Вы находитесь на странице: 1из 13

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

P.P. Povinec

, R. Breier
, L. Coppola
, M. Groening
, C. Jeandel
, A.J.T. Jull
, W.E. Kieser
, S.-H. Lee
L. Liong Wee Kwong
, U. Morgenstern
, Y.-H. Park
, Z. Top
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
University of Arizona, Departments of Physics and Geosciences, Tucson, AZ 85712-1201, USA
University of Toronto, IsoTrace Laboratory, Toronto, M5S 1A7, Canada
Korea Research Institute of Standards and Science, Daejeon, Republic of Korea
International Atomic Energy Agency, Marine Environment Laboratories, MC-98000 Monaco,
Institute of Geological and Nuclear Sciences, Lower Hutt, New Zealand
Musum National d'Histoire Naturelle, LOCEAN/DMPA, F-75231, Paris France
University of Miami, Rosenstiel School of Marine and Atmospheric Sciences, Miami, USA
a b s t r a c t 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
Crozet Basin
Indian Ocean
Anthropogenic radionuclides (
C, and
I) 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 Pacic 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 Pacic 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 20-40S 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.
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 (
H), radiocarbon (
strontium-90 (
Sr), cesium-137 (
Cs), iodine-129 (
I), americium-
241 (
Am), plutonium isotopes (
Pu ), 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
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 Pacic close in fallout from nuclear weapons tests
carried out at Bikini and Enewetak Atolls contributed to radionuclide
inventories in the Pacic 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 Pacic due to the
combined effect of higher precipitation and higher stratosphere-
troposphere exchange of air (Aoyama et al., 2006). Some of the global
fallout radionuclides (e.g.
I) 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
Amand Pu isotopes are more particle reactive (La Rosa et al., 2005), and
suitable for investigation of processes in the water column and
sediments. Signicant portions of these radionuclides in the world
ocean have accumulated at the seaoor as bottom sediments (Bowen
et al., 1980; Hong et al., 1999; Lee et al., 2005; Livingston et al., 2001).
Earth and Planetary Science Letters 302 (2011) 1426
Corresponding author. Tel.: +421 260 295 544; Fax: 421 265 425 882.
E-mail address: povinec@fmph.uniba.sk (P.P. Povinec).
Formerly at the International Atomic Energy Agency, Marine Environment
Laboratories, Monaco.
Present address: Observatoire Oceanologique de Villefranche-sur-mer, La Darse BP
08, 06238 Villefranche-sur-mer, France.
Present address: University of Ottawa, Ottawa K1N 6 N5, Canada.
0012-821X/$ see front matter 2010 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Earth and Planetary Science Letters
j our nal homepage: www. el sevi er. com/ l ocat e/ epsl
We shall focus in this paper on three radionuclides,
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 articially 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.
C levels in the environment have also been disturbed by
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 Lvseth, 1965). Because of its long half-life
(5730 y) and subsequent formation of
in the air, specic
stratosphere-troposphere-biosphere mixing, exchange of carbon
dioxide with surface ocean, and sequestration of carbon dioxide into
the deep ocean,
C became the most frequently studied environ-
mental 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).
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
He ingrowth from tritium decay; Clarke et al., 1976; Schlosser et al.,
1999; Top, 1999), and accelerator mass spectrometry for
C (e.g. Jull
et al., 2008; Key et al., 1996;2002; Povinec, 2005; Povinec et al., 2008;
Tuniz et al., 1998) and
I (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
H and
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 rst
measurements of
H and
C were made through the Geochemical
Ocean Sections (GEOSECS) project (1977-1978) (Broecker et al.,
1986; stlund and Brescher, 1982; Stuiver and Ostlund, 1983), and
followed by the WOCE eld project (1990-1998; www.eWOCE.org) in
the Indian Ocean. Unfortunately, the
H and
C measurements were
not carried out in the western Indian Ocean south of 34S.
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
chlorophyll region in the world. It is contributing to the regulation of
the atmospheric CO
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 Inter-
national 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 anthropo-
genic 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 (
I) and stable (
O) isotopes in surface and deep waters of
the southern Indian Ocean. Results on
Pu and
Am in
surface waters and plankton have been published earlier (Lee et al.,
2. Oceanography background
The Indian Ocean is limited northward to 25N by the continent,
and at 60S, where it becomes the Southern Ocean. The 60
S 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) fromthe western Pacic Ocean via the Indonesian
throughow 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 systeminuencing water circulation between
20 and 40S (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) alongthe easternAfricancoast, entering the South
Atlantic around the Cape of Good Hope, forming a branch of the AC that
does not complete the retroection 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 ows poleward along the continental
shelf break from about 22S to 35S, and then turns eastward. The
Leeuwin Current is warm and of relatively low salinity, low dissolved
oxygen and high phosphate content. It transfers a signicant amount
of heat to the south (Tomczak and Godfrey, 1994).
South of the 40
S, 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 Pacic 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
15 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
the Agulhas Front (AF), Subtropical Front (STF) and Subantarctic Front
(SAF) converge into a narrowband 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 60E, the AF re-circulates to the north,
as a part of the southern limb of the anticyclonic IOSG (Park et al.,
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 stratication 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 (
I, and stable
H and
O) 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 CO
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/VMarion Dufresne (CNRS) as a part of
theANTArctic RESearch(ANTARES) IVcruise (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- 48S and 51- 70E, in the
conuence zoneof the AF (Sts. 1, 2, 5and8), STF (St. 7), SAF (Sts. 3and4)
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
H (and stable isotopes),
(poisoned for preventing any biological activity by adding mercury
chloride) and
I 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
H was measured either by the
He 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, NewZealand;
Morgenstern and Taylor, 2009), and at the Isotope Hydrology
Laboratory of the International Atomic Energy Agency (Vienna, Austria;
Groening et al., 2009).
Hconcentration is given in TritiumUnit (1 TUis
the isotopic ratio of one
H atom to 10
protium (
H) atoms,
equivalent to 118 mBq/L of water). The
He ingrowth method can
H levels down to 0.01 TU (precision at 1 is around 0.005
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 (owing 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.
16 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
TU); the liquid scintillation spectrometry after double electrolytic
enrichment and special calibration can reach 0.02 TU (Morgenstern
and Taylor, 2009; Povinec, 2004).

analyses were done using H2O-Zn reduction (Coleman et al.,
analyses were performed using the CO2-H2O equilibra-
tion procedure described in Epstein and Mayeda (1953). The stable
isotopic compositions of hydrogen and oxygen are reported as "delta"
() values in parts per thousand (denoted as ) enrichments or
depletions relative to a standard of known composition

The isotopic results are reported against the international standard
VSMOW (Vienna Standard Mean Ocean Water) as dened by
Gonantini (1978). Typical uncertainties at 1 are1 for
and0.1 for
O. Analyses of H and O stable isotopes of seawater
were carried out at the Lower Hutt and Vienna laboratories.
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 CO
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
were accelerated and analyzed for isotopic composition (Jull et al.,
C concentration is given as
C in , relative to the NIST
(National Institute of Standards and Technology, Gaithersburg, USA)
C oxalic acid standard (Povinec et al., 2004b; Stuiver and Ostlund,

C = F
1 10
where F
(a fraction of modern carbon) is the measured AMS
ratio in a sample normalised to
=-25 , as dened by
Donahue et al. (1990). The precision of AMS measurements was
around5 . 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 addedtothesample, andafter several steps AgI was
obtainedwhichwas usedas a target for AMS measurements carriedout in
the IsoTrace Laboratory. The AMS measurements were normalised with
respect to ISOT-2 reference material (
I =(1.1740.022) 10
I results are given in atom/L. The relative precision of AMS measure-
ments was around10 %.
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 rst time that such a suite of radioactive
C and
I) and stable (
H and
O) 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 identied: saline North
Atlantic Deep Water (NADW) was observed at all water prole
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
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
C and
I water proles (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.
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
H levels for the surface waters.
C proles (Fig. 3b) also show a downward transport of bomb
C to the depth of about 2000 m. The
I water proles
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
H and
I which best
represents global fallout radionuclides due to their direct input to the
Fig. 2. Diagrams of T-S and T-O
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.
17 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
sea. The standing stocks of
Hin the upper 1 km
of the water column
in the IOSGW(St. 8) were 14010 GBq, much higher than in STSWof
St. 7 (405 GBq) and in SASWof St. 3 (244 GBq). In the case of
the standing stocks were much smaller, and the differences between
the stations were within the uncertainties: 112 kBq for St. 8, then
92 kBq for St. 7 and 82 kBq for St. 3.
C, and
I concentrations observed in the western North
Pacic in 1997 during the IAEA97 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 Pacic and IndianOceanproles show
similar features a surface (or subsurface at 100-200 m) maxima, a
sharpdecrease toward1000 m, andnegligible levels indeepandbottom
4.2. Stable isotopes
In addition to physical oceanography parameters and radio-
nuclides, 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
O, with high salinity in the surface and gradual depletion of
O with
gradually freshening toward the depth. SASW (St.3) were most
depleted with
O and the most fresh among the water masses
present (Fig. 4a). Coefcient of determination, R
=0. 86 (P value
b0.0001), indicates that there is a strong correlation between
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) dened 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
H vs.
O diagram
presented in Fig. 4b which shows a typical correlation between these
two isotopes.
Tritium versus
O, and
I vs. dissolved oxygen diagrams
presented in Fig. 5 also revealed presence of different water masses in
the Crozet Basin. Elevated
H and
I concentrations were observed at
St. 8 in surface, mediumdepth and bottom waters, while at Sts. 3 and 7
only bottom waters showed higher concentrations.
5. Discussion
5.1. Radionuclide proles
Following the global atmospheric deposition, higher surface radio-
nuclide concentrations are expected for the Northern Hemisphere than
in the Southern Hemisphere. The peak value in the Southern
Hemisphere at 50S 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
H value observed in the
northern IndianOcean at 20Nwas 1.0 TU(Mulsowet al., 2003; Povinec
et al., 2003b). Following the Weiss and Roether (1980) estimation of the
development of the mean tritiumconcentration in the open ocean with
time, the expected value in 1998-1999 for the 40S latitude would be
60% of that observed for 20N, it means only 0.6.
It is clear that the observed
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
Hin the Indian Ocean (Livingston
Fig. 3.
C (b) and
I (c) water proles at Sts. 3. 7 and 8 compared with nearest
GEOSECS (stlund and Brescher, 1982; Stuiver and stlund, 1983) and WOCE (www.
eWOCE.org) stations.
C and
I obtained for the NW Pacic (IAEA97 cruise;
Povinec et al., 2004b; Povinec et al., 2010) are also shown.
H data were decay
corrected to January 1999 with half-life of 12.32 y.
H concentration is given in Tritium
Unit (1 TU is the isotopic ratio of 1
H atom to10
protium (
H) atoms, equivalent to
118 mBq/L of water).
C concentration is given as
C in , relative to the NIST
standard (Stuiver and stlund, 1983). Relative precisions (at 1 ) are ~2 % for
H, ~0.5
% for
C and ~10 % for
I. For positions of sampling stations see Fig. 1.
18 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
and Povinec; Povinec et al., 2003b), high surface
H concentrations
observed at Sts. 2, 5 and 8 (1.1 TU for the latitude belt 40S 43S,
Table 1) should be associated with a radionuclide transport from the
western Pacic Ocean, where high
H concentrations were observed
(Povinec et al., 2010).
The average
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 Pacic
(stlund and Brescher, 1982).
H levels in subsurface maxima in the
NW Pacic, 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 10
/s of seawater ows from the western Pacic Ocean
via the Indonesian throughow 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 fromthe western Pacic to the Crozet Basin is around 13 years,
comparable with the
H half-life. A similar value has been obtained
from direct oat measurements (Davis, 2005). The oat data (Davis,
2005) also indicate that about 10 years is needed to complete one
loop in the IOSG (Fig. 1). This would mean that
H decayed in the
IOSG waters is renewed by fresh
Hwater coming fromthe western
Pacic Ocean which acts as a tritium source for the Indian Ocean due
to higher fallout levels found in the western North Pacic Ocean
(Aoyama et al., 2006; Povinec et al., 2010).
In general, the bulk of
H in the water column (Fig. 3a) should
exist near surface due to the atmospheric input. Assuming no specic
sources for
H in the study area, the existence of higher
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 stratication. 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
H levels, as discussed later.
Radiocarbon water proles 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
C proles shown in Fig. 3b reect the different
positions of water fronts, similarly as it was in the case of tritium. Peak
C levels observed at St. 8 are comparable with those observed in the
NW Pacic (IAEA97 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).
Table 1
Oceanographic and isotope data of collected seawater samples in Crozet Basin (January-February 1999).
Station Position Depth Temperature Salinity
(m) (C) () () (TU) () (10
1 3242.19'S 7000.00'E 4 19.05 35.619 0.3 0.57 0.720.03 101.2 -
2 4017.98'S 7000.00'E 4 17.01 35.177 0.3 0.34 0.930.03 108.3 -
4 4551.32'S 5500.00'E 4 9.86 33.684 -5.8 -0.49 0.320.02 32.54 -
5 4000.00'S 5157.27'E 4 19.55 35.533 0.2 0.47 1.100.03 110.5 -
6 4800.00'S 6838.50'E 4 9.90 33.640 -6.1 -0.46 0.180.02 23.52 -
3 4600.03'S 6303.58'E 6 9.53 33.6981 -5.2 -0.37 0.2420.017 26.40 0.0660.009
50 9.54 33.7001 -5.7 -0.57 0.1970.019 43.60 0.0580.008
100 6.01 33.8346 -2.4 -0.37 0.2790.023 31.30 0.0620.009
150 5.22 33.8800 -2.9 -0.30 0.2270.022 -26.80 0.1050.010
300 5.00 34.2135 -3.1 -0.26 0.2410.020 -60.20 0.0610.009
500 3.93 34.2484 -4.0 -0.21 0.1780.010 -100.00 0.0600.006
1000 2.95 34.4709 -4.5 -0.25 b0.02 -147.30 0.0130.008
1500 2.48 34.6666 -4.9 -0.24 b0.02 - 0.0580.008
2000 2.24 34.7527 -4.4 -0.26 b0.02 -145.20 0.0530.007
2500 1.84 34.7621 -4.6 -0.20 b0.02 .150.30 0.0330.006
3500 0.85 34.7080 -2.9 -0.17 0.2890.022 - 0.0620.008
4300 0.47 34.6857 -2.7 -0.19 0.3020.024 -162.00 0.0500.005
7 4411.45'S 6324.45'E 6 14.30 34.3431 -3.3 -0.09 0.3990.024 62.00 0.0600.007
50 13.53 34.3513 -3.8 0.08 0.4330.022 - 0.0770.009
100 12.37 34.9364 -2.7 0.20 0.5340.025 - 0.0800.009
150 11.49 34.8774 -3.3 0.26 0.5240.022 51.30 0.1140.009
300 10.50 34.7880 -4.4 -0.09 0.4930.023 25.10 0.0590.007
500 8.38 34.5982 -3.8 -0.12 0.1580.010 -26.70 0.0580.008
1000 3.73 34.3584 -4.5 -0.20 b0.02 -149.30 0.0260.006
1500 2.82 34.5739 -4.8 -0.27 b0.02 -150.90 0.0420.009
2500 2.14 34.7600 -4.9 -0.20 b0.02 -144.90 0.0460.012
3500 1.22 34.7260 - - - -150.16 0.0390.009
4300 0.49 34.6850 -2.5 -0.15 0.3020.024 -156.40 0.0420.022
8 4308.03'S 6231.42'E 6 17.40 35.5319 2.7 0.56 1.230.03 - 0.0790.012
50 17.53 35.5244 -0.1 0.33 1.250.03 - 0.0820.008
100 15.69 35.4892 -1.2 0.20 1.420.03 101.20 0.0790.008
200 14.77 35.4557 1.2 0.22 1.350.03 98.50 0.1520.014
500 13.14 35.2852 -1.3 0.10 1.290.02 10.70 0.0550.007
1000 7.34 34.5270 -2.0 -0.18 0.750.02 -100.300 0.0330.005
1500 3.55 34.4122 -3.0 -0.20 0.640.02 -125.000 0.0750.008
2000 2.72 34.6225 -3.3 -0.27 0.280.01 -150.100 0.0450.006
2500 2.44 34.7330 -2.9 -0.20 0.540.01 - 0.0770.008
3000 2.09 34.7612 - - - - 0.0810.009
4000 1.12 34.7184 -2.7 -0.17 0.370.03 -162.200 0.0340.008
5000 0.44 34.6786 -2.6 -0.15 0.390.03 - 0.0550.006
* Uncertainty at 1 is1 for
H, 0.1 for
O and 5 for
19 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
Iodine-129 water proles 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
I levels. An interesting feature
is that
I levels in the top 1000 m are by about a factor of two lower
in the IOSGW (St. 8) than in the NW Pacic samples (Sts. 2 and 3 in
Fig. 3c). This can be due to the fact that the Pacic stations were under
the inuence of the Kuroshio Current, which carried higher
I levels
released from the Tokai-mura reprocessing facility (Livingston and
Povinec, 2000; Povinec et al., 2010), while the IOSG St. 8 show global
I concentrations.
The differences between the
H and
I levels observed in the
IOSGW and in NW Pacic waters may indicate that a source of these
radionuclides need not be only NW Pacic. Surface Tasman waters
may ow northward along the Australian coast (Ridgway and Dunn,
2007), and supply IOSGW via the Indonesian Throughow with
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)
Cs data
collected along 20S in the Indian Ocean (Povinec et al., submitted for
publication), and along 30S in the Pacic Ocean (Aoyama et al.,
submitted for publication). However, the observed
H concentrations
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 Pacic Ocean Subtropica Gyre (Mittelstaedt et al., 1999;
Nakano and Povinec, 2003). Therefore the
I levels in the IOSGW
should be lower than in NW Pacic waters, where they were
inuenced by radioactive waste discharges from the Tokai-mura
reprocessing plant.
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
H levels
around 30S, comparable with those presented in Fig. 6a. High surface
Hlevels (0.6-0.7 TU, decay corrected to 1999) were also measured at
34S, 57- 62E during the WOCE project. This is consistent with
concentration observed at St. 1 (0.720.03 TU at 3242S, 7000E;
Table 1), located inside the loop formed by the IOSG. High surface
levels (1.1-1.2 TU), observed at Sts. 2, 5 and 8 (located between 40
and 43S), suggest that these stations were close to the main streamof
the IOSG. It is evident that the gyre ow is strongest at 40-43S, west
of 65E, where highest
H levels were observed. This is a clear
33.0 33.5 34.0 34.5 35.0 35.5 36.0
St. 8
St. 7
St. 3
St. 1
St. 2
St. 6
St. 4
St. 5



-0.9 -0.7 -0.5 -0.3 -0.1 0.1 0.3 0.5
St. 8
St. 7
St. 3
St. 1
St. 2
St. 6
St. 4
St. 5
Fig. 4.
O vs. salinity (a) and
O (b) plots for all stations visited.
are expressed in per mill relative to VSMOW (Vienna Standard Mean Ocean Water).
Relative precisions (at 1 ) are 1 for
H, and 0.1 for
O. Coefcient of
determination, R
=0. 86 (P value b0.0001), indicates that there is a strong correlation
O and salinity. All experimental data are well below the Global Meteoric
Water Line (GMWL) dened 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.
-0.6 -0.5 -0.4 -0.3 -0.2 -0.1 -0.0 0.1 0.2 0.3 0.4 0.5 0.6
St. 8
St. 7
St. 3
St. 1
St. 2
St. 6
St. 4
St. 5

3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
St. 8
St. 7
St. 3
6 100
4300 2500
Oxygen (mL/L)




Fig. 5.
H vs.
O (a) and
I 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.
20 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
inuence of the IOSG on the downstream radionuclide transport from
its western boundary associated with the Agulhas Retroection
(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 40-
50S (International Atomic Energy, 2005; Weiss and Roether, 1980).
Fig. 6. Spatial distribution of
H in surface waters (a) and deep waters (along 60E) (b) at Sts. 1- 8 (x-signs), combined with nearest WOCE stations (dots) (www.eWOCE.org), decay
corrected to January 1999. Higher
Hconcentrations were observed at the main stream of the Indian Ocean Subtropical Gyre at ~40S (as well as at ~20S, as indicated by the WOCE
data). Downward tritium transport around 40S can be seen as well. Higher
H levels observed below 3000 m are associated with Antarctic Bottom Water (AABW).
21 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
H water proles across the basin (Fig. 6b), combined with
WOCE data (www.eWOCE.org) southof 34S (decay correctedto1999),
show higher surface and subsurface
H levels at 20S and 40S 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
C in surface waters (a) and deep waters (along 60E) (b) at Sts. 1- 8 (x-signs), combined with nearest WOCE stations (dots) (www.eWOCE.org). Higher
C concentrations were observed at the main stream of the Indian Ocean Subtropical Gyre at ~40S (as well as at ~20S, as indicated by the WOCE data). Downward
C transport
around 40S can be seen as well.
22 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
higher tritiumlevels can be seen in the transect around 40S and 60E at
water depths below 3000 m which should be due the advection of
5.2.2. Radiocarbon
Combining the
C data presented in this work together with
WOCE data (www.eWOCE.org) we constructed a
C distribution
map for the southern Indian Ocean (Fig. 7a). We can see high
concentrations observed along the subtropical gyre (20 - 40S) at the
top 500 m, and sharp decrease south of 40S, as conrmed by
H data
as well (Figs. 3 and 6).
C water prole data (WOCE and ANTARES IV) presented in
Fig. 7b showa clear penetration of bomb
C around 40S (transect at
60E) 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.
concentrations largely decrease below the 1000 m water depth at
which CDW (Circumpolar Deep Water) appears, formed by the deep-
water circulation and ventilation south of 40S (Fine, 1993; You,
2000). The depletion of
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 ows into
the southwest Indian Ocean, south of Africa between 35S and 40S.
One branch of NADW then moves to the Southwest Indian Ridge
(Toole and Warren, 1993). Considering very low
Hconcentrations (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 CDWby 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
H minima.
Except for the AF, STF and SAF, which control radionuclide
distribution in surface and subsurface waters, several other water
masses can be identied in the basin using isotopes (Fig. 5a, b). The
intermediate layer (300-500 m) with relatively low
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 princi-
pal. 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
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
Hconcentrations observed in
deep waters at St. 8, as the NIDW originates from the northern Indian
Ocean where comparable
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 Throughow region, where deep tidal
mixing could incorporate the tritium-rich Pacic water. Lower
levels found at 2000 m water depth may be due to intrusion of NADW
carrying low radionuclide concentrations.
H concentrations were also observed in bottom
waters at Sts. 3, 7 and 8 (Fig. 3a), as well as at the WOCE stations at
34S (www.eWOCE.org). As sinking of surface water to the bottom at
mid-latitudes is impossible due to strong thermohaline stratication
(You, 2000), an injection of AABW might be considered as the most
plausible source of
H (and low levels of
I, Fig. 3c) in bottom
waters. In fact, belowthe 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 ows 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
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
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
H levels found in bottom waters at Sts. 3, 7 and 8. As
the NADW has very low
H concentrations, and the NIDW is located
well above the bottomlayer, the only plausible explanation for higher
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
H and
O in comparison with surface waters, as
shown in Fig. 4b.
I data (Fig. 3c) have also been used for the identication of
water masses present in the Crozet Basin. The results plotted in Fig. 5b
as the
I concentration vs. dissolved oxygen showthe presence of all
water masses as identied in Fig. 5a with
H data. The surface waters
representing fronts (AF, STF and SAF) have highest
I concentra-
tions. The intermediate waters, represented by AAIW, show very
I concentrations, thus eliminating the differences observed
in the surface layer. The same is true for the deep and bottom waters.
I levels observed at St. 8 at 1500 and 2500 m water depths
are probably due to the presence of NIDW. The lowest
I levels found
at 1000 mwater depth may be due to intrusion of NADWcarrying low
radionuclide concentrations. The separation of NIDW from NADW is,
however, not so clear as it was presented in the case of
H (Fig. 5a).
Low, but measurable
I 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

i =1
i +1
+ W

i +1

+ 2W
+ 2W

( )
where I
is the inventory of the radionuclide R in the seawater column
), N is the number of sampling depths, W
is the radionuclide
concentration in seawater at depth i, d
is the i-th sampling depth of
seawater, and d
is the total water depth to the bottom. The inventories
are giveninBq/m
H: 1 TU=0.118 kBq/m
, andfor
I: 10
L=1.4 Bq/m
Under normal conditions the radionuclide water column inven-
tories should primarily depend on the geographical position of the
sampling station, especially on its latitude. In this case the dominant
23 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
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 Pacic via the Indonesian Seas, and its accu-
mulation in the subtropical gyre should increase their inventories
in the gyre. We shall focus on the evaluation of
H and
I 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
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
Hinventory (33030 kBq/m
) was observed at
St. 8 (430S; 6232E), situated close to the southern gyre stream. In
contrast, the Sts. 3 and 7 situated more south showed lower
inventories (around 60 kBq/m
). This difference in the
H inventories
is mainly caused by differences in water fronts present in the region.
H inventories in top 500 m for SASW (St. 3) and STSW (St. 7)
were 131 kBq/m
and 252 kBq/m
, respectively, when compared
with 788 kBq/m
observed for IOSGW(St. 8). A similar trend can be
also seen for AAIW: at Sts. 3 and 7 the
H inventories were 10
1 kBq/m
and 121 kBq/m
, respectively, while the inventory at St.
8 was 591 kBq/m
. The bottom waters at all three stations were
under the inuence of AABW showing similar
H inventories: 14
1 kBq/m
for St. 3 and 7, and 182 kBq/m
for St. 8.
H inventories estimated for the Crozet Basin can be
compared with WOCE (www.eWOCE.org) and GEOSECS (stlund
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 (33S, 48E), St. 39 (34S, 53E), St.
44 (34S, 57E) and St. 50 (34S, 62E) showed
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
H concentration
comparable with the WOCE stations. The GEOSECS (1978) sampling
St. 428 (38S, 58E; 1300 m water depth) located inside the IOSG loop
(the closest station to St. 8) showed
H inventory of 556 kBq/m
(data were decay corrected to 1999), what is lower than the inventory
observed at St. 8 (for the same water depth), situated, however, at
43S on the main IOSG loop.
The estimated
H inventories in the southern Indian Ocean are
higher at least by a factor of three than expected fromglobal fallout for
similar latitude belts. They are also higher than
H inventories
estimated for the northern Indian Ocean (Mulsow et al., 2003), and
comparable (for the normalized water depth of 1000 m) with
inventories in the western Pacic (Povinec et al., 2010). The
inventories conrm 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.
I inventories in the water columnof the Crozet Basin showsimilar
features as the
Hinventories, i.e. a higher inventoryat St. 8(0.420.08
), and smaller inventories at Sts. 3 and 7 (0.300.06 mBq/m
and 0.310.06 mBq/m
, respectively). Unfortunately, there are no
more data available for
I in the water column of the southern Indian
Ocean. Therefore, the only comparison we can do is with
I data from
the IAEA97 cruise in the NW Pacic Ocean (Povinec et al., 2010). The
I inventories at Sts. 2 and 3 in the NWPacic Ocean (384 mBq/m
and 515 mBq/m
, respectively) are about two orders of magnitude
higher than the inventory at St. 8. The difference is, as we already
pointedout, due to
I releases fromtheTokai-murareprocessingplant.
The estimated radionuclide inventories conrm 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
I 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.
I activity ratios in surface waters (b100 m) of Sts. 3, 7 and
8 are (3.20.5)10
, (5.30.8)x10
and (13.72.1)10
, 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 (b0.8x10
), St. 8 shows the highest ratio
, associated with
I minimum. Waters in the depth
interval of 1500-2500 shows measurable
I ratio (61)10
at St. 8, associated with NIDW, while for Sts. 3 and 7 we can report
only limit (b0.3x10
), associated with NADW. Bottom waters
associated with AABW show comparable ratios from (41)10
St. 3) to (61)10
(for St. 8).
We can also compare
Sr activity ratios in surface waters of
the Crozet Basin where
Sr (half-life 28.78 y) data are available (Lee
et al., 2009). While IOSG Sts. 2, 5 and 8 show comparable activity
ratios (13020, 11015, and 13020, respectively), St. 1 located in
the IOSG loop shows a lower ratio (749). The STF St. 7 gives the
ratio of 8913 and Sts. 3, 4 and 6 of SAF show the ratios of 18030,
29050 and 300140, respectively. The differences are mainly due
to lower
Sr concentrations observed at the SAF stations, however,
they are not signicant as the precision of
Sr measurements was
lower (Lee et al., 2009) than of the
H measurements.
6. Conclusions
Several observations have been made in this study which can be
summarized as follows:
(i) By comparing observed radionuclide (
C, and
I) levels
in the southern Indian Ocean with those found in the western
Pacic Ocean it has been found that they are of common origin
due to the transport of water masses from the western Pacic
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 con-
straints given by different water masses present in the region.
(iii) Highradionuclide concentrations observedintheIndianOceanin
the latitudinal belt of 20-40S are associated with Indian Ocean
Subtropical Gyre which acts as a reservoir of radionuclides.
(iv) High
C levels observed in bottom waters for transect around
40S and 60E document that the southern Indian Ocean is
important for sink of anthropogenic carbon.
(v) While latitudinal (32 - 48S) 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 inuenced 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
The southern Indian Ocean appears to have been acting on a time
scale of several decades as a nal reservoir of contaminants
transported from the northern Indian Ocean and central western
Pacic 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 reects the complex dynamics and advection of
24 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
different water masses, which makes the basin one of the most
interesting oceanographic places in the World Ocean.
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 Interna-
tional Atomic Energy Agency is grateful to the Government of the
Principality of Monaco for support provided to its Marine Environ-
ment Laboratories. PPP acknowledges support provided by the Slovak
Scientic Agency VEGA (grant No. 1/108/08) and the EU Research &
Development Operational Program funded by the ERDF (project No.
Aoyama, M., Hirose, K., Igarashi, Y., 2006. Re-construction and up-dating our
understanding on the global weapons tests
Cs fallout. J. Environ. Monit. 8,
Aoyama, M., Hirose, K., Hamajima, Y., Komura, K., Povinec, P.P., Sanchez-Cabeza, J.A.,
2010. Long term behaviour of
Cs in the Pacic 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 Pacic. Radiocarbon 43, 857867.
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, 379389.
Bayer, R., Schlosser, P., 1991. Tritium proles in the Weddel Sea. Mar. Chem. 35,
Belkin, I.M., Gordon, A.L., 1996. Southern Ocean fronts from the Greenwich Meridian to
Tasmania. J. Geophys. Res. 101, 36753696.
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, 27772784.
Bowen, V.T., Noshkin, V.E., Livingston, H.D., Volchok, H.L., 1980. Fallout radionuclides in
the Pacic Ocean; vertical and horizontal distributions, largely from GEOSECS
stations. Earth Planet. Sci. Lett. 49, 411434.
Broecker, W.S., 1991. The great ocean conveyor. Oceanography 4, 7989.
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, 1433114344.
Burchuladze, A.A., Chud, M., Eristavy, I.V., Pagava, S.V., Povinec, P., ivo, A., Togonidze,
G.I., 1989. Anthropogenic
C variations in atmospheric CO2 and wines.
Radiocarbon 31, 771776.
Clarke, W.B., Jenkins, W.J., Top, Z., 1976. Determination of tritium by mass-spectrometric
measurements of
He. Int. J. Radiat. Isot. 27, 512522.
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, 993995.
Cooper, L.W., Hong, G.-H., Beasley, T.M., Grebmeier, J.M., 2001. Iodine-129 concentrations
inmarginal seas of the North Pacic and Pacic-inuencedwaters of the Arctic Ocean.
Mar. Poll. Bull. 42, 13471356.
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
Th. Deep Sea Res. I 52, 5168.
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, 299313.
Craig, H., 1961. Isotopic variation in meteoric waters. Science 133, 17021703.
Davis, R.E., 2005. Intermediate-depth circulation of the Indian and South Pacic Oceans
measured by autonomous oats. J. Phys. Oceanogr. 35, 683707.
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, 224228.
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, 68816894.
Epstein, S., Mayeda, T., 1953. Variation of O
content of waters from natural sources.
Geochim. Cosmochim. Acta 4, 213224.
Fine, R.A., 1985. Direct evidence using tritiumdata for throughowfromthe Pacic into
the Indian Ocean. Nature 315, 478480.
Fine, R., 1993. Circulation of Antarctic intermediate water in the South Indian Ocean.
Deep-Sea Res. 40, 20212042.
Ganachaud, A., Wunsch, C., 2000. Improved estimates of global ocean circulation, heat
transport a mixing from hydrographic data. Nature 408, 453457.
Gonantini, R., 1978. Standards for stable isotope measurements in natural compounds.
Nature 271, 534536.
Gordon, A.L., 1985. Indian-Atlantic Transfer of Thermocline Water at the Agulhas
Retroection. Science 227 (4690), 10301033.
Gordon, A.L., Fine, R.A., 1996. Pathways of water between the Pacic and Indian Oceans
in the Indonesian seas. Nature 379, 146149.
Gordon, A.L., Susanto, R.D., Vranes, K., 2003. Cool Indonesian througows as a
consequence of restricted surface layer ow. Nature 425, 824828.
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, 118125.
Haine, T.W.N., Watson, A.J., Liddicoat, M.I., 1998. Dickson, R.R., 1998. The ow of
Antarctic bottom water to the southwest Indian Ocean estimated using CFCs. J.
Geophys. Res. 103, 2763727654.
Hong, G.-H., Lee, S.H., Kim, S.H., Chung, C.S., Baskaran, M., 1999. Sedimentary uxes of
Sr- 90, Cs-137, Pu-239, Pu-240 and Pb-210 in the East Sea (Sea of Japan). Sci. Total
Environ. 238, 225240.
Hou, X., Hansen, V., Aldahan, A., Possnert, G., Lind, O.C., Lujanien, G., 2009. A review on
speciation of
I in environmental and biological samples. Anal. Chim. Acta 632,
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, 4546.
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. 241262.
Key, R., 1996. WOCE pacic ocean radiocarbon program. Radiocarbon 38, 415423.
Key, R., Quay, P., Jones, G., McNichol, A., von Reden, K., Schneider, R., 1996. WOCE AMS
radiocarbon.1. Pacic Ocean results (P6, P16 and P17). Radiocarbon 38, 425518.
Key, R., Quay, P., Schlosser, P., McNichol, A., von Reden, K., Schneider, R., Elder, K.,
Stuiver, M., Ostlund, H., 2002. WOCE radiocarbon IV: Pacic Ocean results; P10,
P13N, P14C, P18, P19 & S4P. Radiocarbon 44, 239392.
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
at IsoTrace. Nucl. Instr. Meth. B29, 7276.
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, 427436.
Lee, S.-H., Povinec, P.P., Wyse, E., Pham, M.K., Hong, G.-H., Chung, C.S., Kim, S.-H., Lee, H.J.,
2005. Distributionandinventories of
AmandPuisotopes insediments of
the Northwest Pacic Ocean. Mar. Geol. 216, 249263.
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 OceanANTARES IV
Results. J. Oceanogr. 65, 397406.
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, 6980.
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,
Livingston, H.D., Povinec, P.P., 2000. Anthropogenic marine radioactivity. Ocean Coast.
Manage. 43, 689712.
Livingston, H.D., Povinec, P.P., 2002. A millennium perspective on the contribution of
global fallout radionuclides to ocean science. Health Phys. 82, 656668.
Livingston, H.D., Povinec, P.P., Ito, T., Togawa, O., 2001. The behaviour of plutonium in
the Pacic Ocean. In: Kudo, A. (Ed.), Plutonium in the Environment. Elsevier,
Amsterdam, pp. 267292.
Metzl, N., Tilbrook, B., Poisson, A., 1999. The annual fCO
cycle and the air-sea CO
in the sub-Antarctic Ocean. Tellus 51B, 849861.
Michel, R.L., 1978. Tritium distributions in Weddell Sea water masses. J. Geophys. Res.
83, 61926198.
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), 301310.
Morgenstern, U., Taylor, C.B., 2009. Ultra Low-level tritium measurement using
electrolytic enrichment and LSC. Isot. Environ. Health Stud. 45, 96117.
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, 27612775.
Nakano, M., Povinec, P.P., 2003. Oceanic general circulation model for the assessment of
the distribution of
Cs in the world ocean. Deep Sea Res. II 50, 28032816.
Nydal, R., Lvseth, K., 1965. Distribution of radiocarbon from nuclear tests. Nature 206,
stlund, H.G., Brescher, R., 1982. Tritium Laboratory data Report No. 2., University of
Miami, Miami, USA.
Park, Y.-H., Gambroni, L.L., Charriaud, E., 1993. Frontal structure, water masses and
circulation in the Crozet Basin. J. Geophys. Res. 98, 12,36112,385.
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, 18231842.
Pham, M.K., Betti, M., Povinec, P.P., Almov, V., Biddulph, D., Gastaud, J., Kieser, W.E.,
Lpez Gutirrez, J.M., Possnert, G., Sanchez-Cabeza, J.A., et al., 2010. Certied
25 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426
reference material IAEA-418:
I 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, 32893305.
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, 19151929.
Pollard, R.T., et al., 2009. Southern Ocean deep-water carbon export enhanced by
natural iron fertilization. Nature 457, 577589.
Povinec, P.P., 2004. Developments in analytical technologies for marine radioactivity studies.
In: Livingston, H.D. (Ed.), marine radioactivity. Elsevier, Amsterdam, pp. 237294.
Povinec, P.P., 2005. Ultra-sensitive radionuclide spectrometry: Radiometrics and mass
spectrometry synergy. J. Radioanal. Nucl. Chem. 263, 413417.
Povinec, P.P., Oregioni, B., Jull, A.J.T., Kieser, W.E., Zhao, X.-L., 2000. AMS measurements
C and
I in seawater around radioactive waste dump sites. Nucl. Instr. Meth.
B172, 672678.
Povinec, P.P., Badie, C., Baeza, A., Barci-Funel, G., Bergan, T.D., Bojanowski, R., Burnett, W.,
Eikenberg, J., Field, 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. Certied reference
material for radionuclides in seawater IAEA-381 (Irish Sea water). J. Radioanal. Nucl.
Chem. 251, 369374.
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 Pacic Ocean results of oceanographic and radionuclide
investigations of the water column. Deep-Sea Res. II 50, 26072638.
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, 27512760.
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 Pacic and
Indian Oceans GLOMARD database. J. Environ. Radioactiv. 76, 113137.
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 Pacic Ocean 24 years
after GEOSECS. Radiocarbon 46, 583594.
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 Pacic and Indian
Oceans. WOMARS results. J. Environ. Radioactiv. 81, 6387.
Povinec, P.P., Betti, M., Jull, A.J.T., Vojtyla, P., 2008. New isotope technologies for
environmental physics. Acta Phys. Slovaca 58, 1154.
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,
Sr and
I in
the Pacic and Indian Oceans. Nucl. Instrum. Meth. Phys. Res. B 268, 12141218.
Povinec, P.P., M. Aoyama, M. Fukusawa, K. Hirose, K. Komura, J.-A. Sanchez-Cabeza, J.
Gastaud, M. Jekovsk, I. Levy-Palomo, I. Skora, 2010b. Proles of
Cs in South
Indian Ocean water along the 20S 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, 207214.
Rahmstorf, S., 2006. Thermohaline Ocean Circulation. In: Elias., S.A. (Ed.), Encyclopedia
of Quaternary Sciences. Elsevier, Amsterdam.
Raisbeck, G., You, F., 1999.
I in the oceans: origins and applications. Sci. Total Environ.
237 (238), 3142.
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.
I as a new environmental tracer or
geochronometer for biogeochemical or hydrodynamic processes in the hydro-
sphere and geosphere: The central role of organo-iodine. Sci. Total Environ. 321,
Schlosser, P., Bayer, R., Boenisch, G., Cooper, L.W., Ekwurzel, B., Jenkins, W.J., Khatiwala, S.,
Prman, 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), 1530.
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. 431452.
Schmitz, W.J., 1995. On the interbasin-scale thermohaline circulation. Rev. Geophys. 33,
Speich, S., Blanke, B., Cai, W., 2007. Atlantic meridional overturning circulation and the
Southern Hemisphere supergyre. Geophys. Res. Lett. 34, L23614. doi:10.1029/
Stuiver, M., Ostlund, H.G., 1983. GEOSECS Indian Ocean and Mediterranean radiocar-
bon. Radiocarbon 25, 129.
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, 19732019.
Top, Z., 1999. Tritium inventory of the Black Sea. J. Environ. Radioactiv. 43, 221229.
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, 622637.
Weiss, W., Roether, W., 1980. The rates of tritium input to the world oceans. Earth
Planet. Sci. Lett. 49, 435436.
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,
26 P.P. Povinec et al. / Earth and Planetary Science Letters 302 (2011) 1426