2020年长江上游和中下游先后发生特大洪水，其中干流编号洪水全部发生在上游，构成了长江流域洪水的主要部分。首先回顾2020年洪水及洪灾情况，然后根据历史上几次特大洪水过程和历年实测资料，分析长江上游洪水特征、洪灾类型及特点，最后提出新时代长江流域洪水整体防御战略及山洪灾害防治战术。研究表明：金沙江洪水是长江上游洪水基础部分，岷江、嘉陵江和干流区间是洪峰的主要来源，三者洪水遭遇是产生上游特大洪水的主因，上游洪水又是全流域特大洪水的基础和重要组成部分。目前造成洪灾死亡人数最多的是山洪以及山洪引起的地质灾害，财产损失最大的是中下游及湖泊地区。未来堤防仍然是防洪的基础，提高沿江城市防洪标准主要手段是控制性水库的联合优化调度，而减少洪涝灾害损失最有效的途径是给洪水以空间的自然解决方案等非工程措施。

In 2020， the upstream and mid-downstream of the Yangtze River experienced massive floods， with the major mainstream floods occurring in the upper Yangtze River. In this study， we first reviewed the floods and their related losses， and then analyzed the characteristics of flooding disasters in the upper Yangtze River based on several catastrophic floods in history. Finally， we proposed the integrated strategies for flood defense and the control tactics for flash floods in the Yangtze River Basin in the new era. Our results show that the floods in the Jinsha River underlay the floods in the upper Yangtze River， whereas the flood peaks were primarily attributed to the inflow from the Minjiang River， the Jialing River and the mainstream interval. The co-occurrence of floods in the aforementioned three tributaries led to the mega-floods in the upper Yangtze River， which formed the foundation and an important component of the basin-wide floods. At present， flash floods and accompanied geological disasters caused the largest number of deaths， and the biggest property losses occurred in the middle-lower reaches and the lake areas. In the future， the embankments will remain the basis of flood control. The main means to improve flood control standards of urbans long the river will rely on the joint optimal operation of reservoirs. In addition， the most effective way to reduce flood losses will be non-engineering measures such as natural solutions that give space to floods.

该文较系统地评述了干旱与高温热浪在定义、 标准与类型、 地理分布、 时间和强度变化特征、 危害性质和程度及其影响、 形成机制和产生原因、 对全球气候变暖响应的表现形式和程度、 监测和预测与预警方法与技术、 减灾技术及应对策略和防御措施等8个方面, 既对比了两者间在以上8方面的明显的差异, 也指出两者间还有密不可分的内在联系,当高温天气频繁发生,大气降水量就会明显减少。高温加快了土壤的蒸散速度,加大了土壤水分和植株水分的散失,这种关系在夏季尤其显著。从而造成干旱的发生或加重严重程度。

复合极端事件是导致社会或环境风险的多种驱动因子和/或致灾因子的组合，对人类社会和生态系统造成的影响往往比单个极端事件更严重、更具破坏性。首先简要论述了复合极端事件的定义和内涵，包括先决条件事件、多变量事件、时间复合事件以及空间复合事件；然后详细综述了复合极端事件的时空演变特征、复合极端事件变化的影响因素和未来复合极端事件情景预估3个方面的研究进展；最后针对目前研究中面临的问题，提出今后研究关注的重点，包括复合极端事件的变量/指标选取及阈值确定、复合极端事件因子间依赖关系及相互作用、复合极端事件模拟性能评估和未来情景预估以及复合极端事件影响的动态过程及致灾机制。

Climate extremes threaten human health, economic stability, and the safety of both natural and built environments. Compound extreme events are combinations of multiple climate drivers and/or hazards that contribute to societal or environmental risks, and their impacts on human society and natural ecosystems are often more serious and destructive than those of a single extreme event. Understanding the changes in compound extreme events is important for adaptation, mitigation strategies, and disaster risk management. Here, the definitions and connotations of compound extreme events are briefly discussed, including preconditioned, multivariate, temporal, and spatial compounding events. Subsequently, the progress in compound extreme event research is discussed in terms of temporal and spatial evolution characteristics, influencing factors, and future scenario projections. Given the problems in current research, we suggest that future studies should focus on studying compound extreme events regarding variable/index selection and threshold determination, dependence and interaction analysis among drivers and/or hazards, simulation performance evaluation and future projections, and their dynamic processes and disaster-causing mechanisms. Compound extreme events are expected to increase in frequency and intensity in a warming world, and many regions are projected to experience an increase in the probability of compound events with greater global warming. Therefore, we must improve our understanding of the causes and drivers of compound and cascade events.

全球变暖增加了极端气候事件的发生概率，复合极端事件比单个极端事件对社会经济与环境造成的影响更严重。基于中国639个气象站点1961-2020年逐日降水和最高气温数据，采用加权平均降水指数（WAP）和相对阈值法识别出洪水和热浪事件，分析中国7日内洪水-热浪复合极端事件总频次及发生概率等时空变化特征，对比不同时段单个极端事件与复合事件两者之间变化差异，并探讨了洪水-热浪中热浪最长持续时间和平均强度变化，为我国防洪减灾方案提供参考。研究结果表明：①近60年几乎所有站点都发生过洪水-热浪复合极端事件，并表现出明显空间差异，出现频次较高的区域主要分布在长江流域、东南诸河流域、珠江流域及内陆河流域；②1961-2020年洪水-热浪事件总频次及中国区域平均发生概率均呈显著上升趋势（P<0.01），且不同年代以来上升速率越来越大；不同年份复合事件出现的频次和概率差异较大，2016年频次、概率最高（261次、17.4%），1985年频次、概率最低（8次、0.5%），发生概率与频次呈现出高度一致性；③与单个发生的极端事件相比，洪水-热浪复合事件发生频次增幅更大，1981-2000年、2001-2020年洪水-热浪事件总次数与1961-1980年的比值由1.26增加至2.96，且在长江流域东部、珠江流域东南部及东南诸河流域表现最为突出；④与热浪前没有洪水的热浪事件相比，洪水-热浪事件中热浪的最长持续时间相对较短、平均强度相对较小，不同流域空间差异表现不明显。

The probability of extreme climate events is increasing under global warming， and compound extreme events have more serious impacts in society and environment than individual extreme events. By using daily precipitation observation and daily maximum temperature data from 639 meteorological stations over China during 1961-2020， this paper adopts the index of weighted average of precipitation （ WAP） and the method based on relative threshold to identify flood and heatwave events， and investigates total frequency and probability of compound extreme flood-heatwave events occurrence within seven days， discusses the temporal variation characteristics of compound events in the different periods compared with individual extreme events， and analyzes the duration and average magnitude of heatwave in flood-heatwave event， this study can provide reference for the flood control and disaster reduction scheme in China. The research results are as follows： ① In recent 60 years， the compound flood-heatwave events occur at almost all stations， and show obvious spatial differences. The regions with high frequency of compound extreme events occurrence are mainly distributed in Yangtze， Southeast， Pearl and Northwestriverbasins； ② The total frequency and the regional average occurrence probability of flood-heatwave events over China showed significant increasing trends （ P<0.01） from 1961 to 2020， which became larger and larger among different decades.The frequency and probability of compound events occurrence in different years varied greatly with the highest values in 2016 （261 times， 17.4%） and the lowest values in 1985 （8 times， 0.5%）， and the frequency and probability of occurrence has significant correlation； ③ Compared with individual extreme events， the increasing rate of flood-heatwave events is relatively significant.The ratio of the total number of flood-heatwave events during 1981-2000 and 2001-2020 against the total frequency of compound events during 1961-1980 increased from 1.26 to 2.96， and the increasing rate was most prominent in eastern Yangtze， Southeastern Pearl and Southeastern river basins； ④ Compared with heatwave events without flood before them， the longest duration and average magnitude of heatwave in flood-heatwave events were relatively short and small respectively， which show insignificant spatial differences among different basins.

近年来频发的极端天气气候事件引起了广泛关注，其灾害过程往往源自多个因素的相互作用，给区域安全和风险防范带来诸多挑战。论文结合文献计量，在系统梳理复合极端事件相关研究的基础上，重点阐述了复合极端事件的概念内涵、类型特征和主要驱动因素，并归纳了复合事件时空关联分析和危险性评估的主要方法。结果表明：① 近年来研究的事件类型丰富多样，不同时空关联、不同要素组合类型事件的研究不断涌现；② 研究内容体系日益完善，概念特征、关联关系、成因机制和危险性评估的研究持续推进；③ 研究技术手段不断发展，以Copula为核心的联合概率统计建模实现由二维向多维、由静态向动态发展，以耦合水文水动力学模型和海洋模式为代表的数值模拟精细度不断提高。但在部分重难点问题方面仍需进一步深入研究，包括时间继发型和空间异地型复合事件复杂时空关联结构的诊断建模，天气系统、大尺度环流因子和人类活动多因素对复合事件综合影响的研究，复合事件危险性情景及多维联合概率分析等。此外，未来亟需探究气候变化下复合事件边缘分布和关联结构的非平稳变化及其对复合事件风险的影响。

In recent years, the frequent extreme weather and climate events have attracted wide attention. Their disastrous process often stems from the interaction of multiple factors, which brings many challenges to regional security and risk prevention. Starting from a bibliometric analysis, this article systematically reviewed the progress of research on compound extreme events by focusing on the conceptual features, classification, and driving factors of such events, and summarizing the main methods for the spatial-temporal dependance analysis and hazard assessment of compound events. The review found that: 1) The research on compound extreme events has developed rapidly in recent years, and the types of events studied have become increasingly rich and diverse. 2) The research system has been established and increasingly improved, with significant advance in the research on the conceptual characteristics, dependance, causative mechanism, and risk assessment. 3) The research techniques were constantly evolving. Statistical modeling for joint probability based on Copula has developed from two-dimensional to multidimensional, and from static to dynamic; the precision of numerical simulations represented by the coupled hydrological-hydrodynamic and ocean models has been continuously improved. But further in-depth studies are still needed, especially for some key and difficult problems, such as diagnosing and modeling the complex dependance structure of temporally and spatially compounding events, the synthetic effect of weather system, large-scale circulation and human activity impact on the formation of compound events, hazard scenarios and multidimensional joint probability analysis of compound events, and so on. In addition, it is urgent to explore the non-stationary changes of the marginal distribution and dependance structure of compound events under climate change and their impact on the risk of compound events in the future.