A Comparison of Cloud and Aerosol Measurements from OCO-2 and CALIPSO

Synergistic studies of various A-train sensors with the recently launched OCO-2 observatory are now underway. OCO-2 is meant to measure carbon dioxide from hyperspectral measurements of near-infrared reflected sunlight, but can only obtain high accuracy in clear scenes. In this study, we compare the results of the OCO-2 O₂-A band cloud screening algorithm to CALIPSO observations, to determine their level of agreement in terms of both cloud and aerosol properties. Both CALIPSO optical depth as well as vertical information were analyzed in the context of the OCO-2 O₂-A cloud screening algorithm. In general, we find that for high optical depth CALIPSO measurements, OCO-2 identified most (~70%) of scenes as cloudy. The opposite was also true: for low optical depth CALIPSO measurements, OCO-2 identified most of the scenes as clear. Further, it was found that OCO-2 misidentifies ~40% of scenes containing thick, low clouds, while properly identifying nearly all scenes containing high clouds. These preliminary results confirm expectations derived from previous simulation-based work. Future studies include the dependence of these results on cloud phase, optical thickness detection limits, and potential solar zenith angle dependencies.