I) Cyclones versus tornadoes

I1) How are tropical cyclones different from tornadoes ?
I2) Why do tropical cyclones spawn tornadoes?
I3) Quel est le pourcentage de cyclones tropicaux pouvant générer des tornades ?
I4) What parts of a tropical cyclone are most favored for tornado formation?
I5) How long after landfall are TC tornadoes a threat?
I6) Are TC tornadoes weaker than midlatitude tornadoes?
I7) What is the largest known outbreak of TC tornadoes?
I8) What is the deadliest single TC-spawned tornado?
I9) What is the most damaging single TC-spawned tornado?
I10) Why are TC tornadoes especially difficult to deal with?
I11) How does the damage from hurricanes compare to tornadoes?


Subject I1) How are tropical cyclones different from tornadoes?

Contributed by Chris Landsea

While both tropical cyclones and tornadoes are atmospheric vortices, they have little in common.


Tornadoes have diameters on the scale of 100s of meters and are produced from a single convective storm (i.e. a thunderstorm or cumulonimbus). A tropical cyclone, however, has a diameter on the scale of 100s of *kilometers* and is comprised of several to dozens of convective storms. Additionally, while tornadoes require substantial vertical shear of the horizontal winds (i.e. change of wind speed and/or direction with height) to provide ideal conditions for tornado genesis, tropical cyclones require very low values (less than 10 m/s [20 kt, 23 mph]) of tropospheric vertical shear in order to form and grow. These vertical shear values are indicative of the horizontal temperature fields for each phenomenon: tornadoes are produced in regions of large temperature gradient, while tropical cyclones are generated in regions of near zero horizontal temperature gradient. Tornadoes are primarily an over-land phenomena as solar heating of the land surface usually contributes toward the development of the thunderstorm that spawns the vortex (though over-water tornadoes have occurred). In contrast, tropical cyclones are purely an oceanic phenomena - they die out over-land due to a loss of a moisture source. Lastly, tropical cyclones have a lifetime that is measured in days, while tornadoes typically last on the scale of minutes. For more information on tornadoes you can go to the Storm Prediction Center's FAQ maintained by Roger Edwards.

An interesting side note is that tropical cyclones at landfall often provide the conditions necessary for tornado formation. As the tropical cyclone makes landfall and begins decaying, the winds at the surface die off quicker than the winds at, say, 850 mb. This sets up a fairly strong vertical wind shear that allows for the development of tornadoes, especially on the tropical cyclone's right side (with respect to the forward motion of the tropical cyclone). For the southern hemisphere, this would be a concern on the tropical cyclone's left side - due to the reverse spin of southern hemisphere storms. (Novlan and Gray 1974)

Subject I2) Why do tropical cyclones spawn tornadoes?

Contributed by Bill McCaul

Tropical cyclones spawn tornadoes when certain instability and vertical shear criteria are met, in a manner similar to other tornado-producing systems. However, in tropical cyclones, the vertical structure of the atmosphere differs somewhat from that most often seen in midlatitude systems. In particular, most of the thermal instability is found near or below 10,000 feet altitude, in contrast to midlatitude systems, where the instability maximizes typically above 20,000 feet. Because the instability in TC's is focussed at low altitudes, the storm cells tend to be smaller and shallower than those usually found in most severe midlatitude systems. But because the vertical shear in TC's is also very strong at low altitudes, the combination of instability and shear can become favorable for the production of small supercell storms, which have an enhanced likelihood of spawning tornadoes compared to ordinary thunderstorm cells (Novlan and Gray 1974, Gentry 1983, McCaul 1991).



Subject I3) What percentage of tropical cyclones spawn tornadoes?

Contributed by Bill McCaul

Almost all tropical cyclones making landfall in the United States spawn at least one tornado, provided enough of the TC's circulation moves over land. This implies that Gulf coast landfalling TC's are more likely to produce tornadoes than Atlantic coast TC's that "sideswipe" the coastline. The rate at which TC's produce tornadoes (waterspouts) over the ocean is unknown, although Doppler radars have identified many cases where storm cell rotation suggestive of the presence of tornadoes was observed over water, and there have been a number of cases where TC-spawned waterspouts have been witnessed from shore, with some of these coming ashore as tornadoes (McCaul, 1991); see also the website:

http://www.tornadoproject.com/alltorns/tcyclone2.htm




Subject I4) What parts of a tropical cyclone are most favored for tornado formation?

Contributed by Bill McCaul

In the northern hemisphere, the right-front quadrant (relative to TC motion) of recurving TCs is strongly favored. In the southern hemisphere, the left-front quadrant presumably is favored, although there is little research on this point. Most of the tornadoes form in outer rainbands some 50-300 miles from the TC center, but some have been documented to occur in the inner core, or even in the TC eyewall (Novlan and Gray, 1974; McCaul, 1991).


Subject I5) How long after landfall are TC tornadoes a threat?

Contributed by Bill McCaul and Bart Hagemeyer

TC's may spawn tornadoes up to about three days after landfall. Statistics show that most of the tornadoes occur on the day of landfall, or the next day. However, many of the largest outbreaks have occurred two days after TC landfall, as the TC remnants interact with midlatitude weather systems. The most likely time for tornadoes is during daylight hours, although they can occur during the night too (McCaul, 1991).


Subject I6) Are TC tornadoes weaker than midlatitude tornadoes?

Contributed by Bill McCaul and Bart Hagemeyer

In general, it appears that TC tornadoes are somewhat weaker and briefer than midlatitude tornadoes. During the period 1948-1986, the percentage of TC tornadoes that reached F2 or greater intensity on the Fujita scale was 26% (McCaul et al., 2004), while during a roughly comparable period (1950-1976), the corresponding percentage for all U.S. tornadoes was 36% (Kelly et al., 1978)(Gentry 1983; McCaul 1991) .

In Florida, in particular, the most significant tornadoes tend to occur with "hybrid" cyclones or tropical cyclones with some hybrid influence. This usually means greater westerly shear in the storm environment which is believed to be favorable for stronger, long-lasting tornadoes. Hurricane Agnes in 1972 was a minimal category 1 hurricane with considerable hybrid influence and it produced the most F2 and greater tornadoes in a single day in Florida history (Hagemeyer 1997; Hagemeyer and Spratt, 2002).


Subject I7) What is the largest known outbreak of TC tornadoes?

Contributed by Bill McCaul, Chris Vaccaro and the National Hurricane Center
2004's Hurricane Ivan caused a multi-day outbreak of 127 tornadoes, with the bulk of the tornadoes on 17 September in the mid-Atlantic region, some two days after Ivan's landfall in Alabama. State-by-state tornado counts from Ivan include Florida with 22, Georgia 25, Alabama 8, South Carolina 7, North Carolina 4, Virginia 40, West Virginia 3, Maryland 9, and Pennsylvania 9. There were 26 tornadoes on 15 September, 32 on 16 September, 63 on 17 September, 2 on 18 September, and 4 on 19 September. At least 7 people were killed and 17 injured by these tornadoes.

The previous record was during Hurricane Beulah, which spawned a reported 115 tornadoes in southeast Texas during the first several days after its landfall in September 1967 (Orton, 1970). Frances of 2004 is close behind in third place, with 106 tornadoes, and Rita of 2005 is in fourth place with 92. For a list of the top 25 tornado producing TCs, see the website:

http://www.tornadoproject.com/alltorns/tcyclone2.htm

While it is difficult to predict which TCs will produce large tornado outbreaks, there is evidence suggesting that the likelihood of a major outbreak increases for TCs that are large, intense, are recurving and entering the westerlies, have forward speeds from about 8-18 mph, and are interacting with old, weakened frontal boundaries. In addition, the TC's right-front quadrant must receive significant exposure to land, and this strongly favors TCs making landfall on the Gulf coast as opposed to those grazing the Carolinas (McCaul, 1991; McCaul et al., 2004).


Subject I8) What is the deadliest single TC-spawned tornado?

Contributed by Bill McCaul

One of the tornadoes spawned in October 1964 by Hurricane Hilda killed 22 people in Larose, LA ( Novlan and Gray 1974).




Subject I9) What is the most damaging single TC-spawned tornado?

Contributed by Bill McCaul

One of the tornadoes produced by Hurricane Allen in 1980 did about $50 million damage (1980 dollars; about $127 million damage in 2005 dollars) in the Austin, TX, area. More recently, Hurricane Cindy spawned a strong tornado that damaged the Atlanta Motor Speedway and other nearby areas to the tune of some $71.5 million in July 2005.


Subject I10) Why are TC tornadoes especially difficult to deal with?

Contributed by Bill McCaul

TC tornadoes are often spawned by unusually small storm cells that may not appear particularly dangerous on weather radars, especially if the cells are located more than about 60 miles from the radar. In addition, these small storms often tend to produce little or no lightning or thunder, and may not look very threatening visually to the average person. Furthermore, the tornadoes are often obscured by rain, and the storm cells spawning them may move rapidly, leaving little time to take evasive action once the threat has been perceived. ( McCaul et al. 1996, Spratt et al. 1997).




Subject I11) How does the damage from hurricanes compare to tornadoes?

Contributed by Chris Landsea

Even though winds from the strongest tornadoes far exceed that from the strongest hurricanes, hurricanes typically cause much more damage individually and over a season. (The strongest tornadoes - those of Fujita Tornado Damage Scale 4 and 5 - have estimated winds of 207 mph [333 kph] and higher, while the strongest hurricanes - those of Saffir-Simpson Hurricane Scale 4 and 5 - have winds of 131 mph [210 kph] and higher.) Hurricanes in the continental U.S. cause on average about $3 billion per landfall and about $5 billion annually (Pielke and Landsea 1998). The roughly 1000 tornadoes that impact the continental U.S.each year cause about ten times less - about $500 million in total ( Brooks and Doswell 2001). The top 30 most damaging hurricanes in the last 100 years (normalized to account for higher population, wealth and inflation) have each caused over $2.9 billion (Jarrell et al. 2001). In comparison, only the most damaging tornado in the last 100 years or so - if it hit today - would cause about $2.9 billion in damage: the May 1896 St. Louis tornado (Brooks and Doswell 2001).

Hurricanes tend to cause much more destruction than tornadoes because of their size, duration and variety of ways to damage items. The destructive circular eyewall in hurricanes (that surrounds the calm eye) can be tens of miles across, last hours and damage structures through storm surge, rainfall-caused flooding, as well as wind impacts. Tornadoes, in contrast, tend to be a mile or smaller in diameter, last for minutes and primarily cause damage from their extreme winds.

References:

Brooks, H. E., and C. A. Doswell, III, 2001: Normalized damage from major tornadoes in the United States: 1890-1999. Wea. Forecasting , 16, 168-176.

Jarrell,J.D., M. Mayfield, E.N. Rappaport, and C.W. Landsea, 2001: "The Deadliest, Costliest, and Most Intense United States Hurricanes from 1900 to 2000 (and other Frequently Requested Hurricane Facts)" NOAA Technical Memorandum NWS/TPC-1.