Source apportionment of urban PM2.5 using positive matrix factorization with vertically distributed measurements of trace elements and nonpolar organic compounds

doi:10.1016/j.apr.2021.03.007

Atmospheric Pollution Research

Volume 12, Issue 4, April 2021, Pages 200-207

https://doi.org/10.1016/j.apr.2021.03.007 Get rights and content

Highlights

•
We explored vertical distribution of source-specific contributions of urban PM_2.5
•
The main contributors were industrial emission, biomass burning and oil combustion.
•
The main contributors showed non-significant variation in the vertical distribution.
•
Traffic related contributions to PM_2.5 declined with height in autumn and winter.
•
Environmental tobacco smoke (ETS)/lubricant showed vertical difference in autumn.

Abstract

There are limited source apportionment studies conducted to assess vertical distributions of fine particulate matter (PM_2.5) and its contributing sources in urban regions. Additionally, none of these vertical studies used inorganic and organic tracers simultaneously. The present study aimed to explore the vertical variation of source-specific contributions to PM_2.5 by modeling with both inorganic and organic markers in vertically stratified samples in the Taipei metropolis (Taiwan). One hundred and five samples were collected from three floor-levels at one building and analyzed for trace elements and nonpolar organic compounds (NPOCs). Seven source factors, i.e., industrial emission, biomass burning, PAH related, oil combustion, soil dust, traffic related, and environmental tobacco smoke (ETS)/lubricant, were retrieved by positive matrix factorization (PMF). The major contributors, including industrial emission (29%), biomass burning (23%), and oil combustion (22%), could be involved with regional transported aerosols and showed non-significant variation in the vertical distribution. In contrast, both traffic related and ETS/lubricant factors could be the contributors to the statistically significant difference of PM_2.5 concentrations between floor-levels.

Introduction

Air pollution is of continuous concern worldwide due to its impacts on humans and the environment. Strong associations of adverse health outcomes, such as allergies, asthma, respiratory symptoms, and cardiovascular disease, with exposure to airborne fine particulate matter (PM_2.5) have been reported (Bell et al., 2013; Kim et al., 2015; Lu et al., 2015). PM_2.5 is a mixture of numerous chemicals that can come from diverse sources. To effectively reduce PM_2.5 exposure, receptor models have been developed for source apportionment of PM_2.5 and associated health risks (Cooper and Watson, 1980; Hopke et al., 1976; Paatero and Tapper, 1994). In recent decades, the multivariate solution (e.g., positive matrix factorization or PMF) has been frequently utilized in source apportionment studies (Belis et al., 2013; Hopke, 2016; Paatero and Tapper, 1994; Viana et al., 2008). One advantage of PMF is that it requires less prior information about pollution sources than other source apportionment models, and therefore making it become the most commonly applied receptor model nowadays (Hopke, 2016).

Most PMF studies were based on trace elements, inorganic ions, and/or carbon fractions (organic and elemental carbon, OC/EC) collected at the receptor site. Although temperature-resolved carbon sub-fractions have been utilized to improve source identification, the improvement was limited in traffic related pollution (Sahu et al., 2011. Collecting additional characteristic species such as organic molecular markers could be useful for better characterizing organic aerosols that are unable to be apportioned with inorganic species (Hopke, 2016; Lin et al., 2010; Watson et al., 2015). For example, the commonly used tracers for primary organic aerosols are a set of nonpolar organic compounds (NPOCs) that include alkanes (linear and branched), hopanes, steranes, and polycyclic aromatic hydrocarbons (PAHs) (Ho et al., 2008; Xu et al., 2013). In recent years, there is an increasing number of studies applying both inorganic and organic (i.e., NPOCs) speciation data into PMF modeling, providing more markers for interpretation of source profiles (Han et al., 2018; Liao et al., 2015; Wang et al., 2016b, 2019; Wong et al., 2019).

Urban PM_2.5 pollution is a particularly critical issue because of both high population density and numerous anthropogenic pollution sources. There is a challenging task for source apportionment of PM_2.5 in metropolitan areas where high-rise apartment buildings are common. More than half of the population in the Taipei metropolis are estimated to live above the second floor (Wu and Lung, 2012). Using data collected at a certain height could result in biased exposure estimates at different residential heights. Although there has been an increasing interest in assessing vertical distributions of PM_2.5 in urban areas, only limited receptor modeling studies were conducted based on the data collected at different building floors (Pongpiachan, 2013; Wang et al., 2016a; Wu et al., 2015). These studies revealed vertical characteristics of potential PM_2.5 sources, but none of them estimated source-specific contributions to PM_2.5 using inorganic and organic tracers simultaneously.

Our previous study combined vertically distributed data of trace elements and PAHs from multiple buildings to improve the identification of potential PM_2.5 sources and their vertical distribution (Liao et al., 2020). The findings demonstrated the enhancement of source interpretation using the combined data set. Nonetheless, modeling pooled data from multiple buildings retrieved mixed source profiles. In addition, the small sample size (nine samples per floor per building) may result in large uncertainty in estimating source contributions.

To address these issues, the aim of the present study is to investigate the vertical variation of source-specific contributions to PM_2.5 through a one-building field campaign. We analyzed both inorganic (i.e., trace elements) and organic (i.e., NPOCs) markers in vertically stratified samples collected from three floor-levels at one building in a metropolitan area. A special emphasis is put on evaluating the vertical variability of the source contribution estimates by modeling with elemental compositions and NPOCs in PM_2.5 in the metropolitan area.

Section snippets

Field campaigns

The map of the study region (the Taipei metropolis, Taiwan) is shown in Fig. 1. The study area has a high population density and is a dynamic commercial hub. In addition, there are also several potential sources of air pollution nearby. One building that has more than ten floors (with balconies facing a major road) was selected to collect vertically stratified samples. The 24-h PM_2.5 samples were simultaneously collected from balconies situated at three different floor-levels, that is, low-

Source apportionment

Table S1 summarizes PM_2.5 monitoring results from several studies conducted in the Taipei metropolis. In general, a decreasing trend of average PM_2.5 mass concentrations over time was observed. In the present study, the average concentration of PM_2.5 throughout the study period was 12.2 μg m⁻³, which is slightly higher than the recommended exposure level in the WHO Air Quality Guidelines (10 μg m⁻³ annual mean). Among the 28 variables shown in Table 1, OC was the major constituent of PM_2.5,

Conclusions

We have characterized the vertical variability of the source contribution estimates by modeling with both inorganic (i.e., trace elements) and organic (i.e., NPOCs) markers in PM_2.5 in the Taipei metropolis. Data analyses of source contribution estimates indicated that industrial emission and oil combustion were more likely transported from distant areas, whereas biomass burning might come from multiple source origins. Additionally, the statistically significant difference of PM_2.5

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported in part by research grants from the Ministry of Science and Technology of Taiwan (MOST 106-2221-E-002-021-MY3 and 109-2119-M001–009-A) and the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education of Taiwan (NTU 109L9003).

References (61)

C.A. Belis et al.
Critical review and meta-analysis of ambient particulate matter source apportionment using receptor models in Europe
Atmos. Environ.
(2013)
M.L. Bell et al.
Evidence on vulnerability and susceptibility to health risks associated with short-term exposure to particulate matter: a systematic review and meta-analysis
Am. J. Epidemiol.
(2013)
S.G. Brown et al.
Methods for estimating uncertainty in PMF solutions: examples with ambient air and water quality data and guidance on reporting PMF results
Sci. Total Environ.
(2015)
M.-T. Cheng et al.
Particulate matter characteristics during agricultural waste burning in Taichung City, Taiwan
J. Hazard Mater.
(2009)
J.C. Chow et al.
The IMPROVE_A temperature protocol for thermal/optical carbon analysis: maintaining consistency with a long-term database
J. Air Waste Manag. Assoc.
(2007)
J.A. Cooper et al.
Receptor oriented methods of air particulate source apportionment
J. Air Pollut. Contr. Assoc.
(1980)
M. Eeftens et al.
Development of land use regression models for PM2.5, PM2.5 absorbance, PM10 and PMcoarse in 20 European study areas; results of the ESCAPE project
Environ. Sci. Technol.
(2012)
B. Gugamsetty et al.
Source characterization and apportionment of PM10, PM2.5 and PM0.1 by using positive matrix factorization
Aerosol and Air Quality Research
(2012)
D. Han et al.
Non-polar organic compounds in autumn and winter aerosols in a typical city of eastern China: size distribution and impact of gas–particle partitioning on PM2.5 source apportionment
Atmos. Chem. Phys.
(2018)
J.S. Han et al.
Size-resolved source apportionment of ambient particles by positive matrix factorization at Gosan background site in East Asia
Atmos. Chem. Phys.
(2006)

S.S. Ho et al.

Evaluation of an in-injection port thermal desorption-gas chromatography/mass spectrometry method for analysis of non-polar organic compounds in ambient aerosol samples

J. Chromatogr. A

(2008)

P.K. Hopke

Review of receptor modeling methods for source apportionment

J. Air Waste Manag. Assoc.

(2016)

P.K. Hopke et al.

The use of multivariate analysis to identify sources of selected elements in the Boston urban aerosol

Atmos. Environ.

(1976)

C.Y. Hsu et al.

Ambient PM2.5 in the residential area near industrial complexes: spatiotemporal variation, source apportionment, and health impact

Sci. Total Environ.

(2017)

S.C. Hsu et al.

Variations of Cd/Pb and Zn/Pb ratios in Taipei aerosols reflecting long-range transport or local pollution emissions

Sci. Total Environ.

(2005)

S. Janhall et al.

Vertical distribution of air pollutants at the gustavii cathedral in goteborg, Sweden

Atmos. Environ.

(2003)

M. Kalaiarasan et al.

Traffic-generated airborne particles in naturally ventilated multi-storey residential buildings of Singapore: vertical distribution and potential health risks

Build. Environ.

(2009)

I.G. Kavouras et al.

Iso- and anteiso-alkanes: specific tracers of environmental tobacco smoke in indoor and outdoor particle-size distributed urban aerosols

Environ. Sci. Technol.

(1998)

E. Kim et al.

Source identification of Atlanta aerosol by positive matrix factorization

J. Air Waste Manag. Assoc.

(2003)

K.H. Kim et al.

A review on the human health impact of airborne particulate matter

Environ. Int.

(2015)

M.J. Kleeman et al.

Lubricating oil and fuel contributions to particulate matter emissions from light-duty gasoline and heavy-duty diesel vehicles

Environ. Sci. Technol.

(2008)

E. Lee et al.

Application of positive matrix factorization in source apportionment of particulate pollutants in Hong Kong

Atmos. Environ.

(1999)

H.-T. Liao et al.

Vertical distribution of source apportioned PM2.5 using particulate-bound elements and polycyclic aromatic hydrocarbons in an urban area

J. Expo. Sci. Environ. Epidemiol.

(2020)

H.-T. Liao et al.

Source and risk apportionment of selected VOCs and PM2.5 species using partially constrained receptor models with multiple time resolution data

Environ. Pollut.

(2015)

H.-T. Liao et al.

Source apportionment of PM2.5 size distribution and composition data from multiple stationary sites using a mobile platform

Atmos. Res.

(2017)

H.-T. Liao et al.

Trajectory-assisted source apportionment of winter-time aerosol using semi-continuous measurements

Arch. Environ. Contam. Toxicol.

(2020)

H.T. Liao et al.

Source apportionment of urban air pollutants using constrained receptor models with a priori profile information

Environ. Pollut.

(2017)

L. Lin et al.

Review of recent advances in detection of organic markers in fine particulate matter and their use for source apportionment

J. Air Waste Manag. Assoc.

(2010)

F. Lu et al.

Systematic review and meta-analysis of the adverse health effects of ambient PM2.5 and PM10 pollution in the Chinese population

Environ. Res.

(2015)

L.L. Ming et al.

PM2.5 in the Yangtze River Delta, China: chemical compositions, seasonal variations, and regional pollution events

Environ. Pollut.

(2017)

Cited by (12)

Optical properties and potential sources of water-soluble and methanol-soluble organic aerosols in Taipei, Taiwan
2022, Atmospheric Environment
Citation Excerpt :
As seen in Fig. 6 (right), the CBPF was used to analyse the possible direction of sources of K+, which shows that the high concentrations of K+ primarily occurred under low wind speed. This implied that the light-absorbing substances were clearly contributed by the local sources, such as the activities of adjacent ports (Liao et al., 2021) or crematorium (Kulshrestha et al., 1998). There have been previous studies reporting the high correlation between K+ and OC in Kaohsiung and Manila ports (Tseng et al., 2021).
The optical properties and sources of brown carbon (BrC) have been poorly constrained in climate models due to the variability of spatiotemporal characteristics, impeding the accurate understanding of its impact on air quality and climate. In this study, daily PM_2.5 samples, which were collected from January to November 2021 in urban Taipei, Taiwan, and seasonal variations of optical properties of water-soluble and methanol-soluble organic carbon (WSOC and MSOC) were evaluated. The light absorption coefficients at 365 nm (Abs₃₆₅) of both extracts, which strongly correlated with WSOC and MSOC mass concentrations, displayed distinct seasonal variations with the highest in winter and the lowest in summer. The Absorption Ångström Exponent of WSOC and MSOC ranged from 4.16 to 7.75 and 4.03–6.83, with averages of 6.05 ± 0.56 and 5.29 ± 0.61, respectively. The mass absorption efficiency (MAE₃₆₅), which normalizes the Abs₃₆₅ of both extracts to the mass of WSOC and MSOC, showed significant seasonal difference with the high MAE_{365, WSOC} of 0.96 ± 0.29 m² g⁻¹ in winter and the lowest in summer of 0.49 ± 0.07 m² g⁻¹, whereas contrasting with the largest MAE_{365, MSOC} of 0.99 ± 0.46 m² g⁻¹ in summer and the lowest in winter of 0.66 ± 0.28 m² g⁻¹, respectively. Fossil fuel combustion, such as traffic emission, and biomass burning, such as crematorium, were identified to be important contributors to light-absorbing substances. The estimated fractional radiative forcing by WSOC and MSOC to elemental carbon was most significant during winter (8.15 ± 3.77%) and spring (13.90 ± 4.38%), respectively, which may greatly affect the atmospheric photochemistry and climate. This study suggests that the impact of BrC in Taiwan on the local and regional air quality and climate is non-negligible.
Characterization, source apportionment and associated health risk assessment of respirable air particulates in Metro Manila, Philippines
2022, Atmospheric Pollution Research
Citation Excerpt :
Smoke from biomass burning is the second major source of particulate pollution at the NAMRIA and Valenzuela sites. This observation was also noted in similar source apportionment studies where biomass burning is one of the most relevant sources of PM2.5 in cities (Scotto et al., 2021; Jiang et al., 2021; Liao et al., 2021). Biomass burning is mainly associated with burning biogenic materials.
Coarse (PM₁₀) and fine (PM_2.5) aerosols were sampled from June to November 2016 at three different sites in Metro Manila Philippines (Valenzuela, NAMRIA, and MMDA). PM_2.5 average levels in all sites exceeded the World Health Organization annual and daily limits, indicating that the ambient air qualities in all sites were unhealthy. Black Carbon (BC), a fingerprint of incomplete combustion, constituted 31–46% of PM_2.5. Multi-element analysis of PM indicated that the sulfur, lead, and zinc had substantial contributions to PM_2.5, with the highest level found in Valenzuela. Receptor modeling highlighted six important sources: vehicular emissions, biomass burning, industrials, secondary sulfur, sea spray, and fine soil. Among the sources, vehicular emission was the most dominant source, comprising 20–35% of the apportioned sources. An industrial source, predominantly with Zn and Pb, was seen in Valenzuela, contributing about 5.4% of its PM_2.5. The health risks of fine aerosol, BC, and major elements were evaluated also in this study. Elements studied showed negligible potential (HQ < 1) to cause non-carcinogenic health effects and very low (<1 × 10⁻⁶) carcinogenic health effects to children and adults. The observed trend for carcinogenic risks of PM_2.5 in children was Valenzuela > MMDA > NAMRIA while a medium carcinogenic risk was observed for adults at all sites (CR > 1 × 10⁻⁵). The trend for PM_2.5 average HQ values in both age groups was Valenzuela > MMDA > NAMRIA. Our results showcased varying levels and sources of PM and its components that can impact the health upon exposure and should be addressed in future policies to safeguard public health.
Molecular and elemental marker-based source apportionment of fine particulate matter at six sites in Hong Kong, China
2022, Science of the Total Environment
Citation Excerpt :
BS-DISP is the combination of BS and DISP by providing resampled matrices with displayed elements fij in factor profiles to simultaneously examine both random error and rotational ambiguity (Paatero et al., 2014) . The compositional data from all the sampling sites were combined for PMF analysis with the assumption that the sites were under the influence of a common set of pollution sources (Liao et al., 2021). This is reasonable considering the compact size of Hong Kong (ca. 30 × 30 km).
Source apportionment of PM_2.5 was performed using positive matrix factorization (PMF) based on chemical speciation data from 24-h filters collected throughout 2015 at six sampling sites of varying urban influences in Hong Kong. The input data include major inorganic ions, organic and elemental carbon, elements, and organic tracers. Nine factors were resolved, including (1) secondary sulfate formation process, (2) secondary nitrate formation process, (3) industrial emissions, (4) biomass burning, (5) primary biogenic emissions, (6) vehicle emissions, (7) residual oil combustion, (8) dust, and (9) aged sea salt. The PMF-resolved factor contributions in conjunction with air mass back trajectories showed that the two major sources for PM_2.5 mass, secondary sulfate (annual: 41%) and secondary nitrate (annual: 9.9%), were dominantly associated with regional and super-regional pollutant transport. Vehicular emissions are the most important local source, and its contributions exhibit a clear spatial variation pattern, with the highest (6.9 μg/m³, 24% of PM_2.5) at a downtown roadside location and the lowest (0.4 μg/m³, 2.0% PM_2.5) at two background sites away from city centers. The ability of producing a more reliable source separation and identifying new sources (e.g. primary biogenic source in this study) was a direct advantageous result of including organic tracers in the PMF analysis. PMF analysis conducted on the same dataset in this study but without including the organic tracers failed to separate the biomass burning emissions and industrial/coal combustion emissions. PMF analysis without the organic tracers would also over-apportion the contribution of vehicular emissions to PM_2.5, which would bias the evaluation of the effectiveness of vehicle-related control measures. This work demonstrates the importance of organic markers in achieving more comprehensive and less biased source apportionment results.
Composition, sources, and health risk assessment of particulate matter at two different elevations in Delhi city
2022, Atmospheric Pollution Research
Citation Excerpt :
The input datasets were obtained after pooling the samples from both the elevations together assuming that the chemical profiles of the sources are alike at both the elevations. This approach of pooling data together proved to improve the statistical significance of the results (Liao et al., 2021). The following criteria were used in obtaining the optimal solution: correlation coefficient between the modeled and measured PM concentration (r2), bootstrap mapping for the base run (generally greater than 80%), and changes in Q robust values (%dQ) when constraints are applied (generally less than <5%) (Almeida et al., 2020; Kim and Hopke, 2007).
The PM₁₀ and PM_2.5 source apportionment and health risk assessment were performed for two different elevations (lower elevation (LE) ∼5–10 m and higher elevation (HE) ∼30–45 m) at four different locations of Delhi city during January 2017–March 2017. The measured 24-h average PM₁₀ and PM_2.5 concentrations at different locations were found between 134.7 and 257.7 μg/m³ and 78.7–121.1 μg/m³, respectively. The 24-h average PM₁₀ and PM_2.5 concentrations at the study sites were exceeding the national (NAAQS) and WHO limits by more than 1.3 and 3 times, respectively. The PM mass was enriched with carbonaceous matter, ions, crustal and trace elements and their concentrations (except, Sr, Ni, S, As, V, Sb, Ga- the trace elements associated with coal and heavy oil combustion & NO₃⁻ - due to high nitrate formation ability at greater heights) were found higher in LE. The source apportionment study was performed with positive matrix factorisation (PMF). PMF analysis identified vehicular emission (PM_10, 9.6–24.3% & PM_2.5, 12.1–25.3%), secondary inorganics (PM₁₀, 6.4–22.5% & PM_2.5, 11–23%), crustal source (PM_10, 9–52% & PM_2.5, 3.8–10.7%), fuel oil combustion (PM_10, 3–21% & PM_2.5, 9–23%), biomass burning (PM_10, 7.4–28.6% & PM_2.5, 6–50.5%), and coal combustion (PM₁₀, 13–17% & PM_2.5, 14–19.1%) as the main sources of PM₁₀ and PM_2.5 at the study sites. A significant difference in source contribution between the elevations was observed for coal combustion, fuel oil combustion, biomass burning, and crustal sources. The contribution of vehicular emission and secondary inorganics estimates were broadly similar at both elevations. Coal and fuel oil combustion contribution was found relatively higher at HE. Further, health risk due to exposure to toxic heavy metals in PM_2.5 was assessed. Non-carcinogenic and carcinogenic risks (averaged of four sites) for both children and adults were exceeding the acceptable limit by more than 1.13 times. Results also showed that the public residing at HE is more susceptible to have greater health risks than LE at Delhi city. Sources contribution to carcinogenic risk was assessed and the results indicated that coal combustion (41%) contribution was highest followed by crustal sources (22%), fuel oil combustion (17%), vehicular emission (12%), biomass burning (4%) and secondary inorganics (4%).
Identifying low-PM<inf>2.5</inf> exposure commuting routes for cyclists through modeling with the random forest algorithm based on low-cost sensor measurements in three Asian cities
2022, Environmental Pollution
Citation Excerpt :
The downtown area of each city, ranging between 26.5 and 46.0 km2, served as the study area. In Taipei, the sources of PM2.5 included vehicular emissions, transported regional, secondary pollution sources, industrial emissions, biomass burning, oil combustion, and soil resuspension (Kuo et al., 2014; Liao et al., 2021). Secondary pollution, vehicular emissions, industrial emissions, heavy oil combustion, soil and dust resuspension, as well as incinerators, were identified as the fine particle sources in Osaka (Funasaka et al., 2020).
Cyclists can be easily exposed to traffic-related pollutants due to riding on or close to the road during commuting in cities. PM_2.5 has been identified as one of the major pollutants emitted by vehicles and associated with cardiopulmonary and respiratory diseases. As routing has been suggested to reduce the exposures for cyclists, in this study, PM_2.5 was monitored with low-cost sensors during commuting periods to develop models for identifying low exposure routes in three Asian cities: Taipei, Osaka, and Seoul. The models for mapping the PM_2.5 in the cities were developed by employing the random forest algorithm in a two-stage modeling approach. The land use features to explain spatial variation of PM_2.5 were obtained from the open-source land use database, OpenStreetMap. The total length of the monitoring routes ranged from 101.36 to 148.22 km and the average PM_2.5 ranged from 13.51 to 15.40 μg/m³ among the cities. The two-stage models had the standard k-fold cross-validation (CV) R² of 0.93, 0.74, and 0.84 in Taipei, Osaka, and Seoul, respectively. To address spatial autocorrelation, a spatial cross-validation approach applying a distance restriction of 100 m between the model training and testing data was employed. The over-optimistic estimates on the predictions were thus prevented, showing model CV-R² of 0.91, 0.67, and 0.78 respectively in Taipei, Osaka, and Seoul. The comparisons between the shortest-distance and lowest-exposure routes showed that the largest percentage of reduced averaged PM_2.5 exposure could reach 32.1% with the distance increases by 37.8%. Given the findings in this study, routing behavior should be encouraged. With the daily commuting trips expanded, the cumulative effect may become significant on the chronic exposures over time. Therefore, a route planning tool for reducing the exposures shall be developed and promoted to the public.
Vertical variation of source-apportioned PM<inf>2.5</inf> and selected volatile organic compounds near an elevated expressway in an urban area
2024, Environmental Science and Pollution Research

View all citing articles on Scopus

: Peer review under responsibility of Turkish National Committee for Air Pollution Research and Control.

View full text

Source apportionment of urban PM2.5 using positive matrix factorization with vertically distributed measurements of trace elements and nonpolar organic compounds

Highlights

Abstract

Introduction

Section snippets

Field campaigns

Source apportionment

Conclusions

Declaration of competing interest

Acknowledgments

Critical review and meta-analysis of ambient particulate matter source apportionment using receptor models in Europe

Atmos. Environ.

Evidence on vulnerability and susceptibility to health risks associated with short-term exposure to particulate matter: a systematic review and meta-analysis

Am. J. Epidemiol.

Methods for estimating uncertainty in PMF solutions: examples with ambient air and water quality data and guidance on reporting PMF results

Sci. Total Environ.

Particulate matter characteristics during agricultural waste burning in Taichung City, Taiwan

J. Hazard Mater.

The IMPROVE_A temperature protocol for thermal/optical carbon analysis: maintaining consistency with a long-term database

J. Air Waste Manag. Assoc.

Receptor oriented methods of air particulate source apportionment

J. Air Pollut. Contr. Assoc.

Development of land use regression models for PM2.5, PM2.5 absorbance, PM10 and PMcoarse in 20 European study areas; results of the ESCAPE project

Environ. Sci. Technol.

Source characterization and apportionment of PM10, PM2.5 and PM0.1 by using positive matrix factorization

Aerosol and Air Quality Research

Non-polar organic compounds in autumn and winter aerosols in a typical city of eastern China: size distribution and impact of gas–particle partitioning on PM2.5 source apportionment

Atmos. Chem. Phys.

Size-resolved source apportionment of ambient particles by positive matrix factorization at Gosan background site in East Asia

Atmos. Chem. Phys.

Evaluation of an in-injection port thermal desorption-gas chromatography/mass spectrometry method for analysis of non-polar organic compounds in ambient aerosol samples

J. Chromatogr. A

Review of receptor modeling methods for source apportionment

J. Air Waste Manag. Assoc.

The use of multivariate analysis to identify sources of selected elements in the Boston urban aerosol

Atmos. Environ.

Ambient PM2.5 in the residential area near industrial complexes: spatiotemporal variation, source apportionment, and health impact

Sci. Total Environ.

Variations of Cd/Pb and Zn/Pb ratios in Taipei aerosols reflecting long-range transport or local pollution emissions

Sci. Total Environ.

Vertical distribution of air pollutants at the gustavii cathedral in goteborg, Sweden

Atmos. Environ.

Traffic-generated airborne particles in naturally ventilated multi-storey residential buildings of Singapore: vertical distribution and potential health risks

Build. Environ.

Iso- and anteiso-alkanes: specific tracers of environmental tobacco smoke in indoor and outdoor particle-size distributed urban aerosols

Environ. Sci. Technol.

Source identification of Atlanta aerosol by positive matrix factorization

J. Air Waste Manag. Assoc.

A review on the human health impact of airborne particulate matter

Environ. Int.

Lubricating oil and fuel contributions to particulate matter emissions from light-duty gasoline and heavy-duty diesel vehicles

Environ. Sci. Technol.

Application of positive matrix factorization in source apportionment of particulate pollutants in Hong Kong

Atmos. Environ.

Vertical distribution of source apportioned PM2.5 using particulate-bound elements and polycyclic aromatic hydrocarbons in an urban area

J. Expo. Sci. Environ. Epidemiol.

Source and risk apportionment of selected VOCs and PM2.5 species using partially constrained receptor models with multiple time resolution data

Environ. Pollut.

Source apportionment of PM2.5 size distribution and composition data from multiple stationary sites using a mobile platform

Atmos. Res.

Trajectory-assisted source apportionment of winter-time aerosol using semi-continuous measurements

Arch. Environ. Contam. Toxicol.

Source apportionment of urban air pollutants using constrained receptor models with a priori profile information

Environ. Pollut.

Review of recent advances in detection of organic markers in fine particulate matter and their use for source apportionment

J. Air Waste Manag. Assoc.

Systematic review and meta-analysis of the adverse health effects of ambient PM2.5 and PM10 pollution in the Chinese population

Environ. Res.

PM2.5 in the Yangtze River Delta, China: chemical compositions, seasonal variations, and regional pollution events

Environ. Pollut.

Source apportionment of urban PM_2.5 using positive matrix factorization with vertically distributed measurements of trace elements and nonpolar organic compounds