5:15 PM - 6:30 PM
[ACG36-P01] Evaluation of GCOM-C/SGLI L2 aerosol products: aerosol over the ocean, land aerosol by near-ultraviolet
Keywords:GCOM-C, SGLI, aerosol, remote sensing
Atmospheric aerosol significantly impacts the Earth in multiple ways, such as its radiative effect, climate processes, chemical reactions, and air pollution. While Aerosol Optical Depth (AOD) shows how much aerosols prevent light transmission in the atmosphere, the retrievals from Second-generation Global Imager (SGLI) observations are available among other Atmospheric data products. SGLI is a payload sensor onboard the Global Change Observation Mission Climate (GCOM-C) satellite for monitoring global climate change.
JAXA disseminates three types of SGLI Level 2 (L2) aerosol products: aerosol over the ocean, land aerosol by near-ultraviolet, and aerosol by polarization. We evaluated the first two products, retrieved with the second version algorithm released in June 2020, over different regions in this research. Conventional statistical parameters, such as the linear correlation coefficient, root mean square error, and mean bias error, are calculated. The error of estimation, given by the Moderate Resolution Imaging Spectroradiometer (MODIS) science team through their validations, is utilized, too. The biases with respect to ground truth measurements are analyzed as functions of the truth and retrievals. These diagnostic and prognostic error dependencies can be useful for the algorithm developers and users. We also perform comparisons with MODIS AODs, which have been produced over two decades. The results can be utilized to advance aerosol retrieval algorithms and to support the interpretation of the SGLI L2 AODs.
JAXA disseminates three types of SGLI Level 2 (L2) aerosol products: aerosol over the ocean, land aerosol by near-ultraviolet, and aerosol by polarization. We evaluated the first two products, retrieved with the second version algorithm released in June 2020, over different regions in this research. Conventional statistical parameters, such as the linear correlation coefficient, root mean square error, and mean bias error, are calculated. The error of estimation, given by the Moderate Resolution Imaging Spectroradiometer (MODIS) science team through their validations, is utilized, too. The biases with respect to ground truth measurements are analyzed as functions of the truth and retrievals. These diagnostic and prognostic error dependencies can be useful for the algorithm developers and users. We also perform comparisons with MODIS AODs, which have been produced over two decades. The results can be utilized to advance aerosol retrieval algorithms and to support the interpretation of the SGLI L2 AODs.