Articles | Volume 9, issue 9
Atmos. Meas. Tech., 9, 4687–4700, 2016
Atmos. Meas. Tech., 9, 4687–4700, 2016

Research article 21 Sep 2016

Research article | 21 Sep 2016

AerGOM, an improved algorithm for stratospheric aerosol extinction retrieval from GOMOS observations – Part 1: Algorithm description

Filip Vanhellemont1, Nina Mateshvili1, Laurent Blanot3, Charles Étienne Robert1, Christine Bingen1, Viktoria Sofieva2, Francis Dalaudier4, Cédric Tétard1, Didier Fussen1, Emmanuel Dekemper1, Erkki Kyrölä2, Marko Laine2, Johanna Tamminen2, and Claus Zehner5 Filip Vanhellemont et al.
  • 1Belgisch Instituut voor Ruimte-Aëronomie/Institut d'Aéronomie Spatiale de Belgique, Brussels, Belgium
  • 2Finnish Meteorological Institute, Helsinki, Finland
  • 3ACRI-ST, Sophia Antipolis, France
  • 4Laboratoire Atmosphères, Milieux, Observations Spatiales, Guyancourt, France
  • 5ESA-ESRIN, Frascati, Italy

Abstract. The GOMOS instrument on Envisat has successfully demonstrated that a UV–Vis–NIR spaceborne stellar occultation instrument is capable of delivering quality data on the gaseous and particulate composition of Earth's atmosphere. Still, some problems related to data inversion remained to be examined. In the past, it was found that the aerosol extinction profile retrievals in the upper troposphere and stratosphere are of good quality at a reference wavelength of 500 nm but suffer from anomalous, retrieval-related perturbations at other wavelengths. Identification of algorithmic problems and subsequent improvement was therefore necessary. This work has been carried out; the resulting AerGOM Level 2 retrieval algorithm together with the first data version AerGOMv1.0 forms the subject of this paper. The AerGOM algorithm differs from the standard GOMOS IPF processor in a number of important ways: more accurate physical laws have been implemented, all retrieval-related covariances are taken into account, and the aerosol extinction spectral model is strongly improved. Retrieval examples demonstrate that the previously observed profile perturbations have disappeared, and the obtained extinction spectra look in general more consistent. We present a detailed validation study in a companion paper; here, to give a first idea of the data quality, a worst-case comparison at 386 nm shows SAGE II–AerGOM correlation coefficients that are up to 1 order of magnitude larger than the ones obtained with the GOMOS IPFv6.01 data set.

Short summary
The GOMOS instrument on Envisat has delivered a valuable aerosol extinction data set for the Earth's upper troposphere and stratosphere, from 2002 to 2012. However, at many optical wavelengths, data quality was not optimal. This article describes the AerGOM retrieval algorithm that was built to solve the problem and presents a first look at the reprocessed GOMOS data, clearly demonstrating the improvement. Multi-wavelength studies of atmospheric aerosol–cloud properties will now be possible.