Uncertainty assessment in baseflow nonpoint source pollution prediction: The impacts of hydrographic separation methods, data sources and baseflow period assumptions

Effective management of nonpoint source (NPS) pollutants requires an understanding of their potential surface runoff and baseflow pathways. In this study, an integrated method was developed for separating baseflow and surface NPS pollutants at the watershed scale. Then, the uncertainty of the prediction of baseflow NPS pollutants was explored by comparing four hydrographic separation methods (the United Kingdom Institute of Hydrology (UKIH), the Lyne-Hollick (LH) method, the Eckhardt (ECK) method and the Soil and Water Assessment Tool (SWAT)), three data sources (monitoring data, SWAT and the LOADEST model) and two baseflow period assumptions. A real case study was performed in a typical catchment, the Three Gorges Reservoir Area, China. The results showed that the baseflow annually contributed 30.1% and 23.0% of the total nitrogen (TN) and phosphorus (TP) load, respectively, indicating that baseflow was identified as the key means of transport of NPS pollutants, especially for NPS-N during the dry season. The water quality data source represents the largest uncertainty source in baseflow NPS pollutant estimation, and the simulated baseflow loads of TN and TP were the largest when LOADEST and SWAT were used as water quality data sources, respectively. The baseflow NPS estimations were more uncertain for TN during the non-dry season and TP during the dry season. The results of this study could provide implications for the prediction and management of NPS pollutants at the watershed scale, especially for groundwater-polluted catchments.