Live Q&As

The best place to take refuge is below-ground, in a basement or dedicated storm shelter if possible. Above ground, a small room (such as an interior closet) that is away from windows and outside walls that may be penetrated by flying debris offers the best protection. Security-cam video from last week's tornado strike at the St. Louis airport last Friday evening, when a concourse at Lambert Field lost its roof and was turned into something resembling a wind tunnel filled with shards of flying glass, is an excellent illustration of why a large open space where winds can accelerate and carry heavy projectiles or area near glass windows can be extremely dangerous.

While tornado warning lead times are gradually increasing (now up to 13 minutes lead on average), we have a persistent problem with false alarms--warnings issued but with no actual tornado ever touching down. This problem is rooted in detection methods that emphasize mid-level rotation identified by Doppler radars located far apart, as opposed to bona fide low-level rotation at the ground. What's more, the exact processes that focus broad thunderstorm rotation (termed a "mesocyclone") down to the tornado scale are still not well known, and a subject of active research supported by the National Science Foundation. I'm sorry that this warning uncertainty is a challenge for your situation. My best advice is that, on days where very severe weather is forecast and "watches" are issued earlier in the day, that you try to move to a safer place (perhaps with a friend) for that general period of time.

Most members of the scientific community agree that global climate change is occurring. Some of the most profound and obvious effects are initially appearing at higher latitudes (i.e. over Earth's polar regions), which are well removed from the more temperate midlatitude zones where severe thunderstorms are most prone to occur. As the science stands today, we cannot identify any clear causal link between a warming climate and increased numbers or violence of tornadoes. Why is that? First, tornadoes are relatively rare and ephemeral events. Any long-term changes in tornado activity, if such trends are indeed occurring, are exceedingly difficult to tease-out of imperfect historical records. Maps of tornado activity are relatively accurate today, but become increasingly sketchy as one looks back more than a couple of decades. Significant changes in patterns or reported numbers of tornadoes shown in these maps can be largely interpreted as resulting from shifts in U.S. population, as well as impacts our ongoing communications revolution. After all, someone has to be present to see a tornado or experience its damage, and then have the ability to report that event for it to be recorded for posterity.

The pattern of events that you describe, in which the tornado is adjacent to an area of heavy rain and hail, is a common one but by no means the only way that tornadoes occur. Occasionally, and I would say more frequently across the southern U.S., tornadoes are ?rain-wrapped? and there would be no such let-up in precipitation before high winds strike. As such, my best advice is to keep your ear to the media or your local NOAA Weather Radio if that is available in your area. When in doubt, move to a place of shelter and remain there until the major threat has passed.

To be sure, your region (Tennessee, and points south/west as well) has been pummeled this month. Climatologically speaking, the most active month for tornadoes is May, although this April (during which we have already seen 300+ confirmed tornadoes, an apparent record for the month) has been exceptional. I would like to say that it's unlikely May will continue to be this remarkably active, but my best advice is stay tuned...

I already addressed the large-scale weather pattern that is providing the support (or "cause") for these storms. As to other regions that may be effected, in the short term the severe risk will be shifting eastward toward the eastern seaboard/mid-Atlantic region tomorrow. Current indications are that severe weather will not be as intense or widespread tomorrow morning as this front sweeps eastward across the DC region, but isolated severe reports (mainly high winds, hail and of course heavy rains) may still occur up and down the I-95 corridor. More generally, as spring advances into summer, the band of maximum thunderstorm (and hence tornado) activity tends to shift northward. By the time we reach August, central regions of the U.S. (our "tornado alley") will most likely experience oppressive heat and less-severe "airmass" type thunderstorms, while the Prairie Provinces of Canada will see the most severe weather. That is, assuming patterns progress in the usual way?

Tornadoes do occur on several other continents--across southeastern Australia for example, but the particular juxtaposition of terrain and water in the U.S. provides a uniquely favorable environment. In particular, the proximity of a major north-south mountain range (the Rockies, which favors development of dry midlevel air that tends to fuel very strong downdrafts) and the Gulf of Mexico (an ever-present source of energy in the form of moisture that condenses to release heat and drive strong updrafts), all in a latitude range frequented by strong upper-level jetstreams amounts to something of a "perfect storm" for severe (supercell-type) thunderstorm formation.

Tornadoes are comparatively rare and ephemeral events at any given point on the map, even in the heart of tornado alley. At one time or another, they have been recorded every state in the Union. I might add that, while I?m not sure what part of Northern California you lived in, even the central valley of CA has experienced these storms. Tornado numbers and distributions also exhibit remarkable variations from year-to-year. While New Mexico is not frequently hit, that does not mean it cannot be.

For the most intense (and therefore rare) "EF-5" class tornadoes, the only way to dependably survive a direct hit is to be below ground--preferably in a reinforced tornado shelter. Barring that, your best bet is to be in a small, interior room (closets are ideal) far away from windows and exterior walls that may be penetrated by flying debris. I will quickly add that an automobile is one of the *worst* places to be in a tornado.

Yes, absolutely. The core of strongest jet-stream winds provides a key source of wind shear (the tendency of winds to turn and increase with height). What's more, these winds are to some degree carried earthward (in intense downdrafts) and appear to play an important role in focusing rotation on a broad scale (the "mesoscyclone" that may be 10 or more miles across) into a tornado-sized vortex. Google "rear-flank downdraft" if you want to know more about this process.

I appreciate your sense of humor. It's certainly true that scientific knowledge (and hence expert advice) has evolved over the years and will continue to do so. Remember, even below ground being in a small enclosed space provides your best protection from falling debris.

The standards for issuing tornado warnings by the National Weather Service (namely an expectation that tornadic winds at ground level are imminent or at least very likely to occur within a short period) have not changed, but our ability to detect broader thunderstorm rotation (usually via Doppler weather radar) that points to this elevated risk has. As such, the number of warnings has indeed climbed, as has the average lead time (now 13 minutes or better) between warning issuance and ultimate tornado development, when such actually occurs. Unfortunately, as you note, this has come at the expense of a high "false alarm" rate and that is something we at the National Science Foundation are hoping to address through active research. The Verification of the Origins of Rotation in Tornadoes Experiment, phase 2 (VORTEX2, or 'V2' for short) represents a $9M investment by the National Science Foundation over a period of four years (2008-11, but ongoing) to improve our understanding of the dynamics of tornado formation. V2 involved a total of 10 weeks of data collection (5 weeks each during 2009 & 2010) involving 10+ universities and several non-profit organizations. A total of 40 instrumented vehicles, a roving armada of sorts, were used to construct a highly mobile mesh of sophisticated observations of storm structure and surrounding atmospheric conditions in an effort to better distinguish those rotating "supercell" thunderstorms that produce tornadoes from the vast majority that fail to do so. You can learn more about this exciting project at: http://www.vortex2.org/ Also, an IMAX film (entitled "Tornado Alley") profiling activities in VORTEX2 has been released. I believe it will be coming to the theater at Smithsonian Air & Space this fall, and is already playing at many spots around the country.