Retrieval of tropospheric aerosol, NO2, and HCHO vertical profiles from MAX-DOAS observations over Thessaloniki, Greece: intercomparison and validation of two inversion algorithms

Atmospheric Measurement Techniques, Mar 2022

In this study we focus on the retrieval of aerosol and trace gas vertical profiles from multi-axis differential optical absorption spectroscopy (MAX-DOAS) observations for the first time over Thessaloniki, Greece. We use two independent inversion algorithms for the profile retrievals: the Mexican MAX-DOAS Fit (MMF) and the Mainz Profile Algorithm (MAPA). The former is based on the optimal estimation method (OEM), while the latter follows a parameterization approach. We evaluate the performance of MMF and MAPA, and we validate their retrieved products with ancillary data measured by other co-located reference instruments. The trace gas differential slant column densities (dSCDs), simulated by the forward models, are in good agreement, except for HCHO, where larger scatter is observed due to the increased spectral noise of the measurements in the UV. We find an excellent agreement between the tropospheric column densities of NO2 retrieved by MMF and MAPA (slope=1.009, Pearson's correlation coefficient R=0.982) and a good correlation for the case of HCHO (R=0.927). For aerosols, we find better agreement for the aerosol optical depths (AODs) in the visible (i.e., at 477 nm) compared to the UV (at 360 nm), and we show that the agreement strongly depends on the O4 scaling factor that is used in the analysis. The agreement for NO2 and HCHO near-surface concentrations is similar to the comparison of the integrated columns with slightly decreased correlation coefficients. The seasonal mean vertical profiles that are retrieved by MMF and MAPA are intercompared, and the seasonal variation in all species along with possible sources is discussed. The AODs retrieved by the MAX-DOAS are validated by comparing them with AOD values measured by a CIMEL sun photometer and a Brewer spectrophotometer. Four different flagging schemes were applied to the data in order to evaluate their performance. Qualitatively, a generally good agreement is observed for both wavelengths, but we find a systematic bias from the CIMEL sun photometer and Brewer spectrophotometer measurements, due to the limited sensitivity of the MAX-DOAS in retrieving information at higher altitudes, especially in the UV. An in-depth validation of the aerosol vertical profiles retrieved by the MAX-DOAS is not possible since only in very few cases is the true aerosol profile known during the period of study. However, we examine four cases, where the MAX-DOAS provided a generally good estimation of the shape of the profiles retrieved by a co-located multi-wavelength lidar system. The NO2 near-surface concentrations are validated against in situ observations, and the comparison of both MMF and MAPA revealed good agreement with correlation coefficients of R=0.78 and R=0.73, respectively. Finally, the effect of the O4 scaling factor is investigated by intercomparing the integrated columns retrieved by the two algorithms and also by comparing the AODs derived by MAPA for different values of the scaling factor with AODs measured by the CIMEL sun photometer and the Brewer spectrophotometer.

Article PDF cannot be displayed. You can download it here:

https://amt.copernicus.org/articles/15/1269/2022/amt-15-1269-2022.pdf

Retrieval of tropospheric aerosol, NO2, and HCHO vertical profiles from MAX-DOAS observations over Thessaloniki, Greece: intercomparison and validation of two inversion algorithms

Atmos. Meas. Tech., 15, 1269–1301, 2022 https://doi.org/10.5194/amt-15-1269-2022 © Author(s) 2022. This work is distributed under the Creative Commons Attribution 4.0 License. Retrieval of tropospheric aerosol, NO2, and HCHO vertical profiles from MAX-DOAS observations over Thessaloniki, Greece: intercomparison and validation of two inversion algorithms Dimitris Karagkiozidis1 , Martina Michaela Friedrich2 , Steffen Beirle3 , Alkiviadis Bais1 , François Hendrick2 , Kalliopi Artemis Voudouri1 , Ilias Fountoulakis4,1 , Angelos Karanikolas5,6 , Paraskevi Tzoumaka7 , Michel Van Roozendael2 , Dimitris Balis1 , and Thomas Wagner3 1 Laboratory of Atmospheric Physics, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece Belgian Institute for Space Aeronomy (BIRA-IASB), 1180 Brussels, Belgium 3 Max Planck Institute for Chemistry, 55128 Mainz, Germany 4 Institute for Astronomy, Astrophysics, Space Applications and Remote Sensing, National Observatory of Athens (IAASARS/NOA), 15236 Athens, Greece 5 Physikalisch-Meteorologisches Observatorium Davos, World Radiation Center (PMOD/WRC), Dorfstrasse 33, 7260 Davos Dorf, Switzerland 6 ETH Zurich Institute for Particle Physics and Astrophysics, Hönggerberg campus, Stefano-Franscini-Platz 5, 8093 Zurich, Switzerland 7 Municipality of Thessaloniki, Department of Environment, 54642 Thessaloniki, Greece 2 Royal Correspondence: Dimitris Karagkiozidis () Received: 9 July 2021 – Discussion started: 31 August 2021 Revised: 24 January 2022 – Accepted: 29 January 2022 – Published: 11 March 2022 Abstract. In this study we focus on the retrieval of aerosol and trace gas vertical profiles from multi-axis differential optical absorption spectroscopy (MAX-DOAS) observations for the first time over Thessaloniki, Greece. We use two independent inversion algorithms for the profile retrievals: the Mexican MAX-DOAS Fit (MMF) and the Mainz Profile Algorithm (MAPA). The former is based on the optimal estimation method (OEM), while the latter follows a parameterization approach. We evaluate the performance of MMF and MAPA, and we validate their retrieved products with ancillary data measured by other co-located reference instruments. The trace gas differential slant column densities (dSCDs), simulated by the forward models, are in good agreement, except for HCHO, where larger scatter is observed due to the increased spectral noise of the measurements in the UV. We find an excellent agreement between the tropospheric column densities of NO2 retrieved by MMF and MAPA (slope = 1.009, Pearson’s correlation coefficient R = 0.982) and a good correlation for the case of HCHO (R = 0.927). For aerosols, we find better agreement for the aerosol optical depths (AODs) in the visible (i.e., at 477 nm) compared to the UV (at 360 nm), and we show that the agreement strongly depends on the O4 scaling factor that is used in the analysis. The agreement for NO2 and HCHO near-surface concentrations is similar to the comparison of the integrated columns with slightly decreased correlation coefficients. The seasonal mean vertical profiles that are retrieved by MMF and MAPA are intercompared, and the seasonal variation in all species along with possible sources is discussed. The AODs retrieved by the MAX-DOAS are validated by comparing them with AOD values measured by a CIMEL sun photometer and a Brewer spectrophotometer. Four different flagging schemes were applied to the data in order to evaluate their performance. Qualitatively, a generally good agreement is observed for both wavelengths, but we find a systematic bias from the CIMEL sun photometer and Brewer spectrophotometer measurements, due to the limited sensitivity of the MAX-DOAS in retrieving information at higher altitudes, especially in the UV. An in-depth validation of the aerosol vertical profiles retrieved by the MAX-DOAS is not possi- Published by Copernicus Publications on behalf of the European Geosciences Union. 1270 D. Karagkiozidis et al.: MAX-DOAS retrieval of vertical profiles over Thessaloniki, Greece ble since only in very few cases is the true aerosol profile known during the period of study. However, we examine four cases, where the MAX-DOAS provided a generally good estimation of the shape of the profiles retrieved by a co-located multi-wavelength lidar system. The NO2 near-surface concentrations are validated against in situ observations, and the comparison of both MMF and MAPA revealed good agreement with correlation coefficients of R = 0.78 and R = 0.73, respectively. Finally, the effect of the O4 scaling factor is investigated by intercomparing the integrated columns retrieved by the two algorithms and also by comparing the AODs derived by MAPA for different values of the scaling factor with AODs measured by the CIMEL sun photometer and the Brewer spectrophotometer. 1 Introduction The planetary boundary layer (PBL), also called atmospheric boundary layer, is defined as the lowermost layer of the troposphere that is directly influenced by the terrestrial surface. The PBL height, at midlatitudes, expands typically up to 1– 2 km during daytime (von Engeln and Teixeira, 2013), and its composition has a strong impact on weather, climate, and air quality. The increasing interest of understanding the PBL’s structure and dynamics is apparent in various research fields, from air pollution analysis to weather prediction, and thus, continuous ground-based monitoring of both chemical composition and aerosol content of the PBL with high temporal resolution is of great importance. Thessaloniki is a Mediterranean city, and it is the second largest city of Greece, located in the northern part of the country. Thessaloniki hosts approximately 10 % of the country’s total population with more than 1 million inhabitants (Hellenic Statistical Authority, 2011). With approximately 20 % of the country’s industrial activity, it is considered one of the largest urban agglomerations in the Balkans (Moussiopoulos et al., 2009). The air pollution sources in Thessaloniki are mainly industrial activities in the western part of the city as well as road transport and domestic heating during the cold period of the year, while the air quality of the city is affected by local topographic and meteorological characteristics (Poupkou et al., 2011; Kassomenos et al., 2011). Nitrogen oxides (NOx = NO + NO2 ), formaldehyde (HCHO) and aerosols are considered major atmospheric pollutants contained in the PBL of the city. Nitrogen dioxide (NO2 ) and HCHO are two important trace gas species of the atmosphere that play a critical role in tropospheric photochemistry (Seinfeld et al., 1998), participating in the formation of tropospheric ozone (O3 ), while aerosols can have a strong influence on air quality and climate through effects on radiation (IPCC, 2007). Both NO2 and HCHO are toxic to humans in high concentrations and can lead to severe health conditions. HCHO is a short-lived Atmos. Meas. Tech., 15, 1269–1301, 2022 (...truncated)


This is a preview of a remote PDF: https://amt.copernicus.org/articles/15/1269/2022/amt-15-1269-2022.pdf
Article home page: https://doaj.org/article/0e8be73daa4e410693935e73ac1d740f

D. Karagkiozidis, M. M. Friedrich, S. Beirle, A. Bais, F. Hendrick, K. A. Voudouri, I. Fountoulakis, I. Fountoulakis, A. Karanikolas, A. Karanikolas, P. Tzoumaka, M. Van Roozendael, D. Balis, T. Wagner. Retrieval of tropospheric aerosol, NO2, and HCHO vertical profiles from MAX-DOAS observations over Thessaloniki, Greece: intercomparison and validation of two inversion algorithms, Atmospheric Measurement Techniques, 2022, pp. 1269-1301, Issue 15, DOI: 10.5194/amt-15-1269-2022