The atmospheric content of desert aerosols has a high variability on a multi-annual scale. This long term variability can be interpreted as a response to changes in environmental conditions in source region. Thus the correlation between the increase in occurrences of haze and reduced rainfall observed in the Sahel region between the 70s and the 80s could be explained by an increase in dust emissions in this region in response to a decrease in the vegetation cover. Correlations between the increase of aerosol content and index of drought in the Sahel have also been highlighted on decadal time scales. In the South Pacific area, studying the contents of sediment cores in mineral particles from wind (Atacama Desert in the North of Chile) helped to highlight a decadal variability of the same type.
In order to interpret and quantify the relative weight of different processes responsible for the interannual variability of the mineral dust content, and to study the environmental impacts of desert aerosols at the regional level, the LISA has initiated the development of regional and continental models. This is the case of the CHIMERE-Dust model, adapted from the Chemistry and Transport model CHIMERE to perform decadal simulations of the mineral dust cycle with an hourly time step. In the case of the Sahel, the trend analysis also relies on long-term observations from the Sahelian Dust Transect deployed during the AMMA campaign, and on a deposition network around the Mediterranean during the CHARMEX project. The preset time frame for these observations is the decade, so as to integrate the natural variability of the system and hope to detect the impact of the evolution of the use of land or of climatic conditions.
Figure 1: Sliding average over 30 days of the concentrations in aerosols measured in Senegal (M'Bour), Mali (Cinzana) and Niger (Banizoumbou) from 2006 to 2008. A similar seasonal cycle is systematically recorded at the three stations, which are superimposed peak concentrations related to events of emissions or transport intensity. These peaks are often observed at the three stations, which suggest a transport phenomenon from a regional to a continental-scale.