Identification of Heavy Metal Sources In Agricultural Soils of Qianjin County, China Using An Advanced Finite Mixture Distribution Model Integrated with Positive Matrix Factorization

Abstract

Characterization of heavy metal sources is essential to assess their potential contamination pathways and environmental risks. This study proposes an advanced approach that integrates finite mixture distribution model (FMDM), positive matrix factorization (PMF), cluster analysis (CA), and indicator kriging (IK) to identify the quantitative sources and hazardous areas of heavy metals (HMs) in soils. The conventional and proposed integrated approach were applied to a heavy metal dataset from Qianjin County, China, which belongs to a geological transition zone with drastic variation in HMs concentrations. The conventional FMDM-PMF approach failed to draw valid results due to discrepancies between FMDM and PMF, whereas the advanced approach successfully identified the sources of heavy metals by dividing the study area into two sub-regions. The hazardous areas were mapped using IK based on determined thresholds, revealing a high risk of Cd and Hg contamination. The results highlighted the urgent need to limit the input of Cd from agricultural activities and the input of Hg from industrial emissions into agricultural soils in the study area. Overall, the integrated approach proved to be an effective method for source apportionment and delineation of heavy metal-contaminated areas in soils with great spatial variability.

Keywords: heavy metal, Identification of sources, FMDM, PMF, Mapping of hazardous areas