Factorial Kriging Analysis as a tool for explaining the complex spatial distribution of metals in sediments.

Rivers flowing through urbanized and industrial areas are usually greatly damaged by anthropogenic activities discharging contaminants. Characterizing the spatial distribution of pollutants in sediments is of high importance for selecting a suitable remediation operation, but is a complex task because this spatial variability is the result of various physical and chemical mechanisms occurring at different scales. Factorial Kriging Analysis (FKA) was applied on data collected in a canalized river (Scarpe, France) for that purpose, because this geostatistical technique allows to decompose a given variable into components of different spatial correlations and map them separately. This decomposition is meaningful provided that it can be related to physical phenomena occurring at the identified spatial scales. FKA applied to Cd and Zn concentrations in sediments of the Scarpe river proved to be effective, allowing their mapping to be decomposed in a first map related to a short-range spatial correlation corresponding to hot spots interpreted as the impact of industrial and urban inputs located along the canal, and a second map related to a long-range spatial variability associated with long pollutant plumes interpreted as the effect of one major upstream pollutant input.