Linking Ni and Cr concentrations to soil mineralogy: does it help to assess metal contamination when the natural background is high?

In soils from serpentinitic areas the natural background of Ni and Cr is so high that the assessment of contamination by comparing metal concentrations with some fixed thresholds may give unreliable results. We therefore sought a quantitative relation between serpentines and Ni and Cr concentrations in uncontaminated soils, evaluated if the approach may help in establishing a baseline, and discussed if additional anthropogenic inputs of Ni and Cr can be realistically individuated in these areas. We analysed the total, acid-extractable and exchangeable concentrations of Ni and the total and acid-extractable concentrations of Cr in 66 soil horizons, belonging to 19 poorly developed and uncontaminated Alpine soils. The soils had different amounts of serpentines, depending on the abundance of these minerals in the parent material. We calculated an index of abundance of serpentines in the clay fraction by XRD and related total metal contents to the mineralogical index. We then tested the regressions on potentially contaminated soils, developed on the alluvial plain of the same watershed. We found extremely high total concentrations of Ni (up to 1,887 mg kg(-1)) and Cr (up to 2,218 mg kg(-1)) in the uncontaminated soils, but only a small proportion was extractable. Total Ni and Cr contents were significantly related to serpentine abundance (r (2) = 0.86 and 0.74, respectively). The regressions indicated that even small amounts of serpentines induced metal contents above 200 mg kg(-1), and the 95% confidence limits were 75 and 111 mg kg(-1) of Ni and Cr, respectively. When the regressions were tested on the potentially contaminated soils, a good estimate was obtained for Cr, while the Ni concentration was overestimated, probably because of some leaching of this element. The concentrations of Ni and Cr that can be expected in soils because of the presence of small amounts of serpentines are comparable to the amounts accumulated in the soil because of diffuse contamination and potentially contaminated soils had metal concentrations falling in the range expected from the presence of natural sources. Only in the case of very severe contamination events, the identification of anthropogenic sources adding to the natural background would be feasible.