ABSTRACT The hyperspectral SCanning Imaging Absorption spectroMeter for Atmospheric CHartographY (SCIAMACHY) instrument, launched by Envisat in March 2002, measures atmospheric scattered radiance within the ultraviolet (UV) to near infrared range and has three geometric observation modes: nadir, occultation, and limb. This study describes a retrieval scheme to obtain middle and upper stratospheric ozone profiles from SCIAMACHY limb-scattered solar radiance. Owing to the low ozone concentration at high altitudes and rapid changes in ozone absorption cross-sections in the Hartley – Huggins band, a fundamental objective of this study was to establish a simple and effective retrieval scheme to derive reliable ozone profiles at high altitudes from UV spectral bands. In this study, short UV wavelengths of 302, 312, and 322 nm and a long wavelength of 331 nm were used for wavelength pairing to simplify the retrieval algorithm. Moreover, we employed the weighted multiplicative algebraic reconstruction technique (WMART), which is straightforward to understand, and easy to implement, combined with a radiative transfer model (i.e. SCIATRAN), and produced stable results. The retrieval error estimates for the upper stratosphere were provided, and the total retrieval error of the ozone profile was usually < 18% in the altitude range of 30–53 km. The results were compared with the SCIAMACHY v3.5 product and validated against profiles derived from Optical Spectrometer and InfraRed Imaging System (OSIRIS) v7.1, and Atmospheric Chemistry Experiment Fourier Transform Spectrometer (ACE-FTS) v4.0 for 30–53 km altitude. The retrieved ozone profiles agreed with data from SCIAMACHY v3.5 within − 10%–5%, except at an altitude of 45 km, where the bias became larger (approximately −15%). The comparison of retrieved profiles with OSIRIS v7.1 showed differences within ± 10% for 30–53 km altitude. The errors were mostly less than − 5% below 43 km compared with ACE-FTS v4.0, but there was a large negative deviation in the upper stratosphere.
Read full abstract