264
Theory of the Ventilated Thermocline
large excursions of the outcrop lines through the seasons present us with the
problem of deciding which position of the outcrop line should be used for a
steady theory. Should the average outcrop position be used or is some
particular position of the outcrop line more relevant than others to the steady
structure of the thermocline?
The key idea was supplied by Stommel (1979), who gave not only the
essential explanation for the steadiness of the structure beneath the mixed layer
but also an implied prescription as to which position of the outcrop line should
be used for a steady theory. Stommel pointed out that the southward
movements of the outcrop lines in the gyre are observed to be much more rapid
than the southward motion of the fluid leaving the mixed layer and entering the
thermocline. He examined the motion of the outcrop line of the 20-21 °C water
in the eastern North Atlantic during the course of the seasons and noted that
the southward progression of the outcrop line occurs at an average rate of over
7 cmjs, much larger than the average geostrophic flow. He concluded that any
fluid which subducts earlier in the year, during the summer and fall, would be
swallowed up by the advancing and deepening mixed layer during the course of
the advancing winter. This process is shown schematically in Fig. 4.12.2. In the
figure the mixed layer depth is shown at two times, summer with a shallow
depth hms, and late winter (March) when its depth is greatest and equal to hmw·
A sloping line marked 1/J shows the idealized path of fluid pumped downward
from the Ekman layer after it enters the thermocline. The outcrop line for the
particular density surface shown is at ()s in the summer and at 8w, its most
southern position, in winter. The position of a fluid element leaving the mixed
layer is labeled with an S for its summer position and W for its winter position.
As the fluid downwells it is overtaken by the deepening mixed layer and
remains encapsulated within the mixed layer until the late winter when fluid at
the base of the mixed layer is able to exit the mixed layer and enter the
thermocline. Thus the only fluid, at that latitude, which enters the thermocline
is fluid whose density surface outcrops at 8w in late winter. Thus, in spite of the
large variation of the surface density at that location over the course of the year
only the density associated with the late winter position of the outcrop affects
the ventilation of the permanent thermocline. Stommel likened the process,
whereby only water of a particular density class is able to enter the thermocline
at a given position, to the role of the imaginary "demons" in Maxwell's theory
of gases. The deepening mixed layer, allowing only a narrow range of density
to subduct at any geographical location, acts as a selective gatekeeper, and it
has become common to refer to the selection process of subducted density as
"Stommel's demon."
The simple argument of Stommel's suggests that a steady theory of the
thermocline can be constructed by choosing the outcrop lines to be those
observed in late winter rather than the average position of the outcrop lines.
The kinematics of Stommel's demonstration is highly simplified. It is twodimensional and quasistatic in its construction of flow paths. More
sophisticated kinematic analyses by Cushman-Roisin (1987), Marshall et a!.
Theory of the Ventilated Thermocline
large excursions of the outcrop lines through the seasons present us with the
problem of deciding which position of the outcrop line should be used for a
steady theory. Should the average outcrop position be used or is some
particular position of the outcrop line more relevant than others to the steady
structure of the thermocline?
The key idea was supplied by Stommel (1979), who gave not only the
essential explanation for the steadiness of the structure beneath the mixed layer
but also an implied prescription as to which position of the outcrop line should
be used for a steady theory. Stommel pointed out that the southward
movements of the outcrop lines in the gyre are observed to be much more rapid
than the southward motion of the fluid leaving the mixed layer and entering the
thermocline. He examined the motion of the outcrop line of the 20-21 °C water
in the eastern North Atlantic during the course of the seasons and noted that
the southward progression of the outcrop line occurs at an average rate of over
7 cmjs, much larger than the average geostrophic flow. He concluded that any
fluid which subducts earlier in the year, during the summer and fall, would be
swallowed up by the advancing and deepening mixed layer during the course of
the advancing winter. This process is shown schematically in Fig. 4.12.2. In the
figure the mixed layer depth is shown at two times, summer with a shallow
depth hms, and late winter (March) when its depth is greatest and equal to hmw·
A sloping line marked 1/J shows the idealized path of fluid pumped downward
from the Ekman layer after it enters the thermocline. The outcrop line for the
particular density surface shown is at ()s in the summer and at 8w, its most
southern position, in winter. The position of a fluid element leaving the mixed
layer is labeled with an S for its summer position and W for its winter position.
As the fluid downwells it is overtaken by the deepening mixed layer and
remains encapsulated within the mixed layer until the late winter when fluid at
the base of the mixed layer is able to exit the mixed layer and enter the
thermocline. Thus the only fluid, at that latitude, which enters the thermocline
is fluid whose density surface outcrops at 8w in late winter. Thus, in spite of the
large variation of the surface density at that location over the course of the year
only the density associated with the late winter position of the outcrop affects
the ventilation of the permanent thermocline. Stommel likened the process,
whereby only water of a particular density class is able to enter the thermocline
at a given position, to the role of the imaginary "demons" in Maxwell's theory
of gases. The deepening mixed layer, allowing only a narrow range of density
to subduct at any geographical location, acts as a selective gatekeeper, and it
has become common to refer to the selection process of subducted density as
"Stommel's demon."
The simple argument of Stommel's suggests that a steady theory of the
thermocline can be constructed by choosing the outcrop lines to be those
observed in late winter rather than the average position of the outcrop lines.
The kinematics of Stommel's demonstration is highly simplified. It is twodimensional and quasistatic in its construction of flow paths. More
sophisticated kinematic analyses by Cushman-Roisin (1987), Marshall et a!.