Superchannels for observing meteorological parameters

Abstract

The AIRS instrument has a large number (2378) of potential channels. For use for observing meteorological parameters, several methods have been proposed and/or used. These include selecting a subset of using eigenvectors, and using channels, which are averages of channels which view similar atmospheric features. The channels are selected using a constraint on the wavelength range to be covered, then selecting all the channels that have similar transmittances to be combined in one Super channels have a number of features that make them attractive. They use all the information to reduce the noise and are efficient to use since both rapid transmittance models and equivalent Planck functions can be generated. This means that it requires the same effort to calculate the radiance for one super channel as for a single AIRS channel. Super channels and Planck function have been calculated for AIRS instrument and a rapid transmittance model is being used to generate coefficients to allow a rapid calculation of the corresponding transmittances. Results using the super channels compared to measures of the truth from forecast models and radiosondes will be shown.