Derivation of an observation-based map of North African dust emission

Highlights: • North African dust emission and emission frequency from six models are examined. • There is a power law relationship between emission and emission event frequency. • The lognormal distribution of surface winds gives rise to this power law relation. • Annual North Africa dust emission is estimated via satellite retrievals of emission frequency. • We estimate that 82% of all North Africa dust emission is between 15° and 20°N. Abstract: Changes in the emission, transport and deposition of aeolian dust have profound effects on regional climate, so that characterizing the lifecycle of dust in observations and improving the representation of dust in global climate models is necessary. A fundamental aspect of characterizing the dust cycle is quantifying surface dust fluxes, yet no spatially explicit estimates of this flux exist for the World’s major source regions. Here we present a novel technique for creating a map of the annual mean emitted dust flux for North Africa based on retrievals of dust storm frequency from the Meteosat Second Generation Spinning Enhanced Visible and InfraRed Imager (SEVIRI) and the relationship between dust storm frequency and emitted mass flux derived from the output of five models that simulate dust. Our results suggest that 64 (±16)% of all dust emitted from North Africa is from the Bodélé depression, and that 13 (±3)% of the North African dust flux is from a depression lying in the lee of the Aïr and Hoggar Mountains, making this area the second most important region of emission within North Africa.