Glen Romine
EF5
Mike,
If I understand what you mean by spinning horizontally, then you are talking about rotating air columns that are horizontal? You have seen these many times really, for instance when you look at cloud streets. These are caused by organized horizontal roll vortices. Of course, you don't always get organized horizontal roll vortices, but only when the vertical stability and vertical change in wind speed with height are in certain regimes (and the moisture is appropriate to put clouds on the top of the rolls only).
Still, if you have wind speed changing with height, this vertical speed shear, if somehow tilted (by say an updraft or downdraft) will put some component of that 'spin' into the vertical. Another way to think of this is the momentum of air at different heights. For a typical scenario, the westerly winds increase with increasing height. Let's assume for simplicity the air at the ground is stationary. Then, if you displaced this air from near the ground to some height above, the displaced air is stationary, and the mid-level air will have to flow around it, (so looking to the east) with cyclonic spin on the right side and anti-cyclonic spin on the left side. This idea gets more complicated with a veering wind profile with height, but the net effect is still that an updraft in an environment with vertical shear will have a tendency to rotate. This is rotation on a relatively large scale - the entire updraft, so the tornado cyclone is a much smaller circulation embedded within the larger mesocyclone circulation.
Generally, the mesocyclone circulations, since they are larger, tend to have longer life spans (the bigger the circulation, the more stable it is). There has been some research on what controls the life span of mesocyclones, but not much. The embedded tornado cyclones are typically much shorter lived and transient. When you see the clear slot isolating a small component of the updraft - where you typically find the wall cloud and sometimes obvious cloud base rotation, you are seeing a component of the tornado cyclone. As noted above, these may sometimes have a shroud of dust in them, as the dust becomes sort of trapped in the swirl of air in the tornado cyclone. Researchers are generally stuck on several components. What initiated the RFD surge (visually the development of the clear slot)? What controls the size of the tornado cyclone and the amount of circulation in it? What storm processes favor the tornado cyclone circulation being maintained for longer windows of time? Once the tornado cyclone is in place, what controls whether this circulation will tighten up into a tornado or not. When is the tornado a sub-vortex within the tornado cyclone vs. at the center of it? How are all of the above controlled by aspects of the larger scale environment?
There are some researchers that believe tornadoes must serve some distinct purpose (like Chuck), and finding this will yield some insight into when a tornado is 'necessary'. I'm not sure I buy into that camp yet, as it still seems more like an accident to me, though I do think we are making steady progress on recognizing when one of these 'accidents' is likely to occur. It might not be possible to ever accurately predict most tornadoes, though the odds of reasonably predicting the stronger ones is probably more achievable, since these are much more common with mesocyclones, and there is a larger body of active researchers looking at mesocyclone tornadoes.
That mentioned, I would note that there is no barrier to mesocyclones producing tornadoes from the same mechanism that an ordinary updraft would (confusingly now called a landspout tornado). In fact, they are probably better at it since the updraft rotation enhances the strength of the updraft leading to more effective stretching and concentration of the 'spin' along the boundary. While it is generally believed that a supercell is capable of creating a tornado without the aid of external sources of rotation (such as a boundary), there is no knowledge of how often mesocyclone tornadoes are 'supplemented' by other rotation sources that made the difference in whether that storm produced a tornado or not.
