Atmospheric convection, dynamics and topography shape the scaling pattern of hourly rainfall extremes with temperature globally

Precipitation extremes are expected to intensify under climate change as ground temperature rises with a rate similar to the air's water holding capacity similar to 7%/K (Clausius-Clapeyron). Recent studies have been inconclusive on the robustness and global consistency of this behavior. Here, we use hourly weather stations, 40 years of climate reanalysis and two convection permitting models to unravel the global pattern of rainfall extremes scaling with temperature at the hourly scale and identify hotspots of divergence from thermodynamical expectations. We show that in high- and mid-latitudes precipitation extremes closely follow a Clausius-Clapeyron scaling, while divergence occurs over the tropics and subtropics. Local features of atmospheric convection, larger-scale dynamics and orography, affect the dependence of extreme rainfall on surface temperature. In regions with deep convection, persistent large-scale dynamics and complex orography, hourly rainfall extremes diverge from expectations from the atmosphere's water holding capacity, suggests a global analysis of station data, reanalyses and convection-permitting models.