USING PSEUDO-MONOCHROMATIC RADIANCE FOR THE RETRIEVAL OF ATMOSPHERIC TEMPERATURE AND MOISTURE PROFILES

This study provides a new approach for processing observed hyper-spectral radiance data. It uses a transformation matrix to convert an instrument radiance spectrum into a pseudo-monochromatic radiance spectrum. One common approximation in the atmospheric profile remote sensing field is to approximate instrument channel radiances through spectral convolution of the monochromatic radiative transfer equation (RTE). When determining the atmospheric parameters from the observed radiance spectrum, a physical retrieval algorithm is used. Physical retrieval algorithms are based on the inversion of the RTE. A fast forward radiative transfer model (RTM) is used to calculate instrument channel radiance when applying a physical retrieval algorithm to atmospheric profiles. Many difficulties are involved when developing a fast RTM for instrument channel radiance. One example is that Beer's law is only valid for monochromatic radiation, not for spectrally integrated channel radiance. The new approach described in this paper uses pseudo-monochromatic radiance spectrum in the RTE for physical retrieval. The pseudo- monochromatic radiance spectrum is produced by an empirical transform of the instrument channel spectrum to a monochromatic equivalent spectrum. Eigenvetor regression is used to produce the empirical transformation. Although the transformation does not produce the monochromatic without error, it is shown that this transformation error in the radiance spectrum is generally well below the instrument noise level for most of the spectral channels as a result of the spectral radiance measurement noise filtering effect of the eigenvector transformation. One major advantage of this approach is that it eliminates the need to build different fast radiative transfer models for different observing instruments, since the retrieval of geophysical parameters is based on the inversion of the monochromatic radiative transfer model. A different transformation matrix is required for different instrument spectral channel characteristics. Simulation studies show that the accuracy of the atmospheric temperature and moisture profiles retrieved from monochromatic radiance spectra are improved by approximately 10% relative to the accuracy of profiles retrieved from typical hyperspectral instrument channel radiance spectra when using an eigenvector regression algorithm.