Abstract:A typical suburban farmland of Nanjing city was selected as a case area. Positive matrix factorization (PMF) model was applied to identify the main sources of heavy metals in the topsoils. Correlation between measured and predicted heavy metals were analyzed by Linear Regression in order to investigate the accuracy and feasibility of PMF model. The results were also compared with those obtained through Pb stable isotopic technique, which could further verify the PMF ones. The results indicated that: (1) Mean values of Cd, As, Hg, Pb, Cu, Zn, and Cr in farmland topsoils were greater than their respective soil background values in Nanjing. Concentrations of Cd and Cu in some soil samples were greater than the GradeⅡof the Standard according to the Soil Environmental Quality of China. (2) Accumulation of the heavy metals in soils were related to the overuse of fertilizers for agricultural production for several years, surrounding industrial production, traffic pollution, and soil parent material. (3) The results of the correlation analysis implied that r2values of Cd, Hg and Cr were more than 90%, besides the r2values of other heavy metals were more than 70%, which showed that there were significant correlation between the values of the measured and predicted heavy metals. (4) The contribution of atmospheric deposition to heavy metal accumulation in the soils through PMF model and Pb stable isotopic technique were 32.1% and 36.8%, respectively, which firmly verified that PMF model can be effectively applied to source apportionment of heavy metals in the soils.