Emission ensemble approach to improve the development of multi-scale emission inventories

Many studies have shown that emission invento ries are one of the inputs with the most critical influences on the results of air quality modelling. Comparing emission in ventories among themselves is, therefore, essential to build confidence in emission estimates. In this work, we extend the approach of Thunis et al. (2022) to compare emission inventories by building a benchmark that serves as a refer ence for comparisons. This benchmark is an ensemble that is based on three state-of-the-art EU-wide inventories: CAMS REG, EMEP and EDGAR. The ensemble-based methodol ogy screens differences between inventories and the ensem ble. It excludes differences that are not relevant and iden tifies among the remaining ones those that need special at tention. We applied the ensemble-based screening to both an EU-wide and a local (Poland) inventory. The EU-wide analysis highlighted a large number of in consistencies. While the origin of some differences between EDGAR and the ensemble can be identified, their magnitude remains to be explained. These differences mostly occur for SO2 (sulfur oxides), PM (particulate matter) and NMVOC (non-methane volatile organic carbon) for the industrial and residential sectors and reach a factor of 10 in some instances. Spatial inconsistencies mostly occur for the industry and other sectors. At the local scale, inconsistencies relate mostly to differ ences in country sectorial shares that result from different sectors/activities being accounted for in the two types of in ventories. This is explained by the fact that some emission sources are omitted in the local inventory due to a lack of appropriate geographically allocated activity data. We iden tified sectors and pollutants for which discussion between local and EU-wide emission compilers would be needed in order to reduce the magnitude of the observed differences (e.g. in the residential and industrial sectors). The ensemble-based screening proved to be a useful ap proach to spot inconsistencies by reducing the number of necessary inventory comparisons. With the progressive res olution of inconsistencies and associated inventory improve ments, the ensemble will improve. In this sense, we see the ensemble as a useful tool to motivate the community around a single common benchmark and monitor progress towards the improvement of regionally and locally developed emis sion inventories.