Estimation of Potential Population Level Effects of Contaminants on Wildlife

The objective of this project is to provide DOE with improved methods to assess risks from contaminants to wildlife populations. The current approach for wildlife risk assessment consists of comparison of contaminant exposure estimates for individual animals to literature-derived toxicity test endpoints. These test endpoints are assumed to estimate thresholds for population-level effects. Moreover, species sensitivities to contaminants is one of several criteria to be considered when selecting assessment endpoints (EPA 1997 and 1998), yet data on the sensitivities of many birds and mammals are lacking. The uncertainties associated with this approach are considerable. First, because toxicity data are not available for most potential wildlife endpoint species, extrapolation of toxicity data from test species to the species of interest is required. There is no consensus on the most appropriate extrapolation method. Second, toxicity data are represented as statistical measures (e.g., NOAEL s or LOAELs) that provide no information on the nature or magnitude of effects. The level of effect is an artifact of the replication and dosing regime employed, and does not indicate how effects might increase with increasing exposure. Consequently, slight exceedance of a LOAEL is not distinguished from greatly exceeding it. Third, the relationship of toxic effects on individuals to effects on populations is poorly estimated by existing methods. It is assumed that if the exposure of individuals exceeds levels associated with impaired reproduction, then population level effects are likely. Uncertainty associated with this assumption is large because depending on the reproductive strategy of a given species, comparable levels of reproductive impairment may result in dramatically different population-level responses. This project included several tasks to address these problems: (1) investigation of the validity of the current allometric scaling approach for interspecies extrapolation an d development of new scaling models; (2) development of dose-response models for toxicity data presented in the literature; and (3) development of matrix-based population models that were coupled with dose-response models to provide realistic estimation of population-level effects for individual responses