GOME Ozonprofile: Weiterentwicklung und Validierung

Abstract

Height-resolved ozone information on a global scale is required forthe detection of changes in the atmospheric ozone distribution and for researchinto the underlying chemical and dynamical processes. The Global OzoneMonitoring Experiment (GOME) makes an important contribution in thisfield. GOME aboard ESA´s ERS-2 satellite measures the reflected andbackscattered radiation from the Earth in the ultraviolet and visiblespectral range at moderate spectral resolution in nadir viewinggeometry. Vertical ozone profiles can be derived fromtop-of-atmosphere nadir observations using the FUll Retrieval MethodFURM, which is based upon an advanced Optimal Estimation inversionscheme, using the Kozlov-information-matrix method. In the frameworkof this thesis this algorithm was adapted for routineretrieval of GOME ozone profiles. The calibration of the GOMEinstrument was discussed with respect to its impact on ozone profileretrieval, presenting an effective calibration error correctionscheme. The choice of a-priori ozone profiles was investigated, showingthat not only the mean profiles but also there statisticalproperties reflecting the variability of ozone are important. Anup-to-date, statistically correct ozone climatology is needed.A-priori temperature and pressure profiles have to be taken fromassimilated meteorological fields. The ozone profiles have beenvalidated on a global scale by comparison with measurements of theHALogen Occultation Experiment (HALOE). A good agreement within 10%between 15 and 35 km altitude for most seasons and regions isachieved. Total ozone calculated from integrated profiles have beenvalidated with Dobson measurements in the Northern mid-latitudes, theyagree to within -2% with seasonal variation. With the results ofthis work, the GOME ozone profile retrieval algorithm FURM can be (andalready is) used for routine observation and scientific investigationof the ozone layer.