This study introduces an original methodology combining ensemble boosting and conditional probability theory to estimate return periods of rare climate extremes without relying on long climate simulations. The method is rigorously developed, validated on a red-noise process, and applied to CESM2 simulations, including an analysis of the 2021 Pacific Northwest heatwave. Importantly, this framework enables linking probability estimates to specific climate storylines, allowing an assessment of the odds that an extreme event like the one examined will occur in the future.

This study adopts the concept of atmospheric rivers, which is often applied in the extratropics, in an innovative manner to study the synoptic moisture fluxes in tropical East Africa, a region that stands out in terms of its inter-annual rainfall variability with large socio-economic impacts. In doing so, the authors are able to provide critical insights into the mechanisms underlying this rainfall variability and connect inter-annual changes to shorter time scales and specific synoptic events, including tropical cyclones and phases of the Madden-Julian Oscillation.