全球变暖背景下极端降水事件的变化特征及归因分析
Characteristics and Attribution Analysis of Extreme Precipitation Events in the Context of Global Warming

全球变暖通过热力学–动力学协同效应深刻改变极端降水事件的演变规律，本研究基于多源观测和气候模式，系统揭示极端降水事件的时空分异特征及驱动机制。结果表明：全球陆地极端降水强度以1.3%~2.5%/decade的速率显著增强，热带地区超短历时(1~3小时)强降水增幅达8%~12%/？C，超越Clausius-Clapeyron理论预期，而中高纬度暴雨带向极地迁移，季风区呈现“北扩南缩”的极化特征，如东亚夏季风北界延伸1.5个纬度，长江流域梅雨期暴雨日数增加15%。归因分析表明，人类活动贡献了极端降水长期趋势的60%~85%，其中温室气体强迫通过提升大气持水能力(7%/？C)主导热带地区75%的强度增长，气溶胶辐射效应则削弱东亚1960~2000年极端降水增幅约30%。研究创新构建“观测–模式–机理”耦合框架，量化湿位涡异常与水汽辐合效率的协同机制，揭示城市化热岛效应使京津冀小时极端降水频率上升25%的局地反馈过程，但气候模式对对流参数化和气溶胶–云相互作用的模拟偏差仍导致±15%的归因不确定性。成果为极端降水风险动态评估、基础设施韧性设计及气候适应政策制定提供科学依据，未来需发展千米级地球系统模式与人工智能融合技术，攻克水文极值临界点预测难题，支撑全球气候治理行动。
Global warming profoundly changes the evolution law of extreme precipitation events through thermodynamic-dynamic synergy effect. Based on multi-source observation and climate model, this study systematically reveals the spatial and temporal differentiation characteristics and driving mechanism of extreme precipitation events. The results show that the global land extreme precipitation intensity increased significantly at a rate of 1.3%~2.5%/decade, and the ultra-short duration (1~3 hours) in tropical areas increased by 8%~12%/？C, beyond the Clausius-Clapeyron theory expectation, while the middle and high latitude rainstorm zones migrated to the polar region, the monsoon areas showed the polarization characteristics of “north expansion and south contraction”, such as the north east Asian summer wind extends by 1.5 latitudes, and the number of rainstorm days in the Yangtze River basin increased by 15%. The attribution analysis shows that human activities contribute 60%~85% of the long-term trend of extreme precipitation. Greenhouse gas forcing increases 75% by improving atmospheric water holding capacity (7%/？C), and the extreme precipitation in 1960~2000 by about 30% by aerosol. Research innovation to build “observation-mode-mechanism” coupling framework, quantitative wet vortex anomaly and water convergence efficiency of coordination mechanism, reveals the urbanization heat island effect makes Beijing-Tianjin-Hebei hours extreme precipitation frequency rising 25% of the local feedback process, but the climate model of convection parameterization and aerosol-cloud interaction simulation deviation still leads to ±15% attribution uncertainty. The results provide a scientific basis for dynamic risk assessment of extreme precipitation, infrastructure resilience design and climate adaptation policy