Wednesday, August 5, 2009: 8:00 AM
Santa Ana, Albuquerque Convention Center
Background/Question/Methods Spatial boundaries for ecological risk assessments often must balance administrative and physical boundaries with ecological considerations of how species use the environment. This paper considers how this balance was achieved for ecological risk assessments completed for the Los Alamos National Laboratory (LANL) in New Mexico and the Hanford Site in Washington state. The areas considered in the LANL assessment are located in canyons that historically received contaminated liquid discharges; these liquid discharges have come to be located along tens of kilometers in this semiarid alluvial system. The sites addressed in the Hanford Site risk assessment are spread across most of the 140,000 ha site, but most of the waste sites are concentrated in eight operational areas of approximately 260 ha in total. For both LANL and the Hanford Site a variety of ecological assessment endpoints were developed to address risks at the local and broad scale. Both measures of exposure and effect were considered at local scales, generally 1 ha in area, and some of these measures were integrated over larger scales to evaluate risks from contaminants to species with larger home ranges (100 ha or larger). This information was useful in evaluating risks associated with particular contaminant releases and waste sites and also useful for documenting the general ecological condition.
Results/Conclusions There are both technical challenges for implementing broad scale assessments and policy implications for implementing broad spatial scale ecological risk assessments at these sites. Some of the technical challenges include adequacy of contaminant exposure data for broad areas, habitat quality under current and future conditions, and reconciling protective assumptions with more realistic information on exposures and potential for adverse effects from contaminants. There are policy implications of focussing assessments on the waste site scale versus larger aggregate decision area scales.