Assessment of the performance of ground-based and SARAH-3 solar radiation data over Eastern Africa

Solar radiation can be a potential source of cost-effective and sustainable energy, offering the promise of a clean and safe environment in the East African region. However, reliable solar radiation data remain limited, hindering accurate assessment of the available solar potential and characterisation of the local and regional climate. To address this issue, this study used in situ measurements from 11 ground-based meteorological stations to comprehensively evaluate the performance of the satellite-derived solar radiation product, SARAH-3, across Kenya, Tanzania, Uganda, and Rwanda. Statistical distribution, degree of linear association, explanatory strength, infra-daily, daily, and seasonal cycles were used to assess the degree of similarity. Our findings revealed a significant agreement between surface reference data and SARAH-3 Global Horizontal Irradiance (GHI) data and a significant linear association between bias and the Orographic Complexity Index (OCI) of the neighbour area of the observation points with Pearson’s correlation coefficients (R) of −0.53. At half-hour resolution, in situ measurements and satellite-based solar radiation data revealed high R values, ranging from 0.70 to 0.92, with significant explanatory capacity between 54% and 86%. Percentage bias (PB) was markedly lower than 5% for most stations, except in Rwanda, where PB reached 25%. The distributions of the data sets were generally skewed in the same direction, with a slight difference. Linear association and explanatory capacity improved significantly on the daily time scale, with R exceeding 0.90 and coefficients of determination greater than 78% at all stations. Furthermore, PB and Mean Absolute Error (MAE) were generally low on the monthly scale, except in Rwanda, particularly in Mukarange, where they reached 24.61% and 42.92, respectively. SARAH-3 successfully reproduces infra-daily, daily, and cycles, although its spatial performance is strongly influenced by terrain complexity. These findings confirm the potential of SARAH-3 for solar energy production and climate-related studies in the East African region.