^.":tec States
Environmental Protection
Agency
Office of Science ana Tecnnoiogy ane
Office of Researcn and Development
Wasnington, DC 20460
Equilibrium Partitioning
Sediment Guidelines (ESGs)
for the Protection of Benthic
Organisms: Metal Mixtures
(Cadmium, Copper, Lead, Nickel,
Silver, and Zinc)
DRAFT
m^^^S^ffr^^^^i
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Foreword
Under the Clean Water Act (CWA), the U.S. Environmental Protection Agency (EPA) and the
States develop programs for protecting the chemical, physical, and biological integrity of the
nation's waters. To meet the objectives of the CWA, EPA has periodically issued ambient water
quality criteria (WQC) beginning with the publication of "Water Quality Criteria, 1972" (NAS,
1973). The development of WQC is authorized by Section 304(a)( 1) of the CWA. which directs
the Administrator to develop and publish "criteria" reflecting the latest scientific knowledge on
(1) the kind and extent of effects on human health and welfare, including effects on plankton, fish.
shellfish, and wildlife, that may be expected from the presence of pollutants in any body of water.
including ground water; and (2) the concentration and dispersal of pollutants on biological
communiry diversify, productivity, and stability. All criteria guidance through late 1986 was
summarized in an EPA document entitled "Quality Criteria for Water, 1986" (U.S. EPA. 1987a).
Updates on WQC documents for selected chemicals and new criteria recommendations for other
pollutants have been more recently published as "National Recommended Water Quality Criteria-
Correction" (U.S. EPA. 1999a). EPA will continue to update the nationally recommended WQC
as needed in the future.
In addition to the development of WQC and to continue to meet the objectives of the CWA, EPA
has conducted efforts to develop and publish equilibrium partitioning sediment guidelines (ESGs) .
for some of the 65 toxic pollutants or toxic pollutant categories. Toxic contaminants in bottom
sediments of the nation's lakes, rivers, wetlands, and coastal waters create the potential for
continued environmental degradation even where water column contaminant levels meet
applicable water quality standards. In addition, contaminated sediments can lead to water quality
impacts, even when direct discharges to the receiving water have ceased. These guidelines are
authorized under Section 304(a)(2) of the CWA, which directs the Administrator to develop and
publish information on, among other things, the factors necessary to restore and maintain the
chemical, physical, and biological integrity of all navigable waters.
The ESGs and associated methodology presented in this document are EPA's best recommendation
as to the concentrations of a substance that may be present in sediment while still protecting
benthic organisms from the effects of that substance. These guidelines are applicable to a variety
of freshwater and marine sediments because they are based on the biologically available
concentration of the substance in the sediments.. These ESGs are intended to provide protection to
benthic organisms from direct toxicity due to this substance. In some cases, the additive toxicity ;
for specific classes of toxicants (e.g., metal mixtures or polycyclic aromatic hydrocarbon
mixtures) is addressed. The ESGs do not protect against synergistic or antagonistic effects of
contaminants or bioaccumulative effects to benthos. They are not protective of wildlife or human -
health endpoints.
EPA recommends that ESGs be used as a complement to existing sediment assessment tools, to
help assess the extern of sediment contamination, to help identify chemicals causing toxicity, and
to serve as targets for pollutant loading control measures. EPA is developing guidance tp assist in
the application of these'guidelines in water-related programs of the States and this Agency.
This document provides guidance to EPA Regions, States, the regulated community, and the
public. It is designed to implement national policy concerning the matters addressed. It does not,
however, substitute for the CWA or EPA's regulations, nor is it a regulation itself. Thus, it
cannot impose legally binding requirements on EPA, States, or the regulated communiry. EPA
and State decisionmakers retain the discretion to adopt approaches on a case-by-case basis that
ill
-------
Foreword
differ from this guidance where appropriate. EPA may change this guidance in the future.
This document has been reviewed by EPA's Office of Science and Technology (Health and
Ecological Criteria Division. Washington, DC) and Office of Research and Development (Mid-
Continent Ecology Division, Duluth, MN; Atlantic Ecology Division, Narragansert, RI), and
approved for publication.
Nfemion of trade names or commercial products does not constitute endorsement or
recommendation of use.
Front cover image provided by Wayne R. Davis and Virginia Lee.
i.v
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Contents
Foreword [[[ ................................. iii
Acknowledgments ......................... . ................................................ LX
Executive Summary [[[ .xi
Glossary of Abbreviations [[[ xiii
Section 1
Introduction ........ . [[[ 1-1
1.1 General Information ...................................... . [[[ .. 1-1
1.2 Applications of Sediment Guidelines [[[ 14
1.3 Overview
Section 2
Partitioning of Metals in Sediments .............................................. :-i
2.1 Metal Toxicity in Water-Only and in Interstitial Water of Sediment Exposures ......... 2-1
2.1.1 Toxicity Correlates to Metal Activity ..................................... '. ....................... 2-1
2.1.2 Toxicity Correlates to Interstitial Water Concentration ................................. 2-3
22 Solid-Phase Sulfide as the Important Binding Component ........................................ 2-3
2.2.1 Metal Sorption Phases .................... . [[[ 2-8
222 Titration Experiments ....................... . [[[ 2-8'
2.2.2.1 Amorphous FeS [[[ 2-8
2.2.2.2 Sediments [[[ 2-10
2.2.3 Correlation to Sediment AVS ................................... .' .................................... 2-11
22.4 Solubility Relationships and Displacement Reactions ................................ 2-11
'225 Application to Mixtures of Metals ..... [[[ 2-12
Section 3
Toxicity of Metals in Sediments ..................... . ............................. 3-1
3.1 General Information [[[ . ................................... 3-1
3.1.1 Terminology ..... : [[[ '. ................. . ................ 3-1
32 Predicting Metal Toxicity: Short-Term Studies [[[ 3-1
32.1 Spiked Sediments: Individual Experiments ................ , ................................... 3-1
322 Spiked Sediments: All Experimental Results Summarized .............................. 3-5
-------
Contents
3.4 Predicting Toxicity of Metals in Sediments >17
3.4.1 General Information 3-17
. 3.4.2 EqP Theory for SEM. AV5, and Organic Carbon 3-19
3.4.3 Data Sources 3-20
3.4.4 Acute Toxicity Uncertainty 3-20
3.4.5 Chronic Toxicity Uncertainty 3-22
3.4.6 Summary 3-22
Section 4
Derivation of the ESGs for Metals 4-1
4.1 General Information ....4^1 '
42. Sediment Guidelines for Multiple Metals 4-1
4.2.1 ' AVS Guidelines 4-2
4.2.2 Interstitial Water Guidelines 4-2
4.2.3 Summary •.... : 4-3
4.3 Example Calculation of ESGs for Metals and EqP-Based Interpretation 4-3
4.4 ESG for Metals vs. Environmental Monitoring Databases 4-5
4.4.1 Data Analysis 4-5
4.4.1.1 Freshwater Sediments 4-5
4.4.1.2 Saltwater Sediments • 4-7
4.5 Bioaccumulation 4-7
Section 5
Sampling and Analytical Chemistry 5-1
5.1 General Information 5-1
52 Sampling and Storage 5-1
5.2.1 Sediments .' 5-2
5.22 Interstitial Water •. 5-2
5.3 Analytical Measurements 5-3
5.3.1 Acid Volatile Sulfide 5-4
5.32 Simultaneously Extracted Metals 54
5.3.3 Total Organic Carbon 54
5.3.4 Interstitial Water Metal 54
k
Section 6
Guidelines Statement 6-1
6.1 AVSGuideline .- 6-1
62 Interstitial Water Guideline 6-1
Section 7
References : 7-1
Appendix A A-I
Appendix B ; B-I
VI
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Tables
Table 2- 1. Cadrnium binding capacity and AVS of sediments [[[ 2-11
Table 2-2. Metal sulfide solubility products and ratios .......................... [[[ 2-13
Table 3-1. .Toxicity of sediments from freshwater and saltwater lab-spiked sediment tests, field locations,
and combined lab-spiked and field sediment tests [[[ 3-8
Table 3-2. Summary of the results of full life-cycle and colonization toxicity tests conducted in the
laboratory and field using sediments spiked with individual metals and metal mixtures .................. 3-15
Table 3-3. Test-specific data for chronic toxicity of freshwater and saltwater organisms compared to
................................... : [[[ 3-26
Table 4- 1 . Water qual'ity criteria ( WQC) criteria continuous concentrations (CCC) based on the
dissolved concentration of metal ............................... '.. ............ :..". [[[ 4-2
Table 4-2. ESGs tor metal mixtures: Example calculations for three sediments [[[ 4-4
Figures
Figure 2-1. • Acute toxicity to grass shrimp (Palaemonetes pugio) of total cadmium and cadmium activity
with different concentrations of the complexing ligands NTA and chloride as salinity ...................... 2-2
Figure 2-2. Acute toxicity of total copper and copper activity to the dinoflagellate Gonyaulax tamarensis
with and without the complexing ligand EDTA [[[ 2-3
Figure 2-3. Specific growth rates of a diatom (Thalassiosira pseudonana) and a unicellular algae
(Monochrysis lutheri) versus total copper and copper activity for a range of concentrations
of the complexing ligands Tris and natural DOC in river water 24
Figure 2-4. Copper accumulation in oysters (Crassostrea virginica) versus total copper and copper
activity with different levels of the complexing ligand NTA 2-5
Figure 2-5. Mean survival of the amphipod Rhepoxynius abronius versus dissolved cadmium concentration
for 4-day toxicity tests in seawater and 0- and 4-day tests in interstitial water 2-6
Figure 2-6. Mortality versus interstitial water cadmium activity for sediments from Long Island Sound,
Ninigret Pond, and a mixture of these two sediments 2-7
Figure 2-7. Toxicity. of copper to Hyalella azteca versus copper concentrations in a water-only exposure
-------
Contents
Figure 3-1. Percentage mortality of amphipods (Ampelisca abdita and Rhepoxynius hudsoni) exposed to
sediments from Long Island Sound. Ninigret Pond, and a mixture of these two sediments as a
function of the sum of the concentrations of metals in sediments expressed as dry weight,
interstitial water cadmium activity, and the sediment cadmium/AVS ratio 3-3
Figure 3-2. Concentrations of individual metals in interstitial water of sediments from Long Island Sound
and Ninigret Pond in the mixed met-is experiment as a function of SEM/AVS ratio 3-4
Figure 3-3. Percentage mortality of freshwater and saltwater benthic species in 10-day toxicity tests in
sediments spiked with individual metals (Cd. Cu. Pb, Ni, Ag. or Zn) or a metal mixture
(Cd. Cu, Ni', and Zn) 3-6
Figure 3-4. Percentages of the 184 spiked sediments from Figure 3-3 that were nbntoxic or toxic over
various intervals of concentrations of metal based on sediment dry weight (^mol/g), IWTU,
and SEM/AVS " 3-7
Figure 3-5. Percentage mortality of amphipods. oligochaetes, and polychaetes exposed to sediments from
four freshwater and three saltwater field locations as a function of the sum of the molar
concentrations of SEM minus the molar concentration of AVS (SEM-AVS) 3-H
Figure 3-6. Percentage mortality of freshwater and saltwater benthic species in 10-day toxicity tests in
spiked sediments and sediments from the field 3-12
Figure 3-7. Comparison of the chronic toxicity of sediments spiked with individual metals or metal
mixtures to predicted toxicity based on SEM-AVS '. 3-18
Figure 3-8.. Percent mortality versus SEM-AVS and (SSEM-AVS)//^. for saltwater field data without Bear
Creek and Jinzhou'Bay, freshwater field data, freshwater spiked data, and saltwater spiked data ... 3-21
Figure 3-9. Percent mortality versus (SEM^^-AVS)//^ for each metal in spiked sediment tests using
Ampelisca, Capitella, Neamhes, Lumbriculus. and Helisoma •. 3-23
Figure 3-10. Percent mortality versus (SEM^-AVSV/Q,. for silver and (LSEM-AVS)//^. for a mixture
experiment using Cd, Cu, Ni, andZn '. 3-24
Figure 3-11. Comparison of the chronic toxicity of sediments spiked with individual metals or metal
mixtures to predicted toxicity based on (SEM-AVS)//^ 3-25
Figure 4-1. • SEM-AVS values versus AVS concentrations in EMAP-Great Lakes sediments from
Lake Michigan. Plot (A) shows all values; plot (B) has the ordinate limited to SEM-AVS
values between-10 and+10 Mmol/g 4-6 .
Figure 4-2. SEM-AVS values versus AVS concentrations in EMAP-Estuaries Virginian Province;
REMAP-NY/NJ Harbor Estuary; NOAA NST-Long Island Sound; Boston Harbor;
and Hudson-Raritan Estuaries , 4-8
Figure 4-3. (ZSEM-AVS)//^. versus AVS concentrations in EMAP-Estuaries Virginian Province;
REMAP-NY/NJ Harbor Estuary; NOAA NST-Long Island Sound; Boston Harbor; and
Hudson-Raritan Estuaries •• • 4-9
via
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Acknowledgments
Coauthors
David J. Hansen HydroQual, Inc.. Mahwah. NJ; Great Lakes Environmental Center,
Traverse City, MI (formerly with U.S. EPA)
Dominic M. Di Toro Manhattan College, Riverdale NY; HydroQual, Inc., Mahwah. NJ
Walter J. Berry* ' U.S.EPA.NHEERL. Atlantic Ecology Division. Narragansett.RI
Gerald T. AnkJey U.S. EPA. NHEERL, Mid-Continent Ecology Division, Duluth, MN
Joy A. McGrath HydroQual. Inc., Mahwah. NJ
Laurie D. De Rosa HydroQ.ual. Inc., Mahwah. NJ
Heidi E. Bell* U.S. EPA. Office of Water. Washington, DC
MaryC. Reiley -. U.S. EPA, Office of Water. Washington. DC
Christopher S. Zarba U.S. EPA. Office of Research and Development. Washington. DC
Significant Contributors to the Development of the Approach and Supporting Science
Herbert E. Allen University of Delaware, Newark. DE
Gerald T. Ankley U.S. EPA. NHEERL. Mid-Continent Ecology Division. Duluth, MN
Dominic M. Di Toro Manhattan College, Riverdale NY; HydroQual, Inc., Mahwah, N'J
David J. HanSen HydroQual, Inc., Mahwah, NJ; Great Lakes Environmental Center.
Traverse City, MI (formerly with U.S. EPA)
Landis Hare Universite du Quebec. Sainte-Foy, Quebec, Canada
John D. Mahony Manhattan College. Riverdale. NY
RichajdC. Swartz • Environmental consultant (formerly with U.S. EPA)
»
Christopher S. Zarba U.S. EPA, Office of Research and Development. Washington, DC
Technical Support and Document Review
Robert M. Burgess U.S. EPA. NHEERL. Atlantic Ecology Division. Narragansett. RI
Tyler K. Linton Great Lakes Environmental Center, Columbus, OH
David R. Mount U.S. EPA. NHEERL, Mid-Continent Ecology Division, Duluth. MN
Robert L. Spehar U.S. EPA. NHEERL, Mid-Continent Ecology Division, Duluth. MN
* Principal U.S. EPA contact
IX
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Executive Summary
This equilibrium partitioning sediment guideline (ESG1 document recommends a sediment
concentration for mixtures of cadmium, copper, lead, nickel, silver, and zinc that is EPA's best
estimate of the concentration of the mixture that will protect benthic organisms from the direct
toxicity of these metals in sediments. The equilibrium partitio.ning (EqP) approach was chosen
because it accounts for the varying biological availability of these metals in different sediments
and allows for incorporation of the appropriate biological effects concentration. This provides for
derivation of a guideline that is causally linked to these specific metals, applicable across
sediments, and appropriately protective of benthic organisms.
.Equilibrium partitioning theory predicts that these metals partition in sediment between acid
volatile sulfide (AVS. principally iron monosulfide), interstitial (pore) water, benthic organisms.
and other sediment phases such as organic carbon. Biological responses of benthic organisms to
these metals in sediments are different across sediments when the sediment concentrations are
expressed on a dry weight basis, but similar when expressed on a ZSEM-AVS or interstitial water
basis. The difference between the sum of the molar concentrations of simultaneously extracted
metal (SSEM. the metal extracted in the AVS extraction procedure) minus the molar concentration
of AVS accurately predicts which sediments are not toxic because of these metals. The use of
(SSEM-AVS)//^. reduces variability associated with prediction of when sediments will be toxic.
The ESG for mixtures of the metals cadmium, copper, lead, nickel, silver, and zinc is based on the
solid phase and interstitial water phase of sediments. In sediments, these metals should not cause
direct toxicity to benthic organisms if the ISEM-AVS is sO.O. Alternatively, sediments containing
these metals should not cause direct toxicity to benthic organisms if the sum of the dissolved
interstitial water concentrations for each of the metals (ZM. J divided by their respective water
quality criteria final chronic value (FCV) is s 1.0. Uncertainty bounds on ISEM-AVS and (2SEM-
A\'S)/fx can be used to identify sediments where toxicity, because of these metals, is unlikely.
uncertain, or likely.
These sediment guidelines apply to sediments having AVS concentrations sO. 1 umol/g. They are
not intended to protect against additive, synergistic, or antagonistic effects of other contaminants
or bioaccumulative effects to aquatic life, wildlife, or humans. It is the position of the Agency and
the EPA Science Advisory Board (SAB) that the use of ESGs as stand-alone, pass-fail criteria is
not recommended for all applications and should frequently trigger additional studies at sites
under investigation.
EPA has developed both Tier I and Tier 2 ESGs to reflect the differing degrees of data availability
and uncertainty. Requirements for a Tier 1 ESG include a Kow, FCV, and sediment toxicity tests to
verify EqP assumptions. In comparison, a Tier 2 ESG requires a Kov and a FCV or secondary
chronic value (SCV); sediment toxicity tests are recommended but not required. The ESGs derived
for metal mixtures in this document, as well as the ESGs for dieldrin, endrin. and polycyclic
aromatic hydrocarbon (PAH) mixtures represent Tier 1 ESGs (U.S. EPA. 2000d,e,f). Information on
how EPA recommends ESGs be applied in specific regulatory programs is described in the
"Implementation Framework for the Use of Equilibrium Partitioning Sediment Guidelines (ESGs)"
(EPA.:OOOc).
XI
-------
Equilibrium Partitioning Sediment Guidelines lESGs): Metal Mixtures
Glossary of Abbreviations
Ag Silver
Ag,S Silver monosulfide
AVS Acid volatile sulfide
CCC Criteria continuous concentration
Cd Cadmium
{Cd2*} Activity of ionic cadmium (mol/L)
[Cd-*] Concentration of ionic cadmium (mOl/L)
[Cd]A . Concentration of added cadmium {mol/L)
[Cd]B Concentration of bound cadmium (mol/L)
[CdS(s)] Concentration of solid-phase cadmium sulfide (mol/L)
Cs Concentration of contaminant in sediment
C' Sediment LC50 Concentration
Cu Copper
CWA Clean Water Act
DOC Dissolved organic carbon
EDTA Ethlyenediaminetetra-acetic acid
EMAP Environmental Monitoring and Assessment Program
EPA U.S. Environmental Protection Agency
EqP Equilibrium partitioning
ESG(s) Equilibrium partitioning sediment guideline(s)
fx Fraction of organic carbon in sediment
PCV Final chronic value
Fe Iron
{Fe2*} Activity of ionic iron (mol/L)
[Fe2*] Concentration of ionic iron (mol/L)
[FeS(s)j Concentration of solid-phase iron sulfide (mol/L)
[FeS(s)]i Concentration of initial solid-phase iron sulfide (mol/L)
FeS Iron monosulfide
XIII
-------
Glossary
K,
ftS
K,
MS
voc
K.
GF.AA Gas Furnace Atomic Absorption Spectrophotometry
HECD U.S. EPA, Health and Ecological Criteria Division
[WGU Interstitial water guidelines unit
rWTU Interstitial water toxic unit
Solubility product for FeS(s) [(mol/L)2]
Solubility product for MS(s) [(mol/L)2]
Organic carbon-water partition coefficient
• Sediment-interstitial water partition coefficient
K^f Solubility product constant
LC50 Concentration estimated to be lethal to 50% of the test organisms within
a specified time period
M:* Divalent metal—cadmium, copper, lead, nickel, silver, or zinc
MOH* Metal hydroxide
MS Metal sulfide-
Mn Manganese
{M:*} Divalent metal activity (mol/L)
[M;*] Concentration of ionic metal (mol/L)
[M] x Concentration of added metal (mol/L).
[M]B Concentration of bound metal (mol/L)
[MJ Dissolved metal concentration in the interstitial water
[MS(s)] Concentration of solid-phase metal sulfide (mol/L)
[MT] Total cold extractable metal (mol/L)
NA Not applicable, not available
NAS National Academy of Sciences
M Nickel
NOAA National Oceanographic and Atmospheric Administration
NOEC No observed effect concentration
NST National Status and Trends monitoring program
NTA Nitrilotriacetic acid
NTIS National Technical Information Service-
Pb Lead
OEC Observed effect concentration
xiv
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
ORD U.S. EPA. Office of Research and Development
OST U.S. EPA. Office of Science and Technology
POC Participate organic carbon
REMAP Regional Environmental Monitoring and Assessment Program
S- Sulfide ion
(S:'} Activity of sulfide (mol/L) '.
[S2'] Concentration of sulfide (mol/L)
SAB U.S. EPA Science Advisory Board
SD Standard deviation
SEM • Simultaneously extracted metals
[SEM,.] Simultaneously extracted metals, concentration of the combined metals
Cumol/g)
[SEMCJ Simultaneously extracted metals. Cd concentration (/^mol/g)
[SEMCJ Simultaneously extracted metals. Cu concentration (^mol/g.)
(SEMpJ Simultaneously extracted metals. Pb concentration (^mol/g)
[SEM.J Simultaneously extracted metals, Ni concentration (^mol/g)
[SEM^ ] Simultaneously extracted metals. Ag concentration Cumol/g)
[SEMJ S imultaneously extracted metals, Zn concentration Gumol/g)
TIE Toxicity identification evaluation
TOC Total organic carbon
WQG Water quality criteria
Zn . Zinc
[ECd(aq)] . Concentration of total dissolved Cd2* (mol/L)
[SFe(aq)] Concentration of total dissolved Fe:* (mol/L)
[2M(aq)] Concentration of total dissolved M2* (mol/L)
[LS(aq)] Concentration of total dissolved S2' (mol/L)
XV
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 1
Introduction
1.1 General Information
Under the Clean Water Act (CWA). the U.S.
Environmental Protection Agency (EPA) is responsible
for protecting the chemical, physical, and biological
integrity of the nation's waters. In keeping with this
responsibility, EPA published ambient water quality
criteria (WQC) in 1980 for 64 of the 65 toxic pollutants
or pollutant categories designated as toxic in the CWA.
Additional water quality documents that update criteria
for selected consent decree chemicals and new criteria
have been published since 1980. These WQC are
numerical concentration limits that are EPA's best
estimate of concentrations protective of human health
and the presence and uses of aquatic life. Although
these WQC play an important role in ensuring a
healthy aquatic environment, they alone are not
sufficient to ensure the protection of environmental or
human health.
Toxic pollutants in bottom sediments of the
nation's lakes, rivers, wetlands, estuaries, and marine
coastal waters create potential for continued
environmental degradation even where water column
concentrations comply with established human health
and aquatic life WQC. In addition, contaminated
sediments can be a significant pollutant source that
may cause water quality degradation to persist, even
when other pollutant sources are stopped (U.S. EPA
1997a,b,c;Larsson, 1985, Salomons etal., 1987;
Burgess and Scott, 1992). The scarcity of defensible
sediment guidelines and the single chemical nature of
those available make it difficult to (1) accurately assess
the extent of the ecological risks of contaminated
sediments. 1,2) establish pollution prevention
strategies, and (3) identify, prioritize, and implement
appropriate .cleanup activities and source controls.
As a result of the need for guidance to assist
regulatory agencies in making decisions concerning
contaminated sediment problems and their prevention,
a research team was established from the EPA Office of
Science and Technology (OST) and Office of Research
and Development (ORD) to review alternative
approaches (Chapman. 1987), All of the approaches
reviewed had both strengths and weaknesses, and no
single approach was found to be applicable for
sediment guidelines derivation in all situations (U.S.
EPA, 1989). The equilibrium partitioning (EqP)
approach was selected, and first applied for nonionic
organic chemicals, because it presented the. greatest
promise for generating defensible national chemical-
specific sediment guidelines applicable across a broad
range of sediment types. The term equilibrium
partitioning sediment guidelines (ESGs) refers to
numerical concentrations for individual chemicals or
mixtures of chemicals that are applicable across the
range of sediments encountered in practice. The three
principal observations that established the EqP method
of deriving sediment guidelines for nonionic organic
chemicals were as follows:
1. The concentration of nonionic organic chemicals in
sediments, expressed on an organic carbon basis,
and in interstitial waters, correlates to observed
biological effects on sediment-dwelling organisms
across a range of sediments.
2. Partitioning models can relate sediment
concentrations for nonionic organic chemicals on
an organic carbon basis to freely-dissolved
concentrations in interstitial water.
3. The distributions of sensitivities of bemhic and
water column organisms to chemicals are similar;
thus, the currently established WQC final chronic
values (FCVs) can be used to define the acceptable
effects concentration of a chemical freely-dissolved
in interstitial water.
Because of their widespread release and persistent
nature, metals such as cadmium, copper, lead, nickel.
silver, and zinc are commonly elevated in aquatic
sediments. These metals, in addition to nonionic
organic chemicals, are of potential concern to aquatic
environments. Thus, there have been various
proposals for deriving sediment guidelines for
protecting benthic communities using measurement of
total sediment metals followed by comparison wi'th
background metal concentrations, or in some cases, an
effects-based endpoint (Sullivan et al., 1985; Persaud et
al.; 1989; Long and Morgan. 1990; Ingersoll et al., 1996;
MacDonald et al., 1996). An important limitation to
these types of approaches is that the causal linkage
1-1
-------
Introduction
between the measured concentration of metals and the
observed toxictty cannot be established, in pan
because of the procedures used to derive correlative
values, and because values derived are based on total
rather than bioavailable metal concentrations. That is,
for any given total metal concentration, adverse
toxicological effects may or may not occur, depending
on the physicochemical characteristics of the sediment
of concern (Tessier and Campbell, 1987;Luoma, 1989;
DiToroetal., 1990).
Many researchers have used elaborate sequential
extraction procedures to identify sedimentary
physicochemical fractions with which metals are
associated in an attempt to understand the biological
availability of metals in sediments (Tessier etal., 1979;
Luoma and Bryan. 1981). Key binding phases for
metals in sediments included iron and manganese
oxides and organic carbon. Shortcomings with these
approaches have limited their application largely to
aerobic sediments instead of anaerobic sediments,
where metals are often found in the greatest
concentrations (see Section 2).
In developing ESGs for metals that causally link
metals concentrations to biological effects and that
apply across all sediments, it is essential that
bioavailability be understood. Therefore, the EqP
approach was selected as the technical basis for
deriving ESGs for metals. Different studies have shown
that although total (dry weight) metal concentrations in
anaerobic sediments are not predictive of
bioavailability, metal concentrations in interstitial water
are correlated with observed biological effects (Swartz
et al.. 1985; Kemp and Swartz. 1986). However, as
opposed to the situation for nonionic organic
chemicals and organic carbon (see Di Toro et al., 1991),
se.diment partitioning phases controlling interstitial
water concentrations of metals were not readily
apparent. A key partitioning phase controlling cationic
metal activity and metal-induced toxicity in the
sediment-interstitial water system is acid volatile
sulfide (AVS) (Di Toro et al., 1990,1992). AVS binds, on
a molar basis, a number of cationic metals of
environmental concern (cadmium, copper, lead, nickel,
silver, and zinc), forming insoluble sulfide complexes
with minimal biological availability. (Hereafter in this
document, the use of the term "metals" will apply only
to these six metals.)
The data that support the EqP approach for
deriving sediment guidelines for nonionic organic
chemicals were reviewed by Di Toro et al. (1991) and
U.S. EPA (1997a; 2000a). The utility of the EqP
approach for deriving sediment guidelines for metals
(U.S. EPA. 1994a) was reviewed and endorsed by EPA's
Science Advisory Board (SAB) in 1994 and 1999 (U.S.
EPA, 1995a, 1999b). The data that support the EqP
approach for deriving sediment guidelines for metals
presented in this document were taken largely from a
series of papers published in the December 1996 issue
of Environmental Toxicology and Chemistry by
Ankley et al. (1996), Berry et al. (1996), DeWitt et al.
(1996), Di Toro et al. (1996a.b). Hansen et al. (I996a,b).
Leonard et al. (1996a), Liber et al. (1996). Mahony et al.
(1996), Peterson et al. (1996), and Sibley et al. (1996). In
addition, publications by Di Toro et al. (1990.1992)'.
Ankley et al. (1994), U.S. EPA (1995a). and Berry et al.
(1999) were of particular importance in the preparation
of this document.
The same three general principles observed in
applying the EqP approach to nonionic organic
chemicals listed above also apply with only minor
adjustments to deriving ESGs for mixtures of the
cationic metals—cadmium, copper, lead, nickel, silver.
and zinc:
1. The concentrations of these six metals in
sediments, normalized to the concentration of AVS
and simultaneously extracted metals (SEM) (the
metals extracted with AVS) in sediments and
dissolved in interstitial waters, correlate with
observed'biological effects to sediment-dwelling
organisms across a range of sediments (Di Toro et
al.. 1992).
2. Partitioning models can relate sediment
concentrations for cationic divalent metals (and
monovalent silver) on an AVS basis to the absence
of freely-dissolved concentrations in interstitial
water.
3. The distributions of sensitivities of benthic and
water column organisms to organic chemicals and
metals are similar (U.S. EPA, 2000a); thus, the
currently established WQC FCVs can be used to
define the acceptable effects concentration of the
metals freely dissolved in interstitial water.
The EqP approach, therefore,.assumes that,(l) the
partitioning of the metal between sediment AVS (or any
other binding factors controlling bioavailability) and
interstitial water approximates equilibrium; (2)
organisms receive equivalent exposure from interstitial
water-only exposure or from exposure to any other
equilibrated sediment phase: either from interstitial
water via respiration, sediment via ingestion, or
1-2
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
sediment-integument exchange, or from a mixture of
exposure routes; (3) for the cationic metals cadmium.
copper, lead, nickel, zinc, and silver, partitioning of
metal between the solid phase and interstitial water can
be predicted based on the relative concentrations of
AVS and SEM; (4) the WQC FCV concentration is an
appropriate effects concentration for freely-dissolved
metal in interstitial water; and (5) the toxicity of metals
m interstitial water is no more than additive.
For the first time, the Agency is publishing ESGs
that account for bioavaiiability in sediments and the
potential for effects of a metal mixture in the aquatic
environment, thus providing an ecologically relevant
benchmark. Two equally applicable ESGs for metals, a
solid phase and an interstitial.water phase, are
described. The solid-phase AVS ESG is defined as the
£ [SEM.] i [AVS] (total molar concentration of
simultaneously extracted metal is less than or equal to
the total molar concentration of acid volatile sulfide).
Note that cadmium, copper, lead, nickel, and zinc are
divalent metals so that one mole of each metal can bind
with one mole of AVS. The molar concentrations of
these metals are compared with AVS on a one-to-one
basis. Silver, however, exists predominantly as a
monovalent metal, so that silver monosulfide (Ag2S)
binds two moles of silver for each mole of AVS.
Therefore, SEMV| by convention will be defined as the
molar concentration of silver divided by two. [Ag]/2,
which is compared with the molar AVS concentration.
The interstitial water phase ESG is S(M. J/[FC V. J s 1
i the sum of cadmium, copper, nickel, lead, and zinc of
the concentration of each individual metal dissolved!in
the interstitial water divided by the metal-specific FCV
based on dissolved metaj is less than or equal to one;
note that at present EPA Iocs not have an FCV for
silver). This latter value is termed an interstitial water
guidelines unit (IWGU). A requirement of the IWGU
approach is that the toxicities of interstitial water metal ;
concentrations be additive. The data presented in this
document support the additivity of the toxicity of metal
mixtures in water.
Importantly, both the solid-phase AVS ESG and
interstitial water ESG are no-effect guidelines; that is,
they predict sediments that are acceptable for the
protection of benthic organisms. These ESGs, when
exceeded, do not unequivocally predict sediments that
are unacceptable for the protection of benthic
organisms. The solid-phase AVS guideline avoids the
methodological difficulties of interstitial water sampling
that may lead to an overestimate of exposure and
provides information.on the potential for additional
metal binding. Because the AVS guideline does not
include other metal-binding phases of sediments, the
interstitial guideline is also proposed. The use of both
the AVS and interstitial water guidelines will.improve
estimates of risks of sediment-associated metals. For
example, the absence of significant concentrations of
metal in interstitial water in toxic sediments having
SEMs AVS and in nontoxic sediments having
SEM>AVS demonstrates that metals in these sediments
are unavailable. The (2SEM-AVS)//^. correction,
although not an ESG, can be used to refine the
prediction of sediments where protection of benthic
organisms is acceptable, uncertain, or unacceptable.
ESGs based on the EqP approach are developed
using the latest available scientific data and are suitable
for providing guidance to regulatory agencies because
they are
• Numeric values
• Chemical-specific
• Applicable to most sediments
• Predictive of biological effects
• Protective of benthic organisms
It should be emphasized that these guidelines are
intended to protect benthic organisms from the direct
effects of these six metals in sediments that are
permanently inundated with water, intertidal. or
inundated periodically for durations sufficient to permit
development of benthic assemblages. They do not
apply to occasionally inundated soils containing
terrestrial organisms. These guidelines do not address
the possibility of bioaccumulation and transfer to upper
trophic level organisms, or the synergistic, additive, or
antagonistic effects of other substances. The ESGs
presented in this document represent EPA's best
recommendation of the concentration of mixtures of
cadmium, copper, lead, nickel, silver, and zinc in
sediment that will not adversely affect most benthic
organisms. ESG values may be adjusted to account for
future data or site-specific considerations (U.S. EPA,
2000b).
This document includesIthe theoretical basis and
the supporting data relevant to the derivation of an
ESG for cadmium, copper, lead, nickel, silver, and zinc
and their mixture. An understanding of the "Guidelines
for Deriving Numerical National Water Quality Criteria
for the Protection of Aquatic Organisms and Their,
Uses" (Stephan et al., 1985); Response to Public • .
Comment (U.S. EPA. 1985a); "Ambient Water Quality
Criteria forCadmium" (U.S. EPA. 1985b); "Ambient
1-3
-------
Introduction
Water Quality Catena for Copper" (U.S. EPA. 1985c);
"Ambient Water Quality Criteria—Saltwater Copper
Addendum" (U.S. EPA. 1995c); "Ambient Water Quality
Criteria for Lead" (U.S. EPA. I985d); "Ambient Water
Quality Criteria for Nickel" (U.S. EPA. 1986); "Ambient
Water Quality Catena for Silver" (U.S. EPA. 1980): and
"Ambient Water Quality Criteria for Zinc" (U.S. EPA.
1987b) is necessary in order to understand the
following text, tables, and calculations. Guidance for
the acceptable use of ESG values for metal mixtures is
.contained in the "Implementation Framework for Use of
Equilibnum Partitioning Sediment Guidelines" (.U.S.
EPA.20000.
1.2 Applications of Sediment Guidelines
ESGs are meant to be used with direct toxicity
testing of sediments as a method of evaluation. They
provide a chemical-by-chemical specification of what
sediment concentrations are protective of benthic
aquatic life. The EqP method should be applicable to
nomonic organic chemicals with a KQW above 3.0.
Examples of other chemicals.to which this methodology
applies include endrin, dieldrin, and poiycyclic aromatic
hydrocarbon (PAH) mixtures.
EPA has developed both Tier I and Tier 2 ESGs to
reflect the differing degrees of data availability and
uncertainty. The minimum requirements to derive a Tier
1 ESG include (1) an octanol-water partitioning
coefficient (KQV.) of the chemical, measured with
current experimental techniques, which appears to
remove the large variation in reported values; (2)
derivation of the FCV, which should also be updated to
include the most recent toxicological information; and
(3) sediment toxicity "check" tests to verify EqP
predictions.' Check experiments can be used to verify
the utility of EqP for a particular chemical. As such, the
ESGs derived for nonionic organics, such as dieldrin
and endrin. metal mixtures, and PAH mixtures represent
Tier I ESGs (U.S. EPA, 2000d,e.f). In comparison, the
minimum requirements for a Tier 2 ESG include a Afow
for the chemical (as described above) and the use of
either a FCV or secondary chronic value (SCV). The
performance of sediment toxicity tests is recommended,
but not required for the development of Tier 2 ESGs.
Therefore, in comparison to Tier 1 ESGs, the level of
protection provided by the Tier 2 ESGs would b«
associated with more uncertainty due to the use of the
SCV and absence of sediment toxicity tests. Examples
of Tier 2 ESGs for nonionics are found in U.S. EPA
.(2000g). Information on how EPA recommends ESGs be
applied in specific regulatory programs is described in
the "Implementation Framework for the Use of
Equilibnum Partitioning Sediment Guidelines (ESGs)"
(EPA,2000c). ' -
1.3 Overview
Section 1 provides a brief review of the EqP
methodology as it applies to the individual metals
cadmium, copper, lead, nickel, silver, and zinc and their
mixture. Section 2 reviews published experimental
results that describe the toxicity associated with the
partitioning and bioavailability of these metals in
interstitial water of freshwater and marine sediments.
Section 3 reviews the results of acute and chronic
toxicity tests conducted with spiked and field
sediments that demonstrate that the partitioning and .
bioavailability of metals in sediments can be used to
accurately predict the absence of toxicity of sediment-
associated metals. Section 4 describes the AVS
guideline and interstitial water guideline approaches for
the derivation of the ESG for individual metals and
mixtures of metals. Published WQC values for five of
these six dissolved metals (the silver FCV is not
available) are.summarized for use in calculating IWGUs
as required in the interstitial water ESG approach. The
ESG for metals is then compared with chemical
monitoring data on environmental occurrence of SEM.
AVS, and interstitial metals in sediments from Lake '
Michigan, the Virginian Province from EPA's
Environmental Monitoring and Assessment Program
(EMAPX and the National Oceanic and Atmospheric
Administration (NOAA) National Status and Trends
monitoring program (NST). Section 5 describes
recommended procedures for sampling, handling, and
analysis of metals in sediments and interpretation of
data from the sediment samples that is needed if the
assessments of risks of sediment-associated metals are
,to be appropriately based on the EqP methodology.
Section 6 concludes with EPA's guidelines statement
for a mixture of the metals: cadmium, copper, nickel,
lead, silver, and zinc. The references cited in this
document are listed in Section 7. •
1-4
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 2
Partitioning of Metals in Sediments
2.1 Metal Toxicity in Water-Only and
in Interstitial Water of Sediment
Exposures
The EqP approach t'or establishing sediment
guidelines (i.e.. ESGs) requires that the chemicals be
measured in phases that relate to chemical activity in
sediment.. The information provided in this section
demonstrates that biological effects correlate to metal
activity. Also, it demonstrates that biological response
in sediment exposures is the same as in water-only
exposures when sediment exposure is assessed on the
basis of interstitial water concentrations. This is
fundamental to satisfying the EqP approach for both
metals and nonionic organic chemicals.
A direct method for establishing sediment
guidelines for metals would be to apply the WQC FCV
to measured interstitial water concentrations. The
validity of this approach depends both on the degree
to which the interstitial water concentration represents
free metal activity, and on whether free metal activity
can be accurately measured in surface waters and
water-only toxicity tests used to derive WQC. and in
interstitial water of field sediments and sediments
spiked with metals in the laboratory. For most metals,
free metal activity cannot be directly measured at WQC
concentrations. Therefore, present WQC are not based
on free metal activity; rather, they are based on
dissolved metals. However, many dissolved metals
readily bind'.to dissolved (actually colloidal) organic
carbon (DOC) forming complexes that do not appear to
be bioavailable (Bergman and Dbrward-King,-1997).
Hence, sediment guidelines based on interstitial water
concentrations of metals may be overly protective in
cases where not all dissolved metal is bioavailable.
By implication, this difficulty extends to any
complexing ligand that is present in sufficient quantity.
Decay of sediment organic matter can cause
substantial changes in interstitial water chemistry. In
particular, bicarbonate increases because of sulfate
reduction, which increases the importance of metal-
carbonate complexes and further complicates the
question of the bioavailable metal species (Stumm and.
Morgan. 1996):
Sampling sediment interstitial water for metals is
not a routine procedure. The least invasive technique
employs a diffusion sampler that has cavities covered
with a filter membrane (Hesslein. 1976; Carignan. 1984;
Carignan et al., 1985; Allen et al.. 1993; Bufflap and
Allen, 1995). The sampler is inserted into the sediment
and the concentrations on either side of the membrane
equilibrate. Because the sampler is removed after
equilibration, the concentrations of metals inside the
sampler should be equal to the concentrations of freely-
dissolved metals in the interstitial water. The time
required for equilibration, typically several days.
depends on the size of the filter membrane and the
geometry of the cavity.
An alternative technique for separating interstitial
water is to obtain an undisturbed sediment sample as a
whole sediment or core that can be sliced for vertical
resolution, filter or centrifuge the sample, and then filter
the resultant interstitial water twice. For anaerobic
sediments, this must be done in a nitrogen atmosphere
to prevent precipitation of iron hydroxide, which would
scavenge the metals and yield artificially low dissolved
concentrations of metals (Troup, 1974; Allen et al..
1993).
Although eithertechnique is suitable for research
investigations, they require more than the normally
available sampling capabilities. If solid-phase chemical
measurements were available from which interstitial
water metal activity could be deduced, this would
obviate the need for interstitial water sampling and
analysis, circumvent the need to deal with complexing
ligands, and provide fundamental insight into metal-
binding phases in sediments needed to predict
bioavailability. The recommended procedures for
suitable sampling, handling, and analytical techniques
for interstitial water and sediments are provided in
Section 5 of this document.
2.1.1 Toxicity Correlates to Metal Activity
A substantial number of water-only exposures
indicate that biological effects can be correlated to
divalent metal activity {M1*}. Although other forms of
metal may also be bioavailable (e.g.. MOH*), DOC and
2-1
-------
Parti tiooing
certain other Iigand-complexed tractions of the metal
render it unavailable to organisms. Results from some
of these exposures are summarized below.
Acute toxicity of various concentrations of
cadmium to grass shrimp (Palaemonetes pugio) has
been determined in water containing the complexing
ligand mtnlotnacetic acid (NTA) or chloride (as
salinity), each of which forms cadmium complexes
(SundaetaL 1978). The concentration response
curves as a function of total cadmium are quite different
• at varying concentrations of NTA and chloride (Figure
2-1. A and B). However, if the organism response is
evaluated with respect to measured Cd'* activity, a
single concentration-response relationship results
(Figure 2-1, C and D). Comparable results have been
reported by Anderson and Morel (1978) for the
dinoflagellate Gonyaulax tamarensis exposed to
copper-ethylene diamine tetra-acetic acid (EDTA)
complexes (Figure 2-2. A and C). Likewise, Allen etal. .
(1980) observed that when the concentration of zinc is
held constant and the concentration of the complexing
ligand NTA is varied, growth (cells/mL) ofMicrocystis
aeruginosa decreases as the. addition of NTA is
increased (Figure 2-2B). The authors correlated the
effect to free zinc activity as shown in Figure 2-2D. A
single concentration-response relationship is shown
for the diatom. Thalassiosira pseudonana, and the
• unicellular alga. Monochrysis lutheri, exposed to
copper and the complexing ligand Tris (Sunda and
Guillard, 1976) as well as copper and DOC from natural
river water (Sunda and Lewis, 1978) when exposure
concentration is expressed as metal activity (Figure 2-3,
A, B. C, and D, respectively).
Metal bioavailability, as measured by metal
accumulation into tissues of organisms, has also been
examined (Zamuda and Sunda, 1982). Uptake of copper
by oysters is correlated not to total copper
concentration (Figure 2-4A), but to copper activity
(Figure 2-4B).
The implication to be drawn from these experiments
is that the partitioning model required for establishing a
sediment guideline should predict dissolved metal'
activity in interstitial water, and that the guideline
based on dissolved metal would be conservative. The
following subsection examines the utility of this idea.
100
: 3
75
I- 50
NTA(M)
1 1 If
3 1 10'
« 1 1 NT1 •
5.0 «.0
Total Cadmium (-log Cd,)
100
I to
3 6g
g 40
4.0 7.0
Cadmium Activity (p[Cd*1)
100
50
Total Cadmium (-log CdJ
too
M
M
40
20
D
S«U«tty(%.>
• 1.4
» lit
• 2M
» 2U
"
Cadmium Activity (p(Cd*l)
Figure 2-1. Acute toxicity to grass shrimp (Palaemonetes pugio) of total cadmium (top) and cadmium
activity (bottom) with different concentrations of the complexing ligands NTA (left) and
chloride as salinity (right) (figures from Sunda et al., 1978).
'2-2
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
2.1.2 Toxicity Correlates to Interstitial
Water Concentration
This subsection presents early data that first
indicated the equivalence of interstitial water
concentrations and water-only exposures. Many more
data of this sort are presented in Section 3. Swartz et
al. (1985) tested the acute toxicity of cadmium to the
marine amphipod Rhepoxynius abronius in sediment
and water. An objective of the study was to determine
the contributions of interstitial and particle-bound
cadmium to toxicity. A comparison of the 4-day LC50
value of cadmium in interstitial water (1.42 mg/L) with
the 4-day LC50 value of cadmium in water without
sediment (1.61 mg/L) indicated no significant difference
between the two (Figure 2-5). The LC50 represents the
100
90
80
70
60
50
40
30
20
10
0
With EDTA
Without EDTA
56789
Total Copper (-log (CuT))
10
7.
10
6.
104H
10
O AAPw I.I MFs EDTA
via T.M.
IO-7 M Zn
10-" M NTA
10-"
u
Davj
- I 0
-2.S
-40
-6.0
- I.Ox 10-6
-S.Ox 10-6
- 1.0 \ 10-5
MNTA
M NTA
M NTA
M NTA
M NTA
M NTA
16 20
too
.90
80
70
60
SO
40
30
20
10
0
10
2.
~
in
- 10
•5
U
v
10 11 12 13 14
Copper Activity (pCu)
,5.
103
XO
a A
1C'n nH 7Q
Q -EDTA 13.46
A -VTA 7.64
x -ODS -631
0 -CMOS 5.:5
• - Builder M 433
* • Control
1.0 2.0 3.0 4.0 5.0
Free Zinc (M/L x 10')
Figure 2-2. Acute toxicity of total copper'(A) and copper activity (C) to the dinoflagellate Gonyaulas
tamarensis with and without the completing ligand EDTA (figures from Anderson and
Morel, 1978).. Toxicity of zinc to Microcystis aeruginota showing growth of cells/mL versus
time with different levels of the complexing ligands EDTA and NTA (B) and number of cells
at 5 days as a function of free zinc concentration (D) (figures from Allen et al., .1980).
2-3
-------
Partitioning
* s
3 '"
A JILL; ;
"rt < : < 1 ^ no m (
t j; IBM ro JH ••
A . J • B '.0 mM mi
2.1)0
1.00
3 4 1 5.6
Total Copper (-log CuT)
-7.4
0.00
B
* 10". River WJICT
* JW. River W«ter
A 40% River WIKT
4.00 5.00 6.00
ToCaJ Copper (-log Cur)
7.00
t s
» '5
A
•P-
2.00
1.00
9 10
Copper Activity (pCu)
o.oo
• tO".Rjver W»ter
• JOV, River Witer
A 10", River Water
6.00
7.00 8.00
Copper Activity (pCu)
9.00
Figure 2-3. Specific growth rates of a diatom (Thalassiosira pseudonana) (left) and a unicellular algae (Monochrysis
lutheri) (right) versus total copper (top) and copper activity (bottom) for a range of concentrations of
the complexing ligands Tris (left; from Sunda and Guillard, 1976) and natural DOC in river water
(right; from Sunda and Lewis, 1978).
chemical concentration estimated to cause lethality to
50% of the test organisms within a specified time
period.
Experiments were performed to determine the role
of AVS in cadmium-spiklcl sediments using the
amphipods Ampelisca abdita and Rhepoxynius
hudsoni (Di Toro et al., 1990).- Three sediments were
used: a Long Island Sound sediment with high AVS, a
Ninigret Pond sediment with low AVS concentration.
and a 50/50 mixture of the two sediments. Figure 2-6
presents a comparison of the observed mortality in the
three sediments with the interstitial water cadmium
activity measured with a specific ion electrode. Four- .
day water-only and 10-day sediment toxicity tests were
performed. The water-only response data for A. abdita
and R. Hudsoni are included for comparison although
these data represent a shorter duration exposure.
These experiments also demonstrate the equivalence of
organism response to metal concentrations in •
interstitial water and in water-only exposures.
An elegant experimental design was employed by
Kemp and Swartz (1986) to examine the relative acute
toxicity of particle-bound and dissolved interstitial
cadmium. They circulated water of the same cadmium
concentration through different sediments. This
resulted in different bulk sediment concentrations, but
the same interstitial water concentrations. They found
no statistically significant difference in organism
response for the different sediments. Because the
interstitial water concentrations were the same in each
treatment, that is. the circulating water concentrations
established the interstitial water concentrations, these
experiments confirmed the hypothesis of equal
response to concentrations in water-only and
interstitial water.
A series of 10-day toxicity tests using the
amphipod Hyalella azieca was performed to evaluate
bioavailability of copper in sediments from two sites
highly contaminated with this metal: Steilacoom Lake.
WA, and Keweenaw Watershed, MI (Ankiey et al..
2-4
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
s
_o
1?
c >>
1 "5*
, ' —
£ M
|3
•
•**
£
u
-5?
_a
QJ
-D ^
C *^
.2 n
11
3 2
w "5
u J
u
Q.
a.
9
~
Z
I5U
100
50
A
O A
0 *
O
-
*
0
oih o • - * '
0.01 0.1 1.0 10.0
Total Copper Concentration (umol)
200
180
160
140
120
100
80
60
40
20
0
- B Total NTA (urnol)
Winter
010.0
-
Summer
1 -. ff-,
v A i n
1 I " 8-1
T A
"*
4 —
'
1 .
t
$ A *i
av— ^L
T ^ '.
1 1 1
8 9 10 11
Copper Activity (pCu)
Figure 2-4. Copper accumulation in oysters (Crassostrea virginica) versus.total copper (A) and copper activity (B)
with different levels of the complexing ligand NTA (figures from Zamuda and Sunda, 1982).
2-5
-------
Partitioning
20
^ IS
6
Z
3
10
I
8-
•c
Seawater
Interstitial
Water
\
\
I
1 2 3 4
Dissolved Cadmium Concentration (mg/L)
Figure 2-5. Mean survival of the amphipod Rhepoxynius abronius versus dissolved cadmium concentration for 4-
day toxicity tests- in seawater (symbols) and 0- and 4-day tests (bars) in interstitial water (figure from
Swartz et ah, 1985).
1993). A water-only, 10-day copper toxicity test also
was conducted with the same organism. The mortality
resulting from the water-only test was strikingly similar
to that from the Keweenaw sediment tests when related
to interstitial water (Figure 2-7). The LC50 values show
strong agreement for the water-only (31 Mg/L) and the
Keweenaw sediment test (28 Mg/L) using the average of
day 0 and day 10 interstitial water concentrations.
Steilacoom Lake 10-day interstitial water concentrations
were less than the 7 Mg/L detection limit and were
consistent with the observed lack of toxicity to
H. azteca.
The data presented in this subsection, and the data
in Section 3, demonstrate that in water-only exposures.
metal, activity and concentration can be used to predict
toxicity. The results of the four experiments above
demonstrate that mortality data from water-only
exposures can be used to predict sediment toxicity
using interstitial water concentrations. Therefore, the
metal activity or dissolved concentration in interstitial
water would, 6e an important component of a
partitioning model needed to establish sediment
guidelines. To complete the partitioning model, one
would need to identify the solid metal-binding
2-6
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
2
~
i
100
80
£ 60
40
20
• LI Sound
O Mixture
O Ninigret Pond
Water-Only Exposure
A Rkepoxyniiu
& Ampeiisca
i ' ' * ' »
-5.00
•3.00
•1.00
1.00
3.00
Log,,Cd" Activity (mg/L)
Figure 2-6. Mortality versus interstitial water cadmium activity for sediments from Long Island Sound, Ninigret
Pond, and a mixture of these two sediments. Water-only exposure data are from separate experiments
with both Ampeiisca abdita and Rhepoxynius hudsoni. The line is a joint fit to both water-only data
sets (figure from Di Toro et ah, 1990).
LUU
80
S? 60
^^
>.
•
J 40
20
1
' _ • 0» •
O O 0
O Water-Only Exposure • •
• Sediment Exposure *
- O
•
—
•
•
•
• o
10 100 10(
Copper Gig/L)
10
Figure 2-7. Toxicity of copper to Hyalella azteca versus copper concentrations in a water-only exposure (c) and
interstitial water copper concentrations in sediment exposures (•) using Keweenaw Watershed
sediments (figure from Ankley et al., 1993).
2-7
-------
Partitioning
phase(s). The following subsection presents data that
identifies solid-phase sulftdes as the important metal-
binding phase.
2.2 Solid-Phase Sulfide as the Important
Binding Component
Modeling metal sorption to oxides in laboratory
systems is well developed, and detailed models are
available for cation and anion sorption (see Stumm
[1987] and Dzombak and Morel [ 1990] for summaries).
The models consider surface compiexation reactions as
well as electrical interactions by means of models of the
double layer. Models for natural soil and sediment
panicles are less well developed. However, studies
suggest that the models available for cation and anion
sorption can be applied to soil systems (Allen et al.,
1980; Barrow and Ellis. 1986a.b.c; Spositoetal., 1988).
Because the ability to predict partition coefficients is
required if interstitial water metal concentrations are to
be inferred from the total concentration, some practical
model is required. This subsection presents the state
of the science in theoretical development of metal
partitioning behavior in sediments.
2.2.1 Metal Sorption Phases
The initial difficulty selecting an applicable
sorption model is that available models are complex and
many of the parameter estimates may be specific to
individual soils or sediments. However, the success of
nonionic chemical sorption models based on organic
carbon suggests that some model of intermediate
complexity based on an identification of the dominant
sorption phases may be more generally applicable.
A development in this direction has already been
presented (Jenneetal., 1986;DiToroetal.. 1987). The
basic idea was that instead of considering only one
sorption phase, as is assumed for nonionic
hydrophobic chemical sorption, multiple sorption
phases must be considered. The conventional view of
metal speciation in aerobic soils and sediments is that
metals are associated with the exchangeable, carbonate
and iron (Fe) and manganese (Mn) oxide forms, as well
as organic matter, stable metal sulfides, and a residual
phase. In pxic soils and freshwater sediments, sorption
phases have been identified as paniculate organic
carbon (POC) and the oxides of Fe and Mn (Jenne,
1968.1977; Oakley etal., 1980; Luoma and Bryan. 1981).
These phases are important because they have a large
sorptive capacity. Furthermore, they appear as
coatings on the particles and occlude the other mineral
components. It was thought that they provided the
primary sites for sorption of metals. These ideas have
been applied to metal speciation in sediments.
However, they ignore the critical importance of metal
sulfide interactions, which dominate speciation in the
anaerobic layers of the sediment.
2.2.2 Titration Experiments
The importance of sulfide in the control of metal
concentrations in the interstitial water of marine •
sediments is well documented (Boulegue etal., 1982;
Emerson etal., 1983; Davies-Colley etal., 1985; Morse
et al.. 1987). Metal sulfides are very insoluble, and the
equilibrium interstitial water metal concentrations in the
presence of sulfides are small. If the interstitial water
sulfide concentration, S-', in sediments is large, then
the addition of metal. M-*, to the sediment would
precipitate metal sulfide (MS) following the reaction
:' - MS(s)
(2-1)
This appeared to be happening during a spiked
cadmium sediment toxicity test (Di Toro et al.. 1990)
because a visible bright yellow cadmium sulfide
precipitate formed as cadmium was added to the
sediment. However, interstitial water sulfide activity,
{S2"}, measured with a sulfide electrode unexpectedly
indicated that there was insufficient dissolved sulfide
present in the unspiked sediment.
The lack of a significant quantity of dissolved
sulfide in the interstitial water and the evident
formation of solid-phase cadmium sulfide suggested
the following possibility. The majority of the sulfide in
sediments is in the form of solid-phase iron sulfides.
Perhaps the source of the sulfide is 'from the solid-
phase sulfide initially present. As cadmium is added to ~
the sediment, this causes the solid-phase iron sulfide to
dissolve, releasing sulfide that is available for formation
of cadmium sulfide. The reaction is
CdJ* + FeS(s) - CdS(s) + Fe:>
Cadmium titrations with amorphous FeS and with
sediments were performed to examine this possibility.
2.2.2.1 Amorphous FeS
*
A direct test of the extent to which this reaction
takes place was performed (Di Toro et al.. 1990). A
2-8
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
quantity of freshly precipitated iron sulfide was titrated
by adding dissolved cadmium. The resulting aqueous
cadmium activity, measured with the cadmium electrode,
versus the ratio of cadmium added [CdJ to the amount
of FeS initially present (FeS(s)]l is shown in Figure 2-8.
The plot of dissolved cadmium versus cadmium added
illustrates the increase in dissolved cadmium that
occurs near [CdJ/[FeS(s)], = 1. It is interesting to note
that these displacement reactions among metal sulfides
have been observed by other investigators (Phillips
and Kraus. 1965). The reaction was also postulated by
Pankow (1979) to explain an experimental result
involving copper and synthetic FeS.
These experiments plainly demonstrate that solid-
phase amorphous iron sulfide can be readily displaced
by adding cadmium. As a consequence, the source of
available sulfide must be taken into account when
evaluating the relationship between solid-phase and
aqueous-phase cadmium in sediments.
0.
0.0
0.5
1.0
1.5
2.0
Cadmium Added (^mol Cd/Mmol FeS)
Figure 2-8. Cadmium durations of amorphous FeS. The x-axis is the amount of cadmium added normalized fay
FeS initially present. The y-axis is total dissolved cadmium. The lines connecting the data points are
an aid to visualizing the data. The different symbols represent replicate experiments (figure.from Di
Toro ef ah, 1990).
2-9
-------
Parti tiooing
A direct confirmation that the removal of cadmium
was through the displacement of iron sulfide is shown
m Figure 2-9. The supernatant from a titration of FeS
by Cd:' was analyzed for both iron and cadmium. The
solid lines are the theoretical expectations based on the
stoichiometry of the reaction.
2.2.2.2 Sediments
A similar titration procedure has been used to
evaluate the behavior of sediments taken from four
different marine environments: sediments from Black
Rock Harbor and the Hudson River, and the sediments
from Long Island Sound and Ninigret Pond used in the
toxicity tests (Di Toro et ah, 1990). The binding
.capacity for cadmium is estimated by extrapolating a
straight line fit to the dissolved cadmium data. The
equation is
[ICd(aq)] = max' {m([CdJ - [Cd,])}
(2-3)
where [ZCd(aq)] is the total dissolved cadmium. [Cd J
is the cadmium added, [CdB] is the bound cadmium, and
m is the slope of the straight line. The different
sediments exhibit quite different binding capacities
II
e *"
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
for cadmium, listed in Table 2-1. ranging from
approximately 1 ^mol/g to more than lOO.umol/g. The
question as to whether this binding capacity is
explained by the solid-phase sulfide present in the
samples is addressed in subsequent sections of this
document.
2.2.3 Correlation to Sediment AVS
The majority of sulfide in sediments is in the form
of iron monosulfides l.mackinawite and greigite) and
• iron bisulfide (pyrite). of which the former is the most
reactive. These sediment sulfides can be classified into
three broad classes that reflect the techniques used for
quantification (Berner, 1967; Goldhauber and Kaplan.
1974; Morse et aL 1987). The most labile fraction. AVS.
is associated with the more soluble iron monosulfides.
The more resistant sulfide mineral phase, iron pyrite. is
not.soluble in the cold acid extraction used to measure
AVS. Neither is the third compartment, organic sulfide.
which is associated with the organic matter in
sediments ganders etal.. 1983).
The possibility that acid, volatile sulfide is a direct
measure of the solid-phase sulfide that reacts with
cadmium is examined in Table 2-1. which lists the
sediment-binding capacity for cadmium and the
measured AVS for each sediment, and in Figure 2-10,
which indicates the initial AVS concentration. The
sediment cadmium-binding capacity appears to be
somewhat less than the initial AVS for the sediments
tested. However, a comparison between the initial AVS
of the sediments and that remaining after the cadmium
titration is completed suggests that some AVS is lost
during the titration experiment (Table 2-1). In any case.
the covariation of sediment-binding capacity and AVS
is clear. This suggests that measurement of AVS is the
proper quantification of the solid-phase sulfides that
can be dissolved by the addition of ionic cadmium. '
The chemical basis for this is examined below.
2.2.4 Solubility Relationships and
Displacement Reactions
Iron monosulfide. FeS(s), is in equilibrium with
aqueous-phase sulfide and iron via the reaction
FeS(s)-Fe-*+S2' (24)
If cadmium is added to the aqueous phase, the result, is
Cd2* + FeS(s) - Cd2* + Fe2* -t- S2' (2-5)
As the cadmium concentration increases, [Cd:*] [S:']
williexceed the solubility product of cadmium-sulfide
and CdS(s) will start to form. Since the cadmium sulfide
is more insoluble than iron monosulfide, FeS(s) should
start to dissolve in response to the lowered sulfide
concentration in the interstitial water. The overall
reaction is
Cd2* + FeS(s) - CdS(s) + Fe2*
(2-6)
The iron in FeS(s) is displaced by cadmium to form
soluble iron and solid cadmium sulfide, CdS( s). The
consequence of this replacement reaction can be seen
using the analysis of the M(II)-Fe(II)-S(-II) system with
both MS(s) and FeS(s) presented in Di Toro et al.
(1992). M2* represents any divalent metal that forms a
sulfide that is more insoluble than FeS. If the added
Table 2-1. Cadmium binding capacity and AVS of sediments
Sediment
Black Rock Harbor
Hudson River
LI Sound*1
Mixture11'
Ninigret Pond
Initial AVS*
(>tmol/g)
175(41)
12.6(2.80)
' 15.9(3.30)
5.45 (— )
2.34(0.73)
Final AVSb
(Mtnol/g)
—
_
13.9(6.43)
3.23(1.18)
0.28(0.12)
Cd Binding Capacity" .
(Mmol/gj
114(12)
8.58 (2.95)
4.57 (2.52)
—
1.12(0.42)
'Average (SO) AVS of repeated measurements of the stock.
'Average (SD) AVS after the sediment toxicity experiment.
.•From Equation 2-3.
•JFrom original cadmium experiment.
'SO/SO mixture of LI Sound and Ninigret Pond.
Source: Di Toro et al.. 1990.
2-11
-------
Partitioning
metal. [M]A, is less than the AVS present in the
sediment then the ratio of metal activity to total metal in
the sediment-interstitial water system is less than the
ratio of the MS to FeS solubility product constant
(2-7)
This general result is independent of the details of
the interstitial water chemistry. In particular, it is
independent of the Fe:* activity. Of course, the actual
value of the ratio (M:*}/[M]^ depends on aqueous
speciation, as indicated by Equation 2-6. However, the
ratio is still less than the ratio of the sulfide solubility
products.
This is an important finding because the data
presented in Section 2.1.1 indicate that toxicity is
related to metal activity, {M:*}. This inequality
guarantees that the metal activity, in contrast to the
total dissolved metal concentration, is regulated by the
iron sulfide-metal sulfide system.
The metal sulfide solubility products and the ratios
are listed in Table 2-2. For example, the ratio of
cadmium activity to total cadmium is less than 10"'oa*.
For nickel, the ratio is less than 10'3". By inference,
this reduction in metal activity will occur for any other
metal that forms a sulfide that is significantly more •
insoluble than iron monosulfide. The ratios for the
other metals in.Table 2-2 (Cu. Pb, Ag, and Zn) indicate
that metal activity for these metals will be very small in
the presence of excess AVS.
2.2.5 Application to Mixtures of Metals
A conjecture based on the sulfide solubility
products for the metals listed in Table 2-2 is that the
sum of the molar concentrations of metals should be
compared with AVS. Because all these metals have
lower sulfide solubility parameters than FeS, they
would all exist as metal sulfides if their molar sum (and
using [Ag]/2 because it is monovalent) is less than the
AVS. For this case
(2-8)
no metal toxicity would be expected, where (M^, is the
total cold acid extractable i metal molar concentration
in the sediment (divided by 2 for silver). On the other
hand, if their molar sum is greater than the AVS
AVS
AVS
AVS
1.0
-1
"5k
I
a
BR Harbor
LI Sound
Hudson River
Ninigret Pond
0.6
0.4 -
0.2
0.0
,0.1
1.0 10.0
Cadmium Added (/xmol Cd/g dry wt)
100.0
1000.0
Figure 2-10. Cadmium titration of sediments from Black Rock Harbor, Long Island Sound, Hudson River, and
Ninigret Pond. Cadmium added per unit dry weight of sediment versus dissolved cadmium. Arrows
are the measured AVS concentrations for the four sediments (figure from Di Toro et al., 1990).
2-12
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Table 2-2. Metal sulfide solubility products and ratios
Metal Sulfide
FeS
NiS
ZnS
CdS
PbS
CuS
Ag-S
Logi0%/
-3.64
-9.23
-9.64
-14.10.
-14.67
-22.19
-36.14
i
LOgioAjp
-22.39
-27.98
-28.39
-32.85
-33.42
-40.94
-54.71 '.
Lo?.<0(K^KM)
—
-5.59
-6.00
-10.46
-11.03
-18.55
-3132
•'Solubility products. K , for the reaction M2* «• US' - MSi's) + H* for FeS (mackinawiie). NiS (rrullerue). ind CdS
(.greenockite) from Emerson et al. (1983). Solubility products for ZnS (wurtziie),- PbS (galena). CuS (covellite). ind Ag,S
(acanthite) and pK, = 18.57 for the reaction HS' - H* +• Si- from Schoonen and Barnes (1988).
bK.p for ihe reaction M:* * S:' - MS(s) is computed from log Kip, and pK,.
concentration, then a portion of the metals with the
largest sulfide solubility parameters would exist as free
metal and potentially cause toxicity. For this case the
following would be true
(2-9)
These two equations are precisely the formulas that
could be employed to determine the extent of metal
toxicity in sediments assuming additive behavior and
neglecting the effect of partitioning to other sediment
phases. Whether the normalized sum is less than or
greater than 1.0 discriminates between nontoxic and
potentially toxic sediments. The additivity does not
come from the nature of the mechanism that causes
toxicity. Rather, it results from the equal ability of the
metals to form metal sulfides with the same
stoichiometric ratio of M and S (except silver).
The appropriate quantity of metals to use in the
metals and AVS comparison is referred to as SEM, that
is. the metal extracted with the cold acid used in the
AVS procedure. This is the appropriate quantity to use
because some metals form sulfides that are not labile in
the AVS extraction (e.g., nickel, copper). If a more
rigorous extraction were used to increase the fraction of
metal extracted that did not also capture the additional
sulfide extracted, then'the sulfide associated with the
additional metal release would not be quantified. This
would result in an erroneously high metal value relative
toAVS(DiToroetal., 1992).
The above discussion is predicated on the
assumption that all the metal sulfides behave similarly
to cadmium sulfide. Furthermore, it has been assumed
that only acid-soluble metals are reactive enough to
affect the free metal activity. That is, the proper metal
concentration to be used is the SEM. Both of these.
hypotheses were tested directly with benthic organisms.
using sediment toxicity tests. Results of these
sediment-spiking experiments with cadmium, copper.
lead, nickel, silver, zinc, and a mixture of these metals
are presented in Section 3.
2-13
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 3
Toxicity of Metals in Sediments
3.1 General Information
This section summarizes data from acute and
chronic toxicity tests that demonstrate that absence of
sediment toxicity caused by metals can be predicted by
(a) the use of interstitial water concentrations of metals
or (b'i comparison of molar concentrations of AVS and
SEM. Furthermore, they demonstrate that use of
(,ZSEM-AVS)//OC reduces the variability associated
with prediction of when sediments will be toxic. The
ability to predict toxicity of metals in sediments,
through a fundamental understanding of chemical
bioavailability, is demonstrated using results oftoxicity
tests with benthic organisms in spiked or field
sediments. A wide variety of individual benthic
species having different habitat requirements have
been tested in 10-day experiments in spiked and field
sediments, including the following: anoligochaete
(Lumbriculus variegatus), polychaetes (Capitella
capitata and Neanthes arenaceodentata), amphipods
(A. abdita. R. hudsoni. Leptocheirus plumulosus, and
Hyalella ayeca), a harpacticoid copepod (Amphiascus
tenuiremis), a midge (Chironomus tentans), and a
gastropod (Heiisoma sp.). In addition, the approach
was tested in life-cycle tests with L plumulosus and C.
tentans. Many other benthic species were tested in
freshwater and saltwater benthic colonization studies.
3.1.1 Terminology».
Early studies on use of AVS in prediction of
biological effects (e.g., Di Toro et al.. 1990) involved
the ratio of SEM to AVS. expressed as SEM/AVS. The
ratio appeared more useful in the early laboratory tests
because it caused concentration-response data from
spiking experiments with different sediments to fall on
the same line (Di Toro et al., 1990,1992; Casas and
Crecelius. l994;Peschetal., 1995; Berry etal., 1996).
Later studies, however, showed several advantages to
the use of the difference, expressed as SEM-AVS
(Hansenetal., 1996a). The two expressions—
SEM/AVS <> 1 and SEM-AVS a 0—are functionally
equivalent. Both indicate an excess of AVS over SEM.
The advantages to using SEM-AVS are that it does not
get very large when AVS is very low (as the ratio does).
and that it can be used to develop partitioning
relationships that include other phases, such as total
organic carbon (TOC) (see Section 3.4; see also the
discussion in Section 3.2.5). For these reasons, the use
of the SEM-AVS difference' is the recommended method.
and it will be used throughout the rest of this document
except in the discussion of the historical development
of AVS theory that follows. In the ensuing discussion,
SEM/AVS ratios are presented because they were
originally presented in this form.
3.2 Predicting Metal Toxicity:
Short-Term Studies
3.2.1 Spiked Sediments: Individual
Experiments
A key to understanding the bioavailability of
sediment-associated contaminants was provided by
Adams et al. (1985), who observed that the effects of
kepone. a nonionic organic pesticide, were similar
across sediments when toxicity was related to
interstitial water concentrations. Swartz et al. (1985)
and Kemp and Swartz (1986) first observed that metal
concentrations in interstitial waters of different
sediments were correlated with observed biological
effects. .However, as opposed to the situation for
nonionic organic chemicals-and organic carbon (see Di
Toro et al.. 1991), the sediment-partitioning phases that
controlled interstitial water concentrations of metals
and metal-induced sediment toxicity were initially not
apparent.
Di Toro et al. (1990) first investigated the
significance of sulfide partitioning in controlling metal
bioavailability and metal-induced toxicity in marine
sediments spiked with cadmium. In these experiments.
the operational definition of Cornwell and Morse (1987)
was used to identify that fraction of amorphous sulfide,
or AVS, available to interact with cadmium in the
sediments. Specifically, AVS was defined as the sulfide
liberated from wet sediment when treated with cold IN
HC1 acid. Di Toro et al. (1990) found that, when
expressed on a dry weight basis, the toxicity of
cadmium in sediments in 10-day tests with the
amphipods R. hudsoni or A. abdita was sediment
3-1
-------
Toxicity of. Metals in Sediments
specific (Figure 3-1 A; from Di.Toro et al.. 1990).
Toxicity increased with increasing cadmium
concentration, but the concentration-response
relationships were different for each sediment. Thus, it
would not be possible to predict whether a particular
sediment would be toxic or not. If the cadmium
concentration is expressed on an interstitial water basis
(Figure 3-IB), however, concentration response is not
sediment specific. Similar results are observed when
cadmium concentration is expressed as SEM/AVS
(Figure 3- 1C). Note that when the ratio of umol
Cd/.umol AVS was less than 1.0. the sediments were not
toxic, and when the ratio was greater than 1.0. the
sediments became increasingly toxict Studies by
Carlson etal. (1991) with cadmium-spiked freshwater .
sediments yielded similar results: when there was more
AVS than total cadmium, significant toxicity was not
observed in 10-day tests with an oligochaete
(L variegaius) or snail (Helisoma sp). Di Toro et al.
(1992). in their studies with nickel-spiked sediments
using A. abdita and field sediments contaminated with
cadmium and nickel using the freshwater amphipod
H. axeca. provided further support to the importance
of AVS in controlling metal bioavailability in sediments.
These studies suggested that it may be feasible to
derive an ESG for mixtures of metals by direct
comparison of molar AVS concentrations to the molar
sum of the concentrations of cationic metals
(specifically, cadmium, copper, lead, nickel, and zinc)
extracted with the AVS (i.e.. ZSEM). They observed
that expression of metals concentrations based on the
sum of SEM concentrations is required because a
significant amount of nickel sulfide is not completely
soluble in the AVS extraction. Hence. AVS must be
used as the measure of reactive sulfide and the sum of
SEM as the measure of total reactive metal.
. Casas and Crecelius (1994) further explored the
relationship of SEM and AVS, 'interstitial water
concentrations, and toxicity by conducting 10-day
toxicity tests with the marine polychaete C. capitaia
exposed to sediments spiked with zinc, lead, and
copper. As was true in earlier studies, elevated
interstitial water metal concentrations were observed
only when SEM concentrations exceeded those of AVS.
Sediments were not toxic when SEM concentrations
were less than AVS and when the concentrations in
interstitial water were less than the water-only LC50
values. Green et al. (1993) reported results of another
spiking experiment supporting this general EqP
approach to deriving an ESG for metals. In their study,
metal-sulfide partitioning was not directly quantified,
but it was found that toxicity of cadmium-spiked marine
sediments to the meiobenthic copepod -4. tenuiremis
was predictable based on interstitial water, but not
sediment dry weight cadmium concentrations. Further
spiking experiments by Pesch et al. (1995) demonstrated
that 10-day survival of the marine polychaete
/V. arenceodentata was comparable to controls in
cadmium- or nickel-spiked sediments with more AVS
than SEM.
Berry et al. (1996) described experiments in which
A. abdita were exposed for 10 days to two or three
sediments spiked either singly, or in combination, with '
cadmium, copper, lead, nickel, and zinc. As in previous
studies, significant toxicity to the amphipod did not
occur when AVS concentrations exceeded those of
SEM. They compared observed mortality with
interstitial water metal concentrations expressed as
interstitial water toxic units (IWTUs)
IWTU = [MJ/LC50
(3-D
where [Md] is the dissolved metal concentration in the
interstitial water, and the LC50 is the concentration of
the metal causing 50% mortality of the test species in a
water-only test. If interstitial water exposure in a
sediment test is indeed equivalent to that in a water-
only test, then 1.0 IWTU should result in 50% mortality
of the test animals. Berry et al. (1996) reported that
significant (>24%) mortality of the saltwater amphipod
occurred in only 3.0% of sediments with less than 0.5
IWTU, whereas samples with greater than 0.5 IWTUs .
were toxic 94.4% of the time. Berry et al. (1996) also
made an important observation relative to interstitial
water metal chemistry in their mixed-metals test;
chemical equilibrium calculations suggest that the
relative affinity of metals for AVS should be silver>
copper>tead>cadmium>zinc>nickel (Emerson et al.,
1983; Di Toro et al., 1992); hence, the appearance of the
metals in interstitial water as AVS is exhausted should
occur in an inverse order. For example, zinc would
replace nickel in a monosulfide complex and nickel
would be liberated to the interstitial water, and so on.
Berry et al. (1996) observed this trend in sediments
spiked with cadmium, copper, nickel, and zinc (Figure
3-2). Furthermore, an increase in the concentration of a
metal in a sediment with a low sulfide solubility product
constant (Af,J theoretically would displace a '
previously unavailable and nontoxic metal with a higher
K , making tbat metal available to bind to other
sediment phases or enter interstitial water to become
toxic. Berry et al. (1999) exposed the saltwater
amphipod A. abdita to sediments spiked with silver.
When AVS was detected in the sediments, they were
not toxic and interstitial water contained no detectable
3-2
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
u
o
100
80
60
40
20
A
• LI Sound
• Mixture
O Nigret Pond
10 100 1000 10000
Sediment Cadmium Og Cd/g dry wt)
100000
o
100
80
60
40
20
B
— Water Only
Exposure
— Ampelisca
& Rhepoxynius
MAX-
75%-
50%-
25%-
MIN-
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000
Cadmium Activity (mg Cd'TL)
100
^ 80
>> 60
1 <°
S
20
0
0.
. c F *
• LI Sound • 4
- • Mixture /
O Nigret Pond /•
*.*»** . .%...,
01 0.10 1 10 10
Sediment Cadmium (/zmol Cd/^m AVS)
0
',
Figure 3-1. Percentage mortality of amphipods (Ampelisca abdita and Rhepoxynius hudsoni) exposed to sediments
from Long Island Sound, Ninigret Pond, and a mixture of these two sediments as a function of the
sum of the concentrations of metals in sediments Expressed as: (A) dry weight, (B) interstitial water
cadmium activity, and (C) the sediment cadmium/AVS ratio (figures from Di Toro et al., 1990).
3-3
-------
Toxicity of Metals in Sediments
3
C
3
ii
I
10000-
1000-
100-
10-
1-
o.i-
0.01-
Log g,p
Ni -27.98
Zn -28.39
Cd -32.85
u -40.94
Cu—»
Zn—*
0.01
0.1
10
100
1000
SEM/AVS
10000
o
u
i
"5
S
1000-
100-
10-
1-
0.1-
0.01.
B
0.01
0.1
10
100
1000
SEM/AVS
Figure 3-2. Concentrations of individual metals in interstitial water of sediments from Long Island Sound '(A) and
Ninigret Pond (B) in the mixed metals experiment as a function of SEM/AVS ratio. Concentrations
below the interstitial water detection limits, indicated by arrows, are platted at one-half the detection
limit. Ksr is the sulfide solubility product constant (figures from Berry et al., 1996).
3-4
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
silver. For sediments that contain no detectable AVS.
any SEM silver that is detected is dissolved interstitial
silver, because silver sulfide and silver chloride
precipitate are not extracted using the standard AVS
procedure.
3.2.2 Spiked Sediments: All Experimental
Results Summarized
This summary includes data from amphipods
•exposed in 10-day toxicity tests to saltwater sediments
spiked with cadmium, copper, lead, nickel, silver, or zinc
and their mixtures (Di Toro et al.. 1990: Berry et al.. 1996,
1999): polychaetes exposed to sediments spiked with
cadmium, copper, lead, nickel, or zinc (Casas and
Crecelius. 1994; Pesch et al.. 1995); copepods exposed
to sediments spiked with cadmium (Green et al., 1993:
measured interstitial cadmium but not-AVS); and
freshwater tests using oligochaetes and snails exposed
to sediments spiked with cadmium (Carlson etal.. 1991).
Seven species (freshwater and saltwater) and sediments
from seven different locations were described. AVS
concentrations ranged from 1.9 to 65.7 ^tmol/g dry
weight, and TOC ranged from 0.15% to 10.6% in these
sediments.
Overall, the results of these experiments
demonstrate that predictions of the toxicity of
sediments spiked with metals using the total metal
concentration on a dry weight basis are not based on
scientific theories of bioavailability and will have
considerable error (Figures 3-3 A and 3-4A). Sediments
having s24% mortality are considered nontoxic as
defined by Berry et al. (1996), which is indicated by the
horizontal line in Figure 3-3. Furthermore, the
concentration range where it is 90% certain that the
sediment may be either toxic or nontoxic, shown as
dashed lines in Figure 3-3, is almost two orders of
magnitude for dry weight metals, a little over an order
of magnitude for IWTUs, and only a half order of
magnitude for SEM/AVS (see Section 3.4 for a
description of the derivation of the uncertainty limits).
The uncertain range for dry weight metals is
approximately equal to the sum o'f the uncertainty range
for SEM/AVS plus the range in the AVS concentrations
of the spiked sediments in the database. If sediments
with a lower AVS concentration had been tested,
effects would have occurred at a lower dry weight
concentration, and if sediments with lower or higher
AVS concentrations had been.tested, the uncertainty
range would increase. Importantly, the uncertainty
range for IWTUs or SEM/AVS would likely not be
altered.
Even given the above, it is visually tempting to
select a cutoff at a dry weight concentration of 1.0
umol/g to indicate the separation of sediments that are
toxic or nontoxic. This would be inappropriate because
toxicity of metals in sediments when concentrations are
expressed as dry weights have been shown to be
sediment specific (Figure 3-1 A). Also, had sediments
with lower or higher AVS concentrations been tested.
the cutoff would'have been at lower or higher dry
weight concentrations. However, to further
demonstrate the risks of establishing a dry weight
cutoff, the data from the 184 spiked sediments in Figure
3-3 were re-analyzed. A visually based cutoff of 1.0
yumol/g dry weight, and theoretically based cutoffs of
0.5 IWTU and 1.0 SEM/AVS were selected. Sediment
concentrations were numerically ordered. Those with
concentrations less than the cutoffs were divided into
three groups containing approximately the same
number of sediments (15, 22, or 25 sediments per group
for dry weight metal concentrations, IWTUs. and SEM/
AVS, respectively). Similarly, sediments containing
greater concentrations were divided into six groups (21.
16. or 14 sediments per group for dry weight metal
concentrations. IWTUs, and SEM/AVS. respectively).
The percentages of nontoxic (s 24% mortality) and toxic
(>24% mortality) sediments in each group are plotted in
a stacked bar plot (Figure 3-4). Not surprisingly,
because the distribution was visually selected, most
sediments haying less than 1.0 ^mol/g dry weight metal
were not toxic. The same was true for the
lexicologically selected cutoffs of 0.5 IWTUs and SEM/
AVS ratios of 1.0. The advantage of using IWTUs and
SEM/AVS becomes more clear when the sediments
above the cutoffs are considered. For dry weight metal
concentrations, more of the sediments in the first four
sediment groups (up to 26.8 Mmol/g dry weight) were
nontoxic than were toxic. It was only in the two
sediment groups that contained the highest
concentrations, >27.6 ^mol/g dry weight, that toxic •
sediments predominated after the first two sediment
groups. In contrast, toxic sediments predominated in
only the first two sediment groups above the IWTU
cutoff and after the first sediment group above the
SEM/AVS ratio cutoff.
In some cases, the dry weight metal concentrations
required to cause acute mortality in these experiments
were very high relative to those often suspected to be
of lexicological significance in field sediments (e.g.,
Figures 3-lA and 3-3A). This has sometimes been
interpreted as a limitation of the use of SEM and AVS to
predict metal-induced toxicity. However, the range of
AVS in these sediments spiked with metals is similar to
3-5
-------
Toxicitv of Metals in Sediments
>>
. O.OBODSO ooo «D
a o °p 0
o
0 '
b
0
0.
o
i O 0
o & Q G a o
«O % •
Nil 1 1 1 1 1 ftl
tOD OO OBO -
0 ^*
-
O
III 1 t 1 1 1 1 1I
0.01 0.1 1 10 100 1000
Total Metal or SEM G_mol/g dry wt)
51%
25%
24%
100
80
60
40
20
o ;
i i 1 1 HIM i i i . in
B •«
I
i
_ i
a o-
Vfo'^aoK'l,
i i i i 1 Mill 1
oo oo o«
a
0
0
0
0 0
s
-o-o- -
JO L <£l | | |fl| li
1 1 1 1 1 III 1 1 1 1 1 1 II
o £>o oo ocxnoj
o o o ,
° 0
0
-
o
_,
;
o
BI 1 1 M III 1 1 1 1 1 1 II
0.01 0.1 1 10 100 1000
Interstitial Water Toxic Units
64%
27% 9%
100
/•».
»^ 80
2 60
«
I 40
20
1 I I 1 If 111 1 1 1 1 Illlt 1 III Mill U 1 | 1 Hill
c . 'oL ,
•1 o g
f ' °
1 a
1 | ,
o o o ' •
of ^S'ae^VgP0'
1 1 1 1 Mill 1 III lilt
\
.
0
"
. 0.001 0.01 0.1 1 10 100 1000
SEM/AVS
Figure 3-3. Percentage mortality of freshwater and saltwater benthic species in 10-day toxicity tests in sediments
spiked with individual metals (Cd, Cu, Pb, Ni, Ag, or Zn) or a metal mixture (Cd, Cu, Ni, and
Zn). Mortality is plotted as a function of: (A) the sum of the concentrations of the respective metal
or metal mixture in ymol metal per gram dry weight of sediment; (B) IWTU; and (C) SEM/AVS
ratio. Data below the detection limits are plotted at IWTU=0.01 and SEM/AVS=0.001. Heavy
dashed lines are the theoretically based cutoffs of 0.5 FWTU and a SEM/AVS ratio of 1.0. Light
vertical dashed lines are the 90% uncertainty bound limits derived as in Section 3.4. The percentage
of the total number of sediments (n = 184) within the bounded limits is provided above each of the
three panels for the purpose of comparison (silver data from Berry et al., 1999; all other data •
modified after Berry et al., 1.996).
3-6
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
VI
e
o
13
u
y
M
A
5
<_
9
a
j
O
ac
a
V)
u
O
(^
O
&
2
u
a
Range of Total Metal or SEM
Range of IWTUs
Range of SEM/AVS
Figure 3-4. ' Percentage of the 184 spiked sediments from Figure 3-3 that were nootoxic or toxic over various
intervals of (A) concentrations of metal based on sediment dry weight (^mol/g), (B) IWTU, and (C)
SEM/AVS.
3-7
-------
Toxicitv of Metals in Sediments
that of sediments commonly occurring in the field. The
important point here is that even a sediment with only a
moderate concentration of AVS has a considerable
capacity for sequestering metals as a metal sulfide, a
form that is not bioavailable (Di Toro et al.. 1990).
In contrast, the combined data from all available
freshwater and saltwater spiked-sediment experiments
support the use of IWTUs to predict mortality of
.benthic species in spiked-sediment toxicity tests
(Figure 3-3B). Mortality in these experiments was
sediment independent when plotted against IWTUs.
Sediments with IWTUs of <0.5 were generally not toxic.
Of the 96 sediments with IWTUs <0.5,96.9<7c were not
toxic, whereas 76.4% of the 89 sediments with IWTUs
>0.5 were toxic (Table 3-1). This close relationship
between IWTUs and sediment toxicity in sediments
spiked with metals was also observed in studies with
field sediments contaminated with metals (see Section
3.2.3 below), as well as sediments spiked with nonionic
organic chemicals (Adams et al., 1985; Swanz et al..
1990; Di Toro et al., 1991), and field sediments
contaminated with nonionic organic chemicals (Hoke et
al., l994;Swartzetal., 1994).
Table 3-1. Toxicity of sediments from freshwater and saltwater lab-spiked sediment tests, field locations, and
combined lab-spiked and field sediment tests as a function of the molar concentrations of SEM and
AVS (SEM/AVS or the SEM-AVS), interstitial water toxic units (IWTUs), and both SEM/AVS or SEM-
AVS and FVVTUs
Percent of Sediments
Study Type/Parameter
Laboratory Spike:
SEM/AVS or SEM-AVSC
rwrud
SEM/AVS or SEM-AVSC; IWTU11
Field:
SEM/AVS or SEM-AVSC
[WTUd
SEM/AVS or SEM-AVSC; lWTUd
Lab-Spike and Field:
SEM/AVS or SEM-AVS6
IWTUd
SEM/AVS or SEM-AVSC; IWTUd
Value
sl.Oor sO.O
>l.0or>0.0
<0.5
*0.5
sl.Oor sO.O: <0.5
>1.0or>0.0; aO.5
sl.OorsO.O
>l.0or>0.0
<0.5
*0.5
sl.Oor sO.O;<0.5
>1.0or>0.0; *0.5
sl.OorsO.O
>1.0or>0.0
<0.5
zQ.S
sl.Oor sO.O; <0.5
>1.0or>0.0; *0.5
n
101
95
96
89
83
78
•57
79
79
53
49
45
158
174
175
142
132
123
Nontoxic
98.0
26.3
96.9
23.6
97.6
14.1
-98.2
59.5
98.7
45.3
100.0
33.3
98.1
42.0
97.7
31.7
98.5
21.1
Toxic
2.0
73.7
3.1
76.4
2. -l
85.9
1.8
-10.5
1.3
. 54.7 .
0.0
66.7
1.9
58.0
2.3
68.3
1.5
78.9
aNontoxtc sediments s24% mortality.
bToxic sediments >24% mortality.
cAn SEM/AVS ratio of s I 0 or an SEM-AVS difference of sO.O indicates an excess of sulfide and probable nontoxic
sediments. 'An SEM/AVS ratio of >1 0 or an SEM-AVS difference of >0.0 indicates an excess of metal and potentially
loxic sediments.
dAn IWTU of <0.5 indicates a probable nontoxic interstitial water concentration of less than one-half of the water-only
LC50 of the same duration. An 1WTU of i0.5 indicates a possibly toxic interstitial water concentration of greater than
one-half of the water-only LCSO of the same duration.
Source: Modified from Hansen'et al., I996a.
3-8
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
The interstitial water metal concentrations in
spiked-sediment studies were most often below the
limit of analytical detection in sediments with SEM/AVS
ratios below 1.0(Berry etal.. 1996). Above an SEM/
AVS ratio of 1.0. the interstitial metals concentrations
increased up to five orders of magnitude with
increasing SEM/AVS ratio. This increase of several
orders of magnitude in interstitial water metals
concentration with an increase of only a factor of two
or three in sediment concentration is the reason why
mortality is most often complete in these sediments.
and why the chemistry of anaerobic sediments controls
the toxicity of metals to organisms living in aerobic
microhabitats. It also explains why toxicities of
different metals in the same sediment to different
species when expressed on the basis of sediment
metals concentration are so similar. Interstitial water
metals were often below or near detection limits when
SEM/AVS ratios were only slightly above 1.0,
indicating the presence of other metal-binding phases
'in sediments.
The combined data from all available freshwater
and saltwater spiked-sediment experiments also support
the use of SEM/AVS ratios to predict sediment toxicity
to benthic species in spiked-sediment toxicity tests. All
tests yield similar results when mortality is plotted
against SEM/AVS ratios (Figure 3-3C). Mortality in
these experiments was sediment independent when
plotted on an SEM/AVS basis. With the combined data,
9S.O?cofthe 101 metals-spiked sediments with SEM/
AVS ratios s 1.0 were not toxic, whereas 73.7% of the 95
sediments with SEM/AVS ratios > 1.0 were toxic (Table
3-D.
The overall data show that when both SEM/AVS
ratios and IWTUs are used, predictions of sediments •
that would be toxic were improved. Of the 83 sediments
with SEM/AVS ratios s 1.0 and IWTUs <0.5,97.6% were
not toxic, whereas 85.9% of the 78 sediments with SEM/
AVS ratios > 1.0 and IWTUs 2 0.5 were toxic (Table 3-1).
These results show that SEM/AVS and IWTUs are
accurate predictors of the absence of mortality in
sediment toxicity tests; however, predictions of
sediments that might be toxic are less accurate. The '.
fact that a significant number of sediments (26.3%)
tested had SEM/AVS ratios of > 1.0 but were not toxic
indicates that other binding phases, such as organic
carbon (Mahony et al., 1996), may also control
bioavailability in anaerobic sediments.
Organism behavior may also explain why some
sediments with SEM/AVS ratios of > 1.0 were not toxic.
Many of the sediments that had the highest SEM/AVS
ratios in excess of 1.0 that produced little or no
mortality were from experiments using the polychaete
/V. arenaceodentata (see Pesch et al., 1995). In these
experiments, this polychaete did not burrow into some
of the test sediments with the highest concentrations.
thereby limiting its exposure to the elevated
concentrations of metals in the interstitial water and
sediments. This same phenomenon may also explain
the low mortality of snails. Heliosoma sp., in freshwater
sediments with high SEM/AVS ratios. These snails are
epibenthic and crawl onto the sides of test beakers to
avoid contaminated sediments (G.L. Phipps, U.S. EPA.
Duluth, MN, personal communication). Increased
mortality was always observed in sediments with SEM/
AVS ratios->5.9 in tests with the other five species.
Similarly, a significant number of sediments (23.6%)
with iO.5 IWTUs were not toxic. This is likely the
result of interstitial water ligands, which reduces the
bioavailability and toxicity of dissolved metals;
sediment avoidance by polychaetes or snails; or
methodological problems in contamination-free
sampling of interstitial water. Ankley et al. (1991)
suggested that a toxicity correction for the hardness of
the interstitial water for freshwater sediments is needed
to compare toxicity in interstitial water with that in
water-only tests. Absence of a correction for hardness
would affect the accuracy of predictions of metal-
induced sediment toxicity using IWTUs. Furthermore.
a significant improvement in the accuracy of metal-
induced toxicity predictions using IWTUs might be
achieved if DOC binding in the interstitial water is taken
into account. Green et al. (1993) and Ankley et al.
(1991) hypothesized that increased DOC in the
interstitial water reduced the bioavailability of cadmium
in sediment exposures, relative to the water-only
exposures. Green et al. (1993) found that the LC50
value for cadmium in an interstitial water exposure
without sediment was more than twice that in a water-
only exposure, and that the LC50 value for cadmium in -
interstitial water associated with sediments was more
than three times that in a water-only exposure.
3.2.3 Field Sediments
In addition to short-term laboratory experiments
with spiked sediments, there have been several .
published studies of laboratory toxicity tests with
metal-contaminated sediments from the field. Ankley et
al. (1991) exposed L variegatus and the amphipod H.
ayeca to 17 sediment samples along a gradient of
cadmium and nickel contamination from a freshwater/
.3-9
-------
Toxicitv of Metals in Sediments
estuarme site in Foundry Cove. NY. In 10-day toxicity
tests. H. a-tecd mortality was not significantly different
from controls in all sediments where SEM (cadmium
plus nickel) was less than AVS. Mortality was greater
than controls only in sediments with more SEM than
AVS. L varieqatus was far less sensitive to the
sediments than H. a-teca. which correlates with the
differential sensitivity of the two species in water-only
tests with cadmium and nickel.
In 10-day toxicity tests with the saltwater
amphipod A. abdita in these same sediments, Di Toro
et al. (1992) observed that metals concentrations
ranging from 0.1 to 28 jumol SEM/g sediment were not
toxic in some sediments, whereas metals concentrations
ranging from 0.2 to 1.000 ^mol SEM/g sediment were
lethal in other sediments. These results indicate that
the bioavailable fraction of metals in sediments varies
from sediment to sediment. In contrast, the authors
also observed a clearly discernible mortality-
concentration relationship when mortality was related
to the SEM/AVS molar ratio (i.e., there was no
significant mortality where SEM/AVS ratios were <1.0,
mortality increased in sediments having SEM/AVS
ratios of 1.0 to 3.0. and there was 100% mortality in .
sediments with ratios > 10). The sum of the IWTUs for
cadmium and nickel ranged from 0.08 to 43.5.
Sediments with sO.5 IWTUs were always nontoxic,
those with >2.2 IWTUs were always toxic, and two of
seven sediments with intermediate IWTUs (0.5 to 2.2)
were toxic. Molar concentrations of cadmium and
nickel in the interstitial water were similar. However, ,
cadmium contributed over 95% to the sum of the toxic
units because cadmium is 67 times more toxic to A.
abdita than nickel. The latter illustrates the utility of
interstitial water concentrations of individual metals in
assigning the probable cause of mortality in benthic
species (Hansenetal., I996a).
In tests with the same sediments from Foundry
Cove, Pesch et al. (1995) observed that 6 of the '17
sediments tested had SEM/AVS ratios <1.0 and iWTUs
<0.5, and none of the 6 were toxic to the polychaete N.
arenaceodentata. Interestingly, the other 11 sediments
containing SEM/AVS ratios > 1.0 were also not toxic.
The results are not surprising given that in these
particular tests only one sediment had >0.5 IWTUs, N.
arenaceodentata is not sensitive to cadmium and
nickel, and the polychaetes did not burrow into
sediments containing toxic concentrations of these
metals.
. Ankley et al. (1993) examined the significance of .
AVS as a binding phase for copper in freshwater
sediments from two copper-impacted sites. Based on
interstitial water copper concentrations in the test
sediments, the 10-day LC50 for H. ayeca was 31 Mg/L;
this compared favorably with a measured LC50 of 28
Mg/L in a 10-day waterronly test. Sediments having
SEM/AVS ratios < 1.0 were not toxic. They also
observed no toxicity in several sediments with
markedly, more SEM than AVS. suggesting that copper
was not biologically available in these sediments.
Absence of copper in interstitial water from these
sediments corroborated this lack of bioavailability.
This observation suggested the presence of binding
phases in addition to AVS for copper in the test
sediments. Two studies suggest that an important
source of Che extra binding capacity in these sediments
was organic carbon (U.S. EPA, 1994a; Mahony et al..
1996).
Hansenet al. (I996a).investigated the biological
availability of sediment-associated divalent metals to A.
abdita and H. a-teca in sediments from five saltwater
locations and one freshwater location in the United
States. Canada, and China using 10-day lethality tests.
Sediment toxicity was not related to dry weight metals
concentrations. In the locations where metals might be
likely to cause toxicity, 49 sediments had less SEM than
AVS and <0.5 IWTUs. and no toxicity was observed. In
contrast, one-third of the 45 sediments with more SEM
than AVS and >0.5 IWTUs were toxic (Table 3-1).
Hansen et al. (1996a) made an observation that is'
important to interpretation of toxicity of sediments from
field locations, particularly those from industrial
harbors. They observed that if sediments with SEM/
AVS ratios <1.0 are toxic, even if metals concentrations
on a dry weight basis are very high, the toxicity is not
likely to be caused by metals. Furthermore, it is
incorrect to use such data to reach the conclusion that
the EqP approach is not valid. This is because when
SEM/AVS ratios were < 1.0, there was an almost
complete absence of toxicity in both spiked sediments
and field sediments where metals were the only known
source of contamination and IWTUs for metals were
<0.5. When metals concentrations expressed as the sum
of the IWTUs are used in conjunction with SEM/AVS
ratios, they together provide insight that can explain
apparent anomalies between SEM/AVS ratios < 1.0 and
sediment toxicity in field sediments. Joint use of both
SEM/AVS rdtios and interstitial water concentrations is
also a powerful tool for explaining absence of toxicity
when SEM/AVS ratios are > 1 .'0. Overall, when
freshwater and saltwater field sediments were tested in
the laboratory, 100% were not toxic when SEM/AVS
was s 1.0 and IWTUs were <0.5. and 66.7% were toxic
3-10
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
when SEM/AVS was > 1.0 and IWTUs were aO.5 (Table
3-1).
Therefore, because AVS can bind divalent metals in
proportion to their molar concentrations, Hansen et al.
(1996a) proposed the use of the difference between the
molar concentrations of SEM and AVS (SEM-AVS)
rather than SEM/AX'S ratios used previously. The molar
difference provides important insight into the extent of
additional available binding capacity and the
magnitude by which AVS binding has been exceeded
(Figure 3-5). Further, absence of organism response
when. AVS binding is exceeded can indicate the
potential magnitude of other important binding phases
in controlling bJoavailability. Figure 3-5 shows that for
most nontoxic freshwater and saltwater field sediments,
1 to 100 umol of additional metal would be required to
exceed the sulftde-binding capacity (i.e., SEM-AVS =
-100 to -1 jmol/g). In contrast, most toxic field
sediments contained 1 to l.OOOwmolof metal beyond
the binding capacity of sulfide alone. Data on nontoxic
field sediments whose sulfide-binding capacity is
exceeded (SEM-AVS is > 1.0 umoUg) indicate that other
sediment phases, in addition to AVS. have significance
in controlling metal bioavailability. In comparison to
SEM/AVS ratios, use of SEM-AVS differences is
particularly informative where AVS concentrations are
low, such as those from Steilacoom Lake and the
Keweenaw Watershed, where the SEM-AVS difference
is numerically low and SEM/AVS ratios are high
(Ankley et al., 1993). For these reasons. SEM-AVS is
used instead of the SEM/AVS ratio almost exclusively
for the remainder of this document.
3.2.4 Field Sites and Spiked Sediments
Combined
Figure 3-6 and Table 3-1 summarize available data
from freshwater and saltwater sediments spiked with
individual metaJs or metal mixtures, freshwater field
sites, and saltwater field sites on the utility of metals
concentrations in sediments normalized by dry weight.
IWTUs. and SEM-AVS. These data explain the.
a
5
100
80
60
40
o o
o
D D
e
AQ
0 1
SEM-AVS (^mol/g dry wt)
10
100
1000
Figure 3-5. Percentage mortality of amphipods, oligochaetes, and polychaetes exposed to sediments from four
freshwater and three saltwater field locations as a function of the sum of the molar concentrations of
SEM minus the molar concentration of AVS (SEM-AVS). Sediments having s24% mortality are
considered nontoxic as defined by Berry et al. (199$), which is indicated by the horizontal dotted line
in the figure. The vertical dotted line at SEM-AVS = 0.0 Mmol/g dry wt indicates the boundary
between sulfide-bound unavailable metal and potentially available metal. The different symbols
represent field sediments from different locations (figure from Hansen et al., 1996a).
3-11
-------
Toxicity of Metals in Sediments
?
•»•
;' £
"a
~
9
-
1 '
!
c?
^
&
—
•2
o
g
5
"5
•2
1
100
80
60
40
20
0
100
80
60
40
20
0'
0
100
fcgo
60
40
20,
<
O.
*
j
1 1 i 1 1 1 III 1 1 1 1 1 IIII 1 1 1 1 1 till 1 1 1 1 1 Nil 1 1 1 1 1 1 1
.A o o0<.«.o«f^ 4M.Ooo.oO.
o Spiked Sediment ° o 0 o % o
' 0 Field Sediment ° o
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
bioavailability and acute toxicity of metals in sediments
(HansenetaL 1996a; Berry etal., 1999). This analysis
contains all available data from' 10-day lethality tests
where mortality. [WIUs. SEM. and AVS are known from
experiments with sediments toxic only because of
metals. The relationship between benthic organism
mortality-and total dry weight metals concentrations in
spiked and field sediments is not useful to causally
relate metal concentrations to organism response
(Figures 3-4A and 3-6A). The overlap is almost four
orders of magnitude in the bulk metals concentrations
that cause no toxicity and those that are 100% lethal for
these sediments where metals are the only source of
toxicity (see discussion in Section 3.2.2).
Data in Figure 3-6B show that over all tests, the
to-xicity of sediments whose concentrations are
normalized on an IWTU basis are typically consistent
with the IWTU concept; that is. if IWTUs are s 1.0.
then sediments should be lethal to s50% of the
organisms exposed, and significant mortality probably
should be absent at <0.5 IWTUs. Of the spiked and
field sediments evaluated that had IWTUs <0.5.97.7%
of 175 sediments were nontoxic (Table 3-1). For the 142
sediments having IWTUs aO.5.68.3% were toxic.
However, and as stated above, given the effect on
toxicity or bioavailability of the presence of other
binding phases (e.g., DOC) in interstitial water, water
quality (hardness, salinity, etc.), and organism behavior.
it is not surprising that many sediments having IWTUs
>0.5 are not toxic.
Data in Figure 3-6C show that over all tests,
organism response in sediments whose concentrations
are normalized on an SEM-AVS basis is consistent with
metal-sulfide binding on a mole to mole basis as first
described by Di Toro et al. (1990), and later
recommended for assessing the bioavailability of
metals in sediments by Ankley et al. (1994). Saltwater
and freshwater sediments either spiked with metals or
from field locations with SEM-AVS differences sO.O
were uniformly nontoxic (98.1% of 158 sediments)
(Table 3-1). The majority (58.0%) of 174 sediments
having SEM-AVS >0.0 were toxic. It is not surprising
that many sediments having SEM-AVS >0.0 are not
toxic given the effect on toxicity or. bioavailability of the
presence of other sediment phases that also affect
bioavailability (see Section 3-4; Di Toro et al., 1987,
2000; Mahony etal., 1996).
Over all tests, the data in Figure 3-6 indicate that
use of both IWTUs and SEM-AVS together did not
improve the accuracy of predictions of sediments that
were nontoxic (98.5% of 132 sediments; Table 3-1).
However, it is noteworthy that 78.9% of the 123
sediments with both SEM-AVS >0.0 and IWTUs aO.5
were toxic. Therefore, the approach of using SEM-AVS,
IWTUs, and especially both indicators to identify
sediments of concern is very useful.
The results of all available data demonstrate that
using SEM, AVS, and interstitial water metals
concentrations to predict the lack of toxicity of
cadmium, copper, lead, nickel, silver, and zinc in
sediments is certain. This is very useful, because the
vast majority of sediments found in the environment in
the United States have AVS concentrations that exceed •
the SEM concentration (SEM-AVS <0.0) (see Section
4.4). This may incorrectly suggest that there should be
little concern about metals in sediments on a national
basis, even though localized areas of biologically
significant metal contamination do exist (Wolfe et al..
1994; Hansen et al.. I996a; Leonard et al.. 1996a). It is
potentially important that most of these data are from
field sites where sediment samples were collected in the
summer. At this time of year, the seasonal cycles of
AVS produce the maximum metal-binding potentials
(Boothman and Helmstetter. 1992; Leonard etal.. 1993).
Hence, sampling at seasons and conditions when AVS
concentrations are at a minimum is a must in
establishing the true overall level of concern about
metals in the nation's sediments and in evaluations of
specific sediments of local concern.
Predicting which sediments with SEM-AVS >0.0
will be toxic is presently less certain. Importantly, the
correct classification rate seen in these experiments is
high; that is, the accuracy of predicting which
sediments were toxic was 58.0% using the SEM and
AVS alone, 68.3% using IWTUs. and 78.9% using both
indicators. An SEM-AVS >0.0, particularly at multiple
adjacent sites, should trigger additional tiered
assessments. These might include characterization of
the spatial (both vertical and horizontal) and temporal
distribution of chemical concentration (AVS and SEM)
and toxicity, measurements of interstitial water metal. .
and toxicity identification evaluations (TIEs). In this
context, the combined SEM-AVS and IWTU approach
should be viewed as only one of the many sediment
evaluation methodologies.
3.2.5 Conclusions from Short-Term Studies
EPA believes that results from tests using
sediments spiked with metals and sediments from the
field in locations where toxicity is associated with
metals demonstrate the value of explaining the
3-13
-------
Toxicitv of Metals in Sediments
biological availability of metals concentrations
normalized by SEM-AV'S and IWTUs instead of dry
weight metal concentrations. Importantly, data from
spiked-sediment tests strongly indicate that metals are
not the cause of most of the toxicity observed in field
sediments when both SEM-AVS is sO.O and IWTUs are
<0.5 (Table 3-D. Expressing concentrations of metals in
sediments on an SEM-AVS basis provides important
insight into the available additional binding capacity of
sediments and the extent to which sulfide binding has
been exceeded.
SEM-AVS and interstitial water concentrations of
metals can aid in identifying the specific metal causing
toxicity. For example, the metal(s) in excess of AVS can
be identified by subtracting from the molar
concentration of AVS the molar concentrations of
specific metals in theSEM in order of their sulfide
solubility product constants (ATs ,) in the SEM.
Alternatively, interstitial water concentrations of metals
can be used to identity a specific metal causing
sediment toxicity using the toxic unit concept, if
appropriate water-only toxicity data for the tested
species are available (Hansen et al., 1996a).
Predictions of sediments not likely to be toxic,
based on use of SEM-AVS and IWTUs for all data from
freshwater or saltwater field sediment and spiked-
sediment tests, are extremely accurate (98.5%) using
both parameters. Predictions of sediments likely to be
toxic are less accurate. Nevertheless; SEM-AVS is
extremely useful in identifying sediments of potential
concern. Data were summarized from amphipod tests
using freshwater and saltwater laboratory metals-
spiked sediments and field sediments where metals
were a known problem by comparing the percentage of
sediments that were toxic with the SEM-AVS
concentration (tests with polychaetes and gastropods
were excluded because these organisms avoid
exposure) (Hansen, 1995). Seventy percent of the
sediments in these amphipod studies with an SEM-AVS
concentration of iO.76 Mmol of.excess SEM/g were
toxic. The corresponding values for 80%, 90%, and
100% of the sediments being toxic were 2.7,16, and 115
A^mol of excess SEM/g, respectively.
Of course, SEM. AVS, and IWTUs can only predict
foxicity or the lack of toxicity caused by metals in
sediments. They cannot be used alone to predict
toxicity of sediments contaminated with toxic
concentrations of other contaminants. However, SEM
and AVS have been used in sediment assessments to
rule out metals as probable causative agents of toxicity
(Wolfe et al.. 1994). Also, the use of SEM and AVS to
predict biological availability and toxicity of cadmium,
copper, lead, nickel, silver, and zinc is applicable only to
anaerobic sediments that contain AVS; binding factors
other than AVS control bioavailability in aerobic
sediments (Di Toro et al., 1987; Tessier et al.. 1993).
Measurement of interstitial water metal may be useful
for evaluations of these and other metals in aerobic and
anaerobic sediments (Ankley et al., 1994). Even with
these caveats, EPA believes that the combined use of
SEM, AVS. and interstitial measurements is preferable
to all other currently available sediment evaluation
procedures to causally assess the implications to •
benthic organisms of these six metals associated with
sediments (see discussion in Section 5, Sampling and
Analytical Chemistry, for further guidance).
3.3 Predicting Metal Toxicity:
Long-Term Studies
Taken as a whole, the short-term laboratory
experiments with metal-spiked and field-collected
sediments present a strong argument for the ability to
predict the absence of metal toxicity based on sediment
SEM and AVS relationships and/or interstitial water
metal concentrations. However, if this approach is to
serve as a valid basis for ESG derivation, comparable
predictive success must be demonstrated in long-term
laboratory and field experiments where chronic effects
could be manifested (Luoma and Carter, 1993; Meyer et
al.. 1994). This demonstration was the goal of
experiments described by Hare et al. (1994), DeWitt et
al. (1996), Hansen et al. (I996b), Liber et al. (1996). and
Sibley et al. (1996). An important experimental
modification to these long-term studies, as opposed to
the short-term tests described in Section.3.2, was the
collection of horizon-specific chemistry data. This is
required because AVS concentrations often increase,
and SEM-AVS differences decrease, with an increase in ,
sediment depth (Howard and Evans, 1993; Leonard et
al., 1996a); hence, chemistry performed on homogenized
samples might not reflect the true exposure of benthic
organisms dwelling in surficial sediments (Luoma and
Carter, 1993; Hare etal.. 1994; Peterson etal., 1996).
3.3.1 Life-Cycle Toxicity Tests
DeWitt et al. (1996) conducted an entire life-cycle
toxicity test with the marine amphipod L plumulosus
exposed for 23 days to cadmium-spiked estuarine '
sediments (Table 3-2). The test measured effects on
survival, growth, and reproduction of newborn
3-14
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Table 3-2. Summary of the results of full life-cycle and colonization toricity tests conducted in the laboratory and
field using sediments spiked with individual metals and metal mixtures
Toxicity Test
LifeCvcle:
Leptocheirus
plumulosus
Chironomus
tentans
Colonization:
Laboratory-
saltwater
Field-saltwater
Field-freshwater
Field-freshwater
Measured SEM-AVS1
Dura- (/»mol/g)
tion b
Metal(s) (days) NOEC(s) OEC(s)
Cadmium 28 -3.5. -2.0. 8.9. 15.6
0.78. 2.0
Zinc 56 -2.6. -1.4, 21.9,32.4
6.4
Cadmium 118 . -13.4 8.0,27.4
Cadmium. 120 -0.31. —
copper, -0.06.
lead. 0.02
nickel, zinc
Cadmium -365 -0.07,0.08. 2.2
0.34
Zinc ' 368 -3.6. -3.5. —
-2.9,
-2.0. 1.0J
Effect
Mortality 100%
Larval mortality 85%-
100%
Weight, emergence, and
reproduction reduced.
Fewer polychaetes. shifts
in community
composition, fewer
species, bivalves absent,
tunicates increased
No effects observed
Reduced Chironomus
salinanus numbers
Bioaccumulation
No effects observed
Reference
De Witt etal:.. 1996
SibleyetaL 1996
Hansen et al..
1996b .
Boothman et al.:
2000
Hare etal., 1994
Liber etal.. 1996
aSEM-AVS differences are used instead of SEM/AVS ratios 'to standardize across the studies referenced. An SEM-AVS difference of
sO.O is the same as an SEM/AVS ratio of s 1.0. An SEM-AVS difference of >0.0 is the same as an SEM/AVS ratio of >1.0.
bNOECs = no observed effect concentrations); all concentrations where response was not significantly different from the control.
cOECi = observed effect ccracentration(s); all concentrations where response was significantly different from the control.
^Occasional minor reductions in oligochaetes (Naididae).
amphipods relative to interstitial water and SEM/AVS
normalization. Seven treatments ot'Cd were tested: 0
(control). -3.5. -2.0.0.78.2.0.8.9,and IS.eSEM^-AVS
differences (measured concentrations). Gradients in
AVS concentration as a function of sediment depth
were greatest in the control treatment, decreased as the
SEMCd ratio increased, and became more pronounced
over time. Depth gradients in SEMCd-AVS differences
were primaiily.caused by the spatial and temporal
changes in AVS concentration, because SEMCd
concentrations changed very little with time or depth. .
Thus in most treatments SEMCd-AVS differences were
smaller at the top of sediment cores than at the bottom.
This is expected because the oxidation rate of iron
sulfide in laboratory experiments is very rapid (100% in .
60 to 90 minutes) but for cadmium sulftde it is slow
(10% in 300 hours) (Mahony et al., 1993; Di Toro et al..
1996a). Interstitial cadmium concentrations increased
in a dramatic stepwise fashion in treatments having a
SEM-AVS difference of * 8.9 ^mol of excess SEM. but
were below the 96-hour LC50 value for this amphipod in
lesser treatments. There were no significant effects on
survival, growth, or reproduction .in sediments
containing more AVS than cadmium (-3.5 and -2.0
Amol/g) and those with a slight excess of SEMC(1 (0.78
and 2.0 Mmol/g), in spite of the fact that these samples
contained from 183 to 1.370Mg cadmium/g sediment.
All amphipods died in sediments having SEM; AVS
3-15
-------
Toxicitv of Metals in Sediments
differences >8.9 utnol excess SEM/g. These results are
consistent with predictions of metal bioavailability from
10-day acute tests with metal-spiked sediments (i.e.,
that sediments with SEMCd-AVS differences sO.O are
not toxic, interstitial water metal concentrations are
related to organism response, and sediments with
SEMCJ-AVS differences >0.0 may be toxic).
Sibley et al. (1996) reported similar results from a
56-day life-cycle test conducted with the freshwater
• midge C. tentans exposed to zinc-spiked sediments
(Table 3-2). The test was initiated with newly hatched
larvae and lasted one complete generation, during
which survival, growth, emergence, and reproduction
were monitored. In sediments where the molar
difference between SEM and AVS (SEM-AVS) was <0.0
(dry weight zinc concentrations were as high as 270
mg/'kg). concentrations of zinc in the-sediment
interstitial water were low and no adverse effects were
observed for any of the biological endpoints measured.
Conversely, when SEM-AVS was 21.9 and 32.4 ^mol of
excess SEM/g. interstitial water concentrations of zinc
increased (being highest in surficial sediments), and
reductions in survival, growth, emergence, and
reproduction were observed. Over the course of the
study, the absolute concentration of zinc in the
interstitial water in these treatments decreased because
of the increase in sediment AVS and loss of zinc from
twice-daily renewals of the overlying water.
3.3.2 Colonization Tests
Hansen et al. (1996b) conducted a 118-day benthic
colonization experiment in which sediments were spiked
to achieve nominal cadmium/AVS molar ratios of 0.0
(control); 0.1,0.8, and 3u and then held in the
laboratory in a constant flow of unfiltered seawater
(Table 3-2). Oxidation of AVS in the surficial 2.4 cm of
the control treatment occurred within 2 to 4 weeks and
resulted in sulftde profiles similar to those occurring in
sediments in nearby Narragansett Bay, RI (Boothman
and Helmstetter. 1992). In the nominal 0.1 cadmium/
AVS treatment, measured SEM^ was always less than
AVS (SEM-AVS = -13.4 ^mol AVS/g in the surficial 2.0
cm), interstitial cadmium concentrations (<3 tolO/ug/L)
were less than those likely to cause biological effects,
and no significant biological effects were detected. In
the nominal 0.8 cadmium/AVS treatment (SEM-AVS =
8.Oumol SEM/g), measured SEMC(J commonly exceeded
AVS in the surficial 2.4 cm of sediment, and interstitial
cadmium concentrations (24 to 157 Mg/L) were
sufficient to be of lexicological significance to highly
sensitive species. In this treatment, shifts in the
presence or absence of organisms were observed over
all taxa. and there were fewer macrobenthic polychaetes
(Mediomastus ambiseta, Strebtospio benedicti. and
. Podarke obscura) and meiofaunal nematodes. In the
nominal 3.0 cadmium/AVS treatment (SEM-AVS of 27.4
Mtnol SEM/g). concentrations of SEMCd were always
greater than AVS throughout the sediment column.
Interstitial cadmium ranged from 28.000 to 174,000 ng/L.
In addition to the effects observed in the nominal .0.8
cadmium/AVS treatment, the following effects were
observed: (a) sediments were colonized by fewer
macrobenthic and polychaete species and
harpacticoids, (b) the sediments had lower densities of
diatoms, and (c) bivalve molluscs were absent. Overall
treatments, the observed biological responses were
consistent with predicted possible adverse effects
resulting from elevated SEMCd-AVS differences in
surficial sediments and interstitial water cadmium
concentrations.
Boothman et al. (2000) conducted a field
colonization experiment in which sediments from
Narragansett Bay, RI, were spiked with an equimolar
mixture of cadmium, copper, lead, nickel, and zinc at
nominal SEM/AVS ratios of 0.1,0.8, and 3.0; placed in
boxes; and replaced in Narragansett Bay (Table 3-2).
The AVS concentrations decreased with time in surface
sediments (0 to 3 cm) in all treatments where the
nominal SEM/AVS ratio was < 1.0 (SEM-AVS decreased
from -0.31 to -0.06 ^mol SEM/g in the surficial 2.0 cm)
but did not change in subsurface (6 to 10 cm)
sediments or in the entire sediment column where
nominal SEM/AVS ratios exceeded 1.0 (SEM-AVS = 0.02
/urnol AVS/g). SEM decreased with time only where
SEM exceeded AVS. The concentration of metals in
interstitial water was below detection limits when there
was.more AVS than SEM. When SEM exceeded AVS.
significant concentrations of metals were present in.
interstitial water, and appeared in the order of their
sulfide solubility product constants. Interstitial water
concentrations in these sediments decreased with time.
although they exceeded the WQC in interstitial water
for 60 days for all metals, 85 days for cadmium and zinc,
and 120 days for the entire experiment for zinc. Benthic
faunal assemblages in the spiked-sediment treatments
were not different from those of the control treatment.
Lack of biological response was consistent with the
vertical profiles of SEM and AVS.. AVS was greater
than SEM in all surface sediments, including the top 2
cm of the 3.0 nominal SEM/AVS treatment, because of
oxidation of AVS and loss of SEM. The authors
speculated that interstitial metal was likely absent in the
surficial sediments in spite of data demonstrating the
3-16
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
presence of significant measured concentrations.
Interstitial water in the 3.0 nominal SEM/AVS treatment
was sampled from sediment depths where SEM was in
excess, rather than in the surficial sediments. Important
to the biological data are the surficial sediments', where
settlement by saltwater benthic organisms first occurs.
Also, there was a storm event that allowed a thin layer
of clean sediment-to be deposited on top of the spiked
sediment (W.S. Bobthman, U.S. EPA, Narragansett, RI.
personal communication). These data demonstrate the
importance of sampling sediments and interstitial water
•in sediment horizons where benthic organisms are
active.
Hare et al. (1994) conducted an approximately
1 -year field colonization experiment in which
uncontaminated freshwater sediments were spiked with
cadmium and replaced in the oligotrophic lake from
which they originally had been collected (Table 3-2).
Cadmium concentrations in interstitial waters were very
low at cadmium-AVS molar differences <0.0. but
increased markedly at differences >0.0. The authors
reported reductions in the abundance of only the
chironomid Chironomus saiinanus in the 2.2 .umol
excess SEM/g treatment. Cadmium was accumulated by
organisms from sediments with surficial SEM
concentrations that exceeded those of AVS. These
sediments also contained elevated concentrations of
cadmium in interstitial water.
Liber et al. (1996) performed a field colonization
experiment using sediments having 4.46 ^mol of sulfide
from a freshwater mesotrophic pond (Table 3-2).
Sediments were spiked with 0.8,1.5.3.0,6.0, and 12.0
umol of zinc, replaced in the field, and chemically and
biologically sampled ov» 12 months. There was a
pronounced.increase in AVS concentrations with
increasing zinc concentration; AVS was lowest in the
surficial 0 to 2 cm of sediment with minor seasonal
variations. With the exception of the highest-spiking
concentration (approximately 700 mg/kg, dry weight),
AVS concentrations remained larger than those of SEM.
Interstitial water zinc concentrations were rarely
detected in any treatment, and were never at
concentrations that might pose a hazard to benthic
macroinvertebrates. The only observed difference in
benthic community structure across the treatments was
a slight decrease in the abundance of Naididae
oligochaetes at the highest spiking concentration. The
absence of any noteworthy biological response was
consistent with the absence of interstitial water
concentrations of biological concern. The lack of
biological response was attributed to an increase in
concentrations of iron and manganese sulfides
produced during periods of diagenesis. which were
replaced by the more stable zinc sulfide. which is less
readily oxidized during winter months. In this
experiment, and theoretically in nature, excesses of
sediment metal might be overcome over time because of
the diagenesis of organic material. In periods of
minimal diagenesis. oxidation rates of metal sulfides. if
sufficiently great, could release biologically significant
concentrations of the metal into interstitial waters. The
phenomenon should occur metal by metal in order of
their sulfide solubility product constants.
3.3.3 Conclusions from Chronic Studies
Over all full life-cycle and colonization toxicity
tests conducted in the laboratory and field using
sediments spiked with individual metals and metal
mixtures (Table 3-2). no sediments with an excess of
AVS tSEM-AVS s 0.0) were toxic (Figure 3-7). •
Conversely, all sediments where chronic effects were
observed, and 7 of 19 sediments where no effects were
observed, had an excess of SEM (SEM-AVS >0.0)
(Table 3-2; Figure 3-7). Therefore, the results from all
available acute and chronic toxicity tests support the
use of SEM-AVS s 0.0 as an ESG that can be used to
predict sediments that are unlikely to be toxic.
3.4 Predicting Toxicity of Metals in
Sediments
3.4.1 General Information
The SEM-AVS method for evaluating toxicity of
metals in sediments (Di Toro et aJ., 1990. 1992) has
proven to be successful at predicting the lack.of metal
toxicity in spiked and field-contaminated sediments
(Berryetal., 1996;Hansenetal., 1996a). However, •
because SEM-AVS does not explicitly consider the
other sediment phases that influence interstitial water-
sediment partitioning, and in spite of its utility in
identifying sediments of possible concern, it was never
intended to be used to predict the occurrence of
toxicity. The proposed sediment quality criteria for
metals using SEM, AVS. and IWTUs in Ankley et al.
(1996)—now referred to as ESGs or equilibrium
partitioning sediment guidelines—were constructed as
"one-tailed" guidelines. They should be used to
predict the lack of toxicity but not its presence. Thus
the problem of predicting the onset of toxicity in metal-
contaminated sediments remained unsolved.
3-17
-------
Toxicitv of Metals in Sediments
Predicted
Nontoxic
Toxicity Uncertain
• Experimental OEC
D Experimental NOEC
1
-20
-10
10
20
30
40
SEM-AVS (^mol/g)
Figure 3-7. Comparison of the chronic toxicity-of sediments spiked with individual metals or metal mixtures to
predicted toxicity based on SEM-AVS (data from Table 3-2). Horizontal dashed line separates
experimental observed effect concentrations (solid columns) from no observed effect concentrations
(shaded columns). Values at SEM-AVS s 0.0 Mmo!/^ are predicted to be nontoxic. Values at SEM-AVS
are indicative of sediments that are likely to be toxic or toxicity is uncertain.
3-18
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
This section introduces a modification of the SEM-
AVS procedure in which the SEM-AVS difference is
normalized by the fraction of organic carbon./oc, in a
sediment. This section is largely taken from Di Toro et
al, (2000). Their publication should be consulted for
additional information about the utility of the/^.
procedure and comparison of this procedure with the .
sediment guidelines of Long et al. (1995a) and
MacDonald et al. (1996). The (2SEM-AVS)//^
procedure significantly improves prediction of mortality
by accounting for partitioning of metals to sediment
organic carbon, as well as the effect of AVS. In
addition, the approach used by Di Toro et al. (2000) to
derive (SSEM-AVS)//^ uncertainty bounds for
identifying sediments that are likely to be toxic, are of
uncertain toxicity, or are nontoxic has applicability to
SEN!/AVS ratios. SEM-AVS differences, and IWTUs.
Although not used as an ESG. the uncertainty bounds
should be useful in prioritizing sediments of concern
for further evaluations.
3.4.2 EqP Theory for SEM, AVS, and
Organic Carbon
The EqP model provides for the development of
causal sediment concentrations that predict toxicity or
lack of toxicity in sediments (Di Toro etal., 1991). The
sediment concentration Cs that corresponds to a
measured LC50 in a water-only exposure of the test
organism is
(3-2)
where Cs' is the sediment LC50 concentration (^g/kg
dry wt), K (L/kg) is the partition coefficient between
interstitial water and sediment solids, and LC50 is the
concentration causing 50% mortality (^g/L). For
application to metals that react with AVS to form
insoluble metal sulfides. Equation 3-2 becomes
(3-3)
where AVS is the sediment concentration of acid
volatile sulfides. Equation 3-3 simply states that
because AVS can bind the metal as highly insoluble •
sulfides. the concentration of metal in a sediment that
will cause toxicity is at least as great as the AVS that is
present. The sediment metal concentration that should
be employed is the SEM concentration, because any
metal that is bound so strongly that IN of hydrochloric
acid cannot dissolve it is not likely to be bioavailable
(Di Toro et al.. 1992). Of course, this argument is
theoretical, which is why so much effort has been
expended to demonstrate experimentally that this is
actually the case (Di Toro etal., 1992; Hare etal.. 1994;
Berry et al.. 1996; Hansen et al.. .1996a; Sibley et al..
1996). Therefore, the relevant sediment metal
concentration is SEM, and Equation 3-3 becomes .
The basis for the AVS method is to observe that if
the second term in Equation 3-4 is neglected, then the
critical concentration is SEM = AVS, and the criterion
for toxicity or lack of toxicity is SEM-AVS $ 0.0 (jumol/g
drywt).
The failure of the difference to predict toxicity
when there is an excess of SEM is due to neglect of the
partitioning term KpLC50. Note that ignoring the term
does not affect the prediction of lack of toxicity in that
it makes the condition conservative (i.e., smaller
concentrations of SEM are at the boundary of toxicity
and no toxicity).
The key to improving prediction of toxicity is to
approximate the partitioning term rather than ignore it.
In sediments, the organic carbon fraction is an
important partitioning phase, and partition coefficients
for certain metals at certain pHs have been measured
(Mahony et al., 1996). This suggests that the partition
coefficient K? in Equation 3-4 can be expressed using
the organic carbon-water partition coefficient. KQC.
together with the fraction organic carbon in the
sediment,/,-^
~f K
OC
Using this expression in Equation 3-4 yields
SEM = AVS +/oc KOC LC5°
Moving the known terms to the left side of this
equation yields
SEM-AVS
/c
oc
(3-5)
(3-6).
0-7)
If both Kjy. and LC50 are known, then Equation 3-7 can
be used to predict toxicity.
The method evaluated below uses (ESEM-AVS)/
ffy. as the predictor of toxicity and evaluates the critical
concentrations (the right side of Equation 3-7) based
on observed SEM, AVS, fx, and toxicity data. If
multiple metals are present, it is necessary to use the
total SEM
3-19
-------
Toxicity of Metals in Sediments
= S[SEM:]
(3-8)
to account for all the metals present. Note that (2SEM-
AVSV/OC is the organic carbon-normalized excess SEM
OC
for which we use the notation
foe
,3.9,
3.4.3 Data Sources
Data from toxicity tests using both laboratory-
spiked and fietd-collected sediments were compiled
from the literature. Four sources of laboratory-spiked
tests using marine sediments (Casas and Crecelius.
1994; Pesch etal.. 1995; Berry etal.. 1996. 1999) and one
using freshwater sediments (Carlson etal., 1991) were
included. Two sources for metal-contaminated field
sediments were included (Hansen et al.. 1996a; Kemble
et al.. 1994). The field data from the sediments where
metals were not the probable cause of toxicity (Bear
Creek and Jinzhou Bay) (Hansen et al., I996a) were
excluded. Data reported included total metals, SEM,
AVS, /QC, and 10- or 14-day mortality. In Hansen et al.
( I996a), data were reported for five saltwater and four
freshwater locations, but organic carbon
concentrations were not available for freshwater field
sediments from three locations. Organic carbon data
for the Keweenaw Watershed were obtained separately
(E.N. Leonard. U.S. EPA. Duluth, MN, personal
communication).
Laboratory-spiked and field sediment data were
grouped for analysis. Mortality data were compared .
against the SEM- AVS difference and the SEM-AVS
difference divided by the f^. For each comparison,
two uncertainty bounds were computed: a lower-bound
concentration equivalent to a 95% chance that the
mortality observed would be less than 24% (the
percentage mortality considered to be toxic) (see-Berry
et al., 1996) and an upper-bound concentration
equivalent to a 95% chance that the observed mortality
would be greater than 24%. The lower-bound
uncertainty limit was computed by evaluating the
fraction of correct classification starting from the
lowest x-axis value. When the fraction correct dropped
to below 95%, the 95th percentile was interpolated.
The same procedure was applied to obtain the upper-
bound uncertainty limit. These uncertainty bounds are
the concentration range where it is 90% certain that the
sediment may be either toxic or not toxic.
3.4.4 Acute Toxicity Uncertainty
Mortality in the laboratory-spiked and field-
contaminated sediment tests were both organism and
metal independent when plotted against the SEM-AVS
difference (Figure 3-8A). The horizontal dashed line
indicating 24% mortality is shown for reference. The
90% lower and upper uncertainty bound limits for the
SEM-AVS difference are from 1.7 and 120 ,umol/g. a
factor of 70. Thus, it appears that for both laboratory
and spiked-sediment data, toxicity is likely when the
SEM-AVS difference is > 1 20, uncertain when the
difference is from 1.7 to 120^mol/g, and not likely when
the difference is < 1.7 ,umol/g.
Although use of SEM-AVS differences to predict
toxicity is not based on any theoretical foundation, use
of SEM-AVS sO.O to predict lack of toxicity is based on
the equilibrium partitioning model (Di Toro et al., 199 1)
and the chemistry of metal-sulfide interactions. The
stoichiometry of the uptake of divalent metals by AVS
is such that 1 molof AVS will stabilize I molofSEM.
except for silver, where the ratio is 2: 1. hence the use of
the difference of 0.0 Mmol/g dry weight to predict lack
of toxicity. In fact it is the very low solubility of the
resulting metal sulfides that limits the interstitial water
concentrations to below toxic levels regardless of the
details of the sediment chemistry (e.g.. pH. iron
concentration) as has been demonstrated in this
document and detailed in the Appendix in Di Toro et al.
(1992).
•The (SSEM-AVSV/Qc approach provides an
equivalent theoretical basis that is needed to derive an'
appropriately normalized sediment concentration that
predicts occurrence of toxicity that is causally linked tcr
bioavailable metal. When percent mortality is plotted
against the organic carbon-normalized excess SEM
(SSEM-AVSy/oc) for the same data as contained in
Figure 3-8A, toxicity is likely when the (ISEM-AVSV/gc
is >3,000 Minol/gQj., uncertain when the concentration is
between I30and3,000/imol/goc, and not likely when
the concentration is < 130 Mmol/gQ,. (Figure 3-8B). Thus.
the width of the uncertainty bound is a factor of 70 for
SEM-AVS differences and 23 for (ESEM-
If the (SSEM-AVSy/oc approach improves
predictions of sediment toxicity caused by metals, the
3-20
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
100
so
60
O
20
UNI ii i mm ii i nun ii i niiiiii i luuiiii i
A
-
•
-
•
V
: ° 's|j&Si
MM^^^K
"
•
HIM i i i HUM i i i IHIII i i i nun i i i mini i i i
1 1 1 1 Mill 1 1 IIIIIH1 1 1 1 (till 1 II Mill) 1 1 1 Mllll
1
1
^••*^*^^<^>"* * •
V* ' .•*
,'*> °
• ,. 1
1
• * 1
• a 1
«L v ' 'xa V
fO? • • O
»•£ V 1
1* * "* *
1 1
in ilium i ii mill i ii mill i mum i iiiinii
1 1 1 Mill
•-
•
-
•
~
-
-
1 1 I (fill
SEM-AVS Cumol/g)
i
u
100
80
60
40
20
0
B
-
V
1 V — 534* §""*•"
•
1 1
1 1
f/TF :
•1 • 1.
1 1
1 a 'v ' .
i • i •
i a • i ' *
° 9' * ' eu] '*
• •• • •
-100,000 -10,000 .1,000 -100 -10 -101 10 100 1,000 10,000 100,000 1,000,000
(SSEM-AVSV/oc
Figure 3-8. Percent mortality versus SEM-AVS (A) and (ZSEM-AVSV/,,,. (B) for saltwater Held data without Bear
Creek and Jinzhou Bay (D), freshwater field data (V), freshwater spiked data (•), and saltwater spiked
data (•); silver data excluded. Vertical dashed lines are the 90% uncertainty bound Limits .(figure from
Di Toro et ah, 2000).
3-21
-------
Toxicitv of Metals in Sediments
uncertainty bounds should narrow and the percentages
of sediments where toxicity predictions are uncertain
should decrease. If the uncertainty bound analysis is
not conducted, and SEM-AVS>0.0 is used as proposed
in Sections 3.2 and 3.3, predictions of sediment toxicity
for the 267 spiked sediments are classified as uncertain
for 47.2rr of the sediments. Using the uncertainty
bounds on SEM-AVS of 1.7 to 120 umoL/g as described
in this section results in reduction in the percentage of
sediments where toxicity predictions are uncertain, to
34.1%. Use of (ESEM-AVSV/oc with uncertainty
bounds of 130 to 3.000 Mmol/g^ results in further
reduction in the percentage of sediments where toxicity
predictions are uncertain, to 25.5%.' Therefore, use of
the uncertainty limits of the (SSEM-AVS)/^.. approach
classifies 33.7% more sediments as toxic or nontoxic
than using the uncertainty limits of SEM-AVS, and 85%
more than use of SEM-AVS without uncertainty limits.
This improvement highlights the advantages of using
(SSEM-AV'S)//OC in assessing toxicity of metal-
contaminated sediments.
Use of (ISEM-AVS l/f^ uncertainty limits applies
to all the metals regardless of their identity. Figure 3-9
presents the spiked-sediment data categorized by
identity of the metal. The field-contaminated data
cannot be included because the identity of the metal
causing toxicity cannot be unambiguously determined.
There is no apparent difference for any of the metals in
the region of overlapping survival and mortality data
between I30and 3.000Jumol/goc.
It is interesting to note that organic carbon
normalization appears not to work for silver. The
spiked-sediment test data are presented in Figure 3- 10A
(Berry etal., 1999). Nofwhat there is almost a complete
overlap of mortality and no mortality data. This
suggests that organic carbon is not a useful
normalization for silver partitioning in sediments.
Perhaps this is not surprising because the role of sulfur
groups is so prominent in the qomplexation chemistry
of silver (Bell and Kramer. 1999).
To not depend on the identity of the metal is an
advantage in analyzing naturally contaminated
sediments in that it is difficult to decide which metal is
potentially causing the toxicity. Of course it can be
done using the sequence of solubilities of the metal
sulfides or interstitial metal concentrations (Oi Toro et
al.. 1992; Ankley etal., 1996). The metal-independent
method can be tested using the results of an
experiment with an equimolar mixture of cadmium,
copper, nickel, and zinc (see Figure 3- 10B). The area of
uncertainty falls within the carbon-normalized excess
SEM boundaries above.
3.4.5 Chronic Toxicity Uncertainty
The results of chronic toxicity tests with metals-
spiked sediments can also be compared to CSSEM-
AVSV/oc (Figure 3-11; Table 3-3). Note that Figure 3-11'
indicates a category for "predicted toxic." Significant
chronic effects were observed in only 1 of the 19
sediments, where the uncertainty analysis of acute
toxicity tests indicated that effects were not expected
at(2SEM-AVS)//oc <130Jumoi/goc. The concentration
in the sediment where chronic effects were observed
but not expected, i.e., (SSEM-AVSV/^. = 28 umol .
excess SEM/g^. The previous analysis of the results
of chronic toxicity tests using SEM-AVS indicated that
concentrations of SEM exceeded AVS in 7 of 19
nontoxic sediments. Sediment concentrations based
on (SSEM-AVS V/QJ, placed these sediments in the
uncertain toxicity category. Importantly, use of (SSEM-
AVSV/Qc to classify sediments resulted in six of these
same seven sediments being correctly classified as
probably nontoxic. Chronic effects were observed in
six of the seven sediments where predictions of effects
are uncertain.(130 to 3,000 Aimol/g,-^). This suggests
that chronic toxicity tests with sensitive benthic
species will be a necessary part of the evaluations of
sediments predicted to have uncertain effects.
3.4.6 Summary
The uncertainty bounds on SEM-AVS differences
and organic carbon-normalized excess SEM ((SSEM-
AVSy/Qc) can be used to identify sediments that are
likely to be toxic, are of uncertain toxicity, or are
nontoxic. Use of (SSEM-AVSV/Qj. as a correction
factor for excess SEM is attractive because it is based
on the theoretical foundation of equilibrium
partitioning. Likewise, it reduces the uncertainty of the
prediction of toxicity over that of SEM-AVS
differences.
3-22
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
>,
u
9
1XU
100
80
60
40
20
0
-20
-10,
Copper
-
-
_
o
- ^o v
Av
1
-
-
.
-
000 -100 0 100 10,000 1,000
120
100
80
60
40
20
0
-20
. Lead
-
•
-
ol o
- W
mil km, k..., ki.,,1
1
V| O
O
•
o
o
, ""o4" -
V
1
1.,,^ ,,,,^l,,nJ ., ,..J ,,,nJ MM.
-100
100
10,000 l.OQO.OOO
u
3
120
too
80
60
40
20
0
•20
•
•Mil •!!! 1 Mill 1 Will 1 |
Cadmium
•
-
•
-
•
~
- {r£~
• |T~^ T T
- Sf'^B^V
.
mi, •nil , mil — n , ,
f 1
1 •wip CD O
1 rf^
*J^
1
i T a
i
1 O
i a
'o
• '^ T
1
,,!„• ,t,«llll,« , UK lll« , 1MB
120
100
80
60
40
20
0
-20
.
. Nickel
•
-
'
.
•
—
' 00
• • a
•nil Mini •^tll) Irf11!!!
1
<» oo0oa
i 'a
i
i
i
. a
1
i
0 '
tf 11 1 1 1
1
urn '"f^ intH 'f"™ tiniv .
-
•
-
m
•
_
•
•
in
liU
100
80
60
40
20
0
'. Zinc
•
-
• £
-<*W
1
v bo o
1
7 i o
i
i
1,
10 1
1 0 |
V ' '
1 1
•
-
-
•
•
Figure 3-9. Percent mortality versus (SEMMtul-AVSV/oc for each metal in spiked sediment tests using Ampelisca
(o), CapiteOa (7), Ntanthts (Q), Lumbriculus (•), and Htlisoma (»). Vertical dashed lines are the 90%
uncertainty bound limits (figure from Di Toro et aL, 2000).
3-23
-------
Toxicitv of Metals in Sediments
120
100
80
60
40
20
0
. A: Silver
.
-
.
-
0 0
1
°p
• (j Q
o
Q
II II 1 1 1 Mil 1 1 1 1 I Illl 1 1 1 1 1 Illl 1 1 1 1
» oc
Illl Milt 1 111 Dill
bo
o
>
o
0
1 1 1 1 1 Illl 1
1, 1 1 1 Illl 1 Illl Illl 1 1 1 1 I III
o
.
^
-
.
•
.
•
1 1 ',
2
i 9
120
100
80
60
40
20
0
-?n
- B: Mixture
-
-
-
' •
*••• * ••
"
mi 1 1 i i km 1 1 i i • Inn i i i i Inn 1 1 i i 1
. ^
"
1 ill mill i i 1 1 mil
'
• •*
1 ,
Illl Illl 1 Illl Illl 1 1 1 1 Mil
-
-
-
b
"
1 J
i i urn i i i Mini i iii mi
-10,000 -1,000 -100 -10 0 10 100 1,000 10,000 100,000 1,000,000
(ZSEM-AVSV/oc
Figure 3-10. Percent mortality versus (SEMA|-AVSV/OC for silver (A) and (ISEM-AVSV/oc for a mixture experiment
using Cd, Cu, Ni, and Zn (B; see Berry et aL, 1996). Vertical dashed lines are the 90% uncertainty
bound limits determined from Figure 3-8B (figures from Di Toro et ai., 2000).
3-24
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Predicted
Toxic
• Experimental OEC
H Experimental NOEC
•
j
1
i
i
i
i
!
1
1
• » !
! i • • •
i
i
i
iiniii t i IIIIM i i mini i i mini 11 i ii IIHIII i i i HUB
uncertain
0
i i i inn i
<«-*>
i
i nun
-10000 -1000 -100 -10
10
100 1000 10000
Figure 3-11. Comparison of the chronic toxicity of sediments spiked with individual metals or metal mixtures
to predicted toxicity based on (SEM-AVS)//,,,. (data from Table 3-3). Horizontal dashed line
separates experimental observed effect concentrations (solid columns) from no observed effect
concentrations (shaded columns). Values at (SEM-AVS)//^ s 130 wmol/g^ are predicted to be
nontoxic. Values between 130 and 3,000 ^mol/g^ lie where the prediction of toxicity is uncertain.
and values greater than 3,000 Atmol/j^. are predicted to be toxic.
3-25
-------
Toxicitv of Metals in Sediments
Table 3-3. Test-specific data for chronic toxicity of freshwater and saltwater organisms compared to
(ISEM-AVS)//^.
(ZSEM-AVSX/oc1
Toxicity Test
MetaJ(s)
OEC(s)c
Colonization: '
Laboratory-saltwater
Cadmium
800. 2740
Reference
Lite Cvcle:
Leptocheirus
plumulosus
Chironomus tentans
Cadmium
Zinc
0.030
0.038
-117, -66.7, 26, 63.3
-68, -36.8, 168
297, 520
576, 847
DeWitt et al.. 1996
Sibleyetal.. 1996
Hansen et al., 1996b
Field-saltwater
Field-freshwater
Field- fresh water
Cadmium.
copper.
lead, nickel.
zinc
Cadmium
Zinc
0.002
0.079
0.111
-155. -30. 10
-0.92.1.08.4.30
-32.7. -31.8. -26.4,
-18.2,9".!
— Boothman et al..
2000
28 Hare et al.. 1994
— Liber etal.. 1996'
concentrations in bold type are those between 130 and 3.000 nmoUgQ^ for which the expectation of effects is
uncertain. Italics indicates concentrations where effects were observed but not expected.
NOECs = no observed effect concentrations): all concentrations where response was not significantly different from the control.
OECs = observed effect concentration(s); all concentrations where response was significantly different from the control.
3-26
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 4
Derivation of the ESGs for Metals
4.1 General Information
Section 4 of this document presents the technical
basis for establishing the ESG for cadmium, copper.
•.lead, nickel, silver, and zinc. The basis of the overall
approach is the use of EqP theory linked to the concept
of maintaining metal activity for the sediment interstitial
water system below concentrations that cause adverse
effects. Extensive toxicological concentration-response
data from short-term and chronic laboratory and field
experiments, with both marine and freshwater sediments
and a variety of species, indicate that' it is possible to
reliably predict absence of metal toxicity based on EqP
theory and derive ESGs for metals in sediments using
either of two approaches. The ESGs for the six metals
that collectively predicts absence of their toxicity in
sediments can be derived by .(a) comparing the sum of
their molar concentrations, measured as SEM, with the
molar concentration of AVS in sediments (solid-phase
AVS guideline); or (b) summing the measured interstitial
water concentrations of the metals divided by their
respective WQC FCVs (interstitial water guideline).
Lack of exceedence of the ESG based on either of these
two procedures indicates that metal toxicity should not
occur.
At present. EPA believes that the technical basis
for implementing these two approaches is supportable.
The approaches have bee^n presented to and reviewed
by the SAB (U.S. EPA. 1994a, 1995a, 1999b).
Additional research required to fully implement
other approaches for deriving art ESG for these metals
and to derive an ESG for .other metals such as mercury,
arsenic, and chromium includes the development of
uncertainty estimates; pan of this would include their
application to a variety of field settings and sediment
types. Finally, the ESG approaches are intended to
protect benthic organisms from direct toxicity
associated with exposure to metal-contaminated
sediments. They are not designed to protect aquatic
systems from metal release associated with, for
example, sediment suspension, or the transport of
metals into aquatic food webs. In particular, studies are
needed to understand the toxicological significance of
the biomagnification of metals that occurs when
predators consume benthic organisms that have
accumulated metals from sediments with more AVS than
SEM (Ankley, 19^6).
The following nomenclature is used in subsequent
discussions of the ESGs' derivation for metals. The
ESG for the metals is expressed in molar units because
of the molar stoichiomeiry of metal binding to AVS.
Thus, solid-phase constituents (AVS. SEM) are in
ptmol/g-dry weight. The interstitial water metal
concentrations are expressed in ,umol/L or Mg/L. either
as dissolved concentrations (M ] or activities (M:'|
(Stumm and Morgan. 1981). The subscripted notation.
Mr "is used to distinguish dissolved aqueous-phase
molar concentrations from solid-phase molar
concentrations with no subscript. For the combined »
concentration. [SEM,.], the units are Atmol of total metal
per-gram of dry weight sediment. Note also that when
(SEMA ] is summed and/or compared with AVS. one-
half the molar silver concentration is applied.
One final point should be made with respect to
nomenclature. The terms nontoxic and having no
effect are used only with respect to the six metals
considered in'this document. Toxicity of field-
collected sediments can be caused by other chemicals.
Therefore, avoiding exceedences of the ESG for metals
does not mean that the sediments are nontoxic. It only
ensures that the six metals being considered should not
cause direct toxicity to benthic organisms. Moreover.
as discussed in. detail below, exceedence of the
guidelines for trie six metals does not necessarily
indicate that metals will cause toxicity. For these
reasons, EPA strongly recommends the combined use "
of both AVS and interstitial water measurements;
toxicity tests; TEEs; chemical monitoring.in vertical,
horizontal, and temporal scales; and other assessment
methodologies as integral pans of any evaluation of
the effects of sediment-associated contaminants
(Ankley et al.. 1994; Lee et al.. 2000).
4.2 Sediment Guidelines for Multiple
Metals
It is neither sufficient nor appropriate to derive an
ESG that considers each metal separately, because
metals almost always occur as mixtures in field
sediments and metal-sulfide binding is interactive.
4-1
-------
Derivation of the ESG for Metals
4.2.1 AVS Guidelines
Results of calculations using chemical equilibrium
models indicate that metals act in a competitive manner
when binding to AVS. That is. the six metals—silver.
copper, lead, cadmium, zinc, and nickel— will bind to
AVS and be converted to their respective sulfides in
this sequence (i.e.. in the order ot' increasing solubility).
Therefore, they must be considered together. There
cannot be a guideline for just nickel, for example,
.because all the other metals may be present as metal
sulfides. and therefore, to some extent, as AVS. If these
other metals are not measured as a mixture, then the
ZSEM will be misleadingly small, and it might appear
that S[SEM]<[AVS] when in fact this would not be true
if all the metals are considered together. It should be
noted that EPA currently restricts this discussion to the
six metals listed above; however, in situations where
other sulfide-fotming metals (e.g., mercury) are present
at high concentrations, they also must be considered.
The equilibrium model used to derive the ESG for a
mixture of the metals is presented below (see Ankley et
al.. 1996, for details). If the molar sum of SEM for the
six metals is less than or equal to the AVS, that is. if
S (SEMJ <, [AVS]
where
= [SEMCd] + [SEMC
[SEMVi
Vi
then the concentrations of the mixtures of metals in the
sediment are acceptable for protection of benthic
organisms from acute or chronic metal toxicity.
4.2.2 Interstitial Water Guidelines
The application of the interstitial water guideline to
multiple metals is complicated, not by the chemical
interactions of the metals in the sediment-interstitial
water system (as in the gase with the AVS guideline),
but rather because of possible toxic interactions. Even
if the individual concentrations do not exceed the water
quality criteria continuous concentration (CCC) of each
metal presented in Table 4- 1 , the metals could exert
additive effects that might result in toxicity (Biesinger
et al.; 1986; Spehar and Fiandt, 1986; Enserink'et al..
1991; Kraaket al., 1994). Therefore, in order to address
this potential additivity, the interstitial water metal
concentrations are converted to interstitial water
guideline units (IWGUs). This conversion is done by
dividing the individual metal interstitial water
concentrations by their respective WQC FCVs and
summing these values for all the metals. IWGUs are
conceptually similar to toxic units; however, the term
IWGU was adopted because it is derived using the FC V.
Table 4-1. Water quality criteria (WQC) criteria continuous concentrations (CCC) based on the dissolved
concentration of metal*
Metal
Saltwater CCC
(Mg/U
Freshwater CCC
Cadmium
Copper
Lead
Nickel
' Silver
Zinc
. 9.3
3.1
8.1
8.2
CFC
81
O.S
0.79 l[e(12 ,
0 997re'08440"n(lunln«"|i*1 I643)i
NAe
0.
•"These WQC CCC values are for use in the interstitial water guidelines approach for deriving ESGs based on the dissolved metal
concentrations in interstitial water (U.S. EPA, 1995b).
bFor example, the freshwater CCC at a hardness of SO. 100, and 200 mg CaCO/L are 0.62. 1.0. and 1.7 Mg cadmium/L. 6.3. 10. and
20 ug copper/L; 1.0. 2.5. and 6.1 u% lead/L; 87. 160, and 280 Mg nickel/L: and 58. 100, and 190 Mg zinc/L.
CCF = conversion factor to calculate the dissolved CCC for cadmium from the total CCC for cadmium: CFs»l.l01672-{(ln
hardness)(0.041838)).
dThe saltwater CCC for copper is from U.S. EPA (199SO.
'The silver criteria are currently under revision to reflect water quality factors that influence the criteria such as hardness. DOC.
chloride, and pH. among other factors.
4-2
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
which is intended to be a "no effect" concentration
(i.e.. toxicity wpuld not usually be expected at 1.0
IWGUs). ' '
For freshwater sediments, the FCVs are hardness
dependent for all of the divalent metals under
consideration, and thus, need to be adjusted to the
hardness of the interstitial water of the sediment being '
considered. Because there are no FCVs for silver in
freshwater or saltwater, this approach is not applicable
to sediments containing significant concentrations of
silver (i.e.. ESEM>AV'S). Because silver has the
smallest solubility product (see Table 2-2) and the
greatest affinity for AVS. it would be the last metal to be
released from the AVS or the first metal to bind with
AVS. Therefore, it is unlikely that silver would occur in
the interstitial water of any sediment with measurable
AVS (Berry etal.. 1996).
For the ith metal with a total dissolved
concentration, [M. J, the IWGU is
(4-2)
where
[Mi.dJ _ [Mcn.dl [MCu.d]
[FCVCdd]
4.2.3 Summary
In summary, the sediment guidelines for these six
metals are not exceeded, and benthic organisms are
sufficiently protected, if the sediment meets either one
of the following guidelines.
S.-tSEMJ s [AVS]
or
.
'FCVj.d
(4-1)
(4-2)
If the AVS or interstitial water ESGs are exceeded.
there is reason to believe that the sediment migkt be
unacceptably contaminated by these metals. Further
evaluation and testing would, therefore, be necessary
to assess actual toxicity-and its.causal relationship to
the metals of concern. If data on the sediment-specific
SEM, AVS. and organic carbon concentrations are
available, the uncertainty bounds'for (ZSEM-AVS V/^
described in Section 3.4 could be used to further '.
classify sediments as those in which metals are not
likely to cause toxicity, metal toxicity predictions are
uncertain, or metal toxicity is likely. For sediments in
which toxicity is likely or uncertain, acute and chronic
tests with species that are sensitive to the metals
suspected to be of concern, acute and chronic
sediment TIEs, in situ community assessments, and
seasonal and spatial characterizations of the SEM, AVS,
and interstitial water concentrations would be
appropriate (Ankley etal., 1994).
4.3 Example Calculation of ESGs for
Metals and EqP-Based Interpretation
To assist users of these ESGs for mixtures of
metals, example calculations for deriving solid-phase
and interstitial water ESGs are provided in Table 4-2.
For each of the three sediments, the calculations began
with measured concentrations (in bold) of AVS (,ug/g).
SEM( (^g/g), and interstitial water metal 0/g/L). All
other values were calculated. The specific
concentrations in each of the these sediments were
selected to provide examples of how the chemical
measurements are used with the ESG to determine the
acceptability of a specific sediment and how the risks
of sediment-associated metals can be evaluated within
the technical framework of the EqP approach.
Sediments are arranged in the table in decreasing order
of their sulfide solubility product constants (see
Section 2.2.5).
Sediment A contains relatively high concentrations
of metals in the SEM, between 14.2and I6.5,ug/gfor
copper, lead, and zinc. However, because there is
sufficient AVS (0.96 ^mol/g) in the sediment, the solid-
phase ESG is -0.343 (Mmol/g). and there is no metal
detected in the interstitial water. This sediment is
acceptable for protection of benthic organisms from
direct toxicity of the metals in the sediment. Silver was
not measured in this sediment. However, because AVS
is present, any silver in the sediment is not of-
toxicological concern and none should occur in
interstitial water. One final consideration is the need for
detection limits for metals in the sediment that are
significandy below their respective WQC FCVs. For
this sediment there were no detectable metals in the
interstitial water and ZIWGU was <0.46.
4-3
-------
Derivation of the ESG for Metals
Table 4-2. ESGs for metal mixtures: Example calculations for three sediments
Sediment Concentration
Sediment Analyte ^S/g1
A
IS EM
B
AVS
SEMNi
SEMz,,
SEMCJ
SEMPb
SEMCu
SEM,g
= 0,6 17 umol/g: SEM-AVS
AVS
SEMNi
30.8
2.85
16.5
0.05
14.2
16.0
—
= -0.343
1310
34.0
SEMz,, 2630
ZSEM
C
ZSEM
SEMCJ
SEMPb
SEMCu
SEMA,
82.9
282
227
NDC
Mmol/g
0.96
0.048
0.25
0.001
0.068
0.25
—
umol/g
40.8
0.58
40.2
0.74
1.36
3.58
NDc
Interstitial Water Concentration
Metal (M(.d) Mg/L
— —
Nickel ND' (<0.8)
Zinc ND' (<5.0)
Cadmium ND' (<0.2)
Lead ND' (<0.7)
Copper ND' (<0.6)
Silver ~
— —
Nickel 4.8
Zinc 43.2
Cadmium ND' (<0.01)
Lead ND' (<0.10)
Copper NDC (<0.05)
Silver ND' (<0.01)
FCV" IWGU
—
3.2
81
9.3
8.1
3.1
-
—
160
100
1.0
2.5
11
—
= 46.5 umol/g; SEM-AVS = 5.71 Mmol/g
AVS
SEMNi
SEMZn
SEMca
SEMp,
SEMc,
SEMA|
= 10.28 Mmol/g; SEM-AVS
146
269
12.4
573
66.2
4.44
ND'
= 5.71 MI
4.57
4.58
0.19
5.12'
0.32
0.07
NDc
— —
Nickel 26~3'
Zinc 4.3
Cadmium 24.9
Lead ND' (<0.10)
Copper NDC (<0.05)
Silver ND' (<0.0i)
—
87
58
0.62
1.0
6.3
—
—
<0.10
<0.06
<0.02
<0.09
<0.19
-
ZIWGU <0.46
—
0.03
0.43
<0.01
<0.04
<0.005
—
ZIWGU -0.46
—
0.30
0.07
40.1
<0.10
<0.008
—
mol/g ZIWGU -40.47
1 Molecular weight): sulfur. 32.06: nickel. 38.7: zinc. 65.4; cadmium. 112: lead. 207: capper. 63.5: silver. 108.
b Saltwater sediment: sediment A. Freshwater sediment]: sediment B. interstitial hardness 100 mg/L; sediment C. 50 mg/L.
c NO = not detected.
Sediment B is from a superfund site heavily
contaminated with all of the metals (ESEM = 46.5
Atmol/g). but most severely with zinc (2,630 Mg/g)-
There is an excess of SEM in this sediment (SEM--AVS =
5.71 ,umol/g). Importantly for sediment B, the interstitial
concentrations of the metals were ail less than the
WQC FCVs and the ZIWGU was < 1.0 (-0.46).
Therefore, this sediment is acceptable for protection of
benthic organisms from direct toxicity of this mixture of
metals in the sediment. It should be noted that, if
interstitial metal concentrations had not been
quantified, the sediment would have exceeded the ESG
and additional testing would be advisable. A possible
explanation for the absence of significant metals in the
.4-4
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
interstitial water of this sediment is its higher organic
carbon concentration (fx = 0.05). The (ISEM-AVSV/^.
of 114 ^mol excess SEM/g^ for this sediment is,
therefore, predicted to be nontoxic because it is < 130
-moI excess SEM/g^. (see Section 3.4.4).
Sediment C is heavily contaminated with
approximately equimolar concentrations of cadmium
and nickel. It exceeds the ESG for metals for both solid
and interstitial water phases. The ESEM (10.2S ^mol/g)
exceeds the AVS (4.57 Mmol/g); therefore. SEM-AVS =
5.71 ^mol excess SEM/g. a concentration identical to
that of sediment B. Although lead and copper are
found in the sediment, they are not.found in detectable
concentrations in the interstitial wat6r. This is because
they have the lowest sulfide solubility product
constants and the sum of their-SEM concentrations
(0.39 -mol/g) is less than AVS. If the dry weight
concentrations of metals had been analyzed, silver and
additional copper and nickel might have been detected.
Silver will not be detected in the SEM or interstitial
water when AVS is present (see Section 3.2.1). Nickel.
cadmium, and zinc occur in interstitial water because in
the sequential summation of the SEMr concentrations
in order of increasing sulfide solubilities, the
concentrations of these metals exceed the AVS.
Therefore, these three metals are found in the SEM that
is not a metal sulfide and in the interstitial water, and
contribute to the SIWGU (-40.47) as well as to the
overall exposure of benthic organisms. Because only
cadmium concentrations exceed the WQC FCV, any
effects observed in toxicity tests or in faunal analyses
with this sediment should principally be a result of
cadmium. This sediment is low in organic carbon
concentration (TOC = 0.2%\fx = 0.002). The organic
carbon-normalized concentration (SSEM-AVS//^.) of
2.S55 .umoi excess SEM/gg,. was within the uncertainty
bounds of 130 to 3,000 ^mol excess SEM/g^,
suggesting that additional evaluations should be
conducted (see Section 3.4.4).
4.4 ESG for Metals vs. Environmental
Monitoring Databases
This section compares the ESG based on AVS or
IWGUs with chemical monitoring data from freshwater
and saltwater sediments in the United States. This
comparison of AVS-SEM and interstitial water
concentrations is used to indicate the-frequency of
sediments in the United States where metals toxicity is
unlikely. When data were available in the monitoring
programs, (ZSEM-AVS)//^ is used to indicate
sediments where toxicity is unlikely, likely, or uncertain.
When toxicity or benthic organism community health
data are available in conjunction with these
concentrations it is possible to speculate as to
potential causes of the observed effects. These data.
however, cannot be used to validate the usefulness of
the AVS approach because sediments that exceed the
guidelines are not always toxic, and because observed
sediment toxicity may be the result of unknown
substances.
4.4.1 Data Analysis
Three monitoring databases were identified that
contain AVS. SEM, and/^ information; one also had
data on concentrations of metals in interstitial water.
Toxicity tests were conducted on all.sediments from
these sources. The sources are the Environmental
Monitoring and Assessment Program (EMAP)
(Leonard et a}.. 1996a)! the National Oceanographic and
Atmospheric Administration National Status and
Trends monitoring program (NOAA NST) (Wolfe et al..
1994; Long et al.. 1995b. 1996), and the Regional
Environmental Monitoring and Assessment Program
(REMAP) (Adams et al., 1996).
4.4.1.1 Freshwater Sediments
The AVS and SEM concentrations in the 1994
EMAP database from the Great Lakes were analyzed by
Leonard et aJ. (1996a). A total of 46 sediment grab
samples and 9 core samples were collected in the
summer from 42 locations in Lake Michigan. SEM. AVS.
TOC, interstitial water metals (when sufficient volumes
were present), and 10-day sediment toxicity to the
midge C. tentans and the amphipod H. azteca were
measured in the grab samples (the concentrations are
listed in-Appendix A).
The AVS concentrations versus SEM-AVS
differences from Appendix A are plotted in Figure 4-1.
Grab sediment samples containing AVS concentrations
below the detection limit of 0.05 Aimol/g AVS are plotted
at that concentration. Forty-two of the 46 samples
(91 %) had SEM-AVS differences greater than 0.0.
Thiny-six of these had less than 1.0 ^mol of excess
SEM/g sediment; and none had over 5.8 ^mol excess
SEM/g sediment. Sediments with SEM concentrations
in excess of that for AVS have the potential to be toxic
because of rrfetals. However, the majority of sediments
with an excess of SEM had low concentrations of both
AVS and SEM.. For 20 of these; Lake Michigan
sediments, interstitial water metals concentrations were
measured. The sum of the IWGUs for cadmium, copper.
4-5
-------
Derivation of the ESG for Metals
Ofi
e
a
to
1
3
to
5
U
to
10
-10
-20
-30
-40
0.01
10
B
-10
0.01
0.1
10
100
0.1 1 10
Acid Volatile Sulfide (/zmpl/g)
100
Figure 4-1. SEM-AVS values versus AVS concentrations in EMAP-Great Lakes sediments from Lake Michigan.
Data are from surflcial grab samples only. Plot (A) shows all values; plot (B) has the ordinate limited to
SEM-AVS values between -10 and +10 wmoi/g.
4-6
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
lead, nickel, and zinc was always less than 0.4 (Leonard
et al.. 1996a). In 10-day toxicity tests using C. tentans
and H. a:teca. no toxicity was observed in 81% of the
21 sediments not exceeding the ESQ. Leonard et al.
11996ai concluded that when toxicity was observed it
was not likely from metals, because of the low
interstitial uater metals concentrations. These data
demonstrate the value of using both SEM-AVS and
[WGUs to evaluate the risks of metals in sediments.
4.4.1.2 Salnvater Sediments
', Saltwater data from a total of 398 sediment samples
from 5 monitoring programs representing the eastern
coast of the United States are included in Appendix B.
The EMA? Virginian Province database I U.S. EPA.
1996) consists, in part, of 127 sediment samples
collected from August to mid-September 1993 from
randomly selected locations in tidal rivers and small
and large estuaries from the Chesapeake Bay to
Massachusetts (Strobel et al., 1995). The NOAA data
are from Long Island Sound. Boston Harbor, and the
Hudson River Estuary. Sediments were collected from
63 locations in the coastal bays and harbors of Long
Island Sound in August 1991 (Wolfe etal., 1994).
Sediment samples from 30 locations in Boston Harbor
were collected in June and July 1993 (Long et al.. 1996).
Sediment samples from 38 locations in the Hudson
River Estuary were collected from March to May 1991
(Long etal.. 1995b). Sediment samples were collected
in the REMAP program from 140 locations from the
New York/New Jersey Harbor Estuary System (Adams
etal.. 1996). Allot" the above sediment grab samples
were from approximately the top 2 cm of undisturbed
sediment.
For saltwater sediments, the molar concentration of
AVS typically exceeds that for SEM (SEM-AVS sO.Q
Mfnol/g) for most of the samples across the entire range
of AVS concentrations (Figure 4-2). A total of 68 of the
398 saltwater sediments (17%) had an excess of metal.
and only 4 of the 68 (6%) had Over 2 Mmol excess
SEM/g. As AVS levels increase, fewer and fewer
sediments have SEM-AVS differences that are positive;
none occurred when AVS was >8.1 ^mol/g. Interstitial
water metal was not measured in these saltwater
sediments. Only 5 of the 68 sediments (7%) having
excess of up to 0.9 /imol SEM/g were toxic in 10-day
sediment toxicity tests with the amphipod A. abdita,
whereas 79 of 330 sediments (24%) having an excess of
AVS were toxic. Toxicity was not believed to be metals
related in the 79 toxic sediments where AVS was in
excess over SEM. Metals were unlikely the cause of
toxicity .in those sediments having an excess of SEM
because there was only sO.9 umol excess SEM/g.
Finally, the absence of toxicity in sediments having an
excess of SEM of up to 4.4 ^mol/g indicates significant
metal-binding potential over that of AVS in some
sediments. Organic carbon concentrations from 0.05%
to 15.2% (average 1.9%) provide for some of this
additional metal binding.
Organic carbon, along with SEM and AVS, was
measured in these 398 saltwater sediments. Therefore.
the (ZSEM-AVSV/oc concentrations of concern can be
compared with the organic carbon-normalized
concentrations of SEM-AVS differences (Figure 4-3).
No sediments containing an AVS concentration in
excess of 10 /*mol/g had an excess of SEM; that is. all
(SSEM-AVS.)//^ values were negative. Excess of SEM
relative to AVS became more common as sediment AVS
decreased. None of the sediments contained greater
than 130 ,umol excess SEM/g^.. the lower uncertainty
bound from Section 3.4. This indicates that metals
concentrations in'all of the sediments monitored in the
summer by EPA EMAP and REMAP and by NOAA are
below concentrations of concern for benthic organisms.
4.5 BioaccumuJation
The data appear to suggest that, for these
sediments collected from freshwater and marine
locations in the United States, direct toxicity caused by
metals in sediments is expected to be extremely rare.
Although this might be true, these data by themselves
are inconclusive. Importantly, it would be inappropriate
to use the data from the above studies to conclude that
metals in sediments are not a problem. In all of the
above studies, the sediments were conducted in the
summer when the seasonal biogeochemical cycling of
sulfur should produce the highest concentrations of
iron monosulftde, which might make.direct metal-
associated toxicity less likely than in the winter/spring
months. .Accurate assessment of the extent of the
direct ecological risks of metals in sediments requires
that sediment monitoring occur in the months of
minimum AVS concentration; typically, but not always.
in November to early May. These yet-to-be-conducted
studies must monitor, at a minimum, SEM, AVS./g,.,
interstitial water metal, and toxicity.
Bioaccumulation of metals from sediments when
SEM is less than AVS was not expected based on EqP
theory. However, there is a significant database that
de'monstrates that metals concentrations in benthic
organisms increase when metals concentrations in
4-7
-------
Derivation of the ESG for Metals
50
3JO
O
-50
^ -100
a
C/3 '
-ISO
-200
0.01
O EMAP
a REMAP
O NOAA
0.1
a
o
10
100
1000
10.0
-10.0
0.01
0.1 1 10
Acid Volatile Sulfide Cumol/g)
100
1000
Figure 4-2. SEM-AVS values versus AVS concentrations in EMAP-Estuaries Virginian Province (U.S. EPA, 1996);
REMAP-NY/NJ Harbor Estuary (Adams et al., 1996); NOAA NST-Long Island Sound (Wolfe et al.,
1994); Boston Harbor (Long et al., 1996); and Hudson-Raritan Estuaries (Long et al., 1995b). Plot A
shows all values; plot B has the ordinatt limited to SEM-AVS values between -10 and -t-10 umol/g (see
data in Appendix B).
4-8
-------
5000
A
s
do
"a
a
<*
~9
a
-5000
-10000
-15000 •
-20000
O EMAP
D REMAP
O NOAA
0.01
0.1
10
100
1000
500
-500
-1000
0.01
Acid Volatile Sulfide (umol/g)
1000
1000
Figure 4-3. (SSEM-AVSV/OC versus AVS concentrations in EMAP-Estuaries Virginian Province iU^. EPA. 1996);
REMAP-NY/NJ Harbor Estuary (Adams et al., 1996); NOAA NST-Long Island Sound (Wolfe et al..
1994); Boston Harbor (Long et al., 1996); and Hudson-Raritan Estuaries (Long et al., 199Sb). Plot A
shows all values; plot B has the ordinate limited to (ZSEM-AVS)//^ values between -10 and +10 ^mol/g
(see data in Appendix B).
4-9
-------
Derivation of the ESG for Metals
sediments on a dry weight basis increase (Ankley,
1996). This has caused considerable debate (Lee et al..
2000) because it suggests that metal bioavailability may
be related to dry weight metals concentrations,.and if
the increase in bioaccumulated metal is related to
effects, then effects may be related to dry weight metals
concentrations. Most importantly, these studies, and
all other AVS-related testing, has overwhelming
demonstrated that toxic effects of metals are absent in
sediments when SEM is less than AVS, even when
'bioaccumuiation is observed, and that toxicity is not
related to dry weight metals concentrations. This
suggests that the bioaccumulated metals may not be
toxicologically available or of sufficient concentration
•in the organism to cause effects. In addition, these
metals do not biomagnify to higher trophic levels in
aquatic ecosystems (Suedel et al., 1994). Therefore, an
ESG based on the difference between the
concentrations of SEM and AVS is appropriate for
protecting benthic organisms from the direct effects of
sediment-associated metals, and not for protecting
against metal bioaccumuiation.
4-10
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 5
Sampling and Analytical Chemistry
5.1 General Information
This section provides guidance on procedures for
sampling, handling, and analysis of metals in
sediments, and on the interpretation of data from the
. sediment samples that are needed if the assessments of
the risks of sediment-associated metals are to be
appropriately based on the EqP methodology. The
design of any assessment should match the goal of the
specific assessment and how evaluation tools such as
ESGs are to be applied. The EPA program-specific
guidance on how ESGs for metals and nonionic organic
chemicals will be implemented is detailed in the
"Implementation Framework for Use of Equilibrium
Partitioning Sediment Guidelines (ESGs)" (U.S. EPA,
2000c).
Results of the short- and long-term laboratory and
field experiments conducted to date using sediments
spiked with individual metals and mixtures of metals
represent convincing support for the conclusion that
absence (but not necessarily presence) of metal toxicity
can be reliably predicted based on metal-sulfide
relationships or interstitial water metal concentrations.
In contrast, much confusion exists on how to use this
convincing evidence to interpret the significance of
metals concentrations in sediments from the field.
Using these observations as a basis for predicting
metal bioavailability, or deriving an ESG. raises a
number of conceptual and practical issues related to
sampling, analytical measurements, and effects of
additional binding phases. Many of these were
addressed by Ankley et al. (1994). Those most salient
to the proposed derivation of the ESGs are described
below.
5.2 Sampling and Storage
Accurate prediction of exposure of benthic
organisms to metals is critically dependent on sampling
appropriate sediment horizons at appropriate times.
This is because of the relatively high rates of AVS
oxidation caused by natural processes in sediments and
the requirement that oxidation must be avoided during
sampling of sediments and interstitial water. In fact, the
labile nature of iron monosulfides has led some to
question the practical utility of using AVS as a basis for
an EqP-denved ESG for metals (Luoma and Carter, 1993;
Meyer etal., 1994). For example, there have been many
observations of spatial (depth) variations in AVS
concentrations, most of which indicate that surficial
AVS concentrations are less than those in deeper
sediments (Boothman and Helmstetter. 1992; Howard
and Evans. L993; Brumbaugh etal., L 994; Hare etal..
1994; Besser et al.. 1996; Hansen et al.. 1996b: Leonard
et al.. 1996a; Liber et al.. 1996). This is likely because of
oxidation of AVS (principally FeS) at the sediment
surface, a process enhanced by bioturbation (Peterson
etal.. 1996).
•In addition to varying with depth, AVS can vary
seasonally. For example, in systems where overlying
water contains appreciable oxygen during cold-weather
months, AVS tends to decrease, presumably because of
a constant rate of oxidation of the AVS linked to a
decrease in its generation by.sulfate-reducing bacteria
(Herlihy and Mills, 1985; Howard and Evans. 1993;
Leonard etal.. 1993). Because of potential temporal and
spatial variability of AVS, it appears that the way to
avoid possible underestimation of metal bioavailability
is to sample the biologically "active" zone of sediments
at times when AVS might be expected to be present at
low concentrations. EPA recommends that, at a
minimum, AVS and SEM measurements be made using
samples of the surficial (0 to 2.0 cm) sediments during
the period from November to early May. Minimum AVS
concentrations may not always occur during cool-
weather seasons; for example, systems that become
anaerobic during the winter can maintain relatively
large sediment AVS concentrations (Liber et al., 1996).
Therefore, AVS, SEM, and interstitial metal
concentrations may need to be determined seasonally.
Importantly, the biologically active zones of some
benthic communities may be within only the surficial
first few millimeters of the sediment, whereas other
communities may be biologically active at depths up to
a meter. In order to determine the potential for exposure
to metals, sediment and interstitial water samples from
multiple sediment horizons may be required.
The somewhat subjective aspects of these
sampling recommendations have been of concern.
Multiple sediment samples are necessary because of
the dynamic nature of the metal-binding phases in
5-1
-------
Sampling and Analytical Chemistry
sediments. Depending on the depth of bioturbation.
the possible oxidation rates of specific metal sulfides,
and the extent of possible metal concentrations, the
horizontal and vertical resolution of the needed
monitoring is likely to be site specific. Even if neither
of the sediment guidelines is violated in extensive
monitoring programs, metals concentrations on a dry
weight basis may be high and widely distributed. This
may be a good reason to conduct monitoring studies to
determine the extent of metal bioaccumulation in
benthic food chains. Furthermore, if the ultimate fate of
'the sediments is unknown, risk assessments to evaluate
future risks caused by dynamic processes may be
desirable.
Research suggests that the transient nature of AVS
may be overstated relative to predicting the fate of all
metal-sulfide complexes in aquatic sediments.
Observations from the Duluth EPA laboratory made in
the early 1990s indicate that AVS concentrations in
sediments contaminated by metals such as cadmium
and zinc tended to be elevated over concentrations
typically expected in freshwater systems (G.T. Ankley,
U.S. EPA. Duluth. MN, personal communication). The
probable underlying basis for these observations did
not become apparent, however, until a recent series of
spiking and metal-sulfide stability experiments. The
field colonization study of Liber et al. (1996)
.demonstrated a strong positive correlation between the
amount of zinc added to test sediments and the
resultant concentration of AVS in the samples. In fact,
the initial design of their study attempted to produce
test sediments with as much as five times more SEM^
(nominal) than AVS; however, the highest measured
SEMZn/AVS ratio achieved was only slightly larger than
1. Moreover, the expected surficial depletion and
seasonal variations in AV'S were unexpectedly low in
the zinc-spiked sediments. These observations
suggested that zinc sulfide, which composed the bulk
of AVS in the spiked sediments, was more stable than
the iron sulfide present in the control sediments. The
apparent stability of other metal sulfides versus iron
sulfide also has been noted in laboratory spiking.
experiments with freshwater and saltwater sediments
(Leonard et al., 1995; DeWitt et al.'. 1996; Hansen et al..
1996b; Peterson et al., 1996; Sibley et al., 1996;
Boothmanetal.,2000).
In support of these observations, metal-sulfide
oxidation experiments conducted by Di Toro et al.
(1996b) have confirmed that cadmium and zinc form
more stable sulfide solid phases than iron. If this is
also, true for sulfide complexes of copper, nickel, silver.
and lead, the issue of seasonal/spatial variations in
AVS becomes of less concern because most of the
studies evaluating variations in AVS have focused on
iron sulfide (i.e., uncontaminated sediments). Thus,
further research concerning the differential stability of
metal sulfides, from both temporal and spatial
perspectives, is definitely warranted.
5.2.1 Sediments
At a minimum, sampling of the surficial 2.0 cm of
sediment between November and early May is
recommended. A sample depth of 2.0 cm is appropriate
for monitoring. However, for instances such as
dredging or in risk assessments where depths greater
than 2 cm are important, sample depths should be
planned based on particular study needs. Sediments
can be sampled using dredges, grabs, or coring, but
mixing of aerobic and anaerobic sediments must be
avoided because the trace metal speciation in'the
sediments will be altered (see Bufflap and Allen. 1995,
for detailed recommendations to limit sampling
artifacts). Coring is generally less disruptive, facilitates
sampling of sediment horizons, and limits potential
metal contamination and oxidation if sealed PVC core
liners are used.
Sediments not immediately analyzed for AVS and
SEM must be placed in sealed airtight glass jars and
refrigerated or frozen. Generally, enough sediment
should be added to almost fill the jar. If sediments are
stored this way, there will be little oxidation of AVS
even after several weeks.. Sampling of the stored
sediment from the middle of the jar will further limit
potential effects of oxidation on AVS. Sediments
experiencing oxidation of AVS during storage will
become less black or grey if oxidized. Because the rate '
of metal-sulfide oxidation is markedly less than that of
iron sulfide, release of metal during storage is unlikely..
5.2.2 Interstitial Water
Several procedures are available to sample
interstitial water in situ or ex situ. Carignan et al. (1985)
compared metals concentrations in interstitial water
obtained by ex situ centrifugation at 11,000 rpm .
followed by filtration (0.45 Mm and 0.2 or 0.03 Mm) and
in situ diffusion samplers with 0.2 Mm polysulfone
membranes. For the metals of concern in this guideline
document, concentrations of nickel and cadmium were
equivalent using both methods, and concentrations of
copper and zinc were higher and more variable using
5-2
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
centnfugation. They recommended using in situ
dialysis for studying trace constituents in sediments
because of its inherent simplicity and the avoidance of
artifacts that can occur with the handling of sediments
in the laboratory.
More recently, Bufflap and Allen (1995) reviewed
four procedures for collection of interstitial water for
trace metals analysis. These included ex situ
squeezing, centrifugation. in situ dialysis, and suction
filtration. These authors observed that each method
has its own advantages and disadvantages.
Importantly, interstitial water must be extracted by
centrifugation or squeezing in an inert atmosphere until
acidified, because oxidation will alter metal speciation.
Artifacts may be caused by temperature changes in ex
situ methods that may be overcome by maintaining
temperatures similar to those in in situ methods.
Contamination of interstitial water by fine panicles is
.important in all methods, because differentiation of
paniculate and dissolved metal is a function of the pore
size of the filter or diffusion sampler membrane. The
use of 0.45 .urn filtration, although an often accepted
definition of /'dissolved" metals, may result in
differences from laboratory to laboratory. Use of
suction filtration devices is limited to coarser
sediments, and they do not offer depth resolution.
Use of diffusion samplers is hampered by the time
required for equilibrium (7 to 14 days) and the need for
diver placement and retrieval in deep waters.
Acidification of interstitial water obtained by diffusion
or from suction filtration must occur immediately to limit
oxidation. Bufflap and Allen (1995) conclude that in
situ techniques have less potential for producing
sampling artifacts than ex situ procedures. They
concluded that, of the in situ procedures, suction
filtration has the best potential for producing artifact-
free interstitial water samples directly from the
environment. Of the ex situ procedures, they
concluded that centnfugation under a nitrogen.
atmosphere followed immediately by filtration and
acidification was the simplest technique likely to result
in an unbiased estimate of metal concentrations in
interstitial water. At present. EPA recommends
filtration of the surface water through 0.40 to 0.45 um
polycarbonate filters to better define that fraction of
aqueous metal associated with toxicity (Prothro. 1993).
This guidance applies to interstitial water. Thurman
(1985) equates the organic carbon retained on a 0.45 urn
glass-fiber filter to suspended organic carbon, so that
this filtration procedure under nitrogen atmosphere
followed immediately by acidification is acceptable for
interstitial waters. However, in studies comparing
collection and processing methods for trace metals.
sorption to filter membranes or the filtering apparatus
does occur (Schults et al.. 1992). These authors later
presented a method combining longer centrifugation
times with a unique single-step interstitial water
withdrawal procedure that has potential for minimizing
metal losses by eliminating the need for filtration
(Ozretich and Schults, 1998).
EPA recommends use of dialysis samplers to
obtain samples of interstitial water for comparison of
measured concentrations of dissolved metals with
WQC. This is primarily because diffusion samplers
obtain interstitial water with the proper in situ
geochemistry, thus limiting artifacts of ex situ sampling.
Furthermore. EPA has found that in shallow waters.
where contamination of sediments is most likely,
placement of diffusion samplers is easily accomplished
and extended equilibration times are not a problem.
Second, EPA recommends use of centrifugation under
nitrogen and 0.45 urn filtration using polycarbonate
filters for obtaining interstitial water from sediments in
deeper aquatic systems. Care must be taken to ensure
that filters or the filter apparatus do not remove metal
from or add metal to the interstitial water sample to be
analyzed. Perhaps most imponantiy. the extremely large
database comparing interstitial metals concentrations
with organism responses from spiked- and field-
sediment experiments in the laboratory has
demonstrated that, where the IWTU concept predicted
that metals concentrations in interstitial water should
not be toxic, toxicity was not observed when either
dialysis samplers or centrifugation were used (Berry et
al., 1996; Hansen et al., 1996a). Therefore, it is likely
that when either methodology is used to obtain
interstitial water for comparison with WQC. if metals
concentrations are below 1.0 IWGU. sediments should
be acceptable for protection of benthic organisms. The
exception is for some silver-spiked freshwater and
saltwater sediments that were toxic in spite of the
absence of interstitial silver. It is for this reason that
IWGUs are not used as ESGs for silver (see Sections
4.2. land 4-2.2).
5.3 Analytical Measurements
An important aspect to deriving ESG values is that
the methods necessary to implement the approach must
be reasonably standardized or have been demonstrated
to produce results comparable to those of standard
.5-3
-------
Sampling and Analytical Chemistry
methodologies. From the standpoint of the proposed
metals ESGs. a significant amount of research has gone
into defining methodologies to obtain interstitial water
and sediments (see Section 5.2 above), to extract SEM
and AVS from sediments, and to quantify AVS, SEM.
and the metals in interstitial water.
5.3.1 Acid Volatile Sulfide
The SEM/AVS extraction method recommended by
EPA is that of Allen et al. (1993). In terms of AVS
quantification, a number of techniques have been
successfully utilized, including gravimetric (Di Toro et
al.. 1990; Leonard etal., 1993),colorimetric(Comwell
and Morse, 1987), gas chromatography-
photoionization detection (Casks and Crecelius. 1994;
Slotton and Reuter, 1995), and specific ion electrodes
(Boothman and Helmstetter. 1992; Brouwer and
Murphy. 1994; Brumbaugh etal.. 1994; Leonard etal..
1996b). Allen et al. (1993) report a detection limit for
50<7c accuracy of 0.01 ^mol/g for a 10 g sediment sample
using the colorimetric method. Based on several
studies. Boothman and Helmstetter (1992) report a
detection limit of 1 ^mol AVS, which translates to 0.1
Mmol/g dry weight for a 10 g sediment sample using the
ion specific electrode method.
5.3.2 Simultaneously Extracted Metals
SEMs are operationally defined as metals
extracted from sediment into solution by the AVS
extraction procedure. The dissolved metals in this
solution are also operationally defined as the metal
species that pass through filter material used to remove
the residual sediment. Common convention defines
"dissolved" as metal species <0.45 Mm in size. SEM
concentrations measured in sediments are not
significantly different, however, using Whatman #1
filter paper alone (
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 6
Guidelines Statement
The procedures described in this document
indicate that, except possibly where a locally,
commercially, or recreationally important species is very
sensitive, benthic organisms should be acceptably
protected in freshwater and saltwater sediments if at
least one of the following two conditions are satisfied:
the sum of the molar concentrations, of SEM cadmium,
copper, lead, nickel, silver, and zinc is less than or equal
to the molar concentration of AVS (Section 6.1), or the
sum of the dissolved interstitial water concentration of
cadmium, copper, lead, nickel, and zinc divided by their
respective WQC FCV is less than or equal to 1.0
(Section 6.2). The AVS guideline is intended to apply
to sediments having >0.1 ymol AVS/g. The two
conditions are detailed in Section 4.2 and are repeated
below.
6.1 AVS Guideline
S [SEMJ s [AVS]
where
S, [SEM.,] = [SEMCJ] H- [SEMCu] H- [SEMn] + [SEMN,]
6.2 Interstitial Water Guideline
[MCu.dl
[FCVCddj [FCVClLdJ
[MNi.d]
[FCVNiidj
It is repeated here that the interstitial water
guideline applies only to the five metals: cadmium.
copper, lead, nickel, and zinc. Silver is not included in
this guideline because the FCV for silver is not
available.
The ESG approaches are intended to protect
benthic organisms from direct toxicity associated with
exposure to metal-contaminated sediments. They are
not designed to protect aquatic systems from metals
release associated, for example, with sediment
suspension, or the transport of metals into the food
web from either sediment ingestion or ingestion of
contaminated benthos. Furthermore, these ESGs are
not intended to protect against additive or synergistic
efforts of other contaminants or bioaccumulative
effects to aquatic life, wildlife, or humans.
6-1
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Section 7
References
Adams WJ. Kimerle RA, Mosher RG. 1985. Aquatic
safety assessment of chemicals sorbed to sediments. In
Card well RD. Purdy R. Bahner RC, eds, Aquatic
Toxicology and Hazard Assessment: Seventh
Symposium. STP 854. American Society for Testing and
Materials. Philadelphia. PA. pp 429-453.
Adams DA. O'Connor JS. WeisbergSB. 1996. Sediment
quality of the NY/NJ Harbor System—An investigation
under the Regional Environmental Monitoring and
Assessment Program (R-EMAP). Draft Final Report.
Division of Environmental Science and Assessment,
U.S. Environmental Protection Agency, Edison. NJ.
Allen HE. Hal RH. Brisbin TD. 1980. Metal speciation
effects on aquatic toxicity. Environ Sci Technol 14:441-
443.
Allen HE, Fu G. Deng B. 1993. Analysis of acid-volatile
sulfide (AVS) and simultaneously extracted metals
(SEM) for the estimation of potential toxicity in aquatic
sediments. Environ Toxicol Chem 12:1-13.
Anderson DM. Morel FM. 1978. Copper sensitivity of
Gonyaulax tamarensis. Limnol Oceanogr 23:283-295.
Ankley GT 1996. Evaluation of metal/acid volatile
sulfide relationships in the prediction of metal
bioaccumulation by benthic macroinvertebrates.
Environ Toxicol Chem 15:2138-2146.
Ankley GT, Di Toro DM, Hansen D, Berry WJ. 1996.
Technical basis and proposal for deriving sediment
quality criteria for metals. Environ Toxicol Chem
15:2056-2066.
. Ankley GT. Di Toro DM, Hansen DJ, Mahony JC,
Swartz RC, Hoke RA,.Thomas NA. Garrison AW, Allen
HE, Zarba CS. 1994. Assessing the potential
bioavailability of metals in sediments: A proposed
approach. Environ Manage 18:331-337.
Ankley GT. Mattson VR. Leonard EN, West CW,
Bennett JL. 1993. Predicting the acute toxicity of copper
in freshwater sediments: Evaluation of the role of acid-
volatile sulfide. Environ Toxicol Chem 12:315-320.
Ankley GT. Phipps GL, Leonard EN, Benoit DA.
Mattson VR, Kosian PA, Cotter AM, Dierkes JR.
Hansen DJ. Mahony JD. 1991. Acid volatile sulfide as a
factor mediating cadmium and nickel bioavailability in
contaminated sediments. Environ Toxicol Chem
10:1299-1307.
Barrow NJ, Ellis AS. 1986a. Testing a mechanistic
model. III. The effects of pH on fluoride retention by a
soil. J Soil Sci 37:287-295.
Barrow NJ, Ellis AS. 1986b. Testing a mechanistic
model. IV. Describing the effects of pH on zinc retention
by a soil. J Soil Sci 37:295-302.
Barrow NJ. Ellis AS. 1986c. Testing a mechanistic
model. V. The points of zero salt effect for phosphate
retention and for acid/alkali titration of a soil. / Soil Sci
37:303-310.
Bell RA, Kramer JA. 1999. Structural chemistry and
geochemistry of silver-sulfur compounds: Critical
review. Environ Toxicol Chem 18:9-22.
Bergman HL, Dorward-King EJ. 1997. Reassessment of
Metals Criteria for Aquatic Life Protection. Society of
Environmental Toxicology and Chemistry, Pensacola,
FL
BernerRA. 1967. Thermodynamic stability of
sedimentary iron sulfides. Am J Sci 265:773-785.
Berry WJ, Cantwell M, Edwards P, Serbst JR. Hansen
DJ. 1999. Predicting the toxicity of sediments spiked
with silver. Environ Toxicol Chem 18:40-48.
Berry WJ, Hansen DJ, Mahony JD, Robson DL. Di Toro
DM, Shipley BP. Rogers B, Corbin JM. Boothman WS.
1996. Predicting the toxicity of metals-spiked laboratory
sediments using acid-volatile sulfide and interstitial
water normalization. Environ Toxicol Chem 15:2067-
2079.
Besser JM, Ingersoll CG. Giesy JP. 1996. Effects of
spatial and temporal variability of acid volatile sulfide
on the bioavailability of copper and zinc in freshwater
sediments. Environ Toxicol Chem 15:286-293.
7-1
-------
References
Biesinger KE. Chnstensen CM. Fiandt JT. 1986. Effects
of metal salt mixtures on Daphnia magna reproduction.
Ecotoxicol Environ Saf 11:9-14.
Boothman WS. Hansen DJ. Berry WJ. Robson D,
Helmstefter A. Corbin J. Pratt SD. 2000. Variation of
acid volatile sult'ides and metals and biological
response in field-exposed sediments. Environ Toxicol
Chem (in press).
Boothman WS. Helmstetter A. 1992. Vertical and
seasonal variability of acid volatile sulfides in marine
sediments. Research Report. Environmental Monitoring
and Assessment Program, U.S. Environmental
Protection Agency, Narragansett, RI.
Boulegue-J, Lord III CJ, Church TM. 1982. Sulfur
speciation and associated trace metals (Fe, Cu) in the
interstitial waters of Great Marsh, Delaware. Geochim
Cosmochim Acta 46:453-464.
Brouwer H. Murphy TP. 1994. Diffusion method for the
determination of acid-volatile sulfides (AVS) in
sediment. Environ Toxicol Chem 13:1273-1275.
Brumbaugh WG. Ingersoll CG. Kemble NE. May TW,
Zajcek JL. 1994. Chemical characterization of sediments
and interstitial water from the upper Clark Fork River
and Milltown Reservoir, Montana. Environ Toxicol
Oi
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Di Toro DM. Mahony JJ. Hansen DJ. Scott KJ. Hicks
MB. Mayr SM. Redmond MS. 1990. Toxicity of
cadmium in sediments: The role of acid volatile sulfide.
Environ Toxicol Chem 9:1487-1502.
Di Toro DM. Zarba CS. Hansen DJ. Berry WJ, Swartz
RC. Cowan CE. Pavlou SP. Allen HE. Thomas NA,
Paquin PR. 1991. Technical basis for establishing
sediment quality criteria for nonionic organic chemicals
using equilibrium partitioning. Environ Toxicol Chem
10:1541-1583.
Dzombak DA. Morel FMM. 1990. Surface
Complexaiion Modeling. Hydrous Ferric Oxide. John
Wiley. New York. NY.
Emerson S. Jacobs L, Tebo B, 1983. The behavior of
trace metals in marine anoxic waters: Solubilities at the
oxygen-hydrogen sulfide interface. In Wong CS. ed.
Trace Metals in Sea Water. Plenum Press, New York.
NY
Enserink EL. Maas-Diepeveen JL. van Leeuwen CS.
1991. Combined toxicity of metals: An ecotoxicological
evaluation. Water Res 25:679-687.
Goldhauber MB. Kaplan ER. 1974. The sulfur cycle. In
Goldberg ED. ed. The Sea—Marine Chemistry, Vol 5.
John Wiley. New York. NY, pp 569-655.
Green AS. Chandler GT. Blood ER. 1993. Aqueous-,
interstitial water-, and sedirnent-phase cadmium:
Toxicity relationships for a meiobenthic copepod. .
Environ Toxicol Chem 12:1497-1506.
Hansen. DJ. 1995. Assessment tools that can be used
for the National Sediment Inventory. Memorandum to
C. Fox. U.S. Environmental Protection Agency, Office of
Water, February 28,1994,
Hansen DJ. Berry WJ. Mahony JD. Boothman WS,
Robson DL. Ankley GT. Ma D, Yan Q, Pesch CE. 1996a.
Predicting toxicity of metal-contaminated field
sediments using interstitial water concentrations of
metals and acid-volatile sulfide normalization. Environ
Toxicol Chem 15:2080-2094.
Hansen DJ, Mahony JD, Berry WJ. Benyi SJ. Corbin
JM. Pratt SD, Able MB. I996b. Chronic effect of
cadmium in sediments on colonization by benthic
marine organisms: An evaluation of the role of
interstitial cadmium and acid-volatile sulfide in
biological availability. Environ Toxicol Chem 15:2126-
2137.
Hare L, Carignan R. Huerta-Diaz MA. 1994. An
experimental study of metal toxicity and accumulation
by benthic invertebrates': Implications for the acid
volatile sulfide (AVS) model. Limnol Oceanogr
39:1653-1668.
Herlihy AT. Mills AL. 1985. Sulfate reduction in
freshwater sediments receiving acid mine drainage.
Appl Environ Microbiol 49:179-186.
Hesslein RH. 1976. An in situ sampler for close interval
interstitial water studies. Limnol Oceanogr 30:221 -227.
Hoke RA. Ankley GT. Cotter AM, Goldenstein'T.
Kosian PA. Phipps GL. Vandermeiden FM. 1994.
Evaluation of equilibrium partitioning theory for
protecting acute toxicity of field-collected sediments
contaminated with DDT. DDE. and ODD to the
amphipod Hyalella az.teca. Environ Toxicol Chem
13:157-166.
Howard DE, Evans RD. 1993. Acid-volatile sulfide
(AVS) in a seasonally anoxic mesotrophic lake:
Seasonal and spatial changes in sediment AVS. Environ
Toxicol Chem 12:1051 -1058.
Ingersoll CO, Haverland PS, Brunson EL, Canfield TJ.
Dwyer FJ, Henke CE. Kemble NE. Mount DR. Fox RG.
1996. Calculation and evaluation of sediment effect
concentrations for the amphipod Hyalella a-teca and
the midge Chironomus riparius. J Great Lakes Res
22:602-623,
Jenne EA. 1968. Controls on Mn, Fe, Co. Ni. Cu. and Zn
concentrations in soils and water—The significant role"
of hydrous Mn and Fe oxides. In Advances in
Chemistry. American Chemical Society. Washington.
DC, pp 337-387.
Jenne EA. 1977. Trace element sorption by sediments
and soil—Sites and processes. In Chappell W, Petersen
K, eds. Symposium on Molybdenum in the
Environment, Vol 2. Marcel Dekker, New York. NY. pp
425-553.
7-3
-------
References
Jenne EA. Di Toro DM. Allen HE. Zarba CS. 1986. An
activity-based model for developing sediment criteria
for metals: A new approach. Chemicals in the
Environment. International Conference. Lisbon.
Portugal.
Kemble NE. Brumbaugh WG. Brunson EL, Dwyer FJ,
' Ingersoll CG. Mounds DP. Woodward DF. 1994. Toxicity
of metal-contaminated sediments from the upper Clark
Fork River. Montana, to aquatic invertebrates and fish
•in laboratory exposures. Environ Toxicol Chem
13:1985-1997.
Kemp PF; Swartz-RC. 1986. Acute toxicity of interstitial
and particle-bound cadmium to a marine infaunal
amphipod.
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Luoma SN. Carter JL. 1993. Understanding the toxicity
of contaminants in sediments: Beyond the bioassay-
based paradigm. Environ Toxicol Chem 12:793-796.
MacDonald DD. Carr RS. Calder FD, Long ER, Ingersoll
CG. 1996. Development and evaluation of sediment
quality guidelines for Florida coastal waters.
Ecotoxicol 5:253-278.
Mahony JD. Di Toro DM. Gonzalez AM. Curto M. Dilg
M. 1996. Partitioning of metals to sediment organic
•carbon. Environ ToxicolChem 15:2187-2197.
Mahony JD. Di Toro DM. Koch R. Berry W, Hansen D.
1993. Vertical distfibution of AVS and SEM in bedded
sediments, biological implications and the role of metal
sulfide oxidation kinetics. Abstract. 14th Annual
Meeting. Society of Environmental Toxicology and
Chemistry. Houston, TX. November 14-18.1993. p 46.
Meyer JS. Davison W. Sundby B, Oris JT, Lauren DJ,
Fbrstner U. Hong J. Crosby DG. 1994. The effects of
variable redox potentials. pH and light on
bioavailability in dynamic water-sediment
environments. In Hamelink JL. Landrum PF, Bergman
HL. Benson WH, eds. Bioavailability: Physical,
Chemical and Biological Interactions. CRC Press,
Boca Raton. FL. pp 155-170.
Morse JW. Millero FJ. Cornwell JC, Rickard D. 1987.
The chemistry of the hydrogen sulfide and iron sulfide
systems in natural waters. Eanh Sci Rev 24:1-42.
National Academy of Sciences (NAS). 1973. Water
quality criteria, 1972. EPA-R3-73-033. National Academy
of Sciences, U.S. Environmental Protection Agency,
Washington, DC. • .
Nelson DW. Sommers LE. 1996. Total carbon, total
organic carbon and organic matter. In Sparks D, ed,
Methods of Soil Analysis: Part 3. Chemical Methods.
Soil Science Society of America and American Society
of Agronomy, Madison, WI.
Oakley SM. Williamson KJ, Nelsop PO. 1980. The
geochemical partitioning and bioavailability of trace
m'etals in marine sediments. PhD Thesis. Water
Research Institute. Oregon State University, Corvallis,
OR.
Ozretich RJ. Schults DW. 1998. A comparison of
interstitial water isolation methods demonstrates
centrifugation with aspiration yields reduced losses of
organic constituents. Chemosphere 36:603-615.
Pankow JF. 1979. The dissolution rates and mechanisms
of tetragonal ferrous sulfide (mackinawite) in anoxic
aqueous systems. PhD thesis. California Institute of
Technology, Pasadena. CA.
Persaud D. Jacqumagi R, Hayton A. 1989. Development
of provincial sediment quality guidelines. Draft Report.
Water Resources Branch, Ontario Ministry of the
Environment. Toronto, ON, Canada.
Pesch CE, Hansen DJ. Boothman WS. Berry WJ.
Mahony JD. 1995. The role of acid volatile sulfide and
interstitial water metal concentrations in determining
bioavailability of cadmium and nickel from
contaminated sediments to the marine polychaete
Neanthes arenaceodentata. Environ Toxicol Chem
14:129-141.
Peterson GS, Ankley GT, Leonard EN. 1996. Effect of
bioturbation on metal-sulfide oxidation in surficial
freshwater sediments. Environ Toxicol Chem 15:2147-
2155.
Phillips HO, Kraus KA. 1965. Adsorption on inorganic
materials VI. Reaction of insoluble sulfides with metal
ions in aqueous media. J Chrom 17:549-557.
ProthroMG. 1993. Office of Water policy and technical
guidance on interpretation and implementation of
aquatic life metals criteria. Letter to Water Management
Division directors. Office of Water, U.S. Environmental
Protection Agency, October 1.1993.
Salomons W, De Rooij NM. Kerdijk H. Bril J. 1987.
Sediments as sources of contaminants? Hydrobiol
149:13-30.
Schoonen MAA. Barnes HL. 1988. An approximation of
the second dissociation constant for H,S. Ceochim
, Cosmochim Acta 52:649-654.
Schults DW, Ferraro SP, Smith LM, Roberts FA.
Poindexter CK. 1992. A comparison of methods for
collecting interstitial water for trace organic compounds
and metals. Wat Res 26:989-995.
7-5
-------
References
Sibley PK. Ankley GT. Cotter AM. Leonard EN. 1996.
Predicting chronic toxicity of sediments spiked with
zinc: An evaluation of the acid volatile sulfide (AVS)
model using a life-cycle test with the midge
Chironomus tentans. Environ Toxicoi Chem 15:2102-
2112.
Slotton DO. Reuter JE. 1995..Heavy metals in intact and
resuspended sediments of a California reservoir, with
emphasis on potential bioavailability of copper and
zinc. .War Freshwater Res 46:257-265.
Spehar RL. Fiandt JT. 1986. Acute and chronic effects
of water quality criteria-based metal mixtures on three
aquatic species. Environ Toxicoi Chem 5:917-931.
Sposito G. De Wit JCM. Neal RH. 1988, Selenite
adsorption on alluvial soils: III. Chemical modeling. J
SoilSaSoc.4m 52:947-950.
Stephan CE. Mount DI. Hansen DJ. Gentile JH.
Chapman GA. Brungs WA. 1985. Guidelines for
deriving numerical national water quality criteria for the
protection of aquatic organisms and their uses. PB85-
227049. National Technical Information Service (NTIS),
Springfield. VA.
Strobel CJ, Buffum HW. Benyi SJ, Petrocelli EA.
Reifstick DR, Keith DJ. 1995. Statistical summary
EMAP-Esruanes: Virginian Province 1992-1993. EPA/
620/R-94/-26. U.S. Environmental Protection Agency,
Narragansett, RI.
Stumm W. 1987. Aquatic Surface Chemistry—Chemical
Processes a: the Panicle-Water Interface. John Wiley,
New York. NY.
Stumm W, Morgan JS. 1981. Aquatic Surface
Chemistry—An Introduction Emphasizing Chemical
Equilibria in Natural Waters. John Wiley, New York,
NY
Stumm W, Morgan JS. 1996. Aquatic Surface
Chemistry—Chemical Equilibria, and Rates in Natural
Waters. 3rd ed. John Wiley, New York. NY.
Suedel BC, Boracxek JA. Peddicord RK. Clifford PA,
Dillon TM. 1994. Trophic transfer and biomagnification
potential of contaminants in aquatic ecosystems.
Environ Toxicoi Chem 136:21-89.
Sullivan J. Ball J, Brick E, Hausmann S. Phiiarski G.
Sopcich D. 1985. Report of the technical subcommittee
on determination of dredge material suitability for in-
water disposal. Wisconsin Department of Natural
Resources. Madison. WI.
Sunda WG, Engel DW, Thuotte RM. 1978. Effect of
chemical speciationof toxicity of cadmium to grass
shrimp, Palaemonetes pugio: Importance to free
cadmium ion. Environ Sci Technol 12:409-413
Sunda WG, Guillard RRL. 1976. The relationship
between cupric ion activity and the toxicity of copper
to phytoplankton. J Mar Res 34:511-529.
Sunda WG, Lewis JM. 1978. Effect of complexation by
natural organic ligands on the toxicity of copper to a
unicellular alga, Monochrysis lutheri. Limnol
Oceanogr 23:870-876.
Swartz RC, Cole FA, Lamberson JO, Ferraro SP. Schults
DW, Deben WA, Lee IIH. Ozretich RJ. 1994. Sediment
toxicity, contamination, and amphipod abundance at a
DDT- and dieldrin-contaminated site in San Francisco
Bay. Environ Toxicoi Chem 13:949-962.
Swartz RC. Dittsworth GR. Schults DW, Lamberson JO.
1985. Sediment toxicity to a marine infaunal amphipod:
Cadmium and its interaction with sewage sludge. Mar
Environ Res 18:133-153.
Swartz RC, Schults DW, DeWitt TH. Dittsworth GR.
Lamberson JO. 1990. Toxicity of fluoranthene in
sediment to marine amphipods: A test of the
equilibrium partitioning approach to sediment quality
criteria. Environ Toxicoi Chem 9:1071-1080. .
Tessier A, Campbell PGC. 1987. Partitioning of trace
metals in sediments: Relationships with bioavailability. -
Hydrobiologia 149:43-52.
Tessier A. Campbell PGC. Bisson M. 1979. Sequential
extraction procedure for the speciation of paniculate
trace metals. Anal Chem 51:844-851.
Tessier A. Couillard Y, Campbell PGC, Auclair JC, 1993.
Modeling Cd partitioning in oxic lake sediments and Cd
concentrations in the freshwater bivalve Andonata
grandis. Limnol Oceanogr38:1-17. •
'7-6
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Thurman EM. 1985. Organic Geochemistry of Natural
Waters. Kluwer. Academic Publishers Group, Martinus
Nijhoff/Dr. W. Junk Publishers, Boston. MA. pp 497.
Troup BN. 1974. The interaction of iron with
phosphate, carbonate, and sulftde in Chesapeake Bay
interstitial waters: A thermodynamic interpretation. PhD
thesis. Johns Hopkins University. Baltimore, MD.
U.S. Environmental Protection Agency. 1980. Ambient
water quality criteria for silver—1980. EPA-440/5-80-
071. Office of Water Regulations and Standards,
Washington, DC.
U.S. Environmental Protection Agency. 1985a.
Appendix B—Response to public comments on
"Guidelines for deriving numerical national water
quality criteria for the protection of aquatic organisms
and their uses." July 19, 1985. Federal Register
50:30793-30796.
U.S. Environmental Protection Agency. 1985b. Ambient
water quality criteria for cadmium—1984. EPA 440/5-85-
032. Office of Water Regulations and Standards, '
Washington, DC.
U.S. Environmental Protection Agency. 1985c. Ambient
water quality criteria for copper—1984. EPA 440/5-85-
031. Office of Water Regulations and Standards.
Washington. DC.
U.S. Environmental Protection Agency. 1985d. Ambient
water quality catena for lead—1984. EPA 440/5-85-027.
Office of Water Regulations and Standards,
Washington. DC.
U.S. Environmental Protection Agency. 1986. Ambient'
water quality criteria for nickel—1986. EPA 440/5-85-
004. Office of Water Regulations and Standards,
Washington, DC.
U.S. Environmental Protection Agency. 1987a. Quality
criteria for water, 1986. EPA 440/5-86-001. Office of
Water Regulations and Standards, Washington. DC.
U.S. Environmental Protection Agency. r987b. Ambient
water quality criteria for zinc—1987. EPA 440/5-85-032.
Office of Water Regulations and Standards,
Washington, DC.
U.S. Environmental Protection Agency. 1989. Sediment
classification methods compendium. PB92-231679.
NTIS, Springfield. VA.
U.S. Environmental Protection Agency. 1994a.
Equilibrium partitioning approach to predicting metal
bioavailability in sediments and the derivation of
sediment quality criteria for metals. Vol 1. EPA 822-D-94-
002. Briefing Report to the Science Advisory Board.
Office of Water. Office of Research and Development.
Washington. DC.
U.S. Environmental Protection Agency. 1994b. Methods
for the determination of metals in environmental
samples. Supplement I. EPA-600-R-94- 111. Office of
Research and Development. Washington, DC.
U.S. Environmental Protection Agency. 1994c. Interim
guidance on determination and use of water-effect
ratios for metals. EPA-823-B-94-001. Office of Water,
Washington, DC.
U.S. Environmental Protection Agency. I995a. Review
of the agency's approach for developing sediment
criteria for five metals. EPA-SAB-EPEC-95-020. SAB
Report. Office of Water, Washington. DC.
U.S. Environmental Protection Agency. 1995b. Water
quality criteria for dissolved metals. Federal Register
60(86):22231-22237.
U.S. Environmental Protection Agency. 1995c. Ambient
water quality criteria—Saltwater copper addendum.
Office of Water. Office of Science and Technology.
Washington. DC. p 12.
U.S.Environmental Protection Agency. 1995d. Method,
1669: Sampling ambient water for trace metals at EPA
water quality criteria levels. EPA-821-R-95-034. Office of
Water. Engineering and Analysis Division (4303),
Washington. DC.
U.S. Environmental Protection Agency. I995e. Method
1638: Determination of trace metals in ambient waters
by inductively coupled plasma-mass spectrometry.
EPA-821 -R-95-031. Office of Water. Engineering and
Analysis Division (4303), Washington. DC.
7-7
-------
References
U.S. Environmental Protection Agency. 1995f. Method
1639: Determination of trace metals in ambient waters
by stabilized temperature graphite furnace atomic
absorption. EPA-821-R-95-032. Office of Water,
Engineering and Analysis Division (4303), Washington.
DC
U.S. Environmental Protection Agency. 1996. EMAP-
Estuaries Virginian Province Data 1990-1993. Available
from: EMAP home page, http://www.epa.gov/emap.
Accessed 1997 Sept 17.
U.S. Environmental Protection Agency. 1997a. The
incidence and severity of sediment contamination in
surface waters of the United States, Vol 1: National
sediment quality survey. EPA 823-R-97-006. Office of
Water, Washington. DC.
U.S. Environmental Protection Agency. 1997b. The
incidence and seventy of sediment contamination in
surface waters of the United States, Vol 2: Data
summaries for areas of probable concern. EPA 823-R-97-
007. Office of Water, Washington, DC.
U.S. Environmental Protection Agency. 1997c. The
incidence and severity of sediment contamination in
surface waters of the United States, Vol 3: National
sediment contaminant point source inventory. EPA 823-
R-97-008. Office of Water, Washington. DC.
U.S. Environmental Protection Agency. 1999a. National
Recommended Water Quality Criteria—Correction.
EPA-822-2-99-001. Apnl 1999. Washington, DC.
U.S. Environmental Protection Agency. 1999b.
Integrated approach to assessing the bioavailability
and toxicity .of metals in surface waters and sediments.
EPA-822-E-99-001. Report to the U.S. EPA Science
Advisory Board. April 6-7,1999. Office of Water and
Office of Research and Development, Washington, DC.
U.S. Environmental Protection Agency. 2000a.
Technical basis for the derivation of equilibrium
partitioning sediment guidelines (ESGs) for the
protection of benthic organisms: Nonionic organics.
EPA-822-R-00-001. Office of Science and Technology,
Washington, DC.
U.S. Environmental Protection Agency. 2000b. Methods
for the derivation of site-specific equilibrium
partitioning sediment guidelines (ESGs) for the
protection of benthic organisms: Nonionic organics.
EPA-822-R-00-002. Office of Science and Technology,
Washington, DC.
U.S. Environmental Protection Agency. 2000c.
Implementation framework for use of equilibrium
partitioning sediment guidelines (ESGs). Office of
Science and Technology, Washington, DC.
U.S. Environmental Protection Agency. 2000d.
Equilibrium partitioning sediment guidelines (ESGs) for
the protection of benthic organisms:.Dieidrin. EPA-822-
R-00-003. Office of Science and Technology,
Washington, DC.
U.S. Environmental Protection Agency. 2000e.
Equilibrium partitioning sediment guidelines (ESGs) for
the protection of benthic organisms: Endrin. EPA-822-
R-00-004. Office of Science and Technology,
Washington, DC.
U.S. Environmental Protection Agency. 2000f.
Equilibrium partitioning sediment guidelines (ESGs) for
the protection of benthic organisms: PAH mixtures.
Office of Science and Technology, Washington, DC.
U.S. Environmental Protection Agency. 2000g.
Equilibrium partitioning sediment guidelines (ESGs) for
the protection of benthic organisms: Nonionics
compendium. EPA-822-R-00-06. Office of Science and
Technology and Office of Research and Development,
Washington, DC.
Wolfe DA, Bricker SB, Long ER, Scott KJ. Thursby GB. •
1994. Biological effects of toxic contaminants in
sediments from Long Island Sound and environs.
Technical Memorandum. NOS ORCA 80. National
Oceanic and Atmospheric Administration. Office of
Ocean Resources Conservation and Assessment. Silver
Spring, MD.
ZamudaCD, Sunda WG. 1982. Bioavailability-of
dissolved copper to the American oyster Crassostrea
virginica: Importance of chemical speciation. Mar Dial
66:77-82.
.7-8
-------
Appendix A
Lake Michigan EMAP Sediment Monitoring Database
-------
MISSING FACES FINAL REPORT
1.0 Introduction: Rationale for study (Dave)
Description of project (Dave) ,
2.0 Existing Findings related to served and unserved families: (CDM)
2.1 Introduction (O'Brien, Keane)
2.2 Head Start PIR (O'Brien, Keane)
2.4 FACES exit/dropout data (O'Brien, Keane)
2.5 FACES Staff: Coordinators and Family Service Workers (O'Brien, Keane)
2.6 Family/Household Data Bases
Description of data sets, Tables, highlight salient findings, focusing primarily on
eligible enrolled and non-enrolled families
2.3 FACES parent interviews (O'Brien, Keane)
2.6.1 NSLY79 (Keane)
2.6.2 SIPP (Keane)
2.6.3 PSID (Keane)
2.6.4 Summary of Family/Household Data Bases (O'Brien, Keane)
Tables with summary of findings across data sets
2.7 Summary of Report on Existing findings
3.0 Fall '99 and Spring '00 Site Visits:
3.1 Introduction (Bob)
3.2 Program Descriptions
Site Descriptions with risk factors (Linda)
Eligibility criteria (Bob)
3.6 Community Agency Interviews (MAD)
3.3 Focus Group Discussions (MAD, Linda)
3.4 Record Reviews (Dave)
3.5 Waiting List Reviews (Dave)
3.7 Parent Interviews
Contact Process (MAD)
* Instrument, qualitative summary of responses (MAD)
4.0 Feasibility Assessment . .
4.1 Introduction (Bob)
4.2 Review of feasibility issues (Bob, MAD, Dave)
waiting list development
sample identification
data collection from non-participant families
Agency contacts/recommendations
5.0 Conclusions
Appendices:
D. Secondary Data Base Descriptor Matrices
A. FocUs Group Moderator Guides
B. Record Review Form / Record Review Tables
C. Parent Interview
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Concentrations of SEM, AVS, TOC, and IWGU for cadmium, copper, lead, nickel, and zinc in 46 surficial samples
from Lake Michigan
IWGU
TOC
Samole l0;i)
1 0 13
2 463
3 3 36
•4 4.39
5 0.92
6 437
7 5 27
3 008
9 4.27
10 2.11
'.11 1.39
12 0.41
13 2.37
14 368
.15 0.23
16 0,07
17 3 51
13 0.40
19 1.73
20 069
21 2.51
22 1.17
23 0.13
24 1.03
25 0.63
26 0.30
27 0.29
23 0.21
29 0.11
30 005
3 1 0.27
32 4.95
33 0.54
34 6.75
35 0.18
36 0. 1 5
37 0.56
38 0.10
39 0.06
40 2.68
41 0.16
42 r.SO
.43 1'.29
44 0.05
45 0.14
46 057
J AVS Limit of
SEM AVS
(Umol/g) [umol/g)
053
346
2. "3
3.55
0 14
2.32
1.20
0 17
1.47
025
1.12
0.74
1.17
1.56
1:32
0.-17
Q.75
0.97
l."4
0.70
0 19
0.59
0.21
0.62
0.13
0.15
025
0.12
0.20
0.04
0.85
1.17
0.44
1.37
0.26
0.06
0.17
0.22
0.06
5.83
0.16
0.56
1.02
0.06
0.16
0.66
0.031
0.35
006
0.05j
0.03*
1.13
0.13
0.03*
4.49^
0.0j|
0.03*
0.07
O'.IS
. 0.03*
• 0.44
0.05
0.08
0.03J
0.15,
0.03
0.05
0.03*
0.034
0.03
0.20,
0.03*
0.03'
0.03
0.06
0.03*
0.03*
1.66
0.12
0.09
0.031
0.05
0.05
0.12,
0.03
0.03*
0.07
0.03*
2.25
0.03*
0.05
0.03*
SEM-AVS
Cumol/g)
0.51
3.11
2.72
3.50
0.12
1.69
107
0.15
-3.02
0.23
1.10
0.67
0.99
1.54
0.38
0.12
0.67
0.95
1.59
0.68
0.14
0.57
0.19
0.60
-0.07
0.13
0.23
0.10
0.14
0.02
0.83
•0.49
0.32
1.28
0.24
0.01
0.12
0.10
0.04
5.81
0.09
0.54
-1.23
0.04
Oil
0.64
Cadmium
3
0029
0.018
0.018 .
0.0002"
0.024
0.029
0.115
0.050
—
c
0.0002
c
0.0002.
0.0002"
—
0.018
—
0.079
—
b
—
—
—
—
—
c
0.0002.
0.0002"
—
—
0.012
—
0.018
—
—
—
—
—
0.003
_
0.006 .
0.0002"
__
—
Copper
d
0.003
0.308
0.266
0034
0.049
0.003,
0.003
0.034
—
—
0.070
—
0.003
0.119
—
0.060
—
0.013
—
—
—
—
—
—
—
—
0.155
0.003
—
—
0.036
—
0.041
—
—
—
—
—
0.119
_
0.003d
0.028
—
—
Lead
—
0.00004
0.002
0.0004
0.0008
0.0002
0.0001°
0.001
0.0008
—
—
0.002
—
0.0004
0.0002
—
0.000.8
—
0.0008
—
—
—
—
—
—
—
e
0.0001
0.0004
—
—
0.0004
—
0.0002
—
—
—
—
—
0.001
— .
0.0006
0.002
—
—
Nickel
—
0.005
0.003
0.003
0.006
0.004
0.006
0.006
0.004
—
f
0.0005
—
0.006
0.004
— '
0.008
—
0.010
—
—
—
—
—
—
— •
— '
0.011
0.007
—
—
0.002
— .'
0.0 17
—
—
—
—
r
0.0005
__
0.008 f
0.0005
—
—
Zinc
—
0.003
0.029
0.006
0.032
0.020
0.020
0.055
0.026
—
—
o.oot
—
0.015
0.050
—
0.058
—
0.020
—
—
—
—
—
—
—
—
0.0003
0.0003
—
—
0.020
—
0.012 •
—
—
—
—
—
0.020
_
0.015
0.044
—
—
—
Sum
—
0.040
0.360
0.293
0.073
0.097
0.058
0.180
0.115
—
—
0.074
—
0.025
0.173
—
0.145
—
0.123
—
—
—
—
—
—
—
0.167
0.011
—
—
0.070
—
0.088
—
—
—
—
—
0.144
0.033
0.075
—
—
"a Survival
Hyalella
azteca
92.5
90
92.5
100
0
975
92.5
95
95
77.5
97.5
—
97.5
96.5
90
100
100
95
975
975
75
97.5
575
72.5
95
—
35
75
30
97.5
97.5
97.5
100
95
95
95
—
60
97.5
90
62.5
75
100
'82.5
' —
TO
Chironomus
'.entans
40
90
90
97.5
90
100
100
87.5
100
37.5 .
100 .
—
97.5
92.5
37.5
100
100
100
97.5
97.5
92.5
100
65
57.5
90
—
35
72.5
32.5
100
975
95
100
90
100
92.5
—
55
100
95
65
95
55
72.5
675
Detection =0.03 M
-------
Appendix B
Saltwater Sediment Monitoring Database
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Concentrations of SEM. AYS. toxicitv. and TOC for EMAP, NOAA NST, and REMAP databases
Study-1
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA'
EMAP-VA
EMAP-VA
EMAP-VA'
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-.VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA.
EMAP-VA '
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
SEM
i- mo l/g)
0-289
1.500
0066
0.134
0.266
0.266
1.292
0.347
0.750
0.2 1 2
0.49?
0.624
0.032
0.988
0.604
0.031
1.597
1.065
0.189
0.018
0.079
0.421
0.798
0.903
1.202
0. 1 59
0.246
0.687
0.699
1.663
0.083
0.740
0.878
0.044
0.910 '
0.567
0.734
2.171 •
3.423
0.197
0.162
2.803
0.472
2.079
0.445
2.228.
0.847
1.402
1.425
0.263
2.936
0.394
3.074
2.555
0.452
0.173
' 0.578
AVS
f^mol/g)
1.400
0.742
0.029
0.028
3.740
1080
1.230
0.087
0.948
0.283
0.490
13.400
0.024 .
81.100
3.340 •
0.331
•72.400'
8.480
6.460
0.034
• 0.976
3.210
68.000
3.150
67.700
3.310
4.870
2.420
0.430
116.000
1.300
0.976
1.220
0.025
3.430
0.621
25.000 .
'5.610
138.000
0.892
3.590
11.900
12.500
26.600
0.056
15.100
17.300
52.700
22.300
0.079
29.600
0.031
10.400
0.402
0.480
0.20.1
0.257
SEM-AVS
(umol/g)
-1.111
0.758
0.037
0.106 .
-3.474
-0.814
0.062
0.260
-0.198
-0.071
0.007
-12.776
0.008
-80.112
-2. "36
-0.300
-•0.803
-7.415
.-6.271
-0.016
-0.897
-2.789
-67.202
-2.247
-66.498
-3.151
-4.624
-1.733
0.269
-114.337
-1.217
-0.236
-0.342
0.019
-2.520
-0.054
-24.266
- -3.439
-134.577 '
-0.695
-3.428
-9.097
-12.028
-24.521
0.389
-12.872
-16.453
-51.298
-20.875
0.184
-26.664
0.363
-7.326
2.153
-0.028
-0.028'
- 0.321
Survival0 Significance-
"c *c
100
98
99
103
99
102
'107
102
99
108
103
•113
101
101
107
98
102
93
103
99
97
111
104
99
105
104
106
93
91
100
99
101
98
106
104
104 '
' 107
102
100
107
82
101
101
94
106
103
99
109
88 '
84
100
87
104
96
100
98
101
0
0
• 0
0
0
0
0
0
0
0
0
0
0
0
0-
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0 -.
0
0
0
0
0
0
0
0
0
0
0
0
0
o.
0
0
0
0
0
0
0
0
0
0
0
0
TOC
"c
0.60
2.68
0.17
0.14
0.49
0.56
1.80
0.30
0.95
0.37
1.00
1.58
0.11
3.36
1.38
0..09
4.19
3.P
0.32
0.15
0.14
0.49
2.84
2.S5
2.2S
0.51
O.'l
l.'O
2.05
4.12
0.14
2.30
2.84 '
0.15
3.00
. 0.76
2.21
2.57
4.14
0.37
0.81
2.36
2.77
3.18
0.20
2.92
2.38
2.T0
3.14
0.27
4.15 .
0.18
2.47
118
1.07
0.22
0.65
B-l
-------
Appendix B
StudvJ
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
•EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA
EMAP-VA-
EMAP-VA' .
EMAP-VA
EMAP-VA
EMAP-VA
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
SEM
0.020 .'•
0.088
2.220
0.813
0.851
0.701
1.113
0.601
1.505
0.701
0.717
2.163'
0.616'
2.368
1.278
2.253
0.865
0.950
1.113
AVS
i^mol/2)
3.460
17.800
0.228
0.705
12.900
3.460
2.2*0
54.600
68.000
61.800
35.600
35.600
0.836
0.692
0.227
14.600
6.080
1.200 '.
0.026
'.0.074
0.087
1.120
5.120
. 0.090
0.090
0.174
0.611
4.050
28.200
52.700
12.300
6.140
0.024
0.025
3.460
6.210
29.700
0.259
4.150
59.600
0.381
0.029
3.600
3.510
6.440
18.730
5.630
13.090
65.310
6.940
19.990
4.710
59.590
3.880
16.520
14.950
SEM-AVS
(^mol/2)
-3.251
-.12.389
•1.070
0.334
-11.940
3.909
-0.890
-50.341
-59.771
-58.265
-33.057
-33.476
-0.648
-0.463
1.593
-11.132
-4.458
-0.507
0.268
0.104
0.136
-0.881
-4.319
0.661
0.209
0.167
-0.406
-1.635
-27.568
-51.184
-9.051
-5.678
0.019
0.025
-2.283
-5.586
-28.901
-0.239
-4.062
-57.380
' 0.432
0.822
-2.899
-2.397
-5.839
-17.225
-4.930
-12.373
-63.147
-6.324
-17.622
•3.432
-57.337
-3.015 •
-15.570 •
-13.837
Survival"
ac
96
100
100
102
94
87 ' .
97 ' .
76
43
99
33
0
108
95
'104
102
102
99
'. 95
81
104
88
92
'. 102
104
105
95
100
88
85
103 •
108
100
102
100
104
100
96 .
100
74
93
87
100
96
96
93
93
93
92
92
91 .
91
91
91
90
89
Significance-
"c
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0'
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0"
0
0
0
0
0
0 -
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
TOC
ac
0.36
2.78
0.51
0.30
1.91
1.86
0.25
2.47
4.98
3.19
2.50
2.15
0.35
0.46
1.90
2.08
2.02
1.11
0'.33
0.42
0.43
0.31
1.88
0.66
0.43
0.99
O.'l
2.25
3.35
-.01
3.29
2.19
0,18
0.17
1.83
2.25
4.10
0-.30
0.25
2.18. .
0.98
0.57
0.74
1.12
1.43
2.56
0.77
2.05
3.22 .
0.31
3.02
1.81
2.51
1.32
1.52
2.00
B-2
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Study-1
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI .
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- LI
NOAA- BO
NOAA- BO
.NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
SEM
i^mol/g)
1.026
1.446
2 -77
0:11
2.665
1313
1.235
2.1<»8
3.624
3.594 '
1.342
2:462
0.964
0.332
2.311
0.623
O.S96
0.544
0.641
0.355
0.222
2.262
1.307
1.963
2.^85
4.333
1.927
0.004
3.S31
0.808
1.-S3
2.622
0.597
1.181
. 1.862
2.726
2.102 •
2.471
1.870
1.607
4.942
2.705
2.087
1.514
2.629
3.194
0.872
1.080
0.123'
2.914
2.218
2.609
3.650
1.634
1.267
2.892
AVS
<-mol/g.)
0.850
12.480
29.720
0.090
'8.900
35.050
2.080
14.690
21.800
27.410
37.970
46.450
1.000
4.010
. 79.890
6.610
16.370 .
2.170
2.060
1.390
4.180
39.960
0.380
51.820
61.020
16.080
3.710
24.580
9.250
0.960
40.630
61.840
1.090
3.730 '
50.390
62.760
33.630
7.220
17.120
1-7.810
100.800
83.010
26.730
30.880
32.050
35J90
25.810
11.300
5.310
2.893
2.369
43.959
101.984
5.237 .
3.256
80.584
SEM-AVS
umol/g)
0.1 "6
-11.034
-26.943
0.121
-76.235
-32.237
-0.844
-12.492
-18.176
-23.816
-36.628
-43.988
-0.036
-3.678
-".579
-5.987
-15.475
-1.626
-1.419
-1.035
•3.958
-37.698
0.927
-49.857
-58.235
-11.747
-1.783
-24.576
-5.419
-0.152
-38.847
-59.218
-0.493
-2.549
-48.528
-60.034 '
-31.528
-4.749
-15.250
-16.203
-95.858
-80.305
-24.643
-29.366
-29.421
-32.196
-24.938
-10.220
• -5.187
0.021
-0.151
-4T.350
-98.334
-3.603 '
-1.989
-77.692
Survival6 Significance- TOC
ac ac G( •
88 0 1.63
88 0 2.05
87 0 2.81
87 0 0.54
87 0 3.33
86 ' -. 0 3.83
84 0 1.5S-
84 0 ' 2.80
83 0 2.48
82 0 " 2.59
82 0 1.85
82- 0 3.18
81 0 ' 1.60
. 81 0 1.29
81 0 '369
80 . 0 0.67
80 0. 1.11
79 i Q..27
79 1 1.56 '
79 1 0.64
77. 1 0.45
77 1 2.67
76 1 1.56
76 1 3.46
76 . I 3.81
75 1 3.48
75 1 1.60
74 1 2.87
73 1 3.08
71 • I 1.19
70 ' 1 2.50
70 1 3.49
• 69 1 . • 0.76
68 1 '0.91
67
• 67
' 64
63 '
61
59
54
53
47
42
39.
37
34
16
10
8
15 '
26
29
36
32
2.81 •
2.81
. . 3.42
2.80
3.29.
2.07
3.15
3.62
3.45
2.69
2.68
3. '17
. 1.83
1.91'
0.22
3.05
2.89
3.74
1.83
1.72
1.53'
83 0 6.98
B-3
-------
Appendix B
Study'
NOAA- BO
NOAA- BO
NOAA- BO
NNOAA- 80
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
•NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA-HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
N'O.AA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
SEM
2.5 1 1
0.661
2.458 '
1,3-2
0.959
1430
0."S4
0.943
1.683
1.753
2.447 •
1 339 '
1.296
1.697
1.390
2.310
0.399
2.481
l."36
0.958
9.192
1.525
0.67?
5.037
4.202
1.174
1.855
3.092
2.997
2.581
2.369
5.442
2.618
5.061
2.376
6.998
4.480
4.662
5.896
3.103
1.662
3.512
0.273
0.335
1.664
2.674
5.532
4.029
4.614
3.379
4.240
4.303
5.209
4.801
4.697
2.600
AVS
umol/g)
2.241
13.490
23.077
48.062
53.288
7.599 '
22.486
8.331
42.399
17.697
10.958
68.306
56.838
9.039
43.801
51.857
3.899
19.604
148.969
18.622
120.622
31.842
5.679
69.320
21.980
27.540 .
• 14.170
51.770
79.710
61.050
28.080
25.900
1.080
12.240
4.390
63.450'
20.780
23.720
51.580
59.780
7.230
25.840
0.050
0.036
18.760
3.630
29.210
18.440 •
20.530
30.120
19.320
22.570
14.570
35.370
54.710
56.730
SEM-AVS
fumol/2)
0.270
-12.829
-20.619
-46.190
-52.329
-5.119
-21.702
-7.888
-40.716
-15.944
-8.511
-66.467
-55.542
-7.392
-42.411
-49.547
-3.500
-17.123
-147.233
-17.664 '
-111.430
-80.317
-5.001
-64:283
-17.778
-26.366
-12.315
-48.678
-76.713
-58.469
-25.211
-20.458
1.538
.-7.179
-2.014
-56.452
-16.300-
-19.058 '
-45.684
-56.677
•5.568
-22.328
•0.223
0.299
-17.096
•0.956
-23.678'
-14.411 '
-15.916
. -26.741
.-15.080
-18.267
•9.361
-30.569
-50.013
-54.130
Survival6
86
87
87
89
90
90
91
91
92
94
'94
95
96
97.
97
97
99
99
99
99
100
102
103.
0
41
11
18
101
112
119
81 .
95 •
109
97
. . 108
0
20
14
i
77
19
0
91
93
69
3
96
51
91
88
101
102
101
70
38
37
Significance-'
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
0
0
o
0
1
0
0
1
1
0.
1
0
0
.0
0
0
1
I
1
TOC
2.12
1.00 '.
3.15
3.25
2.39
4.45
1.88
1.78
3.41-
1.41
4.45
2.54
3.05
' 2.68
3.2"
3.35
0.80
3.31
2.94
1." •
4.61
2.96
1.45
5.02
3.47
1.33
4.44
3.36
3.09
136
2.50
2.20
2.67
2.98
2.49
1.98 .
2.98 .
3.19
4.-S
3.99
2.61
4.44
0.07
0.07
0.69
1.00
3.18
' 120
1.94
2.80-
3.15
3.02
3.21
2.98
.347
1.47
B-4
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Studs J
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- HR
NOAA- BA
REMAP-BA
REMAP-BA .
REMAP-BA
REMAP-BA
REMAP-BA
REM.AP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REM.AP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REM.AP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REM.AP-BA
REMAP-BA
REMAP-BA
REM.AP-BA
REM.AP-BA
REMAP-BA
REM.AP-BA
REMAP-BA
REMAP-BA
REM.AP-BA
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB '
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
• SEM
i.* mo 1/2)
1.013
. 1.527
0.505
3.341
3.449
o::o
0.341
O.S88
o.-;:
0.362
2.138 '
3.008
0.151
0.115
0.543
0.103
0.167
0.0"3
0.294
0.120
0.109
0.185
0.120
0.347
0.120
2.2"5
0.344
0.258
0.119
0.258
0.494
0.109
0.266
0.327
. 0.230
2.026
14.550
3.332
3.763
0.357
0.524
0.244
1.247
2.478
1.744
0.131
0.846
4.399
3.884
0.673
3.150
0.270
0.162
2.880
0.323
0.413
AVS
iumol/2)
10.160
15.130
0.630
43.920
3-.860
0.950
0.156
12.971
4.948
0.936
3.295
3.941 '
0.555
.0.156
0. 1 56
0.156
0.932
0. 1 56
0. 1 56
0.156
0. 1 56
0.156
0. 1 56
0. 1 56
0. 1 56
16.592
0.012
0.343
0.156
0.156
0. 1 56
0.156
0.156
0.393
6.400 '
.47.7.93
389.857
243.322
201.687
10.923
3.974
4.502
48.130
47.376
0.156
1.184
0.927
116.954
237.650
21.769 •
43.975
4.491
0.873
153.755
1.684
3.056
SEM-AVS
iumol/g)
-9.147
-13.603
-0.125
-40.579
-34.411
-0.680
0.185
-12.083
•4.226
-0.574
-1.157
-0.933
-0.404
. -0.041
0.387
-0.053
-0.765
-0.083
0.138
-0.036
-0.047
0,029
-0.036
0.191
-0.036
-14.317
0.332
-0.085
-0.037
0.102
0.338
-0.047
0.110
-0.066
-6.170
-45.767
-375.307 '
-239.990
-197.924
-10.566
-3.450
-4.258 •
-46'. 883
-44.898
1.588
-1.053
-0.081
-112,555
-233.766
-21.096
-40.825
-4.221
-0:711
-150.875
-1.361
-2.643
Survival5
<*
29
68
105
86
76
96
84
92
85
98
95
95
96
99
94
85
97 •
99
91
84
92
90.
88
89
81
69
91
94
84
91 '
86
89
86
93
83 .
51
• ' ' 0
' 37
• 79
95
98
84
91
36
69
94
73
93
89
77 ,
91
91
98
92
93
94
Significance-
"c
1
1
0
0
1
0
' 0
0
0
0
0
0
0
0
0
0
. 0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
1
1
. 1
0
0
o •
0
1
1 .
0
1
0
0
I
a
0 .
0
0
0
0
TOC
<*
0.77 .
0.95
• 0.25
2.55
3.63
0.26
0.06
4.05
0.40
0.26
0.43
0.18
0.15
' 0.08
0.07
0.05
0.16
0.05
0.34
0.83
0.92
4.48
0.83
1.26
0.62
LSI
3.S5
O.'7
2.23
0.88
2.10
4.07
1.06
0.29
0.19
0.77
1.52
0.83
0.97
0.26
0.35
0.2"
0.54
1.12
1.14
0.21
K58.
. 6.55
. 8.45'
4.11
5.47
0.74
1.40
7.'0
0.20
1.20'
B-5
-------
Appendix B
Study'
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
RE.MAP-LS
REMAP-LS
RE.MAP-LS
RENLAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REM.AP-LS
REMAP-LS
REM.AP-LS
REMAP-LS
REM.AP-LS
RE.MAP-LS
RE.MAP-LS
REM.AP-LS
RE.MAP-LS
REMAP-LS
RE.MAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REM.AP-LS
REMAP-LS
REM.AP-LS
RENLAP-LS
REM.AP-LS
RE.MAP-LS
RE.MAP-NB
REMAP-NB
REM.AP-NB
REM.AP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REM.AP-NB
REM.AP-NB
REM.AP-NB
REM.AP-NB
REMAP-NB
REMAP-NB
RE.M.AP-NB
REMAP-NB
RE.MAP-NB
RE.MAP-NB
RE.M.AP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
R£MAP-NB
SEM
(-mol/g)
' 0.377
0.099
1.100
0.209
0.213
0954
1. '59
0.711
1.915
2.136
2.480
0.606
3.289
• 3.241
•' 0.616
1.506
2.485
1.894
3.149
0.632
1.057
0.638
1.087
3.711
2.990
8.894
1.277
3.925
5.632
6.809
7.645
4.012
3.905
0.942
3.515
2.216.
• 3.323 •
3.391
3.443
2.466
2.294
5.768
1.013
2.479
0.554
5.222
, 5.116
14.791
4.917
0.398
4.855
3.290
5.822
9.167 .
6.214
.0.794 .
AVS
f.-mol/z)
3.056
0.686
58.945
1.466
0.780
1.542
6.498
10.240
12.596
17.605
23.523
2.501
91.7-3.
56.100' •
1.070
26.201
28.248
25.394
64.643
1.310
4647
0.218
0.312
17.184
59.256
60.816
23.266
42.727
114.770
135.354
150.012
43.663
26.229
6.531
7.134
11.243
7.573
' 4.820
' 3.982
20.273
11.046
5.028
11.079
25.687
2.634
22.617
7.352
109.780
0.530
0.218
9.606
10.105 '
51.460
93.563
42.415
2.651
SEM-AVS
i^mol/si
-2.679
•0.587
-57.845
-1.257
-0.567
-0.588
-3.739
-9.529
-10.681
-15.419
-21.043
-1.895
-88.484
-52.859
-0.454
-24.695
-25.763
-23.500
-61-.494
-0.678
-3.590
0.420
0.7-5
-13.473
-56.266
-51.922
-21.989
-38.802
-109.138
-128.545
-142.367
-39.651
-22.324
-5.589
-3.619
-9.027
-4.250
-1.429
-0.539
-17.807
-8.752
0.740
•10.066
-23.208 '
-2.080
-17.395
-2.236
-94.989
4.387
0.180
-4.751
-6.815
-45.638
-84.396
-36.201
-1.857
'.Survival"
"c
92
93
96
93
95
83
96
97
97
95
99
98
95
97
95
96
96 '
93
93 '
87
90
92
90
88
80
85
92
90
86
91
92
86 '
89
84
87
86
85
83
'95
82
84
75
90
83
84
83
9
8
89
94
83
60 ,
41 •
25
68
93
Significance-'
ac
0
0
0
• o
0
0
0
. 0
•o
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0'
0
0
0
0
0
d
0
0
0
0
0
I
0
0 '.
0
0
1-
1 •
0
0
0
1
1
1 ;
1 ;
0
TOC
ac
1.30
0.75
3.86
0.58
0.69
0.26
0.45
0.56
0.21
0.27
0.32
0.25
0.7'
1.14
0.15
0.95
025
0.98
0.90
1.51
2.44
3.52
7.36
3.99
5.24
3.63
3.18
3.85
4.29
4.36
6.04
3.73
3.93
0.67
0.-5
1.22
1.25
1.05
0.88
1.40
0.95
1.77
0.76
0.99
0.60
1.48
1.45
9.15
3.10-
2.42 '
• 2.62
• 5.70
1 V^
6.48
3.24
2.36
B-6
-------
Equilibrium Partitioning Sediment Guidelines (ESGs): Metal Mixtures
Studva
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REM.AP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REM.AP-RB
REM.AP-RB
REMAP-RB
REM.AP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REM.AP-RB
REM.AP-RB
REM.AP-RB
REM.AP-RB
REM.AP-RB
REM.AP-RB
REM.AP-RB
REMAP-UH
REMAP-UH
R£MAP-UH
REMAP-UH
REMAP-UH
RZMAP-UH
REMAP-UH
R£MAP-UH
REMAP-UH
REMAP-UH
REMAP-UH .
REMAP-UH .
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
R£MAP-UH
REMAP-UH
R£M.\P-UH
R£MAP-UH
REMAP-UH
SEM
umol/g)
4.985
5.280
2.268
6.6'8
2.333
0.333
0.756
0.582
1.012
1.596
0.326
2."09-
• 5.485 .
3.596
5 329
0.337
0.986
0.856
5.364
I. "06
0.371
0.193
0.869
1.288
1.650
2.422
0.512
4.198
5.081
6.095
8.471
3.370
1.198 •
2.127 '
1.360
1.197 »
1.975 »
2.829
2..830 '
1.385
1.519
3.186
2.086
1.799
0.930
0.459"
0.889
0.833
1.317
2.480
0.626
1.500 .
0.723'
4.158
•2.241
2.907
AVS
(amol/2)
43.663
1.934
6.300
T.559
45.222
22.315
1.216
0.821
0.567
0.447
0.156
3.120 .
14.666
19.503
'4.321
2.901
0.156
0. 1 56
39.700
23.515
4.210
0.156
19.617
0.593
0.624
0.156
0.156
4.086
36.490
5.957
8.078
17.247
0.156
12.446
1.790
3.373
'17.136
25.189
56.401
44.588
11.549
86.235
11.713
12.631
10.093
0.156
2.623
2.464
15.563
32.123
9.949
. 5.427 .
1.341
13.504
27.788
29.285
SEM-AVS
i^mol/2)
-38.678
3.346
-4.032
-10.881
-42.389
-21.982
-0.460
-0.239
0.445
1.149
0.170
-0.411
-9.181 .
-15.907
1.008
-2.564
0.830
0.700
-34.336
-21.809
-3.839
0.037
-18.T48
0.695
1.026
2.266
0.356
0.112
-31.409
0.138
0.393
•13.877
1.042
-10.319
-0.430
-2.176
-15.161
-22.360
-53.571
-43.203
-10.030
-83.049
-9,627
-10.832
-9.163
0.303
-1.734
-1.631
-14.246
-29.643
-9.323
-3.927
-0.618
-9.346
-25.547
-26.378
Survival"
^
53
83 .
16
77
54
93
92
94
94
95
93
70
92
62
• 91
97
96
96
91
93
91
92
85
92
91
98
93
90
89
4 '
91
94
94
83
99
92
.45
84
96
88
82
93
82
37
89
98
95
86
88
87
97 «
89 '
89
96
70
95
Significance-
ac
\
0
1
1
1
0
• 0
0
0
0
0
1
0
1
0
0 •
0
0
0
0
0
0
0
0
0
0
0
0
0
. 1
0
0
0
6
0
0
1
0
0
0
0
0 .
0
1
0
o •
0
0
0
0
0
0
o ,•
0
I
0
TOC
at
3.90
6.10
1.99
15.20
2.02
1.23
0.33
0.30
0.30
0.17
0.08
0.42
2.29
0.88
0.97
0.53
0.12
0.51
1.17
3.21
3.54
2.52
2.39
2.44
2.68
2.60
0.42
2.63
2.08
3.03
5.30
3.91
1.03
3.43
1.26
5.85
2.33
0.91
1.21
1,03
1.06
1.39
0.79
1.06
0.43
0.13
0.21
4.96
2.56
• 3.06
2.58
2.71
3.89
4.78
2.66
5.15
B-7
------- |