Attribution of Anthropogenic Impact on Extreme Mei-yu Rainfall in Taiwan

Mei-yu rainfall provides vital water resources to Taiwan, but this quasi-stationary system has produced increasingly severe flooding in recent decades. We investigate the influence of anthropogenic climate change on two extreme Mei-yu events (11–12 June 2012 and 2–3 June 2017) using a pseudo-global-warming (PGW) approach, forcing the Cloud-Resolving Storm Simulator (CReSS) model with anthropogenic climate change signals derived from differences between 15 CMIP6 historical and natural simulations over May–June 2000–2014. Five sensitivity experiments per event, each initialized from a different analysis and reanalysis product (ERA5, CFSv2, FNL, MERRA-2, JRA-55), test the variability of the results. We find that anthropogenic forcing drives two competing effects: increased moisture enhances frontal rainfall, while a weakened southwesterly flow suppresses it. The balance between these effects depends on the synoptic setting, making simulated changes in Mei-yu frontal precipitation under warming somewhat case-dependent. In the 2012 event (moderate front, strong flow), enhanced moisture dominated and led to increased rainfall over Taiwan. In the 2017 event (intense front, moderate flow), the simulated frontal position instead controlled the outcome, with an offshore shift reducing rainfall over Taiwan. These contrasting outcomes suggest that projecting future Mei-yu rainfall risk requires not only accounting for mean-state warming but also considering event-specific synoptic controls—complicating flood-risk assessments for Taiwan.