Runoff Simulation of Yulong Kashi River Basin Based on CN05.1 Dataset Driven SWAT Model

To investigate the simulation performance of the SWAT model driven by the CN05.1 dataset in areas lacking meteorological stations and the response of basin runoff to climate change, the Yulong Kashi River Basin is selected as the study area. This paper uses the aforementioned dataset to drive the SWAT model to conduct monthly runoff simulations of the basin from 1991 to 2020. On this basis, 25 climate scenario models are set up, and runoff is simulated by importing meteorological data under these 25 climate scenarios into the models. The results show that the CN05.1+SWAT model is suitable for runoff simulation in the Yulong Kashi River Basin, with good agreement between the simulated results and the measured data. The coefficient of determination (<italic>R²</italic>), Nash-Sutcliffe efficiency coefficient (<italic>C</italic>_NS), and percentage bias (<italic>σ</italic>_P) of both calibration and validation period meet the evaluation criteria for runoff simulation. <italic>R²</italic>, <italic>C</italic>_NS and <italic>σ</italic>_P during the calibration period (1991—2005) are 0.85, 0.86, and -10.35%, respectively. During the validation period (2006—2020), they are 0.76, 0.79 and -14.75%, respectively. The CN05.1 dataset provides continuous meteorological data with high resolution in a long time series for hydrological simulations in the Yulong Kashi River Basin, which lacks meteorological observations. Therefore, this dataset can be used for future research and in other applications. The CN05.1 dataset can accurately reflect the underlying surface and surface atmospheric characteristics of the Basin. Annual runoff in the Basin shows a positive correlation with both precipitation and temperature, whose change has a more significant impact on runoff. The relationship among precipitation, temperature, and runoff can be expressed by a plane function, with an <italic>R²</italic> of 0.995 7, variance of 0.000 2, and root mean square error of 0.012 3.