Research article
Monitoring and source apportionment of trace elements in PM2.5: Implications for local air quality management

https://doi.org/10.1016/j.jenvman.2017.02.059Get rights and content

Highlights

  • Fine particles (PM2.5) samples were collected hourly at five sites for one year.

  • Spatiotemporal variations of trace elements provided information on source profiles.

  • Secondary aerosols, traffic, and coal burning were dominant sources.

  • Policies prioritizing air quality were proposed to mitigate PM2.5 pollutants.

  • Positive matrix factorization could help inform decisions on air quality management.

Abstract

Fine particulate matter (PM2.5) samples were collected simultaneously every hour in Beijing between April 2014 and April 2015 at five sites. Thirteen trace elements (TEs) in PM2.5 were analyzed by online X-ray fluorescence (XRF). The annual average PM2.5 concentrations ranged from 76.8 to 102.7 μg m−3. TEs accounted for 5.9%–8.7% of the total PM2.5 mass with Cl, S, K, and Si as the most dominant elements. Spearman correlation coefficients of PM2.5 or TE concentrations between the background site and other sites showed that PM2.5 and some element loadings were affected by regional and local sources, whereas Cr, Si, and Ni were attributed to substantial local emissions. Temporal variations of TEs in PM2.5 were significant and provided information on source profiles. The PM2.5 concentrations were highest in autumn and lowest in summer. Mn and Cr showed similar variation. Fe, Ca, Si, and Ti tended to show higher concentrations in spring, whereas concentrations of S peaked in summer. Concentrations of Cl, K, Pb, Zn, Cu, and Ni peaked in winter. PM2.5 and TE median concentrations were higher on Saturdays than on weekdays. The diurnal pattern of PM2.5 and TE median concentrations yielded similar bimodal patterns. Five dominant sources of PM2.5 mass were identified via positive matrix factorization (PMF). These sources included the regional and local secondary aerosols, traffic, coal burning, soil dust, and metal processing. Air quality management strategies, including regional environmental coordination and collaboration, reduction in secondary aerosol precursors, restrictive vehicle emission standards, promotion of public transport, and adoption of clean energy, should be strictly implemented. High time-resolution measurements of TEs provided detailed source profiles, which can greatly improve precision in interpreting source apportionment calculations; the PMF analysis of online XRF data is a powerful tool for local air quality management.

Introduction

Beijing is the capital city of China with a population of over 21.7 million in 2015 (BJSTATS, 2016). With rapid economic development and urbanization, Beijing has experienced serious air quality problems characterized by high concentrations of fine particulate matter (PM2.5) (Cheng et al., 2013). The annual average PM2.5 concentration was 85.9 μg m−3 (BJEPB, 2016), which was considerably higher than the National Ambient Air Quality Standard (NAAQS) of 35 μg m−3 (Class Ⅱ Annual Average Level). High PM2.5 masses significantly influence air quality and urban visibility and adversely affect human health (Pui et al., 2014). In response to the severity of particulate air pollution, a series of air quality control measures have been proposed, including the reduction of emission sources, regulations on vehicular emission standards, improvements to industrial and energy structures, and reductions in coal burning. The design and assessment of air quality control strategies require a thorough understanding of major sources of PM2.5.

Numerous source apportionment (SA) methods, such as positive matrix factorization (PMF) (Gao et al., 2014, Zhang et al., 2013), enrichment factor (Lü et al., 2012), principal component analysis (Li et al., 2013), and chemical mass balance (Wang et al., 2009b), have been implemented to identify the main sources that emit PM2.5 or its precursors. Over the last decade, extensive studies of particulate matter (PM) in Beijing have revealed that major sources include coal burning, vehicular emissions, biomass burning, secondary aerosols, soil dust, and industrial emissions (Song et al., 2006, Song et al., 2012, Zhang et al., 2013). Minor sources, including Chinese cooking emissions (Wang et al., 2009b), cigarette smoke, and vegetative detritus (Zheng et al., 2005), also contribute to total PM concentrations. Previous PM2.5 chemical profiles have been typically based on offline filter collections of 12–24 h periods, which involves labor-intensive and time-consuming sampling collection and chemical analysis processes. Small-scale fluctuations in emissions or incursions of polluted air masses (e.g., plume events) cannot be identified as well, and thus additional information on source profiles cannot be acquired. These limitations restrict the application of SA to air quality strategies.

To overcome these limitations, automatic observation instruments have provided alternative methods to obtain high time-resolution measurements of air samples by collecting a large volume of data with quick and easy processes and providing data at sampling intervals similar to meteorological variations. Data from these instruments can provide temporal variations of contributing sources and meteorology (Lioy et al., 1989). A large number of samples is vital for multivariate models to produce reliable receptor model results (Kim Oanh et al., 2009). Among various online measurements of PM chemical profiles, measurements of trace elements (TEs) cover a large proportion. Although TEs generally account for a few percent of the total mass of PM2.5, they are characteristic chemical species associated with PM emissions from both anthropogenic and natural sources. Given their high source specificity and atmospheric stability, TEs have been employed as effective tracers and used in numerous SA calculations for identifying and apportioning the source contributions to ambient aerosol (Visser et al., 2015, Yu et al., 2013). Despite their practicability, simultaneous online measurements and SA analyses of TEs at multiple sites in Beijing have been seldom reported.

This work primarily aimed to develop a procedure for capturing spatiotemporal variations across a range of environments to gather precise information on PM2.5 source profiles and provide appropriate data-based policy recommendations for local air quality measures. In this study, the concentrations of PM2.5 mass and its chemical species of 13 TEs were detected at five sites in Beijing between April 2014 and April 2015. Online X-ray fluorescence (XRF) analysis was utilized for elemental analysis. Information on spatial and temporal (seasonal, weekly, and diurnal) variations of PM2.5 and TEs is reported together with an evaluation of the likely SA of PM2.5 using PMF modeling. The results are important in assessing long- and short-term air quality, applying PMF analysis of online XRF data to capture different PM2.5 sources, and establishing effective air quality measures.

Section snippets

Sampling sites

Beijing is located at 39°56′N and 116°20′E on the northwestern edge of the Great North China Plain and surrounded by the Taihang Mountains in the west and Yanshan Mountains in the north and northeast, approximately 100 and 50 km far from the urban region (Guinot et al., 2007). The city is 160 km away from the Bohai Sea via Tianjin in the southeast. This region is affected by predominant westerly wind and Asian monsoon climate featuring warm and humid southern winds in summer and cold and dry

TE concentrations in PM2.5

The highest annual average PM2.5 concentrations were observed at Site 4, followed in descending order by Sites 5, 3, and 2; the lowest concentrations were recorded at Site 1 (background site). All sites showed annual average PM2.5 concentrations that were 2.1–2.9 times higher than 35 μg m−3. The values reported in this study were comparable to the PM2.5 values of 83 μg m−3 measured at several sites in Beijing (Gao et al., 2014). The number of days exceeding 75 μg m−3 (Class II 24 h average

Conclusions

PM2.5 samples were collected continuously every hour between April 2014 and April 2015 at five sites in Beijing. The characterization of PM2.5 elemental composition was achieved via online XRF, and sources contributing to PM2.5 were identified via PMF. The study provided a practical methodology using environmental monitoring data for local air quality management.

During the sampling period, the annual average PM2.5 concentration ranged from 76.8 ± 78.1 to 102.7 ± 99.9 μg m−3 and exceeded NAAQS

Acknowledgements

This work was supported by the National Science & Technology Pillar Program [grant number 2015BAK12B02] and the National Natural Science Foundation of China [grant number 71573149]. We thank Lei Jiang for his technical support in the drawing of sampling sites.

References (59)

  • P.L. Kinney et al.

    Traffic impacts on PM2.5 air quality in Nairobi, Kenya

    Environ. Sci. Policy

    (2011)
  • C. Klausbruckner et al.

    A policy review of synergies and trade-offs in South African climate change mitigation and air pollution control strategies

    Environ. Sci. Policy

    (2016)
  • P.H. Kuo et al.

    Retrospective assessment of air quality management practices in Taiwan

    Atmos. Environ.

    (2009)
  • S.L. et al.

    Size distribution of chemical elements and their source apportionment in ambient coarse, fine, and ultrafine particles in Shanghai urban summer atmosphere

    J. Environ. Sci.

    (2012)
  • H.J. Lee et al.

    Spatial and temporal variability of fine particle composition and source types in five cities of Connecticut and Massachusetts

    Sci. Total. Environ.

    (2011)
  • Z.G. Li et al.

    Distributions, sources and pollution status of 17 trace metal/metalloids in the street dust of a heavily industrialized city of central China

    Environ. Pollut.

    (2013)
  • P.J. Lioy et al.

    The effect of sampling duration on the ability to resolve source types using factor analysis

    Atmos. Environ.

    (1989)
  • G. Liu et al.

    Chemical composition and source apportionment of the ambient PM2.5 in Hangzhou, China

    Particuology

    (2015)
  • W. Maenhaut

    ”Global Change“ related and other atmospheric aerosol research at the University of Gent, and the role of PIXE therein

    Nucl. Instrum. Meth. B

    (1996)
  • K. Na et al.

    Characterization and source identification of trace elements in PM2.5 from Mira Loma, Southern California

    Atmos. Res.

    (2009)
  • T. Okuda et al.

    Source identification of nickel in TSP and PM2.5 in Tokyo

    Jpn. Atmos. Environ.

    (2007)
  • J. Pey et al.

    Monitoring of sources and atmospheric processes controlling air quality in an urban Mediterranean environment

    Atmos. Environ.

    (2010)
  • D.Y.H. Pui et al.

    PM2.5 in China: measurements, sources, visibility and health effects, and mitigation

    Particuology

    (2014)
  • Y. Song et al.

    Source apportionment of PM2.5 in Beijing by positive matrix factorization

    Atmos. Environ.

    (2006)
  • D.G. Streets et al.

    Air quality during the 2008 Beijing Olympic Games

    Atmos. Environ.

    (2007)
  • C. Sun et al.

    Restricting driving for better traffic and clearer skies: did it work in Beijing?

    Transp. Policy

    (2014)
  • Y.L. Sun et al.

    The air-borne particulate pollution in Beijing—concentration, composition, distribution and sources

    Atmos. Environ.

    (2004)
  • A.M. Taiwo et al.

    A review of receptor modelling of industrially emitted particulate matter

    Atmos. Environ.

    (2014)
  • J.H. Tan et al.

    Source apportionment of size segregated fine/ultrafine particle by PMF in Beijing

    Atmos. Res.

    (2014)
  • Cited by (0)

    View full text