Abstract:In this study, the positive definite matrix factorization (PMF) receptor model and Monte Carlo simulation were used to evaluate the heavy metal pollution and health risk in the soil of a smelter. The data show that the contents of As, Cu, Pb, Ni and Zn in soil are 4.6, 179.6, 45.31, 8.6 and 38.9 times of the soil background value in Hubei Province, respectively, and the ground accumulation index is judged as light pollution. The ecological risk assessment showed that 89% of Cu and 32% of As were at extremely high risk level. PMF analysis shows that industrial sources are the main pollution sources (40%), followed by natural sources and mixed sources (33% and 27% respectively). The health risk assessment (HRA) shows that the cancer risk of adults and children is acceptable, and the main contribution element is Ni. Among the non-carcinogenic risks, the probability of adults exceeding the threshold is 6.32%, and that of children is 49.5%, which is mainly caused by Pb. Therefore, As is determined to be the priority control factor. Parameter sensitivity analysis shows that skin contact and oral intake are the main contact routes between adults and children, and skin adhesion coefficient and body weight have significant effects. The combination of the two models provides an accurate basis for soil pollution prevention and control.