Hydrological modeling with physics-based models in the oueme basin: Issues and perspectives for simulation optimization

Abstract

Oueme Basin, threatened by climate change and upstream energy projects, is limited by the quality of the adapted meteorological data through the assessment of their impacts with some models. Study focus: This study focuses on the ParFlow-CLM model and describes its ability to simulate the water balances with good (observation data) and bad (observation transformed and reanalysis corrected and uncorrected data) meteorological forcing. New hydrological insights for the region: The model correctly simulates all the terms of water balance, mainly evaporation and water table depth (WTD) with KGE always higher than 0.70. It is less sensitive to temperature, shortwave (SW), and longwave (LW) forcing. With those three variables, ParFlow-CLM simulates none significative difference in water balance terms using the reanalysis forcings on which we note an overestimation of temperature and an underestimation of LW. The same results are obtained with the other so-called bad meteorological forcings. On the other hand, the precipitation of the latter significantly influences the water balance. The annual water balance and especially evapotranspiration shows 683 ± 70 mm for observations versus 753 ± 63 mm and 662 ± 44 mm for ERA5 and MERRA2 respectively. The corrections applied on these variables seem to work better on MERRA2 (734 mm or 51 mm difference with observations) compared to ERA5 (760 mm or 77 mm difference with observations). The temperate, SW, LW variables of the reanalyses can be used for hydrological simulations with ParFlow-CLM and probably with other physics-based models over the Beninese basins but the precipitation of these must be corrected before use.