the ocean. First, there is the overturning (or
thermohaline) circulation. Cold deep and bottom
waters generally flow equatorward and are balanced by an equal and opposite transport of warm
surface waters. The Ekman transport associated
with the direct wind forcing is an important component of this surface transport. Second, there are
the gyre circulations operating in the horizontal
plane. The intense western boundary currents of
the subtropical gyres carry warm water poleward.
After loss of heat to the atmosphere, the cooled
water returns equatorward in the broad eastern
return flows. Finally, time-variable ocean eddies
can carry heat poleward.
As usual in ocean circulation, these three mechanisms are not independent and estimating the
amount of heat carried by each depends on the
definitions chosen. While all of the mechanisms
operate simultaneously, the vertical overturning is
dominant in the North Atlantic Ocean (Hall and
Bryden, 1982). Indeed, the southward flow of deep
cold water in the Atlantic results in a northward
ocean heat transport at all latitudes in this ocean
(Fig. 6.1.7, see Plate 6.1.7, p. 492, Bryden and
Imawaki). In the North Pacific, where there is no
deep water formation, the heat transport by the subtropical gyre is relatively more important (Bryden
et al., 1991). In the Southern Ocean, ocean eddies
are thought to be the prime mechanism for the poleward heat transport (overcoming an equatorward
transport associated with the northward advection
of near-surface water; Rintoul et al., Chapter 4.6).
In coarse-resolution coupled atmosphere–ocean
climate models, the ocean modules have generally
poorly simulated the observed ocean heat transport because they underestimate the strength of
the gyres and do not model the eddy transports.
As a result, large ‘unphysical’ flux corrections at
the air–sea interface have been needed to maintain
a ‘realistic’ climate. Improvements in the simulation of both the atmosphere and ocean heat transports in some recent models have reduced or even
eliminated the need for flux corrections. This
progress has resulted from several improvements
in the ocean models as well as from increased resolution (see Wood, Chapter 2.3).
The ocean also stores significant quantities of
carbon with about a third of the carbon dioxide
released by the burning of fossil fuels sequestered
in the ocean. The same mechanisms that store
and transport heat and fresh water also store and
transport carbon. Wallace (Chapter 6.3) differentiates between the ocean transport of anthropogenic
(or excess) carbon and the transport of natural
(preindustrial) carbon. In preindustrial times the
Atlantic Ocean is thought to have transported
carbon southwards, whereas recent calculations
indicate there is now a northward transport of
anthropogenic carbon in the Atlantic.
1.2.6 Climatic and oceanic variability
In the previous sections, we have introduced the
main physical processes important for understanding the ocean’s role in climate. We now briefly
introduce a number of climate phenomena in
which the ocean plays a significant role. For a more
detailed description the reader is referred to other
chapters in this book and the references cited.
1.2.6.1 El Niño-Southern Oscillation (ENSO)
This energetic coupled ocean–atmosphere phenomenon is perhaps the best-known example of interannual climate variability (Philander, 1990). Under
normal conditions the trade winds maintain a pool
of high-temperature water in the western equatorial
Pacific. This warm reservoir drives intense atmospheric convection that provides abundant rainfall in
such areas as Papua New Guinea and Indonesia. In
simplest terms, El Niño occurs when a relaxation of
the trade winds allows warm water to dominate the
normally cool eastern equatorial Pacific, taking
with it the zone of intense convection. El Niño’s
local influences include increased rainfall and suppression of the coastal upwelling and marine productivity off the coast of Peru and dry conditions in
the western equatorial Pacific. Because this equatorial Pacific convection is the principal driver of the
large-scale meridional overturning circulation of the
global atmosphere, ENSO has a near-global impact
through changes in this atmospheric circulation.
ENSO events typically have a 2-year duration
encompassing the warm El Niño and following
cold La Niña phase, but they occur irregularly
with repeat cycles as short as 2 years or as long as
10. The early 1990s were an unusually long period
of moderately warm El Niño-like conditions in the
eastern and central tropical Pacific (Trenberth and
Hoar, 1996) followed by what was considered by
some measures to be the largest El Niño event ever
documented (Trenberth and Coughlan, 1998;
Coughlan, 1999). This peaked at the end of 1998
SECTION 1 THE OCEAN AND CLIMATE
24
thermohaline) circulation. Cold deep and bottom
waters generally flow equatorward and are balanced by an equal and opposite transport of warm
surface waters. The Ekman transport associated
with the direct wind forcing is an important component of this surface transport. Second, there are
the gyre circulations operating in the horizontal
plane. The intense western boundary currents of
the subtropical gyres carry warm water poleward.
After loss of heat to the atmosphere, the cooled
water returns equatorward in the broad eastern
return flows. Finally, time-variable ocean eddies
can carry heat poleward.
As usual in ocean circulation, these three mechanisms are not independent and estimating the
amount of heat carried by each depends on the
definitions chosen. While all of the mechanisms
operate simultaneously, the vertical overturning is
dominant in the North Atlantic Ocean (Hall and
Bryden, 1982). Indeed, the southward flow of deep
cold water in the Atlantic results in a northward
ocean heat transport at all latitudes in this ocean
(Fig. 6.1.7, see Plate 6.1.7, p. 492, Bryden and
Imawaki). In the North Pacific, where there is no
deep water formation, the heat transport by the subtropical gyre is relatively more important (Bryden
et al., 1991). In the Southern Ocean, ocean eddies
are thought to be the prime mechanism for the poleward heat transport (overcoming an equatorward
transport associated with the northward advection
of near-surface water; Rintoul et al., Chapter 4.6).
In coarse-resolution coupled atmosphere–ocean
climate models, the ocean modules have generally
poorly simulated the observed ocean heat transport because they underestimate the strength of
the gyres and do not model the eddy transports.
As a result, large ‘unphysical’ flux corrections at
the air–sea interface have been needed to maintain
a ‘realistic’ climate. Improvements in the simulation of both the atmosphere and ocean heat transports in some recent models have reduced or even
eliminated the need for flux corrections. This
progress has resulted from several improvements
in the ocean models as well as from increased resolution (see Wood, Chapter 2.3).
The ocean also stores significant quantities of
carbon with about a third of the carbon dioxide
released by the burning of fossil fuels sequestered
in the ocean. The same mechanisms that store
and transport heat and fresh water also store and
transport carbon. Wallace (Chapter 6.3) differentiates between the ocean transport of anthropogenic
(or excess) carbon and the transport of natural
(preindustrial) carbon. In preindustrial times the
Atlantic Ocean is thought to have transported
carbon southwards, whereas recent calculations
indicate there is now a northward transport of
anthropogenic carbon in the Atlantic.
1.2.6 Climatic and oceanic variability
In the previous sections, we have introduced the
main physical processes important for understanding the ocean’s role in climate. We now briefly
introduce a number of climate phenomena in
which the ocean plays a significant role. For a more
detailed description the reader is referred to other
chapters in this book and the references cited.
1.2.6.1 El Niño-Southern Oscillation (ENSO)
This energetic coupled ocean–atmosphere phenomenon is perhaps the best-known example of interannual climate variability (Philander, 1990). Under
normal conditions the trade winds maintain a pool
of high-temperature water in the western equatorial
Pacific. This warm reservoir drives intense atmospheric convection that provides abundant rainfall in
such areas as Papua New Guinea and Indonesia. In
simplest terms, El Niño occurs when a relaxation of
the trade winds allows warm water to dominate the
normally cool eastern equatorial Pacific, taking
with it the zone of intense convection. El Niño’s
local influences include increased rainfall and suppression of the coastal upwelling and marine productivity off the coast of Peru and dry conditions in
the western equatorial Pacific. Because this equatorial Pacific convection is the principal driver of the
large-scale meridional overturning circulation of the
global atmosphere, ENSO has a near-global impact
through changes in this atmospheric circulation.
ENSO events typically have a 2-year duration
encompassing the warm El Niño and following
cold La Niña phase, but they occur irregularly
with repeat cycles as short as 2 years or as long as
10. The early 1990s were an unusually long period
of moderately warm El Niño-like conditions in the
eastern and central tropical Pacific (Trenberth and
Hoar, 1996) followed by what was considered by
some measures to be the largest El Niño event ever
documented (Trenberth and Coughlan, 1998;
Coughlan, 1999). This peaked at the end of 1998
SECTION 1 THE OCEAN AND CLIMATE
24