Translator Disclaimer
Paper
22 December 2004 Validation of AMSU-B water vapor profile retrievals with GOES infrared observations in the 6.5-micron band
Author Affiliations +
Abstract
One-dimensional variational (1DVAR) retrievals of humidity profiles over ocean are generated from AMSU-B observations and a Navy Operational Global Atmospheric Predication System (NOGAPS) background. Retrievals of water vapor profiles from AMSU-B are validated by intercomparing GOES water vapor channel observations with simulated GOES brightness temperatures calculated by the RTTOV-7 forward model from the background and retrieved atmospheric profiles. Brightness temperatures simulated from the retrieved profiles matched the GOES observations much more closely than those calculated from the background profile. After screening out clouds, GOES Imager 6.7-mm brightness temperatures from GOES-10 were reproduced with a correlation of 0.94, a bias of 2.97 K, and a standard deviation of -1.60 K. The impact of AMSU-B retrievals assimilated using NAVDAS is tested by comparing two data assimilation runs. 500-mb and 300-mb specific humidity fields are compared to GOES imagery. Large-scale moisture features such as the Intertropical Convergence Zone (ITCZ) and South Pacific Convergence Zone (SPCZ) at 500 mb are sharper and better defined in the AMSU-B run, agreeing qualitatively with GOES observations. However, differences in simulated GOES water vapor channel brightness temperatures from the two model runs are very small, due to the limited impact of amsu-b in the upper troposphere in this run.
© (2004) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Clay B. Blankenship "Validation of AMSU-B water vapor profile retrievals with GOES infrared observations in the 6.5-micron band", Proc. SPIE 5654, Microwave Remote Sensing of the Atmosphere and Environment IV, (22 December 2004); https://doi.org/10.1117/12.578956
PROCEEDINGS
8 PAGES


SHARE
Advertisement
Advertisement
Back to Top