PCBs, PBDEs, and PAHs in Toronto air: Spatial and seasonal trends and implications for contaminant transport

Science of the Total Environment 429 (2012) 272–280

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

PCBs, PBDEs, and PAHs in Toronto air: Spatial and seasonal trends and implications for contaminant transport Lisa Melymuk a, Matthew Robson b, Paul A. Helm c, Miriam L. Diamond a, b,⁎ a b c

Dept. of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Canada Dept. of Geography and Program in Planning, University of Toronto, Toronto, Canada Environmental Monitoring and Reporting Branch, Ontario Ministry of the Environment, Toronto, Canada

a r t i c l e

i n f o

Article history: Received 25 January 2012 Received in revised form 8 April 2012 Accepted 8 April 2012 Available online 11 May 2012 Keywords: Urban areas Persistent organic pollutants Sources Urban fate and transport Atmospheric transport

a b s t r a c t The distributions of polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), and polycyclic aromatic hydrocarbons (PAHs) in the atmosphere of Toronto, Canada and the surrounding suburban/rural area were examined. A series of temporally- and spatially-distributed air samples was collected over a 1-year period with a high-volume active air sampler at one downtown site and polyurethane foam passive air samplers at 19 sites. Passive sampler air concentrations of ΣPAHs ranged from 0.27 to 51 ng/m3. Concentrations of ΣPCBs ranged from 6.0 to 1300 pg/m3, and concentrations of ΣPBDEs ranged from 0.47 to 110 pg/m3. All compounds exhibited the highest concentrations in the urban core, and lowest concentrations in the surrounding rural areas, however the exact ratio depended on location since concentrations varied considerably within the city. Results from the application of a radial dilution model highlighted the influence of the central business district (CBD) of the city as a source of contaminants to the surrounding environment, however the radial dilution comparison also demonstrated that sources outside the CBD have a significant influence on regional contaminant concentrations. A strong relationship between temperature and partial pressure of the gas-phase PCBs, low molecular weight PBDEs and less-reactive PAHs suggested that their dominant emissions originated from temperature-controlled processes such as volatilization from local sources of PCBs, PAHs and PBDEs at warm temperatures, condensation and deposition of emissions at cold temperatures, and ventilation of indoor air with elevated concentrations. The relationship between temperature and atmospheric PAH concentrations varied along the urban–rural gradient, which suggested that in highly urbanized areas, such as downtown Toronto, temperaturerelated processes have a significant impact on air concentrations, whereas winter emissions from domestic heating have a greater influence in areas with less impervious surface coverage. © 2012 Elsevier B.V. All rights reserved.

1. Introduction/background The distribution of semi-volatile organic contaminants (SVOCs) in the urban environment has been the subject of extensive study (inter alia Du et al., 2009; Gasic et al., 2009; Gingrich et al., 2001; Harner et al., 2004; Motelay-Massei et al., 2005; Persoon et al., 2010). These studies have illustrated the complicated characteristics of urban distributions of SVOCs, stemming from a high density of varied emission sources combined with the inter-compartmental transport and meteorological effects that control SVOC movement. The literature is consistent in identifying elevated urban concentrations of a wide array of chemical compounds and that the resulting “urban plume” is a chemical source to surrounding regions (Hodge and Diamond, 2010; Hornbuckle and Green, 2003; Offenberg and Baker, 1999; Zhang et al., 1999). These observations provide an incentive for ⁎ Corresponding author at: Dept. of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Canada. E-mail address: [email protected] (M.L. Diamond). 0048-9697/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2012.04.022

us to better understand urban contaminant dynamics. In particular, gaps in understanding remain with respect to the relative importance of source types, the nature of seasonal variations in both sources and concentrations that are specific to urban areas, and the spatial distribution of sources within cities. The contaminants evaluated in this study were polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), and polycyclic aromatic hydrocarbons (PAHs). This suite of contaminants was chosen to give a broad picture of contaminant dynamics in the urban region, covering compounds with different chemical sources, physical–chemical properties and use/emission histories. The compounds have a range of volatilities, from those found almost entirely in the gas phase to those found entirely in the particle phase, with the compounds of intermediate volatility shifting between primarily gas and primarily particle phases with seasonal changes in temperature. PCB manufacturing has been banned and its use restricted since the 1970s (Diamond et al., 2010), while PBDEs, in use since 1965 (Vonderheide et al., 2008), have recently been subjected to regulations and reductions in manufacturing (Ward et al., 2008). PAHs have

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

273

a long record of emissions from anthropogenic sources (Environment Canada, 2009; Gabrieli et al., 2010; U.S. EPA, 2011). The goal of this paper is to provide a coherent picture of SVOC distributions and transport in the urban atmosphere through a comparison of source types and identification of the major environmental processes governing atmospheric fate. We used passive and active air samples to provide both spatial and seasonal resolution of concentration patterns, and characterized the spatial concentration gradients within the urban region using a radial dilution model.

March 2008 (Winter), April to June 2008 (Spring), and July to October 2008 (Summer). Samplers were located in trees at a height of 3–4 m above ground level. Sampling rates were obtained from a calibration study using 10 separate passive air samplers and a low-volume sampler conducted at the 0 km site, as described by Melymuk et al. (2011b). A key finding from this calibration study was that PUFpassive samplers collect both gas- and particle-phase compounds and thus passive sampler results are best treated as indicative of bulk air concentrations.

2. Sample collection

2.3. High-volume air samples

2.1. Site selection

A high-volume air sampler (Tisch Environmental) was used to collect samples at the 0 km site, with a sampling train consisting of a glass fiber filter (GFF) and two PUF plugs. The site was located on the roof of a small 3-storey building on the University of Toronto campus in downtown Toronto. Approximately 475 m 3 of air was collected over a 24 hour period, every 12 days from October 2007 to October 2008. A meteorological station collecting temperature, wind speed and direction, relative humidity and precipitation was co-located with the high-volume air sampler.

Nineteen sites were selected along two transects centered over Toronto, Canada, spanning the Greater Toronto Area (GTA) (Fig. 1). The sites ranged from high-density commercial/residential land-use, to low density suburban residential areas adjacent to multi-lane highways, to rural/agricultural land. Samples were mainly collected in parks or residential gardens. Meteorological data were obtained for each site from the nearest Environment Canada weather station (Environment Canada, 2010). Additional site characteristics are given in Table S1 of the Supplementary information (SI). 2.2. Passive air samples Passive air samples were collected at each of the 19 sites from October 2007 to October 2008. All passive air samples were collected using “dome”-type polyurethane foam (PUF) samplers (Harner et al., 2004) as depicted in Fig. S1 and with details listed in Melymuk et al. (2011b). Passive samplers were deployed in four 3-month deployment periods: October to December 2007 (Autumn), January to

N80

N40

3. Analysis Details of the extraction, clean-up, analysis and QA/QC have been published by Melymuk et al. (2011b), and are given in full in the SI. Briefly, samples were extracted via pressurized fluid extraction using dichloromethane. Samples were split into two fractions: one for PCB and PBDE analysis and one for PAH analysis. The PCB/PBDE fraction was washed with concentrated H2SO4, and then purified on an Al2O3– AgNO3 column. The PAH fraction was purified on a silica column. 86 PCBs, 27 PBDEs and 15 PAHs were quantified via gas chromatography– mass spectrometry (GC-MS). Non-detects were treated as zero, as per Harrad and Hunter (2006). Although this may result in low values for the sums of compounds at rural sites, we chose to treat non-detects as zero because the relatively higher detection limits of the hepta- to decaBDEs would excessively bias the results towards high molecular weight PBDEs. The compounds, instrument conditions, QA/QC and detection limits for each class are summarized in the SI. 4. Results and discussion

Toronto

4.1. ΣPAH concentrations E40 N20

N10

E20

N5 E10 W10

W5

E5 S5

W20

N1 E5 E1

W40

W60

W1

0KM

W5

Fig. 1. Sampling locations. The star on the inset map indicates the middle of Toronto's central business district. Site names refer to the direction (north, south, east or west) and approximate distance (in km) of each site from the central (0 km) sampling point.

Passive sampler-measured ΣPAH concentrations ranged from 0.27 ng/m 3 (N80 site in summer) up to 51 ng/m 3 (E10 — spring) (Table 1). Active air sampler measurements of ΣPAH in bulk air, taken at the 0 km site, ranged from 3.0 ng/m 3 (Oct. 9th) to 50 ng/m 3 (July 5th) (Fig. 2). There was a good agreement between the concentrations measured by the passive and active samplers at the 0 km site, with average passive and active sampler concentrations of 15 ± 5.2 and 14 ± 11 ng/m 3, respectively. These concentrations were within the range reported for other sites in the Great Lakes region of North America (Motelay-Massei et al., 2004; Sun et al., 2006b) and urban areas in Western Europe (Harrad and Laurie, 2005; Ras et al., 2009) and China (Wang et al., 2010). 4.2. ΣPCB concentrations Passive sampler-measured concentrations of ΣPCBs ranged from 6.0 pg/m 3 (N80 — spring and summer) to 1300 pg/m 3 (0 km — summer) (Table 1). At the 0 km site active air sampler measurements of ΣPCBs in bulk air ranged from 72 pg/m 3 (Oct. 9th) to 3800 pg/m 3 (July 17th) (Fig. 2). The average ΣPCB concentration measured by high-volume air sampler at the 0 km site was 1100 ± 890 pg/m 3, which compared well with the passive air sampler concentration of

274

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

Table 1 Concentrations of SVOCs in air. Concentrations reflect an arithmetic mean and standard deviation of 4 passive samples, 1 per season collected over a 1 year period. Concentrations for each season are given in Table S3 of the SI. Site

Distance from CBD and transect

PAHs (ng/m3)

PCBs (pg/m3)

PBDEs (pg/m3)

0 km E1 W1 N1 E5 W5 N5 S5 E10 W10 N10 E20 W20 N20 E40 W40 N40 W60 N80

1.4 km 1.7 km east 2.7 km west 3.2 km north 3.0 km east 4.6 km west 7.0 km north 4.4 km south 6.6 km east 8.9 km west 11.3 km north 14.6 km east 16.1 km west 20.1 km north 33.6 km east 33.4 km west 41.0 km north 48.2 km west 72.3 km north

15 ± 5.2 11 ± 2.4 22 ± 17 12 ± 5.9 12 ± 5.5 19 ± 9.9 6.6 ± 2.1 3.4 ± 1.1 28 ± 17 8.4 ± 2.2 7.8 ± 2.3 9.2 ± 3.3 11 ± 2.2 6.4 ± 2.1 5.3 ± 2.5 6.3 ± 1.4 4.4 ± 1.2 6.5 ± 2.1 1.1 ± 0.94

970 ± 280 380 ± 190 220 ± 42 130 ± 40 133 ± 27 380 ± 350 130 ± 35 79 ± 25 140 ± 59 160 ± 100 94 ± 20 220 ± 130 480 ± 170 76 ± 21 74 ± 24 57 ± 16 45 ± 29 88 ± 38 25 ± 32

73 ± 27 37 ± 18 40 ± 19 9.1 ± 0.67 19 ± 13 37 ± 16 8.6 ± 6.1 3.0 ± 3.0 18 ± 18 31 ± 32 11 ± 4.0 28 ± 19 9.0 ± 10 10 ± 3.5 29 ± 20 10 ± 11 8.0 ± 8.1 21 ± 27 3.7 ± 3.9

4.3. ΣPBDE concentrations Passive sampler-measured concentrations of ΣPBDE ranged from 0.47 pg/m 3 (N80 — summer) to 110 pg/m 3 (0 km — spring) (Table 1). At the more rural sites only BDE-28 and -47 were at levels above the method detection limit. BDE-209 was detected in 25% of passive samples, all of which were from sites b20 km from the central business district (CBD) of Toronto. However, we were unable to calculate concentrations as BDE-209 as sampling rates could not be determined from the calibration study (Melymuk et al., 2011b). The active air samples, which include BDE-209 in the total, ranged from 4.6 pg/m 3 (Dec. 26th) to 3000 pg/m3 ΣPBDE (Dec. 14th) (Fig. 2), with the December 14th concentration identified as an outlier at the 99.9% level based on a Grubbs test. The cause of the very high December 14th value is unclear; it is the only outlier of the dataset. Sporadic high concentrations of PBDEs have previously been measured in the southern Ontario region in both air (Gouin et al., 2006) and precipitation (Melymuk et al., 2011a). The average concentrations at the 0 km site measured by the passive and active air samplers were 72± 27 and 140 ± 520 pg/m 3, respectively (excluding BDE-209). The average active air sample concentration was influenced by the very high concentration measured on Dec. 14th, 2007. Without this outlier, the average high-volume air sampler concentration was 43 ± 42 pg/m 3, which was within the range of the passive sampler concentrations. The concentrations of ΣPBDEs in air at the rural site were within the range measured by Gouin et al. (2005) and Su et al. (2009) at background/rural sites. The higher urban concentrations were similar to those measured in other North American (Hoh and Hites, 2005; Venier and Hites, 2008) and European cities (Jaward et al., 2004; Moeckel et al., 2010), but significantly lower than those measured in Chinese cites (Chen et al., 2006).

60

Concentration (ng/m3)

PAHs 50 40 30 20 10 0

PCBs

3 2.5

4.4. Seasonal variations in concentrations

2 1.5 1 0.5

Concentration (ng/m3)

0 0.25

30 ↑3.0 ng/m3

PBDEs

25 20

0.2

15 10

0.15

5 0.1

0 -5

0.05

-10

Average air temperature (°C)

Concentration (ng/m3)

3.5

980 ± 280 pg/m 3. The lower range of air concentrations was similar to concentrations reported at rural sites in North America and Europe (Jamshidi et al., 2007; Li et al., 2010; Sun et al., 2006a) and the average concentration of ΣPCBs in the urban air was similar to concentrations at urban sites in Europe (Cetin et al., 2007; Jamshidi et al., 2007; Mari et al., 2008) and North America (Harner et al., 2004; Hsu et al., 2003; Motelay-Massei et al., 2004), but were significantly lower than those measured in Chicago (Hu et al., 2010; Sun et al., 2006a) and Jersey City (Totten et al., 2006).

-15

0

Gas Phase

Particle Phase

Air Temp

Fig. 2. Bulk air concentrations of ΣPCBs, ΣPAHs and ΣPBDEs measured by active air sampler and average air temperature at the 0 km site.

Temperatures over the sampling period varied from a monthly average 23 °C in July to −3.3 °C in February at the 0 km site, and there was a 3 °C difference in annual average temperature between the 0 km and 80 km north site. Temperature data from the co-located meteorological station was used to investigate concentration trends with temperature and season at the 0 km site. Air concentrations measured by the active air sampler at the 0 km site generally increased with air temperature (Fig. 2) resulting in strong negative relationships between inverse temperature (1/T) and partial pressure (ln P) for several congeners and compounds in each of the compound classes (Table S5). The partial pressure (P) of the gas-phase compound was determined from the concentrations measured by active air sampler. The vapor pressure range over which the 1/T vs. ln P relationship held was similar for all compound classes: the relationship held for PCB congeners with vapor pressures b2.5 × 10 − 5 Pa, for the less reactive PAHs with vapor pressure b5.8 × 10− 5 Pa, and for PBDEs with vapor pressures b6.8 × 10− 5 Pa (vapor pressures are from Li et al. (2003), Paasivirta et al. (1999), and Wong et al. (2001)). The similar maximum vapor pressures suggest that, for an average urban air mass, despite differences in chemical sources/uses, temperature-controlled

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

volatilization and/or condensation of emissions is independent of the sorption of the molecule to a particular substrate such as atmospheric particles or impervious surfaces. As passive sampler data represented bulk concentrations and comprised only four measurements per site, we were unable to apply detailed temperature analyses, however it was possible to identify broad seasonal trends in concentrations at each site. Specific results from the analyses of temperature and seasonal variation for both active and passive air samples are discussed below by compound class. 4.4.1. PCBs and PBDEs Strong relationships (pb 0.0001) between 1/T and ln P were observed for the tri- to octa-CB congeners in active air samples at the 0 km site (Table S5), suggesting that temperature-dependent air-surface exchange processes such as volatilization or condensation had a considerable role in influencing atmospheric concentrations of these compounds at the 0 km site. The slopes of the 1/T vs. ln P relationship from this study agreed well with those from previous reports (listed in Table S5). Bulk passive air concentrations of PCBs had a consistent seasonal trend, with summer≈ spring> autumn> winter for all sites and congeners, consistent with the active air sampler observations, with the exception of tri-CBs along the west transect. Furthermore, the seasonal shift was greatest for the hexa- and hepta-CBs, for which air concentrations were on average 3 to 4 times higher in summer than in winter. This was consistent with the shift in PCB vapor pressures given the seasonal temperature changes which occurred in the GTA: the vapor pressure of a tri-CB increases 18-fold over the −3 to 22 °C temperature range, whereas the vapor pressures of hexa-and hepta-CBs increase 35-fold (Li et al., 2003), thus leading to proportionally greater volatilization of hexa- and hepta-CBs with increasing temperature. PBDE concentrations were more variable than PCBs. Lower molecular weight (di-tetra) PBDE congener concentrations did exhibit significant (p b 0.05) relationships with inverse temperature, up to BDE-66 (Table S5), but this relation did not hold for higher molecular weight PBDEs. The ln P vs 1/T relationships were not as strong for PBDE congeners as they were for PCBs. Similarly, a temperature trend in passive air sampler measurements was only observed for BDE-47. Sites along the north–south transect and sites from W5 eastwards had BDE-47 concentrations that were on average 5× higher in spring/summer than in winter, however concentrations to the west of Toronto's CBD did not vary with temperature. The triCBs also lacked any relationship with temperature at sites west of Toronto. This region has a higher density of industrial facilities than the north and east of Toronto, and thus may have emission sources that differ from those in residential and commercial-dominated urban areas (Diamond and Hodge, 2007). We attribute the apparent weak influence of temperature on PBDEs to a combination of factors, particularly (1) the inability to isolate seasonal variations due to low concentrations of PBDEs, (2) difficulties in using PUF-PAS to reliably sample particle-bound substances (Melymuk et al., 2011b), (3) debromination of higher molecular weight congeners into lower molecular weight PBDEs (Schenker et al., 2008), a loss pathway which is greater at higher temperatures and higher incidences of solar radiation, and (4) the possibility that some releases of PBDEs occur via particle abrasion, where the PBDEs may not be exchangeable with the gas-phase (Webster et al., 2009). Elevated summer concentrations may be due to volatilization from local sources such as transformers, sealants, consumer products (Alcock et al., 2003; Batterman et al., 2009; Hsu et al., 2003; Robson et al., 2010) or re-volatilization from environmental reservoirs (Cetin and Odabasi, 2007; Desborough and Harrad, 2011; Kobližková et al., 2009). In addition to temperature-dependent volatilization and/or condensation processes, differences in building ventilation rates may contribute to observed concentration patterns of PCBs and PBDEs. Higher spring/summer building ventilation rates may lead to elevated outdoor concentrations at higher temperatures. For example, Wallace

275

et al. (2002) found that building ventilation rates increased significantly when windows were opened in warmer months, and this has been found to result in lower indoor concentrations of both gas-phase compounds and particulate matter through transfer of these compounds from indoors to outdoors (Fromme et al., 2007; Goyal and Khare, 2009; Schlink et al., 2004). Given the elevated indoor concentrations of PCBs and PBDEs (Rudel et al., 2008; Zhang et al., 2011), some of the increase in outdoor PCB and PBDE concentrations during warmer seasons may be due to release of indoor air from buildings (Currado and Harrad, 2000; Halsall et al., 1995). 4.4.2. PAHs Five PAHs (fluorene, phenanthrene, fluoranthene, pyrene and chrysene), all with 4-rings or fewer, exhibited significant (p b 0.05) negative relationships between ln P and 1/T (Table S5), suggesting that their main emissions at the 0 km site were from temperature-controlled local processes that lead to higher air concentrations at warmer temperatures, as was observed for PCBs and low molecular weight PBDEs. Acenaphthylene, acenaphthene and anthracene had positive or no relationships with inverse temperature, implying other influences on air concentrations. Higher molecular weight (5- and 6-ring) PAHs did not have sufficient gas-phase concentrations to establish a relationship with temperature. A seasonal/temperature trend was also observed for PAHs in the passive air samples. Nine sites (0 km, E1, N1, E5, E10, E20, E40, W5 and W10), hereafter referred to as Group 1, had high spring/summer concentrations of the less reactive gas-phase PAHs (e.g. 3- and 4-ring PAHs). Conversely, seven sites (N5, N10, N20, N40, N80, W40, and S5 — Group 2) had highest concentrations in winter. The relationship between the active air sampler measurements and temperature, as well as the seasonal trends from the Group 1 sites, are different from the typical relationship between PAH concentrations and temperature. Many populated temperate areas have higher PAH concentrations in winter, generally attributed to increases in emissions from heating during colder weather, and reductions in atmospheric degradation of PAHs (Baek et al., 1991; Finlayson-Pitts and Pitts, 2000; Menichini, 1992; Ravindra et al., 2008). However, the reverse seasonal trend — higher PAHs air concentrations in summer than in winter, has also been found in some North American cities, including Toronto (Gustafson and Dickhut, 1997; Motelay-Massei et al., 2005). Our results suggest that seasonal trends depend on the degree of urbanization of an area. Fossil fuel emissions from mobile sources such as vehicle emissions are the largest source of PAHs to air in cities (Aceves and Grimalt, 1993; Dvorska et al., 2011; Harrad and Laurie, 2005; Prevedouros et al., 2004), and in the GTA, vehicle traffic is relatively constant year-round (Vallurupalli, 1997). However, we suggest that three factors are leading to the observed seasonal trends. First, domestic heating in Toronto is not likely a large contributor to overall PAH levels, as 94% of Toronto fossil fuel space heating comes from natural gas (Office of Energy Efficiency, 2011) which has relatively low PAH emissions compared to other domestic heating sources, such as oil, coal or wood (Gingrich and Macfarlane, 2002; Hobson and Thistlethwaite, 2003; Kakareka et al., 2005). Second, urban areas have potential volatilization sources which could lead to increased PAH concentrations in air at warmer temperatures. These include volatilization from PAH-containing pavement materials, which is believed to be of particular importance in urban areas given the high impervious surface coverage (Gustafson and Dickhut, 1997; Mahler et al., 2005; Motelay-Massei et al., 2005; Rogge et al., 1997), contaminated soils (Cabrerizo et al., 2011), or other urban surfaces (Bozlaker et al., 2008; Meyer et al., 2009; Wild and Jones, 1995). Coal-tar pavement sealants have recently been identified as having the potential to contribute significantly to atmospheric PAH concentrations (Van Metre et al., 2012) and in Toronto, coal-tar sealcoats are estimated to cover 24 km 2, with PAH concentrations of up to 24,000 mg/kg (Diamond et al., 2011). Third, the partitioning of vehicle emissions also has the potential to lead to increased summer concentrations of PAHs. Vehicle exhausts are

276

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

emitted at a high temperature and subsequently follow two pathways: (1) direct emission into air, either in gas- or particle-phases or (2) at colder temperatures condensation to particles as exhaust cools followed by deposition to surfaces (Clements et al., 2009). At warmer temperatures higher PAH air concentrations can come from less condensation and deposition of vehicle exhausts and volatilization of PAHs from surfaces on which they previously condensed or deposited. In urban areas these potential sources include surface films (Diamond et al., 2000) and street dust, both of which act as transient repositories of PAHs with high proportions originating from vehicle exhausts (Dong and Lee, 2009; Liu et al., 2003; Oda et al., 2001). The films equilibrate rapidly with ambient air concentrations but could be a short term source via volatilization as temperatures increase (Csiszar et al., 2012). Thus, our results suggest that in urban areas with high vehicle traffic, high impervious surface coverage and natural gas heating, temperaturedependent processes such as condensation and volatilization play a large role in controlling the distribution of PAHs and can result in elevated summer air concentrations. Conversely, in rural areas with low traffic and impervious surface coverage, and possibly greater use of oil, coal or wood heating (Dvorska et al., 2011; Gingrich and Macfarlane, 2002), emissions from fossil fuel combustion such as domestic heating contribute more to PAH air concentrations, leading to higher winter concentrations. The passive and active sampling data support this hypothesis. Group 1 sites, with highest concentrations of the less reactive gas-phase PAHs in spring/summer, were largely urban sites, and eight of those nine sites had the highest percentage of impervious surface coverage in a 1 km radius of the site. Group 2 sites, with the highest winter concentrations, were largely suburban and rural sites with the lowest percent impervious land coverage in a 1 km radius. In contrast to the other PAH compounds, acenaphthene and acenaphthylene concentrations showed positive relationships between ln P and 1/T, indicating that as temperatures decreased the partial pressures of the compounds increased. Anthracene concentrations did not have a significant relationship between temperature and partial pressure. Acenaphthene, acenaphthylene and anthracene have the shortest atmospheric lifetimes of all PAHs studied (Bunce and Dryfhout, 1992), leading us to hypothesize that for these PAHs atmospheric degradation explains more of the temporal concentration patterns than emissions, and thus disrupts the ln P vs 1/T relationship. An examination of the change in the ratios of anthracene (ANT) to phenanthrene (PHE), where PHE is a relatively stable PAH, further supports the hypothesis of significant rapid losses of reactive PAHs in the urban summer atmosphere, presumably due to reaction with OH radicals formed through a photochemical reaction sequence (Bunce and Dryfhout, 1992). To examine this hypothesis solar radiation data were obtained for the sampling period from a site 25 km from the CBD (UTM Meteorological Station, 2011) (Fig. S2) and compared with the ANT/PHE ratios from the active sampler data. 24-hour average solar irradiance was ~2 × higher in summer than in winter, reaching a peak of 350 W/m 2 on July 5th. The ANT/PHE ratio was significantly correlated with solar radiation (p b 0.0001), with lower ANT/PHE ratios occurring with higher solar radiation. Thus while PAH emissions may peak in areas with high impervious surface coverage in summer, this is counteracted by a significant increase in atmospheric degradation, resulting in relatively constant concentrations of anthracene over the 1 year period. 4.5. Other atmospheric influences The influence of meteorological parameters other than temperature on concentrations was also investigated through the relationship between the active air sampler data and the meteorological data from the co-located weather station. Previous research identified significant influences of wind speed, wind direction, relative humidity (Harrad and Mao, 2004) and weekly rainfall (Tham et al., 2008) on atmospheric SVOC concentrations. We found a consistent inverse

relationship between average wind gust speed and air concentrations of the lower molecular weight compounds (all PCBs, 3- and 4-ring PAHs, and di- through tetra-BDEs). However gust speed and temperature tend to inversely co-vary in Toronto, as higher gust speeds coincide with colder temperatures (Fig. S3). A multiple regression analysis (as per Harrad and Mao, 2004) showed that while the relationship between temperature and concentration held, the relationship between gust speed and concentration was not significant when the influence of temperature was considered. However, it should be noted that this does not mean that such a relationship does not exist, but rather that the influence of wind speed was less important than that of temperature. Similarly, there was a statistically significant (pb 0.05) inverse relationship between the duration of precipitation in the week prior to sampling and concentrations of bulk PCBs and PAHs, however the duration of precipitation events was also inversely related to temperature, and thus the duration of precipitation was not statistically significant when the influence of temperature was taken into account. No other consistent relationships were observed with any other meteorological parameters. We also considered the relationship between concentrations, precipitation and particulate matter, using our data, precipitation volumes (Environment Canada, 2010), and particulate matter concentrations (PMcoarse and PMfine) measured by the National Air Pollutant Surveillance (NAPS) Network over 24 hour periods at a site ~300 m from the 0 km sampling site (Environment Canada, 2008). Both PMcoarse and PMfine were inversely related to the amount of precipitation in the 24 h prior to sampling, indicating rainout/ washout of particles during rain events. However, SVOC concentrations did not follow this pattern, i.e., neither gas- nor particle-phase concentrations decreased following precipitation events. Reductions in air concentrations of PCBs and PAHs following precipitation events have been observed at rural sites (Dickhut and Gustafson, 1995; Simcik et al., 1997) and PCB, PBDE, and PAH concentrations were elevated in precipitation at the 0 km site relative to suburban and rural sites (Melymuk et al., 2011a), which suggests that SVOCs are being removed through wash-out/rain-out. However, consistent with our results, Skrdlikova et al. (2011) estimated that b10% of atmospheric PAHs are washed out during precipitation events. Thus, the combination of a relatively low washout of SVOCs and elevated urban emissions may explain the lack of relationship between atmospheric concentrations and precipitation volume or duration in our data. This suggests that in urban areas continuous emissions of SVOCs overshadow any losses via washout/rainout. 4.6. Urban–rural gradients The relative differences between urban and rural sites have often been represented by ratios of urban to rural concentrations (e.g. Harner et al., 2004; Harrad and Hunter, 2006; Hoh and Hites, 2005). In our study the use of these ratios clearly demonstrates elevated urban concentrations with respect to those at the rural site. Urban concentrations (0 km site) were 13, 39 and 19 times higher than the rural (N80) concentrations for ΣPAHs, ΣPCBs and ΣPBDEs, respectively. The urban–rural ratios of PCBs and PAHs in air were significantly larger than was previously observed for the Toronto area (Harner et al., 2004; Motelay-Massei et al., 2005) and other urban–rural comparisons (Hodge and Diamond, 2010). Although these ratios are useful methods for comparing urban–rural differences using only a few sampling sites, the ratios depend largely on the choice of sites and sampling period, illustrated by the 8-fold difference in air concentrations within a 2 km radius of the CBD. Given the heterogeneity of concentrations within an urban area, these ratios are not a generalizable characteristic of cities. Rather than simple ratios, urban–rural gradients can be better characterized using distance-dependent functions which incorporate data from all sites along the transects. Due to the elevated

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

concentrations in the vicinity of the CBD, we chose an exponential function that relates air concentrations with the distance from the CBD to test the hypothesis that the CBD is the source of chemical concentrations to the study region. A radial dilution model, m

ð3:1Þ

specifies the air concentration (Cr) resulting from dilution at a given radius r, to a point source with air concentration C0. This type of model has been used previously with spatially-distributed ambient air concentrations to identify the point source origin of atmospheric emissions (McDonald and Hites, 2003). We used the radial dilution equation with a known source region (the CBD) to compare the degree to which that region influences concentrations for the three compounds across the GTA. A value of m = −2 corresponds to the inverse square dilution of a chemical mass originating from a point source which decreases in concentration due only to outward dispersion in three dimensions and as such, does not incorporate the influence of atmospheric boundary layer height, nor any loss processes such as atmospheric degradation. Using the logarithmic form of the equation: ð3:2Þ

the parameter m is the slope of the linear regression between ln(Cr/ C0) and ln(r). The parameter m can then be compared among transects, compound classes, and seasons, providing more information about the urban-to-rural transition and reducing possible bias resulting from considering only the most extreme sites, as with the ratio characterization. A significant relationship was found between ln(Cr/C0) and ln(r) for all compound classes (Fig. 3), where r is the distance in km from the CBD and C0 is the annual average concentration in the CBD, which was estimated from S5 air concentrations by assuming that the CBD was the main source of SVOCs to the S5 site, and determining C0 from the radial dilution equation based on that assumption. The S5 site was located on an island 4.4 km south of the CBD and is largely reflective of over-lake concentrations, with limited SVOC sources between the CBD and S5 (Fig. 1). The resultant slopes of the regression for all compounds indicated declining concentrations of all compound classes along the urban–rural gradient (Fig. 3). PCBs decreased most rapidly with increasing distance from the urban area, dropping off by 94% at 20 km from the CBD (approx. edge of Toronto), while PAHs and PBDEs decreased by 90% and 79%, respectively. The spatial distributions varied seasonally. Significant radial dilution was observed for spring and summer PCB, PAH, and PBDE concentrations (Table S6). However, despite the more than three-fold

70

a) PAHs

60

Concentration (ng/m3)

lnðC r =C 0 Þ ¼ m⋅ lnr

concentration differences between the most extreme sites (0 km and N80), m was not statistically significant for any compounds in autumn and winter, which suggested a lack of a consistent relationship between concentration and distance during the colder seasons. The difference in seasonal distributions may be related to a combination of temperature-dependent urban emission sources in spring/summer and seasonal differences in winds. As demonstrated earlier, air concentrations in Toronto were positively correlated with temperature, and higher wind speeds occur at colder temperatures (Fig. S3), leading to greater advective dilution of SVOCs from Toronto in winter. Thus, spatial homogeneity increases in autumn and winter causing urban–rural gradients to dissolve. Seasonal gradients had an added complication for PAHs, as emissions from domestic heating potentially lead to a proportionally greater increase in the suburban/ rural concentrations in autumn and winter. As mentioned above, comparing the theoretical inverse-square (m = − 2) radial dilution model and observed slopes from the linear regression of ln(Cr/C0) vs. ln(r) tests the hypothesis that the CBD is the sole source of SVOCs to the GTA, with slopes closer to − 2 supporting the hypothesis. However, given that the radial dilution model assumes equal hemispherical dispersion and does not include the influence of the height of the atmospheric boundary layer, it is

Measured Theoretical (m=-2) Estimated distribution (m=-0.79)

50 40 30 20 10 0

b) PCBs

1400

Concentration (ng/m3)

Cr ¼ C0r

277

Measured Theoretical (m=-2) Estimated distribution (m=-0.94)

1200 1000 800 600 400 200 0

0.5 0 -0.5

1

2

3

4

ln[Cr/C0]

-1 -1.5 -2 -2.5 -3

Concentration (ng/m3)

ln[r] 0

c) PBDEs

90

70 60 50 40 30 20

-3.5

10

-4

0

-4.5

Measured Theoretical (m=-2) Estimated distribution (m=-0.54)

80

0 PAHs (ng/m3) y = -0.79x R² = 0.22

PCBs (pg/m3) y = -0.94x R² = 0.29

PBDEs (pg/m3) y = -0.54x R² = 0.28

Fig. 3. A logarithmic relationship exists between distance from CBD and air concentrations. Slopes were significant at p b 0.05 based on a linear regression t-test.

20

40

60

80

Distance (km) Fig. 4. The comparison between measured concentrations, the models extrapolated from the linear regression parameters, and the theoretical radial dilution distribution using a radial dilution model centered over Toronto's central business district, for (a) ΣPAHs, (b) ΣPCBs, and (c) ΣPBDEs.

278

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

likely that the true value lies between inverse square dilution (m = −2) and vertical isotropy (m = −1). The atmospheric boundary layer does not have vertically uniform concentrations of SVOCs (Farrar et al., 2005; Li et al., 2009), and can limit vertical dispersion at a certain height. While the model does not account for this limitation on vertical dispersion, it is appropriate as an illustrative representation to compare spatial distributions among compound classes. All three compound classes had elevated suburban concentrations compared with the theoretical inverse-square (m = −2) radial dilution model (Fig. 4) suggesting contributions from sources not located within the CBD. The slopes were significantly different (ΣPCBs: −0.94 ± 0.07, ΣPAHs: −0.79± 0.06, ΣPBDEs: −0.54 ± 0.06), with the slope for PCBs closest to −2. The results indicate that the CBD has the greatest influence on PCB concentrations whereas PBDE and PAH sources are more distributed throughout the urban/suburban area. Given the diffuse and mobile nature of PAH and PBDE sources (consumer products, vehicles, impervious surfaces, heating) it is not surprising that there are sources from outside the CBD. Conversely, estimates of PCB distributions in Toronto identify PCBs as largely held in the CBD, with a PCB source density 10 times greater in the CBD than in the rest of the city (Diamond et al., 2010). Comparison of individual transects also provided insight into the spatial distribution of sources across the study region (Table S6). The distance–concentration regression calculated for the north transect was significant for all compounds, but was less successful at representing the east–west transect, likely reflecting the distributions of population, road networks and industrial regions around the Toronto area (Fig. S4). A higher density of urban development along the east–west transect is expected to result in more sources outside the CBD along this transect than the north–south transect, and thus a weaker relationship with distance. Differences in slopes of individual PAH compounds and PCB congeners were consistent with fractionation patterns found by others (inter alia Motelay-Massei et al., 2005). The slopes for tri- and heptaCBs were −0.63 and −1.2 respectively, which translated into ~10% decrease in the contribution of hepta-CBs to ΣPCBs at a distance of 80 km from the CBD. Similarly, higher molecular weight PAHs (e.g. benzo[b]fluoranthene, benzo[a]pyrene, benzo[ghi]perylene) decreased with greater distance from the urban area, with significantly steeper slopes (average m = −0.85 ± 0.34) than those of 2-, 3- and 4-ring PAHs (average m = −0.40 ± 0.10). This difference in slopes reflected a modest (~5%) but statistically significant shift (pb 0.05) in concentrations in winter and autumn (Fig. S5). The lack of fractionation of PAHs in warmer seasons may be due to atmospheric degradation, as particleassociated PAHs have lower degradation rates than gas-phase compounds (Baek et al., 1991; Finlayson-Pitts and Pitts, 2000) and degradation rates are highest in summer (Bunce and Dryfhout, 1992). Thus, while lower molecular weight PAHs have a greater atmospheric transport potential, in spring and summer some compounds can undergo rapid atmospheric degradation which minimizes the fractionation pattern. Fractionation was not observed for PBDEs along the north– south gradient. Despite the differences between regression parameters and the radial dilution model, the results emphasize the density of sources of all compounds near the CBD, as well as the small scale over which concentrations significantly decreased. Concentrations at 5 km from the CBD, well within the Toronto urban region, were ~50% lower than at 1 km from the CBD, and were closer to rural concentrations than to CBD concentrations. Thus, although SVOC sources exist beyond the CBD, the highest concentrations are constrained to only a relatively small area around the CBD. 5. Conclusions An analysis of air concentrations obtained from a high volume air sampler located in downtown Toronto and 19 passive air samplers

distributed along two urban–rural transects showed strong intraurban variability at a local scale overlaid by a general trend of high urban–low rural concentrations. Concentrations of ΣPCBs, ΣPBDEs and ΣPAHs decreased exponentially with distance from the central business district (CBD) such that they were 94, 79 and 90% lower, respectively, at 20 km from the CBD (approx. edge of Toronto). Application of a simple three-dimensional radial dilution model to the measured data, with the CBD as the center-point, gave insight into the balance that exists between sources and processes in an urban area. This radial dilution model best accounted for the drop off of PCB concentrations, suggesting that the CBD was the major geographic source of PCBs, but reinforced that a combination of spatially distributed sources affected concentrations, and thus should be considered in estimates of source distributions or emissions, particularly for PAHs and PBDEs. Urban atmospheric concentrations of PCBs and lower molecular weight PBDEs and PAHs were strongly correlated with temperature which we suggest is due to temperature-dependent emission processes such as volatilization at warm temperatures from local sources and/or surfaces acting as temporary sinks, removal of emissions from air at cold temperatures due to gas-to-particle condensation followed by deposition, and/or releases from indoor air at warm temperatures. PAH concentrations reflected a balance between temperature-dependent volatilization from impervious surfaces, condensation of combustion sources, atmospheric degradation and low winter emissions from natural gas heating. Acknowledgments Funding was provided by a Great Lakes Atmospheric Deposition fund grant to Diamond, Helm, Blanchard and Backus, the Ontario Ministry of the Environment, a Natural Science and Engineering Research Council (NSERC) of Canada Discovery grant to Diamond, and an NSERC PGS-D to Melymuk. Many thanks for the helpful advice provided by Liisa Jantunen and Susan Csiszar. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.scitotenv.2012.04.022. References Aceves M, Grimalt JO. Seasonally dependent size distributions of aliphatic and polycyclic aromatic hydrocarbons in urban aerosols from densely populated areas. Environ Sci Technol 1993;27:2896–908. Alcock RE, Sweetman AJ, Prevedouros K, Jones KC. Understanding levels and trends of BDE-47 in the UK and North America: an assessment of principal reservoirs and source inputs. Environ Int 2003;29:691–8. Baek SO, Field RA, Goldstone ME, Kirk PW, Lester JN, Perry R. A review of atmospheric polycyclic aromatic hydrocarbons — sources, fate and behavior. Water Air Soil Pollut 1991;60:279–300. Batterman SA, Chernyak S, Jia CR, Godwin C, Charles S. Concentrations and emissions of polybrominated diphenyl ethers from US houses and garages. Environ Sci Technol 2009;43:2693–700. Bozlaker A, Muezzinoglu A, Odabasi M. Atmospheric concentrations, dry deposition and air–soil exchange of polycyclic aromatic hydrocarbons (PAHs) in an industrial region in Turkey. J Hazard Mater 2008;153:1093–102. Bunce NJ, Dryfhout HG. Diurnal and seasonal modeling of the tropospheric half-lives of polycyclic aromatic hydrocarbons. Can J Chem (Rev Can Chim) 1992;70:1966–70. Cabrerizo A, Dachs J, Moeckel C, Ojeda MJ, Caballero G, Barcelo D, et al. Ubiquitous net volatilization of polycyclic aromatic hydrocarbons from soils and parameters influencing their soil–air partitioning. Environ Sci Technol 2011;45:4740–7. Cetin B, Odabasi M. Particle-phase dry deposition and air–soil gas-exchange of polybrominated diphenyl ethers (PBDEs) in Izmir, Turkey. Environ Sci Technol 2007;41:4986–92. Cetin B, Yatkin S, Bayram A, Odabasi M. Ambient concentrations and source apportionment of PCBs and trace elements around an industrial area in Izmir, Turkey. Chemosphere 2007;69:1267–77. Chen LG, Mai BX, Bi XH, Chen SJ, Wang XM, Ran Y, et al. Concentration levels, compositional profiles, and gas-particle partitioning of polybrominated diphenyl ethers in the atmosphere of an urban city in South China. Environ Sci Technol 2006;40:1190–6.

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280 Clements AL, Jia YL, Denbleyker A, McDonald-Buller E, Fraser MP, Allen DT, et al. Air pollutant concentrations near three Texas roadways, part II: chemical characterization and transformation of pollutants. Atmos Environ 2009;43:4523–34. Csiszar SA, Diamond ML, Thibodeaux LJ. Modeling urban films using a dynamic multimedia fugacity model. Chemosphere 2012;87:1024–31. Currado GM, Harrad S. Factors influencing atmospheric concentrations of polychlorinated biphenyls in Birmingham, UK. Environ Sci Technol 2000;34:78–82. Desborough J, Harrad S. Chiral signatures show volatilization from soil contributes to polychlorinated biphenyls in grass. Environ Sci Technol 2011;45:7354–7. Diamond M, Goosey E, Csiszar S, Verkoeyon S, Catching C. Reconnaissance study of coal tar sealcoat application in Toronto and an estimate of related PAH emissions. Report prepared for Environment Canada by the Diamond Environmental Group. Toronto: University of Toronto; 2011. Diamond ML, Gingrich SE, Fertuck K, McCarry BE, Stern GA, Billeck B, et al. Evidence for organic film on an impervious urban surface: characterization and potential teratogenic effects. Environ Sci Technol 2000;34:2900–8. Diamond ML, Hodge E. Urban contaminant dynamics: from source to effect. Environ Sci Technol 2007;41:3796–800. Diamond ML, Melymuk L, Csiszar SA, Robson M. Estimation of PCB stocks, emissions, and urban fate: will our policies reduce concentrations and exposure? Environ Sci Technol 2010;44:2777–83. Dickhut RM, Gustafson KE. Atmospheric washout of Polycyclic Aromatic Hydrocarbons in the Southern Chesapeake Bay region. Environ Sci Technol 1995;29:1518–25. Dong TTT, Lee BK. Characteristics, toxicity, and source apportionment of polycyclic aromatic hydrocarbons (PAHs) in road dust of Ulsan, Korea. Chemosphere 2009;74: 1245–53. Du S, Wall SJ, Cacia D, Rodenburg LA. Passive air sampling for polychlorinated biphenyls in the Philadelphia metropolitan area. Environ Sci Technol 2009;43:1287–92. Dvorska A, Lammel G, Klánová J. Use of diagnostic ratios for studying source apportionment and reactivity of ambient polycyclic aromatic hydrocarbons over Central Europe. Atmos Environ 2011;45:420–7. Environment Canada. National Air Pollution Surveillance Program, 2011. Ottawa: Government of Canada; 2008. Environment Canada. National Pollutant Release Inventory. Ottawa: Government of Canada; 2009. Environment Canada. National Climate Data and Information Archive; 2010. Farrar NJ, Harner T, Shoeib M, Sweetman A, Jones KC. Field deployment of thin film passive air samplers for persistent organic pollutants: a study in the urban atmospheric boundary layer. Environ Sci Technol 2005;39:42–8. Finlayson-Pitts BJ, Pitts JNJ. Airborne polycyclic aromatic hydrocarbons and their derivatives: atmospheric chemistry and toxicological implications. Chemistry of the upper and lower atmosphere. San Diego: Academic Press; 2000. Fromme H, Twardella D, Dietrich S, Heitmann D, Schierl R, Liebl B, et al. Particulate matter in the indoor air of classrooms — exploratory results from Munich and surrounding area. Atmos Environ 2007;41:854–66. Gabrieli J, Vallelonga P, Cozzi G, Gabrielli P, Gambaro A, Sigl M, et al. Post 17th-century changes of European PAH emissions recorded in high-altitude alpine snow and ice. Environ Sci Technol 2010;44:3260–6. Gasic B, Moeckel C, Macleod M, Brunner J, Scheringer M, Jones KC, et al. Measuring and modeling short-term variability of PCBs in air and characterization of urban source strength in Zurich, Switzerland. Environ Sci Technol 2009;43:769–76. Gingrich S, Macfarlane R. Air pollution from wood-burning fireplaces and stoves. City of Toronto, Toronto: Toronto Public Health; 2002. Gingrich SE, Diamond ML, Stern GA, McCarry BE. Atmospherically derived organic surface films along an urban–rural gradient. Environ Sci Technol 2001;35:4031–7. Gouin T, Harner T, Daly GL, Wania F, Mackay D, Jones KC. Variability of concentrations of polybrominated diphenyl ethers and polychlorinated biphenyls in air: implications for monitoring, modeling and control. Atmos Environ 2005;39: 151–66. Gouin T, Thomas GO, Chaemfa C, Harner T, Mackay D, Jones KC. Concentrations of decabromodiphenyl ether in air from Southern Ontario: implications for particlebound transport. Chemosphere 2006;64:256–61. Goyal R, Khare M. Indoor–outdoor concentrations of RSPM in classroom of a naturally ventilated school building near an urban traffic roadway. Atmos Environ 2009;43: 6026–38. Gustafson KE, Dickhut RM. Particle/gas concentrations and distributions of PAHs in the atmosphere of southern Chesapeake Bay. Environ Sci Technol 1997;31:140–7. Halsall CJ, Lee RGM, Coleman PJ, Burnett V, Harding-Jones P, Jones KC. PCBs in UK urban air. Environ Sci Technol 1995;29:2368–76. Harner T, Shoeib M, Diamond M, Stern G, Rosenberg B. Using passive air samplers to assess urban–rural trends for persistent organic pollutants. 1. Polychlorinated biphenyls and organochlorine pesticides. Environ Sci Technol 2004;38:4474–83. Harrad S, Hunter S. Concentrations of polybrominated diphenyl ethers in air and soil on a rural–urban transect across a major UK conurbation. Environ Sci Technol 2006;40:4548–53. Harrad S, Laurie L. Concentrations, sources and temporal trends in atmospheric polycyclic aromatic hydrocarbons in a major conurbation. J Environ Monit 2005;7:722–7. Harrad S, Mao HJ. Atmospheric PCBs and organochlorine pesticides in Birmingham, UK: concentrations, sources, temporal and seasonal trends. Atmos Environ 2004;38: 1437–45. Hobson M, Thistlethwaite G. Emission factors programme task 7 — review of residential & small-scale commercial combustion sources. Abingdon, Oxfordshire, UK: Department for Environment, Food and Rural Affairs, UK; 2003. Hodge E, Diamond M. Sources, fate and effects of contaminant emissions in urban areas. In: Harrad S, editor. Persistent organic pollutants. Chichester, U.K.: Blackwell Publishing; 2010

279

Hoh E, Hites RA. Brominated flame retardants in the atmosphere of the east-central United States. Environ Sci Technol 2005;39:7794–802. Hornbuckle KC, Green ML. The impact of an urban-industrial region on the magnitude and variability of persistent organic pollutant deposition to Lake Michigan. Ambio 2003;32:406–11. Hsu YK, Holsen TM, Hopke PK. Locating and quantifying PCB sources in Chicago: receptor modeling and field sampling. Environ Sci Technol 2003;37:681–90. Hu DF, Lehmler HJ, Martinez A, Wang K, Hornbuckle KC. Atmospheric PCB congeners across Chicago. Atmos Environ 2010;44:1550–7. Jamshidi A, Hunter S, Hazrati S, Harrad S. Concentrations and chiral signatures of polychlorinated biphenyls in outdoor and indoor air and soil in a major UK conurbation. Environ Sci Technol 2007;41:2153–8. Jaward FM, Farrar NJ, Harner T, Sweetman AJ, Jones KC. Passive air sampling of PCBs, PBDEs, and organochlorine pesticides across Europe. Environ Sci Technol 2004;38: 34–41. Kakareka SV, Kukharchyk TI, Khomich VS. Study of PAH emission from the solid fuels combustion in residential furnaces. Environ Pollut 2005;133:383–7. Kobližková M, Růžičková P, Čupr P, Komprda J, Holoubek I, Klánová J. Soil burdens of persistent organic pollutants: their levels, fate, and risks. Part IV. Quantification of volatilization fluxes of organochlorine pesticides and polychlorinated biphenyls from contaminated soil surfaces. Environ Sci Technol 2009;43:3588–95. Li NQ, Wania F, Lei YD, Daly GL. A comprehensive and critical compilation, evaluation, and selection of physical–chemical property data for selected polychlorinated biphenyls. J Phys Chem Ref Data 2003;32:1545–90. Li YF, Harner T, Liu LY, Zhang Z, Ren NQ, Jia HL, et al. Polychlorinated biphenyls in global air and surface soil: distributions, air–soil exchange, and fractionation effect. Environ Sci Technol 2010;44:2784–90. Li YM, Zhang QH, Ji DS, Wang T, Wang YW, Wang P, et al. Levels and vertical distributions of PCBs, PBDEs, and OCPs in the atmospheric boundary layer: observation from the Beijing 325-m meteorological tower. Environ Sci Technol 2009;43:1030–5. Liu QT, Diamond ML, Gingrich SE, Ondov JM, Maciejczyk P, Stern GA. Accumulation of metals, trace elements and semi-volatile organic compounds on exterior window surfaces in Baltimore. Environ Pollut 2003;122:51–61. Mahler BJ, Van Metre PC, Bashara TJ, Wilson JT, Johns DA. Parking lot sealcoat: an unrecognized source of urban polycyclic aromatic hydrocarbons. Environ Sci Technol 2005;39:5560–6. Mari M, Schuhmacher M, Feliubadalo J, Domingo JL. Air concentrations of PCDD/Fs, PCBs and PCNs using active and passive air samplers. Chemosphere 2008;70:1637–43. McDonald JG, Hites RA. Radial dilution model for the distribution of toxaphene in the United States and Canada on the basis of measured concentrations in tree bark. Environ Sci Technol 2003;37:475–81. Melymuk L, Robson M, Diamond ML, Bradley LE, Backus S. Wet deposition loadings of organic contaminants to Lake Ontario: assessing the influence of precipitation from urban and rural sites. Atmos Environ 2011a;45:5042–9. Melymuk L, Robson M, Helm PA, Diamond ML. Evaluation of passive air sampler calibrations: selection of sampling rates and implications for the measurement of persistent organic pollutants in air. Atmos Environ 2011b;45:1867–75. Menichini E. Urban air pollution by polycyclic aromatic hydrocarbons — levels and sources of variability. Sci Total Environ 1992;116:109–35. Meyer T, Lei YD, Muradi I, Wania F. Organic contaminant release from melting snow. 2. Influence of snow pack and melt characteristics. Environ Sci Technol 2009;43: 663–8. Moeckel C, Gasic B, Macleod M, Scheringer M, Jones KC, Hungerbuhler K. Estimation of the source strength of polybrominated diphenyl ethers based on their diel variability in air in Zurich, Switzerland. Environ Sci Technol 2010;44:4225–31. Motelay-Massei A, Harner T, Shoeib M, Diamond M, Stern G, Rosenberg B. Using passive air samplers to assess urban–rural trends for persistent organic pollutants and polycyclic aromatic hydrocarbons. 2. Seasonal trends for PAHs, PCBs, and organochlorine pesticides. Environ Sci Technol 2005;39:5763–73. Motelay-Massei A, Ollivon D, Garban B, Teil MJ, Blanchard M, Chevreuil M. Distribution and spatial trends of PAHs and PCBs in soils in the Seine River basin, France. Chemosphere 2004;55:555–65. Oda J, Nomura S, Yasuhara A, Shibamoto T. Mobile sources of atmospheric polycyclic aromatic hydrocarbons in a roadway tunnel. Atmos Environ 2001;35:4819–27. Offenberg JH, Baker JE. Influence of Baltimore's urban atmosphere on organic contaminants over the northern Chesapeake Bay. J Air Waste Manage Assoc 1999;49:959–65. Office of Energy Efficiency. Comprehensive energy use database, 1990 to 2008. Ottawa: Natural Resources Canada; 2011. Paasivirta J, Sinkkonen S, Mikkelson P, Rantio T, Wania F. Estimation of vapor pressures, solubilities and Henry's law constants of selected persistent organic pollutants as functions of temperature. Chemosphere 1999;39:811–32. Persoon C, Peters TM, Kumar N, Hornbuckle KC. Spatial distribution of airborne polychlorinated biphenyls in Cleveland, Ohio and Chicago, Illinois. Environ Sci Technol 2010;44:2797–802. Prevedouros K, Jones KC, Sweetman AJ. Modelling the atmospheric fate and seasonality of polycyclic aromatic hydrocarbons in the UK. Chemosphere 2004;56:195–208. Ras MR, Marce RM, Cuadras A, Mari M, Nadal M, Borrull F. Atmospheric levels of polycyclic aromatic hydrocarbons in gas and particulate phases from Tarragona Region (NE Spain). Int J Environ Anal Chem 2009;89:543–56. Ravindra K, Sokhi R, Van Grieken R. Atmospheric polycyclic aromatic hydrocarbons: source attribution, emission factors and regulation. Atmos Environ 2008;42: 2895–921. Robson M, Melymuk L, Csiszar SA, Giang A, Diamond ML, Helm PA. Continuing sources of PCBs: the significance of building sealants. Environ Int 2010;36:506–13.

280

L. Melymuk et al. / Science of the Total Environment 429 (2012) 272–280

Rogge WF, Hildemann LM, Mazurek MA, Cass GR, Simoneit BRT. Sources of fine organic aerosol .7. Hot asphalt roofing tar pot fumes. Environ Sci Technol 1997;31:2726–30. Rudel RA, Seryak LM, Brody JG. PCB-containing wood floor finish is a likely source of elevated PCBs in residents' blood, household air and dust: a case study of exposure. Environ Health 2008;7. Schenker U, Soltermann F, Scheringer M, Hungerbuhler K. Modeling the environmental fate of polybrominated diphenyl ethers (PBDEs): the importance of photolysis for the formation of lighter PBDEs. Environ Sci Technol 2008;42:9244–9. Schlink U, Rehwagen M, Damm M, Richter M, Borte M, Herbarth O. Seasonal cycle of indoor-VOCs: comparison of apartments and cities. Atmos Environ 2004;38:1181–90. Simcik MF, Zhang HX, Eisenreich SJ, Franz TP. Urban contamination of the Chicago coastal Lake Michigan atmosphere by PCBs and PAHs during AEOLOS. Environ Sci Technol 1997;31:2141–7. Skrdlikova L, Landlova L, Klánová J, Lammel G. Wet deposition and scavenging efficiency of gaseous and particulate phase polycyclic aromatic compounds at a central European suburban site. Atmos Environ 2011;45:4305–12. Su YS, Hung H, Brice KA, Su K, Alexandrou N, Blanchard P, et al. Air concentrations of polybrominated diphenyl ethers (PBDEs) in 2002–2004 at a rural site in the Great Lakes. Atmos Environ 2009;43:6230–7. Sun P, Basu I, Hites RA. Temporal trends of polychlorinated biphenyls in precipitation and air at Chicago. Environ Sci Technol 2006a;40:1178–83. Sun P, Blanchard P, Brice KA, Hites RA. Trends in polycyclic aromatic hydrocarbon concentrations in the Great Lakes atmosphere. Environ Sci Technol 2006b;40:6221–7. Tham YWF, Takeda K, Sakugawa H. Polycyclic aromatic hydrocarbons (PAHs) associated with atmospheric particles in Higashi Hiroshima, Japan: influence of meteorological conditions and seasonal variations. Atmos Res 2008;88:224–33. Totten LA, Panangadan M, Eisenreich SI, Cavallo GJ, Fikslin TJ. Direct and indirect atmospheric deposition of PCBs to the Delaware river watershed. Environ Sci Technol 2006;40:2171–6. U.S. EPA. Toxics release inventory program. United States Environmental Protection Agency; 2011. UTM Meteorological Station. UTMMS Weather Station Data. Mississauga, Ontario: Department of Geography, University of Toronto Mississauga; 2011.

Vallurupalli RK. New Methods to Estimate Average Annual Daily Traffic using Data from Permanent Count Stations. Department of Civil Engineering. M.A.Sc. University of Toronto, Toronto, 1997. Van Metre PC, Majewski MS, Mahler BJ, Foreman WT, Braun CL, Wilson JT, et al. PAH volatilization following application of coal-tar-based pavement sealant. Atmos Environ 2012;51:108–15. Venier M, Hites RA. Atmospheric deposition of PBDEs to the Great Lakes featuring a Monte Carlo analysis of errors. Environ Sci Technol 2008;42:9058–64. Vonderheide AP, Mueller KE, Meija J, Welsh GL. Polybrominated diphenyl ethers: causes for concern and knowledge gaps regarding environmental distribution, fate and toxicity. Sci Total Environ 2008;400:425–36. Wallace LA, Emmerich SJ, Howard-Reed C. Continuous measurements of air change rates in an occupied house for 1 year: the effect of temperature, wind, fans, and windows. J Expo Anal Environ Epidemiol 2002;12:296–306. Wang XY, Li QB, Luo YM, Ding Q, Xi LM, Ma JM, et al. Characteristics and sources of atmospheric polycyclic aromatic hydrocarbons (PAHs) in Shanghai, China. Environ Monit Assess 2010;165:295–305. Ward J, Mohapatra SP, Mitchell A. An overview of policies for managing polybrominated diphenyl ethers (PBDEs) in the Great Lakes basin. Environ Int 2008;34:1148–56. Webster TF, Harrad S, Millette JR, Holbrook RD, Davis JM, Stapleton HM, et al. Identifying transfer mechanisms and sources of decabromodiphenyl ether (BDE 209) in indoor environments using environmental forensic microscopy. Environ Sci Technol 2009;43:3067–72. Wild SR, Jones KC. Polynuclear aromatic hydrocarbons in the United Kingdom Environment — a preliminary source inventory and budget. Environ Pollut 1995;88:91-108. Wong A, Lei YD, Alaee M, Wania F. Vapor pressures of the polybrominated diphenyl ethers. J Chem Eng Data 2001;46:239–42. Zhang HX, Eisenreich SJ, Franz TR, Baker JE, Offenberg JH. Evidence for increased gaseous PCB fluxes to Lake Michigan from Chicago. Environ Sci Technol 1999;33:2129–37. Zhang XM, Diamond ML, Robson M, Harrad S. Sources, emissions, and fate of polybrominated diphenyl ethers and polychlorinated biphenyls indoors in Toronto, Canada. Environ Sci Technol 2011;45:3268–74.