Abstract :
A method is presented for creating a set of optimal channels for microwave sounding of the atmosphere. The method finds the combinations of center frequencies and bandwidths that provide optimal skill at retrieving a profile of atmospheric characteristics, such as the air temperature or water vapor concentration. The measure of optimality is a function of the degrees of freedom for the signal. The method is capable of considering several environmental regimes (e.g. latitude, season, surface type, scan angle) simultaneously while performing the optimization. Applications of the method are presented for satellite sounding at frequencies near 60, 118, and 183 GHz. The optimized channel sets compare favorably to channel sets from planned operational sensors, in terms of optimality measures and retrieval performance. For tropospheric sounding, optimal channels have spectrally broader passbands than have typically been used in satellite sounding instruments. For water vapor sounding on the 183-GHz line, the number of channels needed to achieve the full retrieval potential is shown to depend on the radiometer noise level and on the vertical structure of the water vapor profiles.
