ARTICLE IN PRESS
Deep-Sea Research I 51 (2004) 71–82
Denmark Strait water circulation traced by heterogeneity in
neodymium isotopic compositions
Francois Lacan*, Catherine Jeandel
LEGOS, Observatoire Midi-Pyr!en!ees, 18 av Edouard Belin, Toulouse 31401, Cedex 4, France
Received 27 May 2003; accepted 5 September 2003
Abstract
Seawater neodymium isotopic composition (eNd ) and rare earth element (REE) concentrations were measured along
four hydrologic sections within the East Greenland Current (EGC), between the Fram and the Denmark Straits, during
summer 1999. EGC intermediate waters, between 77 N and 70 N, displaying similar hydrological characteristics as
Denmark Strait Overflow Water (DSOW), had very constant REE characteristics and eNd value of 10.9. This
constancy reveals the absence of lithogenic input, from the East Greenland margin, into intermediate waters in this
area, corroborating previous dissolved aluminum data. The DSOW was characterized by eNd ¼ 8:471:4: This value
can be explained by the imprint of lithogenic formations, mainly basaltic, bordering the Denmark Strait on the
intermediate waters described above. However, granitic Precambrian formations seem to contribute, although more
slightly, to defining the DSOW Nd signature. This double influence could explain the heterogeneity of the Denmark
Strait waters. These results provide a better understanding of the DSOW Nd signature, allowing a better use of this
tracer in the study of present and past North Atlantic Deep Water dynamics. Atlantic Water was present at a station
located at the mouth of Nansen Fjord, on the western side of Denmark Strait. We suggest that this water reaches the
fjord intermittently as isolated water lenses or eddies detached from the northward flowing branch of the Irminger
current.
r 2003 Elsevier Ltd. All rights reserved.
Keywords: Denmark Strait; Circulation; Neodymium; Water masses; Weathering; Isotopic tracer; Denmark Strait overflow water
1. Introduction
Neodymium (Nd) is a rare earth element (REE).
At the earth surface it occurs mainly in the solid
phase. Its isotopic composition (Nd IC) is the
143
Nd/144Nd ratio, expressed as eNd ; which is the
deviation of this ratio in the sample from a
*Corresponding author. Present address: Woods Hole
Oceanographic Institution, Woods Hole, MA 02543, USA.
Tel.: +1-508-289-3798; fax: +1-508-457-2193.
E-mail address: fl[email protected] (F. Lacan).
reference value (CHUR, Chondritic Uniform
Reservoir, 143 Nd=144 Nd ¼ 0:512638; which represents the present day average earth value):
143
ð Nd=144 NdÞmeasured
eNd ¼
1
104 :
ð143 Nd=144 NdÞCHUR
The Nd IC of a rock depends on both its age
and its samarium to Nd ratio. eNd ranges from 40
in old continental formations (e.g. Greenland,
Taylor et al., 1992) to +10 in new volcanic
.
formations (e.g. Iceland; O’Nions and Gronvold,
0967-0637/$ - see front matter r 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.dsr.2003.09.006
ARTICLE IN PRESS
72
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
1973). The Nd found in the ocean is lithogenic (i.e.
derived from the lithosphere, including continental
shelf and slope sediments supplied by erosion). It
is brought to the ocean in dissolved or particulate
phases by rivers, the atmosphere or remobilization
from margin sediments. Within the ocean, most of
the Nd is found in the dissolved phase; the
particulate phase represents only 5–10% of the
total content (Jeandel et al., 1995). Although input
processes are still not completely understood, it
seems that dissolved/particulate exchanges at the
margins play a preponderant role (Jeandel et al.,
1998; Lacan and Jeandel, 2001; Tachikawa et al.,
2003). In the vicinity of lithogenic sources, eNd is
thus used as a tracer of particle transport and
dissolved-particulate exchanges (Grousset et al.,
1988; Jeandel et al., 1995; Tachikawa et al.,
1999a, b). Away from them, it is conservative
and used as a tracer of water masses (Jeandel,
1993; Jeandel et al., 1998; Lacan and Jeandel,
2001; Piepgras and Jacobsen, 1988; Piepgras and
Wasserburg, 1987). Nd residence time in the ocean
is around 500–1000 years (Tachikawa et al.,
1999a, 2003). That makes eNd suitable for tracing
intra- and inter-ocean currents.
eNd is widely used in paleo-oceanography. Signal
variations stored in sediments (mostly metalliferous) are classically interpreted as reflecting past
variations in circulation. Recently, several studies
have suggested that fluctuations in the Nd sources,
via variations in erosion rates, could also affect eNd
records (Frank et al., 2001; Tachikawa et al., 2003;
Vance and Burton, 1999).
The Nd IC of the North Atlantic Deep Water
(NADW) has been particularly well studied in
paleoceanography (Abouchami et al., 1999; Burton et al., 1997; Rutberg et al., 2000; Vance and
Burton, 1999). The production of this water mass
initiates the conveyor belt circulation and thus is a
key factor controlling our climate (Broecker,
1991). NADW is formed out of four sources:
Denmark Strait Overflow Water (DSOW), Island
Scotland Overflow Water (ISOW), Labrador Sea
Water (LSW) and Lower Deep Water (LDW,
derived from Antarctic Bottom Water, AABW).
Those are formed by convection in high latitudes:
the DSOW and the ISOW in the Nordic Seas
(Greenland, Island and Norwegian Seas), the LSW
in the Labrador Sea and the AABW around the
Antarctic continent (Dickson and Brown, 1994).
The few data available before this work show that
the Nd ICs of these sources are very distinct:
eNd E 8 in the Nordic Seas overflow waters,
eNd E 14:7 in the LSW and eNd E 11:7 in the
AABW (Piepgras and Wasserburg, 1987). eNd
variations in the past or present NADW could
thus be used to evaluate the relative contribution
of each of its source waters and to draw
conclusions about fluctuations in the conveyor
belt activity, if it can be ascertained that the endmember values are constant.
To further develop this potential application of
eNd ; we need to better understand how these waters
acquire their signature. So far, only three Nd IC
measurements were available for the Nordic Seas,
which is insufficient to constrain the signature of
the DSOW and ISOW. One of the goals of the
Signature/GINS cruise (IFRTP) was to further
address these questions. The cruise was carried out
on board the R/V Marion Dufresne, in July and
August 1999, in the North Atlantic and more
particularly in the Greenland and Iceland Seas.
Hydrological parameters (y; S; [O2] and currents)
and chemical tracers (CFC, SF6, 129I, all REE and
Nd IC) were measured. In this paper, we present
Nd ICs and concentrations of seawater samples
from the East Greenland Current (EGC), along
the east Greenland slope, between the Fram and
Denmark Straits. We use these results to explain
the eNd signature of the DSOW and to better
constrain the circulation in the Denmark Strait.
2. Hydrological context, sampling, analytical
procedures
The Nordic Seas are those lying between
Greenland and Norway: the Norwegian, the
Greenland and the Iceland Seas (Fig. 1). Their
deep openings with the oceans are the Fram Strait
in the north, the Denmark Strait in the southwest
and the Faroe Shetland Channel in the southeast.
Their main hydrological structures are (1) an
eastern boundary current, the Norwegian Atlantic
Current, which continues as the West Spitsbergen
Current to Fram Strait and the Arctic Ocean,
ARTICLE IN PRESS
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
73
Fig. 1. Studied area. The large black arrows represent the main hydrodynamic structures: the EGC, the Greenland Sea and Iceland Sea
gyres. Black white-edged segments indicate hydrological sections. The dashed lines bordering the coasts indicate their geological type.
The white black-striped lines indicate granitic Precambrian formations (eNd around 35), whereas the black white-striped lines indicate
basaltic ones (eNd around +8). Note the basaltic nature of the formations bordering the Denmark Strait.
bringing warm and saline Atlantic Water (AW) to
the north; (2) a western boundary current, the
EGC, carrying waters from the Fram Strait,
through the Greenland and Iceland Seas, towards
the Denmark Strait; and (3) three cyclonic gyres
in the Greenland, Iceland and Norwegian Seas
(Fig. 1; Hansen and Østerhus, 2000; Poulain and
Warn-Varnas, 1996).
The EGC flows parallel to the isobaths. It
extends from the continental shelf beyond the
slope into the deep (>3000 m) basin. Its mean
annual transport is 2173 Sv (1 Sv=106 m3 s1;
Woodgate et al., 1999) which includes the strong
internal cyclonic circulation of the Greenland Sea.
It carries waters of different origins, some from the
Arctic Ocean, some from the Atlantic Ocean by
way of the Norwegian Atlantic Current and the
West Spitsbergen Current, and some from the
Greenland and Iceland gyres (Rudels et al.,
1999a, b).
Seawater samples were collected along four
sections within the EGC, above the slope, between
the Fram and Denmark Straits (Fig. 1). A station
was also occupied at the mouth of the Nansen
Fjord, in the northwest of Denmark Strait. We
analyzed the Nd ICs and REE concentrations of
unfiltered samples in most cases. Some surface and
bottom samples were filtered through 0.45 mm
filters. The analytical procedures are described in
Lacan and Jeandel (2001). The reproducibility of
the Nd IC measurement was 0.5 eNd unit, blank
values were 700 pg (3.5% of the most depleted
sample and 2% on average). Reproducibility of
the REE concentration measurements was better
than 10% for all REE and better than 5% for Nd,
blanks values were better than 8% for all REE and
better than 3% for Nd.
3. Denmark Strait Overflow Waters
3.1. East Greenland Current
The DSOW are the densest waters overflowing
the Denmark Strait. Their origins are the topic of
numerous studies and still unclear. In addition to
earlier hypotheses, most of the recent publications
suggest that the precursors of the DSOW are
carried within the EGC from the north of the
Nordic Seas (Mauritzen, 1996; Rudels et al.,
1999a, b; Strass et al., 1993).
During the Signature/GINS campaign, the
DSOW had the following hydrological characteristics: 0:5oyo0 C, 34:87oSo34:90 and
28:01osy o28:05 kg m3. Those values were chosen because they characterized the deepest water
mass of the strait, separated from the overlying
layer by marked temperature, salinity and density
gradients. We determined the Nd ICs of seawater
samples having similar characteristics within the
EGC upstream of the Denmark Strait (Table 1).
74
Table 1
Physical (y; S; sy ), chemical ([O2]) and geochemical ([Nd], eNd ) data used in this work. All data come from the Signature/GINS cruise (summer 1999)
y ( C)
S
sy
(kg m3)
O2
(mg l1)
[Nd]
(ppta)
eNd
0.69
0.69
33.16
33.16
26.65
26.65
13.37
13.37
3.6
3.4
9.2
10.3
0.2
0.2
21 (filtered)
21
123
208
482
608
805
1840 (filtered)
1840
1.25
1.25
3.12
2.60
1.33
0.61
0.14
0.74
0.74
33.24
33.24
34.89
34.96
34.92
34.88
34.89
34.92
34.92
26.74
26.74
27.79
27.89
27.96
27.98
28.01
28.07
28.07
12.23
12.23
9.48
9.95
10.02
9.78
10.03
9.73
9.73
—
3.4
2.8
2.6
2.5
2.5
—
—
—
9.5
9.8
11.5
11.6
11.8
10.7
11.2
10.1
10.7
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
5
21 (filtered)
103
405
505
2016
2512 (filtered)
2512
4.60
4.56
2.21
0.20
0.00
1.00
1.06
1.06
34.68
34.72
34.97
34.90
34.89
34.90
34.90
34.90
27.47
27.50
27.93
28.01
28.02
28.07
28.08
28.08
8.90
7.51
10.30
10.53
10.64
9.85
9.94
9.94
2.5
2.3
2.6
—
2.4
—
2.6
2.8
11.5
12.2
11.3
11.2
11.2
10.5
10.2
10.8
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
50
900
0.94
0.46
34.33
34.89
27.61
28.04
11.67
10.09
3.2
2.4
10.9
10.6
0.2
0.2
2000
754
0.46
34.89
28.04
9.98
2.6
—
72 550 N,
12 580 W
2690
707
2635
0.54
1.12
34.88
34.90
28.04
28.08
10.38
10.00
2.6
3.1
10.9
10.2
0.2
0.3
70 000 N,
18 000 W
1655
174
404
573
1007
1596
0.75
0.16
0.35
0.67
0.91
34.83
34.90
34.89
34.90
34.91
27.92
28.01
28.04
28.06
28.08
10.49
10.35
10.20
9.90
9.76
2.5
2.4
2.4
2.5
3.8
—
—
10.6
10.7
11.4
—
—
0.3
0.2
0.2
Bottom
depth (m)
SGN 33 (77 N)
77 110 N,
04 230 W
1575
SGN 32 (77 N)
77 020 N,
03 450 W
1895
SGN 30 (77 N)
76 450 N,
02 200 W
2560
SGN46 (73 N)
72 520 N,
16 110 W
1520
SGN45 (73 N)
72 520 N,
15 510 W
SGN42 (73 N)
SGN50 (70 N)
Pressure (dbar)
20
50
2s (eNd )
—
ARTICLE IN PRESS
Location
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
Station
1280
98
200
431
698
1001
0.10
0.34
0.05
0.42
0.65
34.83
34.83
34.88
34.89
34.90
27.97
27.99
28.01
28.04
28.06
11.56
11.39
10.45
10.07
9.89
2.4
2.6
2.4
2.4
2.4
—
—
11.1
10.8
10.8
—
—
0.2
0.2
0.2
SGN52 and
SGN53 (Nansen
Fjord)
68 120 N,
29 380 W
375
6
80
183
200
251
303
2.32
1.35
3.31
4.31
1.02
0.20
30.53
33.13
34.52
34.77
34.56
34.57
24.37
26.65
27.47
27.57
27.69
27.74
11.31
11.97
10.15
9.68
10.46
10.54
13.5
14.6
3.6
3.0
6.8
11.9
4.1
3.1
10.2
10.9
4.3
2.2
0.2
0.2
0.2
0.3
0.2
0.2
SGN 54 (West
Denmark Strait)
66 100 N,
27 310 W
500
5
59
120
352
442
479
3.81
1.41
1.39
0.16
0.03
0.04
31.40
33.65
34.19
34.83
34.87
34.87
24.94
27.07
27.51
27.96
28.01
28.01
11.02
11.59
11.08
10.55
10.29
10.18
6.0
14.7
—
3.5
2.8
3.3
10.5
11.2
11.3
10.7
7.2
9.2
0.2
0.2
0.2
0.2
0.2
0.3
SGN55 (Center
Denmark Strait)
66 050 N,
27 150 W
625
11
51
97
372
550
610
610 (filtered)
5.71
2.62
6.80
0.15
0.12
0.12
0.12
33.45
34.39
35.13
34.84
34.88
34.88
34.88
26.36
27.43
27.55
27.97
28.02
28.02
28.02
10.47
10.76
8.86
10.85
10.24
10.20
10.20
4.0
5.9
2.7
—
2.4
2.5
—
11.0
11.5
13.2
8.8
—
10.0
8.3
0.2
0.4
0.2
0.2
—
0.2
0.2
SGN 56 (East
Denmark Strait)
66 000 N,
26 590 W
560
10
54
120
182
302
345
523
5.16
0.46
3.58
6.29
4.45
0.25
0.51
32.52
33.30
34.40
35.06
34.91
34.53
34.90
25.70
26.75
27.35
27.56
27.66
27.74
28.05
10.79
12.40
10.40
9.20
9.27
10.77
9.99
4.3
4.6
3.0
2.8
2.9
3.0
4.1
10.4
9.3
11.5
12.1
11.1
7.7
7.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
Geochemical measurements were performed on unfiltered samples unless otherwise stated. 2sðeNd Þ is two times the standard deviation of the eNd measurement; it defines
the 95% confidence interval.
a
ppt: part per trillion, 1 ppt=1012 g g1.
ARTICLE IN PRESS
70 000 N,
17 000 W
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
SGN48 (70 N)
75
ARTICLE IN PRESS
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
76
Results are:
77 N section: eNd ¼ 11:170:3 (4 unfiltered
samples)
73 N section: eNd ¼ 10:870:2 (2 unfiltered
samples)
70 N section: eNd ¼ 10:870:2 (5 unfiltered
samples)
They show that these waters, thought to be
sources of the DSOW, have a very precisely
defined Nd IC (eNd ¼ 10:970:2; 11 unfiltered
samples), homogeneous and remarkably constant
all the way from 77 N to 70 N. Nd concentrations
are also remarkably constant: ½Nd ¼ 2:470:1 ppt
(part per trillion, 1 ppt=1012 g g1, 11 unfiltered
samples). The absence of variation of the Nd
signature of these waters reflects the absence of
lithogenic input from Greenland, whose Nd IC is
around –35 (ranging from –15 to –45, Brueckner
et al., 1998; Taylor et al., 1992; Thrane, 2002;
Winter et al., 1997). These results are consistent
with dissolved aluminum measurements carried
out in the same area in 1988–1989 by Measures
and Edmond (1992), who showed that lithogenic
enrichments in this region affect deep, but not
intermediate, waters of the EGC. The constancy of
Nd concentrations (4% standard deviation) and
other conservative REE concentrations (3–7%
standard deviation, cf. electronic supplement)
corroborates these results.
3.2. Denmark Strait
The section located at the Denmark Strait sill
comprises three stations (West, Center and East).
Nd IC profiles are shown in Fig. 2, potential
temperature and salinity profiles in Fig. 3 and the
data are listed in Table 1. We discuss here the
densest waters, which compose the overflow; the
upper parts of these profiles are discussed in
Section 4. Four unfiltered samples were collected
in the DSOW flowing just above the sill: two
samples at station West (479 and 442 m), one
sample at station Center (610 m) and one sample
at station East (523 m). In contrast to the
homogeneity of its precursors in the EGC, DSOW
display heterogeneous Nd ICs, ranging between
Fig. 2. eNd profiles in the Denmark Strait and the Nansen
Fjord. Note the subsurface minima, associated with the AW,
found at all stations except station West. The dotted ellipse
shows the DSOW samples. Note the scattering of the DSOW
data and their rather radiogenic values. The unbroken ellipse
shows the ‘‘r-DSOW’’ (radiogenic-DSOW). The filled circle
represents the filtered bottom sample from station Center.
10.0 and 7.2, with an average value of
8.471.4. Nd concentrations are also heterogeneous, ranging between 2.5 and 4.1 ppt, with an
average value of 3.270.7 ppt. The mean values are
identical to those reported for one sample taken in
1981 downstream of the Denmark Strait
(eNd ¼ 8:670:5; ½Nd ¼ 3 ppt; Piepgras and
Wasserburg, 1987).
Nd ICs and concentrations differ greatly from
those found in the EGC. Two hypotheses can
explain this difference. Either the waters sampled
in the EGC received Nd inputs between 70 N and
the Denmark Strait, or they mixed with waters
coming from the Iceland Sea (as suggested by
Swift et al., 1980). Although most recent publications suggest that the precursors of the DSOW are
carried within the EGC from the north of the
Nordic Seas (Mauritzen, 1996; Rudels et al.,
1999a, b; Strass et al., 1993), the hypothesis of an
Iceland Sea contribution can not be dismissed. As
we could not collect the Iceland Sea end-member,
we are unable to address this question in the
present work. We will therefore assume in the
following that the DSOW originate in the EGC
upstream of 70 N and check if this is consistent
with our data. Consequently, we will estimate if
ARTICLE IN PRESS
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
77
Fig. 3. Potential temperature and salinity profiles in the Denmark Strait and the Nansen Fjord. Note the subsurface potential
temperature and salinity maxima, associated with the AW, found at all stations except station West.
Nd inputs can cause the shifts (1) from eNd ¼
10:9 within the EGC to eNd ¼ 8:4 in the
Denmark Strait and (2) from a strong homogeneity to a pronounced heterogeneity.
This estimation requires knowledge of the
geological nature of the surrounding formations.
Whereas the Greenland coast is composed mostly
of granitic Precambrian materials between the
Fram Strait and 70 N, the EGC becomes bordered
by basaltic structures between 70 N and Denmark
Strait. On its eastern side, these borders are first the
Kolbeinsey ridge (see Fig. 4, eNd E þ 10; Mertz
et al., 1991) and then the Icelandic slope (eNd E þ 8;
O’Nions and Gr.onvold, 1973). On its western side,
it is bordered by the Blosseville coast (eNd E þ 4 to
+6; Bernstein et al., 1998; Hansen and Nielsen,
1999) down to Kangerlussuaq fjord (68 100 N, just
a few kilometers downstream of Denmark Strait,
see Fig. 1), then again by granitic Precambrian
formations (eNd E 35; Taylor et al., 1992).
3.3. Enrichment of the DSOW
3.3.1. Unradiogenic enrichment
At 610 m at the station Center, we measured both
filtered (eNd ¼ 8:3) and unfiltered (eNd¼ 10:0)
samples. This difference could be explained by the
occurrence of particles, resuspended from the
sediment, of which eNd is lower than 10.0.
Unfortunately, we do not have the concentration
of the filtered sample, which prevents us from
calculating the Nd IC of these particles.
At station West, the two deepest samples (480
and 440 m, unfiltered) display strictly identical
hydrological characteristics (y; S; sy and [O2], see
Table 1), indicating that they come from the same
water mass. However, their Nd ICs differ by two
eNd units and their Nd concentrations by 0.5 ppt
(eNd ¼ 9:2 and 7.2; ½Nd ¼ 3:3 and 2.8 ppt, at
480 and 440 m respectively). This suggests that,
compared to the 440 m sample, the 480 m sample
received an input of unradiogenic Nd (i.e. Nd
characterized by relatively negative eNd values),
which is consistent with the relative position of the
two samples; the closest to the bottom being more
likely to be enriched by remobilized sediment. The
isotopic composition of the input Nd can be
calculated from
eNdfinal ½Ndfinal eNdinitial ½Ndinitial
eNdinput ¼
;
½Ndfinal ½Ndinitial
ð1Þ
where ‘‘initial’’ refers to the water mass that did
not undergo lithogenic input, ‘‘input’’ refers to the
ARTICLE IN PRESS
78
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
Fig. 4. Denmark Strait area. The black arrow shows the EGC. The black white-edged segments and circle show the 70 N and
Denmark Strait sections and the Nansen Fjord station, respectively. The white-shaded area schematizes the contact area between the
DSOW and the margins. The white-striped areas indicate basaltic formations, the others being granitic Precambrian ones.
input material and ‘‘final’’ refers to the water mass
that underwent the lithogenic input. This leads to
an input Nd IC of 20. This value is very close to
the measurement of a surface sediment sampled
south of Cape Farewell: eNd ¼ 23 (Sinko, 1994).
It is also in agreement with our deduction that the
station Center bottom sample particles must have
eNd o 10:0:
In view of the geological context, this enrichment could be explained by the presence of
granitic Precambrian formations, located in the
west of the Denmark Strait south of 68 N and
north of 70 N, from where sediments could have
been transported. The influence of these sediments,
added to that of the surrounding basaltic formation, could explain the unradiogenic calculated
values of 20. Assuming values of eNd ¼ 35
and ½Nd ¼ 35 ppm (part per million, 1 ppm=
106 g g1) and eNd ¼ þ8 and ½Nd ¼ 8 ppm for
the granitic Precambrian and basaltic formations
respectively (note that the concentrations mentioned here are for rocks rather than seawater;
Mertz et al., 1991; Bernstein et al., 1998; Thrane,
2002), the input material should be composed of
E30% of granitic Precambrian material. Because
the granitic Precambrian formations occur only in
the west of the strait, these inputs would then be
restricted to the waters flowing at the bottom and
the west of the strait.
3.3.2. Radiogenic enrichment
The filtered sample (bottom of station Center)
together with the samples that did not undergo
unradiogenic enrichment (440 m station West and
bottom of station East) are a relatively homogeneous group (identified as ‘‘r-DSOW’’ in Fig. 2,
‘‘r’’ for radiogenic, i.e. characterized by relatively
high eNd values). These waters reach the entrance
of the strait with a mean Nd IC of 7.670.6 and a
mean concentration of 3.5 ppt while their characteristics at 70 N were eNd ¼ 10:9 and
½Nd ¼ 2:4 ppt. This variation could not result
from water mass mixing within the EGC, since all
samples within the EGC have eNd values between
12.2 and 9.2 (cf. Table 1). As above, it could be
explained by lithogenic input. Eq. (1) yields an
input Nd IC of 0.4, which corresponds to an
input material composed of E95% of basaltic
material. This illustrates that mixing of sediments
with different signatures could result in an heterogeneous Nd imprint on water masses transported
within the strait.
As most of our measurements were performed
on unfiltered samples, we are unable to investigate
ARTICLE IN PRESS
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
thoroughly if these enrichments correspond to
resuspension, dissolution or both. However, from
the single pair of filtered/unfiltered DSOW samples (bottom of station Center), it seems that the
unradiogenic enrichments are rather in the particulate phase, while the radiogenic ones are rather
in the dissolved one.
The good agreement between the average Nd IC
and concentration in the DSOW presented here
(eNd ¼ 8:471:4; ½Nd ¼ 3:270:7 ppt, 4 samples,
taken within the strait) and the data published by
Piepgras and Wasserburg, 1987(eNd ¼ 8:670:5;
½Nd ¼ 3 ppt, 1 sample, taken downstream of the
strait in 1981), suggests that the DSOW signature
becomes homogeneous and keeps the imprint
mainly due to the basaltic formations as the water
mass sinks from the Denmark Strait sill along the
east Greenland slope, down to the Irminger Sea
depths. Homogenization could result from strong
mixing within the overflow.
3.3.3. Fluxes required to account for Nd
enrichments
In the following calculations we assess whether
the fluxes required to increase the Nd concentration between the EGC upstream of 70 N (2.4 ppt)
and the mean value in the Denmark Strait (3.2 ppt)
are realistic. Recent estimations of the DSOW flux
based on in situ high resolution current measurements suggest that the flux of DSOW with
sy > 28:01 kg m3 has a mean value of approximately 0.7 Sv (Girton et al., 2001). The increase in
concentration described above then requires a Nd
flux of 18 ton yr1. DSOW precursors, within the
EGC upstream of 70 N, flow between 400 and
1000 m. As they approach the Denmark Strait,
they contact the sea floor and the sides of the
strait. We estimated this contact area to be around
20,000 km2 (Fig. 4). For a value of 8 ppm Nd in
surrounding basaltic formations (Bernstein et al.,
1998; Mertz et al., 1991) and a value of 1.5 g cm3
for the sediment density (Carter and Raymo,
1999), the Nd flux calculated above corresponds
to the remobilization of a 0.07 mm thick sediment
layer per year. Considering the Nd concentration
error bars and the hypothesis stated above, this
result has to be considered cautiously. It is not an
estimation of a sediment remobilization rate, but
79
rather an estimation of the likelihood of our
hypothesis on the origin of the relatively radiogenic DSOW signature. Though this value is large,
it represents only a fraction (4–6%) of sediment
accumulation rates recorded in sheltered neighboring areas (1.2–1.8 mm yr1, for instance, at Nansen
Fjord mouth; Jennings and Weiner, 1996).
Furthermore, the high bottom velocities in the
Denmark Strait are likely to resuspend/remove all
of the sediment supplied to the region (C.
Innocent, pers. comm., 2001). There are two
reasons why DSOW precursors within the EGC
are not subject to any input upstream of 70 N
whereas they suddenly undergo important input
between 70 N and the Denmark Strait. First, the
contact area between the water mass and the
margin upstream of 70 N is restricted to the
western edge of the water mass, which flows along
the Greenland slope between 400 and 1000 m. As it
approaches the Denmark Strait, this contact area
increases suddenly, since the sea floor and the
Iceland slope contact the lower and eastern
boundaries of the water mass. Second, the
geological nature of the margins shifts from a
granitic Precambrian type to a basaltic one, the
former being much less soluble than the latter.
This supports the remobilization of a fraction of
mainly basaltic material bordering the strait as an
explanation for the shift in the Nd isotopic
signature between 70 N and the Denmark Strait.
4. AW crossing the strait towards Greenland
Nd IC profiles in the Denmark Strait display
minima between 100 and 200 m depth at stations
Center and East (eNd ¼ 13:2 and 12.1 respectively, see Fig. 2). These minima are associated
with temperature and salinity maxima (station
Center, 97 m, y ¼ 6:8 C, S ¼ 35:13; see Fig. 3),
and with dissolved oxygen minima ([O2]=
8.86 mg l1). These characteristics clearly identify
AW, carried by the Irminger Current towards the
Iceland Sea.
The Nansen Fjord station is located in the west
of Denmark strait, along the Greenland coast,
within a basaltic area. Temperature and salinity
profiles (see Fig. 3) show the presence of very fresh
ARTICLE IN PRESS
80
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
water within the first 20 m (y > 2 C, So31), then
Polar Water (PW; station Nansen, 80 m, y ¼
1:35; S ¼ 33:13) down to 150 m depth, then
AW (station Nansen, 200 m, y ¼ 4:31; S ¼ 34:77)
down to 250 m depth, and finally a colder less
saline water mass at the bottom (station Nansen,
303 m, y ¼ 0:2 C, S ¼ 34:57). The eNd profile is
shown in Fig. 2.
PW of the Nansen fjord has an Nd IC of –3.1
and a very high Nd concentration of 14.6 ppt.
Comparing with a PW sample upstream in the
EGC (73 N section, 50 m depth, eNd ¼ 10:9;
½Nd ¼ 3:2 ppt), we notice a large increase in eNd
and Nd concentration. Once again, this increase
could result from water mass mixing, lithogenic
input or both. Owing to the homogeneity of
the Nd isotopic signature within the EGC (cf.
Section 3.3.2), mixing should have an insignificant
impact on the signature of this water mass. This
implies that the variation is mainly due to
lithogenic input. We estimated that this enrichment is characterized by an eNd of –0.9 (cf.
Eq. (1)). This value is in excellent agreement with
that calculated above for the radiogenic enrichment of the DSOW (eNd ¼ 0:4). It also suggests
that the enrichment is composed mainly of basaltic
material, which is consistent with the basaltic
nature of the fjord area (between 70 N and
68 N). Since the basaltic rocks are found only
north of the fjord, the PW likely receives lithogenic
inputs along this coast. The bottom water of the
fjord also has a radiogenic signature and a high
Nd concentration (eNd ¼ 2:2; ½Nd ¼ 11:9 ppt),
suggesting that it too received a radiogenic
enrichment.
On the other hand, intermediate waters in the
fjord are characterized by eNd and [Nd] minima
(eNd ¼ 10:9 and ½Nd ¼ 3 ppt). These characteristics, as well as their y and S maxima, clearly
identify AW. The presence of AW here is
surprising since the only known current carrying
AW in the Denmark Strait is the Irminger
Current, which is restricted to the eastern part of
the strait. The fact that its Nd IC and concentration are not very different from those of the AW
measured in the strait (station Center, 97 m, eNd ¼
13:2; ½Nd ¼ 2:7 ppt], Fig. 2 and Table 1) shows
either that it comes from the Irminger Current and
crosses the strait, or that it flows north-eastward
along the east Greenland Coast in a coastal
counter current. Coastal counter currents are
particularly subject to lithogenic inputs, but the
absence of any concentration increase or granitic
Precambrian influence on the eNd value precludes
this possibility. The AW most likely comes from
the northward flowing branch of the Irminger
Current. It is detached from the current north of
the sill and drifts westward onto the Greenland
shelf. The flow in the Denmark Strait being mainly
southwestward, a limb of the Irminger Current
crossing the whole strait without interruption is
hardly imaginable. Moreover, the absence of AW
at station West suggests that its transport is due to
discontinuous structures. One therefore must
invoke a small structure (at the strait scale), which
is able to cross the strait without being entrained
by the EGC. It is therefore probable that the AW
found at the mouth of the Nansen Fjord crossed
the strait within eddies. Rudels et al., 1999a, b)
observed highly variable AW extensions across the
strait, sometimes going far onto the Greenland
shelf, during repeated measurements in August–
September 1997.
5. Conclusions
During the Signature/GINS cruise (IFRTP, July
August 1999), Denmark Strait Overflow Waters
(DSOW) had the following characteristics:
0:5oyo0; 34:87oSo34:90; 28:01osy o28:05;
eNd¼ 8:471:4 and ½Nd ¼ 3:270:7 ppt. Despite
the heterogeneity of their Nd isotopic compositions and concentrations, the mean values are in
good agreement with one comparable historical
data (Piepgras and Wasserburg, 1987).
Waters having similar hydrological characteristics upstream of the Denmark Strait within the
East Greenland Current (EGC), between 77 N
and 70 N, had different Nd characteristics. Those
were very homogeneous (eNd ¼ 10:970:2;
½Nd ¼ 2:470:1 ppt, 11 unfilered samples). This
constancy suggests the absence of lithogenic input
from the highly unradiogenic East Greenland
margin into intermediate waters (flowing between
400 and 1000 m), corroborating conclusions based
ARTICLE IN PRESS
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
on dissolved aluminum measurements (Measures
and Edmond, 1992).
The shift from the characteristics observed in
the waters flowing within the EGC upstream of
70 N to those of the DSOW could be explained by
an input of lithogenic Nd on these waters or by the
mixing with waters coming from the Iceland Sea.
Although, we could not address the question of an
Iceland Sea contribution in the present work, we
showed that the remobilization of mainly basaltic
material bordering the strait can explain the
observed shift. A contribution, although minor,
of granitic Precambrian derived sediments is likely
to be responsible for the observed heterogeneity of
this water mass signature. This heterogeneity likely
vanishes as the water mass sinks downstream of
the strait in the Irminger Sea. However the DSOW
seems to keep the imprint mainly due to the
basaltic formations.
Atlantic Water (AW) was identified at the
Nansen Fjord mouth. Nd signatures helped to
conclude that this water mass probably crosses the
strait within eddies from the Irminger Current.
This work confirmed the occurrence of heterogeneities within water masses in highly energetic
areas, either within straits or at the ocean
boundaries. This emphasizes the need of increasing
the sampling spatial and temporal resolution in
such areas, as well as measuring both filtered and
unfiltered samples. The occurrence of AW as north
as the Nansen Fjord, characterized by a very
distinct Nd signature, open the possibility of
investigating the northward extension of the North
Atlantic Drift in the Holocene by the analysis of
high resolution cores, as those taken during the
Image V cruise.
Acknowledgements
We thank the three anonymous reviewers for
their helpful suggestions. We acknowledge J.C.
Gascard for having organized the physical part of
the Signature/GINS program. We thank the
French Institute for Polar Research and Technology and the captain and crew of the R/V ‘‘Marion
Dufresne’’ for having carried out the cruise. A.
Gouzy is acknowledged for his intensive contribu-
81
tion to the sampling and preconcentration of
Nd on board. We are grateful to M. Souhaut,
M. Roy-Barman and P. Brunet for their help
with the analytical work. R. Francois, G. Reverdin, N. Ayoub, J.D. Milliman, J.T. Andrews and
C. Hillaire-Marcel are deeply thanked for their
help in interpreting the data. We also acknowledge
R. Schlitzer for providing the free Ocean Data
View software.
References
Abouchami, W., Galer, S.J.G., Koschinsky, A., 1999. Pb and
Nd isotopes in NE Atlantic Fe–Mn crusts: proxies for trace
metal paleosources and paleocean circulation. Geochimica
et Cosmochimica Acta 63, 1489–1505.
Bernstein, S., Kelemen, P.B., Tegner, C., Kurz, M.D.,
Blusztajn, J., Brooks, C.K., 1998. Post-breakup basaltic
magmatism along the East Greenland Tertiary rifted
margin. Earth and Planetary Science Letters 160, 845–862.
Broecker, W.S., 1991. The great ocean conveyor. Oceanography 4, 79–89.
Brueckner, H.K., Gilotti, J.A., Nutman, A.P., 1998. Caledonian eclogite-facies metamorphism of Early Proterozoic
protoliths from the North-East Greenland Eclogite Province. Contribution to Mineralogy and Petrology 130, 103–
120.
Burton, K.W., Ling, H.-F., O’Nions, K., 1997. Closure of the
Central American Isthmus and its effect on deep-water
formation in the North-Atlantic. Nature 386, 382–385.
Carter, S.J., Raymo, M.E., 1999. Sedimentological and mineralogical control of multisensor track data at sites 981 and
984. Proceedings of the Ocean Drilling Program, Scientific
results 162, 247–257.
Dickson, R.R., Brown, J., 1994. The production of North
Atlantic Deep Waters: sources, rates and pathways. Journal
of Geophysical Research 99 (C6), 12319–12341.
Frank, M., Whiteley, N., Kasten, S., Hein, J.R., Halliday,
A.N., O’Nions, R.K. (2001). NADW Export to the Eastern
South Atlantic and Southern Ocean over the Past 14 Myr
Deduced from Nd and Pb Isotopes in Ferromanganese
Crusts. European Union of Geosciences XI, Strasbourg,
France.
Girton, J.B., Sanford, T.B., K.ase, R.H., 2001. Synoptic sections
of the Denmark Strait Overflow. Geophysical Research
Letters 28 (8), 1619–1622.
Grousset, F.E., Biscaye, P.E., Zindler, A., Prospero, J., Chester,
R., 1988. Neodymium isotopes as tracers in marine
sediments and aerosols: North Atlantic. Earth and Planetary Science Letters 87, 367–378.
Hansen, H., Nielsen, T.F.D., 1999. Crustal contamination in
Palaeogene East Greenland flood basalts: plumbing system
evolution during continental rifting. Chemical Geology 157,
89–118.
ARTICLE IN PRESS
82
F. Lacan, C. Jeandel / Deep-Sea Research I 51 (2004) 71–82
Hansen, B., Østerhus, S., 2000. North Atlantic-Nordic Seas
exchanges. Progress in Oceanography 45, 109–208.
Jeandel, C., 1993. Concentration and isotopic composition of
neodymium in the South Atlantic Ocean. Earth and
Planetary Science Letters 117, 581–591.
Jeandel, C., Bishop, J.K., Zindler, A., 1995. Exchange of Nd
and its isotopes between seawater small and large particles
in the Sargasso Sea. Geochimica et Cosmochimica Acta 59,
535–547.
Jeandel, C., Thouron, D., Fieux, M., 1998. Concentrations and
Isotopic compositions of Nd in the Eastern Indian Ocean
and Indonesian Straits. Geochimica et Cosmochimica Acta
62, 2597–2607.
Jennings, A.E., Weiner, N.J., 1996. Environmental change in
eastern Greenland during the last 1300 years: evidence from
foraminifera and lithofacies in Nansen Fjord 68 N. The
Holocene 6, 179–191.
Lacan, F., Jeandel, C., 2001. Tracing Papua New Guinea
imprint on the central Equatorial Pacific Ocean using
neodymium isotopic compositions and Rare Earth
Element patterns. Earth and Planetary Science Letters
5779, 1–16.
Mauritzen, C., 1996. Production of dense overflow waters
feeding the North Atlantic across the Greenland-Scotland
Ridge. Part 1: evidence for a revised circulation scheme.
Deep-Sea Research I 43, 769–806.
Measures, C.I., Edmond, J.M., 1992. The distribution of
aluminium in the Greenland Sea and its relationship to
ventilation processes. Journal of Geophysical Research 97
(C11), 17787–17800.
Mertz, D.F., Devey, C.W., Todt, W., Stoffers, P., Hofmann,
A.W., 1991. Sr–Nd–Pb isotope evidence against plume—
asthenosphere mixing north of Iceland. Earth and Planetary
Science Letters 107, 243–255.
.
O’Nions, R.K., Gronvold,
K., 1973. Petrogenetic relationships
of acid and basic rocks in Iceland: Sr-isotopes and rareearth elements in late and postglacial volcanics. Earth and
Planetary Science Letters 19, 397–409.
Piepgras, D.J., Jacobsen, S.B., 1988. The isotopic composition
of neodymium in the North Pacific. Geochimica et
Cosmochimica Acta 52, 1373–1381.
Piepgras, D.J., Wasserburg, G.J., 1987. Rare earth element
transport in the western North Atlantic inferred from
isotopic observations. Geochimica et Cosmochimica Acta
51, 1257–1271.
Poulain, P.-M., Warn-Varnas, A., 1996. Near-Surface circulation of the Nordic seas as measured by Lagrangian drifters.
Journal of Geophysical Research 101 (C8), 18237–18258.
.
Rudels, B., Eriksson, P., Gronvall,
H., Hietala, R., Launiainen,
J., 1999a. Hydrographic observations in Denmark Strait in
fall 1997 and their implications for entrainment into
the overflow plume. Geophysical Research Letters 26 (9),
1325–1328.
Rudels, B., Friedrich, H.J., Quadfasel, D., 1999b. The Arctic
Circumpolar Boundary Current. Deep Sea Research II 46,
1023–1062.
Rutberg, R.L., Hemming, S.R., Goldstein, S.L., 2000. Reduced
North Atlantic deep Water flux to the glacial Southern
Ocean inferred from neodymium isotope ratios. Nature 405,
935–938.
Sinko, J.A. (1994). Les flux de particules issues des eaux de
surface de l’Atlantique Nord depuis 250,000 ans: variabilit!e
et implications pal!eoclimatiques, Vol. I. University of
Bordeaux, Talence, France, p. 329.
Strass, V.H., Fahrbach, E., Schauer, U., Sellmann, L., 1993.
Formation of Denmark Strait Overflow Water by mixing in
the East Greenland Current. Journal of Geophysical
Research 98 (C4), 6907–6919.
Swift, J.H., Aagaard, K., Malmberg, S.-A., 1980. The
contribution of the Denmark Strait overflow to the deep
North Atlantic. Deep-Sea Research 27A, 29–42.
Tachikawa, K., Jeandel, C., Roy-Barman, M., 1999a. A new
approach to Nd residence time in the ocean: the role of
atmospheric inputs. Earth and Planetary Science Letters
170, 433–446.
Tachikawa, K., Jeandel, C., Vangriesheim, A., Dupr!e, B.,
1999b. Distribution of rare earth elements and neodymium
isotopes in suspended particles of the tropical Atlantic
Ocean (EUMELI site). Deep-Sea Research I 46, 733–756.
Tachikawa, K., Athias, V., Jeandel, C., 2003. Neodymium
budget in the modern ocean and paleoceanographic
implications. Journal of Geophysical Research 108 (C8),
3254.
Taylor, P.N., Kalsbeek, F., Bridgwater, D., 1992. Discrepancies
between neodymium, lead and strontium model ages from
the Precambrian of southern East Greenland: evidence for a
proterozoic granulite-facies event affecting Archaean
gneisses. Chemical Geology 94, 281–291.
Thrane, K., 2002. Relationships between Archaean and
Palaeoproterozoic crystalline basement complexes in the
southern part of the East Greenland Caledonides: an ion
microprobe study. Precambrian Research 113, 19–42.
Vance, D., Burton, K., 1999. Neodymium isotopes in planktonic foraminifera: A record of the response of continental
weathering and ocean circulation rates to climate change.
Earth and Planetary Science Letters 173, 365–379.
Winter, B.L., Johnson, C.M., Clark, D.L., 1997. Strontium,
neodymium, and lead isotope variations of authigenic and
silicate sediment components from the Late Cenozoic Arctic
Ocean: implications for sediment provenance and the source
of trace metals in seawater. Geochimica et Cosmochimica
Acta 61, 4181–4200.
Woodgate, R.A., Fahrbach, E., Rohardt, G., 1999. Structure
and transports of the East Greenland Current at 75 N from
moored current meters. Journal of Geophysical Research
104 (C8), 18059–18072.