Air quality in Europe continues to remain poor in many areas, with regulation limits often exceeded by many countries. The EU Life-IP PREPAIR Project, involving administrations and environmental protection agencies of eight regions and three municipalities in Northern Italy and Slovenia, was designed to support the implementation of the regional air quality plans in the Po Valley, one of the most critical areas in Europe in terms of pollution levels. In this study, four air quality modelling systems, based on three chemical transport models (CHIMERE, FARM and CAMx) were applied over the Po Valley to assess the sensitivity of PM2.5 concentrations to NOx and NH3 emission reductions. These two precursors were reduced (individually and simultaneously) from 25% up to 75% for a total of 10 scenarios, aimed at identifying the most efficient emission reduction strategies and to assess the non-linear response of PM2.5 concentrations to precursor changes. The multi-model analysis shows that reductions across multiple emission sectors are necessary to achieve optimal results. In addition, the analysis of non-linearities revealed that during the cold season, the efficiency of PM2.5 abatement tends to increase by increasing the emission reductions, while during summertime, the same efficiency remains almost constant, or slightly decreases towards higher reduction strengths. Since the concentrations of PM2.5 are greater in winter than in summer, it is reasonable to infer that significant emission reductions should be planned to maximise reduction effectiveness.
This paper presents a novel machine learning methodology able to extend the results of detailed local emission inventories to larger domains where emission estimates are not available. The first part of this work consists in the development of an emission inventory of elemental carbon (EC), black carbon (BC), organic carbon (OC), and levoglucosan (LG) obtained from the detailed emission estimates available from the Project LIFE PREPAIR for the Po Basin in north Italy. The emissions of these chemical species in combination with particulate primary emissions and gaseous precursors are very important information in source apportionment and in the impact assessment of the different emission sources in air quality. To gain a better understanding of the origins of atmospheric pollution, it is possible to combine measurements with emission estimates for the particulate matter fractions known as EC, BC, OC, and LG. To identify the sources of emissions, it is usual practice to use the ratio of the measured EC, OC, TC (Total Carbon), and LG. The PREPAIR emission estimates and these new calculations are then used to train the Random Forest (RF) algorithm, considering a large array of local variables, such as taxes, the characteristics of urbanization and dwellings, the number of employees detailed for economic activities, occupation levels and land cover. The outcome of the comparison of the predictions of the machine learning implemented model (ML) with the estimates obtained for the same areas by two independent methods, local disaggregation of the national emission inventory and Copernicus Air Modelling Service (CAMS) emissions estimates, is extremely encouraging and confirms it also as a promising approach in terms of effort saving. The implemented modelling approach identifies the most important variables affecting the spatialization of different pollutants in agreement with the main emission source characteristics and is suitable for harmonization of the results of different local emission inventories with national emission reporting.