Estimation of Streamflow and Sediment Loads in the White Volta Basin Under Future Climate Projections

Abstract

Among the major sub-catchments of the Volta Basin is the White Volta Basin which is shared between Burkina Faso and Ghana in West Africa. Recent socio-economic developments in the two riparian countries have placed immense pressure on the quantity and quality of the water resources in the basin. Information on streamflow and sediment loads in the basin will therefore play a vital role in sustainably developing and managing the water resources of the basin. This study therefore seeks to estimate long-term time-series of sediment loads by evaluating empirical relationships between turbidity, streamflow and suspended sediment concentration using regression models. An evaluation of the impact of climate projections on streamflows and sediment loads in the White Volta Basin using the Soil Water Assessment Tool (SWAT) coupled with an ensemble of three Regional Climate Models (RCMs) under the CORDEX-Africa Project was also undertaken. Regression analysis showed that a simple linear regression equation derived by relating turbidity and suspended-sediment concentration with a validated model efficiency of 93% was the better model for estimating long-term suspended sediments loads in the White Volta Basin. The SWAT model was calibrated and validated for the periods 1991-2003 and 2004-2013 respectively with Nawuni hydrometric station as the watershed outlet. The model was also spatially validated at Pwalugu and Nasia. Analysis of the water balance of the basin shows that 4.90% of the simulated mean annual precipitation is converted to surface runoff while approximately 85% evapotranspires. The results also show that, based on the period 1990-2010, the White Volta Basin contributes annually an estimated 8.2106 metric tons of sediment load into the Volta Lake barring any deposition between Nawuni and the lake. The RCM-GCMs ensembles were used to downscale precipitation and minimum and maximum temperature for the reference period (1990-2010) and the future period (2031-2050). All the RCM-GCMs ensembles forced by the Representative Concentration Pathways (RCP4.5 and 8.5) projects a rise in temperature by 2.3°C and 2.7°C respectively. The RCM-GCMs however predicted mixed projections for precipitation. The downscaled climate data was then used as inputs into the calibrated SWAT model to simulate and compare the water balance of the study area for the reference and future periods. Similar to the simulated precipitation, projected surface runoff and actual evapotranspiration have mixed results based on the RCM-GCM ensemble used for the simulation. Surface runoff is however projected to increase on average by 23.8% and 27.8% for RCP4.5 and RCP8.5 respectively whereas evapotranspiration is projected to decline on average by 1.5% and 1.0% based on RCP4.5 and RCP8.5 respectively. The results also show that irrespective of the scenario, annual sediment loads in the basin is projected to increase on average by 24.7% and 26.3% under RCP4.5 and RCP8.5 respectively. Finally, extreme value analysis of maximum precipitation for the wet months of July, August and September showed that the magnitude of extreme rainfall events are likely to increase in the future with direct consequence for the generation and transport of sediment loads in the basin.