Detailed analysis of MAX-DOAS measurements in Bremen : spatial and temporal distribution of aerosols, formaldehyde and nitrogen dioxide

Abstract

In this thesis, spatial and temporal tropospheric inhomogeneities in the distribution of nitrogen dioxide (NO2), formaldehyde (HCHO) and aerosols are investigated. The analysis was done on a three years dataset (2015 - 2017) of ground-based multi-axis differential optical absorption spectroscopy (MAX-DOAS) measurements in Bremen. MAX-DOAS measurements were investigated in three different azimuthal viewing directions in order to analyse lateral changes in the distribution of NO2 and HCHO in Bremen. A clear seasonality was found and explained by anthropogenic and biogenic emissions for NO2 and HCHO, respectively. While no significant azimuthal variability for HCHO was found, NO2 differs strongly for the westerly and southerly directions due to lateral inhomogeneities and a more frequent pointing towards the sun which has a strong impact on the results. In order to localize possible dominant emission sources of NO2 within the area of Bremen, the onion peeling approach was successfully applied by usage of the three fitting windows in the ultra-violet (UV) and visible (vis) spectral range. Strong emitters could be identified having a large impact on average NO2 results. A major challenge for the analysis of trace gases in the troposphere is the usually insufficient knowledge of aerosols, which might have a large impact on spectroscopic measurements. The novel MAX-DOAS profiling algorithm BOREAS was developed and its accuracy is validated with the help of synthetic data as well as ancillary measurements of the CINDI-2 field campaign (Cabauw, the Netherlands, 2016). In contrast to other algorithms, BOREASa aerosol information are retrieved by minimizing the difference of O4 optical depths of measurements and forward modelling calculations. The resulting aerosol extinction coefficient profiles were used for the retrieval of vertical trace gas concentration profiles. In this thesis, several retrieval modes and various ways of improving the regularization between measurement and a priori constraints as well as the selection of proper a priori profiles by use of a priori pre-scaling were investigated. The BOREAS algorithm was finally applied to the full MAX-DOAS dataset, and three years of aerosol and trace gas vertical profiles from the measurement location Bremen are presented and discussed with the help of in-situ as well as AERONET measurements. Seasonal, weekday and diurnal cycles for aerosols and NO2 were found which could be attributed to near surface emissions mainly from traffic and power plants. The seasonal cycle of HCHO is found to be dominated by biogenic emissions in summer, in addition to a smaller fraction of anthropogenic emissions in winter. While NO2 and aerosols are mainly focussed in layers close to the surface, larger HCHO concentrations could also be observed in the complete planetary boundary layer showing the need for the analysis of vertical concentration profiles of trace gases in the troposphere.