Spectral band and detector optimization for atmospheric sounding interferometers

In the specification and selection of spectral bands and detectors for optical remote sensing instruments, important consideration must be given to the primary scientific applications of the data. In this study, the line-by-line atmospheric code FASCODE and a Lincoln-developed detector model are used to identify the spectral band extent and peak location for Hg/sub 1-x/Cd/sub x/Te detector responsivity that result in the highest sounding sensitivity to atmospheric temperature and water vapor. The goal of this work was to minimize the background photon level that contributes to noise by reducing the spectral band extent and to relocate the peak detector response to the spectral region of greatest interest. An optimization score for each wavelength was computed as the product of terms accounting for sensitivity to the parameter of interest (temperature or water vapor) and insensitivity to the other, normalized by the product of the detector noise and the vertical height of the atmospheric contribution function. By examining the optimization score as a function of wavelength, reduced spectral coverage was seen to be possible without sacrificing sounding sensitivity. These reductions, along with the movement of the detector peak to above the spectral band low wavenumber edge, allowed an improvement of 15 to 33% in noise level and 10 to 15% in sounding performance as compared to the original specification.<
>