Universität zu Köln

Backes, Pascal Martin ORCID: 0000-0003-3449-2093 (2023). Polluter group specific emission optimisation for regional air quality analyses using four-dimensional variational data assimilation. PhD thesis, Universität zu Köln.

PDF Pascal_Martin_Backes_Dissertation.pdf - Published Version Download (56MB)

Abstract

Pollutants in the atmosphere, such as nitrogen oxides and particulate matter, pose a threat to the environment and human health. In addition to natural sources, anthropogenic emissions contribute significantly to air pollution. Since emission rates cannot be measured directly, their estimates provided by research institutes and national environmental agencies are subject to considerable uncertainty. However, for accurate air quality forecasts using atmospheric chemical transport models such as the regional European Air pollution Dispersion - Inverse Model (EURAD-IM), reliable emission data are crucial. To correct the emission data of inventories based on observations of trace gas and aerosol concentrations in the atmosphere, the EURAD-IM comprises a four dimensional variational data assimilation system (4D-Var) that allows for simultaneous optimisation of initial concentrations and species-dependent emission corrections. In order to improve the knowledge about the sources of air pollution, in this work, a new approach is developed and implemented in the data assimilation system of the EURAD-IM to correct emissions individually for source categories such as road transport, industry and agriculture. For the distinction between the emissions of different source categories, the new approach exploits the spatial separation of emission sources of different categories as well as their characteristic diurnal emission profiles and chemical compositions. Assuming a fixed chemical composition of the emissions of the source categories within the grid cells, a full correlation between the emission corrections of the different chemical species is introduced. Furthermore, an anisotropic diffusion operator is implemented that increases the spatial correlation between the road traffic emission corrections of the grid cells along roads. To investigate the ability of the new development to distinguish between emissions of different sectors, two different types of simulations are performed. In identical twin experiments based on synthetic observations, scenarios with increased industrial and agricultural emissions and a simultaneous decrease in road transport emissions are simulated. The data assimilation system based on the new approach is able to reproduce the emission changes in the experiments for large parts of the model domain through the determined sector specific emission corrections. Furthermore, a study is performed in which the emissions within a two-week period in North Rhine-Westphalia are analysed using real observation data. It is shown that in this scenario a distinction of industrial and power plant emissions versus road transport emissions is possible through the sector specific emission optimisation. Moreover, changes in agricultural emissions can be specified due to their high NH3 fraction. For all observed species, i.e. O3, NO2, SO2, PM10 and PM2.5, the agreement of the simulated with the observed concentrations is comparable to that of a reference simulation using the current EURAD-IM data assimilation system. An improvement of the results is expected through additional observation data, especially of CO and CO2 concentrations.

Item Type:	Thesis (PhD thesis)
Translated title:	Title Language Verursachergruppenspezifische Emissionsoptimierung für regionale Luftqualitätsanalysen unter Benutzung vier-dimensionaler variationeller Datenassimilation German
Creators:	Creators Email ORCID ORCID Put Code Backes, Pascal Martin UNSPECIFIED orcid.org/0000-0003-3449-2093 UNSPECIFIED
URN:	urn:nbn:de:hbz:38-704923
Date:	2023
Publisher:	Verlag des Forschungszentrums Jülich
Place of Publication:	Jülich
ISBN:	978-3-95806-717-2
Language:	English
Faculty:	Faculty of Mathematics and Natural Sciences
Divisions:	Außeruniversitäre Forschungseinrichtungen > Forschungszentrum Jülich
Subjects:	Natural sciences and mathematics Earth sciences
Uncontrolled Keywords:	Keywords Language Data Assimilation; 4D-var; sector specific emission optimisation; GNFR sectors; atmospheric chemistry transport model; EURAD-IM English Datenassimilation; 4D-var; sektorspezifische Emissionsoptimierung; GNFR-Sektoren; atmosphärisches Chemietransportmodell; EURAD-IM German
Date of oral exam:	12 July 2023
Referee:	Name Academic Title Fuchs, Hendrik Prof. Dr. Crewell, Susanne Prof. Dr.
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/70492