response in the production of the convectively
formed mode waters (Greenland Sea Deep
Water, Labrador Sea Water and Eighteen Degree
Water) appears to have been particularly important. Driven by wind and buoyancy forcing
associated with the changing NAO, convective
activity at all three sites evolved over decades
to a long-term extreme state – in phase but of
differing sign – during the NAO minimum
of the 1960s, at which time the ventilation of
the Greenland Sea and Sargasso was at a maximum and that of the Labrador Sea was tightly
capped. Since then all three centres evolved
rather more rapidly towards their opposite
extreme states by the 1990s, during which convection in the Greenland Sea and Sargasso was
suppressed but vertical exchange in the Labrador
Sea reached deeper than previously observed.
4 In the West Atlantic, these coordinated changes
of opposite sign in the production and characteristics of the mode waters had important
repercussions on the strength of the Atlantic
gyre circulation. The strength of the main North
Atlantic Current is determined by the gradient
of potential energy anomaly (PEЈ) across the
mutual boundary between the Atlantic subtropical and subpolar gyres, which in turn reflects
the density and heat content of the mode waters
in each gyre. From the observed PEЈ differences
between the centres of the two gyres, Curry and
McCartney (2001) calculate that the long-term
increase in the NAO index to the mid-1990s
was accompanied by a 30% increase in the
0–2000 dbar east-going baroclinic mass transport along the gyre–gyre boundary, from 50 Sv
in 1970 to 65 Sv in the mid-1990s. Both subpolar and subtropical gyres contributed equally
to this change. If this link with the NAO proves
valid, then the North Atlantic gyre circulation
during WOCE is likely to have been at its
strongest for more than a century.
5 The Nordic Seas are important as a source of
change for both the climatically sensitive Arctic
Ocean and for the northern overflows that form
the headwaters of the meridional overturning
circulation (MOC). Over the period (post 1982)
for which both the temperature and salinity of
the deep waters are known with adequate precision, we have observed a progressive, almost linear, increase in the temperature and salinity of
the deep Greenland Sea to unprecedented values
by the mid-1990s, by which time GSDW properties lay quite outside their defined range. This
change is attributed to an altered balance between
vertical and horizontal exchange processes, with
a progressive shutdown in deep winter convection in the Greenland Sea (part of the same
coordinated pan-Atlantic pattern of convective
change as described earlier) and a corresponding increase in the horizontal exchange of warm
saline deep waters through Fram Strait. Alternative (perhaps additional) explanations include
the collapse of the ‘domed’ density structure in
the Greenland Sea and the enhanced importance
of diapycnal mixing over the rough topography
of Mohn’s Ridge as a mechanism for deep water
renewal.
6 Variability in the Arctic has long been thought
to play some special role in the processes of
global change. Three main changes in particular
have characterized the WOCE Period of the
1990s. The first was a more intense and more
widespread influence of Atlantic water than previously observed, with a warming (by up to
2°C) and spreading (by ϳ 20%) of the Atlanticderived sublayer across the Eurasian Basin
(Morison et al., 1998a,b, 2001); as the winter
NAO index amplified, the increasingly anomalous southerly airflow that accompanies such a
change over Nordic seas is held responsible for
a progressive warming and perhaps a strengthening of the two streams of Atlantic water that
enter the Arctic Ocean. Second, the cold halocline layer, which acts to insulate the sea ice
from the warm Atlantic layer below, was shown
to have weakened in the Eurasian Basin (Steele
and Boyd, 1998), with profound effects on the
surface energy and mass balance of sea ice in that
region; hydrography, tracers and modelling all
suggest an eastward diversion of Russian river
input in response to the altered atmospheric circulation. Third, we have new observational evidence that the amplifying NAO was accompanied
by an increased annual efflux of ice through the
western Fram Strait (Vinje et al., 1998; Kwok
and Rothrock, 1999); relatively minor changes in
freshwater distribution are assigned a high
importance in current climate models.
7 During the past few decades, the climatic variability over the North and Equatorial Pacific
has hardly been less extreme than that of the
North Atlantic. In mid-latitudes, the NP index
7.3 The World during WOCE
581
Dickson et al.
formed mode waters (Greenland Sea Deep
Water, Labrador Sea Water and Eighteen Degree
Water) appears to have been particularly important. Driven by wind and buoyancy forcing
associated with the changing NAO, convective
activity at all three sites evolved over decades
to a long-term extreme state – in phase but of
differing sign – during the NAO minimum
of the 1960s, at which time the ventilation of
the Greenland Sea and Sargasso was at a maximum and that of the Labrador Sea was tightly
capped. Since then all three centres evolved
rather more rapidly towards their opposite
extreme states by the 1990s, during which convection in the Greenland Sea and Sargasso was
suppressed but vertical exchange in the Labrador
Sea reached deeper than previously observed.
4 In the West Atlantic, these coordinated changes
of opposite sign in the production and characteristics of the mode waters had important
repercussions on the strength of the Atlantic
gyre circulation. The strength of the main North
Atlantic Current is determined by the gradient
of potential energy anomaly (PEЈ) across the
mutual boundary between the Atlantic subtropical and subpolar gyres, which in turn reflects
the density and heat content of the mode waters
in each gyre. From the observed PEЈ differences
between the centres of the two gyres, Curry and
McCartney (2001) calculate that the long-term
increase in the NAO index to the mid-1990s
was accompanied by a 30% increase in the
0–2000 dbar east-going baroclinic mass transport along the gyre–gyre boundary, from 50 Sv
in 1970 to 65 Sv in the mid-1990s. Both subpolar and subtropical gyres contributed equally
to this change. If this link with the NAO proves
valid, then the North Atlantic gyre circulation
during WOCE is likely to have been at its
strongest for more than a century.
5 The Nordic Seas are important as a source of
change for both the climatically sensitive Arctic
Ocean and for the northern overflows that form
the headwaters of the meridional overturning
circulation (MOC). Over the period (post 1982)
for which both the temperature and salinity of
the deep waters are known with adequate precision, we have observed a progressive, almost linear, increase in the temperature and salinity of
the deep Greenland Sea to unprecedented values
by the mid-1990s, by which time GSDW properties lay quite outside their defined range. This
change is attributed to an altered balance between
vertical and horizontal exchange processes, with
a progressive shutdown in deep winter convection in the Greenland Sea (part of the same
coordinated pan-Atlantic pattern of convective
change as described earlier) and a corresponding increase in the horizontal exchange of warm
saline deep waters through Fram Strait. Alternative (perhaps additional) explanations include
the collapse of the ‘domed’ density structure in
the Greenland Sea and the enhanced importance
of diapycnal mixing over the rough topography
of Mohn’s Ridge as a mechanism for deep water
renewal.
6 Variability in the Arctic has long been thought
to play some special role in the processes of
global change. Three main changes in particular
have characterized the WOCE Period of the
1990s. The first was a more intense and more
widespread influence of Atlantic water than previously observed, with a warming (by up to
2°C) and spreading (by ϳ 20%) of the Atlanticderived sublayer across the Eurasian Basin
(Morison et al., 1998a,b, 2001); as the winter
NAO index amplified, the increasingly anomalous southerly airflow that accompanies such a
change over Nordic seas is held responsible for
a progressive warming and perhaps a strengthening of the two streams of Atlantic water that
enter the Arctic Ocean. Second, the cold halocline layer, which acts to insulate the sea ice
from the warm Atlantic layer below, was shown
to have weakened in the Eurasian Basin (Steele
and Boyd, 1998), with profound effects on the
surface energy and mass balance of sea ice in that
region; hydrography, tracers and modelling all
suggest an eastward diversion of Russian river
input in response to the altered atmospheric circulation. Third, we have new observational evidence that the amplifying NAO was accompanied
by an increased annual efflux of ice through the
western Fram Strait (Vinje et al., 1998; Kwok
and Rothrock, 1999); relatively minor changes in
freshwater distribution are assigned a high
importance in current climate models.
7 During the past few decades, the climatic variability over the North and Equatorial Pacific
has hardly been less extreme than that of the
North Atlantic. In mid-latitudes, the NP index
7.3 The World during WOCE
581
Dickson et al.