General approach to the formulation and solution of the multi-parameter inverse problems of atmospheric remote sensing with measurements and constraints of different types

Abstract

A general approach to the formulation and solution of the multi-parameter inverse problems of atmospheric remote sensing with measurements and constraints of different types is considered, which is based on the concept of the presentation of constraints as virtual measurements of finite accuracy. The advantages of the approach are (1) the possibility to account for all kinds of available information (remote and in situ measurements, physical constraints, model predictions); (2) the unified description of different measurements, a priori information and constraints in the retrieval algorithm; (3) the possibility to use measurements and a priori information of different types in any combination and to assess individual contributions to information content. The approach can be considered as a convenient tool for implementation of different synergistic remote-sensing schemes and for utilization of maximum available information on sought parameters for the purpose of increasing the accuracy of atmospheric remote sensing and providing self-consistency of sets of retrieved parameters. The known formulae for degrees of freedom for signal, averaging kernels, error components, and parameters describing spatial resolution are generalized to the case of multiple sought variables, measurements, and constraints, and the peculiar features of such generalization are discussed. The example is given of the application of the approach to the interpretation of the middle atmosphere limb infrared radiance data obtained in CRISTA (Cryogenic Infrared Spectrometers and Telescopes for the Atmosphere) experiments with special emphasis on the constraints describing coupling of the parameters due to physical processes.