Abstract-
As a basic model relevant to ocean circulation I'll discuss "horizontal
convection": a Boussinesq fluid forced by applying a non-uniform
temperature at its top surface, with all other boundaries insulating. I'll
prove that if the viscosity and thermal diffusivity are lowered to zero the
mechanical energy dissipation per unit mass also vanishes. Thus, according
to some religious definitions of turbulence, this form of convection can
never become truly turbulent (even though the Rayleigh and Reynolds' numbers
are very large). Despite this 'anti-turbulence theorem', horizontal
convection exhibits a transition from a steady to an eddying flow which is
controlled by the Rayleigh and Prandtl numbers. I'll briefly review the
related problem of Rayleigh-Benard convection in which experimental evidence
indicates that the energy dissipation per mass again vanishes in the
inviscid limit.
The implications of these results is that the global ocean circulation is energetically sustained by winds and tides, rather than the pole-to-equator surface density gradient. Thus the ocean is not a heat engine. Rather the ocean is a mechanically driven conveyor belt, transporting heat and freshwater almost as passive contaminants. The engine of the conveyor belt is powered by tides and wind stress. With uncertain consequences, climate models ignore this physics, or sweep it into prescribed eddy diffusivities.
GALCIT Home Page
|
|