REFERENCES:

1
Climate variation in the European Arctic during
the last 100 yearsI. Hanssen-Bauer E. J.
Førland, met.no
INTRODUCTION Nordic cooperation has during the la
ter years resulted in several freely available
datasets including monthly climate data from the
Atlantic/European sector of the Arctic. In the
present poster, selected data from the NARP
dataset (http//projects.met.no/narp/data_index.
html) and the NORDKLIM dataset
(http//www.smhi.se/hfa_coord/nordklim/nkds.htm)
are presented and related to April sea-ice cover
in the Nordic Seas and to the North Atlantic
Oscillation (NAO) and Arctic Oscillation (AO)
winter indices. Figure 1 shows the climate
stations applied in the present study.
Figure 1. Stations applied in the present ana
lysis.
TEMPERATURE There are large spatial differences i
n annual mean temperatures from southern to
northern parts of this area (Figure 2, left
panel), but the standardised temperature series
show similarities (Figure 2, right panel). The
first two decades of the century were cold at all
stations with measurements in this period. Still,
Mann-Kendalls non-parametric test revealed that
only a few series (from Iceland and northern
Norway) show a statistically significant warming
(5 level or better) from 1900 to 2002. Førland
et al. (2002) actually concluded that none of the
stations in Figure 1 showed a statistically
significant warming from 1910-1999. The reason is
that the cold period in the beginning of the
century was followed by a rapid warming and some
warm decades before a period of cooling and
finally a new period of warming at the end of the
century. At many stations in the area, the
thermal optimum of the 20th century was
experienced around the 1930s, and the period 1930
to 1960 was generally warmer than the 1961-1990
average. In the north-eastern part of the area
the following cooling was rapid. The 1960s were
very cold, and mainly followed by a warming trend
during the rest of the century. At the
southernmost stations in Iceland and the Faroe
Islands, there was a more gradual cooling to a
minimum around 1980. At the westernmost stations
in Greenland, the cooling continued to around
1990.
PRECIPITATION The measured annual precipitation
in the area spans from about 200mm at the
northernmost stations to about 1500 mm at the
southernmost stations (Figure 3, left panel). The
decadal scale variability is less spatially
consistent for precipitation than for temperature
(Figure 3, right panel). However, Førland et al.
(2002) concluded that all reliable series in the
area show a non-negative precipitation trend
during the 20th century. According to
Mann-Kendall non-parametric test, the positive
trends tend to be statistically significant (5
level or better) in a belt from north to south in
central parts of the area, while they tend not to
be significant in eastern parts of northern
Fennoscandia and at the westernmost stations.
Positive trends of more than 2 per decade were
found at Svalbard Airport, Bjørnøya and Tromsø,
while trends from 1.5 to 2 per decade were found
at Jan Mayen, Akureyri, Vestmannaeyar and
Torshavn. It should however be stressed that
reliable measurements of precipitation are
difficult to obtain under Arctic weather
conditions, and that quality control and
homogenization are hampered by the sparse station
network.
Figure 3. Lowpass filtered annual precipitation
at stations in the Atlantic/ European Arctic
given in mm (left) and in of the 1961-1990
average (above).
Figure 2. Lowpass filtered annual mean
temperature at stations in the Atlantic/ European
Arctic given in oC (left) and standardised
(above).
Atmospheric circulation and air temperature
It is well known that temperature variation in
the Atlantic/European sector of the Arctic partly
is accounted for by the North Atlantic
Oscillation (NAO, Hurrell 1995) or the Arctic
Oscillation (AO, Thomson Wallace 1998).
Figure 4 illustrates the temperature see-saw
In the western part of the area, annual mean
temperatures and NAO/AO winter indices are
negatively correlated, while they are positively
correlated in the eastern part of the area, where
the positive NAO/AO phase is associated with an
increased advection of mild air from
west-southwest. Calculation of correlation
coefficients for running 30-year periods during
the 20th century gave at Nuuk values from-0.4 to
-0.8, and a Vardø values from 0.4 to 0.7, but
there were no obvious trends in the correlation
coefficients. Figure 5 shows that the AO to a
variable degree may account for interdecadal
variability AO may at least partly account for
the negative temperature trend in Nuuk and the
positive trend in Vardø from the 1960s to the
1990s. The antiphase between east and west in
this period is an AO fingerprint, and the AO
index has a positive trend in this period. Rigor
et al. (2000) concluded that more than half the
warming over the eastern Arctic Ocean and the
cooling over the Labrador Sea from the 1970s to
the 1990s was accounted for by the AO. From
around 1910 to the 1930s, on the other hand,
there was a warming both in eastern and western
parts of the Atlantic/European sector of the
Arctic, and there was no significant trend in the
AOI (Figure 5).
Atmospheric circulation and precipitation
The correlation coefficients between annual
precipitation and the AO winter index exceed 0.2
only for stations in Iceland, Faroe Islands and
Bjørnøya (Figure 6). The correlation coeffecient
between winter precipitation and the AOI is
usually somewhat higher, and at Bjørnøya it was
0.5. Local precipitation is, however, strongly
affected by atmospheric circulation on smaller
scales. Hanssen-Bauer and Førland (1998) showed
that, at least for localities where the
orographic influence on the precipitation
distribution is strong, it is possible to model
the local precipitation trends fairly well by
using local gradients in sea level pressure.
Figure 6. Correlation coefficients between the
NAO and AO winter indices and annual
precipitation at different stations from west to
east.
Sea Ice and air temperature Vinje (2001) publishe
d series of April Sea-ice extent in western
(30oW-10oE) and eastern (10oE-70oE) parts of the
Nordic Seas/Barents Sea region from 1864 to 1998.
These series were correlated to annual mean
temperatures from the present dataset (Figure 7).
In most cases, there is an anticorrelation
between the annual mean temperatures and the
April sea-ice extent, and generally the
anticorrelation is most pronounced for the
sea-ice extent in the sector the station is
located. The Greenland stations, however, which
are located west of both sectors, tend to show a
positive correlation with the ice extent in the
eastern sector. We can recognise the Arctic
Oscillation see-saw (e.g. Rigor et al. 2002)
High AOI is associated with high air temperatures
and reduced sea-ice in the Barents Sea region,
but low temperatures and extensive sea-ice in the
Labrador area. Figure 8 shows that the correlat
ion between sea-ice extent in the Nordic Seas and
temperatures in western Greenland not stationary
For 30year periods from the 1960s to the 1990s
there is a positive correlation (0.5) even with
the sea-ice in the western area. For 30-year
periods in the beginning of the century, however,
the correlation was negative (-0.4).
Figure 4. Correlation coefficients between the
NAO/AO winter indices and the annual mean
temperature at different stations from west to
east.
Figure 5. Lowpass filtered series of the winter
AOI and annual mean temperatures in Nuuk and
Vardø.
Take-away message There is more in the European
Arctic than the AO!
REFERENCES Førland E.J., I.Hanssen-Bauer, T. Jón
sson, C. Kern-Hansen, P.Ø. Nordli, O.E. Tveito,
and E. Vaarby Laursen, 2002 Twentieth-century
variations in temperature and precipitation in
the Nordic Arctic. Polar Record, 38 (206),
203-210 Hanssen-Bauer, I. and E.J. Førland 1998
Long-term trends in precipitation and
temperature in the Norwegian Arctic can they be
explained by changes in atmospheric circulation
patterns? Climate Research, 10, 143-153
Hurrel, J.W., 1995 Decadal trends in the North
Atlantic Oscillation Regional temperatures and
precipitation, Science, 269, 676-679.
Rigor, I.G, R.L. Colony S. Martin, 2000 J. of
Climate, 15, 2648 Rigor, I.G, J.M. Wallace R.L.
Colony, 2002 Variations in surface air
temperature observations in the Arctic, 1979-97.
J. of Climate, 13, 896-907 Thompson, D.W.J. og J.
M. Wallace, 1998 The Arctic Oscillation
signature in the wintertime geopotential height
and temperature fields. Geophys. Res. Letters 25,
9, 1297-1300 Vinje, T., 2001 Anomalies and Trend
s of Sea-Ice Extent and Atmospheric Circulation
in the Nordic Seas during the Period 1864-1998.
J. Climate, 14, 255-267
Figure 7. Correlation coefficients (R) between
April sea-ice extent (western/total/eastern) and
the annual mean temperature at different stations.
Figure 8. R for running 30-year periods, between
April sea-ice extent and the Nuuk and Vardø
annual mean temperature.