A framework for the assessment of health risks for bottlenose dolphins (

Recent studies have reported alarmingly high levels of persistent organochlorine contaminants, particularly polychlorinated biphenyls (PCBs), in the tissues of many marine mammals. The levels of PCBs found in marine mammals are often several orders of magnitude greater than the levels found in humans and other terrestrial mammals. These findings are cause for concern, given the current body of experimental and epidemiological evidence that links PCBs to deleterious effects on reproduction and immune system function. Compounding this concern are reports of recent disease outbreaks that have critically reduced many populations of marine mammals and are suspected to have been caused, at least in part, by the high concentrations of pollutants found in the tissues of the stranded animals.; Determination of the potential risks associated with exposure to persistent organochlorine pollutants is essential for the conservation and management of marine mammal species. This is a complex task, requiring the integration of models of population dynamics with estimates of exposure and associated effects. The difficult logistics involved in carrying out studies in the wild, along with limitations imposed on experimental studies of protected species, further complicate the task. These difficulties result in a lack of directly applicable dose-response information and a high degree of uncertainty with regard to parameters that define population dynamics.; This research integrates methods and concepts from the fields of both ecological and human health risk assessment, as well as survival analysis and mathematical modeling, to develop the necessary paradigm for the assessment of health risks in wild marine mammal populations. The methodology produces probabilistic estimates of potential reductions in marine mammal populations due to a specified level of contaminant exposure. A simulation framework is implemented which couples individual-based modeling with two-dimensional Monte Carlo simulation to provide for the inclusion of knowledge uncertainty, as well as natural variability, inherent in marine mammal populations. The framework is demonstrated for the specific assessment of reproductive health risks for bottlenose dolphins (Tursiops truncatus) associated with exposure to polychlorinated biphenyls. Data from multiple sources are integrated into the simulation framework in order to carry out exploratory simulations for the determination of essential risk assessment parameters and parameter relationships. Specifically, measured contaminant burdens from two populations of bottlenose dolphins, age distribution information from a long-running study of a separate dolphin population, and a proposed dose-response model for the quantification of the relationship between PCBs and detrimental health effects are incorporated. Detailed analyses of associated uncertainty and sensitivities for the developed model are presented which indicate that the model is robust to moderate alterations in parameter inputs. Results indicate that uncertainty with regard to measured exposures is relatively inconsequential as compared to uncertainty in estimated dose-response model parameters.