The use of renewable energy sources, like solar, wind and biomass is rapidly increasing in recent years, with solar radiation being a particularly abundant energy source over Northeast Brazil. Thus, the proper quantitative knowledge of the incoming solar radiation is of great importance for energy generation planning in Brazil, serving as basis for developing future projects of photovoltaic power plants and solar energy exploration. This work presents a methodology for mapping the incoming solar radiation at ground level on Northeast Brazil, using a mesoscale atmospheric model (Regional Atmospheric Modeling System - RAMS), calibrated and validated using the automatic surface stations network data from the Ceará State Foundation for Meteorology and Water Resources (Fundação Cearense de Meteorologia e Recursos Hídricos- FUNCEME). The results showed that the model presents systematic errors, overestimating surface radiation, but, after a proper statistical corrections, using a relationship between the model-predicted cloud fraction and the observed radiation at ground level and the estimated at the top of the atmosphere, a 0.92 correlation value, with a confidence interval of 13,5 W/m², is obtained for monthly data. Using this methodology, an estimated 215 W/m² annual average incoming solar radiation over Ceará is found (maximum in October: 260 W/m²).
Solar Energy; Radiative Transfer; Numerical Modeling
