public needs to be evaluated. Conditions for which regulatory and public acceptance of bioavailability information is most likely are described below.
If site conditions, the contaminant of interest, and the default cleanup objectives are similar to those at other sites where remedial action is needed or underway, investment in an assessment of bioavailability processes may be warranted because regulators and the public will have familiarity with the problem. Acquisition of process data and knowledge and application of new measurement tools for bioavailability assessment may help with formation of more cost-effective solutions. An example is the swine test to assess bioavailability of lead in soil, which was applied by EPA to test soils at the Palmerton Zinc Pile Superfund Site (see Box 2-5). The results of the swine testing did not affect the remediation decision at the Palmerton site, as they pointed to acceptable lead soil concentrations in the range estimated by the default assumptions. However, the experience gained at the Palmerton Site and elsewhere led to subsequent applications of the swine testing at approximately 20 other lead-contaminated sites, including several high-volume waste sites. Remediation decisions were influenced by the swine test results at some of these sites (Weis, 2000).
A bioavailability assessment is difficult to justify if a relevant regulatory body has a policy stance against explicit consideration of particular bioavailability processes. Some state environmental agencies and EPA regions have included in guidance to their remediation project managers and risk assessors recommendations or policy directives to refrain from consideration of certain bioavailability processes in estimating exposure (see Table 2-8).
In contrast, some state environmental agencies and EPA regions have developed guidance for consideration of bioavailability processes in risk assessment. EPA Region 10, for example, developed guidance for bioavailability considerations in human health risk assessments for arsenic contaminated soil (see Chapter 2). Washington state has very recently amended its Model Toxics Control Act to allow for incorporation of new scientific information which could be used to modify the “gastrointestinal absorption fraction” and other bioavailability default assumptions (G. McCormack, Washington Department of Ecology, personal communication, 2003). While this has only been done in a few states as of this writing, and for a limited range of bioavailability processes and contaminants, the existence of guidance signifies openness to bioavailability process evaluation.
The preceding chapters have shown that there is a variety of physical, chemical, and biological processes that determine the availability of contaminants in soils and sediments to ecological receptors; that consideration of these bioavailability processes is inherently part of the risk assessment process; that validated measurement techniques and models exist for some bioavailability processes, but not for many others; and that uncertainty about how to measure and