Remobilization of southern African desert dune systems by twenty-first century global warming.

Although desert dunes cover 5 per cent of the global land surface and 30 per cent of Africa, the potential impacts of twenty-first century global warming on desert dune systems are not well understood. The inactive Sahel and southern African dune systems, which developed in multiple arid phases since the last interglacial period, are used today by pastoral and agricultural systems that could be disrupted if climate change alters twenty-first century dune dynamics. Empirical data and model simulations have established that the interplay between dune surface erodibility (determined by vegetation cover and moisture availability) and atmospheric erosivity (determined by wind energy) is critical for dunefield dynamics. This relationship between erodibility and erosivity is susceptible to climate-change impacts. Here we use simulations with three global climate models and a range of emission scenarios to assess the potential future activity of three Kalahari dunefields. We determine monthly values of dune activity by modifying and improving an established dune mobility index so that it can account for global climate model data outputs. We find that, regardless of the emission scenario used, significantly enhanced dune activity is simulated in the southern dunefield by 2039, and in the eastern and northern dunefields by 2069. By 2099 all dunefields are highly dynamic, from northern South Africa to Angola and Zambia. Our results suggest that dunefields are likely to be reactivated (the sand will become significantly exposed and move) as a consequence of twenty-first century climate warming.