Sensitivity of air quality modelling to different emission inventories: a case study over Europe

Abstract. In this paper we run the EMEP air quality model with 3 different emission inventories at 2015, that is to say using CAMS, EMEP emissions, and EDGAR. The EMEP model has been run for the entire year 2015, and resulting concentration results have been compared with background monitoring stations. Results show that the air quality model, run with the 3 emission inventories, provide similar results despite the emission differences. More in details, EDGAR is providing slightly better validation results for PM_2.5, while the EMEP emissions are slightly better to model yearly average NO₂. The main differences among the model applications arise in the Eastern part of Europe, where the values between the officially estimated emissions and those independently estimated by EDGAR are higher. Results suggest that EDGAR, despite being a methodology aimed at global coverage, with independent sources for activity level, technologies and emission factors and generic gridding practices, can be effectively used for air quality modelling in Europe. The EDGAR dataset (v5.0) used in this paper is available at: https://doi.org/10.2904/JRC_DATASET_EDGAR (link: EDGAR v5.0 Global Air Pollutant Emissions, Crippa et al., 2020a).