Climate Change and Land Use Land Cover Change Impacts on the Hydropower Potential in the Bafing Watershed (Senegal River Basin)

Abstract

strategic region for the socio-economic development of these countries. The Senegal River. Basin is divided into three main parts: The upper basin, the valley and the delta. The Bafing watershed is the main tributary of the Senegal River and is located in the Upper Senegal River Basin. The management of the Bafing watershed in time and space is possible thanks to the Manantali hydropower dam. The Manantali dam aims to meet the growing water, energy and agriculture need of the member states (Senegal, Mali, Guinea, Mauritania). The organization for the development of the Senegal River (OMVS) plans to build new hydropower dams (Koukoutamba, Boureya) upstream of the Manantali dam to increase hydropower potential in the Bafing watershed. In the future, water availability and hydropower generation are expected to be profoundly impacted, mainly due to the change in river flow caused by population growth, climate change, and Land use/land cover (LULC) change. In the coming decades, climate change and changes in LULC will further increase the constraints on the already scarce water resources in West Africa. Despite the number of studies and projects carried out on the Bafing watershed, there are not yet studies that have addressed the hydrological and hydropower potential (HPP) responses considering the combined impact of future climate change, LULC change and the future development of planned dams in the Bafing watershed. Therefore, this study aims to fill this gap by investigating the future impacts of climate change, LULC change, and altered water resource management on the water availability and hydropower potential (HPP) in the Bafing watershed. Firstly, two precipitation products (reanalysis (W-era5) and satellite (CHIRPS)) were compared to the observed precipitation of the Bafing Makana station due to insufficient data caused by numerous gaps in the historical time series. This exercise was done to select the best precipitation product to reproduce the observed precipitation. The results showed that W-era5 represents the observed data more accurately than CHIRPS. After, ten downscaled and bias adjusted Global Climate Models from ISIMIP 3b were investigated to determine whether the models satisfactorily replicate the reference climate (temperature and precipitation of W-era5) of the Bafing watershed. The results indicated that the 10 GCMs could successfully replicate the reference climate. Hence, the median of the 10 GCMs (MME) was used to analyze the future trend in the near future (P1:2035-2065) and the far future (P2:2065-2095/2066-2095) compared to the reference period (P0:1984-2014) under ssp 126 and ssp 370. The results indicated that, according to the median (MME), a rise in temperature by 1.4°C and 2.0°C under ssp126 and ssp370 is predicted in the near future. In the far future, the difference between both climate scenarios is much larger and spans from 1.6°C to 3.7°C. Projected precipitation is uncertain in the future. Indeed, precipitation is predicted to increase under ssp126 or decrease under ssp 370 in the near future. In the far future, precipitation is expected to decrease under both scenarios. Secondly, the past and future LULC change was analyzed between 1986 to 2020 and 2020 to 2050. Landsat images and the random forest classification method were used to map LULC of 1986, 2006 and 2020. Future LULC map in 2050 were simulated under business-as-usual assumptions with the Multi-Layer Perceptron and Markov Chain method embedded in the Land Change Modeller software. The LULC change was analyzing using the post classification change detection technique, a pixel-based method. The results showed that between 1986 to 2020, vegetation, settlement, cultivated area and water increased, while the bareground decreased. Between 2020-2050, the results indicated that vegetation, settlement, cultivated area, and water are projected to increase. The Bafing watershed has seen a trend towards "more people, more trees". Thirdly, an eco-hydrological water management model, the Soil and Water Integrated Model (SWIM), was set up and used to generate river discharge and simulate existing and future dams. SWIM model was driven by ten downscaled and bias adjusted GCMs under ssp 126 and ssp 370 and LULC maps (1986, 2020, 2050). The analysis was carried out using a separation method that includes combining the two components (climate and LULC) and adjusting one factor at a time while holding the other constant. The result indicated that SWIM satisfactorily reproduces the observed flow with statistical performance measures (NSE, KGE) between 0.7 and 0.8. Reservoir module also satisfactory reproduce the inflow, outflow, and water level of the Manantali dam. Under the impact of climate change, the result of the SWIM simulation indicated that the inflow and the HPP of the Manantali dam will decrease except in the near future under ssp 126, following the general trend of the precipitation in the future. Under the impact of LULC change, the inflow and the HPP of the Manantali dam will decrease by -5% and -5.7 respectively due to the conversion of bareground (with high runoff coefficients) to vegetation and cultivated area (low runoff coefficients) during the period 1986-2050. Under the effects of climate change and LULC change, the result of the SWIM simulation pointed out that LULC change has less impact on the inflow and the HPP of the Manantali dam than climate change. Investment in future dams has advantages, such as increased water storage, greater hydropower potential and improved flood protection. However, future dams will be negatively affected by climate change in the future (except in the near future under ssp 126), and their operation will lead to a loss in the hydropower potential of the Manantali dam. Therefore, the implementation of adaptation techniques to mitigate the impact of climate change and LULC change, as well as the effects of the environmental and social impacts of these dams is essential. Adaptation techniques can be an optimization program or adopting a new common energy policy promoting an energy mix that prioritizes renewable energies, namely solar and wind. The results of this study provide relevant information to the OMVS for the management of the Bafing watershed.