A Technique for Deriving Column-integrated Water Content Using VAS Split-Window Data

Abstract

An algorithm is examined that uses Visible?Infrared Spin Scan Radiometer (VISSR) Atmospheric Sounder (VAS) 11- and 12-µm (split-window) data to derive column-integrated water content (IWC) at mesoscale resolution. The algorithm is physically based and derives its first-guess information from radiosonde data. The procedure is applied first to a test case dataset and then to the 19 June 1986 study day from the Cooperative Huntsville Meteorological Experiment (COHMEX). Ground truth data for verifying results from the technique include IWC from National Weather Service and COHMEX radiosondes, the Multispectral Atmospheric Mapping Sensor (MAMS), and a special set of VAS soundings (12 channel) using an independent retrieval method. Results from the test case show reasonable accuracy with the root-mean-square errors as low as ±3.8 mm. On the 19 June case study day IWC analyses depict reasonable gradients and exhibit good spatial and temporal continuity. Furthermore, they provide insight into preferred regions for cumulus cloud and thunderstorm formation. On the average, a mean absolute retrieval error of 2.4 mm (an 8.1% error) and a root-mean-square error of ±2.9 mm are obtained on the case study day. These results compare favorably with those from existing VAS IWC techniques. Overall, the findings indicate that the technique has excellent potential to depict mesoscale moisture variations.