A new transfer learning framework is enabling China's Fengyun-4A (FY-4A) geostationary satellite to estimate surface solar radiation (SSR) and its global, direct, and diffuse components with high accuracy, according to a study published in the Journal of Remote Sensing on April 29, 2026. The work addresses a critical need for accurate sunlight data to support the clean-energy transition, as clouds, aerosols, and atmospheric changes can rapidly alter the amount of solar radiation reaching Earth's surface.
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 method that adapts knowledge from the Himawari-8-based Cloud, Atmospheric Radiation and Renewal Energy Application (CARE) product. By using a deep neural network (DNN) pretrained on Himawari-8 data and fine-tuning it with FY-4A Level 1 observations, the model reduces dependence on auxiliary meteorological datasets. The framework uses top-of-atmosphere reflectance and solar–satellite geometry as dynamic inputs, with Bayesian optimization selecting key hyperparameters to improve generalization and efficiency.
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, 117.5, and 83.1 W m⁻² for global, direct, and diffuse radiation, respectively. At the daily mean scale, RMSEs dropped to 28.5, 30.1, and 22.6 W m⁻².
The study's key advance is demonstrating how knowledge from a mature satellite product can be transferred to another platform, turning China's geostationary satellite observations into a more powerful resource for energy and climate applications. The authors noted that the framework allows FY-4A to estimate not only total sunlight but also the direct and diffuse components, which are crucial for determining how solar energy systems perform under clear, cloudy, and hazy conditions. Direct radiation is especially important for concentrating solar power, while diffuse radiation affects photovoltaic output under cloudy or aerosol-rich skies.
By resolving these components separately, the framework offers more actionable information than global radiation alone. The new FY-4A radiation product could help improve photovoltaic site assessment, power forecasting, grid management, climate modeling, and land-surface simulations. The study also demonstrates that transfer learning can help overcome sensor differences and limited ground training data. Looking ahead, the same 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 bodies. The full article is available at DOI: 10.34133/remotesensing.1044. For more information about the research, visit the Journal of Remote Sensing website.
