Accurately identifying nitrogen (N) and phosphorus (P) pollution sources is key to managing water pollution in intensive agricultural watersheds, but the difficulty of tracing nonpoint source pollutants in watersheds is increased due to the combined effect of multiple sources. The typical intensive agricultural watershed of Xingyun Lake was chosen as an example, collected multiple pollution sources in the watershed, multiple isotopes ( δ 15 N-NO 3 , δ 18 O-NO 3 , δ 15 N-NH 4 and δ 18 O-PO 4 ) and analysis methods (SIAR model, probability statistical analysis and perturbation method) were employed to quantify source contributions to nitrate (NO 3 - ), ammonium (NH 4 + ) and phosphate (PO 4 3- ) in the main rivers entering the lake under seasonal changes, and source apportionment accuracy uncertainty and sensitivity were evaluated. The results revealed that domestic sewage (CE) was the main N source for NO 3 - (mean proportion: 31 ± 9%) and NH 4 + (28 ± 7%) in rivers during the rainy season (RS) and dry season (DS), while PO 4 3- originated from P fertilizer (24 ± 9%), livestock manure (22 ± 6%), and soil P (28 ± 6%), CE (26 ± 8%). The nitrification of NH 4 + during the DS sewage provided a "new" NO 3 - source for the watershed. The CE and soil N and P contributions to NO 3 - , NH 4 + and PO 4 3- in rivers exhibited high uncertainty, especially during the DS (mean UI 90 value of CE: 0.60). Sensitivity analysis showed that the relative changes in the contributions of CE to NO 3 - , NH 4 + and PO 4 3- in river water were largest. The results provide new insights for precise N and P pollution control in intensive agricultural watersheds.
Chen et al. (Tue,) studied this question.