Satellite Measurement of Stratospheric Winds and Ozone Using Doppler Michelson Interferometry. Part II: Retrieval Method and Expected Performance

Abstract

This paper is about the retrieval of horizontal wind and ozone number density from measurement simulations for the Stratospheric Wind Interferometer for Transport Studies (SWIFT). This instrument relies on the concept of imaging Doppler Michelson interferometry applied to thermal infrared emission originating from the stratosphere. The instrument and measurement simulations are described in detail in the first of this series of two papers. In this second paper, a summary of the measurement simulations and a data retrieval method suited to these measurements are first presented. The inversion method consists of the maximum a posteriori solution approach with added differential regularization and, when required, iterations performed with the Gauss?Newton method. Inversion characterization and an error analysis have been performed. Retrieval noise estimates have been obtained both from derived covariance matrices and sample inversions. Retrieval noise levels for wind and ozone number density of ?1?3 m s?1 and <1% have been obtained over the altitude range of 20?45 km with Backus?Gilbert resolving lengths of ?1.5 km. Retrieval noise levels over the extended altitude range of 15?55 km are less than 10 m s?1 and 2%. The sensitivity to other error sources has been examined through a few sample realizations. The contributions from these other errors can be as important as or more so than retrieval noise. An error budget identifying contributing wind and ozone error levels to total errors of 5 m s?1 and 5% for altitudes of 20?45 km has been prepared relying on the retrieval errors and knowledge of the instrument design.