Further Studies on the Latitudinal and Temporal Trends of Persistent

Mar 27, 2004 - ... campaign in which PCB data were previously obtained in 1994−1996 and 1998−2000. Comparisons of the masses of chemicals sequeste...
1 downloads 13 Views 163KB Size
Download PDF
Environ. Sci. Technol. 2004, 38, 2523-2530

Further Studies on the Latitudinal and Temporal Trends of Persistent Organic Pollutants in Norwegian and U.K. Background Air FODAY M. JAWARD,† S A N D R A N . M E I J E R , †,‡ E I L I V S T E I N N E S , § GARETH O. THOMAS,† AND K E V I N C . J O N E S * ,† Department of Environmental Science, Institute of Environmental and Natural Sciences, Lancaster University, Lancaster LA1 4YQ, U.K., and Department of Chemistry, Norwegian Institute of Science and Technology, N-7034 Trondheim, Norway

Data are presented for PBDEs, PCBs, and selected organochlorine compounds, measured at background locations by passive air samplers (semipermeable membrane devices, SPMDs) along a latitudinal transect from the south of the U.K. to the north of Norway during 2000-2002. This work is part of an ongoing air sampling campaign in which PCB data were previously obtained in 1994-1996 and 1998-2000. Comparisons of the masses of chemicals sequestered by the SPMDs during these different time intervals are used to investigate spatial and temporal trends. The study yielded examples of compounds that increase, decrease, or remain uniform with latitude, suggestive of differences in the relative importance of deposition versus atmospheric reaction in controlling their long-range atmospheric transport potential. The main constituents of the penta-BDE product were detected at amounts equivalent to 2.0 (range 1.1-2.5) and 1.1 (0.8-1.6) pg m-3 for the U.K. and Norway background sites, respectively. Fractionation of PBDEs was observed, because the amounts of lighter BDEs decreased with latitude, while the heavier molecular weight congeners were quite uniformly distributed. In contrast, the sequestered amounts of the lighter PCBs were uniformly distributed with latitude, with heavier PCBs decreasing. Sequestered amounts of hexachlorobenzene increased with latitude. Preliminary PCB atmospheric clearance rates were derived using the 1994-1996, 1998-2000, and 2000-2002 data. They averaged ca. 3.5 years for all congeners/locations. No systematic differences in congeners or locations were noted, supporting the hypothesis that the underlying trends in European background air are still controlled by primary, rather than secondary, sources.

* Corresponding author phone: 44 1524 593-972; fax: 44 1524 593-985; e-mail: [email protected]. † Lancaster University. ‡ Present address: Department of Environmental Chemistry, IIQAB-CSIC, Jordi Girona 18-26, Barcelona 08034, Catalunya, Spain. § Norwegian Institute of Science and Technology. 10.1021/es035292t CCC: $27.50 Published on Web 03/27/2004

 2004 American Chemical Society

Introduction Concerns over persistent organic pollutants (POPs) in the environment arise because of their susceptibility to longrange atmospheric transport (LRAT), persistence, bioaccumulation tendency, and potential toxicity. This has led to international measures to control their release into the environment (1) and to better understand the processes that influence their fate and transport on a regional and global scale. Environmental monitoring data are required to assess the effectiveness of these source reduction measures, and to yield information on the fate and distribution of these chemicals in the environment. These data need to provide information on the spatial and temporal trends of POPs in different media (2), to provide clues as to the key processes controlling their fate and distribution. Uncertainties remain over their ambient sources, atmospheric transport and fate, and air-surface exchange. Air concentrations vary spatially and temporally, and ultimately influence the global fate of POPs and their entry into food chains. As outlined previously (3), simultaneous measurements of air concentrations in different locations are needed to assess the relative importance of source regions, atmospheric/global fate processes (e.g., fractionation, cold condensation), the LRAT potential of POPs, and global clearance mechanisms. This is possible with passive air sampling techniques (3-8). A range of different passive sampling devices has been utilized to study POP distributions at the local scale (9, 10) and to conduct urban-rural (11) and latitudinal (3, 6) transects. We also recently demonstrated the feasibility of obtaining ambient data on a continental scale, to shed light on large-scale source/sink/transport issues (12, 13). An approach we first used for regional scale monitoring was to deploy semipermeable membrane devices (SPMDs) as passive air samplers (3, 6). SPMDs consist of a polyethylene dialysis bag filled with 1 mL of triolein lipid and have been deployed as an approach to short- and long-term, integrated passive air sampling for POPs (3, 6, 9, 14-17). SPMDs have a relatively high capacity for retaining POPs and a long linear uptake period. Uptake by SPMDs can be envisaged as a threestep process (initial linear uptake, curvilinear portion as equilibrium is approached, and equilibrium between the surface and the gas-phase concentration). Lighter POPs with lower KOA values are expected to reach equilibrium faster, while heavier POPs with higher KOA values are sampled during the linear uptake period. SPMDs sample POPs from both the vapor phase and the particle phase, although they sample the former more efficiently. Once a particle has adhered to the surface of an SPMD, and has effectively been “sampled” by it, compounds associated with that particle can desorb from the particle and be absorbed by the SPMD. Thus, there are still uncertainties associated with particulate sampling by SPMDs. During 1994-1996, we deployed SPMDs as passive air samplers at remote sites on a latitudinal transect from southern England to northern Norway, to look for evidence of the LRAT of PCBs and their potential to undergo global fractionation (6). These sites were carefully selected, away from local sources. Full details of this campaign are given elsewhere (6). Our intention is to continue to use the same sampling network over the long term, to gain insights into how concentrations of PCBs and other POPs change in background air spatially and temporally. We therefore published a follow-up paper, comparing PCB data from a 1998-2000 deployment on the network (3). Comparison of the masses of chemical sequestered by the SPMDs during these different time intervals provided evidence that primary VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2523

FIGURE 1. Location of the sampling sites.

TABLE 1. Sample Location Information

site

lat (°N)

mean annual temp (°C) (3)

1 2 3 4 5 6 7 8 9 10 11

50.75 52.45 54.02 56.10 58.05 58.53 61.25 61.33 64.97 67.38 69.83

9.9 9.8 9.1 9.2 7.1 6.3 2.0 6.5 1.0 4.4 0.0

a

deployment temp range, 1994-1996 (6) min max -7 -11 -6 -4 -7 -14 -31 -12 -28

+32 +33 +31 +32 +30 +31 +26 +28 +28

-31

+26

deployment temp range, 1998-2000 (3) min max -6 -6 -6 -1 -6 -13 -34 -9 -29 -15 -36

+29 +34 +30 +29 +29 +32 +33 +31 +30 +32 +30

deployment temp range, 2000-2002 min max -5 -8 -5 -4 a -15 -36 -10 -29 -16 -31

+32 +32 +30 +25 a +31 +28 a +27 +31 +28

Data not available.

sources probably still control the underlying ambient trends of PCBs. In this paper, we present data from the 2000-2002 campaign, with the objective of (a) examining the latitudinal distribution of PBDEs and a range of organochlorine compounds (hexachlorobenzene (HCB), o,p′-DDT, o,p′-DDE, p,p′-DDT, and p,p′-DDE) in European background air and (b) further examining the time trends of PCBs, and deriving estimates of their atmospheric clearance rates. We also present data on performance reference compounds from these long-term field deployments, an approach that has been proposed to enable site-specific sampler uptake rates to be derived (15, 16, 18).

Materials and Methods SPMDs. Standard U.S. Geological Survey (USGS) SPMDs (8090 cm × 2.5 cm, 75 mm membrane thickness, 1 mL of triolein) were purchased from Environmental Sampling Technologies, St. Joseph, MO. 2524

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004

Sampling Sites and Deployment. Two SPMDs were deployed at each site in Stevenson screen boxes (6, 15). Care was taken not to contaminate the SPMDs during deployment. The sampling sites have been described previously (3, 6) and are shown in Figure 1. The SPMDs were deployed in the summer of 2000 and were collected after two years, in August 2002. Table 1 shows temperature information and the location of the sampling sites, including a comparison with the sampling sites from the 1994-1996 and 1998-2000 studies (3, 6). Mean annual temperatures were compiled using daily temperature data averaged over several decades since the annual temperatures tend to be very similar from year to year (3). Before deployment, SPMDs deployed in U.K. sites were spiked internally with a mixture of 13C12-labeled PCBs (5000 pg each of PCBs 28, 52, 101, 138, 153, and 180) and resealed. Extraction and Cleanup. To be able to compare SPMD concentrations between this study and the previous ones (3, 6), care was taken to deploy, store, and extract the SPMDs

TABLE 2. Percentage Depuration of Permeation Reference Compounds for the U.K. Sites [13C

12]PCB 28 [13C12]PCB 52 13 [ C12]PCB 90/101 [13C12]PCB 138 [13C12]PCB 153 [13C12]PCB 180

site 1

site 2

site 3

site 4

site 5

100 84 50 29 30 20

100 100 82 47 44 22

99 91 57 40 26

98 92 56 27 23 3

98 88 41 15 18 6

in exactly the same way as reported previously. Briefly, after collection the samples were stored in solvent-cleaned metal cans in a freezer until they were extracted. The SPMDs were shaken for 20 s with 100 mL of hexane to remove particulates and exterior lipids (hereafter referred to as the “exterior fraction”). The exterior fraction of each Norway sample was spiked with a recovery spike containing 13C12-labeled PCBs 28, 52, 101, 138, 153, 180, and 209, while the U.K. samples were spiked with PCBs 54 and 155. The Norway SPMDs were then cut open and spiked internally with the [13C12]PCB recovery spike, while the U.K. samples were spiked with PCBs 54 and 155. After resealing, they were dialyzed in hexane in the dark for 2 × 24 h. This dialysate will be referred to as the “interior fraction”. Both fractions went through the same cleanup method, but only the interior fraction was used in the interpretation of the results, and the exterior fraction was merely used as a check, in line with previous studies (3, 6).

SPMD extracts were cleaned on a mixed silica gel/alumina column, followed by size exclusion chromatography, and then fractionated on a mixed silica gel/alumina column as described elsewhere (3, 6). Both fractions were reduced and solvent exchanged to 50 mL of dodecane containing PCB 30, [13C12]PCB 141, and [13C12]PCB 208 as internal standards. All fractions (interior and exterior, F1 and F2) were analyzed by GC-MS with an EI+ source in selected ion monitoring mode (SIM) for PCB and organochlorine compounds. PBDEs were analyzed separately with a Thermo Trace GC-MS system operated in negative chemical ionization source in SIM mode using ammonia as the reagent gas. Details of the instruments, GC temperature programs, and monitored ions are given elsewhere (14, 19, 20). The following compounds were routinely detected in all SPMDs: tri-PCBs 18, 22, 28, and 31; tetra-PCBs 44, 49, 52, 70, and 74; penta-PCBs 87, 90/101, 95, 99, 105, 110, 118, and 123; hexa-PCBs 138, 141, 149, 151, 153/132, and 158; hepta-PCBs 170, 174, 180, 183, and 187; octa-PCBs 194, 199, and 203; HCB; o,p′-DDT; o,p′DDE; p,p′-DDE; p,p′-DDT; PBDEs 28, 47, 49, 99, 100, 153, 154, and 183. QA/QC. Field blanks (SPMDs stored in sealed cans during the deployment period) and laboratory blanks (newly purchased SPMDs) were included at rates of 50% and 15%, respectively. The field blanks were used to correct the results for blank levels and to calculate the limit of detection (LOD). LOD was calculated as the mean plus 3 times the standard deviation of the blank, and data below the limit of detection

TABLE 3. Data from the 2000-2002 Surveya site 1 PCB 28 710 PCB 52 1 100 PCB 90/101 3 200 PCB 118 1 700 PCB 138 3 300 PCB 153/ 132 4 300 PCB 180 910 ∑7 ICES 15 000 ∑31 PCB 35 000 BDE 28 520 BDE 47 2 300 BDE 49 600 BDE 99 1 900 BDE 100 490 BDE 153 340 BDE 154 200 ∑7PBDE 6 400 HCB 9 800 o,p′-DDT nd o,p′-DDE 230 p,p′-DDE 2 300 p,p′-DDT 3 400

site 2

site 3

site 4

site 5

site 6

site 7

site 8

site 9

Sequestered Amounts of PCBs, PBDEs, and Organochlorine Compounds (pg/SPMD) 620 1200 570 770 800 1200 570 780 1 700 3 200 1 500 1 200 1 700 1 200 850 1 000 3 900 6 300 2 100 2 000 4 400 2 300 2 300 1 900 2 100 560 910 620 1 600 610 670 500 3 300 3 500 1 600 1 200 3 600 1 700 1 500 1 100 4 100 4 500 2 100 1 600 4 600 2 500 2 200 1 400 860 490 350 370 870 750 580 200 17 000 20 000 9 100 7 700 18 000 10 000 8 800 6 900 39 000 48 000 21 000 20 000 42 000 25 000 22 000 17 000 1 200 1 500 570 110 55 42 nd 150 1 900 2 600 880 720 1 400 690 730 840 540 1 100 290 180 220 360 220 81 1 600 660 1 500 1 100 1 300 660 660 660 290 310 250 220 240 140 140 110 400 nd 320 300 550 nd 430 nd 190 110 180 200 260 160 190 80 6 100 6 200 4 000 2 800 4 000 2 100 2 400 1 900 9 100 3 100 11 000 14 000 11 000 11 000 9 700 12 000 nd 2 600 nd 1 700 5 600 1 800 2 100 2 100 1 900 180 590 360 660 280 330 270 3 900 1 800 1 900 3 400 8 700 760 1 700 530 7 000 2 500 4 200 1 000 4 700 900 1 200 840

site 10 630 1 100 2 000 690 1 100 1 300 250 7 000 18 000 74 740 130 1 700 280 270 190 3 300 10 000 2 000 470 1600 510

site 11 800 960 1 500 520 730 900 120 5 600 14 000 nd 840 91 1 300 200 390 210 3 000 18 000 1 000 200 880 540

Estimated Concentrations of Penta-BDEs and Selected Organochlorine Compounds (pg m-3) and the p,p-DDE/p,p-DDT Ratio PBDE 28 0.2 0.5 0.6 0.2 0.04 0.02 0.02 nd 0.1 0.03 PBDE 47 0.9 0.8 1.0 0.3 0.3 0.6 0.3 0.3 0.3 0.3 PBDE 49 0.2 0.2 0.4 0.1 0.1 0.1 0.1 0.1 0.03 0.1 PBDE 99 0.8 0.6 0.3 0.6 0.4 0.5 0.3 0.3 0.3 0.7 PBDE 100 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.04 0.1 PBDE 153 0.1 0.2 nd 0.1 0.1 0.2 nd 0.2 nd 0.1 PBDE 154 0.1 0.1 0.04 0.1 0.1 0.1 0.1 0.1 0.03 0.1 ∑7PBDE 2.5 2.4 2.5 1.6 1.1 1.6 0.8 0.9 0.8 1.3 HCB 65 61 20 72 93 72 71 65 77 67 o,p′-DDT nd nd 17.3 nd 11.2 37 12 14 14 13 o,p′-DDE 1.6 12 1.2 3.9 2.4 4.4 1.9 2.2 1.8 3.1 p,p′-DDE 16 26 12 13 22 58 5.1 11 3.6 10 p,p′-DDT 22 47 17 28 6.7 31 6.0 8.0 5.6 3.4 p,p′-DDE/ 0.7 0.6 0.7 0.5 3.3 1.8 0.8 1.4 0.6 3.0 p,p′-DDT a

nd 0.3 0.04 0.5 0.1 0.2 0.1 1.2 120 6.9 1.3 5.9 3.6 1.6

nd ) not detected; bold numbers represent one-half LOD.

VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2525

FIGURE 2. Sequestered amounts of PBDE congeners (pg/SPMD). were rejected. The laboratory blanks were merely included as a method check. In cases where the concentration was found to be below the detection limit, the value of (1/2)LOD was inserted. These values and nondetect values are shown in bold in Tables 3 and SI-1 (Supporting Information). Recoveries were routinely monitored in all samples using the [13C12]PCB spike and PCBs 54 and 155 as described above. Recoveries were very good for all compounds, averaging 80% (range 65-90%) for the interior fraction and 78% (range 6886%) for the exterior fraction. Method recoveries for PBDEs 2526

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004

were carried out, by extracting and cleaning four SPMDs spiked with 5000 pg of a working standard containing all analyzed PBDE congeners. Very good recoveries were obtained, averaging 72% (range 55-80%) for the interior fraction. After blank correction, the results for F1 and F2 of the interior fraction were added together, and for each site the mean sequestered amount of the two SPMDs was calculated for each congener. There was a very good reproducibility between SPMD duplicates at all sites, with variability ranging from 10% to 20%. Selected PCB congeners,

FIGURE 3. Comparison of sequestered amounts (pg/SPMD) of ICES congeners among the 1994-1996 and 1998-2000 studies and this study (2000-2002). PBDEs, and organochlorine compounds are reported in this paper. A table with detailed results at all sites for all PCB congeners is available in the Supporting Information (Table SI-1).

Results and Discussion Permeation Reference Compound (PRC) Data. The losses of 13C12-labeled PCBs 28, 52, 101, 138, 153, and 180 varied from congener to congener and from site to site with the degree of chlorination of the compounds (see Table 2). PCBs 28 and 52 were virtually completely lost during the deployment period (ca. 720 days). This was expected; 50-85% of [13C12]PCBs 28 and 52 were lost over 120 days in a previous investigation (15). Smaller quantities of the higher molecular weight PRCs were lost during the deployment (see Table 2). For example, 18-44% of PCB 153 had depurated. Wind speed

is a potentially important variable in passive air sampling studies, because it will affect the atmospheric boundary layer (ABL) thickness, and hence air-sampler exchange rates, if air-side resistance controls the uptake rate (21, 22). However, site differences were minimized, because the Stevenson screens exert a buffering effect to moderate the wind speed (6, 15). Site 3 is believed to be the most exposed (windiest) of the five U.K. sites, but PRC depuration was no higher at site 3 than at other locations. These data therefore do not resolve whether the addition of PRCs is helpful in correcting for possible site differences in uptake rates, although they do provide analytical QA/QC reassurance. Congener Composition and Spatial Trends of PBDEs. Table 3 presents a summary of the PBDEs sequestered in the samples (pg/SPMD) and the estimated air concentrations (pg m-3) calculated from these amounts, assuming a sampling VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2527

TABLE 4. Atmospheric Clearance Rates Derived for PCBs between 1994 and 2002 (years)a site

PCB 52

PCB 90/101

PCB 118

PCB 138

PCB 153/132

PCB 180

range

1 2 3 4 5 6 7 8 9 11 range

2.9 3.0 2.5 3.2 3.0 2.8 3.8 2.7 2.5 2.9 2.5-3.8

3.3 3.6 3.2 3.2 3.5 4.6 4.4 5.4 4.1 4.9 3.2-5.4

3.3 3.6 1.5 3.3 3.5 5.2 4.5 3.3 4.4 5.9 1.5-5.9

2.7 2.7 2.0 3.5 2.9 3.7 na na 3.2 3.4 2.0-3.7

3.0 3.0 2.2 2.8 3.0 4.3 na na 3.5 3.6 2.2-4.3

3.1 2.9 1.7 2.8 3.9 4.9 na na 3.9 3.4 1.7-4.9

2.7-3.3 2.7-3.6 1.5-3.2 2.8-3.5 2.9-3.9 2.8-5.2 3.8-4.5 2.7-5.4 2.5-4.4 2.9-5.9

a na ) half-lives for PCBs 138, 153/132, and 180 at sites 7 and 8 not calculated because these PCBs were outliers in the 1998-2000 campaign. Half-lives were not calculated for site 10 because no data were available from the 1994-1996 campaign.

FIGURE 4. Sequestered amounts of some organochlorine compounds from the study: (a) comparison of sequestered amounts (pg/SPMD) of HCB between the 1998-2000 and 2000-2002 studies; (b) sequestered amounts of o,p′-DDE and p,p′-DDE (pg/SPMD) vs latitude. rate of 3.5 m3/day (ranging from 3 to 4 m3/day), derived from measured uptake rates (10, 14). There is thus an uncertainty of about 10-20% associated with the sampling rate. So the estimated air concentrations should be treated with caution. SPMDs have long equilibration times (months/years), and the time to reach gas-phase-sampler equilibrium varies widely between POPs. Lighter POPs with lower KOA values reach equilibrium faster. Given their KOA values, PBDE uptake was assumed to still be in the linear portion of the uptake curve (23). BDEs 47 and 99 were the most abundant congeners, although BDEs 28, 49, 100, 153. and 154 were all routinely or regularly detected (Table 3a). These are dominant congeners in the penta-BDE mixture (24). The derived PBDE air concentrations reported in Table 3b averaged 2.0 (range 1.1-2.5) and 1.1 (range 0.8-1.6) pg m-3 for the U.K. and Norway sites, respectively. These are in line with values measured at Mace Head, on the remote west coast of Ireland (0.2-5 pg m-3 in a series of short-term samples) by Lee et al. (25) and by Jaward et al. (12) for other European background locations. PBDE Latitudinal Distribution and Fractionation. The latitudinal distribution of POPs in air is a function of the proximity to sources and LRAT potential; the latter depends on compound reactivity and net depositional behavior. The U.K. has been a major producer and user of PBDEs, and U.K. urban centers gave the highest values in the European passive sampling campaign (12). However, the sites in this study are all selected to be background locations. Site 3 is the only one close to a substantial area of populationsthe Lancaster city center is about 5 km awaysalthough most of the surrounding area is rural. Figure 2 shows the absolute sequestered amounts of PBDE congeners at all sites. This highlights differences between 2528

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004

congeners. The higher molecular weight congeners (BDEs 153 and 154) are quite evenly distributed and show no significant trend with latitude (p > 0.05). In contrast, the amounts of the lighter PBDEs, BDEs 28, 47, 49, and 100, decreased with latitude (p < 0.05 for all congeners). However, at sites 10 and 11, BDE 99 increased. We do not have an explanation for this. It is unlikely that this is caused by a laboratory contamination artifact, as laboratory contamination is usually more apparent for more volatile congeners (26); BDE 47, the other major penta-product, did not show this high increase. If the data for these two sites are treated as outliers, then BDE 99 also decreases with latitude (p < 0.05). There is thus a broad change in the ∑PBDE composition with latitude, as highlighted in Figure 2, a “fractionation”, but with the sequestered amounts of the low molecular weight congeners decreasing with latitude. This is opposite to what has been observed for PCBs (3, 6), where the sequestered amounts of heavier PCBs generally decrease with increasing latitude, which is often related to the distance from the source, and the lighter PCBs are uniformly distributed. In a modeling study, Wania and Dugani (27) suggested that lower brominated congeners have an LRAT potential comparable to that of PCBs, whereas the highly brominated congeners were predicted to have a low potential to reach remote areas. PBDE congeners are heavier than their PCB analogues. However, the C-Br bond in PBDEs is weaker than the C-Cl bond in PCBs. For example, tri-BDEs and hepta-CBs have a molar mass of approximately 400 g/mol, yet the tri-BDEs are predicted to be more volatile and more reactive by over an order of magnitude in air than the hepta-CBs (27). The LRAT potential of lower halogenated congeners is limited by their efficient degradation in the atmosphere, whereas the LRAT potential of the highly halogenated congeners is limited by their removal from the atmosphere, primarily by particle-

bound deposition and low revolatilization from the surface media (27). Congeners of intermediate halogenation appear persistent and volatile enough to remain airborne long enough to undergo greater LRAT. Observations for BDE 183. PBDE 183, a marker for the octa-BDE mixture, was also quantified. Although at present very little data exists on its occurrence in air, relatively high amounts were measured in this study, ranging from 290 to 2400 pg/SPMD (1.9-16 pg m-3). These data demonstrate that components of the octa-BDE formulation, which is principally used in moulded parts of televisions, computers, car parts, etc., are entering the environment (28). Since the source types are different from those of the penta-BDE mixture (principally used in polyurethane foams in textiles/ furnishings) (28), it was not considered with the other congeners in the previous discussions. No defined latitudinal trend was observed which could be related to usage/source rather than physicochemical properties. Observations for the PCB Data. Figure 3 shows a comparison of absolute sequestered amounts of the major ICES PCB congeners (PCBs 52, 90/101, 118, 153, 138, and 180) among the three sampling periods (1994-1996, 1998-2000, and 2000-2002) at all sites. A comparison of PCB 28 is only possible between the 1998-2000 and 2000-2002 periods; no data are available for the 1994-1996 period (6). In all cases, the absolute sequestered amounts of the lighter PCBs are similar, while the heavier PCBs decrease with increasing latitude although uptake of the lighter PCBs (e.g., PCB 28) during the 24 month sampling period is expected to have left linearity and their distribution between the SPMD and air approached/reached equilibrium (6, 15). Some hexa-, hepta-, and octa-CBs at sites 7 and 8 (3) were outliers in the 19982000 campaign, but this was not seen in the present study. Global fractionation of PCBs continued to be observed in this study, as it had been in the previous ones (3, 6). Derivation of Atmospheric Clearance Rates for PCBs. PCB congener data are now available for the sample transect for three sampling intervals, 1994-1996, 1998-2000, and 2000-2002. This enables preliminary estimates of the atmospheric clearance rates to be determined. It was assumed that PCB air concentrations have been subject to a firstorder decline over time; this is consistent with more detailed, site-specific trend data noted in the Great Lakes regions of North America, the U.K., and the Baltic (29-32). Data from this exercise are presented in Table 4; clearance rates ranged between 1.5 and 5.9 years, averaging ca. 3.5 (with a standard deviation of 0.6) years for all congeners/locations. No systematic differences in congeners or locations were noted. This supports the hypothesis that the underlying trends in European background air are still controlled by primary, rather than secondary, sources (3, 32). Organochlorine Compounds. The HCB data are plotted in Figure 4a. As in 1998-2000 (3), the 2000-2002 data show a very high degree of atmospheric mixing for this compound, and higher sequestered amounts in the northernmost locations (i.e., evidence of cold condensation). There was little difference between the two sampling intervals, indicative of the persistence of this compound and/or ongoing atmospheric emissions. Data for the DDT group are given in Table 3a and summarized in Figure 4b. In contrast to the HCB trends, there is a slight decrease in absolute amounts with increasing latitude for o,p′- DDE and p,p′- DDE. This finding is consistent with studies on vegetation by Calamari et al. and Helstrom (4, 33). Indeed, their distribution is similar to that of the heavier PCBs and PBDEs. This probably reflects their lower volatility, higher particulate burden, and therefore lower LRAT potential. Estimated air concentrations for HCB, p,p′-DDE, and p,p′- DDT (see Table 3b) compare very well with recent results from the European passive air sampling network (12)

and with those of Booij et al. (16) at a near-shore location in The Netherlands. The p,p′-DDE/p,p′-DDT ratios calculated (Table 3b) were generally about unity for most sampling sites (except sites 5, 6, 10, and 11), suggesting that there is some fresh p,p′-DDT reaching these sites.

Acknowledgments We are grateful to the site owners for their kind cooperation. F.M.J. is grateful to the Commonwealth Commission for Ph.D. funding.

Supporting Information Available Table showing sequestered amounts of PCBs, PDBEs, and organochlorine compounds for all sites. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) UNECE. Protocol on Persistent Organic Pollutants under the 1979 Convention on Long-Range Transboundary Air Pollution; United Nations Economic Commission for Europe (ECE/EB.Air/ 60): Geneva, Switzerland, 1998. (2) Klecka, G.; et al. Evaluation of persistence and long-range transport of organic chemicals in the environment; SETAC Special Publication Series; SETAC: Pensacola, FL, 2000. (3) Meijer, S. N.; Ockenden, W. A.; Steinnes, E.; Corrigan, B. P.; Jones, K. C. Environ. Sci. Technol. 2003, 37, 454-461. (4) Calamari, D.; Tremolada, P.; Di Guardo, A.; Vighi, M. Environ. Sci. Technol. 1994, 28, 429-434. (5) Kalantzi, O. I.; Alcock, R. E.; Johnston, P. A.; Santillo, D.; Stringer, R. L.; Thomas, G. O.; Jones, K. C. Environ. Sci. Technol. 2001, 35, 1013-1018. (6) Ockenden, W. A.; Sweetman, A.; Prest, H. F.; Steinnes, E.; Jones, K. C. Environ. Sci. Technol. 1998, 32, 2795-2803. (7) Tremolada, P.; Burnett, V.; Calamari, D.; Jones, K. C. Environ. Sci. Technol. 1996, 30, 3570-3577. (8) Jones, K. C.; Barber, J. L. Sample matrices, site selection and sampling techniques, Proceedings of the UNEP Workshop to Develop a Global POPs Monitoring Programme to Support the Effectiveness Evaluation of the Stockholm Convention; UNEP: Geneva, Switzerland, 2003; pp 185-199. (9) Lohmann, R.; Corrigan, B.; Howsam, M.; Jones, K. C.; Ockenden, W. Environ. Sci. Technol. 2001, 35, 2576-2582. (10) Shoeib, M.; Harner, T. Environ. Sci. Technol. 2002, 36, 41424151. (11) Harner, T.; Ikonomou, M.; Shoeib, M.; Stern, G.; Diamond, M. Organohalogen Compd. 2002, 57, 33-36. (12) Jaward, F. M.; Farrar, N. J.; Harner, T.; Sweetman, A. J.; Jones, K. C. Environ. Sci. Technol. 2004, 38, 34-41. (13) Jaward, F. M.; Farrar, N. J.; Harner, T.; Sweetman, A. J.; Jones, K. C. Environ. Toxicol. Chem., in press. (14) Ockenden, W. A.; Prest, H. F.; Thomas, G. O.; Sweetman, A.; Jones, K. C. Environ. Sci. Technol. 1998, 32, 1538-1543. (15) Ockenden, W. A.; Corrigan, B.; Howsam, M.; Jones, K. C. Environ. Sci. Technol. 2001, 35, 4536-4543. (16) Booij, K.; van Drooge, B. L. Chemosphere 2001, 44, 91-98. (17) Wennrich, L.; Popp, P.; Hafner, C. J. Environ. Monit. 2002, 4, 371-376. (18) Huckins, J. N.; Petty, J. D.; Orazio, C. E.; Lebo, J. A.; Clark, R. C.; Haverland, P. S. A Laboratory Study to Demonstrate the Feasibility of the Use of SPMD Permeability Reference Compounds (PRCs) to Correct for the Effects of Fouling on the Uptake of PAHs; A Report to the American Petroleum Institute; U.S. Geological Survey, Columbia Environmental Research Center: Washington, DC, 1994. (19) Thomas, G. O.; Sweetman, A. J.; Parker, C. A.; Kreibich, H.; Jones, K. C. Chemosphere 1998, 36, 2447-2459. (20) Gouin, T.; Thomas, G. O.; Cousins, I.; Barber, J.; Mackay, D.; Jones, K. C. Environ. Sci. Technol. 2002, 36, 1426-1434. (21) Ockenden, W. A.; Jaward, F. M.; Jones, K. C. Sci. World 2001, 1, 557-575. (22) Harner, T.; Farrar, N. J.; Shoeib, M.; Jones, K. C.; Gobas, F. A. P. C. Environ. Sci. Technol. 2003, 37, 2486-2493. (23) Harner, T.; Shoeib, M. J. Chem. Eng. Data 2002, 47, 228-232. (24) Sjodin, A.; Jakobsson, E.; Kierkegaard, A.; Marsh, G.; Sellstrom, U. J. Chromatogr., A 1998, 822, 83-89. VOL. 38, NO. 9, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2529

(25) Lee, R. G. M.; Thomas, G. O.; Jones, K. C. Environ. Sci. Technol. 2004, 38, 699-706. (26) Alcock, R. E.; Halsall, C. J.; Harris, C. A.; Johnston, A. E.; Lead, W. A.; Sanders, G.; Jones, K. C. Environ. Sci. Technol. 1994, 28, 1838-1842. (27) Wania, F.; Dugani, C. B. Environ. Toxicol. Chem. 2003, 22, 12521261. (28) Rahman, F.; Langford, K. H.; Scrimshaw, M. D.; Lester, J. N. Sci. Total Environ. 2001, 275, 1-17. (29) Hillery, B. R.; Basu, I.; Sweet, C. W.; Hites, R. A. Environ. Sci. Technol. 1997, 31, 1811-1816. (30) Simcik, M. F.; Basu, I.; Sweet, C. W.; Hites, R. A. Environ. Sci. Technol. 1999, 33, 1991-1995.

2530

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004

(31) Sweetman, A. J.; Jones, K. C. Environ. Sci. Technol. 2000, 34, 863-869. (32) Bignert, A.; Olsson, M.; Persson, W.; Jensen, S.; Zakrisson, S.; Litzen, K.; Eriksson, U.; Haggberg, L.; Alsberg, T. Environ. Pollut. 1998, 99, 177-198. (33) Hellstrom, A. Ph.D. Dissertation, Swedish University of Agricultural Sciences, Uppsala, Sweden, 2003.

Received for review November 20, 2003. Revised manuscript received February 5, 2004. Accepted February 10, 2004. ES035292T