Atmospheric Dynamical Amplification Intensifies Future Extreme Precipitation Risks

Extreme precipitation events, driven by complex multi-scale atmospheric dynamic interactions and fueled by available moisture, are expected to intensify with climate change, posing increasing risks to human communities and ecosystems. However, current global climate models with coarse resolutions struggle to accurately represent key extreme precipitation-producing phenomena, such as mesoscale convective systems (MCSs). This limitation hinders our ability to generate robust and reliable future projections. To address this challenge, we introduce an ensemble of high-resolution climate simulations that substantially improve the representation of key atmospheric phenomena, particularly MCSs, and associated multi-scale interactions globally. These enhanced simulations significantly improve the simulation of extreme precipitation during the historical period and offer more physically grounded future projections. Under a business-as-usual scenario, our results suggest that extreme daily precipitation over global land could increase by ~39% by the end of the century, primarily driven by enhanced dynamical contributions, which are significantly underestimated in coarse-resolution models by a factor of three. These findings highlight the crucial role of high-resolution climate modeling in constraining future extremes and informing more effective climate risk assessments and adaptation strategies.