with surface temperatures in the central equatorial
Pacific some 3°C above the 1961–90 base level.
An operational system based on the TAO (Tropical Ocean Atmosphere)–Triton moored buoy
array and upper ocean temperature measurements
from volunteer observing ships is now in place to
provide real-time monitoring of the state of the
surface and subsurface equatorial Pacific Ocean
and atmosphere (McPhaden et al., 1998). Models
developed over the past decade and using this data
have demonstrated useful skill in predicting the
evolution of the most recent (1997–98) event. For
a more detailed description of ENSO, see Philander
(1990) and the special issue of Journal of Geophysical Research (The TOGA Decade, June 1998).
A concern in interpreting the WOCE data set is
the extent to which the evolution of the tropical
Pacific during the 1990s imposed biases on the
largely extratropical observations. Most of the
one-time hydrographic sections in the Pacific were
occupied between 1990 and 1994 whereas in the
North Atlantic observations were spread throughout the 1990–98 period. The evaluation of such
biases as may exist has yet to be made.
1.2.6.2 Other tropical sea surface
temperature anomalies
Similar coherent responses of the equatorial
oceans to atmospheric forcing (and related feedbacks) have been documented in the Atlantic
(Mehta and Delworth, 1995) and Indian Oceans
(Saji et al., 1999; Webster et al., 1999). In the
Atlantic there are two competing hypotheses for
the causes of the variability of the meridional temperature gradient (the Atlantic dipole). One view is
that it is caused by regional ocean–atmosphere
positive feedbacks (Chang et al., 1997); the other
that the anomalies on either side of the equator are
due to independent processes in each hemisphere
(Enfield and Mayer, 1997). Just as with El Niño,
these equatorial anomalies have important impacts
on the climate of the adjacent land masses (Birkett
et al., 1999; Markam and McLain, 1977; Bertacchi
et al., 1998). Thus both the development of in-situ
observing systems and of predictive models are
presently receiving considerable attention.
1.2.6.3 Extratropical anomalies – decadal and
longer-term variability
Over both the North Pacific and North Atlantic
there are clearly defined patterns of atmospheric
variability that show persistence over multiyear time
scales. The most prominent of these are the Pacific
Decadal Oscillation (PDO) and the North Atlantic
Oscillation (NAO). Each can be regarded as a
regional manifestation of the wider Arctic Oscillation (AO). The status of knowledge of these matters
is summarized in the CLIVAR Implementation Plan
(World Climate Research Programme, 1998).
In the North Pacific, there has been a general
decrease in winter Sea Level Pressure (SLP) accompanied by increases in air and Sea Surface Temperature (SST) since the mid-1970s. When the
variability within the ENSO frequency band is
removed, the decadal variability within surface and
subsurface temperature fields is an anomaly pattern
that rotates clockwise around the North Pacific
Gyre (Zhang and Levitus, 1997). The oceanic
impacts arising from this variability has been discussed by Polovina et al. (1995).
For the NAO, the accepted index of its state
is the normalized December–March atmospheric
pressure difference between Lisbon and Reykjavik
and this can be analysed in the instrumental record
back to c. 1870. High values indicate a large pressure gradient and consequent enhanced westerly
airflow carrying heat and moisture lost from the
ocean across western Europe. During the period of
WOCE observations the NAO has remained in a
persistently high state. The oceanic impacts of the
NAO are discussed at length in Dickson et al.
(Chapter 7.3).
1.2.6.4 Changes in properties of deep and
intermediate waters
Small changes in air–sea heat fluxes, such as might
result from decadal climate variability and anthropogenic climate change, are difficult to observe.
However, because the ocean is an integrator of
short-term variability, these air–sea flux changes are
manifested as changes in ocean temperatures and
salinities. The WOCE data set serves as an important baseline against which past and future changes
can be assessed. Comparison of WOCE sections
with earlier International Geophysical Year (IGY)
sections has revealed widespread and significant
changes in water mass properties over decadal time
scales. These are comprehensively summarized by
Dickson et al. (Chapter 7.3) and indicate very large
changes in the ocean heat storage. The maintenance
of long time series, whether of hydrographic time
series stations or of multiyear deployments of
1.2 Ocean Processes and Climate Phenomena
25
Clarke, Church and Gould
Pacific some 3°C above the 1961–90 base level.
An operational system based on the TAO (Tropical Ocean Atmosphere)–Triton moored buoy
array and upper ocean temperature measurements
from volunteer observing ships is now in place to
provide real-time monitoring of the state of the
surface and subsurface equatorial Pacific Ocean
and atmosphere (McPhaden et al., 1998). Models
developed over the past decade and using this data
have demonstrated useful skill in predicting the
evolution of the most recent (1997–98) event. For
a more detailed description of ENSO, see Philander
(1990) and the special issue of Journal of Geophysical Research (The TOGA Decade, June 1998).
A concern in interpreting the WOCE data set is
the extent to which the evolution of the tropical
Pacific during the 1990s imposed biases on the
largely extratropical observations. Most of the
one-time hydrographic sections in the Pacific were
occupied between 1990 and 1994 whereas in the
North Atlantic observations were spread throughout the 1990–98 period. The evaluation of such
biases as may exist has yet to be made.
1.2.6.2 Other tropical sea surface
temperature anomalies
Similar coherent responses of the equatorial
oceans to atmospheric forcing (and related feedbacks) have been documented in the Atlantic
(Mehta and Delworth, 1995) and Indian Oceans
(Saji et al., 1999; Webster et al., 1999). In the
Atlantic there are two competing hypotheses for
the causes of the variability of the meridional temperature gradient (the Atlantic dipole). One view is
that it is caused by regional ocean–atmosphere
positive feedbacks (Chang et al., 1997); the other
that the anomalies on either side of the equator are
due to independent processes in each hemisphere
(Enfield and Mayer, 1997). Just as with El Niño,
these equatorial anomalies have important impacts
on the climate of the adjacent land masses (Birkett
et al., 1999; Markam and McLain, 1977; Bertacchi
et al., 1998). Thus both the development of in-situ
observing systems and of predictive models are
presently receiving considerable attention.
1.2.6.3 Extratropical anomalies – decadal and
longer-term variability
Over both the North Pacific and North Atlantic
there are clearly defined patterns of atmospheric
variability that show persistence over multiyear time
scales. The most prominent of these are the Pacific
Decadal Oscillation (PDO) and the North Atlantic
Oscillation (NAO). Each can be regarded as a
regional manifestation of the wider Arctic Oscillation (AO). The status of knowledge of these matters
is summarized in the CLIVAR Implementation Plan
(World Climate Research Programme, 1998).
In the North Pacific, there has been a general
decrease in winter Sea Level Pressure (SLP) accompanied by increases in air and Sea Surface Temperature (SST) since the mid-1970s. When the
variability within the ENSO frequency band is
removed, the decadal variability within surface and
subsurface temperature fields is an anomaly pattern
that rotates clockwise around the North Pacific
Gyre (Zhang and Levitus, 1997). The oceanic
impacts arising from this variability has been discussed by Polovina et al. (1995).
For the NAO, the accepted index of its state
is the normalized December–March atmospheric
pressure difference between Lisbon and Reykjavik
and this can be analysed in the instrumental record
back to c. 1870. High values indicate a large pressure gradient and consequent enhanced westerly
airflow carrying heat and moisture lost from the
ocean across western Europe. During the period of
WOCE observations the NAO has remained in a
persistently high state. The oceanic impacts of the
NAO are discussed at length in Dickson et al.
(Chapter 7.3).
1.2.6.4 Changes in properties of deep and
intermediate waters
Small changes in air–sea heat fluxes, such as might
result from decadal climate variability and anthropogenic climate change, are difficult to observe.
However, because the ocean is an integrator of
short-term variability, these air–sea flux changes are
manifested as changes in ocean temperatures and
salinities. The WOCE data set serves as an important baseline against which past and future changes
can be assessed. Comparison of WOCE sections
with earlier International Geophysical Year (IGY)
sections has revealed widespread and significant
changes in water mass properties over decadal time
scales. These are comprehensively summarized by
Dickson et al. (Chapter 7.3) and indicate very large
changes in the ocean heat storage. The maintenance
of long time series, whether of hydrographic time
series stations or of multiyear deployments of
1.2 Ocean Processes and Climate Phenomena
25
Clarke, Church and Gould