Modeling a Sahelian Water Resource Allocation under Climate Change and Human Pressure: Case of Loumbila Dam in Burkina Faso

Abstract

Alike most of West African countries, Burkina Faso is facing water resources availability issues. The water scarcity resulting from the rainfall decline of the 1970s is ongoing and represents one of the most challenging problems to tackle for the populations in this area of the world. Despite the implementation of adaptation measures mainly through realization of dams, water availability is still a crucial issue. Consequently, populations still lack safe water for multiple purposes. This seems to be exacerbated by the worst impact of climate change which would impose great change in climate variability as projected by several studies. Indeed, as climate change is inherently linked to the hydrological cycle, it may cause significant alterations in water resources. Nonetheless, despite the seriousness of water scarcity problems and its negative consequences on populations, information is still lacking to understand the past and future climate trend and their impact on the future water availability and demand. Therefore, there is a need to understand those processes in a view of adaptation planning. To model water allocation under climate change and human pressure in an ungauged basin, the Loumbila dam which plays a great role in Ouagadougou´s water supply system is taken as a case study to reconstitute inflows, simulate actual water allocation and predict monthly water allocation under different climate change scenarios. For these purposes, an assessment of the past climate variability in the basin was drawn. Then, the dam water balance was estimated and the water allocation was assessed. Four major conclusions are drawn from this thesis. By applying R-climdex and Instat, various indices were calculated based on an approach recommended by the World Meteorological Organization (WMO). Four indices related to precipitation (PRCPTOT, RX1day, CWD, CDD) and other four related to temperature (TXx, TNn, CSDI, WSDI) were selected. The first key findings of this study on the climate variability are that in general the intensity and frequency of extreme precipitation and temperature events are increasing. The total precipitation is decreasing at a change rate of -2.05 while the maximum one day precipitation is increasing at a change rate of 0.35. The rate of change in CWD is -0.01 and the rate of change in CDD is about 0.59. The rainfall season is becoming shorter and the occurrence of dry spells is also increasing with dry spells length varying between 5 and 35 days. The change rate in temperature indices is 0.03, PhD thesis 2 0.02, 0.06 and 0.28 respectively for TXx, TNn, CSDI and WSDI. By setting up a second set of software (ARC GIS) and models (CROPWAT, GR2M and Yates) to estimate the dam water balance, the results show the reconstitution of the dam inflow, the market gardening area and the crops water requirement. It was found that the market gardening around the dam occupied an area of 177ha. Finally, a serial number of sensitivity analysis with the water allocation model (WEAP) under RCP_8.5 and RCP_4.5 scenarios was done. RCMs data bias correction was set up based on delta change method used by Lettenmaier and Gan (1990). The assessment of water needs shows an upward trend of water supply meaning that this site will experience great challenges in the future. The unmet demand for different demand site will increase in the future due to the decrease in water availability and increase in water demand.