SUPERFUND SITE CHARACTERIZATION USING NONPARAMETRIC VARIOGRAM MODELING

The spatial interpolation method of Kriging, originally developed for estimation of ore reserves in mining and prediction in meteorology, has become an important geostatical tool for assessment of contamination in Superfund site characterization. Kriging uses an estimated spatial covariance matrix, or alternatively the variogram, to find the best linear unbiased estimate (BLUE) of contamination at an unsampled location. The modeling of the variogram (also called semi-variogram) is therefore a necessary step that must be performed before using Kriging for spatial interpolation. In this paper, we propose and investigate a simple non-parametric procedure based on ranks for variogram modeling. Some simulated examples are used to illustrate the proposed procedure. The proposed procedure is compared with the usual method of Ordinary Kriging for the Billings Smelter Superfund site data.