Universität zu Köln

Erraji, Hassnae ORCID: 0000-0001-8970-3893 (2026). The potential of assimilating in situ airborne observations within the planetary boundary layer to enhance regional air quality analyses. PhD thesis, Universität zu Köln.

PDF (PhD thesis) Thesis_Hassnae_Erraji_2025_1.pdf - Published Version Download (63MB)

Abstract

Air pollution represents a serious threat to human health, the environment, and climate. One effective way to protect public health and take action to reduce emissions is through numerical model analysis, which depends on observational data to enhance the pollutant’s representation at regional scales. While the observing system network is extensive and diverse, there is a concerning gap in observational data within the Planetary boundary layer (PBL), the most polluted part of the atmosphere, and the location of most emissions. This work aims to evaluate the potential of in situ airborne observations taken within the PBL to improve regional air quality analyses. To achieve this, the impact of assimilating these observations on the horizontal and vertical distribution of air pollutants has been assessed, and the added value of the observations within the PBL has been evaluated by comparing the analyses of different case studies to analyses in which ground-based observations are assimilated. Two types of observations within the PBL are analysed in two separate studies: drone-based and Zeppelin-based measurements. The drone data originate from the MesSBAR campaign in 2021, while the Zeppelin observations, collected in 2020, are used to investigate pollutant distributions in two distinct regions. The EURAD-IM model and its 4D-Var assimilation system are utilised to assimilate both types of observations. A high temporal resolution of 60 seconds and 5~km x 5~km horizontal resolution is employed to align with the high-resolution measurements. A joint optimisation of the initial values and emission rates is applied across all assimilation simulations. For both observation types, the 4D-Var assimilation positively improves the horizontal and vertical representation of pollutants. Despite the drone system being equipped with low-cost sensors, the analysis shows that ground concentrations of nitrogen oxide (NO), nitrogen dioxide (NO2), and ozone (O3) were primarily corrected at the campaign location during nighttime and early morning hours. Furthermore, the assimilation of drone observations leads to substantial adjustments of nitrogen oxides emission rates in the vicinity of the campaign site. The analysis applying Zeppelin observations assimilation indicates that these observations effectively enhance corrections of emissions originating from power plants and industrial sources released at high altitudes. Achieving similar results with ground-based observations is challenging due to their limited capacity to observe these elevated concentrations, except under strong wind conditions and enhanced vertical mixing. The comparison with independent observations reveals that assimilation of Zeppelin observations significantly enhances daytime surface O3 concentrations by up to 54%, whereas the assimilation of ground-based measurements only yields improvements of up to 22%. Conversely, during nighttime, the assimilation of ground-based observations demonstrates superior performance, achieving an enhancement of up to 20%, compared to just 6% for the assimilation of Zeppelin observations. The daily O3 analysis is best when the observations from the Zeppelin and the ground stations are jointly assimilated, leading to additional improvement of up to 6% compared to other simulations. The improvements in the daytime O3 concentrations resulting from the assimilation of Zeppelin observations are attributed to corrections applied to the initial values within the residual layer. Furthermore, in a polluted urban environment, the assimilation of Zeppelin observations results in a decline in performance during the night. Potential factors contributing to this deterioration include the model resolution, the representation of PBL height, and the poor representation of unobserved pollutants. The positive impact of assimilating in situ drone and Zeppelin data on air quality analyses underscores the importance of observations within the PBL. This importance is further reflected in the optimisation of emission rates, particularly for elevated sources such as power plants and industrial facilities. These findings confirm that PBL observations offer substantial added value compared to ground-based measurements. Therefore, there is a need to increase measurements within the PBL to address the existing observational gap and to provide valuable data for assimilation in air quality models.

Item Type:	Thesis (PhD thesis)
Creators:	Creators Email ORCID ORCID Put Code Erraji, Hassnae hassnae.erraji@gmail.com https://orcid.org/0000-0001-8970-3893 UNSPECIFIED
URN:	urn:nbn:de:hbz:38-799221
Date:	2026
Language:	English
Faculty:	Faculty of Mathematics and Natural Sciences
Divisions:	Faculty of Mathematics and Natural Sciences > Department of Geosciences > Institute for Geophysics and Meteorology
Subjects:	Natural sciences and mathematics Earth sciences
Uncontrolled Keywords:	Keywords Language Air quality English Data assimilation English Dispersion modelling English PBL English Chemical transport model English 4D-var English Emissions optimisation English
Date of oral exam:	18 August 2025
Referee:	Name Academic Title Wahner, Andreas Prof. Dr. Löhnert, Ulrich Prof. Dr.
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/79922