Geostrophy thus leads to eastward flow in the narrow sloping region between the two flat parts of the
lower surface, relative to the denser, stagnant fluid
beneath. Flow between the two surfaces is zero
wherever the lower surface is flat, so the two sides
of the sloping region are streamlines. With this
geometry, the observed near-constant mass transport and increasing downstream thickness of the
currents with distance east is automatically satisfied. Using conservation of vorticity and Bernoulli
function, they solved for the latitude, width and
flow distribution in the jet. For realistic inflow conditions, the poleward movement is well simulated,
with peak jet velocity increasing 50% downstream.
Rowe et al. (2000) explore the SCCs further, showing that in individual sections they form sharp
(20 km wide) poleward boundaries of a pool of
constant vorticity water. Net flow within this pool
is westward, not eastward, due to strong westward
transports near the equator. Horizontal convergence/divergence of these flows implies that strong
diapycnal processes are involved within the constant vorticity pool. Kitamura and Suginohara
(1987), following McPhaden (1984), obtained
SSCC-like features in their model, though they were
a factor of 10 too weak; their results were well captured by a linear analytic model. Ishida et al. (1998)
obtained more realistic SCCs in an Ocean General
Circulation Model (OGCM). The fine vertical resolution (55 levels) and horizontal resolution (1/4°)
extending to high latitudes are considered important elements of the simulation, since SCCs are
believed to mark sharp fronts that extend to quite
high latitudes along eastern boundaries (P. Lu,
personal communication).
The flow paths of the SCCs may also be of
importance for understanding the heat budget of
the upper equatorial Pacific, because of the possibility that the SCCs supply much of the Indonesian
Throughflow (e.g. Wijffels, 1993; Lu et al., 1998).
Rossby wave beams and the annual cycle of temperature
below the thermocline
Before leaving the topic of subthermocline zonal
flows along the equator, it may also be noted that
McCreary (1984) combined all vertical modes
with the same meridional wavenumber to form
equatorial ‘beams’, which he showed were capable
of carrying wind-forced energy downward to the
bottom. Rothstein et al. (1988a) found that if realistic mean currents are included, this process is
inhibited for Kelvin waves by critical-layer absorption. On the other hand, Lukas and Firing (1985)
and Kessler and McCreary (1993) report observations of annual-frequency temperatures below the
thermocline that match McCreary’s (1984) model
for the Rossby wave beam. Davis (1998b) finds
the bulk of variability in zonal velocity at 900 m
as measured by floats is accounted for by annual
variability of spatial modes predicted by a model.
Evidently, the nine cross-equatorial sections
undertaken as part of WOCE (e.g. Kawabe and
Taira, 1998; Kaneko et al., 1998; Roden, 1998;
Tsuchiya and Talley, 1998; Wijffels et al., 1998)
will provide a powerful resource for gaining further understanding of the deep equatorial Pacific.
However, it should be noted that the problem
of interpreting and comparing synoptic sections
is particularly difficult below the thermocline,
because the TOGA array provides no guidance to
non-seasonal variability at these depths.
4.3.3 Equatorial Atlantic
4.3.3.1 Introduction
In the tropical Atlantic Ocean there has been a
strong interest in investigating the subsurface circulation during WOCE. Most of the observational
work there in this period consisted of deep hydrographic and current sections, western boundary
and equatorial mooring arrays, and subsurface
floats. The seasonal response of the near-surface
equatorial Atlantic Ocean was studied earlier, to a
large extent during the FOCAL/SEQUAL experiment in 1983–84, which indicated the extent to
which the near-surface variability in temperature
and currents is forced by the seasonal variability of
the wind stress. This data set presented a nice
example of interannual variability, suggesting
coupling mechanisms between the near-equatorial
ocean and the atmosphere, reminiscent of El Niño
(Philander, 1986; Delecluse et al., 1994), which
were investigated further by Zebiak and Cane
(1987), Zebiak (1993) and Servain (1991). Renewed
interest in air–sea coupling mechanisms in the
equatorial Atlantic have originated from the analysis of long time series of surface parameters (in
particular, SST and winds) and from modelling
experiments. This has led to the identification
of extra-equatorial modes of variability with
a very low frequency (decadal) (Servain, 1991),
and possible air–sea mechanisms with a mostly
SECTION 4 THE GLOBAL FLOW FIELD
226
lower surface, relative to the denser, stagnant fluid
beneath. Flow between the two surfaces is zero
wherever the lower surface is flat, so the two sides
of the sloping region are streamlines. With this
geometry, the observed near-constant mass transport and increasing downstream thickness of the
currents with distance east is automatically satisfied. Using conservation of vorticity and Bernoulli
function, they solved for the latitude, width and
flow distribution in the jet. For realistic inflow conditions, the poleward movement is well simulated,
with peak jet velocity increasing 50% downstream.
Rowe et al. (2000) explore the SCCs further, showing that in individual sections they form sharp
(20 km wide) poleward boundaries of a pool of
constant vorticity water. Net flow within this pool
is westward, not eastward, due to strong westward
transports near the equator. Horizontal convergence/divergence of these flows implies that strong
diapycnal processes are involved within the constant vorticity pool. Kitamura and Suginohara
(1987), following McPhaden (1984), obtained
SSCC-like features in their model, though they were
a factor of 10 too weak; their results were well captured by a linear analytic model. Ishida et al. (1998)
obtained more realistic SCCs in an Ocean General
Circulation Model (OGCM). The fine vertical resolution (55 levels) and horizontal resolution (1/4°)
extending to high latitudes are considered important elements of the simulation, since SCCs are
believed to mark sharp fronts that extend to quite
high latitudes along eastern boundaries (P. Lu,
personal communication).
The flow paths of the SCCs may also be of
importance for understanding the heat budget of
the upper equatorial Pacific, because of the possibility that the SCCs supply much of the Indonesian
Throughflow (e.g. Wijffels, 1993; Lu et al., 1998).
Rossby wave beams and the annual cycle of temperature
below the thermocline
Before leaving the topic of subthermocline zonal
flows along the equator, it may also be noted that
McCreary (1984) combined all vertical modes
with the same meridional wavenumber to form
equatorial ‘beams’, which he showed were capable
of carrying wind-forced energy downward to the
bottom. Rothstein et al. (1988a) found that if realistic mean currents are included, this process is
inhibited for Kelvin waves by critical-layer absorption. On the other hand, Lukas and Firing (1985)
and Kessler and McCreary (1993) report observations of annual-frequency temperatures below the
thermocline that match McCreary’s (1984) model
for the Rossby wave beam. Davis (1998b) finds
the bulk of variability in zonal velocity at 900 m
as measured by floats is accounted for by annual
variability of spatial modes predicted by a model.
Evidently, the nine cross-equatorial sections
undertaken as part of WOCE (e.g. Kawabe and
Taira, 1998; Kaneko et al., 1998; Roden, 1998;
Tsuchiya and Talley, 1998; Wijffels et al., 1998)
will provide a powerful resource for gaining further understanding of the deep equatorial Pacific.
However, it should be noted that the problem
of interpreting and comparing synoptic sections
is particularly difficult below the thermocline,
because the TOGA array provides no guidance to
non-seasonal variability at these depths.
4.3.3 Equatorial Atlantic
4.3.3.1 Introduction
In the tropical Atlantic Ocean there has been a
strong interest in investigating the subsurface circulation during WOCE. Most of the observational
work there in this period consisted of deep hydrographic and current sections, western boundary
and equatorial mooring arrays, and subsurface
floats. The seasonal response of the near-surface
equatorial Atlantic Ocean was studied earlier, to a
large extent during the FOCAL/SEQUAL experiment in 1983–84, which indicated the extent to
which the near-surface variability in temperature
and currents is forced by the seasonal variability of
the wind stress. This data set presented a nice
example of interannual variability, suggesting
coupling mechanisms between the near-equatorial
ocean and the atmosphere, reminiscent of El Niño
(Philander, 1986; Delecluse et al., 1994), which
were investigated further by Zebiak and Cane
(1987), Zebiak (1993) and Servain (1991). Renewed
interest in air–sea coupling mechanisms in the
equatorial Atlantic have originated from the analysis of long time series of surface parameters (in
particular, SST and winds) and from modelling
experiments. This has led to the identification
of extra-equatorial modes of variability with
a very low frequency (decadal) (Servain, 1991),
and possible air–sea mechanisms with a mostly
SECTION 4 THE GLOBAL FLOW FIELD
226