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
« ?
) 2
Oco
U.
CD
•
o
A
(0
3
(0
(0
0)
o
3
O)
il
c
_o
s
0)
o
o
-------
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
------- |