A Machine Learning-Based Multiple Cloud Vertical Structure Parameter Prediction Algorithm Only Using OCO-2 Oxygen A-Band Measurements

Measurements of the global cloud vertical structure (CVS) are critical to better understanding effects CVS on climate. Current algorithms based OCO-2 have be combined with top height products from CALIPSO and CloudSat, which no longer available after these two satellites left A-Train in 2018. In this paper, we derive a machine learning-based algorithm using only oxygen A-band hyperspectral measurements simultaneously predict optical depth (COD), pressure (p_top), thickness (CPT) single-layer liquid clouds. For validation real observations, root mean square errors (RMSEs) COD, p_top, CPT 7.31 (versus MYD06_L2), 35.06 hPa, 26.66 hPa 2B-CLDCLASS-LIDAR). The new can also parameters trained p_tops CALIPSO/CloudSat or CODs MODIS. Controlled experiments show that known more conducive prediction than CODs, both obtain best accuracy RMSE = 20.82 hPa. Moreover, comparison OCO2CLD-LIDAR-AUX rely shows our predictions for all clouds, clouds p_top < 900 CPTs > 30