United States
Environmental Protection
Agency
Office of Water (4304)
Office of Science and Technology
Washington, DC 20460
EPA-822-R-94-002
July 1994
Great Lakes Water Quality
Initiative Technical x.pi
Support Document for the
Procedure to Determine
Bioaccumulation Factors -
JUly 1 yy^r ENVIRONMENTAL
PROTECTION
AGENCY
DALLAS. TEXAS
am
Printed on Recycled Paper
-------
DISCLAIMER
This document has been reviewed by the Health and Ecological Criteria Division, Office of
Science and Technology, U.S. Environmental Protection Agency. Publication does not signify
that the contents necessarily reflect the views and policies of the U.S. Environmental Protection
Agency or of any other organization or agency represented by the authors of, or contributors to,
this document. Mention of trade names and commercial products does not constitute endorsement
of their use.
AVAILABILITY NOTICE
This document is available for a fee upon written request or telephone call to:
National Technical Information Service (NTIS)
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
(800) 553-6847
(703) 487-4650
NTIS Document Numbers: PB 94-202009
Paper Copy:
or
Education Resources Information Center/Clearinghouse for Science, Mathematics, and
Environmental Education (ERIC/CSMEE)
1200 Chambers Road, Room 310
Columbus, OH 43212
(614) 292-6717
ERIC Number:
Paper Copy:
-------
Table of Contents
I. INTRODUCTION 1
II. BAFs BASED ON THE CONCENTRATIONS OF THE FREELY DISSOLVED
CHEMICALS IN WATER 1
A. Relationship between BAFs reported on a total and freely dissolved
basis 1
B. Determination of the fraction of the chemical that is freely
dissolved in water 2
C. Derivation of the equation defining ffd 2
III. PREDICTION OF BIOCONCENTRATION FACTORS (BCFs) 6
IV. FOOD CHAIN MULTIPLIERS 9
A. Data for the Mode 9
B. Calculation of the FCMs 11
C. Evaluation of Food Chain Multipliers 12
V. PREDICTION OF BIOACCUMULATION FACTORS (BAFs) FROM BIOTA-
SEDIMENT ACCUMULATION FACTOR (BSAF) MEASUREMENTS 43
A. Biota-sediment Accumulation Factors (BSAFs) 43
B. Relationship of BAFs to BSAFs , 45
C. Calculation of BAFjds from Lake Ontario Data 47
D. Validity of BAF)ds Calculated from BSAFs 48
E. How to Apply the BSAF Method for Predicting BAFjds 50
F. Summary 51
VI. BIOACCUMULATION EQUIVALENCY FACTORS (BEFs) 81
A. Definitions/Symbols 82
B. Calculation of BAFs and TEC from BEFs 83
C. Great Lakes BEFs 84
D. Example of TEC Calculation Using the BEF Method 87
VII. DERIVATION OF BAFs FOR TWENTY TWO CHEMICALS 89
Appendix A. Derivation of the Value of Log Kow for an Organic Chemical ... 93
Appendix B. Derivation of Values of Log Kow for Twenty-two Chemicals .... 96
Appendix C. Derivation of BAFs for Organic Chemicals 104
Appendix D. Derivation of Values for BAF for Mercury 106
-------
TABLE 1: BASELINE BAFs FOR TROPHIC LEVEL 4 BY VARIOUS METHODS . . 109
TABLE 2: RECOMMENDED BAFs FOR THE DERIVATION OF GLI CRITERIA ..113
in
-------
July 1994
GREAT LAKES WATER QUALITY INITIATIVE
TECHNICAL SUPPORT DOCUMENT FOR
THE PROCEDURE TO DETERMINE BIOACCUMULATION FACTORS
I. INTRODUCTION
A. Purpose and Scope
The purpose of this document is to provide the technical information and rational in
support of the proposed procedures to determine bioaccumulation factors (BAFs).
This document contains six sections: 1. Introduction; 2. BAFs based on the
concentration of the freely dissolved chemical in water; 3. Prediction of
bioconcentration factors (BCFs); 4. Food chain multipliers based on the 1993
Gobas model; 5. BAFs from biota-sediment accumulation factor (BSAF)
measurements; 6. Bioaccumulation Equivalency Factors (BEFs); and 7. Derivation
of BAFs for twenty-two chemicals.
Bioaccumulation factors are needed to determine both human health and wildlife
Tier I water quality criteria and Tier II values. Also, they are used to define
Bioaccumulative Chemicals of Concern among the Great Lakes Initiative universe of
pollutants.
II. BAFs BASED ON THE CONCENTRATIONS OF THE FREELY DISSOLVED
CHEMICALS IN WATER
A. Relationship between BAFs reported on a total and freely dissolved basis
The relationship between a BAF reported on the basis of the total concentration of
the chemical in the water, i.e., freely dissolved plus that sorbed to particulate
organic carbon (POO and dissolved organic carbon (DOC), to a BAF reported on
the basis of the freely dissolved concentration of the chemical in the water is as
follows:
BAF* = ffd • BAF|d (1)
where
BAFJ = BAF (L/Kg of lipid) reported on the basis of the lipid-normalized
concentration of chemical in the biota (Kg/Kg lipid) divided by the
total concentration of the chemical in the water (Kg/L);
-------
July 1994
BAF]d = BAF (L/Kg of lipid) reported on the basis of the lipid-normalized
concentration of chemical in the biota (Kg/Kg lipid) divided by the
freely dissolved concentration of the chemical in the water (Kg/L);
ffd = fraction of the total chemical that is freely dissolved in the water.
B. 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 equation 2 with the Kow for the chemical and the DOC and POC
of the water.
fw = 1 / ( 1 + POC • Kow + DOC • Kow / 10 } (2)
where
POC = concentration of particulate organic carbon, Kg of organic carbon/L
of water;
DOC = concentration of dissolved organic carbon, Kg of organic carbon/L
of water;
Kow = n-octanol/water partition coefficient.
C. Derivation of the equation defining fu
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)):
Ci = c:d + POC • Cpoc + DOC • Cdoc (3)
where
Cl* = concentration of freely dissolved chemical in the water, Kg of
chemical/L of water;
CJ, = total concentration of the chemical in the water, Kg of
chemical/L of water;
Cpoc = concentration of chemical sorbed to the particulate organic
carbon in the water, Kg of chemical/Kg of organic carbon;
-------
July 1994
POC =
concentration of chemical sorbed to the dissolved organic
carbon in the water, Kg of chemical/Kg of organic carbon;
concentration of particulate organic carbon, Kg of organic
carbon/L of water;
DOC = concentration of dissolved organic carbon, Kg of organic
carbon/L of water.
The above equation can also be expressed using partitioning relationships as:
Ci = C^d • (1 + POC • Kpoc + DOC • Kdoc) (4)
where
= Cpoc / c;d and Kdoc = Cdoc
KpOC = equilibrium partition coefficient of the chemical between POC and
the freely dissolved phase in the water,
Kdoc = equilibrium partition coefficient of the chemical between DOC and
the freely dissolved phase in the water.
From equation 4, the fraction of the chemical which is freely dissolved in the water
can be calculated using the following equations:
ffd = 1 / (1 + POC • K^ + DOC • Kdoc) (6)
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:
Kdoc - Kow/10 (7)
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:
-------
July 1994
Kpoc - Kow (8)
By substituting equations 7 and 8 into equation 6, the following equation is
obtained:
ffd ~ 1 / { 1 + POC • Kow + DOC • Kow / 10 ) (9)
References
Carter, C.W., and Suffet, I.H. 1982. "Binding of DDT to dissolved humic
materials." Environ. Sci. Techno!., 16, 735-740.
Chin, Y., and P.M. Gschwend. 1992. "Partitioning of polycyclic aromatic
hydrocarbons to marine porewater organic colloids." Environ. Sci. Technol.,
26, 1621-1626.
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.
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.
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.
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.
Gschwend, P.M., and S. Wu. 1985. "On the constancy of sediment-water
partition coefficients of hydrophobic organic pollutants." Environ. Sci.
Techno/., 19, 90-96.
-------
July 1994
Herbert, B.E., P.M. Bertsch, and J.M. Novak. 1993. "Pyrene sorption by water-
soluble organic carbon." Environ. Sci. Technol., 27, 398-403.
Landrum, P.F., 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, 187-
192.
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.
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.
Yin, C., and J.P. Hassett. 1989, "Fugacity and phase distribution of mirex in
Oswego River and Lake Ontario waters." Chemosphere, 19, 1289-1296.
-------
July 1994
III. PREDICTION OF BIOCONCENTRATION FACTORS (BCFs)
Numerous investigations have demonstrated a linear relationship between the
logarithm of the bioconcentration factor (BCF) and the logarithm of the
n-octanol/water partition coefficient (Kow) for lipophilic non-polar organic chemicals
for fish and other aquatic organisms. Isnard and Lambert (1988) have 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 non-polar 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 for non-metabolizable chemicals in most cases predict BCFs
which are larger than the measured BCFs. The losses of the chemicals due to
metabolism are not accounted for in the simple partitioning model (Baron (1990),
de Wolf etal. (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 non-polar 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) recalculated the linear relationship
reported by Mackay (1982) assuming a 5% lipid content and obtained the
following relationship:
log BCF = 1.00 log Kow - 0.08 (1)
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
-------
July 1994
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.
One option for the final GLWQI, is to estimate predicted BCFs using the following
approximation:
BCF{d - Kow (2)
where the BCFjd is the BCF reported on lipid-normalized basis using the freely
dissolved concentration of the chemical in the water. This relationship is
applicable to lipophilic non-polar organic chemicals with log Kows greater than 3
which are either slowly or not metabolized by aquatic organisms.
Equation 2 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 2 will not be 1.0 and 0.0, respectively. The theoretical
basis presented by Mackay (1982) and the experimental data suggest that n-
octanol is a very reasonable surrogate for lipids.
Equation 2 is also supported 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 k, 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 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 U/g of chemical/Kg of organic carbon in the sediment)/(//g of
freely dissolved chemical/L of sediment pore water) (K,oc or Koc) and the lipid/water-
equilibrium partition coefficient (/;g of chemical/Kg of lipid)/U/g of freely dissolved
chemical/L of sediment pore water) (KL) have been demonstrated to be
-------
July 1994
approximately equal to Kow for non-polar organic chemicals in sediments and
benthic organisms, respectively.
References
Baron, M.G. 1990. "Bioconcentration". Environ. Sci. Techno/., 24, 1612-1618.
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.
Gobas, F.A.P.C. 1993. "A model for predicting the bioaccumulation of
hydrophobia 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.
Isnard, P., and S. Lambert. 1988. "Estimating bioconcentration factors from
octanol-water partition coefficients and aqueous solubility". Chemosphere, 17,
21-34.
Mackay, D. 1982. "Correlation of bioconcentration factors". Environ. Sci.
Technol., 16, 274-278.
8
-------
July 1994
IV. FOOD CHAIN MULTIPLIERS
Food chain multipliers (FCMs) for non-polar 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. This model does not predict FCMs but rather
it predicts a) the chemical residues in the organisms and b) the freely dissolved
concentration of the chemical in the water column. With this information,
bioaccumulation factors (BAFs) for each species in the food chain can be
predicted. FCMs can then be calculated from the predicted BAFs using the
following equation:
FCM = BAFjd/Kow (1)
where Kow is the n-octanol/water partition coefficient for the chemical and BAF{d is
the 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 directly from Gobas (1993) (Table 1). The metabolic .transformation
rate constant was set equal to zero. The organic carbon content of the water
column 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
-------
July 1994
column, the following sediment-water column chemical concentration quotient
(n.oc) was calculated for each chemical reported by Oliver and Niimi (1988):
n,oc = no of total chemical/Kg of organic carbon (in the sediment)
ng of freely dissolved chemical/L of water (in the 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:
Freely dissolved fraction = ffd = 1/( 1 + DOC • Kdoc + POC • K^)
Freely dissolved concentration = Cl* = CJ, • fw
where
ffd = fraction of the chemical which is freely dissolved in the water,
DOC = concentration of dissolved organic carbon,
POC = concentration of paniculate organic carbon,
Kdoc = partition coefficient for the chemical between the DOC and the
freely dissolved phase in the water,
Kpoc = partition coefficient for the chemical between the POC phase
and the freely dissolved phase in the water,
CJ, = total concentration of the chemical in the water, and
Cj,d = freely dissolved 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
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 were obtained from Hawker and Connell (1988) for the PCBs, de
Bruijn et al. (1989) for the chlorinated benzenes, p,p-DDT, p,p-DDE, p,p-DDD, a-
HCH, and K-HCH, Pereira et al. (1988) and Chiou (1985) for hexachloro-1,3-
butadiene, McKim et al. (1985) for mirex, and the CLogP program for 2,4,5-
trichlorotoluene, 2,3,6-trichlorotoluene, 2,3,4,5,6-pentachlorotoluene,
octachlorostyrene, and p-chlordane (Leo 1988); for photomirex, its Kow was set
equal to the Kow of mirex. The relationship for determining Kdoc from Kow was
developed from the results reported by Yin and Hassett (1986, 1989), Eadie et al.
10
-------
July 1994
(1990, 1992), Landrum et al. (1984), and Herbert et al. (1993) for partitioning to
DOC.
In Figure 1, the ratios of neoe to Kow are plotted against the log Kow for each
chemical reported by Oliver and Niimi (1988). For the pesticides and PCB
congeners, the ratios of the ntec to Kow were nearly independent of the Kow of the
chemicals, i.e., Pearson correlation coefficients (r) of -0.18 and -0.34 were
obtained for the pesticides and PCBs, respectively. For the chlorinated benzenes,
toluenes, and butadienes, the ratios of Htoc to Kow were slightly dependent upon the
Kow of the chemicals, i.e., Pearson correlation coefficient of -0.52. The average
(standard deviation & number of values) ratios for the nsoc to Kow for 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 ntoc to Kow on Kow for the pesticides
and PCBs and the average ratios above, a value of 25 was selected for this ratio.
The resulting relationship between the concentration of the chemical in the
sediment on an organic carbon basis (C,^) and the freely dissolved concentration
of the chemical in the water column (C^d) is:
Cgoc = 25 • Kow • c:d " (2)
and this relationship is applicable to chemicals with log Kows from 2 to 10.
B. Calculation of the FCMs
The model of Gobas (1993) (MS-DOS version) was run using the input data listed
in Table 1 and the above relationship between the Ctoc and C™ for Kows of 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. To set
the freely dissolved concentration of the chemical to 1 ng/L in the model of Gobas
(1993), the DOC concentration was set equal to 0.0 mg/L.
It also should be noted that 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 ratio of the
concentration of the chemical in the tissue to the concentration of the chemical in
the water column (BAF) obtained using a 1 ng/L concentration of the chemical will
be equal to that obtained using a 150 //g/L concentration of the chemical for a
specified Kow.
11
-------
July 1994
It should be noted that the model of Gobas (1993) takes the inputted total
concentration of the chemical in the water and before doing any predictions,
calculates 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
0.0 mg/L, the total concentration of the chemical inputted to the model becomes
equal to the freely dissolved concentration of the chemical in the water. This
allowed the fixing of the chemical concentration relationship between sediment and
water phases in the model. BAFs were determined by dividing the chemical
residues predicted by the model of Gobas (1993) by the freely dissolved
concentration of the chemical in the water and therefore, are not influenced by the
concentration of DOC inputted to the model.
Listed in Table 2 are the FCMs calculated with equation 1 for the zooplankton,
forage fish, and piscivorous fish.
C. Evaluation of Food Chain Multipliers
BAFs were predicted for each chemical reported by Oliver and Niimi (1988). 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 piscivorous fishes (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 a) growth rates which are much slower than the predator fishes, b) slower
rates of metabolism than the predator fishes for the chemicals of interest, and c)
the feeding preferences for the trophic level 3 fish that includes predation on other
fish. In the development of FCMs, the feeding preferences for smelt (see Gobas
1993) consisted of a mixture of trophic level 2 and 3 organisms, i.e., mysids,
Diporeia sp., and sculpin. This mixture of different trophic levels combined with
bioenergetic factors for the smelt caused the predicted concentrations of the
chemicals and subsequently, the derived FCMs, to be slightly larger than those for
the piscivorous fishes (trophic level 4).
The average differences between the predicted and measured log BAFs were
-0.59, 0.03, -0.15, -0.02, -0.08, and -0.11 for zooplankton, sculpin, alewives,
small smelt, large smelt, and piscivorous fish, respectively.
12
-------
July 1994
References
Chiou, G.T. 1985, "Partition coefficients of organic compounds in lipid-water
systems and correlation with fish bioconcentration factors." Environ. Sci.
Techno!., 19, 57-62.
Cook, P.M., R.J. Erickson, R.L. Spehar, S.P. Bradbury, and G.T. Ankley. "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, March 1993.
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.F. Landrum. 1992. "Distribution
of hydrophobic organic compounds between dissolved and paniculate organic
matter in Green Bay waters." J. Great Lakes Res., 18, 91-97.
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.
Flint, R.W. 1986. "Hypothesized carbon flow through the deep water Lake Ontario
food web." J. Great Lakes Res., 12, 344-354.
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.
Gobas, F.A.P.C. 1993, "A model for predicting the bioaccumulation of
hydrophobic organic chemicals in aquatic food-webs: application to Lake
0 nta rio." Ecological Modelling, 69, 1-17.
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. Techno!., 27, 398-403.
13
-------
July 1994
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, 187-
192.
Leo, A.J. Unified medchem software, version 3.53, Pomona Medicinal Chemistry
Project, 1988.
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.
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. Techno/., 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.
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.
Yin, C. and J.P. Hassett. 1989, "Fugacity and phase distribution of mirex in
Oswego River and Lake Ontario waters." Chemosphere, 19, 1289-1296.
14
-------
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: 0.0 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 cognatos)
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
15
-------
Table 2. Food Chain Multipliers for Zooplankton, Forage Fish and Piscivorous Fish
Trophic Level 2
Log Kow Zooplankton
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
6.6
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Troohic Level 3 Troohic Level 4
Sculpin
1.0
1.0
1.0
1.0
1.0
1.1
1.1
1.1
1.1
1.1
1.2
1.2
1.3
1.4
1.4
1.5
1.7
1.9
2.1
2.3
2.6
3.0
3.5
4.0
4.7
5.4
6.2
7.1
8.2
9.2
10.4
11.5
12.6
13.6
14.6
15.5
16.2
16.8
17.3
Alewives
1.0
1.0
1.0
1.0
1.0
1.0
1.1
1.1
1.1
1.1
1.1
1.2
1.2
1.3
1.3
1.4
1.5
1.7
1.8
2.0
2.3
2.6
2.9
3.3
3.8
4.3
4.9
5.5
6.2
6.9
7.6
8.2
8.9
9.4
10.0
10.4
10.8
11.1
11.3
Smelt"
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.1
1.1
1.1
1.1
1.2
1.2
1-2
1.3
1.4
1.5
1.6
1.8
2.1
2.4
2.8
3.3
3.9
4.7
5.8
7.0
8.6
10.4
12.4
14.6
17.0
19.4
21.7
24.0
25.9
27.6
Piscivorous
Fish
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.1
1.1
1.1
1.1
1.2
1.2
1.3
1.5
1.6
1.9
2.2
2.6
3.2
3.9
4.7
5.8
7.1
8.6
10.2
12.1
14.0
16.0
17.8
19.9
21.7
23.3
24.6
25.6
16
-------
Table 2. Continued.
Trophic Level 2
Log Kow Zooplankton
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
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Troohic Level 3 Troohic Level 4
Sculpin
17.6
17.8
17.9
17.8
17.7
17.3
16.8
16.2
15.8
14.6
13.6
12.6
11.4
10.2
9.0
7.9
6.8
5.8
4.9
4.1
3.4
2.8
2.3
1.8
Alewives
11.5
11.6
11.6
11.5
11.3
11.1
10.8
10.4
9.9
9.4
8.7
8.1
7.4
6.6
5.9
5.1
4.4
3.8
3.2
2.7
2.2
1.8
1.5
1.2
Smelt"
29.0
29.9
30.4
30.5
30.2
29.4
28.3
26.8
24.9
22.8
20.4
18.0
15.5
13.2
11.0
. 9.0
7.3
5.8
4.6
3.6
2.9
2.2
1.8
1.4
Piscivorous
Fish
26.4
26.7
26.7
26.2
25.5
24.3
22.9
21.0
19.0
16.7
14.4
12.1
9.8
7.8
6.0
4.5
3.3
2.4
1.7
1.1
. 0.8
0.5
0.3
0.2
25% of the smelt diet includes sculpin. Therefore, this species is at a trophic
level slightly higher than 3.
17
-------
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., (pg 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
46
47
48
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-TGB
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
70 + 76
56 + 60 + 81
Log Kow
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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
6.17
6.19
Predicted6 Measured"
Log BAF Log BAF
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19.
4.05
4.14
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
6.17
6.19
7.17
7.78
6.38
7.58
7.80
5.88
4.90
5.08
5.05
7.85
5.77
6.38
5.35
5.14
5.33
4.71
4.90
4.07
5.71
6.48
5.69
6.21
5.79
5.69
7.11
7.06
7.47
18
-------
Table 3. Continued.
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
85
86
87
88
89
Chemical"
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
187 + 182
170+190
183
177
174
Log Kow
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
7.19
7.37
7.20
7.08
7.11
Predicted6
Log BAF
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
7.19
7.37
7.20
7.08
7.11
Measured6
Log BAF
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
7.60
8.20
8.16
8.07
7.88
19
-------
Table 3. Continued.
90
91
92
93
94
95
96
97
98
99
100
101
102
Chemical*
178
171
185
173
203 + 1 96
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
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
Predicted6
Log BAF
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
Measured6
Log BAF
8.26
7.69
-0.59
0.40
61
" Chemical abbreviations taken from Oliver and Niimi (1988).
b Predicted BAFs were obtained by taking the product of the FCM and Kow for
each chemical. Because the FCM is set to 1.0 for zooplankton, the predicted
log BAF equals log Kow.
c 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).
20
-------
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.,
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
47
48
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
70 + 76
56 + 60 + 81
Log Kow
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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
6.17
6.19
Predicted6
Log BAF
8.16
8.21
7.38
8.70
8.70
6.39
3.86
3.73
5.25
9.00
6.69
5.85
5.02
4.92
4.96
4.34
4.20
4.29
5.41
5.41
7.57
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
7.36
7.33
7.35
Measured0
Log BAF
7.70
7.95
6.93
8.23
8.15
7.07
4.69
5.05
5.55
8.89
6.54
5.67
4.91
4.57
6.41
6.37
5.97
7.45
7.06
7.48
21
-------
Table 4. Continued.
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
85
86
87
88
89
Chemical"
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
187 + 182
170 + 190
183
177
174
Log Kow
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
7.19
7.37
7.20
7.08
7.11
Predicted15
Log BAF
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
7.38
7.98
8.57
8.43
8.58
8.44
8.33
8.36
Measured0
Log BAF
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
7.13
8.45
8.07
9.15
8.81
8.63
8.24
22
-------
Table 4. Continued.
90
91
92
93
94
95
96
97
98
99
100
101
102
Chemical"
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
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
Predicted6
Log BAF
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
Measured0
Log BAF
9.14
8.52
0.03
0.43
54
' Chemical abbreviations taken from Oliver and Niimi (1988).
b Predicted BAFs were obtained by taking the product of the FCM and Kow for
/ w • W¥
each chemical.
0 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 the chemical in water (jjg of freely dissolved
chemical/L of water).
23
-------
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)/{/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
46
47
48
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
70 + 76
56 + 60 + 81
Log Kow
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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
6.17
6.19
Predicted15 Measured0
Log BAF Log BAF
7.97
8.02
7.22
8.50
8.50
6.28
3.82
3.73
5.20
8.81
6.57
5.76
4.96
4.86
4.90
4.30
4.16
4.25
5.34
5.34
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
7.20
7.17
7.19
7.84
7.98
6.82
8.18
8.09
6.63
4.82
5.00
8.89
6.32
6.53
6.68
6.39
7.57
7.31
7.79
24
-------
Table 5. Continued.
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
85
86
87
88
89
Chemical*
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
187+182
170+190
183
177
174
Log Kow
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
7.19
7.37
7.20
7.08
7.11
Predictedb Measured0
Log BAF Log BAF
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
8.38
8.24
8.39
8.25
8.13
8.16
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
8.15
7.99
8.84
8.46
8.54
8.51
25
-------
Table 5. Continued.
90
91
92
93
94
95
96
97
98
99
100
101
102
Chemical"
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
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
Predicted6
Log BAF
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
Measured0
Log BAF
8.82
8.22
-0.15
0.40
51
Chemical abbreviations taken from Oliver and Niimi (1988).
Predicted BAFs were obtained by taking the product of the FCM and Kow for
each chemical.
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).
26
-------
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., (j/g 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
47
48
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
66
70 + 76
56 + 60 + 81
Log Kow
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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
6.17
6.19
Predicted6
Log BAF
7.97
8.02
7.22
8.50
8.50
6.28
3.82
3.73
5.20
8.81
6.57
5.76
4.96
4.86
4.90
4.30 .
4.16
4.25
5.34
5.34
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
7.20
7.17
7.19
Measured6
Log BAF
7.65
8.22
6.83
8.19
8.21
6.39
4.56
4.77
8.73
6.15
6.57
*
7.46
7.32
7.73
27
-------
Table 6. Continued.
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
85
86
87
88
89
Chemical"
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
187 + 182
170 + 190
183
177
174
Log Kow
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
7.19
7.37
7.20
7.08
7.11
Predicted6 Measured0
Log BAF Log BAF
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
8.38
8.24
8.39
8.25
8.13
8.16
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
8.18
8.01
8.86
8.59
8.54
8.31
23
-------
Table 6. Continued.
90
91
92
93
94
95
96
97
98
99
100
101
102
Chemical"
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
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
Predicted6
Log BAF
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
Measured0
Log BAF
8.79
8.24
-0.02
0.40
48
Chemical abbreviations taken from Oliver and Niimi (1988).
FCMs for alewives were used for the small smelt. Predicted BAFs were
obtained by taking the product of the FCM and Kow for each chemical.
Measured BAFs were determined by dividing the chemical residues on a lipid
weight basis in the organisms (fjQ of chemical/Kg of lipid) by the freely
dissolved concentration of the chemical in water (fjg of freely dissolved
chemical/L of water).
29
-------
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., (fjg 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
47
48
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
70 + 76
56 + 60 + 81
Log Kow
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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
6.17
6.19
Predicted"
Log BAF
8.39
8.44
7.51
8.90
8.90
6.30
3.82
3.73
5.10
9.13
6.66
5.70
4.86
4.78
4.82
4.27 .
4.13
4.22
5.25
5.25
7.74
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
7.49
7.46
7.48
Measured0
Log BAF
8.15
8.38
6.88
8.50
8.43
6.45
4.71
4.82
8.97
6.41
5.88
-.
6.92
7.88
7.71
8.12
30
-------
Table 7. Continued.
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
85
86
87
88
89
Chemical'
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
187+182
170 + 190
183
177
174
Log Kow
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
7.19
7.37
7.20
7.08
7.11
Predicted6
Log BAF
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
8.79
8.66
8.80
8.67
8.56
8.59
Measured0
Log BAF
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
8.45
8.34
9.02
8.85
8.78
8.71
31
-------
Table 7. Continued.
90
91
92
93
94
95
96
97
98
99
100
101
102
Chemical*
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
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
Predicted6
Log BAF
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
Measured6
Log BAF
9.13
8.50
-0.08
0.40
49
Chemical abbreviations taken from Oliver and Niimi (1988).
Predicted BAFs were obtained by taking the product of the FCM and Kow for
each chemical.
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 the chemical in water (jjg of freely dissolved
chemical/L of water).
32
-------
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., U/g 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
45
46
47
48
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
66
70 + 76
56 + 60 + 81
Log Kow
6.91
6.96
6.22
7.50
7.50
5.54
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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
6.17
6.19
Predicted6
Log BAF
8.34
8.38
7.52
8.78
8.78
6.39
3.78
3.69
5.12
8.93
6.74
5.77
4.86
4.78
4.82
4.23.
4.09
4.18
5.27
5.27
7.73
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
6.20
7.50
7.47
7.49
Measured0
Log BAF
8.00
8.46
7.03
8.59
8.53
6.74
4.69
4.93
9.20
6.41
5.81
5.07
6.89
5.75
6.39
5.92
5.32
5.52
6.76
7.79
7.56
7.96
33
-------
Table 8. Continued.
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
85
86
87
88
89
Chemical"
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
187 + 182
170 + 190
183
177
174
Log Kow
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
7.19
7.37
7.20
7.08
7.11
Predicted1"
Log BAF
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
8.15
8.68
8.58
8.69
8.59
8.49
8.52
Measured0
Log BAF
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
8.58
8.43
9.20
9.03
9.01
8.74
34
-------
Table 8. Continued.
90
91
92
93
94
95
96
97
98
99
100
101
102
Chemical"
178
171
185
173
203 + 196
201
194
195
198
205
206
207
209
Average difference
Standard deviation
Number of values
Log Kow
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
8.55
8.52
8.52
8.44
8.81
8.84
8.88
8.78
8.84
8.89
8.87
8.90
8.83
Measured6
Log BAF
9.26
8.56
-0.11
0.40
59
Chemical abbreviations taken from Oliver and Niimi (1988).
Predicted BAFs were obtained by taking the product of the FCM and Kow for
each chemical.
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 the chemical in water (/JQ of freely dissolved
chemical/L of water).
35
-------
— cp
«
CO
co
o « ir
O J £
*- fc- ,*r
® o 5
5 O it
C 3 ®
.= CT .>
112
w' *- o
. eo w
0> C
v. 0)
3 O
O) C
= 8
o
o 55
(0
Q>
Q.
o
O)
o
<3
II 1 1 1 1 I 1
CO
o
o
o
cf
o
o
o
o
o
36
-------
£
0>
5
•o
£
00
CO
o>
£| S
I SS
" a. g
I § I
3 N *-
CO
(Q
0)
CM
U.
U_
<
CD
TJ
o>
3
CO
(/)
0)
•a
OH
" ^B
to
0)
CL
V)
0)
c
a>
O~~
2 Q
O CD
- O<"
(0 -r^
0) Z
f^ ^
§ *
N
i
o
0)
0
Q)
0)
CA
0*
O
ffl
O
Q.
O)
c
*w
3
•o
0)
s ^
~ •£
•0 o
£ ul
Q.
00
u_
<
CD
<\
o>
^ jl
II C
^i
5 I
51
W u
" "c
o o
c
o
0)
o
in
CO
0
CO
UD
to
CO
jva
37
-------
!•§
^J
/A
redicted BAF«
Q.
•o
C
CO
•o
0)
3
CO
CD
CO
0)
3
O)
LJL
^^
00
0)
E
z
•o
c
CO
c >
fl
» 1
CO
LL.
*{f
m
•o
£
•Mi
3
CO
CO
0)
s
1
c
\
•
-
•'
• \
\ •
A '-A
^^^
9 ** ^9
A *4|r
•*.'
*^<
.^
• ^
•
I.I.
3 O) CO !*•
(0
5: O)
o £
3 W 2?
« 0 0 3
m -T rt C T3
•n 7T 9 ® <» ,«
W (J £Q Q) *» W)
f I ~ O 0) •*• Q 0) ML
fll ^Z itf^ '^ ^^
^ og
-^ r^
.1 ~C
S §
J"
f s
C4 o
II "c
00 i
0 ?
o o
e
i
CO
h-
O
co ^
0)
0
m
0
dva 601
38
-------
O)
O «
,2 CO
CO «
• • «^
0) *•
S 1
V CO
£ »-
£ 8 §
"° £ 5
c •> O
Si £
« < o
en
o>
•o
0)
D.
(O
0)
0)
O
oi"
*•
3
CO
O)
in
39
-------
O 03
.2 CO
CO °>
LL. «E»
m 1
S *
o
•o
c
CO
0)
>
c w O
CO = —
•o JS £
« E o
en
CQ
CD
3
cn
CO
LL
CQ
•o
Q>
CQ
Q>
O)
«
0)
Q.
o
° Q
a>
1
CD
O
CL
U
M
U.
<
m
V.
1
O)
co
'£ "
II C
|i
36 o
E 8
W jj
§1
o o
GO r^> co
dva Bon
in
CO
40
-------
CO
CD
"g
£
2
00
1 s
5 u
II C
° i
2 §
W u
I" C
8?
Q O
C
•
§
O)
o
o>
h» (O
dva 601
41
-------
«n
o
•o
o>
"55
o>
Q.
to
o
0)
3
"5
^ Q
Q CD ^
- O<£
CO T^4
0) *
1*
N
0
X2
0
(O
0)
C
0)
O)
c^
ow
0
m
o
a.
o>
"35
3
•o
0)
O
•o
2
a
(0
u.
CD
(ft
S
O
u.
00
si
o o
^E
1
, CO
T3 -5
0) z
Q <0 "O
.= 0) C
T3 JI 03
9 CO ,_
£ c g
•o
GO
601
CO
42
-------
July 1994
V. PREDICTION OF BIOACCUMULATION FACTORS (BAFs) FROM BIOTA-
SEDIMENT ACCUMULATION FACTOR (BSAF) MEASUREMENTS
Biota-sediment accumulation factors 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; 1994). Since
BSAFs are based on field data and incorporate effects of metabolism,
biomagnification, growth, etc., BAF)ds 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{ds from BSAFs also provides a method for validation of all
measured or predicted BAF{ds for nonpolar organic chemicals.
A. 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:
where
Cj = lipid-normalized concentration of the chemical in tissues of the
biota U/g/g lipid).
C,oc = organic carbon-normalized concentration of the chemical in the
surface sediment (//g/g sediment organic carbon).
43
-------
July 1994
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 (equation 2):
Cf Kt Kt
BSAF « —- = Dh •—- * Dh - 2 <2)
r* v K te
Cs Ksoc **
where
Cbd = concentration of freely dissolved chemical (associated with
water) in the tissues of biota (//g/g wet tissue).
Cjd = concentration of freely dissolved chemical (associated with pore
water) in the sediment (fjg/g sediment organic carbon).
K! = lipid-water equilibrium partition coefficient = C|/Ctd.
Ktoc = the sediment organic carbon-water equilibrium partition
coefficient =
D,,, = the disequilibrium (fugacity) ratio between biota and sediment
V80C'
Measured BSAFs may range widely for different chemicals depending on Kf, K,
and the actual ratio of Cbd to C". At equilibrium, which rarely exists between
sediment and pelagic organisms such as fish, the BSAF would be expected to
equal the ratio of Kj/K.,,,. 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^ = Cbd/Cjd) 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
44
-------
July 1994
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 Ctocs.
B. 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 Dte. 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 measured BAFjd and other
chemicals (n = i) for which BAFjds are to be determined. BAF)d for a chemical "i" is
defined as:
(eft,
where
C^,d = concentration of freely dissolved chemical in water
water).
Substitution of C, from equation 1 into C, of equation 3 for the chemical i gives:
(BAF?), = (BSAF), • -^i (4)
45
-------
July 1994
In order to avoid confusion with the equilibrium partition coefficients Ktoe, Kpoc or
Kdoc, the chemical concentration quotient between sediment organic carbon and a
freely dissolved state in overlying water is symbolized by PI,,,,.:
(eft,
Thus the ratio of BAFjds for chemical i and a reference chemical r is:
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:
OU,
OUr
(7)
therefore:
= (Rtf )r • -' (8)
"' (BSAF)r(KJ,
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 BAFjd
calculations: (1) the concentration ratios between sediment and suspended solids
in the water and (2) the degree of equilibrium between suspended solids and Cj,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 C" 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
chemical and the Kows for both the reference and unknown chemicals. Such errors
can be minimized by comparing results from several reference chemicals and
assuring consistent use of C^d values which are adjusted for dissolved organic
46
-------
July 1994
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 paniculate organic carbon concentrations in the water (POC and DOC):
BAF\ = BAFf-fp (9)
where:
1 1
DOC-KJW
(
Further information on calculation of concentrations of freely dissolved chemicals
in water may be found in section II of this document titled "BAFs Based on
Concentrations of the Freely Dissolved Chemical in Water".
C. Calculation of BAFjds 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 food chain multipliers, biomagnification factors and BAFjds 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)ds calculated from
Oliver and Niimi data from samples collected between 1981-1984 and calculating
BAFjds 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 10). 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
47
-------
July 1994
penta, hexa, and hepta 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 1. Table 1 contains the measured and
predicted log BAFjds from the two data sets.
D. Validity of BAF]ds Calculated from BSAFs
Figures 1, 2 and 3 show the relationship of log BAFjds to log Kows for (1) Oliver
and Niimi (1988) BAF]ds determined from measured concentrations of freely
dissolved chemicals in Lake Ontario water in 1984; (2) BAFjds calculated from
BSAFs derived from Oliver and Niimi data; and (3) BAF{ds calculated from EPA
BSAFs for lake trout in Lake Ontario in 1987 (Cook et al., 1994). The diagonal
lines represent a 1:1 ratio of log BAF to log Kow. The PCB congener BAF)ds in all
three sets of data appear quite similar. The EPA BAF)ds 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 BAF)d 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 choosen 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 BAF)ds 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 BAFjds calculated from BSAFs to measured BAF)ds. Figure 4 illustrates the
agreement between log BAF]ds calculated from the Oliver and Niimi water data and
those calculated from the sediment data. The BAF)ds 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
BAFjds that correlate equally well with the BAF)ds calculated from Oliver and Niimi
data (figure 5).
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 1 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
48
-------
July 1994
two data sets and four reference chemicals resulted in either four or eight
determinations of BAFjd for each chemical listed in Table 1. Mean log BAF)ds
(geometric means of BAF{ds) for the 4-8 determinations from Lake Ontario data are
reported in Table 2. 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
Io9 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 C^d.
The greatest test for robustness of the BSAF method for predicting BAFjds that are
applicable throughout the Great Lakes would be a comparison of two totally
independent data sets based on different ecosytems 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
1 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 Kdoe = Kow/10. Despite the complex exposures of Green Bay fish, figures
6 and 7 illustrate log BAF)d - 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 8-11 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 BAFjds 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
2) are plotted against log Kow in figure 12. Only 59 of these chemicals have
measured BAFjds. Correlations between the mean Lake Ontario trout and Green
Bay brown trout BAFjds (figures 13 and 14) indicate that the Green Bay brown
trout estimates are 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
Lake Ontario and Green Bay PCB congener 198 BAF)ds are noticeably different in
49
-------
July 1994
figures 13 and 14 (greatest log BAF)d based on Lake Ontatio BSAFs) and cause
most of the slight deviation of the slope of the linear regession lines from 1.0. The
agreement of the Green Bay and Lake Ontario results demonstrates the general
applicability of BAFjds calculated from BSAFs in predicting bioaccumulation in
Great Lakes fish from estimated Cj|,ds.
E. How to Apply the BSAF Method for Predicting BAF"s
If high quality data are not available for calculating BAF)ds for organic chemicals
that are expected to bioaccumulate, the mean BAFjds reported in Table 2 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:
1. 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
Cj,d used to calculate a BAFf,d should be based on a consistent adjustment of the
concentration of total chemical in water for DOC and POC using equation 10. It is
preferable to choose at least some reference chemicals on the basis of log Kow and
chemical class similarity with the test chemicals.
2. Measured (slow-stir method or equivalent preferred) or estimated Log Kow
values are chosen for each chemical.
3. 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.
4. 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.
50
-------
July 1994
5. With the data from steps 3 and 4, calculate BSAFs for chemicals of interest
and reference chemicals (equation 1).
6. With BSAFs and Kows for each chemical, plus BAFjds for reference
chemicals, calculate BAF)ds using equation 8.
7. Use the BAF{ds 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.
F. Summary
BAF)ds calculated from two different BSAF data sets for Lake Ontario salmonids
are similar and agree well with measured BAFjds 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. BAF|ds 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)ds (geometric
mean of BAF}ds) from 4-8 determinations from Lake Ontario data are summarized
in Table 2.
References
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.
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, March 1993.
Cook, P.M., G.T. Ankley, R.J. Erickson, B.C. Butterworth, S.W. Kohlbry, P.
Marquis and H. Corcoran. 1994. The biota-sediment accumulation factor
(BSAF): evaluation and application to assessment of organic chemical
bioaccumulation in the Great Lakes. In preparation.
51
-------
July 1994
Hawker, D.W. and D.W. Connell. 1988. Octanol-water partition coefficients of
polychlorinated biphenyl congeners. Environ. Sci. Technol. 22:382-387.
Oliver, B.C. and A.J. Niimi. 1988. Trophodynamic analysis of polychlorinated
biphenyl congeners and other chlorinated hydrocarbons in the Lake Ontario
ecosystem, Environ. Sci. Techno/., 22, 388-397.
U.S. EPA. 1990. Lake Ontario TCDD Bioaccumulation Study - Final Report, U.S.
Environmental Protection Agency, Region II, New York, NY.
52
-------
CO
^^
V
en
to
Q.
i
1 1
Ik
CD
"35
U.
V)
to
•o
in
ID
O
•o
o>
m
•3
u
To
U
"L
uT
03
Q
^"
V
a
V
0
,
T-
a>
5
LU CL
0 ».
0)
•
LL j= If)
a*5
O» w Q-
O Q) 14.
^~ ^ 03
o
TJ
LL a w
^ ^ 2
m LU *"
E
•
E T-J
<1 1
en S 2
o
'§
LL •— flj
'-COlf)lf)lf)lf)Lf)COCO^
2 c * c o o "S
.— 45 " ""iJ-o^
C P fc» ^^ ^» ^ m «C W
^ xO.2^.0cCR>E-0
•5 ^ ^ 5 -| £ V "5 "V ° "r'o-1'8
TD ^3 ^3 ^3 C u O) *^ W v O i-i 4->
CMCMCOCMCM^COo
-------
ID
N
S
i
2
•e
u
n
U
o
*t
»
m
£
O
>
m
U
oo oo^ o^ro o co uin^ooi
co co co in cb cb cb CD cb cb cb in r^ r> r»r»r*-r^f^
oqpeN^cNp«-poooo cpppoj CNLO*- en en
Lo'cbcbcbi^cbcbcbcbcb cbr^p^i^ r^r^od r* r-
co ^ cnor^co^ cnoo otstN CN CN »-Oir^oocr)
d d do'odd do odd d d dddd^
LOCO»-LOCNCOIOOCNOO r*cn«-rt COLOLO LO ^
*^ ^^ ^^ (^ ^^ ^^ *"* *** KO ^^ ^O ^O ^Q CO ^^ 00 ^J 'if (^
dddd^ddddd ddd*^ ddci oi ri
CNCNLOO5 tOCNLOOiCO CO^-^- »-CJ1CD OO 00
LOLOLOCO tOLocbi^cb i^r^co r-»Kr^ 06 06
rrinLbinLbininu^inibLOLOininLOininLriin'inLbcbin'cbcbcbcbcbcb
T-r-T-r-^CNCNCNCOn^r^-'t^-t^'LOLOCOCDCDr^r^COCOCOCO
m
O
a.
54
-------
I
Q.
(A
CO
•o
3
v>
ID
W
O
U
uT
03
o
I/I
w
0)
o
m < 03
c in
u. j°
03 <* P3,
.i qj
o -
CM
u. in
O LU 0-
O s_
"~ W
ECN
u. ~ in
03 a8 U
O) w. Ck
.2 ® M-
T3
Q>
LL Z £
«n s I
.2 E
O
03 c8
QJ
O
o-i
1C
u
E
w
6
o3o»-fncocsocnfN«-^«j
co^t
CO CO •* CO O
i- ooinro r^ csco»-o incM i-^O'-'-incncMcnco en
in cNinco CM oo«™r^^ co O in co in co ^ o CM in oo in
px (**, px px p^ pv oo r^ oo oo oo r** oo oo oo oo oo oo oo oo oo
fp ^v 00 00 ^^ CO in Ifi ^f CM ^^ ^f CO CO OO i*™ CO O5 ^^ If) ^0 CO
co c^ ^f CM ^™ ^j en oo o) ^o p^ en p^ co ^p co ^^ ^^ ^^ ^^ ^^ f p
pv px px p^ 00 00 P^ 00 P^ 00 00 00 00 00 00 00 00 00 00 00 00 CO
tn t-cnco t- mr^in^- coco o**<**i-cocMincoto co
00 CO CO CO tp ^^ ^^ ^^ ^^ fp f^ QJ {Q ^rt fQ p^ ^^ ^^ gjj f^ gft
p*. pvp^p*. px oooooooo oooo p^oooooooooooooocn oo
cnp^^t oo«-oocoo)CMCMco»-cocoin ^•incoin«- cMcn
in mcoo co inococn ^-m p^p^inuscMcocofMP'* «~
^t cM^rt co Tfp-.inp COP^ oqcocMpvcMcococNcn P*.
r^ ^ ^ r^ O CMCMt-^'* CO^ dd^J-CMCOCMCO^tCO CM
oacDcbcdcdcbcocbtdcdcdcdcdcDcdcbcdcococdtbcbcbcocorvr^p^px
(O
co
u
a.
55
-------
in
CD ^ OQ
?&*
—• *fc-
0)
O
«*-
O *"
CM
in
O> LL) Cv
O >«_
LL |= in
m v. Q.
O Q> <*_
~.s? 5
u.. £
< < 3
W P-, «
CD ^ S
1
io
«J
V
I
•a
0)
5
u
15
U
w
•Q •«
U.
o
M
V
O
(0
l-o
LL
m
l
o> »_ 5
o H
s
0)
CJ
O?
CM
C*5 ff)
O T-
O CO O5 O) CM
r»» n CN m o
r^ co O «- in r>j oo
«- CN in in *r in ^t
ooooooooooooooooo)
OOOCMCOCOfOCOCM
cocnoocooooorvinincocdcbcdcocd
co
r>i
cocooscnqco
oo oo oo oo en oo
men
oo
en oo
comoqcqcncq^t
06 06 06 06 co in cb
in
en
(6
en
on
oo
GJ
cb
co m
o
m
co »-
op <*
p^ 06
CO f CO CM CO 00
CD O «- CO CO CO
co cn CT> cn cn oo
Pv CM
CO CM
O) 05
cn«-ininors'f>> en
«-Cnr-CT)pv»-P«. CM
O5 00 O) 00 00 CD CO F^
CM
p-
CM
CO
CO
o
CO
CM
CM
CO
«-«-»-CMCOCMCMOCM »-CO«-OOOOOOOOOOOO i- *-
*t co co co CM in
in in n; P-. cq in
^ co ^t- co in ^
CO O
in en
co TJ- P- co t in CM
in co d 6 6 o 6
co
CM
en
•<*
in
in
— o
in
CM
So
06 en
co co
in p
06 en
co
m
06
en co
oo ^
co P^
co
in
co
en
co
p
co
CO
CO
o
a.
+
CO
in
CO
56
-------
CD
1A
•*#
«
o
<0
a
(
u.
00
M
U.
tf>
GO
•o
3
2
1
o
•o
*•*
3
_o
"To
U
w>
uT
00
3
£
M
<0
To
V
»J
S
H
in
LL °
^ j> ^""
CD < CO
g)LU £
£
i ^
U.JO
^
GO _* CD
a*"
,0 oj a-
~ "S
O "
CM
u. in
< •", 03
CD < r^
CL. ^J
ex uu CL.
O x_
~" 05
E CM
LL := in
< Z CD
00 CJ
0)^.0-
o eo H_
~* * ^ Q)
6
LL. 1
< < 3
co Q- "5
m uj £
E
.
I-o
•— Q)
LL Z S
» 00 (O
m fc 2
.> E
6
u.1*
< 2 w
03 o0 55
es k. "
o E
O
1
o
o
"5
o
Q)
^*
CJ
en o in ^™ co co
o ^— CN oo co in
CO CO 00 00 00 00
r- co en
oo r^ oo
00 00 00
«- CM P*. co *}• in
oo ao en in o CM
oo oo oo en en en
in co co
*-m O CM
en en en
co in co r^ CM CM
' CM CM in «- r^ »-
CM r~ r- ^t O r-
O CD CD
oo o in
CO CM ^
CO O CD
Tf CM CM
oo en en
«— oo en co r*> ^f in
en «- «- CN co co co
co CM r*- O o oo co
co o oo o en o o
T~ CM «- eM T- CM CM
+ + + + + -)• +
oo ^ CM r^ o in co
co r^ oo in r^ en en
U) ,
CD +
o
-------
X)
«
Ol
to
a.
M
U.
1
w
m
•o
V
3
W
TO
•>
E
o
i_
i^
TS
0)
.£2
_o
(Q
U
uT
2
|
M
1
_l
4*
(0
w
O
8
>oo
LL
uu £
u. « in
|Ss
x
< CD
^£
CO OL
LU
00
00 < CD
g»u3 of
"" "w
E00
< z ^
CD CD
o * OL
~" = "«
o <-
U-
CO ^^ ^^
^l tJ
OLU O
o
•
P
.. E -o
U- := «
0 > ®
5
O
O
o
"S
CD
6
O»-CN cn tNcnco^t
^^ 00 O) GO ^^ ^* ^^ CD
m oo r~ in o co r^co room oo
^^ CO ^f ^* n CO O5 CO ^^ Uf) lO UD
r^ oo CD oo oo CD -* rt co in t cn
wo^ oo «— oinco
^^ 00 00 00 CD ^^ ^^ CO
§n CN ^- in ^ CN oo co in o n
co ^^ ^^ ^) co c^ u) co r^ oo oo
CD oo co oo oo co in ^t co in ^t co
O^OlOOP^ COCO CM^OO O
oooorvooooco t ^r cn CD m in
^5 ^^ CO ^^ ^3 CO ^^ ^^ ^^ CO CO 00 OJ ^J ^f CO 00 ^P ^^ ^f O) ^O ^^ CO CO CO
^^ O5 Oi ^^ ^^ ^^ ^0 tf5 ifl CO CO ^^ CO 00 O) P^ """ CO CO CO *^ ^^ ^* O5 O5 CO
CD C g C 0 0 "5
I x|!illl!l, lllniss
**- ^ Q> ^i £ o *S ^^ ® $•"£ CJ5
-------
>oo
to i—
— °- "ft
UJ £
> CM
u. «s in
CD
03
O
53
CD °-.
.
01
a
a
!*•
i
M
U.
3
00
tt
M
a
a>
_jo
o
To
U
M
TJ«.
U.
O
a
£
o
3
a
00
-
ffi < CD
_ O- O
-IS
6
m Z H
a O
5 >
ra
.o
o
6
CM en CM
r- CM CM
in cb in
oo m oo
oo o en
n in 03 t ^t
«- o «- in CD
r^ O
in
(O CO (O CO CO
en »- ^t o O
00 CO O) CO ^
ifi ui in cd cb
c co
CO CO
oo CN
§CN oo cs in «-
en co in en op
CD cb K r^ CD r^
o o oo
i*- CD cn
in ^ en co m
p>> co co P>» en
6 CD 6 d O
in co
en en
CM cp
6 6
• ;
co co K r^ CD
in
co co
* co
oo
en
oo
eo»-^f^tcMp». OCOCMO
«-cocor^«-co «-^«-co
«- o co «»• CM
CO CN Tt
en
r>.
in
q
cb
CM «- en CD oo
^. * T •*? ^!
cb in cb cb cb
co en
•«* q
cb cb
co in t-
in q q
cb cb in
cp
cb
co
CM
co o co
»-
ro
CN ^f p<- oq cp in in rt CN CN
inininincbinin'incbcb
en en CM oo
O rf Tt CD
»- co rt
co en in
co co co co
co co
oo
M;
in
co
r>.
in
^- r- CO CO P^ COCO CMTj-O CO in ID O) in
^«-eOO'-; ^P«; ^ft1"*. «*? °? W'O.O
cb in in cb cb cb in cb in ^t cb cb cb cb P^
ininineocoininmeoco
^ * oo ^p
co ix cq en
cb cb cb cb
r-* oo en
oo »- rv
CM CM in en
en in tx co
in in in cb
co CM in en co
r*. co in »t en
cb in cb r>^ cb
«- q in
p»- p^ cb
co
«- en CD
in tv to
^tminininininininininininininininmminincoincocococo
m
m
O
a.
59
-------
.
en
tt
a
a
M
LL.
CD
2
CD
^
£
3
M
a
V
E
0
m
3
U
15
U
A
LL
3
O
w
¥>
tt
a
a
«•
^}
TS
<0
1-
>
L^ (Q
*f 00
CD (T
a<
~ LLJ
li CO
00 u
o<
— CL
LU
^ >
< m
> .i
CQ UL
Ml
1 1
m <
Q& li I
gVU
*
u.1
< Z
03
a»
O
LL
< ^
™ £
O) UJ
o
•
u.1
m j:
c? •
"|
••; '""
^t 4^
00
- >
6
1
at
"5
u
E
OJ
^^
u
00
f™
CL
"S
S
U
a.
*£
oo
CO
.
re
« O CO «- O O
r^> oo r*> oo oo oo
m CD CN Tt
in r» CD o> oo oo
^O ^^ r~j U) CO ^^
^ in r*> r*» CD oo
CO CO O5 C^ ^**
co in in co «-
cb r>« r*» K r*^
O5 CN tO CM O
CO CO CM O CO
^ rx r* rv rv
CM in oo cu o
CO CM CM CM 00
co r*» r>> r-» CD
^t 00 <- 00 CO
co in in CM in
CO 00 O) CM «-
i- CM 00 CO «-
CO CO K CO 00
CM CM 0 CO CM r CO
CO CO (O CD CO CO CO
ao S ao S oo a> m
CO 00
CO CM
in cn
CM CO
r^ r^
" CM CO
CO CD CO
SS5S
r^
[<
CO
CM
P^
^
T_
in
CO
rv
in
CO
fs^
CM
CD
g
CO CO °
00 00 CO
Tt CD
^ ^ CO
06 *~- r^
00
«- in
O • Tt
*- in
to oor^
in Tt m
t*. oo r^
IV 00 CO
in co Tt
rs. rx p.
r* r*. co
CM r- CM
r* oo p>»
CM co CM
in co cn
^^ 00 ^^
oo [fy oo
CO CD CD
*^ ^O ^5
^3 ^3 *^
Tt O S
CO 00 00
O CO ^
CM oo in
00 (^ 00
co co t-
Tt CO O
*~
inTt rv co
^ co in co
00 00 00 CO
in o
CO 00
Tt co co in
00 O CM O
r* oo oo oo
o in
Tt CO
CO CO
in
CO
rt S S £
CD CD CD ID
22SS
00 ^^ O5
OO 00 00
-
co in co
00 00 CO
«- O CM
«- o> o
co »- Tt co
cn co CM ^«
rx ao oo oo
r- CM in co co o
P^ CO O CM O CM
p- r-cococooo
in o co m
co o cn co
px oo P* f-»
co r^ i- «- >
oo cn
px pv
*- CM
Lf) P»
rv. ^
cn CM
O CM
00 CO
cn «-
cn in
P* CO
CO CO
CD CO
T?S
CO
00
*•
00
in
in
00
CO
00
Tt 00
co in
oooo
p*
o
00
O CO
CO 00
00 00
CM
CO
00
cn "
CD P^
ss
00
00
o
00
0
p^
S
60
-------
.Q
Ot
<0
a
i
M
U.
i
M
U.
i
•o
s
3
M
oo
U. ™ *~
JO (« OQ
CT< 0_
~~ 111 £
U. «0 if)
<°°cn
co o 3
j?< i
u- re
< m
CO /m
fecf
00
u» ^™
CO < 00
a. u
n> 1 1 1 f^
auj Q.
£
«
< 1 *~
08 5
o> o_
— >' «.
~ Q>
O ™
U-
<**r\-
ol§
o
•
LL 1
< Z h-
»<*§
O"
^_ ^
p
E "D
U- :~ fl)
2. ^
03 ag |
— .> £
0
,
C3)
IS
u
o
£
O
oo com OfNfMmes
^t r^o) ^tococMcn
0) COCO 0)0)000)0)
rj- eMin COOOCO.-CO
es ID r** *™ P^ co o O
0) OOOO 0)00000)0)
cs COTJ- r*oeoif)O)
O *f O OlOfN^tfN
CD ^-co ococonen
co «too<*coinonTjTv
es coeotop^ooocor^co
00 000000000)0000000)
^ oencD^mo oo^
CO +~ ^ If) P^ P*» +•• P^ CD
00 000000000000 OOOO
» cnr^rj^^onenco
P^ OOOOOOCOOOCOQOOOO)
o in oo Tt os o in TJ- o)
co n CD co CD o oo o co
00 000000000)00 O) 00
•<}•*— oo n co
r> o mo in
CO O) 00 O) 00
£ - 8 - § * 8 8 - £ § £
cQr^Sf^Kf^oococoooSfl)
VkS *W **J
»
CD
O
0^
O) O O) P^ CO CD
CM cr> rj- o oo ^t
03 0) 0) 0) 00 CO
O O CN 00 CO CN
00 0) 0) 00 CO 00
^ p* o O) eo »~
O ^t CO O »- O
CM r- m CD o co n
in CD oo oo oo oo oo
oo o> oo oo oo oo r»
o e*) co co «•-
CD n in e*) *—
00 00 ^0 ^^ QQ
'
i- O ^ in o) co CN
00 O) 00 00 00 00 f*»
co oo CM «- r*
co in co co o
00 00 00 00 00
^^ ^^ ^^ ^^ O5 ^^ 00
O5 O) ^5 C) ^D ^D ^D
0)Oco«-cor^inr^cDco
inincop>«CM«— cMcocMO
cocop*i^i^p«-p^r^p-co
COCMCOtOOO^OOO
CD ^P ^^ ^^ ^^ (O ^^ ^^ ^^ ^^
UO ^^ \fi fp ^^ ^^ ^^ ^^ ^^ ^^
^o ^p r^* co ^o ^™* c^ <"• ^* c^
«-ocop»cN'-»-eo«-p*
UC> po
00 00 O "* C*)
in t- ps <- p^
m co co PS CD
^•fMcooorscNco m
mincDcococococo r»
t ^t CD O) O)
oo ^ o> n o)
U) CO CO P^ CO
O) 00 CO CO
co «- in CD
CO PS P^ P^
0 r. o g PS j- PS r> o, o
ininininincococococo
ls> ^vl ffl /"^ f rt r\ rvl
CMO^^psCOPsOOOO)
+ + + + + + + + + +
+ +
i- CD
^- in
61
-------
.d
tc
0
CK
2
o.
1
M
U.
"*»
u.
S)
m
•o
M
a
0
•fc
•o
0
75
3
U
To
U
*L
uT
2
*<
o
H
0
a
0
0
0
3
a
H
ii m *—
^ CD ^
00 (3 S
.2 £ M-
LU £
u. w in
*J 03
00 O o
0»<( 0-
~LU £
u. re
< m
• i
20 ^»
^U
00
LL ^
m ,^.
5 £
LL
00 Q[ O
OB LU O
O
.
g
0
i
f
ra
E
0
(J
cncoajT-cor^couoco
OO>O5CM^^O5CDCN
co r^ oo cn cn en oo oo a>
incNinr^tNcoTt^^
oo r^* n* co o) o* r^ ^T cj
r^r^-cooocooocoooo)
o5Sro«coo5^
*~ *~
* m in o CD rv r^
O rf Tt CO <- CO 00
00 00 00 00 O> 00 00
00 CO 00 ^t
t in ^ cq
r^ oo oo oo
co n ^ o in co r*>
r^ *— «— CN oo co m
^^ 00 00 00 00 00 00
CO CN CO O)
fv oo r^ oo
fv 00 00 00
oo co o co
O Tf CM CN
00 CO O O)
8§s;£2S£So§
^cococsr^ooooco
cnrvcooooocDOO
+ «-*-CN*-eN«-CNCS
PX + + -1- + + + + +
03 coco^oOLor^mo)
a. +
r^^co^-oor^TtmcnLOcocN^-oocnTfr^
Oi^-cNoooooococor^v-cor^csrscDino}
cor^r^cocococOLninr^cocor^r^mr^co
COOfMCOOOCD^-^tOOLOCM^-^tr-OOCOCD
co^-cninmincon^-cootcnencDcNco
CDrvCOCDCOCDCDininCDCDCOCDCDLOr>-CD
CNOOOOOOOOOOOOOOOO
QDQQQQOODODDQDDQD
OOOUOUOOOOCJCJOOOOO
K o> x x x o.o^^f££££ o. o-O
r^ ob ob ob 05 cb r^poopooobcnoboboj
!wk i^ .r^ ^^ ^^ i.^ m r** r^ i^ fSk fo i^ i^ ft\
CO r^ pv fv CO ^* ' -i *^ r^ r^ uu r" r^ wu
OiCO^cDr^CD ^S^SiSi^SSIi
62
-------
Table 2. Mean BAFjs from Lake Ontario BSAFs for Salmonids - page 1
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
1245tcb
1234tcb
135tcb
124tcb
123tcb
245tct
236tct
pet
PCBs
5
6
8
12
13
16
17
18
22
25
26
logK.
5.40
6.91
6.96
6.22
7.50
6.76
5.54
5.54
5.54
5.66
5.66
3.78
3.69
4.84
7.94
5.73
5.18
4.66
4.60
4.64
4.19
4.05
4.14
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 BAFf
7.33
8.13
8.80
6.74
8.72
8.57
6.96
6.95
7.32
7.79
6.47
5.24
4.60
9.00
5.77
4.82
3.85
5.72
5.97
5.92
5.54
6.04
6.21
6.69
Mean
BAF?
2.12e+07
1.34e+08
6.30e-l-08
5.53e+06
5.29e+08
3.71e+08
9.07e+06
8.81e+06
2.10e+07
6.10e+07
2.96e+06
1.74e+05
3.96e+04
l.Ole+09
5.90e+05
6.63e+04
7.10e+03
.
5.25e+05
9.28e+05
8.31e+05
3.45e+05
1.10e+06
1.63e+06
4.86e+06
53
-------
Table 2. Mean BAF?s from Lake Ontario BSAFs for Salmonids - page 2
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
136
138
141
logK^
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
6.22
6.83
6.82
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
4
4
8
Mean
log BAF?
5.78
6.12
5.88
6.55
6.48
5.98
5.65
6.76
6.63
6.96
7.19
6.89
7.20
7.45
6.93
7.29
.7.11
7.49
7.59
7.52
7.53
7.17
7.58
7.35
6.84
7.48
7.58
7.68
8.28
7.62
8.25
8.27
8.50
8.33
7.97
8.24
7.59
8.33
8.53
8.25
Mean
BAF?
5.97e+05
1.31e+06
7.61e+05
3.54e+06
3.02e+06
9.47e+05
4.446+05
5.77e+06
4.28e+06
9.04e+06
1.546+07
7.67e+06
1.59e+07
2.81e+07
8.44e+06
1.95e+07
1.29e+07
3.08e+07
3.92e+07
3.34e+07
3.39e+07
1.48e+07
3.77e+07
2.23e+07
6.94e+06
3.05e+07
3.84e+07
4.73e+07
1.90e+08
4.13e+07
1.78e+08
1.85e+08
3.17e+08
2.13e+08
9.25e+07
1.72e+08
3.90e+07
2.12e+08
3.39e+08
1.77e+08
64
-------
Table 2. Mean BAFjs from Lake Ontario BSAFs for Salmonids - page 3
Chemical
PCBs
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
87+97
137+176
138+163
156+171+202
182+187
logK.
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
6.29
6.80
6.91
7.18
7.19
Number
BAFs
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
4
4
4
4
4
Mean
log BAF?
8.28
7.93
8.10
8.46
8.85
8.31
8.21
8.61
8.57
8.34
8.58
8.77
8.99
8.88
8.47
8.67
9.18
8.91
8.45
9.54
8.83
8.69
8.78
8.71
8.07
5.72
6.25
6.70
6.85
6.64
6.71
7.23
7.00
7.01
7.39
7.75
7.97
8.36
8.38
8.84
Mean
BAF?
1.91e+08
8.43e+07
1.26e+08
2.89e+08
7.08e+08
2.02e+08
1.63e+08
4.12e+08
3.69e+08
2.19e+08
3.82e+08
5.94e+08
9.83e+08
7.53e+08
2.93e+08
4.63e+08
1.506+09
8.13e+08
2.79e+08
3.47e+09
6.71e+08
4.92e+08
6.02e+08
5.16e+08
1.18e+08
5.21e-07
1.80e+06
4.97e+06
7.14e+06
4.33e+06
5.07e+06
1.71e+07
9.96e+06
l.Ole+07
2.46e+07
5.64e+07
9.30e+07
2.30e+08
2.40e+08
6.85e+08
65
-------
Table 2. Mean BAFfc from Lake Ontario BSAFs for Salmon ids - page 4
Chemical
PCBs
157+200
170+190
195+208
196+203
PCDDs
2378-TCDD
12378-PeCDD
123478-HxCDD
123678-HxCDD
123789-HxCDD
1234678-HpCDD
OCDD
PCDFs
2378-TCDF
12378-PeCDF
23478-PeCDF
123478-HxCDF
123678-HxCDF
123789-HxCDF
234678-HxCDF
1234678-HpCDF
1234789-HpCDF
OCDF
logK-
7.23
7.37
7.64
7.65
7.02
7.50
7.80
7.80
7.80
8.20
8.60
5.80
6.50
7.00
7.50
7.50
7.50
7.50
8.00
8.00
8.80
Number
BAFs
4
8
4
8
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
Mean
log BAFf
8.53
8.92
8.60
8.86
6.89
7.34
7.16
6.77
6.81
6.80
6.57
5.58
.5.72
7.08
6.26
6.65
7.17
7.21
5.92
7.47
6.90
Mean
BAF?
3.37e+08
8.31e+08
3.99e+08
7.22e+08
7.85e+06
2.17e+07
1.44e+07
5.85e+06
6.49e+06
6.24e+06
3.74e+06
3.77e+05
5.22e+05
1.21e+07
1.81e+06
4.42e+06
1.49e+07
1.61e+07
8.26e+05
2.92e+07
7.94e+06
66
-------
67
-------
1
O)
o
8
CO
I
68
-------
-x-
#•
-X-
(0
CQn
0°
0.
«>
c
!*
1 •
O
CO
o
•o
\1D
\co
1
\
£
£
u
s
4-
a
"2
^o
u
CO
u.
0
O
**
O
0
o
0.
0>
o
O)
o
(0
in
K
03
I
O
o
n
°S
o
CO
in
n
is
•5
DC
O)
I
69
-------
70
-------
o
75
CO
•o
o
CD 6
Q.
IL
CO
00
6
I
a
ul
CD
c
O
8 o-
^ UJ
n %
CO
CD
in
0)
3
O)
iZ
_o
13
O
O
Q.
UJ
O
CD
O)
O
O
o
o
o
o
(0
o
•u
!«
w
o
Q.
1
8
i°
2
-1
O
o
So
0.
0>
LL
CO
CD
1
08
O
-^
O
CD
O)
* 22
71
-------
(0
CD
c
0)
0)
o
o
2 i
-o
o
3
w J
LL
DO
CD
O
Q.
CO
0>
3
O)
O)
O
O
O o
2|
UJ O
-------
o
o
o
CD
o
a
li
CD
UJ
73
-------
ffl •§
2T*
CO 5
00 8
•I
<
, «
* ^?
CD <
• S2
O CD
"5 « 03
C "g %
O S. m
•— 3
w 8 «
Pi 1
CD
oo _2 g)
D> <
5T Q.
UJS
o
UJ
74
-------
£
<
CQ
O)
3
O
0)
.*
03
(0
c o
S
CM
£ &
si
75
-------
76
-------
(0
LL
CD
O)
O
0)
.*
(0
"£ $2
OS
o
z o
§
~ »
il
Hct
« II
U)
CD
o
77
-------
u.
ffl
o
c
O
0)
o
0)
LL
ffi
O)
c
CO
0)
CM CD
O)
O
S
73
CO
D
0)
3
O)
£
CT O
O
g
2
<8
•o
_§
O
*
O
i
z
51
i
in
W* II
U. o
si
73
-------
CD
O) m
of
(0
_• CO
8 fe
O CD
«
o s.S
0) <
~Z
o
(0
m
O
Q.
(0
LL
C/)
CD
O
(0
m
O)
(
O 5
o
£
C
CO 5 o
in
ii
Ij
o>
in
CD
CD Js
O c^
9 i
oc o
UJ O
79
-------
80
-------
July 1994
VI. BIO ACCUMULATION 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, bioaccumulation factor (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 (CJ,) in water
(i.e., consistent definition).
- E, [
-------
July 1994
A. 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, t = lipid in tissue, t = whole tissue wet weight, s = dry sediment,
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).
bioaccumulation factors
BAF{ = cjcl , BAF; = cjcl =/,*AIF; w
BAF* = CyC* , BAff = Cf/Cf = /,*JBL4ff <3>
* = C(/C*. A4F* = C,/C* -/I
biota-sediment accumulation factor
(5)
oranic parbon - water partitioning
fraction dissolved /^ = (1 +000*^/10 + POOA^)'1 (6)
fraction bound to oc in water fb = 1-ffd
82
-------
July 1994
TCDD bioaccumulation equivalency factor
(BSAF )t
(BSAF U (aiF»)
B. Calculation of BAFs and TEC from BEFs
The ratio (equation 7) 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 (BAF") 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
congener in the organic carbon of surface sediments are the same as in suspended
organic carbon. The relationship between paniculate organic carbon (POO,
dissolved organic carbon (DOC), Kow and ffd is presented in equation 6. the
importance of each chemical's Kow should be evident. The BEF can be used to
calculate (BAFJ)j and (BAFjd)j. (BAFJ)jS 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} =
(BEF) . « . • (9,
so,
(BAF& * (BEF)i (BAF^ W* (10)
S3
-------
July 1994
and,
I*A**\ (MF* }< (BEF }« (***?)•* ftX
(BAF, h = — = —
because (U(fM)tcdd/(fw)i(fb)tcdd = (Kow)i/(Kow)tcdd :
(12)
A TCDD TEC can be calculated on the basis of wet tissue residue (TEC,) or lipid
normalized residue (TEC|); water concentration of total chemical (TEC*) or freely
dissolved chemical (TECfd). When bioaccumulation is to be predicted on the basis
of freely dissolved chemical (Cl?), the relative differences in BAFfds for PCDD and
PCDF congeners will be less than for their BAPs. This is because ffds for the
higher chlorinated, more hydrophobic congeners are less than ffd for TCDD. Since
the TEC is based on tissue concentration, TEC{ = TEC{d and TEC} = TEC}d. Thus
if (BAF)d)tcdd is the reference bioaccumulation factor:
(BEF\ (BAF*),^ (K^ (TEF\
ttcdd
(13)
(C*)t (BEF), (BAF?)
^
'tcdd
(14)
TECt = TEC, * /,
(15)
C. Great Lakes BEFs
Lake Ontario sediment and fish residue data (Lodge et al.f 1994) provide a basis
for calculation of BEFs. However, very few PCDDs and PCDFs measured as
sediment contaminants are detectable in fish tissue. The table below provides
estimated BEFs calculated from lake-wide average concentrations of toxicologically
important PCDDs and PCDFs in surface sediment and lake trout samples collected
84
-------
July 1994
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.
85
-------
July 1994
Table 1. TCDD Bioaccumulation equivalency factors (BEFs) 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
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-PeCDF
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
log Kw'
7.02
7.50
7.80
7.80
7.80
8.20
8.60
5.80
6.5"
7.0"
7.5"
7.5"
7.5"
7.5"
8.0b
8.0b
8.80
BSAF
0.059
0.054
0.018
0.0073
0.0081
0.0031
0.00074
0.047
0.013
0.095
0.0045
0.011
0.040
0.037
0.00065
0.023
0.001
TCDD BEF
1.0
0.92
0.3
0.12
0.14
0.051
0.0013
0.80
0.22
1.6
0.076
0.19
0.67
0.63
0.011
0.39
0.016
• Burkhard and Kuehl, 1987.
b Estimated based on degree of chlorination and Burkhard and Kuehl, 1987.
86
-------
July 1994
D. Example of TEC 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 C^ds 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 BAF^, = 7.07x105. The
toxicity equivalency concentration (TEC) for fish with with ^ = 0.09 is
approximately:
TEC09| = (7.07x105)[(0.00002)(1.0)(10.5x106)(1.0)/10.5x106 +
(0.0006)(0.8)(0.63x106)(0.1)/10.5x10e +
(0.000015}(0.92)(31.6x106)(0.5)/10.5x106 +
(0.00016)(1.6)(10x106)(0.5)/10.5x106 +
(0.0003)(0.05)(158x10e)(0.01)/10.5x106] = 14.4 + 20.4 + 1.5 -I- 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'og/s = 7.07x105), the TEC is calculated
to be 14.4 + 42.4 + 0.5 -I- 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 a
correlation between TEFs and BEFs (i.e. the more toxic congeners are the most
bioaccumulative, primarily due to slower rates of biotransformation), additional
data suitable for validating the BSAFs used to calculate the BEFs are needed.
References
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.
Cook, P.M., G.T. Ankley, R.J. Erickson, B.C. Butterworth, S.W. Kohlbry, P.
Marquis and H. Corcoran. 1994. The biota-sediment accumulation factor
(BSAF) evaluation and application to assessment of organic chemical
bioaccumulation in the Great Lakes. In preparation.
87
-------
July 1994
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.
88
-------
July 1994
VII. DERIVATION OF BAFs FOR TWENTY TWO CHEMICALS
Literature searches were conducted for data concerning Kow, laboratory-measured
BCFs, and field-measured BAFs for the chemicals. Additional information was
obtained from Burkhard (1994) and Cook (1994).
For most chemicals, the values that were found for Kow were interpreted as
described in Appendix A, but for a few chemicals the values used by Burkhard
(1994) and Cook (1994) were used to allow comparison of results that were
calculated based on the same value for Kow. The derivation of the best value of
KOW f°r each chemical is described in Appendix B.
FCM
Values for the Food Chain Multipliers (FCMs) were obtained by rounding the best
value for log Kow to one decimal digit and then using the FCM given by Burkhard
(1994) for trophic level 4.
Baseline BAFs were calculated from total BCFs given by Stephan (1993) using the
following equation (see Appendix C):
Baseline BAF =
('Ml
where
BCF-f = a total BCF, i.e., a measured BCF that is based on the total
concentration of the chemical in the water and has not been lipid-
normalized.
ffd = the fraction of the chemical in the water that is freely dissolved.
1
f j = the fraction of the biota (with which the BCF was determined) that is
lipid.
Because POC and DOC were not measured in the bioconcentration tests, plausible
worst-case concentrations were assumed. The concentrations of POC and DOC
89
-------
July 1994
were assumed to be 1 mg/L (0.000001 Kg/L) and 10 mg/L (0.00001 Kg/L),
respectively. When these values are used, ffd can be calculated as:
*..- 1
2(KOW)(1Q-6)
Calculations based on BSAFs directly result in baseline BAFs; the values used were
calculated by Cook (1994).
Field-measured BAFs
Most of the field-measured BAFs were calculated from the data of Oliver and Niimi
(1988) by Burkhard (1994); these calculations directly result in a baseline BAF.
When possible for other chemicals, baseline BAFs were calculated from field-
measured total BAFs given by Stephan (1993) using the following equation (see
Appendix C):
Baseline BAF .[_!_] [BAF " 1
'*«' '*!'
where
BAFj = a total BAF, i.e., a measured BAF that is based on the total
concentration of the chemical in water and has not been lipid-
normalized.
ffd = the fraction of the chemical in the water that is freely dissolved.
f, = the fraction of the biota (with which the BAF was determined) that is
lipid.
The concentrations of POC and DOC were assumed to be 0.000000075 Kg/L and
0.000002 Kg/L, respectively, for Lake Ontario, based on data presented by Eadie
et al. (1990). Concentrations of POC and DOC in Lake Siskiwit were assumed to
be similar to those in Lake Superior, which were assumed to be 0.00000004 Kg/L
and 0.000002 Kg/L, respectively, based on data presented by Eadie et al. (1990).
The order of preference of the resulting baseline BAFs, from most preferred to least
preferred, is:
1. A measured BAF that was calculated from field data.
2. A predicted BAF that was obtained using BSAFs.
90
-------
July 1994
3. A predicted BAF that was obtained by multiplying a measured BCF by a FCM.
4. A predicted BAF that was obtained by multiplying a predicted BCF by a FCM,
where Kow is used as the predicted BCF.
For some chemicals, Burkhard (1994) calculated field-measured baseline BAFs for
trophic level 3 (T. L. 3), but could not calculate field-measured baseline BAFs for
trophic level 4 (T. L. 4) because the necessary data were not reported by Oliver
and Niimi (1988). For these chemicals a value for the baseline BAF for trophic
level 4 can be calculated as:
Baseline BAF for T. L. 4 = (Baseline BAF for J' L 3)(FCM for T' L 4)
(FCM for T. L. 3)
If more than one equally preferred value is available for a baseline BAF using the
same method, the geometric mean of the values is the preferred value. The
available baseline BAFs are given in Table 1 .
After sufficient baseline BAFs were calculated for a chemical, the preferred
baseline BAF was used to calculate a desired human health or wildlife BAF-d%1
using the equation (see Appendix C):
BAF'd%, = 1 + (ft) (baseline BAF)
where / = f, x 100 and values of 0.05 and 0.079 were used for-f, for human
health and wildlife, respectively.
To calculate a plausible worst-case value for BAFf%t for each chemical, the Kow
of the chemical was used with estimated concentrations of POC and DOC in Lake
Superior (0.00000004 Kg/L and 0.000002 Kg/L, respectively, based on data
presented by Eadie et al. 1990) to calculate ffd, which was then used in the
following equation:
BAFJ%I
is the BAF that is appropriate for derivation of the criterion.
91
-------
July 1994
For mercury, the derivation of values for BAF-%,ior trophic levels 3 and 4 for
human health and wildlife is described in Appendix D.
The resulting values of BAF-d%t and BAF-%, are given in Table 2.
References
Burkhard, L. 1994. Food Chain Multipliers for the GLWQI. ERL-Duluth.
Cook, P.M. 1994. Prediction of Bioaccumulation Factors (BAFs) from Biota-
Sediment Accumulation Factor (BSAF) Measurements. ERL-Duluth.
Eadie, B.J., N.R. Morehead, an P.F. Landrum. 1990. Three-Phase Partitioning of
Hydrophobic Organic Compounds in Great Lakes Waters. Chemosphere
20:161-178.
Stephan, C.E. 1993. Derivation of Proposed Human Health and Wildlife
Bioaccumulation Factors for the Great Lakes Initiative. ERL-Duluth.
92
-------
July 1994
Appendix A. Derivation of the Value of Log KOW for an Organic Chemical
A valid Kow can be experimentally determined only for an individual chemical; it is
not possible to experimentally determine a valid Kow for a substance that is a
mixture, such as toxaphene, PCBs, and chlordane. A value for Kow can be
assigned to a mixture either by assigning the value for a major component or by
assigning an average of the values for several major components. The arithmetic
average of values of log Kow can be used, or the geometric mean of values of Kow
can be used.
Values were used only if they were obtained from the original authors or from a
critical review that supplied sufficient information. A Med-Chem Star value was
only of concern if the original reference for the value had not been obtained.
Because of potential interference due to radioactivity associated with impurities,
values that were determined by measuring radioactivity in water and/or octanol
were considered less reliable and were moved down one step in the priority below
values that were determined using the same technique but were quantified using
other methods.
The shake-flask technique has been reported to be acceptable only for chemicals
whose Kows are less than 4 (Karickhoff et al. 1979; Kohemann et al. 1979;
Braumann and Grimme 1981; Harnisch et al. 1983; Brooke et al. 1990). Brooke et
al. (1986) compared techniques and decided that the shake-flask technique is
acceptable for chemicals whose 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.
Values of Kow were given priority based on the technique used as follows:
KOW < 4: Priority Technique
1 Slow-stir
1 Generator-column
1 Shake-flask
2 Med-Chem Star value
3 Reverse-phase liquid chromatography on C18 with
extrapolation to zero percent solvent
4 Reverse-phase liquid chromatography on C18 without
extrapolation to zero percent solvent
5 Predicted by the Med-Chem program
93
-------
July 1994
KOW > 4: Priority Technique
1 Slow-stir
1 Generator-column
2 Reverse-phase liquid chromatography on C18 with
extrapolation to zero percent solvent
3 Reverse-phase liquid chromatography on C18 without
extrapolation to zero percent solvent
4 Shake-flask
5 Med-Chem Star value
6 Predicted by the Med-Chem program
Values that seemed to be different from the rest were considered outliers and were
not used.
For each chemical the available value of Log Kow with the highest priority was
considered the best value, except that if more than one such value was available,
the arithmetic mean of those values was used as the best value.
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 > 106. 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.
94
-------
July 1994
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.
95
-------
July 1994
Appendix B. Derivation of Values of Log Kow for Twenty-two Chemicals
The priorities of the methods for determining log Kow are presented in Appendix A.
BENZENE
The values that have the highest priorities are:
2.11 Shake-flask Karickhoff et al. 1979
2.13 Generator-column Miller et al. 1985
2.19 Slow-stir de Bruijn et al. 1989
The mean is 2.14 and is the best value.
CHLORDANE
The value that has the highest priority is:
6.00 RPLC Veithetal. 1979
The value of 5.54 was used because it was used by Burkhard (1994) and Cook
(1994).
CHLOROBENZENE
The values that have the highest priorities are:
2.90 Slow-stir Brooke et al. 1990
2.78 Slow-stir Brooke et al. 1990
2.98 Generator-column Miller et al. 1985
2.80 Shake-flask Voice et al. 1983
2.90 Slow-stir de Bruijn et al. 1989
, The mean is 2.87 and is the best value.
CYANIDE
A value of log Kow was not used for cyanide.
96
-------
July 1994
DDT
The values that have the highest priorities are:
6.20 Slow-stir Brooke et al. 1986
6.31 Slow-stir Brooke et al. 1990
6.91 Slow-stir Brooke et al. 1990
6.38 Slow-stir Stancil 1994
6.91 Slow-stir de Bruijn et al. 1989
The value of 6.91 was used because it was used by Burkhard (1994) and Cook
(1994).
DEHP
The values that have the highest priorities are:
7.14 Slow-stir Brooke et al. 1990
7.45 Slow-stir Brooke et al. 1990
7.45 Slow-stir de Bruijn et al. 1989
The mean is 7.35 and is the best value.
DIELDRIN
The values that have the highest priorities are:
4.54 Slow-stir Brooke et al. 1986
5.34 Slow-stir Stancil 1994
5.40 Slow-stir de Bruijn et al. 1989
The value of 5.4 was used because it was used by Burkhard (1994) and Cook
(1994).
2.4-DIMETHYLPHENOL
The value that has the highest priority is:
2.42 Shake-flask Banerjee et al. 1980
This is the best value.
97
-------
July 1994
2.4-DINITROPHENOL
The value that has the highest priority is:
1.50 Consensus Klein et al. 1988
This is the best value.
HEPTACHLQR
The values that have the highest priorities are:
5.27 RPLC McDuffie 1981
5.44 RPLC Veithetal. 1979
The mean is 5.36 and is the best value.
HEXACHLOROBENZENE
The values that have the highest priorities are:
5.47 Generator-column Miller et al. 1985
5.73 Slow-stir de Bruijn et al. 1989
The value of 5.73 was used because it was used by Burkhard (1994) and Cook
(1994).
HEXACHLOROETHANE
The values that have the highest priorities are:
4.04 RPLC McDuffie 1981
4.05 RPLC Veith et al. 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 mean
of the four values is 4.04 and is the best value.
98
-------
July 1994
LINDANE
The values that have the highest priorities are:
3.69 Slow-stir de Bruijn et al. 1989
3.32 Shake-flask Platford 1981
The value of 3.69 was used because it was used by Burkhard (1994) and Cook
(1994).
MERCURY
A value for log Kow was not used for mercury.
METHYLENE CHLORIDE
The value that has the highest priority is:
1.25 Calculated Med-Chem
This is the best value.
PCBs
Based on data reported by Schulz et al. (1989), congeners 8, 18, 28, 31, 52,
95, 101, 118, 149, and 153 were selected as being the most prevalent.
Burkhard (1994) calculated field-measured BAFs for nine of these, but field data
were not available for congener 8. The arithmetic average of the values of log
Kow Qiven by Burkhard (1994) for the nine congeners is 6.14 and is the best
value.
PENTACHLOROPHENOL
The values that have the highest priorities are:
5.08 RPLC Miyake and Terada 1982
5.01 RPLC Veithetal. 1979
The mean of these is 5.04 and is the best value.
99
-------
July 1994
2.3.7.8-TCDD
The value that has the highest priority is:
7.02 RPLC Burkhard and Kuehl 1986
This is the best value.
TOLUENE
The values that have the highest priorities are:
2.65 Generator-column Miller et al. 1985
2.79 Slow-stir de Bruijn et al. 1989
2.63 Slow-stir Brooke et al. 1990
2.79 Slow-stir Brooke et al. 1990
The mean of these is 2.72 and is the best value.
TOXAPHENE
The value that has the highest priority is:
4.33 Calculated Med-Chem
This is the best value.
1.2.4-TRlCHLOROBENZENE
The values that have the highest priorities are:
3.98 Generator-column Miller et al. 1985
4.05 Slow-stir de Bruijn et al. 1989
3.93 Shake-flask Konemann et al. 1979
4.02 Shake-flask Chiou et al. 1982
These values are close to 4 and the range of the four values is small. The mean
of the four values is 4.00 and is the best value.
100
-------
July 1994
TRICHLOROETHYLENE
The values that have the highest priority are:
2.42 Shake-flask Banerjee et al. 1980
2.53 Generator-column Miller et al. 1985
3.14 Shake-flask Harnisch et al. 1983
The last value is considered an outlier. The mean of the other two is 2.48 and
is the best value.
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. Techno/. 14:1227-1229.
Brooke, D.N., A.J. Dobbs, and N. Williams. 1986. Octanol:Water Partition
Coefficients (P): Measurement, Estimation, and Interpretation, Particularly for
Chemicals with P > 106. 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.
Burkhard, L. 1994. Food Chain Multipliers for the GLWQI. ERL-Duluth.
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.
Chiou, C.T. 1985. Partition Coefficients of Organic Compounds in Lipid-Water
Systems and Correlations with Fish Bioconcentration Factors. Environ. Sci.
Techno/. 19:57-62.
Chiou, C.T., D.W. Schmedding, and M. Manes. 1982. Partitioning of Organic
Compounds in Octanol-Water Systems. Environ. Sci. Techno/. 16:4-10.
Cook, P.M. 1994. Prediction of Bioaccumulation Factors (BAFs) from Biota-
Sediment Accumulation Factor (BSAF) Measurements. ERL-Duluth.
101
-------
July 1994
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. 8:499-512.
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.
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 PQCT Values of Chloro-Substituted Benzenes, Toluenes and Anilines by High-
Performance Liquid Chromatography on ODS-Silica. J. Chromatog. 178:559-
565.
McDuffie, B. 1981. Estimation of Octanol/Water Partition Coefficients for Organic
Pollutants Using Reverse-Phase HPLC. Chemosphere 10:73-83.
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. Techno!. 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.
Schulz, D.E., G. Petrick, and J.C. Duinker. 1989. Complete Characterization of
Polychlorinated Biphenyl Congeners in Commercial Aroclor and Clophen Mixtures
by Multidimensional Gas Chromatography-Electron Capture Detection. Environ. •
Sci. Technol. 23:852-859.
Stancil, F. 1994. Memorandum to M. Reiley.
102
-------
July 1994
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. Can. 36:1040-
1048.
Veith, G.D., K.J. Macek, S.R. Petrocelli, and J. Carroll. 1980. An Evaluation of
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
on the Sorptive Partitioning of Hydrophobic Pollutants in Aquatic Systems.
Environ. Sci. Techno!. 17:513-518.
103
-------
July 1994
Appendix C. Derivation of BAFs for Organic Chemicals
PREDICTING BCFs FOR LOW Kow CHEMICALS
The procedures discussed above for estimating bioaccumulation factors were
developed for highly lipophilic organic chemicals for which accumulation is closely
associated with the lipid of an organism. In order to extend these methods to less
lipophilic chemicals, it is necessary to accommodate cases in which accumulation
is not dominated by partitioning into lipid. Because weakly lipophilic chemicals are
not significantly accumulated via food, bioaccumulation factors are equal to
bioconcentration factors. Thus, this discussion can be restricted to the question
of predicting steady-state bioconcentration factors based on Kow or extrapolations
from other organisms.
An organic chemical accumulated by an organism associates with various
components of the organism; some chemical in the organism is dissolved in its
water; some is partitioned into membranes, fat deposits, and other lipid material;
and some may be bound to various nonlipid organic material. An organic chemical
with Kow *• 1 will distribute among water and different organic phases with similar
concentrations, so that a steady-state bioconcentration factor will be
approximately 1 in the absence of metabolism and significant growth dilution.
More hydrophilic chemicals will also have bioconcentration factors of the order of
one because water is the predominant component in an organism. More lipophilic
chemicals will have larger bioconcentration factors because there would be an
increased concentration in the organic components relative to water.
Bioconcentration factors should therefore be described in terms of two
components: an aqueous portion that is approximately 1 and an organic portion
which is the product of the amount of organic components and the affinity of the
chemical for organic matter relative to water. Lipid will be the most important
organic component unless there are important specific binding reactions or very
low lipid content. An approximate general equation for bioconcentration factors is
therefore:
BCF = 1 + fg • A
where 1 represents the contribution of the organisms aqueous phase to the BCF
and f|- A represents the contribution of the organic components, the fraction lipid
f, being the quantity of the dominant organic component and A being a measure of
the affinity of the chemical for lipid relative to water.
104
-------
July 1994
Prediction of BCF from K^»,
A predicted equilibrium BCF that is applicable to weakly lipophilic hydrophobic
chemicals as well as strongly lipophilic chemicals can therefore be based on the
two-term equation:
BCF = 1 + ft'Kow
where the octanol:water partition coefficient Kow is the prediction for the affinity of
the chemical to the lipid. For the methods for highly lipophilic chemicals discussed
above, the aqueous term in this equation ("1") can be ignored because it is so
much smaller than the organic term, but for weakly lipophilic chemicals it cannot
be ignored. This equation omits the finer details of chemical distribution within an
organism, but it provides a useful approximation than can be applied to a range of
chemicals. Because exposure via fish consumption contributes little to the overall
risk from weakly lipophilic chemicals, the errors in this approximation are of little
real consequence for these chemicals.
Extrapolation of BCF among organisms
For highly lipophilic chemicals the aqueous portion of apcumulation is usually
negligible. Therefore, for these chemicals lipid-normalization of a bioconcentration
factor provide an estimate for the affinity of the chemical for the lipid fraction:
BCF, - 1
A ~
A BCF for organism "2" can then be estimated by multiplying this lipid affinity by
its lipid content and then adding 1:
BCF2 - 7 + ft-2 • A
Again, this treatment is approximate and ignores various aspects of accumulation,
but for weakly lipophilic chemicals approximate values should suffice because
exposure via fish consumption is an unimportant source of risk.
105
-------
July 1994
Appendix D. Derivation of Values for BAF for Mercury
The following rationale is a revision of that used in the derivation of the GLI BAF
for mercury dated 3-3-93.
a. 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. Watras and Bloom (1992) found that
with mercury, 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 shows a consistent predator-prey factor between
fishes. Thus the model used here will provide for bioconcentration at
trophic level 1, and biomagnification at trophic levels 2, 3, and 4.
b. The BCFs for inorganic mercury and methylmercury will remain at 2,998
and 52,175, respectively.
c. 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) + (.83X2,998) = 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.
d. 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.
e. A variety of studies have found predator-prey factor increases in total
mercury in fish from 1.2 to 15, with a mean of about 5.
f. Use of these factors results in:
(11,358)(2.00) = 22,716
(22,716)(1.26) = 28,622
(28,622X5.00) = 143,110
g. 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:
(28,622X0.975) = 27,906
(143,110X0.975) = 139,532
h. It appears that BCFs and BAFs based on whole body and edible portion should
be similar for mercury. Thus for a specific trophic level, the human health and
wildlife BAFs will be the same.
106
-------
July 1994
i. This derivation indicates that for total mercury in the water column the values
of E4f/%for human health and wildlife should be:
Trophic level BAF
3 27,900
4 140,000
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
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 used to
derive the criterion will be 62,400.
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.
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.
107
-------
July 1994
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.
Watras, C.J., and N.S. Bloom. 1992. Mercury and Methylmercury in Individual
Zooplankton: Implications for Bioaccumulation. Limnol. Oceanogr. 37:1313-
1318.
108
-------
00
Q
o
g
H
X
a
D
o
H
9
(P
•4
H
•J
u
H
K
Pi
O
*
g
H
%
H
r?
H
CO
H
•J
n
o
o
TJ fl)
0) C
tO -H
rO rH
x> a)
to
0) RJ
>H XJ
re s
C h "
re re g
* 0) "^
TJ W n
-H fl) O
axi'^
-H 4J -5
>H 0)
Q) E
4J 3
re to
3 C
O
O U
>1
ID rH
TJ C
>H O
re
e» to
0) C
>H re
C 3
O fl) -^
2 J r.
^.-H g
c ^
a) -
TJ 4J_
0) ro-~
to ? fa
» ^
XJ 0) CQ
XI
0) 4-J -
rH 4J
re re
U XI
•H 4J
8*
rO
o o
re C 0) C
•H • O W
«. -H -H 0)
^ rH 4J O
re - re TJ
rH U fa rH
0) -H < 3 ffl
> E CQ e -H
0) fl) 3 X! •
•H X! fl) O 4-> ^J-
O C U
rj -H re 4J o) -H xj >
CL > in XJ Q)
O rH rO -rH
IH O XJ 0) X!
4-> in x: 4-> o
tO Q) 4-> rH -H
m x:
O TJ I
>1
10 fl) _
Q> x: o
0)
>H 0)
re a)
u
0) . . _
rH >
XI <*H fl)
re O
•H C 4J
h re
fl) E C
rH 4-> 3'H
0) U XI
re to
»H »H -H
m o
I i ^H »l I ^J
c u u re
10 O-H (8 X!
•H -H X! 4J -H 4->
xi 4J a, to h
4-i (8 O 0) fl) re
C 4-> 4-» -HO
-H C 4J h -H
0) in re O XJ
to u o x;
fa C «M 4-> >i U
CQ O s fl> -H 4->
10 3 rH rO
Q) fl) fa rH «J 3
XI XI < re 3 CP
EH 4-> CQ > O1 m
4-)
re
TJ
0)
4J
«M
o
0)
4->
re
TJ
0)
to
re
TJ
0)
4->
re
TJ
ft
3
c
0)
2
0)
re
XI
Q)
rH
XJ
re
4J
to
«H
re
TJ
to
•H
4J
c
to
fl)
3
rH
re
o
re
0)
o
i
rH O
3 TJ
o o
U XJ
4J
4J 0)
re E
<
TT CQ •
re
rH H 4->
fl) tJ re
> O TJ
fl)
re
u
TJ
o
o
fa
fa
u
CQ
TJ
0)
u
CQ
TJ
a>
p
u
Q)
0)
4->
4->
o
Xi
XI
4->
-H
to
re
0)
re
u
fa
re
J*
* XJ
CQ
£ rH fa rH
rj 4J o) < a
•H -H W -H E TJ
XJ C «W CQ 4J O fl)
O)
XJ
in
0> 0>
to • ri3
O 0) CQ
H E rH
to xj <
TJ fl) 0) re
O E X! 4J
XJ -H EH
4-> 4J 0)0
0) 0) X! •
£ E '4J H
o to
W Q) C £
fl) T5 > X! U
4-> a> cnfa
O h fl)-H
4J fl) O
00) X! XJ
O- O fa
0) O U
h 4-> O 4-> CQ
re W 5
re 4J 4J TJ
0)0) tO 0)
««H 0) JH
O X! 3
4J 10
4J X! to h re
4-> C re 0)
TJ re c
fl) Cn-H c
>H r» 0)-H
3 t 4J X! >i
TJ fl) 4J rH
0) TJ E a
O O 0 10 -H
OX! fl) -H 4J
EO)roE
to
0)
x; a>
O XI re E XJ
tw rj Q) TJ TJ V
tOfllrOCCTJ O >H fa XI TJ •
fa TJ rH -H -H CnE 0) < O fl) rH
< 3 re re to c o cnx; CQ c re
0)rHXJXJ5>ilW-H TJ rO-H
aihrooo H P E
C 3 O £ aT5 3 fa rO O XJ 0)
•HTJ in in u -H ai < MHOX:
Q) U tO S 5 rH-HO) fa Q) 0)
WO? fl)3h>TJU3hOJ
XJ CU4-> ror04J TJhin (0 3 O
rOXIX! >> 0)hTJ>
0) h Xi 4-1 4-) **XJ tTTJ 0) Q) tO U
X!34J ^ C «w4JQ)rHO
P O fafa£fa-HO)0)UrHrO JS TJ TJ
OlrHEO)
-OS 4Jr(-H
cn aippaipfaoio > in CD
>ihUU-HU 3E>nrOO4J
0)X>3-H-HrH.Hfl)rH O EXJ
•H tDTJTJaTJXJrOfl) TJ O
re re fl) h K 4J >H
4J o E a a-- ;.,
•H 3 C
•H fl)
X! XJ • • •
4J 0) >H O 4J
£ IH h «w O
0) to 3 fl) C
> m a> fa to
0)
TJ W
C-H
3 XI
O 4-1
0) 10
>H 4->
re -H
en
O -H
o *o
o
•-I 4J
c
TJ re
c u
re
O -H
to
0)
3
C rH
re
C-H
a) to
Q)
3 0)
4-> 0)
0) h
X) XI «H
4J O
10
O
0)
4J
CQ
TJ
.^ o)
4J TJ
•H C
O
JH
TJ
rO
-H
O
0)
fl)
re
O
re
u
u
re
>H
o
c
o
TJ 0
o to
0 -H
0)
0
o
0)
>H
a
a)
c
re
XJ -
O 4J XI
4J 4J
>H
TJ Q) 4->
0) 4J
TJ
3
o
)H
3
O
XJ
m
to c
fa-H
0)
TJ
to
ai
m
4J
to
0)
x>
0)
4J
Q)
re
to
O TJ
rH C
0)
Q) CX
x; a,
H <
-------
Tf
o>
O)
J*»
.>
3
->
~*J
r"l
(0
B
•H
O
0)
TJ
fl)
w
O *
H -^
4-1
41-!
o 0)
* >
Q)
tJIrH
0
J U
•H
0) £
£ a
•P O
U
O^-P
•H Ll
•o o
C iOt
XI «H
•C
0) -a
> 14
•H (0
V4 £
a> ,*
•o u
3
0) CQ
14
Q) >i
* xt
to c
rH Q)
«fl >
O-H
•H O>
B
Q) £
£ 0
U fa
U 0)
•H £
C -P
(0
tr tr
M C
O-H
to
rH 3
r-l
(0 C
0)
h £
0 4J
«M
•o
to c
£ <0
0
fa 4J
•H
a) tr
£-H
H TJ
Q)
£
u -P
w
Q
«M£
^
l-o
Ui (0
0) 4J
4J fa
<
in CQ
•-i
<0 0) t,
3 £ o
CTJJ S
0) ^
to a
rH (0
(0 (LJ
O <1> i]
, ,Z. **^
"P S °
B (0 rn
O 10 c
fM
3
U Q) o
£ P
(0 JJ M
Li
^ W n
2*H(-
658-
CQ fa n
a x
X at
«M W
O fa ^
t-> *
0) CQ o
3 a«
rH S
n «w O
0
Q) 0)
£ Q) 3
EH 3 rH
rH (0
• >
fa .p
O 0) (0
CQ £ £
H 4J
•o
0) VJ
4J . o
U rH
s:
0
b,
i.^
ii
^^
\-
u
CQ
I
||
'fi,*.
&T
•5
OQ
II
f_
5"
"S
op
Q)
c
*•*
*H
fl\
M/
(n
WJ
IB
•v
m
**<
•d
0)
to
(0
XI
to
fa
**>
***
CQ
•o
a)
4J
U
•H •
•O 0)
0 rH
rH XI
a<
•* (0
01
9t 0)
»H fc
'-'lO
>J >i
O 0)
O £
U -P
>.£
Xt U
•H
CM £
:*
Q)
rH V4
XI O
10
O-H
•H O*
•o
0) 0)
^ V4
aro
to to
fa fa
< <
CQ CO
CQ
0)
£ C
EH 0
>-l
0)
•H
rH
O
>1
X!
•c
a)
4J
M
0
a
0)
M
(0
4J
(0
T3
O
)H
-l
3
CQ
>i
XI
T3
a)
4J
10
rH
3
U
rH
(0
o
0)
U
0)
5
(0
fa
<
CQ
•o
0)
b
3 •
10 ^
A CO
Q) 00
B a.
H
0) ^
£
4J-H
B
«W -H
O-H
z
.p
W TJ
0 C
f. (0
-------
o
o
•c
0)
Ll fa
3 <
U) CQ
(0
0)
as
o
*.
1 1 0 1 1 1
1 1 0 1 1 1
i i in i i i
^
in
o o o
^,
fa
<
CO
CQ
o o o
o o o
i i * * - i
1 1 O O O 1
I \ t^ r4 o i
1 1 0 CO O 1
». *. ^
O> CO rH
CM
W
fa S
< U
CQ &^
•a x
0)
4-J 1*4
O U
•H CQ
•o e
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
a)
CU
s
o
fa
X
fa
u
CQ
O
S3
U
fa
CO 0000 rH
CO O O O O TT
rH O O O O r»
^ ^ ^ ^
O O O O
VO VO VO VO
Tf *}• ^- Tf
>...«...
CM CM CM CM
O rH rH rH rH O
• . . . . •
rH r» r> r- r- H
^ ^. ^. ^. ^. ^
O> £1 rH inininm oo
O O
•J !£| CM inininm CM
0
0
o
^
1 0
1 0
1 O
^
o
o
rH
O
0
o
w
1 0
1 0
1 0
1
*r
f>
H
1 1
1 1
1 1
O
1 0
1 0
1
o
0
o
».
f»
H
CM
1 t-
1
1 VO
CM
1 rH
1 01
1
1 VO
1
1
1
o
o
o
•b
o
o
CM
^
1-1
CM
1
1
1
O
o
o
«.
o
VO
*9*
»
rH
CO
•
in
rj
•
in
l
l
i
1
l
l
i
o
rH
rH
».
n
n
VO
CM
O
•
rH
CM
Tf
•
CM
1
1
1
1
1
1
1
1
1
1
CM
ro
0
H
O
in
•
rH
1
1
1
1
1
|
I
O
O
o
•.
0
m
CM
«.
rH
O
O
0
^
o
n
r>
•.
rH
00
•
in
VO
n
•
in
o
0
o
^
o
t-«
in
^
CM
O
O
0
•.
0
CTi
in
i
i
i
o
o
o
^
o
. CO
Tj-
>.
in
CM
.
o
rH
n
(^
•
in
JD
O
o
o
».
CJ*
H
1
1
|
1
CM
H
CO
*
in
o
o
H
^
CM
rH
rH
•
H
„
O
•
^"
O
0
rH
V
in
CO
o
0
VO
•»
&\
n
l
l
l
o
o
o\
^
^*
o
•
H
<*
VO
•
n
l
l
l
1
l
1
1
CM
n
in
^
o\
CO
rH
|
1
1
1
1
1
,
1
1
1
1
t
1
1
1
1
1
1
1
1
00
rH
O
•
rH
in
CM
•
rH
0)
<0
>• u
D
0)
c
0)
N
c
V
to
0)
10 CO W
•DEC
*•( E fl ID
O <0 M-H
rH D>4J 0
U
enzene
jQ
O
>_l
0
J2
u
0)
•o
•H
c
Q
^1
u
H
Q
0
1
^i
^
^-
C
-H
Ll
•D
rH
0)
•H
a
ethylphenol
E
•H
Q
1
"d1
•»
CM
i trophenol
c
•i-t
o
If
^
CM
O
rH
jC
o
CO
.p
a
0)
orobenzene
rH
jC
o
(0
X
0)
X
oroethane
rH
JC
o
(0
X
0)
X
S
u
X
m
(0
0;
Q)
c
(0
•o
c
•H
h-3
^
V4
3
U
Li
Q)
2£
0)
T3
-H
D
O
rH
U
0)
c
0)
rH
0)
-------
u
o
o
o
*.
o
o
o
rg
H
,
|
I
1
1
|
1
O
O
o
».
0
o
VO
..
^f
rg
CO
1
1
1
1
1
1
1
O
o
•»
rg
o
o
o
•k
in
00
eg
VD
1
1
1
o
o
o
o
in
00
f^
i
i
l
o
o
o
«.
0
o
o
..
^
p^
rg
rg
1
1
1
1
1
1
O
O
o\
..
o
n
in
fM
in
o
0
O
O
•h
0
o
VO
^
CM
41
1
1
1
1
i
0
o
».
in
fM
fM
"o
O
T
».
rH
n
1
l
l
l
l
j
I
0
o
o
«.
rH
rH
iH
1
1
1
1
1
|
1
m
f)
n
fM
o
o
TJ
to
a
x:
x
iH
3
CQ
•
to
0)
£
O
rH
x:
u
3
E
a)
JQ
o
4J
fa
<
CQ
TJ
0)
^i
g
to
(0
0)
E
TJ W
•H 33
Q) <
•H CM
C
O
•H
4J
(0
3
cr
0)
0)
X!
4-)
O^
C
•H
to
^* ^J
^^
— H |
f>4
^C *i ^
CQ g
TJ «>
0)
3 H
to
10 ^
0) „
E "
1 5,
TJ t
rH n
to
1
X
H
O
IH
to
H
0)
c
0)
IT
c
o
0
CQ
U
CM
4J
C
0)
rH «
«0 >.
> rH
0) Q)
H >
OVH
4->
0) U
to 0)
o a
^ E 10
u- <»
!H * ^
^j -
- H "4*
S O O^
ijj^ _ VH o\
iH
ID —
0)
3 T3
rH H
, (0 10
> xi
0) co X
• •••••• • CQ
3 U
-5 OU
rg
^
o
.
in
rH
O
c
0)
x:
a
o
o
rH
x:
o
(0
c
0)
&
vo
CM
rg
o
.
r>»
Q
O
U
EH
CO
,.
[>»
«.
n
«.
fM
rH
fM
t-
.
rg
Q)
C
Q)
3
rH
0
EH
tH
m
f)
.
^
0)
C
0)
a
(0
x
o
EH
tH
O
O
.
^T
0)
c
0)
N
c
Q)
O
O
rH
X!
u
•H
to
EH
1
..
fM
«.
H
rH
CO
4*
.
fM
0)
c
Q)
rH
>»!
X!
-P
0)
O
l_i
O
rH
fj
O
•H
to
EH
..
to
0)
TJ
O
U
IH
O
to
c
o
•H
.p
(0
c
(0
rH
CU
X
u
^
(0 O
IH
Q)
to to
3-H
to x:
O JJ
^i TJ
rH Q)
X! ±J
(0 (0
X) b
0 4J
L4 tO
a c
o
TJ E
rH C)
3 TJ
O
4J
E a
tO -H
•H ^
rH O
o to
XI 3
10 C
4J 10
Q) E
£ —
•
IH
(0
g
O
l_l
IH
TJ
(1)
4J
m
rH
3
0
rH
10
U
to
10
0)
3
rH
(0
J>
to
•H
x:
EH
x
0 3
CQ
to
C ^i
n xi
Q)
E TJ
(U
U 4->
•H (0
to rH
4J 3
0) U
E rH
O (0
0) U
IT
to
Q) fa
to <
10 CQ
tO TJ
Q) 0)
2 lj
rH 3
(0 tO
£> (Q
Q)
0) E
to
0) TJ
x: c
EH (0
u
U
0) fa m
c *
o 0 -
N X! 5
C 4J 'H
0)
xi o m
o to **
to ^
O TJ •3 •
rH C 0 t
r* irt w
0 >HrH
•H >. Q)
H 0 W >
jj o *H n)
1 . XirH
4 <- H
- O
fM 10 -H
- -H ' X3
H •"' CU
CQ * O
to U H to
O EH -P
(LJ «l* 10
*»J .
fM ^ M
^r tH O
in ^* IH
H to .
* O CO
r~ IH Q) m
i *^i iw * *
n ^ ^
(ij o ^H _j
^j ?M f^
0>*H it
tT ° *
fa rH CU-^
m ° «-T
03 Q) W
XI JJ rH
TJ EH C
0) W »^
>H O ^.
3 -«H ,n
to o
10 £ rH
O rH U ^
E 0) fa ^.
1 > ^
TJ fl) Q) In
rH rH XI rH
0) 4J 1
•HO f^
IH -H TJ 5^
X! C s^
<0 Cu 10
O (4.1
TJ w n n
Q) *J •
•P «-»,,
•" ** "VT
rH «0 tfl It!
3 -H <
o to ffl
rH -H m
10 TJ
o x: H Q)
U Q) w
^•H > 3
rr £ 0) W
O\ > rH (0
O (U
H - 0 E
^- C-H -^
•H x: TJ
TJ Cu Q< fl)
W rH O 4J
10 3 to O
x; o 4J-H
.X W TJ
iH MO)
3COM
CQ -H
-------
O)
MMENDED BAFS FOR THE DERIVATION OF GLI CRITERIA
O
CJ
H
K
••
CM
H
i
are recommended for use in derivation of human health
0)
rH
X)
10
4J
0)
X!
4J
C
-H
C
0)
. >
•H
O1
to
fa
rtj
CO
0)
x:
EH
for use in the derivation of wildlife criteria are given on
TJ
0)
TJ
0)
E
E
0
O
0)
to
fa
(flj
CQ
•
10
•H
J_l
0)
4->
•H
J^
O
D>
•H
(1)
4J
C
O
TJ
0)
to
to
xt
0)
m
•H
rH
TJ
rH
•H
O
<*H
TJ
C
10
x:
4J
rH
1C
x:
c
1C
E
3
x;
o
c
Q)
>
•H
^
to
fa
<
CO
rH
rH
•
0)
1C
a
4J
X
0)
c
Q)
x:
4J
biota.
o
-H
4J
10
3
tr
1C
0)
«j
<4H
O
a)
3
to
to
•H
4J
0)
4J
HH
O
seline BAFs" are based on 100% lipid and on the concentration
in the water. For each chemical, the baseline BAF for
s table is the most preferred value available for the chemica
farthest to the right for the chemical in Table 1. A BAF is
m
?
«
rH
1C
O
E
Q)
x:
o
o
•H
C
10
)tl
o
c
10
J_l
o
fa
rH -H
10 x:
O 4J
-H
E C
0) -H
x:
o c
0)
•o >
0)-H
rH
o *•
to
tO rH
•H 0)
TJ >
0)
SHrH
rH
Q) O
0) -H
rH X«
•in Qi
O
IH Jj
0 4J
0)
3
rH
10
>
0)
x:
4J
^
•
Q)
•
•H
..
rH
0)
rH
X)
1C
H
C
-H
•
rH
r-t
Xt
o
0)
ID
EH
to
J
O
rH
0)
EH
rH
10
O
•H
0)
O
0)
c
0
l_l
0
•4H
H
0)
xl
10
EH
C
-H
c
0)
>
tr
Jj
o
c
1th are for trophic level 4. BAFs in the table that are not
percent lipid. The GLI human health guidelines currently
sume aquatic biota that are in trophic level 4 and that the
5 percent.
ic in c
0) O 10
x: c O-H
0
C >iT3
1C TJ rH -H
E 0) C Du
3 tfl O -H
XI fa +J a
•H ft, 1C
D>CQ x: a)
•P rH
tO 0) X)
fa C >i 10
<-H <4H U
CQ rH -H -H
0) O rH
>-H to a> a
rH 1C CL (X
< XI ID 1C
0)
X!
c
-H
to
3
TJ
0)
4J
1C
rH
3
o
rH
1C
O
to
1C
rH
0)
0)
rH
o
-H
t
4J
1C
•o
•H
a
•H
rH
4->
C
0)
O
0)
a
^j
o
fa
<
CQ
•c
Q)
>
rH
O
to
to
•H
*o
t>
rH
V
Q)
l_l
<4H
0)
H
• •
c
o
•H
4J
10
3
tr
0)
CO
(0
0)
3
rH
1C
TJ
0)
4->
10
-H
4->
to
0)
C ••
•H C
to o
3-H
4J
TJ 1C
0) 3
4J O1
10 (U
rH
3 0)
o x:
rH 4J
1C
O TJ
10 10
1C
e
rH
a> m
Q) O
rH • •
O
o
•H ||
"D.U
o o
rH PM
4J
TJ
4J C
1C 10
at?
•H E
rH
OJ
C II
0)
O U
rH O
Q) O
a
m o
o o
i
•*4J O
4J 10
•H T> M
D> 0) 3
-H TJ O
•O C O
3 10
4J 0
C r4 r4
10 O
O 0)
•H M C
KH 10 O
c o to
O
0)
•H O
tQ H
Q) c a
c m
o x: a)
4J x:
O 4->
0) 4->
TJ 4J 3
Q) (C O
Q) X)
10
CP
TS
C
3
O
0)
10
to c
fa-H
< £
CQ 4J
— C
10 10
M
10 XI
0) ^i
x: o
Hii i
^*
x: E a
4J 3 E
C-H
to
in a) 4-)
0) iH O
rH O C
to
fa
<
CQ -P TJ
to
0)
o o
-------
0)
•H
rH
•o
rH
-H
Jg
C -P
fl O 0)
(0 fl) fl C iH
H E -H O fl
fl O in -P 0)
O 10 fl> »H -P
•H 4J 10 fl fl
E fl)-H >H x:
0) E in fl) W -P
x; 3 o IH fl)
U U 0) -H tQ
C fl) U 0 fl)
»H O 1-H -H H 6 •
•H rH £ O >1 IT
C TJ O 10 C C
O rH IIH -H fl 3
•H O XI X! TJ i
rH fl) -H .p -H X fl)
•O rH M > fl) X!
rH X» fl) -P -H 4J
•H -H TJ fl 13 H
i to x: c o c
to a) .P -H fa a>
box: x:
o a-p tfl •
10 in O 3 C n
TJ-H O Z fl fl
fl) H
> 4J 0) (0 Q)
•H -H 10 • X» >
U ft, <* Q)fl)
a> x: •< -IH .H
TJ CP0 TJ >i-H
D C rH rH O
CP O » fl -H -H
ex: r» E fl) x:
•H 4J 0 fl rH a
fl) rH rH «W O O
x» < fl) to x: in
> rH W fc -P
>i 0) 0) O
rH . rH > W C
.p Q) 0) •••H -H
C >H U rH U) 0)
0) 0) -H 3 X! Q)
in x: xj o c 4J M
M a-H a) n
3 TJ o x: CP M
O fl) M a O (Q
0) C iH 10 rH
H Q) C 4J fl) rH 3
fl M -H -H fl) TJ
Q) in O > fl
fl M 0) O Q) 0)
-H a IH IH arH to
r4 fl 10 fl
fl) fl) >i O
4J H 4-> rH fl >H ^
•H fl fl c x:
w x: o o a,< fl M -P 4-> fl)
4J 03 4J OJ O rH
-H o x: c E to x:
o1 u > to a
O -H -H -H fl) O
IH 10 -P CP C X! »H
fl) fl (0 .p -P
4J TJ 3 fl) CP
fl C CP W iH C C
5 fl fl fl O -H -H
to
fl)
c
•H
rH
Q)
•o •
•H •*
3
Q)
fl) >
IH Q)
•H rH
i— 1
TJ 0
rH *H
•H JS
* a
0
IH (H
iJ -P
O
TJ
Q) C
x: fl
jj
n
fl)
tO rH
3 Q)
fl >
U Q)
fl) rH
X)
U
*H x:
aa
•H O
rH >H
4J
4J
ex:
Q) 4J
P 8
IH XI
0)
O
Ot IH
•
t** 4J
c
JH fl)
o u
fl)
c a
Q)
^ o\
•H •
Cn t*~
0) >i
r4 H
0
10 Q)
ft* CL
< 10
0
0) rH
M-l 4>>
•H C
rH fl)
T5 IH
•H >H
•H 3
s u
to
fa
1^2
0
^>
rH
0)
^
0)
HJ
O
•H
x:
o
o
(^
EH
TJ
•H
Q
•H
rH
-P
e
a>
u
H
0)
Q
o>
•
t^
O
£>
rH
fl
4J
O
-P
O
0
^
O
4*
o
w
•fl1
•
to
to
•H
TJ
•
iH
o
o
o
«.
0
o
g\
«.
r*
m
0)
c
-H
rH
0)
to
fl
0
o
0
o
«.
0
o
o
>.
o
o
rH
10
fa
r-
O
.c
rH
fl
4J
O
O
o
o
[-X
rH
«
fS
•
to
•H
TJ
•
^i
.
00
VO
0
o
0
«
00
^J>
o>
o
o
0
».
o
0
o
«
(N
rH
0
o
0
00
in
VO
o
o
o
00
o
at
^
0
o
0
..
o
o
in
>.
rH
rH
to
m
u
o
o
o
^
o
VO
rH
0
O
O
«.
O
fS
VO
o
o
0
«
o
in
00
»
f^
o
o
0
cT
N
n
0
0
o
o
rH
u
0
o
o
«.
o
o
(»»
^
in
rH,
0
O
U
H
1
00
«.
(•»
«.
n
»
N
•
^r
*"
2
•™*
11
*^
MI
w
f*
f*
^
a\
***
irt
^
to
to
*H
™
^1
i
rH
AI
V)
• ,
H
HH
""
j_4
fl
0*
£>
1 1
•P
.
. \
4-*
• ,
H
ft\
Q)
^
fi
>H
(I)
w
fl)
tf
•5
•*
^-*
g
IH
^*
^^
* 2
^. *•
3
Q) S
•H s.
XI
.H
0) n
fl «H Jj
to o
C 1 fl)-H ^
•H TJ > X! S
a-H-H a
iH a CP O q_i in
3 -H MO
0 rH X! 4J (w
tO O (t, o
fl) -H w <;
M r4X! 0 § s
O fl > vw o
IH in ^
fl) kfO fl)
EH tQ « H jL fl
D fl) r» fl 3
ax: • xs w -P
EH r- M fl B
O W 3 E n
IH « -H 0 B
VO ^ rH
in c >i^ fl)
VO TJ fl XI >
•CO) .0)
f» fl E TJ oj rH
fl)
TJ - O 4J fl) O
c in -H fl ,«j -H
fl 4J rH n X!
-<» 3 fl a
v* E 0 £ 0
^ x: «-< ^ v<
00 tO 4J tO c -*-1
• fl) -H O .H
X> fl 0 n^ O
» fl o m IH
0 .p 0) 0 .
t*- x: •• vo °°
« e +j o
r« -H o m n
TJ O n «H
*H «» C ^
O >i fl O > fa £
•4! *H 14! IH CP O
0 4J 0 O O fa
O rH
O"* Q) t7^ r*4 (0
o a o < fl
rH 10 rH pa O
0) O -P
TJ W TJ TJ -P
0) fl) 0) TJ
w ^ > w TJ C
3 -P rH 3 C O
co IH o to o a
fl a) en fl a to
0) E tO 0) CO fl)
E to -H E 0) w
TJ W >H
*O rH Q) M O
Q) rH >iXi O U
•P fl rH E-l O
B E fl) TJ
rH 0) fl) TJ rH
3 W • rH 3
O TJ IH 0 3 O
rH C 0 O ^
fl fl •> O >
0 TJ - 0
- 0) 0 EH 0
TJ <0 N 0 O 0
h fl)-H r- o -
fl > rH - 0
XJ-H fl M )H 0
X S e in o f*
M Q) M IH ^
3 rH O *H (*1
0 fl c O •« m
u
-------
^
O)
O)
^
3
-)
O
IH
rq
0)
10
EH
C
-H
TJ
0)
4J
j^
O
Ou
0)
TJ
(0
x:
* &
3 .
00 vo
^j II
10
fO
• CT
t_i o
•H rH
TJ H
C 0
«0 IH
n M
•
h> rH
rH 0)
IH 0)
O rH
to u
S -H
CJ «C
t^ 0<
o
Q) >-)
^H j j
1 \
Jj
C 10
0)
0) 10
£ d)
4J >
Q)-H
ja >
0)
10 rH
-H 10
CO TJ
• c
m 10
rH
C
IH -H
o a
rH
38 3
U U
Pu tO
IP
TJ
0)
10
3
U
rH
10
u
to
l_l
Q)
C
Q) •
C
O rH
U Q)
£»
CO 0)
0 H
U
4J-H
C fi
<0 04
rH O
10 H
Q>
M C
CU-H
0) 0)
H H
0 10
E
fi
Q) U
-p x:
i_i
o -
IH to
0)
to >
b-H
< >
(Q 0)
rH
TJ «3
a)
»H TJ
3 C
to 10
10
0) C
E-H
a
O 3
u
c to
10
Q) H
e o
u
•H TJ
4J 10
0) £
6 «^
O H
0) 3
U CO
•o
0)
X)
n
rH
0)
0)
^
U
•H
£
a
o
^4
4J
^4
O
IH
(t,
^
00
0)
c
•H
rH
0)
to
10
XI •
^*
0)
JC rH
4J Q)
>
4J Q)
10 rH
4J U
•H
(0 £
TJ a
c o
0) H
E -P
E
0 M
u o
Q> IH
^
4J
^ 10
t x:
o\
rH C
-^ 10
r]
tX 4^
O
O >H
U 0)
x:
D-H"
Q £
U
EH 0
1 ^
CO 4J
*.
t** IH
- O
n Q)
' M U
CM O C
4J o)
M U H
O 10 0)
fc< ^M ^M
0)
a
•
t^
^
•H
3
•
C
10
a
0)
CO
•
U
•
u
o
4J
E
3
TJ
C
10
H
O
E
0)
33
•
<*
g\
a\
H
•
£
•
a<
^
^
o
o
u
-------
C o en
O Xc 00
••1 iH • O Cn
4j oj w
O OJ 10 H O
O
1 e
« m
-^ 0)
.
(Q
10
•H
•o
>
rH
OJ
OJ
M-t
*"* oo o
*K §
fa X
m «°
W f^
""^ OJ
en
O
en
•
O
o
0
o
o
in
OJ
o
0
0
in
r-
OJ
en
O
en
•
o
0
o
o
o"
in
OJ
o
0
0
in
r-
OJ
en
0
en
•
0
o
0
0
o
in
OJ
o
o
o
in
r^
OJ
0
o
0
*
H
CO
n
CD
0
O
H
en
O
O
o
o
o
o
CT>
VO
00
en
en
CM
o
o
o
o
o
o
^
o
in
in
o
o
o
fc,
r-
a\
in
o
o
Ol
in
o
o
o
cr>
o
vo
o
o
o
o
o
o
fe
o
o
o
*l
in
o
o
o
^
o
vo
o
in
o
o
in
VO
O
O
o
4k.
00
fS
o
in
o\
o
vo
OJ
fa OJ
CQ 0)
0)
c u
OJ O
10 O
<0
CO
OQ
o o o o
o o o o
o o o o
^ ^ «l ^
o o o o
o o o o
in in in in
in in in in
o
o
o
^
o
o
o
*l
o
o
o
o
o
^
o
o
CM
OJ
o
n
O
o
o
CM
o
o
o
^
o
r-
m
o
o
o
o
o
00
o
o
o
00
OJ
in in in in
• * • •
in in in in
oo
OJ
vo
^1
in
OJ
OJ
o
in
vo
r>
•
in
n
in
o
•
•^<
en i in
vo I oj
i
r> I rH
10
* U
O) -H
i- OJ
> e
D
OJ
OJ
N
C
0)
n
OJ
c
10 10 10
•o i c
rl E <0 0)
rH &>4J O
u
benzene
o
o
rH
jg
U
0) Q
•O O
•H 1
C
«o *•
X ^
CJ ^
c
}4
•o
rH
0)
-H
Q
methylphenol
•H
Q
1
^
fe
OJ
ni trophenol
-rl
O
1
^*
^
OJ
O
rH
JB
0
10
•p
a
OJ
X
lorobenzene
jC
0
10
X
0)
loroethane
JC
u
10
X
0)
X
u
u
10
10
OJ
OJ
c
10
•o
c
•H
•J
fa
3
o
fa
0)
tC
OJ
T3
•H
V4
O
rH
U
OJ
c
OJ
0)
-------
in
CM
a\
co
in
CM
o
o
o
o
o
o
rt
O)
O)
3
-3
0
0
o
k
m
n
•*
o
o
o
o
o
VO
o
0
o
..
0
o
0
CM
rH
^*
rH
.
VO
(0
m
u
r-
CM
VO
VD
^
VO
o
o
o\
*.
CM
rH
5J1
O
.
in
r-t
0
c
Q)
JC
a
o
o
rH
JS
U
(0
c
0)
o
o
o
_
rH
O
_j
rn
0
o
o
(N
n
o
0
o
..
0
in
00
r^
CM
O
•
P*
a
a
u
i
00
V
r*
f>
».
0
in
in
..
rH
O
in
in
rH
o
0
0\
..
0
c*>
CM
r^
.
CM
0)
C
0)
3
rH
O
O
0
o
V
0
CM
^
CM
O
0
o
cT
rH
(N
O
0
O
..
0
o
VO
CNJ
^"
d
f)
•
^t
0)
c
0)
£
a
(0
X
o
o
f-
m
^
rH
O
^
in
rH
O
0
^"
«.
rH
f>
O
O
•
«T
Q)
C
0)
N
C
Q)
Xt
O
O
rH
£
U
•H
iH
1
"^
f>J
••
00
rH
00
rH
n
n
n
00
^>
.
CM
0)
c
a)
iH
>^
,X3
4J
Q>
O
>H
O
iH
x:
u
•H
}-l
•G
i-t
(0
x:
x
^
3
(0
•
r4
0)
S
O
rH
A
U
3
E
0)
Xt
o
4J
fa
<
CO
TJ
0)
V4
3
U
10
vV
E
1 •
•O W
•sa
•H a.
i'Q
r-( 4)
XI 4J
10 m
XI M
O 4J
h Ul —
a c %,
O *H
•OB ^>
•H Q) ^
3 T3
0 *£
4J ^
E Q4 CQ
M -H 3,
•H tl
•HO ||
O U
A 3 J
(0 C -r->
*• * b
0) E "J
jg C. cq
• •
O
-H
^
to
*J
c
o
Q)
X
10
J
h
O
iw
0)
j^
0)
3
in
t^
o
o
o
o
o
o
o
o
*
fS
o
o
o
o
o
i
b
------- |