31-11 Pollution of Tropical Island Stream Ecosystems: Relation of Biotic Accumulation to Land Use and Trophic Dynamics

Elissa N. Buttermore , Department of Biology, North Carolina Cooperative Fish and Wildlife Research Unit, North Carolina State University, Raleigh, NC
Thomas Kwak , North Carolina Cooperative Fish and Wildlife Research Unit, U.S. Geological Survey, Raleigh, NC
W. Gregory Cope , Department of Environmental and Molecular Toxicology, North Carolina State University, North Carolina State University, Raleigh, NC
Patrick B. Cooney , Department of Biology, North Carolina Cooperative Fish and Wildlife Research Unit, North Carolina State University, Raleigh, NC
Peter R. Lazaro , Department of Environmental and Molecular Toxicology, North Carolina State University, Raleigh, NC
Damian Shea , Department of Biology, Department of Environmental and Molecular Toxicology, North Carolina State University, Raleigh, NC
Puerto Rico has a history of anthropogenic chemical usage and its human population density is among the highest globally, providing a model environment to study human impacts on tropical island stream ecosystems.  The objective of our study was to quantify occurrences and patterns of aquatic contaminants as related to watershed land-use characteristics and trophic relationships.  We used stable isotope analyses of carbon, nitrogen, and sulfur to elucidate contaminant and trophic dynamics.  We sampled 12 rivers spanning broad watershed land-use patterns.  Additional studies were conducted at four of these sites, involving stable isotope analyses of food web components and contaminant analyses of native fish and shrimp.  Stream ecosystems in Puerto Rico were not severely polluted, with the exception of PCBs and mercury in some fish.  Trophic level and contaminant concentrations were poorly correlated in these ecosystems that are characterized by frequent hydrologic disturbances, nutrient pulses, and marine influences.  Calculation of food web magnification factors was complicated by low levels of contaminants, biased estimates of trophic level from nutrient pollution, and short food chains.  Lipid content was a better predictor of contaminant concentration than trophic level.  These results will provide scientific information needed to guide ecosystem management and human health risk assessment.