Exploring the influence of land-atmosphere interactions on humid heat extremes in a convection permitting model simulation

Ambient humidity reduces the ability of the body to cool down through sweating, adding to the heat 
stress caused by elevated air temperature alone. Indeed, humid heat waves (HHWs) are already a threat
for humans, livestock and wildlife, and their impacts are projected to increase with global warming.
HHWs result from the combination of thermodynamic and dynamic processes interacting on a range of 
time and space scales and whose relative importance may vary according to location and time of year.

Africa is one continent where HHWs, defined here as extremes of wet-bulb temperature (Twb), are 
expected to become more important under global warming. Local-scale humid heat extremes may occur 
within more moderate larger-scale events across much of the continent. Yet, climatological 
characteristics of these smaller-scale events such as location and timing (in year and day) are poorly 
documented in the current climate, due to a lack of high-resolution data and research focus. Moreover, 
a comprehensive understanding of their meso- to synoptic-scale drivers is still lacking. Here, we explore 
these two issues using a 10-year pan-African convection-permitting model simulation that explicitly 
resolves land-atmosphere interactions, and particularly those involving moist processes that are 
instrumental to HHWs.

We find humid heat extremes in semi-arid regions occurring in the core of the rainy season, on length 
scales down to a few tens of kilometers. During HHWs, Twb peaks several hours 
later than the climatological peak in the late morning. This diurnal cycle shift is likely due to HHWs 
typically developing in the aftermath of a rainfall event: the resulting positive anomaly in soil moisture 
induces increased latent heat fluxes, low level divergence, and a reduced PBL height, all ingredients
displaying sharp spatial gradients conducive to locally high Twb values. These results have implications 
for the improvement of localized HHW predictability based on local soil moisture conditions, a key step 
towards climate change adaptation through e.g., early-warning systems.