Page 875 - The Toxicology of Fishes

Reproductive Impairment of Great Lakes Lake Trout by Dioxin-Like Chemicals 855

100
120
Lake trout caught 90
Lake trout from 1% survival of stocked fry
Thousands of Lake Trout Caught 80 natural reproduction 60 Thousands of 6-Year-Old Lake Trout Possible from 1% Survival of Stocked Fry
80
100
70
Consistent with
50
60
40
40
30
20
20
10
0 0
1860 1880 1900 1920 1940 1960
Year
FIGURE 21.10 Association of lake trout population levels, as revealed by annual commercial catch (thousands) prior to
1940, with numbers of 6-year-old lake trout (thousands) predicted to be available for catch based on hypothetical survival
of 1% of fry annually stocked into Lake Ontario. The commercial catch population data show evidence for recruitment up
until about 1936 of additional lake trout from adults that grew from fry stocked earlier. (From Cook, P.M. et al., Environ.
Sci. Technol., 37, 3867–3877, 2003. With permission.)

particular HAHs, having had an influence in the crash and subsequent lack of rehabilitation of lake trout
populations in the lower Great Lakes for three to four decades during the 20th century. In this section,
we evaluate the information covered in this chapter from the perspective of an ecoepidemiological
assessment. The criteria for establishment of cause-and-effect linkages in ecotoxicology have been taken
from the biomedical sciences and adapted for use with fish and wildlife species (Fox, 1991). The
ecoepidemiological criteria put forth by Fox (1991) provide a framework for evaluation of the strength
of evidence for causation of an effect (i.e., lack of recruitment in Great Lakes lake trout populations)
with a putative causal agent (i.e., HAH chemicals). Seven criteria provide the basis for evaluation of the
data or information for causation in this approach. The ecoepidemiological approach was used by Mac
and Edsall (1991) to evaluate the strength of data that HAHs had an effect on lake trout populations in
the Great Lakes. Those authors concluded that the evidence at the time supported the hypothesis that
chemical contaminants had caused reproductive impairment in lake trout populations of the Great Lakes;
however, they went on to state that much of the evidence was circumstantial, with little definitive proof
of such a causal relationship. Since the time of that evaluation, a great deal of additional evidence has
served to strengthen and clarify the causal relationship between HAHs and reproductive impairment of
lake trout in the Great Lakes. A large portion of that evidence has been presented in this chapter and is
summarized here within the ecoepidemiological framework.

Ecoepidemiological Criteria
The seven criteria for evaluation of the relationship between a suspected causal agent and an adverse
effect are (1) probability, (2) time-order, (3) strength of association, (4) specificity, (5) consistency of
association, (6) predictive performance, and (7) coherence. We provide a summary of the existing data
within this context. The hypothesis that we are evaluating with regard to these criteria is that HAHs
were of sufficient concentrations in lake trout populations to cause embryo and fry mortality and
subsequent lack of recruitment in the lower Great Lakes from the 1950s until the early 1980s. In our
evaluation, we will compare and contrast evidence from all of the Great Lakes except Lake Erie. This
is because Lake Erie never had very strong lake trout fisheries or populations due to the relatively shallow
nature of the lake and the hydrological patterns of the lake.