Momentum Decoupling Drives Low‐Wind Events During Heatwaves in China: A Hybrid Explainable Machine Learning and Dynamical Modeling Approach

Abstract Compound heatwave and low‐wind (HW‐LW) events increasingly threaten China's energy security. Analysis of 2000–2023 data reveals these compound extremes account for 18.5% of total heatwaves, with their annual frequencies nearly doubling in recent years. Integrating explainable machine learning with dynamical simulations, we identify momentum decoupling as the primary physical driver of this wind suppression. During the HW‐LW events, adiabatic subsidence controlled by high‐pressure systems warms the upper atmosphere by over 3.0°C. This forms a stable temperature inversion layer that blocks downward kinetic energy transport. Consequently, this suppresses boundary layer development and causes surface wind speeds to drop by up to 4.0 m/s. These results demonstrate that HW‐LW compound events intrinsically starve the surface of wind energy supply. This highlights critical vulnerabilities in power grid systems as such compound extremes intensify globally.