Source identification of heavy metals in pastureland by multivariate analysis in NW Spain

Arable layer of pastureland in Galicia (NW Spain) was monitored for total content of heavy metals, and analysed by multivariate statistical techniques, in order to investigate the different origin that metals may have in pasture soils. Principal component analysis (PCA), cluster analysis (CA) and correlation matrix (CM) were applied to 65 samples in which the total concentrations of Mn, Fe, Zn, Cu, Cr, Co, Ni, Cd and Pb were measured. Four significant components were extracted by PCA, explaining 78.830% of total variance. Mn, Co and Ni (and partially Cu), and Fe and Cr, were associated in two lithogenic components, respectively, while an anthropogenic origin was identified for Cd, Pb and Zn. Zn (and Cu) were mainly associated with soil fertilisation by cattle slurry or with other activities regarding cattle management. Although the origin of Cd and Pb was also attributed to slurry application, other sources like commercial fertilisers, vehicle exhaust or aerial deposition were not discarded as possible contributors. CA confirmed and completed the results obtained by PCA, classifying the data in four groups representing different areas. Group 1 represented samples corresponding to areas were the application of manure was moderate, while Group 2 included samples of lithogenic origin. Highest contents of anthropogenic metals were included in Group 3, although soils in this cluster were not considered as polluted. The last cluster grouped the samples with the lowest content of all the metals analysed, representing areas correctly managed and not affected by other external sources. Finally, the results obtained by CM agreed with PCA and CA, also helping in elucidating individual relationships between metals.