fluxes from the ocean to the atmosphere. The
results of these ocean influences are reduced seasonal variation in maritime areas compared with
continental regions, weaker meridional gradients
over ocean areas and strong longitudinal dependence of yearly average temperature, especially in
the northern North Atlantic (Fig. 1.1.2, see Plate
1.1.2, p. 44). Here the deviation from the zonal
mean temperature exceeds 10°C off Norway. A
major reason for these high temperatures (besides
transport of heat from the subtropical Atlantic
Ocean) is the high salinity in the northern latitude
Atlantic leading to intermittent deep convection
events either near to the sea-ice edge (Greenland
Sea) or in the open ocean regions such as the
Labrador Sea.
Long-term mean ocean–atmosphere interaction
is responsible for the mean differences between
marine and continental climates. Internal variability of the atmosphere–ocean–cryosphere climate
system on all time scales, supplemented by external forcing (volcanic eruptions and solar cycles),
produces the observed climate variability. On very
long time scales, the forcing by the changed orbit
of the earth around the sun (Milankovich cycles)
becomes dominant.
We have good statistical descriptions of climate
variability, at least for the last few decades, for
many places and large continental areas. However,
we lack even the basic understanding of the causes
for some of the multiyear to decadal time scale
ocean-focused phenomena like the Antarctic
Circumpolar Wave in the Southern Ocean and the
North Atlantic Oscillation. Only the ENSO and
the QuasiBiennial Oscillation (QBO) stand out as
being partly understood. Since ENSO is the cause
of a large part of interannual variability in at least
the tropics, the next breakthrough concerning the
understanding of climate variability may well be
related to mid-latitude variability. As well as
ocean–atmosphere interaction, it is likely also to
involve land ice–atmosphere, vegetation–atmosphere
and sea ice–atmosphere interaction since all these
interactions lead to interannual and decadal time
scale variability. For example, understanding the
variability of monsoons will need cooperation
between two WCRP projects – the Climate Variability and Predictability (CLIVAR) study and
the Global Energy and Water Cycle Experiment
(GEWEX) – in coordinated enhanced observing
periods.
1.1.4 Rapid changes related to the oceans
The tendency of the climate system to react
strongly and rapidly to minor changes is now well
established, both from direct observations and
palaeoclimate reconstructions. An example (not
related to the ocean) is the Antarctic ozone hole.
Less than a billionth of all air molecules, the chlorofluorocarbons (CFCs) and halons, have caused
through their chlorine- and bromine-containing
decay products, the complete disappearance of the
ozone in areas with polar stratospheric clouds
(12–20 km height) in the Antarctic during early
spring since the mid-1980s. (Ozone acts as a UV-B
filter and is the third most important greenhouse
gas after water vapour and CO 2 .) This has led
through mixing with mid-latitude air to the weakening of the strong latitudinal gradient of UV-B
radiation. Now the daily UV-B dose reaching the
surface during a sunny day in late spring and early
summer is sometimes as high in New Zealand
and on the Antarctic Peninsula as it is in Darwin,
tropical North Australia (Seckmeyer et al., 1995).
Palaeo-evidence on the instability of the global
ocean conveyer belt stimulated by increased freshwater input from melting ice sheets into the
Atlantic is now abundant (Keigwin et al., 1994).
Also model studies (e.g. Rahmstorf and Willebrand,
1995) have suggested that this may happen for
comparatively slight changes in the freshwater budget of the Atlantic, north of about 30°S. It is suggested that addition of 0.1 Sverdrup (1 Sverdrup:
10
6 m
3 s
91
) of fresh water could completely stop
deep convection in the northern North Atlantic
and remove the 4°C positive sea surface temperature anomaly (compared with the Eastern Pacific
at a latitude of about 50°N). In addition, coupled
ocean–atmosphere GCMs (General Circulation
Models) run under steadily increasing greenhouse
gas concentrations show a spin-down of the
strength of the meridional overturning in the
Atlantic (e.g. Manabe and Stouffer, 1993) and in
the Southern Ocean (Hirst, 1998). It is therefore
urgent that the monitoring of the flow over the
sills from the Nordic Seas into the North Atlantic
and of the oceanic overturning itself, as conducted
during WOCE, be continued in order to validate
the models’ ability to represent adequately realistic
large-scale ocean circulation and its variability.
Another potential rapid change related to the
ocean would be the disappearance of multiyear sea
SECTION 1 THE OCEAN AND CLIMATE
6
results of these ocean influences are reduced seasonal variation in maritime areas compared with
continental regions, weaker meridional gradients
over ocean areas and strong longitudinal dependence of yearly average temperature, especially in
the northern North Atlantic (Fig. 1.1.2, see Plate
1.1.2, p. 44). Here the deviation from the zonal
mean temperature exceeds 10°C off Norway. A
major reason for these high temperatures (besides
transport of heat from the subtropical Atlantic
Ocean) is the high salinity in the northern latitude
Atlantic leading to intermittent deep convection
events either near to the sea-ice edge (Greenland
Sea) or in the open ocean regions such as the
Labrador Sea.
Long-term mean ocean–atmosphere interaction
is responsible for the mean differences between
marine and continental climates. Internal variability of the atmosphere–ocean–cryosphere climate
system on all time scales, supplemented by external forcing (volcanic eruptions and solar cycles),
produces the observed climate variability. On very
long time scales, the forcing by the changed orbit
of the earth around the sun (Milankovich cycles)
becomes dominant.
We have good statistical descriptions of climate
variability, at least for the last few decades, for
many places and large continental areas. However,
we lack even the basic understanding of the causes
for some of the multiyear to decadal time scale
ocean-focused phenomena like the Antarctic
Circumpolar Wave in the Southern Ocean and the
North Atlantic Oscillation. Only the ENSO and
the QuasiBiennial Oscillation (QBO) stand out as
being partly understood. Since ENSO is the cause
of a large part of interannual variability in at least
the tropics, the next breakthrough concerning the
understanding of climate variability may well be
related to mid-latitude variability. As well as
ocean–atmosphere interaction, it is likely also to
involve land ice–atmosphere, vegetation–atmosphere
and sea ice–atmosphere interaction since all these
interactions lead to interannual and decadal time
scale variability. For example, understanding the
variability of monsoons will need cooperation
between two WCRP projects – the Climate Variability and Predictability (CLIVAR) study and
the Global Energy and Water Cycle Experiment
(GEWEX) – in coordinated enhanced observing
periods.
1.1.4 Rapid changes related to the oceans
The tendency of the climate system to react
strongly and rapidly to minor changes is now well
established, both from direct observations and
palaeoclimate reconstructions. An example (not
related to the ocean) is the Antarctic ozone hole.
Less than a billionth of all air molecules, the chlorofluorocarbons (CFCs) and halons, have caused
through their chlorine- and bromine-containing
decay products, the complete disappearance of the
ozone in areas with polar stratospheric clouds
(12–20 km height) in the Antarctic during early
spring since the mid-1980s. (Ozone acts as a UV-B
filter and is the third most important greenhouse
gas after water vapour and CO 2 .) This has led
through mixing with mid-latitude air to the weakening of the strong latitudinal gradient of UV-B
radiation. Now the daily UV-B dose reaching the
surface during a sunny day in late spring and early
summer is sometimes as high in New Zealand
and on the Antarctic Peninsula as it is in Darwin,
tropical North Australia (Seckmeyer et al., 1995).
Palaeo-evidence on the instability of the global
ocean conveyer belt stimulated by increased freshwater input from melting ice sheets into the
Atlantic is now abundant (Keigwin et al., 1994).
Also model studies (e.g. Rahmstorf and Willebrand,
1995) have suggested that this may happen for
comparatively slight changes in the freshwater budget of the Atlantic, north of about 30°S. It is suggested that addition of 0.1 Sverdrup (1 Sverdrup:
10
6 m
3 s
91
) of fresh water could completely stop
deep convection in the northern North Atlantic
and remove the 4°C positive sea surface temperature anomaly (compared with the Eastern Pacific
at a latitude of about 50°N). In addition, coupled
ocean–atmosphere GCMs (General Circulation
Models) run under steadily increasing greenhouse
gas concentrations show a spin-down of the
strength of the meridional overturning in the
Atlantic (e.g. Manabe and Stouffer, 1993) and in
the Southern Ocean (Hirst, 1998). It is therefore
urgent that the monitoring of the flow over the
sills from the Nordic Seas into the North Atlantic
and of the oceanic overturning itself, as conducted
during WOCE, be continued in order to validate
the models’ ability to represent adequately realistic
large-scale ocean circulation and its variability.
Another potential rapid change related to the
ocean would be the disappearance of multiyear sea
SECTION 1 THE OCEAN AND CLIMATE
6