以黄河源区白河流域为研究对象，采用Mann-Kendall方法分析水文要素的演变特征，利用RCCC-WBM水量平衡模型（monthly water balance model developed by research center for climate change）模拟探究流域生态水文要素对气候变化的敏感性。结果表明：1981—2020年白河流域气温增加趋势显著，气温递升率为0.45 ℃/（10 a），潜在蒸散发年际变化不明显，而降水和径流呈现先减少后增加的阶段性变化特征；RCCC-WBM水量平衡模型对白河流域具有较好的月尺度径流模拟效果，率定期和验证期的纳什效率系数（Nash-Sutcliffe efficiency coefficient，ENS）分别为0.69和0.62；在假定的气候变化情景下，降水量增加40%且气温变化?4 ～4 ℃时对应的径流量变化范围为20.2%～84.4%，实际蒸散发量为1.4%～51.0%，即随着降水量的增多，流域气温和蒸散发变化对径流的影响越来越显著。研究结果可为黄河源区的水循环演变规律的揭示提供参考。

The source area of the Yellow River is an important component of the “Chinese Water Tower”, which has the functions of water conservation and recharge, regulating the local regional climate, and plays a positive role in maintaining the water resources and ecological environment of the Yellow River basin. Meanwhile, as a "sensitive area" of climate change, the hydrological situation of the Yellow River source area on the Tibetan Plateau has attracted much attention. Currently, most of the studies on the response of the Yellow River source area to climate change are based on large-scale watersheds and single meteorological and hydrological elements, and there are relatively limited studies on the mechanisms between hydrological elements in typical watersheds in the source area, while the changes in hydrological processes at the sub-basin scale are crucial for future water resources planning and management. Based on the meteorological and hydrological data collected and collated from 1981 to 2020 in the Bai River basin of the Yellow River headwaters area, the hydrological sequence analysis method was used to analyze the interannual change process of hydrological elements; the parameters of the RCCC-WBM model were rate-set, and the runoff of the Bai River basin was simulated in phases for the years from 1981 to 2000 and from 2001 to 2020, respectively. The parameters of the RCCC-WBM model were calibrated to simulate the changes of runoff, actual evapotranspiration (ET) and soil water volume in from 1981 to 2000 and from 2001 to 2020 in the Bai River basin in stages; finally, the sensitivity of runoff, actual ET and soil water volume to climate change in the Bai River basin was analyzed by moderating the changes of precipitation and temperature on the basis of the available data. From 1981 to 2020, the average annual temperature in the Bai River basin showed a significant increasing trend, with an average temperature increase rate of 0.45 ℃ per decade, consistent with the background of global warming. Annual precipitation was relatively low from 1997 to 2010, with a slight rebound trend after 2010, although it was not significant. The annual potential evapotranspiration trend was similar to the annual average temperature, but the overall increasing trend was not significant. The interannual variation of annual runoff depth showed a non-significant decreasing trend, with a period of drought from 2001 to 2008. Precipitation in the Bai River basin peaked in June and September, and runoff, like precipitation, also exhibited a bimodal pattern, albeit with a time lag. The RCCC-WBM model, using meteorological and hydrological data, effectively simulated the runoff process in the Bai River basin, with Nash-Sutcliffe model efficiency coefficients (ENS) of 0.69 during the calibration period and 0.62 during the validation period. Under assumed climate change scenarios, with a 40% increase in precipitation and temperature change ranging from ?4 ℃ to 4 ℃, the corresponding runoff change ranged from 20.2% to 84.4%, and actual evapotranspiration change ranged from 1.4% to 51.0%. This indicated that with increased precipitation, the impact of temperature and evapotranspiration changes on runoff became more significant. With constant temperature, when precipitation changed from 0 to 20%, the change in soil moisture was 13.3%, while with precipitation changes of 20% to 40%, the change in soil moisture was 12.1%. This showed that as precipitation increased, soil moisture also increased, but the change in soil moisture per unit of precipitation became smaller. The RCCC-WBM model simulated the annual distribution of hydrological elements in the Bai River basin for the periods from 1981 to 2000 and from 2001 to 2020. In July, approximately 90% of evaporation and runoff were derived from precipitation. In December, only about 20% of evaporation and runoff came from precipitation, with the remainder supplemented by groundwater runoff and snowmelt runoff. Compared to the earlier period, after 2001, there was an increasing trend in hydrological elements primarily in April and May, while July and December showed a decreasing trend.