Environ. Sci. Technol. 2009, 43, 5633–5639
Organochlorine Pesticides and PAHs in the Surface Water and Atmosphere of the North Atlantic and Arctic Ocean R A I N E R L O H M A N N , * ,†,‡ ROSALINDA GIOIA,§ KEVIN C. JONES,§ L U C A N I Z Z E T T O , §,| C H R I S T I A N T E M M E , ⊥,# Z H I Y O N G X I E , ⊥ DETLEF SCHULZ-BULL,∇ INES HAND,∇ E R I C M O R G A N , †,O A N D L I I S A J A N T U N E N [ Graduate School of Oceanography, University of Rhode Island, Narragansett, Rhode Island 02882, Centre for Chemicals Management Lancaster Environment Centre, Lancaster University, Lancaster, LA1 4YQ, U.K., Department of Structural and Functional Biology, University of Insubria, 3, Via Dunant, 21100 Varese, Italy, GKSS Research Centre Geesthacht, Institute for Coastal Research, Max-Planck Str. 1, D-21502 Geesthacht, Germany, Institute for Baltic Sea Research (IOW), Seestr. 15, Warnemu ¨ nde, Rostock, Germany, and Environment Canada, 6248 Eighth Line, Essa Township, R.R. 1, Egbert, Ontario L0L 1N0, Canada
Received April 24, 2009. Revised manuscript received June 16, 2009. Accepted June 18, 2009.
Surface seawater and boundary layer atmospheric samples were collected on the FS Polarstern during cruise ARKXX in the North Atlantic and Arctic Ocean in 2004. Samples were analyzed for persistent organic pollutants (POPs), with a focus onorganochlorinepesticides,includinghexachlorocyclohexanes (HCHs), chlordanes, DDTs, hexachlorobenzene (HCB), and polycyclic aromatic hydrocarbons. In addition, the enantiomer fractions (EFs) of pesticides, notably R-HCH and cischlordane (CC), were determined. Concentrations of dissolved HCB increased from near Europe (∼ 1-2 pg/L) toward the high Arctic (4-10 pg/L). For dissolved HCB, strongest correlations were obtained with the average air or water temperature during sampling, not latitude. In the western Arctic Ocean, surface waters with elevated concentrations of HCB (5-10 pg/ L) were flowing out of the Arctic Ocean as part of the East Greenland current. In contrast to dissolved compounds, atmospheric POPs did not display trends with temperature. Air-water exchange gradients suggested net deposition for all compounds, though HCB was closest to air-water equilibrium. EFs for R-HCH in seawater ranged from 0.43 to 0.50, except for two samples from 75°N in the East Greenland Sea, with EFs * Corresponding author fax: (401) 874-6811; e-mail: lohmann@ gso.uri.edu. † University of Rhode Island. ‡ Previously Research Center for Ocean Margins, University of Bremen, 28199 Bremen, Germany. § Lancaster University. | University of Insubria. ⊥ Institute for Coastal Research. # Current address: Eurofins GfA GmbH, Geierstrasse 1, D-22305 Hamburg, Germany. ∇ Institute for Baltic Sea Research (IOW). O Current address: Simmons College, Boston, MA. [ Environment Canada. 10.1021/es901229k CCC: $40.75
Published on Web 07/06/2009
2009 American Chemical Society
of 0.31 and 0.37. Lowest EF (0.47) for CC were also at 75°N, other samples had EFs from 0.49 to 0.52. It is suggested that samples from around 75°N in the Greenland Gyre represented a combination of surface and older/deeper Arctic water.
Introduction The Arctic has been strongly affected by the presence of persistent organic pollutants (POPs) and their accumulation in the food-chain (1). Arguably, the transport of POPs to the Arctic led to the Stockholm Convention that defined and banned them or restricted their use (2). Among the targeted POPs are industrial chemicals, such as polychlorinated biphenyls (PCBs), several organochlorine (OC) pesticides, including dichlorodiphenyltrichloroethane (DDT), cis- and trans-chlordane (CC and TC), hexachlorobenzene (HCB), and unwanted byproducts, notably dioxins and furans. Polycyclic aromatic hydrocarbons (PAHs) are only recognized as POPs under the Aarhus Protocol (3) but are of concern due to their toxicity and mutagenicity. PAHs result from the incomplete combustion of carbonaceous materials. Due to direct and indirect photolysis, PAHs are less stable during atmospheric transport than most POPs, which makes them a complementary group of compounds for investigating the role of atmospheric persistence on the presence of POPs in remote oceans. Early work showed increasing concentrations of R-HCH with increasing latitude, illustrating the concept of cold condensation (4). Numerous studies have been undertaken to verify the theories of fractionation and cold condensation, looking at trends of PCBs in air, vegetation, biota, or passive samplers (5-7). While fractionation has been observed in regional data sets, it is not clear whether temperature-driven partitioning processes are indeed key drivers of POPs’ transport and fate. POPs have been extensively monitored in ambient air in the Arctic region (8-10) but less frequently in seawater, despite the importance of oceans for biological exposure and the global burden of many POPs. In fact, numerous processes other than property-changes due to temperature gradients affect the transport and fate of POPs on the global scale (11). For example, the biological pump not only drives CO2 sequestration by the oceans, but also removes POPs from the atmosphere via air-to-water exchange, partitioning into organic matter, and removal via carbon export to the deeper oceans (12). Oceanic deep water formation results in the removal of POPs from the surface to the deep ocean (13). Simultaneously, atmospheric reactions with OH radicals will deplete POPs during their longrange atmospheric transport (14). Once deposited to the ocean via dry and wet deposition (15, 16), POPs will partition into organic matter due to their hydrophobicity and accumulate in the food-chain (17). Microbial degradation of organic matter can also result in the cometabolism of POPs, as has been shown through the stereoselective loss of R-hexachlorocyclohexane (R-HCH) enantiomers with increasing depth (18). In this study, we analyzed marine boundary layer air and surface water samples from across the North Atlantic and “European” Arctic for the occurrence of POPs, notably HCHs, HCB, DDTs, PAHs, and the enantiomeric signature of R-HCH, heptachlor exo-epoxide (HEPX) and CC. Previous publications from this cruise focused on the occurrence of phthalates (19) and PCBs (20) in air, surface water, and their air-water exchange. The cruise ARKXX on the FS Polarstern covered both the Norwegian Coastal and West Spitsbergen current moving Atlantic water west of Norway and Svalbard into the VOL. 43, NO. 15, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5633
Arctic, and the East Greenland current, moving Arctic Ocean water back into the North Atlantic. The cruise track enabled us to determine whether different ocean currents affected atmospheric concentrations of POPs, and to what degree fractionation, reactions and cold condensation affected patterns of POPs in the northern oceans. Finally, we wanted to assess whether the recent reduction in primary emissions of many POPs has resulted in surface waters becoming secondary sources to the atmosphere.
Materials and Methods Cruise track. The cruise track and general sampling conditions are summarized in Supporting Information (SI) Table 1 for water samples and SI Table 2 for air samples. Surface seawater samples were taken characteristic of the Norwegian coastal current, the West Spitsbergen current, and the East Greenland current (see also Figure 1 in ref 20 for sampling locations of seawater samples). Sampling Protocol. The sampling of seawater and air is described in more detail in ref 20. Briefly, water samples were collected from the ship’s intake system located in the keel using a combination of GFF filters and XAD. Samples were stored at -4 °C. All seawater samples were extracted and analyzed at the Institute for Baltic Sea Research, except for HCHs and chlordanes which were analyzed at Lancaster University. Enantiomers of OC pesticides were determined at Environment Canada (except for TC, which could not be quantified due to an interfering peak). OC pesticides in the atmosphere were analyzed at Lancaster University, and atmospheric PAHs at the University of Rhode Island. Two sets of air samples were collected. OC pesticides were sampled and analyzed on the sample set described in ref 20 for PCBs. One additional high-volume (hi-vol) air sampler (Tisch Environmental PUF style sampler) was used to collect air for PAHs at ∼1 m3/min. (See SI for more details on instrumental analysis and quality control). Samplers were only operated under good conditions, i.e., with the air coming from the bow of the ship (20, 21). Ancillary Data. Latitude, longitude, air and water temperature, salinity, air pressure, and biological parameters were averaged for each water sample from 5 min records from PODAS (Polarstern Data System), an on-board management system that collects nautical and scientific parameters (SI Table 1). Physico-Chemical Properties. For OC pesticides and PAHs, internally consistent values of air-water distribution coefficients, Kaw, were taken from ref 22. Internal energies of air-water exchange (Uaw) for OC pesticides were taken from ref 22, and those for PAHs from ref 23 or assumed to be 50 kJ/mol. Salting out (Setschenow) constants were taken as 0.3 (24). Air-Water Exchange. A total of 20 paired air-water samples were used to assess the degree of air-water equilibrium for OC pesticides along the transect. Air-water fugacity gradients were calculated according to fair/fwater)cgas/cdiss/Kaw(T, sal)
(1)
where fair and fwater are the fugacities in air and water, and Kaw (T, sal) is the (water) temperature salinity-corrected Kaw. A fugacity ratio >1 indicates deposition, 75°N. Above 80°N, [HCB]gas were fairly constant ranging from 40 to 50 pg/m3. Concentrations of dissolved HCB ([HCB]diss) in water increased from near Europe toward the high Arctic (Figure 1). [HCB]diss ranged from 1 to 2 pg/L from 62 to 72°N, ∼2 to 5 pg/L at ∼75°N, 3 to 4 pg/L at 78°N, and increased to 4 to 10 pg/L from 79 to 85°N (SI Table 2). HCB was detected in most particulate samples (SI Table 2, Figure 1). Concentrations increased from ∼0.1 pg/L near Europe to 0.3 pg/L in the northernmost samples. [HCB]diss were comparable to previous results (SI Table 5). Atmospheric [HCB]gas did not display significant correlations with temperature. In contrast, dissolved HCB showed a strong trend of increasing concentration with decreasing temperature (Figure 2). [HCB]diss (in pg/L) was significantly correlated with latitude (R2 ) 0.52, P < 0.001, n ) 22), and longitude (R2 ) 0.53, P < 0.001). Strongest correlations were obtained with either average Tair ([HCB]diss ) 6.8 - 0.74 × Tair, R2 ) 0.84, P < 0.001) or Twater ([HCB]diss ) 6.4 - 0.57 × Twater, R2 ) 0.76, P < 0.001) (see Figure 2). Previous studies have suggested that HCB is much closer to attaining equilibrium in the Northern hemisphere than most other POPs (27). Our results support this as far as there were no elevated concentrations close to Europe in air or water. Hence, temperature is the best predictor of [HCB]diss. However, the correlation with Tair is at least partially coincidental, as the northernmost samples were taken under ice cover. Similar results were obtained correlating ln [HCB]diss versus 1/T (i.e., the Clausius-Clapeyron relationship; R2 ) 0.77, P < 0.001), which expresses the change in solubility with T. Our results show higher [HCB]diss along the western side of the North Atlantic. We suggest this could be caused by the cold East Greenland current carrying water with equilibrated, higher [HCB]diss out of the Arctic Ocean. We had no sampling stations further south on the western side, but expect that concentrations would tend to decrease due to air-water exchange, as the water and air increase in temperature. HCB was closest to equilibrium between air and water (Figure 3). Only five samples (n ) 20) had air-to-water gradients indicating net deposition (>3.0), four of which occurred in samples closest to Europe (69-73°N). As samples were taken in the summer, this was probably due to removal fluxes of HCB with settling particles, resulting in disequilibrium between air and water (e.g., ref 12). HCB detected in the particulate phase showed a similar profile to dissolved concentrations, with higher concentrations north and west. Su et al. (9) estimated that HCB is currently undergoing net deposition to the Arctic Ocean, based on average air and water concentrations of 56 pg/m3 and 5.5 pg/L, respectively. HCHs. In the atmosphere, [R-HCH]gas ranged from
