76
Pacific Oceans through the Antarctic Circumpolar Current. In the Northern part of the Indian
and Pacific Oceans, upwelling is taking place. Finally a warm, shallow water current returns to
the Atlantic, through the Indonesian Passage, the Indian Ocean and a branch of the Agulhas
Current - around the southern tip of Africa. This return route is usually called "warm water
route", as opposed to the "cold water route" - Pacific to Atlantic through Drake Passage-,
which is probably of lesser importance (Gordon, 1986; Doos, 1994). This global ocean
circulation scheme is known as the "conveyor belt" (Gordon, 1986, Broecker, 1991).
"
4'
o.
IG.
300
•••
122$
38S0
-80
- 40
o
40
80
Latitude
Figure 17. Meridional streamfunction, 'l', in the Atlantic and Arctic for simulation I (wind-only
simulation). The contour labels are in Sverdrups. The arrows indicate the direction of the
circulation.
The North Atlantic overturning, roughly speaking the production rate of NADW, and the
NADW southward transport, evaluated at 20 0 S, are significantly better in simulation III than in
II, where the seasonal cycle of the forcing is absent (Table 5). The two equatorial upwelling
cells are present in all numerical experiments, but those of run I reach a depth of about 4000 m,
while the depth of those of the other simulations does not exceed 200-300 m, which is certainly
more realistic.
From the model results discussed above, it seems clear that the wind forcing alone cannot
account resonably well for the circulation in the World Ocean, which does not mean that the
wind stress should be neglected. The wind only simulation fails to represent most important
features of the circulation in both the horizontal and meridional planes.
As regards the simulations with wind and thennohaline forcings, the annual means of the
circulation parameters examined in Table 5 are more realistic when the seasonal cycle of the
forcings are taken into account.
As indicated in Table 6, the properties of the water masses are fairly well represented in
simulation Ill, except that the thennocline and intennediate waters - located at depth ranging
from 200 to 1200 m, say - are too warm. This deficiency may be due to the fact that the model
is in general too diffusive and that no isopycnal diffusion scheme is activated.
Pacific Oceans through the Antarctic Circumpolar Current. In the Northern part of the Indian
and Pacific Oceans, upwelling is taking place. Finally a warm, shallow water current returns to
the Atlantic, through the Indonesian Passage, the Indian Ocean and a branch of the Agulhas
Current - around the southern tip of Africa. This return route is usually called "warm water
route", as opposed to the "cold water route" - Pacific to Atlantic through Drake Passage-,
which is probably of lesser importance (Gordon, 1986; Doos, 1994). This global ocean
circulation scheme is known as the "conveyor belt" (Gordon, 1986, Broecker, 1991).
"
4'
o.
IG.
300
•••
122$
38S0
-80
- 40
o
40
80
Latitude
Figure 17. Meridional streamfunction, 'l', in the Atlantic and Arctic for simulation I (wind-only
simulation). The contour labels are in Sverdrups. The arrows indicate the direction of the
circulation.
The North Atlantic overturning, roughly speaking the production rate of NADW, and the
NADW southward transport, evaluated at 20 0 S, are significantly better in simulation III than in
II, where the seasonal cycle of the forcing is absent (Table 5). The two equatorial upwelling
cells are present in all numerical experiments, but those of run I reach a depth of about 4000 m,
while the depth of those of the other simulations does not exceed 200-300 m, which is certainly
more realistic.
From the model results discussed above, it seems clear that the wind forcing alone cannot
account resonably well for the circulation in the World Ocean, which does not mean that the
wind stress should be neglected. The wind only simulation fails to represent most important
features of the circulation in both the horizontal and meridional planes.
As regards the simulations with wind and thennohaline forcings, the annual means of the
circulation parameters examined in Table 5 are more realistic when the seasonal cycle of the
forcings are taken into account.
As indicated in Table 6, the properties of the water masses are fairly well represented in
simulation Ill, except that the thennocline and intennediate waters - located at depth ranging
from 200 to 1200 m, say - are too warm. This deficiency may be due to the fact that the model
is in general too diffusive and that no isopycnal diffusion scheme is activated.