International audienceThe work presented here focuses on the analysis of a turbulent boundary layer saturated with saltating particles. Experiments were carried out in a wind tunnel 15m long and 0.6m wide at the University of Aarhus in Denmark with sand grains 242 μm in size for wind speeds ranging from the threshold speed to twice its value. The saltating particles were analysed using particle image velocimetry (PIV) and particletracking velocimetry (PTV), and vertical profiles of particle concentration and velocity were extracted. The particle concentration was found to decrease exponentially with the height above the bed, and the characteristic decay height was independent of the wind speed. In contrast with the logarithmic profile of the wind speed, the grain velocity was found to vary linearly with the height. In addition, the measurements indicated that the grain velocity profile depended only slightly on the wind speed. These results are shown to be closely related to the features of the splash function that characterizes the impact of the saltating particles on a sandbed. A numerical simulation is developed that explicitly incorporates low-velocity moments of the splash function in a calculation of the boundary conditions that apply at the bed. The overall features of the experimental measurements are reproduced by simulation

We report on wind tunnel measurements on saltating particles in a turbulent boundary layer and provide evidence that over an erodible bed the particle velocity in the saltation layer and the saltation length are almost invariant with the wind strength, whereas over a nonerodible bed these quantities vary significantly with the air friction speed. It results that the particle transport rate over an erodible bed does not exhibit a cubic dependence with the air friction speed, as predicted by Bagnold, but a quadratic one. This contrasts with saltation over a nonerodible bed where the cubic Bagnold scaling holds. Our findings emphasize the crucial role of the boundary conditions at the bed and may have important practical consequences for aeolian sand transport in a natural environment.

No abstract