Despite its narrow territory, the Italian Peninsula encloses many different geographical and topographical districts, from coastal to countryside, hilly, or mountainous areas. This likely generates similarities of urban air pollution among cities placed in different regions but comparable topographical districts. However, a nation-wide linkage among cities sharing similar geographical system, beyond their regional boundaries, is still missing, whereas it would support mitigation measures more targeted at specific citygeography features. Within the Working Group on Aerosol generation sources and processes (WG1) of the Italian Aerosol Society (IAS, https://www.iasaerosol.it/), a dataset (DS) was built by collecting mass concentration and chemical composition data (major and trace species) of urban PM10 related to more than 20 cities (large/medium/small; coastal/port/inland; etc.) spread over Italy within the 2005-2016 decade. Data were provided on voluntary basis by academic institutions and regional environment protection agencies. Time resolution of the DS is season average, for different years and different cities, based on data availability of contributors. Statistical analyses, estimation of missing data and identification of latent clusters were performed on the DS, to physical consistency checks and data validation aims. In addition, source apportionment modeling by Positive Matrix Factorization (PMF) was applied for the first time, to authors’ knowledge, using input data with such uniquely combined time resolution (as above described) and spatialization (nation-wide), thus representing a novelty [1]. In this contribution, key issues related to this novel PMF application are discussed, as well as the variability of source contribution estimates (SCE) with different geographical districts (Fig. 1). The most reliable and robust PMF solution is the seven-factor constrained analysis. Factors were tentatively labelled, according to their characteristic tracers, as Marine, Dust 1, Dust 2, Urban aerosol, Biomass Burning, Nitrate, Sulfate and Heavy Oil Combustion. Insights on the latter two SCE are in-depth discussed in a companion contribution. PMF chemical profiles were checked for physical consistency by different tests of modeling performance. [1] Pietrodangelo et al., (2024), Science of the Total Environment, 908, 167891
The nation-wide dataset of urban PM10 chemical speciation for Italy: a novel PMF application for source apportionment at Country-scale
M. MASIOL;
2024-01-01
Abstract
Despite its narrow territory, the Italian Peninsula encloses many different geographical and topographical districts, from coastal to countryside, hilly, or mountainous areas. This likely generates similarities of urban air pollution among cities placed in different regions but comparable topographical districts. However, a nation-wide linkage among cities sharing similar geographical system, beyond their regional boundaries, is still missing, whereas it would support mitigation measures more targeted at specific citygeography features. Within the Working Group on Aerosol generation sources and processes (WG1) of the Italian Aerosol Society (IAS, https://www.iasaerosol.it/), a dataset (DS) was built by collecting mass concentration and chemical composition data (major and trace species) of urban PM10 related to more than 20 cities (large/medium/small; coastal/port/inland; etc.) spread over Italy within the 2005-2016 decade. Data were provided on voluntary basis by academic institutions and regional environment protection agencies. Time resolution of the DS is season average, for different years and different cities, based on data availability of contributors. Statistical analyses, estimation of missing data and identification of latent clusters were performed on the DS, to physical consistency checks and data validation aims. In addition, source apportionment modeling by Positive Matrix Factorization (PMF) was applied for the first time, to authors’ knowledge, using input data with such uniquely combined time resolution (as above described) and spatialization (nation-wide), thus representing a novelty [1]. In this contribution, key issues related to this novel PMF application are discussed, as well as the variability of source contribution estimates (SCE) with different geographical districts (Fig. 1). The most reliable and robust PMF solution is the seven-factor constrained analysis. Factors were tentatively labelled, according to their characteristic tracers, as Marine, Dust 1, Dust 2, Urban aerosol, Biomass Burning, Nitrate, Sulfate and Heavy Oil Combustion. Insights on the latter two SCE are in-depth discussed in a companion contribution. PMF chemical profiles were checked for physical consistency by different tests of modeling performance. [1] Pietrodangelo et al., (2024), Science of the Total Environment, 908, 167891
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
