Characterization of metal pollution in soils under two landuse patterns in the Angouran region, NW Iran: a study based on multivariate data analysis.

The study presents the application of selected multivariate statistical methods (multivariate analysis of variance, discriminant analysis, principal component analysis) and geostatistical techniques to evaluate soil pollution status in arable lands of the Angouran region, NW Iran. Two representative landuse patterns, cropland and grassland, were selected for the purpose of this study. Seventy soil samples (35 topsoils and 35 subsoils) were collected from the two landuse types and 21 soil parameters including total element content and physicochemical properties were also determined. Results from application of the multivariate analysis of variance showed that the two landuse patterns were not statistically differentiated by subsoil variables, whereas significant differences existed between the two landuse patterns with respect to topsoil variables. Discriminant analysis rendered seven variables (Cu, As, Cd, OM, P, K and total N) as indicator parameters responsible for the discrimination between the two landuse types. Using the principal component analysis (PCA), two main components (PCs) explaining 71.71% of total variance were extracted. PC1, with a high contribution of Ni, Cr, Fe, Mn and clay content was hypothesized as lithogenic component and PC2, with high loadings for the seven discerning variables (Cu, As, Cd, OM, P, K and total N), was considered as an agrogenic component. Geostatistical analyses, including the calculation of semivariogram parameters and model fitting, further supported the PCA results. PC1 was generally characterized by moderate spatial dependence and long-range spatial variation (8000m) influenced by soil parent martial composition, while PC2 was modelled by pure nugget effect probably reflecting the influences of agrogenic activities. The findings of this study could not only expand our knowledge regarding the soil pollution status in the study area, but would also provide decision makers with the information to manage the agrochemical application in the arable lands to improve the sustainability and safety of intensive-farming activities in the study area.