Estimation of the fraction of soil-borne particulates in indoor air by PMF and its impact on health risk assessment of soil contamination in Guangzhou, China☆
Graphical abstract
Introduction
Health risk assessment of soil contamination of land is a necessary tool of environmental management. The main exposure routes for non-volatile contaminants in soil of land are oral ingestion of soil (Ois), dermal contact with soil (Dcs) and inhalation of soil particles (Pis). Among them, Pis is the human respiratory system exposure to soil contaminants through inhalation of air particulates, and might cause potential damage (Kim et al., 2015). Even if the windows and doors are closed, air particulates carrying soil pollutants can enter the room as long as the wind speed exceeds 0.45 m s−1 (Zhang et al., 2018). Although modern concerns are more focused on fine particulate matter (PM2.5) than coarse particulate matter (PM10) (Brown et al., 2013), the PM10 dominate the portion of soil-borne particulates and the PM2.5 just has the low proportion (Harrison et al., 2000; Hwang et al., 2008; Mooibroek et al., 2011). Therefore, PM10, rather than PM2.5, was addressed in all the health risk assessment models of soil contamination of land. The fraction of soil-borne particulates in indoor air PM10 (fspi) is one of the main exposure parameters to estimate the inhalation average daily dose (ADD).
Various environmental exposure models have been developed to assess the health risk of soil contaminants in land, including the Risk-Based Corrective Action (RBCA) model (ASTM, 2020), the Integrated Exposure Uptake Biokinetic (IEUBK) model (U.S.EPA, 1998), the Contaminated Land Exposure Assessment (CLEA) model (Defra, 2002), the CSOIL model (Brand et al., 2007), and the C-RAG model (China MEE, 2019). Among these models, IEUBK, CSOIL and C-RAG models used fspi to calculate the total dose of indoor PM10 from soil inhaled by humans. Admittedly, the accuracy of models is susceptible to toxicity parameters, physical and chemical parameters of contaminant, and exposure parameters. Compared to the other parameters, the exposure parameters such as fspi might result in more uncertainty for the assessment of health risk as fspi may vary depending on the climate, seasonal surface, soil characteristic and housing designs in different regions (U.S.EPA, 2011; Wang et al., 2012). Thus, only with local fspi can ADD be accurately estimated.
In the CSOIL model, the recommended value of fspi was 80% obtained by the method of indoor dust Pb isotopes (Hawley, 1985; Van den Berg, 1994; Swartjes, 2003). The U.S. Environmental Protection Agency (EPA) estimated fspi at 70% in the IEUBK model by analyzing Pb in indoor dust and outdoor adjacent soil (U.S.EPA, 1998). China national environmental protection standards (HJ 25.3–2019) “Technical guidelines for risk assessment of soil contamination of land for construction” also recommended fspi of 80% in C-RAG model. The methods of estimation fspi by Pb isotopes in the CSOIL model and the IEUBK model need high requirements on experimental instruments and cost of sample treatment. Except for soil sources, coal burning and vehicle exhaust also have significant contribution to Pb-210 content in particulate matter (Church et al., 2008; Mattsson et al., 1988; M. Baskaran et al., 2011). All the relevant isotope characteristic values of samples from various emission sources need to be investigated (Zheng et al., 2004). Thus, the methods of fspi estimation by Pb isotopes are very complicated. Looking for a more convenient method to estimate the local fspi in different areas would be helpful to the health risk assessment of soil contaminants.
Local fspi might be estimated by source apportionment which quantifies factor (source) contributions with chemical and statistical methods. Positive matrix factor analysis (PMF) and principal component analysis (PCA) are extensively applied in the source apportionment of particulate matter (PM). PMF could quantify the main factors and contributions with the weight among chemical components of particulate matter and the least square method based on assuming mass conservation and chemical mass balance between emission sources and receptors (Wu et al., 2020). Suryawanshi et al. (2016) illustrated by PMF that the contribution of soil to PM0.6 amounted to 17.5% in India classrooms. Bari et al. (2015) obtained a 16% fraction of soil to residential PM1 by PMF in Canada. Taghvaee et al. (2018) indicated that the fractions of soil to PM2.5 were 2.5% and 8% for some nursing homes and dormitories respectively in Iran. In Xi'an, China, some indoor factories witnessed the 5.4% soil of PM2.5 by PMF (Dai et al., 2019). By PMF some campus indoors had 4.4%, 6% and 18% PM2.5 originated from soil in the northeastern Nanjing China, United States and Seoul Korea respectively (Heo et al., 2021; Carrion-Matta et al., 2019; Niu et al., 2021). The contribution of soil to residential PM2.5 with PMF in Pittsburgh, Massachusetts, Hubei and Shanghai, reached 5%, 17%, 17% and 25%, respectively (Tunno et al., (2016); Matthaios et al., (2021); Fang et al., (2021); Brehmer et al., (2019)). By PCA, Wang et al. (2018) investigated that the hall of residence had 11.26% natural source to PM2.5 in Nanjing China. However, all the above studies just concerned on indoor or outdoor air fine particulates (≤PM2.5). For PM10, the contribution of soil to outdoor PM10 was estimated as 30%–40% by Cheung et al. (2011) and Ho et al. (2003) using outdoor air receptor model. Very few studies investigated the fraction of soil-borne to indoor PM10 by source apportionment except for Ali et al. (2017), who reported a total of 20% of indoor PM10 from soil in campuses, Malaysia by PCA. And evaluating the contribution of soil to indoor air quality would be different from conventional outdoor air receptor model in methodology. Meanwhile, very few studies used PMF to localize fspi for health risk assessment of soil.
In Guangzhou, a megacity with a population of 18.67 million, about 75% of the total population reside in the main urban areas. 36.56% brownfields in the total ninety-three lands investigated from 2013 to 2018 were found soil contaminated over there (Tan et al., 2021). The accuracy of health risk assessment for soil contamination of land is strongly concerned. Therefore, this study will try to apply PMF methodology to estimate the fraction of soil-borne to indoor PM10 in Guangzhou main urban areas, and supply a new method for investigation of local fspi value in the health risk assessment of soil contamination of land. The specific objectives of this study were: (1) to determine the concentrations of chemical components such as metal elements, anions, organic carbon (OC) and elemental carbon (EC) in indoor ambient PM10 samples collected from Guangzhou main urban areas, China; (2) to characterize the sources of indoor PM10 and the fraction of soil-borne to indoor PM10 by source apportionment; (3) to survey the differences between the health risk results calculated with local fspi value and the recommended value by the Chinese C-RAG model.
Section snippets
Sampling strategy and chemical analysis
Sampling areas were set up in different directions and positions of Guangzhou main urban areas, including nine areas (XC, JN, HB, ZX, ZB, HN, YS, HE and ZK) (Fig. 1). Two to five sites were selected for sampling in each area. There was a total of 27 indoor of low-rise floors. The sampling rooms were few of PM10 indoor sources such as smoking, cooking, printer, electrical equipment and newly painted furniture and decorations. The doors were closed except personnel access, and about a third of
Concentrations of aerosol PM10 compositions
The concentrations of PM10 and their compositions were illustrated in Fig. 2. And their mean, max and min values were shown in Table S1. Daily average indoor PM10 concentration was 85.39 μg m−3, which was higher than the standard of WHO daily average indoor PM10 concentration 50 μg m−3 (WHO, 2019) and lower than the limit of Chinese daily average indoor PM10 concentration 150 μg m−3 (China MHP, 2002). The most dominant elements in indoor PM10 samples were Na, Al, Fe, K and Cu. The
Conclusions
Based on sampling and analyzing indoor ambient PM10 samples, a main local exposure parameter (fspi) of health risk assessment was successfully estimated by PMF method. The methodology was the first time to apply in health risk assessment for soil contamination of land, and was more convenient to be used than the existing method of Pb isotopes. The results showed that the primary sources of indoor PM10 in Guangzhou main urban areas were combustion sources (50.53%), vehicular sources (28.17%) and
Credit author statement
Zi-Jie Xu: Investigation; Data curation; Model running; Visualization; Writing – original draft; Huan-Bin Zhu: Investigation; Methodology; Li-Yun Shu: Investigation; Data curation; Xiao-Xia Lai: Investigation; Wei Lu: Investigation; Lei Fu: Investigation; Bin Jiang: Methodology; Tao He: Validation; Investigation; Fo-Peng Wang: Methodology; Investigation; Qu-Sheng Li: Writing – review & editing; Visualization; Conceptualization; Methodology; Supervision; Funding acquisition.
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.
Acknowledgement
This works was supported by the National Natural Science Foundation of China (No. 41673094 and 41977265).
References (76)
Quantitative source apportionment and human toxicity of indoor trace metals at university buildings
Build. Environ.
(2017)Po-210 and Pb-210 as atmospheric tracers and global atmospheric Pb-210 fallout: a Review
J. Environ. Radioact.
(2011)Investigation of sources of atmospheric aerosol at urban and semi-urban areas in Bangladesh
Atmos. Environ.
(2004)The impact of household air cleaners on the chemical composition and children's exposure to PM2.5 metal sources in suburban Shanghai
Environ. Pollut.
(2019)Spatial and temporal variation of chemical composition and mass closure of ambient coarse particulate matter (pm 10-2.5) in the Los Angeles area
Atmos. Environ.
(2011)Residential coal combustion as a source of primary sulfate in Xi'an, China
Atmos. Environ.
(2019)- et al.
Leaching of zinc from tire wear particles under simulated estuarine conditions
Chemosphere
(2011) Chemical characteristics of PM10 during the summer in the mega-city Guangzhou, China
Atmos. Res.
(2014)Particulate matter in the atmosphere: which particle properties are important for its effects on health
Sci. Total Environ.
(2000)Characterization of pm10 and pm2.5 source profiles for fugitive dust in Hong Kong
Atmos. Environ.
(2003)
A review on the human health impact of airborne particulate matter
Environ. Int.
Source apportionment of traffic emissions of particulate matter using tunnel measurements
Atmos. Environ.
Heavy metals and Pb isotopic composition of aerosols in urban and suburban areas of Hong Kong and Guangzhou, South China-evidence of the long-range transport of air contaminants
Atmos. Environ.
Characterization and source apportionment of PM2.5 based on error estimation from EPA PMF 5.0 model at a medium city in China
Environ. Pollut.
Characterisation of PM10 and PM2.5 particulate matter in the ambient air of Milan (Italy)
Atmos. Environ.
Source apportionment of PM10 and PM2.5 in Milan (Italy) using receptor modelling
Sci. Total Environ.
210Pb and 222Rn as guides in adjudicating SO42- and SO2 air concentrations sulphate in the air in Finland 1962. 1985
Sci. Total Environ.
Source apportionment and spatial variability of PM2.5 using measurements at five sites in The Netherlands
Atmos. Environ.
Combustion sources of particles. 1. Health relevance and source signatures
Chemosphere
Chemical characterization and source apportionment of PM10 and PM2.5 in the metropolitan area of Costa Rica, Central America
Atmos. Pollut. Res.
Speciation and origin of PM10 and PM2.5 in Spain
J. Aerosol Sci.
Comparative PM10-PM2.5 source contribution study at rural, urban and industrial sites during PM episodes in Eastern Spain
Sci. Total Environ.
Source apportionment of PM10 and PM2.5 air pollution, and possible impacts of study characteristics in South Korea
Environ. Pollut.
Characterization of carbonaceous aerosols over the urban tropical location and a new approach to evaluate their climatic importance
Atmos. Environ.
Identification and quantification of indoor air pollutant sources within a residential academic campus
Sci. Total Environ.
Source apportionment of ambient PM2.5 in two locations in central Tehran using the Positive Matrix Factorization (PMF) model
Sci. Total Environ.
Mechanisms of fine particulates formation with alkali metal compounds during coal combustion
Fuel
Indoor source apportionment in urban communities near industrial sites
Atmos. Environ.
Characterisation of traffic-generated particulate matter in Copenhagen
Atmos. Environ.
Heavy metal characteristics and health risk assessment of PM2.5 in three residential homes during winter in Nanjing, China. Build
Environ. Times
Verification of traffic emitted aerosol components in the ambient air of Cologne (Germany)
Atmos. Environ.
A partition computing-based positive matrix factorization (PC-PMF) approach for the source apportionment of agricultural soil heavy metal contents and associated health risks
J. Hazard Mater.
Effects of air pollution control measures on air quality improvement in Guangzhou, China
J. Environ. Manag.
Risk assessment and source analysis of soil heavy metal pollution from lower reaches of Yellow River irrigation in China
Sci. Total Environ.
Source apportionment and analysis on ambient and personal exposure samples with a combined receptor model and an adaptive blank estimation strategy
Atmos. Environ.
Characteristics of lead isotope ratios and elemental concentrations in PM10 fraction of airborne particulate matter in Shanghai after the phase-out of leaded gasoline
Atmos. Environ.
Standard Guide for Risk-Based Corrective Action for Contaminated Sediment Sites, E3240
Indoor and outdoor levels and sources of submicron particles (PM1) at homes in Edmonton, Canada
Environ. Sci. Technol.
Cited by (7)
Enhancing the power generation performance of photovoltaic system: Impact of environmental and system factors
2024, Applied Thermal EngineeringSource Analysis Based on the Positive Matrix Factorization Models and Risk Assessment of Heavy Metals in Agricultural Soil
2023, Sustainability (Switzerland)Distribution and Source Resolution of Heavy Metals in an Electroplating Site and Their Health Risk Assessment
2023, Sustainability (Switzerland)
- ☆
This paper has been recommended for acceptance by Dr. Yong Sik Ok.