Risk management decisions for pesticides and threatened and endangered species: The role of uncertainty analysis

Despite data gaps and information shortfalls, government agencies in the United States are expected to produce timely and defensible decisions to regulate pesticide use under the Federal Insecticide, Fungicide, and Rodenticide Act and in compliance with the Endangered Species Act. The decision to register a pesticide is predicated on a conclusion that no unreasonable effects will accrue to the environment, including threatened and endangered species. We recognize that the definition of acceptable risk is a policy judgment stemming from legislative language and judicial interpretation. However, a common risk assessment approach with similar technical underpinnings and a high degree of transparency used by all the agencies would be cost effective and more likely to achieve consensus among interested parties. Quantitative probabilistic risk assessment (PRA) methods can be used to develop risk estimates and to describe the level of confidence in these estimates. PRA methods can also differentiate among the contributions of natural stochasticity, measurement variability, and lack of knowledge. Because this approach enhances transparency and increases understanding of the implications of limited data sets and associated assumptions, we encourage the appropriate agencies to implement PRA methods as a means of reaching common ground when assessing risks of pesticides to listed species.