Comparison of different interpolation methods and sequential Gaussian simulation to estimate volumes of soil contaminated by As, Cr, Cu, PCP and dioxins/furans.

Understanding the spatial distribution of organic and/or inorganic contaminants is crucial to facilitate decision-making of rehabilitation strategies in order to ensure the most appropriate management of contaminated sites in terms of contaminant removals efficiencies and operating costs. For these reasons, various interpolation methods [Thiessen Polygon (TP) method, inverse of distance (IDW) method, ordinary kriging (OK), as well as sequential Gaussian simulations (SGS)] were used to better understand the spatial distribution of As, Cr, Cu, pentachlorophenol (PCP) and dioxins and furans (PCDD/F) found onto a specific industrial site. These methods do not only vary in complexity and efficiency but also lead to different results when using values coming from the same characterization campaign. Therefore, it is often necessary to evaluate their relevance by performing a comparative analysis. The results showed that ordinary kriging (OK) was a better estimator of the mean and more advanced compared to the two other methods of interpolation (TP and IDW). However, it appeared that SGS has the same power than OK but it also permitted to calculate a reliable value of the probabilities of exceeding regulatory cut-offs of contamination.