commercial); management status of the site (i.e., whether remedial activities are
currently being performed and by whom); whether the site is a hazardous waste
facility or Superfund site or whether an accidental spill has occurred at the site; the
frequency of dredging/dredging history at the site; the identity and location of point
source discharges (current and historical) in the vicinity of the site (including the
use of the National Source Inventory); and the presence of endangered species.
5-2
-------
Chapter Five
Table 5-1. National Sediment Inventory Data
Category Summary
Minimum Data Element
DATA. RECORD
In Computerized Format
Location
Sampling Date
Lat/Long
Reach Number
SITE CHARACTERISTICS
Land Use
Management Status of Site
Location of Haz Waste/Superfund
Site
Spill Information
Frequency of Dredging
Point Source Information
Presence of Endangered Species
QA/QC
Source of Information
Lab Methods
Field Methods
SAMPLING PARAMETERS
Sediment Chemistry
Total Organic Carbon
Grain Size
Acid Volatile Sulfides
Biological Data
Fish Advisories
Benthic Abundance
Fish Pathology ,
Necessary
•
•
•
•
•
•
If Available
•
•
•
•
•
•
•
•
«
•
«
•
•
•
•
•
•
Comments
With data dictionary specifying field names, widths,
delimiters, or file structure
Including waterbody name
Conforming to EPA's standards
Urban, industrial, rural, etc.
Remedial action, etc.
i.e., dredging history
Current/historical
Sponsor or client name and address, name of
analytical lab or principal investigator and address
Quality of data to be coded, method detection limits
used in analyses to be included
Quality of data to be coded
Biotoxicity, bioaccumulation
Benthic infauna, community, other indices
5-3
-------
NSI Framework
QA/QC. The only QA/QC information that must accompany the data before they
can be considered for inclusion in the Inventory is information on the source of the
data. The name, organization, address, and telephone number of the individual
who collected the data or who can address questions concerning the data collection
and analysis procedures must be provided. If available, information on the field
and laboratory samples and methods used should also be included with the data.
Information on several types of QA/QC samples and procedures that can influence
the quality of the data and can be used to check the quality of data will be
analyzed for data sets to be included in the Inventory, if available. Although none
of this information is necessary before a data set can be included, evaluation of
such information will provide an indication of the quality of the data used to target
a specific site. If the QA/QC evaluation procedures are unknown or known to be
inadequate, then the data will be coded accordingly. The site should be considered
only potentially contaminated, and additional information gathering and
assessments would be recommended.
A much more limited version of the QA/QC evaluation procedure for historical
databases developed for EPA's Great Lakes National Program Office (GLNPO)
(Schumacher and Conkling, 1990) is envisioned for use as part of the development
of the NSI. In the GLNPO procedure, various QA/QC components were grouped
into five general categories that encompass the major areas of concern in a good
quality assurance program. Each component then received a ranking as to its
perceived importance in the assurance of good-quality data, and each ranking was
given a score. The scores were then summed to provide an overall assessment of
the likely quality of a. database. Table 5-2 presents the QA/QC categories and
components that were evaluated as part of the GLNPO procedure. For the
purposes of the NSI, it is envisioned that the individual databases that make up the
Inventory will be given one of three broad classifications based on an evaluation
of the QA/QC components used:
• Adequate QA/QC used,
• Inadequate QA/QC used, and
• Unknown QA/QC.
The precise method of scoring the quality of a given data set for the purpose of the
NSI is currently under examination. If the user of the data wishes to acquire more
detailed information concerning the quality of data in a data set, the user will have
to contact the database contact directly.
Sampling Parameters. All of the sediment chemistry data will be evaluated to
identify the potential areas and chemicals of concern. Other types of detailed
sampling data to be included in the NSI, if available, include biological data (i.e.,
biotoxicity and bioaccumulation), pollutant source benthic abundance, and/or fish
5-4
-------
Chapter Five
Table 5-2. QA/QC Components Us<5d in the GLNPO Procedure
Category
Component
Accuracy
Precision
Spike Recovery
Blanks
Miscellaneous
Certified Reference Material
Mid-Range Audit Sample
Low-Level Audit Sample
QC Check Sample
Detection Limit QC Check Sample
Field Duplicate
Analytical Duplicate
Preparation Laboratory Duplicate
Standard Duplicate
Matrix Spike Duplicate
Matrix Spike
Surrogate Spike (organics)
Calibration
Reagent
Field
Cross-Contamination
Field Reagent (preservation)
Instrument Calibration
Instrument Detection Limit
Ion Chromatograph Resolution
Chemistry Relationships (expected
correlations among different parameters)
Improper Sampling Technique Method Error
or Problem
Improper Holding Times
Improper Sample Storage Techniques
Lack of Methods Comparability
Among Analytical Laboratories
5-5
-------
NSI Framework
pathology. In addition, information on fish advisories in the vicinity of the site (if
any) will be collected. Tables 5-3 through 5-8 list the data parameters that will be
included in the NSI when available, for sediment chemistry, bioassay,
bioaccumulation, benthic abundance, fish pathology, and fish advisory data,'
respectively. The organization of the information presented in Tables 5-3 through
5-8 is for convenience of presentation and does not necessarily represent the data
structure and format that will be implemented for the NSI.
Inventory Organization
The development of a detailed monitoring database that provides end-user
computing or "ready-to-go" applications software is beyond the scope and
resources of the current effort. In addition, no resources have been identified for
continued training, operation and maintenance, or support. As a result, the
emphasis of the initial phase is to aggregate data from diverse sources, evaluate
selected data, and disseminate data to EPA Regions for review and update. The
NSI, when distributed to EPA Regions, will consist of XBASE- (i.e., dBASE™0
compatible files. By providing the data in a generic form along with a report
summarizing the analysis. Regions will be able to adopt portions of the data into
their existing systems or use EPA-standard software such as dBASE™ to browse
the data. To promote the dissemination of the Inventory to groups outside the
Agency such as universities, the use of CD-ROM technology and INTERNET will
be further investigated and implemented, if feasible.
The structure and organization of the disseminated files will be developed to ensure
that pertinent data collected for a specific sample can be retrieved even though the
data may reside in separate data files. This approach represents a balance between
the competing requirements associated with the remainder of this effort (described
in other chapters) and allowing for enhancements so that other program areas can
use the data from this effort for screening-level analyses as well. During the
review and initial release of the NSI, it is expected that EPA Regions, program
offices, or other groups will adopt portions of the Inventory into their existing
systems or will have the necessary skill levels to use an appropriate PC-based
database management system. Minimum skill levels would be required to browse
data (e.g., users would need to develop indexes and relate files). More advanced
skills would allow more sophisticated analyses.
The specific data to be included in the Inventory will be developed in the future
once the nature of all the data has been determined. The database structure;
format, and data dictionaries of the following efforts will be considered as starting
points for this effort:
• Seattle COE Sediment Inventory
• Region 4 Sediment Inventory
5-6
-------
Chapter Five
Table 5-3. Sediment Chemistry Sample Parameters
Importance of Parameter
Category
Parameters
Necessary If Available
Sample Information
Sample number
Sample depth (upper and lower)
Sampling equipment
X
X
X
Analysis Information
Environment from which sample was taken
(i.e., suspended, bottom, dissolved, etc.)
Replicate number
Name of chemical
CAS number
Concentration measurement for chemical
Units
Sign (+ or -)
Extraction method
Instrument used
Detection limit (if observation is below
detection limit or not detected)
TOC
Grain size
AVS
Other geologic information
X
X
X
X
X
X
X
X
X
X
X
Table 5-4. Bioassay Sample Parameters
Importance of Parameter
Category
Sample Information
Bioassay Conditions
Analysis Information
Parameters
Sample number
Upper/lower depth of core samples
used as exposure medium
Environment from which sample was
taken (e.g., bottom sediment, interstitial
water, elutriate, etc.)
Collection method
Bioassay type
Number of organisms originally present in
each sample replicate
Exposure duration
Taxonomic code
Units used to report concentration
Concentration of dilution used in bioassay
Variable measured (e.g., LC^, count of
live offspring, etc.)
Measure or count
Necessary
X
X
X
X
X
X
X
X
X
If Available
X
X
X
5-7
-------
NSJ Framework
Table 5-5. Bioaccumulatiion Sample Parameters
Importance of Parameter
Category
Parameters
Necessary
If Available
Sample Information
Sample number
Specimen number or composite number
Gear type
Taxonomic code
Number of individuals
Tissue sampled
X
X
X
X
X
X
Analysis Information
Replicate number
Name of chemical measured
Units used to measure chemical
Concentration
Extraction method
Instrument used
Detection limit
Wet or dry weight
X
X
X
X
X
X
X
X
Table 5-6. Benthic Abundance Sample Parameters
Category
Station Information
Bottom
Characterization
Species Abundance
and Biomass Data
— —g— —
Parameters
Sample number
Sieve mesh size
Core grab surface area
Number of samples
Sampling equipment
Sample depth
Bottom type
Taxonomic code
Number of individuals
Wet weight of individuals
=======
Importance
Necessary
X
X
X
X,
========
of Parameter
If Available
X
x ;
X
X
X
x
5-8
-------
Chapter Five
Table 5-7. Fish Pathology Sample Parameters
Category
Station Information
Individual Record
Lesion Record
Parameters
Collection method
Specimen number
Taxonomic code
Sex
Length
Length units
Method used to measure length
Weight
Weight units
Method used to measure weight
Disease
Health status
Pigmentation
Lesion
Severity
Host response
Organ/Suborgan
Importance of Parameter
Necessary If Available
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Table 5-8. Fish Advisory Parameters
Parameter
Importance of Parameter
Necessary
If Available
Species Affected
Sizes (length or weight)
Contaminants
Reach Number (or other location
identification)
Thresholds for Issuing Advisory
X
X
X
5-9
-------
NSI Framework
• Gulf of Mexico Program Sediment Inventory
• NS&T (NOAA)
• EMAP
• ODES
• Puget Sound Ambient Monitoring Program
• Region 5 Sediment Inventory
• USGS
• Great Lakes Sediment Inventory
The key issues associated with this evaluation will be organization, national
consistency, and breadth of data elements.
Sources of Data to Be Included in the National Sediment Inventory
Depending on available resources, data from the following existing computerized
databases will be included in the National Sediment Inventory:
• Select data sets from STORET, e.g,,
' - COE
- USGS . *
- EPA
- States
- BIOACC
• NS&T (NOAA)
• ODES i »
• Region 4 Sediment Inventory
• Region 5 Sediment Inventory
• Gulf of Mexico Program Sediment Inventory
• COE Seattle District Sediment Inventory
• Great Lakes Sediment Inventory
• Environmental Monitoring and Assessment Program (EMAP)
• National Estuary Program (NEP)
• FWS *
5-10
-------
Chapter Five
• MacDonald Database
• USGS
• National Source Inventory
Data Collection Procedures
Only data collected since 1980 that are currently maintained in computerized
format will be included in the NSL Hard copy data will not be included. Because
of limited time and resources, data sets will be prioritized for integration based on
geographic coverage and quality, as well as the types of information they contain.
The major source, in terms of geographic coverage, of sediment chemistry data will
be STORET. EPA will conduct an assessment of existing sediment and related
data in STORET. Data from select data sources in STORET will be transferred
to the Inventory. These would include data from USGS, COE, EPA, state, and
other possible databases housed in STORET. Biological data maintained hi
STORET (i.e., National Study of Chemical Residues in Fish) will also be included
in the Inventory. EPA will also compile the data from the Region 4, Region 5,
COE Seattle District, Great Lakes, and Gulf of Mexico sediment inventories and
biological data currently in ODES and enter these into the Inventory. NOAA will
provide EPA with data from its NS&T program for incorporation into the
Inventory.
Other databases from such programs as EPA^s EMAP and NEP, USGS, and
MacDonald database will also be investigated to determine the feasibility of
including them in the Inventory. This will again be'determined to a large degree
by the available resources as well as the difficulty in obtaining this information,
the difficulty in analyzing the data, and their ^compatibility with the structure of the
Inventory.
Initial Evaluation
Once the National Sediment Inventory is in place, an evaluation of the data will
be conducted to identify those sample observations that exceed the threshold limits
for each contaminant. The results of this evaluation will be a computer-generated
detailed listing of all observations that exceeded the sediment quality threshold
limits. For inland areas, EPA's River Reach System will be used to organize the
report by watersheds. Organization of results for estuaries and open waterways has
not been finalized. It should be noted that any sediment chemistry measurement
that exceeds the threshold limit for a contaminant will be included as long as the
measured value was also greater than the detection limit for that observation,
5-77
-------
NSI Framework
regardless of whether the detection limit was lower or greater than the threshold
limit.
Additional data related to each potential area of concern in which a sediment
sample that exceeded a given threshold was taken will also be included in;the
initial evaluation. The National Source Inventory will be used to identify point and
nonpoint source discharges contributing to sediment contamination. Such
information will include, when available, bioassay, bioaccumulation, benthic
abundance, fish pathology, and fish advisory data. These data will be included to
assist the Regional reviewer in assessing the ancillary data. Biological and other
forms of data can then be used to further justify the inclusion of a site on the list
of high-priority sites. An evaluation of QA/QC data using an approach similar to
that used by GLNPO will also be conducted. The results of this evaluation will
be a ranking of the potential quality of the data hi each data set as good quality,
poor quality, or unknown quality. Overall summary, statistics will also be
developed for the initial evaluation, defining, for example, the total number of
samples per area, the total number of observations per sample, the total number of
observations exceeding threshold limits for each contaminant, the percentage of all
observations exceeding threshold limits, and the total number of reaches affected
nationally and by EPA Region and state.
It is currently envisioned that the initial evaluation will include selected portions
of the following information, as available:
• Reach identification (alternative approaches for estuaries and open
waters are still .under consideration)
- reach name
- reach number
- state
- county
- waterbody name
- waterbody type (river, lake, coastal)
- upstream lat/long
- downstream lat/long
- reach length
- site states (remediation or regulatory action, none)
- dredged? (last two dates dredged)
- land use/land cover (if known)
- industries within reach (names, SIC code, NPDES number)
• Sediment chemistry sampling information
- contaminant, CAS number, and threshold limit
5-12
-------
Chapter Five
- source of data (agency identifier)
- sample location (lat/long)
- Reach File Index
- date of sample
- sample number
- measurement value that exceeded threshold
- units
- magnitude of threshold exceedance (e.g., Ix, 5x, lOx, >10x)
- QA/QC qualifying code (e.g., acceptable, poor, unknown)
- summary information (i.e., total number of threshold exceedances for
each contaminant in the reach)
For each reach in which one or more sediment quality measurements exceed a
threshold limit, the Inventory would include a listing of other monitoring data or
fish advisory information for that reach, if available. When possible, this
information would include selected portions of the following:
• Sediment toxicity/bioassay testing
- source of data (agency identifier)
- location where sediment sample was taken (lat/long)
- date of sampling
- sample number
- species name and code
- test duration *
- type of assay
- minimum value
- maximum value ' *
- median value i ,
- units
- number of samples
- results
- QA/QC qualifying code *
• Bioaccumulation testing
- source of data (agency identifier)
- location where sample was taken (lat/long)
- date of sampling
- sample number
- contaminant measured
- species name and code
- type of sample (i.e., tissue analyzed: whole body, fillet, other organ)
- minimum value
5-13
-------
NSI Framework
- maximum value
- median value
- units
- QA/QC qualifying code
Benthic abundance information
- source of data (identification number)
- location where sample was taken (lat/long)
- sample number
- date of sample
- indices measured
- results
- QA/QC qualifying code
Fish advisory information
- extent of fish advisory (lat/long)
- fish species
- sizes «
- contaminants
- threshold for issuing/lifting advisory
- date advisory started
Fish pathology information
- source of information (agency identifier)
- location of sample (lat/long)
- date of sample
- sample number
- species name and code
- impairment observed
- QA/QC qualifying code
Review of the National Sediment Inventory
During 1993 EPA will compile the preliminary National Sediment Inventory. Each
Region will be sent a Regional Sediment Inventory (e.g., the data used in the
evaluation) and a preliminary report describing the NSI and the assumptions and
procedures used hi developing the preliminary report and a preliminary list of areas
and chemicals of concern. The Regions will also be provided with the NSI
documentation and procedures for conducting their own assessments of the data if
desired. The Regions will be encouraged to correct inaccurate analyses and
5-14
-------
Chapter Five
nominate additional data not part of the original data compilation for inclusion
during a revised analysis. It will also be the responsibility of EPA Regions to
correct source databases or notify data owners of inaccurate data.
The EPA Regions will identify additional computerized databases that can
supplement the information presented in the NSI; for example, additional biological
data or sediment contamination data for areas of the country for which EPA
Headquarters did not have data. If the Regions are able to identify additional
relevant databases, they will acquire copies of these databases on disk and provide
them to Headquarters for possible entry into the NSI.
The Regions will also review the QA/QC information for the data. For those data
in the Inventory for which the quality of the data is unknown, the Regions should
contact the source of the data to determine what QA/QC samples and procedures
were used during sample collection and analysis. Based on their findings, the
Regions can include an analysis of QA/QC information in the analysis. Gathering
the QA/QC data for samples that are included hi the Inventory and for which the
data quality is unknown will take a considerable amount of effort on the part of the
Regions'because, depending on the Region, much of the sediment chemistry data
may come from STORET, which does not contain detailed data quality
information.
Following Regional review, EPA will select and include selected additional data
from the 10 Regions into the NSI. These data will be evaluated a second time in
a more complex manner, and the results will represent the final report. Other
federal agencies, EPA program offices, and regions, as well as states, will be
involved in formulating the approach for the second evaluation.
Each of the identified sites will be categorized as either those for which sufficient
data exist to characterize them as causing high risks or severe effects or those
which may be contaminated but are in need of additional information and further
assessment. This categorization will be based on consideration of a number of
factors, including the following:
• Number of chemicals exceeding threshold limits;
• Number of observations exceeding threshold limits;
• Severity of contamination (i.e., contaminant concentration);
, • . Biological evidence of contamination and impacts to support conclusions
based on sediment chemistry data;
• Fish advisory information; and
5-/J
-------
NSJ Framework
• Quality of data used to identify the site as contaminated.
A final report describing the process used to develop the NSI and the evaluation
process will accompany the Inventory. As with the preliminary report, the final
report will also provide guidance on accessing and evaluating the data in the
Inventory. The report will also explain the assumptions made in categorizing
chemicals of concern and in categorizing sites as being potential or probable
contaminated sites.
Schedule
Figure 5-1 presents the proposed schedule for the completion of milestones related
to the NSI. The design of the Inventory is scheduled to begin in January of 1993,
and completion of the preliminary report and Regional Sediment Inventories is
scheduled for December of 1993. Regional review and comment and update of the
Inventory are planned to occur from January through mid-summer of 1994. The
final National Sediment Inventory, which will incorporate the input from the
Regional Offices, is scheduled for completion in December of 1994.
5-16
-------
Cfl
Eb
1
Design National
Sediment Inventory
nto NS
Enter
eva
Cond
detailed
R
In
§
.s
3
nd
da
evaluation of
5-77
-------
-------
CHAPTER 6
SEDIMENT ASSESSMENT TECHNIQUES
Sediment assessment is a procedure used to interpret the significance of
contaminant levels measured in sediments, accounting for differences in
contaminant bioavailability caused by site-specific properties of sediments (Adams
et al., 1992). Several sediment quality assessment techniques are briefly reviewed
below for their data requirements and surrounding issues. A more complete review
of this topic may be found in USEPA (1992e), Adams et al. (1992), and Long and
Morgan (1990). A number of sediment assessment approaches and several more
"weight-of-evidence" approaches or combinations of approaches have been adopted
by programs, such as the "Green Book" method for dredged material disposal hi
ocean and near coastal waters adopted by EPA's Oceans and Coastal Protection
Division (OCPD) (COE and USEPA, 1991a), the Long and Morgan (1990)
approach adopted by NOAA's National Status and Trends (NS&T) Program, and
the tiered approach used in the Great Lakes region by the International Joint
Commission (UC).
There appears to be no single method for identifying contaminated sediments that
will apply in all cases because of the variability in sediment properties controlling
the bioavailability of contaminants in sediments, the variability in the sensitivity
and behavior of organisms, and the confounding effects of other chemicals. The
need for a timely assessment of existing sediment quality in the United States will
help drive the selection of an approach that will work, given available data and
resources, despite the current level of uncertainty regarding processes controlling
bioavailability and toxicity of compounds in sediments.
As described previously, the first NSI evaluation to identify potential chemicals
and areas of concern will be based on sediment chemistry threshold exceedances.
The second, more complete evaluation will include biological as well as data
quality information. Input from other federal agencies, EPA program offices and
Regions, and states will be included in this process to identify chemicals of
concern and potential and probable areas of concern. The following sediment
quality assessment techniques will be considered in developing these lists. The
process will likely involve a point system similar to Region 5's prioritization of
sites project.
The sediment quality assessment techniques being reviewed for consideration for
use in the development of the National Sediment Inventory include the following:
• Equilibrium Partitioning
• Sediment Quality Triad
• Bulk Sediment Toxicity
6-1
-------
NSJ Framework
• Interstitial Water Toxicity Evaluation
• Apparent Effects Threshold
• Spiked Sediment Toxicity
• Tissue Residue
• Screening-Level Concentration
• Long and Morgan (1990)
MacDonald (1992)
The approaches for deriving sediment quality criteria as reviewed by USEPA
(1992e), Adams et al. (1992), and Long and Morgan (1990) are summarized in
Table 6-1.
Equilibrium Partitioning
In the Equilibrium Partitioning approach, interstitial water concentrations of
individual chemicals are predicted from equilibrium partitioning theory and
compared with water quality criteria derived from chronic water-only exposure to
test organisms. This method is protective of aquatic organisms whose primary
route of exposure to contaminants is through contact with sediment interstitial
water (primarily benthic organisms that burrow in sediment). A key assumption
is that the appropriate toxicological endpoints and sensitivities of benthic organisms
can be considered to be the same as those of the test species for which the final
chronic values were derived. The method can be applied to nonpolar, nonionic
chemicals in sediments having organic carbon contents in the range of 0.2 percent
to about 30 to 40 percent (D.M. DiToro at EPA Science Advisory Board meeting,
Crystal City, Virginia, 10 June 1992).
Sediment Quality Triad
In the Sediment Quality Triad approach, the correspondence between sediment
chemistry, toxicity, and biological effects is used to indicate the spatial distribution
of sediment contamination and define "hot spots" within a site by distinguishing
high levels of biological effects relative to a suitable reference station. This
approach has an advantage over the bulk sediment toxicity approach in that it
considers multiple categories rather than a single category of information and
therefore may be classified as a weight-of-evidence approach. The method can be
6-2
-------
Chapter Six—Inventory Structure
Table 6-1. Data Requirements and Issues Related to Sediment Assessment Techniques
Approach for Sediment
Assessment
Data Requirements
Issues
Equilibrium Partitioning
« bulk chemistry
• organic carbon content
(May be applied to summary data)
Applies to only one class of
compounds, the nonpolar, nonionic
chemicals
Does not account for exposure through
ingestion, which is important for
compounds with high
Does not account for the joint action of
chemicals
Can be applied only to those chemicals
for which a WQC is available or for
which there is a sufficient database on
effects
Does not apply to sediments with
organic carbon content below about
0.2%
Sediment Quality Triad
• bulk chemistry
* toxicity (several species and
Midpoints desirable)
• benthic community (or possibly
bottom fish histopathology)
• organic carbon content
(Requires extensive detailed
monitoring data)
Requires a complete set of consistent
individual monitoring data at each
station
Subjective judgment is required to
develop SQC
Requires the use of a reference site
Does not allow calculation of statistical
confidence intervals for SQC
Does not address causality or the
mechanisms contributing to
bioaccumulation
Bulk Sediment Toxicity
• bulk sediment toxicity (often more
than one test species or bioassay
type)
(Designed for use with detailed
monitoring data.)
Cannot be used to develop SQC
Does not address causality
Requires the use of a reference site
Does not address the mechanisms
contributing to bioaccumulation
6-3
-------
NSI Framework
Table 6-1. (Continued)
Approach for Sediment
Assessment
Interstitial Water Toxicity
Identification Evaluation (TIE)
interstitial water toxicity
• toxicity of chemical fractions of
interstitial water
• data for validation
(Does not lend itself to use of existing
data)
Pore water toxicity tests and TIE
procedures are insufficiently validated
Interstitial water may not be the
primary route of exposure for
organisms ingesting sediments or
compounds with high
Apparent Effects Threshold
bulk chemistry
• field-collected biological effects
data (results of more than one
bioassay type preferable)
* organic carbon (not absolutely
necessary but preferable)
(Requires individual monitoring data
if sediment toxicity is used; may
perhaps be used with summary data if
animals that traverse die entire site,
such as fish, are used)
Requires data showing a wide range in
chemical concentrations and biological ;
effects
Requires use of a reference site having
negligible measured biological effects
Cannot determine which chemicals are
causing the biological effects
Cannot distinguish the harm caused by
individual chemicals in mixtures i
Does not address bioaccumulation
Spiked Sediment Toxicity
• toxicity tests on a range of test
sediment concentrations where the
test sediment was created by
taking sediment from a reference
site and adding chemical to form
a range of sediment
concentrations
(Cannot be used with existing
monitoring data)
Requires establishment of a reference
sediment
Results depend on sediment aging, i.e.,
the elapsed time between spiking and
testing
Results may depend on the amount of
carrier compound used to dissolve the
chemical in the spiking solution
Does not test field conditions and in
situ organisms; may not mimic natural
conditions
6-4
-------
Chapter Six—Sediment Assessment Techniques
Table 6-1. (Continued)
Approach for Sediment
Assessment
Data Requirements
Issues
Tissue Residue Approach
• either WQC or no-observed-
effects level from bioassay and
BCF or criteria for fish tissue
residues
• sediment organic carbon content
• chemistry in water column,
sediment, and biota
• food chain structure
• ratio of dry to wet weight for
animals in food chain
• respiration rate as a function of
water temperature and organism
mass
• lipid content of the animals and
K™ to calculate a BCF
• growth rate of animals
(Can be used with summary data,
although additional parameters in the
summary database would be required)
Approach is most suitable for
contaminants with high K.,^5 and slow
metabolism
The relationship between contaminant
concentrations in sediments and tissue
concentrations is poorly understood
Thermodynamic and toxicokinetic
bioaccumulation models have been
tested for only a few compounds
Causal relationships between tissue
residues and biological effects are not
well understood
Requires FDA action levels or state
standards, which can vary considerably
by state
Requires literature search or laboratory
analysis of respiration rates
Screening-Level Concentration
• species composition
• sediment chemistry
• organic carbon content
(Requires extensive detailed
monitoring data)
Results can be confounded by changes
in the habitat, sediment properties, and
surface water quality
Has received limited application to
compounds other than nonpolar
organics
Requires large sets of detailed
monitoring data
Cannot determine which chemicals are
causing the effects
Long and Morgan (1990) or
NS&T
• matched chemistry and biological
effects data for many species and
sites
• spiked sediment bioassay data
from literature
(Requires detailed monitoring data)
Cannot determine which chemicals or
other factors are causing the effects
Requires large sets of detailed
monitoring data
Results may be confounded by the
effects of mixtures
Does not consider differences in
bioavailability for different sediments
6-5
-------
NSI Framework
Approach for Sediment
Assessment
MacDonald (1992)
Table 6-1. (Continued)
Data Requirements
matched chemistry and biological
effects data for many species and
sites
(Requires detailed monitoring data)
Issues
Cannot determine which chemicals or
other factors are causing the effects
Requires large sets of detailed
monitoring data
Results may be confounded by the
effects of mixtures
Does not consider differences in
bioavailability for different sediments
applied to all types of sediments and chemicals, provides a direct assessment of
sediment quality, can use existing detailed monitoring data, and can be used to
empirically derive Sediment Quality Criteria (SQC) for many chemicals. Three
categories of individual monitoring data are normalized to values collected at a
monitoring station designated as the reference site by dividing the value of the
specific variable measured by its value measured at the reference site In this
manner a ratio-to-reference (RTR) value is calculated for chemical concentration
various toxicity test results, and parameters measuring benthic community structure
or function. (A complete set of data for each monitoring station is necessary.) The
RTR values for each station are summarized by combining them, for each category
of data, into an average. Average values are used to divide sediments into three
categories: contaminant concentrations at which there are no biological effects,
contaminant concentrations at or above which biological effects are always high,'
and a range of chemical concentrations with intermediate levels of biological
effects.
Bulk Sediment Toxicity
In the Bulk Sediment Toxicity approach, a number of bioassays are performed
using field-collected bulk sediment to determine whether the sediments produce
adverse effects on the growth, survival, or behavior of test organisms. The method
provides a direct measure of biological effects in total for whatever mixtures of
chemicals may be present. This method is routinely used to assess disposal
opportunities for dredged material and to assess the quality of sediments below
discharge points or in the vicinity of waste disposal sites (Adams et al., 1992).
The advantages of the approach are that it is relatively inexpensive to perform, it
can be performed on species from a nearby reference site, and it can integrate the
effects of mixtures of contaminants. This method cannot distinguish the chemical
6-6
-------
Chapter Six—Inventory Structure
agent responsible for the observed toxicity. Like the Sediment Quality Triad
approach, this method may be used to define boundaries of problem areas based
on relative sediment toxicity.
Interstitial Water Toxicity Identification Evaluation
The Interstitial Water Toxicity Identification Evaluation (TIE) approach is a
multistage procedure for evaluating the toxicity of sediment-associated chemicals
to aquatic organisms by exposing organisms to interstitial water, i.e., aqueous
solutions extracted by centrifugation or syringe from sediments. Interstitial water
is used based on the assumption that contact with interstitial water is the primary
route of exposure for organisms living in sediment. Once the degree of toxicity
to interstitial water has been evaluated, toxicity identification and evaluation
procedures are used to identify the contaminants) responsible for the toxicity and
to quantify the degree of biological response. The final and most important stage
of the Interstitial Water TIE approach is the confirmation of the suspected
contaminants using correlation of toxicity with contaminant concentrations, spiked
sediment bioassays, or observation of signs of intoxication among different species.
Apparent Effects Threshold
In the Apparent Effects Threshold (AET) approach, field data on biological effects
are compared with sediment concentrations of individual chemicals. The AET is
defined as the concentration above which biological effects are always observed
(based on statistical significance, P £ 0.05). Paired sediment chemistry and
biological effects data spanning a wide range in chemical concentration and
biological response are required. "Impacted" and "nonimpacted" sites are identified
based on whether the biological response of test organisms exposed to sediments
from the site is statistically different from the biological response measured for
sediment from a reference site. Unimpacted sites are selected and sorted by the
concentrations of each chemical of interest. The highest chemical concentration
in the sediments not causing biological effects is the AET value for that chemical
based on a specific biological response. Several different biological endpoints may
be used to obtain a range of AET values. The AET may be used to discriminate
contaminated sediments and to develop numerical SQC.
Spiked Sediment Toxicity
The Spiked Sediment Toxicity method is used to establish the safe sediment
concentration of a chemical by using a dose-response relationship developed from
sediment spike toxicity tests. The toxicity to one or more benthic organisms is
6-7
-------
NSI Framework
measured by exposing them to test sediments to which a range of chemical
concentrations has been added. This method establishes an unequivocal
relationship between individual chemicals and toxicity, may be used to examine the
joint action of several chemicals, has regulatory and scientific precedence, and can
be applied to all chemicals and sediments. This method assumes that exposure
conditions in the laboratory approximate conditions in the field. It may be used
to develop site-specific SQC by using organisms and sediments from a given site.
Tissue Residue
Sediment Quality Criteria for specific chemicals are established by defining a
critical pathway for exposure between contaminants in sediments and the organism
of interest. The critical pathway considers the exposure of benthic organisms to
contaminants through ingestion of sediments and phytoplankton. Bioaccumulation
and subsequent trophic transfer of the chemicals is modeled, taking into account
the growth and energy expenditure of the organism. The uptake of contaminant
across the gills is assumed to be proportional to the respiration rate of the
organism, which must be determined experimentally for all organisms in the food
chain as a function of water temperature and body weight. Safe concentrations of
contaminants in sediments are then back-calculated from acceptable tissue residue
concentrations. Acceptable tissue residues can be based on sublethal effects on
benthos or human health risk as determined from FDA action limits, state
standards, or cancer models. This method is protective of human health and
aquatic life because it takes into account bioaccumulation in fish tissue. The
method accounts for uptake of contaminants due to ingestion of sediments, prey,
and passage of water over the gills. It may be used for more than one class of
chemicals provided that values for the bioconcentration factor (BCF) of that
chemical are available. Without BCF values, however, this method can be applied
only to nonpolar, nonionic compounds. It can provide a site-specific SQC based
on sediment properties and types of organisms present.
Screening-Level Concentration
The Screening-Level Concentration (SLC) approach is a statistical method for
estimating the highest concentration'of a chemical in sediment that will not be
expected to produce an effect on benthic infaunal composition. Synoptic
observations of organic carbon-normalized chemical concentration and naturally
occurring benthic macroinvertebrate fauna are used to evaluate the quality of
sediments at a particular location. Co-occurrence analysis is used to link biological
effects at each site with the chemicals potentially contributing to these effects. For
each organism, a species screening-level concentration (SSLC) is estimated as the
highest concentration of a given contaminant that the organism can tolerate based
6-8
-------
Chapter Six—Inventory Structure
on the record of its presence or absence at the various monitoring stations for that
site. This concentration is estimated by plotting a cumulative frequency
distribution of the total number of stations where the organism is present versus
the organic carbon normalized concentration in the sediment of those stations. The
90th percentile concentration for the chemical becomes the species' SSLC. The
SLC is calculated for the chemical by plotting SSLCs obtained for a large number
of species as a frequency distribution. The SLC is defined as the concentration
above which 95 percent of the SSLCs are found. The method can be used to
derive site-specific SQC.
Long and Morgan (1990)
Long and Morgan (1990) used a weight-of-evidence approach for establishing
informal guidelines for assessing the sediments sampled within the NOAA NS&T
program. In this method, available site-specific sediment criteria, which were
developed using all available methods, were collected for each compound for
harbors, bays, and rivers in coastal marine and estuarine environments throughout
the United States (although most data are from the northeast and west coasts).
Frequently, SQC used were obtained from the equilibrium partitioning approach,
the apparent effects threshold, screening-level concentrations, and spiked sediment
bioassays. The study involved collecting matched individual monitoring chemical
and biological data for areas showing a gradient in concentration and effects. The
data were used to calculate various types of SQC. Spiked sediment bioassay data
were obtained from the literature. The SQC obtained for various sites and by
various methods were ranked from lowest to highest, and the values corresponding
to the 10th and 50th percentiles were described as the effects range low (ER-L)
and effects range medium (ER-M), respectively. Informal SQC were developed
for 43 chemicals or mixtures of chemicals including metals, PCBs, and pesticides.
MacDonald (1992)
MacDonald (1992) built upon the Long and Morgan (1990) approach used to
develop NS&T guidelines by including extensive data from the southeastern United
States and by incorporating data that demonstrated uncertain, or no, biological
effects, as well as those that demonstrated definite effects. The guidelines
developed by MacDonald (1992) are designed to be indicators of the general
relationship between contaminant concentrations and effects, not absolute indicators
of effects. Both the Region 4 and Gulf of Mexico sediment inventories described
in Chapter 3 of this document employed the MacDonald (1992) guidelines for
evaluating data.
6-9
-------
NSI Framework
The fcvo effecte levels generated by the MacDonald (1992) analysis are defined as
the Threshold Effects Level (TEL) and the Probable Effects Level (PEL) The two
effects levels are determined using both the Biological Effects Data Set (BEDS)
which consists of those data associated with definite biological effects, and the No
Biological Effects Data Set (NEEDS), which consists of those data associated with
no significant effects. The TEL is loosely defined as the level below which no
biological effects would be expected due to the single contaminant being
considered The PEL is loosely defined as that level above which biological
effects would nearly always be expected. It is important to note the TELs and
PELs are single chemical guideline levels that by themselves do not take into
account possible effects due to the presence of chemicals for which there are no
guidelines or the effects of multiple chemicals, which may have additive or
synergistic effects. One drawback of the MacDonald effects levels compared to
those determined by Long and Morgan (1990) is that the additional quality
assurance constraints imposed by MacDonald have resulted in effects levels being
determined for fewer chemicals (USEPA, 1992a).
6-10
-------
CHAPTER 7
CONCLUSIONS AND
RECOMMENDATIONS
The final output from the activities described in the previous chapters will be two-
fold. First, the National Sediment Inventory (NSI), which will include an actual
evaluation of detailed monitoring data from several sources, will be developed.
The NSI will also include biological and other data that were the basis for
classifying the contaminated sediment sites. The second output will be an
evaluation of data housed in the NSI and will include a listing of all those
locations across the country which are potentially severely contaminated and those
for which sufficient data exist to classify them as posing a significant risk to
human health and aquatic life.
The evaluation of data in the NSI will represent a snapshot of sediment
contamination problems across the country. It will provide a near-term screening
assessment of the national extent and severity and potential sources of sediment
contamination, thereby fulfilling the mandates of the Water Resources
Development Act of 1992 and contributing to meeting the objectives of EPA's
Contaminated Sediment Management Strategy. Any site included on the list of
potential or probable contaminated sites should be a target for future, more
intensive study, either to justify and recommend remedial or regulatory actions for
those sites which pose an obvious risk to the environment or to gather additional
information for those sites which appear to be severely contaminated but for which
there are insufficient data to reach a definitive conclusion.
By linking contaminated sites with potential sources, the data in the NSI could also
be used to evaluate the contribution to sediment contamination from various
contaminant sources, including point and nonpoint sources, thereby assisting
managers in assessing the need for stricter effluent controls and best management
practices. The Inventory could also help managers prioritize future remediation,
regulatory, or assessment activities; guide decisions regarding the appropriate type
and scale of regulatory action needed to reduce contaminant inputs; and evaluate
the effectiveness of existing technology-based effluent guidelines, water quality-
based controls, and nonpoint source controls. The Inventory could also be used to
identify and prioritize on a local, state, Regional, or national level those specific
chemicals in need of stricter regulation.
EPA recommends that the NSI be developed in a coordinated effort with a
modernized STORET. This approach will facilitate future updating and future
assessments of sediment quality.
EPA also recommends that efforts be made to ensure that future sediment quality
monitoring programs include additional information and parameter measurements
(which may currently be missing from many data sets), which can be used to more
7-1
-------
NSJ framework
accurately assess the potential environmental impacts of sediment contamination
during future assessments. For example, sediment sampling programs should
include the measurement of total or percent organic carbon content, sediment
particle size, sediment reductive capacity, and salinity. The data should also meet
certain minimum data quality objectives, and the results of data quality evaluations
should be reported with the data or, at a minimum, the QA/QC samples and
procedures used should be identified. Ensuring that ongoing and future data
collections contain these minimum data elements should result in the use of less
time and effort to locate relevant data, evaluate their utility for contaminant
assessment, and evaluate conditions at a particular site.
7-2
-------
REFERENCES CITED
Adams, W.J., R.A. Kimerle, and J.W. Barnett. 1992. Sediment quality and aquatic
life assessment. Environ. Sci. Technol. 26(10): 1864-1873.
Baudo, R., and H. Muntau. 1990. Lesser known in-place poUutants and diffuse
source problems. In Sediments: Chemistry and toxicity of in-place
pollutants, ed. R. Baudo, J. Giesy and H. Muntau, Chapter 1. Lewis
Publishers, Chelsea, MI.
Bolton, H.S., R.J. Breteler, B.W. Vigon, J.A. Scanlon, and S.L. Clark. 1985.
National perspective on sediment quality. Prepared for Office of Water
Regulations and Standards, U.S. Environmental Protection Agency,
Washington, DC.
COE and USEPA. 1991. Evaluation of dredged material proposed for ocean
disposal - testing manual. EPA 503/8-91/001. U.S. Army Corps of Engineers
and U.S. Environmental Protection Agency, Washington, DC.
Davies, J.C., S. Gusman, and F. Irwin. 1979. Determining unreasonable risk
under the Toxic Substances Control Act. The Conservation Foundation,
Washington, DC. pp. 1-33.
ERG. 1991. Preliminary cost estimates for selected components of EPA's
Sediments Management Strategy. Prepared for the Risk Assessment and
Management Branch, U.S. Environmental Protection Agency, Washington,
DC.
ERG. 1992. Clean water reauthorization study: Analysis of the costs and benefits
of remediating contaminated sediments. Draft report prepared for U.S.
Environmental Protection Agency, Office of Water, Washington, DC.
Gillett, J.W. 1983. A comprehensive pre-biological screen for ecotoxicological
effects. Environ. Toxicol. Chem. 2:463-476.
Kites, R.A., R£. LaFlamme, and J.G. Windsor. 1980. Polycyclic aromatic
hydrocarbons in marine/aquatic sediments: their ubiquity. In Petroleum in
the marine environment. Advances in chemistry series, No. 185, ed. L.
Petrakis and F.T. Weiss, pp. 289-311. American Chemical Society,
Washington, DC.
-------
Hoffman, EJ. 1985. Urban runoff pollutant inputs to Narragansett Bay:
Comparison to point sources. In Proc. Perspectives on Nonpoint Source
Pollution, sponsored by the U.S. Environmental Protection Agency, Kansas
City, Missouri, May 19-22, pp. 159-164.. EPA 440/5-85-001.
Johanson, E.E., and J.C. Johnson. 1976. Identifying and prioritizing locations for
the removal of in-place pollutants. U.S. Environmental Protection Agency,
Office of Water Planning and Standards, Washington, DC.
JRB Associates. 1984a. Background and review document on the development of
sediment criteria. Prepared for the U.S. Environmental Protection Agency,
Office of Criteria and Standards, under contract no. 68-01-6388.
JRB Associates. 1984b. Initial evaluation of alternatives for development of
sediment related criteria for toxic contaminants in marine waters (Puget
Sound). Phase II: Development and testing of the sediment-water equilibrium
partitioning approach. Prepared for the U.S. Environmental Protection
Agency, Office of Criteria and Standards, under contract no. 68-01-6388.
Krezoski, J.R., and J.A. Robbins. 1985. Vertical distribution of feeding and
particle-selective transport of 137Cs hi lake sediments by lumbriculid
oligochaetes. /. Geophys. Res. 90(C6): 11999-12006.
Livingston, E.H., and J.H. Cox. 1985. Urban stormwater quality management:
The Florida experience. In Proc. Perspectives on Nonpoint Source
Pollution, sponsored by the U.S. Environmental Protection Agency, Kansas
City, Missouri, May 19-22, pp. 289-291. EPA 440/5-85-001.
Long, E.R., and L.G. Morgan. 1990. Potential for biological effects ofsediment-
sorbed contaminants tested in the National Status and Trends Program.
NOAA Technical Memorandum NOS OMA 52. National Oceanic and
Atmospheric Administration, U.S. National Ocean Service, Rockville, MD.
86 pp. NTIS publication No. PB91-172288.
Lyman, W.J., A.E. Glazer, J.H. Ong, and S.F. Coons. 1987. An overview of
sediment quality in the United States. Prepared for the Office of Water
Regulations and Standards, U.S. Environmental Protection Agency,
Washington, DC.
Manheim, F.T. 1991. Summary of the Interagency Meeting on Contaminated
Sediment Database Development, June 12, 1990, Woods Hole,
Massachusetts.
-------
MacDonald, D.D. 1992. Development of an approach to the assessment of
sediment quality in Florida coastal waters. Workshop draft report prepared
for the Florida Department of Environmental Regulation, Honda Coastal
Management Program.
McCaffrey, R.J., A.C. Myers, E. Davey, G. Morrison, M. Bender, N. Luedtke, D.
Cullen, B. Froelich, and G. Klinkhammer. 1980. The relation between pore
water chemistry and benthic fluxes of nutrients and manganese in
Narragansett Bay, Rhode Island. Limnol. Oceanog. 25:31-44.
NAS. 1989. Contaminated marine sediments - Assessment and remediation.
Prepared jointly by the Committee on Contaminated Marine Sediments,
Marine Research Board, Commission on Engineering and Technical Systems,
and National Research Council for the National Academy Press, Washington,
DC.
NOAA. 1987. National Status and Trends Program: Progress report and
preliminary assessment of findings of the Benthic Surveillance Project -
1984, National Oceanic and Atmospheric Administration, Rockville, MD.
OTA. 1987. Wastes in the marine environment. U.S. Congress, Office of
Technology Assessment report, OTA-0-334.
Pait, A.S., A.E. Desouza, and D.R.G. Farrow. 1992. Agricultural pesticide use in
coastal areas: A national summary. National Oceanic and Atmospheric
Administration, National Ocean Science, Office of Ocean Resources
Conservation and Assessment, Strategic Assessments Division, Rockville,
MD.
Ryan, J., and J.H. Cox. 1985. The influence of NPS pollution in Florida
estuaries: A case study. In Proc. Perspectives on Nonpoint Source Pollution,
sponsored by the U.S. Environmental Protection Agency, Kansas City,
Missouri, May 19-22, pp. 172-179. EPA 440/5-85-001.
SAIC. 1985. Removal and mitigation of contaminated sediments. Draft report.
U.S. Environmental Protection Agency, Cincinnati, OH.
Salomons, W., N.M. de Rooij, H. Kerdijk and J. Bril. 1987. Sediment as a source
for contaminants? Hydrobiologia 149:13-30.
Schumacher, B.A., and B.L. Conkling. 1990. User's guide to the quality
assurance/quality control evaluation scale of historical datasets. Prepared
for the U.S. Environmental Protection Agency, Great Lakes National
Program Office, under the Assessment and Remediation of Contaminated
Sediments (ARCS) Program.
-------
Southerland, E., M. Kravitz, and T. Wall. 1992. Management framework for
contaminated sediments (The U.S. EPA Sediment Management Strategy). In
Sediment toxicity assessment, ed. G.A. Burton, pp. 431-370. Lewis
Publishers, Boca Raton, FL.
TRI. 1992. Impact of toxic substances and pesticides on nearshore Gulf of Mexico:
A preliminary comparison (toxicity indices) of estuarine drainage systems
based on releases of toxics from industrial and municipal sites and pesticide
run-off from agricultural operations in 1989. Prepared by Technical
Resources, Inc., for U.S. Environmental Protection Agency, Office of
Research and Development, Gulf Breeze, FL.
USEPA. 1986. Hazard Evaluation Division, standard evaluation procedure:
Ecological risk assessment, U.S. Environmental Protection Agency, Office
of Pesticide Programs, Washington, DC, June 1986. EPA-450/9-85-001.
USEPA. 1989. Guidance on the regulatory uses of equilibrium partitioning-based
sediment criteria, Draft report. U.S. Environmental Protection Agency,
Office of Water Regulations and Standards, Analysis and Evaluation
Division, Washington, DC, September 1989.
USEPA. 1990. Managing contaminated sediments: EPA decision-making
processes. EPA 506/6-90/002. U.S. Environmental Protection Agency,
Sediment Oversight Technical Committee, Washington, DC.
USEPA. 199 la. Guidance for water quality-based decisions: The TMDL process.
EPA 440/4-91-001. U.S. Environmental Protection Agency, Office of Water,
Washington, DC.
USEPA. 1991b. Summary report for contaminated sediment assessments in U.S.
EPA Region IV coastal areas. Prepared by Science Applications
International Corporation for U.S. Environmental Protection Agency Region
IV, Atlanta, GA.
USEPA. 1992a. Evaluation of the Region 4 inventory of coastal sediment sites.
Prepared by Science Applications International Corporation for U.S.
Environmental Protection Agency, Region IV, Atlanta, GA.
USEPA. 1992b. EPA Region 5 inventory of contaminated sediment sites.
Prepared by Science Applications International Corporation for U.S.
Environmental Protection Agency. Region 5, Chicago, IL. Draft, July 1992.
USEPA. 1992c. EPA's contaminated sediment management strategy: A proposal
for discussion. Draft outline prepared by the Contaminated Sediments
Assessment Workgroup of the U.S. Environmental Protection Agency,
Washington, DC. March 1992.
-------
USEPA. 1992d. Proceedings of EPA's contaminated sediment management
strategy forums. Chicago, EL, April 21-22; Washington, DC, May 27-28 and
June 16, 1992. EPA 823-R-92-007.
USEPA. 1992e. Sediment classification methods compendium. Prepared by the
U.S. Environmental Protection Agency, Sediment Oversight Technical
Committee, Washington, DC.
USFWS. 1986. Preliminary survey of contaminant issues of concern on National
Wildlife Refuges. U.S. Fish and Wildlife Service.
-------
------- |