RESUMEN
Aiming at non-point sources pollution in the agricultural areas with large topographic fluctuations and spatial differences in precipitation, a SWAT model was used to evaluate the spatial variations in the critical source areas (CSAs) of total nitrogen (TN) and total phosphorus (TP) under two precipitation scenarios, i.e., heterogeneous precipitation and uniform precipitation. A change in the CSAs identified based on the two precipitation scenarios during the study period were statistically calculated, and the relationship between the CSAs and precipitation variables was discussed. The study results showed that when the total precipitation was the same, the variation tendency of the identified CSAs for TN and TP under the two precipitation scenarios were similar, and very close for a few years. According to the results of the pair t test, the CSAs of TP were not affected by the spatial variation of precipitation, while the change in CSAs for TN was more significant under different precipitation scenarios, which is likely due to the difference in the physical properties of nitrogen and phosphorus. The correlation analysis between the CSAs of TN and TP with precipitation variables showed that the variation in the CSAs of TP was positively correlated with the precipitation variables in the same year, while the variation in the CSAs of TN was strongly related to the precipitation variables of the previous year. The results obtained in this study are of great significance for further exploring the impact of uncertainty of precipitation, which is an important driving factor, on the CSAs of non-point sources pollution and the governance of agricultural non-point sources pollution.