DEVELOPING SEDIMENT QUALITY STANDARDS:
               COMPREHENSIVE SEDIMENT MANAGEMENT IN  PUGET SOUND

                                 C.  C. Krueger

                 United States Environmental  Protection Agency
                            Hazardous Waste Division
                           1200 Sixth Avenue, HW-113
                            Seattle Washington 98101


                                    ABSTRACT


     High concentrations of potentially harmful toxic chemicals  have  been
identified in the sediments of a number of urban-industrial  bays in Puget
Sound.  In these areas, field studies have documented an increased frequency
of fish disease, sediment toxicity,  altered benthic  communities, and
significant bioaccumulation of harmful chemicals in  the edible tissue of fish
and shellfish.  In response to this information, and a growing public concern
about the health of the estuary, the Washington state Department of Ecology
has established a comprehensive strategy for sediment management in Puget
Sound.  As a component of this strategy, the agency  is now in the process of
developing a suite of sediment management standards  for use in a variety of
regulatory programs.  General sediment quality standards are now available in
draft form.  Once finalized and officially adopted,  the standards will be used
to identify and designate sediments that have adverse effects on biological
resources or pose a health risk to humans.  It is anticipated that the general
sediment quality standards will also be used as a basis for limiting
industrial and municipal discharges, thereby preventing future sediment
contamination.  Separate, but related, sediment management standards  are also
being developed for use in establishing cleanup goals for sediment remediation
and in making environmentally safe decisions concerning the disposal  of
contaminated dredged material.  The Department of Ecology will be using the
apparent effects threshold approach, supplemented by the equilibrium
partitioning approach, as the technical basis for the derivation of the
sediment standards.  Field validation studies indicate that these methods can
be used to generate chemical specific standards which are reliable predictors
of adverse environmental impacts associated with sediment contamination in
Puget Sound.  These methods represent tools with potential widespread
application.

                                  INTRODUCTION

     On December 29, 1987, the Washington state legislature adopted the Puget
Sound Water Quality Management Plan (PSWQA plan).  The PSWQA plan,  prepared by
the Washington state Puget Sound Water Quality Authority (PSWQA), in
cooperation with the United States Environmental Protection Agency (U.S. EPA),
identified existing and emerging environmental problems in Puget Sound,
established environmental goals for the restoration and protection  of the
estuary, and detailed specific actions to be taken by federal,  state  and local
governments towards achieving those goals.  The PSWQA plan addressed  a variety
of environmental issues, including point and nonpoint source pollution,
wetlands protection, environmental monitoring, and contaminated  sediments.
This paper summarizes the PSWQA plan's recommendations concerning sediment

-------
contamination, and describes the comprehensive regulatory framework  and
technical approach currently being developed in Washington state  to  identity
and manage contaminated sediments in Puget Sound.

                A COMPREHENSIVE STRATEGY FOR SEDIMENT MANAGEMENT

     The PSWQA plan established a long-term management goal  for sediment
quality in Puget Sound.  The goal, "to reduce and  ultimately eliminate adverse
effects on biological resources and humans from sediment contamination
throughout the Sound by reducing or eliminating discharges of toxic
contaminants and by capping, treating or removing  contaminated sediments.11,
will not easily be achieved.  Sediments in many parts of Puget Sound,
particularly the nearshore, urban and industrial areas, are currently
contaminated with high levels of potentially toxic substances, including heavy
metals, polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons,
and other organic chemicals.  The past and present sources of these
contaminants are diverse, including urban runoff and a variety of industrial
and municipal discharges.  Field studies conducted in Puget Sound in the early
and mid-1980s documented increased prevalence of sediment toxicity,  altered
bottom dwelling communities, and histopathological disease in fish and
shellfish living in association with contaminated  sediments.  Recent evidence
of significant bioaccumulation of cancer-causing chemicals in fish tissue has
prompted several health departments in the Puget Sound region to  advise
residents to limit their consumption of locally harvested seafoods.

     Past efforts to prevent sediment contamination and to identify, manage
and cleanup contaminated sediments in Puget Sound were hampered  by the absence
of adopted state or federal sediment quality standards or criteria.
Recognizing this void, the PSWQA plan identified an overall strategy for
sediment management (PSWQA 1987).  The strategy consists of four key elements:

     (1)  classification of sediments in the Sound that cause adverse
          biological effects,

     (2)  implementation of Soundwide source control to prevent  future
          sediment contamination,

     (3)  provision of rules and sites for the disposal of contaminated and
          uncontaminated dredged material, and,

     (4)  the development of guidelines for use in sediment cleanup actions  at
          heavily contaminated sites.

     In response to this overall strategy, and the goals set forth in  the
PSWQA plan, the Washington state Department of Ecology  (WDOE) initiated the
development of a comprehensive sediment management program for Puget Sound  in
1987.   Integral to this program are the development of  five categories of
sediment management standards: (1) general sediment quality standards,  (2)
effluent particulate discharge limits, (3) disposal standards for unconfined
dredged material, (4) disposal standards for confined dredged material,  and
(5) cleanup guidelines for sediment remedial action at  highly contaminated
sites.  The different standards are currently  in varying stages of
development.

-------
General Sediment Quality Standards

     In December 1989, WDOE Issued the general  sediment quality  standards  in
draft form.  The draft standards are now undergoing public review and
comment.  WDOE anticipates formal  adoption of the general  standards,  by
regulation, in June 1990.  Once adopted, the general sediment quality
standards will serve the dual purpose of: (1) reaffirming  the water quality
goals (i.e., maintenance  and protection of human health and biological
resources) already established for Puget Sound in Washington state's water
quality standards, and (2) identifying specific chemical and biological
criteria (i.e., acceptable levels of chemicals in sediments, bioassay toxicity
limits) which must be met to ensure protection of the beneficial  uses of the
estuary.  As required in the PSWQA plan, the general sediment quality
standards will enable identification and designation of sediments that have
acute or chronic adverse effects on biological  resources or pose a significant
health risk to humans.  It is currently anticipated that the standards will
include chemical, physical and biological tests and clearly defined pass/fail
guidelines for the tests.  In addition, the sediment quality standards will
describe the intended use of the chemical and biological criteria and describe
variance procedures for,,a variety of regulatory programs.   The methods that
will likely be used to identify chemical concentration limits for the general
sediment quality standards are discussed in a later section of this paper.

Effluent Particulate Discharge Limits

     WDOE is also required by the PSWQA plan to develop procedures for
limiting discharges of pollutants to Puget Sound.  The purpose of these
procudures will be to ensure that future effluent discharges do  not result in
violation of the general sediment quality standards, and that the potential
for future sediment contamination is minimized.  It is currently anticipated
that final rules addressing effluent control limits and the relationship
between effluent control and the general sediment standards will be issued by
WDOE in June 1990.  WDOE is currently considering a variety of approaches for
enhanced control of pollutant discharges, including, but not limited to,
reliance on best available technology, numerical limits on the toxicity of the
particulate fraction of the effluent, and numeric limits on the  mass or
concentration of chemicals discharged.  As part of their discharge management
strategy, WDOE is contemplating the establishment of sediment impact zones in
wastewater discharge permits.  Conceptually, sediment impact zones would be
similar to dilution or mixing zones in the water column, and would be
represented by a limited area near a discharge point in which limited
exceedance of sediment quality criteria would be permissible.

Dredged Material Disposal Standards

     The rules governing the management of dredged material in Puget Sound are
currently in various stages of development.  Procedures and guidelines for
identifying disposal sites, and for evaluating relatively uncontaminated
sediment and disposing of the material at unconfined, open-water sites, were
established in 1988 as part of PSDDA, the Puget Sound Dredged Disposal
Analysis.  PSDDA is a cooperative program sponsored jointly by the U.S. Army
Corps of Engineers, the U.S. EPA, WDOE, and the Washington state Department of
Natural Resources.  The PSDDA sediment evaluation procedures, which combine
chemical evaluation of sediment and biological toxicity testing  (PSDDA 1988),
are now being implemented in Puget Sound by federal and state agencies. WDOE

-------
will formally endorse the PSDDA standards for unconfined dredged  material
disposal as part of the state sediment management  strategy in  July 1990.

     Procedures for evaluating moderately to heavily contaminated dredged
material, and guidelines for safe disposal  of this material  in confined upland
or nearshore containment areas, are currently being developed.  As called for
in the PSWQA plan, the "confined disposal standards" are to be used by WDOE
and local governments in the dredged material permit process,  and shall
address treatment and disposal options for contaminated dredged material  both
in water and on land.  WDOE anticipates the adoption of the final confined
disposal standards in July 1990.

Sediment Remedial Action Guidelines

     Finally, the PSWQA plan directs WDOE to develop standards for deciding
when heavily contaminated sediments should be capped, excavated,  or otherwise
treated.  The sediment cleanup remedial action guidelines which will serve
this purpose are currently being developed.  When  finalized, the  guidelines
will specify chemical concentration trigger levels for use in identifying
sites for expedited remedial action.  In establishing these guidelines, WDOE
is including consideration of the roles of source  control and pollution
prevention, natural recovery, and maintenance dredging in sediment cleanup.
It is not anticipated that all sediments which exceed the general sediment
quality standards will be subject to cleanup per the state remedial action
guidelines.  Rather, the PSWQA plan allows that a  distinction be  made between
low to moderate exceedance of sediment quality goals and extreme  circumstances
(i.e., exceedance of sediment remediation trigger  levels) that warrant direct
intervention.  It is anticipated that only the most heavily contaminated, and
thus the highest priority, sediments in Puget Sound will exceed the remedial
action trigger levels and be subject to expedited  cleanup.  It is anticipated
that WDOE will adopt final remedial action guidelines in June 1990.

        EVALUATION OF METHODS FOR DEVELOPING SEDIMENT QUALITY STANDARDS

     When the PSWQA plan first directed WDOE to develop a comprehensive
program for sediment management, it was understood that innovation would be
required due to the absence of formally adopted federal sediment standards or
criteria.  It was also understood that the sediment assessment methods used  in
standards development and source control in Washington state would likely
represent application of new and controversial techniques.

     Over the past decade, a variety of techniques have been developed by
federal, regional, and state agencies in the United States for evaluating the
significance of chemical contamination in sediments.  Many of the early
efforts involved simple comparisons of chemical concentrations at contaminated
sites to concentrations in sediments from relatively unpolluted reference
areas.  Later, more sophisticated methods were developed which incorporated
not only consideration of sediment chemistry, but also attempted to establish
relationships between specific concentrations of individual chemicals  in
sediment and associated adverse biological and human health effects.   As a
first step in developing sediment standards for Puget Sound, WDOE and  U.S. EPA
evaluated a variety of field-based and laboratory-based approaches to
establishing numerical sediment criteria.  The specific methods which  were
considered by WDOE and U.S. EPA are listed below.

-------
Field Based Approaches

  0  Reference Area Comparison - sediment quality values derived based  on
     comparison of chemical concentrations at a site with  concentrations at an
     acceptable reference area (i.e.,  a relatively unpolluted site with
     otherwise similar sediment characteristics).

  0  Field-Collected Sediment Bioassays - sediment management decisions made
     by exposing test organisms in the lab to field collected sediment.
     Comparisons are then made between the mortality or sublethal effects
     observed in field sediments and effects observed  in experiments using
     sediment from a relatively unpolluted reference area.

  0  Screening Level Concentration - sediment quality  values determined using
     field samples to identify the sediment concentration  above which  95%  of
     the enumerated benthic infauna species are present.


  0  Sediment Quality Triad - sediment quality values  qualitatively derived by
     analyzing relationships among contaminant concentrations in sediment, the
     results of sediment toxicity bioassays, and the characteristics of in
     situ benthic communities.

  0  Apparent Effects Threshold - sediment quality values  quantitatively
     derived by using synoptically collected sediment  chemistry, benthic
     infauna effects, and sediment bioassay data to determine the chemical
     concentration in sediment above which statistically significant
     biological effects are always observed.

Laboratorv/Theoreticallv-Based Approaches

  0  Water Quality Criteria/Interstitial Water - concentrations of chemicals
     in sediment interstitial water are measured and compared directly to  the
     U.S. EPA chronic water quality criteria.

  0  Sediment-Water Equilibrium Partitioning - sediment quality values derived
     by using a theoretical model to predict the partitioning of contaminants
     between sedimentary organic carbon and interstitial water.  Predicted
     interstitial water concentrations are then compared directly to  the  U.S.
     EPA chronic water quality criteria.

  0  Sediment-Biota Equilibrium Partitioning - sediment quality values
     determined by using a model to estimate the sediment  concentration of a
     contaminant that would be expected to result in a body burden of the
     contaminant in benthic organisms exceeding a regulatory  limit  (e.g.,  the
     U.S. Food and Drug Administration limits for chemicals in  fish  and
     shellfish).

  0  Spiked Sediment Bioassays - test organisms are exposed to  sediments that
     have been inoculated with known amounts of chemicals. Dose/response
     relationships are then used to identify chemical  concentrations  which do
     not result in adverse biological impacts.

     Generally, the field-based approaches evaluated by WDOE  and  U.S.   EPA
relied on empirical observations of biological and/or  chemical  measurements to

-------
establish sediment quality values.   Approaches such as the Triad  and AET
methods identified relationships between specific concentrations  of chemicals
in field collected sediments and adverse biological responses to  exposed
organisms.  The laboratory/theoretically-based approaches evaluated by WOOE
and U.S. EPA generally relied on extrapolation of federal water quality
criteria to sediments, models of environmental interactions, or extrapolation
of laboratory studies to develop sediment quality values.  A more detailed
description of each of the methods  listed above can be found in Barrick et al.
1989 and U.S. EPA 1989B.

Comparison of Approaches

     In evaluating the various methods for use in sediment standards
development, VIDOE identified eleven criteria that enabled objective assessment
of the approaches with respect to three important attributes:

  0  applicability of the approach  to existing and planned sediment
     management  programs in Puget  Sound, including those identified  in the
     PSWQA plan,

  0  ability of the approach to generate criteria that are reliable predictors
     of adverse environmental effects, and,

  0  feasibility of implementation  of the approach in the near  term (i.e., by
     1990).

Specifically, each approach was evaluated based on the following  eleven
criteria:

  -  data requirements and cost of sediment quality standards development,

  -  cost of routine application as a regulatory tool,

  -  ability to develop chemical-specific sediment quality standards,

  -  ability to develop sediment quality values for a wide range  of chemicals,

  -  current availability of values for a wide range of.chemicals of concern
     in Puget Sound,

  -  ability to incorporate the influence of chemical mixtures  in sediments,

  -  ability to consider adverse effects on a range of biological indicator
     organisms,

  -  extent to which the approach incorporates direct measurement of
     sediment-associated biological effects,

  -  compatibility of approach to use of historical sediment  chemistry data,

  -  ease and extent of field verification in Puget Sound, and,

  -  extent to which the approach provides proof of a cause/effect
     relationship between concentrations of specific chemicals  in sediment and
     adverse biological effects.

-------
     Table 1 summarizes the results of WDOE's  comparative  analysis.  The table
highlights the relative advantages and limitations  of  each of the methods  as
gauged against the eleven evaluation criteria.   For simplicity of
presentation, each approach was assigned a subjective  scoring of "-",  "0",  or
"+" to enable a relative comparison based on each criterion.  A "-"  has been
assigned in cases in which an approach does not meet the conditions  of the
criterion (e.g., relative to a cost criterion,  a method is expensive to
develop), a "0" was assigned in cases in which an approach somewhat  meets  the
conditions of a criterion (e.g., an approach may be moderately expensive to
develop), and a "=" is assigned in cases in which an approach substantially or
fully meets the conditions of a criterion (e.g., an approach is not  expensive
to develop).  N/A is assigned in cases in which a criterion is not applicable
to an approach.  A more complete analysis of the advantages and limitations of
each method and the scoring rationale for each criterion can be found  in
Barrick et al. 1989.

     Of the nine methods reviewed, two were considered by  WDOE to be the most
promising for developing potential sediment quality standards for Puget Sound:
the apparent effects threshold approach (AET)  and the  sediment-water
equilibrium partitioning approach (EP).  Descriptions  of the AET and EP
approaches, and a brief discussion of the technique used to field validate
sediment quality values derived from these methods  are presented below.

The Apparent Effects Threshold Approach

     The AET approach estimates concentrations of a given  sedimentary
contaminant above which statistically significant  (P  0.05) adverse  biological
effects are always expected (U.S. EPA 1989).  The AET  values for  individual
chemicals or groups of chemicals are derived from actual  sampling data,  and
are based on the statistical relationship between the  contaminant level
measured in field collected sediments and the  results  of  biological  tests
conducted on the same sediments.  The approach is appropriate  for use  with any
organic or inorganic contaminant, and does not require a  priori  assumptions
concerning the specific mechanism for interactions  between contaminants  and
organisms (Seller et al. 1986).  AET-based sediment quality values  can be
developed for any biological effects indicator that can  be statistically
evaluated relative to reference conditions.  Two kinds of biological effects
indicators have been used in developing Puget  Sound AET:  sediment  toxicity
bioassays and benthic community evaluations.

     The laboratory bioassays that have been used to date  in Washington state
to develop AETs are the amphipod mortality bioassay (using Rhepoxvnius
abronius), the oyster larvae abnormality bioassay  (using  Crassostera gigas),
and the Microtox bacterial bioluminescense bioassay (using Photobacterium
phosphoreum) (Barrick et al. 1988).  Generally, these  bioassays  involve the
controlled laboratory exposure of test organisms to field  collected sediment
for a fixed period of time, and an assessment  of acute or  sublethal  effects
resulting from the exposure.  Development of AET based on  indigenous biota has
been achieved by classifying and counting organisms found  in  sediment
collected from contaminated areas, and by then comparing the  abundance of
these organisms to conditions in appropriate reference areas.

     Figure 1 illustrates the derivation of a toxicity AET for lead.  Each
square on the figure represents a sediment sample  that was analyzed for lead
(and many other chemicals), and a hypothetical measure of  adverse biological

-------
response (e.g., amphipod mortality).   All  available sediment samples were
classified into two groups: (1)  samples  that did  induce statistically
significant adverse biological  response  (top bar), and (2) samples that  did
not induce statistically significant  adverse biological response  (bottom
bar).  Lead concentrations of samples within each group were then rank ordered
by increasing concentration.  The AET was  established by the station in  Group
2 with the highest concentration that did  not  exhibit statistically
significant biological effects (660 mg/kg  lead in this example).  Above  the
AET, non-contridictory evidence exists that only  significant biological
effects were observed in this data set.   If the lead AET derived  from  this
dataset were applied to an independent dataset, it is expected that a  high
percentage of samples with lead concentrations above 660 mg/kg would be
associated with toxicity to this test organism, or organisms of similar
sensitivity.

     The AET approach focuses on the  fact  that sediments can have
concentrations of a given chemical as high as  that chemicals AET  and still
have no observed biological effects.   Thus,  it is assumed  in the  AET approach
that effects observed at concentrations  below  the AET for  one chemical could
have resulted from unmeasured, covarying chemicals, interactive effects  of
multiple chemicals, or other chemicals present at concentrations  above their
respective AET values.  The occurrence of biologically impacted stations at
concentrations below the AET of a single chemical does not imply  that  AET in
general are not protective against biological  effects.  Rather, the
implication is that single chemicals  may not account for all stations  with
biological effects.  Field validation studies  indicate that a high  percentage
of all stations with biological  effects  can  be accounted for by developing AET
for multiple chemicals.

Sediment-Water Equilibrium Partitioning  Approach

     The EP approach is based on a simple model that describes the  equilibrium
partitioning of a contaminant between sedimentary organic  carbon  and
interstitial water, with little dependence on  other physical or chemical
factors.  A sediment quality value based on  the EP approach is the  sediment
contaminant concentration  (organic carbon normalized) that would  be expected
to result in an uncomplexed interstitial water concentration equivalent to the
corresponding U.S. EPA chronic water quality criterion for that chemical  (U.S.
EPA 1989B).  If the predicted interstitial concentration for a given
contaminant exceeds its respective chronic water  quality criterion,  then  the
sediment would be expected to cause adverse  biological effects.   The  approach
is assumed applicable to a variety of environmental settings,  including
sediments with very low total organic carbon content  (U.S. EPA  1988).

     The primary advantage of the EP approach  is  that  it uses the existing EPA
water quality criteria toxicological  database  as  a means of estimating the
potential for contaminated sediment to cause adverse  biological  effects.
Thus, use of this method does not require incurring the  expense  of collecting
new, site-specific biological data.  For nonpolar organic  compounds,  the  EP
model has a firm theoretical and empirical basis  (U.S. EPA 1988), and field
verification studies indicate that EP-based  sediment  criteria  are reasonably
predictive  (Read et al. 1989).  For ionic, polar  organic contaminants and
metals, the mechanisms controlling the partitioning of contaminants between
sediment and interstitial water are not fully  understood.  Therefore,  the
ability of  interested agencies to develop sediment quality standards  for

-------
metals and polar organic contaminants using  the  EP  approach  is currently
limited.  U.S.  EPA has initiated an extensive research  effort to refine
partitioning models and to expand the utility of the  EP approach to these
groups of compounds.

Overview of Approach  to Field Validation of  Sediment  Quality Values

     Because none of  the available approaches to developing  sediment  quality
standards are fully capable of addressing all concerns  over  interactive
effects among chemicals and the effects of multiple chemicals on organisms,
WDOE and U.S. EPA determined that field verification  using diverse
environmental samples was important to the evaluation of sediment criteria
derived using the AET and EP methods.

     As a means of testing the predictive reliability of the AET and  EP
approaches when applied to field situations, WDOE and U.S. EPA conducted  a
series of field validation studies (Seller et al. 1986, Barrick et  al.  1988,
Read et al. 1989).  In designing these studies,  WDOE  and U.S. EPA acknowledged
that definitive confirmation of chemical-specific predictions would require
additional controlled laboratory spiking studies.  However,  the costs
associated with a large-scale laboratory program addressing  many chemicals was
prohibitive.  Furthermore, a feasible approach was  not  available for
confirming that the results of single chemical laboratory studies could  be
extrapolated, in a meaningful way, to environmental samples  which contained
complex mixtures of chemicals and represented a  wide  range of sediment
conditions.  As an alternative, the agencies selected an approach to  field
validation which relied on the use of AET and EP values to predict  biological
impacts associated with contaminated sediments collected from the field.   Data
from approximately 330 stations, representing 13 embayments  in Puget  Sound,
were compiled into a  single database.  Each  station included in the database
had been subjected to extensive chemical analyses of the sediment,  and
evaluated for sediment associated bioassay toxicity and/or effects  to
indigenous benthic organisms collected form  the  field.

     As the basis for the AET and EP evaluation, sediment chemistry results
for each station were compared to two different  sets  of AET  and EP  values for
a range of contaminants of concern.  In the  first evaluation, a comparison was
made for all chemicals detected in the Puget Sound database  and available for
either of the approaches (i.e., 12 chemicals for the  EP approach  and  60
chemicals for the AET approach).  A second comparison was then made only for
two chemicals,  total  PCBs and phenanthrene,  that were widely detected in Puget
Sound and common to both approaches.  In both the complete and partial
comparisons, exceedance of a chemical specific AET or EP value for  any one
chemical at a station resulted in a prediction that the results of  the
biological assessment at that station would  indicate adverse biological
effects.

     During the comparative analyses, the AET and EP predictions  were
evaluated according to three measures of reliability: sensitivity,  efficiency,
and overall reliability.  Sensitivity was defined as  the proportion of all
stations exhibiting adverse biological effects that were correctly  predicted
as impacted (i.e., all biologically impacted stations were identified as such
by the AET or EP predictions).  Efficiency was defined  as the proportion of
all stations predicted to have adverse biological effects that actually were
impacted (i.e., only biologically impacted stations were identified as such by

-------
AET or EP predictions).  Overall reliability was defined as the  proportion  of
all stations for which correct predictions were made for either  the  presence
or absence of adverse biological effects.  High overall reliability  results
from correct predictions of a large percentage of the impacted stations  (i.e.,
high sensitivity, few false negatives) and correct predictions of a  large
percentage of the nonimpacted stations (high efficiency, few false positives)
(Read et al. 1989).  The concepts of sensitivity, efficiency and overall
reliability are illustrated in Figure 2.

     As independent measures of reliability, sensitivity and efficiency  are
important concepts to consider in selecting an approach to sediment
management.  A sediment standards approach that sets criteria for a  wide range
of chemicals near their analytical detection limits will probably be sensitive
but inefficient.  That is, it will predict a large percentage of sediments
with biological effects but will also predict, as impacted, many biologically
nonimpacted sediments with only slightly elevated chemical concentrations.
Such an approach may be environmentally protective but also may  result in
overregulation that would not be cost effective.  Conversely, a  sediment
standards approach that sets criteria values at the upper end of the range  of
environmental concentrations may be efficient but insensitive.   That is, a
high percentage of the stations with predicted impacts may indeed be
biologically impacted, but the approach may fail to predict other biologically
impacted stations with moderate to high chemical concentrations.  Such an
approach would be cost-effective and defensible in pursuing high priority
remedial actions (i.e., would not result in overregulation) but  would not be
environmentally protective.  The overall reliability of .aiiy method for
establishing sediment criteria addresses both sensitivity and efficiency, and
provides perspective relative to the ability of potential sediment quality
criteria to balance the need for environmental protection and cost effective
environmental regulation.

Results of Field Validation of Sediment Quality Values

     The results of the field validation comparison using all available
sediment quality values indicated that the Puget Sound AET (i.e., mixed
organic carbon/dry weight normalized AET for 60 organic contaminants and
metals) were generally reliable predictors of adverse biological effects.
When evaluated relative to the Puget Sound database, the benthic AET values
for the 60 contaminants demonstrated a 77 percent sensitivity, 100 percent
efficiency, and an overall reliability of 88 percent in predicting the
presence or absence of adverse biological effects in benthic infauna samples
collected from 201 stations.  Similarly, the sensitivity, efficiency and
overall reliability of the amphipod AET values were 55 percent,  100 percent,
and 83 percent, respectively, with regard to accurately predicting the
presence or absence of amphipod mortality in laboratory bioassays performed  on
samples from 287 stations (Read et al. 1989).

     In general, the sediment quality values generated  using the EP approach
were less reliable than the AET-based values when evaluated relative to the
entire Puget Sound database and all available AET.  Taken in combination,  the
EP values  (i.e., organic carbon normalized mean EP values for 12 nonpolar
organic chemicals) demonstrated a 30 percent sensitivity, 65 percent
efficiency and 54 percent overall reliability in predicting adverse impacts  to
in situ benthos.  Similarly, the EP values were 28 percent sensitive, 51
percent efficient, and 63 percent reliable, overall, with respect to

-------
predicting amphipod bioassay mortality.   The somewhat lower overall
reliability of the 12 EP values in comparison with the 60 AET  values  is  likely
attributed to the lower number of detected chemicals for which EP values were
available relative to AET values.  Higher predictive reliability will  likely
be possible using the EP approach when partitioning coefficients for  metals
and polar organic contaminants become available for use in generating EP-based
sediment quality criteria.

     The results of the reliability tests performed using only phenanthrene
and total PCB values for the AET and EP approach indicate that the  EP-based
values are slightly more predictive than the AET values for these two
contaminants.  The demonstrated sensitivity, efficiency and overall
reliability of the EP values in predicting adverse effects to  benthos at 190
stations were 29 percent, 65 percent, and 54 percent, respectively, compared
to a sensitivity, efficiency, and overall reliability of 14 percent,  100
percent, and 52 percent for the benthic AET values.  Similarly, the
sensitivity, efficiency and overall reliability of the EP-based values in
predicting amphipod mortality in laboratory bioassays were 27  percent, 51
percent and 63 percent, respectively, compared to a sensitivity, efficiency
and overall reliability/of 3 percent, 100 percent, and 64 percent for amphipod
AET.  The reduced sensitivity of AET-based sediment quality values  for total
PCBs and phenanthrene, relative to the sensitivity demonstrated when  AET for
all available chemicals were used,  may reflect the fact that  there are a
number of areas in Puget Sound for which other nonionic organic contaminants,
metals or alkylated phenols are in high concentrations in sediments and likely
contributing to site-specific toxicity. .

     The results of the reliability analysis suggest that elements  of both the
AET and EP approaches are useful in making environmental  decisions.  For this
reason, WDOE is proposing that the Puget Sound sediment quality standards
incorporate the best of both methods (i.e., that the standards for  PCBs and
phenanthrene be EP-based and the standards for all other contaminants be
AET-based).  However, the results of the reliability analyses  also  suggest
that available chemical specific sediment criteria alone may  serve  as
incomplete surrogate indicators of biological effects associated with
unmeasured chemicals or the interactive effects of multiple chemicals.  For
this reason, WDOE acknowledges that a combination of biological and chemical
testing will be necessary, in the near term, for effective management of
contaminated sediments.  In addition to EP and AET-based sediment  standards,
WDOE will be requiring confirmatory, site-specific biological  testing as part
of the general sediment quality standards for Puget Sound.

                                  CONCLUSIONS

     Chemical contamination of marine and estuarine sediments  in  developed
coastal areas is a significant problem not only in Puget Sound but  also
throughout the United States and internationally.  The regulatory framework
adopted by EPA, PSWQA and WDOE for sediment standards development and
implementation in Puget Sound will likely be the first such comprehensive
program in the United States and it may well serve as an appropriate model  for
sediment management elsewhere.  The AET and EP approaches, used  as  methods  for
establishing sediment quality standards, also have potential  widespread
application.  The AET and EP-based sediment quality values appear to  be
reasonably predictive in Puget Sound and can be generated  independently  for
other locations.  Supplemented by limited site-specific biological  testing,

-------
AET and EP-based sediment quality values provide a defensible basis for
regulatory decisions concerning the identification, management and cleanup of
contaminated sediments.

     For additional information on the comprehensive sediment management
program being developed by WDOE, please contact Mr. Keith Phillips, Washington
Department of Ecology, Sediment Management Unit, PV-11, Olympia,  Washington
98504.  For additional information on the AET approach, please contact Ms.
Catherine Krueger, U.S. Environmental Protection Agency, Hazardous Waste
Division, HW-113, 1200 Sixth Avenue, Seattle, Washington 98101.  For
additional information on the EP method, please contact Mr. Christopher Zarba,
Environmental Protection Agency, Office of Water Regulations and  Standards,
WH-585, 401 M Street S.W., Washington D.C., 20460.

-------
                                   REFERENCES

Barrick, R., S. Becker,  R.  Pastorok,  P.  Booth,  and  L. Jacobs.   1989.
Contaminated Sediments Criteria Report.   Prepared for the Washington state
Department of Ecology, Sediment Management  Unit by  PTI  Environmental Services,
Bellevue, Washington.

Barrick, R., S. Becker,  R.  Pastorok,  L.  Brown,  and  H. Seller.   1988.  Sediment
Quality Values Refinement:  1988 Update and  Evaluation of Puget Sound AET.
Prepared by PTI Environmental  Services for  Tetra Tech/U.S. Environmental
Protection Agency, Region 10,  Office  of Puget Sound.  PTI Environmental
Services, Bellevue, Washington.

Seller, H., R. Barrick,  and S. Becker.  1986.   Development of Sediment Quality
Values for Puget Sound.   Prepared by  Tetra  Tech,  Inc. for Resource Planning
Associates/U.S. Army Corp of Engineers,  Seattle District, for the Puget Sound
Dredged Disposal Analysis and  the Puget Sound EStuary Program.  Tetra Tech,
Inc., Bellevue, Washington.
                       /
Puget Sound Dredged Disposal Analysis (PSDDA).   1988.   Evaluation Procedures
Technical Appendix.  Prepared  by Tetra Tech,  Inc. for Resource Planning
Associates/U.S. Army Corp of Engineers,  Seattle District, for the Puget Sound
Dredged Disposal Analysis.   Tetra Tech,  Inc., Bellevue, Washington.

Puget Sound Water Quality Authority (PSWQA).  1987.  1987 Puget Sound Water
Quality Management Plan.  Seattle, Washington.

Read, L., M. Jacobson, and R.  Barrick.  1989.   Application of Equilibrium
Partitioning Sediment Quality  Criteria to Puget Sound.   Prepared by PTI
Environmental Services for the Washington Department  of Ecology.  PTI
Environmental Services,  Bellevue, Washington.

U.S. Environmental Protection  Agency  (U.S.  EPA).  1988.  Draft Briefing Report
on the Equilibrium Partitioning Approach to Generating  Sediment Quality
Criteria.  Submitted to the EPA Science Advisory Board  by the U.S.
Environmental Protection Agency, Office of Water Regulation and Standards,
Criteria and Standards Division, Washington D.C.

U.S. Environmental Protection  Agency  (U.S.  EPA).  1989.  Briefing Report to
the EPA Science Advisory Board:  The  Apparent  Effects Threshold Approach.
Prepared by PTI Environmental  Services for Battelle/U.S. Environmental
Protection Agency,  Region 10, Office of Puget  Sound.   PTI Environmental
Services, Bellevue, Washington.

U.S. Environmental Protection  Agency  (U.S.  EPA).  1989B.  Sediment
Classification Methods Compendium (draft final  report).  Prepared by Tetra
Tech, Inc., for the U.S. Environmental Protection Agency, Watershed Protection
Division.  Tetra Tech, Inc., Bellevue, Washington.

-------