Brage NP

Savinov, Vladimir M.

Gabrielsen, Geir Wing

Savinova, Tatiana N.

English

Trace elements in seabirds from the Barents and Norwegian Seas, 1991-1993

Anker-Nilssen, Tycho

The status of marine birds breeding in the Barents Sea region

Hansen, John Richard

Overrein, Øystein

Røye på Svalbard og Jan Mayen: en statusoversikt med vekt på forvaltningsrelaterte kunnskapsbehov

Bakken, Vidar

Strøm, Hallvard

Golovkin, Alexander N.

Bianki, Vitali V.

Tatarinkova, Iverta P.

The Status of marine birds breeding in the Barents Sea region

Norwegian

Årsmelding 1999

Pomponi, Massimo

Bertonati, Claudia

Fuglei, Eva

Wiig, Øystein

Derocher, Andrew E.

2,3-DPG-Hd complex : a hypothesis for an assymetric binding

Gjertz, Ian

Kovacs, Kit M.

Lydersen, Christian

Movements and diving of adult ringed seals (Phoca hispida) in Svalbard

O'Dwyer, Jane

Isaksson, Elisabeth

Vinje, Torgny

Jauhiainen, Tauno

Moore, John

Pohjola, Veijo

Vaikmäe, Rein

van de Wal, Roderik S.W.

GEOPHYSICAL RESEARCH LETTERS, VOL. 27, NO. 8, PAGES 1159-1162, APRIL 15, 2000
Methanesulfonic acid in a Svalbard ice core as an
indicator of ocean climate
Jane O’Dwyer, 1 Elisabeth Isaksson, 1 Torgny Vinje, 1 Tauno Jauhiainen, 2
John Moore, 2 Veijo Pohjola, 3 Rein Vaikmäe, 4 Roderik S. W. van de Wal. 5
Abstract. Methanesulfonic acid (MSA) is an atmospheric
oxidation product of dimethyl sulfide, produced by marine
biota. MSA preserved in a Svalbard glacier between 1920
and 1996 is compared with the sea surface temperature
(SST) and sea-ice extent of the surrounding ocean over the
same period. On decadal timescales high MSA concentra-
tions are found to be associated with warm SST and re-
duced sea-ice extent. MSA appears to be influenced by cli-
matic changes related to variations in the import of warm
Atlantic Water to the Barents Sea. Atlantic Water plays an
important role in the Arctic climate system, therefore MSA
concentrations may indirectly reflect larger-scale changes in
the region and may be useful as a proxy for past climate.
Introduction
Methanesulfonic acid (MSA) in the snow deposited on
a glacier forms a stratigraphic record of atmospheric con-
centrations. The only source of MSA appears to be the
atmospheric oxidation of dimethyl sulfide (DMS), which is
emitted by marine biota [Saltzman et al., 1983, 1986]. The
MSA record has therefore been used as an indicator of the
strength of the marine biogenic source of DMS [Ivey et al.,
1986; Saigne and Legrand, 1987]. MSA concentrations in the
ice will be affected by (i) conditions at the source location in-
fluencing primary productivity, phytoplankton species, and
air-sea exchange of DMS, (ii) atmospheric conditions influ-
encing the oxidation pathways of DMS or altering the source
location, (iii) conditions at the glacier. Thus variations in air
and sea temperatures, precipitation patterns, sea-ice condi-
tions, winds and ocean currents can affect MSA concentra-
tions, and all these parameters are intimately linked through
the complex interactions of the climate system. It has been
suggested that the sulfur cycle itself actively contributes to
climate regulation through the role of the sulfur aerosol in
cloud formation [Charlson et al., 1987]. Although sulfate is
the major component of the cycle, MSA makes a significant
contribution in the marine atmosphere [Bates et al., 1992].
Here we examine the MSA record for the period from
1920 to 1996 in an ice core taken from a glacier on Sval-
bard (Fig. 1). During this period observations of sea-ice
extent and ocean temperature are available for the region
1Norwegian Polar Institute, Tromsø, Norway
2Arctic Center, University of Lappland, Finland
3Department of Earth Sciences, Uppsala University, Sweden
4Institute of Geology, Tallinn Technical University, Estonia
5Institute for Marine and Atmospheric Research, Utrecht Uni-
versity, The Netherlands
Copyright 2000 by the American Geophysical Union.
Paper number 1999GL011106.
0094-8276/00/1999GL011106$05.00
around Svalbard, and these are compared with MSA concen-
trations. By contrast with the century timescale of observa-
tional timeseries, Arctic glaciers contain a record extending
back several thousand years. If changes in MSA concentra-
tion are correlated with changes in climate then ice cores can
give information about climate on these long timescales.
Climatic conditions around Svalbard
The climate around Svalbard is mild for its latitude due
to the presence of warm Atlantic Water flowing northward
in a continuation of the Gulf Stream (Fig. 1). The current
splits south of Svalbard and one branch flows to the west
of Svalbard, through the Fram Strait and into the Arctic
Ocean. A second branch flows eastward into the Barents
Sea where it forms a tongue of warm, saline surface water
[Matishov et al., 1998].
Sea-ice is another important component of the Arctic en-
vironment. Ice cover strongly affects exchanges between the
atmosphere and the ocean, and thus climate. Ice also affects
marine biota [Clarke and Ackley, 1984]. Thick ice prevents
the growth of phytoplankton, but in spring and summer the
melting ice creates a highly productive environment in which
a fresh surface layer inhibits vertical mixing, holding phyto-
plankton in the euphotic zone. Combined with long daylight
hours this leads to high rates of photosynthesis. For exam-
ple, in the Arctic Ocean the highest concentrations of DMS
are found in open water at the ice edge, and the lowest
concentrations beneath heavy pack ice [Leck and Persson,
1996a, b]. The sea-ice extent around Svalbard varies on sea-
sonal, annual and decadal timescales. As shown in Fig. 1, at
the end of a severe winter sea-ice can extend several degrees
south of Svalbard, whilst in other winters the ocean around
Svalbard is almost ice-free [Vinje, 2000].
Mean sea level pressure fields from the NCEP reanalysis
(accessible via FTP from ncardata.ucar.edu/pub/reanalysis/
monthly) imply easterly geostrophic winds over the region
around Svalbard, stronger in winter and weaker in summer.
This suggests that the Barents Sea to the east of Svalbard
will be an important source of the MSA in snow deposited
on Svalbard.
Results
MSA
In 1997 a 124 m long ice core was retrieved from Lomo-
nosovfonna, the highest ice field on Svalbard at 1230 masl
(Fig. 1). Radar measurements indicated a total ice depth of
126.5 m, and temperature measurements from the borehole
indicated that the ice was below the melting point. The
top 36 m of the core were sampled at high resolution [Isaks-
son et al., 1998] and dated using two independent methods
[E. Isaksson et al., manuscript in preparation, 2000]. The
1159
1160 O’DWYER ET AL.: MSA IN A SVALBARD GLACIER AND OCEAN CLIMATE
−20 0 20 40 60
70
75
80
Longitude (oE)
La
tit
ud
e 
(o
N
)
ARCTIC OCEAN
Svalbard
G
re
en
la
n
d
NORDIC
SEAS
BARENTS
SEA
Figure 1. Svalbard and the surrounding region. The ice core
drilling site at Lomonosovfonna on Svalbard is marked *. Arrows
show the northward flow of warm Atlantic water into the Arctic
Ocean and the Barents Sea. Broken lines mark the end-of-winter
ice edge positions in 1980 (dark grey) and 1995 (light grey).
first method used the seasonal variation of the different ion
records, and by counting the layers and using the 137Cs peak
from 1963 (at 18.50–18.95 m depth) as a reference layer, the
ice core was dated back to 1920. The second method as-
sumed a constant accumulation rate of 36 g cm−2 yr−1 with
thinning by pure shear [Nye, 1963]. The methods give ages
which agree to within three years at 36 m depth.
MSA concentration in 280 samples was determined by
suppressed ion chromatography, as described by Jauhiainen
et al. [1999]. The precision of the method and total un-
certainty at the 5 ng g−1 level were 7% and 21% respec-
tively and the accuracy was confirmed by an interlabora-
tory comparison test between three laboratories (p-value
0.227 at 95% confidence). The detection limit of MSA was
0.3 ng g−1. MSA was not found in blank samples prepared
by freezing pure water in PE measuring cylinders, which
were handled in the same way as the ice core samples.
Annual mean MSA concentrations have large interan-
nual and interdecadal variability (Fig. 2a), with a mean of
7.1 ng g−1 and a standard deviation of 7.5 ng g−1. For our
analysis of variability on decadal and longer timescales, the
annual values were smoothed with a five-year running mean.
There are some clear differences from MSA in Greenland
ice cores over the same period. The accumulation rate at
the 20D site in southern Greenland [Whung et al., 1994] is
similar to that at Lomonosovfonna, but MSA concentrations
in the 20D core are less than half those at Lomonosovfonna.
This is also true for the central Greenland GISP site [Legrand
et al., 1997] where there is a lower accumulation rate. Both
the Greenland sites have a recent decrease in MSA (start-
ing around 1900 in southern Greenland and 1945 in central
Greenland), whereas MSA in the Lomonosovfonna core ex-
hibits no significant trend over the twentieth century.
Sea surface temperature
Timeseries of sea surface temperature (SST) in selected
seasons were calculated for regions to the east and west of
Svalbard. The time of year of the timeseries depended upon
the available data. Hydrographic stations were selected that
had been repeated at least 20 times and SST was defined as
the average temperature over the upper 50 m. For each set
of stations annual averages were calculated and the resulting
timeseries were smoothed with a five-year running mean.
In the Barents Sea, in the area 72–76◦N, 30–40◦E, which
lies in the main path of Atlantic Water inflow, timeseries of
winter (January to March) and summer (July to Septem-
ber) SST were calculated using data from Matishov et al.
[1998] (Fig. 2b). Data were available in 32 and 39 years
0
20
40
C
on
c.
   
(n
g.
g−
1 )
MSA (a)      
1
2
3
4
5
6
T
em
pe
ra
tu
re
 (
 o
C
)
SST (b)
1920 1940 1960 1980
0
400
800
1200
Year
A
re
a 
(1
03
 k
m
2 )
Sea−ice area (c)       
Figure 2. (a) Annual mean (grey) and five-year running mean
(black) of MSA concentration. (b) Five-year running means of
SST in the Barents Sea in summer (+) and winter (×), and west
of Svalbard in autumn (o). (c) Five-year running means of sea-
ice extent in April and August to the east (full lines) and west
(broken lines) of Svalbard.
O’DWYER ET AL.: MSA IN A SVALBARD GLACIER AND OCEAN CLIMATE 1161
for winter and summer respectively. Before 1955 the data
is sparse, although there are indications of high SST at the
start of the 1920s and increasing SST from the mid-1940s
to the mid-1950s. The timeseries are nearly continuous af-
ter 1955 and SST variability corresponds closely to that of
MSA, with a decrease in SST from 1955 to 1970, a peak in
1974, and a gradual increase after 1980. High MSA concen-
trations are associated with warm conditions in the Barents
Sea with correlation coefficients (significant at the 99% con-
fidence level) of R = 0.55 for winter SST and R = 0.58 for
summer SST.
To the west of Svalbard data from 30 years between 1959
and 1992 from the Norwegian Polar Institute were used to
calculate autumn (September to November) SST (Fig. 2b).
Comparing with Fig. 2a, MSA appears to lag the SST and
the correlation coefficient, R = 0.23, is not significantly dif-
ferent from zero. Higher coefficients are obtained when the
SST timeseries is lagged, the maximum, R = 0.69 (signif-
icant at the 99% confidence level), occurring with a lag of
three years. This time-lag suggests that ocean conditions
to the west of Svalbard and MSA concentrations may be
indirectly linked by the larger scale ocean circulation.
Sea-ice
Vinje [2000] derived a 400 year record of sea-ice area in
the Svalbard region from ships’ logbooks. Here we use time-
series of ice extent in regions to the east and west of Sval-
bard in winter (April) and summer (August) of each year,
smoothed with a five-year running mean. Fig. 2c shows that
the variations of ice extent are similar in these two regions
and in different seasons, and a comparison with Fig. 2a indi-
cates an inverse relationship with MSA concentration. The
correlations of MSA concentration with the sea-ice extent
are all significant at the 99% confidence level, with coef-
ficients of R = −0.51 and R = −0.56 for winter sea-ice
extent to the east and west of Svalbard respectively, and of
R = −0.66 and R = −0.37 for summer sea-ice extent to the
east and west of Svalbard. High MSA concentrations are
thus associated with reduced ice extent, and the strongest
correlation is with summer ice extent in the Barents Sea.
Discussion
It is difficult to unravel the climatic parameters affecting
MSA concentrations in an ice core. Sea surface temper-
atures, air temperatures, wind fields and sea-ice extent are
not mutually independent and they can all influence the con-
centration of MSA in the ice on Svalbard, probably through
more than one route. Prevailing easterly winds suggest the
Barents Sea as the main source of MSA in snow deposited
on Svalbard and concentrations in the ice core do appear to
be more strongly linked with conditions in the Barents Sea
than with conditions to the west.
No consistent picture emerges from studies of the links
between sea-ice, SST, and MSA in ice cores. In the Antarc-
tic a positive correlation between MSA in a snow core from
Newall Glacier (a coastal site) and sea-ice in the Ross Sea
was found [Welch at al., 1993] and this was explained in
terms of increased growth of phytoplankton in heavy ice
years, giving high atmospheric MSA concentrations [Bunt,
1963], [Bunt and Wood, 1963]. A negative correlation be-
tween MSA in a core from a relatively coastal site on the
Antarctic Peninsula, and the sea-ice in Scotia Bay in the
South Orkney Island [Pasteur et al., 1995] was explained by
lower local DMS emission during heavy sea-ice years due to
the specific circulation conditions in the Weddell Sea. In the
Northern Hemisphere, on Greenland, Whung et al. [1994]
found a negative correlation between North Atlantic SST
and MSA in an ice core from the 20D site, and Legrand et
al. [1997] found a positive correlation between sea-ice and
MSA in an ice core from the GRIP site.
The variation of MSA in the Svalbard ice core is very
different from that in the Greenland ice cores, demonstrat-
ing that trends over the Arctic region are not uniform. In
contrast to the Greenland ice cores we find a negative cor-
relation with sea-ice extent and a positive correlation with
SST, i.e., high MSA concentrations are associated with years
when the surrounding ocean is warm and has reduced sea-ice
extent. SST and sea-ice extent are strongly linked, so it is
not clear from this correlation analysis which has the most
important influence on MSA. In the Arctic Ocean, however,
sea-ice and sea-ice meltwater are seen to strongly affect DMS
production [Leck and Persson, 1996a]. Possible links be-
tween reduced sea-ice extent and high MSA are the larger
area of open water available for biological production, and
the longer period of time for which ice-free water is present
in years with reduced sea-ice extent.
Atlantic Water in the Barents Sea is derived from the
same source as the Atlantic Water which flows to the west of
Svalbard and into the Arctic Ocean (Fig. 1), and both carry
the signal of the Atlantic Water variability. Thus the MSA
concentrations which correlate with SST in the Barents Sea
are also reflecting larger-scale variability in the ocean circu-
lation in the Nordic Seas and Arctic Ocean, an important
component of the region’s climate system. The signals to
the east and west of Svalbard will be out of phase if there is
a difference in the times taken by Atlantic Water to travel
to these two regions and this could explain the time-lag be-
tween the signal in SST west of Svalbard and the signal in
MSA.
The remainder of the core, estimated to cover the last
thousand years, is currently being analysed. Future work
will include an investigation of the complete MSA record
from the core in relation to the reconstructed 400 year record
of Barents Sea ice extent [Vinje, 2000].
Acknowledgments. We are grateful to all the members
of the Lomonosovfonna ice coring project for drilling the core and
providing chemical analysis. Financial support came from the
Norwegian Polar Institute, Netherlands Foundation for the Ad-
vancement of Pure Research, Finnish Academy, Nordic Council
of Ministers, and Estonian Science Foundation grant 2186. This
is Norwegian Polar Institute contribution 357.
References
Bates, T. S., J. A. Calhoun and P. K. Quinn 1992: Variations in
the methanesulfonate to sulfate molar ratio in submicrometer
aerosol particles over the South Pacific Ocean. J. Geophys.
Res., 97, 9859–9865.
Bunt, J. S. 1963: Diatoms of antarctic sea-ice as agents of primary
production. Nature, 199 1255–1257.
Bunt, J. S., and Wood, E. F. J. 1963: Microalgae and antarctic
sea-ice. Nature, 199, 1254–1255.
Charlson, R. J., J. E. Lovelock, M. O. Andreae and S. G. War-
ren 1987: Oceanic phytoplankton, atmospheric sulphur, cloud
albedo and climate. Nature, 326, 655–661.
Clarke, D. B., and S. F. Ackley 1984: Sea ice structure and bio-
logical activity in the antarctic marginal ice zone. J. Geophys.
Res., 89, 2087–2095.
1162 O’DWYER ET AL.: MSA IN A SVALBARD GLACIER AND OCEAN CLIMATE
Isaksson, E., R. van de Wal, M. Thomassen, V. Pohjola, J. Moore,
T. Jauhiainen, R. Vaikmäe, J. Ivask, T. Martma, J.-F. Pin-
glot, H. Meijer, and R. Mulvaney 1998: An ice core record
from Svalbard with seasonal signals preserved. Eos Vol. 79
(45), supplement; American Geophysical Union Fall Meeting,
December 1998, p. 278.
Ivey, J. P., D. M. Davies, V. Morgan, and G. P. Ayers 1986:
Methanesulphonate in Antarctic ice. Tellus, 38B, 375–379.
Jauhiainen, T., J. Moore, P. Perämäki, J. Derome and K. Derome
1999: Simple procedure for ion chromatographic determination
of anion and cations at trace levels in ice core samples. Anal.
Chim. Acta, 389, 21–29.
Leck, C. and C. Persson 1996a: The central Arctic Ocean as a
source of dimethyl sulfide. Seasonal variability in relation to
biological activity. Tellus, 48B, 156–177.
Leck, C. and C. Persson 1996b: Seasonal and short-term variabil-
ity in dimethyl sulfide, sulfur dioxide and biogenic sulfur and
sea salt aerosol particles in the arctic marine boundary layer
during summer and autumn. Tellus, 48B, 272–299.
Legrand, M., C. Hammer, M. De Angelis, J. Savarino, R. Delmas,
H. Clausen, and S. J. Johnson 1997: Sulfur-containing species
(methanesulfonate and SO4) over the last climatic cycle in the
Greenland Ice Core Project (central Greenland) ice core. J.
Geophys Res., C102, 26,663–26,679.
Matishov, G., A. Zyev, V. Golubev, N. Adrov, V. Slobodin, S.
Levitus and I. Smolyar 1998: Climatic Atlas of the Barents Sea
1998: Temperature, salinity, oxygen. NOAA Atlas NESDIS 26,
CDROM.
Nye, J.F. 1963: Correction factor for accumulation measured by
the ice thickness of the annual layers in an ice sheet. J. Glaciol.,
4, 141–150.
Pasteur, E., R. Mulvaney, D. A. Peel, E. S. Saltzman, and P.-
Y. Whung 1995: A 340-year record of biogenic sulphur from
Weddell Sea area, Antarctica. Ann. Glaciol., 21, 169–174.
Saigne, C., and M. Legrand 1987: Methanesulfonic acid in
Antarctic ice. Nature, 330, 240-242.
Saltzman, E. S., D. L. Savoie, R. G. Zika, and J. M. Prospero
1983: Methane sulfonic acid in the marine atmosphere. J. Geo-
phys. Res., 88, 10,897–10,902.
Saltzman, E. S., D. L. Savoie, J. M. Prospero, and R. G. Zika
1986: Methanesulfonic acid and non-sea salt sulfate in Pacific
air: regional and seasonal variations. J. Atm. Chem., 4, 227-
240.
Vinje, T. 2000: Anomalies and trends of sea ice extent and at-
mospheric circulation in the Nordic Seas during the period
1864–1998. To appear in J. Clim..
Welch, K. A., P. A. Mayewski, and S. I. Whitlow 1993: Methane-
sulfonic acid in coastal Antarctic snow related to sea-ice extent.
Geophys Res. Lett., 20 443–446.
Whung, P.-Y., E. S. Saltzman, M. J. Spencer, P. A. Mayewski,
and N. Gundestrup 1994: Two-hundred-year record of biogenic
sulfur in a south Greenland ice core (20D). J. Geophys. Res.,
99, 1147–1156.
J. O’Dwyer and E. Isaksson, Norwegian Polar Institute, N-9296
Tromsø, Norway. (email: jane.odwyer@npolar.no)
T. Vinje, Norwegian Polar Institute, PB 8100 Dep, N-0032
Oslo, Norway.
T. Jauhiainen and J. Moore, Arctic Center, University of Lap-
pland, 96101 Rovaniemi, Finland.
V. Pohjola, Department of Earth Sciences, Uppsala University,
S-752 36 Uppsala, Sweden.
R. Vaikmäe, Institute of Geology, Tallinn Technical University,
EE0001 Tallinn, Estonia.
R. S.W. van de Wal, Institute for Marine and Atmospheric
Research, Utrecht University, Princetonplein 5, 3584 CC, Utrecht
The Netherlands.
(Received October 06, 1999; revised February 07, 2000;
accepted February 23, 2000.)


Methanesulfonic acid in a Svalbard ice core as an indicator of ocean climate

Aanestad, M.

Berg, J.P.

Hormones and metabolites of arctic foxes (Alopex lagopus) in response to season, starvation and re-feeding

Not a member yet