Elucidating the differentiation of soil heavy metals under different land uses with geographically weighted regression and self-organizing map.

Intensive anthropogenic activity has triggered serious heavy metal contamination of soil. Land use and land cover (LULC) changes bear significant impacts, either directly or indirectly, on the distribution of heavy metal in soils. A total of 180 samples were acquired from various land covers at different depths, namely surface soils (020 cm) and subsurface soils (20-40 cm). Spatial interpolation, geographically weighted regression (GWR) and self-organizing map (SOM) were used to discern how variations in the spatial distributions of soil heavy metals were caused by human activities for different land uses, and how these pollutants contributed to environmental risks. The medium concentrations of Cd, Cr, Cu, Pb and Zn in surface soil all exceeded the corresponding local background values in flat cropland and developed area soil. The overall ecological risk level of the study varied from low to medium. The GWR model indicated that the land use intensity had a certain influence on the accumulation of heavy metals in the surface soil. K-means clustering of the SOM revealed that the type of LULC also contributed to the redistribution of heavy metals in the surface soil.