4.4
Tropical–Extratropical Oceanic Exchange
Pathways
Zhengyu Liu and S. G. H. Philander
247
OCEAN CIRCULATION AND CLIMATE
Copyright © 2001 Academic Press
ISBN 0-12-641351-7
All rights of reproduction in any form reserved
CHAPTER
4.4.1 The role of diffusion and advection
The salient feature of the thermal structure of the
ocean is the thermocline, which is particularly
sharp and shallow in low latitudes. The processes
that maintain the thermocline are presently of considerable interest because those processes appear
to be of central importance to interdecadal climate
variability and global climate changes (Cane et al.,
1997; Gu and Philander, 1997; Liu, 1998; Zhang
et al., 1998a).
The first studies of the thermocline were motivated by Ellis’ (1751) measurements that revealed
extremely cold water at depth in the tropical
Atlantic. Rumford (1800) speculated that cold
currents from polar regions must sustain the low
temperatures in low latitudes. Lenz (1845), who
first noticed that the depth of the thermocline
decreases with decreasing latitude in the tropics
(see Fig. 4.4.1a), inferred that there is equatorial
upwelling sustained by a pair of thermohaline
cells, one in each hemisphere (see Warren, 1981b,
for a historical review of this topic). A century
later, these ideas are quantified (Robinson and
Stommel, 1959; Robinson and Welander, 1963) in
a model in which the thermocline depends on a
balance between the downward diffusion of heat
and the upwelling of cold water.
Over the past two decades, oceanographers have
found that, to account for the observed thermocline
structure, diffusive thermocline models require a
diffusivity much larger than that which is observed.
This prompted them to explore thermocline models
in which horizontal advection rather than diffusion
is of primary importance (Luyten et al., 1983;
Pedlosky, 1987a; Liu, 1994; McCreary and Lu,
1994). In these models, subduction of surface
waters in the subtropics, which in part is maintained by poleward Ekman drift in the surface
layers, feeds subsurface, equatorward geostrophic
flow along isopycnals. Upwelling, especially at the
equator, closes the circulation. Evidence for such a
circulation comes from hydrographic sections and
tracer data (Tsuchiya, 1968; Wyrtki, 1975; Fine
et al., 1981, 1987; Quay et al., 1983; Roemmich,
1987; Tsuchiya et al., 1989; Johnson and McPhaden,
1999). If it is assumed that water parcels conserve
their potential vorticity fѨ z as they travel equatorward, then major features of the thermocline can
readily be explained. Because the Coriolis parameter f decreases with decreasing latitude, conservation of potential vorticity requires that the vertical
density gradient Ѩ z increase. The reduction of layer
thickness for all the subduction layers further
requires a shoaling of thermocline as the equator
is approached. This tropical–extratropical oceanic
exchange is a means by which a change in surface
conditions in the extratropics can affect the structure of the tropical thermocline, in turn affecting
tropical ocean–atmosphere interactions, and hence
can contribute to climate fluctuations and climate
changes at decadal and longer time scales.
This chapter is a review of recent research on
tropical–extratropical exchanges in the Pacific
Ocean. Section 4.4.2 concerns the relatively shallow exchange of thermocline waters between the
tropics and subtropics. Section 4.4.3 focuses on
the exchange of colder, intermediate waters between
the tropics and subpolar regions. Section 4.4.4
Tropical–Extratropical Oceanic Exchange
Pathways
Zhengyu Liu and S. G. H. Philander
247
OCEAN CIRCULATION AND CLIMATE
Copyright © 2001 Academic Press
ISBN 0-12-641351-7
All rights of reproduction in any form reserved
CHAPTER
4.4.1 The role of diffusion and advection
The salient feature of the thermal structure of the
ocean is the thermocline, which is particularly
sharp and shallow in low latitudes. The processes
that maintain the thermocline are presently of considerable interest because those processes appear
to be of central importance to interdecadal climate
variability and global climate changes (Cane et al.,
1997; Gu and Philander, 1997; Liu, 1998; Zhang
et al., 1998a).
The first studies of the thermocline were motivated by Ellis’ (1751) measurements that revealed
extremely cold water at depth in the tropical
Atlantic. Rumford (1800) speculated that cold
currents from polar regions must sustain the low
temperatures in low latitudes. Lenz (1845), who
first noticed that the depth of the thermocline
decreases with decreasing latitude in the tropics
(see Fig. 4.4.1a), inferred that there is equatorial
upwelling sustained by a pair of thermohaline
cells, one in each hemisphere (see Warren, 1981b,
for a historical review of this topic). A century
later, these ideas are quantified (Robinson and
Stommel, 1959; Robinson and Welander, 1963) in
a model in which the thermocline depends on a
balance between the downward diffusion of heat
and the upwelling of cold water.
Over the past two decades, oceanographers have
found that, to account for the observed thermocline
structure, diffusive thermocline models require a
diffusivity much larger than that which is observed.
This prompted them to explore thermocline models
in which horizontal advection rather than diffusion
is of primary importance (Luyten et al., 1983;
Pedlosky, 1987a; Liu, 1994; McCreary and Lu,
1994). In these models, subduction of surface
waters in the subtropics, which in part is maintained by poleward Ekman drift in the surface
layers, feeds subsurface, equatorward geostrophic
flow along isopycnals. Upwelling, especially at the
equator, closes the circulation. Evidence for such a
circulation comes from hydrographic sections and
tracer data (Tsuchiya, 1968; Wyrtki, 1975; Fine
et al., 1981, 1987; Quay et al., 1983; Roemmich,
1987; Tsuchiya et al., 1989; Johnson and McPhaden,
1999). If it is assumed that water parcels conserve
their potential vorticity fѨ z as they travel equatorward, then major features of the thermocline can
readily be explained. Because the Coriolis parameter f decreases with decreasing latitude, conservation of potential vorticity requires that the vertical
density gradient Ѩ z increase. The reduction of layer
thickness for all the subduction layers further
requires a shoaling of thermocline as the equator
is approached. This tropical–extratropical oceanic
exchange is a means by which a change in surface
conditions in the extratropics can affect the structure of the tropical thermocline, in turn affecting
tropical ocean–atmosphere interactions, and hence
can contribute to climate fluctuations and climate
changes at decadal and longer time scales.
This chapter is a review of recent research on
tropical–extratropical exchanges in the Pacific
Ocean. Section 4.4.2 concerns the relatively shallow exchange of thermocline waters between the
tropics and subtropics. Section 4.4.3 focuses on
the exchange of colder, intermediate waters between
the tropics and subpolar regions. Section 4.4.4