In Situ Based Performance of Satellite- and Reanalysis-derived Hourly Global Horizontal Irradiance Products over West Africa

Global horizontal irradiance (GHI) is the primary driver for photovoltaic (PV) technology. For PV system design and monitoring, hourly and sub-hourly GHI from reanalysis and satellite-based data are frequently used, especially in data-poor regions like Sub-Saharan Africa. However, the use of these datasets is uncertain and need to be assessed in detail. In this study, we evaluated the performance of state-of-the-art reanalysis and satellite-based datasets (ERA5, CAMS, MERRA-2 and SARAH-2) with in situ measurements of hourly GHI for the main climatic zones (Guinea, Savannah, and Sahel) in West Africa. The in situ measurements come from novel regional and national meteorological networks consisting of 51 automatic weather stations in Burkina Faso and Ghana. The performance assessment was done for the year 2020 for different weather conditions (cloudy, clear and all sky). The effects of clouds and aerosols were also investigated. Moreover, a new overall performance measure is introduced for joint evaluation of different standard measures, such as the root-mean-square error and the index of agreement. The results show that the data from SARAH-2 performs best under cloudy-sky conditions, while ERA5 performs worse under all atmospheric conditions. The low performance under cloudy skies for all datasets is the result of a large bias observed during the Harmattan period, when the region has a high concentration of aerosols. The average diurnal variation of GHI shows good agreement between the in situ measurements, the satellite and the reanalysis data under clear and all-sky conditions, but an overestimation under cloudy skies at some stations. The new overall performance value clearly indicates hourly GHI from SARAH-2 is the best alternative for assessing solar energy in West Africa.