DUST EMISSION MECHANISMS IN THE CENTRAL SAHARA: NEW INSIGHTS FROM REMOTE FIELD OBSERVATIONS

In boreal summer, satellite retrievals show that the central Sahara is the dustiest place in the world. Fennec, an international consortium led by the University of Oxford, is the first field campaign to provide systematic high resolution ground-based observations of the region. Fennec supersite-1 in south-west Algeria is within kilometres of the average summer global dust maximum, and was the location of two intensive observation periods (June 2011 and June 2012). The findings provide a unique insight into the mechanisms of dust emission and transport.

In both June 2011 and June 2012, density currents emanating from deep convection produced the most dust at the supersite (c. 50%), followed by dust advection (i.e. dust transportation over the site), monsoon surges, low level jet emission and lastly dry convective plumes (< 2%). Numerical models without explicit convection are known to struggle to represent density currents, but given their primacy it is very important that density currents be incorporated in dust modelling efforts over the region. Density currents frequently occur at night and are associated with cloud cover; therefore it is likewise important that appropriate instruments are chosen for monitoring dust in the central Sahara, particularly if the measurements are then used to validate models or satellite algorithms. Sun photometer aerosol optical depth retrievals for example are only available for 23% (41%) of the time density current emission occurs at the supersite in June 2011 (June 2012), because retrievals cannot be made at night or under cloud. Such measurements therefore suffer a large systematic bias. Overall, the number of dust event hours in June 2012 was up to 171 h (61%) more than in June 2011. This is consistent with a more active monsoon in 2012, which promotes more density currents and monsoon surges.

Low level jets lead to a surprisingly low amount of dust emission. This is because, although they are present on most mornings, the jets have to be very strong (roughly >16m/s) for enough momentum to be mixed down to the surface to lead to dust emission. This mix-down can also be inhibited by a shallow convective boundary layer. Model simulation of low level jets appears to hinge on whether they are embedded in the moist southerly monsoon flow or the dry northerly Harmattan winds.