Explorative study on GOME-2 total column ozone retrievals and the validation with ground-based and balloon measurements

Abstract

Abstract. In this work we present an extensive sensitivity study of retrieved total ozone columns from clear sky Global Ozone Monitoring Experiment 2 (GOME-2) measurements between 325 and 335nm which are corrected for instrument degradation. Employing an algorithm based on the scaling of a reference ozone profile with the extension to analytically calculate total column averaging kernels, allows us to investigate the impact of the choice of the reference profile on the retrieved total ozone column, since it represents a regularization of the retrieval. It introduces an error to the retrieved column with respect to the true column typically in the order of 1% depending on the reference scaling profile. However, a proper interpretation of the retrieved column using the total column averaging kernel avoids this error, which is demonstrated by a validation of GOME-2 total ozone columns with collocated ozonesonde and ground-based total ozone column measurements. Globally, we report a bias of 0.1% and a SD of 2.5% for 647 collocations with ground-based and ozonesonde measurements at different geolocations in the period of 2007 to 2010. Futhermore, an extended validation solely based on ground-based observations and a strict cloud filtering shows that the use of pseudo spherical scalar radiative transfer is fully sufficient for the purpose of this retrieval. Polarization of light by atmospheric scattering affects the retrieval accuracy only marginally and thus can be ignored. Finally, we study the effect of instrument degradation on the retrieved total ozone columns for the first four years of GOME-2 observations and discuss the efficiency of the proposed radiometric correction.