Search

Increasing watershed sediment yields are an important problem in Africa but the sources of these sediment yields have only very rarely been investigated. This study therefor aims to discriminate subsurface and surface sources of fine grain sediments in a representative mesoscale (580 km2) West African savanna watershed. We used a sediment source fingerprinting approach for source apportionment including geochemical and radionuclide (137Cs, 210Pbex, 7Be) composite signatures where 7Be was used as a tracer for the first time in the African environment. Two field campaigns were conducted collecting a total of 258 geochemical and 66 isotope samples. We found that subsurface source categories, dominantly river bank, contributed an unexpected high share of 43% (geochemistry) and 45% (radionuclides) to the sampled fine‐grained sediments. Pairwise comparison of the averaged frequency distributions for predicted source proportions using five geochemical signatures with the frequency distribution generated using the single radionuclide signature suggested that the two distributions are not statistically different. Extrapolating our measured contribution of subsurface erosion to areas with similar yields in comparable environmental settings, we can assume that subsurface sources are an important component of sediment loss across large areas of West Africa. Subsurface erosion, primarily associated with bank rather than gully erosion, is likely to increase in the future with projected rises in runoff due to land use and climate change. Source tracing studies need to be undertaken more widely across Africa to help mitigation planning for sediment‐related and land degradation problems.

This study investigates the effect of land use and land cover (LULC) and climate change on catchment hydrology and soil erosion in the Dano catchment in south-western Burkina Faso based on hydrological and soil erosion modeling. The past LULC change is studied using land use maps of the years 1990, 2000, 2007 and 2013. Based on these maps future LULC scenarios were developed for the years 2019, 2025 and 2030. The observed past and modeled future LULC are used to feed SHETRAN, a hydrological and soil erosion model. Observed and modeled climate data cover the period 1990–2030.The isolated influence of LULC change assuming a constant climate is simulated by applying the seven LULC maps under observed climate data of the period 1990–2015. The isolated effect of climate scenarios (RCP4.5 and 8.5 of CCLM4–8) is studied by applying the LULC map of 1990 to the period 1990–2032. Additionally, we combined past modeled climate data and past observed LULC maps. Two chronological and continuous simulations were used to estimate the impact of LULC in the past and in the future by gradually applying the LULC maps. These simulations consider the combined impact of LULC and climate change.The simulations that assumed a constant climate and a changing LULC show increasing water yield (3.6%–46.5%) and mainly increasing specific sediment yield (−3.3%–52.6%). The simulations that assume constant LULC and climate as changing factor indicate increases in water yield of 24.5% to 46.7% and in sediment yield of 31.1% to 54.7% between the periods 1990–2005 and 2006–2032. The continuous simulations signal a clear increase in water yield (20.3%–73.4%) and specific sediment yield (24.7% to 90.1%). Actual evapotranspiration is estimated to change by between −7.3% (only LUCC) to +3.3% (only climate change). When comparing observed LULC and climate change alone, climate change has a larger impact on discharge and sediment yield, but LULC amplifies climate change impacts strongly. However, future LU

Soil erosion is recognized as one main reason for soil degradation in West Africa. However, predictions on the impact of climate change on soil erosion are rare for most West African countries including Burkina Faso.This study assesses the impact of climate change on water resources and soil erosion in a small catchment (126 km2) in southwestern Burkina Faso. Climate data from an ensemble of six regional (RCM) and global (GCM) climate models were used to run the physically based spatially distributed hydrological and soil erosion model SHETRAN. The Representative Concentration Pathways (RCPs) 4.5 and 8.5 were selected as future climate scenarios.Bias corrected precipitation and temperature required for the calculation of potential evapotranspiration were used as input for the SHETRAN model to simulate total discharge and specific suspended sediment yield (SSY). Discharge and SSY from simulations run with climate data were able to reproduce discharge and SSY from a simulation that used observed precipitation and temperature from the historical period (1971–2000).The impact of climate change on hydrology and soil erosion was assessed by comparing the historical period with the future climate scenarios (2021–2050). Most of the used climate models predict an increase of temperature between 0.9 °C and 2.0 °C. The bias correction did not alter the climate change signal of temperature. Large uncertainties among the RCMs-GCMs exist regarding the climate change signal of future precipitation. Some climate models predict an increased (5.9%–36.5%) others a decreased (6.4%–10.9%) or mixed signal. The applied bias correction did not reverse the climate change signal in most cases but it influenced magnitude and timing of precipitation. The ensemble mean suggests an increased discharge between 19.5% (RCP 8.5) and 36.5% (RCP4.5) and an increased SSY of the same order. In general, the climate change signal and the corresponding discharge and SSY predictio

This study presents the calibration and validation of the physically based spatially  distributed hydrological and soil erosion model SHETRAN for the Dano catchment, Burkina Faso.  A sensitivity analysis of six model parameters was performed to assess the model response and to  reduce the number of parameters for calibration. The hydrological component was calibrated and  validated using observed discharge data of two years. Statistical quality measures (R2, NSE, KGE)  ranged from 0.79 to 0.66 during calibration and validation. The calibrated hydrological component  was  used  to  feed  the  erosion  modeling.  The  simulated  suspended  sediment  load  (SSL)  was  compared with turbidity‐based measurements of SSL of two years. Achieved quality measures are  comparable to other SHETRAN studies. Uncertainties of measured discharge and suspended  sediment  concentration  were  determined&amp