The sensitivity of the colour of dust in MSG-SEVIRI Desert Dust infrared composite imagery to surface and atmospheric conditions

Abstract. Infrared Desert Dust composite imagery taken by the Spinning Enhanced Visible and InfraRed Imager (SEVIRI), onboard the Meteosat Second Generation (MSG) series of satellites above the equatorial East Atlantic, has been widely used for more than a decade to identify and track the presence of dust storms from and over the Sahara Desert, the Middle East, and southern Africa. Dust is characterised by distinctive pink colours in the Desert Dust false-colour imagery, however the precise colour is influenced by numerous environmental properties, such as the surface thermal emissivity and skin temperature, the atmospheric water vapour content, and the quantity and height of dust in the atmosphere. This paper is the follow-up to Banks et al. (2018), which analysed the sensitivity of the colour of the dust in the imagery to its infrared optical properties. The previous paper introduced a modelling system combining dust concentrations simulated by the aerosol transport model COSMO-MUSCAT (COSMO: COnsortium for Small-scale MOdelling; MUSCAT: MUltiScale Chemistry Aerosol Transport Model) with radiative transfer simulations from the RTTOV (Radiative Transfer for TOVS) model, in order to simulate the SEVIRI infrared measurements and imagery. Investigating the sensitivity of the synthetic infrared imagery to the environmental properties over a six month summertime period from 2011 to 2013, it is confirmed that water vapour is a major control on the apparent colour of dust, obscuring its presence when the moisture content is high. Of the three SEVIRI channels used in the imagery (8.7, 10.8, and 12.0 μm), the channel at 10.8 μm has the highest atmospheric transmittance and is therefore the most sensitive to the surface skin temperature. A direct consequence of this sensitivity is that the background desert surface exhibits a strong diurnal cycle in colour, with light blue colours possible during the day and purple hues prevalent at night. In dusty scenes, the clearest pink colours arise from high-altitude dust in dry atmospheres. Elevated dust influences the dust colour primarily by reducing the contrast in atmospheric transmittance above the dust layer between the SEVIRI channels at 10.8 and 12.0 μm, thereby boosting red and pink colours in the imagery. Hence the higher the dust altitude, the higher the threshold column moisture needed for dust to be obscured in the imagery: for a sample of dust simulated to have an AOD at 550 nm of 2–3 at an altitude of 3–4 km, the characteristic colour of the dust may only be impaired when the total column water vapour is particularly moist (g 39 mm). Meanwhile dust close to the surface (altitude