Source apportionment of ambient particle number concentrations in central Los Angeles using positive matrix factorization (PMF)

Mohammad Hossein Sowlat,Sina Hasheminassab,Constantinos Sioutas

doi:10.5194/acp-16-4849-2016

Mohammad Hossein Sowlat, Sina Hasheminassab + Show 1 more

Open Access

https://doi.org/10.5194/acp-16-4849-2016

Copy DOI

Abstract

Abstract. In this study, the positive matrix factorization (PMF) receptor model (version 5.0) was used to identify and quantify major sources contributing to particulate matter (PM) number concentrations, using PM number size distributions in the range of 13 nm to 10 µm combined with several auxiliary variables, including black carbon (BC), elemental and organic carbon (EC/OC), PM mass concentrations, gaseous pollutants, meteorological, and traffic counts data, collected for about 9 months between August 2014 and 2015 in central Los Angeles, CA. Several parameters, including particle number and volume size distribution profiles, profiles of auxiliary variables, contributions of different factors in different seasons to the total number concentrations, diurnal variations of each of the resolved factors in the cold and warm phases, weekday/weekend analysis for each of the resolved factors, and correlation between auxiliary variables and the relative contribution of each of the resolved factors, were used to identify PM sources. A six-factor solution was identified as the optimum for the aforementioned input data. The resolved factors comprised nucleation, traffic 1, traffic 2 (with a larger mode diameter than traffic 1 factor), urban background aerosol, secondary aerosol, and soil/road dust. Traffic sources (1 and 2) were the major contributor to PM number concentrations, collectively making up to above 60 % (60.8–68.4 %) of the total number concentrations during the study period. Their contribution was also significantly higher in the cold phase compared to the warm phase. Nucleation was another major factor significantly contributing to the total number concentrations (an overall contribution of 17 %, ranging from 11.7 to 24 %), with a larger contribution during the warm phase than in the cold phase. The other identified factors were urban background aerosol, secondary aerosol, and soil/road dust, with relative contributions of approximately 12 % (7.4–17.1), 2.1 % (1.5–2.5 %), and 1.1 % (0.2–6.3 %), respectively, overall accounting for about 15 % (15.2–19.8 %) of PM number concentrations. As expected, PM number concentrations were dominated by factors with smaller mode diameters, such as traffic and nucleation. On the other hand, PM volume and mass concentrations in the study area were mostly affected by sources with larger mode diameters, including secondary aerosols and soil/road dust. Results from the present study can be used as input parameters in future epidemiological studies to link PM sources to adverse health effects as well as by policymakers to set targeted and more protective emission standards for PM.

Highlights

Numerous epidemiological studies have provided compelling evidence linking exposure to ambient particulate matter (PM) with increased risk of respiratory and cardiovascular diseases, hospitalization, and premature mortality (Brunekreef and Forsberg, 2005; Dockery and Stone, 2007; Miller et al, 2007; Pope et al, 2004, 2002; Gauderman et al, 2015)
In the current version of the model, i.e., positive matrix factorization (PMF) 5.0, the method used for calculating the S / N ratio has been updated compared to the previous versions, resolving the disadvantages associated with the previous method of S / N calculation
The present study was the first attempt to characterize major sources of PM number concentrations and to quantify their contributions using the PMF receptor model applied on PM number size distributions in the range of 13 nm to 10 μm combined with several auxiliary variables, including black carbon (BC), elemental carbon (EC)/organic carbon (OC), PM mass, gaseous pollutants, meteorological, and traffic flow data, in central Los Angeles

Summary

Introduction

Numerous epidemiological studies have provided compelling evidence linking exposure to ambient particulate matter (PM) with increased risk of respiratory and cardiovascular diseases, hospitalization, and premature mortality (Brunekreef and Forsberg, 2005; Dockery and Stone, 2007; Miller et al, 2007; Pope et al, 2004, 2002; Gauderman et al, 2015). According to the most recent global burden of disease study, over 3 million premature deaths occur annually all around the globe due to exposure to ambient PM (Lim et al, 2013). It should, be noted that most of these epidemiological studies have related the aforementioned health outcomes with solely the mass concentrations of PM and, do not adequately represent submicron particles (Ogulei et al, 2007), mainly because this PM fraction contributes negligibly to total ambient PM mass (Delfino et al, 2005; Vu et al, 2015). Several studies have found that PM number concentrations (mostly UFPs) can be associated with adverse effects on human health, for cardiovascular diseases (Delfino et al, 2005; Peters et al, 1997; Wichmann et al, 2000)

Methods

Results

Conclusion