Application of ERA-Interim, empirical models, and an artificial intelligence-based model for estimating daily solar radiation

Solar radiation plays a pivotal role in the energy balance at Earth's surface, evaporation, snow melting, water requirements of plants, and hydrological control catchments. In this work, performance ERA-Interim (a reanalysis dataset) was examined to estimate solar Ahvaz, BandarAbbas, Kermanshah weather stations representing even spatial distribution over Iran using eight empirical models an artificial intelligence-based model (SVM: Support Vector Machine). calibration set, SVM exhibited best with RMSEs 249, 299 437 J.cm−2.day−1 aforementioned stations, respectively. validation reduced errors estimates by 2.5 7.3 percent compared Ahvaz station (Abdallah model, RMSE = 242 J.cm−2.day−1) (Angstrom-Prescott 315 J.cm−2.day−1), During BandarAbbas station, Bahel Abdallah (RMSE 309 Angstrom-Prescott 310 312 showed relatively similar performance. The results also that dataset can be comparatively suitable alternative some models, where or input parameters are not directly measured, ​​of 382.81, 320.82 414.1 respectively (in phase); although its error rates significant substituting it for is recommended.