Modelling and prediction of Ouémé (Bénin) river flows by 2040 based on GR2M approach

Abstract

The study described in this paper has consisted of simulating the real future behaviour of the Ouémé river basin's water flow by 2040. The approach of the Rural Engineering model, with two variables, at a monthly time step named GR2M has been adopted owing to its robustness. This model is an indispensable tool for studying the evolution of water resources in the medium and long term. For our approach, the projected data, from 2016 to 2040, were generated using the Auto-Regressive Integrated Moving Average model. Observed data were then used for the calibration and the validation of the GR2M model. The results obtained showed that the GR2M model is a very satisfactory tool for simulating the transformation of rainfall data into flows on the one hand and an impact of future climate change resulting in a decrease in annual average flows between 11.90% and 46.37% by 2040 on the other hand. The quality parameters revealed very interesting values obtained from the model on the three representative sites of the Ouémé basin with Nash-Sutcliffe more than 70% and determination coefficient more than 0.75.