Source apportionment and risk assessment of soil heavy metals (HMs) are essential for pollution control. However, inherent limitations in receptor models hinder accurate source apportionment, impacting outcomes of source-oriented risk assessment. A hybrid model was employed by combining two receptor models, absolute principal component score/multiple linear regression (APCS/MLR) and positive matrix factorization (PMF) models. Four primary pollution sources were identified, followed by an assessment of source-oriented environmental and human health risks in the Huangshui River Basin. Results revealed that Cr, Cd, and Ni average concentrations surpassed their background by 3.1, 2.1, and 1.9 times, respectively. Additionally, 54.29%, 13.34%, and 18.09% of Igeo values for Cr, Cd, and Ni were classified as “moderately contamination” or higher. Natural sources significantly influence Cu (90.2%) and As (71.4%). Cr (63.4%) and Ni (77.9%) mainly originated from agriculture and industry, respectively. Transportation sources emerged as the primary contributors to Pb (59.6%), Zn (48.6%), and Cd (47.2%). The environmental risk level in this study area remained acceptable. Ecological risk was mainly attributed to industrial activities (46.6%) and transportation (37.6%), with Cd being the predominant metal responsible for this risk. Although the noncarcinogenic risk was negligible for all populations, the carcinogenic risk demands attention, particularly concerning children. Industrial sources (67.0%) and Ni were identified as the main contributors to the carcinogenic risk. This study represents an attempt to develop a hybrid model, providing an effective approach to combine models for the accurate apportionment of pollution sources and the reduction of risks.
Read full abstract