Alternate Modeling Approaches for Contaminant Fate in Soils: Uncertainty, Variability, and Reliability. T. E. McKone, Risk Sciences Program, Environmental Toxicology Department, University of California, Davis, CA 95616
The fate of organic and metal contaminants in soils is an issue that is relevant to uncontrolled hazardous waste sites (Superfund sites) and to waste classification for wastes that must be placed in RCRA facilities. Addressing long-term potential human exposures requires the use of models. Because these models must project waste behavior into the future, there is much uncertainty that derives from these predictions. This means that there is a large statistical variance with regard to predictions of the time behavior of the waste inventory. Some of this variance derives from uncertainty about chemical properties and transformation half-lives. Some derives from the variation in landscape properties among sites. The purpose of the paper is to consider two alternate modeling approaches and the reliability with which each approach can represent the variability of contaminant behavior in the California landscape. One modeling approach is more deterministic and based on an exact solution of the dispersion/advection differential equations that describe contaminant fate in the contaminated soil zone. The second approach is based on a mixed stochastic and mass-balance approach using a box model that has less resolution in the soil layers. The first modeling approach is rather complex and difficult to use in regulatory context, whereas the second approach is transparent and easily used by decision makers. By using an analysis-of-variance approach, the reliability with which these two models represent variability of contaminant behavior in the California landscape is compared. Using trichloroethylene (TCE) in California soil as a case study, we find that, relative to the more complex model, the simple model is accurate in capturing both the correct cumulative distribution and the proper rank regression trends with respect to both variability and uncertainty.
This work was performed in part under the auspices of the U.S. Department of Energy (DOE) through Lawrence Livermore National Laboratory under Contract W-7405-Eng-48 and in part at the University of California, Davis Risk Sciences Program with funding provided by California Office of Environmental Health Hazard Assessment.