266
Theory of the Ventilated Thermocline
analysis of data from the "f3 triangle" region (a relatively small triangle
centered at 27°N, 32°30'W in the North Atlantic which we would identify
with the eastern edge of the ventilalted region) found that advection of
potential vorticity alone is sufficient to explain the local potential vorticity
balance. Their estimate, using salinity as a tracer, of the lateral mixing
coefficient, KH, was 0.5 x 10 7 cm 2 js. For scales of the order of 1000 km this
would give a ratio of the diffusion of q to its advection of 0.05 for velocities of
the order 1 cmjs and hence negligible.
There have been several attempts to directly apply the theory to the actual
oceanic thermocline circulation.
We discussed in Section 4.11 the study by Huang (1989b) of the North
Atlantic thermocline circulation using the continuous model with a spatially
varying mixed layer. The predictions of the theory for the structure of the
isopycnals are very encouraging. If Fig. 4.11.5c, for example, is compared with
the observations of the same density surface shown in Fig 4.1.3, the agreement
is seen to be very good. The variation in depth of the (Je = 26.8 surface is well
captured by the theory. Ventilation is clearly important for this outcropping
surface in the theory which predicts a large area covered by ventilated
streamlines. Observations of chemical tracers in the ocean reveal that this
density surface is indeed well ventilated. Sarmiento et al. (1982) presents maps
of both salinity and tritium on this surface, and long tongues of both tracers
stretching from the outcrop line into the gyre clearly illustrate the ventilation
process which is the heart of the theory. A very beautiful representation of the
ventilation process, as revealed in the tritium tracer data, has been described by
Jenkins (1994). Figure 4.12.3 shows the age of the tracer tritium on several
density surfaces in the North Atlantic. Tritium, produced in the atmosphere by
the detonation of nuclear weapons in the 1960s, enters the ocean and is
advected with the flow. Its mutation to 3 He gives a measure of its age at any
location. In the figure isolines of age of the tracer on selected surfaces are
shown. Examining the (Je = 26.8 surface again, we see the gradual increase in
age as we follow the fluid away from the outcrop line. Jenkins argues that the
distribution of the tritium tracer is consistent with flow trajectories on the
subducted interfaces.
Armi and Stommel (1983) examined in detail the structure of a small
portion of the subtropical gyre in the eastern North Atlantic. We commented
above on their conclusion that conservation of potential vorticity is apparently
satisfied in the location of the f3 triangle region. In addition, they point out that
on the (Je = 26.5-27.5 density layer which outcrops in the subtropical gyre
where the Ekman velocity is downward, the potential vorticity gradient is weak
even though this layer is clearly ventilated. This layer outcrops in a region
where the outcrop line slopes from northwest to southeast, and the relative
uniformity of the potential vorticity on this surface is quite possibly due to the
subduction of uniform q water as described at the end of Section 4.9.
Theory of the Ventilated Thermocline
analysis of data from the "f3 triangle" region (a relatively small triangle
centered at 27°N, 32°30'W in the North Atlantic which we would identify
with the eastern edge of the ventilalted region) found that advection of
potential vorticity alone is sufficient to explain the local potential vorticity
balance. Their estimate, using salinity as a tracer, of the lateral mixing
coefficient, KH, was 0.5 x 10 7 cm 2 js. For scales of the order of 1000 km this
would give a ratio of the diffusion of q to its advection of 0.05 for velocities of
the order 1 cmjs and hence negligible.
There have been several attempts to directly apply the theory to the actual
oceanic thermocline circulation.
We discussed in Section 4.11 the study by Huang (1989b) of the North
Atlantic thermocline circulation using the continuous model with a spatially
varying mixed layer. The predictions of the theory for the structure of the
isopycnals are very encouraging. If Fig. 4.11.5c, for example, is compared with
the observations of the same density surface shown in Fig 4.1.3, the agreement
is seen to be very good. The variation in depth of the (Je = 26.8 surface is well
captured by the theory. Ventilation is clearly important for this outcropping
surface in the theory which predicts a large area covered by ventilated
streamlines. Observations of chemical tracers in the ocean reveal that this
density surface is indeed well ventilated. Sarmiento et al. (1982) presents maps
of both salinity and tritium on this surface, and long tongues of both tracers
stretching from the outcrop line into the gyre clearly illustrate the ventilation
process which is the heart of the theory. A very beautiful representation of the
ventilation process, as revealed in the tritium tracer data, has been described by
Jenkins (1994). Figure 4.12.3 shows the age of the tracer tritium on several
density surfaces in the North Atlantic. Tritium, produced in the atmosphere by
the detonation of nuclear weapons in the 1960s, enters the ocean and is
advected with the flow. Its mutation to 3 He gives a measure of its age at any
location. In the figure isolines of age of the tracer on selected surfaces are
shown. Examining the (Je = 26.8 surface again, we see the gradual increase in
age as we follow the fluid away from the outcrop line. Jenkins argues that the
distribution of the tritium tracer is consistent with flow trajectories on the
subducted interfaces.
Armi and Stommel (1983) examined in detail the structure of a small
portion of the subtropical gyre in the eastern North Atlantic. We commented
above on their conclusion that conservation of potential vorticity is apparently
satisfied in the location of the f3 triangle region. In addition, they point out that
on the (Je = 26.5-27.5 density layer which outcrops in the subtropical gyre
where the Ekman velocity is downward, the potential vorticity gradient is weak
even though this layer is clearly ventilated. This layer outcrops in a region
where the outcrop line slopes from northwest to southeast, and the relative
uniformity of the potential vorticity on this surface is quite possibly due to the
subduction of uniform q water as described at the end of Section 4.9.