Quantifying rainfall variability and potential hazards of extreme events in Beijing through stochastic simulations

The city of Beijing, China. Beijing is affected by heavy rainfall in summer, which may trigger urban floods. These risks make long rainfall records valuable for extreme rainfall analysis, given the complex spatiotemporal patterns and high natural variability of rainfall; however, such records are unavailable. To address this limitation, we propose using a gridded rainfall generator capable of capturing natural climate variability and enabling the simulation of yet-unobserved extreme rainfall events. The AWE-GEN-2d was used to extend the rainfall sample by simulating 30 realizations of 30 years each at 1 km and hourly intervals based on steady-state climate assumptions. It was calibrated and validated with weather stations, local high-resolution gridded rainfall estimates, and the Merra-2 climate reanalysis product. We then analyzed the natural variability and extreme characteristics of rainfall based on the stochastically generated data. Rainfall variability increases with higher spatiotemporal resolution and peaks at the event-based level. For extreme events, variability grows markedly with longer return periods, and rainfall amounts for the 100-year return period can exceed station records by up to 53 % through event-based IDF curves. This study emphasizes the heterogeneity of rainfall fields and reveals the hazards of yet-unobserved extreme rainfall events under a stationary climate, providing critical insights for improving urban flood management strategies in highly urbanized regions like Beijing. • Rainfall generators are suggested to be integrated with extreme rainfall analysis. • This enables the representation of natural rainfall variability on multiple scales. • As a result, the “true” potential range of extreme rainfall can be revealed. • Unseen but plausible extreme rainfall events may greatly exceed observations. • We demonstrate the framework for short-duration summer extreme rainfall in Beijing.