Analyse und Interpretation von Satellitenmessungen im nahen infraroten Spektralbereich mit dem Schwerpunkt Kohlenmonoxid

Carbon monoxide (CO) plays an important role in the Earth's atmosphere. Through its reaction with the hydroxyl radicals (OH) (Logan et al., 1981), CO affects the lifetime of atmospheric methane (CH4), and non-methane hydrocarbons (NMHCs). A main product of this oxidation is carbon dioxide (CO2). Therefore, containing no direct green-house potential, CO still has an indirect effect on the global warming. CO is also one of the most important health hazardous pollutants, which can cause diseases of different degrees of complexity. The nadir near-infrared measurements of scattered and reflected solar radiation by SCanning Imaging Absorption spectroMeter for Atmospheric CHartographY (SCIAMACHY) instrument on board the ENVISAT satellite contain information about CO concentration in all atmospheric layers including the boundary layer, closest to the location of main CO sources. However, the retrieval of CO total column from the radiometric measurements in this spectral region is complicated as the CO overtone lines are weak, and overlapped by strong absorptions of water vapour and methane. Moreover, several known instrumental issues, like an ice layer on the detector and degradation of the detector pixels with time, additionally complicate the retrieval of CO vertical column from the of SCIAMACHY measurements in channel 8. In the scope of this work, the WFM-DOAS (Weighting Functions Modified Differential Optical Absorption Spectroscopy) retrieval algorithm, developed at the University of Bremen, have been improved in order to establish the retrieval of a multi-year CO dataset from SCIAMACHY nadir measurements. The modifications have led to an improved CO fit quality, i.e., to an overall much smaller fit residual. An error analysis and sensitivity studies based on the simulated measurements have shown that the error is generally less than 10%, which is comparable to the required precision for space-based CO measurements. However, due to high instrument noise, the error of the real measurements has been found to be much higher and considerably less stable. The retrieved CO columns have been validated by comparison with ground-based Fourier Transform Spectroscopy (FTS) measurements. A good agreement within 10-20% was found for nearly all considered stations. Furthermore, high correlation between the SCIAMACHY CO and CO from independent space-based total columns measurements performed by the MOPITT (Measurements of Pollution in the Troposphere) instrument onboard the Terra satellite indicates a good performance of the SCIAMACHY CO measurements globally. The overall difference of about 10% can be well explained by the moderate sensitivity of the thermal-infrared MOPITT measurements to lower atmospheric layers.Detailed analysis of the obtained CO dataset has been has been carried out on country level. Due to the presence of strong anthropogenic sources and prevailing west wind conditions, a positive difference of CO concentration is expected from the west to the east side of the United Kingdom. The analysis shows that SCIAMACHY is able to capture the positive 5% west-to-east CO gradient over the UK. These results are consistent with the direct airborne measurements during the AMPEP campaign, which estimated the CO concentration enhancement from the west to the east coast of the UK to be about 10-100 ppb, corresponding to the total column enhancement of 1-10% within the 1 km boundary layer. Over much stronger sources, such as a large biomass burning events, the quantitative potential of SCIAMACHY CO data is expected to be much higher due to much higher levels of CO signal and respectively more available ("good") satellite measurements. To use this fact for further quantitative investigation, the SCIAMACHY simultaneously measurements of CO, nitrogen dioxide (NO2) and formaldehyde (HCHO) over biomass burning events in 2004, were analysed in the scope of the bottom-up emission estimation Excess Mixing Ratios (EMR) method. Good agreement has been found between the calculated SCIAMACHY (Delta CO)/(Delta HCHO) and (Delta CO)/(Delta NO2) and the ER values from referenced literature