Global pilot study for persistent organic pollutants (POPs) using PUF disk passive air samplers

Environmental Pollution 144 (2006) 445e452 www.elsevier.com/locate/envpol

Global pilot study for persistent organic pollutants (POPs) using PUF disk passive air samplers Tom Harner a,*, Karla Pozo a, Todd Gouin b, Anne-Marie Macdonald a, Hayley Hung a, Jill Cainey c, Andrew Peters d a

b

Science & Technology Branch, Environment Canada, 4905 Dufferin Street, Toronto, ON, M3H 5T4, Canada Canadian Environmental Modelling Centre, Trent University, Peterborough, ON, Canada; Burlington, ON, Canada c Cape Grim Baseline Air Pollution Station, Tasmania 7330, Australia d Bermuda Biological Station for Research, Bermuda Received 22 September 2005; accepted 20 December 2005

Seasonal sampling of ambient POPs at global background sites is logistically feasible and highlights spatial difference in compound distribution. Abstract Polyurethane foam (PUF) disks were deployed at global background sites, to test logistical issues associated with a global monitoring network for persistent organic pollutants (POPs). a-HCH, exhibited relatively high and uniform concentrations (17e150 pg/m3) at temperate and arctic sites with elevated concentrations associated with trans-Pacific inflow. Concentrations were much lower (<5 pg/m3) in Bermuda, Chile and Cape Grim. Concentrations for g-HCH, the main component of lindane, were spatially similar to the a-HCH pattern but lower in magnitude (typically, <10 pg/m3). Chlordane concentrations (sum of cis-chlordane, trans-chlordane and trans-nonachlor) were also low (<10 pg/m3). Dieldrin concentrations were in the range 2e25 pg/m3 at most sites but elevated in Bermuda. Back trajectories suggest that advection from Africa and the US may contribute. Endosulfan, a popular current-use pesticide, exhibited highest concentrations ranging from tens to hundreds of pg/m3. There was good agreement between duplicate samplers at each site and PUF disk-derived air concentrations agreed with high volume data. Few logistical/analytical problems were encountered in this pilot study. Crown Copyright Ó 2006 Published by Elsevier Ltd. All rights reserved. Keywords: Passive air samplers; POPs; Persistent organic pollutants; Spatial trends

1. Introduction On May 17, 2004, the Stockholm Convention on persistent organic pollutants (POPs) was ratified (Stockholm Convention on Persistent Organic Pollutants, 2005, http://www.pops.int/). This document which is coordinated through the United Nations Environment Program (UNEP) is intended to reduce or eliminate the use, discharges and emissions of POPs. Initially, * Corresponding author. ARQM Atmospheric Environment Service, Environment Cananda, 4905 Dufferin Street, Downsview, Ontario M3H 5T4, Canada. Tel.: þ1 416 739 4837; fax: þ1 416 739 5708. E-mail address: [email protected] (T. Harner).

a set of 12 chemicals were identified as priority POPs e they include: nine pesticide classes (aldrin, chlordane, dieldrin, endrin, heptachlor, hexachlorobenzene, mirex, toxaphene, and DDT (dichlorodiphenyltrichloroethane), one industrial chemical class (PCBs, polychlorinated biphenyls) and PCDD/Fs (polychlorinateddibenzodioxins/furans) that are associated with various industrial/combustion emissions. Article 16 of the Stockholm Convention deals with its ‘‘effectiveness evaluation’’. The intent is that after 4 years of entry into force, the effectiveness of the Convention will be assessed through a global monitoring program. Regional and global environmental transport of POPs will also be addressed. To assist member countries with this task, UNEP Chemicals published

0269-7491/$ - see front matter Crown Copyright Ó 2006 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2005.12.053

446

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

a guidance document for the global monitoring of POPs (Guidance for a global monitoring programme for persistent organic pollutants). In this document, four environmental media are recommended for investigating POP levels in the environment: air, bivalves (for monitoring water), other biota and human milk. The recommended strategy for air combines conventional high volume air samplers with passive air samplers (PAS). Conventional high volume samplers have been used in several international, long-term monitoring programs (e.g. Northern Contaminants program in Canada, European Monitoring and Assessment Program, and the CanadaeUS, Integrated Atmospheric Deposition Network (IADN), to name a few) which provide useful data for assessing temporal trends of POPs. However, because of the high financial costs and logistical requirements (e.g. need for electricity, trained operators) of running such networks, few examples exist on a global scale as evidenced by the scarcity of air concentration data for POPs in most parts of the world. PAS can help to fill this data gap. PAS are small, relatively inexpensive, simple to deploy and do not require electricity. Their use has already been demonstrated in several spatial studies at local, regional, and continental scales (Harner et al., 2004; Pozo et al., 2004; Jaward et al., 2004; Shen et al., 2004). Further, because PAS data continuously integrate the air burden of POPs, they have an added benefit and may complement the sometimes intermittent information provided by high volume air sampling networks. Several passive air sampler types have been recently developed and applied to air monitoring of POPs. These include SPMDs (semipermeable membrane devices) (Ockenden et al., 2001 and references within), polyurethane foam (PUF) disks (Shoeib and Harner, 2002; Pozo et al., 2004; Jaward et al., 2004; Harner et al., 2004; Motelay-Massei et al., in press), samplers employing XAD-resin (Wania et al., 2003; Shen et al., 2004), and polymer-Coated Glass (POGs) (Harner et al., 2003; Farrar et al., 2005). This paper reports PUF disk-derived air concentrations for organochlorine pesticides (OCPs) from a pilot study where duplicate PUF disks were deployed at several global remote/ background sites for periods of 2e7 months. The purpose of the study was to test logistical issues and other practical requirements for conducting a global sampling campaign. This was in preparation for a larger scale deployment at over 50 sites under the Global Atmospheric Passive Sampling (GAPS) Study which is now in progress (Pozo et al., 2005). Results from this pilot study also address scientific questions concerning the detection of target compounds and reproducibility of the samplers. Lastly, seasonality in air concentrations of OCPs was also evaluated at stations where samplers were deployed over consecutive periods. 2. Material and methods 2.1. PUF disks samplers These samplers consist of a foam disk (14 cm diameter; 1.35 cm thick; surface area, 365 cm2; mass, 4.40 g; volume, 207 cm3; density, 0.0213 g cm
3;

PacWill Environmental, Stoney Creek, ON) positioned in a stainless steel sampling chamber consisting of two domes (Fig. 1). This ‘‘flying saucer’’ design protects the foam disks from direct precipitation, sunlight and coarse particle deposition. Air is allowed to flow over the sampling surface through a w2.5 cm gap between the two domes. The uptake of POPs by PUF disks has been previously characterized (Shoeib and Harner, 2002; Pozo et al., 2004) and shown to sample mainly the gas-phase at w3e5 m3/day although chemicals associated with very fine particles may also be collected. Tuduri et al. (in this issue) showed a small wind dependency on sampling rates at low wind speeds (less than 5 m/s) with increasing wind dependency at higher wind speeds. This wind effect was also observed by Pozo et al. (2004) in the field deployment of PUF disks samplers in mountainous regions of Chile. In this study sampling rates determined using depuration compounds (isotopically labelled chemicals that are spiked into the PUF disks prior to deployment to assess mass transfer kinetics) were w4  1.1 m3/day except at one extremely windy mountain site where duplicate samplers showed sampling rates about two times higher.

2.2. Sampler deployment Beginning in 2002, duplicate PUF disk samplers were deployed at several remote sites (see Table 1, Fig. 2). Deployment times varied from 2 to 7 months depending on logistical constraints such as availability of site contacts to retrieve samplers. Deployment periods are summarized in Table 2. Prior to exposure, PUF disks were pre-cleaned by Soxhlet extraction for 24 h using acetone and then for another 24 h using petroleum ether. PUF disks were stored in amber glass jars with Teflon-lined lids. Field blanks were also collected by installing them in the sampler chambers and then immediately removing and storing as a sample. Duplicate samplers were deployed at each site for all deployment periods.

2.3. Analysis Samples were analyzed for 19 OCPs including: a-, b-, g-, d-HCHs, aldrin, heptachlor, heptachlor epoxide, cis-chlordane, trans-chlordane, trans-nonachlor, endosulfan I, endosulfan II, endosulfan sulphate, o,p0 -DDE, p,p0 -DDE, o,p0 -DDD, p,p0 -DDD, o,p0 -DDT, p,p0 -DDT (Ultra Scientific, North Kingstown, RI, USA). Analysis of PUF disk extracts was by gas chromatographyemass spectrometry (GCeMS) on a Hewlett-Packard 6890 GC-5973 MS. OCPs were determined in negative chemical ionization (NCI). Conditions for NCI analysis and selection of target/qualifier ions are described in Pozo et al. (2004).

3. Results and discussion 3.1. Quality assurance quality control Method recoveries previously determined for OCPs were satisfactory (Pozo et al., 2004) and no recoveries were applied to the data. Field blanks were treated as samples and analyzed for the suite of OCPs. Results were below instrument detection limits (Table 2) so no blank correction was necessary. 3.2. Air parcel back trajectories To aid in the interpretation of the data, air parcel back trajectories were calculated using the Canadian Meteorological Centre Trajectory Model (2005). Three-day back trajectories at 100 m above ground level were calculated daily during the deployment periods. Results were compiled into ‘‘spaghetti plots’’ that help to identify the frequency with which air arrived at the sampling site from particular sectors or

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

447

mounting bracket stainless steel dome

PUF disk

air circulation

• • • •

sunlight precipitation wind effects particles

Fig. 1. Schematic of PUF disks passive air sampler e ‘flying saucer’ design.

regions. Spaghetti plots for all sites and deployment periods are provided in the Supplementary Information.

3.3. Air concentrations of OCPs e spatial trends Volumetric air concentrations were derived by dividing the amount of chemical collected on the PUF disks by the product of the deployment period and an average PUF disk sampling rate (excluding 1 outlier) of 4 m3/day previously derived by Pozo et al. (2004). For future studies, depuration compounds (DCs) will be added to the PUF disks prior to deployment to assess site-to-site differences in sampling rates and produce data that is more quantitative in nature (Pozo et al., 2004). Table 2 reports air concentrations (mean of duplicates) for several OCPs during various deployment periods. These are also highlighted in Fig. 2 where mean concentrations (for all deployment periods at each site) for selected OCPs are shown. These are compared against results from recent PUF disk studies at background sites in Chile (Pozo et al., 2004) and at urban sites in Toronto (Harner et al., 2004; Motelay-Massei et al., in press).

3.3.1. Hexachlorocyclohexanes (HCHs) a-HCH, the main component of technical HCH, was once a widely used pesticide that is now banned in most parts of the world. Air concentrations of a-HCH were in the range of <1 pg/m3 to as high as 150 pg/m3. Highest concentrations were observed in Canadian sub-arctic (Little Fox Lake) followed by the Arctic (Alert) and the high altitude, west coast site at Whistler Mountain. These sites all receive inputs from Asia which has been implicated as potential source of a-HCH and other OC pesticides (Bailey et al., 2000; Koziol and Pudykiewicz, 2001; Harner et al., 2005) for North America. However, due to the long duration of the passive air sample, it is not possible to correlate concentrations with transport from a specific source region. Variability in a-HCH levels at Little Fox Lake and Whistler Mountain show a relation with air parcel back trajectories (see Supplement 1B) with elevated concentrations occurring during periods 1 and 2 for Fox Lake and during periods 1, 2 and especially 4 at Whistler when trajectories were longer and exhibited the most trans-Pacific character. Emissions of a-HCH in Asia may be the result of revolatilization from soils contaminated by heavy historical

Table 1 Site information Site

Coordinates

Description

Alert Fox Lake, Yukon Whistler Mountain, BC Bermuda

82  61  50  32 

Cape Grim

40  410 S 144  410 E

Toronto (urban)a Chileb

43  400 N 79  220 W 18e52  S 69e73  W

Remote site in the high Arctic. Remote site, receiving trans-Pacific inputs. Mountain site (2180 m a.s.l., receiving trans-Pacific inputs). Background site in the North Atlantic Ocean with some advective inputs from the southern US/Central America. Remote site in the Tasman Sea, south Indian Basin. Some advective inputs from southeastern Australia. Urban/industrial with influences from regional agriculture. Background/remote mountain sites receiving inputs from the south Pacific ocean.

a b

300 200 110 220

N N N N

62  200 W 135  380 W 123  030 W 64  420 W

Results presented previously in Harner et al., 2004 and Motelay-Massei et al., in press. Results presented in Pozo et al., 2004.

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

448

106 112

25 pg/m3

-HCH

25 pg/m3

-HCH

462

50 3 pg/m

Dieldrin

100 pg/m3

-ENDO

Fig. 2. Mean air concentrations (pg/m3) of selected OCPs at global pilot study sampling sites. Deployment times are summarized in Table 2. a-HCH (hexachlorocyclohexane), g-HCH, dieldrin and endosulfans (sum of Endo1 and Endo2).

usage (Li et al., 2000) and/or from isolated continued use of aHCH in the region (Harner et al., 2005). g-HCH, the main component of lindane, was once a globally used pesticide. In recent years due to concerns regarding its abundance in the Arctic environment, its use has diminished. Spatial trends for g-HCH are similar to those observed for a-HCH although concentrations are, for the most part, smaller (Table 2). The high concentrations of g-HCH for Toronto are associated with regional use for corn production (Garmouma and Poissant, 2004). High volume air sampling for OCPs and other semi-volatile compounds have been conducted at Alert since 1992 as part of the Canadian Northern Contaminants Program (NCP). Reported monthly mean concentrations (pg/m3) for a-HCH and g-HCH are in the range w25e75 and 5e20, respectively (Hung et al., 2005) and agree very well with PUF disk values of 41 and 8, respectively (Table 2). 3.3.2. Chlordanes These are banned pesticides that occur on the UNEP POPs list. Chlordane consists of several components, the most abundant being cis- and trans-chlordane and trans-nonachlor, which are reported here. Air concentrations of chlordanes were low at most sites at a few pg/m3 (Table 2). The value for trans-chlordane at Alert (0.81 pg/m3) agrees well with the range of monthly mean values reported at Alert of 0.2e 1 pg/m3 (Hung et al., 2005). Higher concentrations of tens

of pg/m3 in Toronto are attributed to past urban uses of chlordane. Chlordane was a popular pesticide used on house foundations and home gardens and lawns (Harner et al., 2004). 3.3.3. Dieldrin The spatial trend for dieldrin was almost opposite to that observed for the HCHs with highest concentrations reported in Bermuda at 60e75 pg/m3 over three sampling periods (Fig. 2, Table 2). This may reflect re-emissions from previous regional use and or advective transport from source regions in North America and/or Africa (trans-Atlantic transport). Back trajectories for Bermuda identify the southern to northeastern US as a potential source region during periods 1 and 2 (Table 2) whereas during period 3 trajectories stem from the east (possible trans-Atlantic inputs from Africa). Jantunen et al. (2000), reported dieldrin concentrations in the range 6e 65 pg/m3 in Alabama during a 1-year air sampling study in 1996/1997. Mean dieldrin concentrations reported for several sites operated under the New Jersey Atmospheric Deposition Network (NJADN) were in the range 10e75 pg/m3 (Gioia et al., 2005). During the summer of 1988, Knap and Binkley (1991) measured dieldrin (and other organochlorine pesticides) in the troposphere above the North Atlantic, close to Bermuda, at altitudes between 300 and 10,000 feet a.s.l. The mean concentration of dieldrin was 16  6 pg/m3. Dieldrin showed a slight decrease in concentration away from North America. Previous studies have clearly established the

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

449

Table 2 PUF disk-derived air concentrations at pilot study sites (pg/m3) Sample start date

Mean T (  C)

a-HCH

Sample end date

Duration (d)

Fox Lake #1 July 25, 2002 #2 Sept. 29, 2002 #3 Dec. 8, 2002 Mean

Sept. 29, 2002 Dec. 8, 2002 Jul. 6, 2003

86 70 210

9 2 8

Bermuda #1 Dec. 2, 2003 #2 Mar. 9, 2004 #3 Jun. 10, 2004 Mean

Mar. 9, 2004 Jun. 10, 2004 Oct. 8, 2004

120 100 120

19 16 25

Alert #1 #3 Mean

May 25, 2002 Oct. 2003

Nov. 27, 2002 Mar. 2004

186 180


10
27

Cape Grim #1 Jun. 13, 2003 #2 Sept. 15, 2003 #3 Dec. 12, 2003 #4 Mar. 12, 2004 Mean

Sept. 15, 2003 Dec. 12, 2003 Mar. 12, 2004 Jun. 10, 2004

90 90 90 90

11 12 15 12

1.4 BDL 1.4 1.4 1.4

Whistler Mountain #1 Sept. 12, 2002 #2 Jan. 8, 2003 #3 Jun. 2, 2003 #4 Sept. 17, 2003 Mean

Jan. 8, 2003 June 2, 2003 Sept. 17, 2003 Mar. 22, 2004

118 145 105 180


9
3 5
7

28 17 25 51 30

Toronto (mean for urban sites during Jan. 2002e2003, n ¼ 4)a Chile (mean for remote sites during Dec. 2003eJan. 2004, n ¼ 5)b

137 148 33 106 4.2 2.1 3.1 3.1 40 43 41

60 5

g-HCH 22 23 9 18

TC

CC

TN

1.7 1.9 0.09 1

5.7 5.6 1.6 4

3.5 4.0 1.5 3

0.72 0.10 0.15 0.32

0.9 0.8 1.0 0.91

1.6 1.3 1.8 1.5

1.5 2.3 1.9 1.9

8.0 8.0 8.0

0.5 1.2 0.81

1.7 3.7 2.7

1.4 3.0 2.2

0.50 0.20 0.73 0.18 0.40

0.7 0.6 0.3 1.3 0.70

0.18 BDL 0.13 0.15 0.15

0.79 0.84 0.48 0.69 1

2.3 1.5 1.8 2.9 2.1

1.5 1.0 1.1 1.6 1.3 20 1.6

BDL 0.89 BDL 0.66 0.78 5.7 5.9 5.5 7.5 6.1 79 7

31 2.5

26 5

Dieldrin 7.1 6.2 5.9 6

Endo1

Endo2

70 72 41 61

BDL BDL BDL

54 75 60 63

14 36 16 22

BDL 0.45 BDL 0.45

8 12 10.1

24 81 52

BDL BDL

117 17 41 719 223

29 20 2.6 62 29

1.6 1.7 2.0 3.6 2.2

36 51 56 321 116

BDL BDL 0.35 BDL

38 1.9

104 36

ND 2.3

18 13 8 25 16

Abbreviations: BDL, below detection limit; IDL, instrumental detection limit; HCH, hexachlorocyclohexane; TC, trans-chlordane; CC, cis-chlordane; TN, transnonachlor; endo, endosulfan. The IDL for OCPs present the following values (pg): a-HCH: 0.07; g-HCH: 0.02; p,p0 -DDE: 0.09; CC: 0.03; TC: 0.01; TN: 0.01; endo I: 0.11; endo II: 0.11. Note: aldrin, b-HCH, and d-HCH, were not detected in any of the samples analyzed due to low values and/or interferences. Heptachlor and heptachlor epoxide were detected sporadically but not reported here. a From Motelay-Massei et al., 2005. b From Pozo et al., 2004.

occurrence of atmospheric inputs of organic contaminants, organic carbon and dust to the North Atlantic region from Africa (Panshin and Hites, 1994; Conte and Weber, 2002; Kaufman et al., 2005). Panshin and Hites (1994) observed significantly elevated concentrations of PCBs in the atmosphere of Bermuda during the summer when air parcels transiting Bermuda originated in Africa and the eastern Atlantic. Dieldrin was one of the most abundant OC pesticides detected in the atmosphere near Senga Bay, Lake Malawi, in southern Africa with a mean concentration of 80 pg/m3 and a maximum of 222 pg/m3 (Karlsson et al., 2000). 3.3.4. Endosulfan Endosulfans were the most abundant of the pesticides reported here, reflecting their wide-spread current use around the world (Fig. 2, Table 2). Global use of endosulfan has increased over the past two decades and is now at w13 kt per year (Li and Macdonald, 2005). India is currently the world’s largest consumer of endosulfan which may partly explain the elevated concentrations at Whistler Mountain during period 4 when advective trans-Pacific flow was greater (as discussed

previously for a-HCH). Concentrations at the pilot study sites ranged from tens of pg/m3 to hundreds of pg/m3 with strong seasonality observed at some sites (discussed later). Endosulfans consist of two isomers e Endo1 the more abundant isomer and Endo2 (Table 2). High volume air concentrations for endosulfans at Alert were w4 pg/m3 during the mid 1990s and exhibited an increasing temporal trend over the period 1992e1997 (Hung et al., 2002). The higher PUF disk value of 24e52 pg/m3 reported here for the time period 2002e2004 may reflect this expected increase in air concentrations and/or a disconnect between the high volume and passive data. More recent high volume air measurements are not available for comparison. 3.4. Reproducibility Results for duplicate, co-located PUF disk air samplers are presented graphically in Fig. 3 for the first sampling period at each site (OCP concentrations for sampler 1 are plotted against the same compounds in sampler 2). Agreement between duplicates was good and generally within 10% for all

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

450

20

1:1 40

Bermuda

20

0

0

20

40

Air Conc. (pg/m3)

L. Fox Lake

Air Conc. (pg/m3)

60

15 10 5 0

60

0

80 60

Whistler Mtn.

40 20 0

10

15

20

30

100

0

20

40

60

80

100 120

Air Conc. (pg/m3)

Air Conc. (pg/m3)

120

Cape Grim

5

Air Conc. (pg/m3)

Air Conc. (pg/m3)

20

10

0

Air Conc. (pg/m3)

0

10

20

30

Air Conc. (pg/m3)

Fig. 3. Reproducibility of air concentrations (pg/m3) derived from co-located PUF disk samplers. (OCPs result for sampler 1 are plotted against results for sampler 2 for the first deployment period).

sites and sampling periods. This small difference can be attributed to analytical variability which is expected to be of similar magnitude. The good agreement between co-located samplers adds confidence to the pending results for the expanded global monitoring program e GAPS- where a single sampler is deployed at most sites (Pozo et al., 2005). 3.5. Seasonality By integrating over months, PUF disks samplers are ideal for investigating differences in air concentrations between seasons. Seasonal differences in air concentrations will depend on many factors e the regional usage pattern, physical chemical properties and meteorological factors for instance. For most of the OCPs investigated here seasonal differences were not pronounced. This is largely due to the fact that many of the target OCPs are now banned or severely restricted and hence do not exhibit temporal variability associated with application periods for agricultural purposes. Endosulfan, however, is an exception as this is still a globally popular high volume pesticide. Fig. 4 compares seasonal trends for a-HCH and endosulfan 1 at three sites e Bermuda, Cape Grim and Toronto. Air concentrations for a-HCH show minimal seasonality at each site and are generally within a factor of about two. Lack of strong seasonality of a-HCH is typical as reported in various monitoring programs for POPs (e.g. NCP data (Hung et al., 2002) and measurements conducted under the Integrated Atmospheric Deposition Network, IADN www.msc.ec.gc.ca/iadn/2001).

The relative constancy of a-HCH air concentrations is attributed to its persistence and relatively high volatility in air that allow it to achieve uniform regional/continental distributions. Further, a-HCH is a banned pesticide so there should be no seasonal usage of a-HCH associated with agriculture for instance. SEndosulfans, however, do show strong seasonality especially at Cape Grim and Toronto (Motelay-Massei et al., in press). These sites are expected to be heavily impacted by seasonal use for agriculture which is consistent with the timing of the observed peak concentrations. Furthermore, high endosulfan concentrations at Cape Grim agree well with the results of the air parcel back trajectories. Highest endosulfan concentrations occurred during periods 1 and 4 when back trajectories exhibited the highest occurrence of inputs from the highest density agricultural regions in the southeastern states of New South Wales and Victoria. 3.6. Logistical issues Very few logistical problems were encountered during this pilot study. Samplers were sent to each site along with an instruction sheet explaining how to mount the chamber and insert, replace and store PUF disks. Site operators carried out these tasks with relative ease and few problems were reported. The return shipment of samples to Environment Canada also occurred without any problems. Lastly, the field blanks were very low which lends confidence to the entire sampling protocol and analytical method.

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

-HCH

Endo I 40

Air Conc. (pg/m3)

Air Conc. (pg/m3)

10

Bermuda

8 6 4 2

30 20 10 0

0 Dec.2, 2003

Mar.9, 2004

Dec.2, 2003

June.10, 2004

Air Conc. (pg/m3)

Air Conc. (pg/m3)

4 3 2 1 0

200

Jun.13, Sept.15, Dec.12, Mar.12, 2003 2003 2003 2004 800

Air Conc. (pg/m3)

Air Conc. (pg/m3)

400

0

80 60 40 20 0

June.10, 2004

600

Jun.13, Sept.15, Dec.12, Mar.12, 2003 2003 2003 2004

100

Toronto

Mar.9, 2004

800

5

Cape Grim

451

Autumn

Winter

Spring

600 400 200 0

Autumn

Winter

Spring

Fig. 4. Seasonality of air concentrations for a-HCH and endosulfan 1 at selected sites.

3.7. Implications

Acknowledgements

The results of the PUF disk global pilot study show great promise for the use of these samplers for assessing POPs trends in the atmosphere on a global scale. Such data is useful for several purposes including: identifying emissions regions for POPs; investigating the occurrence and long-range transport of POPs at the local, regional and global scales; testing global transport models for POPs; and perhaps most importantly for investigating temporal changes in air concentrations of POPs. This last aspect is especially relevant in the context of ‘‘effectiveness evaluation’’ of the Stockholm Convention as discussed previously. PUF disk passive air samplers provide a cost-effective and simple tool for addressing this need. As a result of the success of this pilot study, the Global Atmospheric Passive Sampling (GAPS) Study was initiated in December 2004 at more than 50 sites around the world on all seven continents. Under this 1-year study, PUF disk samplers (deployed for 3-month integration periods) and XAD-based samplers (deployed for 1 year) will be analyzed for several target classes including OCPs, polychlorinated biphenyls and other important classes of emerging pollutants (Pozo et al., 2005).

Many thanks to site operators and colleagues who assisted with sample deployment and collection e Juniper Buller and Kurt Anlauf (Whistler), Jochen Mueller, Laurie Porter and Chris Rickard (Cape Grim), Maria Vavro and Sandy Steffen (Alert) and Peter Sedwick (Bermuda). Thanks also to Jacinthe Racine (CMC) for preparing air parcel back trajectory maps.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.envpol.2005.12.053. References Bailey, R., Barrie, L.A., Halsall, C.J., Fellin, P., Muir, D.C.G., 2000. Atmospheric organochlorine pesticides in the western Canadian Arctic: evidence of transpacific transport. J. Geophys. Res. 105, 11805e11811. Canadian Meteorological Centre Trajectory Model. http://iweb.cmc.ec.gc.ca.

452

T. Harner et al. / Environmental Pollution 144 (2006) 445e452

Conte, M.H., Weber, J.C., 2002. Long-range transport of terrestrial biomarkers to the western North Atlantic. Global Biogeochem. Cycles 16, 1142. Farrar, N.J., Harner, T., Shoeib, M., Sweetman, A.J., Jones, K.C., 2005. Field deployment of thin film passive air samplers for persistent organic pollutants: a study in the urban atmospheric boundary layer. Environ. Sci. Technol. 39, 42e48. Garmouma, M., Poissant, L., 2004. Occurrence, temperature and seasonal trends of a and g-HCH in air (Que´bec, Canada). Atmos. Environ. 38, 369e382. Gioia, R., Offenberg, J.H., Gigliotti, C.L., Totten, L.A., Du, Songyan, Eisenreich, S.J., 2005. Atmospheric concentrations and deposition of organochlorine pesticides in the US mid-Atlantic region. Atmos. Environ. 39, 2309e2322. Guidance for a Global Monitoring Programme for Persistent Organic Pollutants. http://www.chem.unep.ch/gmn/GuidanceGPM.pdf. Harner, T., Farrar, N.J., Shoeib, M., Jones, K.C., Gobas, F.A.P.C., 2003. Characterization of polymer-coated glass as a passive air sampler for persistent organic pollutants. Environ. Sci. Technol. 37, 2486e2493. Harner, T., Shoeib, M., Diamond, M.L., Stern, G.A., Rosenberg, B., 2004. Using passive air samplers to assess urban-rural trends for persistent organic pollutants. 1. Polychlorinated biphenyls and organochlorine pesticides. Environ. Sci. Technol. 38, 4474e4483. Harner, T., Shoeib, M., Kozma, M., Gobas, F.A.P.C., Li, S.M., 2005. Hexachlorocyclohexanes and endosulfans in urban, rural, and high altitude air samples in the Fraser Valley, British Columbia: evidence for trans-Pacific transport. Environ. Sci. Technol. 39, 724e731. Hung, H., Halsall, C.J., Blanchard, P., Li, H.H., Fellin, P., Stern, G., Rosenberg, B., 2002. Temporal trends of organochlorine pesticides in the Canadian arctic atmosphere. Environ. Sci. Technol. 36, 862e868. Hung, H., Blanchard, P., Halsall, C.J., Bidleman, T.F., Stern, G.A., Fellin, P., Muir, D.C.G., Barrie, L.A., Jantunen, L.M., Helm, P.A., Ma, J., Konoplev, A., 2005. Temporal and spatial variabilities of atmospheric polychlorinated biphenyls (PCBs), organochlorine (OC) pesticides and polycyclic aromatic hydrocarbons (PAHs) in the Canadian Arctic: results from a decade of monitoring. Sci. Total Environ. 342, 119e144. Integrated Atmospheric Deposition Network. http://www.msc.ec.gc.ca/iadn/ 2001. Jantunen, L.M., Bidelman, T.F., Harner, T., Parkhurst, W.J., 2000. Toxaphene, chlordane and other organochlorine pesticides in Alabama air. Environ. Sci. Technol. 34, 5097e5105. Jaward, F., Farrar, N.J., Harner, T., Sweetman, A., Jones, K.C., 2004. Passive air sampling of PCBs, PBDEs, and organochlorine pesticides across Europe. Environ. Sci. Technol. 38, 34e41. Karlsson, H., Muir, D.C.G., Teixiera, C.F., Burniston, D.A., Strachan, W.M.J., Hecky, R.E., Mwita, J., Bootsma, H.A., Grift, N.P., Kidd, K.A., Rosenburg, B., 2000. Persistent chlorinated pesticides in air, water and precipitation from the Lake Malawi area, southern Africa. Environ. Sci. Technol. 34, 4490e4495.

Kaufman, Y.J., Koren, L., Remer, L.A., Tanre´, D., Ginoux, P., Fan, S., 2005. Dust transport and deposition observed from the Terra-Moderate Resolution Imaging Spectroradiometer (MODIS) spacecraft over the Atlantic Ocean. J. Geophys. Res. 110 (D10), D10S12. Knap, A.H., Binkley, K.S., 1991. Chlorinated organic-compounds in the troposphere over the western North Atlantic Ocean measured by aircraft. Atmos. Environ. 25, 1507e1516. Koziol, A.S., Pudykiewicz, J.A., 2001. Global-scale environmental transport of persistent organic pollutants. Chemosphere 45, 1181e1200. Li, Y.F., Scholtz, M.T., Heystvan, B.J., 2000. Global gridded emission inventories of a-hexachlorocyclohexane. J. Geophys. Res. 105, 6621e 6632. Li, Y.F., Macdonald, R.W., 2005. Sources and pathways of selected organochlorine pesticides to the Arctic and the effect of pathway divergence on HCH trends in biota: a review. Sci. Total Environ. 342, 87e106. Motelay-Massei, A., Harner, T., Shoeib, M., Diamond, M.L., Stern, G.A., Rosenberg, B., 2005. 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. 39, 5763e5773. Ockenden, W.A., Corrigan, B., Howsam, M., Jones, K.C., 2001. Further developments in the use of semipermeable membrane devices as passive air samplers: application to PCBs. Environ. Sci. Technol. 35, 4536e4543. Panshin, S.Y., Hites, R.A., 1994. Atmospheric concentrations of polychlorinated biphenyls at Bermuda. Environ. Sci. Technol. 28, 2001e2007. Pozo, K., Harner, T., Shoeib, M., Urrutia, R., Barra, R., Parra, O., Focardi, S., 2004. Passive-sampler derived air concentrations of persistent organic pollutants on a northesouth transect in Chile. Environ. Sci. Technol. 38, 6529e6537. Pozo, K., Harner, T., Wania, F., Muir, D.C.G., Barrie, L.A., Jones, K.C., 2005. Global atmospheric passive sampling (GAPS) study. Organohalogen Compd. 1323e1326. Shen, L., Wania, F., Lei, Y.D., Teixeira, C., Muir, D.C.G., Bidleman, T.F., 2004. Hexachlorocyclohexanes in the North American atmosphere. Environ. Sci. Technol. 38, 965e975. Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air samplers for persistent organic pollutants. Environ. Sci. Technol. 36, 4142e4151. Stockholm Convention on Persistent Organic Pollutants. http://www.pops.int/. Tuduri, L., Harner, T., Hung, H. Polyurethane foam (PUF) disk passive air samplers: wind effect on sampling rates. Environ. Pollut., in this issue. Wania, F., Shen, L., Lei, Y.D., Teixeira, C., Muir, D.C.G., 2003. Development and calibration of a resin-based passive sampling system for monitoring persistent organic pollutants in the atmosphere. Environ. Sci. Technol. 37, 1352e1359.