CADMIUM, COPPER, LEAD, NICKEL, SILVER AND




                     ZINC:




            Proposed Sediment Guidelines




                 for the Protection




               of Benthic Organisms:




                Technical Basis and




                  Implementation
       U.S. Environmental Protection Agency:




        Office of Science and Technology and




         Office of Research and Development

-------
                               CONTENTS
Section                                                                Page
1     INTRODUCTION	1
      1.1   GENERAL INFORMATION  	1
      1.2   OVERVIEW OF DOCUMENT	7

2     PARTITIONING OF METALS IN SEDIMENTS  	8
      2.1   METAL TOXICITY IN WATER-ONLY AND IN INTERSTITIAL
           WATER OF SEDIMENT EXPOSURES	  .... 8
           2.1.1  Toxicity correlates to metal activity	9
           2.1.2  Toxicity correlates to interstitial water concentration	11
      2.2   SOLID PHASE SULFTDE AS THE IMPORTANT BINDING
           COMPONENT	19
           2.2.1  Metal Sorption Phases	19
           2.2.2  Titration Experiments	20
                 2.2.2.1      Amorphous FeS:	21
                 2.2.2.2      Sediments 	24
           2.2.3  Correlation to Sediment AVS	24
           2,2.4  Solubility Relationships and Displacement Reactions	25
           2.2.5  Application to Mixtures of Metals	28

3     TOXICITY OF METALS IN SEDIMENTS	30
      3.1   GENERAL INFORMATION	30
      3.2   PREDICTING METAL TOXICITY:  SHORT-TERM	30
           3.2.1  Spiked sediments: Individual experiments	 30
           3.2.2  Spiked Sediments: All experimental results summarized	34
           3.2.3  Field sediments	40
           3.2.4  Field Sites and Spiked Sediments Combined	42
      3.3   PREDICTING METAL TOXICITY:  LONG-TERM STUDIES	46
           3.3.1  Life cycle toxicity tests  	47
           3.3.2  Colonization tests	49

4     DERIVATION OF SEDIMENT GUIDELINES FOR METALS	52
      4.1   GENERAL INFORMATION	52
      4.2   SINGLE METAL SEDIMENT GUIDELINES	53
           4.2.1  AVS Guidelines	54
           4.2.2  Interstitial Water Guidelines	'.	55
      4.3   MULTIPLE METALS GUIDELINES	56

-------
                           CONTENTS (Continued)
Section
Page
4.3.1  AVS Guidelines	57
      4.4   ESG FOR METALS VS. ENVIRONMENTAL MONITORING
           DATABASES   	60
           4.4.1 Data Analysis . . .	60

5     IMPLEMENTATION	65
      5.1   CONSIDERATIONS IN PREDICTING METAL TOXIOTY	65
      5.2   SAMPLING AND STORAGE	65
           5.2.1 Sediments	67
           5.2.2 Interstitial Water	67
      5.3   ANALYTICAL MEASUREMENTS	69
           5.3.1 Acid Volatile Sulfide	70
           5.3.2 Simultaneously Extracted Metal	70
           5.3.3 Interstitial Water Metal	70
      5.4   ADDITIONAL BINDING PHASES	71
      5.5   PREDICTION OF THE RISKS OF METALS IN SEDIMENTS
           BASED ON EqP	72

6     GUIDELINES STATEMENT  	73

7     REFERENCES	75
Appendix A: Glossary of definition of abbreviations and equations
Appendix B: Solubility Relationships for metals sulfides
Appendix C: Lake Michigan EMAP sediment monitoring database
Appendix D. Saltwater sediment monitoring database
                                     11

-------
Table 2-1.


Table 2-2.

Table 3-1.
Table 3-2.
Table 4-1.
                                      TABLES
Cadmium binding capacity and AVS of sediments
(from Di Toro et al., 1990)	
                                                                                 Page
Metal sulfide solubility products
24

28
Toxicity of sediments from saltwater (SW) and freshwater (FW)
field locations, spiked-sediment tests and combined field and
spiked-sediment tests as a function of the difference between the
molar concentrations of SEM and AVS (SEM-AVS), interstitial
water toxic units (IWTUs) and both SEM-AVS and IWTUs
(modified from Hansen et al., 1996a)	
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	
                                                                                   35
                                                                                   48
Water quality criteria (WQC) criteria continuous concentrations
(CCC) based on the dissolved concentration of metalib. These
WQC CCC values are for use in the Interstitial Water Guidelines
approach for deriving sediment guidelines based on the dissolved
metal concentrations in interstitial water	
                                                                                   56
                                          111

-------
                                       FIGURES

                                                                                  Page
Figure 2-1.    Acute toxicity to Palaemonetes of total cadmium (top) and
              cadmium activity (bottom) with different concentrations of the
              complexing ligands NTA (left) and chloride as salinity (right)
              (This figure is from Sunda et al., 1978)	10

Figure 2-2.    Acute toxicity to a dinoflagelate (left) of total copper (top) and
              copper activity (bottom), with and without the complexing ligand
              EDTA (this portion of the figure is from Anderson and Morel,
              1978).  Toxicity of zinc to Microcystis aeruginosa (right)
              showing growth as cells/ml versus time with different levels of
              the complexing  ligands EDTA and NTA (top) and number of
              cells at five days as a function of free zinc concentration (bottom)
              (this portion of the figure is from Allen et al., 1980)	12

Figure 2-3.    Specific growth rate of a diatom (left) (this portion of the figure is
              from Sunda and Guillard, 1976) and Microchrysis lutheri (right)
              versus total copper (top) and copper activity  (bottom) for a range
              of concentrations of the complexing ligands Tris and natural
              DOC (this portion of the figure is from Sunda and Lewis, 1978)	13

Figure 2-4.    Body burdens of copper in oysters (Crassostrea virginica) versus
              total copper (top) and copper activity (bottom) with different
              levels of the complexing ligand NTA (this figure is from Zamuda
          *    and Sunda, 1982)	14

Figure 2-5.    Rhepoxynius abronis  mean survival versus dissolved cadmium for
              4-day toxicity tests in seawater (symbols) and interstitial water at
              time 0 and 4 days (bars) (this figure is from  Swartz et al., 1985)	15

Figure 2-6.    Mortality versus interstitial water cadmium activity for  sediments
              from Long Island Sound, Ninigret Pond and a mixture of these
              two sediments (this figure is from Di Toro et al.,  1990). Water-
              only exposure data for Ampelisca abdita and Rhepoxynius
              hudsoni. The line is  a joint fit to both water-only data sets	17
                                           IV

-------
                                FIGURES (Continued)

                                                                                  Page
Figure 2-7.   Toxicity of copper to Hyalella azteca versus copper
             concentrations in a water-only exposure (open symbols) and
             interstitial water copper concentrations in sediment exposures
             (closed symbols) using Keweenaw Waterway sediments (this
             figure is from Ankley et al.,  1993)	18

Figure 2-8.   Cadmium durations of amorphous FeS.  X-axis is cadmium added
             normalized by FeS initially present (this figure is from Di Toro et
             al., 1990).  Y-axis is total  dissolved  cadmium. The lines
             connecting the data points  are an aid to visualizing the data	22
Figure 2-9.   Concentrations of Fe2* and Cd2* in supernatant from titration of FeS by
             Cd2* (personal communication with Di Toro 1992)	23

Figure 2-10.  Cadmium titration of sediments from Black Rock Harbor, Long
             Island Sound, Hudson River and Ninigret Pond (this figure is
             from Di Toro et al., 1990). Cadmium added per unit dry weight
             of sediment versus dissolved cadmium	26

Figure 3-1.   Concentrations of individual metals in interstitial water of
             sediments from Long Island Sound (top)  and Ninigret Pond
             (bottom) in the mixed metals experiment as a function of
             SEM/AVS ratio (this figure is from Berry et al.,  1996).
             Concentrations below the IW detection limits, indicated by
             arrows, are plotted at one half the detection limit. K^ is  the
             sulfide solubility product constant	,	33

Figure 3-2.   Percentage mortality of saltwater and  freshwater benthic  species
             in 10-day toxicity tests in sediments spiked with individual metals
             (Ag, Cd, Cu, Ni, Pb, 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 cadmium, copper, lead, nickel and zinc in
             ,umoles metal per gram dry weight sediment; (b)  SEM/AVS ratio:
             and (c) interstitial water toxic units (silver data from Berry et al.,
             in review, all other data modified after Berry et al., 1996a).
             Species tested  include: the oligochaete (Lumbriculus variegatus),
             polychaetes ( Capitella capitata and Neanthes arenaceodentata),
             amphipods (Ampelisca abdita and Hyalella azteca), harpacticoid
             copepod (Amphiascus tenuiremis) and gastropod (Helisoma sp.).

-------
                                FIGURES (Continued)

                                                                                 Page

             Data below the SEM detection limit are plotted at SEM/AVS =
             0.01. Data below the detection limit of metals in interstitial
             water are plotted at IWTU = 0.01	36

Figure 3-3.   Percentage mortality of saltwater and freshwater benthic species
             in 10-day toxicity tests in spiked sediments (open symbols) and
             sediments from the field (closed symbols)  (silver data from Berry
             et al., in review, all other data modified after Hansen et al.,
             1996a).  Mortality is plotted as a function of  (a) the sum of the
             concentrations of cadmium, copper, lead, nickel and zinc in
             /^moles metal per gram dry weight sediment;  (b) SEM/AVS
             ratio; and (c) interstitial water toxic units.  Species tested include:
             the oligochaete Lumbriculus variegatus, polychaetes (Capitella
             capital a and Neanthes arenaceodentata), the harpacticoid
             (Amphiascus tenuiremis), amphipods (Ampelisca abdita and
             Hyalella azteca) and the snail (Helisoma sp.).  Data below the
             SEM detection limit are plotted at SEM/AVS =  0.01.  Data
             below the detection limit of metals in interstitial water are plotted
             at IWTU =  0.01  	  37

Figure 3-4.   Percentage mortality of amphipods, oligochaetes and polychaetes
             exposed to sediments from three saltwater and four freshwater
             field locations as a function of the sum of the molar
             concentrations of SEM minus the molar concentration of AVS
             (SEM-AVS) (from Hansen et al., 1996a):  The vertical dashed
             line at SEM-AVS =  0.0 indicates the boundary between sulfide-
             bound unavailable metal and potentially available metal	45

Figure 4-1.   SEM minus  AVS values versus AVS concentrations in EMAP-
             Great Lakes sediments from Lake Michigan.  Data are from
             surficial grab samples only (this figure is taken from Leonard et
             al., 1996, see data in Appendix C). The upper plot shows all
             values, the lower plot has the ordinate  limited to SEM minus
             AVS values between -10 and +10	61

Figure 4-2.   SEM minus AVS values versus AVS concentrations in EMAP-
             Estuaries Virginian Province (U.S. EPA, 1996)  , REMAP-
             NY/NJ Harbor Estuary (Adams et al.,  1996) and the NOAA NST
             Long Island Sound (Wolfe et al., 1994), Boston Harbor (Long et
                                          VI

-------
                   FIGURES (Continued)

                                                                   Page
al., 1996), and Hudson-Raritan Estuaries (Long et al., 1995);
(see data in Appendix D for sources). The upper plot shows all
values, and the lower plot has the ordinate limited to SEM minus
AVS values between -10 and +10	63
                            vu

-------
                                      SECTION 1

                                   INTRODUCTION

1.1    GENERAL INFORMATION

       Under the Clean Water Act (CWA) the U.S. Environmental Protection Agency (U.S.
EPA) is responsible for protecting the chemical, physical and biological integrity of the
nation's waters.  In keeping with this responsibility, the U.S. 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 the U.S. EPA's best estimate of
concentrations protective of human health and the presence and uses of aquatic life. While
these WQC play an important role in assuring 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 the 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
(USEPA 1997a, b, c). The scarcity of defensible sediment guidelines and the single chemical
nature of those available make it difficult to accurately assess the extent of the ecological risks
of contaminated  sediments, to establish pollution prevention strategies, and to identify,
prioritize and implement appropriate clean up activities and source controls.

       As a result of the need for a procedure to assist regulatory agencies in making decisions
concerning contaminated sediment problems and their prevention, a U.S.  EPA Office of
Science and Technology and Office of Research and Development research team was
established 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 for non-ionic 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 EqP sediment guidelines (ESG)
refers to numerical concentrations for individual 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 non-ionic organic chemicals were:

       1.      The concentration of non-ionic organic chemicals in sediments, expressed on an
              organic carbon basis, and in interstitial water, correlate to observed biological
              effects on sediment dwelling organisms across a range of sediments.

       2.      Partitioning models can relate sediment concentrations for non-ionic organic
              chemicals on an organic carbon basis to  freely dissolved concentrations in
              interstitial water.

       3.      The distribution of sensitivities of benthic and water column organisms to
              chemicals are similar, thus, the currently established WQC final chronic values
              (FCV) can be used to define the acceptable effects concentration of a chemical
              freely dissolved in interstitial water.

       Due to their wide-spread release and persistent nature, metals such as cadmium,
copper, lead, nickel, silver and zinc are commonly elevated in aquatic sediments. These metals
are a potential aquatic environmental concern in addition to nonionic organic chemicals. Thus,
there have  been various proposals for deriving sediment guidelines or standards for protecting
benthic communities from metal toxicity.  Many such attempts have featured measurement of
total sediment metals followed by comparison to background metal concentrations, or in some
cases an effects-based endpoint (Ingersoll et al., 1996;  Long and Morgan.  1991; MacDonald et
al., 1996; Persaud et al., 1989; Sullivan et al.,  1985).  An important limitation to these types
of approaches is that causality can not be established in part because of the procedures used to
derive correlative values and because values derived are based on total rather than bioavailable
metal concentrations; i.e., for any given total metal concentration, adverse toxicological effects
may or may not occur, depending upon physico-chemical characteristics of the sediment of
concern (Di Toro et al., 1990; Luoma et al., 1989; Tessier and Campbell,  1987).

-------
       Considerable research has used elaborate sequential extraction procedures to identify
sedimentary physico-chemical fractions with which metals are associated in an attempt to
understand the biological availability of metals in sediments (Tessier et al., 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 found in the
greatest concentrations. (See Section 2.)

       In developing ESG for metals that are causally-based and applicable across sediments, it
is essential that bioavailability be understood.  Different studies have shown that while 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 non-ionic
organic chemicals and organic carbon (Di Toro et al., 1991), sediment 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).  Acid
volatile sulfide binds, on a molar basis, a number of cationic metals of environmental concern
(cadmium, copper, nickel, lead, 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 the six metals  cadmium, copper, lead, nickel, silver and zinc.)

        The data that support the EqP approach for deriving sediment guidelines for non-ionic
organic chemicals are reviewed by Di Toro et al. (1991) and U.S. EPA, (1997a).  Recently
EPA evaluated the potential utility of the EqP approach for deriving sediment guidelines for
metals (U.S.  EPA, 1994a), which was reviewed by EPA's Science Advisory Board (U.S.
EPA, 1995a).  The data that support the EqP approach for deriving sediment guidelines for
metals presented in this document are 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. (1996a);  DeWitt et al. (1996); Di Toro et al. (1990; 1992; 1996a,b) Hansen et al.
(1996a,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) and the U.S. EPA (1995a) were of particular importance in the
preparation of this document.

-------
       The same three principals observed in applying the EqP approach to non-ionic organic
chemicals listed above, also apply with only minor adjustments to deriving ESG 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 acid volatile sulfide (AVS) and simultaneously extracted metals
              (SEM; the metals extracted with AVS) in sediments, and in interstitial waters
              correlate to observed biological effects on sediment dwelling organisms across a
              range of sediments.

       2.      Partitioning models can relate sediment concentrations for divalent cationic
              metals  (and silver) on an AVS basis to the absence  of freely dissolved
              concentrations in interstitial water.

       3.      The distribution of sensitivities of benthic and water column organisms to
              organic chemicals and metals are similar (U.S. EPA, 1998a), thus, the currently
              established WQC final chronic values (FCV) can be used to define the
              acceptable effects  concentration of the metals freely dissolved in interstitial
              water.

       The EqP approach, therefore, assumes that: (1) the partitioning of the metal between
sediment AVS (or any other binding factors controlling bioavailability) and interstitial water is
at 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, sediment-integument exchange, or from a mixtures of
exposure routes; (3) for the cationic metals cadmium,  copper, lead, nickel, silver and zinc, no-
effects concentrations in sediments can be predicted using the difference between the total
molar concentration of SEM for  these metals and the total  molar concentration of AVS. This
difference is the amount of either excess metal or excess AVS.  So long as the molar
concentration of AVS equals or exceeds the sum of the molar concentrations of these metals,
the sediment is not expected to cause acute or chronic toxicity in benthic organisms; and (4)
the WQC FCV concentration is an appropriate effects concentration for freely dissolved metal
in interstitial water  and the toxicity of metals in interstitial water is no more than additive.

-------
       Two equally applicable ESG for metals, a solid phase and an interstitial water phase,
are proposed. For the first time, the Agency is publishing ESG that account for bioavailability
in sediments and potential for effects of a mixture in the aquatic environment.  The mixtures
approach for these six metals will provide an ecologically relevant benchmark by resolving the
longstanding lexicological problem of their interdependent geochemistry.  The solid phase
ESG is defined as the E[SEM]-[AVS]sO (total molar concentration of simultaneously extracted
metal - total molar concentration of acid volatile sulfide is less than or equal to zero).  Note
that cadmium, copper, lead and nickel are divalent metals so that one  mole of each metal can
bind with one mole of A VS. The molar concentrations of these metals are compared to 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 SEMAg by
convention will be defined as the  molar concentration of silver divided by two, [Ag]/2, which
is compared to the molar AVS concentration.  The interstitial water phase ESG is
£[Miid]/[FCVijd] <. 1 (the sum over all of the six metals of the concentration of each individual
metal dissolved in the interstitial water/ the metal-specific Final Chronic Value based on
dissolved metal is less than or equal to one). This later value is termed interstitial water
guidelines toxic units (IWGTUs). The IWGTUs approach by definition requires that the IW
metals are additive.  The data presented in this document supports additivity.

       Importantly, both the solid phase ESG and interstitial water ESG are no-effect
guidelines; i.e., they predict sediments that are acceptable for the protection of benthic
organisms.  These ESG when exceeded do not predict sediments  that  are unacceptable for the
protection of benthic organisms.  The solid phase (SEM-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.   The use of
both the solid phase 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 SEM-AVS >0 and in nontoxic sediments having
SEM-AVS < 0 demonstrates that metals in these sediments are unavailable.  Because of the
known spatial and temporal cycling of the important metal-binding phases in sediments,
Section 5 of this document provides  implementation guidance on sediment collection, handling
and analysis that will improve estimates of risk.

-------
       The ESG developed using the latest available scientific data are suitable for providing
guidance to regulatory agencies because they are:

       1.     numeric values,
       2.     chemical specific,
       3.     causal,
       4.     applicable to most sediments and
       5.     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 associated with sediments. ESG are intended to
apply to sediments permanently inundated with water, intertidal sediment and to sediments
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 question of possible contamination of upper trophic level
organisms or the synergistic, additive or antagonistic effects of other substances.  The ESG
presented in this document represent the U.S. EPA's best recommendation at this time of the
concentration of a metals mixture (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, 1998c).
        *

       This document presents the theoretical basis and the supporting data relevant to the
derivation of the ESG for the metals cadmium, copper lead, nickel, silver and zinc. 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 for Cadmium" (U.S. EPA,
1985b); "Ambient Water Quality Criteria 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, 1985d); "Ambient Water Quality Criteria  for Nickel" (U.S.
EPA, 1986);  "Ambient Water Quality Criteria for Zinc" (U.S. EPA, 1987); "Ambient Water
Quality Criteria for Silver" (U.S. EPA, 1980) is necessary in  order to understand the following
text, tables and  calculations.  Guidance for the acceptable use  of ESG values for metals
mixtures is contained in "Users Guide  for Multi-Program Implementation of Sediment
Guidelines" (U.S. EPA, 1998b).

-------
1.2    OVERVIEW OF DOCUMENT

       Section 1: "Introduction" provides a brief review of the EqP methodology as it applies
to the individual, and mixtures of, the divalent metals cadmium copper, lead, nickel and zinc
and the monovalent form of silver. Section 2: " Partitioning of Metals in Sediments" reviews
published experimental results that describe the partitioning and bioavailability of these metals
in freshwater and marine sediments.  The role of AVS, SEM and interstitial water
concentrations of metals is described.  Section 3: "Toxicity of Metals in Sediments" 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 toxicity of sediment-associated metals.  Section 4: "Derivation of
Sediment Guidelines for Metals" describes the SEM-AVS and interstitial water guidelines toxic
unit approaches for the derivation of the ESG for individual metals and mixtures of metals.
Published WQC values for dissolved metal for five of these six metals (the silver FCV for
freshwater is not available)  are summarized for use in the Interstitial Water Guidelines
Approach. The  ESG for metals is then compared to  chemical monitoring data on the
environmental occurrence of metals and AVS in sediments from Lake Michigan, the Virginian
Province from EPA's Environmental  Monitoring and  Assessment Program (EMAP) and
NOAA's National Status and Trends.  Section 5: "Implementation"describes procedures for
collection^ handling, analysis of sediments and interpretation of data from sediment samples
required if the assessments of the risks of sediment-associated metals are to be accurate.
Section 6: "Guidelines Statement" concludes with EPA's guidelines statement for the metals
cadmium, copper, nickel, lead, silver and zinc.  The references cited in this document are
listed in Section 7.

-------
                                     SECTION 2

                     PARTITIONING OF METALS IN SEDIMENTS

2.1    METAL TOXICITY IN WATER-ONLY AND IN INTERSTITIAL WATER OF
       SEDIMENT EXPOSURES

       The equilibrium partitioning methodology for establishing sediment guidelines requires
that the chemical concentration be measured in the bioavailable phase and that the chemical
potential of the chemical be determined. This Section demonstrates that biological effects
correlate to metal activity. Secondly it demonstrates that biological response is the same for
water only exposures and for sediment exposures using the interstitial water concentrations.
Therefore, for both metals and non-ionic chemicals this fundamental tenant of the Equilibrium
Partitioning model is satisfied.

       A direct approach to establishing sediment guidelines for metals would be to apply  the
water quality criteria final chronic values to measured interstitial water concentrations.  The
validity of this approach depends on the degree to which interstitial water concentration
represents free metal activity and can be accurately measured in both systems.  For most
metals, free metal activity can not be measured at water quality criteria concentrations and
present water quality criteria are not based on activity.  Many metals readily bind to dissolved
(actually colloidal) organic carbon (DOC), and DOC complexes do not appear to be
bioavailable (Bergman and Dorward-King,  1997). Hence guidelines based on interstitial water
concentrations of metals may be overly protective with the direct use of the concentration of
metals in interstitial water.

       By implication this difficulty extends to any complexing ligand that is present in
sufficient quantity.  The decay of sediment  organic matter can cause substantial changes in
interstitial water chemistry. In particular, bicarbonate increases due to sulfate reduction.  This
increases the importance of the metal-carbonate complexes and further complicates the  question
of the bioavailable metal species (Stumm and Morgan, 1996).

       The sampling of sediment interstitial water for metals is not a routine procedure. The
least invasive technique employs a diffusion sampler which has cavities covered with a filter

-------
membrane (Allen et al., 1993; Bufflap and Allen, 1995; Carignan, 1984; Carignan et al.,
1985; Hesslein, 1976).  The sampler is inserted into the sediment and the concentrations on
either side of the membrane equilibrate.  When the sampler is removed the cavities contain
filtered interstitial water samples.  Since  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 depends on the
interstitial size  of the membrane and the geometry of the cavity and usually exceeds one day.

       An alternate technique to separate the interstitial water is to obtain a sediment core,
slice it, filter or centrifuge the slice and then filter the resultant interstitial water twice.  For
anaerobic sediments this must be done in a nitrogen atmosphere to prevent the precipitation of
iron hydroxide which would scavenge the metals and yield artificially low dissolved
concentrations (Allen et al., 1993; Troup, 1974).

       Although either of these techniques are 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, it 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.

2.1.1  Toxicity correlates to metal activity

       A substantial number of water only exposure experiments  disscussed below point to the
fact that biological effects can be correlated  to the divalent metal activity {M2*}.  The claim is
not that the only bioavailable form is  M:~ - for example MOFT may also be bioavailable - but
that the DOC and  certain  other ligand-complexed fractions are not bioavailable.

       The acute toxicity of cadmium to grass shrimp (Palaemonetes) has been determined at
various concentrations of chloride and the complexing ligand NTA, both of which form
cadmium complexes (Sunda et al., 1978). The concentration response curves as a function of
total cadmium (Figure 2-1, top panels) are quite different at varying  concentrations of chloride,
indexed by salinity, and NTA. However, if the organism response is evaluated with respect to
Cd2*  activity in the solution then the data become a single concentration-response relationship

-------
              ACUTE TOXICITY OF CADMIUM TO
                GRASS SHRIMP (Palaemonetes)
               EFFECT OF NTA COMPLEXATION

                (AFTER W.G. SUNDA et al., 1978)
                                                           ACUTE TOXICITY OF CADMIUM TO
                                                            GRASS SHRIMP (Palaemonetes)
                                                                EFFECT OF SALINITY
                                                             (AFTER W.G. SUNDA et al., 1978)
       100
     I
     en
     i-
     UJ
     u
     o:
     ui
     OL
                                        x1(T
                       5.0                6.0
                  TOTAL CADMIUM (-LOG CdT)
                                                        100
                                                     50
• Salinity (0/00)
A— 4.8 + 0.4
•— 8.4 + 0.2
T — 16.3 + 0.3
•— 20.0 + 0.3
•— 28.9 + 0.6
»


^ « 	 .
7t



J. .//
:,• . y /
                                                                TOTAL CADMIUM (-LOG CdT)
     
-------
                                                                                     11

(bottom panels). Comparable results have been reported for copper-EDTA complexes
(Anderson and Morel,  1978) for which concentration-response correlates to Cu2* activity
(Figure 2-2, left top and bottom).

       When the concentration of zinc is held constant and the concentration of the complexing
ligand NT A is varied, the effect on growth of the phytoplankter Microcystis decreases as NT A
added increases (Figure 2-2, right top and bottom; Allen et al., 1980).  The cell density
increases rather than decreases in time and reaches control levels at the highest NTA
concentration (left top and bottom panels). The data can all be correlated to free zinc activity
as shown (right top and bottom panels).  Similar results for diatoms exposed to copper and the
complexing ligand Tris (Figure 2-3, top; Sunda and Guillard, 1976). Variations  in Tris
concentrations and pH  produce markedly  different growth rates (left top and bottom) which can
all be correlated to the  Cu2"1" activity (right).  A similar set of results have been obtained by
Sunda and Lewis (1978) with DOC from  river water as the complexing ligand (Figure 2-3,
right top and bottom).

       Metal bioavailability as measured  by metal accumulation into tissues of organisms has
also also been examined (Zamuda and Sunda, 1982).  Uptake of copper by oysters is  correlated
not to total copper concentration (Figure 2-4, top) but to copper activity (bottom).

       The implication to be drawn from these experiments is that the partitioning model
required for establishing sediment guidelines should predict dissolved metal in  interstitial
water.  The following subsection examines the utility of this idea.

2.1.2  Toxicity  correlates to interstitial water concentration

       This subsection presents some early data that first  indicated the  equivalence of
interstitial water concentrations and water only exposures. Much more data  of this sort are
presented in Section  III of this document.  Swartz et al.(1985) tested the acute toxicity of
cadmium to the marine amphipod Rhepoxynius abronius, in sediment and seawater. 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 of cadmium in interstitial water (1.42
mg/L) with the 4-day LC50 of cadmium in seawater without sediment (1.61  mg/L) resulted in
no significant difference between the two (Figure 2-5).

-------
      CO
      111
      o
      UJ
      §
      111
      o
      tr
      UJ
      a.
             ACUTE TOXICITY OF COPPER
                TO A DINOFLAGELLATE
            (FROM ANDERSON AND MOREL, 1978)
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
                  WITH EDTA
WITHOUT EDTA
                 56789
              TOTAL COPPER (-LOG (Cur))
                                              CHRONIC TOXICITY OF ZINC
                                            ON MICROCYSTIS AERUGINOSA
                                                 (FROM ALLEN, et. al.. 1980)
                                                   10
                                                     7.
                      10
                        6.
                                               UJ  10-
                                               O
                      104H
                                                   10
                                             O AAP «rf1:1 M F»:EOTA wto T
                                                 + 4.8 x 10-7 M Zn
                                                 + 1.0 x 10-7 UNTA
                                                 » 2.5 * 10-7 M NTA
                                                 + 4.0 x 10-7 MNTA
                                                 + 60 x 10-7 M NTA
                                                 »1.0x 10-6 MNTA
                                                 » 5.0 X 10-« M NTA
                                                 + 1.0x10-5 MNTA
                                                       8
                                        12   16
                                       DAYS
20
      iu
      O
      UJ
      g
       ui
       O
       K
       UJ
       Q.
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                                                  10°
                                               I
                                               in
                  <  10'
                  UJ
                  O
                  u.
                  O
                  Q£
                  3
                       ,5.
              9    10   11    12    13   14
               COPPER ACTIVITY ( p Cu )
                                                  103
                                            K' at pH 7.0
                                           -EDTA   13.46
                                           - NTA     7.64
                                           -ODS     6.31
                                           -CMOS    5.25
                                           - Builder M  4.33
                                           -Control
                                            0  1.0  2.0  3.0  4.0  5.0
                                               FREE ZINC (moles/liter x 107)
  Figure 2-1.  Acute toxicity to a dinoflagelate (left) of total copper (top) and copper activity
  (bottom), with and without the complexing ligand EDTA (this portion of the figure is from
   Anderson and Morel, 1978).  Toxicity of zinc to Microcystis aeruginosa (right) showing
growth as cells/ml versus time with different levels of the complexing ligands EDTA and NTA
  (top) and number of cells at five days as a function of free zinc concentration (bottom) (this
                     portion of the figure is from Allen et al.,  1980).

-------
              CHRONIC TOXICITY OF COPPER
                      TO A DIATOM

               (FROM SUNDA AND GUILLARD, 1976)
                                         CHRONIC TOXICITY OF COPPER
                                          TO MONOCHRYSIS LUTHERI
                                          (FROM SUNDA AND LEWIS, 1978)

3 2
— fm
rf ^*
ftp J5


5o
9S5 1
t£_ "*

d Z
O ""
UJ
(o 0
pHBI-82
x 05 mM tit 1 . |
.0io ? L.JI— — """"
A 20 ' 1 	 T M
• 30 ' I/1
o 50 ' /
• 100 „• 1 / 1
PH82 . ,X / 1

PH77 , 1 J.--""T.^«i:- "»-''' '
O 10mM*i» 1— rn tr-"T <•*'»••*'* b — -•
-pM87 . / 4-''/° | '
• 1CTmMtri»i '.•' /
i /* i
> • ' .
1 ' a
! •
li / 0
L /
F '
• /
I ^'
, , _
2.00
Z
£
*J
Jp
i S

§Z 1.00
2 ^
O <2

uTQ
0
UJ
Q.
OT 000
l 1
• 10% River V* tor
A 90% River \Ahter




.






§ §















3 4 5 5.6 -7.4 4.00 5.00 6.00 7.00
TOTAL COPPER(-LOG CuT) TOTAL COPPER (-LOG CuT)

*^
— 2.0
P
$
|
2 1 o
0£
o
o
ik
o
UJ
& o
,
1 1 ^SMV
^~i.. "i' **
1
i u*4-n^
«-twW* t~i~
•v*^9 vTj1' J 1
.Jw
1 *
-Of
JJ_
1
'
111!
8 9 10 11
^.uu
Z

UJ
b
1°
§z 1.00
2Q
o w
o>
CO
u
UJ
0.
OT 0.00
6.0
1 1
• 10% Rivtr AhMr
• 30%Riv«rV*t»r
A90%Rrverv\h«Br

.^-«
X'
•


_
^^
• •
4 ^"'^
"** , • 1














0 7.00 8.00 9.00
COPPER ( pCU)
                                                                      COPPER ACTIVITY (pCu)
Figure 2-3. Specific growth rate of a diatom (left) (this portion of the figure is from Sunda and Guillard, 1976) and Microchrysis
lut^i (right) versus total copper (top) and copper activity (bottomj^ a range of concentrations of the complexing ligands TrisA
                     and natural DOC (this portion of the figuWs from Sunda and Lewis, 1978).                    ™

-------
                       UPTAKE OF COPPER BY OYSTERS
z
o
|£
Z
UJ
0
o
o
or
UJ
a.
a.
O
o

1OU
g
§ 100
3
>j
a
O)
"3 50
O
o>
^
•-•

O A
o if


o

Q




a * '
oi o • A '
0.01 0.1 1.0 10.C
                           TOTAL COPPER CONC. (jlM)
,_
Z
0
UJ
^
Q
f
o>?
Z M
o 2
§ °
3.2
li
u
^
or
UJ
a.
a.
O
u
^_
UJ
z
!
200'


180
160

140
120

100

80
60
40


20

0


TOTAL NTA (uM)

Winter
A 1.0
0 10.0
Summer
*0.2
I A 1.0
I • 8-1

i"~ *
'

-f—
%!
I


• i
^ 1 "ht I
I -•; • At
I

i i i
                    8          9         10        11
                            COPPER ACTIVITY ( pCu )
Figure 2-4. Body burdens of copper in oysters (Crassostrea virginica) versus total copper
(top) and copper activity (bottom) with different levels of the complexing ligand NTA (this
                    figure is from Zamuda and Sunda, 1982).

-------
                         COMPARISON OF WATER AND

                             SEDIMENT EXPOSURE

                           (AFTER R.C. SWARTZ et al., 1985)
        20
        15
  CO
  Ill
  5
  CO



  X

  g
  HI
  X
  OZ
         10
                                                    SEAVWTER

                                                    INTERSTITIAL

                                                    WATER
                                         t=4 t=0
                                  2
                        DISSOLVED CADMIUM CONC. (mg/L)
 Figure 2-5. Rhepoxynius abronis mean survival versus dissolved cadmium for 4-day toxicity

tests in seawater (symbols) and interstitial water at time 0 and 4 days (bars) (this figure is from

                              Swartzet al., 1985).

-------
                                                                                      16
       Experiments were performed to determine the role of acid volatile sulfides in cadmium
spiked 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
to the interstitial water cadmium activity measured with a specific ion electrode. Four-day
water only and 10-day exposure sediment toxicity tests were performed. The water-only
response data for Ampelisca and Rhepoxynius are included for comparison although they
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 particule bound  and dissolved interstitial cadmium.  They
circulated water of the same cadmium concentration through different sediments.  This resulted
in differing bulk sediment concentrations, but the same interstitial water concentrations. They
found no statistically significant difference in organisim response for the different sediments.
Since the interstitial water concentrations were  the same in each treatment - the circulating
water concentrations established the interstitial  water concentrations - these experiments
confirmed the equal response to concentrations in water-only and interstitial water hypothesis.
        *.
       A series of 10-day toxicity tests using the amphipod Hyalella azteca were performed to
evaluate the bioavailability of copper in sediments from two sites highly contaminated with this
metal: Steilacoom Lake, Washington and Keweenaw Watershed,  Michigan (Ankley et al.,
1993).  A water-only, 10-day copper toxicity test was also 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 LCSOs show
strong agreement for the water-only (31 ug/L) and the Keweenaw sediment test (28 ug/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 ug/L detection limit and were
consistent with the observed lack of toxicity  to  H.  azteca (Ankley et al, 1993).

       The data presented in this subsection, and data to be presented in Section 3 of this
document, 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

-------
 §
 ^
 0)
a.
100
 80
 60
 40
 20
         • Water Only Exposure
         O Sediment Exposure
                                   10
                                                         100
1000
                                     Copper (ug/L)
     Figure 2-7. Toxicity of copper to Hyalella azteca versus copper concentrations in a water-
       only exposure (open symbols) and interstitial water copper concentrations in sediment
    exposures (closed symbols) using Keweenaw Waterway sediments (this figure is from  Ankley
                                       etal.,1993).

-------
                                                                                     19
data from water-only exposures can be used to predict sediment toxicity using interstitial water
concentrations. Therefore, the metal activity or concentration in interstitial water would be an
important component of a partitioning model needed to establish sediment guidelines.  The
solid metal-binding phases of this partitioning model need to be identified. The following
subsection presents data that identifies solid phase sulfides 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 the articles in Stumm, (1987) and
Dzombak and Morel, (1990) for recent summaries.]  The models consider surface
complexation reactions as well as electrical interactions via models of the double layer.
Models for natural soil and sediment particles are less well developed. However, recent
studies suggest that similar models can be applied to soil systems (Allen et al., 1980; Barrow
and Ellis, 1986a,b,c; Sposito et al., 1988). Since the ability to predict partition coefficients is
required  if interstitial water metal concentration is to be inferred from the total concentration,
some practical model is required. This subsection presents state of the science in the
theoretical development of metals partitioning in sediments.

2.2.1  Metal Sorption Phases

       The initial difficulty that one confronts in selecting an applicable sorption model is that
the available models are quite complex and many of the parameter estimates may  be specific to
individual soils or  sediments. However, the success of organic carbon based non-ionic
chemical sorption models suggests that some model of intermediate complexity that is based on
an identification of the  sorption phases may be more generally applicable.

       A start in this direction has been presented (Di Toro et al., 1987; Jenne et al., 1986).
The basic idea was that instead of considering only one sorption phase as is assumed for
non-ionic hydrophobic  chemical sorption, multiple sorption phases must be considered. The
conventional view  of metals speciation in aerobic soils and sediments is that metals are
associated with the exchangeable, carbonate and Fe and Mn oxide forms, as well  as organic
matter, stable metal sulfides. and a residual phase.  In oxic soils and freshwater sediments

-------
               -  r/(
-------
                                                                                    21

         Cd2*  + FeS(s) - CdS(s) + Fe2*                                         (2-2)

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 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, [Cd]A, to the amount of FeS initially present,  [FeS(s)]it is
shown hi Figure 2-8. The plot of dissolved cadmium versus cadmium added illustrates the
increase in dissolved cadmium that occurs near [Cd]A / [FeS(s)]j = 1.  A similar experiment
has been performed for amorphous MnS with comparable results. 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 and manganese
sulfide can readily be displaced by adding cadmium.  As a consequence it is a source of
available sulfide which must be taken into account in evaluating the relationship between solid
phase and aqueous  phase cadmium in sediments.

      A direct confirmation that the removal of cadmium was via the displacement of iron
sulfide is shown in Figure 2-9.  The supernatant from  a titration of FeS by  Cd:* was analyzed
for both cadmium and iron. The solid lines are the theoretical expectation  based on the
stoichiometry of the reaction (Equation 2-9) (DiToro et al.. 1990).

2.2.2.2       Sediments

      A similar titration procedure has been used to evaluate the behavior of sediments taken
from four quite  different marine environments: sediments from Black Rock Harbor, Hudson
River and the sediments from Long Island Sound and Ninigret Pond used in the toxicity tests.

-------
                     CADMIUM TITRAT1ON OF IRON SULFIDE

                   1.0
                   0.8 -
             *S>


             •o
             o
             o
             IU


             O
             w
                   0.
                     0.0      0.5      1.0       1.5      2.0



                       CADMIUM ADDED (umol Cd/umol FeS)
Figure 2-8.  Cadmium titrations of amorphous FeS.  X-axis is cadmium added normalized by

  FeS initially present (this figure is from Di Toro et al., 1990). Y-axis is total dissolved

     cadmium.  The lines connecting the data points are an aid to visualizing the data.

-------
            I
            u.
             o c
            II
            4_l w
             O
             o
             o
            o
                      RESULTS OF FeS + Cd** = CdS •»• Fe**
                              Concentration of Fe**
                                 Analysis of Filtrate
600
 -25
                      0  0.2  0.4 0.6 0.8  1   1.2  1.4  1.6  1.8
                               Cd**/FeS Molar Ratio
            u
            "8
            o
            o
            o
            U
700
600
500
400
300
200-
100
                              Concentration of Cd**
                                 Analysis of Filtrate
   0   0.2  0.4 0.6 0.8  1  1.2  1.4  1.6  1.8
             Cd'VFeS Molar Ratio
                        Theory    « 1.14       -1-15
                        1-18      -*> 1-19, w/Buffer
Figure 2-9.  Concentrations of Fe2+ and Cd2* in supernatent from titration of FeS by Cd2H
                   (personal communication with Di Toro 1992).

-------
                                                                                    24
The binding capacity for cadmium is estimated by extrapolating a straight line fit to the
dissolved cadmium data. The equation is:

         [SCd(aq)]  =  max{0,m([Cd]A -  [Cd]B)}                                  (2-3)
where [SCd(aq)] is the total dissolved cadmium, [Cd]A is the cadmium added, [Cd]B is the
bound cadmium, and m is the slope of the straight line. The sediments exhibit quite different
binding capacities for cadmium, listed in Table 2-1, ranging from approximately 1 /umol/gm to
more than 100 /zmol/g. The question is whether this binding capacity is explained by the solid
phase sulfide present in the samples.

        Table 2-1.  Cadmium binding capacity and AVS of sediments (from Di Toro et
                                       al.. 1990).
                             Initial AVS     Final AVS     Cd Binding Capacity(e)
     	Sediment	(Atmol/g)(a)     Qimol/g)(b)	(/zmol/g)	
      Black Rock Harbor      175.(41.)         -               114.(12.1)
      Hudson River           12.6(2.8)         -               8.58 (2.95)
      LISound(c)              15.9(3.3)     13.9(6.43)          4.57(2.52)
      Mixture(c)                5.45 (-)      3.23(1.18)
      Ninigret Pond(c-d)       2.34(0.73)     0.28(0.12)          1.12(0.42)
      (a)Average (Standard Deviation) AVS of repeated measurements of the stock
      (b)Average (Standard Deviation) AVS after the sediment toxicity experiment
      (c)From original cadmium experiment (Di Toro et al.,  1990).
      W)50/50 mixture of LI Sound and Ninigret Pond
      (c)From Equation (2-3)

       2.2.3  Correlation to Sediment AVS

       The majority of sulfide in sediments is in the form of iron monosulfides  (mackinawite
and greigite) and iron bisulfide (pyrite), of which the former are the most reactive. These
sediment sulfides can be classified into three broad classes which reflect the techniques used
for quantification (Morse et al., 1987; Berner, 1967; Goldhauber and Kaplan, 1974).  The
most labile fraction, acid volatile sulfide (AVS), is associated with the more soluble iron and

-------
                                                                                     25
manganese monosulfides. The more resistant sulfide mineral phase, iron pyrite, is not soluble
in the cold acid extraction used to measure A VS.  Neither is the third compartment, organic
sulfide associated with the organic matter in sediments (Landers et al., 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, Table 2-1,
suggests that some AVS is lost during titration experiment.  In any case, the covariation of
sediment binding capacity and AVS is clear.  This suggests that AVS is the proper
quantification of the solid phase sulfides that can be dissolved by 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
concentration via the reaction:

         FeS(s) - Fe2*  +  S2'                                                    (2-4)
If cadmium is added to the aqueous phase, the result is:


         Cd2' - FeS(s) ~ Cd2'  + Fe2' +  S:-                                    (2-5)


As the cadmium concentration increases, [Cd2*][S2~] will exceed the solubility product of
cadmium sulfide and CdS(s) will start to form.  Since 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:

              ' + FeS(s) - CdS(s)  +  Fe2*                                          (2-6)

-------
                      CADMIUM TITRATION OF SEDIMENTS
                                AVS
                               AVS
AVS
        o
        o
        UJ

        3
        o
        (0
        (0
        Q
1.0




0.8




0.6




0.4
0.2
                                 T        T  T
                   * BR HARBOR


                   • LI SOUND



                   * HUDSON RIVER


                   + NINIGRETPOND
                              1.0         10.0           100.0


                        CADMIUM ADDED (umol Cd/gm dry wt)
                                                           1000.0
Figure 2-10. Cadmium titration of sediments from Black Rock Harbor, Long Island Sound, Hudson River and Ninigret Pond (this

     figure is from Di Toro et al., 1990). Cadmium added per uniidry weight of sediment versus dissolved cadmium.

-------
                                                                                        27
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 an analysis of the
M(H)-Fe(II)-S(-II) system with both MS(s) and FeS(s) present in Appendix B.  M(II) represents
any divalent metal that forms a sulfide that is more insoluble than FeS.  If the added 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
products:
                        KMS/KFeS                                                   (2-7)
This is a general result that is independent of the details of the interstitial water chemistry.  In
particular it is independent of the Fe2* activity.  Of course the actual value of the ratio
{M2*}/[M]A 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 since the data presented in Section 2.1 indicates that
toxicity is related to metal activity, {M2*}.  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 sulfide solubility  products and the ratios are listed in Table 2-2.  The ratio of
cadmium activity to total cadmium is less than 10 'I046.  For nickel the ratio is less than 10"5 59.
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, Zn, Cd, Pb, Cu and Ag indicate that metal activity for these metals will be very
small in the presence of excess AVS.

-------
                                                                                     28
                        Table 2-2. Metal sulfide solubility products.
Metal Sulfide
FeS
NiS
ZnS
CdS
PbS
CuS
Ag2S
log
K.I
-3.64
-9.23
-9.64
-14.10
-14.67
-22.19
-36.14
log
-22.39
-27.98
-28.39
-32.85
-33.42
-40.94
-54.71
Log

-5.59
-6.00
-10.46
-11.03
-18.55
-32.32
 "Solubility products, K^, for the reaction M2+ + US' - MS(s) + IT for CdS (greenockite),
 FeS (mackinawite), and NiS (millerite) from Emerson et al., 1983. Solubility products for CuS
 (covellite), PbS (galena), ZnS (wurtzite), and Ag2S (acanthite) and pK2 = 18.57 for the reaction
 HS~ ~ H* + S2' from Schoonen, M.A.A. and H.L. Barnes,  1988.  K^ for the reaction M2+ +
 S2" ~ MS(s) is computed  from log Kjp 2 and pK2.
2.2.5  Application to Mixtures of Metals
         t
       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 to A VS.  Since all these
metals have lower sulfide solubility parameters than FeS, they would all exist as metal sulfides
if their molar sum (assume [Ag]/2 because it is monovalent) is less than the AVS.  For this
case
           [AVS]

no metal toxicity would be expected where [MT]j is the total cold acid extractable ith metal
molar concentration in the sediment (divide by 2 for silver).  On the other hand if their molar
sum is greater than the AVS 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:

-------
                                                                                      29
          [AVS]
                  > 1                                                              (2-9)
These two equations are precisely the formulas that one would employ 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.

       The appropriate quantity of metals to use  in the metals/AVS  ratio  is referred to as
"simultaneously extracted metal" or SEM.  This  is the metal which is extracted in the  cold acid
used in  the AVS procedure. This is the appropriate quantity to use because some metals form
sulfides which 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 which 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 to AVS ratio (Di Toro et al.,
1992).

       The above discussion is predicated on the assumption that all the metal  sulfides behave
similarly to cadmium sulfide.  Further 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 metals mixtures will be presented in Section III which follows.

-------
                                                                                   30
                                     SECTION 3

                       TOXICITY OF METALS IN SEDIMENTS

3.1    GENERAL INFORMATION

       This Section of the guidelines document summarizes data from acute and chronic
toxicity tests that demonstrate that the absence of sediment toxicity due to metals can be
predicted by use of the interstitial water concentrations of metals or by comparison of the
molar concentrations of AVS and SEM. This ability to predict the toxicity of metals in
sediments, through a fundamental understanding  of their bioavailability, is in sharp contrast
with the absence of a causal correlation between  metals induced sediment metal-induced
toxicity and metals concentrations based on the dry weight of sediment (Berry et al., 1996a; Di
Toro et al., 1990; 1992; Luoma, 1989, Berry et  al., in review) and the use of sequential
extractions (Ankley et al., 1996).

3.2    PREDICTING METAL TOXICITY:  SHORT-TERM

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 non-ionic 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 are correlated with observed
biological effects.  However, as opposed to the situation for non-ionic organic chemicals and
organic carbon (Di Toro et al.. 1991), sediment  partitioning phases controlling interstitial
water concentrations of metals and  metal-induced sediment toxicity were not known.

       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, the  AVS was  defined as the sulfide liberated from wet sediment

-------
                                                                                    31
by treatment with IN HC1 acid.  Di Toro et al. (1990) found that when the molar concentration
of AVS in the test sediments was larger than that of the molar concentration of total cadmium
(i.e., when the cadmium:AVS ratio was less than 1, or the cadmium:AVS difference was less
than 0), interstitial water concentrations of the metal were small and no toxicity was observed
in 10-d tests with the amphipods Rhepoxynius hudsoni or Ampelisca abdita. Studies by
Carlson et al. (1991) with cadmium-spiked freshwater sediments yielded similar results; when
there was more AVS than total cadmium, significant toxicity was not observed in 10-d tests
with oligochaetes (Lumbriculus variegatus) or snails (Helisoma sp). Based upon these initial
studies, another study with nickel-spiked sediments using A. abdita and field sediments
contaminated with cadmium and nickel using the freshwater amphipod Hyalella azteca, Di
Toro et al.  (1992) provided further support to the importance of AVS in controlling metal
bioavailability in sediments. Based on these studies they suggested that it may be feasible to
derive ESG for metals by direct comparison of molar AVS concentrations to the molar sum of
the concentrations of cationic metals (specifically, cadmium, copper, nickel, lead and zinc)
extracted with the AVS; i.e., £SEM. They observed that the expression of metals
concentrations based on the £SEM 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 £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-d toxicity tests with the marine
polychaete  Capitella capitata exposed to sediments spiked with zinc, lead or 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 concentration in interstitial water were less
than the water-only LC50. Toxicity was often observed when these were exceeded.  Green et
al. (1993) reported results of another spiking experiment supporting the general EqP approach
to deriving sediment guidelines 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 Amphiascus tenuiremis was predictable based upon interstitial water (but
not sediment dry wt) cadmium concentrations. Further spiking experiments by Pesch et al.
(1995) demonstrated that 10-d survival of the marine polychaete Neanthes arenceodentata was
comparable to controls in cadmium- or nickel-spiked sediments  with more AVS than SEM.

-------
                                                                                     32
       Berry et al.(1996a) described experiments in which A. abdita was exposed for 10 days
to two or three sediments spiked either singly, or in combination, with cadmium, copper,
nickel, lead 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
to interstitial water metal concentrations expressed as toxic units (IWTU):
          IWTU=[Md]/LC50                                                       (3_1)

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. (1996a)
reported that significant (>24%) mortality of the saltwater amphipod occurred in only 3.0% of
sediments with less than 0.5 IWTU, while samples with greater than 0.5 IWTU were toxic
94.4% of the time. Berry et al. (1996a) 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 > lead > 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 (e.g., zinc would replace nickel .in a monosulfide complex and nickel  would be
liberated to the interstitial water, etc).  Berry et al. (1996a) observed this trend in sediments
spiked with cadmium, copper , nickel and zinc (Figure 3-1).  Furthermore, an increase in the
concentration in a sediment of a metal with a low sulfide solubility product constant (Ksp)
theoretically  would displace a previously unavailable and nontoxic metal, with a higher K


-------
                    IW Metal vs SEM/AVS: SW Mixed Metals - LIS
         10000
o
o

0>
      JD
      c
                 -•- Zn  -28J9

                 -•- Cd  -32.85

                 -A- Cu  -40.94
                                                  10
                                                       100
1000
                                        SEM/AVS
                      IW Metal vs SEM/AVS: SW Mixed Metals - NIN
                                                                         1000
 Figure 3-1.  Concentrations of individual metals in interstitial water of sediments from Long
Island Sound (top) and Ninigret Pond (bottom) in the mixed metals experiment as a function of
    SEM/AVS ratio (this figure is from Berry et al., 1996).  Concentrations below the IW
  detection limits, indicated by arrows, are plotted at one half the detection limit.  K^ is the
                          sulfide solubility product constant.  •

-------
                                                                                     34
 3.2.2  Spiked Sediments: All experimental results summarized

       Berry et al. (1996a) summarized the results of all of the above studies using 10-day
 toxicity tests with saltwater sediments spiked with cadmium, copper, lead, nickel, silver or
 zinc and metal mixtures using amphipods ( Di Toro et al.,  1990; Berry et al., 1996a; 1996b),
 polychaetes exposed to sediments spiked with cadmium, copper, lead, nickel or zinc (Casas
 and Crecelius, 1994; Pesch et al.,  1995) and copepods exposed to sediments spiked with
 cadmium (Green et al.,  1993), and freshwater tests using oligochaetes and snails exposed to
 sediments spiked with cadmium (Carlson et a., 1991).  These data describe tests with seven
 freshwater  and saltwater species and sediments from seven locations, with AVS concentrations
 ranging from 1.9 to 65.7 /xmol/g dry wt and TOC ranging  from 0.15 to 10.6%  (Green et al.,
 1993 measured interstitial cadmium but not AVS).  Similar results from 10-day tests with
 amphipods in two marine sediments are summarized in Berry et al., in review.

       Overall, the results of these experiments demonstrate that it is not possible to predict
 the toxicity of sediments spiked with metals using the total metal concentration on a dry
 weight basis (Figure 3-2a). Sediments having z 24% mortality are considered nontoxic as
 defined by Berry et al.,  1996a,  and as indicated by the horizontal line in each panel of
 Figures 3-2 and 3-3. Much of this variability is caused by the fact that the relationship
 between mortality and total metal concentrations in tests was sediment specific as it was in the
 cadmium results shown  in Figure 2-10.  The dry weight metal concentrations required to cause
.acute mortality in these  experiments are very high relative to those often suspected to be of
 toxicological significance in field sediments. This has sometimes been interpreted  as a
 limitation of the use of SEM  and AVS to predict metal-induced toxicity.  However, the range
 in AVS in these sediments spiked with metals is similar to  sediments commonly occurring in
 the field. The important point 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 which is not
 bioavailable (Di Toro et al.,  1990).

       In contrast, the combined data from all available freshwater and saltwater spiked-
 sediment experiments supports the use of IWTU to predict mortality of benthic species in
 spiked sediment toxicity tests (Figure 3-2b). Mortality in these experiments was sediment
 independant when plotted against IWTU.  Sediments with  IWTUs of < 0.5 were generally not
 toxic. Of the 96 sediments with IWTU < 0.5, 96.9% were not toxic, while 76.4% of the 89

-------
                                                                                35
Table 3-1. - Toxicity of sediments from saltwater (SW) and freshwater (FW) field locations,
spiked-sediment tests and combined field and spiked-sediment tests as a function of the
difference between the molar concentrations of SEM and AVS (SEM-AVS), interstitial water
toxic units (IWTUs) and both SEM-AVS and IWTUs (modified from Hansen et al., 1996a).
Study
Type Parameter
Lab-Spike SEM-AVS
(FW&SW)
IWTU

SEM-AVS, IWTU

Field SEM-AVS
(FW & SW)
IWTU

SEM-AVS, IWTU

All SEM-AVS

IWTU

SEM-AVS, IWTU


Value
sO2
>03
< 0.5
* 0.5
z O2, < 0.5
> O3, z 0.5
*02
>03
< 0.5
* 0.5
s O2, < 0.5
> O3, * 0.5
<02
> O3
< 0.5
> 0.5
< 0:, < 0.5
> O3, s 0.5

n
101
95
96
89
83
78
57
79
79
53
49
45
158
174
175
142
132
123
Percent of J
Nontoxic1
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
Sediments
Toxic1
2.0
73.7
3.1
76.4
2.4
85.9
1.8
40.5
1.3
54.7
0.0
66.7
1.9
58.0
2.3
68.3
1.5
78.9
1 Nontoxic sediments < 24 percent mortality.  Toxic sediments > 24 percent mortality.
2 SEM-AVS <0 is the same as an SEM/AVS ratio of < 1.0.
3 SEM-AVS >0 is the same as an SEM/AVS ratio of > 1.0.

-------
                      *
                       o
                      I"
                      a
                      C
                      o
 100


 80


 60


 40


 20


  0
                                                               CUDQBDOOO ODDOB
                                                                °       v
                                                               O
                                                               o
                                                            0  O     O
                                '   A'MWft
                              0.001      0.01      0.1        1        10       100      1000
                                              Total Metal or SEM (umol/g)
i i i nun i 1 1 mm i i i
100 _ B

80.


60 _



40 _

0 0,
20 _ U CU U


«T» O ^Q0< O O.
0 i i 111110"$ XAUWMfti1
inn i i i mm i i i mm
o ODOO oo o go

0 0
- 0
o

o

o
0 0
0
t o

o o ° g o
o •

/o o
i i i mill
03 OQDO.fl
o o.
O -.
o



_

Q

_






i i nun
                              0.001      0.01      0.1       1        10       100      1000
                                             Interstitial Water Toxic Units
                      C
                      o
100

 80

 60

 40
                                    1 1 HUM   1 III Illll   1  III INI
                                                                i i nun   i i i mill
                                                                        OO O
                                                                                  O  O
                                                                 O
                                                                  O
                             0.001     0.01
                    0.1       1        10

                          SEM/AVS
                                                                             100
                                                                                      1000
Figure 3-2.  Percentage mortality of saltwater and freshwater benthic species in 10-day toxicity tests in sediments
spiked with individual metals (Ag, Cd, Cu, Ni, Pb, 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 cadmium, copper, lead, nickel and zinc in ^moles
 metal per gram dry weight sediment; (b) SEM/AVS ratio; and (c) interstitial water toxic units (silver data from
Berry et al., in review, all other data modified after Berry et al., 1996a). Species tested include: the oligochaete
(Lumbriculus variegatus), polychaetes ( Capitella capitata and Neanthes arenaceodentaia), amphipods (Ampelisca
 abdita and Hyalella aztecd), harpacticoid copepod (Amphiascus tenuiremis) and gastropod (Helisoma sp.).  Data
  below the SEM detection limit are plotted at SEM/AVS = 0.01. Data below the detection limit of metals in
                               interstitial water are plotted at IWTU  = 0.01.

-------
#
                   ta
                   i
                                O Spiked Sediment
                                O Reid Sediment
100


 80


 60


 40


 20
                           0.001     0.01      0.1       1        10       100
                                           Total Metal or SEM (umol/g)
                                                                1000
                  1
                   o
                                                   25       50
                                                   SEM-AVS
                                                 75
                                                        100
                                                         125
                   
-------
                                                                                    38
sediments with IWTU > 0.5 were toxic (Table 3-1). This close relationship between IWTU
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), sediments spiked with non-
ionic organic chemicals  (Di Toro et al., 1991; Adams et al.,  1985; Swartz et al., 1990) and
field sediments contaminated with non-ionic organic chemicals (Hoke et al., 1994; Swartz et
al., 1994).

          The interstitial water metal concentrations in all spiked-sediment studies were
usually below the limit of analytical detection in sediments with SEM/AVS ratios below 1.0
(Berry et al., 1996a).  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 orders of
magnitude increase in interstitial water metals concentration with only a factor of two or three
increase in sediment concentration is 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 micro-habitats. It also explains why the toxicity of different metals in
sediments to different species is 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 metals binding phases in sediments.

          •The combined data from all available saltwater and freshwater spiked sediment
experiments supports 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 ratio  (Figure 3-2c).  Mortality in these experiments was sediment
independent when plotted on an SEM/AVS basis.  With the combined data, 98.0% of the  101
metals-spiked  sediments with SEM/AVS ratios < 1.0 were not toxic, while 73.7% of the 95
sediments with SEM/AVS ratios > 1.0 were toxic (Table 3-1).

          These overall data show that when both SEM/AVS ratio and IWTU are used,
predictions of which sediments would be toxic were improved. Of the 83 sediments with
SEM/AVS ratios <  1.0 and IWTU < 0.5, 97.6% were not toxic, while 85.9% of the 78
sediments with SEM/AVS ratios > 1.0 and IWTU * 0.5 were toxic (Table 3-1). ( Note: Table
3-1 uses SEM-AVS instead of SEM/AVS ratios. An SEM-AVS of sO is the same as an
SEM/AVS ratio  of < 1.0.  An SEM-AVS of >0 is the same as an SEM/AVS ratio of > 1.0.)

-------
                                                                                    39
          These results over all experiments show that SEM/AVS and IWTU are accurate
predictors of the absence of mortality in sediment toxicity tests, however, predictions of which
sediments 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.  While the  SEM/AVS model of bioavailability accurately predicts which
sediments will not be toxic, a model which utilizes  SEM/AVS ratios or (SEM-AVS) (Hansen et
al., 1996a) and incorporates other binding phases might more accurately predict which
sediments will be toxic (Di Toro et al., 1987; Mahony et al., 1996).
                                                                          %

          Organism behavior may also explain why sediments with SEM/AVS ratios of >
1.0 were not toxic. Many of the sediments which had the highest SEM/AVS ratios in excess
of 1.0 that produced little or no mortality were from experiments using the polychaete,
Neanthes arenaceodentata (see Pesch et al., 1995, Figure 8).  This appeared to be related, in
part, to the ability of this polychaete to avoid burrowing into the test sediments, thereby
limiting their 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.t in freshwater sediments with high SEM/AVS ratios. These snails are epibenthic and also
have the ability to avoid contaminated sediments (G. Phipps, personal comm.).  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 with £ 0.5 IWTUs were not toxic.
This is likely the result of IW ligands which reduce the bioavailability and toxicity of
dissolved metals, sediment avoidance by polychaetes or snails, or methodological problems in
contamination-free  sampling of IW.  Ankley et al.  (1991) suggested that a toxicity correction
for the hardness of the IW is needed to compare toxicity in IW to that in water-only tests.
Absence of a hardness-correction might affect the accuracy of prediction of metal-induced
sediment toxicity using IWTUs in freshwater.  Further, a  significant improvement in the
accuracy of metal-induced toxicity predictions using IWTUs might be achieved if DOC binding
in the IW is taken into account.  Green et al. (1993) and Ankley et al. (1991) hypothesized that
increased DOC in the IW reduced the bioavailability of cadmium in their sediment exposures,
relative to the water-only exposures. Green et al. (1993)  found that the LC50 value for
cadmium in  an IW exposure without sediment was more  than twice that in a water-only

-------
                                                                                    40
exposure, and that the LC50 value for cadmium in IW 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 Hyalella azteca
to 17 sediment samples along a gradient of cadmium and nickel contamination frofti a
freshwater/estuarine  site in Foundry Cove, NY. In 10-d toxicity tests, H. azteca mortality was
absent 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. Lumbriculus variegatus
was far less sensitive to the sediments than H. azteca, 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 from Foundry Cove, Di Toro et al. (1992) observed metals concentrations from 0.1
to 28 emotes SEM/g sediment were not toxic in some sediments, whereas 0.2 to  1000 umoles
SEM/g were lethal in other sediments indicating that the bioavailable fraction of metals in
sediments varies from sediment to sediment. In contrast, they observed a clearly discernable
mortality-concentration relationship when mortality was related to the SEM/AVS molar ratio
(i.e., there was no signficant mortality where SEM/AVS ratios were  < 1.0, mortality increased
in sediments having SEM/AVS ratios  1.0-3.0, and there was  100% mortality in sediments with
ratios  > 10). The sum of the interstitial water toxic units (IWTU) for cadmium and nickel
ranged from 0.08 to  43.5.  Sediments with < 0.5 IWTUs were always nontoxic, those with
>2.2 IWTUs \vere 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 percent to the sum of the toxic units because
cadmium is 67 times more toxic to A.  abdita than nickel illustrating the utility of interstitial
water concentrations of individual metals  in assigning the probable cause of mortality in
benthic species (Hansen et al., 1996a).

          In tests with the same sediments from this location, Pesch et al. (1995) observed
that six of 17 sediments tested had SEM/AVS ratios < 1.0, interstitial water toxic units <0.5,

-------
                                                                                    41
and none of the six were toxic to the polychaete Neanthes arenaceodentata.  None of the 11
sediments that contained SEM/AVS ratios > 1.0 were toxic.  This was not surprising because
only one sediment had >0.5 IWTUs and because this polychaete is not sensitive to cadmium
and nickel and can avoid 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 upon interstitial water
copper concentrations in the test sediments, the  10-d LC50 for H. azteca was 31 ^g/L; this
compared favorably with a measured LC50 of 28 //g/L in a 10-d water-only 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.  Recent studies suggest that an
important source of the extra binding  capacity in these sediments was organic carbon (Mahony
et al., 1996; U.S. EPA, 1994a).

          Hansen et al. (1996a) investigated the biological availability of sediment-associated
divalent metals to A. abdita and H. azteca 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 49 sediments
evaluated where metals were the likely cause of toxicity (i.e., those with less SEM than AVS
and those with less than 0.5 IWTU), no toxicity was observed. One third of the 45 sediment
samples with more SEM than AVS and more than 0.5 IWTU were toxic.

          Hansen et al. (1996a) made an observation that is important to the interpretation of
the toxicity of sediments from field locations, particularly those from industrial harbors.  They
observed that if these sediments are toxic and SEM/AVS ratios are < 1.0 non-metals associated
toxicity should always be suspected even if metals concentrations are very high on a dry weight
basis.  Further, they stated that the use of such data to reach the conclusion  that this EqP
approach is not valid is incorrect.  This is because when SEM/AVS ratios were less than 1.0
there was an almost complete absence of toxicity in spiked sediments, and field sediments
where metals were the only known source of contamination and  IWTUs for metals were
<0.5. Metals concentrations, when expressed  on a sum of the IWTU basis can, therefore.

-------
                                                                                    42

provide insight that in part may explain apparent anomalies between SEM/AVS ratios and the
observed toxicity of these sediments and sediments from other field sites.  The joint use of both
SEM/AVS ratios and interstitial water concentrations are powerful tools for explaining  the
presence of toxicity when SEM/AVS ratios are < 1.0 and the absence of toxicity when
SEM/AVS ratios are > 1.0.  Over all saltwater and freshwater field sediments tested in the
laboratory, 100% were not toxic when SEM-AVS was sO.O and IWTUs were <;0.5 and 66.7%
were toxic when SEM-AVS was  >0.0 and IWTUs were >0.5 (Table 3-1).

3.2.4      Field Sites  and Spiked Sediments Combined

           Figures 3-3, a,b, and c and Table 3-1 summarize available data from saltwater and
freshwater sediments spiked with individual metals or metal mixtures, saltwater field sites and
freshwater field sites on the utility of metals concentrations in sediments normalized by dry
weight, interstitial water toxic units (IWTUs) or SEM/AVS ratios to explain the bioavailability
and acute toxicity of metals in  sediments.  Data are from Hansen et al., 1996a and Berry et
al., in review. This analysis contains all available data from 10-day lethality tests where
mortality, IWTUs, and SEM/AVS ratios are known from experiments with sediments most
certainly 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 (Figure 3-3a).  The overlap among
bulk metals concentrations which cause no toxicity and those which are 100 percent lethal is
almost four orders of magnitude.

           Data in Figure 3-3b 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 in  recommendations for
assessing the bioavailability of metals in sediments proposed by Ankley et al.  (1994).
Saltwater and freshwater sediments spiked with metals and from field locations with SEM/AVS
ratios < 1.0 were uniformly (98.1 percent of 158 sediments) nontoxic (Figure 3-3b; Table 3-1).
The majority (58.0 percent) of 174 sediments having SEM/AVS ratios >  1.0 were  toxic.
Given the effect on toxicity or  bioavailability of the presence of other sediment phases  that also
affect bioavilabity (Di Toro et  al., 1987; Mahony et al., 1996) it is not surprising that many
sediments having SEM/AVS ratios > 1.0 are not toxic.

-------
                                                                                    43
          Data in Figure 3-3c show that over all tests, the toxicity of sediments whose
concentrations are normalized on an IWTU basis are typically consistent with the interstitial
water toxic unit concept; that is if IWTUs are £ 1.0 sediments should be lethal to £ 50 percent
of the organisms exposed; significant mortality probably should be absent at <  0.5 IWTU
(Figure 3-3c). Of the spiked and field sediments evaluated which had IWTUs < 0.5, 97.7
percent of 175 sediments were nontoxic (Table 3-1).  For the  142 sediments having IWTUs *
0.5, 68.3 percent were toxic (Table 3-1).  Given the effect on toxicity or bioavailability of the
presence of other binding phases (e.g., DOC) in interstitial water, water quality (hardness or
salinity) and organism behavior, it is not surprising that many sediments having IWTUs  * 0.5
are  not toxic.

          Over all tests, the data in Figure 3-3a, b, and c indicate that the use of both IWTUs
and SEM/AVS ratios together did not improve the accuracy of predictions of sediments that
were nontoxic (98.5 percent of 132 sediments; Table 3-1). However, it is noteworthy that
78.9 percent of the 123 sediments with both SEM/AVS > 1.0 and IWTUs z 0.5 were toxic
(Table 3-1). Therefore, the approach of using SEM/AVS ratios, 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 which sediments that contain cadmium, copper, lead,
nickel, silver and zinc will not be toxic is quite certain.  This  is very useful, because the vast
majority of sediments found in the environment in the U.S. have SEM/AVS ratios sl.O
suggesting that there should be little concern about metals in sediments (Wolfe et al.,  1994;
Hansen et al., 1996a; Leonard et al.,  1996a; Section 4 of this document) on a national
basis.even though localized areas of biologically significant metal contamination do exist.
However, a very important consideration is that most of these data are from field sites where
sediment samples were collected in the summer.  This is the time of the year when the seasonal
cycles of AVS produce the maximum metal-binding potentials (Boothman and Helmstetter,
1992; Leonard et al., 1993).  Hence, sampling at seasons and conditions when AVS is at
minimal values is a must in establishing the true level of overall concern about metals in
sediments and in evaluations of specific sediments..  Predicting which of the sediments with
SEM/AVS  > 1.0 will be toxic is presently less certain.  Importantly, the correct classification
rate seen in these experiments (accuracy of predicting which sediments were toxic was 58.0%
using the SEM/AVS ratio alone, 68.3% using IWTUs and 78.9% using both indicators) is
high. An SEM/AVS ratio > 1.0, particularly at multiple adjacent sites, should trigger
additional tiered assessments which might include characterization of the spatial (both vertical

-------
                                                                                    44

and horizontal) and temporal distribution of chemical concentration (AVS and SEM) and
toxicity * measurements of interstitial water metal and toxicity identification evaluations
(TIE's).  In this context, the SEM, AVS, IWTU approach should be viewed as only one of the
many sediment evaluation methodologies.

          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/AVS 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 3-4). Further, absence of
organism response when AVS binding  is exceeded can indicate the potential magnitude of
importance of other binding phases in controlling bioavailability. Figure 3-4 shows that for
most nontoxic saltwater and freshwater field sediments, one to 100 //moles of additional metal
would be required to exceed the sulfide binding capacity (i.e., SEM-AVS =  -1 to -100
^moles/g). In contrast, most toxic field sediments contained 1.0 to 1000 //moles of metal
beyond the binding capacity of sulfide alone. Data on nontoxic field sediments whose sulfide
binding capacity is exceeded (SEM-AVS is > 0.0 ^moles/g) indicates that other sediment
phases, in addition to AVS, have great significance in controlling metal bioavailability.  In
comparison to SEM/AVS ratios, the 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 etal., 1993).

          EPA believes that results  from tests using sediments spiked with metals and
sediments from the field in locations where toxicity  is metals-associated demonstrate the value
in explaining the biological availability of metals concentrations normalized by SEM/AVS ratio
and IWTUs instead of dry weight metals 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 ratios are < 1.0 and IWTU are < 0.5.  Expressing
concentrations of metals in sediments on an SEM-AVS basis provides important insight into
available additional binding capacity of sediments and the extent to which sulfide binding has
been exceeded. It, along with measurement of interstitial water concentrations of metals, can
potentially identify the specific metal causing toxicity. This can theoretically be accomplished
by subtracting the metals-specific molar concentrations in order of their sulfide solubility
product constants (Ksp). 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

-------
IUU

80-

g 60-
15
5 -
20-
o -
u *
-1C




D


A A a o ' ' •""" 	
»* Xf A* ^A
*^ jL*1,* o
	 *&-*£-[&— ^^^-•rfArar
)0 '-10 --1
o cant— fjy- [_h
0 O
0
D D • "
o

A
AD
o d
" " A\0 A ^V
4 , W A A X
1 mO ^ A A*A 4-4. if
to 1 10 100 1000
                                              SEM-AVS OL/mol/g dry wt)
Figure 3-4.  Percentage mortality of amphipods, oligochaetes and polychaetes exposed to sediments from three saltwater and four
  freshwater field locations as a function of the sum of the molar concentrations of SEM minus the molar concentration of AVS
 (SEM-AVS) (from Hansen et al., 1996a):  The vertical dashed line at SEM-AVS = 0.0 indicates the boundary between sulfide-
                                bound unavailable metal and potentially available metal.

-------
                                                                                    46
tests are extremely accurate (98.5 percent) using both parameters (Table 3-1).  While the
predictions of sediments likely to be toxic are less accurate, the use of SEM-AVS is extremely
useful in identifying sediments of potential concern (Table 3-1).  Hansen (1995) summarized
data 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 to the SEM-AVS concentration.  (Tests with polychaetes and
gastropods were excluded because these organisms avoid exposure.) Seventy percent of the
sediments in these amphipod studies with an SEM-AVS concentration of *0.76 ^umoles 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 /^moles of excess SEM/g, respectively.

          Of course,  SEM, AVS and IWTUs can only predict toxicity or the lack  of toxicity
due to metals in sediments. They cannot be used alone to predict the toxicity of sediments
contaminated with toxic concentrations of other contaminants. However, SEM/AVS ratios
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 the biological
availability and toxicity of cadmium, copper, lead, nickel, silver and zinc is applicable only to
anaerobic sediments that contain AVS.  In aerobic sediments binding factors other then AVS
control bioavailability (DiToro etal.,  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 theses caveats, EPA believes that the use of
SEM, AVS and interstitial measurements in combination are superior to all other currently
available sediment evaluation procedures to causally assess the implications of these six metals
associated with sediments. (See discussion in Section 5 "Implementation" 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 an absence of
metal toxicity based upon sediment SEM:AVS relationships and/or interstitial water metal
concentrations.  However, for this approach 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..

-------
                                                                                   47
1994). This demonstration was the goal of experiments described by Hare et al. (1994),
DeWitt et al. (1996), Hansen et al. (1996b), 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 ratios decrease, with
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 et al., 1994;
Peterson etal., 1996).

3.3.1     Life cycle toxicity tests

          DeWitt et al. (1996)  conducted an entire lifecycle toxicity  test with the marine
amphipod Leptocheirus plumulosus exposed for 28 d to cadmium-spiked estuarine sediments
(Table 3-2).  The test began with newborn amphipods and measured effects on survival,
growth and reproduction relative to interstitial water and SEM/AVS normalization. Seven
treatments of Cd were tested: 0  (control), 0.34, 0.74, 1.31, 1.55, 2.23  and 4.82 molar
SEMCd/AVS ratios (measured concentrations). Gradients in AVS concentration as a function
of sediment depth were greatest in the control treatment, decreased as the SEM^/AVS ratio
increased and became more pronounced  over time.  Depth gradients in  SEMCd/AVS were
primarily- due to the spatial and  temporal changes in AVS concentration, because SEMCd
concentrations changed very little with time or depth. Thus, in most treatments  SEMCd/AVS
ratios were higher 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) and for cadmium sulfide is quite slow (10% in 300 hours) (Mahony et al., 1993).
Interstitial cadmium concentrations increased in a dramatic step-function fashion in treatments
having SEM/AVS ratios >2.23; and were below the 96-h LC50 for this amphipod in lesser
treatments. There were no significant effects on survival, growth or reproduction in sediments
containing more AVS than cadmium (SEM/AVS ratios 0.34, 0.74, 1.31 and 1.55), in spite of
the fact that  these samples contained from 183 to  1370 ^g cadmium/g sediment. All
amphipods died in sediments having SEM/AVS ratios 2.23.  These results are consistent with
predictions of metal bioavailability from acute tests with metal-spiked sediments (i.e., that
sediments with SEMCd/AVS ratios <> 1 are not toxic, interstitial water metal concentrations are
related to organism response and sediments with SEMCd/AVS  ratios >  1 may be toxic).

-------
 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.
Toxicity Test Metal
Life-cycle:
Leptocheirus Cadmium
plumulosus
Chironomus Zine
tentans
Colonization;
Laboratory- Cadmium
saltwater
Field-saltwater Cadmium, copper, nickel,
7.ine
Field- Cadmium
freshwater
Field- Xinc
freshwater
Duration Measured
(dayS) NOECs

28 -3.4, -2.0,
0.78, 1.9
56 -4.3, -2.6,
-1.4,6.4

118 -13.4
120 <0
-0.45, -0.25
0.5b
368 -3.6, -3.5,
2.9, -2.0
SEM-AVS3
OECs

8.9, 15.6
21.9,
32.2

8.0, 27.4
-
, 4.5"
1.0
Effect

Mortality 100%
Larval mortality 85-100%.
Weight and emergence reduced.

Fewer polychaetes, shifts in
community composition, fewer
species, bivalves absent,
tunicates increased.
No effects observed.
Reduced Chironomus numbers.
Bioaccumulation.
Occasional minor reductions in
Naididae oligochaetes.
Reference

DeWitt et
al., 1996
Sibley et
al., 1996

Hansen et
al., 1996
Boothman
etal., 1996
Hare et al.,
1994
Liber et
al., 1996
8 SI-M- AVS differences are used instead of SEM/AVS ratios to standardize across the studies referenced. An SEM-AVS difference
of <0 is the same as an SP.M/AVS ratio of < 1.0.  An SEM-AVS difference of >0 is the same as an SEM/AVS ratio of >1.0.
11 Nominal concentrations.

-------
                                                                                    49
          Sibley et al. (1996) reported similar results from a 56-d life-cycle test conducted
with the freshwater midge Chironomus tentans exposed to zinc-spiked sediments (Table 3-2).
The test was initiated with newly hatched larvae and lasted through one complete generation
during which survival, growth, emergence and reproduction were monitored. In sediments
where the molar difference between SEM and AVS was  < 0, at dry wt zinc concentrations 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 exceeded 0, AVS and interstitial water concentrations of zinc increased with
increasing treatments (being highest in surficial sediments) and reductions in survival, growth,
emergence and reproduction were observed. Over the course of the study, the abselute
concentration of zinc in the interstitial water in these treatments decreased due to increase in
sediment AVS and loss of zinc due to  twice daily renewals of overlying water.

3.3.2     Colonization tests

          Hansen et al. (1996b) conducted a  118-d benthic colonization experiment in which
sediments were spiked to achieve nominal cadmium/AVS molar ratios of 0.0 (control), 0.1,
0.8 and 3.0 and 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 within two to four weeks
resulted in sulfide 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
SEMCd was always less than AVS, interstitial cadmium concentrations (<3-10 /xg/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, measured SEM^ commonly exceeded AVS in the
surficial 2.4 cm of sediment and interstitial cadmium concentrations (24-157 /^g/L) were
sufficient to be of toxicological significance to highly sensitive species.  In this treatment.
shifts in the presence or absence over  all taxa,  and fewer macrobenthic polychaetes
(Mediotnastus ambiseta, Streblospio benediai and Podarke obscura) and unidentified
meiofaunal nematodes, were observed. In the nominal 3.0 cadmium/AVS treatment,
concentrations of SEMCd were always greater than AVS throughout the sediment column.
Interstitial cadmium ranged from 28,000 to 174,000 /xg/L. In addition to the effects observed
in the nominal 0.8 cadmium/AVS treatment, sediments in the 3.0 cadmium/AVS treatment
were colonized by fewer macrobenthic species, polychaete species and harpacticoids; had
lower densities of diatoms; lacked bivalve molluscs; and exhibited other impacts. Over all

-------
                                                                                  .  50
treatments, the observed biological responses were consistent with predicted possible adverse
effects resulting from elevated SEMCd/AVS ratios  in surficial sediments and interstitial water
cadmium concentrations.

          Boothman et al. (1996) conducted a field colonization experiment in which
sediments from Narragansett Bay, RI were spiked with an equi-molar mixture of cadmium,
copper, 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).  AVS concentrations decreased with time in surface
(0-3 cm) sediments in all treatments where SEM< AVS, but did not change in subsurface (6-10
cm) sediments or in the entire sediment column in the SEM > AVS treatment.  SEM decreased
with time only where SEM exceeded AVS.  The concentration of metals in interstitial water
was below detection limits when SEM was less than AVS.  When SEM exceeded AVS,
significant concentrations of metals were present in interstitial water in order of their sulfide
solubility product constants.  Interstitial water concentrations in these sediments decreased with
time exceeding the WQC in interstitial  water for 60 days for all metals, 85 days for cadmium
and zinc, and for the entire experiment (120 days) for zinc. Benthic faunal assemblages in the
spiked sediment treatments were not different from 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 SEMrAVS treatment, due to
the oxidation of AVS and loss of SEM. The authors speculate that interstitial metal was likely
absent in the surficial sediments in spite of data demonstrating the presence of significant
measured concentrations of interstitial metal. This is because the interstitial water in the
nominal 3.0 SEM/AVS treatment was sampled from sediment depths where SEM was in
excess. It is in surficial sediments where settlement by saltwater benthic organisms first occurs.
Also,  there was a storm event which allowed a thin layer of clean sediment to be deposited on
top of the spiked sediment (Boothman.  USEPA. personal communication).  These data
demonstrate the importance of sampling of sediments and  interstitial water in sediment
horizons where benthic organisms arc active.

          Hare et al. (1994) conducted an approximately 1-yr 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 ratios < 1.0, but
increased markedly  at ratios  > 1.0.  They reported reductions in  the abundance of only the

-------
                                                                                      51
chironomid Chironomus salinarius in the nominal 10.0 SEM/AVS treatment. Cadmium was
accumulated by organisms from sediments with surficial SEM concentrations were greater than
those of A VS. 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 /xmole sulfide/gram from a freshwater mesotrophic pond (Table 3-2). Sediments
were spiked with 0.8, 1.5, 3.0, 6.0 and 12.0 //mole zinc/gram, replaced in the field and
chemically and biologically sampled over 16 mo. There was a pronounced increase in AVS
concentrations with increasing zinc concentration, AVS was lowest in the surficial Q.-2 cm of
sediment with minor seasonal variations.  With the exception of the highest spiking
concentration (ca., 700 mg/kg, dry wt), 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.  This absence
of any noteworthy biological response was consistent with the absence of interstitial water
concentrations of biological concern. This was attributed to the increase in concentrations of
iron and manganese sulfides,  produced during periods of diagenisis, 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 due to the diagenisis of organic material.  In periods of minimal diagenisis, the oxidation
rates of metal sulfides, if sufficiently great, could release biologically significant concentrations
of the metal into interstitial waters. This phenomenon should occur metal-by-metal in order of
their sulfide solubility product constants.

-------
                                                                                    52
                                      SECTION 4

               DERIVATION OF SEDIMENT GUIDELINES FOR METALS

4.1       GENERAL INFORMATION

          Section 4 of this document presents the technical basis for establishing ESG for
copper, cadmium, nickel, lead, 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 effects  levels.  Extensive toxicological data from short-term ana long-term
laboratory and field experiments, with both marine and freshwater sediments and a variety of
species indicates that it is possible to reliably predict an absence of metal toxicity based upon
EqP theory.  ESG for all six metals collectively can be derived using two procedures: (a)  by
comparing the sum of their molar concentrations, measured as SEM, to the molar
concentration of AVS in sediments (AVS Guideline); or (b) by comparing the measured
interstitial water concentrations of the metals to WQC  final chronic values (FC Vs) (Interstitial
Water Guidelines). These approaches are described in more detail below.  A lack of
exceedence of ESG based upon any one of the two procedures indicates that metal toxicity
should not occur.  Exceedence of either the AVS or Interstitial Water Guidelines is indicative
of a potential problem that would entail further evaluation.

          At present, EPA believes that the technical basis for implementing these two
approaches is supportable.  The Organic Carbon and Minimum Partitioning Approaches as
proposed to the SAB and in Ankley et al.(1996) require additional research prior to their
implementation.  Research issues for these latter two approaches include  the development of
robust partitioning datasets  for the six  metals, as well  as investigation of  factors such as the
importance of other binding phases. The four approaches have been presented to and reviewed
by the Science Advisory Board of EPA (U.S EPA. 1994a; 1995a).

          Additional research required to fully implement other approaches for deriving  ESG
for these metals and to derive ESG for other metals includes the development of uncertainty
estimates associated with any approach; part of this would include their application to a variety
of field settings and sediment types. Research also is  needed to establish the technical basis for
ESG for metals other than the six described herein, such  as mercury, arsenic and chromium.

-------
                                                                                     53
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, for example, with sediment suspension, or the
transport of metals into the food web either from sediment ingestion or the ingestion of
contaminated benthos.  This latter issue, in particular, should be the focus of future research
given existing uncertainty in the prediction of bioaccumulation of metals by benthos  (Ankley,
1996).

          The following nomenclature is used in subsequent discussion  of ESG derivation for
metals. The ESG for the metals are expressed in molar units because of the molar
stoichiometry of metal binding to AVS.  Thus, solid phase constituents (AVS, SEM) are in
moles/g dry wt.  The interstitial water metal concentrations are expressed in //moles/L, either
as dissolved concentrations [MJ or activities {M2*} (Stumm and Morgan, 1981). The
subscripted notation, Md, is used to distinguish dissolved aqueous phase  molar concentrations
from solid phase molar concentrations with no subscript.  For the combined concentration,
[SEMT], the units are moles of metal per volume of solid plus liquid phase (i.e., bulk). Note
also that when [SEMAg]  is summed and/or compared to AVS 1/2 the molar Ag concentration is
applied.

          One  final point should be made with respect to nomenclature. Use of the terms
non-toxic  and having no effect, mean only with  respect to the six metals considered in this
paper. The toxicity of field collected sediments  can be caused by other chemicals.  Therefore,
avoiding exceedences of ESG for metals does not mean that the sediments are non-toxic. It
only ensures that the six metals being considered should not have an undesirable biological
effect. Moreover, as discussed in detail below,  exceedence of the guidelines for the six metals
does not necessarily indicate that metals will cause toxicity. For these reasons,  we strongly
recommend the use of toxicity tests, TIEs, chemical monitoring in vertical, horizontal and
temporal scales, and other assessment methodologies as integral parts of any assessment
concerned with the effects  of sediment-associated contaminants (Ankley  et al., 1994).

4.2       SINGLE METAL SEDIMENT GUIDELINES

          Except in rare instances,  single metal guidelines are not usually applicable to field
situations since there is almost always  more than one metal to be considered.  As will become

-------
                                                                                     54

subsequently clear, it would be technically indefensible to derive guidelines for one metal at a
time because of the competitive nature of AVS binding.  Nevertheless, it is illustrative to
present the logic for single metals as a prelude to the derivation of the multiple metal
guidelines.

4.2.1  AVS Guidelines

       It has been demonstrated that if the SEM of a sediment is less than or equal to the

          [SEM] s [AVS]                                                           (4-1)
AVS then no toxic effects are seen. This is consistent with the results of a chemical
equilibrium model for the sediment - interstitial water system (Di Toro et al., 1992). The
resulting metal activity {M2+} can be related to the total SEM of the sediment and water, and
to the solubility products of the metal sulfide (KMS) and iron sulfide (K^) .  In particular, it is
true that at [SEM] < [AVS] then:
           [SEMT]   K
                     FeS
Because the ratio of metal sulfide to iron sulfide solubility products (KMS/KFeS) is very small
(< 10"5) even for the most soluble of the sulfides, the metal activity of the sediment is at least
five orders of magnitude smaller than the SEM  (see Di Toro et al. (1992) for data sources and
references). This indicates that no biological effects would be expected. Therefore, the
condition [SEM] < [AVS] is a "no effect" ESG.

       The reason we use the term "no effect" is that for the condition [SEM] < [AVS] no
biological impacts are expected. However, for  [SEM]  > [AVS], which might seemingly  be
considered a ESG violation, there are many documented instances where no biological impacts
occur (e.g., because organic carbon partitioning controls metal bioavailability in the interstitial
water, or the species of concern avoid or are insensitive to metals).

-------
                                                                                      55
4.2.2  Interstitial Water Guidelines

       The condition [SEM] £ [AVS] indicates that the metal activity of the sediment -
interstitial water system is low and, therefore, below toxicologically-significant
concentrations. Another way of ensuring this is to place a condition on the interstitial  water
activity directly.  Suppose that we knew the metal activity, denoted by {FCV}, that
corresponded to the [FCV].  Then the ESG corresponding to this effect level is:
           {Mr}<;{FCV}                                                            (4-3)


It is quite difficult, however,  to measure and/or calculate metal activity in a solution phase, at
the low concentrations required, since it depends on the identities, concentrations and
thermodynamic affinities of other chemically reactive species that are present.  Also the WQC
are not expressed on an activity basis. An approximation to this condition is:
         [MdMFCVd]                                                              (4_4)


where [FCVd] is the FCV applied to total dissolved metal concentrations. That is, we require
that the total dissolved metal concentration in the interstitial water [MJ be less than the FCV
applied as a dissolved guideline. Although this requirement ignores the effect of chemical
speciation on both sides of the equation - compare Equations (4-3) and (4-4) - it is the
approximation that is currently being suggested by EPA for the WQC for metals (Prothro,
1993). That is, the WQC should be applied to the total dissolved - rather than the total acid
recoverable - metal concentration (Table 4-1; U.S. EPA.1995b).  Hence, if this second
condition is satisfied it is consistent with the level of protection afforded by the WQC.

       In situations where the SEM exceeds the AVS ([SEMJ >  [AVS]), but the interstitial
water total dissolved metal is less than the final chronic value ([MJ <  [FCVJ), this sediment
would not violate the guidelines.  These cases occur when significant binding to other phases
occurs.  It should be noted that using the FCV for metals in freshwater samples requires that
the hardness of the interstitial water be measured since the WQC vary with hardness.

-------
                                                                                    56

4.3    MULTIPLE METALS GUIDELINES


       As described in the previous subsection, from a practical standpoint it is insufficient
and inappropriate to consider each metal separately. Because of the interactive nature of
metal-sulfide binding, this is of particular concern for the AVS guidelines.


Table 4-1: Water quality criteria (WQC) criteria continuous concentrations (CCC) based on the
dissolved concentration of metalab. These WQC CCC values are for use in the Interstitial
Water Criteria Toxic Unit (TWCTU) approach for deriving sediment guidelines based on the
dissolved metal concentrations in interstitial water.
Metal
Cadmium
Copper6
Lead
Nickel
Silver
Zinc
a (U.S.EPA, 1995b).
b Rounded all criteria
Saltwater CCC, Mg/L
9.3
3.1
8.1
8.2
NAf
81

to two significant figures.
?V»\»/ot^»- C**f~*r* o* o Kotvlrioee r\f *
Freshwater CCC, vz/V
£pre(0.7852(ln(hardne«)J-3.49(h]
NAf
0 9«6re<08473lln(bardness)1't'0'76l4)l

;n inn O«H onn m« r-or'n /T «••» n SA
   0.94, and 1.6 ^g cadmium/L, 6.2, 12, and 20/^g copper/L, 1.0, 2.5, and 6.1^g lead/L, 88,
   160, and 280^g nickel/L, and 58, 108, and 187 ^g zinc/L.
   CF= Conversion factor to calculate the dissolved CCC for cadmium from the total CCC for
   cadmium: CF=1.101672-[(ln hardness)(0.041838)]
   The  saltwater CCC for copper is from the "Ambient Water Quality Criteria- Saltwater
   Copper Addendum" (U.S. EPA. 1995c).
   The  silver criteria are currently under revision to reflect water quality factors that influence
   the criteria such as hardness, and pH  and any other factors.  Since 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 to the AVS so it is unlikely that
   silver would occur in the interstitial water. However in sediments contaminated with silver
   the user should be aware of the limitations in the above criteria for silver.  AVS Guidelines
   can be applied, however, the Interstitial Water Guidelines can not.  If the AVS Guideline is
   exceeded (£SEM >  AVS) and the sediment is contaminated with silver, further testing and
   evalutions would be warranted to access toxicity.

-------
                                                                                      57
4.3.1     AVS Guidelines

   The results of calculations using chemical equilibrium models indicate that metals act in an
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 of increasing solubility). Therefore, they must be considered
together.  There cannot be a guideline for just nickel, for example, since 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 SEM, then the £SEM win be misleadingly small, and it may appear that
[£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 forming metals (e.g.,  mercury) are present at high
concentrations, they also must be considered.
   The equilibrium model prediction of the metal activity is similar to the single metal example
when a mixture of the metals is present. If the molar sum  of SEM for the six metals is less
than or equal to the AVS, that  is:
                                                                                  (4-5)

then:

                                                                                  (4-6)
          [SEMJT]   KFeS


where [SEMJT] is the total SEM (^moles/L(bulk)) for the ith metal. Thus the activity of each
metal, {Mj}, is unaffected by the presence of the other sulfides.  This can be understood as
follows.  Suppose that the chemical system starts initially as iron and metal sulfide solids and
that the system proceeds to equilibrium by each solid dissolving to some extent. The iron
sulfide dissolves until the solubility product of iron sulfide is satisfied.  This sets the sulfide
activity.  Then each metal sulfide dissolves until reaching its solubility.  Since so little of each
dissolve relative to the  iron sulfide, the interstitial water chemistry is not appreciably changed.
Hence, the sulfide activity remains the same and the metal activity adjusts to meet each
solubility requirement.  Therefore, each metal sulfide behaves independently of one another.

-------
                                                                                      58
The fact that they are only slightly soluble relative to iron sulfide is the cause of this behavior.
Thus, the AVS Guidelines are easily extended to the case of multiple metals.

4.3.2      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 case
with the AVS Guideline), but rather because of their possible toxic interactions. Even if the
individual concentrations do not exceed the FCV of each metal (FCVj), the metals could exert
additive effects that might result in toxicity (Biesinger et al., 1986; Spehar and Fiandt, 1986;
Enserink et al.,  1991; Kraak et al., 1994).  Therefore, to address this potential additivity, the
interstitial water metal concentrations are converted to toxic units  (TUs) and these are summed.
Since FCVs are used as the no effects concentrations these TUs are referred  to as interstitial
water guidelines toxic units (IWGTUs). 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., SSEM> AVS).  Since 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 so it is unlikely that
silver would occur in the interstitial water. For the ith metal with  a total dissolved
concentration [Mid], the IWGTU is:

           I\VCTU=  M
                    [FCV
A lack of exceedence of the ESG requires that the sum of the IWGTUs be less than or equal to
one:
                                                                                   (4-8)
Hence, the multiple metals guideline is quite similar to the single metal case (Equation 4-4)
except that it is expressed as summed IWGTUs.

-------
                                                                                    59

  To summarize, the proposed ESG are as follows.  The sediment passes the ESG for the six
metals if either of these conditions is satisfied:

(a)        AVS Guideline:

                                                                                 (4-5)
where

£ JSEMJ = [SEMCJ  + [SEMCd] + [SEMpJ + [SEMNi] + [SEMZn] + [l/2SEMAg]

(b)        Interstitial Water Guideline:
                    T*1                                                          (4-8)
             Lrv-vi.
-------
                                                                                 60
characterizations of the SEM, AVS and interstitial water concentrations would be appropriate
(Ankleyetal., 1994).

4.4       ESG FOR METALS VS. ENVIRONMENTAL MONITORING DATABASES

   The purpose of this  Section is to compare ESG based on SEM-AVS or IWGTUs to chemical
monitoring data from freshwater and saltwater sediments in the United States. This
comparison of AVS-SEM and interstitial water concentrations can indicate the extent of metals
contamination in the United States. 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.

4.4.1     Data Analysis

   Three sources were  identified which contain both AVS and SEM databases; one also had
data on concentrations  of metals in interstitial water. Toxicity tests were also conducted on all
sediments from these sources. The databases are from the Environmental Monitoring and
Assessment Program (EMAP) (Leonard et al., 1996a), National Oceanic and Atmospheric
Administration, National Status and Trends Program (NOAA NS&T) (Wolf et al.,  1994; Long
et al., 1995; 1996) and from the Regional Environmental Monitoring and Assessment Program
(REMAP)'(Adams et al., 1996).

Freshwater sediments:

   The AVS and SEM  concentrations in the 1994 EMAP database from the Great Lakes were
analyzed by Leonard et al.  (I996a). Forty-six sediment grab samples and nine core samples
were collected in the summer from forty-two locations in Lake Michigan. SEM, AVS, TOC,
interstitial water metals (when sufficient volumes were present) and  10 day sediment toxicity to
the midge Clrironomus tentans and the amphipod Hyallela azteca were measured in sediments
collected by the grab (Appendix C).

   The AVS concentrations vs.  SEM-AVS differences from Appendix C are plotted in Figure
4-1. Grab sediment samples containing AVS concentrations below the detection limit of 0.05
umol/g AVS are plotted at that  concentration. Forty-two of the 46 (91 percent) samples had

-------
                10
               -10
            1
               -20
             at
             CO
               -30
               -40
                 0.01
                                                                     (a)
0.1             1             10
  Acid Volatile Sulfide (ymol S/g)
100
                               0.1             1             10
                                 Acid Volatile Sulfide (pmol S/g)
                                         100
   Figure 4-1. SEM minus AVS values versus AVS concentrations in EMAP-Great Lakes
sediments from Lake Michigan.  Data are from surficial grab samples only (this figure is taken
  from Leonard et al., 1996, see data in Appendix C).  The upper plot shows all values, the
    lower plot has the ordinate limited to SEM minus AVS values between -10 and +10.

-------
                                                                                   62
SEM-AVS differences greater than 0. Thirty-six of these had less than 1.0 umol of £SEM
metal per gram sediment; and none had over 5.8 ^moles/g of excess metal. In theory,
sediments  with SEM concentrations in excess of that for AVS have the potential to be toxic due
to metals.  However, the majority of exceedances occur in places where the AVS is very small
and the amount of SEM is also very small.  For these Lake Michigan sediments, a closer look
at both interstitial water metal and toxicity test results is needed. Measurement of the
concentrations of metals in interstitial water can be used to determine if the excess metals are
bound to other sediment phases, therefore, prohibiting toxicity due to interstitial metal.
Interstitial water guidelines toxic units (IWGTU) can be calculated for each metal as the
interstitial water concentration divided by the final chronic value for that metal.  Interstitial
water volumes were sufficient to measure metals concentrations in 20 of the samples. The sum
of the IWGTU for cadmium, copper,  lead, nickel and zinc in these sediments was less than 0.4
(Leonard et al., 1996a). In 10-d toxicity tests using Chironomus tentans and Hyalella azteca,
no toxicity was observed 81 % of the 21 sediments not exceeding the ESG. They conclude that
for the toxic sediments that did not exceed the metals ESG, the observed toxicity is not likely
due to metals. Further, these sediments are unlikely to be contaminated by metals (Leonard et
al., 1996a).  These data demonstrate the value of using both SEM-AVS and IWGTUs to
evaluate the risks of metals in sediments.

Saltwater sediments:

  Saltwater data from a total of 398 sediment samples from five monitoring  programs
representing the eastern coast of the United States from Chesapeake Bay to Massachusetts are
included in Figure 4-2.  The EMAP Virginia Province database (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 el al., 1995).  The NOAA data is from Long Island Sound, Boston
Harbor and the Hudson River Estuary. Sediments were collected from 63 locations in the
coastal bays and harbors of the Long  Island Sound in August, 1991 (Wolfe et al., 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 et al., 1995).  Sediment samples were collected in
the REMAP program from  140 locations from the New York/ New Jersey Harbor Estuary
System (Adams et al.,  1996). All of  the above sediment grab samples were  from
approximately the top 2 cm of undisturbed sediment.

-------
                                                                                    64
   For saltwater sediments, molar concentration of AVS typically exceeds that for SEM (SEM-
AVS  < 0) 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 percent) had an excess of metal, and only 4 of
the 68 (6 percent) had over 2 umol/g excess SEM.  As AVS levels increase above this
concentration fewer and fewer sediments have SEM-AVS differences that are positive; none
occurred when AVS was >8.1 umol/g. Unlike the sediments from the freshwater EMAP
survey in Lake Michigan, interstitial water was not measured in these saltwater sediments.
Only five of the 68 sediments (7 percent) having excess of up to 0.9 umol/g SEM were toxic in
10-d sediment toxicity tests with the amphipod Ampelisca abdita,  whereas 79 of 330 (24
percent) sediments having an excess of AVS were toxic. The data support the interpretation
that (1) toxicity was NOT metals-related  in the 79 sediments where AVS was in excess over
SEM; (2) metals might have caused the toxicity in the five toxic sediments having an excess of
metal, but even in the absence of measurements of interstitial water metals concentrations, we
speculate that metals toxicity is unlikely because there was only sO.9 umol/g excess SEM (the
molar concentration SEM most often exceeds that of AVS,  in sediments having AVS
concentrations s 1 umol/g); and (3) the absence of toxicity in sediments having an excess of
SEM of up to 4.4 umol/g indicates that significant metal-binding potential over that of AVS
existed in some sediments.   Organic carbon concentrations of from 0.05% to 15.2% (average
1.9 percent) provides for some of this additional metal-binding.

   The data above appear to suggest that in the United States direct toxicity caused by metals in
sediments is extremely rare.  While this might be true,these data by themselves are
inconclusive and it would be inappropriate to use the data from the above studies to reach this
conclusion.  All of the above studies were conducted in the summer when the seasonal
biogeochemical cycling of sulfur should produce the highest concentrations of iron monosulfide
which should 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
November to early May.  These yet to be conducted studies must monitor at a minimum SEM,
AVS, interstitial water metal and toxicity. The data presented here are not intended to be used
to draw conclusions about toxicity due to resuspension or bioaccumulation.

-------
                                                                                    65

                                     SECTION 5

                                 IMPLEMENTATION

5.1    CONSIDERATIONS IN PREDICTING METAL TOXICITY

       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 of the conclusion that an absence (but not necessarily a presence) of metal toxicity can
be reliably predicted based upon metal :sulfide relationships and/or interstitial water metal
concentrations.  In contrast, much confusion exists in the use of this convincing evidence to
interpret the significance of metals concentrations in sediments from the field when toxicity
and benthic community structure measurements are available.  In addition, the use of these
observations as a basis for predicting metal bioavailability, or deriving 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); the most
salient to the proposed derivation of ESG are described below.

5.2    SAMPLING AND STORAGE

       Accurate prediction of exposure of benthic organisms to metals is critically dependent
upon sampling appropriate sediment horizons at appropriate times. This is because of the
relatively high rates of AVS oxidation due to natural processes in sediments and the
requirement that oxidation must be avoided during sampling of sediments and interstitial water.
In fact it is this seemingly labile nature that has led some to question the practical utility of
using AVS as a basis for EqP-derived ESG for  metals (Luorria and Carter,  1993; Meyer et al..
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 (Besser et al., 1996; Boothman and Helmstetter,1992; Brumbaugh et al.,
1994; Hansen et al., 1996b; Hare et al., 1994;  Howard and Evans, 1993; Leonard et al.,
1996a; Liber et al., 1996 ).  This likely is due to oxidation of AVS at the sediment surface, a
process that is enhanced by biorurbation (Peterson et al., 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 due to a

-------
                                                                                    66
constant rate of oxidation of the AYS linked to a decrease in its generation by sulfate-reducing
bacteria (Herlihy and Mills, 1985; Howard and Evans, 1993; Leonard et al., 1993). Because
of potential temporal and spatial variability of AVS, it appears the way to avoid possible
under-estimation of metal bioavailability is to sample the biologically "active" zone of
sediments at times when AVS might be expected to be present at small concentrations.  We
recommend that at a minimum AVS and SEM measurements be made using surficial (0-2.0
cm) sediments during the period from November to early May in aerobic aquatic ecosystems.
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, seasonal measurements of AVS,
SEM and interstitial metal concentrations may need to be determined.  Importantly,  the
biologically active zones of some benthic communities may be only the surface few millimeters
in depth and in other communities up to a meter. Therefore, for sufficient characterization,
multiple sediment horizons may require sampling of interstitial water, SEM and AVS to
determine the potential for exposure to metals.

       The somewhat subjective aspects of these sampling recommendations have been of
concern. However, recent research suggests that the transient nature of AVS may be over-
stated relative to predicting the fate of all metal-sulfide complexes in aquatic sediments.
Observations from the Duluth EPA laboratory made in the early 1990s indicated 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,
unpublished data). 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
nominally as much as five-times more SEM (zinc) than AVS; however, the highest measured
SEM/AVS ratio achieved was only slightly larger than 1. Moreover, the expected surficial
depletion and seasonal variations in AVS were unexpectedly low  in the zinc-spiked sediments.
These observations suggested  that zinc sulfide, which comprised the bulk of AVS in the spiked
sediments, was more stable than the iron sulfide  that presumably  was the source of most of the
AVS 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

-------
                                                                                     67

(Boothman et al., 1996; De Witt et al., 1996; Hansen et al.,  1996b; Leonard et al., 1995;
Peterson et al., 1996; Sibley et al., 1996).

        In support of these observations,  recent 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 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, both from a
temporal and spatial perspective,  is definitely warranted.

5.2.1  Sediments

       At a minimum, sampling of the surficial 2.0 cm of sediment in between November and
early May is  recommended. A sample depth of 2.0 cm is more appropriate for remediation
and monitoring. In some instances such as for dredging or where depths greater than 2 cm are
important than 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 air-
tight glass jars and  refrigerated or frozen.  Generally, 50 ml or more  of sediment should be
added to nearly 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

-------
                                                                                       68
centrifugation at 11,000 rpm followed by double filtration (0.45 ^m and 0.2 or 0.03 /^m) and
in situ diffusion samplers with a 0.002 /urn interstitial size.  For the metals of concern in this
guidelines document, concentrations of nickel and cadmium were equivalent using both
methods and concentrations of copper and zinc were higher and more variable using
centrifugation. They recommended the use of in situ dialysis for the study of 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 the collection of
interstitial water for trace metals analysis. Thesejncluded ex situ squeezing and^cgngifiigatiQn^
and in situ dialysis and suction filtration.  This paper should be read by those jelecting a
interstitial water sampling method. JThey observed that each method has its own advantages
and disadvantages, and that each user must make their own choice given the inherent errors of
each method. 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 to those in situ. Contamination of interstitial water by fine particles is
important in all methods as differentiation of paniculate and dissolved metal is a  function of
interstitial size. The use of 0.45 //m filtration, while an often accepted definition of dissolved,
may result in laboratory to laboratory discrepancies.  The use of suction filtration devices is
limited to coarser sediments, and they do not offer depth resolution. The use of diffusion
samplers is hampered by the time required for equilibrium (7-14 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 centrifugation 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 0.4 to 0.45 n
polycarbonate filters to better define that fraction of aqueous metal associated with toxicity
(Prothro, 1993).  Thurman equates the organic carbon retained on a 0.45 micrometer glass-
fiber filter to suspended organic carbon so that this filtration procedure under nitrogen

-------
                                                                                     69

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 filtering apparatus has been identified when losses occur (Ozretich and
Schults,  1998).  Ozretich and Schults, 1998 have recently presented a method combining
longer centrifugation times with a unique single-step IW withdrawal procedure which has the
potential for minimizing metal losses by eliminating the need for filtration.

       In contrast to the above  recommendations, EPA recommends the 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. Further,
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. Secondly, EPA recommends the use of centrifugation under nitrogen and double
0.45/^m filtration using polycarbonate filters for obtaining interstitial water from sediments in
deeper aquatic systems.  Probably most  importantly, 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 interstitial water toxic unit
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.,
1996a; 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
IWGTU sediments should be acceptable for protection of benthic organisms.

5.3    ANALYTICAL MEASUREMENTS

       An important aspect to deriving "global"  ESG values is that the methods necessary to
implement the approach must be reasonably  standardized or have been demonstrated to
produce  results that are comparable to those of standard methodologies.  From the standpoint
of the proposed metal ESG, 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.

-------
                                                                                    70
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 et al., 1993), colorimetric (Cornwall and
Morse, 1987), gas chromatography - photoionization detection (Casas and Crecelius, 1994;
Slotton and Reuter,  1995) and specific ion electrodes (Boothman and Helmstetter,1992;
Brouwer and Murphy, 1994; Brumbaugh et al., 1994; Leonard et al., 1996b).  Allen et al.,
1993 report a limit of detection for 50% accuracy of 0.01 umol/g for a 10-g sediment sample
using the colorimetric method. Based on several studies Boothman reports a detection limit of
1 umol AVS which translates to 0.1 umol/g dry weight for a 10 g  sediment sample using the
ion specific electrode method (personal communication).

5.3.2  Simultaneously Extracted Metal

       Simultaneously extracted metals are operationally-defined as metals extracted from
sediment into solution by the acid volatile sulfide extraction procedure.  The "dissolved"
metals in this solution are also operationally defined as the metal species which pass through
filter material used to remove the residual sediment, and thus are defined by the interstitial size
of the filtration material used.  Common convention defines "dissolved" as metal species
<0.45-^m,in size. SEM concentrations measured in sediments are not significantly different,
however, using Whatman 1 filter paper alone (< 11-pun nominal interstitial size) or in
combination with a 0.45-/im filter (W. Boothman, unpublished data). SEM  solutions generated
by the AVS procedure can be analyzed for  metals, commonly including cadmium, copper,
lead, nickel, silver and zinc by routine atomic spectrochemical techniques appropriate for
environmental waters (e.g. inductively coupled plasma atomic emission or graphite furnace
atomic absorption spectrophotometry) (U.S. EPA, 1994b).  Because of the need to determine
metals at relatively low concentrations, additional consideration must be given to preclude
contamination during collection, transport and analysis (U.S. EPA, 1995d,e,f,g).

5.3.3  Interstitial Water Metal

       Interstitial water can be analyzed for the metals cadmium, copper, lead, nickel, silver
and zinc by routine atomic spectrochemical techniques appropriate for environmental waters

-------
                                                                                     71
(e.g. inductively coupled plasma atomic emission or graphite furnace atomic absorption
spectrophotometry) (U.S. EPA, 1994b). Because of the need to determine metals at
concentrations at or below the threshold of  biological effects (i.e., WQC concentrations);
additional consideration must be given to preclude contamination during collection, transport
and analysis (U.S. EPA, 1995d,e,f,g).  (See guidance on clean chemistry techniques in U.S.
EPA, 1994c.) Generally, detection limits should be at sO.l IWGTU, because the toxic unit
contributions of each of the metals must be summed.

5.4    ADDITIONAL BINDING PHASES

       Although AVS is an important binding phase for metals, there clearly are other
physico-chemical factors that influence  metal partitioning in sediments.  In aerobic systems, or
those with low productivity (i.e., where the absence of organic carbon limits sulfate reduction),
AVS plays little or no role in determining interstitial water concentrations of metals.  For
example, Leonard et al. (1996a) found that a relatively large percentage of surficial sediments
from open areas in Lake Michigan did not contain detectable AVS.  In fact the great majority
(42 of 46) of samples analyzed by Leonard et al. (1996a) contained less AVS than SEM, yet
interstitial water metal concentrations of cadmium, copper, nickel, lead and zinc were
consistently small or non-detectable.  Even in sediments where concentrations of AVS are
significant, other partitioning phases may provide additional binding capacity for SEM (e.g.,
Ankley et al., 1993; Calamono et al., 1990; Slotton and Reuter, 1995).  In aerobic sediments
both organic carbon and  iron and manganese oxides control interstitial water concentrations of
metals (Calamono et al,. 1990; Jenne, 1968; Luoma and Bryan, 1981; Tessier et al.,  1979).  In
anaerobic sediments, organic carbon appears to be an important additional binding phase
controlling metal partitioning, in particular for cadmium, copper and lead (U.S. EPA, 1994a).

       Even in substrates with very little metal binding capacity (e.g., chromatographic sand).
surface adsorption associated with cation exchange capacity will control interstitial water metal
concentrations to some degree (Hassan  et al., 1996).  Although an ideal ESG model for metals
would incorporate all possible metal binding phases, current knowledge concerning
partitioning/capacity of phases other than AVS is insufficient for practical application of a
multiple phase model for deriving ESG in this sediment guidelines document.

-------
                                                                                   7.2
5.5    PREDICTION OF THE RISKS OF METALS IN SEDIMENTS BASED ON EqP

       It is important to repeat that conclusions about sediment toxicity based on SEM-AVS
concentrations pertain only to cadmium, copper, lead, nickel, silver and zinc.  (1) When the
molar concentration of AVS exceeds that of SEM (negative SEM-AVS) sediment toxicity due
to these metals is unlikely and any observed toxicity is most likely from some other cause.  This
is important because toxicity observed in sediments having an excess of AVS is often
incorrectly assumed to disprove the EqP metals theory. The correct conclusion is that some
factor other than metals caused the effect. This can be further substantiated if the toxic unit
concept is applied to metal concentrations measured in interstitial water; the absence of
significant concentrations of metals coupled with the  negative SEM-AVS are powerful
evidence that  metals are an unlikely cause of the effect.  (2) Sediments can only be toxic from
the metals cadmium, copper, lead, nickel, silver and  zinc when the molar concentrations of
SEM exceed those of AVS (SEM-AVS differences are positive).  Measurements of interstitial
water concentrations of metals are invaluable in demonstrating that the sediments are toxic
because of metals, and these measurements will provide insights into the specific metal(s)
causing the observed toxicity.  (3) It is not uncommon for toxicity to be absent in sediments
having concentrations of SEM that exceed those of AVS (SEM-AVS  is positive).  This is
because other metal binding phases in  sediments often reduce the concentrations of bioavailable
metal.  (4) When sediments are toxic,  and SEM-AVS is greater than 0.0, the toxicity may or
may not be metals-related.  Often sediments having SEM-AVS of up to 10 ^moles SEM/g  are
not toxic because the excess metals are associated with other binding phases. Measurements of
interstitial water concentrations of metals are invaluable in demonstrating an absence or
presence of bioavailable metal.

-------
                                                                                   73
                                     SECTION 6

                             GUIDELINES STATEMENT

       The procedures described in this document indicate that, except possibly where a
locally important species is very sensitive, benthic organisms should be acceptably protected in
freshwater and saltwater sediments if any one or both of the following two conditions is
satisfied: (a) If 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 or (b) the sum of the
dissolved interstitial water concentration of cadmium, copper, lead, nickel, silver and zinc
divided by their respective WQC is  less than or equal to 1.0.

(a)     AVS Guidelines:

            ^ [SEMjMAVS]                                                      (4-5)

where

I JSEMj =  [SEMcJ + [SEMCd] + [SEMpJ +  [SEMNj] + [SEMZJ  + [l/2SEMAg]

(b)     Interstitial Water Guidelines

               [M,J    .
                                                                                .(4-8)
              !•' *•" v i.dJ

where



       [M il       IH H!    [Mrill     [Mi* il     [MVH]    [M7  J     [M '  .]
       1  i.dj       l  Cu.d1    l  Cd.J'     l  Pb.d'     i   Ni.d'    l  Zn.dJ     l  Au.dJ
      [FCVJ     [FCVCud][FCVCdd]  [FCVpbd]   [FCVNJ  [FCV2ndl  [FCVAgd]

If any one of these two conditions are violated, this does not mean that the sediment violates the
ESG and is unacceptable.  For example, if SEM exceeds AVS, or if the AVS in a sediment is non-
detectable, then condition (a) will be violated.  However, if there is sufficient sorption to particles,

-------
                                                                                          74
or organic carbon or other binding phases so that condition (b) is satisfied, then the sediment meets
the guideline and benthic organisms are acceptably protected from metals-induced sediment
toxicity.

       If both of these conditions are violated, or if the AVS Guideline is violated and the
sediment is contaminated with silver then there is reason to believe that the sediment may be
unacceptably contaminated by these metals. Further testing and evaluations would, therefore, be
useful in order to assess actual toxicity and its causal relationship to the five metals.  These may
include acute and chronic tests with species that are sensitive to the metals suspected to be causing
the toxicity.  Also, in situ community assessments, sediment TIEs and seasonal characterizations
of the SEM, AVS and interstitial water concentrations would be appropriate (Ankley et al., 1994).

       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, for example, with sediment suspension, or the
transport of metals into the food web either from sediment ingestion or the ingestion of
contaminated benthos. This latter issue, in particular, should be the focus of future research given
existing uncertainty in the prediction of bioaccumulation of metals by benthos (Ankley, 1996).

       It is repeated here that these guidelines apply only  to the six metals discussed in this
document, copper cadmium, lead, nickel,  zinc and silver.  Procedures for sampling and analytical
methods for interstitial water and sediments are discussed  in Section 5, Implementation.

-------
                                                                                    75

                                      SECTION 7

                                    REFERENCES

Adams, D.A., O'Connor, J.S. and Weisberg.S.B. 1996. Draft Final Report Sediment Quality of
       the NY/NJ Harbor System, An Investigation under the Regional Environmental Monitoring
       and Assessment Program (R-EMAP). Monitoring and Assessment Branch, Division of
       Environmental Science and Assessment. United States Environmental Protection Agency,
       Region 2. Edison, New Jersey.

Adams, W.J., R.A. Kimerle and R.G. Mosher.  1985. Aquatic safety assessment of chemicals
       sorbed to sediments.  In Aquatic Toxicology and Hazard Assessment: Seventh Symposium.
       R.D. Cardwell, R. Purdy and R,C. Banner, Eds. STP 854.  American Society for Testing
       and Materials, Philadelphia, PA, USA, pp. 429-453.

Allen,  H.E.,  R.H. Hall and T.D. Brisbin.  1980. Metal speciation effects on aquatic toxicity.
       Environ. Sci. Technol. 14:441-443.

Allen,  H.E., G. Fu and B. Deng.  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, D.M. and F.M.M. Morel. 1978.  Copper sensitivity of Gonyaulax. lamarensis.
       Limnol. Oceanogr. 23:283-295.

Ankley, G.T.. G.L. Phipps. E.N. Leonard, D.A. Benoit.  V.R. Mattson. P.A. Kosian, A.M.
       Cotter, J.R. Dierkes, D.J.  Hansen and J.D. Mahony.  1991. Acid volatile sulfide  as a
       factor mediating cadmium  and nickel bioavailability in contaminated sediments.  Environ.
       Toxicol. Chem. 10:1299-1307.

Ankley, G.T., V.R. Mattson, E.N. Leonard, C.W. West and J.L. Bennett.  1993.  Predicting the
       acute toxicity of copper in  freshwater sediments: Evaluation of the role of acid-volatile
       sulfide. Environ. Toxicol. Chem.  12:315-320.

-------
                                                                                       76
Ankley, G.T., D.M. Di Toro, D.J. Hansen, J.C. Mahony, R.C. Swartz, R.A. Hoke, N.A.
       Thomas, A.W. Garrison, H.E. Allen and C.S. Zarba. 1994. Assessing the potential
       bioavailability of metals in sediments: A proposed approach. Environ. Manage. 18:331-
       337.

Ankley, G.T.  1996. Evaluation of metal/acid volatile sulfide relationships in the prediction of
       metal bioaccumulation by benthic macroinvertebrates.  Environ. Toxicol. Chem. 15:2138-
       2146.

Ankley, G. T., D.M. Di Toro, D.J. Hansen and W.J. Berry. 1996. Technical basis and proposal
       for deriving sediment quality criteria for metals. Environ. Toxicol. Chem. 15:2056-2066.

Barrow, N.J. and A.S.  Ellis.  1986a.  Testing a mechanistic model. III.  The effects of pH on
       fluoride retention by a soil.  J. of Soil Science 37:287-295.

Barrow, N.J. and A.S.Ellis. 1986b.  Testing a mechanistic model.  IV. Describing the effects
       of pH on zinc retention by a soil. J. of Soil Science 37:295-302.

Barrow, N.J. and A.S.  Ellis.  1986c.  Testing a mechanistic model. V.  The points of zero salt
       effect for phosphate retention and for acid/alkali titration of a soil. J. of Soil Science
       37:303-310.

Bergman, H.L. and  E.J. Dorward-King. 1997. Reassessment of Metals Criteria for Aquatic Life
       Protection. Society of Environmental Toxicology and Chemistry (SETAC) Press:
       Pensacola, FL.

Berner, R.A.  1967.  Thermodynamic stability of sedimentary iron sulfides.  Am. J. Sci.  265:773-
       785.

Berry, W.J., D.J. Hansen, J.D. Mahony, D.L. Robson, D.M. Di  Toro,  B.P. Shipley, B. Rogers,
       J.M. Corbin and W.S. Boothman. 1996a. Predicting the toxicity of metals-spiked
       laboratory sediments using acid-volatile sulfide and interstitial  water normalization.
       Environ. Toxicol. Chem.  15:2067-2079.

-------
                                                                                      77

Berry, W., M. Cantwell, P. Edwards, J. Serbst and D. Hansen. 1996b. Abstract: Predicting the
       toxicity of silver-spiked sediments using interstitial water metal and acid volatile sulfide
       normalizations. Society of Environmental Toxicology and Chemistry. Annual Meeting.
       Vancouver, B.C., Canada, p.89-90.

Besser, J.M., C.G..Ingersoll and J.P. Giesy.  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.

Biesinger, K.E., G.M. Christensen and  J.T. Fiandt. 1986.  Effects of metal salt mixtures on
       Daphnia magna reproduction.  Ecotoxicol. Environ. Safety 11:9-14.

Boothman, W.S. and A. Helmstetter.  1992.  Vertical and seasonal variability of acid volatile
       sulfides in marine sediments. Research Report to the U.S. EPA Environmental Monitoring
       and Assessment Program, Narragansett, RI, USA.

Boothman, W., D. Hansen, W. Berry, D. Robson, A. Helmstetter and J. Corbin. 1996. Abstract:
       Variation of acid volatile sulfides and metals and biological response in field-exposed
       sediments. Society of Environmental Toxicology and Chemistry Annual Meeting.
       Washington, D.C.

Boulegue, J., C.J. Lord III and T.M. Church. 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. and T.P. Murphy.  1994.  Diffusion method for the determination of acid-volatile
       sulfides (AVS) in sediment. Environ. Toxicol. Chem. 13:1273-1275.

Brumbaugh, W.G., C.G. Ingersoll, N.E. Kemble, T.W. May and J.L. Zajcek.  1994. Chemical
       characterization of sediments and interstitial water from the upper Clark Fork River and
       Milltown Reservoir, Montana.  Environ.  Toxicol. Chem.  13:1971-1983.

-------
                                                                                     78
Buffiap, W.E. and H.E. Allen.  1995.  Sediment interstitial water collection methods: A review.
       Water Res. 29: 65-177. Calder, E.R. Long and C.G. Ingersoll. 1996.  Development and
       evaluation of sediment quality guidelines for Florida coastal waters. Ecotoxicology, In
       Press.

Calamono, W., W. Ahlf and U. Forstner.  1990. Exchange of heavy metals between sediment
       components and water. In Metal Speciation in the Environment. J.A.C. Broekart, S.
       Gucer and F. Adams, Eds. Springer-Verlag, Berlin, Germany, pp. 503-522.

Carignan, R. 1984. Interstitial water sampling by dialysis: Methodological notes. Limnol.
       Oceanogr. 20:667-670.

Carignan, R., F. Rapin and A. Tessier. 1985. Sediment interstitialwater sampling for metal
       analysis: a comparison of techniques. Geochemica et Cosmochem. Acta 49:2493-2497.

Carlson, A.R., G.L. Phipps, V.R. Mattson, P.A. Kosian and A.M. Cotter.  1991.  The role of
       acid volatile sulfide in determining cadmium bioavailability in sediments.  Environ.
       Toxicol. Chem. 10:1309-1319.

Casas, A.M. and E.A. Crecelius. 1994. Relationship between acid  volatile sulfide and the
       toxicity of zinc, lead and copper in ma rine sediments.  Environ. Toxicol. Chem. 13:529-
       436.

Chapman, G.A. 1987. Establishing sediment criteria for chemicals-regulatory perspective.  In:
       Fate and Effects of Sediment-Bound Chemicals in Aquatic Systems. Editors: K.L.
       Dickson, A.W. Maki and W.A. Brungs. Pergamon Press, New York.  pp. 355-376.

Cornwell, J.C. and J.W.  Morse. 1987. The characterization of iron sulfide minerals in anoxic
       marine sediments.  Mar. Chem. 22:193-206.

Davies-Colley, R.J., P.O. Nelson and K.J. Williamson.  1985. Sulfide control of cadmium and
       copper concentrations  in anaerobic estuarine sediments. Mar. Chem. 16:173-186.

-------
                                                                                    79
De Witt, T.H., R.C. Swartz, DJ. Hansen, W. J. Berry and D. McGovern. 1996. Interstitial
       metal and acid volatile sulfide predict the bioavailability of cadmium during a full life cycle
       sediment toxicity test using the estuarine amphipod, Leptocheirus plumulosus.  Environ.
       Toxicol. Chem.  15:2095-2101.

Di Toro, D.M., F. Harrison, E. Jenne,  S. Karickoff, and W. Lick. 1987. Synopsis of Discussion
       sesson 2: Environmental fate and compartmentalization. In K. L. Dickson, A.W. Maki
       andW.A. Brungs, eds., Fate and Effects of Sediment-Bound Chemicals in Aquatic
       Systems. Pergamon, New York,  NY, USA, pp. 136-147.

Di Toro, D.M., JJ. Mahony, D.J. Hansen, K.J. Scott, M.B. Hicks, S.M. Mayr and M.S.
       Redmond.  1990.  Toxicity of cadmium in sediments:  The role of acid volatile sulfide.
       Environ. Toxicol. Chem. 9:1487-1502.

Di Toro, D.M., C. Zarba, D.J. Hansen, R.C. Swartz, C.E.  Cowan, H.E. Allen, N.A. Thomas,
       P.R. Paquin and W.J. Berry. 1991.  Technical basis for establishing sediment quality
       criteria for non-ionic organic chemicals using equilibrium partitioning. Environ. Toxicol.
       Chem.  10:1541-1583.

Di Toro, D.M., 1992. Personal communication relating to Di Toro et al.,  1990 on titration of
       FeSbyCd2*.

Di Toro, D.M., J.D. Mahony, D.J. Hansen, K.J. Scott, A.R. Carlson and  G.T. Ankley. 1992.
       Acid volatile sulfide predicts the acute toxicity of cadmium and nickel in sediments.
       Environ. Sci. Techno!. 26:96-101.

Di Toro, D.M., J.D. Mahony, A.M. Gonzalez. 1996a Particle oxidation model of synthetic FeS
       and sediment acid volatile sulfide.  Environ. Toxicol. Chem. 15:2156-2167.

Di Toro, D.M., J.D. Mahony, D.J. Hansen and W.J. Berry. I996b.  A model  of the oxidation of
       iron and cadmium sulfide in sediments.  Environ.  Toxicol. Chem. 15:2168-2186.

Dzombak,  D.A. and F.M.M. Morel. 1990.  Surface Complexation Modeling, Hydrous Ferric
       Oxide.  New York, NY, John Wiley & Sons.

-------
                                                                                      80
Emerson, S., L. Jacobs and B. Tebo. 1983. The behavior of trace metals in marine anoxic
       waters: solubilities at the oxygen-hydrogen sulfide interface. In Trace Metals in Sea
       Water. C. S. Wong.

E. Boyle, K.W. Bruland and J.D. Burton, Eds. Plenum Press, NY, USA, pp. 579-608.

Enserink, E.L., J.L. Maas-Diepeveen and C.S. van Leeuwen.  1991.  Combined toxicity of
       metals: An ecotoxicological evaluation.  Water Res. 25:679-687.

Goldhaber, M.B. and I.R. Kaplan. 1974.  The sulfur cycle.  In The Sea,  Vol. 5, Marine
       Chemistry. E.D. Goldberg, Ed.  J. Wiley & Sons, NY., USA,  pp. 569-655.

Green, A.S., G.T. Chandler and E.R. Blood.  1993. Aqueous-, interstitial-water-, and sediment-
       phase cadmium: Toxicity relationships for a meiobenthic copepod. Environ. Toxicol.
       Chem. 12:1497-1506.

Hansen, D.J. 1995. Memorandum to Catherine Fox. Subject: Assessment tools that can be used
       for the National Sediment Inventory. February 28,  1994.

Hansen, D.J., W.J. Berry, J.D. Mahony, W.S. Boothman, D.L. Robson, G. T. Ankley, D. Ma,
       Q Yan and C.E. Pesch. 1996a.  Predicting toxicity of metal-contaminated field sediments
       using interstitial water concentrations of metals and acid-volatile sulfide normalization.
       Environ. Toxicol. Chem. Environ. 15:2080-2094.

Hansen, D.J., J.D. Mahony, W.J. Berry, S.J. Benyi, J.M. Corbin, S.D.  Pratt and M.B. Able.
       1996b. 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., R. Carignan and M.A. Huerta-Diaz.   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.

-------
                                                                                      81
Hassan, S.M., A.W. Garrison, H.E. Allen, G.T. Ankley and D.M. Di Toro.  1996.  Using the
       distribution coefficients of five trace metals in sandy sediments to select the lowest capacity
       natural adsorbent. Environ. Toxicol. Chem.  15:2198-2208.

Herlihy, A.T. and A.L. Mills. 1985. Sulfate reduction in freshwater sediments receiving acid
       mine drainage. Appl. Environ. Microbiol. 49:179-186.

Hesslein, R.H. 1976. An in situ sampler for close interval interstitial water studies. Limnol.
       Oceanogr. 30:221-227.

Hoke, R.A.,  G.T. Ankley, A.M. Cotter, T. Goldenstein, P.A. Kosian, G.L. Phipps and P.M.
       VanderMeiden. 1994. Evaluation of equilibrium partitioning theory for protecting acute
       toxicity of field-collected sediments contaminated with DDT, DDE, and DDD to the
       amphipod Hyalella azteca. Environ. Toxicol. Chem. 13:157-166.

Howard, D.E. and R.D. Evans. 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, C.G., P.S. Haverland, E.L. Brunson, T.S. Canfield, F.S. Dwyer, C.E. Henke and
       N.E. Kemble.  1996. Calculation of sediment effect concentrations. J. Great Lakes Res.
       In Press.

Jenne, E.A. 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, USA. pp. 337-387.

Jenne, E.A.  1977.  Trace element sorption by sediments and soil - sites and processes.  In:
       Symposium on Molybdenum in the Environment. Vol.2,  pp. 425-553. Editors: W.
       Chappell and K. Petersen.  M. Dekker, Inc., New York.

Jenne, E.A.,  D.M. Di Toro, H.E. Allen and  C.S. Zarba. 1986.  An activity-based model  for
       developing sediment criteria for metals:  A new approach.  Chemicals in the environment.
       International Conference, Lisbon, Portugal.

-------
                                                                                      82
Kemp, P.P. and R.C. Swartz.  1986. Acute toxicity of interstitial and particle-bound cadmium to
       a marine infaunal amphipod. Mar. Environ. Res. 26:135-153.

Kraak, M.H.S., D. Lavy,H. Schoon, M. Toussaint, W.H.M. Peelers and N.M. vanStraalen.
       1994.  Ecotoxicity of mixtures of metals to the zebra mussel Dreissena potymorpha.
       Environ. Toxicol. Chem. 13:109-114.

Landers, A.H., M.B. David and M.J. Mitchel,  1983.  Analysis of organic and inorganic sulfur
       constituents in sediments, soils, and water. Int. J. Environ. Anal. Chem. 14:245-256.

Leonard, E.N., V.R. Mattson, D.A. Benoit, R.A. Hoke and G.T. Ankley.  1993.  Seasonal
       variation of acid volatile sulfide concentration in sediment cores from three northeastern
       Minnesota lakes.  Hydrobiologia  271:87-95.

Leonard, E.N., V.R. Mattson and G.T. Ankley.  1995. Horizon-specific oxidation of acid
       volatile sulfide in relation to the toxicity of cadmium spiked into a freshwater sediment.
       Arch. Environ. Contain. Toxicol.  28:78-84.

Leonard, E.N., G.T. Ankley and R.A. Hoke.  1996a.  Evaluation, of metals in marine and
       freshwater surficial sediments from the Environmental Monitoring and Assessment
       Program relative to proposed sediment quality criteria for metals.  Environ. Toxicol.
       Chem.  15:2221-2232.

Leonard, E.N., A.M. Cotter and G.T. Ankley.  1996b. Modified diffusion method for analysis of
       acid volatile sulfides  and simultaneously extracted metals in freshwater sediment.  Environ.
       Toxicol. Chem.  In Press.

Liber, K., D. Call. T.' Markee, K. Schmude, M. Balcer, F. Whiteman and G.T. Ankley.  1996.
       Effects of acid volatile sulfide on zinc bioavailability and toxicity to benthic
       macroinvertebrates in spiked-sediment field experiments.  Environ. Toxicol. Chem.
       15:2113-2125.

-------
                                                                                    83
Long, E.R. and L.G. Morgan. 1991.  The potential for biological effects of sediment-sorbed
       contaminants tested in the national status and trends program.  NOAA Technical
       Memorandum NOS OMA 52.

Long, E.R., D.A. Wolfe, K.J. Scott, G.B. Thursby, E.A. Stem, C. Peven, T. Schwartz. 1995.
       Magnitude and Extent of  Sediment Toxicity in the Hudson-Raritan Estuary. National
       Oceanic and Atmospheric Administration Technical Memorandum NOS ORCA 88
       NOAA/NOS Office of Ocean Resources, Conservation, and Assessment, Silver Spring,
       MD USA.

Long, E.R., G.M. Sloane, R.S. Carr, K.J. Scott, G.B. Thursby, T.L. Wade. 1996.  Sedment
       Toxicity in Boston Harbor: Magnitude and Extent, and Relationships with Chemical
       Toxicants. National Oceanic and Atmospheric Administration Technical Memorandum
       NOS ORCA 88 NOAA/NOS Office of Ocean Resources, Conservation, and Assessment,
       Silver Spring, MD USA. REMAP

Luoma, S.N. 1989.  Can we determine the biological availability of sediment-bound trace
       elements?  Hydrobiologia 176/177:379-396.

Luoma, S.N. and W. Bryan.  1981.  A statistical assessment of the form of trace metals in
       oxidized sediments employing chemical extractants.  Sci. Total Environ. 17:165-196.

Luoma, S.N. and J.L. Carter. 1993. Understanding the toxicity of contaminants in sediments:
       Beyond the bioassay-based paradigm. Environ Toxicol. Chem. 12:793-796.

MacDonald, D.D.. R.S. Carr. F.D. Calder,  E.R.  Long and C.G. Ingersol.  1996. Development
       and evaluation of sediment quality guidelines for Florida coastal  waters. Ecotoxicology
       5:253-278.

Mahony, J.D.. D.M. Di Toro, R. Koch. W. Berry and D. Hansen.  1993.  Vertical distribution of
       AVS and SEM in bedded  sediments, biological implications and  the role of metal sulfide
       oxidation kinetics.  Abstract.  14th Annual Meeting.  Soc. Environ. Toxicol. Chem.  p.46.

-------
                                                                                      84
Mahony, J.D., D.M. Di Toro, A.M. Gonzalez, M. Curto and M. Dilg.  1996.  Partitioning of
       metals to sediment organic carbon.  Environ. Toxicol. Chem.  15:2187-2197.

Meyer, J.S., W. Davison, B. Sundby, J.T. Oris, D.J. Lauren, U. Forstner, J. Hong and D.G.
       Crosby.  1994. The effects of variable redox potentials, pH and light on bioavailability in
       dynamic water-sediment environments. In Bioavailability: Physi-cal, Chemical and
       Biological Interactions.

J.L. Hamelink,  P.P. Landrum, H.L. Bergman and W.H. Benson, Eds.  CRC Press, Boca Raton,
       FL, USA,  pp. 155-170.

Morse, J.W., F.J. Millero, J.C. Cornwell and D. Rickard.  1987. The chemistry of the hydrogen
       sulfide and iron sulfide systems in natural waters. Earth Sci. Rev. 24:1-42.

Oakley, S.M., K.J. Williamson and  P.O. Nelson.  1980.  The geochemical partitioning and
       bioavailability of trace metals in marine sediments. Water Res. Inst., Oregon State Univ.,
       Corvallis, Oregon.  1-84.

Ozretich, R.J. and D.W. Schults.  1998. A comparison of interstitial water isolation methods
       demonstrates centrifugation with aspiration yields reduced losses of organic constituents.
       Chemosphere 36:603-615.

Pankow, J.F. 1979.  The dissolution rates and mechanisms of tetragonal ferrous sulfide
       (Mackinawite) in anoxic aqueous systems. Ph.D thesis.  California Inst.  of Technology,
       Pasadena, Calif. 1-146.

Persaud, D., R. Jacqumagi and A. Hayton.   1989.  Development of provincial sediment quality
       guidelines. Ontario Ministry of the Environment. Toronto, ON, Canada.

Pesch, C.E., D.J. Hansen, W.S. Boothman,  W.J.  Berry and J.D. Mahony. 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. 14:129-14.

-------
                                                                                       85
Peterson, G.S., G.T. Ankley and E.N. Leonard. 1996.  Effect of bioturbation on metal-sulfide
       oxidation in surficial freshwater sediments.  Environ. Toxicol. Chem.  15:2147-2155.

Phillips, H.O. and K.A. Kraus. 1965.  Adsorption on inorganic materials VI. Reaction of
       insoluble sulfides with metal ions in aqueous media.  J. Chromatog 17:549-557.

Prothro, M.G.  1993. Office of Water Policy and Technical Guidance on Interpretation and
       Implementation of Aquatic Life Metals Criteria. Letter from M.G. Prothro, Acting
       Assistant Administrator for the Office of Water, to Water Management Division Directors,
       U.S. EPA Regions I - X. October 1, 1993.

Schoonen, M.A.A. and Barnes, H.L. 1988. An approximation of the second dissociation
       constant for H2S.  Geochim.  Cosmochim-Acta 52:649-654.

Sibley, P.K., G.T. Ankley, A.M. Cotter and E.N. Leonard.  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. Toxicol. Chem.  15:2102-
       2112.

Slotton, D.G. and I.E. Reuter.  1995.  Heavy metals in intact and resuspended sediments of a
       California reservoir, with emphasis on potential bioavailability of copper and zinc.  Mar.
       Freshwater Res. 46:257-265.

Spehar, R.L. and J.T. Fiandt.  1986. Acute and chronic effects of water quality criteria-based
       metal mixtures on three aquatic species.  Environ. Toxicol. Chem.  5:917-931.

Sposito, G,  J.C.M. dc Wit and R.H. Neal.  1988. Selenite adsorption on alluvial soils:  III.
       Chemical modeling. Soil Sci. Soc. Am. J.  52:947-950.

Stephan, C.E., D.I. Mount, D.J. Hansen, J.H.  Gentile, G.A. Chapman and W.A. Brungs. 1985.
       Guidelines for deriving numerical national water quality criteria for the protection of
       aquatic organisms and their uses. PB85-227049. National Technical Information Service,
       Springfield, VA.  98pp.

-------
                                                                                     86
Strobel, C.J., H.W. Buffum, S.J. Benyi, E.A. Petrocelli, D.R. Reifstick and D.J. Keith. 1995.
       Statistical summary EMAP-Estuaries Virginian Province 1992-1993. U.S. EPA/620/R-94/-
       26.  U.S. Environmental Protection Agency, Narragansett, RI. 72 pp.

Stumm, W. and J.S. Morgan.  1996.  Aquatic Chemistry, Chemical Equilibria and Rates in
       Natural Waters, Third Edition. John Wiley & Sons, Inc. New York.

Stumm, W. and J.S. Morgan.  1981.  Aquatic Chemistry -  an Introduction Emphasizing Chemical
       Equilibria in Natural  Waters.  John Wiley & Sons, New York, NY, USA.

Stumm, W.  1987.  Aquatic  Surface Chemistry, Chemical Processes at the Particle-Water
       Interface. John Wiley & Sons, New York.

Sullivan, J., J. Ball, E. Brick, S Hausmann, G. Philarski and D. Sopcich. 1985. Report of the
       technical subcommittee on determination of dredge material suitability for in-water
       disposal. Wisconsin Department of Natural Resources, Madison, WI, USA.

Sunda, W.G. and R.R.L. Guillard. 1976. The relationship between cupric ion activity and the
       toxicity of copper to phytoplankton. J. Mar. Res. 34:511-529.

Sunda, W. G. and J.M. Lewis.  1978. Effect of complexation by natural organic ligands on the
       toxicity of copper to a unicellular alga, Monochrysis lutheri. Limnol. Oceanogr. 23:870-
       876.

Sunda, W.G., D.W. Engel and R.M.  Thuotte.  1978. Effect of chemical speciation of toxicity of
       cadmium to  grass shrimp, Palaemonetes pugio: importance to free cadmium ion. Environ.
       Sci. Techno!. 12:409-413.

Swartz, R.C., G.R. Dittsworth, D.W. Schults and J.O.  Lamberson.  1985.  Sediment toxicity to a
       marine infaunal amphipod: cadmium and its interaction with sewage sludge. Mar.
       Environ. Res.  18:133-153.

-------
                                                                                      87

Swartz, R.C., D.W. Schultz, T.H. DeWitt, G.R. Dittsworth and J.O. Lamberson. 1990. Toxicity
       of fluoranthene in sediment to amphipods: A test of the equilibrium partitioning approach
       to sediment quality criteria. Environ. Toxicol. Chem. 9:1071-1080.

Swartz, R.C., F.A. Cole, J. O. Lamberson, S.P. Ferraro, D.W. Schults, W.A. DeBen, Henry
       Lee II and R.J. Ozretich. 1994. Sediment toxicity, contamination, and amphipod abundance
       at a DDT- and dieldrin-contaminated site in San Francisco Bay. Environ. Toxicol. Chem.
       13:949-962.

Tessier, A., P.G.C. Campbell and M. Bisson.  1979.  Sequential extraction procedure for the
       speciation of paniculate trace metals. Anal. Chem. 51:844-851.

Tessier, A. and P.G.C. Campbell.  1987.  Partitioning of trace metals in sediments: Relationships
       with bioavailability.  Hydrobiologia 149:43-52.

Tessier, A., Y.  Couillard, P.G.C. Campbell and J.C. Auclair.  1993.  Modeling Cd partitioning  in
       oxic lake sediments and Cd concentrations  in the freshwater bivalve Andonata grandis.
       Limnol.  Oceanogr.  38: 1-17.

Thurman, E.M.1985. Organic Geochemistry of Natural Waters. Kluwer Academic Publishers
       Group, Martinus Nijhoff/Dr W. Junk Publishers, Boston, MA. Pp. 497.

Troup,' B.N. 1974. The interaction of iron with phosphate, carbonate, and sulfide in Chesapeake
       Bay interstitial waters: A thermodynamin interpretation. Ph.D thesis. Johns Hopkins
       University, Baltimore, Maryland. 1-114.

U.S. Environmental Protection Agency.  1980.  Ambient Water Quality Criteria for Silver  - 1980.
       Office of Water Regulations and Standards. EPA-440/5-80-071.

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". Uly  19, 1985. Fed.  Regist. 50:30793-30796.

-------
U.S. Environmental Protection Agency.  19855.  Ambient Water Quality Criteria for Cadmium -
       1984.  Office of Water Regulations and Standards. EPA 440/5-85-032.

U.S. Environmental Protection Agency.  1985c.  Ambient Water Quality Criteria for Copper -
       1984.  Office of Water Regulations and Standards. EPA 440/5-85-031.

U.S. Environmental Protection Agency.  1985d.  Ambient Water Quality Criteria for Lead -1984.
       Office of Water Regulations and Standards. EPA 440/5-85-027.

U.S. Environmental Protection Agency.  1986.  Ambient Water Quality Criteria for Nickel -
       1986.  Office of Water Regulations and Standards. EPA 440/5-85-004.

U.S. Environmental Protection Agency.  1987.  Ambient Water Quality Criteria for Zinc - 1987.
       Office of Water Regulations and Standards. EPA 440/5-85-032.

U.S. Environmental Protection Agency.  1989.  Sediment classification methods compendium.
       Watershed Protection Division, U.S. EPA. 280pp.

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. Briefing Report to the Science Advisory Board.  EPA 822-D-94-002. U. S.
       Environmental Protection Agency. Office of Water, Office of Research and Development,
       Washington, DC, USA.

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,
       USA.

-------
                                                                                     89
U.S. Environmental Protection Agency. 1995a. An SAB report: Review of the agencies approach
       for developing sediment criteria for five metals.  EPA-SAB-EPEC-95-020. Office of
       Water, Washington, DC, USA.

U.S. Environmental Protection Agency. 1995b. Water quality criteria for dissolved metals.
       Federal Register Notice, 60(86):22231-22237.

U.S. Environmental Protection Agency. 1995c. Ambient water quality criteria- saltwater copper
       addendum. April 14, 1995. U.S. Environmental Protection Agency, Office of Water,
       Office of Science and Technology, Washington, D.C.

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 20460.

U.S. Environmental Protection Agency. 1995e. 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 20460.

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
       20460.

U.S. Environmental Protection Agency. 1995g. Method 1636: Determination of hexavalent
       chromium by ion chromatography. EPA-821-R-95-029. Office of Water, Engineering and
       Analysis Division (4303). Washington. DC 20460.

U.S. Environmental Protection Agency.  1996. EMAP-Estuaries Virginian Province Data 1990-
       1993.  Available from: EMAP Home Page WWW site, http://www.epa.gov/emap via the
       INTERNET.  Accessed 1997 Sept 17.

U.S. Environmental Protection Agency.  1997a. The Incidence & Severity of Sediment
       Contamination in Surface Waters of the United States. Vol 1: National Sediment Quality
       Survey (EPA 823-R-97-006), Washington, D.C.

-------
                                                                                      90
U.S. Environmental Protection Agency. 1997b.  The Incidence & Severity of Sediment
       Contamination in Surface Waters of the United States. Vol 2: Data Summaries for Areas
       of Probable Concern (EPA 823-R-97-007), Washington, D.C.

U.S. Environmental Protection Agency. 1997c.  The Incidence & Severity of Sediment
       Contamination in Surface Waters of the United States. Vol 3: National Sediment
       Contaminant Point Source Inventory (EPA 823-R-97-008), Washington, D.C.

U.S. Environmental Protection Agency. 1998a. Technical basis for establishing sediment quality
       criteria for non-ionic organic chemicals by using equilibrium partitioning. (In preparation)

U.S. Environmental Protection Agency. 1998b. Users guide for multi-program implementation of
       sediment quality criteria.  (In preparation)

U.S. Environmental Protection Agency. 1998c. Guidelines for deriving site-specific sediment
       quality criteria for the protection of benthic organisms.  (In preparation)

Wolfe, D.A., S.B. Bricker, E.R.  Long, K.J. Scott and G.B. Thursby.  1994.  Biological effects of
       toxic contaminants in sediments from Long Island Sound and environs.  National Oceanic
       and Atmospheric Administration Technical Memorandum NOS ORCA 80 NOAA/NOS
       Office of Ocean Resources Conservation and Assessment, Silver Spring, MD USA.

Zamuda, C.D. and W.G. Sunda.  1982. Bioavailability of dissolved copper to the American
       oyster Crassostrea virginica: Importance of chemical speciation.  Mar. Biol. 66:77-82.

-------
APPENDIX A

-------
                                    APPENDIX A
                        Glossary of Abbreviations and Equations

ACR        Acute-Chronic Ratio
Ag          Silver
AVS        Acid Volatile Sulfide
ASTM       American Society for Testing and Materials
Cd          Cadmium
Cd           Freely dissolved interstitial water concentration of contaminant
Cp           Total interstitial water concentration of contaminant
Cs           Concentration of contaminant in sediment
CSoc          Concentration of contaminant in sediment on an organic carbon basis
CCC        Criteria Continuous Concentration
CFR        Code of Federal Regulations
CLOGP      Computer program for generating partition coefficients
CMC        Criteria Maximum Concentration
CV          Coefficient of Variation                 :
CWA        Clean Water Act
DOC        Dissolved Organic Carbon
EDTA       Ethlyene  diamine tetraacetic acid
EMAP       Environmental Monitoring and Assessment Program
ESG         Equilibrium Partitioning Sediment Guidelines
ESGOC       Organic carbon-normalized Equilibrium Partitioning Sediment Guidelines
foe          Fraction of organic carbon in sediment
EqP   '       Equilibrium partitioning
FAV        Final Acute Value
FCV        Final Chronic Value
{Fe2+}       activity of Fe:" (mol/L)
[Fe2+]        concentration of Fe2" (mol/L)
[FeS(s)]      concentration of iron sulfide (mol/L)
[FeS(s)]j      initial iron sulfide concentration in the sediment (mol/L)
FeS          Iron monosulfide
FPV         Final Plant Value
FRV        Final Residue Value
GC/EC      Gas Chromatography/Electron Capture
GC/MS      Gas Chromatography/Mass Spectrometry

-------
KP
KSP
LC50

L
[M]A
[MS(s>]
m3  or cu m
GFAA        Graphite Furnace Atomic Absorption
IWGTU      Interstitial Water Guidelines Toxic Unit
IWTU        Interstitial Water Toxic Unit
KFeS          solubility product for FeS(s) [(mol/L)2]
KMS          solubility product for MS(s) [(mol/L)2]
              Organic carbon: water partition coefficient
              Octanol: water partition coefficient
              Sediment: water partition coefficient
              Solubility product constant
              Concentration estimated to be lethal to 50 percent of the test organisms within a
              specified time period.
              Liter
              divalent metal activity (mol/L)
              concentration of M2+ (mol/L)
              concentraton of added metal (mol/L)
              concentration of solid-phase metal sulfide (mol/L)
              Cubic meter
              Microgram
yum           Micrometer
yumole        Micromole
mg           Milligram
mg/1          Milligram per liter
ml            Milliliter
mm          Millimeter
NA           Not Applicable, Not Available
ND           Not Determined, Not Detected
ng            Nanogram
Ni            Nickel
NOAA  .     National Oceanographic and Atmospheric Administration
NOEC        No Observed Effect Concentration
NST          National Status and Trends monitoring program
NTA         Nitrilotriacetic acid
Pb            Lead
pH           Negative logarithm of the effective hydrogen ion concentration
OEC         Observed Effect Concentration
POC          Particulate Organic Carbon

-------
APPENDIX B

-------
ppb
ppm
ppt
REMAP
{s2-}
[s2-]
[SEM]
ISEM]Cd
[SEM]Cu
[SEM]Ni
[SEM]Zn
SAB
SD
SLC
SEM
SOP
STORET
TDS
TOC
TU
TVS
U.S. EPA
WQC
Zn
a:-
TTF.2
Parts per billion
Parts per million
Parts per trillion
Regional Environmental Monitoring and Assessment Program
activity of S2' (mol/L)
concentration of S2" (mol/L)
simultaneously extracted metal concentration Gtmol/g)
simultaneously extracted Cd concentration (/zmol/g)
simultaneously extracted Cu concentration (/xmol/g)
simultaneously extracted Ni concentration (/zmol/g)
simultaneously extracted Pb concentration (/zmol/g)
simultaneously extracted Zn concentration (/zmol/g)
U.S. EPA Science Advisory Board
Standard Deviation
Screening Level Concentration
Simultaneously Extracted Metals
Standard Operating Procedure
EPA's computerized water quality data base
Total Dissolved Solids
Total Organic  Carbon
Toxic Unit
Total Volatile  Solids
United States Environmental Protection Agency
Water Quality  Criteria
Zinc
{Fe2*}/[2Fe(aq)]
{M2+}/[2m(aq)J
[SFe(aq)]
[SM(aq)]
[SS(aq)]
activity coefficient of Fe~
activity coefficient of M:+
activity coefficient of S2"
concentration of total dissolved Fe(II) (mol/L)
concentration of total dissolved M(II) (mol/L)
concentration of total dissolved S(II) (mol/L)

-------
           a.,. = [S2-]/[£S(aq)]
 (B-8)
 are the ratios of the divalent species concentrations to the total dissolved M(II), Fe(II), and S(-
 II) concentrations,  [SM(aq)], [SFe(aq)], and [SS(aq)], respectively. [MS(s)J and [FeS(s)] are
 the concentrations of solid-phase metal and iron sulfides at equilibrium. [FeS(s)]j is the initial
 iron sulfide concentration in the sediment, and [M]A is the concentration of added metal.

        The solution of these five equations can be obtained as follows.  The mass balance
 Equations B-3 and  B-4 for M(II) and FE(H) can be solved for [MS(s)j and [FeS(s>] and
 substituted in the mass balance Equation B-5 for S(II):
                                                       = [Ml
 (B-9)
'The mass action Equations B-l and B-2 can be used to substitute for [Fe2+] and [M2+], which
 results in a quadratic equation for [S2~]:
                                                                 = [ML
(B-10)
 The positi\'c root can he accurately approximated by:
                                                                                   (B-ll)
 which results from ignoring the leading term in Equation B-10.  This is legitimate because the

-------
                                     APPENDIX B
                         Solubility Relationships for Metal Sulfides
       Consider the following situation: a quantity of FeS is titrated with a metal that forms a
more insoluble sulfide.  We analyze the result using an equilibrium model of the M-(II)-Fe(II)-
S(-II) system. The mass action laws for the metal and iron sulfides are
         YM>[M21Ys,[S2-]  = K^                                               (B-l)
                                                                                 (B-2)
where [M2+], [Fe2"1"] and [S2"] are the molar concentrations; Y^.,]^. andys2- are the activity
coefficients; and KMS and KFeS are the sulfide solubility products.  The mass balance equations
for total M(H), Fe(II) and S(-II) are
                        [MS(s)] = [M\A                                           (B-3)
            "'
           «"  Fe"[Fe1  + [FeS(s)] = [FeS(s)].                                     (B-4)
               S:-[S2-]  + [MS(s>] - [FeS(s)]  = [FeS(s)]r                             (B-5)
where
           
-------
              2*
           {M*}
                                                                                  (B-16)
The magnitude of the term in parentheses can be estimated as follows.  The first term in the
denominator is always greater than or equal to 1, PFe2t>.  1. because it is the reciprocal of two
terms both of which are less than or equal to 1, Equation B-14.  They are «Fe2+_< 1, which is
the ratio of the divalent to total aqueous concentration, and yFe2+ _< 1, which is an activity
coefficient.  The second term in the denominator cannot be negative, PM2*KMS/KFeS  >  0, since
all of its terms are positive.  Thus, the denominator of the expression in parentheses is always
greater than 1, pFe2+  4- PM2+KMS/KFtS >  1.  Therefore, the expression in parentheses is always
less than 1. Hence, the magnitude of the ratio of metal activity to total added metal is bounded
from above by ratio of the sulfide solubility products:
(Me2*}/[M]
                          KMS/KFeS
(B-17)
This results applies if [FeS]j > [M]A so that excess [FeS(s)] is present.

       If sufficient metal is added to exhaust the initial quantity of iron sulfide, then [FeS(s)]
= 0.  Hence, the iron sulfide mass action equation (B-2) is invalid and the above equation no
longer applies.  Instead, the only solid-phase sulfide is metal sulfide and
           [MS]  -  [FeS],
                                                                       (B-18)
so that, from the metal mass balance equation
                            2-
                          N.  ([M]A - [FeS(s).)
                                                                        (B-19)
this completes the derivation of Equations 2-8 and 2-9.

-------
term in parentheses in Equation B-10 is small relative to [M]A due to the presence of the sulfide
solubility products. As a result, [S2'] is also small since it is in the denominator.  Hence, the
leading term in Equation B-10 must be small relative to [M]A and can safely be ignored.

       The metal activity can now be found from the solubility equilibrium Equation B-l:
         = YM2*[M2'] = K,
MS
                                                                       -I
                                               [M]
so that
           {M2>}
                   "
                                       MS
                                                                                 (B-13)
where
and
                                                                                 (B-14)
                                                                                 (B-15)
Equation A-13 can be expressed as

-------
                                                        APPENDIX C
^Concentrations of SEM. AVS, TOC, and IWCTU for cadmium, copper, lead, nickel, and zinc in 46 surficial samples from Lake Michigan
Sam- TOC
pie (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
K5

27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
0.18
4.63
3.36
4.89
0.92
4.37
5.27
0.08
4.27
2.11
1.89
0.41
2.87
3.68
0.28
0.07
3.51
0.40
1.73
0.69
2.51
1.17
0.13
1.03
0.63
0.30
0.29
0.21
0.11
0.05
0.27
4.95
0.54
6.75
0.18
0.15
0.56
0.10
0.06
2.68
0.16
1.80
1.29
0.05
0.14
0.57
SEM
(umol/g)
0.53
3.46
2.78
3.55
0.14
2.82
1.20
0.17
1.47
0.25
1.12
0.74
1.17
1.56
1.32
0.17
0.75
0.97
1.74
0.70
0.19
0.59
0.21
0.62
0.13
0.15
0.25
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
AVS
(umol/g)
0.03
0.35
0.06
0.05
0.03
1.13
0.13
0.03
4.49
0.03
0.03
0.07
0.18
0.03
0.44
0.05
0.08
0.03
0.15
0.03
0.05
0.03
0.03
0.03
0.20
0.03
0.03
0.03
0.06
0.03
0.03
1.66
0.12
0.09
0.03
0.05
0.05
0.12
0.03
0.03
0.07
0.03
2.25
0.03
0.05
0.03
SEM-
AVS
0.51
3.11
2.72
3.50
0.12
1.69
1.07
0.15
-3.02
0.23
1.10
0.67
0.99
1.54
0.88
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
0.11
0.64

Cadmium

0.029
0.018
0.018
0.0002
0.024
0.029
0.115
0.050
-
-
0.0002
-
0.0002
0.0002
-
0.018
.
0.079
-
-
-
-
.
-
-
-
0.0002
0.0002
-
-
0.012
-
0.018
-
-
-
.
-
0.003
-
0.006
0.0002
-
-
-

Copper

0.003
0.308
0.266
0.034
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
-
0.003
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
-
0.0008
-
0.0008
.
.
-
-
.
.
-
-
0.0001
0.0004
-
-
0.0004
.
0.0002
-
-
-
.
- •
0.001
-
0.0006
0.002
-
.
-
IWCTU
Nickel
.
0.005
0.003
0.003
0.006
0.004
0.006
0.006
0.004
-
-
0.0005
.
0.006
0.004
-
0.008
.
0.010
.
-
-
-
.
.
-
-
0.011
0.007
-
.
0.002
-
0.017
.
-,
. -
.
-
0.0005
-
0.008
0.0005
-
-
-
% Survival
Zinc

0.003
0.029
0.006
0.032
0.020
0.020
0.055
0.026
-
-
0.001
-
0.015
0.050
-
0.058
-
0.020
-
-
-
-
-
.
-
-
0.0003
0.0003
-
-
0.020
-
0.012
-
-
-
-
-
0.020
-
0.015
0.044
-
-
-
Sum Hyalella Chironomus
azteca tentans
.
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
-
.
-
92.5
90
92.5
100
0
97.5
92.5
95
95
77.5
97.5
-
97.5
96.5
90
100
100
95
97.5
97.5
75
97.5
57.5
72.5
95
. .
35
75
80
97.5
97.5
97.5
100
95
95
95
:
60
97.5
90
62.5
75
100
82.5
.
70
40
90
90
97.5
90
100
100
87.5
100
87.5
100
-
97.5
92.5
87.5
100
100
100
97.5
97.5
92.5
100
65
57.5
90
.
35
72.5
82.5
100
97.5
95
100
90
100
92.5
-
55
100
95
65
95
55
72.5
-
67.5
   aAVSLOD=0.05umS/g
   b Insufficient pore-water volume for metals analysis
   c Cadmium LOD=0.01 ug/L (0.0002 IWCTU)
            LOD =0.2 ug/L (0.0003 IWCTU)
          LOD=0.1 ug/L (0.0001 IWCTU)
   f Nickel LOD =0.5 ug/L (0.0005 IWCTU)
   Source: Columns for Sample. TOC, SEM, AVS, SEM-AVS and IWCTU taken directly from Leonard et al.. 1996a.  Column for survival
   from personal communication with Leonard. 1998.

-------
APPENDIX C

-------
                                APPENDIX D
Concentrations of SEM, AVS, Toxicity and TOC for EMAP, NOAA NS &T and REMAP Databases
STUDY'
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
EMAP-VA
SEM
umol/&
.289
1.500
.066
.134
.266
.266
1.292
.347
.750
.212
.497
.624
.032
.988
.604
.031
1.597
1.065
.189
.018
.079
.421
.798
.903
1.202
.159
.246
.687
.699
1.663
.083
.740
.878
.044
.910
.567
.734
2.171
3.423
.197
.162
2.803
.472
2.079
.445
2.228
.847
1.402
1.425
.263
2.936
.394
3.074
2.555
.452
.173
.578
.209
AVS
umol/
1.400
.742
.029
.028
3.740
1.080
1.230
.087
.948
.283
.490
13.400
.024
81.100
3.340
.331
72.400
8.480
6.460
.034
.976
3.210
68.000
3.150
67.700
3.310
4.870
2.420
.430
116.000
1.300
.976
1 .220
.025
3.430
.621
25.000
5.610
138.000
.892
3.590
11.900
12.500
26.600
.056
15.100
17.300
52.700
22.300
.079
29.600
.031
10.400
.402
.480
.201
.257
3.460
SEM-AVS
ugmol/g
-1.111
.758
.037
.106
-3.474
-.814
.062
.260
-.198
-.071
.007
-12.776
.008
-80.112
-2.736
-.300
-70.803
-7.415
-6.271
-.016
-.897
-2.789
-67.202
-2.247
-66.498
-3.151
-4.624
-1.733
.269
-114.337
-1.217
-.236
-.342
.019
-2.520
-.054
-24.266
-3.439
-134.577
-.695
-3.428
-9.097
-12.028
-24.521
.389
-12.872
-16.453
-51.298
-20.875
.184
-26.664
.363
-7.326
2.153
, -.028
-.028
.321
-3.251
SURVIVAL"
%
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.
96.
SIGNIFICANCE*
%
o
o
w
o
o
o
o
0
o
o
0
V
o
o
V
o
o
o
o
o
o
o
o
o
V
o
o
o
V
o
V
o
o
o
v
o
* w
o
o
o
\l
o
o
\J
o
o
o
\J
o
o
V
o
o
o
o
o
o
o
o
o
\J
o
n
V
o
o
V
o
v
0-

0
TOC
r/\
.ou
2£O
.00
17
. I /
4Q
.**y
^A
.JO
1.80
in
. j\/
o*.
-SO
17
.J/
i nn
I .UU
i 
4b . 7fc
7 Ifi
^ . JO
2 70
*f.l\J
1 M
J . In
77
./ /
4 1<
n . i j
Ifi
. 1 5
7 47
X.n /
7 Ifi
X. 1 O
i m
i .\j i
77
. iZ
£<
.Oj
.36

-------
APPENDIX D

-------
STUDY1
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
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- BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
SEM
1.753
2.447
1.839
1.296
1.697
1.390
2.310
.399
2.481
1.736
.958
9.192
1.525
.678
5.037
4.202
1.174
1.855
3.092
2.997
2.581
2.869
5.442
2.618
5.061
2.376
6.998
4.480
4.662
5.896
3.103
1.662
3.512
.273
.335
1.664
2.674
5.532
4.029
4.614
3.379
4.240
4.303
5.209
4.801
4.697
:.600
1.013
1.527
.505
3.341
3.449
.270
.341
.888
.722
.362
2.138
3.008
.151
AVS
umol/
17.697
10.'958
68.306
56.838
9.089
43.801
51.857
3.899
19.604
148.969
18.622
120.622
81.842
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
.050
.036
18.760
3.630
29.210
18.440
20.530
30.120
19.320
22.570
14.570
35.370
54.710
56730
10.160
15.130
.630
43.920
37.860
.950
.156
12.971
4.948
.936
3.295
3.941
.555
SEM-AVS
ugmol/g
-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
-25.21 1
-20.458
1.538
-7.179
-2.014
-56.452
-16.300
-19.058
^5.684
-56.677
-5.568
-22.328
.223
.299
-17.096
. -.956
-23.678
-14.411
-15.916
-26.741
-15.080
-18.267
-9.361
-30.569
-50.013
-54.130
-9.147
-13.603
-.125
-40.579
-34.411
-.680
.185
-12.083
-4.226
-.574
-1.157
-.933
-.404
SURVIVAL"
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.
2.
77.
19.
0.
91.
93.
69.
3.
96.
51.
91.
88.
101.
102.
101.
70.
38.
37.
29.
68.
105.
86.
76.
96.
84.
92.
85.
98.
95.
95.
96.
SIGNIFICANCE1 TOC
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1.
1.
1.
1.
0
0
0
0
0
0
0
0







0
0
1.
1.
0
-1.
0
0
0
0
0





0
0
1.
0
0
0
0
0
0
0
0

1.41
4.45
2.54
3.05
2.68
3.27
3.35
.80
3.31
2.94
1.77
4.61
2.96
1.45
5.02
3.47
1.88
4.44
3.86
3.09
2.86
2.50
2.20
2.67
2.98
2.49
1.98
2.98
3.19
4.78
3.99
2.61
4.44
.07
.07
.69
1.00
3.18
2.20
1.94
2.80
3.15
3.02
3.21 '
2.98
3.47
1.47
.77
.95
.25
2.55
3.63
.26
.06
4.05
.40
.26
.43
.18
.15

-------
STUDY*

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- U
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
NOAA- BO
NOAA- BO
NOAA- BO
NNOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
NOAA- BO
SEM
umoI/E
2.813
1.235
2.198
3.624
3.594
1.342
2.462
.964
.332
2.311
.623
.896
.544
.641
.355
.222
2.262
1.307
1.963
2,785
4.333
1.927
.004
3.831
.808
1.783
2.622
.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
.872
1.080
.123
2.914
2.218
2.609
3.650
1.634
1.267
2.892
2.511
.661
2.458
1.872
.959
2.480
.784
.943
1.683
AVS
umol/
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
.380
51.820
61.020
16.080
3.710
24.580
9.250
.960
40.630
61.840
1.090
3.730
50.390
62.760
33.630
7.220
17.120
17.810
100.800
83.010
26.730
30.880
32.050
35.390
25.810
11.300
5.310
2.893
2.369
43.959
101.984
5.237
3.256
80.584
2.241
13.490
23.077
48.062
53.288
7.599
22.486
8.831
42.399
SEM-AVS
ugmol/g
-32.237
-.844
-12.492
-18.176
-23.816
-36.628
-43.988
-.036
-3.678
-77.579
-5.987
-15.475
-1.626
-1.419
-1.035
-3.958
-37.698
.927
-49.857
-58.235
-11.747
-1.783
-24.576
-5.419
-.152
-38.847
-59.218
-.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
.021
-.151
-41.350
-98.334
-3.603
-1.989
-77.692
.270
-12.829
-20.619
-46.190
-52.329
-5.119
-21.702
-7.888
-40.716
SURVIVAL"
%
86.
.
84.
83.
82.
82.
82.
81.
ffl
ol.
81.
80.
80.
79.
"7Q
17.
7O
/y.
77.
77.
7A
/O.
76.
76.
75.
75.
74.
*y\
/j.
71
/I.
70.
70.
/rn
oy.
Afl
Oo.
67.
67.
64.

-------
STUDY'
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB .
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
REMAP-NB
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
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
SEM
umol/£
1.894
3.149
.632
1.057
.638
1.087
3.711
2.990
8.894
1.277
3.925
5.632
6.809
7.645
4.012
3.905
.942
3.515
2.216
3.323
3.391
• 3.443
2.466
2.294
5.768
1.013
2.479
.554
5.222
5.116
14.791
4.917
.398
4.855
3.290
5.822
9.167
6.214
.794
4.985
5.280
2.268
6.678
2.833
.333
.756
.582
1.012
1.596
.326
2.709
5.485
3.596
5.329
.337
.986
.856
5.364
1.706
.371
AVS
umol/
25.394
64.643
1.310
4.647
.218
.312
17.184
59.256
60.816
23.266
42.727
1 14.770
135.354
150.012
43.663
26.229
6.531.
7.134
1 1 .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
.530
.218
9.606
10.105
51.460
93.563
42.415
2.651
43.663
1.934
6.300
17.559
45.222
22.315
1.216
.821
.567
.447
.156
3.120
14.666
19.503
4.321
2.901
.156
.156
39.700
23.515
4.210
SEM-AVS SURVIVAL" SIGNIFICANCE* TOC
u&mol/g % %
-23.500
-61.494
-.678
-3.590
.420
.775
-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
-.539
-17.807
-8.752
.740
-10.066
-23.208
-2.080
-17.395
-2.236
-94.989
4.387
.180
-4.751
-6.815
-45'.638
. -84.396
-36.201
-1.857
-38.678
3.346
-4.032
-10.881
-42.389
-21.982
-.460
-.239
.445
1.149
.170
-.411
-9.181
-15.907
1.008
-2.564
.830
.700
-34.336
-21.809
-3.839
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.
53.
83.
16.
77.
54.
93.
92.
94.
94.
95.
93.
70.
92.
62.
91.
97.
96.
96.
91.
93.
91.
0
0
0
0
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
1.
1.
0
0
0
1.
1.
1.
1.
0
1.
0
1.
1.
1.
0
0
0
0
0
0
1.
0
1.
0
0
0
0
0
0
0
.98
.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
.67
.75
1.22
1.25
1.05
.88
1.40
.95
1.77
.76
.99
.60
1.48
1.45
9.15
3.10
2.42
2.62
5.70
2.22
6.48
3.24
2.36
3.90
6.10
1.99
15.20
2.02
1.23
' .33
.30
.30
.17
.08
.42
2.29
.88
.97
.53
.12
.51
1.17
3.21
3.54

-------
STUDY'       SEM          AVS    SEM-AVS
            jimol/£	SHP.2LL	HfiE?l/£.
SURVIVAL"
SIGNIFICANCE" TOC
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-BA
REMAP-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
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-JB
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
REMAP-LS
.115
.543
.103
.167
.073
.294
.120
.109
.185
.120
.347
.120
2.275
.344
.258
.119
.258
.494
.109
.266
.327
.230
2.026
14.550
3.332
3.763
.357
.524
.244
1.247
2.478
1.744
.131
.846
4.399
3.884
.673
3.150
.270
.162
2.880
.323
.413
.377
.099
1.100
.209
.213
.954
2.759
.711
1.915
2.186
2.480
.606
3.289
3.241
.616
1.506
2.485
.156
.156
.156
.932
.156
.156
.156
.156
.156
.156
.156
.156
16.592
.012
.343
.156
.156
.156
.156
.156
.393
6.400
47.793
389.857
243.322
201.687
10.923
3.974
4.502
48.130
47.376
.156
1.184
.927
116.954
237.650
21.769
43.975
4.491
.873
153.755
1.684
3.056
3.056
.686
58.945
1.466
.780
1.542
6.498
10.240
12.596
17.605
23.523
2.501
91.773
56.100
1.070
26.201
28.248
-.041
.387
-.053
-.765
-.083
.138
-.036
-.047
.029
-.036
.191
-.036
-14.317
.332
-.085
-.037
.102
.338
-.047
.110
-.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
-.081
-112.555
-233.766
-21.096
-40.825
.-4.221
-.711
-150.875
-1.361
-2.643
-2.679
-.587
-57.845
-1.257
. -.567
-.588
-3.739
-9.529
-10.681
-15.419
-21.043
-1.895
-88.484
-52.859
-.454
-24.695
-25.763
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.
92.
93.
96.
93.
95.
83.
96.
97.
97.
95.
99.
98.
95.
97.
95.
96.
96.
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
0
0
1.
1.
0
1.
0
0
1.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.08
.07
.05
.16
.05
.34
.83
.92
4.48
.83
1.26
.62 .
1.81
3.85
.77
2.23
.88
2.10
4.07
1.06
.29
.19
.77
1.52
.83
.97
.26
.35
.27
.54
1.12
1.14
.21
1.58
6.55
8.45
4.11
5.47
.74
1.40
7.70
.20
1.20
1.30
.75
3.86
.58
.69
.26
.45
.56
.21
.27
.32
.25
.77
1.14
.15
.95
.25

-------
  STUDY*
                  SEM
 AVS    SEM-AVS
.HinPi'L _  SJgjnoI/g
SURVIVAL11
SIGNIFICANCE*  TOC
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-RB
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
REMAP-UH
.193
.869
1.288
1.650
2.422
.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
.930
.459
.889
.833
1.317
2.480
.626
1.500
.723
4.158
2.241
2.907
.852
2.294
2.995
2.981
.677
.156
19.617
.593
.624
.156
.156
4.086
36.490
5.957
8.078
17.247
.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
.156
2.623
2.464
15.563
32.123
9.949
5.427
1.341
13.504
27.788
29.285
1.591
53.955
33.995
44.910
10.323
.037
-18.748
.695
1.026
2.266 •
.356
.112
-31.409
.138
.393
-13.877
1.042
-10.319
-.430
-2.176
-15.161
-22.360
-53.571
-43.203
-10.030
-83.049
-9.627
-10.832
-9.163
.303
-1.734
-1.631
-14.246
-29.643
-9.323
-3.927
-.618
-9.346
-25.547
-26.378
-.739
-51.661
-31.000
-41.929
-9.646
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.
93.
15.
88.
94.
91.
0
0
0
0
0
0
0
0
1.
0
0
0
0
0
0
1.
0
0
0
0
0
0
1.
0
0
0
0
0
0
0
0
0
0
1.
0
0
1.
0
0
0
2.52
2.39
2.44
2.68
2.60
.42
2.63
2.08
3.03
5.30
3.91
1.03
3.43
1.26
5.85
2.33
.91
1.21
1.03
1.06
1.39
.79
1.06
.43
.13
.21
4.96
2.56
3.06
2.58
2.71
3.89
4.78
2.66
5.15
2.03
4.37
3.55
2.97
3.32
a)  Sources: EMAP-VA is U.S. EPA. 1996
             NOAA-LI is  Wolfe el al., 1994
             NOAA-BO is Long et al., 1996
             NOAA-HR is Long el al.. 1995
             REMAP is Adams el al.. 1996
b)  Conclusion of signifigance varies for three databases.
    EMAP significance based on percent survival of control
    NOAA significance based on percent survival less than 80%
    REMAP significance based on  percent survival less than 80%
c)  Significance:  0 - No significant toxicity
             1 - Significant toxicity

-------