[Self-organizing Map （SOM） and Positive Matrix Factorization （PMF） Models for Source Apportionment of Metals in the...

Evaluating the metal pollution status and ecological risks in wetland sediments, as well as identifying the sources, is essential for safeguarding the ecological functionality and sustainable development of wetlands. This study examined a coastal wetland park, converted from a dike fishery and farmland, to assess potential metal pollution. Within the park, 26 surface sediment samples were collected, and the concentrations of metals （Li, Be, V, Cr, Co, Cu, Zn, Cd, Pb, As, and Mn） were analyzed. Using the geo-accumulation index, pollution load index, and potential ecological risk index, we assessed metal pollution in the area. Additionally, the sources and contributions of metals were deciphered through correlation-clustering analysis, self-organizing map （SOM）, and positive matrix factorization （PMF）. The results indicated： ① The average concentrations of metals in the sediment exceeded the background values of soils in Fujian Province, with Cd and Zn showing significant variability and evident anthropogenic influence. ② The pollution assessment results revealed that the park generally exhibited low to moderate pollution level, Li and Zn were identified as the most accumulated metals, while Cd presented a higher potential ecological risk. ③ The combination of correlation-clustering, SOM, and PMF categorized the sources of the 11 metals into three groups： Cr, Cu, Zn, and Pb were predominantly from traffic sources （29.72%）； Li, Be, Mn, V, Co, and As were mainly from natural sources （44.89%）； and Cd and Zn were substantially from agricultural activities （25.39%）.