A new climatology of aerosols in the middle and upper stratosphere by alternative analysis of GOMOS observations during 2002-2006

International audienceRecent works have shown how the study of stratospheric background aerosol (i.e. in periods uninfluenced by major volcanic eruption) seems more complex as it is now performed by more accurate means. We propose a re-analysis of Global Ozone Monitoring by Occultation of Stars (GOMOS) level 1b data for the period August 2002-July 2006, using the LPC2E processor algorithm, which was developed for the retrieval of aerosol extinction in the middle and upper stratosphere. The main differences with regard to the 'official' algorithm are the correction of chromatic scintillation, the spectral domain (which has been restricted to the 400-700 nm region), the use of a Differential Optical Absorption Spectroscopy (DOAS) method for species retrieval, and the use of a fourth-order polynomial to reproduce the wavelength dependence of extinction. Since GOMOS observations are performed using stars of different magnitude and colour, discrepancy in signal-to-noise ratio between several profiles exists, and a data selection concerning standard deviation of aerosol extinction and other parameters becomes necessary. In the middle stratosphere, aerosol extinction profiles obtained with the LPC2E processor seem to be in better agreement with the SAGE III observations and sparse balloon-borne measurements than the 'official products'. We present global coverage of the 500 nm extinction values from around 15-60 km, and the wavelength dependence in the 400-675 nm spectral range which gives information about the nature of the particles. The well-known layer of liquid aerosols can be observed in the lower stratosphere, where the value of extinction is greater for blue than for red wavelengths, as is typical for small droplets. In the middle stratosphere, relatively high extinction values are found, probably due to the presence of solid particles above 30 km at all latitudes. The presence of soot and interplanetary material in the middle atmosphere is discussed, as well as seasonal patterns common to the several years of analysis, such as the stratospheric cleansing of aerosols above 30 km during polar winters