三江源区地处青藏高原腹地，是黄河、长江与澜沧江的源头地区，被誉为“中华水塔”，影响着下游地区约7亿人口的生存和发展，在中国乃至整个东南亚的水资源安全和生态环境保护中都发挥着极其重要的作用。近几十年，全球气候变暖以及土地利用变化对该源区的水文过程产生了重要影响，研究变化环境下三江源区的径流响应对于实现区域水资源的合理管理和可持续发展战略目标具有重要的科学指导意义。然而，高海拔流域的水文过程错综复杂，目前对该源区径流响应的机理分析还存在不足。同时，作为全球气候变化的敏感区，有必要对该源区未来水资源的演变趋势展开研究。鉴于此，本文耦合大尺度陆面水文模型VIC（Variable Infiltration Capacity）与度日冰川算法（耦合后的模型称为VIC-Glacier），以黄河、长江与澜沧江源区内7个子流域为重点，实现对各子流域1986–2100年0.25° × 0.25°分辨率逐月径流模拟，在此基础上综合分析各流域总径流和三大径流组分（降雨径流、融雪径流和冰川径流）的历史演变特征和未来变化规律，量化三大径流组分对总径流的贡献。本论文的主要研究结果如下：

（1）量化了三江源区各流域历史径流变化及三大组分贡献，发现1986–2018年，黄河和长江源区各流域总径流呈显著增加趋势（20.3–32.5 mm/10a），澜沧江源区无明显变化。三江源区各流域均以降雨径流为主（总径流占比85–95.5%）。从年内分布来看，融雪径流主要发生在4–6月，冰川径流主要发生在7–8月。

（2）基于N维概率密度函数转换的多变量联合校正方法MBCn（Multivariate Bias Correction）可以有效提升气候模式在三江源区的模拟性能。在此基础上探究了三江源区未来气候的时空演变规律，发现2019–2100年，三江源区各流域持续增温增湿。具体而言，年均气温、日最高气温、日最低气温和降水量分别以0.09–0.72°C/10a、0.11–0.71°C/10a、0.09–0.77°C/10a和3.9–33.8 mm/10a的速率显著增加。相较于1986–2018年基准期，降水在未来近期2021–2050年变化幅度较小，在未来远期2071–2100年的不同情景下分别增加8–11%（SSP1-2.6）、14–18%（SSP2-4.5）和21–32%（SSP5-8.5）。从季节尺度来看，降水和气温的增加主要发生在冬春季（11月–次年3月），而在未来近期，降水预估在6–9月份会减少。

（3）明晰了三江源区各流域未来总径流及其三大组分的变化规律。2019–2100年三江源区各流域总径流呈显著增加趋势（2.9–21.5 mm/10a）。相较于基准期，三江源区各流域总径流在未来近期约减少2–18%；未来远期黄河源区各流域总径流增加0.3–8.5%，长江源区增加15–45%，澜沧江源区增加0.6–28.6%。从季节尺度来看，相较于基准期，三江源区未来近期各流域总径流在7–9月份减少（0.5–11.5 mm）；未来远期，长江和澜沧江源区各流域总径流在7–9月份增加（0.1–16.1 mm），而黄河源区减少（0.1–6.4 mm）。2019–2100年三江源区各流域降雨径流对总径流的贡献以平均速率0.11–1.0%/10a逐渐增加，未来各流域总体以降雨径流为主导。

Located in the hinterland of the Qinghai-Tibet Plateau, the Three-River Source Region (TRSR) serves as the origin of the Yellow, Yangtze, and Lancang Rivers, earning it the title of “China’s Water Tower”. This designation highlights its critical role in supporting the livelihoods of approximately 700 million downstream inhabitants and in safeguarding water resources security and ecological environment protection across China and of Southeast Asia. In recent decades, the TRSR has experienced significant impacts from global warming and changes in land use. Understanding the runoff response within this region amidst changing environmental conditions is pivotal for steering the water resources management towards sustainable development goals. However, comprehending the complex hydrological process of high-altitude watershed, particularly within the TRSR, remains a great challenge. Moreover, given its sensitivity to global climate change, investigating the future projections of water resources in the TRSR is imperative. To address these concerns, this study employed the VIC (Variable Infiltration Capacity) hydrologic model coupled with the degree-day factor algorithm, known as VIC-Glacier. The study focused on seven sub-basins in the source regions of the Yellow, Yangtze and Lancang Rivers, and simulated the monthly runoff of each sub-basin with the resolution of 0.25° × 0.25°during 1986–2100, conducting a comprehensive analysis of historical runoff variation and corresponding future projections, including contributions from rainfall, snowmelt and glacier melts. The main results of this study are as follows:

(1) The historical runoff variation and the contributions of the three major components in the TRSR were quantified. The total runoff in the source regions of the Yellow and Yangtze River showed a significant increasing trend (20.3–32.5 mm/10a) during 1986–2018, while there was no significant change in the source region of the Lancang River. Rainfall runoff accounted for 85–95.5% of the total runoff amount. Seasonally, snowmelt runoff mainly occurred from April to June, and glacier runoff primarily occurred from July to August.

(2) The Multivariate Bias Correction algorithm (MBCn), based on N-dimensional probability density function transform, effectively improve the simulation performance of global climate model in the TRSR. On this basis, the temporal and spatial evolution law of the future climate in the TRSR was explored. During 2019–2100, the temperature and humidity of each basin in the TRSR will continue to increase. Specifically, the annual mean temperature, daily maximum temperature, daily minimum temperature and precipitation in the sub-basins of the TRSR have shown significant increases. Specifically, the rates of increase are projected by 0.09–0.72°C/10a, 0.11–0.71°C/10a, 0.09–0.77°C /10a and 3.9–33.8 mm/10a, respectively. Compared with the base period of 1986–2018, precipitation is projected to exhibit slight changes in the near future (2021–2050), with anticipated increases of 8–11% (SSP1-2.6), 14–18% (SSP2-4.5) and 21–32% (SSP5-8.5) in the long term (2071–2100). Seasonally, precipitation and temperature increases mainly occur in winter and spring (November to March of the following year), while a decrease in precipitation is expected from June to September in the near future.

(3) The variations of the future total runoff and its three major components in the TRSR were clarified. During the period 2019–2100, the total runoff of each basin in the TRSR is projected to exhibit a significant increasing trend (2.9–21.5 mm/10a). Compared to the base period, a decrease of 2–18% in the total runoff of each basin is expected in the near future. In the long term, total runoff in the source regions of the Yellow, Yangtze and Lancang Rivers is projected to increase by 0.3–8.5%, 15–45%, 0.6–28.6%, respectively. Seasonally, compared to the base period, a decrease in total runoff is expected from July to September (0.5–11.5 mm) in the near future. In the long term, total runoff in the source regions of the Yangtze and Lancang River is projected to increase from July to September (0.1–16.1 mm), while a decrease is expected in the source region of the Yellow River (0.1–6.4 mm). During the period 2019–2100, the contribution of rainfall runoff to total runoff in the TRSR is expected to gradually increase at an average rate of 0.11–1.0%/10a, with rainfall runoff emerging as the dominant component in the future.