Cutting-Edge Hybrid Algorithm Improves Aerosol Monitoring on China's FY-4A Satellite----Institute of Atmospheric Physics

Cutting-Edge Hybrid Algorithm Improves Aerosol Monitoring on China's FY-4A Satellite

Date：2024-01-26

【 Big Mid Small 】

Accurate measurements of atmospheric aerosols are pivotal in understanding Earth's radiation balance, climate change, and air quality. Aboard China's state-of-the-art geostationary meteorological satellite FY-4A, the Advanced Geostationary Radiation Imager (AGRI) scans China every 5 minutes, providing crucial data for monitoring aerosol spatiotemporal variations. However, the inflexibility of traditional physical retrieval algorithms, coupled with the insufficient number of ground-based sunphotometer sites, poses challenges in meeting the extensive sample requirements for machine learning in aerosol optical depth (AOD) retrieval.

AGRI aboard satellite FY-4A. (Image by National Satellite Meteorological Center, China Meteorological Administration)

In response to these challenges, a recent study published in the journal Engineering introduces an innovative AOD retrieval algorithm that combines deep learning and transfer learning. The algorithm incorporates key concepts from the dark target and deep blue algorithms to facilitate feature selection for machine learning. The process involves two main steps: Firstly, using 10-minute AOD obtained by the advanced imager on board the Japanese Himawari satellite as the target to develop a deep neural network (DNN) with residual networks, and secondly, fine-tuning the DNN parameters using sun photometer AOD data from 89 ground stations.

Independent validation confirms that the algorithm is highly accurate in estimating AGRI aerosol levels. The results show a strong correlation with expected values, indicating the algorithm's reliability in predicting aerosol optical depth. The majority of the data aligns closely with the anticipated range, emphasizing the algorithm's precision in real-world applications.

This multidisciplinary study brought together expertise from various fields, and it was conducted through a collaborative effort by the Institute of Atmospheric Physics (IAP) at the Chinese Academy of Sciences, National Satellite Meteorological Center of China Meteorological Administration, Harbin Institute of Technology, the National Institute of Disaster Prevention, Ministry of Emergency Management, and Solar Consulting Services in the United States.

"Our study showcases the significant potential of merging the physical approach with deep learning in geoscientific analysis." said the lead author Dr. FU Disong from IAP. "The proposed algorithm holds promise for application to other multi-spectral sensors aboard geostationary satellites."

References:

D. Fu, H. Shi, C.A. Gueymard, D. Yang, Y. Zheng, H. Che, X. Fan, X. Han, L. Gao, J.　Bian, M. Duan, X. Xia, A Deep-Learning and Transfer-Learning Hybrid Aerosol Retrieval Algorithm for FY4-AGRI: Development and Verification over Asia, Engineering (2024), https://doi.org/10.1016/j.eng.2023.09.023.

D. Fu, H. Shi, C. A. Gueymard, et al. FY-4A REGC AGRI AOD[DS/OL]. V2. Science Data Bank, 2024[2024-01-11]. https://doi.org/10.57760/sciencedb.12395.

Media contact:

Ms. LIN Zheng

Email: jennylin@mail.iap.ac.cn

Tel: 86-10-82995053

http://english.iap.cas.cn/