Perfluorinated acids as novel chemical tracers of global circulation of ocean waters

Available online at www.sciencedirect.com

Chemosphere 70 (2008) 1247–1255 www.elsevier.com/locate/chemosphere

Perfluorinated acids as novel chemical tracers of global circulation of ocean waters Nobuyoshi Yamashita a,*, Sachi Taniyasu a, Gert Petrick b, Si Wei c, Toshitaka Gamo d, Paul K.S. Lam c, Kurunthachalam Kannan e a

National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan b Leibniz-Institute of Marine Sciences, Department of Marine Chemistry, Du¨sternbrooker Weg 20, D-24105 Kiel, Germany c Department of Biology and Chemistry, City University of Hong Kong, Kowloon, Hong Kong, SAR, China d Ocean Research Institute, University of Tokyo, Nakano-ku, Tokyo 164-8639, Japan e Wadsworth Center, New York State Department of Health and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Empire State Plaza, P.O. Box 509, Albany, NY 12201-0509, USA Received 16 May 2007; received in revised form 19 July 2007; accepted 26 July 2007 Available online 12 September 2007

Abstract Perfluorinated acids (PFAs) such as perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) are global environmental contaminants. The physicochemical properties of PFAs are unique in that they have high water solubilities despite the low reactivity of carbon-fluorine bond, which also imparts high stability in the environment. Because of the high water solubilities, the open-ocean water column is suggested to be the final sink for PFOS and PFOA. However, little is known on the distribution of PFAs in the oceans around the world. Here we describe the horizontal (spatial) and vertical distribution of PFAs in ocean waters worldwide. PFOS and PFOA concentrations in the North Atlantic Ocean ranged from 8.6 to 36 pg l1 and from 52 to 338 pg l1, respectively, whereas the corresponding concentrations in the Mid Atlantic Ocean were 13–73 pg l1 and 67–439 pg l1. These were completely different from the surface waters of the South Pacific Ocean and the Indian Ocean (overall range of <5–11 pg l1 for PFOS and PFOA). Vertical profiles of PFAs in the marine water column were associated with the global ocean circulation theory. Vertical profiles of PFAs in water columns from the Labrador Sea reflected the influx of the North Atlantic Current in surface waters, the Labrador Current in subsurface waters, and the Denmark Strait Overflow Water in deep layers below 2000 m. Striking differences in the vertical and spatial distribution of PFAs, depending on the oceans, suggest that these persistent acids can serve as useful chemical tracers to allow us to study oceanic transportation by major water currents. The results provide evidence that PFA concentrations and profiles in the oceans adhere to a pattern consistent with the global ‘‘Broecker’s Conveyor Belt’’ theory of open ocean water circulation.  2007 Elsevier Ltd. All rights reserved. Keywords: PFOS; PFOA; Perfluorinated chemicals; Global circulation; Deep seawater; Chemical tracer

1. Introduction It is clear from the discovery of perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) in wildlife around the world that perfluorinated acids (PFAs) have become global environmental pollutants (Giesy and *

Corresponding author. Tel.: +81 29 861 8335; fax: +81 29 861 8335. E-mail address: [email protected] (N. Yamashita).

0045-6535/$ - see front matter  2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2007.07.079

Kannan, 2001). The physicochemical properties of PFAs are different from those of several established global pollutants such as polychlorinated biphenyls (PCBs). In efforts to determine the mechanism of global transport and distribution of PFAs, global ocean monitoring of PFAs was initiated in 2002 at the National Institute of Advanced Industrial Science and Technology (AIST) in collaboration with the Ocean Research Institute (ORI) of Tokyo University, Japan, and the Leibniz-Institute of Marine Sciences in

1248

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

Kiel, Germany (Yamashita et al., 2005). Concentrations of PFAs in surface waters of the North Atlantic Ocean and the Mid Atlantic have been found to be higher than the concentrations found in the South Pacific Ocean (Yamashita et al., 2005). Atmospheric transport of volatile fluorotelomer alcohols (FTOHs) or other precursor compounds, and subsequent oxidative reactions in the atmosphere, have been hypothesized to be the major sources for some of the PFAs found in remote regions (Ellis et al., 2004; Wallington et al., 2006). Relatively higher concentrations of PFAs in precipitation (Scott et al., 2006; Taniyasu et al., 2006) than in surface waters suggest dispersion of PFAs by an atmospheric route, through wet deposition. However, other mechanisms of transport, such as by ocean currents (Prevedouros et al., 2006), have been suggested as alternative routes of transport of PFAs. To date, the mechanisms of global distribution of PFAs remain elusive. In this study, the vertical and horizontal distribution of PFAs in ocean waters was examined, to enable us to describe the transport of PFAs by ocean currents. We propose that PFAs represent excellent tracers of global circulation of oceanic waters, due to their persistence, water solubility, and measurability. The strong carbon-fluorine bond (bond energy of 485 kJ mol1) imparts stability to PFAs. The water solubilities of PFOS and PFOA are 570 and 3400 mg l1, respectively (OECD, 2002; USEPA, 2002). Because of these high water solubilities, the open-ocean water column is suggested to be the final sink for PFOS and PFOA (Prevedouros et al., 2006). Concentrations of PFAs in open-ocean waters have been reliably measured at parts-per-quadrillion levels, by a combination of solid phase extraction and high performance liquid chromatography (HPLC)-tandem mass spectrometry (Yamashita et al., 2004; Taniyasu et al., 2005). This method has been validated for and applied to the global survey of PFAs in open ocean waters. The detection limits for PFOS and PFOA were on the order of one femtomole and twelve femtomole, respectively, in 1 l of seawater. The dynamic range of the analytical method for PFOS and PFOA was on the order of 108. Such a high level of analytical sensitivity is comparable to that available for other radiochemical tracers such as, tritium, that are widely used in oceanography (Gamo et al., 2001). Furthermore, PFOA is relatively less bioaccumulative than are lipophilic contaminants such as PCBs (Nakata et al., 2006). Accordingly, the influence of seasonal variations in biological productivity and particle scavenging on the concentration of PFOA in ocean waters is expected to be minimal. The three attributes described above for PFOS and PFOA make clear the suitability of these two compounds as chemical tracers of ocean water circulation. Some of the currently used chemical tracers, such as sulfur and ferrates, are biologically available to flora and fauna in oceans, and the levels of these chemicals in ocean waters can thus be altered seasonally by variations in primary or secondary production (Bullister et al., 2006; Steiner et al., 2006).

Here, we report the horizontal and vertical distribution of PFAs in ocean water samples collected during a number of research cruises in several oceans. Vertical profiles of PFAs in the marine water column were associated with the global ocean circulation theory. Our results provide key information to explain global dispersion of PFAs in the oceans. 2. Methods Open ocean and offshore surface and subsurface water samples were collected between 2002 and 2006 from 62 locations (Table 1; Supplementary Information) around the world. Nine water-column samples, at several depths of up to 5500 m, were collected from the Labrador Sea, the Mid Atlantic Ocean, the South Pacific Ocean, and the Japan Sea. Seawater were collected by rosette-type sampler equipped with Go-Flo or Niskin bottles during several international ocean research cruises (M55, M59, M68/2, CG161/62, SO177, KT-05-11, KH-03-1, KH-04-5, KH02-4) by the Leibniz-Institute of Marine Sciences (Germany), ORI, and Ibaraki University (Japan). Water samples were stored in clean 1 l polypropylene bottles and were kept frozen until analysis. Samples were thawed at room temperature, and a solid phase extraction method using WAX cartridge (Waters Co.) developed by Taniyasu was used for the determination of PFAs by HPLCtandem mass spectrometry (HPLC-MS/MS) (Yamashita et al., 2004; Taniyasu et al., 2005). HPLC-MS/MS, composed of a HP1100 liquid chromatograph (Agilent Technologies, Palo Alto, CA) interfaced with a Micromass (Beverly, MA) Quattro Ultima Pt mass spectrometer was operated in the electrospray negative ionization mode. Ten microliters aliquot of the sample extract was injected into a guard column (Ace 3 C18, 2.1 mm i.d., 3 lm; Advanced Chromatography Technologies) connected sequentially to a Ace 3 C18 column (2.1 mm i.d. · 150 mm length, 3 lm; Advanced Chromatography Technologies). MS/MS was operated under multiple reaction monitoring (MRM) mode, and the parameters were optimized for transmission of the [M–K] or [M–H] ions. Several PFAs including C2–C18 perfluorocarboxylic acids, C3–C10 perfluoroalkylsulfonates and FTOH, were determined in the water samples. In this report, we discuss the results for three PFAs, namely PFOS, PFOA, and perfluorobutanesulfonate (PFBS). Procedure blanks and travel/ field blanks were included during each of the sampling cruises, and the limits of quantification were determined as: 1 pg for PFOS, 6 pg for PFOA, and 0.2 pg for PFBS in 1 l of water sample. Field blanks contained 800 ml of Milli-Q water were taken in a polypropylene bottle and transported to sampling locations. Although the concentrations of target fluorochemicals in field blanks were similar to those in procedural blanks in most cases, any sample sets that were found to have notable contamination in field blanks were eliminated from further analysis. Recoveries of target PFAs were examined by spiking one hundred pg of each compound in 1 l of Milli-Q water. Recoveries

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

1249

Table 1 Locations of water sampling from various oceans and seas Sampling site Offshore seawater Offshore Japan

Offshore Angola Open ocean water North Atlantic Ocean (M59 and CG161/62 cruises)

AO1 AO2 Mid-Atlantic Ocean (M55 and M68/2 cruises)

AO3 AO4 AO5 Central to Western Pacific Ocean (Ibaraki University, KH-03-1 and KH-04-5 cruises)

South Pacific Ocean (KH-04-5 cruise) PO1 PO2

South China Sea (KH-02-4 and SO177 cruises)

Depth

Latitude

Longitude

Date

Reference (number of layers)

(0–2 m) (0–2 m) (0–2 m) (0–2 m)

3457 0 N 3322 0 N 3259 0 N 3122 0 N

13933 0 E 13536 0 E 13349 0 E 13143 0 E

2002-Oct 2002-Oct 2002-Oct 2002-Oct

*

Surface (0–2 m)

1225 0 S

1237 0 E

2004-May

#

Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) 45–3500 m 15–2750 m

4950 0 N 5526 0 N 5706 0 N 5324 0 N 4824 0 N 4313 0 N 4418 0 N 4701 0 N 4820 0 N 5634 0 N 5941 0 N

5238 0 W 5326 0 W 5513 0 W 5018 0 W 4501 0 W 4912 0 W 3721 0 W 2424 0 W 1410 0 W 5248 0 W 3940 0 W

2003-Aug 2003-Sep 2003-Sep 2003-Sep 2003-Sep 2003-Sep 2003-Sep 2003-Sep 2003-Sep 2004-Sep 2004-Sep

*

Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) Surface (0–2 m) 100–5460 m 100–5500 m 100–5450 m

1012 0 N 1000 0 N 1000 0 N 1000 0 N 000 0 N 1059 0 N 1135 0 N 2317 0 N 2547 0 N 2703 0 N

5219 0 W 4335 0 W 3511 0 W 2905 0 W 2441 0 W 2300 0 W 1729 0 W 6419 0 W 6459 0 W 6435 0 W

2002-Oct 2002-Oct 2002-Oct 2002-Oct 2002-Oct 2002-Nov 2002-Nov 2004-Mar 2004-Mar 2004-Mar

*

Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface Surface

2331 0 N 2617 0 N 2000 0 N 0 1630 0 N 802 0 N 358 0 S 800 0 S 1508 0 S 2000 0 S 2215 0 S 2600 0 S 2829 0 S 2559 0 S 2200 0 S 1300 0 N 039 0 S 005 0 N

124 40 0 E 128 31 0 E 14000 0 W 12300 0 W 9457 0 W 9500 0 W 8549 0 W 8549 0 W 10059 0 W 10800 0 W 11959 0 W 12747 0 W 14000 0 W 14159 0 W 15057 0 E 15731 0 E 17000 0 W

2002-Oct 2002-Oct 2003-Jul 2003-Jul 2003-Jul 2003-Jul 2003-Jul 2003-Jul 2003-Jul 2003-Aug 2003-Aug 2003-Aug 2003-Aug 2003-Aug 2004-Dec 2004-Dec 2004-Dec

*,**

Surface (0–2 m) Surface (0–2 m) 0–3730 m Surface (0–2 m) 0–5400 m Surface (0–2 m)

6000 0 S 6321 0 S 6712 0 S 5922 0 S 5001 0 S 3959 0 S

14959 0 E 17008 0 E 17240 0 W 16957 0 W 16959 0 W 16959 0 W

2004-Dec 2004-Dec 2004-Dec 2004-Dec 2004-Dec 2004-Dec

#

Surface Surface Surface Surface Surface Surface

1128 0 N 850 0 N 917 0 N 1330 0 N 1430 0 N 3959 0 S

12150 0 E 12148 0 E 12013 0 E 11930 0 E 11800 0 E 16959 0 W

2002-Dec 2002-Dec 2002-Dec 2002-Dec 2002-Dec 2004-Dec

*,**

Surface Surface Surface Surface

(0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m)

(10 m) (10 m) (10 m) (10 m) (10 m) (8 m)

* * *

* * * * * * * * # #

(11 layers) (13 layers)

*,** *,** *,** * * * #

(9 layers) (8 layers) # (10 layers) #

*,** * * * * * * * * * * * * # # #

# #

(9 layers)

# #

(14 layers)

*

*,** *,** *,** *,** #

(continued on next page)

1250

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

Table 1 (continued) Sampling site

Depth

Latitude

Longitude

Date

Reference (number of layers)

Japan Sea (KT-05-11 cruise) JS1 JS2

Surface (10 m) 10–3250 m 10–3440 m

3919 0 N 4043 0 N 4412 0 N

13624 0 E 13634 0 E 13854 0 E

2005-May 2005-May 2005-May

#

Indian Ocean

Surface Surface Surface Surface Surface Surface Surface Surface

1715 0 S 3032 0 S 4924 0 S 5858 0 S 5458 0 S 6347 0 S 6757 0 S 6642 0 S

11421 0 E 11432 0 E 11441 0 E 10643 0 E 8353 0 E 8212 0 E 7559 0 E 7027 0 E

2005-Dec 2006-Mar 2005-Dec 2005-Dec 2005-Dec 2005-Dec 2006-Jan 2006-Jan

***

# * ** ***

(0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m) (0–2 m)

# #

(12 layers) (12 layers)

*** *** *** *** *** *** ***

This study. Yamashita et al. (2005). Yamashita et al. (2004). Wei et al. (in press).

of PFOS, PFBS, PFOA, 13C4-PFOS and 13C4-PFOA through the whole procedure (n = 10) were 95(±4)%, 96(±8)%, 107(±7)%, 91(±2)% and 91(±6)%, respectively. Further quality assurance and quality control (QA/QC) protocols were based on participation in a round-robin test of the International Organization for Standardization (ISO/TC147/SC2/WG56, PFOS and PFOA; protocol number CD25101.2). Temperature and salinity gradients were obtained during the sampling process at each site. 3. Results and discussion Concentrations of PFOS and PFOA measured in the surface waters of several oceans are shown in Fig. 1. PFOS and PFOA concentrations in the North Atlantic Ocean ranged from 8.6 to 36 pg l1 and from 52 to 338 pg l1, respectively, whereas the corresponding concentrations in the Mid Atlantic Ocean were 13–73 pg l1 and 67– 439 pg l1. The surface waters of the North and MidAtlantic Ocean are more highly contaminated with PFOS and PFOA than are the surface waters of the South Pacific Ocean and the Indian Ocean (overall range of <5 to 11 pg l1 for PFOS and PFOA). The highest concentrations of PFOS and PFOA found in the Atlantic Ocean were 15 and 88 times, respectively, higher than the lowest concentrations determined in the South Pacific Ocean and the Indian Ocean. The off-shore waters of the eastern North Pacific Ocean, along the Asia-Pacific Rim, contained 10–50-fold higher concentrations of PFOA than the water collected in the western North Pacific Ocean. Concentration of PFOA in the eastern Pacific Ocean ranged from 10 to 60 pg l1 whereas that in the western Pacific Ocean ranged from 140 to 500 pg l1. Such a remarkable longitudinal (east–west) difference in the concentrations of PFOA in the Pacific Ocean was unexpected because many PFAs have been produced and used in large quantities in North America (Prevedouros et al., 2006); thus the concentrations in the eastern North Pacific Ocean are expected to be high. Ocean water collected from the southern Indian Ocean

contained trace levels of PFAs. PFAs were not found in the surface waters of the Antarctic Ocean (7027 0 E to 17240 0 W), which is explicable by the isolation of the Antarctic circumpolar water from the water masses in the Northern Hemisphere. There was striking match between our finding to marine birds collected from the Southern Ocean and the Antarctic contained only trace levels of PFOS, whereas the marine birds from the Northern Hemisphere contained high levels of PFOS and other PFAs (Tao et al., 2006). Since the ocean is a three dimensional environment, general monitoring survey of PFAs in the surface water provide very limited information of global kinetics. It’s clear that comprehensive survey of not only in surface water but also deep water in Oceans and Seas are necessary to reconstruct the global kinetics model of PFAs. Therefore, we collected water column samples at several depths, from the Labrador Sea, the Mid-Atlantic Ocean, the South Pacific Ocean, and the Japan Sea, to enable an understanding of the hydrodynamics of PFAs, and to elucidate the global oceanic circulation. The vertical distribution of PFOA in selected oceans is presented in Fig. 1. Further details of the vertical profiles of PFOS, PFOA and PFBS along with temperature and salinity gradients of water columns, are shown in Fig. 2. The vertical profile of PFAs in the two water columns from the Japan Sea (JS1 and JS2) showed gradual decrease in PFA concentrations from surface to subsurface layers (Fig. 2). Concentrations of PFOS in surface water samples were 15 pg l1, and decreased to 2 pg l1 at a depth of 1000 m. At depths below 1500 m, concentrations were at or below the detection limit of 1 and 6 pg l1, for PFOS and PFOA, respectively. The vertical profiles of PFOS and PFOA concentrations were similar to the profiles for the widely used chemical tracer, tritium, in the Japan Sea (Gamo et al., 2001) (Fig. 2). This agreement suggests that PFAs are good chemical tracers for oceanographic studies. The high water solubility of tritium makes it a suitable chemical tracer for the study of global ocean circulation, and PFAs share this characteristic.

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

1251

Fig. 1. Concentrations of PFOS and PFOA in 62 surface-water samples and nine water-column samples collected from several oceans worldwide (modified from original figure by W. Broecker, 1991). Bars represent concentrations in pg l1. Red diamonds refer to surface-water sampling sites. Watercolumn sampling locations are represented as A, B, C and D.

The vertical profile of PFOS and PFOA concentrations in the Labrador Sea (AO1 and AO2) showed a remarkable difference from the JS1 and JS2 profiles from the Japan Sea (Fig. 2). Concentrations of PFOS and PFOA in the AO1 water column samples were relatively constant throughout the depth, except in subsurface water samples and water below 2000 m, in which PFOA concentrations increased. The AO2 water column samples from the Labrador Sea showed high concentrations of PFOS and PFOA at the surface and then uniform concentrations down to 2000 m. Temperature and salinity measurements for the AO1 and AO2 water columns suggested that the water mass from surface to a depth of 2000 m was well-mixed (Fig. 2). The Labrador Sea is known as the critical location for global circulation of ocean water and several investigations have revealed the role of this sea in moderating Earth’s climate (Dickson et al., 2002; Smith et al., 2005; Kieke et al., 2006, 2007) through its convective formation of the water mass known as Labrador Sea Water (LSW); this is the key element of the global thermocline circulation. The vertical profile of PFAs described above shows a trend consistent with the known hydrodynamics of the Labrador Sea. A simplified illustration of the water currents in the North Atlantic Ocean around the Labrador Sea and the vertical profiles of concentrations of three PFAs in

the AO1 and AO2 water columns are presented in Fig. 3. The overflow and descent of cold, dense water from the sills of the Denmark Strait (Denmark Strait Overflow Water or DSOW) and Faroe–Shetland Channel into the North Atlantic Ocean are the principal means by which the deep oceans are ventilated, and the Labrador Sea is the site at which deep sea waters originate. Concentrations of PFAs in the AO1 water column from the Labrador Sea were low in the subsurface layers, at depths of 100–200 m. The profile of water temperature and salinity indicated influx of the Labrador Current (LC) that contained melt-waters having low salinity. This influx of water markedly decreases the concentrations of PFAs in the subsurface waters (100–200 m). Concentrations of PFBS and PFOA increased at depths below 2000 m. This pattern suggests the presence of an independent deep-water current, namely, the DSOW. PFOA was the predominant PFA throughout the depth of the AO1 water column. The DSOW lies below the convective current of the LSW, but the waters of the two currents do not mix. High concentrations of PFAs were found in the surface waters of the AO2 water column collected from the North Atlantic Ocean (Fig. 3). Significant amount of PFAs from the North Atlantic Current (NAC) appears to have contributed to the high concentrations of PFAs

1252

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

A : Labrador Sea PFOS concentrations [pg/L] 0

10

20

30

40

PFOA concentrations [pg/L]

50

0

50

100

PFBS concentrations [pg/L]

150

0

20

40

60

80

Salinity [psu]

Temperatyre [ºC]

100

0

2

4

6

8

10

34.7

34.8

34.9

35

35.1

0 500

depth [m]

1000 1500 2000

PFOS

2500

3500 4000

PFOA

PFBS AO1 AO2

AO1 AO2

3000

LOQ: 2 pg/l

AO1 AO2

AO1

AO1 AO2

LOQ: 7 pg/L

B : Middle Atlantic Ocean 10

20

30

40

50

0

50

100

Salinity [psu]

Temperature [ºC]

PFOA concentrations [pg/L]

PFOS concentrations [pg/L] 0

0

150

10

20

30

33

34

35

36

37

0 500 1000 1500 depth [m]

2000 2500 3000

PFOS

3500 4000 4500 5000 5500

PFOA AO3 AO4 AO5

AO3 AO4 AO5

LOQ: 0.4 pg/l

AO3 AO4 AO5

AO3 AO4 AO5

LOQ: 7 pg/L

C : South Pacific Ocean 0

PFOS concentrations [pg/L] 10 20 30 40 50

0

PFOA concentrations [pg/L] 50 100 150

-5

Temperature [ºC] 5 15

25

33

34

Salinity [psu] 35 36

37

0 500 1000 1500 depth [m]

2000 2500 3000 3500

PFOS

4000

5000 5500

PFOA PO1 PO2

4500

PO1 PO2

LOQ: 2 pg/l

PO1 PO2

PO1 PO2

LOQ: 6 pg/L

D : Japan Sea PFOA concentrations [pg/L]

PFOS concentrations [pg/L] 0

10

20

30

40

50

0

50

100

Tritium [Bq/L]

150

0

0.1

0.2

0.3

Temperature [ºC]

0.4

0.5

0

10

Salinity [psu]

20

30

33

34

35

36

37

0 500

depth [m]

1000 1500 2000

PFOS

2500

PFOA JS1 JS2

3000 3500

Tritium JS1 JS2

Station CM12, 1984

JS1 JS2

JS1 JS2

4000

LOQ: 1 pg/l

LOQ: 6 pg/L

Gamo et al. (2001)

Fig. 2. Vertical profiles of PFAs in ocean water columns from (A) the Labrador Sea, (B) the Mid-Atlantic Ocean, (C) the South Pacific Ocean, and (D) the Japan Sea. Temperature and salinity profiles are shown. Tritium profile (Gamo et al., 2001) is shown for the Japan Sea water column.

in the surface-layer of AO2. The eastern Greenland Sea is also known for convection of surface NAC, resulting in contaminated subsurface DSOW. It can be concluded that the PFAs transported from surface waters of the NAC sink

to deep water through the There are two locations of the eastern Greenland Sea PFA concentration of the

convection of surface water. deep sea water formation in and the Labrador Sea. The AO2 water column showed

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

1253

AO1

concentrations [pg/l] 0

deep water C: Convection

LSW : Labrador Sea Water LC : Labrador Current DSOW : Denmark Strait Overflow Water DWBC: Deep Western Boundary Current NAC : North Atlantic Current EGC : East Greenland Current

EG C

E: Entrainment

60

80

LSW

1000

C

100

LC

500

1500 2000

DSOW (DWBC)

2500 3000

DSOW

LC

40

0

depth [m]

surface water

20

3500 4000

E

PFOA

0

20

40

60

80

C

NAC

depth [m]

500

AO2 LSW

100

0

E

AO1

PFBS

concentrations [pg/l]

AO2

DWBC

PFOS

1000

+

1500

LSW

2000 2500 3000

NAC

DSOW

3500 4000 PFOA

PFOS

PFBS

Fig. 3. Vertical profiles of PFAs in ocean water columns from the Labrador Sea (AO1) and the North Atlantic Ocean (AO2) with proposed water currents and hydrodynamics.

relatively complex vertical profile and there appeared to be a mixed influence of entrainment of the NAC and convection of the LSW. In the AO2 water column, concentrations of PFBS were higher than the concentrations of PFOS and PFOA. This is different from the pattern for the AO1 water column, suggesting that the sources of PFAs differ between the AO1 and AO2 water columns. While PFOS and PFOA have been in production for more than 60 years, PFBS was introduced into the global market in the early 2000s, as a replacement for PFOS (Prevedouros et al., 2006). The occurrence of PFBS in the water column of the Labrador Sea suggests input of this compound to the oceans after 2000 or earlier input by other source such as impurity. The distribution of chlorofluorocarbons (CFCs), another class of chemical tracers, in the Labrador Sea has been studied (Kieke et al., 2006, 2007). It was estimated that, in the Labrador Sea, the rate of formation of water mass during winter time convection is one million m3 per second (Kieke et al., 2006). Concentrations of PFOS and PFOA in surface waters in the AO1 water column were 20 and 55 pg l1, respectively. If the volume of water sinking into the LSW is assumed to be constant, then 620 kg (1.7 kg d1) of PFOS and 1460 kg (4 kg d1) of PFOA

can be estimated to be transferred into the deep oceans every year by this mechanism. The quantity of emissions of PFAs from industrial activities was estimated to be 3200–7300 tons (Prevedouros et al., 2006). Based on this estimate and the sinking rate of water in the Labrador Sea, approximately 1% (68 tons) of the total emissions of PFOA has been transferred into the deep sea water over the past 60 years. At this rate of distribution, more than 4500 years would be needed for transfer of all of the current emissions of PFOA into the deep sea water in the Labrador Sea. This calculation is based on a rough estimate of the global inventory of PFOA currently available and the sinking rates of CFCs. However, the use of CFCs as chemical tracers in the study of hydrodynamics in the oceans is limited because of the saturation equilibrium of CFCs, between the atmosphere and oceans (Kieke et al., 2007). On the other hand, PFAs, due to their high water solubility, are not expected to reach saturation concentrations in water, should therefore be better chemical tracers for LSW research. Hence, it can be concluded that PFAs have several essential features as oceanographic chemical tracers; ‘high water solubility (no escape into air)’, ‘persistency’, ‘traceability in open ocean water’ and ‘less bioavailability’. These features are different

1254

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255

from other traditional chemical tracers in oceanography, such as sulfur and ferrates. A report by Lohmann et al. (2006) showed sinking of polychlorinated biphenyls (PCBs) by several deep water formations. They estimated the total PCB flux due to sinking in the deep water of the Labrador Sea to be 140 kg per year. This value was four times lower than the value for PFOS and ten times lower than that for PFOA, estimated in our study, even though the rate of deep water formation was five times faster in their calculation. It can be concluded that non-volatile and high soluble nature of perfluoroacids results in faster elimination into deep oceans compared to semi-volatile persistent organic pollutant such as PCBs. Three water columns (AO3, AO4, AO5) from the MidAtlantic Ocean showed vertical PFA profiles similar to those of the Japan Sea columns JS1 and JS2 (Fig. 2). In each of the three Mid-Atlantic water columns, there was a considerable difference in PFA concentrations between the surface and middle layers of the water column, below 800 m. This pattern was consistent with a marked change in gradient, of both temperature and salinity, below 800 m. Concentrations of PFAs were almost negligible in the deepest layers, below 4000 m. The latter finding suggests a lack of direct vertical transport of PFAs from surface to bottom waters. This pattern is clearly different from that found in the Labrador Sea. The vertical profile of PFAs in the Mid-Atlantic Ocean seems to be typical of the profiles found for coastal and off-shore regions of industrialized countries (Yamashita et al., 2005). We can conclude that PFAs discharged into the surface waters in the Mid-Atlantic Ocean have long residence times due to the isolation of the surface waters from the deep waters, and due also to the circulation of the water mass by the Gulf Stream and the North Atlantic Drift. These factors provide an explanation for the fact that the highest concentrations of PFAs are found in the Mid-Atlantic Ocean. The above results also suggest possible transportation of PFAs discharged into LSW and DSOW to deep water in the southern Atlantic Ocean by Deep Western Boundary Current (DWBC) in Fig. 3. The estimated period of water mass transport from DSOW to the southern Atlantic Ocean is 100 years at the minimum (personal communication with Prof. Wallace and Dr. Tanhua). The production and usage of PFAs started in the early 1960s and the initial input, namely ‘pollution front’ of PFAs entered into the global circulation through deep water currents is still on their way to the southern Atlantic Ocean. The vertical profiles of PFAs in two water columns, PO1 and PO2 from the South Pacific Ocean were completely different from those found for the other oceans and seas studied here. Concentrations of PFAs were consistently low (<10 pg l1) or below the limit of detection, from surface to bottom. These water columns are samplings of ocean currents that are more than 1000 years old, in terms of global circulation, as well as surface coastal streams derived from Antarctic circumpolar waters. Despite the fact that PO1 and PO2 are different masses of water, as evidenced

by differences in the temperature and salinity profiles between PO1 and PO2, PFA concentration profiles were constant throughout both columns. Absence of PFAs in the water columns of the South Pacific Ocean showed that there is no direct input of PFAs to this remote region. Thus, the open ocean water in the South Pacific is less contaminated by PFAs than the water masses in other oceans. The pollution sources to the Pacific Ocean differ from the sources to the Atlantic Ocean. The results presented above provide evidence that PFA concentrations and profiles in the oceans adhere to a pattern consistent with the global ‘‘Broecker’s Conveyor Belt’’ theory of open ocean water circulation (Broecker, 1991), as described in Fig. 1. Striking differences in the vertical and spatial distribution of PFAs, depending on the oceans, suggest that these persistent acids can serve as useful chemical tracers to allow us to study oceanic transportation by major water currents. The unique physicochemical properties of PFAs make them ideal tracers for this purpose. Acknowledgements We thank the staff of ORI in Japan, Leibniz-Institute of Marine Sciences in Germany, Ibaraki University in Japan, City University of Hong Kong, and Dr. Yuichi Horii at the Wadsworth Center in the USA, for sample collection. We also thank Prof. Douglas Wallace and Dr. Toste Tanhua at the Leibniz-Institute of Marine Sciences for their critical comments. References Broecker, W.S., 1991. The great ocean conveyor. Oceanography 4, 79–89. Bullister, J.L., Wisegarver, D.P., Sonnerup, R.E., 2006. Sulfur hexafluoride as a transient tracer in the North Pacific Ocean. Geophys. Res. Lett. 33, L18603. Dickson, B., Yashayaev, I., Meincke, J., Turrell, B., Dye, S., Holfort, J., 2002. Rapid freshening of the deep North Atlantic Ocean over the past four decades. Nature 416, 832–837. Ellis, D.A., Martin, J.W., De Silva, A.O., Mabury, S.A., Hurley, M.D., Andersen, M.P.S., Wallington, T.J., 2004. Degradation of fluorotelomer alcohols: a likely atmospheric source of perfluorinated carboxylic acids. Environ. Sci. Technol. 38, 3316–3321. Gamo, T., Momoshima, N., Tolmachyov, S., 2001. Recent upward shift of the deep convection system in the Japan Sea, as inferred from the geochemical tracers tritium, oxygen, and nutrients. Geophys. Res. Lett. 28, 4143–4146. Giesy, J.P., Kannan, K., 2001. Global distribution of perfluorooctane sulfonate in wildlife. Environ. Sci. Technol. 35, 1339–1342. ISO, 2007. Water quality – determination of perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) – Method for unfiltered samples using solid phase extraction and liquid chromatography/mass spectrometry. ISO/CD25101.2 (Committee draft of ISO/TC147/SC2/ WG56). Kieke, D., Rhein, M., Stramma, L., Smethie, W.M., LeBel, D.A., Zenk, W., 2006. Changes in the CFC inventories and formation rates of Upper Labrador Sea Water, 1997–2001. J. Phys. Oceanogr. 36, 64–86. Kieke, D., Rhein, M., Stramma, L., Smethie, W.M., Bullister, J.L., LeBel, D.A., 2007. Changes in the pool of Labrador Sea Water in the subpolar North Atlantic. Geophys. Res. Lett. 34, L06605.

N. Yamashita et al. / Chemosphere 70 (2008) 1247–1255 Lohmann, R., Jurado, E., Pilson, M., Dachs, J., 2006. Oceanic deep water formation as a sink of persistent organic pollutant. Geophys. Res. Lett. 33, L12607. doi:10.1029/2006GL025953. Nakata, H., Kannan, K., Nasu, T., Cho, H.S., Sinclair, E., Takemura, A., 2006. Perfluorinated contaminants in sediments and aquatic organisms collected from shallow water and tidal flat areas of the Ariake Sea, Japan: Environmental fate of perfluorooctane sulfonate in aquatic ecosystems. Environ. Sci. Technol. 40, 4916–4921. OECD, 2002. Hazard assessment of perfluorooctane sulfonate (PFOS) and its salts. Prevedouros, K., Cousins, I.T., Buck, R.C., Korzeniowski, S.H., 2006. Sources, fate and transport of perfluorocarboxylates. Environ. Sci. Technol. 40, 32–44. Scott, B.F., Spencer, C., Mabury, S.A., Muir, D.C.G., 2006. Poly and perfluorinated carboxylates in north American precipitation. Environ. Sci. Technol. 40, 7167–7174. Smith, J.N., Jones, E.P., Moran, S.B., Smethie, W.M., Kieser, W.E., 2005. Iodine 129/CFC 11 transit times for Denmark Strait Overflow Water in the Labrador and Irminger Seas. J. Geophys. Res. 110, C05006. Steiner, N., Denman, K., McFarlane, N., Solheim, L., 2006. Simulating the coupling between atmosphere-ocean processes and the planktonic ecosystem during SERIES. Deep-Sea Res. Part II-Top. Stud. Oceanogr. 53, 2434–2454. Taniyasu, S., Kannan, K., So, M.K., Gulkowska, A., Sinclair, E., Okazawa, T., Yamashita, N., 2005. Analysis of fluorotelomer alcohols, fluorotelorner acids, and short- and long-chain perfluorinated acids in water and biota. J. Chromatogr. A 1093, 89–97.

1255

Taniyasu, S., Rostkowski, P., So, M.K., Miyake, Y., Lam, P.K.S., Falandysz, J., Yamashita, N., 2006. Ultra trace analysis of persistent perfluorinated chemicals in the wet deposition matrixes. In: Proceeding of the 232nd ACS Meeting, San Francisco, CA, pp. 840–843. Tao, L., Kannan, K., Kajiwara, N., Costa, M.M., Fillmann, G., Takahashi, S., Tanabe, S., 2006. Perfluorooctanesulfonate and related fluorochemicals in albatrosses, elephant seals, penguins, and Polar Skuas from the Southern Ocean. Environ. Sci. Technol. 40, 7642–7648. USEPA, 2002. Revised draft Hazard assessment of perfluorooctanoic acid and its salts. Wallington, T.J., Hurley, M.D., Xia, J., Wuebbles, D.J., Sillman, S., Ito, A., Penner, J.E., Ellis, D.A., Martin, J., Mabury, S.A., Nielsen, O.J., Andersen, M.P.S., 2006. Formation of C7F15COOH (PFOA) and other perfluorocarboxylic acids during the atmospheric oxidation of 8:2 fluorotelomer alcohol. Environ. Sci. Technol. 40, 924–930. Wei, S., Chen, L., Yamashita, N., Taniyasu, S., Gamo, T., Du, J., So, M. K., Yeung, W.Y., Lam, P.K.S., in press. The latitudinal distribution of perfluorinated acids in surface waters from coastal areas of Asia to Antarctica. Mar. Pollut. Bull. Yamashita, N., Kannan, K., Taniyasu, S., Horii, Y., Okazawa, T., Petrick, G., Gamo, T., 2004. Analysis of perfluorinated acids at partsper-quadrillion levels in seawater using liquid chromatography-tandem mass spectrometry. Environ. Sci. Technol. 38, 5522–5528. Yamashita, N., Kannan, K., Taniyasu, S., Horii, Y., Petrick, G., Gamo, T., 2005. A global survey of perfluorinated acids in oceans. Mar. Pollut. Bull. 51, 658–668.