关键词: 气候变化, 复合高温?极端降水事件（CHPEs）, 危险性评估, 贝叶斯模型平均（BMA）, Copula

Abstract: China’s coastal regions are potential high-risk areas for compound heat and extreme precipitation events (CHPEs). Based on the NEX-GDDP-CMIP6 dataset, this study applies Bayesian model averaging (BMA) to optimize multi-model ensemble projections and uses Copula functions to investigate the spatiotemporal evolution of CHPE hazards, including frequency, intensity, and duration, as well as changes in joint return periods under different future scenarios. The results show that BMA simulations outperform the conventional multi-model ensemble mean. CHPE hazards along coastal China are projected to increase in the future, with the largest increase occurring under the high-emission scenario. Frequency and intensity show particularly pronounced increases; under the high-emission scenario, CHPE frequency and intensity are projected to be 5.96 and 101.23 times those of the baseline period, respectively. Spatially, CHPE hazards exhibit a clear pattern of being higher in the south and lower in the north, with high-hazard areas are mainly concentrated along the South China and Southeast China coasts, as well as in Hainan and Taiwan. Historically rare high-intensity compound events are projected to experience substantially shortened joint return periods under future scenarios. By integrating BMA with Copula-based joint probability analysis in the spatiotemporal assessment of CHPE hazards, this study not only addresses the limitation of conventional multi-model ensemble means in neglecting differences in model performance, but also further reveals the joint risk amplification caused by the concurrent intensification of heat and extreme precipitation. These findings provide a scientific basis for risk projection and adaptation management of compound extreme events in coastal China.

Key words: Climate change, Compound heat-extreme precipitation events (CHPEs), Hazard assessment, Bayesian model averaging (BMA), Copula