Modeling climate and land use change impacts on water resources in the Dano catchment (Burkina Faso, West Africa).

This study investigated the impacts of climate and land use changes on water resources in the Dano catchment combining hydrological processes understanding, hydrological simulations and climate and land use scenarios application. The catchment covers about 195 km2 and is located in the Southwest of Burkina Faso in West Africa. The study area is characterized by an annual population growth of about 3% over the past decades.
Based on intensive field investigations on soil hydraulic properties, instrumentation and monitoring of hydro-meteorological variables such as discharge, soil moisture, groundwater level, precipitation, temperature etc. the distributed and physically based hydrological simulation model WaSiM was successfully calibrated and validated for the catchment. Achieved model statistical quality measures (R2, NSE and KGE) ranged between 0.6 and 0.9 for total discharge, soil moisture, and groundwater level, indicating a good agreement between observed and simulated variables.
Land use change assessment in the catchment over the period of 1990-2013 exhibited a decrease of natural and semi-natural vegetation at an annual rate of about 2%. Conversely, cropland, and to a lesser extend urban areas, have increased. Land conversion was attributed to population growth, changing in farming practices and environmental conditions.
Four land use maps were used to build land use scenarios corresponding to different levels of land use change in the catchment. Application of the land use scenarios to the calibrated and validated hydrological model indicated that, compared to the land use status in 1990, the current situation leads to an increase in total discharge of about 17% and a decrease of actual evapotranspiration of about 5%. The results of simulations further showed that the increase in total discharge is related to high peak flow, suggesting an alteration of flood risk.
Following field measurements that showed infiltration rates 1.2 times higher under semi-natural vegetation compared to cropland, land use change related effects on soil infiltration rate was integrated in the modeling of LULC change impact assessment. Model results with a refined soil (integrating this additional information) and a classic soil indicated similar trends in water balance components as a result of land use change. However slight differences of 0.5 to 20 mm/year in the water balance component were noticed between the two soil parameterization approaches. The integration of land use related effects on soil properties was suggested to render LULC change scenarios more plausible.
The projected climate change signal in the catchment was analyzed using the representative concentration pathways 4.5 and 8.5 of six datasets of the COordinated Regional climate Downscaling Experiment-project. Compared to the reference period of 1971-2000, the climate models ensemble consistently projects an increased temperature of 0.1 to 2.6°C over the period 2021-2050. However, an agreement was not found among climate models with regards to precipitation change signal as projections for annual rainfall ranged between -13 and +18%.
The application of the climate models ensemble in WaSiM showed future discharge change signals very similar to the projected precipitation change. Individual climate models showed opposite annual discharge change signals ranging from -40 to +50 %. On average, the climate models ensemble suggested a 7 % increase in annual discharge under RCP4.5 and a 2 % decrease under RCP8.5. The analysis of the catchment sensitivity to precipitation and temperature change indicated that discharge is more related to precipitation than to temperature as the environmental system of the catchment is water limited and not energy limited.