【目的】将SWAT(soil and water assessment tool)模型运用于重庆三峡库区长江支流的流域径流过程研究，为流域水资源的合理调控提供理论依据。【方法】以位于重庆市西部的璧南河流域为研究对象，构建SWAT模型对该流域的径流过程进行模拟。【结果】1) 率定期与验证期流域月均径流量模拟值与实测值吻合度较好，率定期和验证期的Nash-Suttcliffe效率系数、确定性系数和相对误差接近一致，表明SWAT模型在璧南河流域具有很好的适用性。2) 2013—2021年璧南河流域年平均径流总量为1.79×108 m3，多年平均径流深为346.04 mm。各子流域产流主要集中在汛期(6—10月)，非汛期产流量不足；季节产流量分配不均匀，夏秋季节产流量高，春冬季节产流量低；流域产流量整体呈下降趋势。3) 流域径流深在空间上呈现东西两侧高、中间低的分布特征，多年径流深变化率呈现出高产流量区域的产流量变化率低、低产流量区域的产流量变化率高的空间分布规律；产流量高值区主要集中在吴滩、三合、广普、建龙等镇(街道)，产流量低值区主要位于陈食、丁家、正兴、来凤、青杠等镇(街道)，与流域地势两侧高中间低以及坡度、降水量两侧大中间小的空间分布特征相符。4) 在模拟未来气候变化情景中，璧南河流域降水量变化与径流量变化呈正相关关系，与气温变化呈负相关关系；与气温变化相比，流域径流量对降水量变化的敏感性更高，且降水量是影响未来璧南河流域径流变化的主要因素。【结论】SWAT模型可用于璧南河流域径流过程的模拟；可以考虑通过水土保持生态建设、水存储和调水基础设施建设、农田水利建设等措施提高流域水资源利用率，同时调整和优化流域土地利用结构，从而实现未来气候变化背景下璧南河流域水资源的科学管理。
［Purposes］In order to apply SWAT model to the study of runoff process in Binan River Basin and provide theoretical basis for the rational regulation of water resources in the basin. ［Methods］Taking the Binan River Basin, a primary tributary of the Yangtze River on the left bank as an example, the SWAT model was constructed to simulate the runoff process of the basin. ［Results］1) The simulated value of monthly runoff in the validation period and the validation period is in good agreement with the measured value, with the validation period R2=0.81, Ens=0.80, |Re|=6.36%, the validation period R2=0.90, Ens=0.82, |Re|=7.85%, which shows that the SWAT model has good applicability in the Binan River basin. 2) The total annual average runoff of the basin from 2013 to 2021 is 1.79×108 m3, multi-year average runoff depth is 346.04 mm, and the runoff of each sub-basin is mainly concentrated in the flood season from June to October. The runoff in non-flood season is insufficient, and the distribution of seasonal runoff is uneven. The runoff in summer and autumn is high, while the runoff in spring and winter is low, and the overall runoff is in a downward trend. 3) the runoff depth of the basin is characterized by high in the east and west and low in the middle in space. The change rate of the runoff depth over the years shows the spatial distribution pattern of low change in the high-yield flow area and high change in the low-yield flow area. The high-yield area is mainly concentrated in Wutan, Sanhe, Guangpu Jianlong and other towns or subdistricts. The low-yield area is mainly located in Chenshi, Dingjia, Zhengxing, Laifeng, Qinggang and other towns or subdistricts, which is related to the low, middle and high gradient on both sides of the basin terrain. The distribution characteristics of large precipitation on both sides and small precipitation in the middle are consistent. 4) Under the background of future climate change, the sensitivity of runoff to precipitation change is higher than that of temperature change. When the precipitation remains unchanged, the average annual runoff depth decreases by 4.39 mm every time the temperature increases by 1 ℃. When the temperature decreased by 1 ℃, the average annual runoff depth increased by 3.37 mm. When the temperature remained unchanged, the annual runoff depth decreased by 6.73 mm for every 1% decrease in precipitation. Watershed precipitation increases by 1%, the average annual runoff depth increases by 9.13 mm, precipitation is the main factor affecting the future runoff change in Binan River basin, so it can carry out related soil and water conservation projects, optimize the land use structure, improve the utilization rate of water resources to achieve the future climate change under the scientific management of water resources in the basin. ［Conclusions］The distributed hydrological model SWAT can be used to simulate the hydrological process of the Binan River basin, and the research results can provide certain theoretical guidance for the management and control of basin water resources.