This project has two purposes: to develop the chemical basis for establishing sediment quality criteria for arsenic and chromium for arsenic and chromium; and to construct coupled water column-sediment fate and transport models. Arsenic and chromium in aquatic sediments are currently listed as contaminants at 113 and 142 Superfund sites, respectively. It is of critical importance, therefore, to have reliable methods for determining the sediment concentrations at which the metals pose an environmental and human health risk. Currently available methods are based on the concentration of total arsenic and chromium that ignore bioavailability that are known not to be predictive of toxicity. We intend to use the Equilibrium Partitioning (EqP) model, that is currently being used by EPA, in order to generate these criteria. This approach involves first: determining the solid phase or phases that regulate pore water concentrations; and third: determining the potential for remobilization (and future exposure) of sediment bound metal. For water column animals and their human consumers, the extent to which metals are released from sediments to the overlying water, and the extent to which the reverse process occurs, are critical components in a comprehensive analysis of the risk environmental and human health posed by these metals. We intend to develop sediment models first and then coupled water column-sediment models for arsenic and chromium. The purpose of the sediment model is to computer the flux of metal from the sediment to pore water, and ultimately, to the overlying water. This release is determined in large measure by the rate of oxidation of the reduced solid phase metal species. Once sediments are judged to pose an environmental or human health risk, it is necessary to project future exposure concentrations in the sediments and overlying water and to evaluate the efficacy of remedial actions. This step requires the use of mathematical models to described the combined effects of transport and chemical/biochemical reactions. We intend to construct these models and to apply them to data sets that are available in the literature.
Showing the most recent 10 out of 158 publications
