Fine Particulate Matter Sources Identification through PMF-Hysplit Modeling Based on Metals and Carbonaceous Compounds

RAFAELA SQUIZZATO, Caroline Fernanda Hei Wikuats, Thiago Nogueira, Edmilson Dias de Freitas, Maria de Fatima Andrade, University of São Paulo

     Abstract Number: 121
     Working Group: Biomass Combustion: Outdoor/Indoor Transport and Indoor Air Quality

Abstract
In South America, few studies involve the chemical composition of fine particulate matter (PM_2.5), especially in cities with less than 300,000 inhabitants, where live about 23 % of the world's population. In Brazil, those cities can have their air quality impacted by local and distant sources of air pollution. The lack of studies in these areas limits possible actions that aim to improve the control of atmospheric emissions. The present study was carried out in the southeastern region of Brazil, from June 12 to August 25 of 2019 (dry season), an area of frequent fires due to the planting of sugarcane for ethanol production (biofuel) and long-range transport of forest fires. Daily data on the elemental composition of PM_2.5 and carbonaceous compounds (OC, EC, and BCe) were considered for the identification of profiles and source contributions by applying the Positive Matrix Factorization (PMF) model. The main sources were soil/dust resuspension, biomass burning, and vehicular (light and heavy vehicles) emissions. The Hysplit trajectory model showed that the highest concentrations of particulate occurred when the trajectories were from the North, although the predominance of winds in the site was from the southeast. Preliminary results show that PM_2.5 concentrations were relatively low (average 9 ± 4 µg m^-3) during the sampled period. Still, unstable atmospheric conditions may have contributed to the dispersion of pollutants, altering the air quality at the station. During the sampling period, pollution plumes were detected from burning areas in Central and North Brazil and from the Metropolitan Region of São Paulo (RMSP), located 200 km away from the sampling site in the East/Southeast direction.