Identification of sources of heavy metals in agricultural soils using multivariate analysis and GIS
2013 (English)In: Journal of Soils and Sediments, ISSN 1439-0108, E-ISSN 1614-7480, Vol. 13, no 4, 720-729 p.Article in journal (Refereed) Published
Purpose Heavy metals in agricultural soils readily enter the food chain when taken up by plants, but there have been few investigations of heavy metal pressure in farming areas with low background concentrations. This study was carried out in a cultivation area of Northeast China that has undergone decades of intensive farming, with the aim of identifying the sources of accumulated heavy metals in agricultural soils using multivariate analysis and geographic information system (GIS). Materials and methods In 2011, concentrations of total iron (Fe), manganese (Mn), copper (Cu), nickel (Ni), lead (Pb), zinc (Zn), cadmium (Cd), chromium (Cr) and cobalt (Co), as well as soil pH and organic matter, were measured at 149 sites in arable soils in the study area. The principal component analysis (PCA) was employed to extract hidden subsets from the raw dataset in order to detect possible sources. Metal contents in soils from various croplands were further investigated using analysis of variance. With the Kriging interpolation method, GIS was used to display the PCA results spatially to explore the influence of land use on heavy metal accumulation. Results and discussion Most of the studied metals in arable soils of the study area were shown to have low concentrations, except for Cd (0.241 mgkg−1). According to the results of the PCA analysis, Fe, Mn, Pb, Zn, Cd, and Co formed the first component (PC1) explaining 40.1 % of the total variance. The source of these metals was attributed to farming practices (“anthropogenic” factor). Cu, Ni, and Cr fell into the second component (PC2), heavy metals that derived from parent rock materials (“lithogetic” factor). This component describes 24.6 % of the total variance. Compared to paddy lands, soils in drylands had greater accumulations of all the etals in PC1, which can be explained by a higher rate of phosphorus fertilizer application and a longer farming history. Conclusions Owing to the natural low backgrounds, soils in the study area were safe from heavy metal pollution with a contamination risk of Cd the only exception. Multivariate analysis and GIS were effective means in helping to identify the sources of soil metals and addressing the land use influence on soil metals accumulation. This work can support the development of strategy and policies to aid in the prevention of widespread heavy metal contamination in area with characteristics similar to those of the study area.
Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2013. Vol. 13, no 4, 720-729 p.
Arable soils, Geographic information system (GIS), Heavy metal, Principal component analysis (PCA), Sources identification
IdentifiersURN: urn:nbn:se:umu:diva-70384DOI: 10.1007/s11368-012-0637-3OAI: oai:DiVA.org:umu-70384DiVA: diva2:621364