Improvement of CO2 retrieval algorithm with modified aerosol parameters using GOSAT measurements over East-Asia

Abstract:

The carbon dioxide (CO₂) retrieval algorithm based on optimal estimation method was developed and its performance was analyzed. In sensitivity analysis using simulated radiance spectra at difference surface and atmospheric conditions, the aerosol-related parameters such as total Aerosol Optical Depth (AOD) and aerosol optical properties are the most important factor in CO₂ retrieval, resulting in errors up to 6.5 ppm by inaccurate aerosol optical information. These errors are caused by the simplified aerosol assumptions in the forward model, which only represent a subset of three potential aerosol optical properties, and can be also increased in CO₂ retrieval using real-spectra. As aerosols in the atmosphere are highly variable in their amount, vertical distribution and optical properties, their effect can be under-constrained (Frankenberg et al., 2012).

In this study, to reduce the errors caused by the simplified aerosol assumptions, modification of aerosol-related parameters in the forward model have been applied in CO₂ retrieval algorithms, presenting by 12 parameters consisting of the vertical profile in terms of central height and width, aerosol size distribution and refractive index parameters as their optical parameters. The CO₂ retrievals with two different approaches in handling aerosol information have been analyzed using the Greenhouse Gases Observing SATellite (GOSAT) spectra over East-Asia and evaluated through the comparison with collocated ground-based observations at several Total Carbon Column Observing Network (TCCON) sites. These results can improve the accuracy of CO₂ retrieval algorithm taking into account aerosol information and provide useful information to reduce uncertainty and increase data availability.