Evaluating the Effects of Diet and Trophic Structure on Mercury Accumulation in Fish Using a System Dynamics Modeling Approach

Spatial and temporal trends of mercury concentrations in fish from a contaminated stream in Tennessee are not directly correlated with analogous trends in mercury concentrations in water and sediment. Likewise remediation measures aimed at lowering concentrations in sediment and water have not had the intended effect on fish concentrations. To address the hypothesis that differences in diet composition among species and ages explain much of the observed variation, we created a dynamical food web model of mercury accumulation. The model simulates uptake of methyl and inorganic mercury for multiple fish species and age groups simultaneously over several years at five locations within the 25-km stream.  Diets and consumption rates varied among grazers, insectivores, omnivores, and piscivores depending on species, age class, prey availability, and temperature. Mercury in diet items varied annually, seasonally, and spatially. Simulations of mercury accumulation for up to 10-year periods were calibrated and validated with over 20 years of chemical and biological data from the creek. Inclusion of variation in diet in the model demonstrated that changes in community structure with increasing stream size can result in non-intuitive differences in mercury accumulation in fish. The model is being used to evaluate various mitigation and recovery options.