Source-to-sink dynamics of aeolian landscapes

If rivers and glaciers play a critical role in sediment transfers and landscape dynamics on Earth, their action on drylands remains limited. In these arid regions, the sediment transport by the wind then takes on its full extent and becomes a major player in shaping continental surfaces and in mass exchanges between continents, atmosphere and ocean. The wind thus mobilizes sand sheets and dust plumes and models aeolian landscapes (erosive depressions, fields of yardangs and dunes), while impacting climate, life and human activities. A better knowledge of current and past sediment transport by the wind is therefore crucial to understand and manage aeolian processes. Accordingly, this PhD thesis aims to characterize and quantify aeolian landscape dynamics and related sediment fluxes from their sources in erosion to the sinks where they accumulate over multiple space and time scales (from 101 to 106 meters or years). The continuous improvement of remote sensing resources, wind data and morphochronological dating methods, coupled with a better understanding of dune dynamics, currently bring an increasing number of theoretical and observational constraints on aeolian morphodynamics. By taking advantage of these recent opportunities, a novel source-to-sink analysis strategy is developed for wind-blown sand, at the interface between transport physics, geomorphology, morphochronology and climatology. Using this strategy, sand routes and fluxes are documented throughout deserts, as well as the morpho-sedimentary dynamics of associated landscapes over a few decades to a few million years. We thus hope to provide an original and quantitative description of the contribution of aeolian processes to the Earth system, while developing tools for a better understanding of their impact in arid zones.