Numerical prediction of solar radiation in South Korea: assessment of WRF-Solar and WRF

This study aims to comparatively evaluate the performance of the Weather Research and Forecasting (WRF) model and the WRF-Solar model for forecasting global horizontal irradiance (GHI), a key factor in photovoltaic (PV) power generation, over South Korea during summer 2023. Experiments were conducted using WRF (CNTL) and WRF-Solar (EXP_R and EXP_F): All experiments used the same physical parameterization schemes, except for those specialized in WRF-Solar—Thompson–Eidhammer microphysics and Deng shallow cumulus (used in EXP_R and EXP_F), and Fast All-sky Radiation Model for Solar applications (FARMS) schemes (used only in EXP_F).

The GHI forecasts were evaluated across all (ALL), clear (CLR), cloudy (CLD), and overcast (OVC) sky conditions using in-situ and remote sensing data. Both EXP_R and EXP_F outperformed CNTL, with RMSE improvement of 7.2 − 19.9 % for EXP_R and 6.8 − 37.0 % for EXP_F. All experiments performed similarly under CLR; however, EXP_F outperformed EXP_R in other conditions, achieving up to 37.0 % improvement under OVC. The spatial distribution analysis also indicated improvements in WRF-Solar experiments, with EXP_F outperforming EXP_R. Additionally, WRF-Solar showed better cloud detection performance than WRF. In conclusion, this study demonstrates the first successful application of WRF-Solar in South Korea and highlights its potential to enhance energy management and solar power planning.