Tuesday, April 05, 2011

How likely is an F6 tornado?

kw: analysis, tornadoes

Tornado season is upon us, or at least upon the central US. Though the stronger storms occur in July and August, any spring thunderstorm spawned by colliding air masses can produce a tornado. The more common thermal thunderstorms don't do so.

I got to thinking about the strongest storm possible, and I wondered if a tornado will ever be classified F6, or even greater. When Dr. Ted Fujita first devised the F scale, ranging from F0 to F5, he considered that an F6, which he called "Inconceivable Tornado", was probably not possible energetically. Thus the final scale stops at F5, which has no defined top wind velocity. In effect, F numbers greater than five were defined out of existence! Secondarily, the scale is based on the level of damage. Since an F5 typically leaves nothing behind but plowed ground, there is no way to determine after the fact whether the top wind speed was at the low end or high end of the "F5 window", or above it.

However, that hasn't stopped others from having a second look. The CSG Tornado Table returns to the original formula for the lowest wind speed for each F number:

V (mph) = 14.1(F+2)1.5

The threshold for an F6 by this scale is 318 mph (511 kph). These folks extend the scale to F12 with a wind speed threshold of Mach 1, or 738 mph (1188 kph). You can make wind like that in a wind tunnel, but nothing even close has been observed in the natural atmosphere.

Let's review the complex way extreme wind velocity is produced.

This image, from Jon Merage's Gallery of weather photos shows a multi-vortex tornado, probably an F3, at a very early stage, before it picked up enough dirt to obscure its structure. Three of the sub-vortices Fujita called "suction vortices" can be seen.

The whirling air in a suction vortex can reach 100 mph (160 kph) relative to the center of the vortex. In the most powerful tornadoes, the main vortex's wind speed is 100-150 mph (160-240 kph), so the outer edge of a suction vortex—in an F5 that is standing still—can reach 250 mph. But they don't stand still. A typical tornado moves along at about 30 mph (less than 50 kph), but they can zoom across the landscape at 60-70 mph (95-110 kph), so it is possible for all these to add up to very local wind speeds as high as 320 mph (515 kph).

With this in mind, it is just possible that perhaps two storms in history reached F6 velocities. Thus, while such storms seem not just possible, but may be historical, they are very, very rare. What keeps the lid on?

There is a literal lid on the power of tornadic thunderstorms: the thickness of the troposphere. A feature of the largest thunderstorms that illustrates this visually is the anvil, shown here in a photo from Spacelab (see this WeatherQuestions item for more info). It shows the top of a thunderstorm that has bumped up into the stratosphere and been forced to spread out. In the troposphere, where all weather occurs, temperature nearly always decreases with altitude. Above this, the temperature in the stratosphere rises with altitude, which prevents rising air masses from rising further. The rising, moist air masses that spawn thunderstorms get their energy from the cooling with altitude, and are weakened when they hit the stratosphere. A very strong rising air mass may punch a kilometer or two into the stratosphere, but that's about the limit.

Will global warming increase the chances for more F5 storms, and possibly some F6-scale tornadoes? It is not likely. Ground-level heating will be the greatest at high latitudes, and there will be little effect nearer the equator. If anything, the troposphere might increase in thickness farther to the north, at the expense of the atmosphere near the equator, thus making the "tornado belt" wider but less intense overall. I am not sure the northern Canadians and Siberians (and possibly the gauchos in southern Argentina) will welcome a push of tornado-potential weather to greater latitudes. But it will take a lot of warming to make a significant difference in the extent of each continent that can support frequent tornadoes.

F6: possible? Yes. Plausible? Less so. Based on history, about one tornado per century just might qualify.

No comments: