Application of geostatistics to soil and groundwater samples contaminated by petroleum products - a case study

Abstract

A geoenvironmental site investigation/characterization survey took place in a contaminated costal area in the surroundings of an active oil refinery having an associated petrochemical complex, located north of Porto, in the north-western part of Portugal. In this context, several soil and groundwater samples were collected, respectively in trenches and boreholes, and chemically analyzed allowing the characterization of their chemical pollution, in terms of the different contamination products and respective concentrations, namely BTEX, TPH, PHA's and lead, as well as of some physical/chemical soil properties and charac-teristics, namely density, pH and organic matter content. At this stage of the inves-tigation, a multivariate statistical analysis has been previousely carried out. The present study is focused on comparing alternative geostatistical procedures, namely different kriging methods as well as strategies, used to characterize the qualitative and quantitative spatial contamination dispersion in the investigated area leading to different mapped models prior to the definition of remediation strategies. In addition, it is also summarized a risk analysis study, also based on geostatistical procedures, leading to a partition of the studied zone in terms of their assessed relative remediation needs.