One-dimensional variational (1D-Var) retrieval of middle to upper tropospheric humidity using AIRS radiance data

Abstract

A one-dimensional variational analysis (1D-Var retrieval) of tropospheric humidity was conducted using hyper-spectral radiance data from the Atmospheric Infrared Sounder (AIRS). For the vertical range of the atmosphere between 200 and 600 hPa, the same high-resolution retrieval of humidity profiles as for clear-sky conditions was possible over low clouds if the cloud height was lower than 800 hPa. Global analyses from a global data assimilation system were used for initial profiles, and clear conditions over 800 hPa height were determined from AIRS radiance data. Results of analyses for 50 days of global radiosonde matchup data from 21 December 2008 to 8 February 2009 revealed that our 1D-Var calculations derived humidity profiles were closer to the sonde profiles than those of a global analysis at a height over 600 hPa. Furthermore, the results of 1D-Var retrieval often represented high and supersaturated relative humidity around the supposed ice clouds. The altitudes of the high humidity region agreed with the height of ice clouds that had been detected by CloudSat/CALIPSO. As well as possibly improving the humidity profiles in a numerical model by data assimilation, it is expected that these humidity analyses using AIRS radiance data will provide additional information for the study of ice clouds in the middle to upper troposphere.