In rough terms, tornadoes are not hard to understand. The updraft in the thunderstorm creates a low pressure beneath it, causing a massive inflow. Air responding to a low pressure will tend to converge straight toward the low pressure from all directions, with the kinetic energy falling off with the square of the distance from the source of the low pressure. But if external factors offset the inflow, the air will spiral inward. The centrifugal force that emerges will then oppose the low pressure. This means that the low pressure has to get its air from somewhere else. As the low pressure and the centrifugal force are in equilibrium, the pressure at the center of the vortex equals the pressure at the source of the low pressure. In essence, the low pressure has been "piped" through the vortex. At the end of the "pipe," if the air is not spiraling inward, then we go back to the simple case, where air converges in a straight line, and the kinetic energy falls off with the square of the distance from the end of the pipe. But if the air flowing toward the end of the pipe is also spiraling inward, then once again the centrifugal force opposes the low pressure, and the low pressure has to be satisfied by air from elsewhere. In this way, a low pressure can extend a great distance away from its source through a vortex, where the only limiting factor is the very slight amount of friction resulting from the rotation, dissipating the energy as we move away from the source.

The hard part is understanding why tornadoes rotate so robustly at the surface of the Earth, considering the vast amount of friction encountered at the solid boundary. A vortex 350 m tall, 35 m across, and rotating at 45 m/s, will only lose about 1,000 watts of power due to friction in the air. But the same vortex will lose about 1,000,000 watts of power to skin friction at a solid boundary. So it will take 1,000 times more energy to rotate the vortex at the surface than it takes in the entire 350 meters above the surface. And in a free vortex, there is no source for such energy. Energy dissipates away from its source due to friction — it doesn't increase by 3 orders of magnitude when it hits an opposing force. So nominally speaking, the air speed at the surface should be less than ¹⁄1,000 of the speed above the surface. If the air speed 350 m above the surface is 45 m/s, the air speed at the surface should be less than .045 m/s. In other words, the air at the surface should not be moving at all. And with no movement, there will be no vortex. In actuality, no single force dictates the final answer — all of the forces present will have their effects, and we have to add it all up to get the eventual result.

• The low pressure moves air inward.

• An external factor offsets the air, initiating a cyclonic inflow.

• Skin friction at the solid boundary greatly reduces the air speed at the surface. If the surface is perfectly smooth, the effect will be limited to the air directly at the surface, but an uneven surface will create turbulence that will extend the effect well above the surface.

• The reduction in air speed reduces the centrifugal force, and the air flows more directly toward the low pressure.

• Without the opposition of the centrifugal force, the low pressure is relieved at the boundary.

• Less low pressure means less centripetal force, resulting in a widened vortex at the boundary, with slow rotation.

• Once the air enters the vortex and begins its ascent toward the source of the low pressure, the pressure will decrease with altitude.

• As the pressure decreases, the centripetal force increases, which reduces the radius of the vortex as the air gets nearer to the source of the low pressure.

The following images clearly illustrate the behavior of a free vortex when encountering a solid boundary. Note the turbulent flow at the surface, which resolves into a wide vortex above the surface, which then tightens into a narrow vortex in the direction of the flow. (See this for a high-resolution image of a vortex at the same location as in Figure 1.)

Tornadoes, on the other hand, do the exact opposite. They start out with a small radius at the surface. Then the radius expands, in the direction of the flow, sometimes to the point that the vortex disintegrates into a turbulent flow. So the lowest pressure, and the fastest wind speeds, are at the surface, where the friction is the greatest. As the air approaches the source of the low pressure, and where there is no skin friction, the centripetal force relaxes, and the air slows down. This doesn't make sense.

Furthermore, when a free vortex encounters an obstacle, it simply reorganizes elsewhere, since fluids always follow the path of least resistance. And yet a tornado is not perturbed by obstacles, even when the obstacle is larger than the tornado itself. Either the tornado maintains its general form while riding over the obstacle, or the tornado removes the obstacle. Since the amount of friction goes up quite dramatically when obstacles are encountered, a vortex that is relatively unperturbed by obstacles is inexplicable in fluid dynamic terms.

Figure 7 represents the airflows in free vortexes compared to tornadic vortexes. Aside from the fact that they are both rotating columns of air, they are different in every respect.

Clearly, other factors are present in tornadoes, which make them fundamentally different from free vortexes. So what is the nature of those factors?

To actually get laboratory airflows that match those in tornadoes, other forces have to be introduced. The best work done to date was with an apparatus similar to that depicted in Figure 8.

Figures 9 and 10 show the results, using different "swirl ratios" (i.e., the amount of angular momentum imparted into the flow before it passes through the hole).

In the 1^st panel of Figure 9, a small amount of angular momentum at the base creates a perfectly straight, laminar vortex. In the 2^nd panel of Figure 9 (and also in Figure 10), with a larger swirl ratio, we see a phenomenon known as "vortex breakdown." With a higher degree of angular momentum imparted into the vortex by the louvers in the base of the apparatus, the air is subjected to more friction as it moves through the stationary air in the central chamber. The friction reduces the angular velocity in the vortex, and the laminar flow becomes prone to turbulence. The turbulence then allows the surrounding air, not subject to any centripetal force (because it is not rotating) to flow downward into the vortex, seeking the extreme low pressure at the base. A "downdraft" inside the vortex relieves the low pressure, and thereby reduces the centripetal force. This results in the rapid widening of the vortex just prior to its breakdown. Note that even in tightly-controlled conditions, this configuration is extremely unstable. In the 3^rd panel, with an even higher swirl ratio, the vortex breakdown occurs at soon as the air exits the hole. And in the 4^th panel, the turbulence is so robust that it shrouds the vortex.

Note the similarity between the vortexes in the 2^nd, 3^rd, and 4^th panels and the following photographs of real tornadoes.

Hence if there is a force that can oppose the low pressure, and create a bottleneck in the flow, an extreme low pressure develops, away from the source of the low pressure. In the direction of the flow, the air speed relaxes, and in extreme cases, the laminar flow gives way to turbulence. Without such a bottleneck, only a normal, free vortex is possible, with little or no rotation at the surface. In the apparatus depicted in Figure 8, the bottleneck is created by a piece of plywood with a hole in it, which holds the air down until it achieves the centerline, where suddenly it can respond vigorously to the low pressure. But how could such a bottleneck be created in nature?

There is only one other force operative in the atmosphere, so it's the only candidate: electromagnetism. Then the question becomes: how does electromagnetism hold the air down until it gets to the vortex, and how is that force removed such that the air can finally respond vigorously to the low pressure? In other words, if we were to build something that had the effect of a piece of plywood with a hole in it, using only electromagnetic forces, how would we do it?

Electromagnetism is composed of two parts: the electric force, and the magnetic force. We can eliminate the magnetic force as a possibility, because air is only infinitesimally responsive to magnetism. That leaves the electric force. So how could electric charges attract the air to the surface?

If the air is bearing an electric charge, it will induce an opposite charge in the Earth, and then there will be an attractive force between them. Since the Earth has more mass than the atmosphere, the Earth will stay where it is, and the air will be drawn toward the surface.

Then the question is: how much charge would it take to overpower how much low pressure, in order to keep the inflow at the surface, establishing such a bottleneck?

First let's consider the force of the electric field that is pulling the air toward the ground. The number of charged particles in the tornadic inflow has been estimated at one part per billion (2.14 · 10¹⁴ charged particles/m³), and the charge per particle has been estimated at 3.2 · 10^-17 C. This means a space charge of 0.0068 C/m³ in the tornadic inflow. In an electric field of 5 kV/m, this yields 27 N/m³ of force.

Next we'll look at the low pressure drawing the air toward the mesocyclone. At 20 °C, air weighs 1.2041 kg/m³, which means 11.8002 N/m³ of gravitational force. The lowest pressure ever recorded anywhere in a tornadic system was 100 mb below ambient, which means a 10% reduction in density. 10% of 11.8002 means a loss of 1.18002 N/m³ of gravitational force. Put another way, that's 1.18 N/m³ of buoyancy.

Finally, we can clearly see that if the downward force is 27 N/m³ and the upward force is 1.18 N/m³, the air will stay at the surface until the electric charges are neutralized.

If the tornadic inflow is charged, can we determine the sign?

Since the tornadic inflow is clear, we know that it does not contain liquid or solid water particles (i.e., rain, sleet, or hail) — it's all in the gaseous state. There are three different molecules abundant in the air: N₂, O₂, and H₂O. Of these, only the H₂O is stable with a net negative charge. At 100% relative humidity, H₂O represents 1% of the air by volume. Actually RH readings from tornadic inflow are more like 20%, meaning that the H₂O is something like 0.2% of the air by volume. It's hard to believe that a negative charge distributed in 0.2% of the air would exert a force more powerful than the low pressure and the friction acting on the air. But all matter, including N₂ and O₂, can become positively-charged, as any molecule can have a deficiency of electrons. Hence it is reasonable to proceed on the assumption that only a positive charge could be distributed throughout enough of the molecules in the air to exert a powerful body force.

So we have a low pressure under the thunderstorm's updraft, that is drawing in air from all around. We have an offset in the inflow that is inducing rotation. And we also have a positive charge in the inflow, which is attracting the inflow to the surface. Hence the air would otherwise curve gracefully upward into the updraft (whether or not it is also spiraling inward), but because of the electric force, it sticks to the surface. Friction at the surface then creates the bottleneck, resulting in an extreme low pressure, and high wind speeds at the mouth of the vortex.

So how does the inflow ultimately break away from the surface and ascend within the tornado? Does the low pressure finally overpower the electric force? In other words, is the electric force a leaky piece of plywood, that will allow the air to be pulled away if there is a sufficient degree of low pressure? Or does the EM plywood have a hole in it? The robustness of the inflow strongly suggests that somehow the electrostatic attraction is eliminated inside the vortex.

Freeing the air from its electrostatic attraction to the Earth would necessitate that the charge be neutralized. Neutralizing a positive charge would, of course, take electrons. So we need a flow of electrons down from the cloud and through the tornado, that will neutralize the positive charge, freeing it from its electrostatic attraction to the Earth. And there is, in fact, just such a flow of electrons. An average tornado has an electric current in the range of 100 ~ 250 amps, and this has been confirmed by several methods.

Indeed, this electric current led researchers in the 1960s to believe that tornadoes were discharge vortexes. And while they could prove the existence of the current inside the tornado, they couldn't explain where it went, since only a small percentage of the current can be detected in the Earth. They also could not explain why these discharge vortexes show no preference for highly-conductive features on the surface, such as rivers & streams, railroad tracks, etc., which any electric current would certainly do. This left the researchers with more questions than answers, and eventually, the entire electromagnetic paradigm was abandoned.

Now we can piece it all together. The current is present inside the tornado, but the electrons are not flowing into the ground. Rather, the electrons are flowing into the positively-charged air. So we shouldn't expect large telluric currents, nor any preference for conductive features in the surface. And this current explains the extremely robust inflow to the tornado, despite the enormous amount of friction at the surface — the electric current is releasing the inflow from its attraction to the surface. So it's the hole in the plywood.

The only remaining question is: why hasn't this powerful positive charge in the tornadic inflow been detected?

It's possible that it has been detected, but without the present hypothesis in hand, the data would not have been interpreted as a space charge in the tornadic inflow. The principle instrument for studying electromagnetism in the atmosphere is the electric field meter, which measures electrostatic potentials. Under a supercell thunderstorm, the field meters do typically show a positive charge aloft. This is unusual for a thunderstorm, because the main positive charge region in the cloud is at the top, while the negative charge carriers in the storm (rain, sleet, and hail) tend to be closer to the Earth. But this has not led researchers to conclude that the tornadic inflow is charged. The assumption is that all of the charge is inside the cloud, and that a field meter at the surface will measure the potential between the charges in the cloud and an induced opposite charge in the Earth. If we were to measure the space charge in the air just above the surface, we would expect to find it opposite in sign to the induced charge in the Earth, but this does not indicate that the air itself was already bearing an electric charge. We would call it just an artifact of charge separations in the presence of an electric field. So space charge studies are not regularly performed.

Still, instinct tells us that if there is a body force acting on the air that is binding the inflow to the surface, and if that force is electric, a respectable percentage of the molecules would have to be charged. Otherwise, the few that were charged would have little effect on the rest of the air. And if just a single-digit percentage of the air molecules are positively-charged, we would expect the electric field to be enormous. Yet the electric field under a supercell is relatively weak by thunderstorm standards. So we must return to the data, to find what we're missing.

There is nothing in the data from tornadoes over land that suggests an answer, but photography of waterspouts does. In Figure 14, we immediately recognize the cyclonic pattern as evidence of inflow to the vortex, which is what we would expect. But darker water means faster winds, so this is evidence of a discrete channel of air flowing into the vortex, and this is not what we would expect.

In fluid dynamics, channeling is evidence of differences in viscosity. If all of the air has the same viscosity, it is all subjected to the same friction. Any air moving faster will experience more friction, so we expect a self-regulated consistency in the inflowing speed. But if some of the air has a lower viscosity, it will experience less friction, and therefore it will tunnel through the higher-viscosity air.

So then the question is: what are the conditions necessary for viscosity differences?

We might think that the air is warmer, as fluids are generally less viscous at higher temperatures. Yet gases actually get more viscous with temperature, though in the relevant range (20~30 °C) the difference is slight.

Since we can expect the air to be relatively homogenous, there is only one other factor that could affect the viscosity: its electric charge. Electrostatic repulsion in charged air prevents the particle collisions that instantiate friction, thereby reducing the viscosity.¹

And this, then, explains why we're not seeing a huge positive charge under the thunderstorm. If all of the charge is concentrated in a discrete channel, we wouldn't expect a powerful field everywhere under the cloud — only inside the channel will the field be unusually powerful.

So why haven't we detected an unusually powerful electric field in the banded inflow?

First, not every electric field study also includes airflow instrumentation, in which inflow bands would be detected, and the correlation between electric fields and airspeeds could be seen.

Second, electric fields under a thunderstorm fluctuate dramatically, as charges shift inside the cloud, and as such charges are neutralized by lightning. While the tornado is active, the thunderstorm typically issues several lightning strikes per minute, which means that in the period of time that an inflow band is passing over an electric field meter, the electrostatic potential will fluctuate at least a couple of times due to lightning strikes, making it more difficult to see a correlation.

Third, to the extent that some correlation between inflow bands and electric fields has been observed, it is typically attributed to triboelectric charging in the particulate matter that is creeping or saltating in the inflow band. The particulate matter itself has not been studied — it is merely assumed that any difference in electric field that was directly proportional to air speed would most likely be due to static electricity, because until now, no one proposed that the air itself was charged.

What we need is a space charge study that will incontrovertibly identify differences in electric charges in the air itself, inside and outside the inflow band. And that study has not been conducted. So absence of evidence in support cannot be cited as evidence against in this case.

And what we do have is plenty of evidence of a force that is not fluid dynamic, that can only be electromagnetism, and that behaves exactly as electromagnetic principles predict.