Elsevier

Atmospheric Environment

Volume 45, Issue 40, December 2011, Pages 7688-7697
Atmospheric Environment

Source identification of ambient PM2.5 for inhalation exposure studies in Steubenville, Ohio using highly time-resolved measurements

https://doi.org/10.1016/j.atmosenv.2010.12.032Get rights and content

Abstract

Recent epidemiological and toxicological studies have suggested that short-term elevations of ambient fine particle mass concentrations (aerodynamic diameter <2.5 μm, PM2.5) can increase cardiac and pulmonary health risks. Thus, examining temporal variations of chemical changes in ambient PM2.5 that could pose the greatest health risks and identifying its sources is critical so that the most toxic categories can be controlled. In this study we collected detailed air quality data in Steubenville, Ohio in August 2006 with the ultimate goal to evaluate associations between cardiovascular (CV) parameters measured in exposed laboratory animals and the chemical and elemental composition of PM2.5. Current approaches using radiotelemetry to measure CV parameters in conscious laboratory animals are capable of collecting continuous recordings. To provide a robust and analogous dataset that can be better matched with CV responses, we have incorporated a highly time-resolved sampling method to characterize trace elements and thereby obtain more robust input data to determine potential emission sources. We applied positive matrix factorization (PMF) to trace element concentrations from 30-minute ambient PM2.5 samples in Steubenville, Ohio, an area designated as a non-attainment area for the PM2.5 National Ambient Air Quality Standards by the Environmental Protection Agency.

The average ambient PM2.5 filter-based mass concentration during the 8-hour summer exposure study period was 26 ± 11 μg m−3. Results from PMF indicated that six major factors contributed to the ambient PM2.5 mass during this time: coal combustion/secondary (39 ± 46%), mobile sources (12 ± 14%), metal coating/processing (10 ± 11%), iron and steel manufacturing (5 ± 5%), Pb factor (5 ± 8%), and incineration/smelting (1 ± 3%). The objectives of this paper are (1) to present chemical composition of ambient PM2.5 and its potential emission sources in Steubenville; and (2) to evaluate the PMF modeling results using observed meteorological data. These semi-continuous sampling approaches to determine potential emission sources have significant advantages over similar analyses using samples averaged over 8–24 h, and are being utilized by our group to determine associations of PM with acute CV responses from animal inhalation toxicology field studies.

Highlights

► Information on potential emission sources impacting Steubenville, Ohio was obtained. ► PMF results indicated that six major factors contributed to the ambient PM2.5 mass. ► These results are being utilized for animal inhalation toxicology field studies.

Introduction

The Ohio River Valley is characterized by numerous coal-fired power plants and other industrial facilities, and these sources are known to be significant contributors to ambient air pollution in the Eastern US (Kim et al., 2007, Pekney et al., 2006). In particular, the Steubenville, Ohio/Weirton, West Virginia area contains several large point sources, including coal-fired power plants, steel mills, coke facilities, metal processing operations, metal coating facilities, metal can industries, and tar production facilities and has been continuously designated as a PM2.5 non-attainment area by the Environmental Protection Agency (EPA) for violations of both 1997 and 2006 National Ambient Air Quality Standards. Furthermore, Steubenville’s air quality has been part of several historic epidemiological studies including the Harvard Six Cities and the American Cancer Society studies and has been consistently associated with an increased risk of morbidity and mortality (Krewski et al., 2003, Laden et al., 2000, Pope et al., 1995, Dockery et al., 1993, Schwartz and Dockery, 1992). More recently, health studies have reported that ambient PM2.5 mass and sulfate concentrations in Steubenville were associated with altered autonomic function (e.g., reduced heart rate variability (HRV)) (Luttmann-Gibson et al., 2006) and that ambient sulfate and ozone levels in the same location were associated with arrhythmia (Sarnat et al., 2006).

The intent of the overall Bi-City Concentrated Ambient Particle Study, which was conducted in Detroit, Michigan and Steubenville, Ohio, was to collect highly time-resolved exposure and outcome data on animals exposed to concentrated ambient PM at these two locations. Specifically, we collected HRV data; HRV varies with cardiac autonomic function, and reduced HRV is a widely-used clinical prognostic marker of increased risk of adverse cardiovascular (CV) effects (Tsuji et al., 1996). In animal studies, HRV data are typically collected using telemetric methods and continuous recordings. However the value of these highly time-resolved measurements is lost when evaluating associations with PM metrics that are typically averaged over 8-24 h. To provide a more analogous dataset that can be better matched with CV responses, a more highly time-resolved sampling method is needed to characterize trace elements; the resulting data can then be used to determine possible associations between elements and CV responses. Moreover, taken together with real-time meteorological data, the elemental data can be utilized in factor analytical or other receptor models to determine potential emission sources. This approach may help identify the sources and components of ambient PM2.5 that are linked with adverse health effects.

During our animal inhalation exposure studies, the semi-continuous aerosol sampler (SEAS) was used to measure highly time-resolved ambient PM2.5 trace elements and subsequently to determine potential sources. The development and use of the SEAS has been described in detail (Pancrasa et al., 2006, Ogulei et al., 2006, Kidwell and Ondov, 2004). Trace elements are particularly useful constituents of ambient PM as they can serve as tracers for specific emission sources, and some do not transform significantly in the atmosphere (Henry, 1997). Moreover, numerous recent studies have reported that metals in PM derived from combustion sources may be closely associated with the adverse health impacts of ambient PM2.5 (Cavallari et al., 2008, Hirshon et al., 2008, Lippmann et al., 2006, Gavett et al., 2003), making semi-continuous monitoring of trace elements particularly important to our understanding of component-specific and source-specific morbidity and mortality.

In the present study, highly-time resolved fine PM samples were collected as part of the Bi-City Concentrated Ambient Particle Study to assess health effects of ambient PM2.5 in regions dominated by different sources. Detroit, MI and Steubenville, OH were selected in order to capture PM dominated by mobile/industrial sources and coal-fired power plant/industrial sources, respectively. Exposure and cardiac function results for Detroit are described by Morishita et al. (2011) and Rohr et al. (in press), respectively. At both locations, associations between heart rate (HR) and HRV and the chemical and elemental composition of ambient PM2.5 were determined. Positive matrix factorization (PMF) modeling was used along with detailed meteorological analysis and particle morphological analysis to estimate the contribution of emission sources to the measured PM. In this paper we report the chemical composition of ambient PM2.5 and its potential emission sources in Steubenville and evaluate the PMF modeling results using observed meteorological data. A subsequent related report will describe the cardiac function findings.

Section snippets

Site description

AirCARE1, a 53-foot mobile air research laboratory was designed and constructed by Michigan State University and the University of Michigan; its construction and capabilities are described in detail elsewhere (Harkema et al., 2004). AirCARE1 was stationed in Steubenville, OH on the campus of Franciscan University (40.379N, 80.620W) overlooking the Ohio River. This monitoring site was selected because of its proximity to numerous anthropogenic air pollution sources in the Ohio River Valley.

Air pollution during the Steubenville summer study

The mean ± standard deviation of ambient PM2.5 mass concentration measured by TEOM during the 13-day exposure period was 25 ± 15 μg m−3 (Table 1). Another recent study in the Steubenville region – the Steubenville Comprehensive Air Monitoring Program (SCAMP) – reported that from 2000 through 2002 the average PM2.5 concentration measured in Steubenville was 18.4 μg m−3. However, since the current study was conducted only during the daylight hours of 13 summer days, the average concentration cannot be

Conclusions

Utilizing highly time-resolved elemental concentration data and receptor modeling, along with source directionality analysis, six factors for the ambient PM2.5 observed in Steubenville were identified and the contributions from each factor were determined. Primary sources included (1) coal combustion/secondary, (2) mobile sources, (3) metal coating/processing, (4) iron and steel manufacturing, (5) Pb factor, and (6) incineration/smelting. The identified factors contributing the most mass to

Acknowledgements

This work was supported by the Electric Power Research Institute (Contract EP-P10983/C5530/56546) and the U.S. Department of Energy’s National Energy Technology Laboratory (DE-FC26-03NT42303). However, any opinions, findings, conclusions, or recommendations expressed herein are those of the authors, and do not necessarily reflect the views of DOE-NETL. The authors would like to thank James Barres, Lori Bramble, Tim Dvonch, Lynne Gratz, Ryan Lewandowski, Frank Marsik, and Emily White for their

References (51)

  • D.P. Connell et al.

    The Steubenville Comprehensive Air Monitoring Program (SCAMP): analysis of short-term and episodic variations in PM2.5 concentrations using hourly air monitoring data

    J. Air Waste Manag. Assoc.

    (2005)
  • D.W. Dockery et al.

    An association between air pollution and mortality in six U.S. cities

    N. Engl. J. Med.

    (1993)
  • G.C. England et al.

    Dilution-based emissions sampling from stationary sources: Part 2-Gas-fired combustors compared with other fuel-fired systems

    J. Air Waste Manag. Assoc.

    (2007)
  • B.D. Garg et al.

    Brake wear particulate matter emissions

    Environ. Sci. Technol.

    (2000)
  • S.H. Gavett et al.

    Metal composition of ambient PM2.5 influences severity of allergic airways disease in mice

    Environ. Health Perspect.

    (2003)
  • Gertler, A.W., Gillies, J.A., Pierson, W.R., Rogers, C.F., Sagebiel, J.C., Abu-Allaban, M., Coulombe, W., Tarnay, L.,...
  • T.J. Grahame

    Does improved exposure information for PM2.5 constituents explain differing results among epidemiological studies?

    Inhal. Toxicol.

    (2009)
  • J.R. Harkema et al.

    Effects of inhaled urban air particulates on normal and hypersecretory airways in rats

    Health Effects Ins. Res. Rep.

    (2004)
  • J.M. Hirshon et al.

    Elevated ambient air zinc increases pediatric asthma morbidity

    Environ. Health Perspect.

    (2008)
  • P.K. Hopke et al.

    PM source apportionment and health effects: 1. Intercomparison of source apportionment results

    J. Exposure Sci. Environ. Epidemiol.

    (2006)
  • A. Iijima et al.

    Emission factor for antimony in brake abrasion dusts as one of the major atmospheric antimony sources

    Environ. Sci. Technol.

    (2008)
  • G.J. Keeler et al.

    Sources of mercury wet deposition in Eastern Ohio, USA

    Environ. Sci. Technol.

    (2006)
  • C.B. Kidwell et al.

    Development and evaluation of a prototype system for collecting sub-hourly ambient aerosol for chemical analysis

    Aerosol Sci. Technol.

    (2001)
  • C.B. Kidwell et al.

    Elemental analysis of sub-hourly ambient aerosol collections

    Aerosol Sci. Technol.

    (2004)
  • D. Krewski et al.

    Overview of the reanalysis of the Harvard Six Cities Study and American Cancer Society study of particulate air pollution and mortality

    J. Toxicol. Environ. Health A

    (2003)
  • Cited by (17)

    • Source directional apportionment of ambient PM<inf>2.5</inf> in urban and industrial sites at a megacity in China

      2020, Atmospheric Research
      Citation Excerpt :

      Thus, this factor was identified as crustal dust. Factor 5 was identified as industrial emission with high loadings of Fe (62.9%), Cd (44.4%), Cu (35.8%), Mn (60.8%), Ni (67.0%), Ti (55.3%), and Cl− (40.1%) (Almeida et al., 2015; Liu et al., 2016; Morishita et al., 2011; Mansha et al., 2012; Yao et al., 2016). Overall, secondary sources (SS), coal and biomass burning (CBB), vehicle exhaust (VE), crustal dust (CD), and industrial emission (IE) were identified as major PM2.5 sources using the PMF model.

    • Characterization and source apportionment of PM<inf>2.5</inf> based on error estimation from EPA PMF 5.0 model at a medium city in China

      2017, Environmental Pollution
      Citation Excerpt :

      Therefore, vehicle exhaust is a major contributor to PM2.5 in Heze. Factor 5 is characterized by the high loadings of Cr (24.3–67.4%), Cu (32.0–48.6%), Fe (31.8–47.7%), Mn (23.4–35.8%), Ni (33.3–65.0%), Ti (33.8–50.9%) and V (27.3–75.5%), which are closely related with the metal manufacturing industries (Almeida et al., 2015; Kabala and Singh, 2001; Liu et al., 2015, 2016; Morishita et al., 2011; Mansha et al., 2012; Yao et al., 2016). Therefore, factor 5 is identified as the mental manufacturing, and the contributions of this factor to PM2.5 vary from 2.7% (autumn) to 5.2% (winter), with an annual contribution of 3.8% (Fig. S13 and Fig. 5).

    • Metal-bearing fine particle sources in a coastal industrialized environment

      2017, Atmospheric Research
      Citation Excerpt :

      These studies led to chemical source profiles database as the U.S. EPA “SPECIATE” database or the recent European “SPECIEUROPE” database (Belis and Pernigotti, 2015). Especially, some PM2.5 sources apportionment studies using Positive Matrix Factorization (PMF) have shown that coal combustion, metal processing and refining, iron and steel manufacturing, waste and sludge incineration and cement/lime production, represent major sources of fine particles (Morishita et al., 2011a, 2011b). It is thus particularly important to characterize these metal-rich fine particles and estimate their source contribution to the ambient air pollution, especially near densely urbanized areas.

    • Chemical composition and source apportionment of ambient PM<inf>2.5</inf> during the non-heating period in Taian, China

      2016, Atmospheric Research
      Citation Excerpt :

      The measured PM2.5 concentrations and the calculated PM2.5 concentrations by PMF showed a higher correlation (R2 = 0.95) (Fig. S3). Factor 1 primarily consisted of Ti, V, Ni, Cr and Mn, and these elements were generally associated with iron/steel or other metal manufacturing (Almeida et al., 2015; Liu et al., 2015; Morishita et al., 2011; Mansha et al., 2012). According to the EFs analysis, the elements Ti, V, Ni, Cr and Mn in August–September, and Ni, Cr and Mn in November were derived mainly from anthropogenic sources (Fig. S2).

    • Chemical composition and source apportionment of the ambient PM<inf>2.5</inf> in Hangzhou, China

      2015, Particuology
      Citation Excerpt :

      Antimony can be released from many sources. For example, it could be derived from brake wear and iron/steel production (Mansha et al., 2012; Morishita et al., 2011; Querol et al., 2007). Therefore, factor 3 probably was a mixture of several anthropogenic emission sources.

    View all citing articles on Scopus
    View full text