Discussion on the TF-Scale.

[Stuart Robinson]

I wanted to open a debate (not a can of worms) on the American F-Scale Vs the European TF-Scale. As some of you may know over in Europe we rate our tornados on the 10 point TF-scale (Meaden 1975) as apposed to the 5 point F-Scale (Fujita 1971).

Could you take some time to read the following slide show that was presented to the 2004 European Conference on Severe Storms.

http://www.torro.org.uk/TORRO/ECSS_Slide_S...ide%20show.html

And some further background to the TF-Scale can be found here
http://www.torro.org.uk/TORRO/severeweathe...scaleorigin.php
http://www.torro.org.uk/TORRO/severeweather/tscale.php

I suppose that my question is? Is it now time to re think the F-Scale and would the US ever adopt the TF-Scale.

I should point out that I am a Executive member of TORRO (I hold the electronic records for the UK Tornado database) Therefore I am biased towards the TF-Scale.

 

[Simone Lussardi]

Nice attempt. I hope there will be a close follow.

By the way, for reference, the only probable known T11 ever estimated by TORRO in Euope was in Italy.

July 24th, 1930. Stayed on the ground 84 minutes, for a path of 80 km. Moved at 87.5 km/h. Intensity T10-T11.

Data from http://www.tornadoit.org/tornadostorici.htm , adapted in english.

:wink:

 

[Aaron Kennedy]

This sums it up:

The scale is quite precise. Not only is it well-suited for the more accurate methods of tornadic wind speed determination, but also for rating weak tornadoes which account for the vast majority of global events.

ROFL

How many tornadoes do we have information for that could provide an accurate assesment of wind speed at the surface?

99% of the cases will be based off damage surveys... and now your scale is simply a damage scale that tries to attach wind speed scales to it.

Care is given to the rating of slow-moving tornadoes, because they will create more intense damage than a faster-moving tornado of the same wind speed.

Maybe. So what do you classify as a slow moving tornado? Once again subjectivity enters the equation.

As it stands now... there are three types of tornadoes (in my eyes). Those that do little damage.... those that make places uninhabitable, and those that lay waste to everything in their path. I don't care if a tornado is a F4 or F5... both are intense tornadoes.

Aaron

 

[Robert Dewey]

I don't really like the F-scale to begin with, so I am already biased. But, with that said, I think the Torro scale is a bit too precise.

 

[Thomas Loades]

I think the Torro Scale puts a little too much reliance on wind speeds, when they don't always correlate neatly with the damage observed.

Also — going by their rating descriptions — T8 and T10 tornadoes do the same thing to frame homes:
T8: “Wooden-framed houses and their contents dispersed over long distancesâ€￾
T10: “Entire frame houses and similar buildings lifted bodily from foundations and carried some distances.â€￾

How do you tell the difference?

 

[Jason Toft]

Not knowing much of anything, but having an opinion, I would say that it would be good to adopt a rating system that is more spread out, like the TF scale.

The only problem would be the lowest ranks would be more guesswork than actual science because it would be hard to measure those small tornadoes because they last for such a short time usually.

(I hope the people here consider me a decent contributor!)

Jason

 

[B Ozanne]

To answer the question...I don't see the F-scale going away anytime soon. Classifying a tornado in a 10-20mph range seems so arbitrary. Velocoties vary so greatly, and the current methods for estimating wind speeds are not that accurate.

Not to be arrogant or anything, but America rules the tornado world. We have the most tornadoes, we have the most research money, and we have the best terrain to study them. I don't see us adapting a European system.

Was it me or was that article full of anti-American rhetoric? I think it is an interesting system, but I was a little turned off by the tone of that presentation.

 

[Robert Dewey]

There is one MAJOR flaw with the F-scale system, and that is the fact that it rates tornadoes with no damage and F-0. It should simply be posted as "unknown"... When reviewing case studies, one can be thrown off by the fact that all of the tornadoes were rated F-0 or F-1 within a "perfect setup", yet in real life, they were all monster wedges. It also heavily skews climatological data...

But as I stated above, I don't really support the Torro scale either.

 

[B Ozanne]

On the other end of the scale though poor construction techniques can mislead people into thinking the tornadoes were much stronger than they really were.

It doesn't take much to blow a house clean off a concrete slab when its only heald down by a few bolts.

 

[Robert Dewey]

Originally posted by B Ozanne
On the other end of the scale though poor construction techniques can mislead people into thinking the tornadoes were much stronger than they really were.

It doesn't take much to blow a house clean off a concrete slab when its only heald down by a few bolts.

True, but when a tornado doesn't even hit a structure, there is nothing to base the wind speeds on. Instead of giving it a low rating, they should just mark it as "unknown", "no damage", or something similar. At least when the tornado hits a really weak house - and makes the tornado appear alot stronger than it actually is - they can at least conduct some kind of structural engineering test (i.e. they at least have somewhere to start). If it hits nothing but trees, they don't even have a place to begin.

I'm not real smart when it comes to wind engineering and the F-scale, so my comments may seem somewhat redundant.

 

[Darren Addy]

I've read over the material and have a couple of observations. Firstly, I appreciate the scientific inquiry and hope that some reasonable discussion can be had without bringing stupidity into it. (WX-CHASE discussion started off with political Europe vs U.S. political crap). :roll:

It seems that everyone agrees that the whole system of classification is subjective and accessment must be done after the fact. Fujita attempts to classify tornadoes based on damage. As pointed out, wind speed estimates are a part of that, but are only part of the picture. How wide was the damage path? How fast was it moving? Ect.

As a newbie, I was taken aback when I read (in Tim Vasquez's Weather Forecasting Handbook) the top 10 myths, number 8:

\"TORNADO INTENSITY IS MEASURED WITH THE FUJITA SCALE.\" The Fujita Scale is a damage scale, not an intensity scale.

I did not realize that. I think that much of the perception damage was probably done by movies like "Twister" where people are describing a tornado visually as an "F2" of "F3".

The Tscale attempts to measure intensity not just damage, which is admirable, but I don't think a great case is made for how well it really does that. Attempting to extend the Beaufort Scale and basing it on that is not a bad thing, but is really no less arbitrary than any other method (IMHO). As was stated at http://www.torro.org.uk/TORRO/severeweathe...scaleorigin.php:

\"The introduction of anemometers led to the necessity for a scale of equivalents between Beaufort numbers estimated by experienced observers and the velocity of the wind in miles per hour.\"

The problem (as observed in the material) is that anemometers don't survive most tornado force winds and experienced observers can't very well estimate the velocity of tornado winds. So it is all an attempt to approximate based on data after the fact. Since the rationale for the Beaufort scale breaks down at higher levels, basing a system on it and extending it to higher levels seems arbitrary (particularly when, as it has been noted, the scale becomes too fine for measuring something so subjective. I feel like it should say: "Train car blown 50 meters: T6. Train car blown 100 meters: T7." )

It seems to me that the Tscale puts the emphasis on measuring/estimating wind speed (and calling that a "true tornado intensity scale") while Fujita recognizes that a wind speed estimate is simply a byproduct of a damage estimate. While the Fujita scale is arbitrary and could use revision, I'm not sure I agree with the premise for the Tscale. In fact, I think a "true tornado intensity scale" has to take into account the width of the damage path. Is a tornado that produces 175 mph winds within a 500 yard path equal in "intensity" to one that produces 175 mph winds in a 1/2 mile wide path? Not in my mind.

PS... there is a typo in the TABLES RELATING THE TORRO AND FUJITA SCALES on this page. Under "Fujita Scale vs TORRO Scale" the F2 TORRO number should be 3.70 (not 13.70), I believe.

Darren Addy
Kearney, NE

 

[Darren Addy]

Originally posted by rdewey
But, with that said, I think the Torro scale is a bit too precise.

I would argue that it isn't any more precise - just uses a more (too) finely graded scale and the only reason that it does so is because it has to when it ties itself to the Beaufort scale.

Problem is that the Beaufort scale uses hard data (anemometer readings) while tornado measurement cannot. So it attempts to "jump the tracks" to the wind speed estimates based on

# visiting the tornado damage site to make non-engineering assessments
# obtaining an engineering assessment of the damage
# using Doppler radar
# applying photogrammetric analysis, and
# directly measuring observed tornadoes

and then attempting to extrapolate the data and convert it back into a Beaufort scale number.

Basing the system on the Beaufort numbers (which progress nicely) gives the illusion of accuracy at tornado velocities, but ignores the disconnect in the data collection along the way (the way the number is arrived at).

Darren Addy
Kearney, NE

 

[Luca Ronca]

Originally posted by B Ozanne
To answer the question...I don't see the F-scale going away anytime soon. Classifying a tornado in a 10-20mph range seems so arbitrary. Velocoties vary so greatly, and the current methods for estimating wind speeds are not that accurate.

Not to be arrogant or anything, but America rules the tornado world. We have the most tornadoes, we have the most research money, and we have the best terrain to study them. I don't see us adapting a European system.

Was it me or was that article full of anti-American rhetoric? I think it is an interesting system, but I was a little turned off by the tone of that presentation.

I agree. The Fujita scale is now a standard and should be so for all chasers and scientists and not only in the US. The world is going to be smaller now and we always attempt to create a new standard.... In Europe some scientists make a big work only with words not fact. In the research fields of science we have a lot of people that go out of Europe; funds are mirages, and where they go? In the US.
I'm Italian and I see every day a bad information about weather, we dont'have a nationwide radar coverage, and some national institutes don't give free access to basic informations.
In America I feel more action, people want to do something and they do; here in Europe we have great scientist but we are far away from conclusions :cry:
Sorry if I made some mistakes in writing
Ciao
Luca

 

[Stuart Robinson]

I am glad that a sensible debate has evolved out of this thread and really I do believe that as the boundaries of Tornado research pushes forward, a more finer resolution of tornado scale rating will be required, if not a must, if we are going to improve and verify our forecasts and models worldwide.

Aaron mentioned that really there were only 3 real tornado ratings which I suppose is all that would be required from a news reporting / insurance point of view but from a world wide scientific point of view we need more scale resolution. Even in the USA a vast majority of the tornados are rated less than TF-6. Given the 2005 reports received so far I don’t believe that there has been a F4 reported yet.

Looking at a wider picture (and that is where I am coming from) the F-scale has suited the USA for the last 30 years but it is not really any good for the worldwide study on tornados and research where there are many eminent scientists are involved on this world wide work.

As far as I know tornados occur on every continent ( I am not sure of any reports from Antarctica however ) and now there is some evidence of vortexes occurring on Mars! So we need an international rating scale, one that we will all understand both as chasers and scientists ~ and one that will cater for the 1000 yearly tornados in the US, and all the tornadoes in Europe, Asia and Australia and wherever.

Finally Bill Mentions some comments about the tornado world. (Which we all live in – but an awful lot of it occurs outside of the USA) IMO the most destructive and powerful tornados on this planet are to be found over the Khulna, Bangladesh, Chittagong triangle of North East India where many many people are killed every year.

So we do need a international scale now, one that will cater for all vortex, even they occur with the US, Europe, Asia or even Mars! but it must be based on sound scientific and mathematical parameters.

 

[Steve Burrows]

We use the F-scale down here.

Just as a note New Zealand records the 2nd largest number of tornadoes a year behind America. These are typically f0-f1 although last year we did have an f-3 up North!!
I have yet to see my first!