Improving the representation of glyoxal in the global EMAC model using TROPOMI retrievals and air-borne campaign data

Glyoxal (CHOCHO) is the simplest and one of the most abundant atmospheric α-dicarbonyls. In the atmosphere, it is mainly produced by the oxidation of other non-methane volatile organic compounds originating from natural and anthropogenic sources. Against photooxidation, it has a rather short lifetime of about 1 to 3 hours. Due to its high solubility, it quickly partitions into cloud droplets and deliquescent aerosols where it is known to form oligomers leading to secondary organic aerosols (SOA). In order to assess its significance for air pollution, it is thus necessary that global atmospheric chemistry models satisfactorily predict its abundance.

On board of the Sentinel-5 Precursor satellite, the TROPOspheric Monitoring Instrument (TROPOMI) provides tropospheric glyoxal columns. These tropospheric columns are generated with an improved version of the BIRA-IASB scientific retrieval algorithm relying on the Differential Optical Absorption Spectroscopy (DOAS) approach. By combining these retrievals with glyoxal measurements obtained during multiple air-borne campaigns using the High Altitude and Long Range Research Aircraft (HALO), we evaluate the capabilities of the ECHAM/MESSy Atmospheric Chemistry (EMAC) model to reproduce the global distribution and abundance of glyoxal. In its standard configuration, EMAC uses the detailed Mainz Organic Mechanism (MOM) to represent gas-phase chemistry. Additionally, we use the Model of Emissions of Gases and Aerosols from Nature (MEGAN) and the Emissions Database for Global Atmospheric Research (EDGAR, v4.3.2) to represent natural and anthropogenic emissions, respectively. When analysing an EMAC simulation using this standard configuration, we find that EMAC tends to overestimate tropospheric glyoxal columns over continental regions close to strong natural (e.g., Amazon Basin) and anthropogenic emission sources (e.g., China). At the same time, EMAC tends to underestimate glyoxal columns over tropical oceanic regions.

In this study, we perform a series of sensitivity simulations and demonstrate that the model bias over continental regions is mainly resolved by including detailed aqueous-phase chemistry from the Jülich Aqueous-phase Mechanism of Organic Chemistry (JAMOC) and by reducing biogenic emissions towards the latest estimates. In addition, by implementing additional glyoxal precursors from oceanic sources, we demonstrate that the model bias over the tropical ocean is reduced. Following the more realistic model representation of glyoxal levels, we present a revised tropospheric glyoxal budget.