Article

Urban Organic Aerosol Exposure: Spatial Variations in Composition and Source Impacts

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

We conducted a mobile sampling campaign in a historically industrialized terrain (Pittsburgh, PA) targeting spatial heterogeneity of organic aerosol. Thirty-six sampling sites were chosen based on stratification of traffic, industrial source density, and elevation. We collected organic carbon (OC) on quartz filters, quantified different OC components with thermal-optical analysis, and grouped them based on volatility in decreasing order (OC1, OC2, OC3, OC4, and pyrolyzed carbon (PC)). We compared our ambient OC concentrations (both gas and particle phase) to similar measurements from vehicle dynamometer tests, cooking emissions, biomass burning emissions, and a highway traffic tunnel. OC2 and OC3 loading on ambient filters showed a strong correlation with primary emissions while OC4 and PC were more spatially homogenous. While we tested our hypothesis of OC2 and OC3 as markers of fresh source exposure for Pittsburgh, the relationship seemed to hold at a national level. Land use regression (LUR) models were developed for the OC fractions, and models had an average R2 of 0.64 (SD=0.09). The paper demonstrates that OC2 and OC3 can be useful markers for fresh emissions, OC4 is a secondary OC indicator, and PC represents both biomass burning and secondary aerosol. People with higher OC exposure are likely inhaling more fresh OC2 and OC3, since secondary OC4 and PC varies much less drastically in space or with local primary sources.
Content may be subject to copyright.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The authors reported that OC 2 , and to a lesser extent OC 3 , contributed to a significant fraction of total carbon (TC) in all of the combustion source samples, whereas OC 4 contributed negligibly to TC in all of the combustion source samples, implying its secondary nature. More recently, Li et al. (2018) demonstrated that ambient concentrations of OC 2 -OC 3 fractions are highly correlated with the total OC concentrations from traditional combustion source measurements (i.e., gasoline and diesel exhaust, cooking, and biomass burning), which indicates the primary combustion origin of these fractions, while OC 4 was not statistically significantly correlated with total OC in any of the combustion source filters, which suggested its secondary origin. ...
... Table 1 presents summary statistics of the parameters used in the input matrices at each of the sites. As mentioned in section 2.2.1, OC 1 has a higher vapor pressure in comparison to the other OC fractions (Chow et al., 1993;Li et al., 2018) which favors the partitioning of this species mostly in the vapor phase with very low concentrations (in most cases below our detection limit) in the particle phase. The percentage of concentration values lower than the LOD for this OC volatility fraction was > 90% in our dataset, which prompted us to exclude this OC fraction from our analysis. ...
... On the other hand, this factor contributed to less than 50% (ranging from 32 to 46%) of OC 4 at all sites. Previous studies have indicated that OC 2 and OC 3 are associated with combustion sources, including traffic, as the emissions from these sources are rich in semi-volatile organic compounds (SVOCs), whereas OC 4 is the least volatile OC fraction and is mostly associated with SOA (Cao et al., 2005;Li et al., 2018). In addition, as shown in Fig. S5, OC 2 and OC 3 were highly correlated with total OC concentrations from combustion source test measurements (such as gasoline and diesel Soleimanian, et al. ...
Article
In this study, we used the positive matrix factorization (PMF) model to apportion the sources of organic carbon volatility fractions (OC x ) as well as total OC concentrations in five different locations across the Los Angeles Basin, including West Long Beach, Anaheim, central Los Angeles (CELA), Rubidoux, and Fontana over the period from July 2012 to June 2013. Total OC as well as OC volatility fractions (OC 1 -OC 4 ), measured with the thermal-optical analysis as part of the fourth Multiple Air Toxics Exposure Study (MATES IV) by the South Coast Air Quality Management District (SCAQMD), in combination with gaseous and particulate source tracers (such as NO x , O 3 , particulate sulfate, K ⁺ /K ratio, and biomass-burning originated black carbon (BC bb )) were used as inputs to the PMF model. A 3-factor solution, including traffic, secondary organic aerosols (SOA), and biomass burning, was found to be the most physically interpretable solution. Average total OC concentrations showed an upward trend from the sites closer to the coast (i.e., 3.7 ± 1.9 μg m ⁻³ at West Long Beach) to inland downwind sites (i.e., 4.8 ± 1.8 μg m ⁻³ at Fontana), especially in the warm season, suggesting the major impact of SOA formation on total OC concentrations. Source apportionment results indicated that traffic is the dominant contributor to OC 2 and OC 3 fractions, especially at the sites that are near major primary sources such as CELA and West Long Beach, with corresponding contributions of 60 ± 1.0% and 79 ± 1.7% to OC 2 and 53 ± 0.9% and 64 ± 1.3% to OC 3 , respectively. On the other hand, SOA was found to be the dominant contributor to OC 4 fraction, especially at the receptor sites located further inland, with corresponding contributions of 66 ± 1.0% in Rubidoux and 56 ± 0.7% in Fontana. Our results also indicated that traffic is the dominant source of total OC concentrations, with an average contribution of 53 ± 2.4% at all the sites, followed by SOA formation and biomass burning, contributing to 40 ± 1.8% and 7 ± 0.8% of total OC concentrations, respectively. The contribution of traffic and biomass burning to total OC concentrations increased during the cold season, while that of SOA became more significant during the warm season when photochemical activities peak. Results from the present study provided important insight on the sources and spatio-temporal variations of OC volatility fractions as well as total OC concentrations in PM 2.5 across the Los Angeles Basin.
... 26 PM components that arise from different processes and sources can have correspondingly different spatial patterns in urban environments. 11,27 Previous studies have shown significant intraurban concentration variations in POA factors such as COA and HOA. 11,28 While LUR models of PM 2.5 mass concentrations have become common, 29 very few LUR studies have focused on the components that comprise PM. ...
... Li et al. developed LURs for fractions of particulate organic carbon (OC) based on volatility, and found certain fractions to have strong connections with source categories (e.g., semivolatile OC and traffic). 27 However, none of these studies use explicitly source-resolved PM measurements as a starting point, nor do they have speciated measurements for all of the major components of fine PM massinorganic species, OA factors, and black carbonlike we do here. ...
... This strategy has been used in previous LUR studies (see Saraswat et al. 48 and Li et al. 27,32 ). CRM data are assigned to grid cells corresponding to times when our mobile laboratory was within each cell. ...
Article
This study presents land-use regression (LUR) models for submicron particulate matter (PM1) components from an urban area. Models are presented for mass concentrations of inorganic species (SO4, NO3, NH4), organic aerosol (OA) factors, and total PM1. OA is source-apportioned using positive matrix factorization (PMF) of data collected from aerosol mass spectrometry deployed on a mobile laboratory. PMF yielded a three-factor solution: cooking OA (COA), hydrocarbon-like OA (HOA), and less-oxidized oxygenated OA (LO-OOA). This study represents the first time that LUR has been applied to source-resolved OA factors. We sampled a roughly 20 km2 area of West Oakland, California, USA, over 1 month (mid-July to mid-August, 2017). The road network of the sampling domain was comprehensively sampled each day using a randomized driving route to minimize temporal and spatial bias. Mobile measurements were aggregated both spatially and temporally for use as discrete spatial observations for LUR model building. LUR model performance was highest for those species with more spatial variability (primary OA factors: COA R2 = 0.80, HOA R2 = 0.67) and lowest for secondary inorganic species (SO4 R2 = 0.47, NH4 R2 = 0.43) that were more spatially homogeneous. Notably, the stepwise selective LUR algorithm largely selected predictors for primary OA factors that correspond to the associated land-use categories (e.g., cooking land-use variables were selected in cooking-related PM models). This finding appears to be robust, as we demonstrate the predictive link between land-use variables and the corresponding source-resolved PM1 components through a subsampling analysis.
... Identification of the mortality-related types of ambient PM 2.5 can enable the development of a focused intervention strategy of placing appropriate preventive measures for reducing the generation of source-specific PM 2.5 and subsequently diminishing PM 2.5 -related mortality. Studies have indicated that sources of PM 2.5 may vary and are likely to contribute to the accumulation of various toxic compounds that are suspended in the air, such as sulphur oxides (SO x ), carbon monoxide (CO), particulates, and nitrogen oxides (NO x ) [8][9][10] , which may then contribute to various health problems and subsequently an increase in the global burden of disease. Policy makers have also set global no-threshold limits for exposure to ambient PM 2.5 (i.e., daily exposure less than 25 μg/m 3 while annual exposure less than ≤ 10 μg/m 3 ) 9 ; yet, there is still an ongoing discussion of the need to harmonize ambient air quality standards since these standards vary greatly among regions and countries 11,12 . ...
... Since estimations of the dose-response relationship for global data may not be linear, the penalized spline smoothing function was used to determine the non-linear relationship between the discriminated ambient PM 2.5 and the under-five and maternal mortality (Figs. 4,5,6,7,8,9). However, the results on the dose-response relationship of the global biomass PM 2.5 and the under-five mortality and maternal mortality indicated a slight increase in the risk of under-five deaths (Fig. 8) and maternal deaths (Fig. 8) after surpassing a biomass PM 2.5 concentration of approximately 33 μg/m 3 , suggesting higher levels of exposure than the current global standards, which require daily exposure to be less than 25 μg/m 3 while annual exposure should be less than ≤ 10 μg/m 3 . ...
... Finally, the dose-response relationship was determined using the penalized spline and the generalized additive mixed-effects model (GAMM), and year and country were taken as the random effects because of the nonparametric relationship that was exhibited in the global data (Figs. 4,5,6,7,8,9). The degrees of freedom were estimated using generalized cross-validation (GCV). ...
Article
Full-text available
Globally, it was estimated that maternal and under-five deaths were high in low-income countries than that of high-income countries. Most studies, however, have focused only on the clinical causes of maternal and under-five deaths, and yet there could be other factors such as ambient particulate matter (PM). The current global estimates indicate that exposure to ambient PM2.5 (with ≤ 2.5 microns aerodynamic diameter) has caused about 7 million deaths and over 100 million disability-adjusted life-years. There are also several health risks that have been linked PM2.5, including mortality, both regionally and globally; however, PM2.5 is a mixture of many compounds from various sources. Globally, there is little evidence of the health effects of various types of PM2.5, which may uniquely contribute to the global burden of disease. Currently, only two studies had estimated the effects of discriminated ambient PM2.5, that is, anthropogenic, biomass and dust, on under-five and maternal mortality using satellite measurements, and this study found a positive association in Africa and Asia. However, the study area was conducted in only one region and may not reflect the spatial variations throughout the world. Therefore, in this study, we discriminated different ambient PM2.5 and estimated the effects on a global scale. Using the generalized linear mixed-effects model (GLMM) with a random-effects model, we found that biomass PM2.5 was associated with an 8.9% (95% confidence interval [CI] 4.1–13.9%) increased risk of under-five deaths, while dust PM2.5 was marginally associated with 9.5% of under-five deaths. Nevertheless, our study found no association between PM2.5 type and global maternal deaths. This result may be because the majority of maternal deaths could be associated with preventable deaths that would require clinical interventions. Identification of the mortality-related types of ambient PM2.5 can enable the development of a focused intervention strategy of placing appropriate preventive measures for reducing the generation of source-specific PM2.5 and subsequently diminishing PM2.5-related mortality.
... In particular, mobile sampling enables deployment of high-time-resolution measurements that can identify specific PM sources. For example, Li et al. (2018) showed that emissions of primary OA and BC drive much of the spatial variation in PM 2.5 observed in Pittsburgh. Apte et al. (2017) used mobile BC measurements to identify hotspots associated with vehicle traffic and industrial activities. ...
... Data were collected as part of the Center for Air, Climate, and Energy Solutions (CACES) air quality observatory (Zimmerman et al., 2018). The mobile laboratory is an instrumented Nissan 2500 cargo van, previously described by Li et al. (2016Li et al. ( , 2018. Figure 1A shows a map of the sampling domain. We sampled on all streets in the domain that were open to public traffic. ...
Article
Full-text available
We investigated spatial and temporal patterns in concentration and composition of sub-micron particulate matter (PM1) in Oakland, California in the summer of 2017 using an aerosol mass spectrometer mounted in a mobile laboratory. We performed ∼160 hours of mobile sampling in the city over a 20-day period. Measurements are compared for three adjacent neighborhoods with distinct land uses: a central business district (downtown), a residential district (West Oakland), and a major shipping port. The average organic aerosol (OA) concentration is 5.3μgm−3 and contributes ∼50% of the PM1 mass. OA concentrations in downtown are, on average, 1.5μgm−3 higher than in West Oakland and Port. We decomposed OA into three factors using positive matrix factorization: hydrocarbon-like OA (HOA; 20% average contribution), cooking OA (COA; 25%) and semi-volatile oxidized OA (SV-OOA; 55%). The collective 45% contribution from primary OA (HOA + COA) emphasizes the importance of primary emissions in Oakland. The dominant source of primary OA shifts from HOA-rich in the morning to COA-rich after lunch time. COA in downtown is consistently higher than West Oakland and Port due to a large number of restaurants. HOA exhibits variability in space and time. Morning-time HOA concentration in downtown is twice that in Port, but Port HOA increases more than two-fold during mid-day, likely because trucking activity at the Port peaks at that time. Despite the expectation of being spatially uniform, SV-OOA also exhibits spatial differences. Morning-time SV-OOA in downtown is roughly 25% (∼0.6μgm−3) higher than the rest of Oakland. Even as the entire domain approaches a more uniform photo-chemical state in the afternoon, downtown SV-OOA remains statistically higher than West Oakland and Port, suggesting that downtown is a microenvironment with higher photochemical activity. Higher concentrations of particulate sulfate (also of secondary origin) with no direct sources in Oakland further reflect higher photochemical activity in downtown. A combination of several factors (poor ventilation of air masses in street canyons, higher concentrations of precursor gases, higher concentrations of the hydroxyl radical) likely result in the proposed high photochemical activity in downtown. Lastly, through Van Krevelen analysis of elemental ratios (H/C, O/C) of the OA, we show that OA in Oakland is more chemically reduced than several other urban areas. This underscores the importance of primary emissions in Oakland. We also show that mixing of oceanic air masses with these primary emissions in Oakland is an important processing mechanism that governs the overall OA composition in Oakland. The findings of this study are important because the pollutants we find contributing the most to OA variability, both of primary and secondary origin, are ubiquitous in other urban locations.
... The CH 4 analyzer agreed within 3% of the reference gas concentrations. Weather data from AIMAR 200WX were filtered based on the in-built data quality indicator in the split wind data output (Li et al., 2017). The null measurements and CH 4 concentrations less than 1.5 ppm were filtered. ...
... The handheld CH 4 detector reported 60 ppm-m and the CH 4 analyzer showed >0.2 ppm increase. The leak rate was below the high flow sampler's measurement rage and thus was estimated as half the detection limit (Li et al., 2017). The unlabeled 3rd CH 4 plume was from a natural gas well pad and not from the gathering pipeline system, thus not included in the discussion here. ...
... In particular, mobile sampling enables the deployment of high-time-resolution measurements that can identify specific PM sources. For example, Li et al. (2018) showed that emissions of primary OA and BC drive much of the spatial variation in PM 2.5 observed in Pittsburgh. Apte et al. (2017) used mobile BC measurements to identify hot spots associated with vehicle traffic and industrial activities. ...
... Data were collected as part of the Center for Air, Climate, and Energy Solutions (CACES) air quality observatory (Zimmerman et al., 2018). The mobile laboratory is an instrumented Nissan 2500 cargo van, previously described by H. Z. and Li et al. (2018). Figure 1a shows a map of the sampling domain. ...
Article
Full-text available
We investigated spatial and temporal patterns in the concentration and composition of submicron particulate matter (PM1) in Oakland, California, in the summer of 2017 using an aerosol mass spectrometer mounted in a mobile laboratory. We performed ∼ 160 h of mobile sampling in the city over a 20-day period. Measurements are compared for three adjacent neighborhoods with distinct land uses: a central business district (downtown), a residential district (West Oakland), and a major shipping port (port). The average organic aerosol (OA) concentration is 5.3 µg m−3 and contributes ∼ 50 % of the PM1 mass. OA concentrations in downtown are, on average, 1.5 µg m−3 higher than in West Oakland and port. We decomposed OA into three factors using positive matrix factorization: hydrocarbon-like OA (HOA; 20 % average contribution), cooking OA (COA; 25 %), and less-oxidized oxygenated OA (LO-OOA; 55 %). The collective 45 % contribution from primary OA (HOA + COA) emphasizes the importance of primary emissions in Oakland. The dominant source of primary OA shifts from HOA-rich in the morning to COA-rich after lunchtime. COA in downtown is consistently higher than West Oakland and port due to a large number of restaurants. HOA exhibits variability in space and time. The morning-time HOA concentration in downtown is twice that in port, but port HOA increases more than two-fold during midday, likely because trucking activity at the port peaks at that time. While it is challenging to mathematically apportion traffic-emitted OA between drayage trucks and cars, combining measurements of OA with black carbon and CO suggests that while trucks have an important effect on OA and BC at the port, gasoline-engine cars are the dominant source of traffic emissions in the rest of Oakland. Despite the expectation of being spatially uniform, LO-OOA also exhibits spatial differences. Morning-time LO-OOA in downtown is roughly 25 % ( ∼ 0.6 µg m−3) higher than the rest of Oakland. Even as the entire domain approaches a more uniform photochemical state in the afternoon, downtown LO-OOA remains statistically higher than West Oakland and port, suggesting that downtown is a microenvironment with higher photochemical activity. Higher concentrations of particulate sulfate (also of secondary origin) with no direct sources in Oakland further reflect higher photochemical activity in downtown. A combination of several factors (poor ventilation of air masses in street canyons, higher concentrations of precursor gases, higher concentrations of the hydroxyl radical) likely results in the proposed high photochemical activity in downtown. Lastly, through Van Krevelen analysis of the elemental ratios (H ∕ C, O ∕ C) of the OA, we show that OA in Oakland is more chemically reduced than several other urban areas. This underscores the importance of primary emissions in Oakland. We also show that mixing of oceanic air masses with these primary emissions in Oakland is an important processing mechanism that governs the overall OA composition in Oakland.
... The Airmar weather station was elevated by a polyvinyl chloride (PVC) pipe to a height of 2.5 m above the ground ( Figure 2). This setup helped ensure wind measurements were not affected by vehicle movements [18,[37][38][39][40]. Our sampling line was a 1/4'' OD Teflon tube attached to the weather station. ...
... The Airmar weather station was elevated by a polyvinyl chloride (PVC) pipe to a height of 2.5 m above the ground (Figure 2). This setup helped ensure wind measurements were not affected by vehicle movements [18,[37][38][39][40]. Our sampling line was a 1/4" OD Teflon tube attached to the weather station. ...
Article
The United States Environmental Protection Agency Greenhouse Gas Inventory only recently updated the emission factors of natural gas gathering pipelines in April 2019 from the previous estimates based on a 1990s study of distribution pipelines. Additional measurements are needed from different basins for more accurate assessments of methane emissions from natural gas midstream industries and hence the overall climate implications of natural gas as the interim major energy source for the next decade. We conducted an unmanned aerial vehicle (UAV) survey and a ground-based vehicle sampling campaign targeting gathering pipeline systems in the Utica Shale from March to April in 2019. Out of 73 km of pipeline systems surveyed, we found no leaks on pipelines and two leaks on an accessory block valve with leak rates of 3.8 ± 0.4 and 7.6 ± 0.8 mg/s. The low leak frequency phenomenon was also observed in the only existing gathering pipeline study in Fayetteville Shale. The UAV sampling system facilitated ease of access, broadened the availability of pipelines for leak detection, and was estimated to detect methane leaks down to 0.07 g/s using Gaussian dispersion modeling. For future UAV surveys adopting similar instrument setup and dispersion models, we recommend arranging controlled release experiments first to understand the system's detection limit and choosing sampling days with steady and low wind speeds (2 m/s).
... Mobile monitoring is well suited to quantify pollutant spatial variations at fine (sub-km) length scales. Numerous studies have demonstrated the ability of mobile sampling to resolve pollutant spatial differences near specific sources (e.g., roadways), as well as to map neighborhood-level differences within a city (Apte et al., 2017;Deshmukh et al., 2016;Hankey and Marshall, 2015;Li et al., 2018Li et al., , 2016Steffens et al., 2017;Ye et al., 2018). A major challenge with mobile sampling is to collect enough data in a given location to accurately quantify annual-or other long-term average concentrations, though identifying hotspots or consistent gradients between neighborhoods requires less data (Tan et al., 2014b;Apte et al., 2017;Messier et al., 2018). ...
... However, neither traffic nor restaurants can completely describe the observed spatial variability, and other covariates may be necessary for modeling intracity pollutant variation (Eeftens et al., 2012;Li et al., 2018Li et al., , 2016. Variability in land use seems to be more important as a determining factor for pollutant spatial variations than the absolute activity level. ...
Article
Long- and short-term exposure to airborne pollutants results in adverse health effects. Regulatory monitors can be used to determine if regional concentrations meet regulatory standards of air pollution. As assessments of air pollutant exposure become more spatially resolved, evaluation is needed to assess the spatial representativeness of monitors in different environments. We measured NO2, ultrafine particle concentration (UFP), and PM1 with both stationary and mobile platforms in Pittsburgh, PA in 2016 and 2017. We sampled in eight ∼1 km2 neighborhoods representing different land use and exposure regimes (e.g., urban and suburban, high and low traffic). Mobile sampling was conducted on up to 25 days in each neighborhood to study fine-scale spatial variation in pollutant concentrations. NO2 exhibited within-neighborhood spatial variation, with hotspots elevated by up to a factor of 5 above the regional background. Spatial differences in UFP within the same 1 km2 neighborhoods could be a factor of 2.4 times regional background. PM1 was more regional and less spatially variable. Most neighborhoods exhibited less than 1 μg m-3 spatial variability in PM1. Spatial variability of NO2 and UFP showed moderate correlation (R2 > 0.5) with traditional land use covariates such as traffic volume and restaurant density. We used the Wilcoxon rank-sum test to calculate the fraction of each neighborhood represented by the same underlying concentration distribution. PM1 was the most spatially homogeneous, with 80-100% of each 1 km2 area being statistically similar to a reference location. Quantifying pollutant spatial patterns with high fidelity (e.g., <2 ppb NO2 or <1 μg m-3 PM1) seems unattainable in many urban areas unless the sampling network is significantly dense, with more than one or two nodes per km2.
... In SEA, data on OC fraction bound to PM 2.5 in the haze periods showed that OC4 contributed about 20% and 10% of OC and PM 2.5 respectively. The significance of OC4 in PM 2.5 in the haze periods implies the secondary nature of OC [69,100]. However, studies relating to SOA in haze periods SEA have not been considered yet. ...
Article
Full-text available
Haze is a well-known air pollution phenomenon linked to the severe and persistent particulate matter (PM) episodes in Southeast Asia (SEA), which significantly impacts the environment, health, and economy. This work reviewed for the first time the characteristics of haze episodes in terms of PM concentrations, chemical compositions, and the causes of haze in both Lower (Maritime) and Upper (Mainland) SEA. In addition, we carried out a systematic comparison of the frequency and intensity of haze events through SEA regions in recent years. Our finding indicated that the different trend of haze frequency and intensity between SEA cities are not only due to local air pollution sources such as biomass burning (BB) but also meteorology and long-range transport. Other sources such as secondary aerosols also play an important role in haze formation, but they have not been comprehensively investigated in previous studies. Due to the complicated formation mechanisms and the transportations of haze and its impacts on SEA’s human health and economy, more sophisticated and specific policies are needed to deal with haze issues not only for individual countries but also on a regional scale.
... 1,4 Such intracity variations are often caused by localized primary sources of PM emissions. 5 Studies have also shown evidence that emissions from different sources vary in their negative health effects. 6 Thus, it is important to capture the spatial variability of both the concentration and source contributions of PM in urban environments. ...
Article
Localized primary emissions of carbonaceous aerosol are the major drivers of intra-city variability of submicron particulate matter (PM1) concentrations. We investigated spatial variations in PM1 composition with mobile sampling in Pittsburgh, Pennsylvania, USA, and performed source apportionment analysis to attribute primary organic aerosol (OA) to traffic (HOA) and cooking OA (COA). In high source impact locations, the PM1 concentration is on average 2 μg m-3 (40%) higher than urban background locations. Traffic emissions are the largest source contributing to population-weighted exposures to primary PM. Vehicle-miles travelled (VMT) can be used to reliably predict the concentration of HOA and localized black carbon (BC) in air pollutant spatial models. Restaurant count is a useful but imperfect predictor for COA concentration, likely due to highly variable emissions from individual restaurants. Near-road cooking emissions can be falsely attributed to traffic sources in the absence of PM source apportionment. In Pittsburgh, 28% and 9% of the total population are exposed to >1 μg m-3 of traffic- and cooking-related primary emissions, with some populations impacted by both sources. The source mix in many U.S. cities is similar, thus we expect similar PM spatial patterns and increased exposures in high-source areas in other cities.
... The mass fractions of OC2, and OC3 (i.e., the more volatile fractions of OC) as well as OC4 (i.e., least volatile fraction of OC) were almost the same between the warm and colder phases. It should be noted that OC1 was not included in Fig. S1, since it has a much higher vapor pressure in comparison to the other OC fractions (Chow et al., 1993;Li et al., 2017), resulting in very low particle phase concentrations (in most cases below the detection limit) in the ambient conditions of Los Angeles. Detailed explanations regarding concentration losses of each of the investigated groups (i.e., PAHs, organic acids, n-alkanes, and levoglucosan) in this study are presented in the following sections. ...
Article
The volatility profiles of PM2.5 semi-volatile compounds and relationships to the oxidative potential of urban airborne particles were investigated in central Los Angeles, CA. Ambient and thermodenuded fine (PM2.5) particles were collected during both warm and cold seasons by employing the Versatile Aerosol Concentration Enrichment System (VACES) combined with a thermodenuder. When operated at 50 °C and 100 °C, the VACES/thermodenuder system removed about 50% and 75% of the PM2.5 volume concentration, respectively. Most of the quantified PM2.5 semi-volatile species including organic carbon (OC), water soluble organic carbon (WSOC), polycyclic aromatic hydrocarbons (PAHs), organic acids, n-alkanes, and levoglucosan, as well as inorganic ions (i.e., nitrate, sulfate, and ammonium) exhibited concentration losses in the ranges of 40-66% and 67-92%, respectively, as the thermodenuder temperature increased to 50 °C and 100 °C. Species in the PM2.5 such as elemental carbon (EC) and inorganic elements (including trace metals) were minimally impacted by the heating process – thus can be considered refractory. On average, nearly half of the PM2.5 oxidative potential (as measured by the dichlorodihydrofluorescein (DCFH) alveolar macrophage in vitro assay) was associated with the semi-volatile species removed by heating the aerosols to only 50 °C, highlighting the importance of this quite volatile compartment to the ambient PM2.5 toxicity. The fraction of PM2.5 oxidative potential lost upon heating the aerosols to 100 °C further increased to around 75-85%. Furthermore, we document statistically significant correlations between the PM2.5 oxidative potential and different semi-volatile organic compounds originating from primary and secondary sources, including OC (Rwarm, and Rcold) (0.86, and 0.74), WSOC (0.60, and 0.98), PAHs (0.88, and 0.76), organic acids (0.76, and 0.88), and n-alkanes (0.67, and 0.83) in warm and cold seasons, respectively, while a strong correlation between oxidative potential and levoglucosan, a tracer of biomass burning, was observed only during the cold season (Rcold=0.81).
... Each trip was from 8 a.m. EDT to 4 p.m. EDT. We designed driving routes based on locations of previously identified CH 4 sources and a stratified sampling strategy of selecting both upwind and downwind roads to capture CH 4 spatial gradients Li et al., 2019Li et al., , 2017Robinson et al., 2018). The final routes did not include highways and interstates since we intended to drive slowly to increase the chance of intersecting the plume. ...
Article
Methane (CH4), as the second largest component of greenhouse gases in the atmosphere, can be formed through biogenic or thermogenic processes. A 2015 aircraft survey conducted by the National Oceanic and Atmospheric Administration (NOAA) observed a 2.7 ppm methane plume (40% increase relative to background) at an altitude of 500 m above ground in Western Maryland. Specific sources were not attributed, however, due to rough terrain and a variety of possible sources including coal mines, landfills, and natural gas infrastructure (production wells, storage wells, abandoned wells, pipelines and compressor stations). National Energy Technology Laboratory (NETL) conducted a stationary ambient air monitoring campaign during 2014 Spring/Summer (May to August) and Fall/Winter (November to February) and a 5-day mobile sampling survey in September and October 2018 to identify the potential source(s). The ambient air monitoring laboratory was located in Western Maryland in an area containing production, abandoned, and storage wells. By using cavity ring-down spectroscopy, we observed CH4 concentrations up to 7.4 ppm. The plume events had a thermogenic δ¹³C signature of −35.2 ± 0.6‰. Mobile survey routes were optimized based on locations of previously identified CH4 sources. We observed a 10-min long CH4 plume event while stopping/driving in a township downwind of a compressor station and repeated 10 ppm plumes near the landfill in the northeast of Western Maryland. CH4 emissions from the landfill had a δ¹³C value of −54.0 ± 0.4‰, indicative of biogenic origins. We calculated the CH4 emission flux from the landfill to be 1575 ± 1173 tons/year by using inverse dispersion modeling. The flux estimation agreed with Maryland Department of the Environment inventory if assuming no day to day variation in the emission rate. CH4 sources in Western Maryland include natural gas infrastructure and the landfill. The 2.7 ppm plume observed by the aircraft was most likely from the landfill near Frostburg in the northeast of the study area.
... Table 1 presents the outputs of PCA analysis for both seasons, which resulted in the identification of three factors explaining approximately 99% (warm) and 91% (cold) of total variance in the data. According to Table 1, the first factor was identified as the fossil fuel combustion, due to the high loadings of EC, WIOC, total PAHs, OC 2 , and OC 3 as surrogates of vehicular activities and fresh primary emissions (Alcántara et al., 2008;Cao et al., 2005;Kueseng et al., 2010;Li et al., 2018;Lima et al., 2005;Liu et al., 2014;Schauer, 2003). In the Los Angeles area, fossil fuel combustion was dominated by traffic related emissions (i.e., diesel and gasoline vehicles), along with minor contributions of power plants and local industries consuming petroleum and natural gas (Hasheminassab et al., 2014a(Hasheminassab et al., , 2013Heo et al., 2013). ...
... Several previous studies have documented EC and OC1 as tracers of vehicular emissions (66)(67)(68). In addition, the presence of other OCx (i.e., OC2, OC3), Cu, and Ti in the profile of this factor also corroborates its vehicular origins (52,53,69). The seasonal trend for this factor, as shown in Figure 7, demonstrated significantly higher (Pvalue < 0.05) winter-time contributions in both CELA and Riverside sites, due mainly to the stable meteorological conditions and depressed mixing heights during cold weather which restrict the atmospheric dispersion and dilution of these emissions (70,71). ...
Article
This study aimed to investigate the long-term variations in the contribution of emission sources to ambient PM2.5 organic carbon (OC) in central Los Angeles (CELA) and Riverside using the Chemical Speciation Network (CSN) database in the 2005-2015 period, during which several federal and state PM-based regulations were implemented to reduce tailpipe emissions in the region. The measured concentrations of OC, OC volatility fractions (i.e., OC1, OC2, and OC3), elemental carbon (EC), ozone (O3), sulfate, the ratio of potassium ion to potassium (K+/K ), and selected metal elements were used as the input to the positive matrix factorization (PMF) model. PMF resolved tailpipe emissions, non-tailpipe emissions, secondary organic aerosol (SOA), biomass burning, and local industrial activities as the main sources contributing to ambient OC at both sampling sites. Vehicular exhaust emissions, non-tailpipe emissions, and SOA were dominant sources of OC across our sampling sites, accounting cumulatively for more than 80% of total OC mass throughout the study period. Our findings showed a significant reduction in the absolute and relative contribution of tailpipe emissions to the ambient OC levels in CELA and Riverside over the time period of 2005-2015. The contribution of exhaust emissions to total OC in CELA decreased from 3.5 μg/m3 (49%) in 2005 to 1.5 μg/m3 (34%) in 2015, while similar trends were observed at Riverside during this period. These reductions are mainly attributed to the implementation of several federal, state, and local air quality regulations targeting tailpipe emissions in the area. The implementation of these regulations furthermore reduced the emissions of primary organic precursors of secondary aerosols, resulting in an overall decrease (although not statistically significant, P values ranging from 0.4 to 0.6) in SOA mass concentration in both locations over the study period. In contrast to the tailpipe emissions, we observed an increasing trend (by ~ 4 to 14%) in the relative contribution of non-tailpipe emissions to OC over this time period at both sites. Our results demonstrated the effectiveness of air quality regulations in reducing direct tailpipe emissions in the area, but also underpinned the need to develop equally effective mitigation policies targeting non-tailpipe PM emissions. Keywords: PM-based regulations, PMF source apportionment, Organic carbon, Non-tailpipe emissions, Tailpipe emissions, Los Angeles basin
... Several studies [5][6][7] have shown that air pollutant concentrations have fine-scale spatial variations that are not captured by the regulatory monitors. Recently, Hugh Z. Li et al. (2019) [8] have experimented, by deploying both stationary and mobile platforms in the city of Pittsburgh, that NO 2 exhibited within-neighborhood spatial variations, with hotspots elevated by up to a factor of five above the regional background. ...
Article
Full-text available
Urban air pollution continues to represent a primary concern for human health, despite significant efforts by public authorities for mitigating its effects. Regulatory monitoring networks are essential tools for air pollution monitoring. However, they are sparse networks, unable to capture the spatial variability of the air pollutants. For addressing this issue, networks of low cost stations are deployed, supplementing the regulatory stations. Regarding this application, an important question is where these stations are installed The objective of this study was to generate a site suitability map for the development of a network of low cost multi-sensor stations across a city for a spatially dense urban air quality monitoring. To do that, a site suitability analysis was developed based on two geographical variables properly selected for representing the impact of urban pollutant sources and urban form on the pollutant concentrations. By processing information about emissions patterns and street canyon effects, we were able to identify air quality hotspot areas supposed to show high spatial variability. Low cost monitoring stations, there located, are able to provide that informative content, which is lacking for both regulatory monitoring networks and predictive modelling for high resolution air quality mapping. Keywords: site suitability analysis; optimal low cost sensor deployment; urban air quality monitoring; 3D urban model; street canyon effect; traffic-related pollutant emissions
Article
Volatile organic compounds (VOCs) are precursors for ozone (O3) and secondary particulate matter, which contribute to asthma and cardiovascular diseases. With the technology development of hydraulic fracking, the United States experienced a shale gas boom in the last decade while the public raised concerns about the potential health impacts of co-emitted VOCs and other airborne pollutants. National Energy Technology Laboratory conducted stationary trailer-based ambient monitoring to study the sources of VOCs in Maryland, where the state enacted a moratorium on unconventional natural gas extraction. The campaign had two periods, May to August 2014 (summer) and November 2014 to February 2015 (winter). Ethane was the most abundant VOC, averaging 12.3 ppb (SD = 15.7 ppb) in summer and 21.7 ppb (SD = 21.6 ppb) in winter. The seasonal variation of VOCs indicated different source strengths. The sampling region was in the nitrogen oxides (NOx) limited regime for O3 production, and the O3 concentrations were sensitive to VOC/NOx ratios in the early mornings. We derived a six-factor profile using positive matrix factorization: motor vehicles, industrial, biogenics, coal burning, fugitive and evaporative, and ozone secondary. The fugitive and evaporative factor explained 44.5% of total VOCs, and the motor vehicles factor followed second with 15.5%. Oil and gas activities had a considerable impact on the abundance of VOCs in this region.
Article
In this study, we investigated the organic constituents and toxicological characteristics of PM2.5 liquid suspensions collected by the means of a modified versatile aerosol concentration enrichment system (VACES)/aerosol-into-liquid collector tandem technology. Filter and slurry PM2.5 samples were collected during warm (spanning from mid-August 2019 to early-September 2019) and cold (from mid-December 2019 to early-January 2020) seasons at an urban site in central Los Angeles. The collected samples were chemically analyzed for their total organic carbon (TOC) as well as individual organic species, including polycyclic aromatic hydrocarbons (PAHs), hopanes and steranes, n-alkanes, and organic acids. In addition, the oxidative potential of PM2.5 samples was quantified by means of the 2′,7′-dichlorodihydrofluorescein (DCFH) in vitro assay. Overall, our findings revealed a very good agreement between the mass fractions of organic compounds in the collected PM2.5 slurries and filters, judging by the average slurry-to -filter ratios of 1.17 ± 0.15, 1.13 ± 0.14, 1.13 ± 0.15, and 1.16 ± 0.22 for PAHs, hopanes and steranes, n-alkanes, and organic acids, respectively. Moreover, comparable (Pvalue = 0.2) oxidative potential levels were observed between the collected filters and PM2.5 liquid suspensions. The aqueous PM suspensions collected by the VACES/aerosol-into-liquid collector exhibited slightly higher (although not statistically significant) capture of some semi-volatile organic compounds (SVOCs), probably due to less sampling artifacts and better preservation of these SVOCs collected directly into aqueous suspensions compared to the filter samplers. The outcomes of this study confirm the efficient collection of PM-bound organic species in aqueous slurries by the aerosol-into-liquid collector, further corroborating the ability of this sampler to be used as an attractive alternative to filter samplers for collection of PM2.5 for toxicological studies.
Article
Organic aerosol (OA) is a major component of fine particulate matter (PM2.5) in urban environments. We performed in-motion ambient sampling from a mobile platform with an aerosol mass spectrometer (AMS) to investigate the spatial variability and sources of OA concentrations in Pittsburgh, Pennsylvania, a midsize, largely postindustrial American city. To characterize the relative importance of cooking and traffic sources, we sampled in some of the most populated areas (∼18 km²) in and around Pittsburgh during afternoon rush hour and evening mealtime, including congested highways, major local roads, areas with high densities of restaurants, and urban background locations. We found greatly elevated OA concentrations (10s of μg m–3) in the vicinity of numerous individual restaurants and commercial districts containing multiple restaurants. The AMS mass spectral information indicates that majority of the high concentration plumes (71%) were from cooking sources. Areas containing both busy roads and restaurants had systematically higher OA concentrations than areas with only busy roads and urban background locations. Elevated OA concentrations were measured hundreds of meters downwind of some restaurants, indicating that these sources can influence air quality on neighborhood scales. Approximately 20% of the population (∼250 000 people) in the Pittsburgh area lives within 200 m of a restaurant; therefore, restaurant emissions are potentially an important source of outdoor PM exposures for this large population.
Article
Full-text available
Background: The evidence from observational epidemiological studies of a link between long-term air pollution exposure and diabetes prevalence and incidence is currently mixed. Some studies found the strongest associations of diabetes with fine particles, other studies with nitrogen dioxide and some studies found no associations. Objectives: Our aim was to investigate associations between long-term exposure to multiple air pollutants and diabetes prevalence in a large national survey in the Netherlands. Methods: We performed a cross-sectional analysis using the 2012 Dutch national health survey to investigate the associations between the 2009 annual average concentrations of multiple air pollutants (PM10, PM2.5, PM10-2.5, PM2.5 absorbance, OP(DTT), OP(ESR) and NO2) and diabetes prevalence, among 289,703 adults. Air pollution exposure was assessed by land use regression models. Diabetes was defined based on a combined measure of self-reported physician diagnosis and medication prescription from an external database. Using logistic regression, we adjusted for potential confounders, including neighborhood- and individual socio-economic status and lifestyle-related risk factors such as smoking habits, alcohol consumption, physical activity and BMI. Results: After adjustment for potential confounders, all pollutants (except PM2.5) were associated with diabetes prevalence. In two-pollutant models, NO2 and OP(DTT) remained associated with increased diabetes prevalence. For NO2 and OP(DTT), single-pollutant ORs per interquartile range were 1.07 (95% CI: 1.05, 1.09) and 1.08 (95% CI: 1.05, 1.10), respectively. Stratified analysis showed no consistent effect modification by any of the included known diabetes risk factors. Conclusions: Long-term residential air pollution exposure was associated with diabetes prevalence in a large health survey in the Netherlands, strengthening the evidence of air pollution being an important diabetes risk factor. Most consistent associations were observed for NO2 and oxidative potential of PM2.5 measured by the DTT assay. The finding of an association with the oxidative potential of fine particles but not with PM2.5, suggests that particle composition may be important for a potential effect on diabetes.
Article
Full-text available
Background Studies have shown that long-term exposure to air pollution increases mortality. However, evidence is limited for air-pollution levels below the most recent National Ambient Air Quality Standards. Previous studies involved predominantly urban populations and did not have the statistical power to estimate the health effects in underrepresented groups. Methods We constructed an open cohort of all Medicare beneficiaries (60,925,443 persons) in the continental United States from the years 2000 through 2012, with 460,310,521 person-years of follow-up. Annual averages of fine particulate matter (particles with a mass median aerodynamic diameter of less than 2.5 μm [PM2.5]) and ozone were estimated according to the ZIP Code of residence for each enrollee with the use of previously validated prediction models. We estimated the risk of death associated with exposure to increases of 10 μg per cubic meter for PM2.5 and 10 parts per billion (ppb) for ozone using a two-pollutant Cox proportional-hazards model that controlled for demographic characteristics, Medicaid eligibility, and area-level covariates. Results Increases of 10 μg per cubic meter in PM2.5 and of 10 ppb in ozone were associated with increases in all-cause mortality of 7.3% (95% confidence interval [CI], 7.1 to 7.5) and 1.1% (95% CI, 1.0 to 1.2), respectively. When the analysis was restricted to person-years with exposure to PM2.5 of less than 12 μg per cubic meter and ozone of less than 50 ppb, the same increases in PM2.5 and ozone were associated with increases in the risk of death of 13.6% (95% CI, 13.1 to 14.1) and 1.0% (95% CI, 0.9 to 1.1), respectively. For PM2.5, the risk of death among men, blacks, and people with Medicaid eligibility was higher than that in the rest of the population. Conclusions In the entire Medicare population, there was significant evidence of adverse effects related to exposure to PM2.5 and ozone at concentrations below current national standards. This effect was most pronounced among self-identified racial minorities and people with low income. (Supported by the Health Effects Institute and others.)
Article
Full-text available
Biomass burning injects many different gases and aerosols into the atmosphere that could have a harmful effect on air quality, climate, and human health. In this study, a comprehensive biomass burning emission inventory including domestic and in-field straw burning, firewood burning, livestock excrement burning, and forest and grassland fires is presented, which was developed for mainland China in 2012 based on county-level activity data, satellite data, and updated source-specific emission factors (EFs). The emission inventory within a 1 × 1 km² grid was generated using geographical information system (GIS) technology according to source-based spatial surrogates. A range of key information related to emission estimation (e.g. province-specific proportion of domestic and in-field straw burning, detailed firewood burning quantities, uneven temporal distribution coefficient) was obtained from field investigation, systematic combing of the latest research, and regression analysis of statistical data. The established emission inventory includes the major precursors of complex pollution, greenhouse gases, and heavy metal released from biomass burning. The results show that the emissions of SO2, NOx, PM10, PM2.5, NMVOC, NH3, CO, EC, OC, CO2, CH4, and Hg in 2012 are 336.8 Gg, 990.7 Gg, 3728.3 Gg, 3526.7 Gg, 3474.2 Gg, 401.2 Gg, 34 380.4 Gg, 369.7 Gg, 1189.5 Gg, 675 299.0 Gg, 2092.4 Gg, and 4.12 Mg, respectively. Domestic straw burning, in-field straw burning, and firewood burning are identified as the dominant biomass burning sources. The largest contributing source is different for various pollutants. Domestic straw burning is the largest source of biomass burning emissions for all the pollutants considered, except for NH3, EC (firewood), and NOx (in-field straw). Corn, rice, and wheat represent the major crop straws. The combined emission of these three straw types accounts for 80 % of the total straw-burned emissions for each specific pollutant mentioned in this study. As for the straw burning emission of various crops, corn straw burning has the largest contribution to all of the pollutants considered, except for CH4; rice straw burning has highest contribution to CH4 and the second largest contribution to other pollutants, except for SO2, OC, and Hg; wheat straw burning is the second largest contributor to SO2, OC, and Hg and the third largest contributor to other pollutants. Heilongjiang, Shandong, and Henan provinces located in the north-eastern and central-southern regions of China have higher emissions compared to other provinces in China. Gridded emissions, which were obtained through spatial allocation based on the gridded rural population and fire point data from emission inventories at county resolution, could better represent the actual situation. High biomass burning emissions are concentrated in the areas with more agricultural and rural activity. The months of April, May, June, and October account for 65 % of emissions from in-field crop residue burning, while, regarding EC, the emissions in January, February, October, November, and December are relatively higher than other months due to biomass domestic burning in heating season. There are regional differences in the monthly variations of emissions due to the diversity of main planted crops and climatic conditions. Furthermore, PM2.5 component results showed that OC, Cl⁻, EC, K⁺, NH4⁺, elemental K, and SO4²⁻ are the main PM2.5 species, accounting for 80 % of the total emissions. The species with relatively high contribution to NMVOC emission include ethylene, propylene, toluene, mp-xylene, and ethyl benzene, which are key species for the formation of secondary air pollution. The detailed biomass burning emission inventory developed by this study could provide useful information for air-quality modelling and could support the development of appropriate pollution-control strategies.
Article
Full-text available
Biomass burning (BB) is one of the most important contributors to atmospheric aerosols on a global scale, and wildfires are a large source of emissions that impact regional air quality and global climate. As part of the Biomass Burning Observation Project (BBOP) field campaign in summer 2013, we deployed a high-resolution time-of-flight aerosol mass spectrometer (HR-AMS) coupled with a thermodenuder at the Mt. Bachelor Observatory (MBO, ∼ 2.8 km above sea level) to characterize the impact of wildfire emissions on aerosol loading and properties in the Pacific Northwest region of the United States. MBO represents a remote background site in the western US, and it is frequently influenced by transported wildfire plumes during summer. Very clean conditions were observed at this site during periods without BB influence where the 5 min average (±1σ) concentration of non-refractory submicron aerosols (NR-PM1) was 3.7 ± 4.2 µg m⁻³. Aerosol concentration increased substantially (reaching up to 210 µg m⁻³ of NR-PM1) for periods impacted by transported BB plumes, and aerosol composition was overwhelmingly organic. Based on positive matrix factorization (PMF) of the HR-AMS data, three types of BB organic aerosol (BBOA) were identified, including a fresh, semivolatile BBOA-1 (O ∕ C = 0.35; 20 % of OA mass) that correlated well with ammonium nitrate; an intermediately oxidized BBOA-2 (O ∕ C = 0.60; 17 % of OA mass); and a highly oxidized BBOA-3 (O ∕ C = 1.06; 31 % of OA mass) that showed very low volatility with only ∼ 40 % mass loss at 200 °C. The remaining 32 % of the OA mass was attributed to a boundary layer (BL) oxygenated OA (BL-OOA; O ∕ C = 0.69) representing OA influenced by BL dynamics and a low-volatility oxygenated OA (LV-OOA; O ∕ C = 1.09) representing regional aerosols in the free troposphere. The mass spectrum of BBOA-3 resembled that of LV-OOA and had negligible contributions from the HR-AMS BB tracer ions – C2H4O2⁺ (m∕z = 60.021) and C3H5O2⁺ (m∕z = 73.029); nevertheless, it was unambiguously related to wildfire emissions. This finding highlights the possibility that the influence of BB emission could be underestimated in regional air masses where highly oxidized BBOA (e.g., BBOA-3) might be a significant aerosol component but where primary BBOA tracers, such as levoglucosan, are depleted. We also examined OA chemical evolution for persistent BB plume events originating from a single fire source and found that longer solar radiation led to higher mass fraction of the chemically aged BBOA-2 and BBOA-3 and more oxidized aerosol. However, an analysis of the enhancement ratios of OA relative to CO (ΔOA ∕ΔCO) showed little difference between BB plumes transported primarily at night versus during the day, despite evidence of substantial chemical transformation in OA induced by photooxidation. These results indicate negligible net OA production in photochemically aged wildfire plumes observed in this study, for which a possible reason is that SOA formation was almost entirely balanced by BBOA volatilization. Nevertheless, the formation and chemical transformation of BBOA during atmospheric transport can significantly influence downwind sites with important implications for health and climate.
Article
Full-text available
We measured the gas-particle partitioning of vehicle emitted primary organic aerosol (POA) in a traffic tunnel with three independent methods: artifact corrected bare-quartz filters, thermodenuder (TD) measurements, and thermal-desorption gas-chromatography mass-spectrometry (TD-GC-MS). Results from all methods consistently show that vehicle emitted POA measured in the traffic tunnel is semivolatile under a wide range of fleet compositions and ambient conditions. We compared the gas-particle partitioning of POA measured in both tunnel and dynamometer studies and found that volatility distributions measured in the traffic tunnel are similar to volatility distributions measured in the dynamometer studies, and predict similar gas-particle partitioning in the TD. These results suggest that the POA volatility distribution measured in the dynamometer studies can be applied to describe gas-particle partitioning of ambient POA emissions. The POA volatility distribution measured in the tunnel does not have significant diurnal or seasonal variations, which indicate that a single volatility distribution is adequate to describe the gas-particle partitioning of vehicle emitted POA in the urban environment.
Article
Full-text available
Background: Evidence of health risks associated with ambient airborne fine particles in nonurban populations is extremely limited. Objective: We estimate risk of hospitalization associated with short-term exposures to particulate matter with an aerodynamic diameter <2.5? (PM2.5) in urban and nonurban counties with population ?50,000. Methods: We utilize a database of daily cardiovascular- and respiratory-related hospitalization rates constructed from Medicare National Claims History files (2002-2006), including 28 million Medicare beneficiaries in 708 counties. Daily PM2.5 exposures were estimated using the Community Multiscale Air Quality (CMAQ) downscaler. We use time-series analysis of hospitalization rates and PM2.5 to evaluate associations between PM2.5 levels and hospitalization risk in single pollutant models. Results: We observed an association between cardiovascular hospitalizations and same-day PM2.5, with higher risk in urban counties: a 0.35% (95% posterior interval: -0.71%-1.41%) and 0.98% (0.73%-1.23%) increase in hospitalization risk per 10?g/m(3) increment in PM2.5 was observed in the least urban and most urban counties, respectively. The largest association for respiratory hospitalizations, a 2.57% (0.87%-4.30%) increase per 10?g/m(3) increase in PM2.5, was observed in the least urban counties; in the most urban counties, a 1.13% (0.73%-1.54%) increase was observed. Effect estimates for cardiovascular hospitalizations were highest for smaller lag times, while effect estimates for respiratory hospitalizations increased as more days of exposure were included. Conclusion: In nonurban counties with population ?50,000, exposure to PM2.5 is associated with increased risk for respiratory hospitalizations; in urban counties, exposure is associated with increased risk of cardiovascular hospitalizations. Effect estimates based on a single day of exposure may underestimate true effects for respiratory hospitalizations.
Article
Full-text available
Biomass burning injects many different gases and aerosols into the atmosphere, which could have a harmful effect on air quality, climate change and human health. In this study, a comprehensive biomass burning emission inventory including crop straw domestic combustion and in field burning, firewood and livestock excrement combustion, forest and grassland fire was developed for mainland China in 2012 based on county-level activity data and updated source-specific emission factors (EFs). The emission inventory within 1 × 1 km grid was generated using geographical information system (GIS) technology according to source-based spatial surrogates. A range of key information related to emission estimation (e.g., province-specific proportion of crop straw domestic burning and open burning, detailed firewood combustion quantities, uneven temporal distribution coefficient) was obtained from field investigation, systematic combing of the latest research and regression analysis of statistical data. The established emission inventory includes the major precursors of complex pollution, greenhouse gases and heavy metal released from biomass burning. The results show that the emissions of SO2, NOx, PM10, PM2.5, VOC, NH3, CO, EC, OC, CO2, CH4 and Hg in 2012 were 332.8 Gg, 972.5 Gg, 3676.0 Gg, 3479.4 Gg, 3429.6 Gg, 395.8 Gg, 33987.9 Gg, 367.1 Gg, 1151.7 Gg, 665989.0 Gg, 2076.5 Gg and 3.65 Mg, respectively. Indoor and outdoor burning of straw and firewood combustion are identified as the dominant biomass burning sources. The largest contributing source is different for various pollutants. Straw indoor burning is the major source of SO2, CO, CH4 and Hg emission; firewood contributes most to EC and NH3 emission. Corn, rice and wheat represent the major crop straws, with their total emission contribution exceeding 80 % for each pollutant. Corn straw burning has the greatest contribution to EC, NOx and SO2 emissions; rice straw burning is dominant contributor to CO2, VOC, CH4 and NH3 emissions. Heilongjiang, Shandong, and Henan provinces located in northeast and central-south region of China have higher emissions. Gridded emissions, which were obtained through spatial allocation based on the gridded rural population and fire point data from emission inventory at county resolution, could better represent the actual situation. Higher biomass burning emissions are concentrated in the areas with greater agricultural and rural activity. The temporal distribution shows that higher emissions occurred in April, September, and October during the whole year. There’s regional difference in monthly variation due to the diversity of main planted crop and the climate conditions. Furthermore, PM2.5 component results showed that OC, Cl−, EC, K+, NH4+, K element and SO42− are the main PM2.5 species accounting for 80 % of the total emissions. The species with relatively higher contribution to VOCs emission including ethylene, propylene, toluene, mp-xylene and halocarbons which are key species for the formation of secondary air pollution. The detailed biomass burning emission inventory generated by this study could provide useful information for air quality modelling and support the development of appropriate pollution control strategies.
Article
Full-text available
Mobile monitoring of traffic-related air pollutants was conducted in Pittsburgh, PA. The data show substantial spatial variability of particle-bound polycyclic aromatic hydrocarbons (PB-PAH) and black carbon (BC). This variability is driven in large part by pollutant plumes from high emitting vehicles (HEVs). These plumes contribute a disproportionately large fraction of the near-road exposures of PB-PAH and BC. We developed novel statistical models to describe the spatial patterns of PB-PAH and BC exposures. The models consist of two layers: a plume layer to describe the contributions of high emitting vehicles using a near-roadway kernel, and an urban-background layer that predicts the spatial pattern of other sources using land use regression. This approach leverages unique information content of highly time resolved mobile monitoring data and provides insight into source contributions. The two-layer model describes 76% of observed PB-PAH variation and 61% of BC variation. On average, HEVs contribute at least 32% of outdoor PB-PAH and 14% of BC. The transferability of the models was examined using measurements from 36 hold-out validation sites. The plume layer performed well at validation sites, but the background layer showed little transferability due to the large difference in land use between the city and outer suburbs.
Article
Full-text available
Airborne particulate matter has been associated with cardiovascular and respiratory morbidity and mortality, and there is evidence that metals may contribute to these adverse health effects. However, there are few tools for assessing exposure to airborne metals. Land-use regression modeling has been widely used to estimate exposure to gaseous pollutants. This study developed seasonal land-use regression (LUR) models to characterize the spatial distribution of trace metals and other elements associated with airborne particulate matter in Calgary, Alberta.
Article
Full-text available
The U.S. Environmental Protection Agency (EPA) initiated the national PM2.5 Chemical Speciation Monitoring Network (CSN) in 2000 to support evaluation of long-term trends and to better quantify the impact of sources on particulate matter (PM) concentrations in the size range below 2.5 μm aerodynamic diameter (PM2.5; fine particles). The network peaked at more than 260 sites in 2005. In response to the 1999 Regional Haze Rule and the need to better understand the regional transport of PM, EPA also augmented the long-existing Interagency Monitoring of Protected Visual Environments (IMPROVE) visibility monitoring network in 2000, adding nearly 100 additional IMPROVE sites in rural Class 1 Areas across the country. Both networks measure the major chemical components of PM2.5 using historically accepted filter-based methods. Components measured by both networks include major anions, carbonaceous material, and a series of trace elements. CSN also measures ammonium and other cations directly, whereas IMPROVE estimates ammonium assuming complete neutralization of the measured sulfate and nitrate. IMPROVE also measures chloride and nitrite. In general, the field and laboratory approaches used in the two networks are similar; however, there are numerous, often subtle differences in sampling and chemical analysis methods, shipping, and quality control practices. These could potentially affect merging the two data sets when used to understand better the impact of sources on PM concentrations and the regional nature and long-range transport of PM2.5. This paper describes, for the first time in the peer-reviewed literature, these networks as they have existed since 2000, outlines differences in field and laboratory approaches, provides a summary of the analytical parameters that address data uncertainty, and summarizes major network changes since the inception of CSN.
Article
Full-text available
While the association between PM2.5 mass and mortality has been extensively studied, few national-level analyses have estimated mortality effects of PM2.5 chemical constituents. Epidemiological studies have reported that estimated effects of PM2.5 on mortality vary spatially and seasonally. We hypothesized that associations between PM2.5 constituents and mortality would not vary spatially or seasonally if variation in chemical composition contributes to variation in estimated PM2.5 mortality effects. We aim to provide the first national, season-specific, and region-specific associations between mortality effects of PM2.5 constituents. We estimated short-term associations between non-accidental mortality and PM2.5 constituents across 72 urban U.S. communities from 2000-2005. Using U.S. EPA Chemical Speciation Network data, we analyzed seven constituents that together compose 79-85% of PM2.5 mass: ammonium ion, elemental carbon (EC), nitrate, organic carbon matter (OCM), silicon, sodium ion, and sulfate. We applied Poisson time-series regression models controlling for time and weather to estimate mortality effects. Interquartile range increases in OCM, EC, silicon, and sodium ion were associated with estimated increases in mortality of 0.39% (95% Posterior Interval (PI): 0.08, 0.70%), 0.22% (95% PI: 0.00, 0.44%), 0.17% (95% PI: 0.03, 0.30%), and 0.16% (95% PI: 0.00, 0.32%), respectively, based on single pollutant models. We did not find evidence that associations between mortality and PM2.5 or PM2.5 constituents differed by season or region. Our work indicates that some constituents of PM2.5 may be more toxic than others and therefore regulating PM total mass alone may not be sufficient to protect human health.
Article
Full-text available
Experiments were performed to investigate the gas-particle partitioning of primary organic aerosol (POA) emissions from two medium-duty (MDDV) and three heavy-duty (HDDV) diesel vehicles. Each test was conducted on a chassis dynamometer with the entire exhaust sampled into a constant volume sampler (CVS). The vehicles were operated over a range of driving cycles (transient, high-speed, creep/idle) on different ultralow sulfur diesel fuels with varying aromatic content. Four independent yet complementary approaches were used to investigate POA gas-particle partitioning: artifact correction of quartz filter samples, dilution from the CVS into a portable environmental chamber, heating in a thermodenuder, and thermal desorption/gas chromatography/mass spectrometry (TD-GC-MS) analysis of quartz filter samples. During tests of vehicles not equipped with diesel particulate filters (DPF), POA concentrations inside the CVS were a factor of ten greater than ambient levels, which created large and systematic partitioning biases in the emissions data. For low-emitting DPF-equipped vehicles, as much as 90% of the POA collected on a quartz filter from the CVS were adsorbed vapors. Although the POA emission factors varied by more the order of magnitude across the set of test vehicles, the measured gas-particle partitioning of all emissions can be predicted using a single volatility distribution derived from TD-GC-MS analysis of quartz filters. This distribution is designed to be applied directly to quartz filter data that are the basis for existing emissions inventories and chemical transport models that have implemented the volatility basis set approach.
Article
Full-text available
Field blanks (bQF) and backup filters (quartz-fiber behind quartz-fiber filter; QBQ) have been adopted by US long-term air quality monitoring networks to estimate PM2.5 organic carbon (OC) sampling artifacts. This study documents bQF and QBQ carbon levels for the: 1) Interagency Monitoring of Protected Visual Environments (IMPROVE); 2) Speciation Trends Network (STN; part of the Chemical Speciation Network [CSN]); and 3) Southeastern Aerosol Research and Characterization (SEARCH) networks and examines the similarities/ differences associated with network-specific sampling protocols. A higher IMPROVE sample volume and smaller filter deposit area results in PM 2.5 areal density (μg/cm2 on filter) 3-11 times those of STN/CSN and SEARCH samples for the same ambient PM2.5 concentrations, thus reducing the relative contribution of sampling artifacts from passive OC adsorption. A relatively short (1-15 min) passive exposure period of STN/CSN and SEARCH bQF OC (0.8-1 μg/cm2) underestimates positive and negative OC artifacts resulting from passive adsorption or evaporation of semi-volatile organic compounds on quartz-fiber filters. This is supported by low STN/CSN and SEARCH bQF levels and lack of temporal or spatial variability among the sites within the networks. With a much longer period, ∼7 days of ambient passive exposure, average IMPROVE bQF and QBQ OC are comparable (2.4±0.5 and 3.1±0.8 μg/cm2, respectively) and more than twice levels found in the STN/CSN and SEARCH networks. Sampling artifacts in STN/CSN were estimated from collocated IMPROVE samples based on linear regression. At six of the eight collocated sites in this study, STN/CSN bQFs underestimated OC artifacts by 11-34%. Using a preceding organic denuder in the SEARCH network minimized passive adsorption on QBQ, but OC on QBQ may not be attributed entirely to the negative sampling artifact (e.g., evaporated or volatilized OC from the front filter deposits after sample collection).
Article
Full-text available
Organic carbon (OC) and elemental carbon (EC) are operationally defined by the analysis methods, and different methods give in different results. The IMPROVE Iinteragency Monitoring of protected Visual Environments) and NIOSH (National Institute of Occupational Safety and Wealth) thermal evolution protocols present different operational definitions. These protocols are applied to 60 ambient and sonrce samples from different environments using the same instrument to quantify differences in implemented protocols on the same instrument. The protocols are equivalent for total carbon sampled on quartz-fiber filters. NIOSH EC was typically less than half of IMPROVE EC. The primary difference is the allocation of carbon evolving at the NIOSH 850 degrees C temperature in a helium atmosphere to the OC rather than EC fraction. increasing light transmission and reflectance during this temperature step indicate that this fraction should he classified as EC. When this portion of NIOSH OC is added to NIOSH EC, the IMPROVE and NIOSH analyses are in good agreement. The most probable explanation is that mineral oxides in the complex particle mixture on the filter are supplying oxygen to neighboring carbon particles at this high temperature. This has been demonstrated by the principle of the thermal manganese oxidation method that is also commonly used to distinguish OC from EC, For both methods, the optical pyrolysis adjustment to the EC fractions was always higher for transmittance than for reflectance. This is a secondary cause of differences between the two methods, with transmittance resulting in a lower EC loading than reflectance. The difference was most pronounced for very black filters on which neither reflectance nor transmittance accurately detected further blackening due to pyrolysis.
Article
Full-text available
Integrated, 24-hour, ambient PM2.5 (particulate matter <2.5 μm in aerodynamic diameter) samples were collected at a rural monitoring site in Brigantine, New Jersey, on Wednesdays and Saturdays using Interagency Monitoring of Protected Visual Environments ( IMPROVE) samplers. Particulate carbon was analyzed using the thermal optical reflectance method, which divides carbon into four organic carbon (OC), pyrolyzed organic carbon (OP), and three elemental carbon (EC) fractions. A total of 910 samples and 36 variables collected between March 1992 and May 2001 were analyzed using positive matrix factorization, and 11 sources were identified: sulfate-rich secondary aerosol I (48%), gasoline vehicle (13%), aged sea salt (8%), sulfate-rich secondary aerosol II (7%), nitrate-rich secondary aerosol (6%), sulfate-rich secondary aerosol III (5%), sea salt (4%), airborne soil (4%), diesel emission (3%), incinerator (2%), and oil combustion (1%). Temperature-resolved carbon fractions enhanced source separations including three sulfate-rich secondary aerosols and two traffic-related sources that had different abundances of carbon fractions different between sources. Conditional probability functions using surface wind data and deduced source contributions aid in the identification of local sources. Potential source contribution function (PSCF) analysis shows the regional influence of sulfate-rich secondary aerosols. Backward trajectories indicate that the highly elevated airborne soil impacts at the monitoring site were likely caused by either Asian or Sahara dust storms.
Article
Full-text available
Measurement of ambient particulate organic carbon (POC) with quartz filters is prone to positive and negative sampling artifacts. One approach for estimating these artifacts is to sample with a backup quartz filter placed behind either the main quartz filter or a Teflon filter in a parallel line. Another approach is to use a denuder to reduce the positive artifact in combination with a highly adsorbent backup filter to capture any negative artifact. Results obtained using both of these approaches in parallel for over one year in Pittsburgh, PA are presented in this article. A sampler using an activated carbon monolith denuder has been developed and tested extensively. Transmission losses were found to be negligible, and the denuder is on average 94% efficient at removing gas-phase organics. Denuder breakthrough is corrected for each run using a dynamic blank in parallel with the sample line. Comparisons with the dynamic blank indicate that the denuder almost eliminates the positive artifact on the quartz filter. Negative artifact from the denuded quartz filter is quantified using a carbon-impregnated glass fiber (CIG) backup filter and was found to be small, typically less than 10% of the ambient POC. Compared to the denuded sampler POC, 24 h bare quartz samples showed an almost constant positive artifact of 0.5 μg-C/m3 for samples taken throughout the year-long study period. Sampling for shorter durations (4–6 h) resulted in a larger positive artifact. A quartz filter behind a Teflon filter (QBT) provides a consistent estimate of the positive artifact on the bare quartz filter irrespective of sample duration, though it overcorrects for the positive artifact by 16–20% (attributed to particulate matter volatilizing off the upstream Teflon filter). The quartz behind quartz (QBQ) approach provides a reasonable estimate of the positive artifact on the bare quartz filter for the 24 h samples but not for the shorter samples. A slight seasonal variation is observed in the absolute value of the positive artifact, with higher values observed during the summer months.
Article
Full-text available
Quantitative measurements of environmental factors greatly improve the quality of epidemiologic studies but can pose challenges because of the presence of upper or lower detection limits or inter- fering compounds, which do not allow for precise measured values. We consider the regression of an environmental measurement (dependent variable) on several covariates (independent variables). Various strategies are commonly employed to impute values for interval-measured data, including assignment of one-half the detection limit to nondetected values or of "fill-in" values randomly selected from an appropriate distribution. On the basis of a limited simulation study, we found that the former approach can be biased unless the percentage of measurements below detection limits is small (5-10%). The fill-in approach generally produces unbiased parameter estimates but may pro- duce biased variance estimates and thereby distort inference when 30% or more of the data are below detection limits. Truncated data methods (e.g., Tobit regression) and multiple imputation offer two unbiased approaches for analyzing measurement data with detection limits. If interest resides solely on regression parameters, then Tobit regression can be used. If individualized values for measurements below detection limits are needed for additional analysis, such as relative risk regression or graphical display, then multiple imputation produces unbiased estimates and nominal confidence intervals unless the proportion of missing data is extreme. We illustrate various
Article
Full-text available
Efforts to understand and mitigate the health effects of particulate matter (PM) air pollution have a rich and interesting history. This review focuses on six substantial lines of research that have been pursued since 1997 that have helped elucidate our understanding about the effects of PM on human health. There has been substantial progress in the evaluation of PM health effects at different timescales of exposure and in the exploration of the shape of the concentration-response function. There has also been emerging evidence of PM-related cardiovascular health effects and growing knowledge regarding interconnected general pathophysiological pathways that link PM exposure with cardiopulmonary morbidity and mortality. Despite important gaps in scientific knowledge and continued reasons for some skepticism, a comprehensive evaluation of the research findings provides persuasive evidence that exposure to fine particulate air pollution has adverse effects on cardiopulmonary health. Although much of this research has been motivated by environmental public health policy, these results have important scientific, medical, and public health implications that are broader than debates over legally mandated air quality standards.
Article
To assess the public health benefits of reducing PM2.5 in Tianjin, we created an annual air quality surface with a land use regression (LUR) model conducted at a high spatial resolution (1 km). The predictors included in the final model were population density, road length within a 1000 m buffer, industrial land area within a 2000m buffer and distance to the coast. The fitting R² and the leave-one-out-cross-validation (LOOCV) R² of the PM2.5 LUR models were 0.78 and 0.73, respectively, suggesting that the predicted PM2.5 concentrations fitted well with the measured values for the entire year. Daily air quality surfaces were established based on historic concentration data and interpolation method. We evaluated avoided cases of mortality and morbidity in Tianjin, assuming achievement of China's current air quality daily and annual standards (No. GB3095-2012). Reducing the daily average PM2.5 to the daily Class II standard (75 μg/m³), the avoided emergency department visits, the deaths for cardiovascular disease and the deaths for respiratory disease are 85,000 (95% confidence interval (CI), 17,000–150,000), 2000 (95% CI, 920–3100) and 280 (95% CI, 94–460) per year respectively, and the monetary values are 23–42 million yuan, 180–4800 million yuan and 25–670 million yuan per year in 2015 yuan year respectively. Reducing the annual average PM2.5 to the annual Class II standard (35 μg/m³), the avoided emergency department visits, the deaths for cardiovascular disease and the deaths for respiratory disease are 59,000 (95% CI, 12,000–110,000), 1400 (95% CI, 640–2100) and 200 (95% CI, 66–320) per year respectively, and the monetary values are 16–29 million yuan, 130 to 3400 million yuan and 18 to 480 million yuan per year in 2015 yuan year respectively.
Article
Exposure assessment for elemental components of particulate matter (PM) using land use modeling is a complex problem due to the high spatial and temporal variations in pollutant concentrations at the local scale. Land use regression (LUR) models may fail to capture complex interactions and non-linear relationships between pollutant concentrations and land use variables. The increasing availability of big spatial data and machine learning methods present an opportunity for improvement in PM exposure assessment models. In this manuscript, our objective was to develop a novel land use random forest (LURF) model and compare its accuracy and precision to a LUR model for elemental components of PM in the urban city of Cincinnati, Ohio. PM smaller than 2.5 (PM2.5) and eleven elemental components were measured at 24 sampling stations from the Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS). Over 50 different predictors associated with transportation, physical features, community socioeconomic characteristics, greenspace, land cover, and emission point sources were used to construct LUR and LURF models. Cross validation was used to quantify and compare model performance. LURF and LUR models were created for aluminum (Al), copper (Cu), iron (Fe), potassium (K), manganese (Mn), nickel (Ni), lead (Pb), sulfur (S), silicon (Si), vanadium (V), zinc (Zn), and total PM2.5 in the CCAAPS study area. LURF utilized a more diverse and greater number of predictors than LUR and LURF models for Al, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all showed a decrease in fractional predictive error of at least 5% compared to their LUR models. LURF models for Al, Cu, Fe, K, Mn, Pb, Si, Zn, TRAP, and PM2.5 all had a cross validated fractional predictive error less than 30%. Furthermore, LUR models showed a differential exposure assessment bias and had a higher prediction error variance. Random forest and other machine learning methods may provide more accurate exposure assessment.
Article
Atmospheric chemistry dominates the size distribution and composition of most fine particles inhaled by humans. However, it is important to distinguish between secondary particles - new particles formed in the atmosphere - and secondary mass - molecules formed in the atmosphere that condense to existing particles. In many ways the life stories of particles viewed from the perspectives of particle number concentrations and particle mass concentrations are distinct. Individual particle cores can often be said to have an individual source, while the mass on individual particles comes from myriad sources. This, plus the aforementioned chemical processing in the atmosphere, must be kept in mind when considering the health effects of fine particles.
Article
Organic aerosol (OA) can absorb solar radiation in the low-visible and ultra-violet wavelengths thereby modifying radiative forcing. Agricultural waste burning emits a large quantity of organic carbon in many developing countries. In this work, we improved the extraction and analysis method developed by Chen and Bond, and extended the spectral range of OC absorption. We examined light absorbing properties of primary OA from pyrolysis of corn stalk, which is a major type of agricultural wastes. Light absorption of bulk liquid extracts of OA was measured using a UV–vis recording spectrophotometer. OA can be extracted by methanol at 95%, close to full extent, and shows polar character. Light absorption of organic aerosol has strong spectral dependence (Absorption Ångström exponent = 7.7) and is not negligible at ultra-violet and low-visible regions. Higher pyrolysis temperature produced OA with higher absorption. Imaginary refractive index of organic aerosol (kOA) is 0.041 at 400 nm wavelength and 0.005 at 550 nm wavelength, respectively.
Article
Monitoring street-level particulates is essential to air quality management but challenging in high-density Hong Kong due to limitations in local monitoring network and the complexities of street environment. By employing vehicle-based mobile measurements, Land Use Regression (LUR) models are developed to estimate the spatial variation of PM2.5 and PM10 in the downtown area of Hong Kong. Sampling runs were conducted along routes measuring a total of 30km during selected measurement period of total 14 days. In total, 321 independent variables were examined to develop LUR models by using stepwise regression with PM2.5 and PM10 as dependent variables. Approximately, 10% increases in the model adjusted R(2) were achieved by integrating urban/building morphology as independent variables into the LUR models. Resultant LUR models show that the most decisive factors on street-level air quality in Hong Kong are frontal area index, an urban/building morphological parameter, and road network line density and traffic volume, two parameters of road traffic. The adjusted R(2) of the final LUR models of PM2.5 and PM10 are 0.633 and 0.707 respectively. These results indicate that urban morphology is more decisive to the street-level air quality in high-density cities than other cities. Air pollution hotspots are also identified based on the LUR mapping.
Article
Background: Short-term exposure to ambient PM2.5 concentrations has been associated with increased mortality and morbidity. Determining which sources of PM2.5 are most toxic can help guide targeted reduction of PM2.5. However, conducting multicity epidemiologic studies of sources is difficult because source-specific PM2.5 is not directly measured and source chemical compositions can vary between cities. Objectives: We determine how the chemical composition of primary ambient PM2.5 sources varies across cities. We estimate associations between source-specific PM2.5 and respiratory disease emergency department (ED) visits and examine between-city heterogeneity in estimated associations. Methods: We used source apportionment to estimate daily concentrations of primary source-specific PM2.5 for four US cities. For sources with similar chemical compositions between cities, we applied Poisson time-series regression models to estimate associations between source-specific PM2.5 and respiratory disease ED visits. Results: We found biomass burning, diesel vehicle, gasoline vehicle, and dust PM2.5 was similar in chemical composition between cities, but PM2.5 composition from coal combustion and metal sources varied across cities. We found some evidence of positive associations of respiratory disease ED visits with biomass burning PM2.5; associations with diesel and gasoline PM2.5 were frequently imprecise or consistent with the null. We found little evidence of associations with dust PM2.5. Conclusions: We introduced an approach for comparing chemical compositions of PM2.5 sources across cities and conducted one of the first multicity studies of source-specific PM2.5 and ED visits. Across four US cities, among the primary PM2.5 sources assessed, biomass burning PM2.5 was most strongly associated with respiratory health.
Article
Dynamometer experiments were conducted to characterize the intermediate volatility organic compound (IVOC) emissions from a fleet of on-road gasoline vehicles and small off-road gasoline engines. IVOCs were quantified through gas chromatography/mass spectrometry analysis of adsorbent samples collected from a constant volume sampler. The dominant fraction (>80%, on average) of IVOCs could not be resolved on a molecular level. These unspeciated IVOCs were quantified as two chemical classes (unspeciated branched alkanes and cyclic compounds) in 11 retention-time-based bins. IVOC emission factors (mg kg-fuel-1) from on-road vehicles varied widely from vehicle to vehicle, but showed a general trend of lower emissions for newer vehicles that meet more stringent emission standards. IVOC emission factors for 2-stroke off-road engines were substantially higher than 4-stroke off-road engines and on-road vehicles. Despite large variations in the magnitude of emissions, the IVOC volatility distribution and chemical characteristics were consistent across all tests and IVOC emissions were strongly correlated with non-methane hydrocarbons (NMHCs), primary organic aerosol and speciated IVOCs. Although IVOC emissions only correspond to approximately 4% of NMHC emissions from on-road vehicles over the cold-start unified cycle, they are estimated to produce as much or more SOA than single-ring aromatics. Our results clearly demonstrate that IVOCs from gasoline engines are an important class of SOA precursors and provide observational constraints on IVOC emission factors and chemical composition to facilitate their inclusion into atmospheric chemistry models.
Article
We introduce new method to estimate the mass concentration of particulate organic carbon (POC) collected on quartz filters, demonstrating it using quartz-filter samples collected in greater Pittsburgh. This method combines thermal-optical organic carbon and elemental carbon (OC/EC) analysis and the volatility basis set (VBS) to quantify the mass concentration of semi-volatile POC on the filters. The dataset includes ambient samples collected at a number of sites in both summer and winter as well as samples from a highway tunnel. As a reference we use the two-filter bare-Quartz minus Quartz-Behind-Teflon (Q-QBT) approach to estimate the adsorbed gaseous fraction of organic carbon (OC), finding a substantial fraction in both the gas and particle phases under all conditions. In the new method we use OC fractions measured during different temperature stages of the OC/EC analysis for the single bare-quartz filter in combination with partitioning theory to predict the volatility distributions of the measured OC, which we describe with the VBS. The effective volatility bins are consistent for data from both ambient samples and primary organic aerosol (POA)-enriched tunnel samples. Consequently, with the VBS model and total OC fractions measured over different heating stages, particulate OC can be determined by using the BQ filter alone. This method can thus be applied to all quartz filter-based OC/EC analyses to estimate the POC concentration without using backup filters.
Article
Exposure to ambient air pollution is a major risk factor for global disease. Assessment of the impacts of air pollution on population health and the evaluation of trends relative to other major risk factors requires regularly updated, accurate, spatially resolved exposure estimates. We combined satellite-based estimates, chemical transport model simulations and ground measurements from 79 different countries to produce new global estimates of annual average fine particle (PM2.5) and ozone concentrations at 0.1° × 0.1° spatial resolution for five-year intervals from 1990-2010 and the year 2013. These estimates were then applied to assess population-weighted mean concentrations for 1990 – 2013 for each of 188 countries. In 2013, 87% of the world’s population lived in areas exceeding the World Health Organization (WHO) Air Quality Guideline of 10 μg/m3 PM2.5 (annual average). Between 1990 and 2013, decreases in population-weighted mean concentrations of PM2.5 were evident in most high income countries, in contrast to increases estimated in South Asia, throughout much of Southeast Asia, and in China. Population-weighted mean concentrations of ozone increased in most countries from 1990 - 2013, with modest decreases in North America, parts of Europe, and several countries in Southeast Asia.
Article
Emissions of intermediate-volatility organic compounds (IVOCs) from five on-road diesel vehicles and one off-road diesel engine were characterized during dynamometer testing. The testing evaluated the effects of driving cycles, fuel composition and exhaust aftertreatment devices. On average, more than 90% of the IVOC emissions were not identified on a molecular basis, instead appearing as an unresolved complex mixture (UCM) during gas-chromatography mass-spectrometry analysis. Fuel-based emissions factors (EFs) of total IVOCs (speciated + unspeciated) depend strongly on aftertreatment technology and drive cycle. Total-IVOC emissions from vehicles equipped with catalyzed diesel particulate filters (DPF) are substantially lower (factor of 7 to 28, depending on driving cycle) than from vehicles without any aftertreatment (non-aftertreatment vehicles). Total-IVOC emissions from creep and idle operations are substantially higher than emissions from high-speed operations. Although the magnitude of the total-IVOC emissions can vary widely, there is little variation in the IVOC composition across the set of tests. The new emissions data are combined with published yield data to investigate secondary organic aerosol (SOA) formation. SOA production from unspeciated IVOCs is estimated using surrogate compounds, which are assigned based on gas-chromatograph retention time and mass spectral signature of the IVOC UCM. IVOCs contribute the vast majority of the SOA formed from exhaust from on-road diesel vehicles. The estimated SOA production is greater than predictions by previous studies and substantially higher than POA. Catalyzed DPFs substantially reduce SOA formation potential of diesel exhaust, except at low speed operations.
Article
Land Use Regression (LUR) models typically use fixed-site monitoring; here, we employ mobile monitoring as a cost-effective alternative for LUR-development. We use bicycle-based, mobile measurements (~85 hours) during rush-hour in Minneapolis, MN to build LUR models for particulate concentrations (particle number [PN], black carbon [BC], fine particulate matter [PM2.5], particle size). We developed and examined 1,224 separate LUR models by varying pollutant, time-of-day, and method of spatial and temporal smoothing of the time-series data. Our base-case LUR models had modest goodness-of-fit (adjusted R2: ~0.5 [PN], ~0.4 [PM2.5], 0.35 [BC], ~0.25 [particle size]) and included predictor variables which captured proximity to and density of emission sources. The spatial density of our measurements resulted in a large model-building dataset (n=1,101 concentration estimates); ~25% of buffer variables were selected at spatial scales of <100m, suggesting that on-road particle concentrations change on small spatial scales. LUR model-R2 improved as sampling runs were completed, with diminishing benefits after ~40 hours of data collection. Spatial autocorrelation of model residuals indicated that models performed poorly where spatiotemporal resolution of emission sources (i.e., traffic congestion) was poor. Our findings suggest that LUR modeling from mobile measurements is possible, but that more work could usefully inform best practices.
Article
Land Use Regression (LUR) models have been used increasingly for modeling small-scale spatial variation in air pollution concentrations and estimating individual exposure for participants of cohort studies. Within the ESCAPE project, concentrations of PM2.5, PM2.5 absorbance, PM10, and PMcoarse were measured in 20 European study areas at 20 sites per area. GIS-derived predictor variables (e.g., traffic intensity, population, and land-use) were evaluated to model spatial variation of annual average concentrations for each study area. The median model explained variance (R-2) was 71% for PM2.5 (range across study areas 35-94%). Model R-2 was higher for PM2.5 absorbance (median 89%, range 56-97%) and lower for PMcoarse (median 68%, range 32-81%). Models included between two and five predictor variables, with various traffic indicators as the most common predictors. Lower R-2 was related to small concentration variability or limited availability of predictor variables, especially traffic intensity. Cross validation R-2 results were on average 8-11% lower than model R-2. Careful selection of monitoring sites, examination of influential observations and skewed variable distributions were essential for developing stable LUR models. The final LUR models are used to estimate air pollution concentrations at the home addresses of participants in the health studies involved in ESCAPE.
Article
We compare the relative toxicity of various organic aerosol (OA) components identified by an Aerosol Mass Spectrometer (AMS) based on their ability to generate reactive oxygen species (ROS). Ambient fine aerosols were collected from urban (three in Atlanta, GA and one in Birmingham, AL) and rural (Yorkville, GA and Centerville, AL) sites in the Southeastern United States. The ROS generating capability of the water-soluble fraction of the particles was measured by the dithiothreitol (DTT) assay. Water-soluble PM extracts were further separated into the hydrophobic and hydrophilic fractions using a C-18 column, and both fractions were analyzed for DTT activity and water-soluble metals. Organic aerosol composition was measured at selected sites using a High-Resolution Time-of-Flight AMS. Positive matrix factorization of the AMS spectra resolved the organic aerosol into isoprene-derived OA (Isop_OA), hydrocarbon-like OA (HOA), less-oxidized oxygenated OA, (LO-OOA), more-oxidized OOA (MO-OOA), cooking OA (COA), and biomass burning OA (BBOA). The association of the DTT activity of water-soluble PM2.5 (WS_DTT) with these factors was investigated by linear regression techniques. BBOA and MO-OOA were most consistently linked with WS_DTT, with intrinsic water-soluble activities of 151±20 and 36±22 pmol/min/μg, respectively. Although less toxic, MO-OOA was most widespread, contributing to WS_DTT activity at all sites and during all seasons. WS_DTT activity was least associated with biogenic secondary organic aerosol. The OA components contributing to WS_DTT were humic-like substances (HULIS), which are abundantly emitted in biomass burning (BBOA) and include highly oxidized OA from multiple sources (MO-OOA). Overall, OA contributed approximately 60% to the WS_DTT activity, with the remaining probably from water-soluble metals, which were mostly associated with the hydrophilic WS_DTT fraction.
Article
We used a mobile measurement platform to characterize a suite of air pollutants (black carbon (BC), particle-bound polycyclic aromatic hydrocarbons (PB-PAH), benzene, and toluene) in the city of Pittsburgh and surrounding areas. More than 270 hours of data were collected from forty-two sites which were selected based on analysis in the geographic information system (GIS). Mobile measurements were performed during three different times of day (mornings, afternoons/evenings, and overnight) in both winter (Nov 2011 - Feb 2012) and summer (Jun 2012 - Aug 2012). Pollutant concentrations were elevated in river valleys by 9% (benzene) to 30% (PB-PAH) relative to upland areas. Traffic had strong impacts on measured pollutants. PB-PAH and BC concentrations at high traffic sites were a factor of two and 30% higher than at low traffic sites, respectively. Pollutant concentrations were highest in the morning sessions due to a combination of traffic and meteorological conditions. The highly time-resolved data indicated that elevated pollutant concentrations at high traffic sites were due to short duration plume events associated with high emitting vehicles. High emitting vehicles contributed up to 70% of the near road PB-PAH and 30% of BC; emissions from these vehicles drove substantial spatial variations in BC and PB-PAH concentrations. Many high emitting vehicles were presumably diesel trucks or buses, because plumes were strongly correlated with truck traffic volume. In contrast, PB-PAH and BC in the non-plume background air was weakly correlated with traffic, and their spatial patterns were more influenced by terrain and point source emissions. The spatial variability in contributions of high emitting vehicles suggests that the effect of potential control strategies vary for different pollutants and environments.
Article
Land use regression (LUR) models have been used to model concentrations of mainly traffic related air pollutants (nitrogen oxides (NOx), particulate matter (PM) mass or absorbance). Few LUR models are published of PM composition, whereas the interest in health effects related to particle composition is increasing. The aim of our study was to evaluate LUR models of polycyclic aromatic hydrocarbons (PAH), hopanes/steranes and elemental and organic carbon (EC/OC) content of PM2.5. In 10 European study areas PAH, hopanes/steranes and EC/OC concentrations were measured at 16-40 sites per study area. LUR models for each study area were developed based on annual average concentrations and predictor variables including traffic, population, industry, natural land obtained from geographic information systems. The highest median model explained variance (R2) was found for EC - 84%. The median R2 was 51% for OC, 67% for benzo[a]pyrene and 38% for sum of hopanes/steranes, with large variability between study areas. Traffic predictors were included in most models. Population and natural land were included frequently as additional predictors. The moderate to high explained variance of LUR models and the overall moderate correlation with PM2.5 model predictions support the application of especially the OC and PAH models in epidemiological studies.
Article
Routine air monitoring provides accurate measurements of annual average concentrations of air pollutants, but the low density of monitoring sites limits its capability in capturing intra-urban variation. Pollutant mapping studies measure air pollutants at a large number of sites during short periods. However, their short duration can cause substantial uncertainty in reproducing annual mean concentrations. In order to quantify this uncertainty for existing sampling strategies and investigate methods to improve future studies, we conducted Monte Carlo experiments with nationwide monitoring data from the EPA Air Quality System. Typical fixed sampling designs have much larger uncertainties than previously assumed, and produce accurate estimates of annual average pollution concentrations approximately 80% of the time. Mobile sampling has difficulties in estimating long-term exposures for individual sites, but performs better for site groups. The accuracy and the precision of a given design decrease when data variation increases, indicating challenges in sites intermittently impact by local sources such as traffic. Correcting measurements with reference sites does not completely remove the uncertainty associated with short duration sampling. Using reference sites with the addition method can better account for temporal variations than the multiplication method. We propose feasible methods for future mapping studies to reduce uncertainties in estimating annual mean concentrations. Future fixed sampling studies should conduct two separate 1-week long sampling periods in all 4 seasons. Mobile sampling studies should estimate annual mean concentrations for exposure groups with five or more sites. Fixed and mobile sampling designs have comparable probabilities in ordering two sites, so they may have similar capabilities in predicting pollutant spatial variations. Simulated sampling designs have large uncertainties in reproducing seasonal and diurnal variations at individual sites, but are capable to predict these variations for exposure groups.
Article
Few studies on long-term exposure to air pollution and mortality have been reported from Europe. Within the multicentre European Study of Cohorts for Air Pollution Effects (ESCAPE), we aimed to investigate the association between natural-cause mortality and long-term exposure to several air pollutants. We used data from 22 European cohort studies, which created a total study population of 367 251 participants. All cohorts were general population samples, although some were restricted to one sex only. With a strictly standardised protocol, we assessed residential exposure to air pollutants as annual average concentrations of particulate matter (PM) with diameters of less than 2·5 μm (PM2·5), less than 10 μm (PM10), and between 10 μm and 2·5 μm (PMcoarse), PM2.5 absorbance, and annual average concentrations of nitrogen oxides (NO2 and NOx), with land use regression models. We also investigated two traffic intensity variables-traffic intensity on the nearest road (vehicles per day) and total traffic load on all major roads within a 100 m buffer. We did cohort-specific statistical analyses using confounder models with increasing adjustment for confounder variables, and Cox proportional hazards models with a common protocol. We obtained pooled effect estimates through a random-effects meta-analysis. The total study population consisted of 367 251 participants who contributed 5 118 039 person-years at risk (average follow-up 13·9 years), of whom 29 076 died from a natural cause during follow-up. A significantly increased hazard ratio (HR) for PM2·5 of 1·07 (95% CI 1·02-1·13) per 5 μg/m(3) was recorded. No heterogeneity was noted between individual cohort effect estimates (I(2) p value=0·95). HRs for PM2·5 remained significantly raised even when we included only participants exposed to pollutant concentrations lower than the European annual mean limit value of 25 μg/m(3) (HR 1·06, 95% CI 1·00-1·12) or below 20 μg/m(3) (1·07, 1·01-1·13). Long-term exposure to fine particulate air pollution was associated with natural-cause mortality, even within concentration ranges well below the present European annual mean limit value. European Community's Seventh Framework Program (FP7/2007-2011).
Article
Air pollution in New Delhi, India is a significant environmental and health concern. To assess determinants of variability in air pollutant concentrations we develop land use regression (LUR) models for fine particulate matter (PM2.5), black carbon (BC) and ultrafine particle number concentrations (UFPN). We used 136 hours (39 sites), 112 hours (26 sites), 147 hours (39 sites) of PM2.5, BC and UFPN data respectively, to develop separate morning (0800-1200) and afternoon (1200- 1800) models. Continuous measurements of PM2.5 and BC were also made at a single fixed rooftop site located in a high-income residential neighborhood. No continuous measurements of UFPN were available. In addition to spatial variables, measurements from the fixed continuous monitoring site were used as independent variables in the PM2.5 and BC models. The median concentrations (and interquartile range) of PM2.5, BC and UFPN at LUR sites were 133 (96-232) μg m-3, 11 (6-21) µg m-3 and 40 (27-72) × 103 cm-3 respectively. In addition a) for PM2.5 and BC, the temporal variability was higher than the spatial variability; b) the magnitude and spatial-variability in pollutant concentrations was higher during morning than during afternoon hours. Further, model R2 values were higher for morning (for PM2.5, BC and UFPN, respectively: 0.85, 0.86, and 0.28) than for afternoon models (0.73, 0.69 and 0.23); (c) the PM2.5 and BC concentrations measured at LUR sites all over the city were strongly correlated with measured concentrations at a fixed rooftop site; d) spatial patterns were similar for PM2.5 and BC, but different for UFPN; (e) population density and road variables were statistically significant predictors of pollutant concentrations; and, (f) available geographic predictors explained a much lower proportion of variability in measured PM2.5, BC and UFPN than observed in other LUR studies, indicating the importance of temporal variability and suggesting the existence of uncharacterized sources.
Article
The gas-particle partitioning of the primary organic aerosol (POA) emissions from fifty-one light-duty gasoline vehicles (model years 1987–2012) was investigated at the California Air Resources Board Haagen–Smit Laboratory. Each vehicle was operated over the cold-start unified cycle on a chassis dynamometer and its emissions were sampled using a constant volume sampler. Four independent yet complementary approaches were used to investigate POA gas-particle partitioning: sampling artifact correction of quartz filter data, dilution from the constant volume sampler into a portable environmental chamber, heating in a thermodenuder, and thermal desorption/gas chromatography/mass spectrometry analysis of quartz filter samples. This combination of techniques allowed gas-particle partitioning measurements to be made across a wide range of atmospherically relevant conditions – temperatures of 25–100 °C and organic aerosol concentrations of <1–600 μg m−3. The gas-particle partitioning of the POA emissions varied continuously over this entire range of conditions and essentially none of the POA should be considered non-volatile. Furthermore, for most vehicles, the low levels of dilution used in the constant volume sampler created particle mass concentrations that were greater than a factor of 10 or higher than typical ambient levels. This resulted in large and systematic partitioning biases in the POA emission factors compared to more dilute atmospheric conditions, as the POA emission rates may be over-estimated by nearly a factor of four due to gas-particle partitioning at higher particle mass concentrations. A volatility distribution was derived to quantitatively describe the measured gas-particle partitioning data using absorptive partitioning theory. Although the POA emission factors varied by more than two orders of magnitude across the test fleet, the vehicle-to-vehicle differences in gas-particle partitioning were modest. Therefore, a single volatility distribution can be used to quantitatively describe the gas-particle partitioning of the entire test fleet. This distribution is designed to be applied to quartz filter POA emission factors in order to update emissions inventories for use in chemical transport models.
Article
The concentrations of PM2.5 carbon fractions in rural, urban, tunnel and remote environments were measured using the IMPROVE thermal optical reflectance (TOR) method. The highest OC1 and EC1 concentrations were found for tunnel samples, while the highest OC2, OC3, and OC4 concentrations were observed for urban winter samples, respectively. The lowest levels of most carbon fractions were found for remote samples. The percentage contributions of carbon fractions to total carbon (TC) were characterized by one peak (at rural and remote sites) and two peaks (at urban and tunnel sites) with different carbon fractions, respectively. The abundance of char in tunnel and urban environments was observed, which might partly be due to traffic-related tire-wear. Various percentages of optically scattering OC and absorbing EC fractions to TC were found in the four different environments. In addition, the contribution of heating carbon fractions (char and soot) indicated various warming effects per unit mass of TC. The ratios of OC/EC and char/soot at the sites were shown to be source indicators. The investigation of carbon fractions at different sites may provide some information for improving model parameters in estimating their radiative effects.
Article
A dilution source sampling system was used to quantify the organic air pollutant emissions from commercial-scale meat charbroiling operations. Emission rates of gas-phase volatile organic compounds, semivolatile organic compounds, and high molecular weight particle-phase organic compounds were simultaneously quantified on a single compound basis. Fine particle mass emission rates and fine particle elemental chemical composition were measured as well. Emission rates of 120 organic compounds, spanning carbon numbers from C1 to C29 were quantified including n-alkanoic acids, n-alkenoic acids, carbonyls, lactones, alkanes, aromatics, polycyclic aromatic hydrocarbons, alkenes, and steroids. Ethylene, formaldehyde, and acetaldehyde were found to be the predominant light gas-phase organic compounds emitted from the charbroiling operations. n-Alkanoic acids, n-alkenoic acids, and carbonyls made up a significant fraction of the quantified semivolatile and particle-phase organic compound emissions. Meat charbroiling is one of the few sources identified to date that contributes to the high molecular weight aldehydes measured in the urban atmosphere. Semivolatile and particle-phase organic compounds were collected for quantification by two simultaneous sampling protocols: (1) quartz fiber filters followed by polyurethane foam (PUF) cartridges, and (2) XAD-coated annular denuders followed by quartz fiber filters and PUF cartridges. Good agreement was observed for the total mass emissions collected by the two different sampling procedures; however, the partitioning of the semivolatile organic compounds between the gas phase and particle phase, as measured by the two sampling procedures, showed significant differences for n-alkanoic acids, indicating that significant artifact adsorption of these compounds occurs to the filter in the filter/PUF sampling system.
Article
Aging of aerosol from wood chip combustion in a stoker burner was monitored in an outdoor environmental chamber for 19–27h in order to study the size, volatility and organic carbon (OC) content of the combustion aerosol particles during aging. A scanning mobility particle sizer, a volatility tandem differential mobility analyzer (VTDMA), and a thermal–optical carbon analyzer were utilized. The VTDMA and carbon analyses were performed at the beginning, after 17–24h of aging and at one intermediate point. The size decrease of freshly emitted particles was 6–10% when heated to 360∘C, and was found to depend on the experiment start time. For particles aged for 24h, a 74–86% decrease in particle size at 360∘C was observed. The more volatile OC fraction and the total OC fraction in the particles increased and the less volatile OC fraction decreased with aging. This suggests that during aging more volatile compounds condense on or heavier compounds photodegrade into lighter ones in the particles. Occasionally, new particle formation and growth were observed in the following day. The new particles were found to be composed mainly of volatile material.
Article
Land Use Regression (LUR) models have been used to describe/model spatial variability of annual mean concentrations of traffic related pollutants like nitrogen dioxide (NO2), nitrogen oxides (NOx/) and particulate matter (PM). No models have yet been published of elemental composition. As part of the ESCAPE project, we measured the elemental composition in both the PM10 and PM2.5 fraction sizes at 20 sites in each of 20 study areas across Europe. LUR models for eight a priori selected elements; copper (Cu), iron (Fe), potassium (K), nickel (Ni), sulphur (S), silicon (Si), vanadium (V) and zinc (Zn) were developed. Good models were developed for Cu, Fe and Zn in both fractions (PM10 and PM2.5) explaining on average between 67 and 79% of the concentration variance (R(2)) with a large variability between areas. Traffic variables were the dominant predictors, reflecting non-tailpipe emissions. Models for V and S in the PM10 and PM2.5) fractions and Si, Ni and K in the PM10 fraction performed moderately with R(2) ranging from 50 to 61%. Si, NI and K models for PM2.5) performed poorest with R(2) under 50%. The LUR models are used to estimate exposures to elemental composition in the health studies involved in ESCAPE.
Article
Land use regression models (LUR) frequently use leave-one-out-cross-validation (LOOCV) to assess model fit, but recent studies suggested that this may overestimate predictive ability in independent datasets. Our aim was to evaluate LUR models for nitrogen dioxide (NO2) and particulate matter (PM) components exploiting the high correlation between concentrations of PM metrics and NO2. LUR models have been developed for NO2, PM2.5 absorbance and Copper (Cu) in PM10 based on 20 sites in each of the 20 study areas of the ESCAPE project. Models were evaluated with LOOCV and "hold-out evaluation (HEV)" using the correlation of predicted NO2 or PM concentrations with measured NO2 concentrations at the 20 additional NO2 sites in each area. For NO2, PM2.5 absorbance and PM10 Cu, the median LOOCV R2s were 0.83, 0.81 and 0.76 whereas the median HEV R2 were 0.52, 0.44 and 0.40. There was a positive association between the LOOCV R2 and HEV R2 for PM2.5 absorbance and PM10 Cu. Our results confirm that the predictive ability of LUR models based on relatively small training sets is overestimated by the LOOCV R2s. Nevertheless, in most areas LUR models still explained a substantial fraction of the variation of concentrations measured at independent sites.
Article
Combining evidence from laboratory measurements, field studies, and regional modeling, we propose a new conceptual framework for atmospheric organic aerosol. Measurements with both diesel exhaust and wood smoke indicate that primary organic particulate emissions are semivolatile; they thus partially evaporate with atmospheric dilution. A chemical transport model is modified to account for partitioning of semivolatile emissions. The model predicts that evaporation reduces urban organic aerosol in the northeastern US by up to a factor of two, bringing model results into better agreement with observations. Laboratory experiments show that photo-oxidation of gas-phase semivolatile diesel emissions generates organic aerosol, greatly exceeding the contribution of known secondary organic aerosol precursors. A module describing photochemical aging of semivolatile emissions in the chemical transport model sharply increases predicted regional organic aerosol concentrations, especially downwind of urban areas. The combination of evaporation of primary emissions and photochemical aging of semivolatile emissions dramatically alters the primary- secondary split. In the Eastern US, even urban populations are largely exposed to a regional organic aerosol that has been formed by photochemistry. We hypothesize that photochemical aging of semivolatile emissions is an important contributor to high levels of secondary organic aerosol observed in recent field studies.
Article
Land Use Regression (LUR) models have been used increasingly for modeling small-scale spatial variation in air pollution concentrations and estimating individual exposure for participants of cohort studies. Within the ESCAPE project, concentrations of PM(2.5), PM(2.5) absorbance, PM(10), and PM(coarse) were measured in 20 European study areas at 20 sites per area. GIS-derived predictor variables (e.g., traffic intensity, population, and land-use) were evaluated to model spatial variation of annual average concentrations for each study area. The median model explained variance (R(2)) was 71% for PM(2.5) (range across study areas 35-94%). Model R(2) was higher for PM(2.5) absorbance (median 89%, range 56-97%) and lower for PM(coarse) (median 68%, range 32- 81%). Models included between two and five predictor variables, with various traffic indicators as the most common predictors. Lower R(2) was related to small concentration variability or limited availability of predictor variables, especially traffic intensity. Cross validation R(2) results were on average 8-11% lower than model R(2). Careful selection of monitoring sites, examination of influential observations and skewed variable distributions were essential for developing stable LUR models. The final LUR models are used to estimate air pollution concentrations at the home addresses of participants in the health studies involved in ESCAPE.
Article
Staged tests were conducted to measure the particle and vapor emissions from a CFM56-2B1 gas-turbine engine mounted on a KC-135T Stratotanker airframe at different engine loads. Exhaust was sampled using a rake inlet installed 1-m downstream of the engine exit plane of a parked and chocked aircraft and a dilution sampler and portable smog chamber were used to investigate the particulate matter (PM) emissions. Total fine PM mass emissions were highest at low (4%) and high (85%) load and lower at intermediate loads (7% and 30%). PM mass emissions at 4% load are dominated by organics, while at 85% load elemental carbon is dominant. Quantifying the primary organic aerosol (POA) emissions is complicated by substantial filter sampling artifacts. Partitioning experiments reveal that the majority of the POA is semivolatile; for example, the POA emission factor changed by a factor of two when the background organic aerosol concentration was increased from 0.7 to 4 mu g m(-3). Therefore, one cannot define a single non-volatile PM emission factor for aircraft exhaust. The gas- and particle-phase organic emissions were comprehensively characterized by analyzing canister, sorbent and filter samples with gas-chromatography/mass-spectrometry. Vapor-phase organic emissions are highest at 4% load and decrease with increasing load. Low-volatility organics (less volatile than a C-12 n-alkane) contributed 10-20% of the total organic emissions. The low-volatility organic emissions contain signatures of unburned fuel and aircraft lubricating oil but are dominated by an unresolved complex mixture (UCM) of presumably branched and cyclic alkanes. Emissions at all loads contain more low-volatility organic vapors than POA; thus secondary organic aerosol formation in the aging plume will likely exceed POA emissions.
Article
Although organics constitute approximately 10–70% of the total dry fine particle mass in the atmosphere, their concentrations and formation mechanisms are less well understood than those of other components such as sulfate and nitrate. This is because particulate organic matter is an aggregate of hundreds of individual compounds whose concentrations cannot be characterized by a single analytical technique; more than half of the organic carbon mass has not yet been identified as individual compounds. Moreover, the collection process itself can alter the gas–particle equilibrium of a number of condensable organics resulting in both positive and negative sampling biases. The incomplete characterization of particulate organics coupled with the complexity of the photochemical reactions that produce particulate matter from volatile organic emissions has prevented the development of a first principle simulation approach. These limitations are providing an impetus for numerous scientific studies, proving organics to be the next frontier for particle characterization and simulation. This paper reviews the current state of organic aerosol sampling, analysis, and simulation, examines the limitations of the current technology, and presents prospects for the future. The emphasis is on distilling findings from recent atmospheric, smog chamber, and theoretical studies to provide a coherent picture of what has been accomplished, especially during the last five years, and what problems are ripe for further investigation.
Article
Filter collection of particulate organic carbon is complicated by adsorption and volatilization sampling artifacts which inhibit attempts to accurately assess aerosol concentrations. Insight into the adsorption artifact can be gained by considering that adsorption on sampling filters is conceptually similar to gas-particle partitioning with the filter acting as a model aerosol. We found that significant adsorption of gas-phase organic compounds occurs on quartz fiber filters, and the presence of typical loadings of particulate material on the sampling filter does not significantly affect the magnitude of the adsorption artifact. Also placing a Teflon filter upstream of a pair of quartz fiber filters has no effect on total organic carbon loadings on the final quartz fiber filter, but adsorbed organic carbon loadings are much higher on quartz fiber filters behind Teflon filters than on quartz fiber filters behind quartz fiber filters. Organic carbon sampled from the atmosphere is unlikely to attain equilibrium between that in the gas phase and that adsorbed on the quartz fiber sampling filter. As a result, a quartz fiber filter behind a quartz fiber filter is exposed to a lower gas-phase concentration than a quartz fiber filter behind a Teflon filter. This explains the difference between Teflon-quartz and quartz-quartz backup filters and suggests that Teflon-quartz backup filters provide a better estimate of adsorbed vapor on quartz fiber front filters than do quartz-quartz backup filters. Under typical sampling conditions adsorption is the dominant artifact in the sampling of particulate organic carbon, and longer sampling periods reduce the percentage of collected material that is adsorbed vapor.
Article
Overnight aging experiments with diesel engine exhaust from a diesel power aggregate, with no or 9 kW load, and from a diesel-fueled vehicle were conducted in an environmental chamber. During a 24 h aging period the volatilities of monodisperse particles at 140, 250 and 360 °C heater temperatures were analyzed with volatility tandem differential mobility analysis (VTDMA). The particulate organic to total carbon ratio and organic carbon subfractions at 120, 250, 450 and 550 °C were analyzed with thermal-optical carbon analysis for samples from fresh, 8 or 18 h aged and 24 h aged aerosol. During the experiment also the particle size distribution, ozone and nitrogen oxide concentration, and temperature, relative humidity and total solar and total ultraviolet radiation in the chamber were monitored.After injection, the geometric mean diameter and number concentration of the particles in the chamber were 66–85 nm and 0.9–4.6×105 cm−3, respectively. The particles were seen to grow fast, at a growth rate of 18–47 nm h−1 during the first hour. The fresh particles from the diesel power aggregate contained 37–45% of apparent volume semi-volatile compounds with no load and 10–24% with 9 kW load. The semi-volatile apparent volume fraction at 360 °C for 50 nm particles produced by the diesel power aggregate was 57%. After 24 h of aging, the semi-volatile apparent volume fraction at 360 °C for 100 nm particles was 99%. This suggests that the particles in the 24 h aged aerosol at this size class are no more primary particles but particles that are formed in the chamber through nucleation and subsequent growth.