I've read a lot of science stuff about tornadoes, but never any explanation that even came close to making sense. One thing they guess is that the fierce updrafts and downdrafts during thunderstorms, when they are close together, might cause spinning horizontal tubes that somehow get downturned to touch the earth. That seems a very specious idea, because what would cause the downturn? Why would the tube tend to be shaped like a cone or vortex? And why would all that angular momentum horizontally get shifted to vertical angular momentum in the absence of an overwhelming force? It seems they are grasping at straws to come up with a reasonable weather-related air-related explanation. So we need to apply some clear thinking to the problem.
First of all, it's clear that tornadoes must be the result of a sink of some kind. When we try to simulate tornadoes for a kid's classroom, we connect two soda bottles, one half filled, and swirl the fluid around while it drains from top to bottom. The behavior of tornadoes is exactly like that of the whorl of water and air sucked down to a draining sink. It acts exactly right, even down to the roped out S-shaped tail that happens near the end. So what is actually going on in the sky during a tornado? What is being drained into what?
I've lived in the Southeastern US all my life, and so I've had multiple opportunities to observe in the close vicinity of tornadoes and tornado conditions. Obviously, they happen during thunderstorms, usually in the Spring and Fall. One thing I and many other people have observed is that often the sky turns greenish when tornadoes are about, something I would guess is due to the green line of ionized Oxygen, also known as the airglow. So that suggests that the air may get somewhat ionized when the conditions are right for tornadoes.
Another thing I've observed quite frequently in close proximity to tornadoes and that's a kind of continuous flickering lightning all over the clouds from which tornadoes drop, cloud-to-cloud lightning. Neither the green sky nor the continuous lightning are present 100% of the time, but they are frequently enough associated with tornadoes that they can serve as a clue.
A third observation is that tornadoes often come during a respite from the rain on the trailing edge of the heaviest torrential downpours.
So what is it that's draining down the funnel of a tornado? It's obviously not air. There's nothing sucking air out of the sky and underground during a twister. The most obvious thing I can think of is electric charge. Has that idea been tested? I need to research this better to find what the latest observations are. But electromagnetic forces are enormous, and so they would seem to have what it takes to develop the power output of a big tornado.
Perhaps the torrential rain gathers either negative or positive charge out of the air and releases it into the ground, then a tornado forms in the swirl of air caught up in the rush of the charge returning to neutralize conditions.
I would love to read the signal from a big ground loop, similar to the vehicle loop detectors we use in the roads to control our traffic signals, but much larger, which could tell us what kinds of charge flows happen before, during, and after tornadoes pass. It's tough taking data in such extreme conditions, of course. Most of your test equipment is going to be destroyed by high winds and flying 2x4s, and being a storm-chaser is quite a dangerous trade. Not one for me. You'll find me cowering in the center of the ground floor of whatever sturdy building is available. I've seen the destruction firsthand too many times, whole houses splintered to matchsticks.
But a better understanding of how tornadoes form and develop could help us figure out a way to prevent or ameliorate the kind of disasters we tend to see every Spring and Fall across the middle of the US. That would certainly be something worth doing.
My reading on atmospheric phenomena is 10 or 20 years out of date. So the next step is to read up on the latest. I'll update this entry according to what I find.
Additions: One observation so far that seems to fit is this. "More often than not, overall cloud-to-ground (CG) lightning activity decreases as a tornado reaches the surface and returns to the baseline level when the tornado lifts." I got this from Wikipedia, so I'm going to follow up on the sources it cites. More coming.