Despite the implementation of various field-level measures to mitigate phosphorus loss from agricultural land, effective control at the watershed scale remains challenging due to the complexity of influencing factors. To address this gap, an improved phosphorus surplus index was incorporated into the traditional index method as the primary source factor. Using Erhai Lake in China as a case study, a phosphorus loss risk assessment system tailored to cropping systems in plateau lake basins was developed. Control strategies for reducing phosphorus loss were evaluated using machine learning techniques, scenario analysis, and runoff plot experiments. In the vegetable continuous cropping system, 73% of the area was classified as high-risk, which is 709.2% higher than that of the rice-rapeseed rotation system. The primary factors influencing phosphorus loss risk were the distance between phosphorus sources and nearby rivers, followed by soil erosion. Optimized fertilization (reducing nitrogen and phosphorus inputs by 25% and increasing potassium input by 25% relative to conventional practices) significantly reduced phosphorus loss risk in the rice-rapeseed rotation system. In contrast, optimized fertilization had a limited effect in the vegetable continuous cropping system, where the most effective control measure was establishing a 500 m fertilizer-free buffer zone along riverbanks. These results provide a scientific basis for nutrient management and non-point source pollution control and offer a useful reference for developing phosphorus loss risk assessment systems in other lake basins.
Li et al. (Thu,) studied this question.