[A-CG41] 衛星による地球環境観測

[ACG41-08] Remote sensing of aerosols from satellites during the AI Era

*Chong Shi¹、Husi Letu¹、Chenqian Tang¹、Wenwu Wang¹、Ruijie Yao¹、Teruyuki Nakajima²、Miho Sekiguchi³ (1.Aerospace Information Research Institute, Chinese Academy of Sciences、2.The University of Tokyo, Japan、3.Tokyo University of Marine Science and Technology, Japan)

キーワード：Neural network, Radiative transfer, Aerosol, Satellite remote sensing

In recent years, the artificial intelligence (AI) techniques, such as the neural network approach, have been used in atmospheric remote sensing, due to their robust ability to model complex nonlinear relationships between inputs and outputs. In this talk, we would like to introduce some new results for the retrieval of aerosols by combining the physical-based radiative transfer model and data-driven NN techniques. With this combination, higher retrieval accuracy for the aerosol optical thickness and fine mode fraction, higher calculation efficiency to achieve real-time processing, can be expected, in comparison to the traditional retrieval algorithms. In addition, some preliminary results for the retrieval of aerosol layer height will be introduced from the satellite hyperspectral measurements.