Risk analysis for remediation of contaminated sites: the geostatistical approach

The assessment of the risks associated with contamination by elevated levels of pollutants is a major issue in most parts of the world. The risk arises from the presence of a pollutant and from the uncertainty associated with estimating its concentration, extent and trajectory. The uncertainty in the assessment comes from the difficulty of measuring the pollutant concentration values accurately at any given location and the impossibility of measuring it at all locations within a study zone. Estimations tend to give smoothed versions of reality, with the smoothing effect being inversely proportional to the amount of data. If risk is a measure of the probability of pollutant concentrations exceeding specified thresholds, then the variability is the key feature in risk assessment and risk analysis. For this reason, geostatistical simulations provide an appropriate way of quantifying risk by simulating possible “realities” and determining how many of these realities exceed the contamination thresholds, and, finally, provides a means of visualizing risk and the geological causes of risk. This study concerns multivariate simulations of organic and inorganic pollutants measured in terrain samples to assess the uncertainty for the risk analysis of a contaminated site, an industrial site in northern Italy that has to be remediated. The main geostatistical tools are used to model the local uncertainty of pollutant concentrations, which prevail at any unsampled site, in particular by means of stochastic simulation. These models of uncertainty have been used in the decision-making processes to identify the areas targeted for remediation.