Solar radiation plays a critical role in shaping Earth's climate and sustaining life on the planet. Among its key components, downward surface solar radiation (DSSR), ultraviolet radiation (UVR), and photosynthetically active radiation (PAR) each exert unique influences on energy production, human health, air quality, and ecosystem functioning. However, acquiring accurate, high-resolution radiation data across a vast country such as China has long posed a significant challenge, due to the high cost and complexity of ground-based observations.

To tackle this obstacle, a research team from the Institute of Atmospheric Physics of the Chinese Academy of Sciences has developed an advanced deep learning model named the FT-Transformer. The model reconstructs hourly datasets of DSSR, UVR, and PAR at a 0.1° spatial resolution across China from 2005 to 2023. These datasets were produced by integrating multiple data sources, including ground observations from the Chinese Ecosystem Research Network (CERN), reanalysis products, and satellite remote sensing data.

The findings were recently published in Atmospheric Research.

The team validated the datasets against independent ground measurements and compared them with widely used global products such as CERES, CAMS, and ERA5. Results show that the datasets achieve high accuracy, with coefficient of determination (R²) values ranging from 0.78 to 0.83 at the hourly scale, 0.87 to 0.89 at the daily scale, and 0.96 to 0.98 at the monthly scale, respectively.

The reconstructed datasets reveal distinct long-term trends: DSSR and UVR have increased significantly across most regions of China since 2013. This shift is largely attributed to reduced aerosol emissions driven by China's Air Pollution Prevention and Control Action Plan. In contrast, PAR exhibits an overall downward trend accompanied by strong regional disparities. Further analysis indicates that year-to-year fluctuations in cloud cover and aerosol concentrations account for 80% to 93% of the variability in DSSR and UVR over eastern and central China.

The team noted that these open-access datasets provide valuable resources for researchers and policymakers engaged in solar energy development, climate change research, and ecosystem management.

This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences.