Quantitative source identification and apportionment of heavy metals under two different land use types: comparison of two receptor models APCS-MLR and PMF.

At present, many researchers are increasingly aware of the importance of using models to identify heavy metal (HM) pollution sources. However, on the performance and application of different source identification models to HMs under different land use types had been studied little. In this study, comparison of absolute principal component scores-multiple linear regression (APCS-MLR) and positive matrix factorization (PMF) models and their application characteristics in identifying pollution sources were carried out by using 11 HMs in Zhongwei City farmland and Shizuishan industrial park, Ningxia. The results indicated that HM pollution in farmland mainly came from pesticides, fertilizers, and deposition of the Yellow River, while the pollution in industrial park mainly originated from atmospheric deposition and various industrial productions. The APCS-MLR model had the problem of less identification sources and the difficulty to explain the complex pollution, while the PMF model not only identified more pollution sources, but also distinguished heavy metal-related sources for two different land use types and different industrial production conditions. It is of great significance the formulation of agricultural-related pesticides' and chemical fertilizers' rational use and various industrial production-related raw materials put in and emission control strategies.