Modeling and Variability Assessment of Global Solar Radiation in Gwagwalada, North-Central Nigeria: A Comparison of the Hargreaves-Samani and Angstrom-Prescott Models

Abstract

Reliable estimates of global solar radiation are essential for the design, sizing, and performance evaluation of solar energy systems, particularly in regions where ground-based radiometric data are sparse. This study assessed the variability of global solar radiation and solar irradiance in Gwagwalada, a town in Nigeria's Federal Capital Territory (8.95°N, 7.08°E), using the Hargreaves-Samani and Angstrom-Prescott models. Monthly average maximum and minimum temperature, daylight hours, and measured global solar radiation and irradiance data spanning January 2020 to December 2022 were obtained from the Nigerian Meteorological Agency (NIMET). Model outputs were validated against measured radiation using mean bias error (MBE), root mean square error (RMSE), mean percentage error (MPE), and the coefficient of determination (R²). The Hargreaves-Samani model showed a stronger association with measured radiation (R² = 0.981, RMSE = 1.100 MJ/m²/day, MBE = 0.999 MJ/m²/day) than the Angstrom-Prescott model (R² = 0.902, RMSE = 2.031 MJ/m²/day, MBE = 1.734 MJ/m²/day), indicating that a temperature-based estimation approach is more reliable than a sunshine-duration-based approach at this location. Global solar radiation was consistently higher during the dry season (November to March) and lower during the wet season (April to October), consistent with reduced cloud cover and lower atmospheric moisture in the dry months, with the annual peak occurring in February. Daily global solar radiation ranged from approximately 11.4 to 19.9 MJ/m²/day in the wet season and 17.4 to 21.8 MJ/m²/day in the dry season.  These findings provide a quantitative basis for solar energy system planning in Gwagwalada and add to the limited body of location-specific solar radiation data available for North-Central Nigeria.