气候变化情景下极端降水事件的频次和强度预估呈增加趋势，这会导致全球部分地区极端降雨诱发地质灾害风险的增加。本文基于中国降雨诱发地质灾害易发性模型和不同地貌分区的累积事件降雨量-降雨历时阈值曲线，采用最新的CMIP6全球气候模式多模式集合结果，基于全球温升目标情景的视角，从地质灾害空间易发性和发生频次两方面，探讨温升情景下中国地质灾害危险性的可能变化及其对暴露人口的潜在影响。结果表明，CMIP6多模式集合预估的多年平均降水在温升1.5 ℃和2.0 ℃情景下相比基准时期可能增加5.4%~9.5%，导致中等至极高地质灾害易发区范围预估增加0.33%~0.74%，由于预估的极端降水事件增加，地质灾害发生频次预估增加7.0%~11.2%，进一步综合未来人口空间分布，潜在地质灾害暴露人口可能增加6.20亿人次（18.90%）和4.26亿人次（12.97%）。各地貌分区未来情景下地质灾害危险性预估增加且存在显著的空间异质性，温升2.0 ℃情景下中等至极高易发性范围相比基准时期增加0.71%~1.28%，地质灾害发生频次预估增加1.2%~15.6%，其中，青藏高原区地质灾害危险性增加最明显。综合考虑未来人口的变化，结果显示，由于未来预估人口的明显减少，青藏高原区温升1.5 ℃和2.0 ℃时潜在的地质灾害暴露人口预估减少468万人次~928万人次，而东南丘陵区潜在的地质灾害暴露人口分别增加3.96亿人次和3.00亿人次。本研究的结果可以为制定更有针对性的适应气候变化影响和减轻地质灾害风险措施提供科学指导。

The frequency and intensity of extreme precipitation events are projected to increase under climate change scenarios, which may result in increasing risk of rainfall-induced landslide in some parts of the world. Based on the established national scale rainfall-induced landslide susceptibility model and cumulative rainfall -rainfall duration threshold curves for different geomorphic regions, this study employs the latest CMIP6 global climate model ensemble to assess the changes in landslide hazard in China under global warming in terms of both the spatial landslide susceptibility and frequency resulted from rainfall events exceeding the threshold of landslide occurrence. The results show that the multi-year mean annual precipitation projected by the CMIP6 multi-model ensemble is likely to increase by 5.4% to 9.5% under the 1.5°C and 2.0°C warming scenarios compared to the baseline period, resulting in a projected increase of 0.33% to 0.74% in moderate to very high landslide susceptibility areas, and a projected increase of 7.0% to 11.2% in landslide frequency due to the projected increase in extreme precipitation events. By further combing the projections of future population distribution, the potential exposed population is expected to increase by 620 million (18.90%) and 426 million (12.97%) under the 1.5°C and 2.0°C warming scenarios, respectively. The projected landslide hazards under the future scenarios increase in each geomorphic region, and there exists significant spatial heterogeneity. The range of moderate to very high susceptibility under a 2.0°C temperature rise scenario increases by 0.71%~1.28% compared with the baseline period, and the landslide occurrence frequency is projected to increase by 1.2%~15.6%. The CMIP6 multi-model ensemble projections reveal hotspot areas where landslide susceptibility level and frequency are expected to increase under warming scenarios, including the southeastern Tibetan Plateau, the Tianshan Mountains in the northwest, and the Kunlun Mountains at the border of the Tibetan Plateau. And Qilian Mountains in the southwest mountainous region, the Loess Plateau and the Taihang Mountains in the south of the north-central plain region, and the Changbai area in the eastern plain region, are also key areas where appropriate landslide risk mitigation measures for climate change adaptation are needed. Considering the future population changes, our results show that the potential landslide exposed population in the Qinghai-Tibet Plateau area is expected to decrease by 4.68 million to 9.28 million, respectively, due to the obvious decrease in predicted future population, while the potential landslide exposed population in the southeastern hilly area increases by 396 million and 300 million, respectively, when temperature rises by 1.5°C and 2.0°C.