Integrating geophysical data for mapping the contamination of industrial sites by polycyclic aromatic hydrocarbons: A geostatistical approach

A case study is presented of building a map showing the probability that the concentration in polycyclic aromatic hydrocarbon (PAH) exceeds a critical threshold. This assessment is based on existing PAH sample data (direct information) and on an electrical resistivity survey (indirect information). Simulated annealing is used to build a model of the range of possible values for PAK concentrations and of the bivariate relationship between PAH concentrations and electrical resistivity. The geostatistical technique of simple indicator kriging is then used, together with the probabilistic model, to infer, at each node of a grid, the range of possible values which the PAH concentration can take. The risk map is then extracted for this characterization of the local uncertainty. The difference between this risk map and a traditional iso-concentration map is then discussed in terms of decision-making.