The Makgadikgadi Pans in northern Botswana are the desiccated relicts of a former major inland lake system, with fossil shorelines preserved at five distinct elevations (~995 m, 945 m, 936 m, 920 m and 912 m). These lakes persisted in the Makgadikgadi Basin, which evolved in the Okavango-Makgadikgadi Rift Zone: the south-western extension of the East African Rift System (EARS) into northern Botswana. This paper synthesizes cross-disciplinary evidence, which reveals that the antiquity of this lake complex has been widely underestimated. It presents a Regional Drainage Evolution Model that invokes tectonically initiated drainage reorganizations as the underlying control over lake evolution. Lake formation was initiated by rift-flank uplift along the Chobe Fault, across the course of the Zambezi River, which diverted the regional drainage net into the Makgadikgadi Basin. Filling of the basin initiated a major climatic feedback mechanism that locally increased rainfall and lowered evaporation rates. This progressively enhanced water input to the basin, and most likely led to overtopping of the Chobe Horst barrier during the three highest lake stands, with outflow into the Zambezi River. During this period, the hydrology of the basin would have been closely analogous to modern, shallow Lake Victoria. Fragmentation of the regional drainage network by successive river captures resulted in sequential contractions of the lake to lower elevation shorelines. In turn, resultant decreases in areas of these successive lakes modulated the magnitude of the feedback mechanism. Thus, loss of the Upper Chambeshi catchment caused the lake to drop from the 990 to the 945 m level. Severance of the former link between the Kafue and Zambezi resulted in a further drop to the 936 m shoreline. Inflow declined further after the impoundment of a major lake (Palaeo-Lake Bulozi) on the Upper Zambezi River, causing contraction to the 920 m shoreline. Continued incision of the Zambezi channel into the Chobe horst barrier ultimately terminated input from this river to the Makgadikgadi depression, causing contraction of the lake below 920 m, sustained by the Cuando and Okavango prior to final desiccation. This Regional Drainage Evolution Model contradicts previous proposals that have invoked Late Pleistocene climatic forcing to explain inferred fluctuations in lake levels. The timeframe developed for the drainage reorganizations requires that the lake was initiated by ~1.40 to 0.51 Ma at the most recent (Early – Mid-Pleistocene), while archaeological evidence shows that it had contracted below the 936 m shoreline before 500 ka. This contrasts with ¹⁴C and quartz luminescence dates (generally <100 ka), which require that the 945 m lake stage was extant during much of the Upper Pleistocene. The calcareous radiocarbon dates reflect multiple episodes of calcrete formation, while the young luminescence dates are ascribed to the extensive bioturbation of older Kalahari landforms.

We present a dust plume source inventory for southern Africa. In order to locate and track the local, short‐lived plume events, source and frequency data have been derived from Meteosat Second Generation (MSG) thermal infrared composite data (4 km data using 8.7, 10.8, and 12.0 µm) and Moderate Resolution Imaging Spectroradiometer (MODIS) visible composite data (0.25 km data utilizing 0.620 – 0.670 µm, 0.545 – 0.565 µm, and 0.459 – 0.479 µm). Between January 2005 and December 2008, a total of 328 distinct daytime dust plumes more than 10 km in length were detected. These plumes were attributed to 101 distinct point sources, consisting largely of ephemeral inland lakes, coastal pans as well as dry river valleys in Namibia, Botswana, and South Africa. These data also provided sub‐basin scale source observations for large basins such as Etosha and Makgadikgadi Pans.

Modeled estimates of eolian dust emission can vary by an order of magnitude due to the spatiotemporal heterogeneity of emissions. To better constrain location and magnitude of emissions, a surface erodibility factor is typically employed in models. Several landscape‐scale schemes representing surface dust emission potential for use in models have recently been proposed, but validation of such schemes has only been attempted indirectly with medium‐resolution remote sensing of mineral aerosol loadings and high‐resolution land surface mapping. In this study, we used dust emission source points identified in Namibia with Landsat imagery together with field‐based dust emission measurements using a Portable In‐situ Wind Erosion Laboratory wind tunnel to assess the performance of schemes aiming to represent erodibility in global dust cycle modeling. From analyses of the surface and samples taken at the time of wind tunnel testing, a Boosted Regression Tree analysis identified the significant factors controlling erodibility based on Portable In‐situ Wind Erosion Laboratory dust flux measurements and various surface characteristics, such as soil moisture, particle size, crusting degree, and mineralogy. Despite recent attention to improving the characterization of surface dust emission potential, our assessment indicates a high level of variability in the measured fluxes within similar geomorphologic classes. This variability poses challenges to dust modeling attempts based on geomorphology and/or spectral‐defined classes. Our approach using high‐resolution identification of dust sources to guide ground‐based testing of emissivity offers a valuable means to help constrain and validate dust emission schemes. Detailed determination of the relative strength of factors controlling emission can provide further improvement to regional and global dust cycle modeling.

The sandy croplands in the Free State have been identified as one of the main dust sources in South Africa. The aim of this study was to investigate the occurrence and strength of physical soil crusts on cropland soils in the Free State, to identify the rainfall required to form a stable crust, and to test their impact on dust emissions. Crust strength was measured using a fall cone penetrometer and a torvane, while laboratory rainfall simulations were used to form experimental crusts. Dust emissions were measured with a Portable In-Situ Wind Erosion Laboratory (PI-SWERL). The laboratory rainfall simulations showed that stable crusts could be formed by 15 mm of rainfall. The PI-SWERL experiments illustrated that the PM₁₀ emission flux of such crusts is between 0.14% and 0.26% of that of a non-crusted Luvisol and Arenosol, respectively. The presence of abraders on the crust can increase the emissions up to 4% and 8% of the non-crusted dust flux. Overall, our study shows that crusts in the field are potentially strong enough to protect the soil surfaces against wind erosion during a phase of the cropping cycle when the soil surface is not protected by plants.

(2013). The surface geology and geomorphology around gobabeb, namib desert, namibia. Geografiska Annaler: Series A, Physical Geography: Vol. 95, No. 4, pp. 271-284. doi: 10.1111/geoa.12028