Spatial and temporal scaling of sub-daily extreme rainfall for data sparse places

Abstract Global efforts to upgrade water, drainage, and sanitation services are hampered by hydrometeorological data-scarcity plus uncertainty about climate change. Intensity–duration–frequency (IDF) tables used routinely design water infrastructure so offer an entry point for adapting engineering standards. This paper begins with a novel procedure guiding downscaling predictor variable selection heavy rainfall simulation using media reports of pluvial flooding. We then present three-step workflow to: (1) spatially downscale daily from grid-to-point resolutions; (2) temporally scale series sub-daily extreme rainfalls and; (3) test methods temporal scaling within Regional Climate Model (RCM) simulations under changed conditions. Critically, we compare the moments parameters annual maximum into rainfalls, whilst accounting intermittency. The applied Kampala, Uganda Kisumu, Kenya Statistical Downscaling (SDSM), two RCM covering East Africa (CP4 P25), in hybrid form (RCM-SDSM). demonstrate that Gumbel (and IDF tables) can be reliably scaled durations 3 h observations RCMs. Our RCM-SDSM reduces errors estimates when compared direct output. Credible parameter relationships also found discuss practical aspects applying such workflows other city-regions.