Fluid Dynamics
-------------
This a 35 frame animation loop stored as a multi-image 8-bit color TIFF file(NIH Image ÒstackÓ). it was contributed by Fang Zhong at Los Alamos(zhong@goshawk.LANL.GOV).

Since the frames are very small(96x96), it is a good idea to enlarge the window 2 or 3 times before animating. To do this, click several times within the window with the magnifying glass and then make the window as large as possible by dragging the grow box in the lower right corner.


Description
----------
        I am studying Rayleigh-Benard convection in rotation.  We have a layer of water in rotation. The top temperature of the layer is held  at constant within 0.25 mK rms. The bottom is heated uniformly. When the heat is small, a linear temperature profile is established and the  heat is conducted by molecular collisions through the layer.  A convective motion sets in when the heat is increased over a critical value.  In a convective state, warm fluid moves up and cold fluid moves down. There is a certain structure in the density variation across the water layer.  Our goal is to study how this structure is influenced by the rotation rate.  This basic research will help scientists eventually to understand and model atmosphere and ocean flow since there are always heats moving in them and our earth is under rotation.
        We use different techniques in our research. Socalled shadow-graph technique is used currently. A collimated light is sent into the water layer, reflected back by a mirror at the bottom of the water layer and received by a CCD camera. A resulted image is grabbed and stored on computer hard disk.
        The density variation results in the variation of the refractive index in the water layer which works as many small lenses.  A warm fraction of water diverges light and a cold one converges light. It corresponds to a dark part and a bright part in the resulted image.
	So when you see a pattern in our images, you link it to the variation of a water density and motion.
	For the dark circles you will see in our images, the heat is transferred through vertical votex flows.  This is the direct result of the Coriolis force in a rotating frame.  I am not quite clear now how you can connect this with tornatos.
	The include images show a transient from convective state to turbulent state.
	Different pattern symmetries can be prepared inintially but will be lost eventually through vortex collision and anihilation. The average number of vortices depends on rotation rate and heat input.