Improvement and interpretation of the tropospheric ozone columns retrieved based on SCIAMACHY Limb-Nadir Matching approach

Tropospheric ozone, one of the most important green-house gases and one of the most essential components of photochemical smog, has been monitored from space by different retrieval techniques since the late 1980s. Satellite measurements are well suitable to investigate sources and transport mechanisms of tropospheric ozone, as well as its atmospheric chemistry on regional and global scales. Nevertheless, the retrieval of tropospheric ozone columns (TOCs) from satellite data constitutes a big challenge since approximately 90% of the total ozone amount is located in the stratosphere, and only the remaining 10% is located in the troposphere. The Limb-Nadir Matching technique is one of the methods that has been widely used to re-trieve TOCs from space borne measurements. In previous studies, this approach has been applied to measurements from the SCIAMACHY instrument, which alternates limb and nadir geometry. An accurate tropopause height, retrieved from the ECMWF database, was used to subtract the stratospheric ozone column from the total ozone column. In this thesis, a three-step approach is shown that was developed to improve the current Limb-Nadir Matching TOC retrieval technique, and resulted in the new database version 1.2. Several improvements in the V1.2 TOC data have been achieved. The obtained amount of TOC V1.2 data has increased by a factor of two in comparison to the original dataset. Fur-thermore, the data quality has improved in many aspects. First of all, the V1.2 TOC data set reduces the negative (>10 DU) and positive (~10 DU) biases over tropics and high latitudes, respectively. The reduction is achieved by use of the improved limb ozone data set V3.0, which was tested and validated against the previous version V2.9 in this thesis. The TOC values were also optimized over the midlatitudes by decreasing its positive biases. The yearly averaged V1.2 TOC data set agrees well with ozonesonde measurements within 5 DU globally. More details on the TOC distribution were successfully captured because of the improved accuracy of the data. The clear observation of the spring TOC maxima (~42 DU) over the Arabian Sea (AS) during the pre-monsoon is one of the benefits of using the V1.2 TOC product. In the present thesis, the potential sources of the AS spring ozone pool are investigated by use of multiple data sets (e.g., SCIAMACHY Limb-Nadir-Matching TOC, OMI/MLS TOC, TES TOC, MACC reanalysis data, MOZART-4 model and HYSPLIT model). 3/4 of the enhanced ozone concentrations are attributed to the 0-8 km height range. The main source of the ozone enhancement is considered to be caused by long range transport of pollutants from India (~ 50% contributions to the lowest 4 km, ~ 20% contributions to the 4-8 km height range), the Middle East, Africa and Europe (~30% in total). In addition, the vertical pollution accumulation in the lower troposphere, especially at 4-8 km, was found to be important for the AS spring ozone pool. Local photochemistry, on the other hand, plays a negligible role in producing ozone at the 4-8 km height range. In the 0-4 km height range, ozone is quickly removed by wet-deposition. The AS spring TOC maxima are influenced by the dynamical variations caused by the sea surface temperature (SST) anomaly during the El Nino period in 2005 and 2010 with a ~5 DU decrease. The Limb-Nadir Matching retrieval improvement scheme developed in this thesis leads to a much more accurate TOC product measured by SCIAMACHY and a better understanding of tropospheric ozone distributions