Identification of multilevel priority management areas for diffuse pollutants based on streamflow continuity in a water-deficient watershed

Agricultural diffuse pollution is, on the global scale, one of the most important causes of surface water quality deterioration and eutrophication. However, the identification of agricultural priority management areas (APMAs) for diffuse pollutants is still an ongoing challenge in water-deficient watersheds. To enhance the accuracy of identification of APMA, a modified framework for APMA identification is proposed by combining the streamflow continuity index constructed by Reliability-Resilience-Vulnerability (R-R-V) indices and the SWAT model in this paper. This novel framework was tested in Zhang River Watershed and the main results are as follows. First, the SWAT model is calibrated and validated with a good performance of the Nash-Sutcliffe efficiency (NSE) and coefficient of determination (R2)greater than 0.75 and 0.80 for both the calibration and validation periods, respectively. Second, the wet year is found to be the critical period for diffuse TN and TP output at an annual scale. Specifically, the critical loss period of TN is identified as from April to October on the monthly scale, especially in July and August. However, the critical period of TP loss is mainly concentrated in the range from July to September in Zhang River Watershed. Third, the reaches with a high streamflow continuity index(SCI) are primarily located in mid-downstream part of the watershed. The reaches are mostly characterized by relatively low SCI values due to the lack of streamflow volume in the upper reaches. Fourth, the APMAs for TN and TP for the watershed final outlet connected to the Yuecheng Reservoir are approximately distributed in two areas by considering the SCI estimation, namely the Qingzhang River sub-watershed and the mid-downstream part of the Zhuozhang River sub-watershed. Compared with the results obtained from the modified APMA framework, the main areas of TN and TP APMAs may be overestimated by 27.17 and 10.02 percentage points, respectively, using the original APMAs framework without the SCI evaluation. This revised APMA framework can be easily extended to other water-deficient watersheds for improving the efficacy of the identification and the control of agricultural diffuse pollution.