Spatial Network of Bioaccumulation in Food Webs

Tulane University, New Orleans LA

Investigators

Abstract

Spatial Network of Bioaccumulation in Food Webs PI: Dr. Karen H. Watanabe, Tulane University (watanabe@mailhost.tcs.tulane.edu) Co-PI: Dr. Ronaldo Luna, University of Missouri-Rolla (rluna@umr.edu) Co-PI: Dr. Henry L. Bart, Jr., Tulane University (hank@plato.museum.tulane.edu Traditional formulations of food-web bioaccumulation models assume steady-state conditions and use average environmental input parameters. Yet, there is known variability in environmental conditions such as "hot spots" of pollution and poorly mixed water environments that are not accounted for in the current paradigm of bioaccumulation modeling. Averaging of known spatial variability in environmental conditions is a source of error in model predictions that can be reduced by using site-specific data. While a top predator may traverse a large region for food and be exposed to a variety of contaminant concentrations, the organisms at the lowest trophic levels may be relatively stationary and exposed to the local contaminant concentration. A modeling methodology is needed to incorporate spatial variability in the study of bioaccumulation of contaminants in a food web. This project integrates an aquatic food web bioaccumulation model for hydrophobic organic contaminants with an efficient spatial representation of the environment. In particular, it targets geographic study regions (e.g., rivers and bayous) that can be represented as linear features in a spatial network model using tools available in a geographic information system (GIS). Data stored as different events on the dynamically segmented linear feature will be used in combination with rules to allow the movement of different species within their natural home range. The bioaccumulation model will directly access the spatially referenced environmental data to predict the biological fate of contaminants at different locations in the study region. Food web organisms, sediment and water samples will be collected from a network of bayous in the LaBranche Wetlands in Louisiana to demonstrate the modeling methodology and validate the food web bioaccumulation model. These samples will be analyzed for semi-volatile organic chemicals and the results stored in a spatial database. The project spans three years for the integration of modeling techniques and collection of validation data. The result will be a portable computational tool driven by a GIS engine that improves upon current bioaccumulation modeling methodology.