New approach to the sampling of contaminated sites

In Italy, the areas identified as contaminated sites, during the last 15 years, are more than 3% of the entire territory and, consequently, a remediation action is required. The environmental characterization of a potentially contaminated site is preliminary to any reclamation intervention. The procedures reported in the D.Lgs. 152/2006 (also known as environmental code), prescribe a reasoned or random sampling for the determination of the contaminant concentrations of the samples to be compared with the CSC (concentrations of contamination threshold). The choice of one of these two procedures depends on the knowledge of the story of the site and of the industrial activities carried out within the site during the years. If this information is known, the best choice will be a reasoned sampling. However, the data obtained with this type of sampling are not statistically processable. If nothing or little is known about the site, a random procedure is preferable. In this case the samples are representative just because they are ramdomly selected, and the analytical data coming from the samples can be extended to the entire site through an inductive procedure. In any case, the current practice does not take into account the variability that accompanies the quantitative meausures determined on the samples. This paper describes a new sampling method and analytical result processing based on a statistical approach, in contrast to the deterministic approach, generally used. This method can be applied to the entire site or within each mesh, of variable size, that forms an ideal grid superimposed on the entire bounded area. Each mesh becomes, in this way, a sub-area of the entire site to be considered as an individual portion for remediation. In addition, the areal sampling, in contrast to the punctual sampling, aided by geostatistical methods, permits to extend the quantitative information obtained as average values of the samples, to the potentially contaminated site, with a controllable degree of uncertainty, through the rules of probability calculation. In this way, while considering the specific characteristics of the site, it can be used as a valid and accepted protocol, which establishes, before any risk analysis procedure and with a certain decisional error, if the site is potentially contaminated or not. Such methodology can be also applied, to check the quality of an emergency safety implementation and to follow the trend of a remediation intervention. Finally, a dedicated software is available that allows to upload maps of contaminated sites and to organize a sampling plan sitting in front of a computer. This software named Visual Sample Plan supports the development of a defensible sampling plan, based on statistical sampling theory and the statistical analysis of sample results to support confident decision making. VSP couples site, building, and sample location visualization capabilities with optimal sampling design and statistical analysis strategies. It also helps determine the number of samples needed and performs statistical analyses. Moreover it permits to store and manage all the analytical results coming from the environmental sampling and characterization.