The hydrological cycle of the Niger River basin simulated by the CORDEX-Africa regional climate models

Abstract:

The Niger River Basin (NRB) is a large transnational watershed of ~1.5 million km², whose water resources sustain more than 100 million people of nine countries in West Africa. Evaluating the reliability of climate simulations in the region is essential to support water sustainability and food security under possible future climatic changes and population growth. Here, we assess the ability of a set of state-of-the-art regional climate models (RCMs) of the COordinated Regional climate Downscaling EXperiment (CORDEX)-Africa to reproduce the hydrologic cycle of the NRB. For this aim, we adopt a verification framework based on the mass conservation principle that assumes that the mean annual difference between precipitation and evaporation equals the long-term mean discharge. We focus on four nested sub-basins encompassing different climatic zones with available discharge observations. We found that most RCMs overestimate the mean annual runoff (from +10% to +400%), because of a positive bias in the simulation of precipitation and a weak hydrologic cycle in the evaporation channel. Some exceptions are found in the more humid sub-basin upstream where a few climate simulations are not able to capture the intensity of the West African monsoon. Analyses of the water balance components also revealed that the signature of the RCMs is more significant than that of the driving General Circulation Model, likely due to the specific schemes adopted in the RCMs to parameterize the land-surface processes. This work is useful to increase the utility of regional climate simulations in impact studies supporting the development of water management polices and
planning of hydraulic infrastructures in the basin.