An improved SWAT-based computational framework for identifying critical source areas for agricultural pollution at the lake basin scale

Soil and Water Assessment Tool (SWAT) model has been successfully used to identify critical source areas (CSAs) at the lake basin scale. However, when employing the well-mixed river channel model, SWAT has difficulty in decomposing the pollutant load at watershed outlet in terms of origin location. In previous studies, the river migration process is not included in the CSAs identification framework. The identification criterion for CSAs mainly based pollutant load releasing into river rather than the final pollutant load flowing into lake. This study is used to develop a new computational framework to address this problem. The SWAT model was used to offer a linkage between pollution input in the upstream watershed and pollutant load response at the watershed outlet. Simulation test was utilized to investigate the pollution contributions from different agricultural activities and from different villages. Combined with the agricultural production statistic data, the pollution contribution from each village was separately calculated, and the cumulative curve of pollution distribution was thus drawn to identify CSAs. We illustrated this method in the Lake Erhai basin, Yunnan province, China. CSAs for N pollution were identified, and a comparison with the existing method was made. The quantitative results indicate that there is a large gap between the pollutant loads released to river and flowed to lake. The river migration process was vital for N degradation so as to significantly influence the final CSAs map. The new method improved the ability of supplying a more comprehensive delineation of CSAs.

► We proposed an improved SWAT-based computational framework of CSAs identification at the lake basin scale.
► Comparison results illustrated that the river migration process significantly influences the CSAs map.
► Both the pollutant releasing process and migration process should be included in the CSAs identification framework.