Recent extreme storm events in the San Francisco Bay Area have been labeled “unprecedented” with “never before seen impacts.” We know extreme storms are no longer well represented by historical observations. However, extreme storm characterization is also lacking within typical state-of-the-art global climate models that have coarse scale resolution, creating challenges for long-range planning. Decision makers, planners, and engineers need to understand how extreme events will change over time to know how communities will be impacted. This study fills a critical gap in the regional understanding of how precipitation may change over the coming century, with an emphasis on extreme events and storms commonly used to support hydrologic and hydraulic design criteria.

Our methodology quantifies the impacts of climate change on individual storm events at the spatial scales required for decision making using the Weather Research and Forecasting (WRF) regional climate model. These high-resolution (3km) simulations estimate how the magnitude of an extreme storm could change if a similar event occurs again in a warmer climate. Historical storms were selected from a catalog of high impact storms, including large atmospheric rivers and extratropical cyclones. Differences in the response to anthropogenic warming across these two storm types is investigated, and the applicability of Clausius-Clapeyron scaling relative to temperature changes at different vertical levels is discussed.

Stakeholders were actively engaged during the project to ensure that the findings would translate into actionable science for decision making. The main data product of interest to stakeholders were future condition precipitation Intensity-Duration-Frequency (IDF) curves. IDF curves for San Francisco down to sub-daily temporal resolution were developed from our WRF simulation results by supplementing them with downscaled precipitation projections from the CMIP5 archive and temperature projections from GCMs in the CMIP6 archive. Amplification of precipitation changes in the short duration rainfall relative to storm total changes were observed, which highlight the added value of including regional climate modeling when generating projections of future extreme storms.