Ecological risk assessment of pesticides for terrestrial vertebrates: evaluation and application of the U.S. Environmental Protection Agency's quotient model

The ecological risks of pesticides to wildlife are assessed by the United States Environmental Protection Agency with a simple quotient model. A risk factor is calculated by dividing the estimated dietary concentration by the chemical's dietary toxicity. The risk factor, which can be used to predict population mortality, serves as a regulatory criterion to identify compounds that could pose unacceptable risk. Under the "New Paradigm," the Environmental Protection Agency recently cut by 50% the critical risk factor, yielding more-rigorous decision-making criteria by decreasing acceptable wildlife mortality from 50 to 8.8% following one pesticide application and by increasing the number of pesticides that pose unacceptable risk to wildlife. The quotient model depends on at least one implicit and approximately 11 explicit assumptions; deviations from these assumptions may bias outputs and consequent regulatory decisions. To correct error-prone assumptions we present three modifications that improve ecological risk estimates. We used the agency's quotient model to survey risks to a regional avifauna (Florida birds) in relation to mosquito pesticide exposure (fenthion, malathion, and naled). Certain groups (small-bodied, insectivorous passerines) were predicted to face the highest ecological risk due to taxonomy, diet, and body mass. Finally, we performed a risk assessment for Black-whiskered Vireos (Vireo altiloquus barbatulus) exposed to fenthion. The results revealed that this species could potentially suffer 42% mortality in a population exposed to one application of fenthion. Ecological risk assessments increasingly will be based on mathematical models. To improve predictions of the effects of xenobiotic compounds on wildlife, biologists--especially conservationists--should work with ecotoxicologists to improve current models, develop new models, and provide data to parameterize models. Equally important, conservationists should participate in defining regulatory levels of ecological risk for wildlife.