Quantifying the effect of aerosol scattering on OCO-2 XCO2 retrieval over urban emission plume

Large source regions such as megacities are typically associated with heavy aerosol loading. These cities have observable plumes, containing high concentrations of carbon dioxide (CO₂) and aerosols downstream of prevailing wind. The Orbiting Carbon Observatory-2 (OCO-2) measures the column-averaged dry air mole fraction of CO₂ (XCO₂) from space with three near-infrared bands, the oxygen (O₂) A-band at 0.76 μm and CO₂ bands at 1.6 μm and 2.1 μm. With cross-plume soundings, retrieved OCO-2 XCO₂ data have been used to deduce carbon emissions from cities or power plants. However, the accuracy of XCO₂ retrievals over urban emission plumes, in the presence of high aerosol loading, is unclear. Previous studies have shown that imperfect characterization of atmospheric aerosols is the dominant error source in retrievals of greenhouse gases from near infrared space-based measurements. Also, the retrieved aerosol optical depth (AOD) from the OCO-2 full physics retrieval algorithm is very poorly correlated with the AERONET AOD measurements.

In this study, we compare the AOD and aerosol layer height (ALH) retrieved from OCO-2 against CALIPSO measurements. The differences between two retrievals are used to investigate the impact of improper characterization of aerosol vertical structure on the XCO₂ retrievals. We applied a spectral sorting approach (Zeng et al., 2018, 2019) to retrieve AOD and ALH using O₂ A-band measurements from OCO-2. Using the OCO-2 full physics retrieval algorithm, we evaluate potential improvements for XCO₂ retrievals using AOD and ALH derived with the spectral sorting method. We also estimated urban emissions from corrected XCO₂ retrievals over the associated plumes.