Abstract
Reanalysis products are the most often used data to characterize the Congo Basin climate system due to the lack of observational data. However, although some studies assessed their performance in reproducing the central equatorial Africa rainfall system, there is still a knowledge gap about the uncertainties that feature these datasets. Yet, they are widely used in the process-based evaluation of climate models over the basin. The present study is a first step towards assessing these uncertainties, i.e., the analysis of discrepancies between the ERA5, MERA2, and NCEP2 reanalysis datasets in the simulation of the thermodynamic processes governing the moisture balance. We focus on the Congo Basin’s atmospheric hydrological cycle while decaying the moisture budget into its dynamic and thermodynamic components during rainfall peaks in March–April–May (MAM) and September–October–November (SON). The results show that reanalysis data consistently simulate a prevalence of thermodynamic processes in the moisture provision into the basin during the little rainy season MAM, but instead a prominence of dynamic mechanisms during the great rainy season SON. Apart from MERRA2, there is also an agreement that the evaporation rate weakens in SON, in association with a decrease in the energy required to evaporate soil moisture. However, large uncertainties feature mechanisms contributing to the thermodynamic component during both seasons. Indeed, in MAM, surface solar radiation is the factor maintaining high evaporation in ERA5 and NCEP2 while soil moisture does so in MERRA2. In SON, higher values of soil moisture together with the surface latent heat flux are factors inducing higher evaporation in MERRA2 than in ERA5 and NCEP2. Also, the high rate of evaporation in NCEP2 (3.61 mm/day) than in ERA5 (3.52 mm/day) is associated with the difference in their surface irradiation (23 W. m−2).
DOI Link to Publication