Integrating source-water connectivity into SWAT for hotspot-oriented nitrogen management in hilly catchments

Nitrogen pollution from agricultural and urban source remains a critical challenge for water quality management in hilly catchments. Hotspot identification is essential for prioritizing mitigation strategies. In this study, we integrated a Source-Water Connectivity Index (SWCI) into the Soil and Water Assessment Tool plus (SWAT+) model to enhance the assessment of nitrogen load distributions. The results demonstrated that the SWCI-SWAT+ model exhibited superior performance in simulating nitrate nitrogen (NO₃^--N) and total nitrogen (TN) loads, while the conventional SWAT model overestimated the hotspot areas for NO₃^--N and TN loads by 17.89% and 1.29%, respectively. Land-use analysis indicated that tea plantations were the dominant contributors, accounting for 60.76% of NO₃^--N and 68.64% of TN loads, while steep slopes (>40°) contributed 56.92% and 46.90%, respectively. Soil texture analysis indicated that high sand-content soils dominated NO₃^--N loads (56.79%) and medium sand-content soils contributed most to TN loads (54.80%). Spatially, NO₃^--N hotspots clustered in the western basin due to tea and orchard cultivation, whereas TN hotspots extended to the southern basin, influenced by urban sewage discharge. The analysis identified four spatial patterns linking nitrogen load and concentration. Spatially, high NO₃^--N load and concentration occurred in Sub-basin 1, while high TN levels extended to Sub-basins 5 and 8. The SWCI-SWAT+ framework offers an approach for non-point source pollution control by identifying pollution hotspots and diagnosing mismatches in the load-water quality relationship at the watershed scale.