![[logo] US EPA](http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
2007 CompTox Forum
Abstract - Modeling the Ecological Effects of Endocrine Active Compounds on Fish: Scaling from Individuals to Populations
Kenneth A. Rose[a], Cheryl A. Murphy[b], Peter Thomas[c], Lee A. Fuiman[c], Maria C. Alvarez[c], Ian D. McCarthy[d] and Sandra L. Diamond[e]
Kenneth A. Rose (presenting author)
Professor
Coastal Fisheries Institute and Department of Oceanography and Coastal Sciences
Louisiana State University
Energy, Coast and Environment Building, Baton Rouge, LA 70803
Phone: 225-578-6346
E-mail: karose@lsu.edu
Exposure to endocrine disruptors tends to rely on biomarkers that are static measurements from a dynamic system and, therefore, difficult to interpret. Exposure can cause subtle effects on individual performance that do not easily relate to endpoints of ecological significance. We use two different modeling approaches applied to polychlorinated biphenyls (PCBs), cadmium, methylmercury, and hypoxia to help understand biomarkers and to scale sublethal effects measured on individuals to population-level responses. The first approach is a physiologically-based model that simulates the major biochemical reactions from the secretion of gonadotropin to the production of vitellogenin in a single female fish over a 6-month spawning period. Various intermediate variables in the model correspond to commonly measured biomarkers of endocrine disruption used in field studies. Model predictions of reduced cumulative vitellogenin production and changes in biomarkers for PCB, cadmium, and hypoxia exposure were similar to laboratory-measured values. The model was subsequently applied to biomarkers measured in field-caught fish sampled at normoxic and hypoxic locations in Pensacola Bay, Florida, to help identify multiple stressors. The second modeling approach uses a series of linked statistical, individual-based, and matrix projection simulation models to simulate the effects of endocrine disruption on Atlantic croaker population dynamics. We used results of laboratory experiments on the behavioral effects of PCBs and methylmercury on larval fish escape responses to predators. Laboratory results were statistically analyzed and changes in larval escape abilities and swimming speed were inputted into an individual-based larval fish cohort model. The individual-based model then predicted changes in larval growth and survival, which were used to change to parameters of a stage-within-age matrix projection model. The matrix model simulated the long-term population dynamics under baseline conditions and various exposure scenarios. Taken together, these models provide a basis for better understanding the sometimes complex relationship between static biomarkers and varying exposures, and for placing sublethal effects into an ecologically-relevant and management-oriented context.