A study published in the Journal of Remote Sensing demonstrates a new method for retrieving surface solar radiation (SSR) from China's Fengyun-4A (FY-4A) geostationary satellite. The approach uses transfer learning to adapt knowledge from the Himawari-8 satellite, reducing reliance on ground observations and auxiliary meteorological data. This advance could strengthen solar power forecasting, climate modeling, and sustainable energy planning.
Surface solar radiation is critical for Earth's energy balance, hydrological cycles, and the performance of solar photovoltaic (PV) and concentrating solar power systems. Ground-based radiometric networks offer the most reliable measurements but are sparse, especially over oceans and developing regions. Reanalysis products provide broad coverage but often lack accuracy due to coarse resolution and simplified cloud-aerosol interactions. Satellite observations can fill these gaps, but many existing algorithms are sensor-specific and primarily estimate global radiation without separating direct and diffuse components.
Researchers from the Aerospace Information Research Institute, Chinese Academy of Sciences; Sichuan University of Science and Engineering; and the Institute of Atmospheric Physics, Chinese Academy of Sciences developed a transfer learning strategy that carries radiative knowledge from Himawari-8 to FY-4A. They first built a deep neural network (DNN) model using Himawari-8 Level 1 observations and the Cloud, Atmospheric Radiation and Renewal Energy Application (CARE) product, then fine-tuned the pretrained model with FY-4A data. The model uses top-of-atmosphere reflectance and solar-satellite geometry as inputs, with Bayesian optimization selecting key hyperparameters to improve generalization.
Validation against 33 ground stations from the Baseline Surface Radiation Network (BSRN), Bureau of Meteorology (BOM), and Global Tropical Moored Buoy Array (GTMBA) during 2018–2020 showed strong performance. At representative BSRN sites, FY-4A achieved instantaneous root mean square errors (RMSEs) of 102.2 W m⁻² for global, 117.5 W m⁻² for direct, and 83.1 W m⁻² for diffuse radiation. At daily mean scales, RMSEs dropped to 28.5, 30.1, and 22.6 W m⁻², respectively.
The study highlights how knowledge from a mature satellite product can be transferred to another platform to build new operational capability. By estimating both direct and diffuse radiation, the framework provides more actionable information than global radiation alone. Direct radiation is vital for concentrating solar power, while diffuse radiation affects PV output under cloudy or aerosol-rich skies. The method also reduces reliance on auxiliary meteorological data, making it more practical for near-real-time monitoring. This turns China's geostationary satellite observations into a more powerful resource for energy and climate applications.
Looking ahead, the same transfer learning strategy could be extended to other Chinese geostationary satellites, including Fengyun-4B (FY-4B), supporting more reliable solar-energy monitoring across East Asia and beyond. The study was supported by the National Natural Science Foundation of China and other funding sources. For more details, refer to the original study at https://spj.science.org/doi/10.34133/remotesensing.1044.
