Environ. Sci. Technol. 2009, 43, 1036–1041
Polychlorinated Dibenzo-p-dioxins and Dibenzofurans in the Atmosphere Around the Great Lakes MARTA VENIER,† JOSEPH FERRARIO,‡ A N D R O N A L D A . H I T E S * ,† School of Public and Environmental Affairs, Indiana University, Bloomington, Indiana 47405, and Environmental Chemistry Laboratory, OPP/BEAD, United States Environmental Protection Agency, Stennis Space Center, Mississippi 39520
Received September 17, 2008. Revised manuscript received December 9, 2008. Accepted December 9, 2008.
The atmospheric concentrations of PCDDs and PCDFs were measured in four sites near the shores of the North American Great Lakes. The sites included an urban site (Chicago, Illinois) and three rural/remote sites (Eagle Harbor, Michigan; Sleeping Bear Dunes, Michigan; and Sturgeon Point, New York). Sampling occurred every 24 days between November 2004 and December 2007. The concentration of PCDD/Fs averaged 2.3 ( 0.2 fg WHO TEQ/m3 at Eagle Harbor, 35 ( 3 fg WHO TEQ/ m3 at Chicago, 7.4 ( 1.4 fg WHO TEQ/m3 at Sleeping Bear Dunes, and 13 ( 2 fg WHO TEQ/m3 at Sturgeon Point. The total concentration of the 17 toxic PCDD/F congeners showed a significant seasonal trend at all sites, except Chicago. The date of maximum concentration averaged Dec 6 ( 35 days, which is consistent with residential heating being an important source of PCDD/Fs to the atmosphere. A significant positive relationship between the logarithm of the total concentration of the 17 toxic PCDD/F congeners and the logarithm of the number of people within a 25 km radius around the sampling site was found. We suggest that urban and industrial areas, which are heavily populated, act as sources of PCDDs and PCDFs to the atmosphere.
Introduction Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/ Fs) are well-known toxic compounds. They are nonpolar, lipophilic, and chemically stable (1). Their toxicity varies considerably depending on the exact positions of the chlorine atoms on the rings. Of the 210 possible PCDD/F congeners, only 17, the 2,3,7,8-substituted congeners, are known to be highly toxic and have been assigned toxic equivalency factors by the World Health Organization. The dominant sources of dioxins and furans to the environment are anthropogenic. The major identified sources of environmental releases are grouped into six broad categories: combustion sources, metals smelting, refining and process sources, chemical manufacturing sources, natural sources, and environmental reservoirs (2). Atmospheric transport and deposition are responsible for the ubiquitous presence of PCDD/Fs in soil, lake and ocean sediment, and air (3). PCDD/Fs are present even in the most remote regions of the globe (4). * Corresponding author e-mail:
[email protected]. † Indiana University. ‡ U.S. Environmental Protection Agency. 1036
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 4, 2009
Due to their chemical stability, these compounds make their way up the food chain and bioaccumulate in fatty tissues. Hence, with the exception of occupational settings or accidents, exposure to these chemicals for most people occurs as a result of eating meat, milk, eggs, fish, and related products. The atmosphere is the most relevant transportation medium of these chemicals; nevertheless, few data exist on the atmospheric concentrations of PCDD/Fs, especially at rural and remote sites. The United States Environmental Protection Agency’s (U.S. EPA’s) National Dioxin Monitoring Network provided an extensive database of background concentrations in the United States, but little attention was dedicated to the Great Lakes region (5), which is important to both the United States and Canada. This region hosts one of the world’s largest concentrations of industrial activity, and >10% of the United States population and >25% of the Canadian population live there. In addition, a high percentage of the agricultural and industrial production of both countries is centered in the Great Lakes region (6). Moreover, the Great Lakes form the largest surface freshwater system on earth, containing 84% of North America’s and 21% of the world’s supply. These lakes are particularly vulnerable since their outflow is less than 1% a year, which allows persistent pollutants to accumulate over time. This paper describes measurements of PCDD/Fs in the atmosphere at four sites located on the shores of Lakes Superior, Michigan, and Erie. The sampling site locations ranged from remote (Eagle Harbor, Michigan) to rural (Sleeping Bear Dunes, Michigan, and Sturgeon Point, New York) to urban (Chicago, Illinois). Of particular interest was the comparison between the urban site of Chicago, expected to be a source region, and the rural/remote sites, representing the regional background. These concentration data were also analyzed to determine temporal trends with the ultimate goal of determining if emission controls have been effective or if additional controls are needed. Additionally, a relationship between local human population and PCDD/F atmospheric concentrations was elucidated.
Methods and Materials Sampling Sites. The samples were collected at four sites. The urban samples were collected at the Illinois Institute of Technology in downtown Chicago, Illinois (41° 50′ 04′′ N, 87° 37′ 29′′ W). The three rural or remote samples were collected at Eagle Harbor, Michigan, on the southern shore of Lake Superior (47° 27′ 47′′ N, 88° 08′ 59′′ W), at Sturgeon Point, New York, near the eastern end of Lake Erie (42° 41′ 35′′ N, 79° 03′ 18′′ W), and at the Sleeping Bear Dunes National Lakeshore, on the northeastern shore of Lake Michigan, (44° 45′ 40′′ N, 86° 03′ 31′′ W). These locations are given in Figure 1. Sample Collection. Samples were collected using modified Anderson high-volume samplers calibrated to collect approximately 6500 m3 of air at the rural and remote sites and 1000 m3 at the Chicago site. Sampling events occurred every 24 days for 168 h at the rural and remote sites and for 48 h at Chicago. Particles were collected on quartz fiber filters, and gas-phase compounds were collected on cartridges containing 40 g of XAD-2 resin. Previous experience with these cartridges in our laboratory (7) indicates that breakthrough of semivolatile compounds does not occur to any appreciable extent. Samples were collected from November 1, 2004, to December 31, 2007. Analytical Methods. All sample media precleaning methods followed the Integrated Atmospheric Deposition Net10.1021/es802644w CCC: $40.75
2009 American Chemical Society
Published on Web 01/22/2009
FIGURE 1. Sampling sites for the data reported in this paper. These sites are also used by the Integrated Atmospheric Deposition Network (IADN). work (IADN) Standard Operating Procedures (7). After sampling, the XAD-2 resin and filters for each sample were combined, spiked with isotopically labeled surrogate standards (the 17 13C-labeled dioxins/furans congeners), and Soxhlet extracted for 16 h with toluene. Extracts were cleaned using acid/base silica gel, carbon, and alumina columns. Samples were analyzed for the 17 toxic PCDD/Fs and for the 10 Cl4 to Cl8 PCDD/F homologues at the U.S. EPA’s Environmental Chemistry laboratory located at the Stennis Space Center in Mississippi. Details of the instrumental method can be found elsewhere (8). All analyses were performed on a Micromass Autospec Ultima high-resolution mass spectrometer operated in the selected ion mode at a mass resolution of 10,000. Separation was achieved using a Hewlett-Packard 6890 series gas chromatograph fitted with a 60-m DB-5 MS column. Quality Control. Three field blanks (once each year) per site were collected and processed in a manner identical to that of the samples. The amount of PCDD/Fs in the field blanks was generally low, ranging from 2 to 8% of the average mass found in the samples. Therefore, sample concentrations for the entire study were not blank corrected. Method blanks were examined for the presence of interferences; they were negligible. Analyte concentrations below the detection limit are reported as nondetects. Sample sets were reviewed by U.S. EPA’s QA/QC officer to ensure compliance with the QA/ QC guidelines.
Results and Discussion Ambient PCDD/F Concentrations. All of the measured concentrations are reported in Table S1 of the Supporting Information. In this paper, the notation ΣPCDD/F refers to the sum of the 17 toxic PCDD/F congeners concentrations (in fg/m3), and TEQDF refers to the corresponding toxic equivalent concentration obtained using the 1998 World Health Organization Toxic Equivalent Factors (in fg TEQ/ m3). As can be observed from Figure 2, the average ΣPCDD/F and TEQDF concentrations were highest at the urban site in Chicago (1300 ( 110 fg/m3 and 35 ( 3 fg TEQ/m3; given as average ( standard error), followed by the rural site of Sturgeon Point (740 ( 83 fg/m3 and 13 ( 2 fg TEQ/m3) and the two remote sites of Sleeping Bear Dunes and Eagle Harbor
FIGURE 2. Total PCDD/PCDF concentration (fg/m3) and total TEQDF concentrations (in fg TEQ per m3) at the four Great Lakes sites ordered from west to east. The horizontal lines represent the median, and the dotted lines represent the mean. The boxes represent the 25th and 75th percentiles, and the whiskers represent the 5th and 95th percentiles. The numbers in parentheses represent the number of samples reported at each site. (400 ( 93 fg/m3 and 7.4 ( 1.4 fg TEQ/m3 and 120 ( 13 fg/m3 and 2.3 ( 0.2 fg TEQ/m3, respectively). The statistically supported sequence was: CH > SP > SB ) EH (Tukey’s post hoc test, P < 0.05, for ΣPCDD/F). The levels measured in this study are in good agreement with those reported in similar studies in the United States, VOL. 43, NO. 4, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1037
FIGURE 3. Mean congener concentrations in air at the four Great Lakes sites as a percentage of the ΣPCDD/F concentration at the four Great Lakes sites. The bars represent the standard errors.
TABLE 1. Values of the Coefficients Obtained by Regression of Equation 1 for the ΣPCDD/F Concentrations (in fg/m3) a0, intercepta Eagle Harbor Chicago Sleeping Bear Dunes Sturgeon Point average a
a1, rate const. (days-1)
4.70 ( 0.20 -5.30 ( 0.23 6.49 ( 0.19 5.50 ( 0.20
(One standard error about the mean.
NS -NS NS -b
b
9
a3, period (days)
a4, offset
maximum date
0.46 ( 0.13 -0.67 ( 0.16 0.37 ( 0.13 0.50 ( 0.14
365 ( 21 -365 ( 5 365 ( 22 365 ( 19
6.84 ( 0.67 -6.33 ( 0.47 5.42 ( 0.67 6.20 ( 0.61
Oct 29 ( 39 -Nov 28 ( 27 Jan 20 ( 39 Dec 6 ( 35
NS ) not significant.
and they are lower than those reported for Europe and Asia. In the United States, the mean concentrations detected by the National Dioxin Monitoring Network (NDAMN) ranged from 6.4 to 15 fg WHO-TEQ/m3 in rural areas and from 0.1 to 3 fg WHO-TEQ/m3 in remote areas (5). For urban sites, atmospheric concentrations varied from 21 fg WHO-TEQ/ m3 in Oklahoma City (9) and 19 fg WHO-TEQ/m3 in Houston, Texas (10) to 130 fg WHO TEQ/m3 in Los Angeles, California (11). In Europe, Spanish mean concentrations measured at industrial, high traffic, and rural sites in Catalonia averaged 140, 72, and 28 fg I-TEQ/m3, respectively (12). Atmospheric PCDD/F concentrations varied from 73 fg WHO TEQ/m3 in Athens, Greece (13) and 81 fg WHO-TEQ/m3 in Lancaster, United Kingdom (14) to 27 fg WHO TEQ/m3 in Frederiksborg, Denmark, which is located 30 km north of Copenhagen (15). In Asia, ambient dioxin concentrations ranged from 7.3 to 580 fg I-TEQ/m3 in Hong Kong (16) and from 140 to 500 fg I-TEQ/m3 in Shanghai, China (17). In Japan, atmospheric levels of PCDD/F averaged 50 fg WHO-TEQ/m3 in 2006 (18). On the remote island of Tasmania, atmospheric ΣPCDD/F concentrations ranged from 0.23 to 1.2 fg/m3, with an average of 0.57 fg/m3 (19). The concentrations of dioxins and furans reported here and elsewhere for the United States are generally lower than those at comparable sites in other parts of the world. 1038
a2, amplitude
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 4, 2009
Homologue and Congener Profiles. The composition of the PCDD/F mixture present in the air samples can provide useful hints about the nature of the sources, as well as about the transformation processes that occur during transport from the sources to the sampling sites. For example, it has been shown that homologue patterns are generally quite consistent, except for those close to significant sources (20). For the sites in this study, octachlorodibenzo-p-dioxin (OCCD) was by far the largest contributor to ΣPCDD/F concentrations, accounting for 61-65% of the total (see Figure 3). The second most abundant measured congener was 1,2,3,4,6,7,8-HpCDD, accounting for 16-19% of the total. Among the dibenzofurans, octachlorodibenzofuran (OCDF) and 1,2,3,4,6,7,8-HpCDF were the most abundant measured congeners, accounting for 3.5-4.5% and 3.6-4.5%, respectively, of the total. Although the 17 toxic congeners are the most studied and reported compounds, information is also available from the homologues patterns. For the homologues, a trend of increasing PCDD and decreasing PCDF relative levels as a function of increasing chlorination was observed (see Figure S1). Thus, at all sites, OCDD dominated among the PCDD homologues (ranging between 28-35%), while TCDF dominated among the PCDF (ranging between 10-18%). Interestingly, there was a statistically significant difference
FIGURE 4. Temporal trends of the natural logarithm of ΣPCDD/F (in fg/m3) at the four Great Lakes sites. The curves were fitted using equation 1; if no curve is given, this regression was not significant. between the contribution of both OCDD and TCDF to the total homologue concentration at the urban site in Chicago and that at the rural/remote sites of Eagle Harbor, Sleeping Bear Dunes, and Sturgeon Point taken as a group (Tukey’s post hoc test, P < 0.01). No significant differences could be determined among the latter three sites, suggesting a common source or similar transformation processes occurring during transport through the atmosphere. Similar findings were reported in previous studies of atmospheric samples from other parts of the world (10, 17, 20). When looking at the TEQDF concentrations, 2,3,4,7,8PeCDF and 1,2,3,7,8-PeCDD were the major contributors, accounting for 18-27% and 24-28%, respectively, of the TEQDF. Thus, these two congeners were responsible for ∼50% of the overall toxicity. Other significant contributions to the TEQDF were 1,2,3,4,6,7,8-HpCDD (7-11%); 1,2,3,6,7,8-HxCDD and 1,2,3,7,8,9-HxCDD (5-7%); and 1,2,3,4,7,8-HxCDF, 1,2,3,6,7,8-HxCDF, and 2,3,4,6,7,8-HxCDF (4-6%). Similar patterns were found in air samples collected in Spain (12), Sa˜o Paolo, Brazil (21), Shanghai, China (17), Houston, Texas (10), and 26 worldwide sites (20). Temporal Trends. Following the approach of Sun et al. (22), the natural logarithms of the ΣPCDD/F concentrations (C) were fitted by the following function: ln C ) a0 + a1t + a2sin
(
)
2πt + a4 a3
(1)
where C is the concentration in fg/m3 in a particular sample, t is the time in Julian Days (relative to November 1, 2004) when that sample was taken, a0 is an intercept (unitless), a1 is a first order rate constant (days-1) that describes an exponential decrease or increase of the concentrations over time, a2 is a periodic amplitude (unitless), a3 is the length of the period (days), and a4 is a periodic offset. This curve fitting was done in SigmaPlot 8.0 using the constraint that a3 ) 365.24 days. The results are given in Table 1 and illustrated in Figure 4. All of the rural and remote sites showed a significant periodicity, defined as P < 0.05 for both the amplitude, a2, and the length of the period, a3. The urban site at Chicago did not show any significant periodicity. This observation
FIGURE 5. Atmospheric ΣPCDD/F concentrations (fg/m3) as a function of human population within a 25-km radius of the sampling site in North America (n ) 60). The black line is the linear regression of the following equation: log (PCDD/F) ) y0 + y1 log (pop). The fitted parameters are y0 ) 0.30 ( 0.04 and y1) 1.17 ( 0.18, respectively, and r2 ) 0.536. The symbols are color coded as follows: red ) NDAMN sites, cyan ) NAPS sites, green ) CADAMP sites, yellow ) sites reported here. The black dotted lines represent the 95% confidence limits. may be related to the shorter sampling time at this site compared to the other three (48 h vs 168 h) or to the presence of nonseasonal sources of PCDD/Fs in urban areas, such as medical and municipal waste incineration (2). From the periodic offset, a4, we calculated the date corresponding to the maximum concentration at each site. At all the sites, the highest concentration of ΣPCDD/Fs was observed in the winter: October 29 ( 39 at Eagle Harbor, November 28 ( 27 at Sleeping Bear Dunes, and January 20 ( 39 at Sturgeon Point. The shift in the dates of maximum concentration from late-fall at Eagle Harbor to midwinter at Sturgeon Point may be related to the decreasing latitude of these sites, which determines the start of the coldest months (see Figure 1). In the winter, when temperatures are lower, space heating requirements are highest, suggesting that combustion is an important source of PCDD/Fs to the VOL. 43, NO. 4, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1039
atmosphere. The peak-to-valley ratio is given by exp(2a2), and this value is a measure of the intensity of the seasonal effect. On average, this value was 2.7, with Sleeping Bear Dunes showing a peak-to-valley ratio significantly higher (P < 0.05) than those at the two other sites, which were similar to one another. This average suggests that ΣPCDD/F concentrations were about three times higher in the winter than in the summer, which is a notable seasonal effect. The ΣPCDD/F concentrations did not change significantly at any of the sites as a function of time over the course of this study (P > 0.05 for a1). Although PCDD/F atmospheric concentrations are thought to be generally declining over the past decade or so (3), this trend could not be detected in these atmospheric samples. In another study, the atmospheric half-lives of PCDD/Fs were estimated to be about two years in samples collected in Bermuda over a 4-year period (23). Considering that the data reported here cover a time span of only 3 years, we suspect that a longer time might be necessary before we would be able to detect a change in concentration with time for these compounds. We also investigated the temporal trends of each of the seven PCDD congeners using a similar approach (see Table S2). The PCDF congeners were not included in this analysis since their concentrations were much lower than those of the PCDDs. Chicago did not show any significant periodicity for any of the congeners, which is consistent with urban combustion sources other than residential heating, as discussed above. At the other sites, with a few exceptions (2,3,7,8-TCDD at Eagle Harbor and 1,2,3,4,6,7,8-HpCDD and OCDD at Sturgeon Point), all of the congeners showed a significant periodicity (P < 0.05 for a2 and a3). The lack of periodicity for 2,3,7,8-TCDD at Eagle Harbor is probably related to the relatively low concentrations (close to background levels) measured at that site. The lack of periodicity of two congeners that were already found to characterize urban air masses (24)s1,2,3,4,6,7,8-HpCDD and OCDDsat Sturgeon Point suggests the mixed rural and urban nature of this location. This site exhibited a similar effect for polybrominated diphenyl ethers (PBDEs) (25). Similar to ΣPCDD/Fs, the maximum concentrations for single congeners occurred during the colder months and followed the same latitudinal trend, peaking from middle to late fall at Eagle Harbor to middle winter at Sturgeon Point. The most pronounced seasonal effect was observed at Sleeping Bear Dunes, with peak-to-valley ratios ranging from 3.5 for OCDD to 6.8 for 1,2,3,7,8,9-HxCDD. It is not yet clear what might explain these differences. Finally, as expected given the results for ΣPCDD/Fs, none of the individual congeners showed a significant increase or decrease as a function of time (P > 0.05 for a1). Overall, these findings confirm the seasonal character of the atmospheric concentrations of PCDD/Fs, suggesting that combustion, probably related to space heating in the winter, is a major source of these compounds. In urban centers, non-seasonal sources, such as medical and municipal waste incineration, might also play an important role. At the moment, given the lack of change with time, it is not yet possible to determine if combustion emission controls have been effective in reducing the atmospheric levels of PCDD/ Fs in the Great Lakes region. Relationship between ΣPCDD/F and Population. To verify the hypothesis that urban centers are sources of atmospheric PCDD/Fs, we analyzed the relationship between local population (as a surrogate for total combustion in an urban center) and ΣPCDD/F concentration. In addition to the results reported here, several other locations were included in this analysis. This analysis was restricted to the North American region, and only results from large sampling networks were included. The following sources were employed: the EPA National Dioxin Monitoring Network 1040
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 4, 2009
(NDAMN) which deployed samplers mainly in rural and remote locations around the United States (5); the California Ambient Dioxin Air Monitoring Program (CADAMP) which collected samples predominantly in heavily populated areas of California (26); and the Canadian National Air Pollution Surveillance Network (NAPS) (27) which sampled air throughout Canada. For consistency among the data sets, only data collected in 2002 were used, with the exception of the data reported here. With few exceptions, the method for calculating the local population was the same as described by Hafner et al. (28); hence, only a brief description will be given here. Once the coordinates of each site were determined (they were either available directly from the source or they were determined using mapping devices), the population within a 25-km radius around each sampling site was calculated. Population information was obtained from the Landscan database created by the Oak Ridge National Laboratory (ORNL) Global Population Project. The 2002 database was employed in this analysis (29). Using ArcGIS 9.0, the Landscan raster was projected using the Lambert Azimuthal projection, which preserves the area. Twenty-five-km radius buffers were created around each site and then used as “masks” to extract the population information from the Landscan raster. The full data set used in this analysis is reported in Table S3. The plot of the logarithm of the population versus the logarithm of the concentration of ΣPCDD/Fs (in fg/m3) for the 60 continental sites is shown in Figure 5. A linear regression of these two variables was statistically significant (P < 0.0001), indicating that concentrations of ΣPCDD/Fs increase exponentially with the logarithm of the local population. When the local population doubles, the atmospheric ΣPCDD/F concentration more than doubles (21.17 ) 2.25). There are two outliers that should be mentioned: Saint John, New Brunswick, Canada and Dixon Springs, Illinois. In the first case, there might have been an overestimation of the population for this site; the Landscan database might misrepresent the real situation in rural and sparsely populated regions or at sites close to the coast (28). In the second case, the reported concentration seems particularly high given the nature of the site, but the reason for this observation is not yet clear. Given the evidence discussed above, we suggest that large urban areas and industrialized centers, which are highly populated, act as sources of PCDD/Fs to the atmosphere. Long range transport, together with relatively small local sources, are responsible for the concentrations observed in rural and remote places. A similar trend has also been demonstrated for PBDEs (25) and PAHs (28).
Acknowledgments We thank the Great Lakes Commission for funding (Project Officer Jon Dettling); Ilora Basu and Team IADN for the operation of the network; David Cleverly at the U.S. Environmental Protection Agency for supplying the NDAMN data; Chris Byrne, Tripp Boone, and Craig Vigo (all from the EPA/ OPP/BEAD Environmental Chemistry Laboratory) for QA/ QC, coordination of sample preparation, and for HRMS analyses, respectively. Although the research described in this paper has been partially funded by the U.S. Environmental Protection Agency, it has not been subjected to the Agency’s peer and administrative review and, therefore, may not necessarily reflect the views of the EPA; nor does the mention of trade names or commercial products constitute endorsement or recommendation of use.
Supporting Information Available The full data set of results for the four Great Lakes sampling sites (Table S1); values of the equation 1 regression coefficients for the concentrations of individual PCDD congeners
versus time (Table S2); sampling sites used for the regression between local population and total PCDD/F concentrations (Table S3); and mean homologue distributions at the four sampling sites (Figure S1). This information is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Rappe, C. Sources and environmental concentrations of dioxins and related compounds. Pure Appl. Chem. 1996, 68 (9), 1781– 1789. (2) U.S. EPA. An Inventory of Sources and Environmental Releases of Dioxin-Like Compounds in the U.S. for the Years 1987, 1995, and 2000;EPA/600/P-03/002f, Final Report; U.S. Environmental Protection Agency: Washington, DC, November, 2006. (3) Alcock, R. E.; Jones, K. C. Dioxins in the environment: A review of trend data. Environ. Sci. Technol. 1996, 30 (11), 3133–3143. (4) Hung, H.; Blanchard, P.; Poole, G.; Thibert, B.; Chiu, C. H. Measurement of particle-bound polychlorinated dibenzo-pdioxins and dibenzofurans (PCDD/Fs) in Arctic air at Alert, Nunavut, Canada. Atmos. Environ. 2002, 36 (6), 1041–1050. (5) Cleverly, D.; Ferrario, J.; Byrne, C.; Riggs, K.; Joseph, D.; Hartford, P. A general indication of the contemporary background levels of PCDDs, PCDFs, and coplanar PCBs in the ambient air over rural and remote areas of the United States. Environ. Sci. Technol. 2007, 41 (5), 1537–1544. (6) The Great Lakes, An Environmental Atlas and Resource Book, 3rd ed.; U.S. Environmental Protection Agency and Government of Canada: Washington, DC, 1995. (7) Basu, I. Analysis of PCBs, Pesticides, and PAHs in Air and Precipitation Samples: IADN Project, Sample preparation Procedure; Indiana University: Bloomington, IN, 2000. (8) U.S. Environmental Protection Agency. Method TO-9A - Sampling and Analysis Method for the Determination of Polychlorinated, Polybrominated, and Brominated/Chlorinated Dibenzop-Dioxins and Dibenzofurans in Ambient Air. In Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air, 2nd ed.; U.S. EPA: Washington, DC, 1999. (9) Cleverly, D. H.; Holdren, M.; Stafford, K.; Winters, D.; Riggs, K.; Ferrario, J.; Hartford, P.; Morrison, J.; Holowecky, P.; Wisbith, T. Urban air transect study to investigate urban areas as sources of PCDDs and PCDFs to the environment. Presented at the 23rd International Symposium on Halogenated Organic Pollutants, Boston, MA, 2003. (10) Correa, O.; Rifai, H.; Raun, L.; Suarez, M.; Koenig, L. Concentrations and vapor-particle partitioning of polychlorinated dibenzop-dioxins and dibenzofurans in ambient air of Houston, TX. Atmos. Environ. 2004, 38 (39), 6687–6699. (11) Maisel, B. E.; Hunt, G. T. Background concentrations of PCDDs/ PCDFs in ambient air -- A comparison of toxic equivalency factor (TEF) models. Chemosphere 1990, 20 (7-9), 771–778. (12) Abad, E.; Martinez, K.; Gustems, L.; Gomez, R.; Guinart, X.; Hernandez, I.; Rivera, J. Ten years measuring PCDDs/PCDFs in ambient air in Catalonia (Spain). Chemosphere 2007, 67 (9), 1709–1714.
(13) Mandalakis, M.; Tsapakis, M.; Tsoga, A.; Stephanou, E. G. Gasparticle concentrations and distribution of aliphatic hydrocarbons, PAHs, PCBs and PCDD/Fs in the atmosphere of Athens (Greece). Atmos. Environ. 2002, 36 (25), 4023–4035. (14) Lohmann, R.; Lee, R. G. M.; Green, N. J. L.; Jones, K. C. Gasparticle partitioning of PCDD/Fs in daily air samples. Atmos. Environ. 2000, 34 (16), 2529–2537. (15) Hovmand, M. F.; Vikelsoe, J.; Andersen, H. V. Atmospheric bulk deposition of dioxin and furans to Danish background areas. Atmos. Environ. 2007, 41 (11), 2400–2411. (16) Peng, X. L.; Choi, M. P. K.; Wong, M. H. Receptor modeling for analyzing PCDD/F and dioxin-like PCB sources in Hong Kong. Environ. Monit. Assess. 2007, 12 (3), 229–237. (17) Li, H. R.; Feng, H. L.; Sheng, G. Y.; Lu, S. L.; Fu, J. M.; Peng, P. A.; Man, R. The PCDD/F and PBDD/F pollution in the ambient atmosphere of Shanghai, China. Chemosphere 2008, 70 (4), 576– 583. (18) Environmental Survey of Dioxins in FY 2006; Ministry of the Environment, Government of Japan; http://www.env.go.jp/en/ headline/file_view.php?serial)186&hou_id)652. (19) Gras, J.; Mu ¨ ller, J.; Graham, B.; Symons, R.; Carras, J.; Cook, G. Dioxins in ambient air in Australia; Department of the Environment and Heritage; Canberra, Australia, 2004. (20) Lohmann, R.; Jones, K. C. Dioxins and furans in air and deposition: A review of levels, behavior and processes. Sci. Total Environ. 1998, 219 (1), 53–81. (21) Assuncao, J. V. d.; Pesquero, C. R.; Bruns, R. E.; Carvalho, L. R. F. Dioxins and furans in the atmosphere of Sao Paulo City, Brazil. Chemosphere 2005, 58 (10), 1391–1398. (22) Sun, P.; Basu, I.; Hites, R. A. Temporal trends of polychlorinated biphenyls in precipitation and air at Chicago. Environ. Sci. Technol. 2006, 40 (4), 1178–1183. (23) Baker, J. I.; Hites, R. A. Polychlorinated dibenzo-p-dioxins and dibenzofurans in the remote North Atlantic marine atmosphere. Environ. Sci. Technol. 1999, 33 (1), 14–20. (24) Lohmann, R.; Brunciak, P. A.; Dachs, J.; Gigliotti, C. L.; Nelson, E.; Van Ry, D.; Glenn, T.; Eisenreich, S. J.; Jones, J. L.; Jones, K. C. Processes controlling diurnal variations of PCDD/Fs in the New Jersey coastal atmosphere. Atmos. Environ. 2003, 37, 959–969. (25) Venier, M.; Hites, R. A. Flame retardants in the atmosphere near the Great Lakes. Environ. Sci. Technol. 2008, 42 (13), 4745– 4751. (26) California Ambient Dioxin Air Monitoring Program (CADAMP); available at http://www.arb.ca.gov/aaqm/qmosopas/dioxins/ dioxins.htm. (27) National Air Pollution Surveillance (NAPS) Network, Environment Canada; available at http://www.etc-cte.ec.gc.ca/NapsAnnualRawData. (28) Hafner, W. D.; Carlson, D. L.; Hites, R. A. Influence of local human population on atmospheric polycyclic aromatic hydrocarbon concentrations. Environ. Sci. Technol. 2005, 39 (19), 7374–7379. (29) LandScanTM Global Population Database; available at http://www.ornl.gov/landscan/.
ES802644W
VOL. 43, NO. 4, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1041
