vvEPA
          United States
          Environmental Protection
          Agency
           Office of Water
           4301
EPA-820-B-95-005
March 1995
Great Lakes Water
Quality Initiative
Technical Support
Document for the
Procedure to
Determine
Bioaccumulation
Factors

-------
                           DISCLAIMER

This document has been  reviewed by the Health and Ecological
Criteria Division, Office  of  Science and Technology, U.S.
Environmental Protection Agency,  and approved for publication
as a support document for  the Great Lakes Water Quality
Initiative.  Mention of trade names and commercial products
does not constitute endorsement of their use.
                        ACKNOWLEDGEMENTS

Technical support for preparation of this document was
provided to the Office  of Water by Charles E. Stephan,
Lawrence Burkhard, and  Phil  Cook of the Office of Research and
Development, Environmental Research Laboratory, Duluth. MN.
                      AVAILABILITY NOTICE

This document i's available  for a fee upon written request or
telephone call to:

         National  Technical Information Center  (NTIS)
                  U.S. Department  of Commerce
                      5285 Port Royal Road
                     Springfield, VA  22161
                         (800)  553-6847
                        (703)  487-4650
               NTIS Document Number:   PB95187290

                               or

   Education Resources  Information Center/Clearinghouse  for
Science, Mathematics, and Environmental Education  (ERIC/CSMEE)
                  1200 Chambers Road, Room 310
                      Columbus,  OH  43212
                         (800)  276-0462
                         (614)  292-6717
                      ERIC Number:  DO49
                                           U.S. Environmental Protection Agency
                                           Region 5.Library (PL-12J)
                                           77 West Jackson Boulevard, 12th Floor
                                           Chicago, IL  60504-3590

-------
 GREAT LAKES WATER QUALITY INITIATIVE TECHNICAL SUPPORT DOCUMENT
     FOR THE PROCEDURE TO DETERMINE BIOACCUMULATION FACTORS

                         TABLE OF CONTENTS

                                                                  Page

I.   INTRODUCTION	   1
    A.   Purpose and Scope  	   1
    6.   Overview of Bioaccumulation and Bioconcentration  	   1
    C.   Outline of the Methods for Deriving Baseline BAFs	   2
    D.   GLI BAFs	   2
    E.   Definitions  	   3

II.   DATA REQUIREMENTS AND EVALUATION	   5

III.  DETERMINATION OF BAFs FOR ORGANIC CHEMICALS  	   5
    A.   Lipid Content of Fish Consumed By Humans and Wildlife	   5
    B.   Bioavailability   	   8
        1.  Determination of the Fraction of the Chemical that is Freely
           Dissolved in Water	   9
        2.  Derivation of the Equation Defining ffd 	   9
    C.   Bioconcentration and Octanol-Water Partitioning  	  11
    D.   Food-Chain Biomagnification	  13
        1.  Food-Chain Multiplier	  14
           a.  Data  for  the Model  	  14
           b.  Calculation of the FCMs	  16
           c.  Application of  FCMs  	  17
           d.  Evaluation of FCMs	  18
    E.   Prediction of BAFs from Biota-Sediment Accumulation Factor (BSAF)
        Measurements	  47
        1.  Biota-Sediment Accumulation Factors BSAFs	  47
        2.  Relationship of BAFs to BSAFs 	  48
        3.  Calculation of BAFjds from Lake Ontario Data  	  50
        4.  Validity of BAFjds  Calculated from BSAFs	  51
        5.  How to Apply the  BSAF Method for Predicting BAF]ds  	  53
        6.  Summary	  54

    F.   Bioaccumulation Equivalency Factors (BEFs)  	  81

IV.  DETERMINATION OF BAFs FOR INORGANIC CHEMICALS	  87

V.   CALCULATION OF BASELINE BAFs FOR ORGANIC CHEMICALS  	  87
    A.   Baseline BAF from a Field-Measured BAF 	  87

-------
                     TABLE OF CONTENTS (Continued)

                                                                    Page

    B.   Baseline BAF from Field-Measured BSAF Methodology 	  88
    C.   Baseline BAF from a Laboratory-Measured BCF  	  89
    D.   Baseline BAF from a Octanol-Water Partition Coefficient	  89

VI.  CALCULATION OF BASELINE BAFs FOR INORGANIC CHEMICALS	  90

VII. REFERENCES	.91

Appendix A.  Procedure for Deriving Recommended Values for Log Kow	A-1

Appendix B.  Derivation of Recommended Values of Log Kow  . „	B-1

Appendix C.  Derivation of Basic Equations Concerning Bioconcentration and
            Bioaccumulation of Organic Chemicals	C-1

Appendix D.  Derivation of Baseline BAFs from Field-Measured BAFs and
            Laboratory-Measured BCFs	D-1

Appendix E.  Derivation of Baseline BAFs for Mercury	E-1

Appendix F.  Derivation of Baseline BAFs for PCBs	F-1

Appendix G.  Baseline BAFs for Trophic Level Four by Four Methods	G-1

Appendix H.  Recommended Baseline BAFs for Trophic Levels Three and Four  H-1

Appendix I.  Derivation of Consumption Weighted Mean Percent Lipid for
            Human Health and Wildlife 	1-1

Appendix J.  FORTRAN Source Code for the Model of Gobas (1993)  	J-1

Appendix K.  Determination of BAFs for DDT and Metabolites and
            Biomagnification Factors for the Derivation of Wildlife Criteria . . K-1

-------
I.   INTRODUCTION

    A.  Purpose and Scope

The purpose of this document is to provide the technical information and rationale
in support of the methods to determine bioaccumulation factors. Bioaccumulation
factors, together with the quantity of aquatic organisms eaten and the percent
lipid, determine the extent to which people and wildlife are exposed to chemicals
through the consumption of aquatic organisms.  The more bioaccumulative a
pollutant is, the more important the consumption of aquatic organisms becomes as
a potential source of-contaminants to humans and wildlife.

Bioaccumulation factors are needed to determine both human health and wildlife
Tier I water quality criteria and human health Tier II values. Also, they are used to
define Bioaccumulative Chemicals of Concern among  the Great Lakes Initiative
universe of pollutants. Bioaccumulation factors range from less than one to
several million.

    B.  Overview of Bioaccumulation and Bioconcentration

Aquatic organisms in nature absorb and retain some water-borne chemicals in their
tissues at levels greater than the concentrations of these chemicals in the ambient
water.  This  process is bioaccumulation.  Bioaccumulation can be viewed simply as
the result of competing rates of chemical uptake and  depuration.  However,
bioaccumulation is a very dynamic process, affected by the physical and chemical
properties of the chemical, the physiology and biology of the organism,
anvironmental conditions, and the amount and source of the chemical.  When
uptake and depuration are equal, the ratio of the concentration  of the chemical in
the organism's tissue to the concentration of the chemical in the ambient water is
the bioaccumulation factor (BAF). Thus:

                                         CR
                                BAF =  —                           (1)
                                         Cw


where:  CB   =  concentration of chemical in the aquatic biota.
        Cw   =  concentration of chemical in the ambient water.

The CB is expressed on a mass per mass basis and the Cw is expressed in a mass
per volume basis. For example, the CB and Cw may be in mg/kg and mg/L
respectively; the BAF is expressed in L/kg.  Most Cw values available in the current
literature are total concentrations.  BAFs would be more useful  if the Cw is  limited
to that portion of the total concentration that is available to the organism for
uptake.

                                     1

-------
Bioaccumulation refers to uptake by aquatic organisms of a chemical from all
sources such as diet and bottom sediments as well as the ambient water.
Measured BAFs are based on field measurements of concentrations of the chemical
in biota and water.

Bioconcentration refers to uptake of a chemical by aquatic organisms exposed only
from the water. A bioconcentration factor (BCF) is, as is the BAF, the ratio
between the concentration of the chemical in the aquatic biota and the
concentration in the water.  BCFs are measured in laboratory experiments and have
the same units as BAFs. They are determined as follows:

                                          CB
                                BCF =  —2                            (2)
where:  CB   =  concentration of chemical in the aquatic biota.
        Cw   =  concentration of chemical in the water.

Reported BCFs, measured in the laboratory, are not always determined under
steady-state conditions (i.e., conditions under which the concentrations in the
biota and the surrounding water are stable over a period of time).  Only steady-
state BCFs, either measured directly or extrapolated based on the data, are useful
for the determination of BAFs.  The terms BAF and BCF are defined in this
document to be steady-state BAF and steady-state BCF, respectively.

    C.  Outline of the Methods for Deriving Baseline BAFs

Baseline BAFs shall be derived using the following four methods, which are listed
from most preferred to least preferred:

    1.  A measured baseline BAF for an organic or inorganic chemical derived
        from a field study of acceptable quality;

    2.  A predicted baseline BAF for an organic chemical derived using field-
        measured BSAFs of acceptable quality;

    3.  A predicted baseline BAF for an organic or inorganic chemical derived
        from a BCF measured  in a laboratory study of acceptable quality and a
        FCM;

    4.  A predicted baseline BAF for an organic chemical derived from a Kow of
        acceptable quality and a FCM.

    D.  GLI BAFs

-------
The BAFs used by the GLI include the effects of all routes of chemical exposure,
i.e, from water, sediment, and contaminated food, in the aquatic ecosystem.
These BAFs by including all routes of exposure do not assume simple water-fish
partitioning but rather are an overall expression of the total bioaccumulation using
the concentration of'the chemical in water column as a reference point.' These
BAFs do not ignore contaminated sediments.

Field-measured BAFs  and BAFs derived using the BSAF methodology used in the
final  Guidance include all aspects of the environmental behavior of the chemicals
including metabolism, disequilibrium, volatilization, predator-prey relationships, and
include sources of the chemical from both the benthic and pelagic food webs.
BAFs predicted using FCMs include many but not all of the environmental
processes and interactions affecting bioaccumulative chemicals. The most notable
process not accounted for in the predicted BAFs is metabolism and thus, when
metabolism of the chemical is significant, the predicted BAFs will be larger than
field  derived BAFs.  Thus, well field-measured BAFs  are preferred.

The water column and sediment in any ecosystem are interconnected and  in a
subsequent chapter of this document, the interconnectedness between the
sediment and water column concentrations of the chemicals is shown.  This means
that  residues in  fishes can also be predicted equally  well using the concentration of
the chemical in sediment as a reference point.  In the methodology in the final
Guidance, the concentration of the chemical in the water column  has been selected
as the reference point for bioaccumulation.  The second method for deriving a
baseline BAF uses the interconnectedness between the sediments and the water
column to derive BAFs from field-measured BSAFs.

Sediment contamination in the Great Lakes is not localized except for small areas
in tributaries and harbors which are slowly releasing contaminants to the open
water systems.  Most of the  Great Lakes biomass is associated with the open
waters which have concentrations of bioaccumulative chemicals that are strongly
influenced by surface sediments in depositional basins which act as a source to
benthic organisms and lake water through mixing. The BAFs used in the in the
final  Guidance are reflective of the open waters of the  Great Lakes and include the
effects of all routes of chemical exposure including contaminated the sediments.

    E.  Definitions

Baseline BAF (BAF*d). For organic chemicals, a BAF  that is based on the
concentration of the freely dissolved chemical in the ambient water and takes into
account the partitioning of the chemical within the organism; for inorganic
chemicals, a BAF that is based on the wet weight of the tissue.

Baseline BCF (BCF*d).  For organic chemicals, a BCF  that is based on the

-------
concentration of the freely dissolved chemical in the ambient water and takes into
account the partitioning of the chemical within the organism; for inorganic
chemicals, a BCF that is based on the wet weight of the tissue.

Bioaccumulation. The net accumulation of a substance by an organism as a result
of uptake from all environmental sources.

Bioaccumulation factor (BAF).  The ratio  (in L/kg) of a substance's concentration in
tissue of an aquatic organism to its concentration in the ambient water, in
situations where both the organism and its food are exposed and the ratio does not
change substantially over time.

Bioconcentration. The net accumulation of a substance by an aquatic organism as
a result of uptake directly from the ambient water, through gill membranes or other
external body surfaces.

Bioconcentration factor (BCF). The ratio (in L/kg) of a substance's concentration in
tissue of an aquatic organism to its concentration in the ambient water, in
situations where the organism is exposed through the water only and the ratio
does not change substantially over time.

Biomagnification. The increase in tissue concentration of  poorly depurated
materials in organisms along a series of predator-prey associations, primarily
through the mechanism of dietary accumulation.

Biota-sediment accumulation factor (BSAF). The ratio (in  kg of organic carbon/kg
of lipid) of a substance's lipid-normalized concentration in  tissue of an aquatic
organism to its organic carbon-normalized concentration in the surface sediment, in
situations where the ratio does not change substantially over time, both the
organism and its food are exposed, and the surface sediment is representative of
average surface sediment in the vicinity of the organism.

Depuration. The loss of a substance from an  organism as a result of any active or
passive process.

Food-chain multiplier (FCM). The  ratio of a BAF to an appropriate BCF.

Octanol-water partition coefficient (Kow).  The ratio of the concentration of a
substance in the n-octanol phase to its concentration in the aqueous phase in an
equilibrated two-phase octanol-water system. For log Kow, the log of  the octanol-
water partition coefficient is a base 10 logarithm.

Uptake.  Acquisition by an organism of a substance from  the environment as a
result of any active or passive process.

-------
II.  DATA REQUIREMENTS AND EVALUATION

Data used to calculate BAFs, BSAFs, and BCFs are obtained from EPA criteria
documents, published papers, and other reliable sources. Data should be screened
for acceptability using the criteria in The U.S. Environmental Protection Agency
(EPA) guidelines for deriving aquatic life criteria (Stephan et al. 1985), and
American Society for Testing and Materials guidance (practice E  1022-84) detailing
methods for conducting a flow-through bioconcentration test (ASTM 1990).

In general, the Great Lakes Water Quality Initiative (GLWQI)  BAF methods follow
closely the EPA guidance (Stephan et al. 1985) with the addition of the BSAF
methodology and the Food-Chain Multiplier (FCM) when a predicted BAF is
calculated from a laboratory-measured or predicted BCF. The EPA published draft
guidance on the control of bioaccumulative pollutants in surface  waters which
recommends the  use of FCMs (USEPA 1991 A).

No guidance can  cover all the variations of experimental design and data
presentation found in the literature concerning BAFs, BSAFs, BCFs and Kows.
Professional judgment is needed throughout the BAF development process to
select the best available information and use it appropriately.

III.  DETERMINATION OF BAFs FOR  ORGANIC CHEMICALS

    A.  Lipid Content of Fish Consumed By Humans and Wildlife

An important determinant of bioconcentration of non-polar organic chemicals in
aquatic organisms is lipid content of the organism (see Barron, 1990 and the
references cited by Barron, 1990). In the classic study by Reinert (1970), lipid
normalization of DDT residues in fishes caused the differences between species
and differences between size groups to  become considerably less.  It is now
generally accepted that lipid normalization  of chemical residues is essential in
understanding and predicting the bioconcentration and bioaccumulation of
bioaccumulative chemicals in aquatic organisms (Barron, 1990).  Lipid
normalization is now part of the EPA guidance on bioaccumulation (Stephan et al.
1985, USEPA 1991 A), and is included in the BAF procedure in the final Guidance.

BAFs and BCFs are lipid-normalized by dividing the BAFs or BCFs by the fraction
lipid of the tissue.  Because  BAFs and BCFs for organic chemicals are lipid-
normalized, it does not make any difference whether the tissue sample is whole
body or edible portion, but both the BAF (or BCF) and the percent lipid must be
determined for the same tissue.  The percent lipid of the tissue should be measured
during the BAF or BCF study, but in some cases it can be reliably estimated from
measurements on tissue from other organisms. If percent lipid is not reported for
the test organisms in the original study, it may be obtained from  the author; or, in

-------
the case of a laboratory study, lipid data for the same or a comparable laboratory
population of test organisms that were used in the original study may be used.

A lipid-normalized BAF, of a chemical in tissue shall be calculated using the
following equation:

                              BAF, =  — -—£                          (3)
where:  BAF,    =   lipid-normalized BAF.
        BAFT    =   BAF based on the total concentration of the organic
                     chemical in the tissue of biota (either whole organism or
                     specified tissue) (//g/g).
        f,       =   fraction of the tissue that is lipid.

When deriving water quality criteria for human health and wildlife it is important to
accurately characterize the potential exposure to a chemical.  To do this,
information is needed on several parameters including the quantity of aquatic biota
consumed by humans and wildlife, the percent lipid  in the aquatic biota, the trophic
level of the aquatic biota and the BAF for that  chemical. The quantity of aquatic
biota consumed can be estimated using consumption surveys for humans and,
where available, studies on the feeding habits of wildlife. To estimate BAFs that
can be used in deriving human health and wildlife criteria, a standard percent lipid
value is needed for both humans and wildlife.  The standard  percent lipid value
used in the BAF derivation should, if possible, be a consumption- weighted percent
lipid value. A consumption-weighted percent lipid value is preferred because it
provides a more accurate characterization of the potential exposure to humans and
wildlife than simply assuming humans and wildlife consume all or a subset of the
species within the area of concern (in this case the area of concern is the Great
Lakes Basin). To  estimate a consumption-weighted percent lipid value for humans
and wildlife the following information is needed:  (1) a consumption survey that
documents the type and quantity of aquatic biota consumed by humans and
wildlife; (2) the percent lipid of the aquatic biota consumed by humans and
wildlife; and (3) the trophic level of the aquatic biota consumed by humans and
wildlife.

A consumption survey that documents the type and quantity of aquatic biota
consumed by humans and/or wildlife in conjunction with the percent lipid values
for those  species  will assist in accurately characterizing the potential exposure to
humans and wildlife from consumption of contaminated aquatic biota.  EPA has
published the document "Consumption Surveys for  Fish and  Shellfish.  A Review
and Analysis of Survey Methods" (Feb. 1992,  EPA 822/R-92-001)  which may
assist in conducting and analyzing the results of such surveys.

-------
The second critical piece of information is the percent lipid values of aquatic biota
consumed by humans and/or wildlife. The lipid values used for deriving human
health BAFs should be from aquatic biota collected from the Great Lakes or their
tributaries and be from the edible tissue (e.g., muscle). For wildlife, whole body
lipid data should be used.  Data on the edible tissue is available from the
contaminant monitoring  programs in the various Great Lakes States.  Whole body
lipid data are also available from the State monitoring programs, but is not as
abundant.

Finally, the trophic level  of the biota consumed should be determined. This is
important when attempting to accurately characterize the potential exposure to
humans and wildlife because humans and wildlife consume both trophic level 3 and
trophic level 4 fish and the BAFs for trophic level 3 and trophic level 4 are different
for many pollutants.  If it is assumed that humans consume only trophic level 4
species, then the trophic level 4 BAFs used for deriving human health criteria could
be overestimated or underestimated.  The determination of the appropriate trophic
level for a fish species will depend on the size and age of the fish  being consumed.
Some fish are in trophic  level 3 when young, but in trophic level 4 as adults. Data
on the size and age of fish consumed by humans and/or wildlife will, in most
cases, not be included in a consumption survey.  In these situations, best
professional judgment will need to be exercised when determining the appropriate
trophic level for a fish species.

For the Great Lakes Water Quality Initiative a consumption survey by West et al.
(1993) was used to characterize the consumption patterns of sport anglers in the
Great Lakes Basin (Table 5 of Appendix I).  This study was selected because it
represented the largest consumption survey of sport  anglers  in the Great Lakes
Basin. In addition, it was possible to determine the type and quantity of each
species consumed.

Percent lipid data from the fish contaminant monitoring programs in  Michigan,
Wisconsin, Ohio,  Indiana, New York and Minnesota provided lipid data for edible
tissues (e.g., muscle) of fish from each of the Great Lakes (Tables 1-3 of Appendix
I). Most lipid data are for skin-on fillets because skin-on fillets are the accepted
tissue sample used by most of the Great Lakes fish consumption advisory
programs.

The report "Trophic Level and Exposure Analyses for Selected Piscivorous Birds
and Mammals" (EPA, 1995) was used along with professional judgement to
determine the trophic level of the fish species consumed by the sport anglers.
Each consumed fish species was assigned to either trophic level 3 or trophic level
4 based on data from the report and/or professional judgement.

The data from the West  survey (1993) in conjunction with the data from the fish

-------
monitoring programs and the report on trophic levels of various fish species were
used to determine consumption weighted mean percent lipid values for use in
deriving human health BAFs. The total grams per day of each species consumed
by sport anglers was multiplied by the percent lipid value for that species to
determine the grams, of lipid consumed per day by sport anglers for that species.
The grams of lipid consumed from all species were summed and divided by the
total grams of fish consumed from trophic level 3 and trophic level 4 fish to arrive
at a consumption weighted mean percent lipid value for each trophic level. These
percent lipid values are used to derive BAFs  which are then utilized in calculating
human health criteria.  The mean values for  use in deriving human health BAFs are
1.82 for trophic level 3 fish consumed and 3.10 for trophic level 4 fish consumed
(Table  6 of Appendix I). The values were not rounded to whole numbers because
they are intermediate values that are used in the derivation of human health
criteria.

For wildlife,  an  analysis of the most common prey species consumed by the five
representative wildlife species used to derive wildlife criteria was conducted. The
data allowed only a  gross determination of the type of species consumed by the
five representative species and the percent of prey species consumed from each
trophic level. The analysis did not allow a quantitative determination of the
quantity of the  prey species consumed at each trophic level. Consequently, a
consumption weighted percent lipid value similar to that derived for humans was
not possible. Nonetheless, a percent lipid value for both trophic level 3 and trophic
level 4 were estimated using whole fish lipid data from the U.S. Fish and Wildlife
Service national contaminant biomonitoring program, the Canada Department of
Fisheries and Oceans, the  New York Department of Environmental Conservation,
arid the Michigan Department of Natural Resources (Table 4 of Appendix I). The
trophic levels of the species consumed were determined using the data from the
report  "Trophic Level and Exposure Analyses for Selected Piscivorous Birds and
Mammals" (EPA, 1995).  The mean percent lipid values for wildlife for use in
deriving wildlife BAFs  are 6.46 for trophic level 3 prey species consumed and
10.31  for trophic level 4 prey species consumed (Table 7 of Appendix I).  The
values were not rounded to whole numbers  because they are intermediate values
that are used in the  derivation of wildlife criteria.

     B.   Bioavailability

Baseline BAFs and BCFs for organic chemicals, whether measured or predicted,
shall be based on the  concentration of the chemical that is freely dissolved in the
ambient water in order to account for bioavailability. For the purposes of this
guidance, the relationship  between the total concentration of the chemical in the
water  (i.e., that which is freely dissolved plus that which is sorbed to paniculate
organic carbon  or to .dissolved organic carbon) to the freely dissolved concentration
of the  chemical in the ambient water shall be calculated using the following


                                      8

-------
equation:
                             f d
                                =  I f
                                   V -L
                                      f d
where:  C™          —  freely dissolved concentration of the organic chemical in
                        the ambient water.
        C^      =  total concentration of the organic chemical in the ambient
                    water.
        ffd       =  fraction of the total chemical in the ambient water that is
                    freely dissolved.

    1 .  Determination of the Fraction of the Chemical that is Freely Dissolved in
        Water

The fraction of the chemical that is freely  dissolved in the water, ffd, can be
determined using the following equation with the Kow for the chemical and the
DOC and POC of the water.
             f £d  =
where:  DOC     =  concentration of dissolved organic carbon, kg of organic
                    carbon/L of water.
        Kow     =  octanol-water partition coefficient of the chemical.
        POC     = . concentration of paniculate organic carbon, kg  of organic
                    carbon/L of water.

    2.  Derivation of the Equation  Defining fw

Experimental investigations have shown that hydrophobic organic chemicals exist
in water in three phases, 1) the freely dissolved phase, 2) sorbed to suspended
solids and 3) sorbed to dissolved organic matter (Hassett and Anderson (1979),
Carter and Suffet (1982), Landrum  et al. (1984), Gschwend and Wu (1985),
McCarthy and Jimenez (1985), Eadie et al. (1990,  1992)). The total  concentration
of the chemical in water is the sum of the concentrations of the sorbed chemical
and the freely dissolved chemical (Gschwend and Wu  (1985) and Cook et al.
(1993)):

-------
                                                                       (6)

where:  Cj,d      =  concentration of freely dissolved chemical in the ambient
                    water, kg of chemical/L of water.
        CJJ,       =  total concentration of the chemical in the ambient water, kg
                   ' of chemical/L of water.
        Cpoc      =  concentration of chemical sorbed to the paniculate organic
                    carbon, in the ambient water, kg of chemical/kg of organic
                    carbon.
        Cdoc      =  concentration of chemical sorbed to the dissolved organic
                    carbon in the ambient water, kg of chemical/kg of organic
                    carbon.
        POC     = . concentration of paniculate organic carbon in the ambient
                    water, kg of organic carbon/L of water.
        DOC     =  concentration of dissolved organic carbon in the ambient
                    water, kg of organic carbon/L of water.

The above equation can also be expressed using partitioning relationships as:

            Cwfc  = Cwfd •  (1 + POC • KpOC + DOC • Kdoc )        (7)

where:
        KpOC      =  equilibrium partition coefficient of the chemical between
                    POC and the freely dissolved phase in the ambient water
        Kdoc      =  equilibrium partition coefficient of the chemical between
                    DOC and the freely dissolved phase in the ambient water.

From equation 7, the fraction of the chemical which is freely dissolved in the water
can be calculated using the following equations:

                                          fd
            ffd    1 +  (DOC)  (Kdoc)  +  (POC)  (KpOC)        (1°}
                                     10

-------
Experimental investigations by Eadie et al. (1990, 1992), Landrum et al. (1984),
Yin and Hassett (1986, 1989), Chin and Gschwend (1992), and Herbert et al.
(1993) have shown that Kdoc is directly proportional to the Kow of the chemical and
is less than the Kow.  The Kdoc can be estimated using the following equation:
                               IT    _   "OW                           /ill
                               K ,   ss  	                           VJ-xj
                                 doc     10

The above equation is based upon the results of Yin and Hassett (1986, 1989),
Chin and Gschwend (1992), and Herbert et al. (1993).  These investigations were
done using unbiased methods, such as the dynamic headspace gas-partitioning
(sparging) and the fluorescence methods, for determining the Kdoc.

Experimental investigations by Eadie at al. (1990, 1992) and Dean et al. (1993)
have shown that Kpoc is approximately equal to the Kow of the chemical. The Kpoc
can be estimated using the following equation:
                                    ,  »  Kow                           (12)

By substituting equations 11 and 12 into equation 10 , the following equation is
obtained:
                                         -i
             ffd  =
(DOC) (Kow)
10
- + ( POP } ( K" }
^ \ C\J^, ) \ i^QW /
(13)
    C.  Bioconcentration and Octanol-Water Partitioning

Numerous investigations have demonstrated a linear relationship between the
logarithm of the bioconcentration factor (BCF) and the logarithm of the octanol-
water partition coefficient (Kow) for organic chemicals for fish and  other aquatic
organisms. Isnard and Lambert (1988) listed various regression equations that
illustrate this linear relationship. The underlying assumption for the linear
relationship between' the BCF and Kow is that the bioconcentration process can be
viewed as a partitioning of a chemical between the lipids of the aquatic organisms
and water and that the Kow is an  useful surrogate for this partitioning process
(Mackay (1982)).

The regression equations demonstrating the linear relationship between the
logarithms of the BCF and Kow have been developed using organic chemicals which
are slowly, if at all, metabolized by fishes or other aquatic organisms. For
metabolizable chemicals, the regression equations developed between BCF and Kow


                                     11

-------
for non-metabolizable chemicals in most cases predict BCFs which are larger than
the laboratory-measured BCFs. The losses of the chemicals due to metabolism are
not accounted for in the simple partitioning model (Baron (1990), de Wolf et al.
(1992)).

Mackay (1982) presented a thermodynamic basis for the partitioning process for
bioconcentration and in essence, the BCF on a lipid-normalized basis (and freely
dissolved concentration of the chemical in the water) should be similar if not equal
to the Kow for organic chemicals. Unfortunately, almost all of the reported
regression equations have used BCFs reported on a wet weight basis instead of
lipid-normalized. When regression equations are constructed using BCFs reported
on a lipid-normalized'basis, regression equations are obtained which have slopes
and intercepts which are not significantly different from one and zero, respectively.
For example, de Wolf et al.  (1992) adjusted the relationship reported by Mackay
(1982) to a  100 percent lipid basis (lipid normalized basis) and obtained the
following relationship:

                  log BCF = 1.00 log Kow -  0.08            (14)

For chemicals with large log Kows (i.e., greater than 6.0), reported BCFs are often
not equal to the Kow for non-metabolizable chemicals. As discussed by Gobas et
al. (1989), this non-equality between the  BCF and Kow is not caused by a
breakdown of the BCF-KOW relationship but rather is caused by (1) not accounting
for growth dilution which occurred during the BCF determination, (2) using the
total concentration of the chemical in the water instead  of the  bioavailable (freely
dissolved) concentration of the chemical in calculating the BCF, (3) not allowing
sufficient time in the' exposure to achieve steady-state conditions, and (4) not
correcting for elimination of the chemical  into the feces.  BCFs for non-
metabolizable chemicals are equal to the Kow when the BCFs are reported on lipid-
normalized basis, determined using the freely dissolved concentration of the
chemical in the exposure water, corrected for growth dilution,  determined from
steady-state conditions or determined from accurate measurements of the
chemical's uptake (k,) and elimination (k2) rate constants from  and to the water,
respectively, and determined using no solvent carriers in the exposure.

In the final Guidance, predicted BCFs are estimated using the following
approximation:

                               BCF/d *  Kow                         (15)

where:  BCF)d =     BCF reported on lipid-normalized basis using the freely
                   • dissolved concentration of the chemical  in the water.

This relationship is applicable to organic chemicals which are either slowly  or not


                                     12

-------
metabolized by aquatic organisms and have Kows greater than a 1000. For
chemical with Kows less than a 1000, a slightly different relationship is applicable
for organic chemicals because the portion of the chemical in the organism that is
not associated with lipid becomes significant relative to the associated with the
lipid.  Appendix C contains a complete derivation of this relationship.

Equation 15 implicitly assumes that n-octanol is an appropriate surrogate for lipids
in aquatic organisms. If n-octanol is not an appropriate surrogate for lipids, the
slope and intercept of equation 14 will not be 1.0 and 0.0, respectively.  The
theoretical basis and the experimental data presented by Mackay (1982) suggest
that n-octanol is a  very reasonable surrogate for lipids.

Equation 15 is also supported by and consistent with the food-chain model of
Gobas (1993). For the Gobas model, the BCF]d is equal to Kow when  the growth
rate of the organisms and metabolism rate of the chemical by the organisms are
set equal to zero.  It should be noted that the model does not use the partitioning
process described by Mackay (1982) for bioconcentration. Instead the food-chain
model predicts the k1 and k2 rate constants for the fishes and the bioconcentration
factor is determined by dividing the uptake rate constant from water (k,) by the
elimination rate constant to water (k2).

The above equation is also supported by and consistent with the equilibrium
partitioning theory  being developed by EPA for the derivation of sediment quality
criteria (Di Toro et  al. 1991). Both the sediment organic carbon-water equilibrium
partition coefficient (/c/g of chemical/Kg of organic  carbon in the sediment)/(//g of
freely dissolved chemical/L of sediment pore water)  (K80C or Koc) and the lipid/water
equilibrium partition coefficient U/g of chemical/Kg of \\pld)/(JJQ of freely dissolved
chemical/L of sediment pore water) (KL) have been demonstrated to be
approximately equal to  Kow for organic chemicals in  sediments and benthic
organisms, respectively.

    D.  Food-Chain .Biomagn'rf ication

The importance of  uptake of chemicals through the diet and the potential for a
stepwise increase in bioaccumulation from one trophic level to the next in natural
systems has been recognized for many years (Hamelink et al. 1971).  This
pathway,  involving transfer of a chemical in food through successive trophic levels,
is called biomagnification.  Many researchers have noted that the BAFs of some
chemicals in nature exceed the bioconcentration factors measured in the laboratory
or estimated by log Kow models (e.g., Oliver and Niimi 1983,  Oliver and Niimi
1988, Niimi 1985,  Swackhammer and Hites 1988).  Chemicals exhibiting  this
phenomenon are typically highly lipophilic, have low water solubilities, and are
resistant to being metabolized by aquatic organisms (Metcalf et al. 1975).
                                     13

-------
        1 .  Food-Chain Multiplier

FCMs for organic chemicals were determined using the model of Gobas (1993).
This model includes both benthic and pelagic food chains thereby incorporating
exposures of organisms to chemicals from both the sediment and the water
column. With the model of  Gobas (1993), disequilibrium between the
concentrations of the chemicals in sediment and the water column are included in
the predicted BAFs and the  FCM derived from the predicted BAFs. The
disequilibrium is accounted for by inputting the concentrations of the chemical in
the sediment and water column to the model.  Subsequently, the disequilibrium is
incorporated into the pelagic and benthic food web pathways because the model
predicts the chemical residues in benthic invertebrates by using equilibrium
partitioning and in zooplankton by assuming  that the BCF for zooplankton is equal
to the Kow of chemical after correction for lipid content.  Chemical residues for all
other organisms (e.g., fishes) are determined from the rates of (1) chemical uptake
from food and water, (2) depuration and excretion of the chemical, (3)  dilution due
to growth of the organism, and (4) metabolism. This model requires the
specification of the food chain structure, feeding preferences, temperature of the
ecosystem, organic carbon content of the sediments, organism weights and lipid
contents, and the rate of metabolism of the chemical. Because rates of
metabolism for bioaccumulative chemicals are not known, the rate of metabolism
used in determining the FCMs was zero (i.e., no metabolism).

The model of Gobas (1993) does not predict FCMs but rather it predicts the  BAF
for each species in the food chain. FCMs can be calculated from the predicted
BAFs using the following equation:
                             FCM =                                 (16)
where:   Kow     =  octanol-water partition coefficient.
         BAF{d   =  BAF reported on a lipid-normalized basis using the freely
                    dissolved concentration of the chemical in water.

        a.  Data for the Model

The data of Oliver and Niimi (1988) and Flint (1986) for Lake Ontario were used
for the feeding preferences, weights, and lipid contents for each species in the
food chain (Table 1). The mean water temperature of Lake Ontario was set to 8°C
and the organic carbon content of sediment  was set to 2.7% as reported by Oliver
and Niimi  (1988) (Table 1).  Values for the densities of the lipid and organic carbon
were taken from Gobas (1993) (Table 1). The metabolic transformation rate
constant was set equal to zero.  The organic carbon content of the water column


                                     14

-------
was set to 0.0 kg/L (see b. Calculation of the FCMs).

With the values specified in Table 1, the remaining data needed for the model of
Gobas (1993) are the concentrations of the chemical in the sediment and water
column, and the Kow of the chemical.  The Kow of the chemical is used as the
independent variable in deriving the FCMs and thus only the two chemical
concentrations need to be defined for the model.

To determine the relationship between the total concentration of the chemical in
the sediment and the freely dissolved concentration of the chemical in the water
column, the following sediment-water  column chemical concentration quotient
(n.oc)  was calculated for each chemical reported by Oliver and Niimi (1988):
 H    =   ng of total chemical/Kg of organic carbon (in sediment)
  soc   ng of  freely  dissolved chemical/L of water  (in water column]
The freely dissolved concentrations of the chemicals in the water column were
calculated from the data of Oliver and Niimi (1988) using the equations of
Gschwend and Wu (1985) and Cook et al.  (1993). These equations are:
                    1  +  (DOC) (Kdoc)  +  (POC)  (KpOC)
where:  ffd      =  fraction of the chemical which is freely dissolved in the
                    water;
        DOC    =  concentration of dissolved organic carbon;
        Kdoc     =  partition coefficient for the chemical between the DOC and
                  •  the freely dissolved phase in the water;
        POC    =  concentration of particulate organic carbon;
        KPOC     =  partition coefficient for the chemical between the POC phase
                    and the freely dissolved phase in the water;
        C^d     =  freely dissolved concentration of the chemical in the water;
        CJJ,      =  total concentration of the chemical in the water.

The concentrations in the water reported by Oliver and Niimi (1988) were obtained
by liquid-liquid extraction of aliquots of Lake Ontario water which had passed
through a continuous-flow centrifuge to remove POC. Therefore, the
concentrations in the water reported by Oliver and  Niimi (1988) include both the
freely dissolved chemical and the chemical associated with the DOC in the water


                                    15

-------
sample. The above equations were used to derive the freely dissolved
concentrations of the chemicals in the water by setting the POC  = 0.0 mg/L, DOC
= 2 mg/L, and Kdoc = Kow/10. Kows used to derive the freely dissolved
concentrations are  listed in Appendix B of this document.  The relationship for
determining  Kdoc from Kow was developed from the results reported by Yin and
Hassett (1986, 1989), Eadie et al. (1990, 1992), Landrum et al.  (1984), and
Herbert et al. (1993) for partitioning to DOC.

In Figure 1, the ratios of n.oow to Kow are plotted against the log Kow for each
chemical reported by Oliver and Niimi  (1988).  Visual inspection of Figure 1
suggest that the  ratio of naoow to Kow is not strongly dependent upon the Kow.
Correlation coefficients of the ratio (of l~l.ocw to Kow) against log Kow of 0.02, -0.34,
and -0.55 were obtained for the pesticides, PCB congeners, and the group of
chemicals consisting of the chlorinated benzenes, toluenes, and butadienes,
respectively. The average (standard deviation & number of values) ratios for the
n.oow to KOW f°r pesticides, PCB congeners, pesticides and PCBs combined, and the
group of chemicals consisting of the chlorinated benzenes, toluenes, and
butadienes were  11.8 (8.4 & 9), 25.9 (26.8 & 46), 23.6 (25.3 & 55), and 294
(1188  & 12), respectively.

Based upon  the independence of the ratios of  n.ocw to Kow on Kow for the
pesticides and PCBs (the chemicals of primary concern  in the derivation of food
chain multipliers), a value of 25 was selected for this ratio, the average of the
pesticides and PCBs combined.  The resulting  relationship between the
concentration of  the chemical in the sediment on an organic carbon basis (C80C) and
the freely dissolved concentration  of the  chemical in the water column (Cwd) is:

                                = 25 • K   • Cfd                    (19)
                                  ^D   ^     u
        b.  Calculation of the FCMs

The model of Gobas (1993) (MS-DOS version) was used to determine the FCMs.
A listing of the source code in FORTRAN is provided in Appendix J for the food
web model of Gobas (1993).

The model was run using the data listed in Table 1 with the above relationship
(equation  19) between the C80C and Cwd for Kows 3.5, 3.6, 3.7, 3.8, ..., and 9.0.
The freely dissolved concentration of the chemical in the  water was set to 1 ng/L
and the concentration of the chemical in the sediment was calculated using the
above sediment-water concentration relationship. The model of Gobas (1993)
does not include solubility controls or limitations, and thus, the concentration of
the chemical in the water used with the model is arbitrary for determining the BAFs
(i.e., the  BAF obtained using a 1 ng/L concentration of the chemical will be  equal


                                     16

-------
to that obtained using a 150 //g/L concentration of the chemical for a specified
Kow).

In using the model of Gobas (1993), we have not used his method for accounting
for bioavailability.  In section B of chapter III in this document, the procedure for
determining the freely dissolved concentration of the chemical in the ambient water
is presented. To not use or override the method of Gobas for accounting for
bioavailability, we have set the concentration of the DOC in the model to an
extremely small number, 1.0e-30 L/L.  The model of  Gobas (1993) takes the
inputted total concentration of the chemical in the water and before doing any
predictions, corrects for bioavailability by calculating  the freely dissolved
concentration of the chemical in the water.  The freely dissolved concentration of
the chemical in the water is then used in all subsequent calculations by the model.
By setting the concentration of the  DOC to 1 .Oe-30 L/L, the total concentration of
the chemical inputted to the model  becomes equal to the freely dissolved
concentration of the'chemical in the water because the correction for bioavailability
using the bioavailability method of Gobas is extremely small.

For each value of  Kow inputted to the model,  BAFjds are reported by the  model  for
each organism in the food  web.  Using equation 16, FCMs were calculated for
each organism using the reported BAF]ds.  Listed in Table 3 are the FCMs for
trophic level 2 (zooplankton), trophic level 3 (forage fish), and trophic level 4
(piscivorous fish).  The FCMs for the forage fish, trophic level 3, were determined
by taking the geometric mean of the FCMs for sculpin and alewife. The  FCMs for
the smelt were not used in determining the mean  FCMs for the forage fish because
the diet of this organism includes small sculpin. This diet causes smelt to be at a
trophic level slightly higher than 3 but less than trophic level 4. In contrast, the
diets of the sculpin and alewife were solely trophic level 2 organisms (i.e.,
zooplankton and Diporeia sp).

        c.  Application of FCMs

In the absence of  a field-measured BAF or a predicted BAF derived from  the BSAF
methodology, a FCM shall  be used to calculate the baseline BAF for trophic levels
3 and 4 from a laboratory-measured or predicted BCF.  For an organic chemical,
the FCM used shall be derived from Table 3 using the chemical's log Kow and linear
interpolation. A FCM greater than 1.0 applies to most organic chemicals with a log
Kow of four or more. The trophic level used shall take into account the age or size
of the fish species consumed by the human, avian or mammalian predator because,
for some species of fish, the young are in trophic level 3 whereas the adults are in
trophic level 4.

The FCMs were developed assuming no metabolism of the chemical by any of the
organisms in the food web. Thus, for chemicals where metabolism is significant,

                                     17

-------
the predicted BAFs will be larger than a field-measured BAF or BAF determined
using the BSAF methodology. BAFs predicted using laboratory-measured BCFs
(i.e., the product of the FCM and the laboratory-measured BCF), might be in closer
agreement with the field derived BAFs than the BAFs predicted using predicted
BCFs because laboratory-measured BCFs might include some metabolism in their
determination.  In general, for highly persistent chemicals, the effects of all
metabolic processes can not be easily included in the BCF determination.

The FCMs were determined using a disequilibrium factor of 25 from Kow (equation
16) between the concentrations of the chemical in the sediment on an organic
carbon normalized basis and the freely dissolved concentration of the chemical in
water column.  This disequilibrium is incorporated into the pelagic and benthic food
web pathways in the model of Gobas (1993) and is subsequently reflected in the
BAFs predicted by the model and the resulting FCMs.

        d.   Evaluation of FCMs

Baseline BAFs were predicted using the model of Gobas (1993) for each chemical
reported by Oliver and Niimi (1988).  The predicted BAFs are equal to the product
of the Kow and the FCM determined for that organism.  Baseline BAFs also were
derived from the data of Oliver and Niimi (1988) by dividing the lipid-normalized
concentration of the chemical in the  fish  by the freely dissolved concentration of
the chemical in the water column. The freely dissolved concentration of the
chemical in the water was determined as described above. These results are
summarized in Tables 3 through 8 and Figures 2 through 7.

Measured chemical residues in fishes assigned to trophic level 3 can be higher than
those in trophic level 4 from the same food chain. Potential causes of the higher
concentrations  (on a lipid basis) in the trophic level 3 fish include (1) growth rates
which are much slower than the predator fishes, and (2) differing rates of
depuration and  elimination of the chemical by the predator fishes.

The average differences between the predicted and measured log BAFs were -
0.61, 0.01, -0.17, -0.04, -0.10, and -0.12 for zooplankton, sculpin, alewives,
small smelt, large smelt, and piscivorous fish, respectively.
                                     18

-------
Table 1. Environmental Parameters and Species Characteristics Used with the
        Model of Gobas (1993) for Deriving the Food Chain Multipliers
    Environmental parameters:
            Mean water temperature:  8°C
            Organic carbon content of the sediment: 2.7%
            Organic carbon content of the water column:  1.0e-30 kg/L
            Density of lipids:  0.9 kg/L
            Density of organic carbon: 0.9 kg/L
            Metabolic transformation rate constant: 0.0 day'1

    Species characteristics:
    Phytoplankton
            Lipid content: 0.5%

    Zooplankton: Mysids (Mysis relicta)
            Lipid content:  5.0%

    Diporeia sp.
            Lipid content:  3.0%

    Sculpin  (Cottus cognatus)
            Lipid content:  8.0%
            Weight:  5.4 g
            Diet:  18% zooplankton, 82% Diporeia sp.

    Alewives  (Alosa pseudoharengus)
            Lipid content:  7.0%
            Weight:  32 g
            Diet:  60% zooplankton, 40% Diporeia sp.

    Smelt (Osmerus mordax)
            Lipid content:  4.0%
            Weight:  16 g
            Diet:  54% zooplankton, 21 % Diporeia sp., 25% sculpin

    Salmonids  (Salvelinus namaycush, Oncorhynchus mykiss, Oncorhynchus
    velinus namaycush)
            Lipid content:  11.0%
            Weight:  2410 g
            Diet:  10% sculpin,  50% alewives, 40% smelt
                                     19

-------
Table 2.  Food-Chain Multipliers for Trophic Levels 2, 3 & 4.
Log Kow
2.0
2.5
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.0
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5.0
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
6.0
6.1
6.2
6.3
6.4
6.5
Trophic
Level 2
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Trophic"
Level 3
1.005
1.010
1.028
1.034
1.042
1.053
1.067
1.083
1.103
1.128
1.161
1.202
1.253
1.315
1.380
1.491
1.614
1.766
1.950
2.175
2.4G2
2.780
3.181
3.643
4.188
4.803
5.502
6.266
7.096
7.962
8.841
9.716
10.556
11.337
12.064
12.691
13.228
13.662
Trophic
Level 4
1.000
1.002
1.007
1.007
1.009
1.012
1.014
1.019
1.023
1.033
1.042
1.054
1.072
1.096
1.130
1.178
1.242
1.334
1.459
1.633
1.871
2.193
2.612
3.162
3.873
4.742
5.821
7.079
8.551
10.209
12.050
13.964
15.996
17.783
19.907
21.677
23.281
24.604
                                       20

-------
Table 2.  Continued.
Log Kow
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
9.0
Trophic
Level 2
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Trophic
Level 3
13.980
14.223
14.355
14.388
14.305
14.142
13.852
13.474
12.987
12.517
11.708
10.914
10.069
9.162
8.222
7.278
6.361
5.489
4.683
3.949
3.296
2.732
2.246
1.837
1.493
Trophic
Level 4
25.645
26.363
26.669
26.669
26.242
25.468
24.322
22.856
21.038
18.967
16.749
14.388
12.050
9.840
7.798
6.012
4.519
3.311
2.371
1.663
1.146
0.778
0.521
0.345
0.226
8 The FCMs for trophic level 3 are the geometric mean of the FCMs for sculpin and
alewife.
                                    21

-------
Table 3.  Measured and Predicted BAFs for Zooplankton.  BAFs are reported on a
lipid weight basis using the freely dissolved concentration of the chemical in water
(i.e., (jjg of chemical/Kg of lipid)/(//g of freely dissolved chemical/L of water)).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Chemical"
ppDDT
ppDDE
ppDDD
mirex
photomirex
g-chlordane
alpha-BHC
gamma-BHC
HCBD
OCS
HCB
QCB
1,2,3,5/TeCB
1,2,4,5-TeCB
1,2,3,4-TeCB
1,3,5-TCB
1 ,2,4-TCB
1,2,3-TCB
2,4,5-TCT
2,3,6-TCT
PCT '
8
6
5
12
13
28 + 31
18
22
26
16
33
17
25
24 + 27
32
66
Log Kow
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
5.44
6.20
Predicted15
Log BAF
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
5.44
6.20
Measured0
Log BAF
6.95
7.66
6.34
7.12
7.35
5.93
4.90
5.08
5.05
6.73
5.76
6.38
5.35
5.14
5.33
4.71
4.90
4.07

571






6.48
5.69
6.21


5.79
5.69



7.11
                                      22

-------
Table 3.  Continued.

47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Chemical"
70 + 76
56 + 60 + 81
52
47 + 48
44
74
49
64
42
53
40
41+71
46
45
101
84
118
110
87 + 97
105
95
85
92
82
91
99
153
138
149
146
141
128
151
132
156
136
129
180 .
LogKow
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
6.22
6.73
7.36
Predicted1"
Log BAF
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
6.22
6.73
7.36
Measured6
Log BAF
7.06
7.47
6.10
5.97
6.27
7.02
6.34
6.96
7.01





6.61
7.53
7.37
7.11
7.38
7.36
6.14
7.12

7.50
6.33
6.51
7.50
7.43
7.31
7.93
7.46

6.62
7.08

6.34

7.66
                                    23

-------
Table 3.  Continued.
Predicted11 Measured0
Chemical" LogKow Log BAF Log BAF
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102



187 + 182
170+190
183
177
174
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
8 Chemical abbreviations taken
" DrA^lis%+Ar4 D A C» %AI AI^A ^tl^+oinAr
7.19
7.37
7.20
7.08
7.11
7.14
7.11
7.11
7.02
7.65
7.62
7.80
7.56
7.62
8.00
8.09
7.74
8.18



from Oliver
4 l^«/ +^lyit"»*t •#
7.19 7.60
7.37 8.20
7.20 8.16
7.08 8.07
7.11 7.88
7.14
7.11
7.11
7.02
7.65 8.26
7.62
7.80 7.69
7.56
7.62
8.00
8.09
7.74
8.18
-0.61
0.39
61
and Niimi(1988).
'!•»« r\rr\i4t i*-»+ *\f +ttA Cf*H/l ^nri 1^ -fr\r
      r ICUIL.LOU ur-vro vvcic uuLdiiicu uy iais.iiiy uic piuuuui ui uic r v^ivi aiiu IN.QW lul
      each chemical. Because the FCM is set to 1.0 for zooplankton, the predicted
      log BAF equals log Kow.
      Field-measured BAFs were determined by dividing the  chemical  residues on a
      lipid weight basis in  the organisms (fjg of chemical/Kg of lipid) by the freely
      dissolved concentration of thechemical in water (/jg of freely dissolved
      chemical/L of water).
                                       24

-------
Table 4. Measured and Predicted BAFs for Sculpin. BAFs are reported on a lipid
weight basis using the freely dissolved concentration of the chemical in  water (i.e.,
(JJQ of chemical/Kg of lipid)/(/yg of freely dissolved chemical/L of water)).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Chemical"
ppDDT
ppDDE
ppDDD '
mirex
photomirex
g-chlordane
alpha-BHC
gamma-BHC
HCBD
DCS .
HCB
QCB
1,2,3,5-TeCB
1,2,4,5-TeCB
1,2,3,4-TeCB
1,3,5-TCB
1,2,4-TCB
1,2,3-TCB
2,4,5-TCT
2,3,6-TCT
PCT
8
6
5
12
13
28 + 31
18
22
26
16
33
17
25
24 + 27
32
Predicted6
Log Kow
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
5.44
Measured0
Log BAF
7.67
8.01
7.18
8.14
8.14
7.10
3.83
3.72
5.29
7.48
6.51
5.71
5.00
4.85
4.90
4.31
4.08
4.20
5.43
5.43
7.56
5.67
5.66
5.51
5.89
6.02
6.63
5.91
6.49
6.62
5.83
6.51
5.98
6.63
6.19
6.23
Log BAF
7.47
7.83
6.89
7.77
7.69
7.12
4.69
5.05
5.55
7.77
6.53
5.67


4.91

4.57



6.41





6.37
5.97








                                      25

-------
Table 4.  Continued.

46
47
48
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
Chemical"
66
70 + 76
56 + 60 + 81
70 + 76
56 + 60 + 81
52
47 + 48
44
74
49
64
42
53
40
41+71
46
45
101
84
118
110 '
87 + 97
105
95
85
92
82
91
99
153
138
149
146
141
128
151
132
156
Log Kow
6.20
6.17
6.19
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
Predicted6
Log BAF
7.36
7.33
7.35
7.33
7.35
6.86
6.84
6.77
7.36
6.91
7.05
6.78
6.53
6.62
6.86
6.38
6.38
7.59
7.14
7.99
7.71
7.48
7.90
7.26
7.49
7.56
7.36
7.26
7.60
8.17
8.08
7.92
8.14
8.07
7.99
7.88
7.82
8.42
Measured6
Log BAF
7.45
7.06
7.48
7.06
7.48
6.80
6.15
6.65
7.30
6.77
7.16
7.07





7.30
8.05
7.86
7.44
7.54
7.82
6.98
7.50
7.70
7.60
6.44

8.05
8.06
7.28
8.49
8.11

8.34
7.41

                                     26

-------
Table 4.  Continued.

82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102



Chemical"
136
129
180
187 + 182
170+190
183
177
174
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
6.22
6.73
7.36
7.19
7.37
7.20
7.08
7.11
7.14
7.11
7.11
7.02
7.65
7.62
7.80
7.56
7.62
8.00
8.09
7.74
8.18



Predictedb
Log BAF
7.38
7.98
8.57
8.43
8.58
8.44
8.33
8.36
8.39
8.36
8.36
8.27
8.78
8.78
8.90
8.72
8.78
9.01
9.04
8.87
9.08
0.01
0.42
54
Measured6
Log BAF
7.13

8.45
8.07
9.15
8.81
8.63
8.24




9.14

8.52









      Chemical abbreviations taken from Oliver and Niimi (1988).
      Predicted BAFs were obtained by taking the product of the FCM and Kow for
      each chemical.
      Field-measured BAFs were determined by dividing the chemical residues on a
      lipid weight basis in the organisms U/g of chemical/Kg of lipid) by the freely
      dissolved concentration of the chemical  in water (fjg of freely dissolved
      chemical/L of water).
                                     27

-------
Table 5.  Measured and Predicted BAFs for Alewives.  BAFs are reported on a lipid
weight basis using the freely dissolved concentration of the chemical in water (i.e.,
(jjg of chemical/Kg of lipid)/0t/g of freely dissolved chemical/I of water)).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Chemical"
ppDDT
ppDDE
ppDDD
mirex
photomirex
g-chlordane
alpha-BHC
gamma-BHC
HCBD
DCS
HCB
QCB
1,2,3,5-TeCB
1,2,4,5-TeCB
1 ,2,3,4-TeCB
1,3,5-TCB
1,2,4-TCB
1,2,3-TCB
2,4,5-TCT
2,3,6-TCT
PCT
8
6
5
12
13
28 + 31
18
22
26
16
33
17
25
24 + 27
32
Log Kow
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
5.44
Predictedb
Log BAF
7.49
7.82
7.02
7.95
7.95
6.95
3.82
3.71
5.22
7.31
6.39
5.63
4.94
4.81
4.85
4.29
4.06
4.18
5.36
5.36
7.39
5.59
5.58
5.43
5.80
5.92
6.51
5.82
6.37
6.50
5.74
6.39
5.88
6.51
6.09
6.13
Measured0
Log BAF
7.61
7.86
6.78
7.72
7.63
6.68
4.82
5.00

7.77
6.31









6.53





6.68
6.39








                                       28

-------
Table 5.  Continued.'

46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
Chemical"
66
70 + 76.
56 + 60 + 81
52
47 + 48
44
74
49
64
42
53
40
41+71
46
45
101
84
118
110
87 + 97
105
95
85
92
82
91
99
153
138
149
146
141
128 '
151
132
156
136
129
Log Kow
6.20
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
6.22
6.73
Predicted6
Log BAF
7.20
7.17
7.19
6.72
6.70
6.63
7.20
6.76
6.90
6.64
6.41
6.50
6.72
6.27
6.27
7.41
6.99
7.80
7.53
7.31
7.71
7.10
7.32
7.38
7.20
7.10
7.42
7.98
7.89
7.73
7.95
7.88
7.80
7.69
7.63
8.23
7.22
7.79
Measured6
Log BAF
7.57
7.31
7.79
6.84
6.85
6.86
7.35
6.98
7.30
7.38





7.25
7.90
7.71
7.51
7.89
7.72
7.14
7.67
7.93
7.86
6.74
7.37
7.82
7.89
7.75
8.30
7.96

8.17
7.45

7.25

                                     29

-------
Table 5.  Continued.
                                          Predicted6   Measured0
            Chemical"           Log Kow    Log BAF    Log BAF
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102



180
187 + 182
170+190
183
177
174
178
171
185 .
173
203 + 1 96
201
194
195
198
205
206 '
207
203
Average difference
Standard deviation
Number of values
a Chemical abbreviations taken
" DrArlio+Arl R A Ce \AfArA sttt+oin Ar
7.36
7.19
7.37
7.20
7.08
7.11
7.14
7.11
7.11
7.02
7.65
7.62
7.80
7.56
7.62
8.00
8.09
7.74
8.18



from Oliver
4 Kw •tals'inn +
8.38
8.24
8.39
8.25
8.13
8.16
8.19
8.16
8.16
8.08
8.59
8.59
8.71
8.53
8.59
8.82
8.86
8.68
8.89
-0.17
0.40
51
and Niimi
•ho nmrli ir*
8.15
7.99
8.84
8.46
8.54
8.51




8.82

8.22









(1988).
•fr nf -frho nf*IV/! anrl If -for
                                                                      vow
      each chemical.
      Field-measured BAFs were determined by dividing the chemical residues on a
      lipid weight basis in the organisms (jjg of chemical/Kg of lipid) by the freely
      dissolved concentration of the chemical in water (//g of freely dissolved
      chemical/L of water).
                                     30

-------
Table 6.  Measured and Predicted BAFs for Small Smelt.  BAFs are reported on a
lipid weight basis using the freely dissolved concentration of the chemical in water
(i.e., U/g of chemical/Kg of lipid)/(/sg of freely dissolved chemical/L of water)).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Chemical"
ppDDT
ppDDE
ppDDD
mirex
photomirex
g-chlordane
alpha-BHC
gamma-BHC
HCBD
DCS .
HCB
QCB
1,2,3,5-TeCB
1,2,4,5-TeCB
1,2,3,4-TeCB
1,3,5-TCB
1 ,2,4-TCB
1,2,3-TCB
2,4,5-TCT
2,3,6-TCT
PCT
8
6
5
12
13
28 + 31
18
22
26
16
33
17
25
24 + 27
32
Log Kow
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
5.44
Predicted1"
Log BAF
7.49
7.82
7.02
7.95
7.95
6.95
3.82
3.71
5.22
7.31
6.39
5.63
4.94
4.81
4.85
4.29
4.06
4.18
5.36
5.36
7.39
5.59
5.58
5.43
5.80
5.92
6.51
5.82
6.37
6.50
5.74
6.39
5.88
6.51
6.09
6.13
Measured0
Log BAF
7.43
8.11
6.80
7.73
7.75
6.44
4.56
4.77

7.61
6.14















6.57









                                      31

-------
Table 6.  Continued.

46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
Chemical"
66
70 + 76
56 + 60 + 81
52
47 + 48
44
74
49
64
42
53
40
41+71
46
45
101
84
118
110
87 + 97
105 .
95
85
92
82
91
99
153
138 •
149
146
141
128
151
132
156
136
129
Log Kow
6.20
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
6.22
6.73
Predicted1"
Log BAF
7.20
7.17
7.19
6.72
6.70
6.63
7.20
6.76
6.90
6.64
6.41
6.50
6.72
6.27
6.27
7.41
6.99
7.80
7.53
7.31
7.71
7.10
7.32
7.38
7.20
7.10
7.42
7.98
7.89
7.73
7.95
7.88
7.80
7.69
7.63
8.23
7.22
7.79
Measured6
Log BAF
7.46
7.32
7.73
6.54
6.73
6.40
7.31
6.46
7.14
7.18





7.05
7.90
7.76
7.41
7.79
7.71
6.83
7.41
7.17
7.77
6.40
6.43
7.93
7.87
7.63
8.30
7.84

7.74
7.06



                                     32

-------
Table 6.  Continued.
Predicted6

84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102



Chemical"
180
187+182
170+190
183
177
174
178 "
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
7.36
7.19
7.37
7.20
7.08
7.11
7.14
7.11
7.11
7.02
7.65
7.62
7.80
7.56
7.62
8.00
8.09
7.74
8.18



• Chemical abbreviations taken from Oliver
** Ct* t^Af* ff\r <^l At**l« lf**+ *m*f*^f* m if+f*f4 ff^f <+l"«**h r+w*f
Log BAF
8.38
8.24
8.39
8.25
8.13
8.16
8.19
8.16
8.16
8.08
8.59
8.59
8.71
8.53
8.59
8.82
8.86
8.68
8.89
-0.04
0.40
48
and Niimi
.11 ___i«. i
Measured0
Log BAF
8.18
8.01
8.86
8.59
8.54
8.31




8.79

8.24









(1988).
      obtained by taking the product of the FCM and Kow for each chemical.
      Field-measured BAFs were determined by dividing the chemical residues on a
      lipid weight basis in  the organisms (JJQ of chemical/Kg of lipid) by the freely
      dissolved concentration of the chemical in water U/g of freely dissolved
      chemical/L of water).
                                     33

-------
Table 7.  Measured and Predicted BAFs for Large Smelt. BAFs are reported on a
lipid weight basis using the freely dissolved concentration of the chemical in water
(i.e., (jjg of chemical/Kg of lipid)/{//g of freely dissolved chemical/L of water)).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Chemical"
ppDDT
ppDDE
ppDDD
mirex
photomirex
g-chlordane
alpha-BHC
gamma-BHC
HCBD
OCS
HCB
QCB
1,2,3,5-TeCB
1,2,4,5-TeCB
1 ,2,3,4^TeCB
1,3,5-TCB
1,2,4-TCB
1,2,3-TCB
2,4,5-TCT
2,3,6-TCT
PCT
8
6
5
12
13
28 + 31
18
22
26
16
33
17
25
24 + 27
32
Log Kow
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
5.44
Predicted1"
Log BAF
7.85
8.23
7.26
8.37
8.37
7.17
3.80
3.69
5.13
7.62
6.45
5.55
4.85
4.72
4.77
4.24
4.03
4.15
5.26
5.26
7.72
5.52
5.51
5.35
5.74
5.88
6.60
5.76
6.43
6.59
5.68
6.45
5.84
6.60
6.07
6.11
Measured0
Log BAF
7.93
8.27
6.84
8.04
7.97
6.50
4.71
4.82

7.85
6.40
5.87














6.92









                                      34

-------
Table 7.  Continued.,

46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
Chemical"
66
70 + 76
56 + 60 '+81
52
47 + 48
44
74
49
64
42
53
40
41+71
46
45
101
84
118 •
110
87 + 97
105
95
85
92
82
91
99
153
138
149
146
141
128 •
151
132
156
136
129
Log Kow
6.20
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
6.22
6.73
Predictedb
Log BAF
7.49
7.46
7.48
6.86
6.84
6.77
7.49
6.94
7.11
6.78
6.47
6.59
6.86
6.29
6.29
7.76
7.20
8.20
7.89
7.63
8.11
7.36
7.64
7.73
7.49
7.36
7.77
8.40
8.31
8.13
8.37
8.30
8.20
8.08
8.02
8.65
7.51
8.19
Measured6
Log BAF
7.88
7.71
8.12
6.91
7.22
6.92
7.66
7.03
7.54
7.63





7.35
8.29
8.13
7.81
8.06
8.11
7.17
7.85
7.80
8.14
6.90
7.40
8.24
8.22
7.99
8.66
8.17

8.28
7.67



                                     35

-------
Table 7.  Continued.

84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102



Chemical"
180
187 + 182
170+190
183 .
177
174
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209 .
Average difference
Standard deviation
Number of values
Log Kow
7.36
7.19
7.37
7.20
7.08
7.11
7.14
7.11
7.11
7.02
7.65
7.62
7.80
7.56
7.62
8.00
8.09
7.74
8.18



Predicted15
Log BAF
8.79
8.66
8.80
8.67
8.56
8.59
8.62
8.59
8.59
8.50
8.96
8.98
9.06
8.92
8.98
9.12
9.13
9.05
9.13
-0.10
0.41
49
Measured0
Log BAF
8.45
8.34
9.02
8.85
8.78
8.71




9.13

8.50









      Chemical abbreviations taken from Oliver and Niimi (1988).
      Predicted BAFs were obtained by taking the product of the FCM and Kow for
      each chemical:
      Field-measured BAFs were determined by dividing the chemical residues on a
      lipid weight basis in the organisms (//g of chemical/Kg of lipid) by the freely
      dissolved concentration of the chemical in water (//g of freely dissolved
      chemical/L of water).
                                     36

-------
Table 8.  Measured and Predicted BAFs for Piscivorous Fish. BAFs are reported on
a lipid weight basis using the freely dissolved concentration of the chemical in
water (i.e., (JJQ of chemical/Kg of lipid)/U/g of freely dissolved chemical/L of
water)).

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Chemical"
ppDDT
ppDDE
ppDDD .
mirex
photomirex
g-chlordane
alpha-BHC
gamma-BHC
HCBD
OCS
HCB •
QCB
1,2,3,5-TeCB
1,2,4,5-TeCB
1,2,3,4-TeCB
1,3,5-TCB
1,2,4-TCB
1,2,3-TCB
2,4,5-TCT
2,3,6-TCT
PCT
8
6
5
12
13
28 + 31
18
22
26
16
33
17
25
24 + 27
Log Kow
6.45
6.76
6.06
6.89
6.89
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.65
4.56
4.59
4.17
3.99
4.10
4.93
4.93
6.36
5.07
5.06
4.97
5.22
5.29
5.67
5.24
5.58
5.66
5.16
5.60
5.25
5.67
5.40
Predicted6
Log BAF
7.83
8.19
7.29
8.32
8.32
7.20
3.79
3.68
5.14
7.62
6.53
5.61
4.85
4.70
4.75
4.22
4.01
4.13
5.30
5.30
7.71
5.58
5.57
5.38
5.81
5.96
6.68
5.83
6.51
6.67
5.75
6.53
5.92
6.68
6.16
Measured0
Log BAF
7.78
8.35
7.00
8.13
8.07
6.79
4.69
4.93

8.07
6.40
5.81


5.07











6.89
5.75
6.39

5.92
5.32
5.52


                                      37

-------
Table 8.  Continued.

45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
Chemical"
32
66
70 + 76
56 + 60 + 81
52
47 + 48
44
74
49
64
42
53
40
41+71
46
45
101
84
118
110
87 + 97
105
95
85
92
82
91
99
153
138 .
149
146
141
128
151
132
156
136 •
Log Kow
5.44
6.20
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
5.84
5.53
5.53
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
6.35
6.20
6.13
6.39
6.92
6.83
6.67
6.89
6.82
6.74
6.64
6.58
7.18
6.22
Predicted1"
Log BAF
6.20
7.50
7.47
7.49
6.92
6.90
6.83
7.50
7.00
7.15
6.84
6.55
6.67
6.92
6.38
6.38
7.75
7.24
8.16
7.87
7.63
8.07
7.38
7.64
7.72
7.50
7.38
7.76
8.35
8.26
8.09
8.32
8.25
8.16
8.05
7.99
8.57
7.52
Measured0
Log BAF
6.76
7.79
7.56
7.96
7.01
7.18
6.96
7.66
7.13
7.51
7.49
6.51
6.55



7.45
8.28
8.15
7.79
8.08
8.13
7.25
7.89
8.11
8.13
6.92
7.39
8.32
8.30
7.99
8.73
8.32

8.51
7.56

7.37
                                     38

-------
Table 8.  Continued.
Predicted13 Measured6

83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102



Chemical"
129 .
180
187+182
170+190
183
177
174
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
9 Chemical abbreviations taken
** DrAr1ir*+As-4 D A Ct* \*tart\ s\tt+<*»ir-k AJ>
Log Kow
6.73
7.36
7.19
7.37
7.20
7.08
7.11
7.14
7.11
7.11
7.02
7.65
7.62
7.80
7.56
7.62
8.00
8.09
7.74
8.18



from Oliver
4 !•%%/ +^Ix!«« +
Log BAF
8.15
8.68
8.58
8.69
8.59
8.49
8.52
8.55
8.52
8.52
8.44
8.81
8.84
8.88
8.78
8.84
8.89
8.87
8.90
8.83
-0.12
0.40
59
and Niimi
'I**A nr*\rti if*
Log BAF

8.58
8.43
9.20
9.03
9.01
8.74




9.26

8.56









(1988).
* f\f 4>k>n Cr*kH **r%rl V fnr
    each chemical.
    Field-measured BAFs were determined by dividing the chemical residues on a
    lipid weight basis in the organisms (jjg of chemical/Kg of lipid) by the freely
    dissolved concentration of the chemical in water (jjg of freely dissolved
    chemical/L of water).
                                     39

-------
2  I

£  o
a,  .-

•£  ro
0  e ^
C  ^ CO
c  O co
£  0 o>

—  •* C,
o  « ^
o  -i e
a.  h. .£
 gc?
Z  .£  S
.£  o-  >

S  §  o
« S  -5
 .  re   w
    0>
    o
    c
    o
    o


CO
£
*0fl
._ •
•^*
CO
0)
Q.



CO
0)
C
0
^3
"o
in Q
O CO
1- o
-------












o g
jjj Qp
£ £
2 1
13 -5
m Z
ed and Predict
ooplankton
rom Oliver and
3 N *"
3) &
m LL
0) <
S 03
•o
CM 0)
g 3
3 w
re <°
•5*  00 h-
M
JV

to
1 =
3 2 ^5
^^^ MB ••{
« B O 3
W H c. "n
» 2 Q o «
2 O 03 O> tS 5
OH » O


-------















5
CO
u.
<
m
•o
0
**
u
'•5
£
Q.
•o
C
03
•o
£
3
03
V
^
m
CO

••*
O/
il
















00*
00
f9\
O*
^^

_
2*~

"0
03
.£ §
f5
" £
W 1
(0
U.
<
m
•o
£
3
(!)
03
0
S



» D)
s «, -
3 £ W
^  2 Q § o
2 O ffi o> -g «
OB o °° m Q.
w \ NOW
i '-.^ § °- S
• • -9 g
V* 5
• ~ _
• '•A^
J"
^1 A
• * w
: X
j>. i
xi
*._
.
< \
•
1,1,1,1,1,1,"-.
O> e jj
1 ^
0 W
& u
ill
€ =
S g
•B e

C •
E o
Nf%
u
" 1
CO § S)
o 6
1
i
1
CO

•^
f^«




.s
O) CM
0 •*



lf>





^

o>     oo    i^-    «o     in
            dva BOH

-------









b.
•2
(0
u.
<
ffl
•o
V
*•«
o
TJ
2 «
o. 8
•o .>
|i
3
W
(Q
Q>
•
2
mm
3
O)
il










co"
00
T—
E
2*™"

•o
CO
CD
O
2
(0
u.
ffi
•o
3
CO
(0



CO
C c
Q) •;£ -
= 22
«. o  O^l ^~ A ••• 2 ""
' «5< ^ "i™^*^ O ^^ T3 C3
\ • S c 3 ° £ u.
• Q. O DO Q.
N O tf»
P- V^ V)
oj Q. u-
• ^ \ W ^ £Q
3 \Q ^
9 \ -
*^t
0 .A
^^f
*y& o
•B^fe.
^^^^^^fc .A
• ^ '••%
0'<1._
• \
—
"•-.
\
\
•
.1 , 1 , 1 , 1 • 1 i 1 , "'•
O)




CO





r^




5
o
O)
3
•••

in



*
ff\
dva

-------
       CO
       CO
 (0
LJ.

CO

•o
 0>
3
V)
to
0)
in
3
D)
    •O

    CO
£  a.
 0)  O
 E  .£


=  ^

 1  |
co  •*=
    (A
    Ll-

    CO

    •o


    3
    (0
    CO
    Q>
                                                                               i!
                                                                               * o
                                                                               •e e
                                                                               E S
                                                                               CM 0

                                                                               " 1
                                                                            O
                                                                        in
                                                                       CO
                   O>
                       00
(D
in
CO
                                   dva

-------














k.
C)
to
LL
<
CD
•o
0)
S

•o
CD
V.
D.
•o
C
(0
•o
0)
3

CO
CD
w
•
CO
0)
SM
3
0)
il




















00
rt>
U»
I

•o
c
CO
±i v.
il
f/> —
w O
0) _
o) E
jjj 2
_j «•—
CO
LL

CO
•o
£
mm
3
CO
CO
0>
s



c
^







CO
0)
2
,£•

• To
m
• w
^ Q_
W\ ""
• ^
\ •
V' •
'.
'I.
^
-5^
m^f. ^^
Jb.
^^P'. A*
* *2'
• ^w-.
M -.
^ ^





.




•
l,l,l,

5 O> 00 h-
•
dva

CO
« D)
Si =
Q) «^ -
S w co
^ CD 3
H e *H
!L 0 oo
O CD D) * «
- O<1 =• ~ 2 —
CO -r^ Ow -O o
8)1 O «> h.
C rf< *• **•
0 * ffl 0.
N 0 CO
s °- &
^^ ^x
«Q r?
_L CD
O






"~


^
*•.
*,
XI \
XI -._
•
".
".
•,
*9
'-.
•
1,1,1,'--

C£) LO -d- C*
BOT
O> o j?
"5: ^
0 ~
& »
o a
•§ ^

^ w
•B e
**
e u
7 -2
II C
00 « S
§S
E
n
E
S
£


f".



£
o
»^
o>2
O^r
,
ml





LA




^^
^
4V%
CO
5


-------












v> en~
,2 00
CO 2
LL ^
CD C
"o H
0) ^
o g -o
••§• 5SI ~^
v £ (C
5> co ».
£ ii §
•D CO US
i 1°
II 1
^ to w
s £ <
2 ffi
•o
fs. Q)
£ 3
3 CO
HI CO
.5? o>
LL 2

."
y
_•"
_."
^


.
0



.
• ^'•^
'^
^r
^ ^^ A
"^
«T-^
X^LJ
^^^r!
* •




•





1,1,


0)
2 D)
S c
2 £ 2»
8 lo I 1
2 O ffl o> -5 w _
^5 W T O "O (J
CO W •• • y Q) ii
Q. 0) ffl CX
H 0 co
0 Q. LL
•Q 2
J. ffl
o




>
1

*Y
• •-. *
J**
\
"•.
'•.
<] \
". -
".
\
B \
1,1,1,1,'--
0> o jg
'S **!
i "

II C
1!
•e e
||
II "£
oo 8s>
o o

i
i



i^




^
o
CO ^
o> S
o ^


LO






-
MS
O>
00
in
           dva

-------
    E.  Prediction of BAFs from Biota-Sediment Accumulation Factor (BSAF)
        Measurements

BSAFs may be used for measuring and predicting bioaccumulation directly from
concentrations of chemicals in surface sediment. They may also be used to
estimate BAF)ds (Cook et al., 1993; 1995).  Since BSAFs are based on field data
and incorporate effects of metabolism, biomagnification, growth, etc., BAF"s
estimated from BSAFs will incorporate the net effect of all these factors. The
BSAF approach is particularly beneficial for developing water quality criteria for
chemicals such as polychlorinated dibenzo-p-dioxins, dibenzofurans and certain
biphenyl congeners which are difficult to measure in water and have reduced
bioaccumulation potential due to metabolism.  The calculation of BAF" from
BSAFs also provides a method for validation of all measured or predicted BAFjds
for organic chemicals.

    1.  Biota-Sediment Accumulation Factors BSAFs

BSAFs are measured by relating lipid-normalized concentrations of chemicals in an
organism to organic carbon-normalized concentrations of the chemicals in surface
sediment samples associated with the average exposure environment of the
organism.  The BSAF equation is:
                              BSAF  = 	—                          (20)
where:  C,   =  lipid-normalized concentration of the chemical in tissues of the
                biota (fjg/g lipid).
        CSOG  =  organic carbon-normalized concentration of the chemical in the
                surface sediment (/vg/g sediment organic carbon).

Since BSAFs are rarely measured for ecosystems which are at equilibrium, the
BSAF inherently includes a measure of the disequilibrium of the ecosystem. This
disequilibrium can be assessed for chemicals with log Kow > 3 with the following
relationship:

                           Cfd'K             K
              BSAF a	—  = Db  •	— *  Db  •  2         (21)
                         Csfd'Ksoc          Ksoc

where:  C"  =  concentration of freely dissolved chemical (associated with
                water) in the tissues of biota (jjg/g wet tissue).
        C"  =  concentration of freely dissolved chemical (associated with pore
                water) in the sediment (fjg/g sediment organic carbon).
        K,   =  lipid-water equilibrium partition coefficient =  C,/C™.
                                     47

-------
             =  the sediment organic carbon-water equilibrium partition
                coefficient = C8oc/C8d.
             =  the disequilibrium (fugacity) ratio between biota and sediment
Measured BSAFs may range widely for different chemicals depending on K,, K.oc,
and the actual ratio of C£d to C,d. At equilibrium, which rarely exists between
sediment and pelagic organisms such as fish, the BSAF would be expected to
equal the ratio of K,/K800 which is thought to range from 1 -4.  When chemical
equilibrium between sediment and biota does not exist, the BSAF will equal the
disequilibrium (fugacity) ratio between biota and sediment (D^ = C"/C") times the
ratio of the equilibrium partition coefficients (approximately 2).

The deviation of D^ from the equilibrium value of 1 .0 is determined by the net
effect of all factors which contribute to the disequilibrium between sediment and
aquatic organisms. D^ > 1 can occur due to biomagnification or when surface
sediment has not reached steady-state with water.  D^ <  1 can occur as a result
of kinetic limitations for chemical transfer from sediment to water or water to food
chain, and biological processes, such as growth or biotransformation of the
chemical in the animal and its food chain. BSAFs are most useful when measured
under steady-state conditions  or pseudo-steady-state conditions in which chemical
concentrations in water are linked to slowly changing concentrations in sediment.
BSAFs measured for. systems  with new chemical loadings or rapid increases in
loading may be unreliable due to underestimation of steady-state C80Cs.

    2.  Relationship of BAFs to BSAFs

Differences between BSAFs for different organic chemicals are good measures of
the relative bioaccumulation potentials of the chemicals. When  calculated from a
common organism/sediment sample set, chemical-specific differences in BSAFs
reflect primarily the net effect of biomagnification,  metabolism, and bioenergetic
and bioavailability factors on each chemical's D^.   Ratios of BSAFs of PCDDs and
PCDFs to a BSAF for TCDD (bioaccumulation equivalency factors, BEFs) have been
proposed in the GLWQI for evaluation of TCDD toxic equivalency associated with
complex mixtures of these chemicals (see 58 FR 20802).  The same approach is
applicable to calculation of BAFs for other organic chemicals.  The approach
requires data for a steady-state or near steady-state condition between sediment
and water for both a. reference chemical (r) with a field-measured BAF]d and other
chemicals (n =i) for which  BAF]ds are to be determined.  BAFjd for a chemical "i" is
defined as:


                                  d)   - -±
                                     48

-------
where:  C,   =  lipid-normalized concentration of the chemical in tissues of the
                biota (jjg/g lipid).
        Cj,d  =  concentration of freely dissolved chemical in water (//g///L
                water).

Substitution of C/ from equation 20 into Ct of equation 22 for the chemical i gives:

                                              .  '*-*soc' i
                                                    fd
                                                 (Cw  ) ±
                                                                      (23)
In order to avoid confusion with the equilibrium partition coefficients Ksoc, Kpoc or
Kdoc, the chemical concentration quotient between sediment organic carbon and a
freely dissolved state in overlying water is symbolized by n,oc:

                                         (C   ) •
                                     -     soc' i
                                            fd).
                                           W  '1


Thus the ratio of BAF" for chemical i and a reference chemical r is:


                                                                      (25)
                     (BAF/d)r      (BSAF)r(nsoc)r
If both chemicals have similar fugacity ratios between water and sediment, as is
the case for many chemicals in the open waters of the Great Lakes:
                                              r±                     (26)
                            *  soc' r      *  OW' r

therefore:

              (BAFfd).  =  '-°*^fd^   •  (BSAF)i (Kow)i
The assumption of equal or similar fugacity ratios between water and sediment for
each chemical is equivalent to assuming that for all chemicals used in BAF"
calculations:  (1) the. concentration ratios between sediment and suspended solids
in the water and (2) the degree of equilibrium between suspended solids and C^d
are the same. Thus, errors could be introduced by inclusion of chemicals with non-
steady-state external loading rates or chemicals with strongly reduced Cj[,d due to
rapid volatilization from the water.  Note that BAF]ds calculated from BSAFs will
incorporate any errors associated with measurement of the BAF]d for the reference
                                    49

-------
chemical and the Kows for both the reference and unknown chemicals.  Such errors
can be minimized by comparing results from several reference chemicals, including
those with similar Kows to those of the unknown chemicals, and by assuring
consistent use of C" values which are adjusted for dissolved organic carbon
binding effects on the fraction of each chemical that is freely dissolved (ffd) in
unfiltered, filtered or centrifuged water samples. BAF,s based on total chemical
concentration in water (BAFJ) can be calculated on the basis of  ffd for the dissolved
and particulate organic carbon concentrations in the water (POC and DOC):
                                  = BAF/d-ffd                      (28)
where:
                1  + DOC-Kdoc + POC-Kp^    1 + DOC-KOW

Further information on calculation of concentrations of freely dissolved chemicals
in water may be found in section III.B of this document titled "Bioavailability".

    3.  Calculation of BAF"s from Lake Ontario Data

Two data sets are available to EPA for calculating BAF)ds from BSAFs for fish in
Lake Ontario. The Oliver and Niimi (1988) data set, which has been used
extensively for construction of food chain models of bioaccumulation and
calculation of FCMs, biomagnification factors and BAF]ds from chemical
concentrations determined in organisms and water, also contains surface sediment
data which allows calculation of lakewide average BSAFs. The second data set is
provided by an extensive sampling of fish and sediments in 1987 for  EPA's Lake
Ontario TCDD Bioaccumulation Study (U.S. EPA, 1990)  for the purpose of
determining BSAFs.  These samples were later analyzed for PCDD, PCDF, PCB
congeners and some organochlorine pesticides at ERL-Duluth.  Although  these data
should be submitted for publication within this year, they are needed  here to
provide a unique data set for checking BAF"s calculated from Oliver and Niimi data
from samples collected between 1981-1984 and calculating BAF"s for organic
chemicals not measured by Oliver and Niimi.

BAF]ds for salmonids were calculated for this demonstration of the BSAF ratio
method  using PCB congeners 52, 105 and 118 and DDT as reference chemicals.
Several reference chemicals were used in order to examine the variability
introduced by choice of reference chemical.  The water analyses of Oliver and
Niimi (1988) were adjusted for an estimated  2 mg/L residual dissolved organic
carbon concentration in the centrifuged water (assumed no residual POC) and an
estimated Kdoc = Kow/10 in order to calculate C™ from ffd (equation 30).  Log Kows
for PCBs are those reported by Hawker and Connell (1988). Log Kows for PCDDs
and PCDFs are those estimated by Burkhard and Kuehl (1986) except for the

                                     50

-------
penta, hexa, and hep'ta chlorinated dibenzofurans which were estimated on the
basis of assumed similarity to the trends reported for the PCDDs by Burkhard and
Kuehl (1986).  Log Kows for other chemicals are either as cited in the Appendix B
of this document or noted in Table 9.  Table 9 contains the measured and
predicted log BAFjds from the two data sets.

    4.  Validity of BAF{ds Calculated from BSAFs

Figures 8, 9, and 10 show the relationship of log BAF]ds to log Kows for (1) Oliver
and Niimi (1988) BAFjds determined from measured concentrations of freely
dissolved chemicals in Lake Ontario water in 1984; (2) BAF"s calculated from
BSAFs derived from Oliver and Niimi data; and (3) BAFjds calculated from EPA
BSAFs for lake trout in Lake Ontario in 1987 (Cook et al., 1995). The diagonal
lines represent a 1:1 ratio of log BAF to log Kow.  The PCB congener BAFjds in all
three sets of data appear quite similar. The EPA BAFjds predictions (figure 3)
include a number of chemicals not in the Oliver and Niimi data set. These are the
PCDDs,  PCDFs, chlordanes, nonachlors and dieldrin. Only the dieldrin BAFjd has
been measured elsewhere.  The BAFjds for five of six chlordanes and nonachlors
are much greater than those for PCBs with the same estimated log Kow. Therefore,
the log Kow values chosen here for the chlordanes and  nonachlors may  be
significantly underestimated. The bioaccumulative PCDDs and PCDFs (2,3,7,8-
chlorinated), as expected due to metabolism in fish, have BAFjds 10-1000 fold less
than PCBs with similar Kows. Thus, the BSAF method for measuring BAFjds
appears to work  well for Lake Ontario.

Accuracy of the BSAF method can be best judged on the basis of comparison of
the BAF]ds calculated from BSAFs to field-measured BAF"s. Figure 11  illustrates
the agreement between log BAF]ds calculated from the Oliver and Niimi water data
and those calculated from the sediment data.  The BAFjds for chlorinated benzenes
and toluenes may tend to be underestimated with  BSAFs because the water-
sediment fugacity gradient is altered in comparison to PCBs in response to  rapid
volatilization losses from water.  Use of EPA BSAFs measured from a different set
of fish and sediment samples collected several years after the Oliver and Niimi
samples gives BAF{ds that correlate equally well with the BAF]ds calculated from
Oliver and Niimi data (figure 12).

All of the above correlations were based on the BSAF method using the Oliver and
Niimi measured Lake Ontario salmonid BAF" for PCB congener 52 as the
reference. Very similar correlations result for comparisons of data in Table 9 for
PCB congeners 105, 118 or DDT as reference chemicals.  The BSAF method is
strengthened through use of several reference chemicals with a range of Kows and
greatest likelihood for accuracy in measurements of concentrations in water.  The
two data sets and four reference chemicals resulted in  either four or eight
determinations of BAFjd for each chemical listed in Table  9.  Mean log BAFjds

                                    51

-------
(geometric means of BAFjds) for the 4-8 determinations from Lake Ontario data are
reported in Table 10.  The BAF]d for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) at
7.85 x 106 compares well to 3.03 x 106 estimated by a different method for
TCDD log Kow = 7 by Cook et al. (1993).  The small difference in the two
estimates may be attributable to an underestimate of the sediment-disequilibrium
between sediment and water by Cook et al. (1993) that resulted in an
overestimate of Cj|,d.

The greatest test for- robustness of the BSAF method for predicting BAF"s that are
applicable throughout the Great Lakes would be a comparison of two totally
independent data sets based on different ecosystems and conditions. Such a
comparison can be made for bioaccumulation of PCBs in Lake Ontario fish and
Green Bay fish. The EPA Green Bay /Fox River Mass Balance Study involved
extensive sampling of water, sediment and fish in 1989.  Green Bay is a shallower,
smaller, and more eutrophic body  of water than Lake Ontario.  Measurement of
bioaccumulation in  Green Bay is complicated by the movement and interaction of
biota through gradients of decreasing PCBs, nutrients and suspended organic
carbon which extend from the Fox River to the outer bay and Lake Michigan. Table
9 contains brown trout BAF)ds calculated from  PCB BSAFs measured in the mid-
bay region using PCB congeners 52 and 118 as reference chemicals. The
reference chemical BAF)ds were determined with water and brown trout data from
the same region. Concentrations  of freely dissolved PCBs were calculated, as for
Lake Ontario, on the basis of dissolved organic carbon in the water samples and an
assumed Kdoc = Kow/10. Despite  the complex exposures of Green Bay fish, figures
13 and 14 illustrate log BAF jd - log Kow relationships found in Green  Bay which  are
similar to those from the Oliver and Niimi and EPA Lake Ontario data sets. The
correlations between the PCB BAFjds for Green Bay brown trout and BAF]ds based
on Oliver-Niimi salmonid and water measurements and EPA lake trout BSAFs are
shown in figures 15-18 for reference chemicals PCB 52 and PCB 118, respectively.
Good agreement exists between Green Bay  brown trout predictions and Lake
Ontario measured and BSAF-predicted BAF,ds for both reference chemicals.

The means of log BAFjds calculated for each chemical from two sets of BSAFs and
four reference chemicals for 124 chemicals measured in Lake Ontario trout (Table
10) are plotted against log Kow in  figure 19.  Only 59 of these chemicals have
field-measured BAFjds.  Correlations between the mean Lake Ontario trout and
Green Bay brown trout BAFjds (figures 20 and 21) indicate that the  Green Bay
brown trout may be slightly larger. This may be a sample set artifact associated
with the complex Green Bay fish-water-sediment relationships in Green Bay rather
than an actual site/species/food chain-specific difference in bioaccumulation.  The
agreement of the Green Bay and Lake Ontario results demonstrates the general
applicability of BAF"s calculated from BSAFs in predicting bioaccumulation in
Great Lakes fish from estimated C'ds.
                                     52

-------
    5.  How to Apply the BSAF Method for Predicting BAFJds

If high quality data are not available for calculating BAF]ds for organic chemicals
that are expected to.bioaccumulate, the mean BAF"s reported in Table 10 may be
used. To apply the method for additional chemicals, site-specific determinations,
or biota from different trophic levels than salmonids, the following steps and data
requirements must be completed:

    a. Reliable BAF]ds which have been measured for several reference chemicals
    in biota in the ecosystem must be chosen. The water sample analyses should
    approximate the  average exposure of the organism and its food chain over a
    time period that'is most appropriate for the chemical, organism and
    ecosystem.  Each C" used to calculate a BAF)d should be based on a
    consistent adjustment of the concentration of total chemical in water for DOC
    and POC using equation 30.  It is preferable to choose at least some reference
    chemicals on the basis of log Kow and chemical class similarity with the test
    chemicals.

    b. Measured (slpw-stir method or equivalent preferred) or estimated Log Kow
    values are chosen for each chemical.

    c. Obtain chemical residue and % lipid data for representative samples of the
    tissues of the organisms.  Migration patterns, food chain movement and
    hydrodynamic factors should be considered. For highly bioaccumulative
    chemicals variation of chemical residues in adult fish in the open waters of the
    Great Lakes within an annual cycle is usually slight.

    d. Obtain chemical concentrations and % organic carbon data for surface
    sediment samples.  Sediment sampling sites should be selected to allow
    prediction  of ratios of freely dissolved chemical concentrations in the overlying
    water of the ecosystem region of interest.  A 1 cm layer of surface sediment
    is ideal but 3  cm samples will work if sedimentation rates are large and
    periodic scouring events are not likely.   Although desirable, sediment samples
    do not have to represent the average surface sediment condition in the area of
    the ecosystem affecting the exposure of the organisms for which
    bioaccumulation  is to measured.  Since this is a ratio method, the
    concentrations of each chemical in sediment need only be predictive of the
    ratios of concentrations of the chemicals in the ecosystem water.

    e. With the data from steps 3 and 4, calculate BSAFs for chemicals of
    interest and reference chemicals (equation 21).

    f.  With BSAFs and Kows for each chemical, plus BAF]ds for reference
    chemicals, calculate BAF]ds using equation 27.


                                     53

-------
    g. Use the BAFjds to predict chemical residues in fish and other biota or to
    establish unsafe concentrations of chemicals in water only on the basis of
    chemical concentration expressions for water and organisms that are
    consistent with the BAF)d definition and measurement.

    6.  Summary •

BAF]ds calculated from two different BSAF data sets for Lake Ontario salmonids
are similar and agree well with field-measured BAF]ds of Oliver and Niimi (1988).
The BSAF method allows calculation of BAF]ds for chemicals which have not been
measured in Great Lakes water but are detectable in fish tissues and sediments.
BAF]ds can also be calculated for other fish species and biota at  lower trophic
levels in the food web. BAFjds calculated for PCBs in Green Bay brown trout agree
well with the Lake Ontario salmonid/lake trout values despite differences in
ecosystem, food chain and exposure conditions.  Mean log BAF"s (geometric
mean of BAF"s) from  4-8 determinations from Lake Ontario data are summarized
in Table 10.
                                     54

-------

























n
h
I
n
CQ
O
•O

rl
n
0)
*
§
U
LI
M
«M
•a
o
4J
«
3
U
iH
id
u
"L
b
itf
CQ
4J
§
EH

fl>
A)
Id
O
rl
o

•
on

0)
3
H
tn
o
u. H
Aj CQ
Oj, U

O & IH
<-H a)
H
in
. o
•H H

f^ ~H CQ
2 a,
oiS01
0 "H
rH rH 0)
0 rl
CN
tn
EC,
f4| PQ
""^ ft
0 "«w
1-1 2
-H CM
§ -rl CQ
Z vJ
0)3*
°g 0)
k
•0
.O l»l
^* rt« 3
_2 Di IB
« w a

•H _.

rtj "^J t|
" g 3
o1*^ *
• ,_( Q)
O ^


•H
e-o
r -H 0)
5S -H rl
(X 2j 3
CO i£ (Q
CQ • id
rH 11)
0 E


«
Fn
0
rH





rH
Rt
U
-rl
(U
U



in o P* *(* ro rH in ^*
en in »* oo rH in co in
vo p- eo r- r~ r- r~ VD'


.


%
f ro VD en co o tN VD room eo
m^»inovo«* win POVOP- p-
p- co vo co co f- in<< p-in^« m





p-cN
-------






















^^»
a
CO
•0
0)
3
n
id
£
I
«M
•o
0)
a
rt
u
«J

Ji
fa
i4*
0)
jj
3
O

0
id
0)
M
0
*
en

O
_J
^^
•§
H
in
o
fH
«L 8
O)^ ^
O iwi (W
rH (1)
M
in
• O
•H rH
P| jj
(0 -H CJ
R) 3
O • 
s1s
EQ -
E
-H _.
gj-H *

o1 • °
rH rH g


"iH
.-HO)
pj ._j J.
CO § S
ffl • (B
rH 0)
0 E

*
fTl
Ol
0




rH
16
O
-H
(U
A
O

•
fN fl in<4< vo vo vo




HSSSPS5 S?VoS «S?
LA LC m irt m in LA LA LA LA LA LA P*«* t*+>
vfc* w ui ui ui M/ %u **J \v w w NU r^ r*



^in *& \o OC900 iH n ro
t^co oo^* <7\*firo o vo in

vo ^o vo %p lo vo in r^* r*1 r**
B
Tfeocninvovor) t~t~»-ir» ocsw
^ iH«r pnmin
CNt^i-irHfHin<«i invovoco r~eo^<
O H O O O O O O O O H O O 00





vovoT)ivDvoi^i>-inincoeovOrHcr\rv)(N
ininininin^nininininininvoinvovo



SSSS?^5!J^?SS2SS?




•


in rH CO rH «9" in VO VOOiHPOPO
VO O tN I rH CS



en in vo
t~ H in
r- eo r*-




PO O\ rH in t"*
in o vo r~ co
r* ^« PO l-l O





^< r~ in co t~- r- -in
VO VO VO VD VO VO VO



SSrHr5SS°P?
^H ^^ ^^ ^H ^H ^H ^t ^H




in

-------


















n
h
O)
n
*
to
01
•o
0)
0
a
R)
o
£


IU
•a
o
4J
a)
3
0
tf
u

n
¥
n
3
O


id
.P
id
a>
6
on

0)
rH
J3
A
H
in
o
H
fflx S

O M IH
IH •*• a)
H
in
• o
-H H
< -H CQ
10 2 a!
O)«a
O • «H
r-l rH fl)
O H
in
h
^"^o!
&, |e S
Oi§*
t2 iH *!
O 0}

•0
fw. (U
•jH ^ ij
M 3 §
0 W (0
0>
E

«
•H _j
^ "H 2
""23
._ ^1 Tfi
21 . «j
. ^ Q)
O «


"H
e-o
•HO)
S "ly ij
w § 5
OQ • n)
iH 0)
0 S

t
t-n
(Ji
i-H




r-l
(B
0
-H
£
0)
6


mHt-tnn
00 OD •* VO O
r^ r^ r*- r^ oo




o H H in en CM en
n in co «(< o ts in
00 00 00 CO 00 00 00





H ro en o in
vo in H «• r-
,}( »j< rt< oo cs in ro
vo oo vo t^ «f v\ a\
00000000000000



^f in ro in H
t» cs a\ vo a\
rH H O rH rH




/

r~ o N ro en H N
ro ro ro c~ on in ro
r^ oo oo oo r^ oo oo






^ in in CN ro oo oj
oo cs r» CN ro ro CN

H



CN 00 00 00 VO VD O\
VO VO VO VO VO VO VO

..


S33S3SE
03
U
a





VO O rH in OOOOBCaVDVOrOOOCN rH
oen CNOH^c^^icNroo rHCNinin^ i-i
oo r- cococococDoocoooon coo^aoooaooor-ininvovovovovovo t-




ro o\ in co ^* en o in ro o\ in oo rH rH vo ro ro in on oo ro
co in ro vo co en o ro o oo co in oo vo ro oo ^j< an ro on t-
co oo co ao co oo en ao en co co oo co co oo in vo vo r- vo t-





mvo iHinrHror>ir«rHrH«((< onooesror»^or(O^ir-coino«*< ro
vo ro ooHcocNrooo r-cs cr\Hininot~t^ on CN CN vo
H on voOrHrorovo rocN rHonrHont^Ht^ CN r^ ro o
on co ooonononcnco o\on o\oocncooovovo r^ r^ r** co



CNtn vomHvovooo^ii-it~ rHinrooo^"crvrocNcncNinvoi-ii-4ro CN
inVD rONOlt^CNVDCNt^'J' •inrH<*rOOOOiHt-IVDVO^
-------




















n

n
a
TO
a
Tl
w
O
ll
g
O

g

IB
a
IB
a

o
o»

a>
rH

1
in
o
b H
oa ^ t_>

r?S«S
in
• o
<-H n
§cu
O • «W
rH rH 0)
0 U
 ^*
^ j 03
s
"HCN
h g in
^ -H 09
2 L)
rn'rH'H
M
•a
h *
"li m
03 w is
ovovorocnpo
r*.r*r*r*r*.r*.r*.iotor*ior*.r*r*.ior*.r*.







SlnS?cow?5rHen!5rHt^ovofN'r>
OOOOOOOOOOOOO • 0 O O
• • • O O O O • • -O • -OO -O
OOO • • • O O O O • O O OOO
OOO- O •
o o o













(NtncocooocNvoinootnininmooco

r-


Q Q Q
Q Q Q Q bb.bb.QQ
QQQQU bbQQQQUU
Q U U CJ Q. QQUUUUQiOi
QUXXXK bOUXXXXEK
Cj Q) 9^ 3u 35 ' CH u) fl) MH 3C 3C 33 ' '
O 0. ' ' * CO O O* CU i i i i 00 Cn
H ' oo oo cri r~ H> leooocnoor-co
ioot-t--oovo icocot^r-cot^vor-
cor^^«vor~Tj
-------




















n
n
•s»
n
TO
CQ
•o
0)
9
n
«J
!
*M
•o
0
4J
a
3
U
•1
0
A
b
s
n
4J
9
O
fi
r*
ID
0)
id
4J
id
o
u
0
•
en
0)
i-l
X)
id

>i«
b (0 H
< CQ H
CQ
ft? 8
O < CM
M 01
^
5- * in
™"ffl
0 < ^
rH 0| uj
ta -oo oo- r^t^covo


B



oot^omcN^« voo t^voi^ H )H
X C C C 0 0
fl) fl3 ro TO ^H ^H j^ ^fj
pj ^ *O *Q *O •£? jC^ u u
"^ "^ ^4 V^ W U U cC «£ f} o <3
ri EOOORtrd'i u u u
*O X.OrHrHrHCC







CAOOCNCNHCNCNin
^pinininininininm





rH H rH H rH CN








-------

















n
n
•*x
B
TO
0)
•u
d>
(j
3
B
10
ID
£
g
«w
•0
o
4J
(8
U
H
U

a
¥
n
4J
3
O
W
EH

ID
0>
3
A
O
M
0
»

0)
rH
•8
EH
>i«
fc  o 2



^
uT
hn
r?





10
u
-H
E
0)
u
TJI »}< r~o o rs CD CN in H oooororHCNOVooovo oinr~cocoO'9't- in in
in VD o in vo m ro in en eo r-voe^^eoocnHOo HCN«mmHvoa\coco coocMrHi-icotsco in vo
"* '" inr*r*ir»r*. r*r^r«>«nr*.r*.r*rs. r^ voH^^CMt^ rnvocMvoinr-fMinn^ CMooHHinmvom H H
oo oo rH o ro Tf CN *t<  in cMCMmmcMCMOrric)rHfOrtcMpnr>)rovo^ir^incor--t~- -in
inintninininininininininvoinvDvovovDvovovovDvovDvovovovovovovDvovovovovovo vo




rH rH rH rH rH rH rH rH rH rH rH


*




-------

















n
i
M
n
•0
o
S
0
s
i»
h «J i-
§
o a
S
^•i J2? m
N, CQ
rH 2 tU
U 0)
14
<2
03 O
oa o<
B
CO
b H
S ffl
?§*
r-H *W
• 00
S *"*
^3§
s *
rH O 0)
m"fl! §
Qi
O 0)
r— t W

*H
^:H a
§
Dl • 4-1
O ,-4 0)
•-i O H

I
tn
O
i-H





_J
R)
U
-rl
E
0)
u
o> o ^ 'ro m
C^ (Tl VO VO VO
00 CO 00 OO 00

in vo r- min
cocococor^f^coco


H o t~  ro
CO CO t»t- CO


•
inr^invopo^ieoo
CO CD 00 GO CO CO CO O\


.

CNCOCOCOVOVOatr-l
vovovovovo^vor^




VOOOHVOCTlHrOVO
HHHHi-ttHHrH




•

co vom o (N (N in fs
^ t^-cn <4)ovor>)ro
en coco o\ a\ CD a\ o\

rf tN in VO CO CO rH CO
cs inr~ rHr»rooo

ui CO 00 v\ CO CO O'l w\


o vo TJI vo ro in en
H *»« co o vo ro ro ro



coco «tico^ivoinoro<jr>OHflornr>- vo

eocncooococor^invovovovovot^r^ r-


orovovorHcoco o ^< ro oo
co co co co oo in vo vo r- vo r-
H
VD


cocncocococot^invovovovovor^r^ t-



or»ininor»co o\ ro i-(*J





r'COrHinvof'cT\r~HeNcoiHOVDinH(Nr-
i-IHCNCNM(N
-------
      
M
 en

 O
 rH

 •§
 H
         O  0)
             )H
       b.    H


          ft


       O    0)
KJ -rt FH
«* -iH Q
«2Q


 ^  • *4H
 0 rH  (1)
•H O  rl
 n)
 u
-H

 a)

u
                  CntnCNt-HrHCriVDCS
                  r^oooooooocooocn
                        in
                             o vo r- r-
                             VO H VO 00
                  OO CO CO    00 O\ CO 00
                          VD
                          in
                          00



                          CO
                                      vo

                                      00
                                                                                                                                  vo
                     vo ao
                     m PO
                       ro a\ o o
                       in o vo oo
                     CO CO
                             CO
                                   00 CO
                          in
                          o
t~
at
                                OO
                     0>1 H rH (N
                     vo P-
                                  I VO VO

                                   r~ r-
                                                  csLncoeooooivoinooininininooao
                                                  o	
                                                   • r-r-t-i-coaovot--r-r-r»r-f-coeooo
                  vontNr~oooom
                  r-vooaooaioo
                  HrHOlHCNrHfSN

                  r-eoHcsr-oinvo
                        VO
                        in
                    g
                 Q U
                 Q 0)
                 u a
                 EH  i
                  i  00
                 oo r>
                 r- m
                 ro IN
                 r» H
§§
U U
                                                          EL
                                                          X
                                                                               (K  En U> b
                                                                         bbQQQQ
                                                                                         Q Q
                                                                                         UU
                                                        i   i   i  VO    U
                                                                                         co a\
ao cr\
r- co
vo r~
ro m
cs r»

-------
Table 10. Mean BAF™ s from Lake Ontario BSAFs for Salmonids
Chemical
dieldrin
ddt
dde
ddd
mirex
photomirex
g-chlordane .
t-chlordane
c-chlordane
t-nonachlor
c-nonachlor
alpha-hch
gamma-hch
hcbd
ocs
hcb
pcb
1235tcb
1 245tcb
1 234tcb
135tcb
124tcb
123tcb
245tct
236tct
pet
RGBs
5
6
8
12
13
16
17
18
22
25
26
log Kow
5.30
6.45
6.76
6.06
6.89
6.89
6.00
6.00
6.00
6.00
6.00
3.78
3.67
4.84
6.29
5.60
5.11
4.56
4.50
4.59
4.17
3.99
4.10
4.93
4.93
6.36

4.97
5.06
5.07
5.22
5.29
5.16
5.25
5.24
5.58
5.67
5.66
Number
BAFs
4
8
8
4
8
4
4
4
4
4
4
4
4

4
4
4


4








4

4


8
8
8
8
8
Mean
log BAF{d
7.29
7.73
8.66
6.64
8.17
8.76
7.48
7.46
7.84
8.18
6.87
5.30
4.64

7.41
5.70
4.81


3.86








5.78

6.03


5.98
5.60
6.10
6.27
6.75
Mean
BAFjd
1.93e + 07
5.33e + 07
4.56e + 08
4.39e + 06
1.49e + 08
5.74e + 08
3.00e + 07
2.91e + 07
6.95e + 07
1.53e + 08
7.43e + 06
2.00e + 05
4.34e + 04

2.58e + 07
5.01e + 05
6.47e + 04


7.256 + 03








6.02e + 05

1.06e + 06


9.52e + 05
3.96e + 05
1.27e + 06
1.87e + 06
5.57e + 06
                                 63

-------
Table 10. Mean BAFjds from Lake Ontario BSAFs for Salmonids (continued)
Chemical
PCBs
32
33
40
42
44
45
46
49
52
53
63
64
66
74
77
81
82
83
84
85
87
91
92
95
97
99
100
101
105
110
118
119
126
128
129
130
132
log Kow

5.44
5.60
5.66
5.76
5.75
5.53
5.53
5.85
5.84
5.62
6.17
5.95
6.20
6.20
6.36
6.36
6.20
6.26
6.04
6.30
6.29
6.13
6.35
6.13
6.29
6.39
6.23
6.38
6.65
6.48
6.74
6.58
6.89
6.74
6.73
6.80
6.58
Number
BAFs

4
8
8
4
8
4
8
4
8
4
4
4
4
8
4
4
8
4
4
8
4
8
4
4
4
8
4
8
8
8
8
4
4
8
8
4
4
Mean
log BAF{d

5.84
6.18
5.94
6.61
6.54
6.04
5.71
6.82
6.69
7.02
7.25
6.94
7.26
7.51
6.99
7.35
7.17
7.55
7.65
7.58
7.59
7.23
7.64
7.41
6.90
7.54
7.64
7.73
8.34
7.68
8.31
8.33
8.56
8.39
8.03
8.30
7.65
Mean
log BAF{d

6.84e + 05
1.50e + 06
8.726 + 05
4.06e + 06
3.466 + 06
1.09e + 06
5.08e + 05
6.61e + 06
4.90e + 06
1.04e + 07
1.77e + 07
8.80e + 06
1.83e + 07
3.23e + 07
9.68e + 06
2.24e + 07
1.48e + 07
3.536 + 07
4.50e + 07
3.836 + 07
3.89e + 07
1.696 + 07
4.32e + 07
2.55e + 07
7.95e + 06
3.49e + 07
4.40e + 07
5.436 + 07
2.18e + 08
4.746 + 07
2.04e + 08
2.12e + 08
3.63e + 08
2.44e + 08
1.06e + 07
1.98e + 08
4.47e + 07
                                   64

-------
Table 10. Mean BAF's from Lake Ontario BSAFs for Salmonids (continued)
Chemical
PCBs
136
138
141
146
149
151
153
156
158
167
171
172
174
177
178
180
183
185
189
194
195
197
198
201
205
206
207
209
24 + 27
28 + 31
37 + 42-
47 + 48
41+64 + 71
56 + 60
70 + 76
66 + 95
56 + 60 + 81
84 + 92
log Kow

6.22
6.83
6.82
6.89
6.67
6.64
6.92
7.18
7.02
7.27
7.11
7.33
7.11
7.08
7.14
7.36
7.20
7.11
7.71
7.80
7.56
7.30
7.62
7.62
8.00
8.09
7.74
8.18
5.40
5.67
5.80
5.82
5.87
6.11
6.17
6.17
6.19
6.20
Number
BAFs

4
4
8
8
8
8
8
4
4
4
4
4
8
8
8
8
8
8
4
8
4
4
4
8
8
8
8
8
8
8
4
8
4
4
8
4
4
4
Mean
log BAF?

8.39
8.59
8.31
8.34
7.98
8.16
8.52
8.91
8.37
8.27
8.67
8.63
8.40
8.64
8.83
9.05
8.94
8.53
8.72
9.23
8.97
8.50
9.60
8.89
8.75
8.84
8.77
8.13
5.78
6.31
6.76
6.91
6.70
6.76
7.29
7.06
7.06
7.45
Mean
BAF?

2.44e + 08
3.88e + 08
2.03e + 08
2.18e + 08
9.66e + 07
1.45e + 08
3.31e + 08
8.12e + 08
2.32e + 08
1.87e + 08
4.72e + 08
4.24e + 08
2.51e + 08
4.38e + 08
6.80e + 08
1.13e + 09
8.63e + 08
3.36e + 08
5.30e + 08
1.72e + 09
9.32e + 08
3.20e + 08
3.98e + 09
7.706 + 08
5.64e + 08
6.90e + 08
5.92e + 08
1.35e + 08
5.98e + 07
2.06e + 06
5.706 + 06
8.18e + 06
4.97e + 06
5.82e + 06
1.96e + 07
1.146 + 07
1.16e + 07
2.82e + 07
                                  65

-------
Table 10. Mean BAFJs from Lake Ontario BSAFs for Salmonids (continued)
Chemical log Kow
PCBs
87 + 97. 6.29
137 + 176 6.80
138 + 163 6.91
156+171+202 7.18
182 + 187 7.19
157 + 200 7.23
170+190 7.37
195 + 208 7.64
196 + 203 7.65
PCDDs
2378-TCDD 7.02
12378-PeCDD 7.50
123478-HxCDD 7.80
123678-HxCDD 7.80
123789-HxCDD 7.80
1234678-HpCDD8.20
OCDD 8.60
PCDFs
2378-TCDF 6.50
1 2378-PeCDF 7.00
23478-PeCDF 7.00
123478-HxCDF 7.50
123678-HxCDF 7.50
123789-HxCDF 7.50
234678-HxCDF 7.50
1 234678-HpCDF 8.00
1 234789-HpCDF 8.00
OCDF 8.80
Number
BAFs

4
4
4
4
4
4
8
4
8

4
4
4
4
4
4
4

4
4
4
4
4
4
4
4
4
4
Mean
log BAF?

7.81
8.03
8.42
8.44
8.89
8.59
8.98
8.66
8.92

6.95
7.40
7.22
6.83
6.87
6.85
6.63

6.34
6.28
7.14
6.32
6.70
7.23
7.27
5.98
7.53
6.96
Mean
BAF?

6.46e + 07
1.07e + 08
2.64e + 08
2.76e + 08
7.856 + 08
3.86e + 08
9.53e + 08
4.58e + 08
8.27e + 08

9.00e + 06
2.49e + 07
1.656 + 07
6.71e + 06
7.44e + 06
7.16e + 06
4.296 + 06

2.16e + 06
1.89e + 06
1.38e + 07
2.07e + 06
5.07e + 06
1.70e + 07
1.84e + 07
9.476 + 05
3.35e + 07
9.10e + 06
                                   66

-------
                I
                O)
                o
          r-
          co
in
CO
                   o>


                   1
                   11
                   n
                   c -n
0) sr
w O)
CO C


«!
U_ o
< o
CQ O

-------
 0)
u.
m
5
'£
 o

75
     0c
 ^  LJ.
 *:  <
 (0  CO
  I  CD

     1
O>

 0)
 U)
•o  |
     o

     u.
     CD

     D)
     O
                                                     CO
                                                        I
                                                        O)
                                                        O
           00
           (O
                                                     CO
                                                           CM
                                                           U)

                                                           CD
  CO
                                CO
CO
                                                           
-------
                         o
                         O)
                         o
   O)
   CO
                     CO
CO
LO
CO
                            CM
                            in
                            m
                            o



                            i!
                            < *
                            2
                            T3
P CM

5 ^
0) o
IT O

-------
 (0
LL
    TJ
M  ®
CD t>
 = *
 c £
 O u-

ISi
 « ?

-------
 W
1L
•o
0)

Q.

li.
CD
33
'E
 o
 E
 Of
(0
•—   £

 (5   ?•
-i-  LL.

n«
O  ffl
(0
ffl
o
o
    Q.

 « !^
-J O
  . C
CM .2
 CD  £

 3  o

 0)°

-------
 ^-

 (0
 

-------
 ^•^
 ec
CD
 c
 0)
 0)
O
 3
 O
 c

 o
m
 L T5
 O
<
en
OQ
o
 0)
u.
     (Q
CO
O
CD
CD
O
CL
 0)
 O)
il

-------

-------

-------

-------
u.
m £
"-1 o
Ei
2 *
*- Q.
0) 11.
<»
2s
««. CD
0 %
c m
.2 d
   (0
0)
ir *j o)
O c LL
   OO <
   m CD
 • sj? rn
00 co ^

     CD

     £
     CD
     O 2

-------
 0)
UL
ffl
mo

 CO
      CO
CO    CD
CD    C
TJ    CD

SO  c<
to    ®
0   5

^   O
                         Q
                    10    O

                    So  %*
                    a.    Q
                         Q
                         o
                         Q.
                               o
                               0)

                               co
                              "S
                               o
O
 o

 (0



 o
 (0
Li.
m
 c
 (0
 0
     D)
.    CD
LJ.
 -.
 Q)
                                                        o>
                                                        CO
                                                                 oo
                                                           O
                                                           O)
                                                           o ^
                                                              00
                                                        CO
                       00
                                       CO
                                     LO
                                                              08

                                                              I o

                                                              b •$

                                                              E 5
                                                              2^
                                                              v> O
                                                              CQ 08

                                                                 -"
< O
m Q

-------

-------
 (0
LL
CD
m
 U)

il
     m

     CD  co
     O  LI-


 -  I  w
O  2  CQ
^^^^  ^bM  I

-i*|

H«  0  >•
 rt  -^  ^
 o  <  co
 r-  °-
 =  m  o
5  s  2

 £  o  £
 O  c  m
     On  en
     Q>  O
         CD

         O

         CL
LU
                                                      rr,  ^
                                                      05  UL
                                                      CO
                                                      CD
                                                CO

                                                CD

                                                CD

                                                O

                                                CL



                                                O
                                                         CO

                                                         LL
                                                         CD

                                                         O)
                                                         CO
                                                            00
                                                   m
                                                   O

                                                oi
                                                 ' OJ
                                                      o
                                                      oo
          O>
                          oo
CO
55
   CQ
 C CD

 co > «
   0< o
   co z:

2 5|

   S5 n


   II
   ° ofi

   CD a>
   m E
   o "?

   Q *M
   Q^ 8
   LU O

-------
      F.     Bioaccumulation Equivalency Factors (BEFs)

The use of 2,3,7, 8-tetrachlorodibenzo-p-dioxin (TCDD) toxicity equivalency factors
(TEFs) for assessing the total TCDD toxicity risk from complex mixtures of
polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in aquatic
environments is complicated by the wide range of bioaccumulation potentials
associated with these chemicals. Human and wildlife exposures are related to
residues of each chemical in fish and other aquatic organisms ingested as food.
Each congener's TCDD  equivalent risk is proportional to the product of the
congener's TEF times the concentration of the chemical in the food.  The sum of
all the products provides a TCDD equivalence concentration (TEC) for the food
exposure. When it is necessary to relate water or effluent concentrations of
PCDDs and PCDFs to risk estimates for food exposure, the TEC equals the sum of
the products of the water concentration, BAF and TEF for each congener present.
Note that the BAFs and water concentrations have to be based either on freely
dissolved chemical (C") or on total chemical (C^,) in  water (i.e., consistent
definition).
                        i), (BAF<\ (TEF)] = £ [(C^BAF^TEF)]       (30)
BAFs for PCDDs and- PCDFs have not been measured due to the very small water
concentrations present in contaminated ecosystems.  Concentrations of these
chemicals can be measured in surface sediments to provide a measure of the
relative amounts of each chemical present in association with organic carbon of
the ecosystem.  Furthermore, the relative activities of each chemical and TCDD
should be similar for both sediment organic carbon and organic carbon suspended
in water.  The fugacity gradients of each chemical between sediment and water
may or may not be similar, depending on differences in chemical loading to the
water which are not near steady-state with surface sediment.  The biota-sediment
accumulation factor (BSAF) is a direct measure of each chemical's distribution
between sediment organic  carbon and lipid of associated aquatic organisms.  When
PCDDs and PCDFs have similar sources and distribution patterns between water
and sediment, the BSAFs at a site will provide good measures of the
bioaccumulation potentials relative to TCDD or any other chemical for which a  BAF
has been estimated (Cook et al., 1995).  Systems with steady-state distributions
of the chemicals between sediment and  water are most appropriate for these
measurements of relative bioaccumulation potential.

Definitions/Symbols

The following bioaccumulation terms and symbols are used to derive and apply
TCDD bioaccumulation equivalency factors (BEFs). "C" is used for concentration
and "f" for fraction. Subscripts are used to indicate the mass basis for "C" or "f"
(w  = water, f = lipid in tissue, t =  whole tissue wet weight, s = dry sediment,


                                     81

-------
soc = sediment organic carbon, and ssoc = suspended solids organic carbon);
superscripts are used to indicate the water phase of the chemical (fd = freely
dissolved, b = bound to organic carbon in water, and t = total chemical = fd + b;
and subscripts following parentheses indicate the chemical (tcdd =  2,3,7,8-TCDD
and i = the ith chemical).

      bioaccumulati6n factors

                                                                     (3D
                 BAF? = CJC* ,     BAFf = Cf/Cf = frBAI?            (32)


                  BAF? = C,/C* ,     BAF? = C,/C* = ft*BAF?             <33>

      biota-sediment accumulation factor

                          BSAF = Ct IC^ = ^-^                     (34)
                                           ^5*/«

      organic carbon - water partitioning


                                                   ^                (35)
           fraction dissolved

           fraction bound to oc in water   fb =  1-ffd

      TCDD bioaccumulation equivalency factor
                                (BSAFY     (BAF?).
                       (BEFY = -^ - — « — - ^-                  (36)
Calculation of BAFs and TEC from BEFs

The ratio (equation 36) between each PCDD and PCDF congener's BSAF to that of
TCDD will be called the TCDD bioaccumulation equivalency factor (BEF).  Because
BAFs based on freely dissolved chemical in water (BAFfd) are directly proportional
to Kow which varies among PCDDs and PCDFs, the BEF describes only the BAF
relative to TCDD on the basis of organic carbon bound chemical concentration in
water (BAFb). This assumes that the relative amounts of each PCDD and PCDF

                                    82

-------
congener in the organic carbon of surface sediments are the same as in suspended
organic carbon.  The relationship between paniculate organic carbon (POC),
dissolved organic carbon  (DOC), Kow and ffd is presented in equation 36. the
importance of each chemical's Kow should be evident. The BEF can be used to
calculate (BAFJ), and (BAFfd)j.  (BAF}),s estimated from BEFs, under the condition of
similar sediment/ water fugacity ratios for each chemical, may be  used to predict
bioaccumulation by pelagic fish from estimated C "s regardless of site-specific
differences in chemical distribution between sediment and water.

                                BAF? = BAF\lfb
                      (BEF)  •           .                               (38,
so,

                                  lnfC\  /DvlE'f'V   lf\
                                                                         (39)
and,

                         /DA TT^ \    t DI7ZT \ f DA E**^\   If \ (f \
                                                                        (40)
because (fb)i(f,d,todd/(ffd)i(fb,tcdd =  (Kow)i/(Kow)tcdd :
A TCDD TEC can be calculated on the basis of wet tissue residue (TECt)tcdd or lipid
normalized residue (TEC|)todd; water concentration of total chemical  (TEC*)tcdd or
freely dissolved chemical (TEC*d)tcdd.  When bioaccumulation is to be predicted on
the basis of freely dissolved chemical (Cj|,d), the relative differences in BAFfds for
PCDD and PCDF congeners will be less than for their BAPs.  This is because f,ds
for the higher chlorinated, more hydrophobic congeners are less than ffd for TCDD.
When the TEC is based on concentration of chemicals in tissue, TEC* = TEC'd and
TEC} = TEC{d.  Thus if (BAF]d)todd is the reference bioaccumulation  factor:
                                     83

-------
                                              )^ (Kow)t
                                              'tcdd
                                           (42)
(C* ),-
                                                       ; (1EF),.
                                              I tcdd
(43)
                                                                       (44)
Great Lakes BEFs

Lake Ontario sediment and fish residue data (Lodge et ai., 1994) provide a basis
for calculation of BEFs. However, very few PCDDs and PCDFs measured as
sediment contaminants are detectable in fish tissue. Table 11 below provides
estimated BEFs calculated from lake-wide average concentrations of lexicologically
important PCDDs and PCDFs in surface sediment and lake trout samples collected
in 1987 for the EPA Region II Lake Ontario TCDD Bioaccumulation Study.  Lake
Ontario conditions in 1987 involve sediment as the principal source of these
chemicals to the water and food chain. The BSAFs if  measured under conditions
of steady-state between external chemical loading, water, food chain and surface
sediment would be somewhat larger but BEFs should be similar. Lake Ontario
sediment cores also demonstrate that all PCDD and PCDF congener concentrations
have similar temporal trends during the past four decades and all have water
column concentrations that are strongly controlled by sediment resuspension due
to large declines in loading from sources external to the lake.  Limited comparison
to BEFs calculated from data obtained for other ecosystems confirms these
bioaccumulation potential differences and suggests that this BEF set would be
predictive of bioaccumulation differences for PCDDs and PCDFs for fish in
ecosystems outside the Great Lakes.  Similar results are likely for other persistent
bioaccumulative organic chemicals such as PCBs and chlorinated pesticides.

BEFs for Calculation of TCDD Toxicity  Equivalence Concentrations in Water in
Relation to a GLWQI TCDD Criterion to Protect Human Health

BEFs are measures of bioaccumulation differences between chemicals but do not
incorporate differences in bioavailability attributable to partitioning in water.  Use
of BAFfds and C^ds eliminates bioavailability variation due to partitioning of
chemicals with different hydrophobicities to organic carbon in water. When BAFs,
based on the concentration of total TCDD in water (BAFjs) are used, site-specific
bioavailability differences are incorporated into the BAFJ.  The final Guidance
                                     84

-------
utilizes TCDD BAFJs for protection of human health. Trophic levels three and four,
each with a different fraction lipid, are considered for human exposure.  The TCDD
BEFs presented in Table 11 are based on lake trout (trophic level four).  TCDD
BEFs for trophic level 3 fish such as smelt are not likely to be significantly
different, however they could be calculated and used in the same manner as the
trophic level four TCDD BEFs.  The choice of specific dissolved (DOC) and
paniculate (POO organic carbon concentrations in water for calculation  of TCDD
BAFJs for human health must be considered when applying BEFs to calculate
TCDD toxicity equivalence concentrations in water on the basis of concentrations
of total chemical in water, (TEC^)tcdd, from concentrations of each PCDD and PCDF
congener:
                                    (TEF)X (BEF)X
(45)
                                                  l'-./tytedd
The human health BAFJs for TCDD were calculated for default conditions of DOC
= 2.0 mg/L and POC = 0.04 mg/L.  If the product of (BEF)X times (1-ffd)x/(1-ffd)tcdd
is defined as the BEF for health criteria based on total chemical concentration
(BEF^), equation 46 can be simplified to equation 47.  Table 11 contains TEFs and
BEFJ,s for calculating human health (TECX)tcdds from C^s, either measured or
estimated for the default DOC and POC conditions. TCDD BEF^,s differ only
slightly from TCDD BEFs in proportion to differences in hydrophobicity.
                                  (Cfw)x (TEF)X
(46)
Table 11.  TCDD Bioaccumulation equivalency factors (BEFs) and TCDD
bioaccumulation equivalency factors for human health criteria for total chemical
concentration in water (BEF^s).  The BEFs and BEF,J,s are derived for lexicologically
important PCDDs and PCDFs from lakewide averages of concentrations in Lake
Ontario  lake trout and surface sediment in depositional areas.
Congener
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1, 2,3,7, 8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
OCDD
2,3,7,8-TCDF
log KOW"
7.02
7.50
7.80
7.80
7.80
8.20
8.60
6.5b
BSAF
0.059
0.054
0.018
0.0073
0.0081
0.0031
0.00074
0.047
TCDD BEF
1.0
0.92
0.31
0.12
0.14
0.051
0.012
0.80
TCDD BEF^
1.0
1.13
0.40
0.16
0.18
0.072
0.017
0.48
                                     85

-------
1 ,2,3,7, 8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF

2,3,4,6,7,8-HxCDF-
1,2,3,7,8,9-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDF
7.0b
7.0"
7.5b
7.5b

7.5b
7.5b
8.0b
8.0b
8.80
0.013
0.095
0.0045
0.011

0.040
0.037
0.00065
0.023
0.001
0.22
1.6
0.076
0.19

0.67
0.63
0.011
0.39
0.016
0.22
1.59
0.094
0.23

0.84
0.78
0.015
0.52
0.023
      ' Burkhard and Kuehl, 1987.
      b Estimated based on degree of chlorination and Burkhard and Kuehl, 1987.

Example of a (TEC)tcdd Calculation Using the BEF Method

Projected PCDD and PCDF loadings to a Great Lake result in estimated water
concentrations (C^) of 0.0001, 0.0008, 0.0002, 0.0008 and 0.02 pg/ml for
2,3,7,8-TCDD, 2,3,7,8-TCDF, 1,2,3,7,8-PeCDD, 2,3,4,7,8-PeCDF and
1,2,3,4,6,7,8-HpCDD, respectively.  The concentration of POC is 0.2 mg/L, DOC
is 2.0 mg/L, so the Cj^s for each congener are 0.00002, 0.0006, 0.000015,
0.00016, and 0.0003 pg/L, respectively.  The BAFjd for TCDD is estimated to be
7.85x106 and TEFs are 1.0, 0.1, 0.5 0.5 and 0.01 for each congener,
respectively. At 9% lipid (f,=0.09), the 2,3,7,8-TCDD BAh^91 = 7.07x105. From
equation 46 the TCDD toxicity equivalency concentration for fish with f, =0.09,
(TEC.09,)tcdd, is calculated to be:

      (TEC09,)tcdd  = (7.07x105)[(0.00002)(1.0)(10.5x106)(1.0)/10.5x106 +
(0.0006)(0.8)(0.63x106){0.1 )/10.5x106 +
(0.000015M0.92H31.6x1 Oe)(0.5)/10.5x106 +
(0.00016)(1.6)(10x106)(0.5)/10.5x106  +
(0.0003M0.05M158x106)(0.01)/10.5x106]  =  14.4 + 20.4 +  1.5 + 0.86 + 1.6
=  38.8 pg TCDD eq./g wet fish.

In this hypothetical example 2,3,7,8-TCDD contributes 37% of the TEC. Without
use of the BEF approach (all BAF^9,s = 7.07x105), the TEC is calculated to be
14.4  + 42.4  + 0.5 + 5.7 + 21.2  = 84.2 pg TCDD eq./g wet  fish with TCDD
contributing only 17%.  The overestimation of bioaccumulation for TCDF, PeCDF
and HpCDD leads to a greater TEC estimate.  Since there appears to be an
association between TEFs and BEFs (i.e., the more toxic congeners are the most
                                    86

-------
bioaccumulative, primarily due to slower rates of biotransformation), additional
data suitable for validating the BSAFs used to calculate the BEFs are needed.

IV.   DETERMINATION OF BAFs FOR INORGANIC CHEMICALS

The lipid-BAF relationship does not apply to the determination of BAFs for
inorganic chemicals.  BAF and BCF data for inorganics are not as transferable from
one species, or one tissue, to another as organic data. Bioaccumulation of some
trace metals is substantially greater in internal organs than muscle tissue. For
example, BCFs for rainbow trout liver, kidney, gut and skin, and muscle exposed to
cadmium for 178 days were about 325, 75,  7, and 1 respectively (Giles 1988).
Merlin! and Pozzi (1977) reported that lead bioconcentrated 30 times more in
bluegill liver than in bluegill muscle tissue after eight days. They reported a BCF for
muscle tissue of 0.46.

Because bioaccumulation can differ dramatically between tissues, BAFs or BCFs for
edible tissue of fish should be  used for BAFs to calculate human health criteria.
Similarly, BAFs or BGFs for whole body of fish should be used for the BAFs used
to calculate wildlife criteria.

BAFs or BCFs for inorganic chemicals measured in plants or invertebrate animals
might be one or more orders of magnitude greater than BAFs or BCFs for the edible
tissue of fish (see Table 5 in the EPA criteria documents for cadmium, copper, lead
and nickel; USEPA 1985A, USEPA 1985B, USEPA 1985C, and  USEPA 1986).  For
this reason plant or invertebrate BAFs and BCFs should not be used to calculate
human health criteria and values.  If site-specific conditions warrant, and the
resulting criteria are more stringent, plant or invertebrate BAFs or BCFs could be
used to calculate wildlife criteria.

Mercury and certain  other metals are subject to methylation through microbial
action in nature.  The organo-metallic form of the metal, especially methyl mercury,
is highly bioaccumulative in the muscle tissue of fish (Grieb et al. 1990).

V.    CALCULATION OF BASELINE BAFs FOR ORGANIC CHEMICALS

      A.   Baseline BAF from a Field-Measured BAF

A baseline BAF shall be calculated from a field-measured BAF of acceptable quality
using the following equation:
               Baseline BAF =
                                Measured BAFy
- 1
where:      BAF|  =    BAF based on total concentration in tissue and ambient


                                    87

-------
                        water.
            f i           =    fraction of the tissue that is lipid.
            ffd           =    fraction of the total chemical that is freely
                             dissolved in the ambient water.

The trophic level to which the baseline BAF applies is the same as the trophic level
of the organisms used in the determination of the field-measured BAF.  For each
trophic level, a species mean baseline BAF shall be calculated as the geometric
mean if more than one measured baseline BAF is available for a given species. For
each trophic level, the geometric mean of the species mean baseline BAFs shall be
calculated.

If a baseline BAF based on a field-measured BAF is available for either trophic level
3 or 4, but not both; the baseline BAF for the other trophic level shall be calculated
using the ratio of the FCMs that are obtained by linear interpolation from Table B-1
for the chemical.

      B.    Baseline BAF from Field-Measured BSAF Methodology

A  baseline BAF for organic chemical "i" shall be calculated from a field-measured
BSAF of acceptable quality  using the following equation:

                 ( Baseline BAF)i - (BAF,"), .
where:      (BAF]d)r     =     BAF based on the measurement of freely dissolved
                              reference chemical in the water column.
            (BSAF)j     =     BSAF for chemical "i".
            (BSAF)r-     =     BSAF for the reference chemical "r".
            (Kow)j =     octanol- water partition coefficient for chemical "i".
            (Kow)r =     octanol-water partition coefficient for the reference
                        chemical "r".
The trophic level to which the baseline BAF applies is the same as the trophic level
of the organisms used in the determination of the BSAF.  For each trophic level, a
species mean baseline BAF shall be calculated as the geometric mean if more than
one baseline BAF is predicted from BSAFs for a given species.  For each trophic
level, the geometric mean of the species mean baseline BAFs shall be calculated.

If a baseline BAF based  on a measured BSAF is available for either trophic level 3
or 4,  but not both, the baseline BAF for the other trophic level shall  be calculated
using the ratio of the FCMs that are obtained by linear interpolation  from Table 3
for the chemical.
                                     88

-------
      C.    Baseline BAF from a Laboratory-Measured BCF

A baseline BAF for trophic level 3 and a baseline BAF for trophic level 4 shall be
calculated from a laboratory-measured BCF of acceptable quality and a FCM using
the following equation:
               Baseline BAF = (FCM) ( Measured baseline BCF)


              Baseline BAF - (FCM)
where:      BCF|  =    BCF based on total concentration in tissue and water.
            fj          =    fraction of the tissue that is lipid.
            ffd          =    fraction of the total chemical that is freely
                             dissolved in the ambient water.
            FCM  =    the food-chain multiplier obtained from Table 3 by linear.
                       interpolation for trophic level 3 or 4, as necessary.

For each trophic level, a species mean baseline BAF shall be calculated as the
geometric mean if more than one baseline BAF is predicted from laboratory-
measured  BCFs for a given species. For each trophic level, the geometric mean of
the species mean baseline BAFs shall be calculated.

      D.    Baseline BAF from a Octanol-Water Partition Coefficient

A baseline BAF for trophic level 3 and a baseline BAF for trophic level 4 shall be
calculated from a Kow of acceptable quality and a FCM using the following
equation:

         Baseline BAF = ( FCM )( predicted baseline BCF) = (FCM)(KOW)
where:      FCM  =    the food-chain multiplier obtained from Table 3 by linear
                       interpolation for trophic level 3 or 4, as necessary.
            Kow        =    octanol-water partition coefficient.

VI.    CALCULATION OF BASELINE BAFs FOR INORGANIC CHEMICALS

For most inorganic chemicals, the baseline BAFs for trophic levels 3 and 4 are both
assumed to equal the BCF determined for the chemical with fish (i.e., the FCM is
assumed to be 1 for both trophic levels 3 and 4).  However, a FCM greater than 1
might be applicable to some metals, such as  mercury, if, for example, an
organometallic form of the metal biomagnifies.
                                    89

-------
VII.   REFERENCES

ASTM. 1990.  Standard practice for conducting bioconcentration tests with fishes
      and saltwater bivalve molluscs. Designation E 1022 - 84. Pages 606-622
      in Annual book of ASTM standards. Section 11, Water and Environmental
      Technology, Volume 11.04. American Society for Testing and Materials.

Barber, M.G., L.A. Suarez and R.R. Lassiter.  1988.  Modeling bioconcentration of
      nonpolar organic pollutants by fish.  Environ. Toxicol. Chem.  7: 545-558.

Barber, M.G., L.A. Suarez and R.R. Lassiter.  1991.  Modeling bioaccumulation of
      organic pollutants in fish with an application to PCBs in Lake Ontario
      salmonids. Can. J. Fish. Aquat. Sci. 48: 318-337.

Baron, M.G. 1990. "Bioconcentration."  Environ. Sci. Technol., 24, 1612-1618.

Bligh, E.G. and W.J. Dyer. 1959.  A rapid method of total lipid extraction and
      purification. Can. J. Biochem.  Physiol. 37: 911-917.

Borgmann, U. and D.M. Whittle. 1991. Contaminant concentration trends in Lake
      Ontario lake trout (Salvelinus namavcush): 1977 to  1988. J. Great Lakes
      Res. 17: 368-381.

Burkhard, L.P.  and D.W. Kuehl. 1986. N-octanol/water partition coefficients by
      reverse phase liquid chromatography/mass spectrometry for eight
      tetrachlorinated planar molecules. Chemosphere 15:163-167.

Carter, C.W., and Suffet,  I.H.  1982.  "Binding of DDT to  dissolved humic
      materials."  Environ. Sci. Technol., 16, 735-740.

Chin, Y., and P.M. G'schwend.  1992.  "Partitioning of polycyclic aromatic
      hydrocarbons to marine porewater organic colloids." Environ. Sci. Technol.,
      26, 1621-1626.

Chiou, G.T.  1985, "Partition coefficients of  organic compounds in  lipid-water
      systems and correlation with fish bioconcentration factors."  Environ. Sci.
      Technol., 19, 57-62.

Connell, D.W.  1988. Bioaccumulation behavior of persistent organic chemicals
      with aquatic organisms.  Pages 117-159 In Review  of Environmental
      Contamination and  Toxicology, Volume 101.

Connolly, J.P.  and C.J. Pedersen. 1988.  A thermodynamic-based evaluation of
      organic chemical accumulation in aquatic organisms. Environ. Sci. Technol.


                                    90

-------
      22: 99-103.

Cook, P.M., R.J. Erickson, R.L. Spehar, S.P. Bradbury, and G.T. Ankley. 1993.
      "Interim report on data and methods for assessment of 2,3,7,8-
      tetrachlorodibenzo-p-dioxin risks to aquatic life and associated wildlife."
      EPA/600/R-93/055. U.S. Environmental Protection Agency, Environmental
      Research Laboratory Duluth, MN.

Cook, P.M. Memorandum to Nelson Thomas. 1995.

Dean, K.E., M.M. Shafer,  and D.E. Armstrong. 1993.  "Particle-mediated transport
      and fate of a hydrophobic organic contaminant in southern Lake Michigan:
      the role of major water column particle species.  J. Great Lakes Res., 19,
      480-496.

De Bruijn, J.,  F.Busser, W. Seinen, and J. Hermans.  1989, "Determination of
      octanol/water 'partition coefficients for hydrophobic organic chemicals with
      the "slow-stirring" method."  Environ. Toxicol. Chem., 8, 499-512.

De Wolf, W.,  J.H.M. de Bruijn, W. Seinen, and J. L.M. Hermens.  1992. "Influence
      of biotransformation on the relationship between bioconcentration factors
      and octanol-water partition coefficients."  Environ. Sci. Techno!., 26, 1197-
      1201.

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

Eadie, B.J., N.R. Morehead, and P.P. Landrum. 1990.  "Three-phase partitioning
      of hydrophobic organic compounds in  Great Lakes waters."  Chemosphere,
      20, 161-178.*

Eadie, B.J., N.R. Morehead, J. Val Klump, and P.P. Landrum.  1992.  "Distribution
      of hydrophobic organic compounds between dissolved and paniculate
      organic matter in Green Bay waters."  J. Great Lakes Res.,  18, 91-97.

Ellgehausen, H., J.A. Guth and  H.O. Esser. 1980.  Factors determining the
      bioaccumulation potential of pesticides in the individual compartments of
      aquatic food chains. Ecotox. Environ. Safety 4: 134-157.

Evans, M.S., G.E. Noguchi and  C.P. Rice. 1991. The biomagnification of
      polychlorinated biphenyls, toxaphene,  and DDT compounds in a Lake
      Michigan offshore food web.  Arch. Environ. Contam. Toxicol. 20: 87-93.


                                    91

-------
Flint, R.W. 1986.  "Hypothesized carbon flow through the deep water Lake Ontario
      food web."  J. Great Lakes Res., 12, 344-354.

Giles, M.A. 1988.  Accumulation of cadmium by rainbow trout, Salmo gairdneri.
      during extended exposure. Can. J. Aquat. Sci. 45: 1045-1053.

Gobas, F.A.P.C.  1993.  "A model for predicting the bioaccumulation of
      hydrophobic organic chemicals in aquatic food-webs:  application to Lake
      Ontario." Ecological Modelling, 69,  1-17.

Gobas, F.A.P.C., K.E. Clark, W.Y. Shiu, and D. Mackay.  1989.  "Bioconcentration
      of polybrominated benzenes and biphenyls and related superhydrophobic
      chemicals in fish:  role of bioavailability and elimination into feces."
      Chemosphere, 8, 231-245.

Grieb, T.M., C.T. Driscoll, S.P. Gloss,  C.L. Schofield, G.L. Bowie and D.B.  Porcella.
      1990.  Factors affecting mercury accumulation in fish in the upper Michigan
      Peninsula.  Environ. Toxicol. Chem.  9: 919-930.

Gschwend, P.M., and S. Wu.  1985. "On the constancy of sediment-water
      partition coefficients of hydrophobic organic pollutants."  Environ. Sci.
      Technol., 19, 90-96.

Hamelink, J.L., R.C. Waybrant and R.C. Ball. 1971.  A proposal: exchange
      equilibria control the degree chlorinated hydrocarbons are biologically
      magnified in lentic environments. Trans. Amer. Fish. Soc. 100: 207-214.

Hassett, J.P., and M.A. Anderson.  1979.   "Association of hydrophobic organic
      compounds with dissolved organic matter in aquatic systems."  Environ. Sci.
      Technol., 13, 1526-1529.

Hawker, D.W. and D.W. Connell. 1988. "Octanol-water partition coefficients of
      polychlorinated biphenyl congeners." Environ. Sci.  Technol. 22, 382-387.

Herbert, B.E., P.M. Bertsch, and J.M.  Novak.  1993.  "Pyrene sorption by  water-
      soluble organic carbon."  Environ. Sci. Technol., 27,  398-403.

Isnard, P., and S.  Lambert.  1988.  "Estimating  bioconcentration factors from
octanol-water    partition coefficients and aqueous solubility." Chemosphere,
17, 21-34.

Landrum, P.P., S.R.  Nihart, B.J.  Eadie, and W.S. Gardner.  1984. "Reverse-Phase
      separation method for determining pollutant binding to Aldrich humic and
      dissolved organic carbon of natural waters." Environ. Sci. Technol., 18,


                                     92

-------
      187-192.

Leo, A.J.  Unified medchem software, version 3.53, Pomona Medicinal Chemistry
      Project,  1988.

Lodge, K., P.M. Cook, D.R. Marklund, S.W. Kohlbry, J. Libal, C. Harper, B.C.
      Butterworth and A.G. Kizlauskas. 1994.  Accumulation of polychlorinated
      dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in sediments and
      fishes of Lake Ontario.  In preparation.

Mackay, D.  1982.  "Correlation of bioconcentration factors." Environ. Sci.
      Technol., 16, 274-278.

McCarthy, J.F., and B.D. Jimenez.  1985. "Interaction between polycyclic
      aromatic hydrocarbons and dissolved humic material: binding and
      dissociation." Environ. Sci. Technol.,  19, 1072-1076.

McKim, J., P. Schmieder, and G. Veith. 1985.  "Absorption dynamics of organic
      chemical transport across trout gills as related to octanol-water partition
      coefficient."  Toxicol. Applied Pharmacol., 77,  1-10.

Merlini, M. and G. Pozzi. 1977.  Lead and freshwater fishes: part I - lead
      accumulation and water pH.  Environ Pollut. 12: 167-172.

Metcalf, R.L., J.R. Sanborn, P.Y. Lu and D. Nye. 1975.  Laboratory model
      ecosystem studies of the degradation and fate of radiolabeled tri-tetra-, and
      pentachlorobiphenyl compared with DDE. Arch. Environ Contam. Toxicol. 3:
      151-165.

Neely, W.G., D.R. Branson and G.E. Blau.  1974.  The use of the partition
      coefficient to measure the bioconcentration potential of organic chemicals in
      fish. Environ. Sci.  Technol. 8: 1113-1115.

Niimi, A.J. 1985.  Use of laboratory studies in assessing the behavior of
      contaminants in fish inhabiting natural ecosystems.  Water Poll. Res. J.
      Canada 20: 79-88.

Niimi, A.J. and G.P. Dookhran. 1989.  Dietary absorption efficiencies and
      elimination rates of polycyclic aromatic hydrocarbons (PAHs) by rainbow
      trout (Salmo gairdneri).  Environ. Toxicol. Chem. 8: 719-722.

Niimi, A.J., H.B. Lee and G.P. Kissoon. 1989. Octanol/Water partition coefficients
      and bioconcentration factors of chloronitrobenzenes in rainbow trout (Salmo
      gairdneri).  Environ. Toxicol. Chem.  8: 817-823.


                                    93

-------
Oliver, B.C. and A.J. Niimi. 1983.  Bioconcentration of chlorobenzenes from water
      by rainbow trout: correlations with partition coefficients and environmental
      residues. Environ. Sci Technol. 17: 287-291.

Oliver, B.G. and A.J. Niimi. 1985.  Bioconcentration factors of some halogenated
      organics for rainbow trout:  limitations in their use for prediction of
      environmental residues. Environ. Sci. Technol.  19: 842-849.

Oliver, B.G. and A.JS Niimi.  1988, Trophodynamic analysis of polychlorinated
      biphenyl congeners and other chlorinated hydrocarbons in the Lake Ontario
      ecosystem." Environ. Sci. Technol., 22, 388-397.

Pereira, W.E., C.E. Ostad, C.T. Chiou, T.I. Brinton, L.B. Barber II, O.K. Demcheck,
      and C.R. Demas. 1988, "Contamination of estuarine water, biota, and
      sediment by halogenated organic compounds: a field study."  Environ. Sci.
      Technol., 22, 772-778.

Rasmussen, J.B., D.J. Brown, D.R.S. Lean and J.H. Carey. 1990. Food chain
      structure in Ontario lakes determines PCB levels in lake trout (Salvelinus
      namaycush) and other pelagic fish. Can. J. Fish. Aquat. Sci. 47:
      2030-2038.

Schmitt,  C.J., J.L. Zajicek and P.M. Peterman. 1990.  National contaminant
      biomonitoring program:  residues of organochlorine chemicals in U.S.
      freshwater fish, 1976-1984.  Arch. Environ.  Contam. Toxicol.  19:
      748-781.

Schuuman, G.  and W. Klein. 1988.  Advances in bioconcentration prediction.
      Chemosphere 17: 1551-1574.

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. NTIS

Swackhamer, D.L. and R.A. Hites. 1988. Occurrence and bioaccumulation of
      organochlorine compounds in fishes from Siskiwit Lake, Isle Royale, Lake
      Superior.  Environ. Sci. Technol. 22: 543-548.

Thomann, R.V. 1989. Bioaccumulation model of organic chemical distribution in
            aquatic food  chains.  Environ. Sci. Technol. 23: 699-707.

Thomann, R.V. and J.P. Connolly. 1984. Model of PCB in the Lake Michigan lake
      trout food chain. Environ. Sci. Technol.  18:65-71.

                                     94

-------
West et al. May 1993.  1991-1992 Michigan Sport Anglers Fish Consumption
      Study - Final Report to the Michigan Great Lakes Protection Fund, Michigan
      Department of Natural Resources. University of Michigan, School of Natural
      Resources, Natural Resource Sociology Research Lab. Technical  Report #6.

U.S. EPA. 1985A.  Ambient water quality criteria for cadmium - 1984.  EPA
      440/5-84-032.  U.S.EPA, Office of Water, Washington, D.C.

U.S. EPA. 1985B.  Ambient water quality criteria for copper - 1984. EPA
      440/5-84-03V.  U.S. EPA, Office of Water, Washington, D.C.

U.S. EPA. 1985C.  Ambient water quality criteria for lead - 1984. EPA
      440/5-84-027.  U.S. EPA, Office of Water, Washington, D.C.

U.S. EPA. 1986. Ambient water quality criteria for nickel - 1986. EPA
      440/5-86-004.  U.S. EPA, Office of Water, Washington, D.C.

U.S. EPA. 1990.  Lake Ontario TCDD Bioaccumulation Study - Final Report, U.S.
      Environmental Protection Agency, Region II, New York, NY.

U.S. EPA. 1991 A.  Assessment and control of bioconcentratable contaminants in
      surface waters.  Draft. U.S. EPA, Office of Water, Washington, D.C.

U.S. EPA. 1991B.  Technical support document for water quality-based toxics
      control.  EPA/505/2-90-001 U.S. EPA, Office of Water, Washington, D.C.

U.S. EPA. Consumption Surveys for Fish and Shellfish. A Review and Analysis of
      Survey Methods.  Office of Water. EPA 822\R-92-001.

U.S. EPA. 1995. Trophic Level and Exposure Analyses for Selected Piscivorous
      Birds and Mammals. Volume I: Analyses for Species on the Great Lakes
      Basin and of the Great Lakes Basin and Volume III: Appendices.

Veith, G.D., D.L. DeFoe and B.V.  Bergstedt. 1979.  Measuring and estimating the
      bioconcentration factor of chemicals in fish. J. Fish. Res. Bd. Canada 36:
      1040-1048.

Veith, G.D. and P. Kosian. 1983.  Estimating bioconcentration potential from
      octanol/water partition coefficients.  Chapter 15 in PCBs in the Great Lakes.
      Mackay, D.,  R-. Patterson, S. Eisenreich, and M. Simmons (eds).  Ann Arbor
      Science.

Yin, C. and J.P. Hassett.  1986, "Gas-partitioning approach for laboratory and field
      studies of mirex fugacity in water."  Environ. Sci.  Technol., 20, 1213-1217.


                                    95

-------
Yin, C. and J.P. Hassett. 1989,  "Fugacity and phase distribution of mirex in
      Oswego River and Lake Ontario waters."  Chemosphere, 19, 1289-1296.
                                    96

-------
Appendix A.       Procedure for Deriving Recommended Values for Log KOW

Measured values of Kow have been obtained using the slow-stir, generator-column,
and shake-flask techniques.  The shake-flask technique has been reported to be
acceptable only for chemicals whose log Kows are less than 4 (Karickhoff et al.
1979; Konemann et al. 1979; Braumann and Grimme 1981; Harnisch et al. 1983;
Brooke et al.  1990).  Brooke et al. (1986) reported that the shake-flask technique
is acceptable for chemicals whose log Kows are less than 5, whereas Chessells et
al. (1991) stated that this technique is acceptable for values of log Kow up to about
5.5.  Although the three techniques seem to give about the same values on the
average up to at least a log Kow of 4.5, the slow-stir and generator-column
techniques are given preference in the final Guidance for chemicals whose log Kows
are greater than 4; because phase separation is always a potential problem with
the shake-flask technique, it  is possible that the slow-stir and generator-column
techniques should also  be given preference for chemicals whose log KOWS are less
than 4.

Predicted values of Kow have been based on reverse-phase high performance liquid
chromatography (RPLC) and  thin-layer chromatography (TLC).  Generally, results
obtained using the RPLC technique should  be used in the final Guidance if the
calibration curve is based on measured values of Kow, but not if the calibration
curve is based on values calculated on a basis such as fragment or substituent
constants; the actual values  used in the calibration curve are more important,
however, than the source of the values. Because it is based on more
measurements and seems to have a better scientific  basis, the version of RPLC
that includes extrapolation to zero percent solvent is given preference over the
version that does not include extrapolation to zero percent solvent.  Values based
on TLC are not considered because this technique has not been adequately
investigated.

Calculated values of *KOW have been obtained using a variety of methods, but the
most widely used is the computer program CLOGP.   Calculated values of Kow
should be used only as  a last resort.

Because  of potential interference due to radioactivity associated with impurities,
values of Kow that are determined by measuring radioactivity in water and/or
octanol are less reliable and should be used only as a last resort.

Thus, values  of Kow are given priority based on the technique used as follows:
                                    A-1

-------
Log Kow <  4:     Priority      Technique
                     1        Slow-stir.
                     1        Generator-column.
                     1        Shake-flask.
                     2        Reverse-phase liquid chromatography on C18
                              with extrapolation to zero percent solvent (RPLC-
                              E).
                     3        Reverse-phase liquid chromatography on C18
                              without extrapolation to zero percent solvent
                              (RPLC).
                     4        Calculated by the CLOGP program.

Log Kow > 4:     Priority      Technique

                     1        Slow-stir.
                     1        Generator-column.
                   •  2        Reverse-phase liquid chromatography on C18
                              with extrapolation to zero percent solvent (RPLC-
                              E).
                     3        Reverse-phase liquid chromatography on C18
                              without extrapolation to zero percent solvent
                              (RPLC).
                     4        Shake-flask.
                     5        Calculated by the CLOGP program.

Values that seem to be different from the rest should be considered outliers and
not used.

For each chemical the available value of log Kow with the highest priority should be
the recommended value, except that if more than one such value is available, the
arithmetic mean of log Kows or the geometric mean of Kows should be the
recommended value.- In some cases, another value may be the recommended
value if adequately justified.

A Kow can describe the partitioning of an individual chemical more usefully than it
can describe the partitioning of a mixture, such as toxaphene, PCBs,  or chlordane.
When a measured value is not available for a mixture, a recommended value may
be derived by using  the value for a major component or by calculating a weighted
or unweighted average of the values for various components.  If an unweighted
average is used, the arithmetic average of values of log Kow may be used or the
geometric mean of values of Kow may be used.

Measured and predicted values should be taken from the original publications.

                                    A-2

-------
Values may be referenced to Medchem in some cases, preferably only if Medchem
associates the value with Hansch, Leo, and/or Pamona College; all such values are
assumed to have been determined by the shake-flask technique.

Recommended values for log Kow should be given to three decimal digits (e.g.,
4.321) because these are intermediate values in the calculation of BAFs, criteria,
and permit limits.
References

Braumann, T., and L.H. Grimme. 1981. Determination of Hydrophobic Parameters
    for Pyridazinone Herbicides by Liquid-Liquid Partition and Reversed-Phase High-
    Performance Liquid Chromatography. J. Chromatog. 206:7-15.

Brooke, D.N., A.J. Dobbs, and N. Williams.  1986.  OctanohWater Partition
    Coefficients (P): Measurement, Estimation, and Interpretation, Particularly for
    Chemicals with P >  105.  Ecotoxicol Environ. Safety 11:251-260.

Brooke, D., I. Nielsen, J. de Bruijn, and J. Hermens. 1990. An Interlaboratory
    Evaluation of the Stir-Flask Method for the Determination of Octanol-Water
    Partition Coefficient (Log Pow).  Chemosphere 21:119-133.

Chessells,  M., D.W. Hawker, and D.W. Connell.  1991.  Critical Evaluation of the
    Measurement of the  1 -Octanol/Water Partition Coefficient of Hydrophobic
    Compounds. Chemosphere 22:1175-1190.

Harnisch, M., H.J. Mockel, and G. Schulze.  1983.  Relationship Between Log Pow
    Shake-Flask Values and Capacity Factors Derived  from Reverse-Phase High-
    Performance Liquid Chromatography for /7-Alkylbenzenes and Some OECD
    Reference Substances. J. Chromatog. 282:315-332.

Karickhoff, S.W., D.S. Brown, and T.A Scott. 1979.  Sorption of Hydrophobic
    Pollutants on Natural Sediments. Water Research 13:241-248.

Konemann, H., R. Zelle, F. Busser, and W.E. Hammers. 1979. Determination of
    Log PQCT Values .of Chloro-Substituted Benzenes, Toluenes and Anilines by
    High-Performance Liquid Chromatography on ODS-Silica.  J. Chromatog.
    178:559-565.
                                    A-3

-------
Appendix B. Derivation of Recommended Values of Log Kow

Appendix A describes the procedure for deriving recommended values of log Kow
that are used for chemicals in the final Guidance.  Various techniques that can be
used to measure, predict, and calculate the log Kow of a chemical are given
priorities in Appendix A.  This Appendix B presents the application of the procedure
to various chemicals*and gives the recommended value of log Kow that is used in
the final Guidance for each of the chemicals.

It was inconvenient to repeatedly acknowledge duplicate publication of two sets of
values below; only the original investigators are cited in both cases. Banerjee et al.
(1980) is cited for values that are also reported by Veith et al. (1980). Similarly,
de Bruijn et al. (1990) is cited for values that are also reported by Brooke et al.
(1990).

Except as  noted, all calculated values of log Kow were obtained using version 3.4
of CLOGP.

The notation "(R)n indicates that the value was based on measurement of
radioactivity.

Benzene [CAS#: 71-43-2]
    The values that have the highest  priorities are:
        2.186  Slow-stir            de  Bruijn et al. 1989
        2.13    Generator-column    Miller et al. 1985
        2.114  Shake-flask       Karickhoff et al. 1979
        2.13    Shake-flask       Medchem
        2.130  Shake-flask       Watari et al. 1982
        2.20    RPLC-E          Hammers et al. 1982
        2.23    RPLC-E          Harnisch et al.  1983
        2.18    RPLC            Miyake  and Terada 1982
        2.48    RPLC            Swannetal.  1983
        2.25    RPLC            Rapaport and Eisenreich 1984
        2.39    RPLC            Veith etal. 1979a
        2.13    RPLC            Veith etal. 1980
        2.26    RPLC            de Kock and Lord 1987
        2.121   Shake-flask (R)    Banerjee et al. 1980
        2.1     Consensus       Klein et al. 1988
    The value  used in the final Guidance is  2.138, which is the average of the top
    five values.

Chlordane  [CAS#:  see below]
    There are several relevant CAS numbers:
        CAS#:    57-74-9        Chlordane, mixture of cis and trans

                                    B-1

-------
        CAS#:    5103-71-9      alpha-chlordane; cis-chlordane
        CAS#:    5103-74-2      beta-chlordane; trans-chlordane
        CAS#:    5566-34-7      gamma-chlordane
        CAS#:   12789-03-6      Chlordane, technical
    All of these, and all of their mixtures, are expected to have similar values for
    log Kow, BCF, and BAF.

    The value that has the highest priority is:
        6.00    RPLC           Veithetal. 1979b
    The value used in the final Guidance is 6.00.

Chlorobenzene [CAS#: 108-90-7]
    The values that have the highest priorities are:
        2.784   Slow-stir             Brooke et al. 1990
        2.898   Slow-stir             de Bruijn et al. 1989
        2.98    Generator-column    Miller et al.  1985
        2.80    Shake-flask      Voice et al. 1983
        2.89    Shake-flask      Medchem
        2.840   Shake-flask      Watari et al. 1982
        2.83    RPLC-E          Hammers et al. 1982
        2.94    RPLC           Miyake and Terada 1982
        3.00    RPLC           de  Kock and Lord 1987
        2.8     Consensus       Klein et al. 1988
    The value used in the final Guidance is 2.865, which is the average of the top
    six values.

Cyanide  [CAS,?: 57-12-5]
    A value of log Kow is not used for cyanide.

ODD [CAS#: see below]
    There are several relevant CAS numbers:
        CAS#:    72-54-8            p,p'-DDD; 4,4'-DDD
        CAS#:    53-19-0            o,p'-DDD; 2,4'-DDD
        CAS#:   4329-12-8       m,p'-DDD; 3,4'-DDD
    All of these, and all of their mixtures, are expected to have similar values for
    log Kow, BCF, and BAF.

    The values that have the highest priorities are:
        5.90    Slow-stir             Stancil 1994
        6.217   Slow-stir             de Bruijn et al. 1989
        4.73    RPLC           McDuffie 1981
        5.00    RPLC           de  Kock and Lord 1987
    The value used in  the final Guidance is 6.058, which is the average of the top
    two values.


                                    B-2

-------
DDE [CAS#: see below]
    There are several relevant CAS numbers:
        CAS#:    72-55-9           p,p'-DDE; 4,4'-DDE
        CAS#:   3424-82-6       o,p'-DDE; 4,4'-DDE
    All of these, and all of their mixtures, are expected to have similar values for
    log Kow. BCF, and BAF.

    The values that have the highest priorities are:
        6.57    Slow-stir            Stancil 1994
        6.956   Slow-stir            de Bruijn et al. 1989
        5.89    RPLC            Burkhard et al. 1985
        5.83    RPLC            Veithetal.  1979a
        5.69    RPLC            Veithetal.  1979b
        5.63    RPLC            Swann et al. 1983
        5.89    RPLC            McDuffie 1981
        6.09    RPLC            de Kock and Lord  1987
    The value used in the final Guidance is 6.763, which is the average of the top
    two values.

DDT [CAS#: see below]
    There are several relevant CAS numbers:
        CAS#:    50-29-3        p,p'-DDT; 4,4'-DDT
        CAS#:    789-02-6.       o,p'-DDT; 2,4'-DDT
        CAS#:   33086-18-9      DDT
    All of these, and all of their mixtures, are expected to have similar values for
    log Kow, BCF, arid BAF.

    The values that have the highest priorities are:
        6.198   Slow-stir            Brooke et al. 1986
        6.307   Slow-stir            Brooke et al. 1990
        6.38    Slow-stir            Stancil 1994
        6.914   Slow-stir            de Bruijn et al. 1989
        6.38    RPLC-E          Hammers et al. 1982
        6.06    RPLC-E          Harnisch et al. 1983
        5.84    RPLC-E          Harnisch et al. 1983
        6.4     RPLC-E          Brooke et al. 1986
        5.44    RPLC            Burkhard et al. 1985
        5.13    RPLC            Rapaport and Eisenreich 1984
        5.75    RPLC            Veithetal.  1979b
        5.63    RPLC            de Kock and Lord  1987
        6.36    Shake-flask       Chiou et al. 1982
        5.1      Shake-flask (R)    Platford et al. 1982
        6.2     Consensus       Klein et al. 1988
    The value used in the is 6.450, which is the average of the top four values.


                                   B-3

-------
Dieldrin  [CAS#: 60-57-1]
    The values that have the highest priorities are:
        5.335   Slow-stir            Stancil 1994; U.S. EPA 1991 a
        5.401   Slow-stir            de Bruijn et al. 1989
        4.538   Slow-stir            Brooke et al. 1986
        5.16    Generator-column     U.S. EPA 1991 a
        5.11    RPLC-E          Hammers et al. 1982
        4.65    RPLC            de  Kock and Lord 1987
        5.01    Shake-flask      U.S. EPA 1991 a
    The value of 4.54 is considered an outlier. The value used in the final
    Guidance is 5.299,  which is the average of the first, second, and fourth
    values.

2,4-Dimethylphenol [CAS#: 105-67-9]
    The values that have the highest priority are:
        2.30    Shake-flask      Medchem
        1.99    RPLC            Veithetal. 1980
        2.07    RPLC            Haky and Young 1984
        2.420   Shake-flask (R)   Banerjee et al. 1980
    The value used m the final Guidance is 2.30.

2,4-Dinitrophenol  [CAS#: 51-28-5]
    The value that has the highest priority is:
        1.51    Shake-flask      Medchem
        1.67    Shake-flask      Medchem
        1.54    Shake-flask      Medchem
        1.56    Shake-flask      Medchem
        1.59    Shake-flask      Medchem
        1.55    Shake-flask      Medchem
        1.50    Consensus      Klein et al. 1988
    The value used in the final Guidance is 1.570, which is the average of the top
    six values.

Hexachlorobenzene [CAS#: 118-74-1]
    The values that have the highest priorities are:
        5.47    Generator-column     Miller et al. 1985
        5.731   Slow-stir            de Bruijn et al. 1989
        5.9     RPLC-E          Brooke et al.  1986
        5.66    RPLC-E          Hammers et al. 1982
        5.46    RPLC-E          Harnisch et al. 1983
        5.26    RPLC-E          Harnisch et al. 1983
        6.71    RPLC            Rapaport and Eisenreich 1984
        6.86    RPLC            Burkhard et al. 1985
        7.42    RPLC            Veith  etal. 1979a

                                    B-4

-------
        5.23    RPLC            Veithetal. 1979b
        6.92    RPLC            de Kock and Lord 1987
        5.47    Shake-flask      Harnisch et al. 1983
        5.50    Shake-flask      Chiou et al. 1982
        5.00    Shake-flask      Konemann et al. 1979
        5.2     Shake-flask      Platford et al. 1982
        5.44    Shake-flask      Briggs 1981
        5.312   Shake-flask      Watari et al. 1982
    The value used in the final Guidance is 5.600, which is the average of the top
    two values.

Hexachlorobutadiene  [CAS#: 87-68-3]
    The values with the highest priorities are:
        4.785   Shake-flask      Banerjee et al. 1980
        4.90    Shake-flask      Chiou 1985
    The value used in the final Guidance is 4.842.

Hexachlorocyclohexane (HCCH)  [CAS#: 608-73-1]
        alpha-HCCH      [CAS#: 319-84-6]
        beta-HCCH      [CAS#: 319-85-7]
        delta-HCCH      [CAS#: 319-86-8]
        gamma-HCCH    [see lindane]
    The most useful values that were found are:
        alpha:   3.776   Slow-stir        de Bruijn et al. 1989
        beta:    3.842   Slow-stir        de Bruijn et al. 1989
    The values used in the final Guidance are:
        HCCH:   3.769
        alpha:   3.776
        beta:    3.842
        delta:    3.769
    The value used for HCCH and for delta is the average of the values obtained
    by de Bruijn et al. (1989) for alpha, beta, and gamma.

Hexachloroethane  [CAS#: 67-72-1]
    The values that have the highest priorities are:
        4.04    RPtC            McDuffie 1981
        4.05    RPLC            Veithetal. 1980
        4.14    Shake-flask      Chiou 1985
        3.93    Shake-flask      Veith et al. 1980
    These values are close to 4, and the range of the four values is small.  The
    value used in the final Guidance is 4.040, which is the average of the four
    values.

Lindane (gamma-HCCH)   [CAS#: 58-89-9]


                                   B-5

-------
    The values that have the highest priorities are:
        3.688   Slow-stir            de Bruijn et al. 1989
        3.61    Shake-flask      Medchem
        3.72    Shake-flask      Medchem
        3.32    Shake-flask      Platford 1981,1982
        3.89    RPLC           Veithetal.  1979b
        3.66    RPLC           Saito et al.  1993
        3.00    RPLC           de Kock and Lord  1987
    The value of 3.32 is considered an outlier.  The value used in the is 3.673,
    which is the average of the top three values.

Mercury  [CAS#: 7439-97-6]
    A value for  log Kow is not used for mercury.

Methylene chloride  [CAS#: 75-09-2]
    The value that has the highest priority is:
        1.25    Shake-flask      Medchem
    The value used in the final Guidance is 1.25.

Mirex  [CAS#: 2385-85-5]
    The value that has the highest priority is:
        6.89    RPLC           Veithetal.  1979b
        5.28    Shake-flask      Medchem
        4.650   Calculated       CLOGP
    The value used in the final Guidance is 6.89.

Nonachlor  [CAS#: see below]
    There are several relevant CAS numbers:
        CAS#:     - 3734-49-4      Nonachlor
        CAS#:   5103-73-1       cis-nonachlor
        CAS#:   39765-80-5     trans-nonachlor
    The value that has the highest priority is:
        5.655   Calculated       CLOGP
    The value used in the final Guidance is 6.0, which is the value used for the
    structurally similar chlordane and is considered to be a better value for
    nonachlor than 5.655; this value is used only in connection with the BSAF
    methodology.

Octachlorostyrene  [CAS#: 29082-74-4]
    The value that has the highest priority is:
        6.29    RPLC           Veithetal.  1979b
    The value used in the final Guidance is 6.29.

PCBs

                                    B-6

-------
    See Appendix F of this document.

Pentachlorobenzene  [CAS#: 608-93-5]
    The values that have the highest priorities are:
        5.183  Slow-stir            de Bruijn et al. 1989
        5.03    Generator-column    Miller et al. 1985
        5.06    RPLC-E         Hammers et al. 1982
        5.29    RPLC            Veithetal.  1980
        6.12    RPLC            de Kock and Lord  1987
        5.20    Shake-flask      Chiou 1985
        4.88    Shake-flask      Konemann et al. 1979
        5.167  Shake-flask      Watari et al. 1982
        4.940  Shake-flask (R)   Banerjee et al. 1980
    The value  used in the final Guidance is 5.106, which is the average of the top
    two values.
                  •

2,3,4,5,6-Pentachlorotoluene  [CAS#: 877-11-2]
    The value  that has the highest priority is:
        6.356  Calculated       CLOGP
    The only value available is 6.356; this value is used only in the study of the
    food-chain model.

Photomirex  [CAS#:- 39801 -14-4]
    The value  that has the highest priority is:
        4.537  Calculated       CLOGP
    The value  used in the final Guidance is 6.89, which is the value used for mirex
    and is considered to be a better value for photomirex than  4.537.

2,3,7,8-TCDD   [CAS#:  1746-01-6]
    The value  that has the highest priority is:
        6.42    Slow-stir            Sijm et al. 1989
        6.63    Slow-Stir            Marple et al. 1986
        7.02    RPLC            Burkhard and Kuehl 1986
    As per pages 2-2, 2-3, and 3-9 of U.S. EPA (1993), the value used in the final
    Guidance is 7.02.

1,2,3,4-Tetrachlorobenzene  [CAS#: 634-66-2]
    The values that have the highest priorities are:
        4.635  Slow-stir            de Bruijn et al. 1989
        4.55    Generator-column    Miller et al. 1985
        4.41    RPLC-E         Hammers et al. 1982
        4.75    Shake-flask      Bruggeman et al. 1982
        4.60    Shake-flask      Chiou 1985
        4.46    Shake-flask      Konemann et al. 1979


                                    B-7

-------
        4.375   Shake-flask      Watari et al. 1982
    The value used in the final Guidance is 4.592, which is the average of the top
    two values.

1,2,3,5-Tetrachlorobenzene  [CAS#: 634-90-2]
    The values that have the highest priorities are:
        4.658   Slow-stir            de Bruijn et al.  1989
        4.65    Generator-column    Miller et al. 1985
        4.35    RPLC-E          Hammers et al. 1982
        4.59    Shake-flask      Chiou 1985
        4.50    Shake-flask      Konemann et al. 1979
        4.459   Shake-flask (R)   Banerjee et al. 1980
    The average of the top two values is 4.654; this value is used only in the
    study of the food-chain model.

1,2,4,5-Tetrachlorobenzene  [CAS#: 95-94-3]
    The values that have the highest priorities are:
        4.604   Slow-stir            de Bruijn et al.  1989
        4.51    Generator-column    Miller et al. 1985
        4.52    RPLC-E          Hammers et al. 1982
        4.70    Shake-flask      Chiou 1985
        4.52    Shake-flask      Konemann et al. 1979
        4.555   Shake-flask      Watari et al. 1982
    The value used in the final Guidance is 4.557, which is the average of the top
    two values.

Toluene  [CAS#: 108-88-3]
    The values that have the highest priorities are:
        2.65    Generator-column    Miller et al. 1985
        2.786   Slow-stir            de Bruijn et al.  1989
        2.63    Slow-stir            Brooke et al. 1990
        2.73    Shake-flask      Medchem
        2.77    Shake-flask      Medchem
        2.77    RPLC-E          Harnisch  et al. 1983
        2.78    RPLC-E          Hammers et al. 1982
        2.78    RPLC           Burkhard  et al. 1985
        2.99    RPLC           Veithetal. 1980
        2.89    RPLC           Rapaport and Eisenreich 1984
        2.62    RPLC           Miyake and Terada 1982
        3.00    RPLC           de Kock and Lord 1987
        2.21    Shake-flask (R)   Banerjee  et al. 1980
        2.7     Consensus      Klein et al.  1988
    The value used in  the final Guidance is 2.713, which is the average of the top
    five values.

                                    B-8

-------
Toxaphene  [CAS#: 8001-35-2]
    The value that has the highest priority is:
        4.330   Calculated       CLOGP
    The value used in the final Guidance is 4.330.

1,2,3-Trichlorobenzene   [CAS#: 87-61-6]
    The values that have the highest priorities are:
        4.139   Slow-stir            de Bruijn et al. 1989
        4.04    Generator-column    Miller et al. 1985
        4.11    Shake-flask      Konemann et al.  1979
        4.14    Shake-flask      Chiou 1985
        4.053   Shake-flask      Watari et al. 1982
        3.88    RPLC-E         Hammers et al. 1982
        4.02    RPLC            McDuffie 1981
    The top five values are all close to 4 and the range is small.  The average of
    the top five values is 4.096; this value is used only in the study of the food-
    chain model.

1,2,4-Trichlorobenzene   [CAS#: 120-82-1]
    The values that have the highest priorities are:
        4.050   Slow-stir            de Bruijn et al. 1989
        3.98    Generator-column    Miller et al. 1985
        3.93    Shake-flask      Konemann et al.  1979
        4.02    Shake-flask      Chiou et ai. 1982; Chiou 1985
        3.970   Shake-flask      Watari et al. 1982
        3.96    RPLC-E         Hammers et al. 1982
        4.2?    RPLC            Veithetal.  19795
        4.22    RPLC            de Kock and Lord 1987
        4.20    Consensus      Klein et al. 1988
    The top five values are all close to 4 and the range is small.  The value used in
    the final Guidance is 3.990, which is the average of the top five values;
    currently this value  is only used in the study of the food-chain model.

1,3,5-Trichlorobenzene   [CAS#: 108-70-3]
    The values that have the highest priorities are:
        4.189   Slow-stir            de Bruijn et al. 1989
        4.02    Generator-column    Miller et al. 1985
        4.15    Shake-flask      Konemann et al.  1979
        4.31    Shake-flask      Chiou 1985
        4.190   Shake-flask      Watari et al. 1982
        4.17    RPLC-E         Hammers et al. 1982
    The top five values  are all close to 4 and the range is small.  The average of
    the top five values is 4.172; this value is used only in the study of the food-
    chain model.


                                   B-9

-------
Trichloroethylene  [CAS#: 79-01-6]
    The values that have the highest priorities are:
        2.53     Generator-column    Miller et al. 1985
        3.14     Shake-flask      Harnisch et al. 1983
        2.67     RPLC-E          Harnisch et al. 1983
        2.56     RPLC-E          Harnisch et al. 1983
        2.420   Shake-flask (R)   Banerjee et al. 1980
        2.4      Consensus       Klein et al. 1988
    The value used in the final Guidance is 2.53.  The value of 3.14 is considered
    an outlier.

2,3,6-Trichlorotoluene  [CAS#: 2077-46-5]
    The value that has the highest priority is:
        4.930   Calculated       CLOGP
    The only value available is 4.930; this value is used only in the study of the
    food-chain model.

2,4,5-Trichlorotoluene  [CAS#: 6639-30-1]
    The value that has the highest priority is:
        4.930   Calculated       CLOGP
    The only value available is 4.930; this value is used only in the study of the
    food-chain model.

References

Banerjee, S., S.H. Yalkowsky, and S.C. Valvani.  1980. Water Solubility and
    Octanol/Water Partition Coefficients of Organics. Limitations of the Solubility-
    Partition Coefficient Correlation.  Environ. Sci. Technol. 14:1227-1229.

Bowman, B.T., and W.W. Sans.  1983.  Determination of Octanol-Water
    Partitioning Coefficients (Kow) of 61 Organophosphorus and Carbamate
    Insecticides and Their Relationship to Respective Water Solubility (S) Values.
    J. Environ. Sci. Health 618:667-683.

Briggs, G.G.  1981.  Theoretical and Experimental Relationships between Soil
    Adsorption, Octanol-Water Partition Coefficients, Water Solubilities,
    Bioconcentration Factors, and the Parachor.  J. Agric. Food Chem.  29:1050-
    1059.

Brooke, D.N., A.J. Dobbs, and N. Williams.  1986. Octanol:Water Partition
    Coefficients (P): Measurement, Estimation, and Interpretation, Particularly for
    Chemicals with P >  105.  Ecotoxicol Environ. Safety  11:251-260.

Brooke, D., I. Nielsen, J.  de Bruijn, and J. Hermens. 1990.  An Interlaboratory


                                     B-10

-------
    Evaluation of the Stir-Flask Method for the Determination of Octanol-Water
    Partition Coefficient (Log Pow). Chemosphere 21:119-133.

Bruggeman, W.A., J. van der Steen, and 0. Hutzinger. 1982.  Reversed-Phase
    Thin-Layer Chromatography of Polynuclear Aromatic Hydrocarbons and
    Chlorinated Biphenyls.  J. Chromatog. 238:335-346.

Burkhard, L.P., and D.W. Kuehl.  1986. N-Octanol/Water Partition Coefficients by
    Reverse Phase Liquid Chromatography/Mass Spectrometry for Eight
    Tetrachlorinated Planar Molecules.  Chemosphere 15:163-167.

Burkhard, L.P., D.W. Kuehl, and G.D. Veith. 1985.  Evaluation of Reverse Phase
    Liquid Chromatography/Mass Spectrometry for Estimation of N-Octanol/Water
    Partition Coefficients for  Organic Chemicals.  Chemosphere 14:1551-1560.

Chiou, C.T.  1985. Partition Coefficients  of Organic Compounds in Lipid-Water
    Systems and Correlations with Fish Bioconcentration Factors. Environ. Sci.
    Technol. 19:57-62.
                  •

Chiou, C.T.,  V.H. Freed, D.W. Schmedding, and R.L.  Kohnert.  1977.  Partition
    Coefficients and Bioaccumulation of Selected Organic Chemicals.  Environ.
    Sci. Technol. 11:475-478.

Chiou, C.T.,  D.W. Schmedding, and M. Manes. 1982. Partitioning of  Organic
Compounds in Octanol-Water Systems. Environ. Sci. Technol. 16:4-10.

de Bruijn, J., F.  Busser, W Seinen, and J. Hermens.  1989.  Determination of
    Octanol/Water Partition Coefficients for Hydrophobic Organic Chemicals with
    the "Slow-Stirring" Method. Environ. Toxicol. Chem. 8:449-512.

de Kock, A.C., and D.A. Lord. 1987.  A Simple Procedure for Determining
    Octanol-Water Partition Coefficients Using  Reverse Phase High Performance
    Liquid Chromatography (RPHPLC).  Chemosphere 16:133-142.

Haky, J.E., and  A.M. Young.  1984.  Evaluation of a  Simple HPLC Correlation
    Method for the Estimation of the Octanol-Water Partition Coefficients of
    Organic  Compounds.  J. Liq. Chromatog. 7:675-689.

Hammers,  W.E., G.J. Meurs, and C.L. de Ligny. 1982. Correlations between
    Liquid Chromatographic Capacity Ratio Data on Lichrosorb RP-18 and Partition
    Coefficients in the Octanol-Water System.  J. Chromatog. 247:1-13.

Harnisch, M., H.J. Mockel, and G. Schulze.  1983. Relationship between Log Pow
    Shake-Flask Values and Capacity Factors Derived from Reverse-Phase High-


                                   B-11

-------
    Performance Liquid Chromatography for /7-Alkylbenzenes and some OECD
    Reference Substances.  J. Chromatog. 282:315-332.

Karickhoff, S.W., D.S. Brown, and T.A Scott.  1979.  Sorption of Hydrophobic
    Pollutants on Natural Sediments.  Water Research 13:241-248.

Klein, W., W. Kordel, M. Weib, and H.J. Poremski.  1988.  Updating the OECD
    Test Guideline 107 "Partition Coefficient N-Octanol/Water": OECD Laboratory
    Intercomparison Test on the HPLC Method.  Chemosphere 17:361-386.

Konemann, H., R. Zelle, F. Busser, and W.E. Hammers.  1979. Determination of
    Log POCT Values .of Chloro-Substituted Benzenes, Toluenes and Anilines by
    High-Performance Liquid Chromatography on ODS-Silica. J. Chromatog.
    178:559-565.

Marple, L., B. Berridge, and L. Throop. 1986. Measurement of the Water-Octanol
    Partition Coefficient of 2,3,7,8-Tetrachlorobenzo-p-dioxin. Environ. Sci.
    Technol. 20:397-399.

McDuffie, B.  1981. • Estimation of Octanol/Water Partition Coefficients for Organic
    Pollutants Using Reverse-Phase HPLC. Chemosphere 10:73-83.

Means, J.C., S.G. Wood, J.J. Hassett, and W.L. Banwart.  1980.  Sorption of
    Polynuclear Aromatic Hydrocarbons by Sediments and Soils.  Environ. Sci.
    Technol. 14:1524-1528.

Miller, M.M., S.P. Wasik, G.-L. Huang, W.-Y. Shiu, and D. Mackay.   1985.
    Relationships between Octanol-Water Coefficient and Aqueous Solubility.
    Environ. Sci. Technol. 19:522-529.

Miyake, K., and H. Terada.  1982. Determination of Partition Coefficients of  Very
    Hydrophobic Compounds by High-Performance Liquid Chromatography on
    Glyceryl-Coated Controlled-Pore Glass.  J. Chromatog. 240:9-20.

Platford,  R.F. 1981.- The Environmental Significance of Surface Films II.
    Enhanced Partitioning of Lindane in Thin Films of Octanol on the Surface  of
    Water.  Chemosphere 10:719-722.

Platford,  R.F. 1982.  Pesticide Partitioning in Artificial Surface Films. J. Great
    Lakes Res. 8:307-309.

Platford,  R.F., J.H.Carey, and E.J. Hale.  1982.  The Environmental Significance of
    Surface Films: Part 1 - Octanol-Water Partition Coefficients for  DDT and
    Hexachlorobenzene. Environ. Pollut.  (Series B) 3:125-128.


                                   B-12

-------
Rapaport, R.A., and S.J. Eisenreich. 1984.  Chromatographic Determination of
    Octanol-Water Partition Coefficients (Kow's) for 58 Polychlorinated Biphenyl
    Congeners.  Environ. Sci. Techno).  18:163-170.

Saito, H., J. Koyasu, K. Yoshida, T. Shigeoka, and 8. Koike. 1993.  Cytotoxicity
    of 109 Chemicals to Goldfish GFS Cells and Relationships with 1-
    Octanol/Water Partition Coefficients.  26:1015-1028.

Sijm, D.T.H.M., H. Wever, P.J. de Vries, and A. Opperhuizen.  1989. Octan-1-
    ol/water Partition Coefficients of Polychlorinated Dibenzo-p-dioxins and
    Dibenzofurans: Experimental Values Determined by a Stirring Method.
    Chemosphere 19:263-266.

Stancil, F.  1994.  Memorandum to M. Reiiey.

Swann, R.L., D.A. Laskowski, P.J. McCall, K. Vander Kuy, and H.J.  Dishburger.
    1983.  A Rapid Method for the Estimation of the Environmental  Parameters
    Octanol/Water Partition Coefficient, Soil Sorption Constant, Water to Air Ratio,
    and Water Solubility. Residue Reviews 85:17-28.

U.S. EPA.  1991a. Proposed Sediment Quality Criteria for the Protection of
    Benthic Organisms:  Dieldrin. Office of Water, Washington, DC.

U.S. EPA.  1991b. Proposed Sediment Quality Criteria for the Protection of
    Benthic Organisms:  Endrin.  Office of Water, Washington, DC.

U.S. EPA.  1991c. Proposed Sediment Quality Criteria for the Protection of
    Benthic Organisms:  Fluoranthene.  Office of Water, Washington, DC.

U.S. EPA.  1991d. Proposed Sediment Quality Criteria for the Protection of
    Benthic Organisms:  Phenanthrene.  Office of Water, Washington, DC.

U.S. EPA.  1993. Interim Report on Data and Methods for Assessment of 2,3,7,8-
    Tetrachlorodibenzo-p-dioxin Risks to Aquatic Life and Associated Wildlife.
    EPA/600/R-93/055. National Technical  Information Service, Springfield, VA.

Veith, G.D., N.M. Austin, and R.T. Morris. 1979a.  A Rapid Method for Estimating
    Log P for Organic Chemicals. Water Res. 13:43-47.

Veith, G.D., D.L. DeFoe, and B.V. Bergstedt.  1979b.  Measuring and Estimating
    the Bioconcentration Factor of Chemicals in Fish.  J. Fish. Res. Bd. Can.
    36:1040-1048.

Veith, G.D., K.J. Macek, S.R.  Petrocelli, and J. Carroll. 1980.  An Evaluation of


                                    B-13

-------
    Using Partition Coefficients and Water Solubility to Estimate Bioconcentration
    Factors for Organic Chemicals in Fish. IN: Aquatic Toxicology.  Eaton, J.G.,
    P.R. Parrish, and A.C. Hendricks, Eds. ASTM STP 707.  American Society for
    Testing and Materials, Philadelphia, PA. pp. 116-129.

Voice, T.C., C.P. Rice, and W.J. Weber, Jr.  1983.  Effect of Solids Concentration
    o the Sorptive Partitioning of Hydrophobic Pollutants in Aquatic Systems.
    Environ. Sci. Technol. 17:513-518.

Watari, H., M. Tanaka, and N. Suzuki.  1982.  Determination of Partition
    Coefficients of Halobenzenes in Heptane/Water and 1 -Octanol/Water Systems
    and Comparison with the Scaled Particle Calculation. Anal. Chem. 54:702-
    705.
                                    B-14

-------
Appendix C.  Derivation of Basic Equations Concerning Bioconcentration and
             Bioaccumulation of Organic Chemicals

Introduction

Most work dealing with the bioconcentration and bioaccumulation of organic
chemicals has concerned chemicals whose log K<,ws are greater than 3.  The
purpose of this appendix is to explain why modifications of the equations generally
used with such chemicals are necessary so that the equations also are appropriate
for chemicals whose-Kows, BCFs, or BAFs are less than 1000, and to derive all of
the appropriate equations that are used in the calculation of BAFs for the final
Guidance.

Background

Bioconcentration factors were originally defined as:

                                                                          (1)
                                         r*
                                         Ow
where:
    BCFj   =  a total bioconcentration factor (i.e., a BCF that is based on the total
              concentrations of the chemical in the water and in the aquatic
              biota).
    CB     =  the total concentration of the chemical in the aquatic biota, based
              on the wet weight of the aquatic biota.
    Cw     =  the total concentration of the chemical in the water around the
              aquatic biota.

This is not the nomenclature that was used originally, but it is used here for clarity.

It was subsequently realized that extrapolation of BCFs for  organic chemicals from
one species to another would be more accurate if the BCFs were normalized on the
basis  of the amount of lipid in the aquatic biota. It was also realized that
extrapolation of BCFs for organic chemicals from one water to another would be
more  accurate if the BCFs were calculated on the basis of the freely dissolved
concentration of the organic chemical in the water around the aquatic biota.  Thus
two additional BCFs were defined and used:

                                 BCF]  = -El                               (2)


                                 BCFJ"  = -El                               (3)

where:


                                    C-1

-------
     BCFj    =   the lipid-normalized total bioconcentration factor (i.e., normalized
                 to 100 percent lipid and based on the total concentration of the
                 chemical in the water around the biota).
    Cj       =   the lipid-normalized concentration of the chemical in the aquatic
                 biota.
     BCFJd   =   the lipid-normalized, freely dissolved bioconcentration factor.
     Cw  =   the freely dissolved concentration of chemical in the water around the
             aquatic biota.

The experimental definition of C, is:

    C   =   the total amount of  chemical in the aquatic biota
      1            the amount  of  lipid in the aquatic biota

             (B)(CB) =  (B)(CB) = Cl
                L  •    (f,)(B)    f,
where:
    B   =   the wet weight of the aquatic biota.
    L   =   the weight of the  lipid in the aquatic biota.
    f,   =   the fraction of the aquatic biota that is lipid = LIB.

Using equation 4 to substitute for  C, in  equation 2 and then using equation 1:


                           BCFJ  = 	?i_ = -55!:                        (5)
                                   (Cw)(f,)    f*

If ffd = the fraction of the chemical in the water around the aquatic biota that is
freely dissolved, then:

                                        rtd
                                   ^ = ^                                <6>
                                        Cw

Using equations 4 and 6 to substitute for C, and Cw in equation 3 and then using
equation 1:


                        BCFfd =      CB      _   BCFj                      (7)
                   *    *»^^  f                   1X1/^4
Equations 1,  5, and 7 show the relationships between the three different
bioconcentration factors.
                                     C-2

-------
Theoretical justification for use of both lipid-normalization and the freely dissolved
concentration of the "organic chemical in the ambient water is based on the
concept of equilibrium partitioning, whereas practical justification is provided by
the general similarity of the value of BCFJd for an organic chemical across both
species and waters. It will be demonstrated, however, that a more complete
application of equilibrium partition theory shows that  BCF"  extrapolates well only
for chemicals whose Kows are greater than 1 000, whereas a different BCF
extrapolates well for organic chemicals whose Kows are greater than 1 000 as well
as for chemicals whose Kows are less than 1 000.

Partition Theory and Bioconcentration

Equilibrium partition theory provides the understanding necessary to ensure proper
use of Kows, BCFs, and BAFs in the derivation of water quality criteria for organic
chemicals.  For the purpose of applying partition theory, aquatic biota can be
modelled as consisting of water, lipid, and non-lipid organic matter (Barber et al.
1991).  In this model, an organic chemical in aquatic biota exists in three forms:

    1 .  Chemical that is freely  dissolved  in the water that is in the biota.
    2.  Chemical that is partitioned to the lipid that is in the biota.
    3.  Chemical that is partitioned to non-lipid organic matter in the biota.  The
        total concentration of chemical in the water inside the biota includes
        chemical that is partitioned to lipid and non-lipid organic matter in the
        water.

According to this model:

                      Cl = 
-------
where:
    KLW      =   the lipid-water partition coefficient.

"KLW" (Gobas 1993) is used herein because it is more descriptive than "K^'. which
is used by DiToro et al. (1991).  This partition coefficient is central to the
equilibrium partition approach that is used to derive sediment quality criteria
(DiToro et al. 1991), the Gobas model that is used to derive Food-Chain Multipliers
for the final Guidance,  and the equations given here that are used to derive BCFs
and BAFs for the final Guidance.

In order for equations 8 and 9 to be correct, partition theory requires that the
concentration of the organic chemical in the lipid, CL, be defined as:

Q  _  the amount of chemical partitioned to lipid in aquatic biota
 L             the amount of lipid in the aquatic biota

It is difficult to determine CL experimentally because it is not easy to measure only
the chemical that is partitioned to the lipid (i.e., it is not easy to separate the three
different kinds of chemical that, according to the model, exist in aquatic biota).
Because  all of the organic chemical in the biota is measured when C, is
determined, C, can be  determined easily, and C, is higher than CL.

It is useful to define another bioconcentration factor as:


                                 BCF? = -L                              (10)
                                         pfd
                                         ^W

Because  CL is lower than C,f  BCF™ <  BCFJd .

The only difference between KLW and BCF" is that the denominator in KLW is CWB >
whereas the denominator  in BCF" is C" • When partition theory applies, however,
all phases are in equilibrium and so:
Therefore, when the organic chemical is not metabolized by the aquatic biota and
when growth dilution is negligible:

                                 BCF? = KLW            -                  (12)

Because octanol is a useful surrogate for lipid, a reasonable approximation is that:
where:

                                     C-4

-------
    Kow =   the octanol-water partition coefficient.

Thus:
                         predicted BCF? = KLW = Kow                      <14)

By using equations 9 and 1 1 to substitute for CL and CWB in equation 8:
                  Cl  = (fw)(C#)  4. (f,)(BCFlw)(Cwl) + (fN)(CN)               H5)

By using equation  6 to substitute for Cw in equation 1 5:

                CB = (fw)
-------
    a.  Because of bones and other inorganic matter, the sum of fw + f, + fN
        must be less than 1 .
    b.  fw is usually about 0.7 to 0.9.
    c.  Because f, must be measured if the BAF or BCF is to be useful, f, is
        known for the aquatic biota; it is usually between 0.03 and 0.15.
    d.  The term " ( — -)" is similar to BCF" (see equation 10) and is therefore
                     Cfd
                     W
        probably related to Kow (see equation 14), although the affinity of the
        chemical for non-lipid organic matter is probably much less than its
        affinity for lipid.

Although such considerations aid in understanding "x", the magnitude of "x" in
equation 20 is important only for chemicals whose log Kows are in the range of 1
to 3.  For organic chemicals whose log Kows are about 1 , ffd is about 1 . In
addition, such chemicals distribute themselves so as to have similar concentrations
in water and in the different organic phases in the aquatic biota, which means that
BCFj will be approximately 1 if both metabolism and growth dilution are negligible.
An organic chemical whose log Kow is less than 1 will also have a BCFj on the
order of 1  because water is the predominant component in aquatic biota.  Setting
"x" equal to 1 is about right in the range of log Kows in which it is not negligible
(see also McCarty et al. 1992).

Substituting x = 1  into equation  20:

                        BCF| = (ffd)[  1 +  (fjMBCF?) ]                    (21)

Rearranging gives:   .

                          BCF? = (-Efl  - 1  )(1)                       (22)
BCF? can be called the "baseline BCF" because it is the most useful BCF for
extrapolating from one species to another and from  one water to another for
organic chemicals with both high and low Kows. The baseline BCF is intended to
reference bioconcentration of organic chemicals to partioning between lipid and
water.

Equations 12, 13, and 22 demonstrate that both Kow and
                                  fd
                                    C-6

-------
are useful approximations of the baseline BCFs.  It will probably be possible to
improve both approximations within a few years, but such improvements might not
affect the BCFs substantially and probably will not require changes in the rest of
the equations or the terminology.

When BCFj is greater than 1000, the "-1" in equation 22 is negligible and so this
equation becomes equivalent to equation 7 (i.e., when BCFj is large, BCFJd is a
useful approximation of the baseline BCF).

Bioaccumulation

By analogy with equations 21 and 22:

                        BAFJ = (ffd)[ 1 + (f,)(BAFLfd)  ]                    (23)


                                         -  1 )(1)                      (24)
                                     fd         'l

BAF" can be called the "baseline BAF" because it is the most useful BAF for
extrapolating from one species to another and from one  water to another for
chemicals with both high and low Kows.

It is convenient to define a food-chain multiplier  (FCM) as:

                        FCM  = baseline BAF  =  BAF,"                    (25)
                               baseline BCF    BCF"

Some of the consequences of equation 25 are:

1 .   Substituting equations 22 and 24 into equation 25:
                             FCM =     T " ffd                          (26)
                                    BCF{ - ffd

    Therefore, BAFj = (FCMMBCFt) only when ffd is much less than BAF| and
BCFJ.

2.  When FCM = 1 (as for trophic level 2 in the Gobas model):

                        baseline BAF = baseline  BCF                     (27)
                                    C-7

-------
3.  Predicted baseline BAFs can be obtained using FCMs and the following
    rearrangement of equation 25:

            predicted baseline BAF = (PCM) (baseline BCF)                 (28)

    a.   Using a laboratory-measured BCF in equation 22:

            predicted baseline BAF    = (FCM)( measured BCF")           (29)
                                     l - 1 )()                        (30)
                                   'fd        *t

    b.  Using a predicted BCF in equation 14:

            predicted baseline BAF     = (FCM)( predicted BCF")           (31)

                        = (FCM)(KOW)                                  (32)

    The FCMs used to calculate predicted baseline BAFs must be appropriate for
    the trophic level* of the aquatic biota for which the predicted baseline BAF is
    intended to apply.

Although BAFs can be related to BCFs using FCMs, BAFs and BCFs can also  be
related using Biomagnification Factors (BMFs).  The tow systems are entirely
compatible, but confusion can result if the terms are not  used consistently and
clearly.  Because both systems are used in the final Guidance and elsewhere, it is
appropriate to explain the relation between the two here. The basic difference is
that FCMs always relate back to trophic level one, whereas BMFs always relate
back to the next trophic  level.   In the FCM system:
                     = BCF

            BAFTL2   = (FCMTL2)(BAFTL1)

            BAFTL3 -  = (FCMTL3)(BAFTL1)

            BAFTL4

In the BMF system:
            BAFTL1   = BCF
            BAFTL2'  = (BMFTL2)(BAFTL1)

                                    C-8

-------
            BAFTL3   = (BMFTL3)(BAFU2)

            BAFU4   = (BMFTL4)(BAFTL3)

Therefore:

            BMFT..2

            BMFTL3-

            BMFTL4   = (FCMU4)/(FCMTL3)

Both metabolism and growth dilution can cause BMFs to be less than 1.

Calculation of Criteria

Baseline BCFs and BAFs can be extrapolated between species and waters, but they
cannot be used directly in the calculation of criteria that are based on the total
concentration of the chemical in the water. The BCFs and BAFs that are needed to
calculate such criteria can be calculated from measured and predicted baseline
BCFs and BAFs using the following equations, which are derived from equations
21 and 23:

                  • BCFJ = [ 1  + (baseline BCF)(f,)  ](ffd)                 (33)

                    BAFj = [ 1  + (baseline BAF)(f,)  ](ffd)                 <34>

References

Barber, M.C., L.A. Suarez, and R.R. Lassiter.  1991.  Modelling Bioaccumulation of
    Organic Pollutants in Fish with an Application to PCBs in Lake Ontario
    Salmonids. Can. J. Fish. Aquat. Sci. 48:318-337.

DiToro, D.M., C.S. Zarba, D.J. Hansen,  W.J. Berry,  R.C. Swartz,  C.E. Cowan, S.P.
    Pavlou, H.E. Allen, N.A. Thomas, and P.R. Paquin.  1991.  Technical Basis for
    Establishing Sediment Quality Criteria for Nonionic Organic Chemicals Using
    Equilibrium Partitioning.  Environ. Toxicol. Chem. 10:1541-1583.

Gobas, F.A.P.C.  1993.  A Model for Predicting the  Bioaccumulation of
    Hydrophobic Organic Chemicals in Aquatic Food-Webs: Application to Lake
    Ontario. Ecological Modelling 69:1-17.

McCarty, L.S., D. Mackay, A.D. Smith,  G.W. Ozburn, and D.G. Dixon. 1992.


                                     C-9

-------
Residue-Based Interpretation of Toxicity and Bioconcentration QSARs from
Aquatic Bioassays: Neutral Narcotic Organics.  Environ. Toxicol. Chem.
11:917-930.
                                C-10

-------
Appendix D. Derivation of Baseline BAFs from Field-Measured BAFs and
            Laboratory-Measured BCFs

Some of the more important restrictions on use of field-measured BAFs and
laboratory-measured BCFs in the final Guidance are:

    1 .   A laboratory-measured BCF is not used if it is based on the measurement
         of radioactivity unless the BCF is intended to include metabolites or when
         there is confidence that there is no interference due to metabolites.

    2.   For a chemical for which  log Kow is greater than 4,  a laboratory-measured
         BCF or a field-measured BAF is not used unless the concentrations of POC
         and DOC were measured or can be reliably estimated in the ambient water
         because:
         a.   The higher the Kow, the more the calculated baseline BAF will depend
            on the concentrations of POC and DOC.
         b.   If log Kow is very large and there is fast equilibrium with POC and
            DOC, uptake via ingestion of food particles in a bioconcentration test
            might be substantial, thereby giving a high estimate of the
            bioconcentration factor.
         If reliable values for POC  and DOC are not available and if log Kow is less
         than 4,  the fraction of the toxicant that is not freely dissolved is
         negligible.

    3.   BCFs and BAFs are used only if the percent lipid was measured or could
         be reliably estimated.

Baseline BAFs were not calculated in this appendix from field data reported by
Oliver and Niimi (1988) because baseline BAFs were calculated from these data in
Tables 4, 5 and 8.  The equation presented here is equivalent to that used for
Tables 4, 5, and 8, as demonstrated below with DDE.

The following equation from Section III.B is used to calculate the fraction of the
chemical that is freely dissolved in the ambient water:
                   133ffd =
                                              (POC)(KOW)
where:
    ffd       =  fraction that is freely dissolved.
    DOC     =  concentration  of dissolved organic carbon (kg/L).
    POC     =  concentration  of paniculate organic carbon (kg/L).
    Kow      =  octanol-water  partition coefficient.
                                    D-1

-------
The following equation from Appendix C is used to calculate a measured baseline
BAF from a field-measured BAF:


                  measured baseline BAF = (    T  - 1)(-L)
                                             'fd        'I
where:
    BAFj     =  BAF based on total concentrations of the organic chemical in the
                tissue and in the ambient water.
    f,        =  fraction lipid in the tissue.

The trophic level to which the baseline BAF applies depends on the organisms used
in the determination of the field-measured BAF.

The following equation from Appendix C is used to calculate a measured baseline
BAF from a laboratory-measured BCF:

                                              BCF*       1
               predicted baseline  BAF = (FCM) (—-I - 1)( —)
                                               Md       *l
where:
    BCFj     =  BCF based on total concentrations of the organic chemical in the
                tissue and in the ambient water.
    FCM     =  Food-Chain Multiplier.

The trophic level to which the predicted baseline BAF applies depends on the
trophic level to which the FCM applies.

Benzene

    Based on a predicted BCF and a FCM. See Appendix H.

Chlorobenzene

    Based on a predicted BCF and a FCM. See Appendix H.

Chlordane

    The following field-measured BAFs are available for alpha and gamma
    chlordane:

          BAF         % L        Species            Reference

       1,400,000 '  7.592       R. trout            Oliver and Niimi 1985
                                    D-2

-------
         76,000     7.592       R. trout            Oliver and Niimi 1985

       Geometric mean BAF = 326,190

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

                 DOC = 0.000002   kg/L
                 POC  = 0.000000075 kg/L

   The log Kow derived in Appendix B for chlordane is 6.00.  The resulting value
   of ffd is 0.7843, and then:
      Measured baseline BAF^ = (~- - 1 > < 0.07*592 * = 5'478'115
   A measured baseline BAF of 6, 1 66,000 is derived in Table 8 based on Oliver
   and Niimi (1988); this is considered a better value and is used in the final
   Guidance because it is based on a more comprehensive set of data.

Cyanides

   No appropriate BAF or BCF exists for this chemical.

DDE
   The following field-measured BAF is available:

         BAF        %  L       Species           Reference

       18,000,000   7.592      R. trout           Oliver and Niimi 1985

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

                 DOC = 0.000002    kg/L
                 POC = 0.000000075 kg/L

   The log Kow derived in Appendix B for DDE is 6.763.  The resulting value of ffd
   is 0.3856, and then:
   Measured baseline BAFTL4 = (            - D(Q QJ592 ) = 614,864,290
   A measured baseline BAF of 223,900,000 is derived in Table 8 based on Oliver


                                   D-3

-------
   and Niimi (1988); this is considered a better value and is used in the final
   Guidance because it is based on a more comprehensive set of data.

   The following field-measured BAF is also available:

         BAF        % L       Species            Reference

       11,315,789   11.00       Salmonids         Oliver and Niimi 1 988

   Salmonids are expected to be in trophic level 4.  These data were obtained
   from Lake Ontario, but the water sample was centrifuged before the
   concentration of DDE was measured. Thus the concentrations of POC and
   DOC are expected to be:

                 DOC = 0.000002  kg/L
                 POC  = 0.0     kg/L

   The log Kow derived  in Appendix B for DDE is 6.763. The resulting value of ffd
   is 0.4632, and then:
1   ^I89  - 1><7>--
  0.4632         0. 1 1
     Measured baseline BAF^ = (             - 1><--> = 222,083,394
   The log of this baseline BAF is 8.3465, which is very similar to the value of
   8.35 that is derived in Table 8 for DDE from the same dataset. Thus the
   equation used here is equivalent to that used to calculate the fiel-measured
   BAFs given in Tables 4, 5, 6, 7 and 8 .

Dieldrin

   Based on the BSAF methodology. See Section HUE and Appendix H.

2.4-Dimethylphenol

   Based on a predicted BCF and a FCM.  See Appendix H.

2.4-Dinitrophenol

   Based on a predicted BCF and a FCM.  See Appendix H.

Hexachlorobenzene

   The following field-measured BAFs are available:
                                    D-4

-------
         BAF         % L        Species            Reference

       1,467,000    20.9        L trout            Oliver and Nicol 1982
        494,667     7.592      R. trout            Oliver and Niimi 1983

     Geometric mean BAF = 851,866
     Geometric mean % L =   12.60

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

       DOC = 0.000002   kg/L
       POC = 0.000000075 kg/L

   The log Kow derived in Appendix B for hexachlorobenzene is 5.600. The
   resulting value of ffd is 0.9013, and then:


       Measured baseline BAFTL4 = (8Q" ff*6 - 1)
-------
   The log Kow derived in Appendix B for hexachlorobutadiene is 4.842. The
   resulting value of ffd is 0.9812,  and then:
        Measured b.aseline BAFU4 =  (        - D(         > - 43'937
   This measured baseline BAF of 43,937 is considered a better value for trophic
   level 4 because it is based on concentrations in fish at trophic level 4.

alpha-Hexachlorocvclohexane (alpha-HCCH)

   The following laboratory-measured BCFs are available:

       BCF     % L     Baseline BCF       Reference

        140    3.1       4484            Canton etal. 1975,1978
        124    3.1       3968            Canton et al. 1975,1978
       1600    7.19     22239            Oliver and Niimi 1985
       2400    7.38     32507            Oliver and Niimi 1985

   Because the log Kow derived in Appendix B for alpha-HCCH is 3.776, which is
   less than 4, ffd is assumed to be 1 .0. The baseline BCFs are calculated using
   the equation given above.

       Geometric mean baseline BCF = 1 0650

   The FCM for trophic level 4 for  log Kow = 3.776 is 1 .04, which gives:

       Predicted baseline BAFTL4 = (10650H1.04) = 11076.

   The following field-measured BAF is available:

       BAF          % L        Species            Reference

       700        .  7.592      R. trout            Oliver and Niimi  1985

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of  POC and DOC are expected to be:

                 DOC  = 0.000002   kg/L
                 POC = 0.000000075 kg/L

   The log Kow derived in Appendix B for alpha-HCCH is 3.776.  The resulting
   value of ffd is 0.9984, and then:

                                   D-6

-------
        Measured baseline BAFlL4 = (^L. - D(^^) = 9,222

   A measured baseline BAF of 48,980 is derived in Table 8 based on Oliver and
   Niimi (1988); this is considered a better value and is used in the final Guidance
   because it is based on a more comprehensive set of data.

Hexachloroethane

   The following laboratory-measured BCFs are available:

                     % L        Species            Reference
       510         8.2         R. trout            Oliver and Niimi 1983
       1200         8.7         R. trout            Oliver and Niimi 1983

       Geometric mean BCF = 782
       Geometric mean % Lipid = 8.45

   Because the log Kow derived in Appendix B for hexachloroethane is 4.040, ffd is
   assumed to be 1:0.  Therefore:


            Measured baseline BCF =  {I8-2- - 1)(    1    ) = 9243


   The FCM for trophic level 4 for log Kow = 4.040 is 1.08, which gives:

       Predicted baseline BAFTL4 = (9243M1.08)  = 9982.


   The following field-measured BAF is available:

       BAF          % L        Species            Reference

       1,302        7.592      R. trout            Oliver and Niimi 1983

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the  concentrations of POC and DOC are expected to be:

                 DOC = 0.000002   kg/L
                 POC  = 0.000000075 kg/L

   The log Kow derived in Appendix B for hexachloroethane is 4.040. The
   resulting value of ffd is 0.9970, and then:

                                   D-7

-------
        Measured baseline BAFTL4 = (-±jj^ - 1 )( Q Q J5g2 ) = 17,188

Lindane

   The following laboratory-measured BCFs are available:

       BCF     % L     Baseline  BCF        Reference

        180    7.6        2355            Veith et al. 1979
       420    2.65      15811           Rogers et al. 1 983
       1200    7.19      16676           Oliver and Niimi  1 985
       2000    7.38      27087           Oliver and Niimi  1 985

   Because the log Kow derived in Appendix B for lindane is 3.673, which is less
   than 4, ffd is assumed to be  1 .0. The baseline BCFs are calculated using the
   equation given above.

       Geometric mean baseline BCF = 11 388

   The FCM for trophic level 4 for log Kow = 3.673 is 1.03, which gives:

        Predicted baseline BAFTL4  = (11388)(1.03) =  11730.

   The following field-measured BAF is available:

       BAF           % L       Species            Reference

       1000         7.592       R. trout            Oliver  and Niimi 1 985

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

                 DOC =  0.000002   kg/L
                 POC  = 0.000000075 kg/L

   The log Kow derived  in Appendix B for lindane is 3.673. The resulting value of
   ffd is 0.9987, and then:
        Measured baseline BAF^ . 1, - Dl) - 13,176
   A measured baseline BAF of 85,1 10 is derived in Table 8 based on Oliver and
   Niimi (1988); this is considered a better value and is used in the final Guidance


                                   D-8

-------
    because it is based on a more comprehensive set of data..

Mercury

    See Appendix E.

Methvlene Chloride ,

    Based on a predicted BCF and a FCM. See Appendix H.

Mirex

    The following field-measured BAF is available:

         BAF     •   % L       Species            Reference

       15,000,000  7.592      R. trout            Oliver and Niimi 1985

    Trout are expected to be in trophic level  4. These data were obtained from
    Lake Ontario, in which the concentrations of  POC and DOC are expected to be:

                 DOC  =  0.000002   kg/L
                 POC  = 0.000000075 kg/L

    The log Kow derived  in Appendix B for mirex is 6.89. The resulting value of ffd
    is 0.3190, and then:
   Measured baseline BAFTL4 = ( 1 *          - D(        } = 619'361'730
   A measured baseline BAF of 134,900,000 is derived in Table 8 based on Oliver
   and Niimi (1988); this is considered a better value and is used in the final
   Guidance because it is based on a more comprehensive set of data.

Octachlorostyrene

   The following field-measured BAF is available:

         BAF         % L       Species           Reference

       1,400,000    7.592      R. trout           Oliver and Niimi 1985

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of  POC and DOC are expected to be:


                                   D-9

-------
                 DOC = 0.000002   kg/L
                 POC = 0.000000075 kg/L

   The log Kow derived in Appendix B for octachlorostyrene is 6.29.  The resulting
   value of ffd is 0.6510, and then:
    Measured baseline BAF^ = (  '          - 1 )(__) = 28,326,351
   A measured baseline BAF of 1 17,500,000 is derived in Table 8 based on Oliver
   and Niimi (1988); this is considered a better value and is used in the final
   Guidance because it is based on a more comprehensive set of data.

PCBs

   See Appendix F.

Pentachlorobenzene

   The following field-measured BAFs are available:

        BAF          %  L       Species            Reference

       56,570       20.9       L. trout            Oliver and Nicol 1 982
       16,150        7.592      R. trout            Oliver and Niimi 1 983

         Geometric mean BAF = 30,226
         Geometric mean % L =   12.60

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

                 DOC =  0.000002   kg/L
                 POC  = 0.000000075 kg/L

   The log Kow derived in Appendix B for pentachlorobenzene is 5.106. The
   resulting value of ffd  is 0.9661,  and then:
        Measured baseline BAFTL4 . <        - 1)-) = 248'299
   A measured baseline BAF of 645,700 is derived in Table 8 based on Oliver and
   Niimi (1988); this is considered a better value and is used in the final Guidance
   because it is based on a more comprehensive set of data.

                                   D-10

-------
2.3.7.8-TCDD

   Based on the BSAF methodology. See Section III.E and Appendix H.

1.2.3.4-Tetrachlorobenzene

   The following field-measured BAFs are available:

        BAF          % L       Species            Reference

      69,280       20.9       L. trout            Oliver and Nicol 1982
       8,769         7.592     R. trout            Oliver and Niimi 1983
       7,700      .   7.592     R. trout            Oliver and Niimi 1985

        Geometric mean BAF = 16,724
        Geometric mean % L =    10.64

   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

                 DOC = 0.000002   kg/L
                 POC  = 0.000000075 kg/L

   The log Kow derived in Appendix  B for 1,2,3,4-tetrachlorobenzene is 4.592.
   The resulting value of ffd is 0.9894, and then:


        Measured baseline  BAFTL4 = (16'724 - 1)(	]	) = 158,855
                                  0.9894      0.1064

   A measured baseline BAF of 117,500 is derived in Table 8 based on Oliver and
   Niimi (1988); this is considered a better value and is used in the final  Guidance
   because it is based on a more comprehensive set of data.

1.2.4.5-Tetrachlorobenzene

   The following field-measured BAFs are available:

        BAF          % L       Species            Reference
       31,620       20.9       L. trout           Oliver and Nicol 1982
       5,034         7.592     R. trout           Oliver and Niimi 1983

         Geometric mean BAF = 12,616
         Geometric mean % L =    12.60
                                  D-11

-------
   Trout are expected to be in trophic level 4. These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:

                 DOC = 0.000002   kg/I
                 POC = 0.000000075 kg/L

   The log Kow derived in Appendix B for 1 ,2,4,5-tetrachlorobenzene is 4.557.
   The resulting value of ffd is 0.9902, and then:


        Measured baseline BAF^ = (        - ^"1 = 101'110
Toluene

    Based on a predicted BCF and a FCM. See Appendix H.

Toxaphene

    The following field-measured BAF is available:

         BAF      .      % L       Species          Reference

       1,778,636       8.284      L. trout          Swain et al. 1986

    Trout are expected to be at trophic level 4.  These data were obtained from
    Siskiwit Lake, in which the concentrations of POC and DOC are expected to be
    similar to those in Lake Superior:

                 DOC  = 0.000002   kg/L
                 POC = 0.00000004 kg/L

    The log Kow derived in Appendix B for toxaphene is 4.330.  The resulting value
    of ffd  is 0.9949, and then:
     Measured baseline BAFTL4 = ( 1 '         ~ ^          = 21'580'789
1 ,2.4-Trichlorobenzene

    The following laboratory-measured BCFs are available:

       BCF       % L    Baseline BCF   Reference

      2800       7.6      36829       Veith et al. 1979


                                   D-12

-------
  1600
    85
   349
    39
  1300
  3200
  2300
  3700
   124
   248
   498
   914
   769
   769
  1127
  1365
  1442
   991
   410
  2026
9.12
2.1(e)
3.2(h)
0.7(a)
8.2
8.7
7.19
7.38
1.8
2.2
4.4
5.0
5.2
5.2
5.7
5.8
7.7
8.2
3.79
11.4
17533
4000
10875
5429
15841
36770
31975
50122
6833
11227
11295
18260
14769
14769
19754
23517
18714
12073
10792
17763
                    Kosian et al. 1981
                    Galassi and Calamari 1983
                    Galassi and Calamari 1983
                    Galassi and Calamari 1983
                    Oliver and Niimi 1983
                    Oliver and Niimi 1983
                    Oliver and Niimi 1985
                    Oliver and Niimi 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Geyer et al. 1985
                    Carlson and Kosian 1987
                    Smith etal. 1990
Because the log Kow derived in Appendix B for 1,2,4-trichlorobenzene is 3.990,
which is less than 4, ffd is assumed to be 1.0.  The baseline BCFs are
calculated using the equation given above.

   Geometric mean baseline BCF = 15497

The FCM for trophic level 4 for log Kow  =  3.990 is 1.07, which gives:

   Predicted baseline BAFTL4 = (15497)(1.07) = 16582

A field-measured baseline BAF of 37,154 is given in Table 2 of Section 3 for
sculpin, which is at trophic  level 3. For this chemical, the log Kow is 3.990,
and so the FCM for trophic  level 3 is 1.24 and the FCM for trophic level 4 is
1.07. This results in a baseline BAF of (37,154)(1.07)7(1.24) =  32,060 for
trophic level 4.  •

The following field-measured BAFs are available:

     BAF          % L       Species           Reference
    5,270
     899.5
    1,200
20.9
 7.592
 7.592
L. trout
R. trout
R. trout
Oliver and Nicol 1982
Oliver and Niimi 1983
Oliver and Niimi 1985
                               D-13

-------
         Geometric mean BAF = 1,785
         Geometric mean % L =   10.64

   Trout are expected to be in trophic level 4.  These data were obtained from
   Lake Ontario, in which the concentrations of POC and DOC are expected to be:
                  DOC  = 0.000002   kg/L
                  POC = 0.000000075 kg/L
   The log Kow derived in Appendix B for 1 ,2,4-trichlorobenzene is 3.990. The
   resulting value of ffd is 0.9973, and then:
         Measured baseline BAFTL4 = (         -  ^o/meV  =  16'812

   This measured baseline BAF of 16,812 is considered a better value for trophic
   level 4 because it is based on concentrations in fish at trophic  level 4.

Trichloroethylene

   Based on a predicted BCF and a FCM.  See Appendix H.

References

Canton, J.H., P.A. Greve, W. Slooff, and G.J. van Esch.  1975. Toxicity,
    Accumulation and Elimination Studies of o-Hexachlorocyclohexane (cr-HCH)
    with Freshwater Organisms of Different Trophic Levels.  Water Res.  9:1 163-
    1169.

Canton, J.H., R.C.C. Wegman, T.J.A. Vulto, C.H. Verhoef, and G.J. van Esch.
    1978.  Toxicity-, Accumulation- and Elimination Studies of a-
    Hexachlorocyclohexane (a-HCH) with Saltwater Organisms of Different Trophic
    Levels.  Water Res. 12:687-690.

Carlson, A.R., and P.A. Kosian.  1987. Toxicity of Chlorinated Benzenes to
    Fathead Minnows (Pimephales promelas).  Arch. Environ. Contam. Toxicol.
    16:129-135.

Galassi, S., and D.  Calamari. 1983.  Toxicokinetics of 1,2,3 and 1,2,4
    Trichlorobenzenes in Early Life Stages of Salmo gairdneri.  Chemosphere
    12:1599-1603.-

Galassi, S., D. Calamari, and F. Setti. 1982.  Uptake and Release of p-
    Dichlorobenzene in Early Life Stages of Salmo gairdneri. Ecotoxicol. Environ.
    Safety 6:439-447.
                                   D-14

-------
Geyer, H., I. Scheunert, and F. Korte.  1985.  Relationship between the Lipid
    Content of Fish and Their Bioconcentration Potential of 1,2,4-
    Trichlorobenzene. Chemosphere 14:545-555.

Konemann, H., and K. van Leeuwen. 1980.  Toxicokinetics in Fish: Accumulation
    and Elimination of Six Chlorobenzenes by Guppies.  Chemosphere 9:3-19.

Kosian, P., A. Lemke, K.  Studders, and G. Veith. 1981. The Precision of the
    ASTM Bioconcentration Test.  EPA 600/3-81-022.  National Technical
    Information Service,  Springfield, VA.

Oliver, B.G., and K.D. Nicol.  1982. Chlorobenzenes in  Sediments, Water, and
    Selected Fish from Lakes Superior, Huron, Erie, and Ontario. Environ. Sci.
    Technol. 16:532-536.

Oliver, B.C., and A.J.. Niimi. 1983. Bioconcentration of Chlorobenzenes from
    Water by Rainbow Trout: Correlations with Partition Coefficients and
    Environmental  Residues. Environ. Sci. Technol. 17:287-291.

Oliver, B.G., and A.J. Niimi. 1985. Bioconcentration Factors of Some
    Halogenated Organics for Rainbow Trout: Limitations in Their Use for
    Prediction of Environmental Residues. Environ. Sci. Technol. 19:842-849.

Oliver, B.G., and A.J. Niimi. 1988. Trophodynamic Analysis of Polychlorinated
    Biphenyl Congeners and Other Chlorinated Hydrocarbons in the Lake Ontario
    Ecosystem.  Environ. Sci. Technol. 22:388-397.

Rogers, J.H., Jr., K.L. Dickson, and M.J. DeFoer.  1983. Bioconcentration of
    Lindane and Naphthalene in Bluegills (Lepomis macrochirus). In: Aquatic
    Toxicology and Hazard Assessment:  Sixth Symposium. W.E. Bishop, R.D.
    Cardwell, and B.B. Heidolph, Eds. ASTM STP 802.  American Society for
    Testing and Materials, Philadelphia, PA. pp.  300-311.

Smith, A.D., A.  Bharath,  C. Mallard, D. Orr, L.S. McCarty, and G.W. Ozburn.
    1990.  Bioconcentration Kinetics of Some Chlorinated Benzenes and
    Chlorinated Phenols in American Flagfish, Jordanella floridae (Goode and
    Bean).  Chemosphere 20:379-386.

Swain, W.R., M.D.  Mullin, and J.C. Filkins. 1986.  Long Range Transport of Toxic
    Organic Contaminants to the North American Great Lakes.  IN: Problems of
    Aquatic Toxicology,  Biotesting, and Water Quality Management. R.C. Ryans,
    ed. EPA/600/9-86/024. National Technical Information Service, Springfield,
    VA.  pp. 107-121.
                                   D-15

-------
Veith, G.D., D.L. DeFoe, and B.V. Bergstedt. 1979.  Measuring and Estimating the
    Bioconcentration Factor of Chemicals in Fish.  J. Fish. Res. Bd. Canada
    36:1040-1048.
                                   D-16

-------
Appendix E.  Derivation of Baseline BAFs for Mercury

In the Gobas model, which is used in the derivation of BAFs and FCMs for organic
chemicals, only bioconcentration applies to trophic levels 1 and 2, whereas
biomagnification occurs between trophic levels 2 and 3 and between trophic levels
3 and 4. In their study with mercury, however, Watras and Bloom (1992) found
that biomagnification occurred between trophic levels 1 and 2 and between trophic
levels 2 and 3.  Watras and Bloom (1992) only studied trophic levels 1, 2, and 3,
but a substantial amount of data from other investigators show a biomagnification
factor between fishes.  Thus the model used here with mercury will provide for
bioconcentration at trophic level 1, and biomagnification at trophic levels 2, 3, and
4.

The BCFs for inorganic mercury and methylmercury are 2,998 and 52,175 (U.S.
EPA 1985).  It is possible that  the higher BCFs obtained in the tests with fathead
minnows should not be used because they reflect some bioaccumulation, not just
bioconcentration, due to the fact that this species is a grazer and therefore
possibly ate food that contained mercury.  Accumulation through food is
considered negligible, however, because this species does not do well in chronic
tests unless food is provided; it is unlikely that grazing would provide a substantial
amount of food or  mercury.  It is, of course, also possible that the food provided
for the fish might rapidly sorb mercury from the water; there is no reason to
believe that  such sorption  is substantial or that it occurs more in bioconcentration
tests with one species than with the  other.  Another possibility is that lower BCFs
were obtained with salmonids than with fathead minnows because of growth
dilution.  Several investigators  have determined BCFs for organic chemicals with
small fish, such as guppies, to  reduce or avoid the effects  of growth dihjtion. If
growth dilution occurs, bioconcentration tests with  salmonids would produce BCFs
that are too  low unless calculation of the results accounts for growth dilution.

Based on the data  of Gill and Bruland (1990),  it will be assumed that, on the
average, 17 percent of the total mercury in the Great Lakes is methylmercury and
that 83 percent is inorganic mercury. Thus the weighted average BCF is:
(0.17)(52,175)  +  (0.83)(2,998)  = 11,358. Based on data for phytoplankton,
Watras and Bloom  (1992)  obtained a BCF of about 25,000 for total  mercury at a
pH of 6.1.  This pH is below 6.5 and therefore this  BCF might not be appropriate
for use in the derivation of water quality criteria.

The data of Watras and Bloom (1992) show an increase of about a factor of 2
from trophic level 1 to trophic level 2, and an  increase of about a factor of 1.26
from trophic level 2 to trophic level 3.

A variety of studies have found that total mercury increases from prey fish to
predator fish by factors ranging from  1.2 to 15,  with a mean of about 5:


                                     E-1

-------
                7.7 to 9.2               MacCrimmon et al. 1983
                up to 8.4                Wren et al. 1983
                up to 6 and 13           Skurdal et al. 1985
                8 and 15                Mathers and Johansen 1985
                2.9                     Parks 1988
                6.4                     Cope et al. 1990

The BCF and BMFs derived above result in:
          (11,358){2.00)  =  22,716
          (22,716)(1.26)  =  28,622
          (28,622)(5.00)  = 143,110

The corresponding FCMs are:

          Trophic Level 2: FCM  = 22,716/11,358  = 2.00
          Trophic Level 3: FCM  = 28,622/11,358  = (2.00H1.26)
          Trophic Level 4: FCM  = 143,110/11,358 = (2.00)(1.26)(5.00)

Bloom (1992) concluded that "for all species studied, virtually all (>95%) of the
mercury present is as CH3Hg and that  past reports of substantially lower CH3Hg
fractions may have been biased by analytical and homogeneity variability".
Therefore, it will be assumed that 97.5 percent of the mercury in fish in the Great
Lakes is methylmercury:
          (28,622)(0.975) = 27,906
          (143,110)(0.975) = 139,532

Although McKim et al. (1976)  and Heiskary and  Helwig (1983) found higher
concentrations of mercury in the edible portion of fish than in the whole body,
Huckabee et al. (1974) and Heisinger et al. (1979) found the same concentration
in whole body and muscle tissue.  Thus for a specific trophic level, the human
health and wildlife BAFs will be the same.

This derivation indicates that for total  mercury in the water column the baseline
BAFs should be:

               Trophic level                   Baseline BAF

                      3                         27,906
                      4                        139,532

The difference between trophic levels  3 and 4 is important.

A.  Comparison of field-measured BAFs for mercury with the BAFs derived above
    must properly identify the trophic level of the aquatic biota used in the

                                    E-2

-------
    determination of the field-measured BAF. If field-measured BAFs are
    compared to the BAF derived for trophic level 4, the field-measured BAFs must
    have been determined with aquatic biota that are in trophic level 4. Many of
    the field-measured BAFs for mercury have been determined with aquatic biota
    that is in trophic level 3. It might also be necessary to account for a different
    percent methylmercury in the water column.  In addition, the age of the fish is
    probably important because the concentration of mercury in fish seems to
    increase consistently with age without showing signs of leveling off.

B.  If the aquatic biota consumed by humans and wildlife is incorrectly assigned to
    too high a trophic level on the average, the  resulting criteria will be
    unnecessarily low, but not because the derived BAFs for mercury are too high.
    For example, if all the consumed food is assumed to be trophic level 4, the
    BAF used to derive the criterion will be 139,532.  If, however, the consumed
    food is actually a  1:1 combination  of trophic levels 3 and 4, the BAF of
    139,532 would be used with half of the consumed food, and a BAF of 27,906
    would be used with  the other half of the consumed food.

C.  Identification of the trophic level of some species of fish must take into
    account the age and/or size of the  specific organisms of concern.  Some
    species of fish are in trophic level 3 when they are young, but are in trophic
    level 4 when they are older. The trophic level might also vary from one body
    of water to another, depending on  the food chain.  With both humans and
    wildlife, knowing the species consumed is not necessarily sufficient to allow
    an accurate identification of the trophic level of the consumed food.

EPA has completed a more comprehensive analysis of data concerning the
bioaccumulation of mercury by fish, which is being peer reviewed at this time.
The final Guidance had intended to use the baseline BAFs contained in the initial
draft of the report but  it was decided to wait until the report has been peer-
reviewed and completed.  The initial draft of the report contained higher baseline
BAFs than those derived herein.

References

Bloom, N.S. 1992.  On the Chemical Form of Mercury in Edible Fish and Marine
    Invertebrate Tissue.  Can. J. Fish. Aquat. Sci. 49:1010-1017.

Cope, W.G., J.G. Wiener, and R.G. Rada.  1990.  Mercury Accumulation in Yellow
    Perch in Wisconsin Seepage Lakes: Relation to Lake Characteristics.  Environ.
    Toxicol. Chem. 9:931-940.

Gill, G.A., and K.W.  Bruland. 1990.  Mercury Speciation in Surface Freshwater
    Systems in California and Other Areas. Environ. Sci. Technol. 24:1392-1400.


                                    E-3

-------
Heisinger, J.F., C.D. Hansen, and J.H. Kim.  1979. Effect of Selenium Dioxide on
    the Accumulation and Acute Toxicity of Mercuric Chloride in Goldfish.  Arch.
    Environ. Contam. Toxicol. 8:279-283.

Heiskary, S.A. and D.D. Helwig.  1983. Acid Rain Intensive Study Program.
    Status Report for the 1981 Study Lakes.  Minnesota Pollution Control Agency,
    Roseville, MN.

Huckabee, J.W., C. Feldman, and Y. Talmi.  1974.  Mercury Concentrations in Fish
    from the Great Smoky Mountains National Park. Anal. Chimica Acta 70:41-
    47.

MacCrimmon, H.R., C.D. Wren, and B.L. Gots. 1983.  Mercury Uptake by Lake
    Trout, Salvelinus namaycush, Relative to Age, Growth, and Diet in Tadenac
    Lake with Comparative Data from Other PreCambrian Shield Lakes. Can. J.
    Fish. Aquat. Sci. 40:114-120.

McKim, J.M., G.F. Olson, G.W. Holcombe, and E.P. Hunt. 1976. Long-term
    Effects of Methylmercuric Chloride on Three Generations of Brook Trout
    (Salvelinus fontinafis): Toxicity, Accumulation, Distribution,  and Elimination. J.
    Fish. Res. Bd. Canada. 33: 2726-2739.

Mathers,  R.A., and  P.H. Johansen. 1985. The Effects of Feeding Ecology on
    Mercury Accumulation in Walleye (Stizostedion vitreum) and Pike (Esox lucius)
    in  Lake Simcoe. Can. J. Zool.  63:2006-2012.

Parks, J.W.  1988.   Selected Ecosystem Relationships in the Mercury
    Contaminated Wabigoon-English River System, Canada, and Their Underlying
    Causes.  Water Air Soil Pollut. 42:267-279.

Skurdal, J., T. Qvenild, and O.K. Skogheim.  1985. Mercury Accumulation in Five
    Species of Freshwater Fish in Lake Tyrifjorden, South-East Norway, with
    Emphasis on Their Suitability as Test Organisms.  Environ. Biol. Fish. 14:233-
    237.

U.S. EPA. 1985.  Ambient Water Quality Criteria for Mercury -  1984. EPA 440/5-
    84-026.  National Technical Information Service, Springfield, VA.

Watras, C.J., and N.S. Bloom.  1992.  Mercury and Methylmercury in Individual
    Zooplankton: Implications for Bioaccumulation. Limnol.  Oceanogr. 37:1313-
    1318.

Wren, C.D.,  J.R. MacCrimmon, and B.R. Loescher.  1983.  Examination of
    Bioaccumulation and Biomagnification of Metals in a PreCambrian Shield  Lake.


                                    E-4

-------
Appendix F. Derivation of Baseline BAFs for PCBs

Although a Kow can usefully describe the partitioning of a mixture between octanol
and water, the relation between Kows and BAFs is more uncertain for mixtures than
for individual chemicals.  The additional uncertainty occurs because the
composition of the mixture will differ from one phase to another, due to differential
partitioning and to differences in metabolism by aquatic organisms. The
uncertainty increases as the magnitudes of the differences between the properties
of the individual components of the mixture increase.

Although Burkhard and Kuehl (1986), Burkhard et al. (1985), Chiou et al. (1977),
de Bruijn et al. (1988), Karickhoff et al. (1979), Miller et al. (1985), Rapaport and
Eisenreich (1984), Veith et al. (1979a), and Woodburn et al. (1984) have published
measured values for the log Kow of various PCB mixtures and congeners, the set of
values published by Hawker and Connell (1988) is considered the best for use in
the final Guidance. Similarly, laboratory-measured BCFs and BAFs have been
reported in such publications as Bruggeman et al.  (1981), Gobas and Schrap
(1990), Gobas et al. (1989), Hansen et al. (1971), Oliver and Niimi (1984,  1985),
Snarski and Puglisi (1976), Veith et al. (1979a,b), but the data reported by Oliver
and Niimi (1988) are considered the best for use in the final Guidance.

Hawker and Connell (1988) and Oliver and Niimi (1988) contain Kows and BAFs,
respectively, for individual PCB congeners and so mean values can be calculated
for various of mixtures. Calculation of an arithmetic mean of the logarithms of the
Kows or BAFs is equivalent to calculation of a geometric mean of the Kows or BAFs.
A mean that is calculated by giving each value the same weight is often called an
unweighted mean; alternatively a mean can be calculated by giving a weight of 1
to some values  and giving a weight of 0 to all other values. Another alternative is
to assign weights based on the relative amounts of the congeners in commercial
mixtures or in organisms,  water, and/or sediment, based on data reported in such
publications as Schulz et al. (1989) and Oliver and Niimi (1988).

For the purpose of the final Guidance, it seems most appropriate to assign weights
based on the concentrations in fish in the Great Lakes, because these represent the
congeners that are ingested the most by eating aquatic life from the Great Lakes.
Table F1 contains the relevant information and most of the necessary calculations.
The results are:

                   Mean log Kow =  26/735.25  = 6.5394-19
                                    4,057.3

        Weighted geometric mean Kow = 3,885,000
                                    F-1

-------
                 Mean log BAF^ =   ''    = 7.742575


        Weighted geometric mean BAF^ = 55,281,000


                 Mean log BAFTL4 = 32''1  = 8.066525
        Weighted geometric mean BAF^ = 116,553,000

These mean values are used when generic values are needed for PCBs in the final
Guidance.

By using a log Kow of 6.589, FCMs from Table 2, and equation 32 from Appendix
C, the following results are obtained:

    For trophic level 3:
                  . =  13.94
            predicted BAFTL3 = 54,110,000

    For trophic level 4:

            FCMTL4 =  25.53
            predicted BAF^ = 99,090,000

The weighted geometric mean field-measured BAFs calculated above are higher
than these predicted BAFs.

It is also possible to calculated a "mean" BAFTL4 for PCBs from the data given by
Oliver and Niimi (1988)  for total PCBs in water and salmonids:

          RAF   = (4300 ng/g)(1000 pg/ng)(1000g/l) = 73 47Q OOQ
              714  .     (1100 pg/L)(0.1 1X0.4837)        o.*/vfuuu

where 0.1 1 is the fraction of the salmonids that was lipid and 0.4837 is the
fraction dissolved that is calculated for a chemical with log Kow = 6.589 in Lake
Ontario. This value is lower that both of the above values for BAFTL4.

References

Bruggeman, W.A., UB.J.M. Martron, D. Kooiman, and O. Hutzinger.  1981.
    Accumulation and Elimination Kinetics of Di-, Tri- and Tetra Chlorobiphenyls by
    Goldfish after Dietary and Aqueous Exposure.  Chemosphere 10:81 1-832.


                                    F-2

-------
Burkhard, L.P., and D.W. Kuehl. 1986.  N-Octanol/Water Partition Coefficients by
    Reverse Phase Liquid Chromatography/Mass Spectrometry for Eight
    Tetrachlorinated Planar Molecules. Chemosphere 15:163-167.

Burkhard, L.P., D.W. Kuehl, and G.D. Veith.  1985.  Evaluation of Reverse Phase
    Liquid Chromatography/Mass Spectrometry for Estimation of N-Octanol/Water
    Partition Coefficients for Organic Chemicals. Chemosphere 14:1551-1560.

Chiou, C.T., V.H. Freed, D.W. Schmedding, and R.L. Kohnert. 1977.  Partition
    Coefficients and Bioaccumulation of Selected Organic Chemicals.  Environ.
    Sci. Technol. 11:475-478.

de Bruijn, J., F.  Busser, W. Seinen, and J. Hermens. 1989.  Determination of
    Octanol/Water Partition Coefficients for Hydrophobic Organic Chemicals with
    the "Slow-Stirring" Method. Environ. Toxicol. Chem. 8:449-512.

Gobas, F.A.P.C., and S.M. Schrap.  1990. Bioaccumulation of Some
    Polychlorinated Dibenzo-p-dioxins and Octachlorodibenzofurans in  the Guppy
    (Poecilia reticulata). Chemosphere 20:495-512.

Gobas, F.A.P.C., K.E. Clark, W.Y. Shiu, and D. Mackay.  1989.  Bioconcentration
    of Polybrominated Benzenes and Related Superhydrophobic Chemicals in Fish:
    Role of Bioavailability and Elimination into the Feces. Environ. Toxicol. Chem.
    8:231-245.

Hansen, D.J., P.R. Parrish, J.I. Lowe, A.J. Wilson, Jr., and P.O. Wilson.  1971.
    Chronic Toxicity, Uptake, and Retention of Aroclor 1254 in  Two Estuarine
    Fishes.  Bull. Environ. Contam. Toxicol. 6:113-119.

Hawker, D.W., and D.W. Connell.  1988. Octanol-Water Partition Coefficients of
    Polychlorinated Biphenyl Congeners.  Environ. Sci. Technol. 22:382-387.

Karickhoff, S.W., D.S.  Brown, and T.A Scott. 1979. Sorption of Hydrophobic
    Pollutants on Natural Sediments.  Water Research  13:241-248.

Miller, M.M., S.P. Wasik, G.-L. Huang, W.-Y. Shiu, and D. Mackay. 1985.
    Relationships between Octanol-Water Coefficient and Aqueous Solubility.
    Environ. Sci. Technol.  19:522-529.

Oliver, B.C., and A.J. Niimi.  1984.  Rainbow Trout Bioconcentration of Some
    Halogenated Aromatics from Water at Environmental Concentrations.  Environ.
    Toxicol. Chem. 3:271-277.

Oliver, E.G., and A.J. Niimi.  1985.  Bioconcentration Factors of Some


                                    F-3

-------
    Halogenated Organics for Rainbow Trout: Limitations in Their Use for
    Prediction of Environmental Residues. Environ. Sci. Technol. 19:842-849.

Oliver, B.G., and A.J. Niimi.  1988. Trophodynamic Analysis of Polychlorinated
    Biphenyl Congeners and Other Chlorinated Hydrocarbons in the Lake Ontario
    Ecosystem.  Environ. Sci. Technol. 22:388-397.

Rapaport, R.A., and S.J. Eisenreich.  1984.  Chromatographic Determination of
    Octanol-Water Partition Coefficients (Kow's) for 58 Polychlorinated Biphenyl
    Congeners.  Environ. Sci. Technol. 18:163-170.

Schulz, D.E., G. Petrick, and J.C. Duinker.  1989. Complete Characterization of
    Polychlorinated Biphenyl Congeners in Commercial Aroclor and Ciophen
    Mixtures by Multidimensional Gas Chromatography-Electron Capture
    Detection.  Environ. Sci. Technol. 23:852-859.

Snarski, V.M., and F.A. Puglisi.  1976. Effects of Aroclor 1254 on Brook Trout,
    Salvelinus fontinalis.  EPA-600/3-76-112.  National Technical Information
    Service, Springfield, VA.

Veith, G.D., N.M. Austin, and R.T. Morris.  1979a. A Rapid Method for Estimating
    Log P for Organic Chemicals.  Water Res.  13:43-47.

Veith, G.D., D.L. DeFoe, and B.V. Bergstedt.  1979b.  Measuring and Estimating
    the Bioconcentration Factor of Chemicals in Fish.  J. Fish. Res. Bd. Can.
    36:1040-1048.

Woodburn, K.B., W.J. Doucette, and A.W. Andren.  1984.  Generator Column
    Determination of Octanol/Water Partition Coefficients for Selected
    Polychlorinated Biphenyl Congeners. Environ. Sci. Technol. 18:457-459.
                                     F-4

-------
Table F.1.  Log Kows and BAFs for PCB Congeners
Congener

28+31
18
22 (1.7)
16 (0.3)
33 (0.3)
17 (0.3)
32 (0.3)
66
70+76
56+60+81
52
47+48
44
74
49
64
42
53 (1.5)
40(1.3)
101
84
118
110
87+97
105
95
85
Weigh
t

36.
4.3
0.
0.
0.
0.
0.
160.
140.
74.
62.
60.
45.
38.
31.
28.
10.
0.
0.
270.
260.
250.
230.
200.
110.
80.
58.
Log KQW

5.67
5.24
5.58
5.16
5.60
5.25
5.44
6.20
6.17
6.19
5.84
5.82
5.75
6.20
5.85
5.95
5.76
5.62
5.66
6.38
6.04
6.74
6.48
6.29
6.65
6.13
6.30
Product
(LogKoW)
204.12
22.53
0.00
0.00
0.00
0.00
0.00
992.00
863.80
458.06
362.08
349.20
258.75
235.60
181.35
166.60
57.60
0.00
0.00
1,722.60
1,570.40
1,685.00
1,490.40
1,258.00
731.50
490.40
365.40
BAF
(Scul)
6.37
5.97





7.45
7.06
7.48
6.80
6.15
6.65
7.30
6.77
7.16
7.07


7.30
8.05
7.86
7.44
7.54
7.82
6.98
7.50
BAF
(Ale)
6.68
6.39





7.57
7.31
7.79
6.84
6.85
6.86
7.35
6.98
7.30
7.38


7.25
7.90
7.71
7.51
7.89
7.72
7.14
7.67
Product
ave(Sc+Al)
234.90
26.57
0.00
0.00
0.00
0.00
0.00
1,201.60
1,005.90
564.99
422.84
390.00
303.98
278.35
213.13
202.44
72.25
0.00
0.00
1,964.25
2,073.50
1,946.25
1,719.25
1,543.00
854.70
564.80
439.93
BAF
(Salmon)
6.89
5.75
6.39
5.92
5.32
5.52
6.76
7.79
7.56
7.96
7.01
7.18
6.96
7.66
7.13
7.51
7.49
6.51
6.55
7.45
8.28
8.15
7.79
8.08
8.13
7.25
7.89
Product
(Salmon)
248.04
24.73
0.00
0.00
0.00
0.00
0.00
1,246.40
1,058.40
589.04
434.62
430.80
313.20
291.08
221.03
210.28
74.90
0.00
0.00
2,011.50
2,152.80
2,037.50
1,791.70
1,616.00
894.30
580.00
457.62
                                  F-5

-------
Table F.1. (Continued). Log Kows and BAFs for PCB Congeners
Congener

92
82
91
Weigh
t

53.
29.
29.
Log KQW

6.35
6.20
6.13
Product
(LogKoW)
336.55
179.80
177.77
BAF
(Scul)
7.70
7.60
6.44
BAF
(Ale)
7.93
7.86
6.74
Product
ave(Sc+Al)
414.20
224.17
191.11
BAF
(Salmon)
8.11
8.13
6.92
Product
(Salmon)
429.83
235.77
200.68
                                   F-6

-------
Table F.1.  (Continued).  Log Kows and BAFs for PCB Congeners
Congener

99(20)
153
138
149
146
141
151
132
136
180
187+182
170+190
183
177
174
203+1%
194
SUM
Weigh
t

0.
430.
260.
190.
88.
83.
51.
39.
31.
200.
130.
84.
71.
36.
32.
52.
23.
4,057.3
Log KOW

6.39
6.92
6.83
6.67
6.89
6.82
6.64
6.58
6.22
7.36
7.19
7.37
7.20
7.08
7.11
7.65
7.80

Product
(Log KOW)
0.00
2,975.60
1,775.80
1,267.30
606.32
566.06
338.64
256.62
192.82
1,472.00
934.70
619.08
511.20
254.88
227.52
397.80
179.40
26,735.25
BAF
(Scul)

8.05
8.06
7.28
8.49
8.11
8.34
7.41
7.13
8.45
8.07
9.15
8.81
8.63
8.24
9.14
8.52

BAF
(Ale)
7.37
7.82
7.89
7.75
8.30
7.96
8.17
7.45
7.25
8.15
7.99
8.84
8.46
8.54
8.51
8.82
8.22

Product
ave(Sc+Al)
0.00
3,412.05
2,073.50
1,427.85
738.76
666.91
421.01
289.77
222.89
1,660.00
1,043.90
755.58
613.09
309.06
268.00
466.96
192.51
31,413.95
BAF
(Salmon)
7.39
8.32
8.30
7.99
8.73
8.32
8.51
7.56
7.37
8.58
8.43
9.20
9.03
9.01
8.74
9.26
8.56

Product
(Salmon)
0.00
3,577.60
2,158.00
1,518.10
768.24
690.56
434.01
294.84
228.47
1,716.00
1,095.90
772.80
641.13
324.36
279.68
481.52
196.88
32,728.31
The weights are those reported by Oliver and Niimi (1988) for salmonids. Oliver and Niimi (1988) did
not report the concentrations of congeners 22, 16, 33, 17, 32, 53, 40, and 99 in sculpin and/or alewives.
To avoid irregularities in the treatment of the data, these eight were all assigned weights of zero.  The
actual weights of the eight 'are given in parentheses in the first column.  The total  weight for the eight is
25.7, which means that the total weight of all congeners in salmonids was 4083; the eight constitute about
0.6 percent of the total for all congeners.
The log
             and BAFs are from Tables 4, 5, and 8.
                                              F-7

-------
Appendix G. Baseline BAFs for Trophic Level Four by Four Methods

The purpose of this appendix is to identify how many of the four methods in the
final Guidance have been used to derive baseline BAFs for 31 chemicals for:

    1)  Use in deriving human health criteria for chemicals in Table 3 of part 132
    2)  Use in deriving wildife criteria for chemicais on Table 4 of part 132
    3)  Use in determining the bioaccumulative chemicals of concern in Table 6a
        of part 132.

Baseline BAFs for the other 107 chemicals of initial focus will be available in a
separate document, "Derivation of Human Health and Wildlife Bioaccumulation
Factors for the Great Lakes Initiative."  Because they are referenced to a standard
set of conditions, baseline BAFs for trophic level 4 are used in this appendix,
although those for trophic level 3 could have been used.  For each chemical,
baseline BAFs were derived by each of the four methods whose data requirements
were satisfied.

For inorganic chemicals, the BAF based on the  wet weight of muscle tissue of
species consumed by humans is used as the baseline BAF.

The four baseline BAFs that can be  derived for organic chemicals using the
methods described in the final Guidance are:

1.   A measured baseline BAF that is based on field data that includes the
    measured concentrations of the chemical in tissue of aquatic organisms and in
    the ambient water.
        A  measured baseline BAF is calculated from a field-measured BAFj by
        using the equation:
            ,-       ,.   ,.    _._    . measured BAFj   -.,1.
            Measured baseline BAF =  (	 -  1) (J-)
                                            'fd             't

        Except as noted, the measured baseline BAFs are from Table 8 in Section
        III.D.

2.   A predicted baseline BAF that is based on  BSAF methodology.
        All the baseline BAFs predicted using BSAF methodology are from Table
        10.

3.   A predicted baseline BAF that is based on  a laboratory-measured BCF and a
    Food-Chain Multiplier (FCM); the FCM is 1  for most inorganic chemicals and is
    derived from log Kow for organic chemicals.
        A  predicted baseline BAF is calculated from a measured BCF| by using
        the equation:

                                    G-1

-------
           Predicted baseline BAF      = (FCM) ( measured BCF^  - 1) (1)
                                                     ffd             ^1

       Except as noted, baseline BAFs based on laboratory-measured BCFs and
       FCMs are derived in Appendix D.

4. A predicted baseline BAF that is based on a predicted BCF and a FCM, where
   the predicted baseline BCF equals Kow and the FCM is derived from log Kow.
       A predicted baseline BAF is calculated from a predicted BCF by using the
       equation:

                   Predicted baseline BAF  = (FCM)(Kow)

       A predicted baseline BAF obtained using this equation will equal one
       obtained using the Gobas model.
                   •
Method 1 gives the most preferred baseline BAF, whereas method 4 gives the least
preferred.  Baseline BAFs may be derived using other methods if justified  by good
science. All four procedures can be used with organic chemicals, but only
procedures 1 and 3 can be used with inorganic chemicals. Some measured and
predicted BCFs and BAFs are geometric means.

BAFs less than 10 are rounded to one decimal digit; BAFs between  10 and  1000
are rounded to whole numbers; BAFs greater than 1000 are rounded to four
significant digits; this does not imply anything about the precision or accuracy of
the values.  All BAFs are intermediate values in the calculation of permit limits and
so critical rounding should be performed only  at the permit limit.  For a chemical
with a low  BAF, however, the criterion is controlled by intake via ingestion of
water rather than by ingestion of tissue of aquatic life.  Thus a low  BAF does not
need many digits.

Except as noted, the values given for log Kow are derived in Appendix B using the
procedure described in  Appendix A.

The FCMs for organic chemicals are derived by linear interpolation of the  values
given in Table F.2.
                                    G-2

-------























Tt

5
o
JS
ex
2
H
£

U*

n
0)
CQ
O
9
J
§
U
^
1

3

^
•35
1
1
-^*
P^
U
£
i
•8
ts
T)
£


5,
* j
|H
Ju








2
•


S


1
ui
*
o
£
i
^

•o
o
^
g

i^
•e
e
g.

x-v

IT
e§.







| Chemical
8
»— c

8

oo
pn
ts
r-
vM


























| Benzene
fc
in
o

8
VO
8

§
§"*

VO





•

fe
O
o"
^jT
pn
•
O
«s
vcT





B

|| Chlordane
^
— H

0

i
ts
S


























1 Chlorobenzene


































4)
O
^H

S
*"*
OO
VO
1
o
Tf
ON









pn
•*

X
S5
§
o







Q
Q
Q
pn
rf

VO
8
pn
VO
|
o
ON
pn
in









§
VO
£'
8
8
8

CM






Q
Q
J
pn

3
ts
o
VO
1
0
t-










8
pn
m
8
8
S
ts
s







Q
Q
ON
••f

P

8
in
1
^H
ON








o
O
8
pn
ON
*"*











1 Dieldrin
o
— H

8
i-H
0
ts
1


























1 2,4-Dimethylphenol
8
»— 1

8

in

t-
pn


























|| 2,4-Dinitrophenol
fe
r-

m
m
oo
1
in
§
^
5
pn









o
m


8
8
ts
m
ts







1 Hexachlorobenzene
ON
m
ts

8
ts
oo
rt
|
ON














fc
ON
r>
pn








|| Hexachlorobutadiene
f


3

8
pn
-
VO














2>
r^
S








-------
J

8
    g
    OO
ylene ch
        pn
        vo
        PO
            VO
                 PO
                s
                 o\
                 CQ
                     PO
                     PO
                     in
                     oo
e
                         PO
                              ft
                             S
                             o
                              r
t
                                 (S
                                 VO
                                  8
                                      in
                                      r
                                          (N
                                              PO
                                              JQ
oxaphene
                                                  ts
                                                    g
                                                    V


                                                   f
                                                    8
                                                    o

                                                   1
                                                                            •a"

                                                                            s -S
                                                                  Q W
                                                            _2 <_,
                                                                  to
                                             «*- 2 »G
                                              o 13 H

-------
Appendix H. Recommended Baseline BAFs for Trophic Levels Three and Four

The BAFs given in the table are recommended for use in derivation of human
health criteria. BAFs recommended for use in the derivation of wildlife criteria are
given on the next page.  All BAFs given for human health and for wildlife are based
on wet weight of the tissue of the aquatic biota.

For an organic chemical, "Baseline BAFs" are based on 100% lipid and on the
concentration of freely dissolved chemical in the water.  All BAFs given for human
health are for trophic levels 3 and 4. BAFs in the table that are not baseline BAFs
are based on 1.82 percent lipid for trophic level 3 and 3.10 percent lipid for trophic
level 4.  The human health guidelines in the final Guidance currently specify that
humans consume aquatic biota that are in trophic level 3 and 4 and that the
applicable percent lipid is 1.82 and 3.10, respectively.

To calculate human health and wildlife BAFs for an organic chemical, the Kow of
the chemical shall be used with a POC concentration of 0.00000004 Kg/L and a
DOC concentration of 0.000002 Kg/L from Lake Superior (Eadie et al.) to yield the
fraction freely dissolved:
                         1
    f,
     fd
                    10

                                    1
               (0.000002kB/L)(Kow) + (0.00000004 kg/L)(Kow)


                        1
           1  + (0.00000024 kg/L)(Kow)

    The human health BAFs for an organic chemical shall be calculated using the
    following equations:
        Human Health BAFTL3= [{baseline BAF)(0.0182)+ 1](ffd)
    For trophic level 3:
    Human Health BAF

    For trophic level 4:
                  DApHH
    Human Health bAhTL4= [(baseline BAF)(0.0310)+ 1]{ffd)

where:
    0.0182 and 0.0310 are the standardized fraction lipid values for trophic
    levels 3 and 4, respectively, that are used to derive human health criteria
    and values for the final Guidance.
                                    H-1

-------
    The wildlife BAFs for an organic chemical shall be calculated using the
following equations:
         For trophic level 3:
         Wildlife
RACWL
BAFTL3 = [(baseline BAF)(0.0646)+  1](ffd)
         For trophic level 4:
                RAPWL
         Wildlife BAFTL4 =  [(baseline BAF)(0.1031)+  1](f(d)
    where:
         0.0646 and 0.1031 are the standardized fraction lipid values for trophic
         levels 3 and 4, respectively, that are used to derive wildlife criteria for the
         final Guidance.
                                    Wildlife

Water quality criteria are currently being derived for wildlife for only four chemicals
and so the BAFs are presented here.  Although it is possible that wildlife consume
some aquatic biota that are in trophic level 2, BAFs for the derivation of wildlife
criteria are given here only for trophic levels 3 and 4.  Note that the trophic level
refers to an organism, not to a species, genus, or family, because individuals of
some species are not in the same trophic level for their whole life span. For
example, many species that are in trophic level 4 as adults are in trophic level 3
when they are young.

Wildlife BAFs are given for 6.46 and 10.31 percent lipid because the  wildlife
guidelines in the final Guidance currently specify  6.46 percent lipid for trophic level
3 and 10.31 percent lipid for trophic level 4.
Chemical
DDT
Mercury
PCBs (class)
2,3,7,8-TCDD
Trophic Level 3
Baseline
34,670,000"
27,900
55,280,000'
9,360,000°
BAF6-4a,ti
1,336,000
27,900
1 ,850,000
172,100
Trophic Level 4
Baseline
60,260,000
140,000
116,600,000'
9,000,000
BAF1031%|
3,706,000
140,000
6,224,000
264,100
Fraction
Freely
Dissolved
0.597
-
0.518
0.258
Method*
1
3
1
2
                                      H-2

-------
                                Human Health
Human health BAFs are given for 1.82 and 3.10 percent lipid because the human
health guidelines in the final Guidance currently specify 1.82 percent lipid for
trophic level 3 and 3.10  percent lipid for trophic level 4.
Chemical
Benzene
Chlordane
Chlorobenzene
Cyanide
DDD
DDE
DDT
Dieldrin
2,4-Dimethylphenol
2,4-Dinitrophenol
Hexachlorobenzene
Hexachlorobutadiene
Hexachlorocyclohexane
alpha-Hexachlorocyclohexane
beta-Hexachlorocyclohexane
delta-Hexachlorocyclohexane
Hexachloroethane
Lindane
Mercury
Methylene Chloride
Mirex
Octachlorostyrene
PCBs (class)
Pentachlorobenzene
Photomirex
2,3,7,8-TCDD
1 ,2,3,4-Tetrachlorobenzene
1 ,2,3,5-Tetrachlorobenzene
Toluene
Toxaphene
Trichloroethylene
Trophic Level 3
Baseline
137
7,943 ,000"
747
1
6,839,000"
69,980,00"
34,670,000
4,180,000«l
202
37
2,630,000"
354,800-
77,620'
56.8901
77,620'
77,620'
20,370«
105,900"
21,909
18
55,590,000-
58,880,000-
55,280,000"
467,700-
45,710,000-
9,360,000°
81,280-
135,100«
527
27,510,000"
342
BAFYjj*,
3
116,600
15
1
97,680
532,800
376,400
72,610
5
2
43,690
6,352
1,412
1,035
1,411
1,412
371
1,926
27,900
1
353,400
730,000
520,900
8,248
290,600
48,490
1,467
2,439
11
498,100
7
Trophic Level 4
Baseline
137
6,166,000
740
1
10,000,000
223,900,000
60,260,000
19,300,000''
200
37
2,512,000
43,940
64,570
48,980
64,570
64,570
17,190
85,110
140,000
18
134,900,000
117,500,000
116,600,000"
645,700
117,500,000
9,000,000
117,500
101,110
516
21,580,000
339
BAF310%|
5
154,200
24
1
243,300
2,903,000
1,114,000
571,000
7
2
71,080
1,341
2,000
1,517
1,999
2,000
532
2,636
140,000
2
1,461,000
2,481,000
1,871,000
19,420
1,272,000
79,420
3,610
3,109
17
665,600
12
Fraction
Freely
Dissolved
1.000
0.806
1.000
-
0.785
0.418
0.597
0.954
1.000
1.000
0.913
0.984
0.999
0.999
0.998
0.999
0.997
0.999
-
1.000
0.349
0.681
0.518
0.970
0.349
0.285
0.991
0.991
1.000
0.995
1.000
Method*
4
1
4
-
1
1
1
2
4
4
1
1
1
1
1
1
1
1
3
4
1
1
1
1
1
2
1
1
4
1
4
                                    H-3

-------
* The methods used to calculate the recommended baseline BAFs for trophic level 4 were:
      1 =  A measured baseline BAF was based on a field-measured BAF.
      2 =  A predicted baseline BAF was based on field-measured BSAF methodology.
      3 =  A predicted baseline BAF was based on a laboratory-measured BCF and a Food-Chain
Multiplier (FCM).
      4 =  A predicted baseline BAF was based on a predicted BCF and a FCM.

b This is the geometric mean of measured baseline BAFs for sculpin and alewives (see Tables 4 and 5 ),
both of which are in trophic level 3.

c Cook, P.M.  1995.  Memorandum to C.E. Stephan. March 7.

d This is based on the concentrations  of dieldrin in sediment and fish.  However, the concentration in fish
is probably partially due to exposure of the fish to aldrin, which is converted to dieldrin.  Thus this BAF
is probably not appropriate where there is substantially more or less aldrin.

" This is a measured baseline BAF for sculpin (see Table 4), which is in trophic level 3.

f This is the geometric mean of the measured baseline BAFs  for alpha-HCCH and lindane (gamma-
HCCH).

* This baseline BAF for trophic level 3 was calculated by using the following equation:
                                                     FCM^
                                      ™ ~       ™   FCM^
 where:
             =  Baseline BAF for trophic level 3
             =  Baseline BAF for trophic level 4
            j = Food-Chain Multiplier for trophic level 3
            , = Food-Chain Multiplier for trophic level 4
 The values needed for this calculation are  given in Appendix G.

h See Appendix E.

1 See Appendix F.
                                              H-4

-------
Appendix I.  Derivation of Consumption Weighted Mean Percent Lipid for Human
           Health and Wildlife

                               TABLE 1
              LIPID CONTENT OF EDIBLE PORTION OF FISH

LAKES/SPECIES
SUPERIOR
Bloater Chub
Brown Trout
Carp
Chinook
Chinook
Chinook
Chinook
Coho
Coho
Coho
Herring
Herring
Lake Trout
Lake Trout
Lake Trout
Lake Trout
Rainbow Smelt
Rainbow Trout
Rainbow Trout
Walleye
Whitefish
Whitefish
Yellow Perch
HURON
Brown Trout
Carp
Channel Catfish
Chinook
Coho
Lake Trout
Walleye
ERIE
Carp
Chinook
Channel Catfish
Coho
Lake Trout
Smallmouth Bass
Walleye
Walleye
White Bass
Whitefish
PERCENT LIPID
Xg
















11.34




7.85



7.54
11.37
10.69
1.72
3.96
14.12
1.62

3.44

7.11



2.56



Xa

10.27
6.40
7.84
3.35
2.95
2.96
2.68
7.50
1.39
1.56
9.20
4.58
11.42
10.46
9.21

0.90
2.13
1.24
1.91

7.15
0.92










3.88

4.50
13.00
1.99

1.98
4.42
8.75

N

3
11
9
10
4
5
14
3
8
5
1
6
44
71
28
71
3
3
8
33
10
2
8

20
9
1
44
8
80
10

8
21
10
22
5
19
40
9
8
4

PORTION

F
F
F
Fs
F
F
F
F
F
F
F
D
F
F
F
F
D
F
F
F
F
F
F

F
Fs
Fs
F
F
F
F

Fs
F
Fs
F
F
F
F
Fs
Fs
Fs

SOURCE

WDNR
WDNR
WDNR
MDNR
WDNR
MPCA
MPCA
WDNR
MPCA
MPCA
WDNR
MPCA
WDNR
MPCA
MPCA
MDNR
MPCA
WDNR
MPCA
WDNR
MDNR
MPCA
WDNR

MDNR
MDNR
MDNR
MDNR
MDNR
MDNR
MDNR

MDNR
NYDEC
MDNR
NYDEC
NYDEC
NYDEC
MDNR
OEPA
OEPA
OEPA
                                  1-1

-------

LAKES/SPECIES
ONTARIO
Brown Trout
Channel Catfish Chinook
Coho
Lake Trout
Rainbow Trout
Smallmouth Bass
White Perch

STATEWIDE (Wisconsin)
Bass (largemouth)
Bluegill
Bowfm
Buffalo (bigmouth)
Burbot
Cisco
Crappie
Muskie
Redhorse Suckers
Rockbass
MICHIGAN (Green Bay)
Black Bullhead
Brook Trout
Brown Trout
Carp
Channel Catfish
Chinook
Coho
Lake Trout
Rainbow Trout
Smallmouth Bass
Walleye
White Bass
Yellow Perch
PERCENT LIPID
Xg Xa



































10.40
12.80
2.75
3.38
14.53
9.04
1.85
5.64

0.70
0.83
0.40
8.66
0.86
6.09
0.92
1.53
1.86
0.44

1.10
4.97
9.44
8.17
4.75
4.63
7.70
11.88
6.39
1.34
2.71
3.76
0.76

N

91
47
45
98
120
57
161
33

107
74
1
115
39
14
135
11
72
85

8
9
106
48
15
46
1
28
45
10
67
18
26

PORTION

F
Fs
F
F
F
F
F
F

F
F
F
F
F
F
F
F
F
F

Fs
F
F
F
Fs
F
F
F
F
F
F
F
F

SOURCE

NYDEC
NYDEC
NYDEC
NYDEC
NYDEC
NYDEC
NYDEC
NYDEC

WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR

WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
WDNR
1-2

-------
LAKES/SPECIES
MICHIGAN
Black Bullhead
Bloater Chub
Brook Trout
Brown Trout
Brown Trout
Brown Trout
Brown Trout
Brown Trout
Brown Trout
Carp
Carp
Carp
Channel Catfish
Chinook
Chinook
Chinook
Chinook
Chinook
Chinook
Chinook-Trim
PERCENT LIPID
Xg Xa





5.68





6.82







1.79
0.99

1.80
14.75
4.33
11.96

11.19
11.22
3.88
6.70
20.43

10.68
8.92
4.20
4.92
2.60
1.45
2.46


N

1
92
68
170
46
21
6
5
9
2
16
47
11
275
30
4
5
28
71
10
PORTION

Fs
F
F
F
F
A
D
Fs
F
F
Fs
F
Fs
F
A
D
Fs
F
F
O
SOURCE

WDNR
WDNR
WDNR
WDNR
MDNR
IDEM
IDEM
IDEM
IDEM
IDEM
MDNR
WDNR
WDNR
WDNR
IDEM
IDEM
IDEM
IDEM
MDNR
MDNR
1-3

-------
LAKES/SPECIES
MICHIGAN (con't)
Coho
Coho
Coho
Coho
Coho
Coho
Lake Trout
Lake Trout
Lake Trout
Lake Trout
Lake Trout
Lake Trout
Lake Trout-trim
Longnose Sucker
Longnose Sucker
Longnose Sucker
Northern Pike
Northern Pike
Rainbow Trout
Steelhead
Steelhead
Steelhead
Steelhead
Walleye
Walleye
Walleye
Whitefish
White Sucker
White Sucker
Yellow Perch
Yellow Perch
Yellow Perch
Yellow Perch
Yellow Perch
PERCENT LIPID
Xg Xa





2.42





16.67

9.19


5.59

0.57
3.76




1.63




1.61



0.82


5.96
6.51
1.95
2.80

3.82
17.25
16.58
8.81
12.01

12.71

5.45
4.95

3.00


11.09
7.10
2.77
5.62

1.45
2.19
9.00
2.45

3.00
1.55
1.06

0.95
N

19
8
2
18
36
164
156
13
3
9
60
311
10
2
3
10
2
10
25
17
3
2
6
11
9
9
1
2
10
1
6
9
10
24
PORTION

A
D
Fs
F
F
F
A
D
Fs
F
F
F
O
A
F
F
A
Fs
F
A
D
Fs
F
F
Fs
F
A
A
F
A
D
F
F
F
SOURCE

IDEM
IDEM
IDEM
IDEM
MDNR
WDNR
IDEM
IDEM
IDEM
IDEM
MDNR
WDNR
MDNR
IDEM
IDEM
MDNR
IDEM
MDNR
MDNR
IDEM
IDEM
IDEM
IDEM
MDNR
MDNR
WDNR
IDEM
IDEM
MDNR
IDEM
IDEM
IDEM
MDNR
WDNR
Key to Abbreviations

Percent Lipid:
Xg = geometric mean, contributing program (source) used geometric means to summarize data.
Xa = arithmetic mean, contributing program (source) used arithmetic means to summarize data.

N = Number of fish sampled

Portion:
F = filet, skin on
Fs = filet, skin off
A  = Anterior section through fish
D  = dressed (gutted, head removed)
0 = filet, skin off, visible fat removed (trimmed)
                                             1-4

-------
Source:

MDNR = Michigan Department of Natural Resources. Fish Contaminant Monitoring
Program, Data for Lakes Erie, Huron, Michigan and Superior 1986-1989.

MFC A = Minnesota Pollution Contol Agency.  Minnesota Fish Consumption Advisory
Program, Data for Lake Superior.

IDEM = Indiana Department of Environmental Management, OWM-Biological  Studies,
Data for Lake Michigan.

OEPA = Ohio Environmental Protection Agency. Ohio Dept. of Natural Resources, Data for
Lake Erie.

WDNR = Wisconsin Department of Natural Resources. Data for Lakes Michigan and
Superior and State of Wisconsin.

NYDEC = New York Department of Environmental Conservation. Data for Lakes Erie and
Ontario.
                                       1-5

-------
          APPENDIX I: TABLE 2
LIPID CONTENT OF EDIBLE PORTION OF FISH
LAKES/SPECIES
LAKE SUPERIOR
Bloater chub
Brown trout
Carp
Chinook
Coho
Herring
Lake trout
Rainbow smelt
Rainbow trout
Walleye
Whitefish
Yellow perch
LAKE HURON
Brown trout
Carp
Chinook
Channel catfish
Coho
Lake trout
Walleye
LAKES ST. CLAIR AND ERIE
Carp
Channel catfish
Chinook
Coho
Lake trout
Smallmouth bass
Walleye
White bass
Whitefish
N*

1
1
1
4
3
2
4
1
2
1
2
1

1
1
1
1
1
1
1

1
1
1
1
1
1
2
1
1
MEAN PERCENT LIPID

10.27
6.40
7.84
2.99
3.48
6.89
10.61
0.90
1.69
1.91
7.50
0.92

7.54
11.37
7.72
10.69
3.96
14.12
1.62

3.44
7.11
3.88
4.50
13.00
1.99
2.27
4.42
8.75
                  1-6

-------
LAKES/SPECIES
LAKE MICHIGAN (inc. Green Bay)
Black bullhead
Bloater chub
Brook trout
Brown trout
Carp
Channel catfish
Chinook
Coho
Lake trout
Longnose sucker
Northern pike
Rainbow trout (steelhead)
Smallmouth bass
Walleye
White sucker
White bass
Whitefish
Yellow perch
LAKE ONTARIO
Brown trout
Channel catfish
Chinook
Coho
Lake trout
Rainbow trout
Smallmouth bass
White perch
WISCONSIN (statewide)
Bass (largemouth)
Bluegill
Bowfin
Buffalo (bigmouth)
Burbot
Cisco
Crappie
Muskie
Redborse suckers
Rockbass
N*

2
1
2
7
4
2
7
7
7
3
2
6
1
4
2
1
1
6

1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
MEAN PERCENT LIPID

1.45
14.75
4.65
8.58
11.53
6.84
3.15
4.45
13.70
5.33
1.79
6.12
1.34
2.00
2.03
3.76
9.00
1.36

10.40
12.80
2.75
3.38
14.53
9.04
1.85
5.64

0.70
0.83
0.40
8.66
0.86
6.09
0.92
1.53
1.86
0.44
* Number of state programs reporting data for a species.
                                          1-7

-------
          APPENDIX I: TABLES
LJPID CONTENT OF EDIBLE PORTION OF FISH
Species
Black bullhead
Bloater chub
Bluegill
Bowfin
Brook trout
Brown tout
Buffalo
Burbot
Carp
Channel catfish
Chinook
Cisco
Coho
Crappie
Herring
Lake Trout
Largemouth bas:>
Longnose sucker
Musky
Northern pike
Rainbow smelt
Rainbow trout
Redhorse sucker
Rockbass
Smallmouth bass
Walleye
White perch
White bass
White sucker
Whitefish
Yellow perch
Mean Percent Lipid
1.45
12.51
0.83
0.40
4.65
8.23
8.66
0.86
8.55
9.36
2.90
6.09
3.95
0.92
6.89
13.19
0.70
5.33
1.53
1.79
0.90
5.62
1.86
0.44
1.73
1.95
5.64
4.09
2.03
8.42
1.14
                  1-8

-------
    APPENDIX I: TABLE 4
LIPID CONTENT OF WHOLE FISH
Species
Alewife
Bloater
Bluegill
Bluntnose minnow
Brown bullhead
Brown trout
Channel catfish
Coho salmon
Common carp
Emerald shiner
Freshwater drum
Lake trout
Lake whitefish
Lake herring
Northern pike
Pink salmon
Rainbow smelt
Rainbow trout
Redhorse
Rock bass
Skipjack herring
Slimy sculpin
Smallmouth bass
Spake
Spottail shiner
Sunfish
Walleye
White bass
White perch
White sucker
Yellow perch
LAKE*
Sup.

13.1







.

16.6
10.5


•




9.8
•









Mich.

22.3
1.55##








17.0










1.32##


1.73##



6.8
7.4
Hur.






18.7

10.5


20.5
10.0
















6.0
4.1
st.c.








9.5

















8.1
9.6



Erie








11.6
1.6#
8.4







6.4





2.0#

11.4
9.8

4.9
4.2
Ont.



1.5*
6.1
12.2
11.7

5.8
2.7#

15.3#

6.0





4.8




1.8#



10.2#

5.6
CDF&O**
9.73



3.58
15.44

8.45
8.59


17.25


2.17
1.78
4.78
7.59



6.95

10.12


8.01
10.16

5.15
5.95
MEAN
9.73
17.70
1.55
1.50
4.84
13.82
15.20
8.45
9.20
2.15
8.40
17.33
10.25
6.00
2.17
1.78
4.78
7.59
6.40
4.80
9.80
6.95
1.32
10.12
1.90
1.73
9.17
9.85
10.20
5.71
5.50
            1-9

-------
Footnotes

* Data for the individual lakes from U.S. Fish and Wildlife Service National Contaminant
Biomonitoring Program 1976-1984.

** CDF&O = Canada Department of Fisheries and Oceans.  Percent lipid data for unspecified Great
Lakes. These data are averaged together with the lake-specific data from the U.S. Fish and Wildlife
Service.

# Value includes data from the New York State Department of Environmental Conservation.

## Values are from Michigan Department of Natural Resources

Data Sources:

Canada Department of Fisheries and Oceans, Great Lakes Contaminant Surveillance Program, 1977-
     1985.

New York Department of Environmental Conservation

Schmitt, C.J., J.L. Zajicek and P.H. Peterman.  1990.  National contaminant biomonitoring program:
     residues of organochlorine chemicals in U.S. freshwater fish, 1976-1984.  Arch. Environ.
     Contain. Toxicol. 19: 748-781.

Michigan Department of Natural Resources
                                             1-10

-------
                   APPENDIX I: TABLES
AVERAGE DAILY PER CAPITA ESTIMATES OF FISH CONSUMPTION BY
   SPECIES FROM THE 1991-1992 MICHIGAN SPORT ANGLERS FISH
                   CONSUMPTION STUDY1
SPECIES
Perch (Yellow)
Walleye
Bluegill
Pike (Northern)
Salmon
Bass (Largemouth)
Other3
Trout (Lake)
Trout (Rainbow)
Smelt
Crappie
Trout (Brown)
Trout (Brook)
Catfish (Channel)
Salmon (Coho)
Whitefish
Salmon (Chinook/King)
Sucker (White)
Bass (Small mouth)
Bullhead
Perch/Bluegill
Rockbass
Whitebass
Sunfish
Bass/Bluegill
Burbot
Carp
CONSUMPTION RATE
(grains/person/day)
Mean
3.03
2.59
2.20
0.97
0.95
0.73
0.75
0.70
0.69
0.50
0.49
0.48
0.31
0.29
0.29
0.21
0.20
0.17
0.16
0.13
0.11
0.11
0.07
0.05
0.04
0.03
0.03
Bias-adjusted Mean2
2.63
2.25
1.91
0.84
0.82
0.63
0.65
0.61
0.60
0.43
0.43
0.42
0.27
0.25
0.25
0.18
0.17
0.15
0.14
0.11
0.10
0.10
0.06
0.04
0.03
0.03
0.03
                           Ml

-------
                        APPENDIX I:  TABLE 5 (continued)
    AVERAGE DAILY PER CAPITA ESTIMATES OF FISH CONSUMPTION BY
       SPECIES FROM THE 1991-1992 MICHIGAN SPORT ANGLERS FISH
                             CONSUMPTION STUDY1
SPECIES
Muskie
Buffalo (Bigmouth)
Sucker (Longnose)
Cisco
Bowfm
Redhorse
Walley/Perch
Pike/Perch
CONSUMPTION RATE
(grains/person/day)
Mean
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
Bias-adjusted Mean2
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
1 Source:  Fish Consumption Estimates Based on the 1991-1992 Michigan Sport Anglers Fish
Consumption Survey.  February 21, 1995. USEPA.  Submitted by SAIC to EPA under Contract No.
68-C4-0046.

2 The bias-adjusted mean consumption rate is calculated by multiplying of the actual consumption rate
times the nonresponse bias correction of 0.86834 (1.0 - 0.13174) from West et al. 1991-1992
Michigan Sport Anglers Fish Consumption Study - Final Report to the Michigan Great Lakes
Protection Fund, Michigan Department of Natural Resources.  University of Michigan, School of
Natural Resources, Natural Resource Sociology Research Lab. Technical Report #6. May 1993.

3 Other includes "other single species", "other combinations", and "species not recorded".
                                        1-12

-------
          APPENDIX I: TABLE 6
CALCULATION OF A CONSUMPTION WEIGHTED
     MEAN PERCENT LIPID VALUE FOR
TROPHIC LEVEL 3 FISH CONSUMED BY HUMANS
Species
Bluegill
Crappie
Trout (Brook)
Whitefish
Other
Sucker (White)
Bullhead
Perch/Bluegills
Sunfish
Carp
Buffalo (Bigmouth)
Sucker (Longnose)
Redhorse
Cisco
TOTAL
Bias-Adjusted1
Consumption
(g/day)
1.91
0.43
0.27
0.18
0.162
0.15
0.11
0.10
0.04
0.03
0.02
0.02
0.01
0.01
3.44
Lipid
(%)3
0.83
0.92
4.65
8.42
1.824
2.03
1.45
1.01s
0.83s
8.55
8.66
5.33
1.86
6.09

Size
(cm)'
5-27
13-42
10-40
3^0
—
5-60
> 10
—
—
10-23
25-46
35-60
> 6.5
20-30

Trophic
Level*
2.6 - 3.0
3.0 - 3.4
3.2
3.0 - 3.4
—
2.7 - 2.9
2.7 - 3.2
< 3.57
2.8-3.3
2.2-3.1
2.6 - 3.0
2.4 - 3.0
2.9
3.0-3.1

Assigned
Trophic
Level8
3
3
3
3
3
3
3
3
3
3
3
3
3
3

Product9
1.59
0.40
1.26
1.52
0.29
0.30
0.16
0.10
0.03
0.26
0.17
0.11
0.02
0.06
6.27
                  1-13

-------
      APPENDIX I: TABLE 6 (continued)
CALCULATION OF A CONSUMPTION WEIGHTED
     MEAN PERCENT LIPID VALUE FOR
TROPHIC LEVEL 4 FISH CONSUMED BY HUMANS
Species
Perch (Yellow)
Walleye
Pike (Northern)
Salmon
Bass (Largemouth)
Trout (Lake)
Trout (Rainbow)
Other
Smelt
Trout (Brown)
Catfish (Channel)
Salmon (Coho)
Salmon (Chinook)
Bass (Smallmouth)
Rockbass
Whitebass
Bass/Bluegills
Burbot
Muskie
Pike/Perch
Walleye/Perch
Bowfin
TOTAL
Bias-Adjusted1
Consumption
(g/day)
2.63
2.25
0.84
0.82
0.63
0.61
0.60
0.492
0.43
0.42
0.25
0.25
0.17
0.14
0.10
0.06
0.03
0.03
0.02
0.01
0.01
0.01
10.80
Lipid
(%)3
1.14
1.95
1.79
3.5310
0.70
13.19
5.62
3.104
0.90
8.23
9.36
3.95
2.90
1.73
0.44
4.09
0.855
0.86
1.53
1.305
1.51s
0.40

Size
(cm)'
20-30
30-80
> 10
—
> 20
> 40
> 50
—
—
—
> 45
45-60
—
> 10
> 7.5
> 20
—
> 50
—
—
—
—

Trophic
Level*
3.1-3.8
3.9 - 4.5
4.0
4.0
3.8
4.0 - 4.5
4.0
> 3.5
3.1-3.5
> 3.5"
3.5 - 3.9
4.0 - 4.5
> 3.512
3.4 - 3.9
3.3 - 3.7
3.9
> 3.5"
4.0
> 3.514
> 3.515
> 3.516
4.0

Assigned
Trophic
Level8
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4

Product*
3.00
4.39
1.50
2.90
0.44
8.05
3.37
1.52
0.39
3.46
2.34
0.99
0.49
0.24
0.04
0.25
0.03
0.03
0.03
0.01
0.02
0.01
33.50
                  1-14

-------
Consumption weighted mean percent lipid value for Trophic Level 3 (6.27/3.44) = 1.82
Consumption weighted mean percent lipid for Trophic Level 4 (33.50/10.80) = 3.10

Total grams of lipid consumed per day from Trophic Level 3 (6.27/100)= 0.0627
Total grams of lipid consumed per day from Trophic Level 4 (33.50/100) = 0.3349

24.16% of fish consumed are Trophic level 3
75.84% offish consumed are Trophic level 4

1 The bias-adjusted consumption rate comes from Table 5 of Appendix I.

2 Consumption rate calculated by multiplying bias-adjusted consumption rate "other" category (0.65
g/day) in Table 5 of Appendix I by percent of fish consumed in trophic level 3 (24.16%) or trophic
level 4 (75.84%).

3 Percent lipid values are taken from Table 3 of Appendix I unless otherwise noted.

4 Percent lipid is the overall consumption weighted mean lipid value for trophic level 3 or trophic
level 4.

5 Percent lipid is weighted average of Perch/Bluegill, Smallmouth Bass/Largemouth Bass/Bluegill,
Pike/Perch, or Walleye/Perch.  Lipid value for sunfish assumed to be the same as for bluegill.

6 Size values and Trophic levels taken from: USEPA. 1995. Trophic Level and Exposure Analyses
for Selected Piscivorous Birds and Mammals. Volume I: Analyses for  Species on the Great Lakes
Basin and of the Great Lakes Basin and Volume III: Appendices.

7 Trophic level assumed to be less than 3.5 based on bluegill data.

8 Species were placed in trophic level 4 if the highest value hi the reported range was greater than or
equal to 3.5.  Species were placed hi a trophic level 3 when the highest value in the reported range
was less than 3.5.

9 Product is equal to the bias-adjusted consumption rate for  that species multiplied by the percent lipid
for that species times 100.

10 Percent lipid is  weighted average of Coho and Chinook Salmon.

11 Trophic level assumed to be greater than 3.5 based on other trout data.

12 Trophic level assumed to be greater than 3.5 based on other salmon data.

13 Trophic level assumed to be greater than 3.5 based on bass data.

14 Trophic level assumed to be greater than 3.5 based on knowledge of Muskie  feeding habits.

15 Trophic level assumed to be greater than 3.5 based on Pike/Perch data.

16 Trophic level assumed to be greater than 3.5 based on Walleye/Perch data.
                                             1-15

-------
             APPENDIX I: TABLE 7
    CALCULATION OF A PERCENT LIPID VALUE
FOR TROPHIC LEVEL 3 FISH CONSUMED BY WILDLIFE
Species1
Longnose/White Sucker
Whitefish
Alewife
Common Carp
Lake Herring
Yellow Perch
Smallmouth bass
Bluegill
Redhorse suckers
Trout (brown)
Trout (rainbow)
Sculpin
Sunfish
Rainbow smelt
AVERAGE
Lipid (%)2
5.715
10.25
9.73
9.20
6.00
5.45
1.32
1.55
6.40
13.82
7.59
6.95
1.73
4.78
6.46
Size (on)3
35-60
3-40
5-23
> 23
20-30
< 7
< 7
< 7
< 7
8-18
7-23
< 8
5-10
2-17

Mean
Trophic Level3
2.8
3.2
3.2
2.4
3.1
3.0
3.4
2.8
2.7
3.2
3.2
3.0
3.1
3.1

Assigned
Trophic Level4
3
3
3
3
3
3
3
3
3
3
3
3
3
3

                    1-16

-------
                          APPENDIX I:  TABLE 7 (continued)
                   CALCULATION OF A PERCENT LIPID VALUE
             FOR TROPHIC LEVEL 4 FISH CONSUMED BY WILDLIFE
SPECIES1
Lake trout
Walleye
Bloater chub
Pike (Northern)
Trout (Average of brown and
rainbow trout)
Rock bass
AVERAGE
LIPID (%)2
17.33
9.17
17.70
2.17
10.71
4.80
10.31
SIZE
(cm)3
20-40
15-30
20-30
25
7-23
10-22

MEAN
TROPHIC
LEVEL3
3.8
3.5
3.5
4.0
3.5
3.5

ASSIGNED
TROPHIC
LEVEL4
4
4
4
4
4
4

1 The species selected are those consumed by the 5 representative species used to derive wildlife
criteria and those with available percent lipid data.  Other species consumed by the 5 representative
species but not included in the tables because of lack of lipid data include: trophic level 3 - burbot,
pumpkinseed, blackstripe topminnow, darters, brook silverside, bullhead, blacknose dace, creek chub,
mudminnow, stickleback,'and brook trout.  Source of data: USEPA. 1995. Trophic Level and
Exposure Analyses for Selected Piscivorous Birds, and Mammals. Volume I: Analyses for Species on
the Great Lakes Basin and of the Great Lakes Basin and Volume III: Appendices.

2 Percent lipid taken from Table 4 in Appendix I unless otherwise noted.

3 Size values and trophic levels taken from source cited in footnote 1.

4 Species were placed in a trophic level 3 if the mean value was  less than 3.5. Species  were placed in
trophic level 4 if the mean value was greater than or equal to 3.5.

5 Percent lipid data for white suckers were assumed to be similar for both longnose and white sucker.
                                          1-17

-------
Appendix J.   FORTRAN Source Code for the Model of Gobas (1993)

This source code includes the feeding preferences, lipid content, and weight of the
organisms; temperature-; and sediment organic carbon content used in the final Guidance for
deriving the FCMs.  This code does not include the correction for bioavailability discussed in
the journal article by Gobas.

     real lipid(6),weights(6)
     real residues(6),pref(5,4),Kow
     common Kow,  VF, VL, cf, residues,  pref, c_w, t, ink

c data
c for lipids, residues, and weights
c zoo, dip, scu, ale, sme, pf
     data lipid/0.05,0.03,0.08,0.07,0.04,0.11/
     data weights/0,0,0.0054,0.032,0.016,2.41/
     data residues/0,0,0,0,0,0/

c for pref columns: scu, ale, sme, pf
c for pref row: zoo, dip, scu, ale, sme
     data pref/0.18,0.82,0,0,0,0.60,0.40,0,0,0,
     x    0.54,0.21,0.25,0,0,0,0,0.10,0.50,0.40/

     density_oc=0.9
     density_dip=0.9
c  temperature and sediment organic carbon
     t=8
     soc=0.027

     write(6,*) ' Input log Kow'
     read(5,*) Kow
     Kow =  10**Kow
     c_w  =  1
     c_sed = 25 * Kow * c_w * soc

c  zooplankton
     residues(l)  = lipid(l)*Kow*c_w

c  diporeia
     residues(2)  = c_sed*density_oc/soc*lipid(2)/density_dip

c  sculpin
     VF=weights(3)
     VL=lipid(3)
     ink=l
     call fish
     residues(3)=cf

                                         J-l

-------
c alewives
    VF=weights(4)
    VL=lipid(4)
    ink=2
    call fish
    residues(4)=cf   .

c smelt
    VF=weights(5)
    VL=lipid(5)
    ink=3
    call fish
    residues(5)=cf

c piscivorous fish
    VF=weights(6)
    VL=lipid(6)
    ink=4
    call fish
    residues(6)=cf

    write(6,1215)
1215     format(t26, 'Log BAF ' ,146, 'FCMV
    x  t22,'(h'pid normalized',/,t22,'& freely dissolved)')
    write(6,1220) ((loglO(residues(i)/c_w/lipid(i)),
    x    residues(i)/c_w/lipid(i)/Kow),i= 1,6)
1220     format(t5,'Zooplankton',t22,n0.3,t40,n0.3/
    x    t5,'Diporeia',t22,fl0.3,t40,fl0.3/
    X    t5,'Sculpin',t22,fl0.3,t40,f!0.3/
    x    t5,'Alewives',t22,fl0.3,t40,f!0.3/
    x    t5,'Smelt',t22,fl0.3,t40,fl0.3/
    x    t5,'Piscivorous fish',t22,f!0.3,t40,fl0.3)

    stop
    end

    subroutine fish
    real residues(6),pref(5,4),Kow
    real kl, k2, km, kd, kg, ke
    common Kow, VF, VL, cf, residues, pref, c_w, t, ink

    QW = 88.3*VF**0.6
    QL = QW/100.0
    kl = 1/(VF/QW +  VF/QL/Kow)
    k2 = kl/(VL*Kow)
    ED = l/(5.3e-8*Kow  + 2.3)
    FD = 0.022*VF**0.85*FJCP(0.06*t)
                                        J-2

-------
     kd = ED*FD/VF-
     km =0

c Note the following errors in the manuscript.
c ke is not 0.25*kd
     ke = 0.20*kd
c temperature equations are different
     if(t.lt. 17.5) then
     kg = 0.002*vf**-Q.2
     else
     kg = 0.01*vf**-0.2
     endif

     cf=0
     do 10 i=l,5
     cf=cf+pref(i,ink).*residues(i)
10   continue
     cf = (kl*c_w + kd*cf)/(k2 + ke + km + kg)

     return
     end
                                          J-3

-------
Appendix K. Determination of BAFs for DDT and Metabolites and
             Biomagnification Factors for the Derivation of Wildlife Criteria

I.   DETERMINATION OF A BAF FOR TOTAL DDT AND METABOLITES

In order to calculate an avion class-specific wildlife value for DDT, a BAF for a mixture of
DDT, DDE and DDD representative of the Great Lakes had to be determined. This was
necessary because the study from which the test dose was derived  (Anderson et al., 1975)
was based on exposure-to pelicans from anchovies containing DDE, DDD and DDT.

A BAF for the total DDT mixture (DDTr) was calculated from the BAFs for DDE, DDE,
and DDT derived for the Lake Ontario ecosystem by Oliver and Niimi (1988). There was
no statistically significant difference between the distribution of these compounds for the total
DDT mixture between the Lake Ontario ecosystem and the California coastal ecosystem, the
location of the field study by Anderson et al.  (1975). (In the report by Anderson et al.
(1975), the average composition  of the total DDT mixture in anchovies was 69.4% (8.3%
standard deviation, n = 6, range 60.0 to 80.0%) for DDE and 30.6% for the sum of DDT
and DDD. The distribution of the total DDT mixture in the Great Lakes for forage fish
(i.e., sculpin, alewife, and smelt) taken from  the report of Oliver and Niimi (1988) was
77.5% (6.4%, n = 4, range 71.4 to 84.9%) for DDE and 22.5% for the sum of the DDT
and DDD.

Below is the analysis and calculations carried out to  determine the appropriate BAF for the
total DDT mixture in wildlife prey species in trophic levels three and four. The molecular
weights for DDE, DDD, and DDT were used in these calculations and are 318.0, 320.1, and
354.5 g/mole for each compound, respectively.  This analysis is consistent with Appendix B
of 40 CFR Part  132, Great Lakss Water Quality Initiative Methodology for Deriving
Bioaccumulation Factors. Chemical-specific, fish species-specific data obtained from or
derived from the work of Oliver and Niimi (1988) are presented in Table K.I and Table K.2
below.

Table K.I.  Measured log BAF" and measured residues for DDE, DDD, and DDT in fish
derived from a Lake Ontario ecosystem by Oliver and Niimi (1988).

Sculpin
Alewife
Large1 Smelt
Small2 Smelt
Pise3 fish
Measured
Log BAFj"
DDE
7.83
7.86
8.26
8.11
8.37
DDD
6.89
6.78
6.84
6.80
7.00
DDT
7.47
7.61
7.93
7.43
7.78
Measured Residues
(ng/g)
DDE
190
180
260
180
860
DDD
47
32
21
19
83
DDT
29
35
41
13
80
1 large fish,2 small fishj 3 piscivorous fish
                                       K-l

-------
Table K.2.  Measured residues and the average composition of the DDE, DDD, and DDT
derived from Oliver and Niimi (1988).

Sculpin
Alewife
Large1
Smelt
Small2
Smelt
Pise3 Fish
Measured Residues
(moles/g fish)
DDE
0.597
0.566
0.818
0.566
2.704
DDD
0.147
0.100
0.066
0.059
0.259
DDT
0.082
0.099
0.116
0.037
0.226
Sum
0.826
0.765
0.999
0.662
3.189
Average Composition
of Congeners
(% mole basis)
DDE
72.32
74.01
81.85
85.49
84.79
DDD
17.78
13.08
6.57
8.97
8.13
DDT
9.90
12.91
11.58
5.54
7.08
1 large fish, 2 small fish, 3piscivorous fish

To be consistent with the Gobas model (1993) which was used to derive Food Chain
Multipliers for organic chemicals (as described in this parent document), the prey of trophic
level 3 fish are considered to be sculpin and alewife and the prey of trophic level 4 fish are
considered to be piscivorous fish.  Using the data in Tables K.I and K.2 above, and taking
the geometric means to determine the average values for forage fish (i.e.,  sculpin and
alewife) for DDE, DDD, and DDT, the average percent compositions are 73.2%, 15.4%,
and 11.4%, respectively and the log BAF" values are 7.84, 6.83, and 7.54, respectively.

    The composite log BAFjd for each trophic level can then be determined as presented
below:

     The composite BAF"3 (DDT mixture; trophic level 3) =

    BAF%        =   (.732)(10**7.84) + (.154)(10**6.83) +  (.114)(10**7.54)

                  =   50,642,027 +  1,041,167 + 3,952,800 = 55,635,994

    log BAF&    =  • 7.75
                                        K-2

-------
     The composite BAF"4 (DDT mixture; trophic level 4) =

     BAFJ?4       = '  (.848)(10**8.37) + (.081)(10**7.00) + (.071)(10**7.78)

                  =   198,327,759 + 813,197  + 4,263,920 = 203,404,876

     log BAF£4    =   8.31

The next step is to calculate the BAF based on the total DDT mixture using the appropriate
percent lipids of Great Lakes fish for wildlife species.  The lipid values for wildlife are
6.46% and 10.31% for trophic levels 3 and 4, respectively.  The log KQWS for DDE, ODD,
and DDT are 6.76, 6.06, and 6.45, respectively.

The K
-------
The fraction of the chemical which is freely dissolved for trophic level 4 is:

                         2.0e-6*10**6.72/10 + 0.04e-6*10**6.71)
              =   0.4462

The BAF"^^  for the total DDT mixture for trophic level 3 (based on freely dissolved and
6.46% lipid) is:

                    =  6.46%  * 55635994 =  3594085

                   = 6-56

Adjusting this value for the total chemical in the water results in the following BAF^*^ for
trophic level 3:

                  = • fd  * BAF
                     •  fd(3      6 4
                  =   0.4695 *  3,594,084
                  =   1,687,000

The BAFjQ31| 4 for the total DDT mixture for trophic level 4 (based on freely dissolved and
10.31% lipid) is:

                  =  10.31%  * 203,404,876 = 20,971,043
                  ='7.32

Adjusting this value for the total chemical in the water results in the following BAF|031%/ 4
for trophic level 4:

    BAF{0-3i%M    =   ffd>4 * BAFia31M
                  =   0.4462*  20,971,043
                  = .  9,357,000

Therefore, the final BAFs used to determine the avian wildlife values are 1,687,000 for
trophic level 3 and 9, 357,000 for trophic level 4.
H.  DETERMINATION OF BIOMAGNIFICATTON FACTORS FOR THE
    DERIVATION OF WILDLIFE VALUES FOR THE BALD EAGLE

In the derivation of wildlife criteria for the Great Lakes Water Quality Initiative, five
species were selected as representative of avian and mammalian species resident in the Great
Lakes basin likely to experience the highest exposures to bioaccumulative contaminants
through the aquatic food web.  One of these representative species is the bald eagle.
Estimates of prey species for the bald eagle indicate that approximately eight  percent of a
bald eagle's diet (on a wet weight basis) consists of piscivorous birds (i.e., gulls; EPA,
1995a,b).  A Biomagnification Factor (BMP) is needed to quantify the contribution of
contaminant to the eagle's diet from ingestion of gulls.  The BMFs used for the derivation of

                                        K-4

-------
wildlife criteria for the four chemicals for which wildlife criteria exist in the Great Lakes
Water Quality Initiative are presented in Table K.3. These were derived from the work of
Braune and Norstrom (1989), unless otherwise indicated, who measured the concentrations of
various contaminants in both gulls from Lake Ontario and in the prey fish of the gulls.  The
BMFs presented in Table K. 3 are the ratios of the concentration of a contaminant in the gulls
to the concentration in their prey fish.

The BMP for the total DDT mixture (DDTr) is calculated by using the average percentages
of the various DDT congeners in trophic  level 3 fish presented in the section above.

    BMP (DDTr)  = '  (0.732)(85) + (0.154)(3.2) + (0.114)(3.2)
                   =   62.2 + .49 + .36
                   =   63

Table K.3. Biomagnification factors used to derive wildlife values for the bald eagle.
Chemical
DDE
DDD
DDT
DDTr
Mercury
2,3,7,8-TCDD
PCBs
Biomagnification Factor1
85
3.22
3.2
63
103
30
90
1 All values derived from Braune and Norstrom (1989) unless otherwise indicated.

2 Not reported by Braune and Norstrom and assumed to be similar to that for DDT.

3 Derived by analysis of data in Noreheim and Forslie (1978), Wren et al. (1983), and
Vermeer et al. (1973) and the application of best professional judgment.


References

Braune, B. M. and R. J. Norstrom.  1989. Dynamics of organochlorine compounds in herring
     gulls: HI. Tissue distribution and bioaccumulation in Lake Ontario gulls. Environ.
     Toxicol. Chem. 8:957-968.

Oliver, B.G., and AJ. Niimi. 1988. Trophodynamic analysis of polychlorinated biphenyl
     congeners and other chlorinated hydrocarbons in the Lake Ontario ecosystem.  Environ.
     Sci. Technol. 22:388-397.
                                         K-5

-------
Noreheim, G. and A. Forslie.  1978.  The degree Of methylation and organic distribution in
     some birds of prey.  Acad. Pharmacol. Toxicol.  43:196-204,

Vermeer, K. F.A. J. Armstrong, and D.R.M. Hatch.  1973.  Mercury in aquatic birds at
     Clay Lake, Western Ontario.  J. Wildl. Manage. 37:58-61.

Wren, C.D. H.R. MacCrimmon, and B.R. Loescher.  1983.  Examination of
     bioaccumulation and  biomagnification of metals in a precambrian shield lake. Water,
     Air, and Soil Pollut.  19:277-291.
                                                             U.S. Environmental Protection Agency
                                                             Region 5, Library (PL-12J)
                                        K-6                  H. West Jackson Boulevard,  12th Floor
                                                             Chicago,  IL  60604-3590

-------