流域是以河流为中心的自然-社会复合体，城市流域是指分布在城市行政区划内的流域。流域径流是变化环境的重要响应要素，研究其变化及对未来气候变化响应特征有助于提高人类对变化环境的应对能力。

本研究选取北运河、潮白河、永定河北京市域主要集水区为研究区域。通过分析水文气象观测站近40年水文数据，分析城市流域降水径流变化趋势。提取洪水场次，分析暴雨洪水特征；引入全球气候模式数据，选择SSP1-2.6、SSP2-4.5和SSP5-8.5三种不同强迫情景，进行降尺度与偏差校正处理。通过构建北运河流域的HEC-HMS水文模型，对2030-2040年、2050-2060年和2090-2100年流域径流的响应特征进行了分析。主要结论如下：

（1）北运河流域历史时期年降水量未发生显著变化，年径流量自2004年后以0.33亿m³/a的速率显著增加；潮白河流域年径流量1985-2005年间呈先增加后减小趋势，2016年后以1.32亿m³/a的速率显著增加。永定河流域历史时期径流无显著变化。北运河流域年径流在2011年出现显著性突变，潮白河流域年径流在2004年出现显著性突变，永定河流域年径流未发生显著的模式改变。1985-2005年，北运河流域径流存在12年尺度周期变化；潮白河流域在径流动态变化中存在着5年尺度的周期变化规律；永定河流域径流序列的长期变化过程中周期性表现不显著。北运河与潮白河之间存在双向因果关系，北运河径流对永定河径流的影响更大，潮白河径流与永定河径流之间不存在显著的因果关系。

（2）对CMIP6 CGMs偏差校正后，各气候模式的模拟能力均得到有效提高，且线性标度法（LS法）对降水序列的偏差校正效果更优。多模式集合平均（MME）与观测数据之间的偏差明显降低，并且校正后的模拟结果能够用于气候变化分析及其影响评估。未来时期北运河流域年降水量不同情景之间的趋势变化并不显著且较为复杂，降水呈现波动增加的趋势。

（3）未来径流预测方面，2030-2040年SSP5-8.5情景下的多年最大流量为333.2 m³/s，SSP3-7.0情景下多年最大流量为299.2 m³/s。2050-2060年，各情景多年最大流量的中值由大到小依次为：SSP5-8.5＞SSP1-2.6＞SSP2-4.5。未来时期各个情景最大流量值在季节上的表现为夏季>秋季>春季>冬季。三种情景中，特别是在SSP5-8.5高排放情景下，流量增大幅度明显，甚至在2050年流量相较基准期翻倍增大。在SSP5-8.5情景下，超标准洪水发生频率明显增加，出现极端暴雨洪水事件。

（4）针对气候变化条件下径流响应结果，北运河未来发生洪水多为20年一遇以下（含20年）标准的洪水。在未来气候变化条件下，发生低标准洪水概率大，结合5年周期尺度，五年一遇洪水同频概率较大。需要关注北运河流域2036年、2059年、2096年关键年份洪水情况，及时通过温潮减河、运潮减河削减洪峰，做好防洪预案；构建城市流域防洪调度系统框架，发挥水库联合调度的优势；城市内部提升洪涝防治韧性，提升智慧化应急水平；坚持适应和减缓并重，积极应对气候变化。

Watersheds are natural-social complexes centred on rivers, and urban watersheds are watersheds distributed within urban town zones. Watershed runoff is significant response element of the changing environment, and the study of its changes pattern and response to future climate change can help to improve human resilience to cope with the changing environment.

In this study, the extent of the North Canal, Yongding River and Chaobai River basins within Beijing was selected as the study area. By analyzing the hydrological data of hydrometeorological observation stations over the past 40 years, we analyzed the trends in the precipitation runoff in urban watersheds. In addition, this study extracts flood fields and analyses storm flood characteristics. Three different forcing scenarios, SSP1-2.6, SSP2-4.5 and SSP5-8.5, were selected for downscaling and bias correction. By constructing the HEC-HMS hydrological model for the Northern Canal Basin, the response characteristics of the basin runoff were analysed for the years 2030-2040, 2050-2060 and 2090-2100.The main conclusions are as follows:

(1) The annual precipitation in the North Canal basin did not change significantly during the historical period, and the annual runoff increased significantly at a rate of 0.33 billion m³/a since 2004; The annual runoff of the Chaobai River Basin during the period 1985-2005 changed in a pattern of increasing and then decreasing, and increased significantly at a rate of 132 million m³/a after 2016. There was no significant variation in runoff in the Yongding River basin in the historical period. Regarding the mutation profile, the annual runoff of the North Canal Basin changed significantly in 2011. In addition, the time of significant change in annual runoff in the Chaobai River Basin was 2004. The annual runoff of the Yongding River Basin did not show any significant pattern change. From 1985-2005, the runoff of the North Canal Basin had a 12-year scale cyclic change; the Chaobai River Basin had a 5-year scale cyclic pattern of change in the dynamics of the runoff; and the runoff sequence in the Yongding River basin does not show significant periodicity. There is a bi-directional causal relationship between the North Canal and the Chaobai River, with the North Canal runoff having a greater influence on the Yongding River runoff, and there is no significant causal relationship between the Chaobai River runoff and the Yongding River runoff.

(2) After correcting the bias of CMIP6 CGMs, the simulation ability of each climate model was effectively improved, and the linear scaling method (LS) was more effective in correcting the bias of the precipitation series used. The bias between multi-model ensemble averaging (MME) and observed data is significantly reduced, and the corrected simulation results can be used for climate change analysis and its impact assessment. The trend changes between different scenarios of annual precipitation in the Northern Canal Basin in the future period are not significant and complex, with a fluctuating trend of increasing precipitation.

(3) For future runoff projections, in 2030-2040, the multi-year maximum flow for scenario SSP5-8.5 is 333.2 m³/s, and the multi-year maximum flow for scenario SSP3-7.0 is 299.2 m³/s. The median multi-year maximum flow for each scenario in 2050-2060 is, in descending order, SSP5-8.5 > SSP1-2.6 > SSP2-4.5. The seasonal behaviour of the scenario maximum flow values in the future period is summer > autumn > spring > winter. In all three scenarios, especially in the SSP5-8.5 high-emission scenario, the flow increase is significant and even doubles in 2050 compared to the base period. Under the SSP5-8.5 scenario, there is a significant increase in the frequency of exceeding standard floods, with extreme heavy rainfall flood events occurring.

(4)  In response to the results of the runoff response under climate change conditions, most of the future floods occurring in the North Canal will be floods with a standard of less than one in 20 years (including 20 years). Under future climate change conditions, there is a high probability of a low standard flood, and combined with the 5-year cycle scale, there is a high probability of a 1 in 5 year flood with the same frequency. It is necessary to pay attention to the floods in the North Canal Basin in the critical years of 2036, 2059 and 2096, to cut down the flood peaks in time through the Wenchao Reducing River and the Yunchao Reducing River, and to make a good flood prevention plan; to build a framework for the flood prevention and dispatch system of the urban basin, and to take advantage of the joint scheduling of the reservoirs; to enhance the resilience of the flood prevention and control within the city, and to improve the level of intelligent emergency response; and to insist on the equal importance of adaptation and mitigation, and to respond to the climate change in a proactive manner.