Page 872 - The Toxicology of Fishes

852 The Toxicology of Fishes

The C values were measured from 1-cm section samples of two radionuclide-dated sediment cores.
soc
210
Two radionuclides ( Pb and 137 Cs) were used to estimate the chronologies of the sediment cores
(Goldberg, 1963; Krishnaswami et al., 1971; Robbins and Edgington, 1972), along with two dating
methods, to verify the accuracy of the aging (Appleby and Oldfield, 1978; Robbins, 1978; Robbins and
Herchle, 1993). The aging estimates derived from the two methods were nearly identical, as were the
resolutions provided by each estimation technique (Cook et al., 2003). BSAF values were measured
egg
from samples collected from 1978 to 1988. These empirical values for BSAF intrinsically account for
egg
differences in bioaccumulation for each of the congeners, as these factors are specific to species, life
stage, tissues, and locations. For periods prior to 1972 the BSAF values required adjustment due to
egg
the fact that larger concentrations of the chemicals existed in the water relative to concentrations in the
surface sediments. Because BSAFs are by definition normalized to C , increased uptake of the chemicals
soc
via water (pelagic food chain) with respect to uptake via sediment (benthic food chain) will increase
the BSAF values in proportion to the amount of increased bioaccumulation by lake trout (Burkhard
egg
et al., 2003).
The expected pre-1970s greater BSAF values were modeled based on predicted decreases in the
egg
sediment–water column concentration quotients (∏ socw ) (Cook and Burkhard, 1998; Thomann et al., 1992):
∏ = C ( soc) ( )
fd
C
w
socw
The values of ∏ socw for Lake Ontario were modeled with dynamic mass balance simulations (Endicott
and Cook, 1994). Based on predicted changes in ∏ socw , BASF values prior to 1970 were predicted to
egg
be two to three times greater than more recent values; thus, the pre-1970 BASF values were set at
egg
two times the measured values (1978 to 1988) by Cook et al. (2003). The BASF values ranged from
>10 to 0.3 for the PCBs and 0.27 to <0.001 for PCDDs and PCDFs and were consistent with other
published data and predictions (Russell et al., 1999). The concentrations of all of the dioxin-like HAH
congeners measured in the aged sediment core slices were used to estimate concentrations of each
congener expected to be present in lake trout eggs over the period from approximately 1920 to 1988.
The modeled concentrations of PCBs, PCDDs, and PCDFs in lake trout eggs were verified by comparison
with measured values in lake trout collected in 1988, 1990, and 1991. The modeled concentrations were
in good agreement with the measured amounts. For the periods when no lake trout were present in Lake
Ontario, BSAF-derived lake trout egg concentration estimates were also verified based on modeling
from concentration of HAHs measured in herring gull eggs (Environment Canada, 1991). Again, esti-
mates from both of these methods were in good agreement with one another (Cook et al., 2003).

2,3,7,8-Tetrachlorodibenzo-p-Dioxin (2,3,7,8-TCDD) Toxicity
Equivalence Risk Modeling in Lake Trout from Lake Ontario
Lake trout egg concentrations of HAHs predicted from the sediment core profiles were then multiplied
by congener-specific TEF values (Table 21.2) and summed according to an additive model of toxicity
(van den Berg et al., 1998). The resulting TEQ values estimated in lake trout eggs over the period
egg
from 1920 to 2000 were then compared to toxicity thresholds of TCDD for early-life-stage mortality in
laboratory studies (see Table 21.1). Cook and his colleagues (2003) estimated that lake trout egg TEQ egg
values less than 30 pg/g ww would not result in embryo or sac fry mortality, while concentrations greater
than 100 pg/g ww would result in 100% mortality, and that mortality would be approximately linear
between the concentrations of 30 and 100 pg/g (Figure 21.8). There was a general trend of increasing
TEQ values in lake trout eggs starting from 1930 and increasing to a maximum predicted amount in
egg
the late 1960s (Figure 21.8). Following the 1960s, TEQ values steadily decreased. The 2,3,7,8-TCDD
egg
toxicity equivalence risks posed by AhR agonists over time suggest that mortality in lake trout fry would
have been first expected as early as 1940 and continued to be expressed until the early 1980s, a 40-year
period. Complete mortality of fry would have occurred for approximately three decades from 1945 until
about 1975 (Figure 21.8) based on the estimated TEQ values; thus, regardless of any other stressor
egg