Demography and Modeling To Improve Pesticide Risk Assessment of Endangered Species

The present ecological risk assessment process for pesticides as practiced by the United States Environmental Protection Agency under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) consists of developing short-term toxicity data for a few select species and comparing these data to expected environmental concentrations to develop risk quotients (RQ). Risk quotients are then compared to "levels of concern" (LOC) which vary depending upon the type of pesticide being evaluated and the type of organisms likely to be exposed. The LOC is supposed to account for all of the types of uncertainty associated with the risk assessment. There are several problems associated with this process. For example, populations do not respond the same way to wtoxicant exposures as do individuals. Populations, if thinned may compensate for losses and actually grow faster than expected. Furthermore, exposure to pesticides can result in a proportion of a population dying while the remaining individuals are impaired due to multiple sublethal effects. Another issue is that the few species that are used for the development of toxicity data may not be good representatives of the many species they are supposed to protect. These species are often chosen based on ease of rearing and evaluation, not because they are good representatives of many species. Susceptibility of a population to pesticides is influenced by life history traits and differences in life history traits are not considered in the current risk assessment process. It has also been shown that populations of organisms often exist as mixtures of life stages. The makeup of the population structure can greatly influence susceptibility to pesticides and this is also not considered in the risk assessment process. Furthermore, different life stages may exhibit vastly different susceptibility to toxicants, yet usually only one life stage is evaluated. In this chapter, the current risk assessment process and new approaches to improve risk assessment of pesticides are discussed. The new approach consists of developing population-level measures of toxicant effect that incorporate the total effect (lethal and multiple sublethal effects) followed by population modeling to determine the probability that specific concentrations of pesticides will drive populations to extirpation or whether populations will recover.