Development of GOCI-II Aerosol Retrieval Algorithm: Information Content Analysis

Geostationary Ocean Color Imager (GOCI, or GOCI-I), the world’s unique ocean color imager in geostationary orbit, has been in operation over East Asia since 2011. The second generation of GOCI (GOCI-II) will be launched in 2019 onboard GeoKompast-2B (GK2B) satellite. GOCI-II has 13 spectral bands including wideband as compared to 8 bands of GOCI-I. A new feature of GOCI-II spectral bands compared to GOCI is the inclusion of UV band centered at 380 nm, which is advantageous to retrieve radiative absorptivity of aerosol over darker land surface reflectance than visible band. The purpose of this study is to evaluate and compare the information content of GOCI-II in retrieving aerosol properties compared to GOCI-I. We simulate the GOCI-II observation system using a radiative transfer model, the linearized pseudo-spherical vector discrete ordinate radiative transfer (VLIDORT), and calculate the degree of freedom for signal (DFS) and a posteriori error using the optimal estimation approach. We investigate the information content for surface reflectivity as well as aerosol optical properties to evaluate the feasibility of simultaneous retrieval of them using GOCI-II measurement spectra. The information content analysis results show higher DFS in GOCI-II measurements than in GOCI-I. Particularly, it is found that an impact of UV band utilization with previous GOCI-I spectral bands shows higher increase of the DFS than the utilization of additional visible channels. These results indicate that GOCI-II measurements can increase information content for retrieval of aerosol optical properties.