11:00 AM - 1:00 PM
[ACG38-P04] Comparison with other satellite and ground-based observation with GOSAT-2/TANSO-CAI-2 aerosol properties retrieved by MWPM and the error and bias investigations
Keywords:Aerosol, Satellite remote sensing
The Greenhouse Gases Observing Satellite-2 (GOSAT-2) launched in 2019 has Cloud and Aerosol Imager 2 (TANSO-CAI-2). CAI-2 has ten bands composed of seven wavelengths at 340, 380, 443, 550, 674, 869 and 1630 nm with spatial resolution of 460 m (and 920 m at 1630nm), and performs two-directional observation in forward and backward directions. CAI-2 is characterized by having two ultraviolet (UV) bands that dust and black carbon (BC) particles are sensitive for light absorption.
We have developed aerosol retrieval algorithm called MWPM (Multiple wavelengths and pixels method) (Hashimoto and Nakajima, 2017). The method simultaneously retrieves fine and coarse mode AOT and single scattering albedo (SSA) by using several wavelengths and pixels where surface reflectance is difference spatially. The method is useful for aerosol retrieval over spatially inhomogeneous surface region like an urban area where is a source region of anthropogenic aerosols.
We have applied the algorithm to CAI-2 data and retrieve aerosol properties, such as fine and coarse mode AOT, SSA, BC volume fraction in fine mode particles and Ångström exponent (AE). Furthermore, equivalent value of PM2.5 (ePM2.5) is calculated by derived aerosol properties. It is important to derive accurate aerosol optical properties to calculate and evaluate ePM2.5.
We have compared retrieved aerosol optical properties to ground-based observation, SKYNET and AERONET. In the case that Ag is less than 0.1, root mean square deviation (RMSD) of AOT at 550nm is around 0.1, and correlation coefficient is over 0.7. SSAs at 340 and 380nm are correlated between CAI-2 and SKYNET. CAI-2 AE is basically underestimated compared with AETRONET and SKYENT. We also have compared the results with other satellite data such as VIIRS, MODIS and Himawari-8/AHI aerosol products. CAI-2 AOT is a little overestimate, but in good agreement with those of the other satellites with correlation coefficient over 0.8. We will show the comparison results and discuss the results for modification of the algorithm in the future.
We have developed aerosol retrieval algorithm called MWPM (Multiple wavelengths and pixels method) (Hashimoto and Nakajima, 2017). The method simultaneously retrieves fine and coarse mode AOT and single scattering albedo (SSA) by using several wavelengths and pixels where surface reflectance is difference spatially. The method is useful for aerosol retrieval over spatially inhomogeneous surface region like an urban area where is a source region of anthropogenic aerosols.
We have applied the algorithm to CAI-2 data and retrieve aerosol properties, such as fine and coarse mode AOT, SSA, BC volume fraction in fine mode particles and Ångström exponent (AE). Furthermore, equivalent value of PM2.5 (ePM2.5) is calculated by derived aerosol properties. It is important to derive accurate aerosol optical properties to calculate and evaluate ePM2.5.
We have compared retrieved aerosol optical properties to ground-based observation, SKYNET and AERONET. In the case that Ag is less than 0.1, root mean square deviation (RMSD) of AOT at 550nm is around 0.1, and correlation coefficient is over 0.7. SSAs at 340 and 380nm are correlated between CAI-2 and SKYNET. CAI-2 AE is basically underestimated compared with AETRONET and SKYENT. We also have compared the results with other satellite data such as VIIRS, MODIS and Himawari-8/AHI aerosol products. CAI-2 AOT is a little overestimate, but in good agreement with those of the other satellites with correlation coefficient over 0.8. We will show the comparison results and discuss the results for modification of the algorithm in the future.