RADIOCARBON,
Vo141,
Nr
1,
1999,
p
51-73
©1999 by
the
Arizona
Board of
Regents on behalf of
the University of
Arizona
OCEANIC RADIOCARBON BETWEEN
ANTARCTICA AND
SOUTH
AFRICA ALONG
WOCE
SECTION
16
AT
30°E
Viviane
Leboucherl
James Orr2
Philippe
Jean-Baptiste2
Maurice
Arnold1
Patrick Monfrayl
Nadine
Tisnerat-Labordel Alain
Poisson3
Jean-Claude Duplessyt
ABSTRACT. Accelerator mass
spectrometry
(AMS)
radiocarbon
measurements
were made on 120
samples collected
between Antarctica
and South
Africa
along
30°E
during the WOCE-France
CIVA 1
campaign in February 1993.
Our
principal
objective
was to
complement
the Southern Ocean's sparse existing data set in order to improve the
14C
benchmark used for
validating ocean carbon-cycle models,
which
disagree considerably
in this
region. Measured
14C
is consistent with
the 0-S
characteristics of CIVA1. Antarctic Intermediate Water (AAIW) forming north of the Polar Front (PF) is rich in'4C,
whereas
surface waters south of the PF are depleted in
14C.
A distinct old
14C
signal
was found for the contribution of the Pacific Deep
Water (PDW) to the return flow of Circumpolar Deep Waters (CDW). Comparison to previous measurements shows a 4C
decrease
in
surface waters, consistent
with
northward
displacement of surface waters, replacement by old deep waters
upwelled at the Antarctic Divergence,
and
atmospheric
decline in
14C.
Conversely, an increase was found in deeper layers, in
the AAIW. Large uncertainties, associated
with
previous
methods for separating natural and
bomb
14C
when
in
the Southern
Ocean
south of 45°S, motivated us to develop a new approach that relies
on a simple mixing model and on chlorofluorocarbon
(CFC)
measurements also taken during
CIVA1. This approach leads to inventories for CIVA1 that are equal
to or higher than
those calculated with previous
methods. Differences between
old and new methods are especially high south of approxi-
mately
55°S,
where bomb
14C
inventories are relatively
modest.
INTRODUCTION
The Southern
Ocean extends south
to Antarctica and north
to the Subtropical Front
(STF),
which
meanders between
40°S
and
50°S.
As
such, the
Southern Ocean covers
about
21%
of
the world's
ocean
surface.
Within the Southern Ocean,
large vertical exchanges
take place
between water
masses.
Important processes
include ventilation of deep
waters at the
Antarctic Divergence Zone
(ADZ, ca. 65°S), production
of ventilated
bottom waters
in
the Weddell
Sea and around the
Antarc-
tic continental
slope, and
formation of
Antarctic Intermediate
Water
(AAIW) within the Polar
Fron-
tal
Zone (PFZ) between
the Polar Front
(PF)
and the
Subantarctic
Front
(SAF).
The Southern Ocean
is considered
to be a
major sink for anthropogenic
CO2
(Sarmiento et a1.1992).
However,
for
this
region,
there
is
substantial
disagreement between carbon-cycle
simulations from
different ocean
general
circulation
models
(OGCMs),
which
offer
the only
means to estimate future
oceanic CO2
uptake (On 1996).
Radiocarbon
is
routinely
used to evaluate
the performance
of such
models (Toggweiler et
al. 1989a, 1989b;
Maier-Reimer 1993;
Taylor
1995; On'
1.996). However,
few
14C
data are available
in the remote
Southern Ocean,
where
both the spatial and temporal
changes
in
14C
are
important. Here,
we provide
14C
data to
complement the
sparse established
dataset.
Previous
14C
measurements
include those
from
GEOSECS (Geochemical
Ocean
Sections
Study)
during
1973-1978
(Ostlund and
Stuiver 1980;
Stuiver and
Ostlund
1980; Stuiver
et al. 1981),
from
the
Win-
ter Weddell
Sea
Experiment (WWSP)
in 1986
(Schlosser et al.
1994), and
from
INDIGO
(Indian Gas
Ocean
project)
in
1985-1987
(Ostlund
and
Grall 1988).
The
14C
data
presented here
were
collected
during
CIVA1(Circulation
et Ventilation
Bans
l'Antarctique) of WOCE-France
taken
in
February-
March
1993.
Measurements of
14C
were
made
by
AMS in Gif-sur-Yvette,
France.
t
Laboratoire
des
Sciences du Climat et
de 1'Environnement
(LSCE), Avenue
de la Terrasse,
F-91198
Gif-sur-Yvette
Cedex,
France
2Laboratoire
des
Sciences
du Climat et de
1'Environnement
(LSCE),
DSMICEN SaclaylCEA, L'Orme,
Bat. 709,
F-91191
Gif-sur-Yvette
Cedex,
France
3Laboratoire
de
Physique et Chimie
Marines
(LPCM),
Universite
Pierre et
Marie Curie,
4,
place
Jussieu,
F-75252
Paris
Cedex
05,
France
51
https://doi.org/10.1017/S0033822200019330 Published online by Cambridge University Press
52 V Leboucher et at.
Oceanic
14C
must be separated
into its
natural and anthropogenic components
if
it is
to be useful for
model
validation. Natural
14C
is produced
by
cosmic radiation and can be considered at present
to
be
in
steady state
with
respect
to the vertical turnover
time
of the ocean
(ca.
1000
yr).
Bomb
14C
was
injected in
the atmosphere nearly exclusively
by
atmospheric thermonuclear
weapons
tests during
the 1950s and 1960s.
At its
peak
in
1963, the nuclear contribution nearly doubled prenuclear atmo-
spheric levels of
14C.
Since
then, atmospheric
14C
has decreased. At
present, levels are roughly
10%
higher
than in the prenuclear
period.
Both
14C
components are transferred to the ocean
by
air-sea gas
exchange as
14CO2.
With its
steady-state production and 5730
yr
half-life, the natural component of
14C
is
useful for evaluating modeled deep-ocean circulation fields.
Bomb
14C
has had
much less
time
to
invade
the ocean. Because
its
boundary
conditions have changed rapidly since the late
1950s, bomb
14C
is
useful for evaluating near-surface circulation
fields predicted by
ocean
models.
To
separate natural and
bomb
14C,
we
first
applied
the
method
developed by
Broecker et al. (1995),
which
relies on the near linear
correlation between
natural A14C
and
dissolved silica
throughout
much
of
the ocean. However,
for the stations situated
south of 45°S
(-90% of samples collected
here), uncertainties
associated with
Broecker et
al.'s
method
are
large.
Since
the GEOSECS
cam-
paign in
the 1970s, it
has become less
certain that deep waters
remain
uncontaminated
with
bomb
14C,
as supposed by
Broecker
et al.
(1995)
for GEOSECS
era
samples.
To take
this
possibility into
account,
we
modified
the bomb versus natural
separation methodology
for our samples taken south
of 45°S.
Our
new
method
takes advantage of
CFC-11 and CFC-12 measured
during
CIVAI
and uses
a
simple
ventilation model,
similar
to that used in
previous
tracer studies
(Jenkins
1980;
Andrie
et
al. 1986;
Haine
1996).
METHODS
The
14C
data (Table
1)
were obtained
following
the procedure
described by
Bard et al.
(1987) and
Arnold
et
al.
(1987). Seawater
was collected in 12-L
Niskins.
Once
aboard ship,
seawater
was trans-
ferred
to 250 mL
borosilicate
bottles and
poisoned with 1
mL
of saturated
HgCl2
solution. The boro-
silicate
bottles
were
then sealed.
Back
in
the laboratory,
CO2 was extracted by
adding 2 mL
of 15N
H3P04
to a 100 mL
seawater
aliquot in
a vacuum-tight
system
sparged by
He gas
(flow rate 80
mL
min
1).
Extraction
required
1
h. Water was
removed
by
traps at
-80°C;
CO2 was trapped
at
-180°C.
Subsequently,
CO2
was reduced
to graphite
using
hydrogen in
the presence
of
iron
powder (6-8 h).
The
carbon-iron
mixture
was
divided
into
3
parts, then
pressed
into
an aluminum
target for
accelera-
tor mass
spectrometry
(AMS)
analysis.
Usually, 2 or
3 targets
per sample
were analyzed
to obtain
a
precision
close to
±5%.
The 3
isotopes,
12C,'3C
and
14C,
were
measured
directly in
the
AMS
system
to calculate
A14C
normalized
to a
b13C
of
-25%.
The data
are expressed
as O14C in
per
mil
(Stuiver
and Polach
1977).
RESULTS
To
establish
the
level of
agreement
of our
14C
CIVA
1 measurements
with
previous data, Figure
1
provides
14C
profiles
from station 442
of
GEOSECS
(1978) and
from station 1
of
JADE1(Java
Aus-
tralia
Dynamic Experiment
1989) cruise,
both
taken at
the same
location
(1
°S, 91
°E)
in
the equato-
rial
Indian
Ocean.
The
14C
measurements
from JADEI
were made in
Gif-sur-Yvette
by AMS
just
prior
to
those
of CIVA
1. JADE 1
and CIVA
1
samples
received identical
processing.
The deep por-
tions
of GEOSECS
and JADE1
profiles,
where
bomb
14C
has
not penetrated,
agree to within
the pre-
cision
of
the measurements.
Therefore, our
14C
measurements
by
AMS
are
consistent with
those
from GEOSECS,
determined
by
13-counting.
Furthermore,
the
methodology
used
at
Gif-sur-Yvette
has
been
previously
checked by directly
comparing
13-counting
and AMS measurements
performed
on
the same
samples
from
INDIGO
(Bard et al.
1988).
https://doi.org/10.1017/S0033822200019330 Published online by Cambridge University Press
Oceanic
14C
between
Antarctica
and
South
Africa
53
Table
1 Data
for
the
stations
occupied
during
the
CIVA1
cruise
Potential
Num.
Nat.
Depth
temp.
Salinity
Si02
SC02
CFC-11
CFC-12
of
0140
A'4C
L\'4C
(m)
(°C)
(S,
psu)
Sig 0
(pmol
kg-')
(mmol
kg-')
(pmol
kg-')
(pmol
kg-')
meas.
(%o)
(%)
(%o)
Station
6 68°S,
30°E
10.2
0.2009
33.915
27.218
62.25
2149
6.76
2.92
2
-97 ±
4
-132 35
19.9
0.1145
33.919
27.225
62.25
2152
6.72
2.93
2
-97±4
-132
36
40
-0.1182
33.933
27.249
63.06
2160
6.7
2.86
2 -98
±
4
-133
35
100.6
-1.7636
34.35
27.651
68.77
2217
5.68
2.47
2
-114
± 4
-139
25
150.5
-1.6967
34.371
27.665
70
2210
5.59
2.45
2
-117 ±
4
-140 23
199.7
-0.5536
34.468
27.708
83.5
2240
3.59
1.52
2
-141 ± 4
-154
13
324.4
0.7518
34.642
27.773
99.93
2247
1.14
0.46
2
-165 ±
4
-176
11
698.8
0.7136
34.693
27.816
115.4
2254
0.58
0.23
2
-157
±
4
-163 6
998.8
0.4484
34.687
27.829
124.7
2266
0.46
0.18
2
-157 ±
4
-162
5
1198.8
0.2899
34.683
27.833
130.9
2264
0.41
0.17
2
-164
± 4
-168
4
2247.1
-0.2104
34.666
27.847
136.3
2260
0.56
0.21
2
-158
± 4
4
153
-163
159
6
6
2497.6
-0.2729
-
27.848
135.6
2258
0.64
0.24
2
±
-
-
2997.6
-0.378
34.665
27.851
136.3
2253
0.91
0.36
3
-145 ± 7
-153
9
3495.9
-0.4769
34.655
27.852
137.1
2254
1
0.41
2
-157
±
4
-167
10
3681.3
-0.5667
34.655
27.855
137.1
-
0.99
0.4
2
-162 ±
4
-171
10
Station
8 67°S,
30°E
10.6
0.646
33.866
27.155
60
2158
6.58
2.89
2
-97±4
-130 33
20.2
0.6406
33.869
27.155
60
2164
6.57
2.84
2
-97 ± 5
-130
33
40
-1.505
34.249
27.568
60.77
2197
5.7
2.46
2
-104±4
-131
27
80.8
-1.7041
34.356
27.653
65.5
2216
5.43
2.36
2
-103±4
-136 33
124.7
-1.142
34.412
27.682
71.02
2228
4.62
1.95
2
-111±4
-141
30
175.9
0.4412
34.564
27.729
86.01
2255
2
0.85
2
-130±5
-156
26
322.1
1.1479
34.679
27.778
92.69
2266
0.74
0.29
2
-155±5
-162
7
598.4
1.0146
34.708
27.812
102.5
2263
0.39
0.15
2
-148±4
-152
4
795
0.7838
34.704
27.823
110.1
2267
0.32
0.13
2
-179
±
5
-182
4
997.1
0.5854
34.697
27.83
115.4
2250
0.27
0.09
2
-175
±4
-175
0
1195.8
0.3901
34.687
27.834
118.4
2267
0.33
0.12
2
-154±4
-157
3
1795.5
0.0023
34.672
27.841
124.8
2270
0.53
0.12
2
-154±5
-159
4
2242.4
-0.1795
34.667
27.846
128.6
2263
-
-
2
-154±4
-160
6
2740.9
-0.3279
34.663
27.85
127.8
2260
0.69
0.28
3
-166±8
-173
7
3193.9
-0.4247
34.659
27.851
126.3
2258
094
0.39
2
-160±4
-169
9
3697.2
-0.5206
34.656
27.853
127.1
2260
0.98
0.42
2
-167±4
-176
10
4065.7
-0.6255
34.652
27.855
129.3
2247
0.91
0.36
2
-150±4
-159
9
4065.9
-0.6265
34.653
27.856
130.9
2253
0.94
0.36
2
-154±4
-163
9
Station
12
65°S,
30°E
10.6
1.004
33.829
27.1
54.24
2164
6.45
2.79
2
-117
±
5
-125
8
20.1
1.0006
33.829
27.1
53.44
2155
6.44
2.79
2
-97
± 4
-123
26
39.4
0.6738
33.842
27.15
54,24
2168
6.46
2.81
2
-97
±
5
-124
27
99.8
-0.9583
34.411
27.673
70.95
2225
4.37
1.87
2
-125±5
-141
16
149.7
0.6731
34.57
27.721
85.3
2259
1.84
0.75
2
-140±4
-155
16
200.1
0.9763
34.615
27.74
89.28
2261
1.29
0.49
2
-144±4
-159
15
324.5
1.2006
34.675
27.772
95.66
2255
0.72
0.26
2
-148±4
-155
7
499.8
1.2749
34.715
27.799
102
2264
0.3
0.07
2
-152±4
-152
0
698.9
1.0755
34.718
27.813
-
2266
0.24
0.05
2
-158±4
-158
0
998
0.749
34.706
27.827
113.2
2257
0.2
0.04
2
-152±4
-152
0
1197.1
0.5752
34.699
27.831
116.4
2259
0.2
0.03
2
-161±4
-161
0
2000
-
34.674
-
-
2264
0.23
0.04
2
-164 ± 5
-164
0
3000
-
34.663
-
-
2261
0.4
0.11
2
-155±4
-158
4
4495.9
-0.6169
34.652
27.856
129.2
2253
0.63
0.22
2
-164±4
-170
6
Station
18
61 °S,
30°E
11.3
1.7759
33.686
26.938
44.3
2136
6.32
2.83
2
-74 ±
5
-114 40
20.9
1.6254
33.721
26.973
44.72
2140
6.33
2.85
2
-85±5
-114
30
40.7
-0.0172
33.891
27.211
45.56
2152
6.41
2.85
2
-80
± 5
-116
36
80
-1.62
34.122
27.463
50.6
2176
6.11
2.74
2
-87±5
-121
34
126.5
-1.213
34.198
27.509
57.35
2196
5.56
2.44
2
-91
±
5
-127
36
175.2
0.7429
34.482
27.646
82.22
2243
-
-
2
-124 ± 5
-152
28
252.1
1.4748
34.635
27.72
90.64
2250
0.85
0.33
2
-146 ±
5
-155
8
599.1
1.2922
34.718
27.799
100.3
-
0.36
0.15
2
-143 ± 6
-146
4
799.7
1.0488
34.714
27.812
107.1
-
0.31
0.11
2
-166
± 5
-169
3
998.9
0.8452
34.709
27.823
112.1
--
0.24
0.1
2
-164
±
5
-164
0
1197.6
0.6426
34.7
27.828
116.4
-
0.21
0.09
2
-162±5
-162
0
2244.5
0.0032
34.673
27.843
127.3
-
0.22
0.07
2
-156 ± 5
-156
0
2995.5
-0.2888
34.664
27.849
127.3
-
0.24
0.1
2
-157
±
5
-157
0
4495.7
-0.6071
34.655
27.857
129
2260
0.26
0.09
2
-154±5
-154
0
Station
24
57°S,
30°E
27
10.4
2.2178
33.626
26.853
40.6
2131
6.27
2.84
2
-83
±5
-111
19.8
2.2133
33.625
26.853
40.6
2139
6.27
2.82
2
-77 ± 5
-111 33
40.7
2.2012
33.633
26.862
40.6
2140
6.26
2.84
2
-74 ± 5
-111
37
99
-0.7262
33.92
27.267
50.32
-
6.91
3.06
2
-81 ± 5
-120
39
150.5
-1.6046
34.141
27.476
68.97
2202
6.62
2.91
2
-87 ± 5
-139
52
199.3
-1.4569
34.234
27.549
73.03
2209
6.08
2.68
2
-92
±
5
-143
51
320.7
0.5166
34.596
27.751
99.78
2256
1.6
0.7
2
-141 ±5
-157
15
https://doi.org/10.1017/S0033822200019330 Published online by Cambridge University Press
54
V
Leboucher
et al.
Table
1
Data
for
the
stations
occupied
during
the CIVA
1
cruise
(Continued)
Potential
Num.
Nat.
Depth
temp.
Salinity
Si02
4C
(m)
(°C)
(S,
psu)
Sig
0
kg-')
kg-')
kg-')
kg-')
496.9
0.9574
34.695
27.803
2
5
696.1
0.7337
34.695
27.818
2
5
4
997.1
0.5577
34.695
27.829
2
±5
3
1195.8
0.4318
34.69
27.833
2
±5
3
1992.9
0.0367
34.675
27.842
2
5
0
2995.3
-0.3072
34.663
27.85
2
4
0
4495.2
-0.6165
34.654
27.856
2
4
0
5422.2
-0.7349
34.65
27.858
2
4
3
Station
28
54.3°S,
30°E
10.9
2.8814
33.876
26.998
2
6
19.9
2.8829
33.877
26.998
2
5
39.5
2.7778
33.952
27.066
2
5
101.2
0.0656
34.096
27.378
2
5
151.1
0.6092
34.294
27.456
2
5
197.2
1.3819
34.366
27.511
2
5
323.1
1.9407
34.572
27.632
2
5
7
496.7
1.8816
34.658
27.707
2
5
5
701.6
1.6213
34.689
27.752
2
5
5
995.2
1.5408
34.737
27.795
2
4
0
1399.7
1.0201
-
27.823
2
4
0
2494.4
0.1553
34.68
27.841
2
5
0
3993.7
-0.4818
34.658
27.853
2
4
0
4996.3
-0.621
34.657
27.858
2
4
0
5455.9
-0.6523
34.653
27.858
2
±4
0
Station
38
49.7°S,
30°E
9.3
4.5363
33.874
26.835
2
5
18.8
4.5366
33.874
26.836
2
5
38.2
4.5302
33.876
26.836
2
5
79
3.139
33.982
27.061
2
5
125.9
2.6988
34.011
27.123
2
5
172.9
3.169
34.125
27.173
2
6
325.4
2.6227
34.199
27.28
2
±6
576.9
2.3437
34.36
27.433
2
5
795.4
2.4431
34.525
27.556
2
5
998
2.3759
34.613
27.632
1
8
0
1198
2.3707
34.685
27.691
2
0
2498.4
1.1142
34.735
27.825
0
2
0
3994.5
-0.1069
34.672
27.848
2
5
0
5183.5
-0.4829
34.656
27.853
2
5
0
Station
52
44°S,
30°E
9.8
12.315
34.475
26.125
2
19.4
12.297
34.472
26.128
2
39.7
12.045
34.45
26.156
2
100.5
9.2985
34.596
26.756
2
142.6
8.7958
34.594
26.835
2
197.4
7.916
34.497
26.892
2
323.3
6.3703
34.341
26.983
2
499
5.1576
34.3
27.128
2
±5
698.7
4.0819
34.329
27.244
2
5
1001.2
3.0136
34.421
27.423
2
±5
0
1198.1
2.7672
34.53
27.532
2
4
0
2244.9
2.2081
34.793
27.789
2
0
3496.4
1.0783
34.753
27.841
0
2
0
4497.9
0.1862
34.69
27.846
2
±4
0
Figure
2
shows
a map
including
the
WOCE/CIVA1
transect
between
68°S
and 44°S
along
30°E,
where
14C
samples
were
collected.
Also
indicated
are
other
stations
in
the
same
area
where
14C
has
been
measured
on
previous
campaigns:
GEOSECS,
WWSP,
and INDIGO.
The
6
and
S
along
the CIVA1
transect
(Fig.
3A,B)4
exhibit
signatures
of
the major
fronts
that
charac-
terize
the
Southern
Ocean:
the Antarctic
Divergence
Zone
(ADZ)
between
63°S
and
67°S,
the
Wed-
dell
Gyre
Front
(WF)
at
57°S,
the
Polar
Front
(PF)
at
52°S,
and
the
Subantarctic
Front
(SAF)
at
48°S
(Gordon
1971;
Whitworth
and
Nowlin
1987;
Orsi
et
al.
1993,
1995;
Archambeau
et
al.
1998).
The
4Note:
Color
versions
of Figs.
3,
4,
7
and
9 are
available
at
http://www.radiocarbon.org/Journa1/v41n1/Leboucher/
https://doi.org/10.1017/S0033822200019330 Published online by Cambridge University Press
Oceanic
14C
between Antarctica and South
Africa
55
A14C (%o)
-220
-180 -140 -100 -60
-20 20
0
100
60
1000
2000
a)
3000
4000
5000
Figure 1 Profiles
of measured A14C
from GEOSECS
station 442 and
JADE1 station I
principal water
masses are
likewise apparent.
From the
Antarctic continent
to the
PF,
the
Summer
Surface
Water (SSW)
overlies
the near freezing
Winter Water
(WW). Below
the WW lies a
massive
amount
of
Circumpolar Deep
Water (CDW),
which
is
relatively
warm (>0.1
°C) and old
(with
respect
to ventilation
by contact
with the surface).
Beneath the
CDW and
along the bottom
is found the
Ant-
arctic
Bottom Water
(AABW).
In the AABW,
one can trace
the influence of
the Weddell Sea
Bottom
Water
(WSBW),
which at
its
origin
(the Weddell
Sea)
has a temperature
<-0.7 °C.
Another water
mass, the Subantarctic
Surface
Water
(SASW),
lies
between
the
PF
and
the
SAF. The
SASW
has
higher temperatures
and salinities
than
the
more southern surface waters
and
is remarkable
for its
south-to-north
downward
tilt in
0
and S
isolines, where
AAIW
is
formed.
In
this
northern
part
of the
CIVAI section,
intermediate level
CDW bears
a
high salinity
signature (maximum
34.8 psu)
from its
contact
with North
Atlantic Deep Water
(NADW).
AABW north of the
Atlantic-Indian
ridge
is
warmer and
less
dense than
AABW to the south,
which is closer to the source
of the
cold WSBW. Fur-
ther
description
of CIVAI
hydrography
can be
found in Archambeau
et al. (1998).
Table
1 provides the
CIVAI
14C
measurements.
These
data are
presented as a
composite
latitude
versus
depth section
(Fig. 4A) as well
as
individual profiles
(Fig.
5A,B). The former
reveals
that the
distribution
of the
measured
14C
roughly
follows
CIVAI
hydrography.
A
striking contrast
is
found
across
the PF:
north of the
PF, surface
waters are
enriched
in
14C
(levels
reach 55%0);
south of
the
PF, the upper
200
m
are
depleted
in
14C
(levels from
-70%o
to -140%o).
The Southern
depletion
is
principally
due to old
CDW upwelling
in the
Antarctic
Divergence
Zone between
63°S and
67°S
(Toggweiler
and
Wallace 1995).
However,
winter
ice cover
limits gas
exchange and
thus
may play
https://doi.org/10.1017/S0033822200019330 Published online by Cambridge University Press