Identifying quantitative sources and spatial distributions of potentially toxic elements in soils by using three receptor models and sequential indicator simulation.

Abstract

Determining the reliable source contribution and spatial distribution of potentially toxic elements (PTEs) is a focal issue for soil regulation and remediation. For this purpose, three receptor models, US-EPA positive matrix factorization (EPAPMF), weighted alternating least squares positive matrix factorization (WALSPMF), and non-negative constrained absolutely principle analysis (NCAPCA), were used to a dataset consisting of ten PTEs (As, Cd, Co, Cr, Cu, Hg, Mn, Ni, Pb, and Zn) for source apportionment. Hazardous areas of ten PTEs were delineated using sequential indicator simulation (SIS) and uncertainty analysis. Three factors for ten PTEs were derived by three receptor models with a one-to-one correspondence between the factors. To obtain more appropriate results, the three receptor models were combined to calculate the ensemble-average source contributions. As, Co, Cr, Cu, Mn, and Ni were derived from a natural source with ensemble-average contributions higher than 85.72%. Cd, Hg, Pb, and Zn were contributed by both parent material and anthropogenic influence. More than half of Hg concentrations were associated with atmospheric deposition caused by human emissions. The concentrations of 28.04% for Cd, 20.74% for Hg, 43.49% for Pb, and 23.71% for Zn were associated with human inputs including agriculture practice, industrial activities, and vehicle emissions. The maps of spatial distribution generated by the SIS indicated that parent materials controlled the spatial distributions of As, Cd, Co, Cr, Cu, Mn, Ni, Pb, and Zn. 27.1% and 32.1% of the total area for Cd and Hg were identified as hazardous areas exceeding 1.5 times background values of Shandong province.