An F6 Tornado?

[Bob Hartig]

Originally posted by rdale
\"However, in the beginning, it clearly was conceived as scale that rated both wind speeds and damage, not just damage.\"

Where do you get that from? What part of his paper?

No paper, rdale. I'm an interested amateur, not a researcher or meteorology student. But my entire first paragraph explained my reasoning. I'll recap one part of it: If Fujita was concerned only with rating damage, then why, when he first conceived his scale, did he even bother to postulate ratings from F6 up to F12/Mach 1 that far exceed the point at which everything is blown away? From F5 on up, damage is so total that it's no longer a factor, so the only thing left to rate, at least in theory, was wind speed. You never see that extended scale now, of course--the scale has evolved through practical use in rating damage--but it was a part of the original Fujita scale, and to me it suggests a broader intention for the scale, at least originally, than damage measurement alone.

 

[rdale]

"If Fujita was concerned only with rating damage, then why, when he first conceived his scale, did he even bother to postulate ratings from F6 up to F12/Mach 1 that far exceed the point at which everything is blown away?"

He didn't... He simply divided the speed of sound by 12 and discared F6+. Don't know why. But he did not list F10 damage in his paper.

- Rob

 

[Justin Turcotte]

Doswell has some good comments on the F-scale posted in his "pet-peeves" site:
http://webserv.chatsystems.com/~doswell/pe...Pet_Peeves.html
Scroll down to #11.

 

[Ryan McGinnis]

Originally posted by Chris Lott
According to this, there will never be an F-6. This also shows wind speeds for each category, but clearly the F scale is a damage scale.

http://www.disastercenter.com/tornado/fujipear.htm

I disagree; it does not say that at all. Emphasis mine:

\"The maximum wind speeds of tornadoes are not expected to reach the F6 wind speeds. (words of Fujita)

Words of site author:

\"The scale was original created in 1971 by Dr. Fujita as a way of determining the strength of tornadoes from the damage that they caused.

In other words, damage is a means through which we determine "strength".

Like many of the contributions of Dr. Fujita, it was not widely accepted when first published.

Dr. Fujita and Dr. Pearson two years later published a paper that added in factors related to the width and length of the tornado path, and called the scale the Fujita-Pearson Scale. It was this work that caused the scale to gain acceptance.

The Storm Prediction Center now uses the greatly improved Fujita Scale to determine tornado strength from the damage that the tornado causes after the tornado.

Yet when a spotter or anyone else from the field makes a judgment about the size of a tornado without damage data, but on the width and length of the tornado path, they are making that judgment based on the size of the tornado on the ground, which is the Pearson method.

"Not expected to reach" is not the same thing as "there will never be". It simply means that given the research at the time, F6 strength was considered unlikely to the point where it was not expected to occur. Times change, and with them come new methods for measurement.

Lastly, the site you are quoting from is by no means an authority on the matter, any more than you or I are. One of the few people in the thread who's taken the time to go back and look at some of Fujita's papers has posted excerpts that suggest that Fujita meant for the scale to be BOTH a damage and a wind scale, and that he provided ways to indicate which method a researcher was using to arrive at the scale numbers (i.e., windspeed = uppercase F, damage survey = lowercase f). The argument can be made that the intent of the author is meaningless and that the only thing that matters is how the scale is popularly adopted and applied -- but if that is the case, then there really is no solid "right" and "wrong" to this discussion, because while the climo types definately still stick to damage surveys, it would seem that the bleeding-edge field researchers (we chasers don't count, unless we're scientifically measuring the windspeeds inside the vortex) are leaning towards allowing direct windspeed measurements as well.

It should be noted that the Saffir-Simpson scale and the Beaufort scale are also a measurments of "damage" (or effects), but the National Hurricane Center certainly doesn't wait for a damage survey before declaring a storm hurricane and giving it a Saffir-Simpson scale classification. This is because windspeeds can be reliably measured, or at least guesstimated, through things like dropsondes and in-flight wind instrumentation. Now that some people have possibly figured out how to do this with tornadoes as well *cough*Samaras*cough*, I see no problem with using that information to determine F-scale as well (especially since Fujita envisioned people doing just that and built something into his scale to reflect it) so long as the methodology being used is cross-checked and tested enough that we know the margin of error.

 

[Chris Lott]

You're right. Anything can happen. Back to one of my previous posts in this thread but with a little twist now, a tornado could literally wipe part of Oklahoma off the map for example. Clearly that would be F-5 damage(everything in it's path is destroyed) but if DOW was to clock a wind speed of 319 mph with this "imaginary" storm, I bet meteorologists and scientists and anyone else who could have say so in it would rate the storm an F-6, just because "technically" an F-5 only goes up to 318 mph. No more destruction, but a mere one mph more will get it an F-6 rating if it ever happens. But you hear so many times that after a storm has passed that they have to go assess the damage to determine how strong it was. Has there ever been a storm where someone rated a storm, maybe an F-3 but wind estimates proved otherwise?

 

[Charles Kuster]

I think we have had a discussion about this a long time ago. I remember that at first, I thought that there could be an F6. I thought "if winds are measured at 320mph why wouldn't it be classified as an F6?" Then I was shown/told that the F-scale (for rating purposes) is only based on damage.

Here are some quotes form The Tornado by Thomas Grazulis.

"The process of rating a tornado with the F-scale begins by observing the damage... A wind speed range can then be estimated for this tornado. There is nothing magical about the wind speed numbers; they are only rough estimates" P.133

"Judging whether an F5 rating applies is very difficult. The major difficulty is that the house no longer exists." P.136 (everything is destroyed, no more damage can be done)

"Portable radar readings of wind speeds are irrelevant to the Fujita Scale... If a house was unroofed and portable Doppler noted a 350-mph wind, the tornado should still be rated F2." P.138

As of now, an F6 can't happen. The Fujita scale is not perfect, but it is a great accomplishment and it is the best system we have now.

 

[Shane Adams]

If a tornado caused F5 damage but_was_accurately_clocked_above_319mph, technically, by definition it would be an F6, because regardless of their accuracy/inaccuracy, those numbers are posted next to each rating.

Because the numbers are posted, they must be recognized. Otherwise, why are they there? If an improvement needs to be made to the F-scale, simply leave the damage descriptions alone (since those are what matters according to everyone) and take away the windspeed numbers associated with each rating.

An F6 "inconceivable" tornado doesn't exist, by definition. However an F6 (319-???mph) "inconceivable" tornado, by definition, would exist, if that windspeed was accurately measured.

I don't care if F5 winds are really 185mph as opposed to 300mph. But as long as those little windspeed estimates are sitting behind each rating, there's going to be conversations like these. The F-scale is a damage scale; take the windspeed numbers off and just use the damage descriptions. Problem solved, because you eliminate the need to "guess" but you keep the F-scale (also eliminating the headache of creating and selling a new rating system).

 

[David Brookshier]

A damage scale with maximum allowable windspeed

I agree that the F scale is a damage scale but they made a mistake by including the fact that the maximum allowable windspeed in an F5 is 318 mph and in so doing the day will come when they will be forced to see an F6. We have already heard in this discussion that the Saffir Simpson scale is also a damage scale but NOAA doesn't wait around for damage to occur before rating a hurricane why??? because the windspeeds in that scale can and have been accuratly measured. We have seen advancements in our technology that would make people living 200 years ago die of shock just from one look at a television. It is only a matter of time before we finally gain the technology to accuratly record the exact wind measurements in the vortex of a tornado and when that day comes to where thier is no margin for error when a new teched out super radar clocks windspeeds at 319 mph they will have to resign thier beurocratic ways and make way for the F6 :!: Windspeeds are included in the Fujita scale and they have got to be included for a reason. Five perhaps fifty years from now we can harness the technology to put those numbers in the F scale to good use, an F6 will come and I just hope I live to see the day.
Instability=Success
Break the cap punch the core shoot the nado and the chase goes on.

 

[Scott Blair]

First, we can all agree that the F-scale has major problems that most likely will never be resolved. Now moving into the heart of the discussion...

In theory, the F-scale is a wind scale. It was created due to the desire to have the ability to classify such phenomena by strength. Strength being wind speed, at the time, could only be derived by examining damage from tornadoes. With that said, much work was accomplished in the laboratory and in the field to correlate damage with the wind speeds required to create such damage. The scale was then based on damage to estimate the wind speed of a tornado.

So the question... can there ever be a F6? The answer is a definite NO. Again, even though the F-scale is an estimated wind speed scale, it is based on the damage from a tornado. Therefore, a F5 tornado creates total destruction, foundations swept clean, etc. No damage can exceed total devastation... therefore impossible to rate anything higher.

So what to do if a mobile research radar "records" a wind greater than 318mph? Let's take May 3, 1999 DOW data that recorded a 300+mph wind. First off, it’s my understanding this speed only existed for one rapid scan. Secondly, the speed was a very small pixel on the velocity image, liable for some error. Lastly, line-of-sight propagation would agree that this speed was not at the ground, but slightly elevated above the wind/ground layer. Therefore, do we give allowance to the height of the radar beam above ground to "count"? It’s a simple fact wind speeds in a tornado are stronger above the friction found at the surface where the damage occurs. A lot of uncertainty and potential error is already looming by choosing this route.

The biggest problem facing us by using DOW data to justify a specific rating without damage is the simple climatological inconsistency that will result. Since DOWs are not on every tornado across the nation, the immediate skewing of data occurs by going this route. You use the scale in a way that's inconsistent with the entire national ranking system... which is rating based on damage. Any deviation from this methodology creates a national tornado database that becomes inconsistent in rating.

So what to do if the data is available? Perhaps continue the way it’s always been done. In addition, freedom allows plenty of mention in the event summary description that EX: "while this tornado remained over open countryside and resulted in no damage...mobile radar data recorded estimated winds of 155mph."

In summary, while the F-scale contains a wind speed scale... the wind speeds must be determined by tornado damage. Therefore, regardless of wind speeds remotely sensed or extreme damage... tornadoes will remain classified F0-F5 and should always be based on damage to determine the rating until a new scale commences.

And a side note regarding the hurricane mention... NOAA does in fact wait until after the storm to officially rate the strength of the landfalling system. Take Andrew as one of many examples... ten years before the ‘official’ strength was determined. Numerous research groups collect meteorological data during the storm that is only available after the storm. This serves as the true ground truth NOAA uses to incorporate into their final report and associated landfall strength.

Scott Blair
http://www.targetarea.net/

 

[Shane Adams]

I'm far more interested in the true windspeeds of tornadoes than I am the damage they do and what speed that corelates to. Why even rate tornadoes at all? There's no way to measure windspeeds accurately, and we all have seen the damage they can do. Why do they need to be rated according to damage if there's no strength scale (via windspeeds) associated with them?

In other words, why not just release a public damage report for all tornadoes that hit stuff; "The tornado cleaned foundations of well-built homes" or "the tornado removed roofs". What's the point in assigning a rating if there's no kinetic measurement available to jusitfy that rating? Take the Jarrell, TX tornado. Big controversy exists because it was a slow mover, and many experts think it did "F5" damage simply because it sat over structures for so long, and it really only had "F3" winds. So if the F-scale is so off, why even rate tornadoes? Just describe the damage it did and move on.

 

[Rob Satkus]

Originally posted by Scott Blair


So the question... can there ever be a F6? The answer is a definite NO. Again, even though the F-scale is an estimated wind speed scale, it is based on the damage from a tornado. Therefore, a F5 tornado creates total destruction, foundations swept clean, etc. No damage can exceed total devastation... therefore impossible to rate anything higher.

Although I understand your points, this is one of those things I just can't agree with. I do feel it is highly unlikely an F6 will occur, but I do believe that should one occur you will be able to tell the difference...damage of the type that has never been seen before..I just feel something like that is possible...highly unlikely, but possible. The way things have gone with the hurricane season, I don't think anything can be ruled out.

Regarding the tornado database:
The data base has always been inconsistent and always will be. Even if in the future, all tornadoes are rated only on damage, whose to say those are accurate? NWS personel can't get out and survey every tornado and even when they do, can we be sure they take the Tim Marshall approach and really check the structural integrity of each building? So what harm is there then is rating a few tornadoes using common sense? If there is definitive evidence, whether it be measured wind speeds or quality video that shows violent motion in a tornado, why not rate it higher than F0? That makes more sense to me that rating a mile wide, clearly strong to violent tornado an F0 just because it only blew away grass and ground squirrels.
As kind of an aside for Scott...nice picture in the June Storm Data...and it kind of ties in with the tornado database...SD has it listed as an F0 with a 30 mile path, but the narrative says "four"different" tornado touchdowns over a 30 minute period northwest of Ekalaka Mt....the tornadoes came from different parts of a large mesocyclone" "at times the tornadoes were multiple vortex" Note they say "tornadoes"...so this brings me to another problem with database accuracy...and maybe this should be a different thread...can you give a better account of what happened there? Was it 1 "long tracked tornado" Or 4 tornadoes? The vagueness of SD submissions is most frustrating.

Rob

 

[Chris Nuttall]

Originally posted by Rob Satkus
So what harm is there then is rating a few tornadoes using common sense? If there is definitive evidence, whether it be measured wind speeds or quality video that shows violent motion in a tornado, why not rate it higher than F0? That makes more sense to me that rating a mile wide, clearly strong to violent tornado an F0 just because it only blew away grass and ground squirrels.

Ahh, but there is the crux of the problem. Currently, we have no reliable method for determining accurate wind speeds near the surface in a tornado. New technology is being developed to hopefully solve this problem, but right now, we have nothing that can do this.

So, the questions remain? How do we describe the strength of a tornado? I'm sure most of us would agree that wind speed is the "most proper" indicator. Okay, now, how do we find that wind speed? We have no way to directly measure the wind speed. So, how do we estimate it? Pretty much, the only way is by what's left behind. And that's the damage.

Also, the Saffir-Simpson Scale is not perfect either. How many times have we seen landfalling hurricanes have we seen where the measured wind speeds fall below what the hurricane is "rated" at landfall in an advisory. Granted, there are several reasons for this, but that is another discussion.

Will an F6 happen? With the current F-scale, I'm going to say no. Could it happen, I suppose, but again, I don't think so. Then again, maybe if we get something like in L.A. during "The Day After Tomorrow," I suppose that might be an F6. Sorry, couldn't resist.

Lastly, great discussion. This is good stuff.

 

[Rob Satkus]

Originally posted by Chris Nuttall



Ahh, but there is the crux of the problem. Currently, we have no reliable method for determining accurate wind speeds near the surface in a tornado. New technology is being developed to hopefully solve this problem, but right now, we have nothing that can do this.


.

Well, with the DOW's and other mobile dopplers this can be done now. And they are reliable enough that wind speed measurements have been accepted. Obviously, there's no way to measure the wind speed at the surface, but near the surface, yes. Earlier this year, NWS OUN applied an F3 rating to the Knox county Tx tornado of May 13th...winds of around 177mph (79m/s) were measured about 30m above ground, about 6 miles south of Vera. (From the May issue of Storm Data and personal conversation with a friend at the NWS OUN). My friend said it was a difficult decision to make...several people involved felt as some do here, that a rating should only be given for damage, but ultimately it was agreed that the wind measurement was reliable enough to use as a determining factor. (If anyone involved with that is on the list, please chime in with any info..) Personally, I agree with that...I know others won't and that's fine...given the disparities of the F-scale. Now, I don't know how much difference friction would make from 30m to the ground, especially in a flat area like NW Tx ( at least I think it's flat)...maybe the mathmeticians out there can tell us. But even if you take into account friction, knock a category off the rating and it would still be close.
I have to wonder if there will ever be a reliable replacement to the F-scale, given the big differences in opinion on here about it...
So, anyway I'm not saying that my way is the right way or anyone else is absoluteley wrong...just my take on things.

Rob

 

[Shane Adams]

I think it's great that the May 13 tornado was assigned its rating based on an actual windspeed measurement - that's serious progress. True, this wind measurement may have inaccuracies, but no more than any typical damage assesment (because of the tons of variables including structure type, design, integrity, etc etc). This news is very encouraging IMO, from the viewpoint of an individual that has long-supported the F-scale.

 

[Chris Nuttall]

Okay, let me rephrase. If we can get a wind measurement and determine that the accuracy in margin of error is acceptable, then by all means, we should use it.

Yes, mobile radars like the DOWs can get these measurements on occassion. The problem is that we cannot do this in a regular manner. The DOWs are not going to get data from even 75% of the tornadoes that occur. Thus, we have no reliabe and regular way (maybe that's a better phrase) to obtain these mesaurements.

IMO, until we can get these measurements on a regular basis, damage still remains the primary way of rating a tornado and then guesstimating wind speed from that. But, if we can get measurements from a few and the data is acceptable, then fine. I'm all for supplementing the ratings with that info. That would be awesome.

In the meantime, I'm pretty interested in the research being done at Texas Tech. As has already been mentioned, some of their research shows that the F-scale winds may be as much as 30%-40% too high. IMO, I don't think Fujita ever "calibrated" those wind speeds when he created the scale. I think those were just educated guesses. So, if some revision can be done to the F-scale by calibrating the wind speeds, it would probably be a good idea.

Then, we run into another problem with tornado climo. If things turn out vastly different, then someone will have to go through and "re-rate" everything in the database. Sounds like a job for a few lowly undergrad students. :wink: