Abstract / Summary
Abstract Source attribution of antibiotics in stream sediments remains challenging because similar compounds can originate from multiple sectors and their environmental transformation may alter source-specific compositional patterns after release. This study evaluated whether heavy metal(loid)s could provide complementary information for PMF-based antibiotic source apportionment. Sediment samples from 12 sites in the Muhan Stream watershed were analyzed for 11 antibiotics and 10 heavy metal(loid)s, of which 9 antibiotics and all 10 heavy metal(loid)s were detected. A PMF model based on the 9 detected antibiotics was directly compared with an integrated model incorporating both antibiotics and heavy metal(loid)s. The antibiotic-based model produced a five-factor solution, but several site-specific factor contributions were not fully consistent with the spatial distribution of known pollution sources. The integrated model retained comparable diagnostic performance and stability while providing additional multielement contrasts that reduced ambiguity among source-related factors. In particular, factors associated with the wastewater treatment plant, livestock manure treatment facility, and livestock farms showed more distinct site-specific contribution patterns and greater spatial coherence with known source distributions. These results indicate that heavy metal(loid)s can complement antibiotic profiles and support clearer PMF-based source interpretation. Factor-attributed screening analysis further indicated that manure-related factors accounted for relatively large shares of the estimated ecological risk. This approach may provide a practical basis for source-oriented pollution management in watersheds affected by multiple pollution sources.