Statistical models for mapping solar irradiation across southern Thailand using meteorological and satellite data

A statistical model is developed to calculate the monthly average daily solar irradiation on the ground by considering cloud fraction and temperature. Cloud fraction data were gathered from 29 meteorological stations throughout southern Thailand and from the Aura satellite. Average temperature values were collected from the FLDAS model, through the utilization of interpolation techniques. In addition, solar irradiation data from ground-based pyranometers was collected for 2014 to 2023 from the Songkhla meteorological center to serve as the basis for the modeling ( R 2 = 0.70–0.84). The data from the developed model and the measured data were in good agreement, with a Root Mean Square Deviation (RMSD) score of 6.84%–9.37% (Mean Absolute Percent Error, RMSE-observations of the Standard Deviation Ratio, Nash-Sutcliffe Efficiency, and Peak Irradiance to Noise scores were also very good). Solar irradiation estimated from data acquired by the GLDAS model was in reasonable agreement with the model estimate, with RMSD score of 20.93%. The model was utilized to generate solar irradiation maps for southern Thailand, using kriging, IDW, Spline, Trend, KIB, DIW, LPI, and RBF estimation techniques. The map, which was similar to maps derived from a more complex model, showed that solar irradiation in the region is mainly influenced by mountains, and the southeast and northwest monsoons, which cause dense cloud cover.