10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Advanced Receptor Models as a Tool to Improve the Knowledge of Aerosol Emission Sources at a Hot-Spot Pollution Site (Milan – Italy)
ROBERTA VECCHI, Vera Bernardoni, Alessandro Bigi, Giulia Calzolai, Miriam Elser, Paola Fermo, Alice Forello, Franco Lucarelli, Dario Massabò, Silvia Nava, Andrea Piazzalunga, Rosaria Erika Pileci, Paolo Prati, Sara Valentini, Gianluigi Valli, University of Milan & INFN-Milan (Italy)
Abstract Number: 42 Working Group: Source Apportionment
Abstract Emission sources and processes in atmosphere affect size distribution and chemical composition of atmospheric aerosol which are the main parameters responsible for aerosol effects at both local and global scale. Thus, understanding sources and processes leading to the measured concentrations in atmosphere is mandatory to develop suitable and effective abatement strategies.
Hot-spot pollution areas are peculiar sites where source emissions and meteorological conditions foster particulate matter accumulation, thus very high aerosol concentrations are often registered. The Po Valley (Northern Italy) – where Milan is located – is one of the main hot-spot pollution areas in Europe, especially during wintertime.
In this work, we present results from advanced receptor modelling approaches applied to aerosol collected with different sampling strategies in Milan (Italy) with the aim of improving the information generally obtained by traditional (e.g. Positive Matrix Factorization) receptor modelling.
The first example we will present is related to multi-time receptor modelling (Zhou et al., 2004), which allows a source apportionment study using all data with their own time resolution. This exploits the potentialities of high-time resolved measurements – e.g. allowing to detect sources acting for short periods – at the same time providing for such sources profile information also for components which are not measured with high time resolution.
To this aim, 1-h resolved elemental composition was measured by Particle-Induced X-Ray emission and 4-wavelength light absorption coefficient was determined by the polar photometer PP_UniMI (Bernardoni et al., 2017). In parallel, 24-h (during summer) or 12-h (during winter) resolved data of mass concentration, elements by Energy-Dispersive X-Ray Fluorescence, inorganic ions by ion chromatography, organic and elemental carbon by thermal-optical transmittance method, and levoglucosan by high-performance liquid chromatography coupled to pulsed amperometric detection were determined on 47 mm PTFE (for elements) and quartz fibre filters (for other components).
Another example of advanced source apportionment is the 3-way source apportionment (Ulbrich et al., 2012). In particular, a vector-matrix model (Tucker 1 model) was implemented. In this model, each element of the 3-D input matrix (representing the M species of the aerosol collected in N stages of a cascade impactor during R samplings) is factorised in S (unknown) factors. As reported in Bernardoni et al. (2017b) the model (implemented with Multilinear Engine 2, ME-2) was applied to size segregated aerosol samples collected using a Dekati-SDI cascade impactor (12 stages in the range 45nm-8.5μm) at an urban background station in Milan, Italy, during a winter period. Fourteen samplings were carried out with the cascade impactor, for a total of 168 samples available. All samples were collected on polycarbonate membranes. Elemental composition (Si-Pb) was determined by Energy-Dispersive X-Ray Fluorescence, the main inorganic ions (nitrate, sulphate and ammonium) by ion chromatography (IC), and levoglucosan (marker for wood burning) by high-performance liquid chromatography coupled to pulsed amperometric detection.
Special features of the results obtained with this model will be evidenced, such as the ability of identifying two sources related to traffic (i.e. diesel and gasoline), or the possibility to associate secondary compounds rapidly formed to specific sources (Bernardoni et al., 2017b).
[1] Bernardoni V., Valli G., Vecchi R. (2017). J. Aerosol Sci., 107, 84-93. [2] Bernardoni V., Elser M., Valli G., Valentini S., Bigi A., Fermo P., Piazzalunga A., Vecchi R. (2017b). Environ. Pollut., 231, 601-611. [3] Ulbrich, I. M., Canagaratna M. R., Cubison, M. J., Zhang, Q., Ng, N. L., Aiken, A. C., and Jimenez, J. L. (2012), Atmos. Meas. Tech., 5, 195–224. [4] Zhou L., Hopke P.K., Paatero P., Ondov J.M., Pancras J.P., Pekney N.J., Davidson C.I., (2004). Atmos. Environ. 38, 4909–4920.