PMF model combined with stable isotope technology to track heavy metals accumulated in farmland soils

Source apportionment of heavy metals in farmland soils is necessary to ensure the safety of agricultural products. However, the source apportionment methods have not been effectively evaluated. Here, we took a polluted farmland in Jiaxing city, a typical plain river network area located in the Yangtze River Delta of China, as an example, and developed positive matrix factorization (PMF) model and stable isotope technology (SIT) to identify the sources of heavy metals. The concentrations of Cu, Ni, Cr, Zn, Pb, Cd, As, and Hg in cultivated layer soils were 19.0–38.0, 24.0–48.0, 49.0–71.0, 55.0–85.0, 21.8–51.1, 0.01–0.16, 4.45–11.3, and 0.09–0.68 mg/kg, respectively. Hg exceeded the risk screening value with an exceeding rate of 12.5 %. Heavy metals in farmland soils originated from industrial production (32.0 %), natural source (28.2 %), agricultural activity (25.8 %) and vehicle emission (14.0 %). However, for Hg, only industrial production (37.1 %) and agricultural activity (62.9 %) were identified, and the contributions of natural source and vehicle emission were not obtained from PMF. Hg stable isotope ratios can compensate for the incompleteness of PMF, with agricultural activity contributing the most (51.6 %), followed by natural source (20.9 %), industrial production (15.4 %) and vehicle emission (12.1 %). 79.1 % of Hg was affected by external inputs with agricultural activity accounting for 65.2 %, which was supported by the input of Hg from potential sources (agricultural activity 65.0 %). PMF may have limitation in the source apportionment of specific heavy metal, and the combined application of SIT can solve the limitation.