A combination of finite mixture distribution model with geo-statistical models to study spatial patterns and hazardous areas of heavy metals in cropland soils of the Guanzhong Plain, Northwest China

Abstract

An extensive cropland soil investigation was conducted to determine the pollution thresholds and hazardous zones of heavy metals (HMs) in the Guanzhong Plain, by using an integrated approach that combines finite mixture distribution model (FMDM) and geo-statistical analysis. FMDM results demonstrated that Pb, Cr, Ni, and Cu were fitted by binary mixture distributions representing the background and moderate pollution distributions, and Zn was fitted by a triple mixture distribution representing the background, moderate and high contamination distributions. The moderate pollution thresholds of Pb, Cr, Ni, Zn and Cu calculated by FMDM were 29.75, 80.15, 38.60, 81.48 and 27.10 mg kg−1, whereas the cutoff value of Zn high contamination was 97.49 mg kg−1. The moderately polluted thresholds of all five HMs were higher than their background values in the study area, and lower than the corresponding national standards. The indicator kriging simulation showed Pb, Cr, Ni, Zn had <0.1%, 2.6%, <0.1%, 2.9% of total areas exceed contamination cutoff values, whereas the hazardous area of Cu was contiguous, and covered 17.3% of the total area. Overall, 17.5% of the total area surpassed the moderate contamination threshold. The pollution hot spots and hazardous zones of soil HMs were located in the southern part of the Guanzhong Plain, where population and industrial activities are centralized, indicating that anthropogenic activities played a critical role in HMs accumulation in high-risk regions. The combination of geo-statistical and FMDM delineate the thresholds and hazardous area for HMs pollution reliably, and facilitate the improvement of soil environmental management.