Evaluation of Variability in the Ambient PM2.5 Concentrations from FEM and FRM-like Measurements for Exposure Estimates

Received: May 12, 2020
Revised: September 28, 2020
Accepted: October 28, 2020

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Cite this article:

Mou, C.Y., Hsu, C.Y., Chen, M.J., Chen, Y.C. (2021). Evaluation of Variability in the Ambient PM_2.5 Concentrations from FEM and FRM-like Measurements for Exposure Estimates. Aerosol Air Qual. Res. 21, 200217. https://doi.org/10.4209/aaqr.2020.05.0217

HIGHLIGHTS

• FRM-like PM_2.5 with negligible between-site variability was found in the Taipei City.
• Daily FEM PM_2.5 were dominated by the within-site variability.
• Ambient PM_2.5 was mainly affected by the gaseous pollutants.
• SO₂ was significantly correlated with daily and annual PM_2.5.

Keywords: Ambient PM2.5; Calibration; Within and between variability; Exposure estimates.

1 INTRODUCTION

The elevated levels of fine particulate matter (PM_2.5) are associated with adverse health effects, such as respiratory and cardiovascular morbidity and mortality, which have been reported in many studies (Beelen et al., 2014; Cai et al., 2016; Kaufman et al., 2016). The World Health Organization (WHO, 2014) has indicated that outdoor PM_2.5 accounts for 7 million deaths worldwide every year. Recent literature has documented that exposure to even low levels of PM_2.5 can significantly increase all-cause mortality (Shi et al., 2016). National ambient air quality standards (NAAQS) are usually regulated in many countries to limit air pollution (such as ambient PM_2.5 levels) and protect public health. For instance, the annual (15 µg m^–3) and daily (35 µg m^–3) standards of PM_2.5 have been set by Taiwan EPA while lower guideline limits for outdoor PM_2.5 (annual = 10 µg m^–3, daily = 25 µg m^–3) have been suggested by WHO. The filter-based instruments such as BGI PQ200 and RAAS2.5 single and multi-day samplers are usually used for consistent and repeatable measurements of 24-h PM_2.5 concentrations following the federal reference method (FRM) certified by USEPA (McNamara et al., 2011; Kelly et al., 2017). In addition, the continuous monitoring measurements following the federal equivalent method (FEM) or non-FRM based on optical, beta ray attenuation, or tapered element oscillation microbalance (TEOM) monitoring have also been widely used for timely PM_2.5 measurements in the metropolitan area. In Taiwan, hourly concentrations of FEM PM_2.5 from 76 air quality monitoring stations (AQMSs) of Taiwan EPA (https://data.epa.gov.tw/dataset/aqx_p_15) have been announced since 2006. FEM PM_2.5 concentrations obtained via optical measurements are usually affected by humidity, temperature, size distribution, and chemical composition (Bortnick et al., 2002; Dinoi et al., 2017; Sofowote et al., 2014). Thus, the calibration of the FEM PM_2.5 data with the FRM PM_2.5 measurements through a statistical linear regression model (so-called FRM-like PM_2.5) is necessary. Bortnick et al. (2002) indicated that the resulting FRM-like PM_2.5 measurements using a linear regression model were able to provide more timely reporting of PM_2.5. The calibration approach of the data quality objective (DQO) process consists of a seven-step strategy, as suggested by U.S. EPA (U.S. EPA, 1994).

The FRM-like measurements of PM_2.5 are performed to not only provide a comprehensive assessment of air quality for the public but also to investigate its impact on human health. Numerous population-based epidemiological studies of air pollution usually rely on FRM-like PM_2.5 measurements either by direct methods or using geographical modeling, i.e., microenvironmental exposure, land use regression (LUR), and kriging models to estimate PM_2.5 exposure (Koenig et al., 2005; Meng et al., 2005; Laden et al., 2006; Krewski et al., 2009; Jerrett et al., 2013; Özkaynak et al., 2013; Kioumourtzoglou et al., 2014). However, it is not clear whether the PM_2.5 estimates from FRM-like measurements through a calibration process attenuate exposure variability or lead to non-differential misclassification of exposure for population-based health studies. For instance, a model could result in misleading PM_2.5 concentrations, when it is mainly affected by meteorological conditions such as temperature (Bortnick et al., 2002). The exposure error obtained from ambient PM_2.5 measurements can impact observed health risks, potentially distorting associations and interactions between covariates and outcomes, and leading to invalid inferences. To date, no study has explored the applicability of uncalibrated FEM and calibrated FRM-like measurements of PM_2.5 for exposure estimates in health effect studies. On the other hand, a large number of studies have examined the effect of meteorological conditions and gaseous pollutants on PM_2.5 concentrations (Ito et al., 2007; Zhang et al., 2015). These potential confounders, such as temperature, NO₂, and O_3, are usually included for adjustment in the health effect models (Crouse et al., 2015; Luo et al., 2016). Thus, it is important to understand the magnitude of the impact of these factors on within- and between-site (or -group) variability in concentrations at different time intervals. This information is vital for the future design of studies to improve exposure estimates associated with mortality and morbidity outcomes.

This study aims to evaluate PM_2.5 variability in FRM-like and FEM measurements and identify the important factors affecting within- and between-site (-group) variability in PM_2.5 concentrations in the metropolitan area. This study, a part of the Taiwan Health and Air Pollution study (THAP), can make an effort on the improvement of exposure estimates for population-based health risk analysis.

 
2 METHODS

2.1 Data Sources and Calibration Process

In this study, we selected Taipei and New Taipei cities, (together known as Big Taipei City), having a population of 7 million in 2,225 km² of land, as our study area because it has a high-density of national AQMSs. There are 19 national AQMSs (Fig. 1), including one national park AQMS (Yangming mountain), three traffic AQMS (Sanchong, Yonghe, and Datong), one background AQMS (FugueiCape), and 14 general AQMSs (Tu-cheng (TC), Shi-lin (SL), Zhong-shan (ZS), Gu-ting (GT), Xi-zhi (XiZ), Song-shan (SS), Ban-qiao (BQ), Lin-kou (LK), Tam-sui (TS), Cai-liao (CL), Xin-dian (XD), Xin-zhuang (XinZ), Wanli (WL), and Wan-hua (WH)). Out of these 14 general stations, one is the background station (WL) for the specific purpose of air quality monitoring. After excluding national park and background monitoring stations, the air quality monitoring data from a total of 13 general stations (except WL), and 3 traffic stations in Big Taipei City along with the meteorological observations and site specifications were included for analysis.

Fig. 1. The location of AQMSs in Taipei and New Taipei Cities (SC (San-chong); TC (Tu-cheng); SL (Shi-lin); DT (Da-tong); ZS (Zhong-shan); GT (Gu-ting); YH (Yong-he); XiZ (Xi-zhi); SS (Song-shan); BQ (Ban-qiao); LK (Lin-kou); TS (Tam-sui); CL (Cai-liao); XD (Xin-dian); XinZ (Xin-zhuang); WH (Wan-hua); WL (Wanli); YM (Yaming Nantional Park); FC (FugueiCape)).

Briefly, the FRM calibration procedure has been implemented by the contractors of Taiwan EPA across Taiwan since 2014. Due to limited resources, filter-based FRM measurements at only 5 AQMSs in big Taipei were conducted once every three days. In the analysis, 16 AQMSs shared five FRM measurements of PM_2.5 simultaneously. As a result, 16 resultant calibration equations by the station were generated for every year (Tables S1–S4).

The regression model for the calibration of FRM-like PM_2.5 measurements from 2014 to 2017 in Big Taipei City is shown in the supplementary data. The measurement data, including ambient temperature, relative humidity, wind speed, rainfall, CO, NO, NO₂, NO_x, O₃, SO₂, CH₄, Non-methane hydrocarbon (NMHC), and total hydrocarbon (THC), were obtained from the Taiwan Air Quality Monitoring Network database for 16 stations. Station profiles such as the station types (ambient/traffic), the height of sampling ports (3.5–13.5 m/17.5 m/19.5–21.5 m), the distance to the nearest main road (1–5.6 m/10–15 m/20–100 m), the types of automatic sampling instruments (VEREWA F701/MetOne 1020) were also collected. Moreover, we incorporated the concurrent atmospheric pressure measurements obtained from the weather bureau stations selected based on the distance to the nearest AQMS. As a lot of meteorological information from AQMSs was missing, all the missing data were imputed by the concurrent measurements from the nearest weather bureau station.

 
2.2 Statistical Analysis

The SPSS software was used to evaluate temporal (within-site) and spatial (between-site) variability in PM_2.5 concentrations from FEM and FRM-like measurements at 16 stations. The temporal variation of PM_2.5 is defined as the variability in PM_2.5 over time at each AQMS, whereas the spatial variation is defined as the variability in PM_2.5 across sites at any time. Both FEM and FRM-like PM_2.5 concentrations had skewed distribution and were normalized by logarithmic transformation. The measurements at daily and yearly time resolutions were evaluated to explore the trends in spatial and temporal variations at short-term and long-term scales.

Pearson’s correlation coefficient (r) was used to evaluate the correlations between the variables. Among the variable pairs having r > 0.7, only the determinants considered more logically related to PM_2.5 and/or more consistently and reliably available throughout the duration of the study across selected AQMSs were retained for modeling. The mixed-effects model based on a restricted maximum likelihood estimation procedure was used to examine the relationship between each variable and the log-transformed daily and annual PM_2.5. The determinants were treated as fixed effects in the model. The AQMSs were treated as random effects to account for potential correlation within repeated measurements at the same AQMS. For PM_2.5, the mixed-effects model was specified as (Peretz et al., 2002):

For i = 1, …, k (AQMSs) and j = 1, …, n_i (the repeated day or year of the i^th AQMS), where Y_ij is the log-transformed PM_2.5 concentration; β₀ is an overall intercept for the study area that corresponds to mean background PM_2.5 (log-transformed) when all factors equal zero; β₁, …, β_p are fixed effects; X_ij1, …, X_ijp are values of the variables for the i^th AQMS on the j^th day or year; b₁, …, b_k are AQMSs’ random effects; b_i is the i^th AQMS random effect, which corresponds to the discrepancy between its intercept and the group intercept β₀; and z₁, …, z_k are AQMSs’ indicators. ε_ij is the residual error associated with i^th AQMS on the j^th day or year. b_i and ε_ij were assumed to be independent and normally distributed, with a mean of 0, and the variances of σ²_B (between-site) and σ²_w (within-site), respectively.

Exponential beta [exp(β)] value of determinants strengthens the correlation of variables with PM_2.5 in the model. The statistical significance was set at p < 0.05 based on a two-tailed analysis. We created univariate and multivariate models to evaluate the selective variables statistically affecting PM_2.5. The fitness of each mixed-effects model with selected variables was estimated by the Akaike information criterion (AIC). A lower AIC indicates a better model fit.

 
3 RESULTS AND DISCUSSION

3.1 PM_2.5 Concentrations from the Calibrated and Uncalibrated Measurements

Table 1 shows the descriptive statistics of PM_2.5 concentrations from the FRM-like (calibrated) and FEM (uncalibrated) measurements in the Big Taipei City at 16 AQMSs during 2014–2017. The annual mean concentrations of FRM-like PM_2.5 were 21.5, 18.5, 17.5, and 16.6 µg m^–3 in 2014, 2015, 2016, and 2017, respectively, with all the concentrations exceeding the annual standard (15.0 µg m^–3) set by Taiwan EPA. In general, the mean PM_2.5 concentrations from the FEM measurements were approximately 30% significantly (p < 0.001) higher than that from the FRM-like measurements at all the stations during 2014–2017. The ZS and LK stations had the highest mean PM_2.5 concentrations from the FEM and FRM-like measurements, respectively, throughout the duration of the study. The traffic stations, including SC, DT, and YH, did not show higher PM_2.5 concentrations compared with the concentrations observed at the ambient stations. As shown in Fig. S1, the annual mean concentrations of PM_2.5 from both the FEM and FRM-like measurements significantly decreased by approximately 22% from 2014 to 2017 at all the AQMSs. The improvement in the air quality in terms of PM_2.5 is likely attributed to several effective policies implemented by the Taiwan EPA and Environmental Protection Bureau of Taipei. Few of the policies implemented include eliminating two-stroke motorcycles and old-generation diesel vehicles, promoting electric buses and electric motorbikes, emission control measures for the catering industry, stricter standards for boiler emissions, and implementing low-sulfur jet fuel. As shown in Fig. S1, the PM_2.5 concentrations at CL (31.9 µg m^–3 in 2014 to 20.5 µg m^–3 in 2017) from FEM measurements and at XinZ (23.3 µg m^–3 in 2014 to 14.0 µg m^–3 in 2017) from FRM-like measurements drastically reduced by 35–40% owing to the operation of Taipei mass rapid transit (MRT) for the Xinzhuang reduced by 35–40% owing to the operation of Taipei mass rapid transit (MRT) for the Xinzhuang line in 2014. In addition, the lowest decline in PM_2.5 concentrations was observed at LK (12.5%) and XD (7%) from the FEM and FRM-like measurements, respectively. The FEM and FRM-like measurements presenting such a dissimilarity in the reduction of PM_2.5 levels measured at the stations may mask information on the effectiveness of region-specific ambient air quality management.

3.2 Variations in PM_2.5 Concentrations from the FEM and FRM-like Measurements

Table 2 shows σ_B², σ_W², and total variance (σ_Total²) of daily and annual FEM and FRM-like PM_2.5 concentrations at 16 AQMSs during 2014–2017. The daily FEM PM_2.5 concentrations were dominated by the within-site variability (> 90% of σ_Total²) rather than the between-site variability. When the daily FRM-like data was analyzed in the null model (all factors of fixed effects equal zero), the between-site variability (σ_B² = 0.004; 1.29% of σ_Total²) in PM_2.5 concentrations decreased by ~8% of total variance compared with FEM data (σ_B² = 0.027; 9.84% of σ_Total²) as shown in Table 2. The increase of σ_W² (=0.339; 98.7% of σ_Total²) in FRM-like PM_2.5 could be also observed compared with that in FEM PM_2.5 (σ_W² = 0.245; 90.2% of σ_Total²).

While fitting such linear regression models, the calibration procedure limits the inherent variation in PM_2.5 concentrations obtained from the AQMSs at various locations, which may reduce spatial variation in PM_2.5 concentrations. Basically, the FRM measurements of PM_2.5 at five AQMSs (SL, WH, XZ, BQ, and Taoyuan) were used to calibrate single or multiple FEM measurements at the corresponding AQMSs through linear regression models and coefficient of determination (R², mean = 0.92, range = 0.83–0.97) during 2014–2017 (Tables S1–S4). For instance, the FRM measurement at one station (WH) was shared by the FEM measurements at a maximum of 8 AQMSs for calibration, where R² ranged from 0.86 to 0.97 in 2014 (Table S1). When we compared daily variations in PM_2.5 concentrations from the FRM and FEM measurements during 2014–2017, σ_Total² and coefficient of variation (CV) and σ_W² of PM_2.5 concentrations from the FRM measurements were larger than those from the FEM measurements (Table S5). Hence, it can be stated that the calibration approach of FRM-like PM_2.5 measurements directly decreases the spatial heterogeneity in PM_2.5 concentrations across sites.

For annual PM_2.5, the within-site (σ_W² = 0.016) and between-site (σ_B² = 0.014) variance in the FEM measurements were similar. However, the between-site variance in the FRM-like measurements became negligible (σ_B² = 0%) after the calibration process. The within-site variation (σ_W² = 0.245, accounting for 90.2% of σ_Total²) in daily PM_2.5 concentrations was much higher than that (σ_W² = 0.016, accounting for 52.3% of σ_Total²) in annual PM_2.5 concentrations, indicating the importance of temporal variation in PM_2.5 concentrations for studying short-term PM_2.5 exposure. Che et al. (2015) indicated that the within-group exposure variability is larger than the between-group variability for the daily exposure of children to PM_2.5. Our recent study also reported that the within-subject variability (81.3% of σ_Total²) in personal exposure concentrations of PM_2.5 was higher than the between-subject variability in concentrations (18.7% of σ_Total²) for university students (Hsu et al., 2020). In contrast, long-term (annual) concentrations of PM_2.5 were fairly attributable to the between-site or between-group variability (47.8%). As a result, some advanced exposure models with the spatial variables have been used to estimate PM_2.5 exposure. For instance, the land-use regression model incorporating annual PM_2.5 levels from the monitoring stations and geographic information has been successfully developed to predict ambient PM_2.5 concentrations for exposure estimates in epidemiological studies (Eeftens et al., 2012; Wu et al., 2017). However, PM_2.5 concentrations from FEM measurements are more reliable to be incorporated in geographic-based prediction models, while the annual FRM-like PM_2.5 data has almost zero between-site variability (i.e., σ_B² = 0). To minimize the measurement error, the calibration of FRM-like PM_2.5 is still suggested after the exposure estimates in the health risk analysis have been done.

 
3.3 Factors Determining Variability in PM_2.5 Concentrations

Table 3 shows the definition of variables for 16 AQMSs. The information about these variables was obtained from the Taiwan EPA site and the nearest weather bureau. FEM PM_2.5 concentrations and 12 variables associated with meteorology, gaseous air pollutants, and sampling site characteristics were included in the models. Since FRM-like PM_2.5 concentrations presented an extremely low σ_B², this data was not evaluated in the mixed-effects model.

Table 4 shows the model coefficients for the selected variables correlated with daily FEM PM_2.5 determined by the univariate and multivariate analysis. By performing the univariate analysis in the model, we found that the PM_2.5 concentrations decreased significantly from 2014 to 2017, when compared with the reference concentration in 2017. The meteorological factors including ambient temperature, relative humidity, wind speed, and rainfall were negatively correlated (p < 0.001) with ambient PM_2.5, which explained 18.3% [(0.246–0.201)/0.246] of the within-site variability and 6.63% [(0.027–0.025)/0.027] of the between-site variability (Table S6). The ambient pressure was not included in the model for daily PM_2.5 estimates because of its high correlation with ambient temperature. The factors such as gaseous air pollutants (NO₂, O₃, and SO₂) were positively correlated (p < 0.001) with ambient PM_2.5, explaining 45.8% [(0.246–0.133)/0.246] of the within-site variability and 26.8% [(0.027–0.020)/0.027] of the between-site variability (Table S6). Among sampling site factors, the only variable significantly affecting PM_2.5 concentrations was the type of instrument used for measurement, explaining 0% of the within-site variability but 19.5% [(0.027–0.021)/0.027] of the between-site variability (Table S6). It was observed that the VEREWA F701 sampler presented a higher value of 17% than the MetOne 1020 sampler. By performing the multivariate analysis in the model, we found that all the factors significantly affected concentration variability, explaining 50.2% [(0.245–0.122)/0.245] of the within-site variability and 15.3% [(0.027–0.023)/0.027] of the between-site variability (Table S6). The AIC scores from each mixed-effects model were shown in Fig. 2. We found that the selective model considering all the significant variables can predict 55% of ambient PM_2.5 concentrations. Among all the variables, NO₂, O₃, and SO₂ were the predominant variables, explaining 22.1%, 13.5%, and 16.1% of σ_Total² of PM_2.5, respectively.

Fig. 2. AIC scores for the selected variables associated PM_2.5 concentrations using mixed-effect models.

The meteorological factors have a strong influence on the pollutants, leading to daily fluctuations in the pollutant levels. These factors have been shown to explain 50–60% of σ_Total² of winter concentrations in China in a previous study by Zhang et al. (2015). The elevated PM_2.5 concentrations with lower temperature and higher atmospheric pressure are usually observed because the lower mixing layer and higher frequency of thermal inversion in winter can restrict atmospheric dispersion and thereby trap organic and inorganic matter in the particles. It was observed that the decrease in wind speed due to unfavorable atmospheric diffusion could lead to an increase in PM_2.5 concentrations (Liao et al., 2018). The long-range transport usually observed in Taipei during the cold season may be, in part, attributable to the elevated PM_2.5 concentrations (Kuo et al., 2013; Hsu et al., 2019). The positive correlation of PM_2.5 with gaseous air pollutants in our study confirmed the enhanced levels of PM_2.5 due to the chemical reactions in the atmosphere involving precursor gases of NO₂ and SO₂. O₃ was significantly and positively correlated with PM_2.5, which probably could be explained by the covariance of VOCs. However, the precursor formation and meteorological conditions can add complexity to the relationship between both the air pollutants. Ito et al. (2007) have reported that the correlation between O₃ and PM_2.5 changed with the season (positive in summer and negative in winter). In a previous study by Liu et al. (2013), it was highlighted that the use of different aerosol samplers (MetOne 1020 vs VEREWA-F701) for measuring PM_2.5 concentrations might result in a potential bias of exposure variability. It was not surprising that the height of the sample port, distance to the nearest main road, and station type were not significantly associated with ambient PM_2.5 because the inherent confounders were not available for adjustment. Quang et al. (2012) indicated that the PM_2.5 concentrations decreased with the increase in the height of the office building; however, the vehicle emissions, particle formation, flow patterns around the building envelope, and street profile were the important determinants that should be taken into account for comparison. In addition, the differences in the aerosol instruments (MetOne 1020 and VEREWA F701) used in our study and the variations in the sampling factors (i.e., height of the sample port, distance to the nearest main road, and station type) led to variations in PM_2.5 concentrations. For annual PM_2.5, SO₂ was significantly correlated with PM_2.5 in addition to ambient temperature and time (in years) (Table S7). In the population-based study associated with exposure to PM_2.5, meteorological conditions (i.e., temperature) are often adjusted in the model (Luo et al., 2016). Our results showed SO₂ as an important factor showing a significant positive correlation with both daily and annual PM_2.5. We suggest that SO₂ should be taken into account in the future for health risk analysis associated with PM_2.5 exposure for adjustment.

 
4 CONCLUSIONS

In this study, we highlighted that the FRM-like PM_2.5 measurements with negligible between-site variability might lead to non-differential misclassification of exposure. The gaseous pollutants (such as NO₂, O₃, and SO₂) are significant factors affecting the spatial and temporal variations in ambient PM_2.5 concentrations, which should be incorporated in the health risk assessment model. We indicated that the FEM measurements of PM_2.5, rather than the FRM-like measurements, at the AQMSs are mainly applicable for exposure estimates in epidemiological studies. Then, the resultant exposure estimates of PM_2.5 are further calibrated with the FRM measurements to minimize the measurement errors.

 
ACKNOWLEDGMENTS

We appreciate the measurement data supported from the Environmental Protection Administration Executive Yuan, Taiwan. The authors gratefully acknowledge funding received from the National Institute of Environment and Health Sciences (grant No. EM-105-SP08), National Health Research Institutes (NHRI) in Taiwan.

 
DISCLAIMER

The authors declare no conflict of interest.