Optimal tracer identification for dissolved organic matter (DOM) source tracking in watersheds using point source effluent load data

In this study, we characterized the optical and molecular weight (MW) properties of dissolved organic matter (DOM) with various sources in an agriculture-forestry watershed. We proposed a guideline to identify optimum DOM source tracers for downstream rivers during both rain and non-rain events, utilizing the load of dissolved organic carbon (DOC) from point sources. Six descriptors were pre-selected based on established criteria in the literature, and fifteen pairs of these descriptors were evaluated for their applicability in end-member mixing analysis (EMMA). The results from EMMA provided inconsistent estimates of relative contributions from DOM sources across the fifteen pairs, with optical descriptors outperforming MW-based descriptors and their combinations. The optimal source tracers were determined by comparing relative contributions of DOM from upstream effluent wastewater using DOC load ratios calculated from on-site monitoring data and predictions based on EMMA. The pair of optical descriptors, HIX (humification index) and BIX (biological index), closely matched the measured load ratios with minimal discrepancies (0.4 ± 0.4 %). According to the EMMA results using pairs of HIX and BIX, non-rain events were primarily influenced by oil-cake fertilizer and treated effluent wastewater, while rain event samples were dominated by manure and soils. These findings offer insights into managing non-point organic pollution sources in agricultural-forestry watersheds, contributing to our understanding of carbon and nutrient cycling in aquatic systems. Notably, this study proposes a validation guideline that employs load ratios of point sources, such as effluent wastewater, to enhance source tracking accuracy.