Machine Learning Predicted CCN Concentration for ORACLES ACI Study

These files contain machine learning-predicted CCN concentration at the same resolution as High Spectral Resolution Lidar 2 data for observations from the NASA ORACLES campaign. CCN is predicted at a supersaturation of 0.4%, and the variable is called CCN within the HSRL structure in each of these files. All other variables are those from the corresponding date's HSRL-2 file, available here. The machine learning model used to generate this specific dataset used backscatter at 532 and 1064 nm and depolarization ratio at 532 and 1064 nm as input to predict CCN concentration. The method and training code are further described in Redemann and Gao (2024).