Wind erosion and the associated mineral dust emissions are of particular interest because of their impacts on the Earth system. At a global scale, dust impact the radiative budget and the oceanic carbon cycling. At a local scale, wind erosion can thin down the soil layer and decrease its nutrients content. Dust emissions mostly occur in the arid and semi-arid areas, but their quantification remains challenging in semi-arid areas like the Sahel because of the large seasonal and inter-annual dynamics of surface properties. A bottom-up modeling approach has been used to simulate the Sahelian green vegetation and dust emissions. First, the seasonal dynamics of rangeland vegetation and dust emissions over 4 recent years (2004-2007) has been simulated over the whole Sahel, using a vegetation model and a dust emission model. Model outputs have been compared to satellite observations (particularly MODIS vegetation indices). It is shown that Sahelian dust emissions exhibit a large inter-annual variability, and that vegetation and soil moisture have non negligible impacts on these dust emissions. Rangeland vegetation simulations have also been performed on a longer period (1960-2015), representing green and dry vegetation. They catch well the effect of rainfall interannual variability on Sahelian vegetation, like the droughts in the 1970’s and 1980’s. Secondly, measurements performed at a plot scale have been used to better characterize wind erosion from a rangeland and from a cropland. An additional vegetation model was used to represent Sahelian cultivated plants (e.g. millet). The results focus on the impacts of farming practices on wind erosion in Sahelian croplands. It is shown that the inter-annual variability of the wind erosion due to farming practices is about the same order as the variability induced by climate factors.