Download The biogeochemical cycle of dissolved cobalt in the

MARCHE-02851; No of Pages 14
Marine Chemistry xxx (2011) xxx–xxx
Contents lists available at ScienceDirect
Marine Chemistry
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a r c h e m
The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean
south off the coast of South Africa
Johann Bown a,⁎, Marie Boye a, Alexander Baker b, Eric Duvieilbourg a, François Lacan c,
Frédéric Le Moigne a,1, Frédéric Planchon a,2, Sabrina Speich d, David M. Nelson a
a
Laboratoire des Sciences de l'Environnement Marin UMR6539, Institut Universitaire Européen de la Mer, Technopôle Brest Iroise, Place Nicolas Copernic, 29280 Plouzané, France
School of Environmental Sciences, University of East Anglia, Norwich, NR4 7TJ, United Kingdom
LEGOS, Observatoire Midi Pyrénées, 18 av. Edouard Belin, 31401 Toulouse Cedex 9, France
d
Laboratoire de Physique des Océans, Universite de Bretagne Occidentale — UFR Sciences, 6 av. Le Gorgeu, C.S. 93837, 29238 Brest Cedex 3, France
b
c
a r t i c l e
i n f o
Article history:
Received 7 September 2010
Received in revised form 27 January 2011
Accepted 25 March 2011
Available online xxxx
Keywords:
Dissolved cobalt
Southeastern Atlantic
Southern Ocean
International Polar Year
GEOTRACES
a b s t r a c t
The spatial distribution, biogeochemical cycle and external sources of dissolved cobalt (DCo) were investigated
in the southeastern Atlantic and the Southern Ocean between 33°58′S and 57°33′S along the Greenwich
Meridian during the austral summer 2008 in the framework of the International Polar Year. DCo concentrations
were measured by flow-injection analysis and chemiluminescence detection in filtered (0.2 μm), acidified and
UV-digested samples at 12 deep stations in order to resolve the several biogeochemical provinces of the
Antarctic Circumpolar Current and to assess the vertical and frontal structures in the Atlantic sector of the
Southern Ocean. We measured DCo ranging from 5.73 ± 1.15 pM to 72.9 ± 4.51 pM. The distribution of DCo
was nutrient-like in surface waters of the subtropical domain with low concentrations in the euphotic layer
due to biological uptake. The biological utilization of dissolved cobalt was proportional to that of phosphate in
depletion ratio of ~ 44 μM M−1. In deeper waters the distribution
the subtropical domain with a DCo:HPO2−
4
indicated remineralization of DCo and inputs from the margins of South Africa with lateral advection of
enriched intermediate and deep waters to the southeastern Atlantic Ocean. In contrast the vertical distribution
of DCo changed southward, from a nutrient-like distribution in the subtropical domain to scavenged-type
behavior in the domain of the Antarctic Circumpolar Current and conservative distribution in the Weddell
Gyre. There the cycle of DCo featured low biological removal by Antarctic diatoms with input to surface waters
by snow, removal in oxygenated surface waters, and dissolution and stabilization in the low-oxygenated Upper
Circumpolar Deep Waters. DCo distributions and physical hydro-dynamics features also suggest inputs from
the Drake Passage and the southwestern Atlantic to the 0° meridian along the eastward flow of the Antarctic
Circumpolar Current. Bottom enrichment of DCo in the Antarctic Bottom Waters was also evident, together
with increasing water-mass pathway and aging, possibly due to sediment resuspension and/or mixing with
North Atlantic Deep waters in the Cape Basin. Overall atmospheric input of soluble Co by dry aerosols to the
surface waters was low but higher in the ACC domain than in the northern part of the section. At the highest
latitudes, it is possible that snowfall could be a source of DCo to surface waters. Tentative budgets for DCo in the
mixed layer of the subtropical and the ACC domains have been constructed for each biogeochemical region
encountered during the cruise. The estimated DCo uptake flux was found to be the dominant cobalt flux along
the section. This flux decreases southward, which is consistent with the observations that DCo shows a
southward transition from nutrient-like towards conservative distribution in the mixed layer.
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
The abundance of cobalt (Co) in the upper continental crust
(10 ppm) is much lower than that of iron (3.5 104 ppm), with a Fe/Co
⁎ Corresponding author. Tel.: + 33 2 98 49 88 38.
E-mail address: [email protected] (J. Bown).
1
Now at: National Oceanography Centre, Southampton University of Southampton,
Waterfront Campus European Way Southampton, SO14 3ZH, United Kingdom.
2
Previously at: Royal Museum for Central Africa (RMCF), 3080 Tervuren, Belgium.
crustal abundance ratio of ~3500 (Taylor and McLennan, 1995).
Dissolved cobalt (DCo) concentrations in the open ocean are low,
typically b150 pM (Knauer et al., 1982; Martin et al., 1990; Fitzwater et
al., 2000; Saito and Moffett, 2001; Saito et al., 2004; Ellwood, 2008;
Noble et al., 2008; Pohl et al., 2011). Knowledge of Co distributions in the
ocean has increased with the development of sensitive analytical
techniques that are more accurate at such low concentrations (Vega and
van den Berg, 1997; Cannizzaro et al., 2000; Saito and Moffett, 2001;
Milne et al., 2010; Shelley et al., 2010) but data are still scarce in the
ocean worldwide. Nevertheless oceanic distributions have suggested a
0304-4203/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.marchem.2011.03.008
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
2
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
nutrient-like cycle of DCo (Martin et al., 1993; Saito and Moffett, 2002;
Saito et al., 2004; Jakuba et al., 2008; Noble et al., 2008; Saito and
Goepfert, 2008), with additional (versus phosphate) Co removal from
deep waters by scavenging as observed in the deep oligotrophic
northeast Pacific and western Philippine Sea (Knauer et al., 1982; Wong
et al., 1995).
The elemental composition of phytoplankton suggests that Co is an
important micronutrient for the growth of cyanobacteria (Sunda and
Huntsman, 1995), coccolithophorids and dinoflagellates (Ho et al.,
2003). Cobalt is involved in important biological functions and
molecules such as vitamin B12 and carbonic anhydrase (Droop,
1957; Guillard and Cassie, 1963; Morel et al., 1994; Bertrand et al.,
2007). Furthermore laboratory studies have shown that cobalt can
interact with cadmium and zinc in cellular metabolism (Price and
Morel, 1990; Sunda and Huntsman, 1995; Saito and Goepfert, 2008),
although the specific impact of cobalt on phytoplankton growth and
community composition is still not well understood. In addition to
biological removal, internal processes and external sources such as
remineralization, inputs from the atmosphere, deep inputs from the
continental margin and physical hydro-dynamics may impact the
oceanic distribution of cobalt. Sources of Co to surface waters are
poorly constrained. Knauer et al. (1982) and Wong et al. (1995) have
reported atmospheric inputs, whereas the impact of physical features
and water masses on DCo distribution has been described by Wong et
al. (1995), Saito et al. (2004) and Noble et al. (2008). However DCo
distribution may reflect both the biogeochemical and physical
characteristics of the studied regions (Wong et al., 1995; Saito and
Moffett, 2002; Saito et al., 2004; Noble et al., 2008).
In this study the vertical and meridional distributions of DCo are
presented in the southeastern Atlantic Ocean and the Southern Ocean
south off South Africa. The section includes contrasting biogeochemical domains, with an oligotrophic system in the subtropical northern
region and a High Nutrient Low Chlorophyll region (HNLC; Minas et
al., 1986) in the Antarctic domain in the southern part of the section.
Furthermore, the transect crossed the zone where South Atlantic,
Southern Ocean and Indian waters converge in one of the most
dynamic and variable ocean domains in the world (Wunsch and
Stammer, 1995; Boebel et al., 2003). DCo concentrations are discussed
in combination with biogeochemical and physical features to further
investigate how cobalt is distributed in these key ocean regions and to
explain the distribution of DCo in terms of sources, sinks and internal
processes that could impact its behavior within the water column.
2. Methods
2.1. The cruise
Samples were collected during the multidisciplinary MD166 BONUSGoodHope cruise that took place during the International Polar Year in
the austral summer 2008 (02/13/08–03/24/08) on board the French R/V
Marion-Dufresne II which sailed from Cape Town, South Africa, and
along the Greenwich Meridian to 57°S in the Southern Ocean (Fig. 1).
Two main types of stations were sampled for trace metal analysis.
These stations were defined by the number and depth of operations;
LARGE (L) and SUPER (S) (Fig. 1). The distribution of DCo was studied
at the five SUPER deep cast stations with ~20 sampling depths
between 15 and 4000 m, and at the seven shallower LARGE stations
with ~10 sampling depths between 15 and 2500 m. The position and
designation of each station along the section and the position of the
fronts are shown in Fig. 1. The exact position of the stations and
depths sampled for DCo is reported in Table 1.
2.2. Sample collection
Samples were collected with 12 L acid-cleaned GO-FLO bottles
(General Oceanics®) modified with PTFE rings, attached to a Kevlar
Fig. 1. Location of the stations sampled for DCo during the MD166 BONUS-GoodHope
cruise. Black circles designate the LARGE stations and white circles the SUPER stations.
The positions of fronts are also shown, with the southern branch of the Subtropical
Front (S-STF; ~42°2’S), the SubAntarctic Front (SAF; 44°2’S), the Polar Front (PF; 50°
22.4’S), the Southern ACC Front (SACCF; ~51°52’S) and the Southern Boundary of the
ACC (SBdy; ~55°54.3’S). From their geographical positions, stations L1, S1 and L2 were
in the Subtropical Zone (STZ), S2 was on the northern side of the Subantarctic Zone
(SAZ), stations L3, L4, S3 and L5 were within the Polar Frontal Zone, station L6 was on
the northern flank of PF, S4 was on the SACCF, station L7 was at the Sby, and station S5
was in the northern branch of the Weddell Gyre.
Map from ODV (Reiner Schlitzer).
wire and closed using Teflon-coated messengers. The samples were
then collected in acid-cleaned 250 mL LDPE Nalgene® bottles after
online filtration with a 0.22 μm Sartobran® 300 (Sartorius) cartridge
under pure N2 pressure (filtered 99.99% N2, 1 bar) in a pressurized
clean container (class 100). Each sample was then acidified on board
under a laminar flow hood (class 100) with hydrochloric acid
(ultrapur HCl®, Merck) and stored at pH ~ 2 in double bags at dark
and ambient temperature, until their analyses in the shore-based
laboratory about 18 months after their collection.
2.3. Measurement of dissolved cobalt concentration
2.3.1. Method and equipment
Dissolved cobalt concentrations were measured in acidified and
UV-digested samples by Flow-Injection Analysis (FIA) and chemiluminescence detection, following the method of Shelley et al. (2010).
The method includes UV oxidation of the acidified sample (30 mL) for
3 h in clean silica tubes using a 600 W high-pressure mercury-vapor
lamp and an equilibration time of 48 h before analysis. An IDA-based
resin (Toyopearl AF-650M Chelate) was used instead of immobilized
8-HQ resin (Sakamoto-Arnold and Johnson, 1987). The pre-concentration resin was conditioned and rinsed with ammonium acetate
buffer, and the reaction temperature was maintained at 60 °C using a
thermostatic bath to avoid interference due to bubbles passing
through the detector (Shelley et al., 2010).
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Table 1
Dissolved cobalt concentrations (DCo) and corresponding depths with the position of the
stations and the biogeochemical domains. Errors on the DCo concentrations (Stdv) were
calculated as the standard deviation of triplicate measurements of the same sample. nd means
not determined.
Domain
Station
Latitude
(°S)
Longitude
(°E)
Depth
(m)
DCo
(pM)
Stdv DCo
(pM)
STZ
L1
34.43
14.4
20
40
60
80
200
700
800
1000
1200
2100
20
30
40
70
100
200
300
500
700
850
1000
1200
1400
1600
2000
2700
3050
3500
3800
4000
15
35
45
95
300
600
800
1200
1400
2100
15
30
35
80
120
196
314
461
598
809
1029
1250
1441
1764
2156
2548
2891
3245
3636
3950
30
100
150
270
400
600
1200
1400
2100
30
60
27.1
29.7
30.4
29.9
35.0
50.1
48.6
47.5
54.4
46.1
23.7
15.6
26.3
41.0
44.3
30.5
37.6
42.7
51.8
46.6
48.6
48.8
45.3
46.6
48.9
45.4
53.9
37.8
38.7
41.4
17.0
15.3
19.0
17.7
46.6
54.6
54.0
46.3
45.9
49.2
7.40
5.73
7.50
21.6
26.3
38.2
44.2
52.1
65.1
49.7
41.0
59.6
65.6
43.4
45.1
59.2
44.6
41.8
41.8
36.7
22.8
38.7
50.8
49.4
55.0
47.2
49.5
51.9
47.1
27.3
29.1
2.58
2.55
1.54
0.67
2.70
3.09
2.20
2.39
2.55
0.20
2.15
1.68
1.97
0.23
0.73
1.43
2.80
0.65
0.53
2.35
2.39
1.52
4.64
5.17
2.14
1.35
3.30
1.53
2.58
2.39
0.54
3.19
0.23
1.08
5.63
4.30
3.33
0.95
1.44
2.38
0.67
1.15
1.39
1.19
0.84
0.99
1.14
2.66
4.89
2.21
1.30
6.54
0.86
1.84
1.79
2.05
1.86
4.14
1.65
1.93
1.34
0.83
1.53
2.50
9.41
3.56
1.04
1.21
1.01
nd
1.66
STZ
S1
36.50
13.1
STZ
L2
41.18
9.92
S-STZ
S2
42.47
8.93
SAZ
SAZ
L3
L4
44.88
46.01
6.88
5.87
(continued on next page)
3
Table 1 (continued)
Domain
Station
Latitude
(°S)
Longitude
(°E)
SAZ
L4
46.01
5.87
PFZ
S3
47.55
4.37
PFZ
L5
49.03
2.84
North PF
L6
50.38
1.33
South PF
S4
51.85
0
Sbdy
L7
55.23
0.03
Weddell Gyre
S5
57.55
0.03
Depth
(m)
DCo
(pM)
Stdv DCo
(pM)
100
150
270
400
800
1300
1600
2050
20
30
40
70
100
200
300
450
600
800
1070
1500
2020
2500
3000
3500
3980
40
80
150
170
250
350
700
1000
1600
2200
30
60
100
135
180
300
600
850
1600
2100
30
60
130
160
180
250
300
350
400
500
700
900
1117
1950
2300
2490
30
60
100
120
200
300
650
1000
1500
2100
30
60
120
190
48.2
40.5
52.9
49.2
48.7
45.9
50.3
45.5
35.4
33.6
37.3
35.7
42.9
40.5
63.1
50.1
43.0
48.9
56.3
42.3
39.5
40.7
36.6
33.3
30.5
43.5
43.8
42.3
50.2
49.0
53.9
73.0
49.6
46.0
39.9
39.0
37.8
51.0
58.7
52.0
48.6
48.0
45.1
39.5
36.5
40.3
37.6
48.8
45.6
50.4
56.8
55.2
49.6
50.4
50.2
47.1
38.8
36.5
62.9
36.5
28.6
51.3
45.6
47.7
49.7
56.2
49.5
43.9
46.5
33.1
44.7
36.7
36.0
32.1
36.9
1.90
1.86
1.28
2.01
2.12
2.32
0.50
2.53
2.45
0.85
1.33
1.27
1.80
2.17
3.50
0.30
1.03
1.08
3.34
0.28
1.84
3.22
0.91
3.20
0.27
2.33
2.93
3.23
1.56
0.95
2.25
4.51
1.37
1.10
1.26
1.54
3.27
1.52
2.19
1.72
0.46
1.97
3.06
2.22
1.74
0.87
2.10
1.09
2.55
2.04
1.40
2.27
1.33
6.14
1.66
2.98
1.56
3.40
8.04
3.16
0.95
1.11
nd
2.54
1.88
4.46
3.15
3.79
2.79
1.89
0.59
0.49
0.60
1.86
1.05
(continued on next page)
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
4
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Table 1 (continued)
Domain
Station
Latitude
(°S)
Weddell Gyre
S5
57.55
Longitude
(°E)
0.03
Depth
(m)
DCo
(pM)
Stdv DCo
(pM)
250
350
450
550
750
800
1250
1700
2150
2600
3050
3500
3840
43.4
36.4
39.7
34.4
50.4
37.6
30.1
38.4
32.9
33.3
24.8
29.1
30.5
1.23
2.22
2.74
4.52
nd
4.03
1.91
0.65
1.26
3.01
1.30
1.01
0.32
Briefly, the measurements are based on the oxidation reaction of
pyrogallol with hydrogen peroxide in an alkaline solution in the
presence of cetylmethylammonium bromide (CTAB) and methanol. The
reaction is catalyzed by cobalt, hence chemiluminescent emission in the
visible wavelengths that is proportional to the cobalt concentration is
produced. The flow injection manifold consisted of a peristaltic pump
(205 CA, Watson Marlow) used at 10 rpm, Tygon® pump tubing
(internal diameter of 1.85 mm for all the reagents except for the buffer
line which was 1.52 mm), two micro-electronically actuated, 6-port, 2
position, injection valves (SCIVEX valve and VICI valve from VALCO
instruments), and connecting PTFE tubing of 0.75 mm diameter. The
detection system is consisted of a photomultiplier detector (Hamamatsu, H9319 Series). The injection valves and the photomultiplier
detector were connected to a computer by serial ports and operated
by a Labview® 8.4 interface adapted from Bowie et al. (2005). The
electronic devices were connected to a current modulator (ELLIPSE MAX
600, MGE/UPS Systems). The DCo concentration was calibrated against
two calibration curves made with standard additions of cobalt of 0, 12.5,
25, 50 and 75 pM and performed before and after each series of
8 samples (see below). DCo concentrations and the standard deviation
of the measurement were based on triplicate analyses of each sample
using the peak height of the chemiluminescent signal. At a pump flow
rate of 10 rpm using 1.85 mm internal diameter sample line tubing and
a time of 5 min resulted in 6.25 mL of sample being preconcentrated. A
complete run of triplicate analyses of each sample took ~23 min.
Finally, the reaction buffer solution made up of sodium hydroxide
(0.2 M, Sigma Ultra®, N98%) and 20% v/v methanol (Sigma, for trace
analysis®) was prepared. Sodium hydroxide is required to reach the
optimal chemiluminescent reaction pH of 10.4 and the addition of
methanol leads to an increase of the chemiluminescence signal.
The Toyopearl IDA AF-650M ion exchange resin was mounted in a
1 cm column device with metal-free frits (Global-FIA Inc.) and
washed with 1 M HCl for 3 h and then rinsed with MilliQ for 3 h.
Primary standards of Co(II) with concentrations of 10−6 M and
−8
10 M were prepared weekly in acidified MilliQ water from serial
dilution of a Spectrosol commercial solution (979 μg mL−1, Sigma
Aldrich). Calibration was made daily with DCo standards containing
0, 12.5, 25, 50 and 75 pM of added DCo using the 10−8 M stock
solution diluted in filtered and acidified seawater to pH ~ 2 with HCl
(Ultrapur®, Merck). The seawater used for the calibrations was
collected at station S5 (57.55°S 0.03°E at 30 m) and cleaned by
passing through Toyopearl IDA AF-650M ion exchange resin. The
background concentration of DCo in this seawater (at 0 addition of
DCo) was 3.72 ± 1.36 pM (n = 40).
2.3.3. Analytical performance
Duplicate reagent blanks including the buffer blank were analyzed
by measuring DCo with MilliQ water used instead of sample, at the
beginning and at the end of DCo analyses (8 samples on average). The
mean reagent blank (based on all blank determinations) was 5.90 ±
1.24 pM of DCo (n = 40). DCo concentrations presented in Table 1 are
corrected with their respective reagent blanks.
The limit of detection of the method was estimated as three times
the standard deviation of the mean reagent blank and was thus on
average 3.72 pM of DCo (n = 40).
Each series of samples was calibrated by running one sample
collected during the Sampling and Analysis of iron (SAFe) program,
either the surface “S-SAFe” sample or the deep “D2-SAFe” sample,
along with the samples. SAFe samples were UV-digested for 3 h prior
to analyses and the results of DCo in S and D2 samples are reported in
Table 2. Our measured DCo concentration in the D2-SAFe is within the
range published (Johnson et al., 2007). Our S-SAFe value is slightly
higher but this is more likely a reflection of the detection limit
reported here (3.72 pM), which is close to the S-SAFe consensus
concentration (4.20 ± 1.90 pM, Table 2).
2.4. Hydrography
2.3.2. Preparation of the reagents
In order to reduce airborne contamination, sample handling in the
home laboratory was carried out under a laminar flow hood (ADS
Laminaire, ISO 5 class). All reagents were prepared in 1L LDPE
Nalgene® bottles with MilliQ water the day before the analysis and
kept overnight at room temperature for equilibration.
A rinsing solution composed of HCl (0.05 M, Suprapur®, Merck)
and ammonium acetate (0.05 M, Sigma Ultra®) was used to remove
the seawater matrix ions from chelating resin; those ions would
otherwise have interfered with the chemiluminescence detection.
An ammonium acetate (0.3 M, Sigma Ultra®) buffer solution was
used for in-line pH adjustment of samples up to pH ≥ 5.2 to keep
within the optimal working pH of the chelating resin and ensure
quantitative binding of cobalt. This solution was cleaned on line by
passing through a column filled with IDA AF-650M chelate Toyopearl
resin.
Hydrochloric acid (0.1 M, Suprapur®, Merck) was used to elute the
chelated cobalt from the resin column. The final volume of the eluted
sample was 1.89 mL.
The pyrogallol reagent solution was made up of pyrogallol (0.05 M,
Sigma 99%, ACS reagent), CTAB (0.025 M, Sigma Ultra®, 99%) and 6% v/v,
hydrogen peroxide (35%, Sigma Aldrich). CTAB was used to provide
micellar environment for increasing the sensitivity of the reaction.
Potential temperature (Θ), salinity (S) and dissolved oxygen (O2)
were recorded using a SBE 911+ Seabird® probe, with respectively,
SBE3+, SBE4 and SBE43 sensors (Branellec et al., 2010). Data
recording and validation of potential temperature, salinity and oxygen
are described by Branellec et al. (2010) and by Kermabon and Arhan
(2008).
2.5. Macronutrient analyses
Nutrient samples were collected using a CTD-rosette (SBE 32
Seabird®) equipped with Niskin bottles (Le Moigne, 2008; Le Moigne
et al., in preparation). Silicate and nitrate were analyzed by standard
Table 2
Comparison of dissolved cobalt analyses obtained for UV-oxidized samples by the FIAChemiluminescence method used in the present study with consensus values reported
by the Sampling and Analysis of iron (SAFe) program. Water samples provided by SAFe
from surface waters (SAFe S) and deep waters (SAFe D2) were analyzed. All ± terms
represent standard deviation from average values.
UV oxidized
SAFe S
SAFe D2
DCo consensus value (pM)
DCo measured value (pM)
4.20 ± 1.90
43.1 ± 3.20
6.92 ± 0.40 (n = 6)
47.9 ± 1.94 (n = 6)
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
5
Fig. 2. Potential temperature (°C) versus salinity (psu) in the whole water column (left panel) and in deep waters (right panel) at the LARGE and SUPER stations.
(CTD data recording and validation as described by Branellec et al., 2010; Kermabon and Arhan, 2008).
methods with a Bran+Luebbe AAIII autoanalyser as described by
Tréguer and Le Corre (1979). Phosphate samples were analyzed
manually by spectrophotometry as described by Murphy and Riley
(1962).
2.6. Chlorophyll-a
Chlorophyll-a concentration was determined after vacuum filtration of 1–2 L of seawater onto GF/F filters. The filters were placed in
90% (v/v) acetone/water and homogenized for 5 min (Guéneuguès
and Boye, 2008). Measurements were performed with a TURNER
Designs 10-AN fluorometer (Parsons and Strickland, 1963; Evans and
Anderson, 1987). Detection limit and precision for chlorophyll-a were
respectively 0.005 μg L−1 and 0.075 μg L−1 (Yentsch and Menzel,
1963; Neveux, 1976).
3. Results
3.1. Hydrography
The cruise track crossed the subtropical domain, the Antarctic
Circumpolar Current (ACC) and entered the Weddell Gyre. The
subtropical domain southwest of Africa (stations L1, S1, and L2) extends
to the southern branch of the Subtropical Front (S-STF), which was
located at ~42°2′S during our cruise. In this study, the ACC (stations S2 to
L7) is bounded to the north by the Subantarctic Front (SAF; Orsi et al.,
1995; Belkin and Gordon, 1996) at ~44°50′S, and to the south by the
Southern Boundary (SBy; Orsi et al., 1995) located at ~55°54.3′S during
the cruise. The Weddell Sea Gyre corresponds to the southern end of the
section (Station S5), as shown in Fig. 1. The T–S diagram for the SUPER
and LARGE stations is presented in Fig. 2.
The BONUS-GoodHope cruise crossed the subtropical domain in
the Cape Basin region which is characterized by a very complex
dynamical regime. Here warm and salty Indian Ocean water
anticyclonic eddies commonly interacts strongly with shelf and slope
waters, cyclones and filaments of South Atlantic origin. The anticyclones are generally referred to as “Agulhas rings” and are ejected from
the western boundary current of the South Indian Ocean, the Agulhas
Current, at its Retroflection (Lutjeharms and Vanballegooyen, 1988).
The rings are constituted of Indian central water (Gordon et al., 1992)
characterized by warm (potential temperature N 10 °C) and salty
(S N 34.5) waters in the upper 800 m (Fig. 2). Station S2 is located
just south of the S-STF and north of the SAF (hence north of the ACC).
Its surface waters exhibited salinity and temperature characteristics of
Indian subtropical waters (Fig. 2), which suggests advection from the
S-STF region by a neighboring eddy. Hence this station is discussed
here within the subtropical domain. Closer to Africa (stations L1 and
S1) we observed Antarctic Intermediate Water of Indian Ocean origin
(I-AAIW) between 800 and 1200 m which is transported in the Cape
Basin region via the Agulhas current and the Agulhas rings (Gordon et
al., 1992). I-AAIW is characterized by salinities N34.3 (Fig. 2). Another
variety of AAIW was observed to the south (at L2 and S2) with
salinities b34.3 (Fig. 2) featuring AAIW formed in the subantarctic
region of the southwest Atlantic (A-AAIW; Piola and Gordon, 1989).
Below AAIW, we observed Upper Circumpolar Deep Water (UCDW)
that is characterized by an oxygen minimum (160–180 μmol kg−1) at
Fig. 3. Vertical distributions of nitrate, phosphate, silicate (µM) in the upper 500 m at the LARGE and SUPER stations (Le Moigne, 2008; Le Moigne et al., in preparation).
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
6
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Fig. 4. Chlorophyll-a (µg L-1) in the top 300 m at the LARGE and SUPER stations (Guéneuguès and Boye, 2008).
~ 1200–1500 m depth (Fig. 6). The sampled UCDW did not show welldefined varieties, but its northern part was also likely transported by
the Agulhas Current and rings (L1 and S1), while it may originate from
the southwest Atlantic (A-UCDW; Whitworth and Nowlin, 1987) at L2
and S2. At deeper depths, a diluted variety of North Atlantic Deep
Water (SE-NADW) which has flowed along the southwest African
continental shelf by following the “southeast route” (Arhan et al.,
2003; Gladyshev et al., 2008) has been identified by its salinity
signature N34.8 (Fig. 2), and an oxygen maximum at 2000–3200 m
depth (Fig. 6) at stations S1 and S2. At the bottom, an old variety of
Antarctic Bottom Water (AABW) likely formed in the Weddell Sea
(Reid, 1989; Gladyshev et al., 2008) was observed in the Cape Basin
abyssal plain at S1 below 3500 m depth which was characterized by
low salinity and cold temperatures (Fig. 2).
In the ACC south of the Agulhas Ridge, the Surface Water (SW) was
marked by a southward decrease of temperature from 4 °C (L5) to 2 °C
(S4) and by low salinity (b34), Fig. 2. Below lays the Atlantic variety of
AAIW (A-AAIW) which was characterized by salinities b34.3 (Fig. 2).
The isohaline layer was depressed at the northern edge, along the SAF,
indicating the subduction of A-AAIW, which was further characterized
by an oxygen maximum N250 μmol kg−1 (Fig. 6). The A-AAIW waters
were detected at stations L3 to L5 between depths of 300 and 600 m.
At greater depths the A-UCDW, which was colder than A-AAIW and
marked by lower dissolved oxygen was found in the 1000–1500 m
layer from station L3 to L5 (Fig. 6). North of the Polar Front (PF, Orsi et
al., 1995; Belkin and Gordon, 1996), deep waters (1500–3000 m,
stations L3 to L5) exhibit properties of diluted South West NADW
(SW-NADW), which flows along the continental slope of South
America, down to the Argentinean Basin before being injected in the
ACC in the southwestern Atlantic (Whitworth and Nowlin, 1987). In
the deepest sampled layers at station S3, below 3250 m on the
northern flank of the Mid Atlantic Ridge, we observed a variety of
bottom water fresher and colder than AABW observed in the Cape
Basin Abyssal plain (Fig. 2). South of the PF another variety of UCDW,
which has passed through the Drake Passage (DP-UCDW) was
identified by a core of low oxygen water (Whitworth and Nowlin,
1987) at depths between 250 and 700 m for stations L6, S4 and L7
(Fig. 6). There, deep waters exhibited properties of Lower Circumpolar
Deep Water (LCDW), with lower salinity than SW-NADW.
South of the ACC domain (station S5), the whole water column
was impacted by Weddell Gyre waters: this water is much colder
(Fig. 2) with a younger variety of AABW observed at the bottom than
that at S3 and in the Cape Basin at S1. These Weddell Gyre waters
were characterized by higher dissolved O2 concentrations and lower
Fig. 5. Contour plot of dissolved cobalt concentrations (DCo; pM) versus depth (m) and latitude along the MD166 BONUS-GoodHope section. The positions of LARGE (L) and SUPER
(S) stations and the fronts are indicated. Black dots represent sampling depths for DCo measurements. The colour mapping extrapolation is based on the sampling resolution along
the section of ~2600 km that was achieved with 12 stations separated on average by ~2.2° latitude (~240 km), with a total of 160 sampling depths. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)
Figure from ODV (Reiner Schlitzer).
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
7
Fig. 6. DCo (pM) and O2 (µmol kg-1) versus depth (m) at the LARGE and SUPER stations.
(O2 data recording and validation as described by Branellec et al., 2010; Kermabon and Arhan, 2008).
salinities and temperature (indicating a less important mixing with
nearby waters than the two varieties of AABW sampled in the Cape
Basin and along the Mid-Atlantic Ridge; Fig. 6).
3.2. Biogeochemical features along the section
Contrasting biogeochemical provinces were crossed along the section,
generally characterized by the southward increase of macronutrient
concentrations (Fig. 3) and associated with the frontal boundaries.
The northern part of the subtropical domain (stations L1 and S1)
was characterized by low nitrate and silicate concentrations, typically
≤5 μM (Fig. 3) and with low phosphate ≤0.5 μM (Fig. 3). Here there
was a subsurface chlorophyll-a (Chl-a) maximum with concentrations ranging between 0.3 and 0.5 μg L−1 (Fig. 4). These features
indicate oligotrophic conditions (Le Moigne et al., in preparation). In
the southern part of the subtropical domain (L2 and S2), surface
silicate concentrations remained low (≤5 μM) but nitrate (5–20 μM)
and phosphate (0.5−1.2 μM) increased as compared with the
northern subtropical region (Fig. 3). Surface waters of these stations
(L2 and S2) were also marked by a subsurface maximum in Chl-a
concentrations with the highest values recorded along the section,
ranging from 0.4 μg L−1 at S2 to 0.6 μg L−1 at L2 (Fig. 4).
In the Polar Frontal Zone, on the northern side of the PF, nitrate and
phosphate concentrations increased to 20 μM and 1 μM respectively,
while silicate remained ≤10 μM (Fig. 3). Ammonium exhibited a
subsurface maximum of ~1.25 μM near the PF between 70 and 100 m
depth (Le Moigne et al., in preparation) and Chl-a concentrations
ranged between 0.2 and 0.4 μg L−1 (Fig. 4), while phaeopigments, a
degradation product of chlorophyll-a, displayed the highest values
recorded along the section (Le Moigne et al., in preparation). All these
features indicate a bloom of diatoms, which probably occurred before
the start of our observations in the vicinity of the PF (Le Moigne et al.,
in preparation).
In the HNLC area between SACCF (S4) and Sby (L7), silicate and
nitrate concentrations both increased with values ranging from 20 to
70 μM for silicate and near 30 μM for nitrate as well as phosphate
levels that increased up to 1.5 μM (Fig. 3). Chl-a concentrations were
low (b0.3 μg L−1; Fig. 4). A slight increase of Chl-a concentration to
0.3 μg L−1 (Fig. 4), associated with an increase of phaeopigments and
a slight decrease of silicate concentrations, was recorded at the
southern end of the section at station S5 located in the Weddell Gyre.
3.3. The meridional and vertical distributions of dissolved cobalt
The distribution of DCo along the section is presented in Figs. 5 and 6,
and the data are provided in Table 1. Dissolved cobalt concentrations
ranged from 5.73 ± 1.15 to 72.9 ± 4.51 pM along the section (all ± terms
reported represent one standard deviation of triplicate analyses of a
single sample unless otherwise indicated). The lowest concentration
was recorded in surface waters south of the S-STF (5.73 ± 1.15 pM at
30 m depth at station S2), whereas the highest level was recorded in the
deeper waters north of the PF (72.9 ± 4.51 pM at station L5 at 700 m).
The distribution of DCo at depth exhibited vertical gradients that
corresponded to different biogeochemical provinces. Indeed the
distribution of DCo shows a nutrient-like profile in the subtropical
domain as well as in the subantarctic region and beyond it (e.g., stations
L1 to L4), with surface depletion in the euphotic layer and increasing
concentrations below 250 m depth (Figs. 5–6). The lowest DCo
concentrations detected in the euphotic layer along the whole section
coincided with the Chl-a maximum at stations L2 and S2 (Fig. 5). In these
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
8
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
domains DCo concentrations then decreased again toward the bottom
below 3000 m depth (e.g., as exemplified by the stations S1 and S2,
Figs. 5–6).
In contrast, DCo was not depleted in surface waters of the Polar
Frontal Zone and between PF and SBdy (stations S3 to L7, Figs. 5–6).
Here the concentrations were relatively high in surface waters (from
35.4 ± 2.5 pM at S3 to 51.3 ± 1.1 pM at L7) and in the core of UCDW
(A-UCDW and DP-UCDW) whereas DCo concentrations occasionally
decreased with depth below these layers to values in the range or
slightly lower than the surface concentrations (Figs. 5–6). In the
Weddell Gyre (station S5) the distribution of DCo was fairly
homogeneous throughout the water column, although the lowest
values were observed at the bottom in the AABW (Figs. 5–6).
In the subtropical domain DCo exhibited concentrations ranging
from 15.6 ± 1.68 pM to 44.3 ± 0.73 pM in the surface Indian central
waters at L1 and S1, and the lowest concentrations encountered in the
surface waters of the section at L2 and S2 (Figs. 5–6). At intermediate
depths (e.g., 200–1000 m) DCo was higher than in surface waters,
with a mean DCo of 49.3 ± 2.04 pM (standard deviation for all samples
measured, n = 4) in the I-AAIW and of 55.8 ± 3.68 pM (n = 4) in the
A-AAIW (Fig. 6, Table 3). Beneath the intermediate depths DCo
concentrations were also high, with a mean value of 57.3 ± 2.18 pM
(n = 14) found in the core of the A-UCDW (Figs. 5–6). In deeper
waters DCo had relatively high values, ranging between 40 and 50 pM,
with the exception of higher DCo imprints in the core of the SE-NADW
sampled at S1 (53.9 ± 3.30 pM, at 3050 m depth) and S2 (59.2 ± 2.05
pM at 2548 m depth), Figs. 5 and 6.
In the ACC domain surface DCo concentrations generally increased
southward, from 22.8 ± 3.71 pM at station L3 to 51.3 ± 1.11 pM at L7,
and the depletion of DCo in the euphotic layer became less and less
pronounced southward (Fig. 6). At intermediate and deep depths
north of the Polar Front, DCo concentrations in the A-AAIW and AUCDW were in the same range as at the more northerly stations (L2
and S2), whereas there was no imprint of high levels in the core of the
SW-NADW were DCo could decrease to 39.5 ± 1.84 pM at 2020 m
depth (Fig. 6). South of the Polar Front DCo reached the highest
concentrations at depth in the core of the DP-UCDW (50.0 ± 2.97 pM,
mean and standard deviation of 7 samples), while underlying deep
DCo values could exhibit lower concentrations (i.e. 36.5 ± 3.40 pM at
1117 m depth, S4). A discrete deep concentration showed, however, a
high value at 1950 m at station S4 (62.9 ± 8.04 pM, Fig. 6).
In the Weddell Gyre (S5) DCo concentrations were ~35 pM throughout most of the water-column, and lower in the AABW (28.1 ± 0.87 pM).
Thus, DCo concentrations in the bottom waters of the Weddell Gyre
were lower than those observed in the older variety of the AABW
Table 3
Comparison and ranges of dissolved cobalt (DCo) in the water column of the Southern
Ocean reported for different sectors: (a) this study; (b) Fitzwater et al., 2000; (c)
Sanudo-Wilhelmy et al., 2002; (d) Ellwood et al., 2005; (e) Ellwood, 2008; (f)
Westerlund and Ohman, 1991; (g) Martin et al., 1990. All seasons indicated are for the
Southern Hemisphere.
Location
Position
Season
Depth
(m)
DCo
(pM)
42–57°S; 0–15°E
Summer
72°S–76°S; 180°W
63–65°S; 41–64°W
48–50°S; 20°E
Summer
Summer
Autumn
e
South East Atlantic
Sector
Ross Sea
Weddell Sea
Atlantic sector
across the Polar Front
Subantarctic zone
40–52°S; 155–160°E
Winter
f
Weddell Sea
60–80°S; 10–60°W
Summer
g
Drake Passage
56°S; 65°W
Summer
Gerlarche Strait
64°S; 63°W
Summer
15–200
200–4000
0.5–375
Surface
0–200
200–1000
0–1200
200–1200
50–200
200–4000
0–200
200–1850
0–200
5–59
24–74
5–41
20–172
15–40
30–45
10–40
20–50
12–46
13–58
~25
20–30
58–82
a
b
c
d
flowing north of the Agulhas Ridge (40.0 ± 2.49 pM, mean and standard
deviation of 2 different samples).
4. Discussion
Relatively few studies of DCo distributions in the Southern Ocean
have been reported, but they provide ranges of DCo of the same
general magnitude as that reported here (Table 3). More precisely,
DCo concentrations reported here from the upper 1000 m on both
sides of the PF were somewhat higher (35–75 pM) than those
reported by Ellwood et al. (2005) along the 20°E meridian (15–50
pM). Differences in the pretreatment of the samples might cause such
discrepancy since the samples were acidified (pH ~ 2) before the UVdigestion step in the present study whereas they were UV-digested
without acidification in the work of Ellwood et al. (2005) possibly
yielding lower recovery of DCo. The acidification of the sample
coupled to the UV-oxidation may promote the release of DCo from
strong organic ligands, thus increasing the recovery of DCo (Donat
and Bruland, 1988; Vega and van den Berg, 1997; Saito and Moffett,
2001; Noble et al., 2008; Shelley et al., 2010). In the Weddell Sea
western rim, previously reported DCo concentrations in surface
waters ranged between 18.4 and 71.1 pM, with a mean of 53.0 ±
16.7 pM (Sanudo-Wilhelmy et al., 2002). The concentrations detected
in the present study, further east in the northern side of the Weddell
Gyre, were somewhat lower, with a mean of 36.90 ± 2.70 pM (n = 6)
between 0 and 250 m and 32.10 ± 4.50 pM (n = 8) in deeper waters
(800–3840 m) at station S5 (Fig. 6, Table 1). However the DCo
concentrations we report here compare well with values observed in
the surface waters of the Weddell Sea (30–40 pM), as well as with
deep concentrations, which were fairly uniform around 26 pM
(Westerlund and Ohman, 1991) (Table 3).
In the present study the vertical distribution of DCo evolved
southward from a nutrient-like behavior in the subtropical domain and
the subantarctic zone, to more complex distributions within the Polar
Frontal Zone including scavenging-type behavior, and to a conservative
behavior in the Weddell Gyre region. These distributions reflect the
several biogeochemical and physical processes and the different external
sources that play important roles in the oceanic cycle of DCo.
4.1. Biogeochemical cycle of cobalt in the subtropical region
The distribution of DCo in the subtropical domain was nutrientlike, with low levels in surface waters and the lowest concentrations
recorded where Chl-a levels were highest (Figs. 6 and 4). This
suggests biological uptake of cobalt in this region. Similar surface
depletion of DCo has been reported in the subtropical region of the
South Pacific (Ellwood, 2008), in the northeast Pacific (Martin et al.,
1989), in the North Atlantic (Martin et al., 1993), and in the Sargasso
Sea (Saito and Moffett, 2001), as well as in other oligotrophic systems
(Martin et al., 1989; Saito and Moffett, 2002), upwelling waters (Saito
et al., 2004) or near islands (Noble et al., 2008), all indicative of
biological utilization. For instance the cyanobacteria Prochlorococcus
sp. and Synechococcus sp., which often dominate the picophytoplankton assemblage in oligotrophic regions, have an absolute cobalt
requirement for growth (Saito and Moffett, 2002). Biological uptake
by cyanobacteria can be the dominant removal mechanism for cobalt
in surface waters of the oligotrophic Sargasso Sea (Saito and Moffett,
2001). Similarly uptake by cyanobacteria may be sufficient to cause
the observed depletion of DCo in the oligotrophic region on the
northern side of the section. Furthermore the strongest removal of
DCo was observed in the southernmost subtropical region where
nanoflagellates (b5 μm, mostly unidentified species) dominated the
phytoplankton biomass and dinoflagellates were also abundant (B.
Becker, LEMAR, France, pers. comm.). Dinoflagellates can have a
relatively high Co cellular quota compared to other phyla such as
diatoms and chlorophyceae sp. (Ho et al., 2003), but in the range of the
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Fig. 7. [HPO24 ] (µM) versus DCo (pM) in the top 150 m at the LARGE and SUPER stations
(Le Moigne, 2008; Le Moigne et al., in preparation).
Co quota of cyanobacteria (e.g., Co:C of 0.08 to 1.43 μmol mol−1 for
Synechococcus bacillaris; Sunda and Huntsman, 1995). Nanoflagellates, such as prasinophyceae, can also have Co quota similar than
those of diatoms, although their Co quota are lower compared to
dinoflagellates (Ho et al., 2003). It is thus possible that Co was used by
dinoflagellates and nanoflagellates in addition to cyanobacteria.
Moreover the uptake of DCo may be proportional to that of major
nutrient phosphate, as previously observed in other oceanic areas
(Martin et al., 1989; Saito and Moffett, 2002; Saito et al., 2004; Jakuba
et al., 2008; Noble et al., 2008). In the present study a linear
relationship between DCo and phosphate concentrations was observed in the euphotic layer (0–150 m) of the whole section. Although
the correlation is relatively weak when all data are considered
(r2 = 0.424, n = 45; Fig. 7), a much stronger relationship was detected
in the oligotrophic region (r2 = 0.914; n = 5 at S1) with a DCo:HPO2−
4
depletion ratio of 49.3 μM M−1, as well as in the southern side of the
subtropical domain (r2 = 0.988, n = 5 at S2) with a DCo:HPO2−
4
depletion ratio of 44.0 μM M−1 (Fig. 7). In this region the DCo:HPO2−
4
depletion ratio is in the same range as that reported in the northeast
−1
Pacific (DCo:HPO2−
; Sunda and Huntsman, 1995; Martin
4 ~ 40 μM M
et al., 1989) and near the Hawaiian Islands (37 μM M−1; Noble et al.,
2008). By contrast, a much higher ratio has been reported in the
Equatorial Atlantic (DCo:HPO2−
depletion ratio = 560 μM M−1)
4
where concentrations of phosphate are in the low nanomolar range
(40–140 nmol L−1) and DCo range from 5 to 87 pM at 5 m depth
(Saito and Moffett, 2002) compared to 0.1 to 1.0 μM and 5 to 45 pM for
phosphate and DCo levels respectively observed in the present study.
Such differences in phosphate and DCo concentrations could explain
the contrasted DCo:HPO2−
4 depletion ratio observed in the Equatorial
Atlantic and those found in the subtropical south-west Atlantic.
Furthermore, several factors could account for regional differences in
the DCo:HPO2−
depletion ratio, such as differences in the biological
4
community composition, biochemical substitution of zinc for cobalt in
eukaryotic phytoplankton and changes in the bioavailability of cobalt
(Noble et al., 2008). In deeper waters the nutrient-like distribution of
DCo observed in the subtropical domain (Fig. 6) suggests remineralization of DCo. For instance the greatest increase of DCo with depth
was observed in the mesopelagic zone (200–600 m) of the southern
side of the subtropical region, where remineralization of particulate
matter was found to be intense as exemplified by high levels of
particulate barite (D. Cardinal, RMCA, Belgium, pers. comm.). The
increase in DCo in the mesopelagic zone was not proportional to that
9
of phosphate, suggesting a decoupling between the cobalt and the
phosphate cycles in intermediate and deep waters. Furthermore the
DCo:HPO2−
depletion ratios are lower at depth than in surface layer,
4
indicating that the apparent remineralization of DCo is lower than
that of phosphate in intermediate and deep waters. In turn the
internal cycle of cobalt in the subtropical domain may be described by
combining a nutrient-like cycling like that of major nutrient
phosphate, with a lower regeneration of Co compared to P and/or
with additional (versus P) removal of Co in the mesopelagic zone.
Both later processes would account for the decoupling between Co
and P in deep waters.
The internal processes of Co cycling within the ocean furthermore
interact with external sources. There are potentially at least two such
sources in the subtropical domain, aeolian input of terrestrial dust into
the surface waters and enrichment from continental shelf and slope
waters. Aeolian deposition of cobalt may generate a near-surface DCo
maximum as reported in the western Philippine Sea (N100 pM at
48 m depth; Wong et al., 1995) and recently in the Sargasso Sea at the
Bermuda Atlantic Time-series Study (BATS) site (Milne et al., 2010).
However at the northern stations likely to be the most affected by
aerosol deposition across the section (stations L1, S1, L2 and S2) the
low surface DCo concentrations (≤ 45 pM in the upper 100 m, Fig. 6,
Table 3) do not indicate a significant atmospheric input of cobalt,
originating either from the South Africa or from Patagonia (Li et al.,
2008), unlike dissolved iron (Chever et al., 2010).
Nevertheless at stations L1 and S1, DCo distribution displayed
relative DCo maxima in the top 100 m depth, especially at S1 where
the relative maximum ranged between 40 and 45 pM at 70–100 m
depth (Fig. 6). Although the biological uptake could be lower at those
stations compared to the southern subtropical region (stations L2 and
S2) according to the Chl-a levels (Fig. 4), which in turn may generate a
relative maximum, it is possible that the DCo inputs were comparatively higher in the surface waters of the northernmost stations.
Particulate aluminum distributions indeed shown enrichment in
surface waters at station S1 compared to stations located more in the
south, but it was not possible to discern whether the lithogenic inputs
there resulted from atmospheric deposition and/or from the transportation of particles from the Agulhas Bank originated either from
river discharges or from sediment resuspension above the margin and
the continental slope (Jeandel et al., 2010; F. Lacan, LEGOS, France,
pers. comm.). On the other hand, the stronger depletion of DCo
observed in surface waters near the S-STF (stations L2 and S2) (Fig. 6)
is marked by the highest subsurface maximum Chl-a concentrations
encountered along the section (Fig. 4). There, near-surface DCo
removal by biological utilization could potentially mask any lithogenic
inputs from the atmosphere or from the advection of enriched surface
waters. Nevertheless, analyses of dry aerosols sampled during the
cruise gave an estimation of the dry deposition flux of soluble Co of
0.16–0.33 nmol m−2 d−1 in surface waters of the subtropical domain
(A. Baker, UEA, United Kingdom, pers. comm.). Considering a 4 month
deposition episode, such input would contribute for an increase of
0.40 to 0.80 pM DCo in the mixed layer, hence representing ≤2% of the
concentration measured in the mixed layer. Another study showed
comparable estimates and suggested that the dissolution of cobalt
from a natural dust event may account for an enrichment of only
~0.01 to 0.43 pM DCo to the mixed layer (Thuróczy et al., 2010); that
would represent ≤1.2% of the concentration measured in the mixed
layer of the northern part of the section. Thus, aeolian inputs would
not be a sufficient Co input to generate a distinct surface maximum
unless they were much greater than those occurring during the cruise
or previously estimated by Thuróczy et al. (2010), especially when
phytoplankton uptake is also occurring. In turn the relative DCo
maxima observed in surface waters of the northern subtropical
domain may further originate from the advection of Indian waters
that interact strongly with shelf and slope waters, as well as with river
discharges, of the Agulhas Bank.
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
10
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Maxima of DCo were observed in intermediate and deep waters in the
cores of I-AAIW and SE-NADW, as well as those of A-AAIW and A-UCDW
(Fig. 6), suggesting lateral advection of Co-enriched waters at those
depths. Benthic and sediment inputs, and transport from continental shelf
and slope waters towards the open ocean, may be important sources of
trace metals to the deep ocean (Elderfield et al., 1981; Bowie et al., 2002;
Noble et al., 2008; Zhang et al., 2008b), notably for DCo (Noble et al., 2008;
Huang and Conte, 2009). Moreover deep Co inputs might be stabilized as
Co organic complexes and generate high deep DCo concentrations (Saito
and Moffett, 2002). The organic complexation of DCo along the section
will be examined elsewhere (Bown et al., in prep.). The I-AAIW had
flowed along the southern margin of South Africa (Gordon et al., 1992)
whereas the SE-NADW followed the southwestern African margin before
passing by the southeast Atlantic (Arhan et al., 2003).
Hence it is possible that the high DCo found in these water-masses
was transported from the sediments of the South African margins.
Particulate aluminum (Jeandel et al., 2010) and dissolved iron (Chever
et al., 2010) also displayed high values in these water masses, further
supporting the hypothesis of a significant lithogenic source from the
margins of South Africa in this region. Similarly distinct depth maxima
in DCo were observed in the Pacific waters surrounding Hawaiian
Islands, suggestive of the release of cobalt from areas of comparable
depth that surrounds those islands (Noble et al., 2008).
A-UCDW also exhibited high DCo concentrations that are close to
those found in the A-AAIW; both of these water masses were formed
in the southwest Atlantic (Whitworth and Nowlin, 1987) suggesting a
source of cobalt in the region of formation and/or an enrichment of
DCo between the southwest Atlantic and the southeast Atlantic due to
remineralization at intermediate depths.
4.2. The cycle of cobalt in the ACC and the Weddell Gyre
The distribution of DCo changed from a nutrient-like behavior
north of the Polar Frontal Zone toward a more conservative profile in
the Weddell Gyre (Fig. 6). This trend was mostly attributable to a
southward increase of DCo in the euphotic layer, as there was no
significant meridional gradient in the deep concentrations (Fig. 6).
In the northern ACC domain (stations L3 and L4), depletion of DCo in
surface waters (~25 pM) was observed within the relative Chl-a
maxima (e.g., 0.25 μg L−1 at L3; 0.4 μg L−1 at L4) and there, nanoflagellates still dominated the phytoplankton assemblage (B. Beker,
pers. comm.). Dinoflagellate abundance was slightly higher at L4 and
nearly twice as high at L3 as those observed in the subtropical domain.
Hence biological uptake may generate the nutrient-like distribution of
Co north of the ACC. Here, the uptake of DCo was proportional to that of
phosphate, since the correlation between DCo and phosphate was
significant (r2 = 0.870, n = 4 at L4) with a DCo:HPO2−
4 depletion ratio of
48.0 μM M−1 (Fig. 7). The DCo:HPO2−
depletion ratio observed at
4
station L4 was in the same range than those recorded on the southern
side of the subtropical domain, indicating that DCo and phosphate could
be biologically used in the same proportions in both biogeochemical
provinces.
In the PF region and the Weddell Gyre, there was no significant
2−
correlation between DCo and HPO2−
4 and no clear trend in the DCo:HPO4
depletion ratio associated with the change of DCo behavior. The PF region
has been characterized by a vast post-diatom-bloom condition (Le Moigne
et al., in preparation). Diatoms dominated the phytoplankton biomass (B.
Becker, pers. comm.), and in the Weddell Gyre this assemblage contained
degraded frustules with small or absent chloroplasts also suggesting a late
stage of a relatively minor diatom bloom (Le Moigne et al., in preparation).
These observations suggest that biological uptake, especially by diatoms,
may not be the dominant mechanism for removal of DCo from surface
waters at those latitudes. This observations contrasts with the relatively
strong depletion of silicate (Le Moigne et al., in preparation) and iron
(Chever et al., 2010) observed on the southern side of the Polar Front, but
it is consistent with culture experiments, which have shown that diatoms
use much more iron, and to a lesser extent zinc, than cobalt to grow
(Sunda and Huntsman, 1995). For instance zinc distributions were
nutrient-like in the PF region indicating biological utilization of zinc,
although zinc concentrations still kept at high values (≥2 nM) in the
surface waters south of the PF at the Greenwhich Meridian (Baars and
Croot, in press). So there would be no reason for a high cellular demand for
Co (Ho et al., 2003; Baars and Croot, in press). On the other hand the rather
low DCo observed in the subsurface waters south of the PF suggests that
another process was at work to remove DCo from the surface. For instance
these subsurface waters were characterized by relatively high O2 levels
(N300 μmol kg−1, Fig. 6), at which cobalt may be in the less soluble Co(III)
state (Saito and Moffett, 2002).
The highest DCo concentrations in surface waters were recorded at
the SBdy latitudes (e.g., L7 station, Fig. 6). The low biological uptake of
cobalt may have led to relative cobalt accumulation there, although
snow that fell at those latitudes during the expedition could be a source
of DCo in surface waters. The concentrations of cobalt in snow could
indeed be 4–5 times higher than in surface waters, with mean values of
137 pM measured in snow depositing on the Wasa Shelf in Antarctica
(Westerlund and Ohman, 1991). In addition to the input by snow, dry
aerosol deposition flux was 1.3 to 4 times higher at the polar latitudes
than in the subtropical domain, with soluble Co deposition flux ranging
from 0.42 to 0.66 nmol m−2 d−1 (A. Baker, pers. comm.). Such
deposition could increase DCo concentrations in the mixed layer by
0.60 to 1.40 pM over a 4 months dust deposition event, representing
only ~2–4% of DCo recorded in the mixed layer. Comparison of DCo in
surface waters previously found in the northern (27.1 pM) and southern
(25.1 pM) Drake Passage (Martin et al., 1990; Table 3) with,
respectively, those recorded at L6 (39.0 ± 1.54 pM) and at stations S4
and L7 (45.8 ± 1.0 pM, n = 2) indicates an enrichment from the Drake
Passage to the 0° meridian. It suggests that the transport of Patagonian
dust by the westerly winds and their deposition in the surface ACC can
still be a source of DCo, similar to iron (Cassar et al., 2007). In addition
the higher surface DCo concentrations recorded at Gerlache Strait (58–
82 pM at 15–200 m depth; Martin et al., 1990; Table 3) were probably
due to the proximity of the Antarctic Peninsula, which could be
identified as another potential DCo source in surface waters of the ACC.
At intermediate depths, the nutrient-like profiles of DCo observed
north of the PFZ furthermore indicate remineralization of cobalt in the
nutricline (Figs. 3–6); this is consistent with the particulate barite
distribution (D. Cardinal, pers. comm.). In contrast, intermediate and
deep concentrations of DCo in the central and southern PFZ and
between PF and SBdy decreased with depth, as magnified at the SACCF
(Fig. 6). This suggests that cobalt may be scavenged in the water
column of the central and the southern ACC domains by adsorption
onto living, dead and mineral particulate material (Brown and Parks,
2001). For instance the strongest removal of DCo in deep waters at the
SACCF is in line with the most intense export of particles recorded
234
238
along the section inferred by Th/ U measurements (F. Planchon
and F. Dehairs, VUB, Belgium, pers. comm.). Scavenging-type
distribution has been previously suggested for cobalt in ocean waters
receiving atmospheric dust (Wong et al., 1995).
Dissolved cobalt concentrations were high in the fast eastwardflowing jet of the ACC in the cores of A-AAIW and A-UCDW north of
the PF, as well as DP-UCDW south of the PF (Fig. 6). These waters have
been flowing along the South American continental slope before being
injected in the ACC in the southwestern Atlantic (Whitworth and
Nowlin, 1987), and the DP-UCDW passed through the Drake Passage
(Whitworth and Nowlin, 1987). Those water masses may thus have
received sedimentary inputs from the Drake Passage as well as inputs
from the margins of South America for those flowing north of the PF,
and of the Antarctic Peninsula for those flowing south of the PF. Hence
it is possible that the high DCo measured in these water-masses was
transported from these source regions within the rapidly flowing ACC,
which displays eastward velocities of ~0.16 m sec−1 in the Atlantic
sector (Cunningham and Pavic, 2007; Gladyshev et al., 2008). The
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
transport of DCo from the source regions towards the 0° meridian is
conceivable since the mean oceanic residence time estimate of 40–
150 year for DCo (Saito and Moffett, 2002) is 20 to 75 times longer
than a mean transit time estimate of 2 year from the Drake passage to
1°E (Blanke et al., 2006). Comparison of DCo concentrations measured
in the core of DP-UCDW at S4 and L7 stations (50.0 ± 2.97 pM, n = 7)
with those recorded in the same water-mass at the Drake Passage
(~24 pM, Martin et al., 1990; Table 3) suggests an enrichment of DCo
in the DP-UCDW from the Drake Passage eastward towards the 0°
meridian. Hence in addition to the advection from the source regions,
DCo may also be regenerated along the transit of these water-masses
between the Drake Passage and the 0° meridian, as it may stabilize to a
more soluble redox state in the A-UCDW and DP-UCDW which are
characterized by an oxygen minimum of ~ 170–180 μmol kg−1.
Indirect evidences based on the chemical similarities between Co
and manganese (Mn), especially considering the redox speciation, can
further support the stabilization of Co in less oxygenated waters. For
instance both Co and Mn could be co-oxidized via a common
microbial pathway in regions of high microbial Mn oxidizing activity
(Moffett and Ho, 1996). Furthermore comparable maximum concentrations of Mn and Co have been reported near the oxygen minimum
layer in deep waters (Johnson et al., 1996; Pohl et al., 2011). The
reduction of Mn oxides from margin sediments that intersect the
oxygen minimum, followed by the transport of soluble Mn into the
open ocean could generate such maxima (Johnson et al., 1996). It is
11
therefore possible that the maximum of DCo observed in the A-UCDW
and DP-UCDW resulted from the reduction of Co(III) oxides from
margin sediments and transport of the more soluble Co(II) in those
less oxygenated water masses.
In the bottom waters lower DCo concentrations (20–30 pM) were
observed in the recently formed AABW near the Weddell Gyre
seafloor as compared to the higher values (30–40 pM) observed in the
older variety of AABW that flowed north of the Agulhas Ridge (Fig. 6).
This may reflect enrichment with DCo as AABW flows from south to
north across topographic features like the Agulhas Ridge, possibly
from abyssal sediments as bottom waters flow northward, and
perhaps from hydrothermal vents, which are a source of trace metals
such as iron (Bennett et al., 2008). However there was no record of
high DCo at the bottom of the Mid Atlantic Ridge (Fig. 5), whereas
high values of dissolved iron concentrations suggested hydrothermal
inputs (Chever et al., 2010). This suggests that hydrothermal activity
may not be acting as a significant source of DCo as it is for Ca, Mn and
Fe (German et al., 1991; Tagliabue et al., 2010).
4.3. Comparative biogeochemical Co budgets in the subtropical South
Atlantic and the Southern Ocean
Tentative budgets for DCo in the mixed layer of the subtropical and
ACC domains are presented in Table 4. The budgets are based on the S
stations located in the STZ (S1 and S2), in the PFZ (S3), at the SACCF
Table 4
Summary of cobalt pools and fluxes for the surface mixed layer at each SUPER (S) stations during BONUS-Goodhope expedition.
STZ
MLD (m)
Permanent pycnocline (m)
Mean DCo in the mixed layer (pM)
(1) PCoN 0.4 μm (pmol L−1)
(2) TPOC (fine particles 1–53 μm) + (large particles N53 μm) (μmol L−1)
(2) POCN 53 μm (μmol L−1)
(3) PCoN 53 μm estimation (pmol L−1)
(4) C export from the mixed layer (mmol m−2 d−1)
Fluxes (nmol m−2 d−1)
(A) Vertical diffusive DCo supply
(B) Lateral advective DCo supply
(C) Soluble cobalt supply from dry aerosols
(D) Downward PCo export flux
(E) Cobalt uptake (estimated from cobalt uptake rate by phytoplankton)
Cobalt uptake (estimated from NPP)
Net cobalt uptake (estimated from Co net seasonal depletion)
(F) Vertical advective DCo supply
PFZ
SACCF
Weddell Gyre
S1
S2
S3
S4
S5
54
585
21.9 ± 1.93
2.30 ± 0.18
1.11
0.13
0.28
1.7 ± 0.1
51
775
6.88 ± 1.10
1.69 ± 0.17
3.60
0.40
0.19
1.7 ± 0.2
60
139
35.4 ± 1.54
1.49 ± 0.17
3.57
0.25
0.10
3.3 ± 0.2
88
169
38.9 ± 1.49
0.53 ± 0.18
1.74
0.26
0.08
6.0 ± 1.4
90
110
36.4 ± 0.55
1.27 ± 0.21
2.12
0.55
0.33
3.3 ± 1.4
(+) 0.04
(+) 17–28
(+) 0.2–0.3
(−) 2.5
(−) 47–94
(−) 55–73
nd
Negligible
(+) 0.20
nd
(+) 0.2–0.3
(−) 0.8
(−) 47–94
(−) 55–73
(−) 6.0
Negligible
(+) 0.12 ± 0.02
(−) 0.87
(+) 0.4–0.7
(−) 1.4
(−) 13
(−) 18–36
nd
Negligible
(+) 0.46 ± 0.08
(−) 1.7
(+) 0.4–0.7
(−) 1.8
(−) b 13
(−) 9.0–18
(−) 3.3
(+) 6.4
(+) 0.33 ± 0.06
negligible
(+)0.4–0.7
(−) 2.0
(−) b 13
(−) 9.0–18
nd
Negligible
Note: nd means not determined
(1) 1 data point available in the mixed layer (data from F. Lacan et al.).
(2) 1 data point available in the mixed layer (data from F. Planchon et al.).
(3) Using the relationship: PCoN 53 μm = PCoN 0.4 μm (POCN 53 μm/TPOC) considering that PCo and POC are covarying in each size-fraction of the particulate matter.
(4) Estimated from total
234
234
Th export fluxes (0–100 m) and POC/ Th ratios of sinking particles (N53 μm) (data from F. Planchon et al.).
(A) Calculated from DCo gradients across the pycnocline and a Kz of 0.3 cm2 s−1 (Kawabe, 2008) in the subtropical domain (S1 and S2) and a Kz of 0.66 ± 0.11 cm2 s−1 (Croot et
al., 2007) in the ACC domain (S3 and S4) and the Weddell Sea gyre (S5).
(B) In the PFZ and SACCF: estimated from DCo concentrations north and south at the Drake Passage (Table 3; Martin et al., 1990), a velocity of 0.17 m s−1 (Cunningham and
Pavic, 2007; Gladyshev et al., 2008) and a distance of 10 575 km between Drake Passage and the 0° meridian. It is also considered that mixed layer depths are similar at the
Drake Passage and at stations S3 and S4. This flux is negligible at S5 in the light of surface concentrations previously measured in the Weddell Sea (Table 3). In the STZ, see
details of the estimation of the flux in the text.
(C) Data from A. Baker.
(D) Using the relation: PCoexported
flux = Cexport
(PCoN 53 μm/POCN 53 μm).
(E) In the subtropical domain, the flux was estimated from the mean cobalt uptake rate of Prochlorococcus (Saito and Moffett, 2002) of 14.3 atom cell−1 h−1 obtained at [Co2+] b 1 pM,
and from a cells density ranging between 50 000 and 100 000 cell mL−1 (Partensky et al., 1999; Zhang et al., 2008a). In the Polar Front region (S3), the cobalt uptake rate was
estimated from a fucoxanthine concentration of 63.7 ng L−1 similar to the recorded field values (J. Ras and H. Claustre, pers. comm.) and the corresponding cobalt uptake rate
of 0.21 pmol L−1 d−1 reported by Saito et al. (2010). At S4 and S5, the fucoxanthine concentrations were too low (J. Ras and H. Claustre, pers. comm.) to find a corresponding cobalt
uptake rate. The second estimation of the cobalt uptake flux was based on New Primary Production values in the study area (Field et al., 1998) and an assumed phytoplankton Co/C
ratio of 1.6 × 10−6 (mol mol−1), a value based on data reported for oceanic species (Ho et al., 2003). A third estimate is given for station S2 and S4. At S2 the DCo net seasonal
depletion was estimated from DCo concentration found south of the STF during winter (Ellwood, 2008) whereas at S4 DCo net depletion was estimated from DCo concentration
recorded in Antarctic Winter Waters.
(F) The supply of DCo by upwelling at S4 was defined as vertical velocity (0.13 m d−1, de Baar et al., 1995) deep DCo mean concentration. This term includes contribution from
remineralization.
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
12
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
(S4) and in the Weddell Gyre (S5), since the full set of cruise
parameters (chemical, biological, geochemical and hydrographic) was
available only at those stations. Some fluxes have been estimated
using parameters from the literature. Nevertheless some parameters
cannot be estimated. For example, we are aware of no information on
rates of DCo regeneration in the mixed layer, and the budgets
described below suggest that this was probably a significant source of
DCo.
The estimated cobalt uptake flux was estimated in two ways. One
estimate was derived from phytoplankton biomass, as estimated from
measured photosynthetic pigment concentrations, and literature
values for uptake of Co by phytoplankton (Saito et al., 2010). The
other was derived from satellite-based estimates of annual net
primary productivity in the study area (Field et al., 1998) and an
assumed phytoplankton Co/C ratio of 1.6 × 10−6 (mol mol−1), a value
based on data reported for oceanic species (Ho et al., 2003). These two
independent estimates agree reasonably well (Table 4) and each
indicates that uptake by phytoplankton was the dominant Co flux at
all stations considered. In addition, both estimation methods show Co
uptake by phytoplankton decreasing from north to south on the
section (Table 4). This is in line with the observation that DCo showed
a nutrient-like vertical distribution in the STZ, and that this changed
gradually to a more conservative distribution in the Weddell Gyre.
All the budgets were negative, indicating missing or underestimated source terms, overestimated removal fluxes or some
combination of those two effects. It seems likely that the most
important Co source not accounted for in these budgets is the
biological regeneration of DCo within the upper water column. We
have estimates of seasonal net Co removal (biological uptake minus
regeneration) in the mixed layer at stations S2 and S4, where winter
DCo values are available from comparable surface waters (Ellwood,
2008). At S2, the net removal rate estimated in that way is only ~10%
of the uptake rate estimated by each of the two methods described
above, while at S4 it is ~ 30% of those rates. This excess of biological
uptake over seasonal net removal implies that regeneration in surface
waters is a very important Co flux throughout the study area, ranging
from ~ 70% of biological uptake at the SAACF to ~90% in the STZ. The
budgets become less negative from north to south, reflecting the
decrease of the estimated cobalt uptake flux from the STZ to the
Weddell Gyre.
At station S1 the estimated lateral advective DCo supply ranges
from 17 to 28 nmol m−2 d−1, based on estimated inputs from the
South African continental shelf (Table 4). This flux was estimated
from a hypothetical coastal DCo concentration of 100 pM, based on
coastal concentrations measured by Saito and Moffett (2002) and
Kremling and Streu (2001), a distance of ~650 km between the
continental shelf and station S1 and a surface water velocity ranging
from 0.03 to 0.05 m s−1 according to Agulhas Rings velocities of
translation (Boebel et al., 2003). Supply from the nearshore waters
could thus be a significant source term at this station. Lateral
advective sources appear to be only minor terms in the Co budget at
the other sites considered (Table 4).
Vertical diffusive supply, aeolian supply from dry aerosols and
downward export of particulate Co all appear to be minor terms in the
regional DCo budget (Table 4). Vertical supply via upwelling also
appears to be minor at most stations, but at S4 this flux is estimated to
be sufficient to supply approximately half of the DCo taken up by
phytoplankton (Table 4).
5. Conclusion
The biogeochemical cycle of DCo differs in the subtropical domain
of the southeastern Atlantic and in the Southern Ocean. The biological
uptake of DCo proportional to that of phosphate, and the remineralization of cobalt in the mesopelagic zone mostly generate the
nutrient-like distribution of DCo in the subtropical domain. In the ACC
domain, the uptake by diatoms was apparently not a significant sink
of DCo in surface waters and this feature was also evident in budgets.
There, scavenging of DCo onto particles may occur at intermediate
depths. However the removal by adsorption onto particles was
difficult to discern, as deep profiles were strongly impacted by
different water masses which were identified as potential sources of
DCo.
The external sources of DCo may be driven by margin input and
transport in both the subtropical and ACC domains. South African
margins have indeed been identified as potential sources of DCo into
the intermediate and deep waters of the subtropical domain with
additional input in surface waters at the northernmost stations.
Furthermore DCo transport from source regions in the Drake Passage
and southwestern Atlantic, carried by the eastward flow of the
Antarctic Circumpolar Current, could bring DCo at the Greenwich
Meridian. In addition, A-UCDW and DP-UCDW exhibited the highest
DCo concentrations encountered in this study. Thus, those water
masses could be considered as the main reservoirs of DCo along the
section. Each of those water masses was characterized by relatively
low O2, which may have promoted stabilization of DCo into the more
soluble Co(II) state in the eastward flow of the ACC. Similarly the low
DCo concentrations observed in surface waters both north and south
of the PF may be partly due to the decrease in the solubility of cobalt as
the surface waters there had higher dissolved oxygen concentrations.
Water mass characteristics, especially the absence of oxygenation,
may hence play a crucial role in DCo solubility and residence time thus
requiring further investigations.
Acknowledgments
The Captain P. Courtes, the officers and the crew members of the
French research vessel Marion Dufresne II are warmly acknowledged
for their wonderful work at sea during the IPY MD166 BONUSGOODHOPE cruise. S. Speich and M. Boye were co-chief scientists on
the cruise, and we thank as well their collaborators M. Arhan (LPO,
Brest, FR) and F. Dehairs (VUB, Brussels, Belgium). We are grateful to
the colleagues of our GO-FLO sampling team (Fanny Chever, Bronwyn
Wake, Eva Buciarrelli, Géraldine Sarthou, and Amandine Radic). We
thank Beatriz Becker, Damien Cardinal, Frank Dehairs, Marie Labatut,
Hervé Claustre and Joséphine Ras for providing access to their
unpublished data on phytoplankton species composition, barite
234
238
distributions, Th/ U, pigments and particulate cobalt, respectively.
We also thank the two anonymous reviewers of this paper for
constructive comments which significantly improve this manuscript.
This investigation was supported by the BONUS-GOODHOPE project
and cruise funded by the French LEFE National Program of INSU, the
National Agency of French Research (ANR–07–BLAN–0146–01) and
the Paul Emile Victor Institut. It is a contribution to the International
Polar Year and the international GEOTRACES programs. We also thank
the European COST Action ES801 for funding a Short Term Scientific
Mission to J. Bown to be trained on the FIA cobalt manifold with the
precious help of Rachel Shelley and Maeve Lohan (Plymouth
University, U.K.). The Region Bretagne is supporting the PhD
fellowship of J. Bown.
References
Arhan, M., Mercier, H., Park, Y.H., 2003. On the deep water circulation of the eastern
South Atlantic Ocean. Deep-Sea Research Part I—Oceanographic Research Papers 50
(7), 889–916.
Baars, O., Croot, P.L., in press. The speciation of dissolved zinc in the Atlantic sector of
the Southern Ocean. Deep-Sea Research II. doi:10.1016/j.dsr2.2011.02.003.
Belkin, I.M., Gordon, A.L., 1996. Southern Ocean fronts from the Greenwich meridian to
Tasmania. Journal of Geophysical Research—Oceans 101 (C2), 3675–3696.
Bennett, S.A., Achterberg, E.P., Connelly, D.P., Statharn, P.J., Fones, G.R., Gernian, C.R.,
2008. The distribution and stabilisation of dissolved Fe in deep-sea hydrothermal
plumes. Earth and Planetary Science Letters 270 (3–4), 157–167.
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Bertrand, E.M., Saito, M.A., Rose, J.M., Riesselman, C.R., Lohan, M.C., Noble, A.E., Lee, P.A.,
DiTullio, G.R., 2007. Vitamin B-12 and iron colimitation of phytoplankton growth in
the Ross Sea. Limnology and Oceanography 52 (3), 1079–1093.
Blanke, B., Arhan, M., Speich, S., 2006. Salinity changes along the upper limb of the
Atlantic thermohaline circulation. Geophysical Research Letters 33 (6).
Boebel, O., Lutjeharms, J., Schmid, C., Zenk, W., Rossby, T., Barron, C., 2003. The Cape
Cauldron: a regime of turbulent inter-ocean exchange. Deep-Sea Research Part II—
Topical Studies in Oceanography 50 (1), 57–86.
Bowie, A.R., Whitworth, D.J., Achterberg, E.P., Mantoura, R.F.C., Worsfold, P.J., 2002.
Biogeochemistry of Fe and other trace elements (Al, Co, Ni) in the upper Atlantic
Ocean. Deep-Sea Research Part I—Oceanographic Research Papers 49 (4), 605–636.
Bowie, A.R., Achterberg, E.P., Ussher, S., Worsfold, P.J., 2005. Design of an automated
flow injection-chemiluminescence instrument incorporating a miniature photomultiplier tube for monitoring picomolar concentrations of iron in seawater.
Journal of Automated Methods & Management in Chemistry (2), 37–43.
Branellec, P., Arhan, M., Speich, S., 2010. Projet GoodHope, Campagne BONUSGoodHope, Rapport de données CTD-O2. Rapport Interne Ifremer, OPS/LPO/1002., LPO-IFREMER.
Brown, G.E., Parks, G.A., 2001. Sorption of trace elements on mineral surfaces: modern
perspectives from spectroscopic studies, and comments on sorption in the marine
environment. International Geology Review 43 (11), 963–1073.
Cannizzaro, V., Bowie, A.R., Sax, A., Achterberg, E.P., Worsfold, P.J., 2000. Determination
of cobalt and iron in estuarine and coastal waters using flow injection with
chemiluminescence detection. The Analyst 125 (1), 51–57.
Cassar, N., Bender, M.L., Barnett, B.A., Fan, S., Moxim, W.J., Levy, H., Tilbrook, B., 2007.
The Southern Ocean biological response to aeolian iron deposition. Science 317
(5841), 1067–1070.
Chever, F., Bucciarelli, E., Sarthou, G., Speich, S., Arhan, M., Penven, P., Tagliabue, A.,
2010. Physical speciation of iron in the Atlantic sector of the Southern Ocean, along
a transect from the subtropical domain to the Weddell Sea Gyre. Journal of
Geophysical Research-Oceans 115.
Croot, P.L., Frew, R.D., Sander, S., Hunter, K.A., Ellwood, M.J., Pickmere, S.E., Abraham,
E.R., Law, C.S., Smith, M.J., Boyd, P.W., 2007. Physical mixing effects on iron
biogeochemical cycling: FeCycle experiment. Journal of Geophysical Research—
Oceans 112 (C6).
Cunningham, S., Pavic, M., 2007. Surface geostrophic currents across the Antarctic
circumpolar current in Drake Passage from 1992 to 2004. Progress in Oceanography 73 (3–4), 296–310.
de Baar, H.J.W., Dejong, J.T.M., Bakker, D.C.E., Loscher, B.M., Veth, C., Bathmann, U.,
Smetacek, V., 1995. Importance of iron for plankton blooms and carbon-dioxide
drawdown in the Southern-Ocean. Nature 373 (6513), 412–415.
Donat, J.R., Bruland, K.W., 1988. Direct determination of dissolved cobalt and nickel in
seawater by differential pulse cathodic stripping voltammetry preceded by
adsorptive collection of cyclohexane-1,2-dione dioxime complexes. Analytical
Chemistry 60 (3), 240–244.
Droop, M.R., 1957. Vitamin-B12 in marine ecology. Nature 180 (4594), 1041–1042.
Elderfield, H., McCaffrey, R.J., Luedtke, N., Bender, M., Truesdale, V.W., 1981. Chemical
diagenesis in Narragansett Bay sediments. American Journal of Science 281 (8),
1021–1055.
Ellwood, M.J., 2008. Wintertime trace metal (Zn, Cu, Ni, Cd, Pb and Co) and nutrient
distributions in the subantarctic zone between 40–52°S; 155–160°E. Marine
Chemistry 112 (1–2), 107–117.
Ellwood, M.J., van den Berg, C.M.G., Boye, M., Veldhuis, M., de Jong, J.T.M., de Baar,
H.J.W., Croot, P.L., Kattner, G., 2005. Organic complexation of cobalt across the
Antarctic Polar Front in the Southern Ocean. Marine and Freshwater Research
56 (8), 1069–1075.
Evans, J.R., Anderson, J.M., 1987. Absolute absorption and relative fluorescence
excitation-spectra of the 5 major chlorophyll-protein complexes from spinach
thylakoid membranes. Biochimica et Biophysica Acta 892 (1), 75–82.
Field, C.B., Behrenfeld, M.J., Randerson, J.T., Falkowski, P., 1998. Primary production of
the biosphere: integrating terrestrial and oceanic components. Science 281 (5374),
237–240.
Fitzwater, S.E., Johnson, K.S., Gordon, R.M., Coale, K.H., Smith, W.O., 2000. Trace metal
concentrations in the Ross Sea and their relationship with nutrients and
phytoplankton growth. Deep-Sea Research Part II — Topical Studies in Oceanography 47 (15–16), 3159–3179.
German, C.R., Campbell, A.C., Edmond, J.M., 1991. Hydrothermal scavenging at the MidAtlantic Ridge — modification of trace-element dissolved fluxes. Earth and
Planetary Science Letters 107 (1), 101–114.
Gladyshev, S., Arhan, M., Sokov, A., Speich, S., 2008. A hydrographic section from South Africa
to the southern limit of the Antarctic Circumpolar Current at the Greenwich meridian.
Deep-Sea Research Part I—Oceanographic Research Papers 55 (10), 1284–1303.
Gordon, A.L., Weiss, R.F., Smethie, W.M., Warner, M.J., 1992. Thermocline and
intermediate water communication between the South-Atlantic and Indian Oceans.
Journal of Geophysical Research—Oceans 97 (C5), 7223–7240.
Guéneuguès, A., Boye, M., 2008. Chlorophyll-a and phaeopigments along BONUSGoodHope section. MD166 BONUS-GoodHope cruise report in Biogeochemistry
and Geochemistry (135 pp), pp. 108–110.
Guillard, R.R.L., Cassie, V., 1963. Minimum cyanocobalamin requirements of some
marine centric diatoms. Limnology and Oceanography 8 (2), 161–165.
Ho, T.Y., Quigg, A., Finkel, Z.V., Milligan, A.J., Wyman, K., Falkowski, P.G., Morel, F.M.M.,
2003. The elemental composition of some marine phytoplankton. Journal of
Phycology 39, 1145–1159.
Huang, S., Conte, M.H., 2009. Source/process apportionment of major and trace
elements in sinking particles in the Sargasso sea. Geochimica and Cosmochimica
Acta 73 (1), 65–90.
13
Jakuba, R.W., Moffett, J.W., Dyhrman, S.T., 2008. Evidence for the linked biogeochemical
cycling of zinc, cobalt, and phosphorus in the western North Atlantic Ocean. Global
Biogeochemical Cycles 22 (4).
Jeandel, C., Lacan, F., Labatut, M., 2010. Trace element concentrations of the suspended
particles in the Southern Ocean (Bonus/GoodHope transect). AGU OS10 Scientific
Program., Ocean Sciences Conference (AGU/ASLO), Portland, USA.
Johnson, K.S., Coale, K.H., Berelson, W.M., Gordon, R.M., 1996. On the formation of the
manganese maximum in the oxygen minimum. Geochimica and Cosmochimica
Acta 60 (8), 1291–1299.
Johnson, K.S., Boyle, E., Bruland, K., Coale, K., Measures, C., Moffett, J., Aguilar-Islas, A., B.,
K., Bergquist, B., Bowie, A., Buck, K., Cai, Y., Chase, Z., Cullen, J., Doi, T., Elrod, V.,
Fitzwater, S., Gordon, M., King, A., Laan, P., Laglera-Baquer, L., Landing, W., Lohan,
M., Mendez, J., Milne, A., Obata, H., Ossiander, L., Plant, J., Sarthou, G., Sedwick, P.,
Smith, G.J., Sohst, B., Tanner, S., Van den Berg, S., Wu, J., 2007. The SAFe Iron
intercomparison cruise: an international collaboration to develop dissolved iron in
seawater standards. Eos, Transactions of the American Geophysical Union 88 (11),
131–132.
Kawabe, M., 2008. Vertical and horizontal eddy diffusivities and oxygen dissipation rate
in the subtropical Northwest Pacific. Deep-Sea Research Part I—Oceanographic
Research Papers 55 (3), 247–260.
Kermabon, C., Arhan, M., 2008. SBE9+: validation et réduction des données. Rapport
Interne OPS/LPO 08–04, IFREMER. .
Knauer, G.A., Martin, J.H., Gordon, R.M., 1982. Cobalt in Northeast Pacific Waters. Nature
297 (5861), 49–51.
Kremling, K., Streu, P., 2001. The behaviour of dissolved Cd, Co, Zn, and Pb in North
Atlantic near-surface waters (30°N/60°W–60°N/2°W). Deep-Sea Research Part I—
Oceanographic Research Papers 48 (12), 2541–2567.
Le Moigne, F.A.C., 2008. Distribution des sels nutritifs dans le secteur Atlantique de
l'Océan Austral en fin de période estivale, campagne BONUS-GoodHope. Rapport de
stage Master-2. Université de Bretagne Occidentale, Brest, France. 33 pp.
Le Moigne, F.A.C., Boye, M., Masson, A., Corvaisier, R., Grossteffan, E., Guéneuguès, A.,
Pondaven, P. and Nelson, D.M., in preparation. Biogeochemical features of the
subtropical southeastern Atlantic and the Southern Ocean south off Africa during
the Austral summer of the International Polar Year. In prep. for submission to
Biogeosciences Special Issue.
Li, F., Ginoux, P., Ramaswamy, V., 2008. Distribution, transport, and deposition of
mineral dust in the Southern Ocean and Antarctica: contribution of major sources.
Journal of Geophysical Research—Atmospheres 113 (D10).
Lutjeharms, J.R.E., Vanballegooyen, R.C., 1988. The retroflection of the Agulhas current.
Journal of Physical Oceanography 18 (11), 1570–1583.
Martin, J.H., Gordon, R.M., Fitzwater, S., Broenkow, W.W., 1989. Vertex — phytoplankton iron studies in the Gulf of Alaska. Deep-Sea Research Part a—Oceanographic
Research Papers 36 (5), 649.
Martin, J.H., Gordon, R.M., Fitzwater, S.E., 1990. Iron in Antarctic waters. Nature 345
(6271), 156–158.
Martin, J.H., Fitzwater, S.E., Gordon, R.M., Hunter, C.N., Tanner, S.J., 1993. Iron, primary
production and carbon nitrogen flux studies during the JGOFS North-Atlantic
Bloom Experiment. Deep-Sea Research Part I — Topical Studies in Oceanography 40
(1–2), 115–134.
Milne, A., Landing, W., Bizimis, M., Morton, P., 2010. Determination of Mn, Fe, Co, Ni, Cu,
Zn, Cd and Pb in seawater using high resolution magnetic sector inductively
coupled mass spectrometry (HR-ICP-MS). Analytica Chimica Acta 665 (2),
200–207.
Minas, H.J., Minas, M., Packard, T.T., 1986. Productivity in upwelling areas deduced from
hydrographic and chemical fields. Limnology and Oceanography 31 (6),
1182–1206.
Moffett, J.W., Ho, J., 1996. Oxidation of cobalt and manganese in seawater via a common
microbially catalyzed pathway. Geochimica and Cosmochimica Acta 60 (18),
3415–3424.
Morel, F.M.M., Reinfelder, J.R., Roberts, S.B., Chamberlain, C.P., Lee, J.G., Yee, D., 1994. Zinc and
carbon co-limitation of marine-phytoplankton. Nature 369 (6483), 740–742.
Murphy, J., Riley, J.P., 1962. A modified single solution method for determination of
phosphate in natural waters. Analytica Chimica Acta 27 (1), 31–36.
Neveux, J., 1976. Fluorimetric determination of chlorophyll-a and phaeophytin-a.
Annales De l'Institut Océanographique 52 (2), 165–174.
Noble, A.E., Saito, M.A., Maiti, K., Benitez-Nelson, C.R., 2008. Cobalt, manganese, and
iron near the Hawaiian Islands: a potential concentrating mechanism for cobalt
within a cyclonic eddy and implications for the hybrid-type trace metals. Deep-Sea
Research Part II—Topical Studies in Oceanography 55 (10–13), 1473–1490.
Orsi, A.H., Whitworth, T., Nowlin, W.D., 1995. On the meridional extent and fronts of the
Antarctic circumpolar current. Deep-Sea Research Part I—Oceanographic Research
Papers 42 (5), 641–673.
Parsons, T.R., Strickland, J.D.H., 1963. Discussion of spectrophotometric determination
of marine-plant pigments, with revised equations for ascertaining chlorophylls and
carotenoids. Journal of Marine Research 21 (3), 155–163.
Partensky, F., Hess, W.R., Vaulot, D., 1999. Prochlorococcus, a marine photosynthetic
prokaryote of global significance. Microbiology and Molecular Biology Reviews 63
(1), 106.
Piola, A.R., Gordon, A.L., 1989. Intermediate Waters in the Southwest South-Atlantic.
Deep-Sea Research Part a—Oceanographic Research Papers 36 (1), 1–16.
Pohl, C., Croot, P.L., Hennings, U., Daberkow, T., Budeus, G., von der Loeff, M.R., 2011.
Synoptic transects on the distribution of trace elements (Hg, Pb, Cd, Cu, Ni, Zn, Co,
Mn, Fe, and Al) in surface waters of the Northern- and Southern East Atlantic.
Journal of Marine Systems 84 (1–2), 28–41.
Price, N.M., Morel, F.M.M., 1990. Cadmium and cobalt substitution for zinc in a marine
diatom. Nature 344 (6267), 658–660.
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008
14
J. Bown et al. / Marine Chemistry xxx (2011) xxx–xxx
Reid, J.L., 1989. On the total geostrophic circulation of the South-Atlantic Ocean — flow
patterns, tracers, and transports. Progress in Oceanography 23 (3), 149–244.
Saito, M.A., Goepfert, T.J., Noble, A.E., Bertrand, E.M., Sedwick, P.N., DiTullio, G.R., 2010. A
seasonal study of dissolved cobalt in the Ross Sea, Antarctica: micronutrient
behavior, absence of scavenging, and relationships with Zn, Cd, and P. Biogeosciences 7 (12), 4059–4082.
Saito, M.A., Goepfert, T.J., 2008. Zinc–cobalt colimitation of Phaeocystis antarctica.
Limnology and Oceanography 53 (1), 266–275.
Saito, M.A., Moffett, J.W., 2001. Complexation of cobalt by natural organic ligands in the
Sargasso Sea as determined by a new high-sensitivity electrochemical cobalt
speciation method suitable for open ocean work. Marine Chemistry 75 (1–2),
49–68.
Saito, M.A., Moffett, J.W., 2002. Temporal and spatial variability of cobalt in the Atlantic
Ocean. Geochimica & Cosmochimica Acta 66 (11), 1943–1953.
Saito, M.A., Moffett, J.W., DiTullio, G.R., 2004. Cobalt and nickel in the Peru upwelling
region: a major flux of labile cobalt utilized as a micronutrient. Global
Biogeochemical Cycles 18 (4), GB4030.
Sakamoto-Arnold, C.M., Johnson, K.S., 1987. Determination of picomolar levels of cobalt
in seawater by flow-injection analysis with chemiluminescence detection.
Analytical Chemistry 59 (14), 1789–1794.
Sanudo-Wilhelmy, S.A., Olsen, K.A., Scelfo, J.M., Foster, T.D., Flegal, A.R., 2002. Trace
metal distributions off the Antarctic Peninsula in the Weddell Sea. Marine
Chemistry 77 (2–3), 157–170.
Shelley, R.U., Zachhuber, B., Sedwick, P.N., Worsfold, P.J., Lohan, M.C., 2010.
Determination of total dissolved cobalt in UV-irradiated seawater using flow
injection with chemiluminescence detection. Limnology and Oceanography:
Methods 8, 352–362.
Sunda, W.G., Huntsman, S.A., 1995. Cobalt and zinc interreplacement in marine
phytoplankton: biological and geochemical implications. Limnology and Oceanography 40 (8), 1404–1417.
Tagliabue, A., Bopp, L., Dutay, J.C., Bowie, A.R., Chever, F., Jean-Baptiste, P., Bucciarelli, E.,
Lannuzel, D., Remenyi, T., Sarthou, G., Aumont, O., Gehlen, M., Jeandel, C., 2010.
Hydrothermal contribution to the oceanic dissolved iron inventory. Nature
Geoscience 3 (4), 252–256.
Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental-crust.
Reviews of Geophysics 33 (2), 241–265.
Thuróczy, C.E., Boye, M., Losno, R., 2010. Dissolution of cobalt and zinc from natural and
anthropogenic dusts in seawater. Biogeosciences 7 (6), 1927–1936.
Tréguer, P., Le Corre, P., 1979. Ratios of nitrate, phosphate, and silicate during uptake
and regeneration phases of the moroccan upwelling regime. Deep-Sea Research
Part I — Oceanographic Research Papers 26 (2), 163–184.
Vega, M., van den Berg, C.M.G., 1997. Determination of cobalt in seawater by catalytic
adsorptive cathodic stripping voltammetry. Analytical Chemistry 69 (5), 874–881.
Westerlund, S., Ohman, P., 1991. Cadmium, copper, cobalt, nickel, lead, and zinc in the
water column of the Weddell Sea, Antarctica. Geochimica & Cosmochimica Acta 55
(8), 2127–2146.
Whitworth, T., Nowlin, W.D., 1987. Water masses and currents of the Southern-Ocean
at the Greenwich meridian. Journal of Geophysical Research—Oceans 92 (C6),
6462–6476.
Wong, G.T.F., Pai, S.C., Chung, S.W., 1995. Cobalt in the western Philippine Sea.
Oceanologica Acta 18 (6), 631–638.
Wunsch, C., Stammer, D., 1995. The global frequency-wavenumber spectrum of oceanic
variability estimated from TOPEX/POSEIDON altimetric measurements. Journal of
Geophysical Research—Oceans 100 (C12), 24895–24910.
Yentsch, C.S., Menzel, D.W., 1963. A method for the determination of phytoplankton
chlorophyll and phaeophytin by fluorescence. Deep-Sea Research 10 (3), 221–231.
Zhang, Y., Jiao, N.Z., Hong, N., 2008a. Comparative study of picoplankton biomass and
community structure in different provinces from subarctic to subtropical oceans.
Deep-Sea Research Part II — Topical Studies in Oceanography 55 (14–15),
1605–1614.
Zhang, Y., Lacan, F., Jeandel, C., 2008b. Dissolved rare earth elements tracing lithogenic
inputs over the Kerguelen Plateau (Southern Ocean). Deep-Sea Research Part II —
Topical Studies in Oceanography 55 (5–7), 638–652.
Please cite this article as: Bown, J., et al., The biogeochemical cycle of dissolved cobalt in the Atlantic and the Southern Ocean south off the
coast of South Africa, Mar. Chem. (2011), doi:10.1016/j.marchem.2011.03.008