Tracing of water masses using a multi isotope approach in the southern Indian Ocean

Earth and Planetary Science Letters 302 (2011) 14–26

Contents lists available at ScienceDirect

Earth and Planetary Science Letters j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e p s l

Tracing of water masses using a multi isotope approach in the southern Indian Ocean P.P. Povinec a,⁎,1, R. Breier a, L. Coppola b,2, M. Groening c, C. Jeandel b, A.J.T. Jull d, W.E. Kieser e,3, S.-H. Lee f,1, L. Liong Wee Kwong g, U. Morgenstern h, Y.-H. Park i, Z. Top j a

Comenius University, Faculty of Mathematics, Physics and Informatics, Mlynska dolina F-1, SK-84248 Bratislava, Slovakia CNRS/CNES/IRD/Universite de Toulouse, Laboratoire d'Etudes en Geophysique et Oceanographie Spatiales, Toulouse, France International Atomic Energy Agency, Isotope Hydrology Laboratory, Vienna, Austria d University of Arizona, Departments of Physics and Geosciences, Tucson, AZ 85712-1201, USA e University of Toronto, IsoTrace Laboratory, Toronto, M5S 1A7, Canada f Korea Research Institute of Standards and Science, Daejeon, Republic of Korea g International Atomic Energy Agency, Marine Environment Laboratories, MC-98000 Monaco, h Institute of Geological and Nuclear Sciences, Lower Hutt, New Zealand i Muséum National d'Histoire Naturelle, LOCEAN/DMPA, F-75231, Paris France j University of Miami, Rosenstiel School of Marine and Atmospheric Sciences, Miami, USA b c

a r t i c l e

i n f o

Article history: Received 26 July 2010 Received in revised form 13 November 2010 Accepted 16 November 2010 Available online 22 December 2010 Editor: P. DeMenocal Keywords: tritium deuterium carbon-14 oxygen-18 iodine-129 seawater ANTARES IV Crozet Basin Indian Ocean

a b s t r a c t Anthropogenic radionuclides (3 H, 14 C, and 129I) stemmed from nuclear weapons tests were found in 1999 to be very abundant in the surface of the southern Indian Ocean, comparable to those in the subtropical Northwest Pacific Ocean. The observed radionuclide variations with latitude/longitude in the southern Indian Ocean are not due to deposition patterns of global fallout, but due to transport of water masses from the western Pacific through the Indonesian seas, and different water fronts present in the Crozet Basin of the Indian Ocean. High radionuclide concentrations observed in the latitudinal belt of 20-40°S are associated with the Indian Ocean Subtropical Gyre which acts as a reservoir of radionuclides, maintaining their high concentrations on a time scale of several decades. 14 C data documents that the southern Indian Ocean is an important for sink of anthropogenic carbon. The isotopic tracers reveal the evidence of the most intense surface gradients and presence of several water masses in the southern Indian Ocean, which makes the region one of the most dynamic places of the World Ocean. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Global fallout radionuclides (e.g. tritium (3 H), radiocarbon (14 C), strontium-90 (90Sr), cesium-137 (137Cs), iodine-129 (129I), americium241 (241Am), plutonium isotopes (238Pu, 239,240Pu ), etc.,) have been found as useful tracers for studying the heat and material transport and exchange processes occurring naturally both in the terrestrial (e.g. Hou et al., 2009; Levin and Hesshaimer, 2000; Santschi and Schwehr, 2004) and marine environments (e.g. Livingston and Povinec, 2002; Schlosser et al., 1999). Concentrations of these radionuclides in the sea had risen

⁎ Corresponding author. Tel.: + 421 260 295 544; Fax: 421 265 425 882. E-mail address: [email protected] (P.P. Povinec). 1 Formerly at the International Atomic Energy Agency, Marine Environment Laboratories, Monaco. 2 Present address: Observatoire Oceanologique de Villefranche-sur-mer, La Darse BP 08, 06238 Villefranche-sur-mer, France. 3 Present address: University of Ottawa, Ottawa K1N 6 N5, Canada. 0012-821X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.epsl.2010.11.026

since 1945 and peaked in the Northern Hemisphere in 1963, after large scale atmospheric nuclear weapons tests carried out in 1961-1962 by former Soviet Union at Novaya Zemlya (Livingston and Povinec, 2002). In the equatorial Pacific close in fallout from nuclear weapons tests carried out at Bikini and Enewetak Atolls contributed to radionuclide inventories in the Pacific Ocean as well. The major portion of global fallout deposited in the mid-latitudes of the Northern Hemisphere (UNSCEAR, 2000), in particular, the North-western Pacific due to the combined effect of higher precipitation and higher stratospheretroposphere exchange of air (Aoyama et al., 2006). Some of the global fallout radionuclides (e.g. 3 H, 14 C, 90Sr, 137Cs, 129I) are dissolved in seawater and becomes constituents of seawater, and suitable therefore for studying transport of water masses in the ocean. On the other hand Am and Pu isotopes are more particle reactive (La Rosa et al., 2005), and suitable for investigation of processes in the water column and sediments. Significant portions of these radionuclides in the world ocean have accumulated at the seafloor as bottom sediments (Bowen et al., 1980; Hong et al., 1999; Lee et al., 2005; Livingston et al., 2001).

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

We shall focus in this paper on three radionuclides, 3 H, 14 C and I. Tritium is directly incorporated into the water molecule, usually as HTO, and has suitable half-life (12.32 y), therefore, it is used extensively in oceanographic studies. It is produced both naturally by interactions of cosmic rays with nitrogen and oxygen in the upper atmosphere, and artificially in large amounts from atmospheric nuclear weapons tests. Its concentration peaked in the atmospheric moisture of the Northern Hemisphere in 1963, when it was 1000 times higher than its cosmogenic concentration of 60 TU (Weiss and Roether, 1980). It has also been released in large quantities from nuclear reprocessing facilities (Livingston and Povinec, 2000). The penetration of bomb tritium from surface waters into deeper layers of the ocean has been used to study pathways and time scales of deep and bottom water formation, (e.g. Bayer and Schlosser, 1991; Broecker and Peng, 1982; Broecker et al., 1986). Tritium with its strong interhemispheric concentration gradient (compared e.g. with radiocarbon) is a unique tracer for studying exchange of water masses between the basins. Natural 14 C levels in the environment have also been disturbed by bomb 14 C. In 1963 they were in the carbon dioxide of the Northern Hemisphere by a factor of 2 higher than natural levels (Burchuladze et al., 1989; Nydal and Lövseth, 1965). Because of its long half-life (5730 y) and subsequent formation of 14CO2 in the air, specific stratosphere-troposphere-biosphere mixing, exchange of carbon dioxide with surface ocean, and sequestration of carbon dioxide into the deep ocean, 14 C became the most frequently studied environmental radionuclide, important for better understanding of climate change (e.g. Key et al., 2004). Oceanic radiocarbon data contributed to a better understanding of thermohaline circulation in the World Ocean known as a Great Ocean Conveyor Belt (Broecker, 1991; Rahmstorf, 2006). 129 I has been introduced to the World Ocean from nuclear reprocessing facilities, global fallout and due to its natural production (Raisbeck and You, 1999). Because of its pulsed input from global fallout, releases by reprocessing plants in northwestern Europe, and long-half-life (15.7 million years) it is a powerful tracer to delineate source waters in the North Atlantic (Edmonds et al., 2001). Thanks to recent developments in mass spectrometric analysis (3He ingrowth from tritium decay; Clarke et al., 1976; Schlosser et al., 1999; Top, 1999), and accelerator mass spectrometry for 14 C (e.g. Jull et al., 2008; Key et al., 1996;2002; Povinec, 2005; Povinec et al., 2008; Tuniz et al., 1998) and 129I (e.g. Kilius et al., 1987; Povinec et al., 2000; Raisbeck and You, 1999) it has been possible to analyze these radionuclides with high sensitivity and precision in small seawater volumes (e.g. 0.5 L), and thus investigate their distribution in the water column using direct sampling with Rosette systems. Also, multinational oceanographic sampling program has been emerged to cover the synoptic view of the global ocean. A notable one is the WOCE (World Ocean Circulation Experiment) program conducted in the 1990 s, which represents the most extensive 3 H and 14 C project carried out in the World Ocean (Key, 1996; www.eWOCE.org). Although global fallout radionuclide distribution datasets have been frequently used as tools for tracing water masses in the World Ocean (e.g. Livingston and Povinec, 2002; Schlosser et al., 1999;2001), the Indian Ocean has received only a limited attention (Lee et al., 2009; Povinec et al., 2004a; van Beek et al., 2008). The first measurements of 3 H and 14 C were made through the Geochemical Ocean Sections (GEOSECS) project (1977-1978) (Broecker et al., 1986; Östlund and Brescher, 1982; Stuiver and Ostlund, 1983), and followed by the WOCE field project (1990-1998; www.eWOCE.org) in the Indian Ocean. Unfortunately, the 3 H and 14 C measurements were not carried out in the western Indian Ocean south of 34°S. The Indian sector of the Southern Ocean is a key region for the exchange of water masses between Antarctica and Equatorial regions, playing an important role in the global climate change (Key et al., 2004). The Southern Ocean is the largest oceanic high-nitrate low 129

15

chlorophyll region in the world. It is contributing to the regulation of the atmospheric CO2 via the biological pump (Metzl et al., 1999). The southern Indian Ocean was included therefore in the Worldwide Marine Radioactivity Studies (WOMARS), coordinated by the International Atomic Energy Agency's Marine Environment Laboratories (IAEA-MEL) in Monaco, and carried out in collaboration with several laboratories in Denmark, France, Germany, India, Italy, Japan, South Korea, New Zealand, Sweden, UK and USA (International Atomic Energy, 2005; Povinec et al., 2003a; Povinec et al., 2005). The aim of the project was to study the distribution and behavior of anthropogenic radionuclides in the world ocean. Some of the results obtained for the northern Indian Ocean have already been published (Bhushan et al., 2003; Mulsow et al., 2003; Povinec et al., 2003b). French ANTArctic RESearch (ANTARES) IV cruise in the southern Indian Ocean in 1999 was liaised with the IAEA WOMARS program. The cruise plan (Park et al., 2002) and strategy for sampling radionuclides (Coppola et al., 2005, 2006; Lee et al., 2009) was designed with the aim to study distribution of radioactive and stable isotopes in the Crozet Basin of the southern Indian Ocean characterized with strong ocean currents. In this paper, we report the distribution of radioactive (3 H, 14 C and 129I) and stable (2 H, 18O) isotopes in surface and deep waters of the southern Indian Ocean. Results on 90Sr 239,240Pu and 241Am in surface waters and plankton have been published earlier (Lee et al., 2009). 2. Oceanography background The Indian Ocean is limited northward to 25°N by the continent, and at 60°S, where it becomes the Southern Ocean. The 60oS limit appears to be largely set by political grounds. An important feature of water circulation in the Indian Ocean is a transport of warm water masses (10 - 15 Sv) from the western Pacific Ocean via the Indonesian throughflow to the eastern Equatorial Indian Ocean and to the southern Indian Ocean (Fine, 1985; Gordon and Fine, 1996; Gordon et al., 2003). The south Indian Ocean Subtropical Gyre (IOSG) is the most important current system influencing water circulation between 20° and 40°S (Tomczak and Godfrey, 1994). The gyre is the most intense with a tight, double celled central core focused at the western boundary. A fraction of IOSG water is transported southward by the Agulhas Current (AC) along the eastern African coast, entering the South Atlantic around the Cape of Good Hope, forming a branch of the AC that does not complete the retroflection pattern, also called the Agulhas leakage (Gordon, 1985; Schmitz, 1995). Along the eastern boundary of the Indian Ocean, off the western Australia, the Leeuwin Current flows poleward along the continental shelf break from about 22°S to 35°S, and then turns eastward. The Leeuwin Current is warm and of relatively low salinity, low dissolved oxygen and high phosphate content. It transfers a significant amount of heat to the south (Tomczak and Godfrey, 1994). South of the 40oS, the wind-driven Antarctic Circumpolar Current (ACC), going around Antarctica, transports cold waters from the Atlantic Sector of the Southern Ocean via the southern Indian Ocean to the southern Pacific Ocean. The southern Indian Ocean plays therefore a key role in the exchange of water masses between the Equator and Antarctica (Ganachaud and Wunsch, 2000), and between the Southern Hemisphere basins (Ridgway and Dunn, 2007), important for better understanding of global oceanic processes and the climate (Key et al., 2004; Rahmstorf, 2002). As it is surrounded by highly populated continents in its Northern area, there is also subjected to contamination from land-based sources. The banded structure of the ACC in the southern Indian Ocean consists of several narrow jets associated with sharp hydrographic fronts due to the presence of the Crozet and Kerguelen Plateaus (Park et al., 1993) (Fig. 1). From the north of the Crozet Islands to the downstream area, there exists a very strong triple frontal zone where

16

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

Fig. 1. Sampling stations in the southern Indian Ocean during the ANTARES IV expedition. The main hydrographic fronts in the region: Agulhas Front (AF), Subtropical Front (STF), Subantarctic Front (SAF) and Polar Front (PF) are also shown (flowing from the west to the east). While at Sts. 1, 2, 4 ,5 and 6 only surface water was collected, at Sts. 3, 7 and 8 full water column was sampled.

the Agulhas Front (AF), Subtropical Front (STF) and Subantarctic Front (SAF) converge into a narrow band of 100 to 200 km wide, with sharp across-front changes in temperature, salinity and oxygen content (Belkin and Gordon, 1996; Park et al., 2002). Strong ocean currents formed in this area characterizes the southern Indian Ocean as the most dynamic region of the World Ocean. The dominant current system affecting the circulation in the Crozet Basin is the AF, characterized by warm and saline subtropical waters carried by the Agulhas Return Current (ARC). Extending eastward into the basin, up to 60°E, the AF re-circulates to the north, as a part of the southern limb of the anticyclonic IOSG (Park et al., 2002). The STF forms a boundary between the subtropical surface water and cooler, fresher subantarctic surface water (Pollard et al., 2002). It represents the northernmost frontal jet that passes through Drake Passage and is generally regarded as circumpolar in extent. The SAF is located at the northern boundary of the Polar Frontal Zone (PFZ), a transition region between the SAF and the Polar Front (PF). North of the SAF, there is a subsurface salinity minimum, associated with the subduction of Antarctic Intermediate Water (AAIW) (Park et al., 1993; Pollard et al., 2007). South of the SAF, the lowest salinity water is in the surface layer. Temperature dominates the stratification north of the SAF (Park et al., 1993). The PF is not a part of the ACC main core, but is very close to the Kerguelen Plateau (Fig. 1). The distribution of tracers in the southern Indian Ocean is therefore expected to be controlled by the banded structure of the fronts. It has been therefore a great challenge to collect water samples in such a complex and dynamic region, and to use isotopic tracers (3 H, 14 C, 129I, and stable 2 H and 18O) to investigate movement of water masses in the frontal zones. The oceanographic cruise plan (Park et al., 2002) was designed within the framework of the French Southern Ocean Joint Global Ocean Flux Study (SO-JGOFS) with the main objective to quantify the stocks and the export of biogenic particles in relation to the biological pump of atmospheric CO2 in the Indian sector of the Southern Ocean. The sampling area, characterized with strong physical gradients, was located north of the Crozet Basin. This zone was chosen to study biogeochemical processes in a mesoscale circulation pattern (Coppola et al., 2006; Lee et al., 2009; Park et al., 2002). Radionuclide sampling was carried out in subtropical waters inside the IOSG (Sts. 1, 2, 5 and

8), north of STF (St. 7), and south of SAF, (Sts. 3, 4 and 6) with the aim to study distribution of radionuclides within the water fronts. 3. Samples and methods 3.1. Water sampling Ocean observation and water sampling (January-February 1999) was carried out on board of the R/V Marion Dufresne (CNRS) as a part of the ANTArctic RESearch (ANTARES) IV cruise (Coppola et al., 2004;2006; Park et al., 2002) in the offshore of the northwest of Kerguelen Islands and east of Crozet Islands between 32°- 48°S and 51°- 70°E, in the confluence zone of the AF (Sts. 1, 2, 5 and 8), STF (St. 7), SAF (Sts. 3 and 4) and close to the PF (St. 6) (Fig. 1). The observation on the water temperature, salinity and dissolved oxygen was reported in Coppola et al. (2006). Surface samples for radionuclide analyses were collected about 4 m below the sea level using a water pump. Water column samples (Sts. 3, 7 and 8) were collected at different depths using a Rosette sampling system equipped with 12 L Niskin bottles. One litter water samples were stored separately for 3 H (and stable isotopes), 14 C (poisoned for preventing any biological activity by adding mercury chloride) and 129I analyses in well closed glass bottles. The bottles had double plugs to prevent a penetration of air into the water sample during storage, which was checked by storage and analysis of IAEA Reference Materials, and no differences with time were observed. Temperature and conductivity/salinity were measured using a commercial CTD mounted on the Rosette sampling system. 3.2. Isotope analyses 3 H was measured either by the 3He ingrowth method (Clarke et al., 1976; Top, 1999) at the University of Miami (USA) or by liquid scintillation spectrometry (after an electrolytic enrichment) at the Institute of Geological and Nuclear Sciences (Lower Hutt, New Zealand; Morgenstern and Taylor, 2009), and at the Isotope Hydrology Laboratory of the International Atomic Energy Agency (Vienna, Austria; Groening et al., 2009). 3 H concentration is given in Tritium Unit (1 TU is the isotopic ratio of one 3 H atom to 1018 protium (1 H) atoms, equivalent to 118 mBq/L of water). The 3He ingrowth method can measure 3 H levels down to 0.01 TU (precision at 1 σ is around 0.005

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

17

TU); the liquid scintillation spectrometry after double electrolytic enrichment and special calibration can reach 0.02 TU (Morgenstern and Taylor, 2009; Povinec, 2004). δ2Η analyses were done using H2O-Zn reduction (Coleman et al., 1982). δ18Ο analyses were performed using the CO2-H2O equilibration procedure described in Epstein and Mayeda (1953). The stable isotopic compositions of hydrogen and oxygen are reported as "delta" (δ) values in parts per thousand (denoted as ‰) enrichments or depletions relative to a standard of known composition
 3 δ ¼ Rsample =Rstandard 1 10 : The isotopic results are reported against the international standard VSMOW (Vienna Standard Mean Ocean Water) as defined by Gonfiantini (1978). Typical uncertainties at 1 σ are ± 1 ‰ for δ2H, and ± 0.1 ‰ for δ18O. Analyses of H and O stable isotopes of seawater were carried out at the Lower Hutt and Vienna laboratories. 14 C was analyzed in dissolved inorganic carbon present in seawater using AMS. The samples were prepared either in the University of Arizona or in IAEA-MEL following the procedure described by Donahue et al. (1990) and Liong Wee Kwong et al. (2004). The CO2 sample extracted from seawater was converted into graphite, which was pressed into a sample target holder, and loaded into the AMS machine ion source. The graphite sample was bombarded with Cs+ ions under vacuum, and the sputtered C- ions were accelerated and analyzed for isotopic composition (Jull et al., 2008). 14 C concentration is given as Δ14C in ‰, relative to the NIST (National Institute of Standards and Technology, Gaithersburg, USA) 14 C oxalic acid standard (Povinec et al., 2004b; Stuiver and Ostlund, 1983) 14

Δ C = ðFm −1Þ10

3

where Fm (a fraction of modern carbon) is the measured AMS 14 C/13 C ratio in a sample normalised to δ13CPBD = -25 ‰, as defined by Donahue et al. (1990). The precision of AMS measurements was around ± 5 ‰. The AMS analyses were carried out at the University of Arizona (Jull et al., 2008). Iodine samples were prepared either in the IsoTrace Laboratory or in IAEA-MEL following the procedure described by Povinec et al. (2000). NaI carrier (2-10 mg) was added to the sample, and after several steps AgI was obtained which was used as a target for AMS measurements carried out in the IsoTrace Laboratory. The AMS measurements were normalised with respect to ISOT-2 reference material (129I/127I=(1.174±0.022) 10-11). 129 I results are given in atom/L. The relative precision of AMS measurements was around±10 %. IAEA reference materials (Irish Sea water (IAEA-381, Povinec et al., 2002; Mediterranean Sea water, IAEA-418, Pham et al., 2010) were analysed simultaneously with collected samples to ensure high quality of results. This is the first time that such a suite of radioactive (3 H, 14 C and 129I) and stable (2 H and 18O) isotope tracers have been used in an oceanographic study. 4. Results 4.1. Radionuclides Diagrams of potential temperature versus salinity, and potential temperature versus dissolved oxygen (Fig. 2) clearly show the presence of different water masses in the sampling sites of the Crozet Basin (Fig. 1). Several water masses can be identified: saline North Atlantic Deep Water (NADW) was observed at all water profile stations (3, 7 and 8). Antarctic Bottom Water (AABW) occupies the bottom layers, below NADW. Antarctic Intermediate Water (AAIW) situated above the NADW may be present at stations 7 and 8. Surface

Fig. 2. Diagrams of T-S and T-O2 for Stations 3, 7 and 8 in the southern Indian Ocean (sampling sites as in Fig. 1). Water masses: STSW - Subtropical Surface Water; SASW – Subantarctic Surface Water; AASW – Antarctic Surface Water; AABW – Antarctic Bottom Water; AAIW – Antarctic intermediate water; NADW – North Atlantic Deep Water; NIDW - North Indian Deep Water.

waters are represented either by Subtropical Surface Water (STSW) (St. 7 and 8), or by SASW (St. 3). Further, the potential temperature versus dissolved oxygen plot revealed another water mass of North Indian Deep Water (NIDW), whose oxygen concentration is around 4 ml/L (Park et al., 1993). The observed 3 H, 14 C and 129I water profiles (Fig. 3, Table 1) show typical features – the concentrations were highest at the surface (or subsurface at 100-200 m), then gradually decreasing with depth with sharp decreases at 1000-2000 m water depths, and relatively invariable with depth in deep and bottom waters. 3 H concentrations (Fig. 3a) were below detection limit (0.02 TU) between 1000 and 2500 m at Sts. 7 and 3, however, levels comparable to surface were observed at bottom waters at both stations. In contrast to this, St. 8 shows by a factor of three higher 3 H levels for the surface waters. The 14 C profiles (Fig. 3b) also show a downward transport of bomb produced 14 C to the depth of about 2000 m. The 129I water profiles shown in Fig. 3c differ from the tritium ones due to a larger variability observed both at surface and bottom waters. We calculated the standing stocks of 3 H and 129I which best represents global fallout radionuclides due to their direct input to the

18

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

sea. The standing stocks of 3 H in the upper 1 km3 of the water column in the IOSGW (St. 8) were 140 ± 10 GBq, much higher than in STSW of St. 7 (40 ± 5 GBq) and in SASW of St. 3 (24 ± 4 GBq). In the case of 129I the standing stocks were much smaller, and the differences between the stations were within the uncertainties: 11 ± 2 kBq for St. 8, then 9 ± 2 kBq for St. 7 and 8 ± 2 kBq for St. 3. 3 H, 14 C, and 129I concentrations observed in the western North Pacific in 1997 during the IAEA´97 expedition (Povinec et al., 2003a; Povinec et al., 2010), and in the East Sea (the Sea of Japan) (Cooper et al., 2001) are comparable with the ANTARES IV data in the southern Indian Ocean (Sts. 2, 5 and 8, Fig. 3). The Pacific and Indian Ocean profiles show similar features – a surface (or subsurface at 100-200 m) maxima, a sharp decrease toward 1000 m, and negligible levels in deep and bottom waters. 4.2. Stable isotopes In addition to physical oceanography parameters and radionuclides, also stable isotopes of water have been included in our analyses with the aim of better understanding the characteristics of the various water fronts present in the Crozet Basin. The subtropical waters inside the IOSG (Sts. 8, 1, 2, 5) appeared to be enriched with 18 O, with high salinity in the surface and gradual depletion of 18O with gradually freshening toward the depth. SASW (St.3) were most depleted with 18O and the most fresh among the water masses present (Fig. 4a). Coefficient of determination, R2 = 0. 86 (P value b0.0001), indicates that there is a strong correlation between δ18O and salinity. While surface waters collected at Sts. 2, 5 and 8, located close to the main stream of the IOSG, were enriched in deuterium and oxygen, representing warmer and saltier subtropical waters, the southern St. 7 (STF), and especially Sts. 3, 4 and 6 were depleted in both isotopes (Table 1), representing thus fresher, cooler and less saline PFZ water masses (Fig. 4b). All experimental data are well below the Global Meteoric Water Line (GMWL) defined by Craig (1961), documenting heavy depletion of both isotopes in seawater. In the subtropical ocean the evaporation dominates over precipitation, while at higher latitudes the precipitation becomes higher than the evaporation (LeGrande and Schmidt, 2006). This effect is clearly visible in the δ2H vs. δ18O diagram presented in Fig. 4b which shows a typical correlation between these two isotopes. Tritium versus δ18O, and 129I vs. dissolved oxygen diagrams presented in Fig. 5 also revealed presence of different water masses in the Crozet Basin. Elevated 3 H and 129I concentrations were observed at St. 8 in surface, medium depth and bottom waters, while at Sts. 3 and 7 only bottom waters showed higher concentrations. 5. Discussion 5.1. Radionuclide profiles

Fig. 3. 3H (a), 14 C (b) and 129I (c) water profiles at Sts. 3. 7 and 8 compared with nearest GEOSECS (Östlund and Brescher, 1982; Stuiver and Östlund, 1983) and WOCE (www. eWOCE.org) stations. 3 H, 14 C and 129I obtained for the NW Pacific (IAEA´97 cruise; Povinec et al., 2004b; Povinec et al., 2010) are also shown. 3 H data were decay corrected to January 1999 with half-life of 12.32 y. 3 H concentration is given in Tritium Unit (1 TU is the isotopic ratio of 1 3 H atom to1018 protium (1 H) atoms, equivalent to 118 mBq/L of water). 14 C concentration is given as Δ14C in ‰, relative to the NIST 14 C standard (Stuiver and Östlund, 1983). Relative precisions (at 1 σ) are ~ 2 % for 3 H, ~ 0.5 % for 14 C and ~ 10 % for 129I. For positions of sampling stations see Fig. 1.

Following the global atmospheric deposition, higher surface radionuclide concentrations are expected for the Northern Hemisphere than in the Southern Hemisphere. The peak value in the Southern Hemisphere at 50°S observed in 1965-1967 was only 4 % of that observed in 1963 for the same Northern Hemisphere latitude, with decreasing values towards both the Equator and the South Pole (Weiss and Roether, 1980). In 1998 the average 3 H value observed in the northern Indian Ocean at 20°N was 1.0 TU (Mulsow et al., 2003; Povinec et al., 2003b). Following the Weiss and Roether (1980) estimation of the development of the mean tritium concentration in the open ocean with time, the expected value in 1998-1999 for the 40°S latitude would be 60% of that observed for 20°N, it means only 0.6. It is clear that the observed 3 H levels, combined with WOCE data (decay corrected to 1999), do not follow this pattern (Fig. 6a). As there is not an additional source of 3 H in the Indian Ocean (Livingston

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

19

Table 1 Oceanographic and isotope data of collected seawater samples in Crozet Basin (January-February 1999). Station

Position

1 2 4 5 6 3

32°42.19'S 40°17.98'S 45°51.32'S 40°00.00'S 48°00.00'S 46°00.03'S

70°00.00'E 70°00.00'E 55°00.00'E 51°57.27'E 68°38.50'E 63°03.58'E

7

44°11.45'S 63°24.45'E

8

43°08.03'S 62°31.42'E

Depth

Temperature

(m)

(°C)

4 4 4 4 4 6 50 100 150 300 500 1000 1500 2000 2500 3500 4300 6 50 100 150 300 500 1000 1500 2500 3500 4300 6 50 100 200 500 1000 1500 2000 2500 3000 4000 5000

19.05 17.01 9.86 19.55 9.90 9.53 9.54 6.01 5.22 5.00 3.93 2.95 2.48 2.24 1.84 0.85 0.47 14.30 13.53 12.37 11.49 10.50 8.38 3.73 2.82 2.14 1.22 0.49 17.40 17.53 15.69 14.77 13.14 7.34 3.55 2.72 2.44 2.09 1.12 0.44

* Uncertainty at 1 σ is ± 1 ‰ for δ2H, ± 0.1 ‰ for δ18O and 5 ‰ for

Salinity

35.619 35.177 33.684 35.533 33.640 33.6981 33.7001 33.8346 33.8800 34.2135 34.2484 34.4709 34.6666 34.7527 34.7621 34.7080 34.6857 34.3431 34.3513 34.9364 34.8774 34.7880 34.5982 34.3584 34.5739 34.7600 34.7260 34.6850 35.5319 35.5244 35.4892 35.4557 35.2852 34.5270 34.4122 34.6225 34.7330 34.7612 34.7184 34.6786 14

δ2H*

δ18O*

3

(‰)

(‰)

(TU)

(‰)

(108 atom/L)

0.3 0.3 -5.8 0.2 -6.1 -5.2 -5.7 -2.4 -2.9 -3.1 -4.0 -4.5 -4.9 -4.4 -4.6 -2.9 -2.7 -3.3 -3.8 -2.7 -3.3 -4.4 -3.8 -4.5 -4.8 -4.9 -2.5 2.7 -0.1 -1.2 1.2 -1.3 -2.0 -3.0 -3.3 -2.9 -2.7 -2.6

0.57 0.34 -0.49 0.47 -0.46 -0.37 -0.57 -0.37 -0.30 -0.26 -0.21 -0.25 -0.24 -0.26 -0.20 -0.17 -0.19 -0.09 0.08 0.20 0.26 -0.09 -0.12 -0.20 -0.27 -0.20 -0.15 0.56 0.33 0.20 0.22 0.10 -0.18 -0.20 -0.27 -0.20 -0.17 -0.15

0.72 ± 0.03 0.93 ± 0.03 0.32 ± 0.02 1.10 ± 0.03 0.18 ± 0.02 0.242 ± 0.017 0.197 ± 0.019 0.279 ± 0.023 0.227 ± 0.022 0.241 ± 0.020 0.178 ± 0.010 b 0.02 b 0.02 b 0.02 b 0.02 0.289 ± 0.022 0.302 ± 0.024 0.399 ± 0.024 0.433 ± 0.022 0.534 ± 0.025 0.524 ± 0.022 0.493 ± 0.023 0.158 ± 0.010 b 0.02± b 0.02± b 0.02± 0.302 ± 0.024 1.23 ± 0.03 1.25 ± 0.03 1.42 ± 0.03 1.35 ± 0.03 1.29 ± 0.02 0.75 ± 0.02 0.64 ± 0.02 0.28 ± 0.01 0.54 ± 0.01 0.37 ± 0.03 0.39 ± 0.03

101.2 108.3 32.54 110.5 23.52 26.40 43.60 31.30 -26.80 -60.20 -100.00 -147.30 -145.20 .150.30 -162.00 62.00 51.30 25.10 -26.70 -149.30 -150.90 -144.90 -150.16 -156.40 101.20 98.50 10.70 -100.300 -125.000 -150.100 -162.200 -

0.066 ± 0.009 0.058 ± 0.008 0.062 ± 0.009 0.105 ± 0.010 0.061 ± 0.009 0.060 ± 0.006 0.013 ± 0.008 0.058 ± 0.008 0.053 ± 0.007 0.033 ± 0.006 0.062 ± 0.008 0.050 ± 0.005 0.060 ± 0.007 0.077 ± 0.009 0.080 ± 0.009 0.114 ± 0.009 0.059 ± 0.007 0.058 ± 0.008 0.026 ± 0.006 0.042 ± 0.009 0.046 ± 0.012 0.039 ± 0.009 0.042 ± 0.022 0.079 ± 0.012 0.082 ± 0.008 0.0790.008 0.152 ± 0.014 0.055 ± 0.007 0.033 ± 0.005 0.075 ± 0.008 0.045 ± 0.006 0.077 ± 0.008 0.081 ± 0.009 0.034 ± 0.008 0.055 ± 0.006

H

14

C*

129

I

C.

and Povinec; Povinec et al., 2003b), high surface 3 H concentrations observed at Sts. 2, 5 and 8 (1.1 TU for the latitude belt 40°S – 43°S, Table 1) should be associated with a radionuclide transport from the western Pacific Ocean, where high 3 H concentrations were observed (Povinec et al., 2010). The average 3 H concentration observed in the top 200 m at St. 8 is 1.3 TU, what is close to the value (1.6 TU, decay corrected 1999) observed in 1973 at GEOSECS stations 228 and 229 in the NW Pacific (Östlund and Brescher, 1982). 3 H levels in subsurface maxima in the NW Pacific, sampled in 1997 (Povinec et al., 2010), were between 1.3 and 1.4 TU (Fig. 3a), again in very good agreement with data measured for St. 8. Some 107 m3/s of seawater flows from the western Pacific Ocean via the Indonesian throughflow into the Indian Ocean (Gordon et al., 2003). About 50% of it is transported to the Crozet Basin, and the rest leaks to the South Atlantic (Speich et al., 2007). Using a simple mass balance approach we may estimate that the time needed to transport waters from the western Pacific to the Crozet Basin is around 13 years, comparable with the 3 H half-life. A similar value has been obtained from direct float measurements (Davis, 2005). The float data (Davis, 2005) also indicate that about 10 years is needed to complete one loop in the IOSG (Fig. 1). This would mean that 3 H decayed in the IOSG waters is renewed by “fresh” 3 H water coming from the western Pacific Ocean which acts as a tritium source for the Indian Ocean due

to higher fallout levels found in the western North Pacific Ocean (Aoyama et al., 2006; Povinec et al., 2010). In general, the bulk of 3 H in the water column (Fig. 3a) should exist near surface due to the atmospheric input. Assuming no specific sources for 3 H in the study area, the existence of higher 3 H concentrations in the bottom layer could be related to an intrusion of surface water. In the study area, however, there is no information about sinking of surface water to the bottom due to strong thermohaline stratification. You (2000) reported that only a small net transport of 0.5 Sv occurred across the lower intermediate layer downward. The great difference (10 times more) in the water transport across the upper layer than across the lower layer gives a strong implication for the advection of AABW to the region with elevated 3 H levels, as discussed later. Radiocarbon water profiles presented in Fig. 3b show features expected for the southern Indian Ocean. Radiocarbon is behaving differently in the water column due to exchange processes with the atmosphere (Aramaki et al., 2001; Bard et al., 1988; Stuiver and Ostlund, 1983). The 14 C profiles shown in Fig. 3b reflect the different positions of water fronts, similarly as it was in the case of tritium. Peak 14 C levels observed at St. 8 are comparable with those observed in the NW Pacific (IAEA´97 Sts. 2 and 3; Povinec et al., 2004b), as well as at the close-by GEOSECS (Stuiver and Ostlund, 1983) and WOCE stations (Key et al., 2002).

20

0.4 0.2

1.6 St. 1

St. 1

St. 6

St. 2

St. 7

St. 3

St. 8

1.4

St. 2 St. 3

1.2

St. 4 St. 5

(TU)

δ18Ο (ο/οο)

a

0.6

-0.0

1.0 0.8

3H

a

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

0.6

St. 4

IOSGW

St. 5 NIDW

St. 6 St. 7 St. 8

-0.2

STSW

0.4 -0.4

0.2

AABW

SASW

AAIW NADW

-0.6 33.0

33.5

34.0

34.5

35.0

35.5

0.0 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 -0.0 0.1

36.0

7 5

St. 1

St. 6

St. 2

St. 7

St. 3

St. 8

St. 3

0.12

St. 4 St. 5

3 1 -1

-0.7

-0.5

-0.3

-0.1

0.1

0.3

δ18Ο (ο/οο) Fig. 4. δ18O vs. salinity (a) and δ2H vs. δ18O (b) plots for all stations visited. δ2H and δ18O are expressed in per mill relative to VSMOW (Vienna Standard Mean Ocean Water). Relative precisions (at 1 σ) are 1 ‰ for δ2H, and 0.1 ‰ for δ18O. Coefficient of determination, R2 = 0. 86 (P value b 0.0001), indicates that there is a strong correlation between δ18O and salinity. All experimental data are well below the Global Meteoric Water Line (GMWL) defined by Craig (1961), documenting heavy depletion of both isotopes in seawater. Sts 1, 2, 4, 5 and 6 are surface water samples, at Sts. 3, 7 and 8 full water column was sampled. For positions of sampling stations see Fig. 1.

0.5

0.6

200 STSW

St. 8

SASW

NIDW+NADW

0.08 0.06

0.00 3.5

0.5

IOSGW

St. 7

0.10

0.02

-5 -0.9

0.4

150

0.04

-3

-7

0.3

0.16 0.14

GMWL

(108 atom/L)

9

δ1Η (ο/οο)

b

11

129I

b

0.2

δ18Ο (ο/οο)

Salinity

100

50

150 100

1500 6 50 2500 3000 6 500 500 300 6 1500 2000 300 2000 500 3500 2500 AAIW 5000 1000 1500 4000 2500 4300 AABW 1000 1000 4300

4.0

4.5

5.0

5.5

6.0

50

100

6.5

7.0

Oxygen (mL/L) Fig. 5. 3H vs. δ18O (a) and 129I vs. dissolved oxygen (b) plots for collected samples showing different water masses in the region. IOSGW- Indian Ocean Subtropical Gyre Water, STSW – Subtropical Surface Water, SASW- Subantarctic Surface Water, AAIW – Antarctic Intermediate Water, NADW – North Atlantic Deep Water; NIDW – North Indian Deep Water, AABW – Antarctic Bottom Water. Water depths in meters are also shown in the plot (b). Sts 1, 2, 4, 5 and 6 are surface water samples, and at Sts. 3, 7 and 8 full water column was sampled. For positions of sampling stations see Fig. 1.

in the Tasman Sea are only of global fallout origin (unfortunately no I data are available), which have been accumulating in the western part of the Pacific Ocean Subtropica Gyre (Mittelstaedt et al., 1999; Nakano and Povinec, 2003). Therefore the 129I levels in the IOSGW should be lower than in NW Pacific waters, where they were influenced by radioactive waste discharges from the Tokai-mura reprocessing plant. 129

Iodine-129 water profiles shown in Fig. 3c have minima at 1000 m water depth at all ANTARES IV samples which may be associated with the presence of NADW carrying low 129I levels. An interesting feature is that 129I levels in the top 1000 m are by about a factor of two lower in the IOSGW (St. 8) than in the NW Pacific samples (Sts. 2 and 3 in Fig. 3c). This can be due to the fact that the Pacific stations were under the influence of the Kuroshio Current, which carried higher 129I levels released from the Tokai-mura reprocessing facility (Livingston and Povinec, 2000; Povinec et al., 2010), while the IOSG St. 8 show global fallout 129I concentrations. The differences between the 3 H and 129I levels observed in the IOSGW and in NW Pacific waters may indicate that a source of these radionuclides need not be only NW Pacific. Surface Tasman waters may flow northward along the Australian coast (Ridgway and Dunn, 2007), and supply IOSGW via the Indonesian Throughflow with elevated 3 H levels (around 1.2 TU) observed in the Tasman Sea (www.eWOCE.org). A similar connection has been observed between the SHOTS (Southern Hemisphere Ocean Tracer Study) 137Cs data collected along 20°S in the Indian Ocean (Povinec et al., submitted for publication), and along 30°S in the Pacific Ocean (Aoyama et al., submitted for publication). However, the observed 3 H concentrations

5.2. Spatial distribution of radionuclides 5.2.1. Tritium Surface water samples collected in 1998 as a part of the Indian Ocean transect (Povinec et al., 2003b) also showed higher 3 H levels around 30°S, comparable with those presented in Fig. 6a. High surface 3 H levels (0.6-0.7 TU, decay corrected to 1999) were also measured at 34°S, 57°- 62°E during the WOCE project. This is consistent with 3 H concentration observed at St. 1 (0.72 ± 0.03 TU at 32°42´S, 70°00´E; Table 1), located inside the loop formed by the IOSG. High surface 3 H levels (1.1-1.2 TU), observed at Sts. 2, 5 and 8 (located between 40° and 43°S), suggest that these stations were close to the main stream of the IOSG. It is evident that the gyre flow is strongest at 40-43°S, west of 65°E, where highest 3 H levels were observed. This is a clear

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

21

Fig. 6. Spatial distribution of 3 H in surface waters (a) and deep waters (along 60°E) (b) at Sts. 1- 8 (x-signs), combined with nearest WOCE stations (dots) (www.eWOCE.org), decay corrected to January 1999. Higher 3 H concentrations were observed at the main stream of the Indian Ocean Subtropical Gyre at ~ 40°S (as well as at ~ 20°S, as indicated by the WOCE data). Downward tritium transport around 40°S can be seen as well. Higher 3 H levels observed below 3000 m are associated with Antarctic Bottom Water (AABW).

influence of the IOSG on the downstream radionuclide transport from its western boundary associated with the Agulhas Retroflection (Ridgway and Dunn, 2007).

The atmospheric deposition of global fallout tritium in the Southern Hemisphere was highest during the late sixties in the latitude belt 4050°S (International Atomic Energy, 2005; Weiss and Roether, 1980).

22

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

However, 3 H water profiles across the basin (Fig. 6b), combined with WOCE data (www.eWOCE.org) south of 34°S (decay corrected to 1999), show higher surface and subsurface 3 H levels at 20°S and 40°S latitude

belts. This indicates an accumulation of tritium within the IOSG on a time scale of several decades. The gyre acts as a reservoir, maintaining higher radionuclide concentrations in the region. Simultaneously,

Fig. 7. Spatial distribution of 14 C in surface waters (a) and deep waters (along 60°E) (b) at Sts. 1- 8 (x-signs), combined with nearest WOCE stations (dots) (www.eWOCE.org). Higher 14 C concentrations were observed at the main stream of the Indian Ocean Subtropical Gyre at ~ 40°S (as well as at ~ 20°S, as indicated by the WOCE data). Downward 14 C transport around 40°S can be seen as well.

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

higher tritium levels can be seen in the transect around 40°S and 60°E at water depths below 3000 m which should be due the advection of AABW. 5.2.2. Radiocarbon Combining the 14 C data presented in this work together with WOCE data (www.eWOCE.org) we constructed a 14 C distribution map for the southern Indian Ocean (Fig. 7a). We can see high 14 C concentrations observed along the subtropical gyre (20 - 40°S) at the top 500 m, and sharp decrease south of 40°S, as confirmed by 3 H data as well (Figs. 3 and 6). 14 C water profile data (WOCE and ANTARES IV) presented in Fig. 7b show a clear penetration of bomb 14 C around 40°S (transect at 60°E) down to almost 5000 m (similarly as we could see it in the case of tritium, Fig 6b). Fig. 7 documents that the southern Indian Ocean is important for sink of anthropogenic carbon. The high phytoplankton biomass in the region of the Kerguelen Plateau and the Crozet Islands and Plateau was found to be fueled by the natural iron inputs from shallow topography close to islands (Pollard et al., 2009). 5.3. Water masses The isotopic observations (Fig. 3 and especially Figs. 5, 6 and 7) show the presence of different water masses in the Crozet Basin more clearly than the conventional water mass tracers of potential temperature and salinity/dissolved oxygen presented in Fig. 2. 3 H concentrations largely decrease below the 1000 m water depth at which CDW (Circumpolar Deep Water) appears, formed by the deepwater circulation and ventilation south of 40°S (Fine, 1993; You, 2000). The depletion of 3 H in the CDW in the STF and SAF fronts (Stations 3 and 7, water depth 1000 – 2500 m) may be related to the intrusion of NADW especially to LCDW (Lower CDW), characterized by salinity maximum (Fig. 2a). NADW injects its water into the CDW in the south Atlantic by mixing and entrainment, and then flows into the southwest Indian Ocean, south of Africa between 35°S and 40°S. One branch of NADW then moves to the Southwest Indian Ridge (Toole and Warren, 1993). Considering very low 3 H concentrations (5 to 24 mTU) observed in the South Atlantic bottom water (Jenkins et al., 1983), which is predominantly affected by the NADW, and taking into account the replenishment of the CDW by lateral transport of the NADW from the South Atlantic to the South Indian Ocean on a time scale of 4-10 years, the contribution of the NADW to the CDW may explain the observed 3 H minima. Except for the AF, STF and SAF, which control radionuclide distribution in surface and subsurface waters, several other water masses can be identified in the basin using isotopes (Fig. 5a, b). The intermediate layer (300-500 m) with relatively low 3 H levels represent AAIW, which is best characterized by a minimum in salinity (Fig. 2a). There are differences in AAIW depths because the stations are distributed across the ACC whose strong currents make the isopycnals inclined vertically due to geostrophic equilibrium principal. In the southern Indian Ocean, the isopycnals along which water masses move become deeper when going to the equator, but shallower when going to the pole (Park et al., 1993). So, the shallowest depth of AAIW at St. 3 (300 m) and the deepest depth at St. 8 (500 m) are entirely consistent with the geostropic dynamics. However, the elevated 3 H levels observed at St. 8 at water depths 1500 and 2500 m (Fig. 3a) indicate the presence of other water masses (Fig. 5a). It is known that the NIDW, which is characterized by a deep oxygen minimum (Fig. 2b), exists in the Crozet Basin (Park et al., 1993). This would explain higher 3 H concentrations observed in deep waters at St. 8, as the NIDW originates from the northern Indian Ocean where comparable 3 H levels were observed (Mulsow et al., 2003; Povinec et al., 2003b). There may also be at St. 8 a contribution from the advection via the Agulhas Current system of the deep water coming from the Indonesian Throughflow region, where deep tidal

23

mixing could incorporate the tritium-rich Pacific water. Lower 3 H levels found at 2000 m water depth may be due to intrusion of NADW carrying low radionuclide concentrations. Measurable 3 H concentrations were also observed in bottom waters at Sts. 3, 7 and 8 (Fig. 3a), as well as at the WOCE stations at 34°S (www.eWOCE.org). As sinking of surface water to the bottom at mid-latitudes is impossible due to strong thermohaline stratification (You, 2000), an injection of AABW might be considered as the most plausible source of 3 H (and low levels of 129I, Fig. 3c) in bottom waters. In fact, below the CDW, the salinity and temperature decrease, but oxygen increases sharply to the bottom, indicating the presence of the AABW (Park et al., 1993). A major formation area of the AABW is the Weddell Sea, where it is produced by sinking of cold, dense shelf waters (Rahmstorf, 2002). The AABW then flows around the Conrad Rise and enters the Crozet Basin through the Crozet-Kerguelen Gap (Park et al., 1993). Weddell Sea waters showed in 1975-76 3 H levels of 1 TU for surface and 0.5-0.7 TU for bottom waters (Michel, 1978). This is in agreement with our recent analysis of 3 H in seawater samples collected in 2003 at depths of 100-200 m in the Weddell and Ross Seas (0.4-0.5 TU). The transit time of surface water from the Weddell Sea to deep waters in the Crozet-Kerguelen Gap was estimated using CFCs as 23 ± 5 years (Haine et al., 1998). Hence, 1 TU observed in 1975-76 in Weddell Sea surface water may become 0.27 TU in 1999 in deep waters in the southern Indian Ocean, which is consistent with 3 H levels found in bottom waters at Sts. 3, 7 and 8. As the NADW has very low 3 H concentrations, and the NIDW is located well above the bottom layer, the only plausible explanation for higher 3 H levels found in bottom waters is the presence of the AABW. This is also supported by the stable isotope data as the AABW should be depleted both in 2 H and 18O in comparison with surface waters, as shown in Fig. 4b. The 129I data (Fig. 3c) have also been used for the identification of water masses present in the Crozet Basin. The results plotted in Fig. 5b as the 129I concentration vs. dissolved oxygen show the presence of all water masses as identified in Fig. 5a with 3 H data. The surface waters representing fronts (AF, STF and SAF) have highest 129I concentrations. The intermediate waters, represented by AAIW, show very similar 129I concentrations, thus eliminating the differences observed in the surface layer. The same is true for the deep and bottom waters. Higher 129I levels observed at St. 8 at 1500 and 2500 m water depths are probably due to the presence of NIDW. The lowest 129I levels found at 1000 m water depth may be due to intrusion of NADW carrying low radionuclide concentrations. The separation of NIDW from NADW is, however, not so clear as it was presented in the case of 3 H (Fig. 5a). Low, but measurable 129I levels were found at bottom waters which could be attributed to AABW. 5.4. Radionuclide inventories The inventory of a radionuclide in a seawater column is calculated by interpolating the radionuclide concentration measured at each depth:

IR =

1 2

(

) N    ∑ Wi + 1 + Wi di + 1 −di + 2W1 d1 + 2WN ðdB −dN Þ ;

i=1

where IR is the inventory of the radionuclide R in the seawater column (Bq/m2), N is the number of sampling depths, Wi is the radionuclide concentration in seawater at depth i, di is the i-th sampling depth of seawater, and dB is the total water depth to the bottom. The inventories are given in Bq/m2 (for 3 H: 1 TU= 0.118 kBq/m3, and for 129I: 106 atom/ L = 1.4 μBq/m3). Under normal conditions the radionuclide water column inventories should primarily depend on the geographical position of the sampling station, especially on its latitude. In this case the dominant

24

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

factor affecting the radionuclide water column inventories would be global fallout. However, in the subtropical part of the Indian Ocean a continuous transport of global fallout radionuclides from the central western Pacific via the Indonesian Seas, and its accumulation in the subtropical gyre should increase their inventories in the gyre. We shall focus on the evaluation of 3 H and 129I water column inventories as these radionuclides have had similar input functions to the ocean, as well as very similar behavior in the water column, and at least for 3 H the data can be compared with GEOSECS and WOCE results. These radionuclides could be therefore useful for characterization of water masses and their transport in the region. The highest total 3 H inventory (330 ± 30 kBq/m2) was observed at St. 8 (43°0´S; 62°32´E), situated close to the southern gyre stream. In contrast, the Sts. 3 and 7 situated more south showed lower 3 H inventories (around 60 kBq/m2). This difference in the 3 H inventories is mainly caused by differences in water fronts present in the region. The 3 H inventories in top 500 m for SASW (St. 3) and STSW (St. 7) were 13 ± 1 kBq/m2 and 25 ± 2 kBq/m2, respectively, when compared with 78 ± 8 kBq/m2 observed for IOSGW (St. 8). A similar trend can be also seen for AAIW: at Sts. 3 and 7 the 3 H inventories were 10 ± 1 kBq/m2 and 12 ± 1 kBq/m2, respectively, while the inventory at St. 8 was 59 ± 1 kBq/m2. The bottom waters at all three stations were under the influence of AABW showing similar 3 H inventories: 14 ± 1 kBq/m2 for St. 3 and 7, and 18 ± 2 kBq/m2 for St. 8. The 3 H inventories estimated for the Crozet Basin can be compared with WOCE (www.eWOCE.org) and GEOSECS (Östlund and Brescher, 1982) data, although their sampling sites were not close to the ANTARES IV ones. The WOCE transect I5P in the southern Indian Ocean (1998) with sampling St. 35 (33°S, 48°E), St. 39 (34°S, 53°E), St. 44 (34°S, 57°E) and St. 50 (34°S, 62°E) showed 3 H inventories by about a factor of 2 lower than that observed at St. 8. This is a reasonable result as all WOCE stations were inside the IOSG loop, similarly as St. 1, which showed the surface 3 H concentration comparable with the WOCE stations. The GEOSECS (1978) sampling St. 428 (38°S, 58°E; 1300 m water depth) located inside the IOSG loop (the closest station to St. 8) showed 3 H inventory of 55 ± 6 kBq/m2 (data were decay corrected to 1999), what is lower than the inventory observed at St. 8 (for the same water depth), situated, however, at 43°S on the main IOSG loop. The estimated 3 H inventories in the southern Indian Ocean are higher at least by a factor of three than expected from global fallout for similar latitude belts. They are also higher than 3 H inventories estimated for the northern Indian Ocean (Mulsow et al., 2003), and comparable (for the normalized water depth of 1000 m) with 3 H inventories in the western Pacific (Povinec et al., 2010). The 3 H inventories confirm again our hypothesis that there must be an additional source of tritium (as documented by ANTARES IV and WOCE data), which keeps its inventory in IOSGW high. 129 I inventories in the water column of the Crozet Basin show similar features as the 3 H inventories, i.e. a higher inventory at St. 8 (0.42 ± 0.08 mBq/m2), and smaller inventories at Sts. 3 and 7 (0.30 ± 0.06 mBq/m2 and 0.31± 0.06 mBq/m2, respectively). Unfortunately, there are no more data available for 129I in the water column of the southern Indian Ocean. Therefore, the only comparison we can do is with 129I data from the IAEA´97 cruise in the NW Pacific Ocean (Povinec et al., 2010). The 129 I inventories at Sts. 2 and 3 in the NW Pacific Ocean (38 ± 4 mBq/m2 and 51 ± 5 mBq/m2, respectively) are about two orders of magnitude higher than the inventory at St. 8. The difference is, as we already pointed out, due to 129I releases from the Tokai-mura reprocessing plant. The estimated radionuclide inventories confirm the unique role of the IOSG accumulating high inventories of radionuclides in the southern Indian Ocean. We can expect that the same is true for other radionuclides (e.g. for cesium (Povinec et al., submitted for publication), heavy metals and organic compounds which are dissolved in seawater and behave similarly as iodine.

5.5. Isotope activity ratios We shall compare isotope activity ratios of 3 H and 129I in the water column. As these radionuclides have similar input functions to the ocean, and similar behavior in the water column (however, very different half-lives - 12.32 y vs. 15.7 My), their activity ratios should help to differentiate between water masses present in the region. 3 H/129I activity ratios in surface waters (b100 m) of Sts. 3, 7 and 8 are (3.2 ± 0.5)106, (5.3 ± 0.8)x106 and (13.7 ± 2.1)106, respectively, documenting a presence of different water fronts in the region (SAF, STF and AF, respectively). While at 1000 m water depth of Sts. 3 and 7 we could report only limits (b0.8x106), St. 8 shows the highest ratio (19 ± 3)106, associated with 129I minimum. Waters in the depth interval of 1500-2500 shows measurable 3 H/129I ratio (6 ± 1)106 only at St. 8, associated with NIDW, while for Sts. 3 and 7 we can report only limit (b0.3x106), associated with NADW. Bottom waters associated with AABW show comparable ratios from (4 ± 1)106 (for St. 3) to (6 ± 1)106 (for St. 8). We can also compare 3 H/90Sr activity ratios in surface waters of the Crozet Basin where 90Sr (half-life 28.78 y) data are available (Lee et al., 2009). While IOSG Sts. 2, 5 and 8 show comparable activity ratios (130 ± 20, 110 ± 15, and 130 ± 20, respectively), St. 1 located in the IOSG loop shows a lower ratio (74 ± 9). The STF St. 7 gives the ratio of 89 ± 13 and Sts. 3, 4 and 6 of SAF show the ratios of 180 ± 30, 290 ± 50 and 300 ± 140, respectively. The differences are mainly due to lower 90Sr concentrations observed at the SAF stations, however, they are not significant as the precision of 90Sr measurements was lower (Lee et al., 2009) than of the 3 H measurements. 6. Conclusions Several observations have been made in this study which can be summarized as follows: (i) By comparing observed radionuclide (3 H, 14 C, and 129I) levels in the southern Indian Ocean with those found in the western Pacific Ocean it has been found that they are of common origin due to the transport of water masses from the western Pacific via the Indonesian Seas to the southern Indian Ocean. (ii) The radionuclide variations with latitude observed in the Indian Ocean are not due to the latitudinal atmospheric deposition patterns of global fallout, but due to the redistribution of radionuclides by ocean currents and mixing and spatial constraints given by different water masses present in the region. (iii) High radionuclide concentrations observed in the Indian Ocean in the latitudinal belt of 20-40°S are associated with Indian Ocean Subtropical Gyre which acts as a reservoir of radionuclides. (iv) High 14 C levels observed in bottom waters for transect around 40°S and 60°E document that the southern Indian Ocean is important for sink of anthropogenic carbon. (v) While latitudinal (32° - 48°S) variations of anthropogenic radionuclides in surface waters (b300 m deep) are attributable to different fronts tightly concentrated in the Crozet Basin, the intermediate layer (300-1000 m) is influenced by the AAIW, the deep waters (1000-3000 m) by the NADW and NIDW, and the bottom waters (N3000 m) by the AABW. As a consequence, the AAIW, NADW, NIDW and AABW eliminate the differences observed in the surface layer characterized by the AF, STF and SAF. The southern Indian Ocean appears to have been acting on a time scale of several decades as a final reservoir of contaminants transported from the northern Indian Ocean and central western Pacific Ocean. This has strong environmental consequences for the protection of the marine environment against a contamination from the land-based sources. The observed distribution of isotopic tracers in the Crozet Basin reflects the complex dynamics and advection of

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

different water masses, which makes the basin one of the most interesting oceanographic places in the World Ocean. Acknowledgments Colleagues participating in water sampling, and the Captain and the crew of the R/V Marion Dufresne are acknowledged for their assistance during the ANTARES IV expedition. The authors thank Isabelle Durand for the frontal analysis, as well as the Editor and two anonymous reviewers for constructive comments. This work was carried out in the framework of the IAEA international project Worldwide Marine Radioactivity Studies (WOMARS). The International Atomic Energy Agency is grateful to the Government of the Principality of Monaco for support provided to its Marine Environment Laboratories. PPP acknowledges support provided by the Slovak Scientific Agency VEGA (grant No. 1/108/08) and the EU Research & Development Operational Program funded by the ERDF (project No. 26240220004). References Aoyama, M., Hirose, K., Igarashi, Y., 2006. Re-construction and up-dating our understanding on the global weapons tests 137Cs fallout. J. Environ. Monit. 8, 431–438. Aoyama, M., Hirose, K., Hamajima, Y., Komura, K., Povinec, P.P., Sanchez-Cabeza, J.A., 2010. Long term behaviour of 137Cs in the Pacific Ocean and results of high density observation along P6 line in 2003. Progress in Oceanography, submitted. Aramaki, T., Mizushima, T., Kuji, T., Povinec, P.P., Togawa, O., 2001. Distribution of radiocarbon in the Southwestern North Pacific. Radiocarbon 43, 857–867. Bard, E., Arnold, M., Ostlund, H.G., Maurice, P., Monfray, O., Duplessy, J.-C., 1988. Penetration of bomb radiocarbon in the tropical Indian Ocean measured by means of accelerator mass spectrometry. Earth Planet. Sci. Lett. 87, 379–389. Bayer, R., Schlosser, P., 1991. Tritium profiles in the Weddel Sea. Mar. Chem. 35, 123–136. Belkin, I.M., Gordon, A.L., 1996. Southern Ocean fronts from the Greenwich Meridian to Tasmania. J. Geophys. Res. 101, 3675–3696. Bhushan, R., Dutta, K., Mulsow, S., Povinec, P.P., Somayajulu, B.L.K., 2003. Distribution of natural and man-made radionuclides during the reoccupation of GEOSECS stations 413 and 416 in the Arabian Sea: temporal changes. Deep-Sea Res. II 50, 2777–2784. Bowen, V.T., Noshkin, V.E., Livingston, H.D., Volchok, H.L., 1980. Fallout radionuclides in the Pacific Ocean; vertical and horizontal distributions, largely from GEOSECS stations. Earth Planet. Sci. Lett. 49, 411–434. Broecker, W.S., 1991. The great ocean conveyor. Oceanography 4, 79–89. Broecker, W.S., Peng, W.H., 1982. Tracers in the sea. Eldigo Press, New York. 690 pp. Broecker, W.S., Peng, T.H., Ostlund, H.G., 1986. The distribution of bomb tritium in the ocean. J. Geophys. Res. 9, 14331–14344. Burchuladze, A.A., Chudý, M., Eristavy, I.V., Pagava, S.V., Povinec, P., Šivo, A., Togonidze, G.I., 1989. Anthropogenic 14 C variations in atmospheric CO2 and wines. Radiocarbon 31, 771–776. Clarke, W.B., Jenkins, W.J., Top, Z., 1976. Determination of tritium by mass-spectrometric measurements of 3He. Int. J. Radiat. Isot. 27, 512–522. Coleman, M.L., Sheppard, T.J., Durham, J.J., Rouse, J.E., Moore, G.R., 1982. Reduction of water with zinc for hydrogen isotope analysis. Anal. Chem. 54, 993–995. Cooper, L.W., Hong, G.-H., Beasley, T.M., Grebmeier, J.M., 2001. Iodine-129 concentrations in marginal seas of the North Pacific and Pacific-influenced waters of the Arctic Ocean. Mar. Poll. Bull. 42, 1347–1356. Coppola, L., Roy-Barman, M., Mulsow, S., Povinec, P.P., Jeandel, C., 2005. Low particulate organic carbon export in the frontal zone of the Southern Ocean (Indian Sector) revealed by 234Th. Deep Sea Res. I 52, 51–68. Coppola, L., Roy-Barman, M., Mulsow, S., Povinec, P.P., Jeandel, C., 2006. Thorium isotopes as traces of particles dynamics and deep water circulation in the Indian sector of the Southern Ocean (ANTARES IV). Mar. Chem. 100, 299–313. Craig, H., 1961. Isotopic variation in meteoric waters. Science 133, 1702–1703. Davis, R.E., 2005. Intermediate-depth circulation of the Indian and South Pacific Oceans measured by autonomous floats. J. Phys. Oceanogr. 35, 683–707. Donahue, D.J., Jull, A.J.T., Linick, T.W., Toolin, L.J., 1990. Radiocarbon measurements at the University of Arizona AMS Facility. Nucl. Instr. Meth. Phys. Res. B 52, 224–228. Edmonds, H.N., Zhou, Z.Q., Raisbeck, G.M., Yiou, F., Kilius, L., Edmond, J.M., 2001. Distribution and behavior of anthropogenic 129I in water masses ventilating the North Atlantic Ocean. J. Geophys. Res. 106, 6881–6894. Epstein, S., Mayeda, T., 1953. Variation of O18 content of waters from natural sources. Geochim. Cosmochim. Acta 4, 213–224. Fine, R.A., 1985. Direct evidence using tritium data for throughflow from the Pacific into the Indian Ocean. Nature 315, 478–480. Fine, R., 1993. Circulation of Antarctic intermediate water in the South Indian Ocean. Deep-Sea Res. 40, 2021–2042. Ganachaud, A., Wunsch, C., 2000. Improved estimates of global ocean circulation, heat transport a mixing from hydrographic data. Nature 408, 453–457. Gonfiantini, R., 1978. Standards for stable isotope measurements in natural compounds. Nature 271, 534–536.

25

Gordon, A.L., 1985. Indian-Atlantic Transfer of Thermocline Water at the Agulhas Retroflection. Science 227 (4690), 1030–1033. Gordon, A.L., Fine, R.A., 1996. Pathways of water between the Pacific and Indian Oceans in the Indonesian seas. Nature 379, 146–149. Gordon, A.L., Susanto, R.D., Vranes, K., 2003. Cool Indonesian througflows as a consequence of restricted surface layer flow. Nature 425, 824–828. Groening, M., Auer, R., Brummer, D., Jaklitsch, M., Sambandam, C., Tanweer, A., Tatzber, H., 2009. Increasing the performance of tritium analysis by electrolytic enrichment. Isot. Environ. Health Stud. 45, 118–125. Haine, T.W.N., Watson, A.J., Liddicoat, M.I., 1998. Dickson, R.R., 1998. The flow of Antarctic bottom water to the southwest Indian Ocean estimated using CFCs. J. Geophys. Res. 103, 27637–27654. Hong, G.-H., Lee, S.H., Kim, S.H., Chung, C.S., Baskaran, M., 1999. Sedimentary fluxes of Sr- 90, Cs-137, Pu-239, Pu-240 and Pb-210 in the East Sea (Sea of Japan). Sci. Total Environ. 238, 225–240. Hou, X., Hansen, V., Aldahan, A., Possnert, G., Lind, O.C., Lujaniené, G., 2009. A review on speciation of 129I in environmental and biological samples. Anal. Chim. Acta 632, 181–196. International Atomic Energy Agency, 2005. Worldwide Marine Radioactivity Studies (WOMARS). IAEA-TECDOC-1429. IAEA, Vienna. 187 pp. Jenkins, W.J., Lott, D.E., Pratt, M.W., Boudreau, R.D., 1983. Anthropogenic tritium in South Atlantic bottom water. Nature 305, 45–46. Jull, A.J.T., Burr, G.S., Beck, J.W., Hodgins, G.W.L., Biddulph, D.L., McHargue, L.R., Lange, T.E., 2008. Accelerator mass spectrometry of long-lived light radionuclides. In: Povinec, P. (Ed.), Analysis of Environmental Radionuclides. Elsevier, Amsterdam, pp. 241–262. Key, R., 1996. WOCE pacific ocean radiocarbon program. Radiocarbon 38, 415–423. Key, R., Quay, P., Jones, G., McNichol, A., von Reden, K., Schneider, R., 1996. WOCE AMS radiocarbon.1. Pacific Ocean results (P6, P16 and P17). Radiocarbon 38, 425–518. Key, R., Quay, P., Schlosser, P., McNichol, A., von Reden, K., Schneider, R., Elder, K., Stuiver, M., Ostlund, H., 2002. WOCE radiocarbon IV: Pacific Ocean results; P10, P13N, P14C, P18, P19 & S4P. Radiocarbon 44, 239–392. Key, R.M., Kozyr, A., Sabine, C.L., Lee, K., Wanninkhof, R., Bullister, J.L., Feely, R.A., Millero, F.J., Mordy, C., Peng, T.-H., 2004. A global ocean carbon climatology: Results from Global Data Analysis Project (GLODAP). Glob. Biogeochem. Cycles 18 (4) doi:10.1029/2004GB002247 GB4031. Kilius, L.R., Rucklidge, J.C., Litherland, A.E., 1987. Accelerator mass spectrometry of 129I at IsoTrace. Nucl. Instr. Meth. B29, 72–76. La Rosa, J., Gastaud, J., Lagan, L., Lee, S.H., Levy-Palomo, I., Povinec, P.P., Wyse, E., 2005. Recent developments in the analysis of transuranics (Np, Pu, Am) in seawater. J. Radioanal. Nucl. Chem. 263, 427–436. Lee, S.-H., Povinec, P.P., Wyse, E., Pham, M.K., Hong, G.-H., Chung, C.S., Kim, S.-H., Lee, H.J., 2005. Distribution and inventories of 90Sr, 137Cs, 241Am and Pu isotopes in sediments of the Northwest Pacific Ocean. Mar. Geol. 216, 249–263. Lee, S.-H., Povinec, P.P., Gastaud, J., Oregioni, B., Coppola, L., Jeandel, C., 2009. Radionuclides as tracers of water fronts in the south Indian Ocean—ANTARES IV Results. J. Oceanogr. 65, 397–406. LeGrande, A.N., Schmidt, G.A., 2006. Global gridded data set of the oxygen isotopic composition in seawater. Geophys. Res. Lett. 33, L12604. doi:10.1029/2006GL026011. Levin, I., Hesshaimer, V., 2000. Radiocarbon – a unique tracer of global carbon cycle dynamics. Radiocarbon 46, 69–80. Liong Wee Kwong, L., Povinec, P.P., Jull, A.J.T., 2004. Preparation of graphite targets from small marine samples for AMS radiocarbon measurements. Radiocarbon 46, 133–139. Livingston, H.D., Povinec, P.P., 2000. Anthropogenic marine radioactivity. Ocean Coast. Manage. 43, 689–712. Livingston, H.D., Povinec, P.P., 2002. A millennium perspective on the contribution of global fallout radionuclides to ocean science. Health Phys. 82, 656–668. Livingston, H.D., Povinec, P.P., Ito, T., Togawa, O., 2001. The behaviour of plutonium in the Pacific Ocean. In: Kudo, A. (Ed.), Plutonium in the Environment. Elsevier, Amsterdam, pp. 267–292. Metzl, N., Tilbrook, B., Poisson, A., 1999. The annual fCO2 cycle and the air-sea CO2 flux in the sub-Antarctic Ocean. Tellus 51B, 849–861. Michel, R.L., 1978. Tritium distributions in Weddell Sea water masses. J. Geophys. Res. 83, 6192–6198. Mittelstaedt, E., Osvath, I., Povinec, P.P., Togawa, O., Scott, E.M., 1999. Transport of radionuclides from the Mururoa and Fangataufa atolls through the marine environment. Sci. Total Environ. 237 (238), 301–310. Morgenstern, U., Taylor, C.B., 2009. Ultra Low-level tritium measurement using electrolytic enrichment and LSC. Isot. Environ. Health Stud. 45, 96–117. Mulsow, S., Povinec, P.P., Somayajulu, B.L.K., Oregioni, B., Liong Wee Kwong, L., Gastaud, J., Top, Z., Morgenstern, U., 2003. Temporal (3 H) and spatial variations of 90Sr, 239, 240Pu and 241Am in the Arabian Sea: GEOSECS Stations revisited. Deep Sea Res. II 50, 2761–2775. Nakano, M., Povinec, P.P., 2003. Oceanic general circulation model for the assessment of the distribution of 137Cs in the world ocean. Deep Sea Res. II 50, 2803–2816. Nydal, R., Lövseth, K., 1965. Distribution of radiocarbon from nuclear tests. Nature 206, 1029–1031. Östlund, H.G., Brescher, R., 1982. Tritium Laboratory data Report No. 2., University of Miami, Miami, USA. Park, Y.-H., Gambéroni, L.L., Charriaud, E., 1993. Frontal structure, water masses and circulation in the Crozet Basin. J. Geophys. Res. 98, 12,361–12,385. Park, Y.-H., Pollard, R.T., Read, J.F., Leboucher, V., 2002. A quasi-synoptic view of the frontal circulation in the Crozet Basin during the Antares-4 cruise. Deep Sea Res. II 49, 1823–1842. Pham, M.K., Betti, M., Povinec, P.P., Alfimov, V., Biddulph, D., Gastaud, J., Kieser, W.E., López Gutiérrez, J.M., Possnert, G., Sanchez-Cabeza, J.A., et al., 2010. Certified

26

P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 14–26

reference material IAEA-418: 129I in Mediterranean Sea water. J. Radioanalytical Nucl. Chem. doi:10.1007/s10967-010-0621-6. Pollard, R.T., Lucas, M.I., Read, J.F., 2002. Physical controls on biological zonation in the Southern Ocean. Deep Sea Res. II 49, 3289–3305. Pollard, R.T., Venables, H.J., Read, J.F., Allen, J.T., 2007. Large-scale circulation around the Crozet Plateau controls an annual phytoplankton bloom in the Crozet Basin. Deep Sea Res. II 54, 1915–1929. Pollard, R.T., et al., 2009. Southern Ocean deep-water carbon export enhanced by natural iron fertilization. Nature 457, 577–589. Povinec, P.P., 2004. Developments in analytical technologies for marine radioactivity studies. In: Livingston, H.D. (Ed.), marine radioactivity. Elsevier, Amsterdam, pp. 237–294. Povinec, P.P., 2005. Ultra-sensitive radionuclide spectrometry: Radiometrics and mass spectrometry synergy. J. Radioanal. Nucl. Chem. 263, 413–417. Povinec, P.P., Oregioni, B., Jull, A.J.T., Kieser, W.E., Zhao, X.-L., 2000. AMS measurements of 14 C and 129I in seawater around radioactive waste dump sites. Nucl. Instr. Meth. B172, 672–678. Povinec, P.P., Badie, C., Baeza, A., Barci-Funel, G., Bergan, T.D., Bojanowski, R., Burnett, W., Eikenberg, J., Fifield, L.K., Serradell, V., Gastaud, J., Goroncy, I., Herrmann, J., Hotchkis, M.A.C., Ikaheimonen, T.K., Jakobson, E., Kalimbadjan, J., La Rosa, J.J., Lee, S.-H., Liong Wee Kwong, L., Lueng, W.M., Nielsen, S.P., Noureddine, A., Pham, M.K., Rohou, J.-N., Sanchez-Cabeza, A., Suomela, J., Suplinska, M., Wyse, E., 2002. Certified reference material for radionuclides in seawater IAEA-381 (Irish Sea water). J. Radioanal. Nucl. Chem. 251, 369–374. Povinec, P.P., Livingston, H.D., Shima, S., Aoyama, M., Gastaud, J., Goroncy, I., Hirose, K., Huyhn-Ngoc, L., Ikeuchi, Y., Ito, To., La Rosa, J., Liong Wee Kwong, L., Lee, S.-H., Moriya, H., Mulsow, S., Oregioni, B., Pettersson, H., Togawa, O., 2003a. IAEA'97 expedition to the NW Pacific Ocean – results of oceanographic and radionuclide investigations of the water column. Deep-Sea Res. II 50, 2607–2638. Povinec, P.P., Delfanti, R., Gastaud, J., La Rosa, J., Morgenstern, U., Oregioni, B., Pham, M.K., Salvi, S., Top, Z., 2003b. Anthropogenic radionuclides in the Indian Ocean surface waters – the Indian Ocean transect 1998. Deep Sea Res. II 50, 2751–2760. Povinec, P.P., Hirose, K., Honda, T., Ito, T., Scott, E.M., Togawa, O., 2004a. Spatial distribution of 3 H, 90Sr, 137Cs and 239, 240Pu in surface waters of the Pacific and Indian Oceans – GLOMARD database. J. Environ. Radioactiv. 76, 113–137. Povinec, P.P., Aramaki, T., Burr, G.S., Jull, A.J.T., Liong Wee Kwong, L., Togawa, O., 2004b. Radiocarbon in the water column of the southwestern North Pacific Ocean – 24 years after GEOSECS. Radiocarbon 46, 583–594. Povinec, P.P., Aarkrog, A., Buesseler, K.O., Delfanti, R., Hirose, K., Hong, G.-H., Ito, T., Livingston, H.D., Nies, H., Noshkin, V.E., Shima, S., Togawa, O., 2005. 90Sr, 137Cs and 239, 240Pu concentration surface water time series in the Pacific and Indian Oceans. WOMARS results. J. Environ. Radioactiv. 81, 63–87. Povinec, P.P., Betti, M., Jull, A.J.T., Vojtyla, P., 2008. New isotope technologies for environmental physics. Acta Phys. Slovaca 58, 1–154. Povinec, P.P., Lee, S.-H., Liong Wee Kwong, L., Oregioni, B., Groening, M., Jull, A.J.T., Kieser, W.E., Morgenstern, U., Top, Z., 2010. Tritium, radiocarbon, 90Sr and 129I in the Pacific and Indian Oceans. Nucl. Instrum. Meth. Phys. Res. B 268, 1214–1218.

Povinec, P.P., M. Aoyama, M. Fukusawa, K. Hirose, K. Komura, J.-A. Sanchez-Cabeza, J. Gastaud, M. Ješkovský, I. Levy-Palomo, I. Sýkora, 2010b. Profiles of 137Cs in South Indian Ocean water along the 20°S latitude – an evidence for accumulation of pollutants in the subtropical gyre. Progress in Oceanography, submitted. Rahmstorf, S., 2002. Ocean circulation and climate during the past 120, 000 years. Nature 419, 207–214. Rahmstorf, S., 2006. Thermohaline Ocean Circulation. In: Elias., S.A. (Ed.), Encyclopedia of Quaternary Sciences. Elsevier, Amsterdam. Raisbeck, G., You, F., 1999. 129I in the oceans: origins and applications. Sci. Total Environ. 237 (238), 31–42. Ridgway, K.R., Dunn, J.R., 2007. Observational evidence for a Southern Hemisphere oceanic supergyre. Geophys. Res. Lett. 34, L13612. doi:10.1029/2007GL030392. Santschi, P.H., Schwehr, K.A., 2004. 129I/127I as a new environmental tracer or geochronometer for biogeochemical or hydrodynamic processes in the hydrosphere and geosphere: The central role of organo-iodine. Sci. Total Environ. 321, 257–271. Schlosser, P., Bayer, R., Boenisch, G., Cooper, L.W., Ekwurzel, B., Jenkins, W.J., Khatiwala, S., Pfirman, S., Smethie, W.M., 1999. Pathways and residence times of dissolved pollutants in the ocean derived from transient tracers and stable isotopes. Sci. Total Environ. 237 (238), 15–30. Schlosser, P., Bullister, J.L., Fine, R., Jenkins, W.J., Key, R., Lupton, J., Roether, W., Smethie, W.M., 2001. Transportation and age of water masses. In: Siedler, G.J., Church, J., Gould, J. (Eds.), Ocean Circulation and Climate. Academic Press, London, UK, pp. 431–452. Schmitz, W.J., 1995. On the interbasin-scale thermohaline circulation. Rev. Geophys. 33, 151–174. Speich, S., Blanke, B., Cai, W., 2007. Atlantic meridional overturning circulation and the Southern Hemisphere supergyre. Geophys. Res. Lett. 34, L23614. doi:10.1029/ 2007GL031583. Stuiver, M., Ostlund, H.G., 1983. GEOSECS Indian Ocean and Mediterranean radiocarbon. Radiocarbon 25, 1–29. Tomczak, M., Godfrey, J.S., 1994. Regional Oceanography: An Introduction. Pergamon Press, New York. 422p. Toole, J.M., Warren, B.A., 1993. A hydrographic section across the subtropical South Indian Ocean. Deep-Sea Res. 40, 1973–2019. Top, Z., 1999. Tritium inventory of the Black Sea. J. Environ. Radioactiv. 43, 221–229. Tuniz, C., Bird, J.R., Fink, D., Herzog, G.F., 1998. Accelerator Mass Spectrometry: Ultrasensitive analysis for global science. CRC Press, Boca Raton. 371 pp. UNSCEAR, 2000. Sources and Effects of Ionizing Radiation. United Nations, New York. van Beek, P., Bourquin, M., Reyss, J.-L., Souhaut, M., Charette, M.A., Jeandel, C., 2008. Radium isotopes to investigate the water mass pathways on the Kerguelen Plateau (Southern Ocean). Deep Sea Res. II 55, 622–637. Weiss, W., Roether, W., 1980. The rates of tritium input to the world oceans. Earth Planet. Sci. Lett. 49, 435–436. You, Y., 2000. Implications of the deep circulation and ventilation of the Indian Ocean on the renewal mechanism of North Atlantic Deep Water. J. Geophys. Res. 105, 23895–23926.