A comparison of satellite and sounding derived cloud top temperatures

A comparison of cloud top temperatures derived using automated icing algorithms at NCAR, and balloon-borne sounding observations developed by Wang and Rossow is presented. Results indicate that the Wang and Rossow technique for identifying layers and cloud top temperatures works well for relatively thick, stratiform clouds. It is also found that the technique often fails when broken or thin clouds are present. Distribution of error indicate that the most common errors are <25° C, though large errors are prone to occur where the key factor responsible is cloud thickness.