My suggestion would be to get a weather radio where you enter your home county and only receive the audible warnings when that county is included in a warning. (The NWS encodes the effected counties into every warning bulletin. County codes can be found here: http://www.nws.noaa.gov/nwr/indexnw.htm) That way you will get the general Tornado/Severe Thunderstorm Watch alert but not Tornado Warnings for counties 50 miles to the north or south. Cuts down on the “false alarms” for storms that have no chance of coming your way.

1. The fraction of a thunderstorm’s energy in the wind component (tornadic or non-tornadic) is small. In the energy budget of the thunderstorm, the change of phase of water is ~2 orders of magnitude larger.

2. The increase in tornado reports shown in the other graph is almost entirely in the F0 tornadoes. See this paper.

3. We know that, historically, the probability of a tornado occurring given that a storm is severe is almost entirely a function of the wind shear. I can’t get the image uploaded right now due to computer difficulties, but I’d be happy to e-mail it to you.

4. We know that the the active tornado years of the early 1970s had high shear and low Convective Available Potential Energy (CAPE), the simplest good measure of the thermodynamics, and that the inactive tornado years of the late 1980s had low shear and high CAPE.

I’m guessing there will be an increase in tornado frequency associated with global warming in the region north of the Gulf for the very simple reason that tornadoes are common components of the severe storms typical of the region, these storms are energetically born of warm gulf waters, and gulf waters are warmer with global warming.

However, there are a lot of other ways this could go. There may be a decrease in tornadoes and an increase in severity of straight line winds or just an increase in the number of tornadoes. Or this energy shift could simply all end up as extra hurricanes.

This graph in this post shows an increase in tornado frequency over the same time that we see an increase in various global temperature measures, at the decadal scale. But although the trends match, there is no short term positive correlation (there may even be a negative correlation).

Not surprisingly, this is complex and will be difficult to suss out.

Of all of the severe thunderstorm phenomena (tornado, wind, hail), non-tornadic winds are the most likely to increase since they tend to occur more often in low shear environments.