Use of inorganic and organic markers associated with their directionality for the apportionment of highly correlated sources of particulate matter.

Particulate matter source identification using receptor models is one of the tools applied in air quality management. These models have limitations such as the collinearity effects, hindering their application and interpretation. Positive Matrix Factorization (PMF) models use chemical markers for the definition of likely sources, leaving to users the factors interpretation. This can lead to biased interpretations, as chemical species can be markers for several sources, particularly when there is source similarity. The Region of Greater Vitória, located southeast of Brazil, is a complex site in which similar industrial activities are installed, such as a pelletizing plant and a steel plant, that produce iron pellets and sinter, both iron-agglomerates with similar chemical profiles. To minimize the effects of collinearity between those sources, a new PMF approach is proposed by using inorganic and organic chemical species and the directionality of pollutant using wind roses. The proposed methodology determines the following consolidated markers: elemental carbon (EC) and organic carbon (OC) for vehicular sources; chloride (Cl) and sodium (Na) for sea salt; iron (Fe) for industrial sources. This association was possible by identifying the directionality of the chemical species. Cl a typical sea salt marker also attributed to industrial sintering activities. Some PMF factors showed high OC loadings, a typical marker for both vehicular exhaust and coal burning. The definition of the most appropriate sources for those factors was only possible due to the assessment of the pollutant roses. Pollutant roses generally showed that higher concentrations of potassium (K), a marker of biomass burning, was predominantly associated with winds from an industrial park, and are most likely associated with sintering emissions. Results showed that combining both organic and inorganic markers with the pollutant roses for identification of the directionality of predominant sources improved the interpretation of PMF factor numbers in source apportionment studies.