Enhanced Fujita Ratings Debate Thread

[jiharris0220]

Also interestingly, more than half of tornados with ＞100m/s winds on DOW didnt have EF5 contextual damage at all.

This gives more credence to the claim that tornadoes, even ones that are genuinely below the violent threshold, have far stronger winds than thought.

Along with the fact DOW have recorded several non violent tornadoes with winds exceeding 100m/s just this year alone, even taking into account contextual damage.

And tornadoes that are genuinely ef5 level, likely have winds far exceeding 135m/s (301mph+)

 

[A Guy]

<text>

Nice points - though is there more to this than is contained in that tweet? Because I can't see specific statements that you are making your numbered points against.



I'd also like to address this from Noles:



We have these statements on record:

From John Robinson: https://arktimes.com/arkansas-blog/2014/05/05/meteorologist-defends-ef4-rating-on-vilonia-tornado


From Tom Grazulis: https://ejssm.com/ojs/index.php/site/article/view/14 (in the PDF download)



So the question should be thrown back - why were some involved in the construction of the EF scale so fixated with the idea that houses should not be rated EF5, and why did they get their way?

I'd also add that I think the scale should be aligned around construction practices that actually exist, and that a house being completely swept way should only be EF3 if there is evidence it was actually or effectively unanchored. I'd also add this when it comes to failure modes - yes garage doors are a weak point, but not many recently constructed freestanding homes don't have an integral garage. So the scale should be realigned around the houses that are actually built.

 

[slenker]

This is total conjecture by me, so take this with a grain of salt. I’m a physics undergraduate student, not in an engineering field. I’m no expert by any means and I’d love to hear input on this.

One thing I haven’t seen taken into account in any intensity estimate is the extreme low pressures generated in the cores of violent tornadoes. Now, I understand that attempting to take something like that into account for estimating a wind speed would be practically impossible simply because any calculation done on the force required to slab a house would (probably) become absurdly complicated- with that said, wouldn’t the atmospheric pressure inside a violent tornado cause the air inside the house to want to GTFO as soon as possible? Wouldn’t this allow, on some level, the wind speed to not have to be as rapid to “slab” a home? I assume it would cause the house to be more readily obliterated.

If this is true, it would probably be better to rate tornadoes based on the net force that would have been required to do the damage it did, rather than wind speed due to the inconsistency we see. I want to know if there’s been any papers/studies done on this, it’s something that’s been nagging me.

There’s been EF4 tornadoes that have inflicted EF4 damage with windspeeds that were measured to be in-line with the EF scale itself (I can’t quite remember what tornado this was, but I’m pretty sure it happened- if not, I stand corrected) but also there’s tornadoes like Sulfur 2016 or Greenfield that have winds exceeding EF5 bounds but just simply not inflicting damage representative of it. It probably is just an altitude thing, like tornadic winds varying drastically with changing heights, but part of me believes it could be this.

I want to hear from somebody more knowledgeable than me on this, it’s just something I found interesting.

 

[Shakespeare 2016]

Sorry it took me this long to read all of this. I must say that many points are quite valid and I'm in full agreement with almost all of it. Some comments:

#1 It seems strange to me that winds are always listed as multiples of 5 which is because they can't resolve better than that, yet we get many200's and no 201's or 205's.
#2 Evidence of what they consider "well built" includes toe-nailed studs,m and that evidence is rather easy to find. Very few homes have toe-nailed studs,so finding no holes where the toe-nails would have been means "not well-built" right away. This one parameter can be discovered in an instant so long as there are studs which can definitely be linked to that one site. I have no beef with the NWS on this in most instances, but my experience in building and demo work shows me that end-nailing (the most common practice) also has variables- mainly nail type- and that true 16d common nails resist side-shear approximately as well as toe nailed studs and after being in place several years also resist end-pull-out nearly as well too. However in the last 20+years, almost all homes are built with nail-guns, having thinner diameter shanks and shorter lengths. Those nails (along with CC Sinker type) have good initial pull-out resistance but once moved lose most of their grip. I have built exactly one home where we toe-nailed studs with guns, none by hand, and I've built or helped build at least 70 homes in several different parts of the US using local techniques. They need to alter that DI to allow for true 16d nails then I'd be happy.
#3 If no other rating requires a 70+ yard diameter damage area, then it should be removed as an EF-5 DI/DOD rule. Accept the damage found as is. We've all seen momentary subvortices that probably covered far less area so it's clearly possible for a group of homes to be lower in damage while one or two affected by that subvortice are EF-5.
#4 "Home not hit by debris" would have some validity if you specified exactly what the mass of excluding debris is but they don't. All homes get hit by debris; not all homes have a vehicle thrown at them. If we do such defining we might end up like the low end of the TORRO scale which to me is laughable in it's efforts to exactly define ending up with more questions that answers. I would like a rule on the debris mass "being large enough and at such a velocity to greatly affect wall strength" of something similar.
#5 :Has to be ground scouring: is insane. There are so many different types of soils and conditions that the lack of scouring can be meaningless as anyone who has dug into dry red Georgia clay or dry Texas 'hardpan' can tell you. It's almost like concrete and even 205MPH winds would have a hard time dislodging it. The wet fertile soil of the Imperial Valley in SoCal could probably be scoured by a low-end EF-4, maybe less. Asphalt also has many variations. I think we should use scouring just as an additional contextual indicator when it occurs and take nothing from it when it's absent.
#6 "All trees within 35 yards have to be fully gone and fully debarked does not take into account that buildings or even blown large debris can protect against debarking until the protection goes away, by which time the EF-5 core could have passed. Or the structure ciould have been at the edge of the EF-5 core with the tree(s) just outside of the 35 yard area.
#7 Extreme debris granulation will be affected by the time itis in the core, which is why slow-movers like Jarrel granulate completely while fast-movers likr Philadelphia or Smithville may not. Again, it's presence can mean everything but it's absence means nothing.

In conclusion we need to end the mindset of everything having to be exactly within written parameters or it being discounted, instead allowing an experienced surveyor to override some of the exactness without question when they truly believe the damage was otherwise at a higher level. Show me a precise tornado then I'll accept precise parameters for determining damage levels (but only for that tornado, and not the usual imprecise ones).

I think the windspeeds on the EF-SCALE are too low. I made my own and no I don't round up to the nearest 5 or 10.

 

[A Guy]

This is total conjecture by me <>

From what I've read, the pressure drop in very strong tornadoes is 100 mbar, maybe up to 200 and occurs over a sufficient time period and houses are sufficiently porous that it doesn't really do anything.

 

[Shakespeare 2016]

Nice points - though is there more to this than is contained in that tweet? Because I can't see specific statements that you are making your points against.

I'd also like to address this from Noles:

View attachment 28324

We have these statements on record:

From John Robinson: https://arktimes.com/arkansas-blog/2014/05/05/meteorologist-defends-ef4-rating-on-vilonia-tornado
View attachment 28325

From Tom Grazulis: https://ejssm.com/ojs/index.php/site/article/view/14 (in the PDF download)

View attachment 28326

So the question should be thrown back - why were some involved in the construction of the EF scale so fixated with the idea that houses should not be rated EF5, and why did they get their way?

Kind of like how there can be more 165 mph EF3'S than 166 mph EF4'S just like you say 200 mph EF4'S and not 201 mph EF5'S. When I ask other NWS offices about this you never get an answer back. I don't care what people think but I am very suspicious they are hiding something.

 

[pohnpei]

This gives more credence to the claim that tornadoes, even ones that are genuinely below the violent threshold, have far stronger winds than thought.

Along with the fact DOW have recorded several non violent tornadoes with winds exceeding 100m/s just this year alone, even taking into account contextual damage.

And tornadoes that are genuinely ef5 level, likely have winds far exceeding 135m/s (301mph+)

We just have DOW7 right near the Greenfield when it did high end EF4 damage(some claim near or reached EF5 level) and they got as high as 290mph winds. We don't know it's a raw velocity or Vp and aspect ratio adjusted ratio.

We have another similar case in Spencer SD 1998 when an F4 tornado went through town with DOW3 nearby

The highest 5-s average wind speed had a maximum of about 112 m s−1 on the western edge of town around Third Street. The highest 60-s average wind speed at each point in the town showed a 50-m swath across the south-central portion of the town between Second and Third Streets where this parameter reached a maximum of just over 80 m s−1 (Fig. 8b).

It is worthwhile to note that the observed asymmetric tornado contained ground-relative wind speeds as high as 118 m s−1, slightly into the F5 range, as it passed over the western edge of the town.

One thing noted here is those cases seem to be surprisingly well accord with IF scale wind speed scale.


Those were wonderful cases though we also need to know with one highest non ground level bin on radar being representative of the overall intensify of the tornado could be an oversimplification of the problem.

 

[Shakespeare 2016]

We just have DOW7 right near the Greenfield when it did high end EF4 damage(some claim near or reached EF5 level) and they got as high as 290mph winds. So don't know it's a raw velocity or Vp and aspect ratio adjusted ratio.
We have another similar case in Spencer SD 1998 when an F4 tornado went through town with DOW3 nearby


One thing noted here is those cases seem to be surprisingly welk accord with IF scale wind speed scale.
View attachment 28327View attachment 28328

Those were wonderful cases though we also need to know with one highest non ground level bin on radar being representative of the overall intensify of the tornado could be an oversimplification of the problem.

The winds in the Greensburg tornado were probably as high as 260 to 290 mph and it certainly did EF5 damage.

 

[slenker]

From what I've read, the pressure drop in very strong tornadoes is 100 mbar, maybe up to 200 and occurs over a sufficient time period and houses are sufficiently porous that it doesn't really do anything.

I see a simple google search would have saved me a wall of text. It seems crazy to me that homes are porous enough for that though, especially when a tornado could be approaching at 70 mph.

 

[pohnpei]

This is total conjecture by me, so take this with a grain of salt. I’m a physics undergraduate student, not in an engineering field. I’m no expert by any means and I’d love to hear input on this.

One thing I haven’t seen taken into account in any intensity estimate is the extreme low pressures generated in the cores of violent tornadoes. Now, I understand that attempting to take something like that into account for estimating a wind speed would be practically impossible simply because any calculation done on the force required to slab a house would (probably) become absurdly complicated- with that said, wouldn’t the atmospheric pressure inside a violent tornado cause the air inside the house to want to GTFO as soon as possible? Wouldn’t this allow, on some level, the wind speed to not have to be as rapid to “slab” a home? I assume it would cause the house to be more readily obliterated.

If this is true, it would probably be better to rate tornadoes based on the net force that would have been required to do the damage it did, rather than wind speed due to the inconsistency we see. I want to know if there’s been any papers/studies done on this, it’s something that’s been nagging me.

There’s been EF4 tornadoes that have inflicted EF4 damage with windspeeds that were measured to be in-line with the EF scale itself (I can’t quite remember what tornado this was, but I’m pretty sure it happened- if not, I stand corrected) but also there’s tornadoes like Sulfur 2016 or Greenfield that have winds exceeding EF5 bounds but just simply not inflicting damage representative of it. It probably is just an altitude thing, like tornadic winds varying drastically with changing heights, but part of me believes it could be this.

I want to hear from somebody more knowledgeable than me on this, it’s just something I found interesting.

After years I've finally see someone said what always want to say and agree. It's been an idea in my mind for really long time.

We now have 3 second wind scale from EF scale and instantaneous wind scale from IF scale. They are of vast difference in tornados.(in some cases of above 60% difference) We actually dont know which scale is better for measuring in tornados. It may fits some tornados but not others for each one of these scale. And wind direction inside tornados are overly complicated. Even in hurricanes, we have 1min scale used by NHC and JTWC, 2 min scale used by CMA and 10min scale used by JMA. The wind speed they give for same level strom were very different due to the difference of wind speed scale. (for example, JTWC gave typhoon Haiyan 170kt and JMA gave 125kt but they both had 895hpa pressure)And we have many claims exist that pressure is the more consistent way for measuring the overall intensity. For tornados, the situation would be way way more complicated than even in hurricanes due to the features we all know inside tornados. Also we don't know the response time for the damage done by winds. It probably also changed a lot in different situations.

As we have claims that pressure is more consistent way for measuring hurricane's intensity. The pressure itself don't play a role in the direct damage compared with wind. But the pressure inside tornados do play a big role in the direct damage, maybe even bigger role than winds in many situations for reasons you said about.

For these two reasons above(the consistenteny of measurement, the important role it play into damage), I strongly agree the idea that, at the very least, pressure deficit inside tornados need to be studied much more.

Also using net force would be an even more interesting idea. I've only seen one people come up with this idea before. The only problem is how we measure it.

 

[pohnpei]

After years I've finally see someone said what always want to say and agree. It's been an idea in my mind for really long time.

We now have 3 second wind scale from EF scale and instantaneous wind scale from IF scale. They are of vast difference in tornados.(in some cases of above 60% difference) We actually dont know which scale is better for measuring in tornados. It may fits some tornados but not others for each one of these scale. And wind direction inside tornados are overly complicated. Even in hurricanes, we have 1min scale used by NHC and JTWC, 2 min scale used by CMA and 10min scale used by JMA. The wind speed they give for same level strom were very different due to the difference of wind speed scale. (for example, JTWC gave typhoon Haiyan 170kt and JMA gave 125kt but they both had 895hpa pressure)And we have many claims exist that pressure is the more consistent way for measuring the overall intensity. For tornados, the situation would be way way more complicated than even in hurricanes due to the features we all know inside tornados. Also we don't know the response time for the damage done by winds. It probably also changed a lot in different situations.

As we have claims that pressure is more consistent way for measuring hurricane's intensity. The pressure itself don't play a role in the direct damage compared with wind. But the pressure inside tornados do play a big role in the direct damage, maybe even bigger role than winds in many situations for reasons you said about.

For these two reasons above(the consistenteny of measurement, the important role it play into damage), I strongly agree the idea that, at the very least, pressure deficit inside tornados need to be studied much more.

Also using net force would be an even more interesting idea. I've only seen one people come up with this idea before. The only problem is how we measure it.

The pressure gradient and net force thing could explain some inconsisteneny we see if we only use wind speed. One example is Bridge Creek and Mulhall. They basically had the same value on DOW3 near the same level while damage was such different. Also 13 El Reno type of tornado would very likely had stronger winds than Moore 2013 cause it had much more subvortex. With some subvortex traveling at 80m/s, it would be extremely hard for Moore 2013 type of tornado(lacking subvortex) to beat if we just talking about pure wind speed.

 

[slenker]

For these two reasons above(the consistenteny of measurement, the important role it play into damage), I strongly agree the idea that, at the very least, pressure deficit inside tornados need to be studied much more.

Also using net force would be an even more interesting idea. I've only seen one people come up with this idea before. The only problem is how we measure it.

I suppose I just find it pretty hard to believe that the air within a house is capable of conforming to the atmosphere surrounding it purely from a pressure standpoint, especially when talking about tornadoes that reach forward translational speeds of over 50 mph. I would love to see papers or studies exploring this idea, that may be a major factor within the sledgehammer effect. I can definitely see it not being a factor in slow movers, but with fast violent monsters like Smithville/Hackleburg/Mayfield the wind/pressure gradient has to be unbelievably tight, winds can go from 30 mph to about 250 mph in roughly 15 seconds in those extreme tornadoes. If it isn’t the pressure, then what is that factor that could possibly come into play?

Also, trench-digging and deep ground scouring. I find it very hard to believe that that is purely wind.

Another thing? Extreme super typhoons/cat 5 hurricanes can have wind gusts approaching or even exceeding 200 mph rather easily, yet they do nowhere near the level of wind damage that even a low-end EF4 tornado does. There’s definitely a vertical component to the winds that cause a greater capability of creating damage like that, but again, I find it hard to believe that’s the only factor.

 

[pohnpei]

I suppose I just find it pretty hard to believe that the air within a house is capable of conforming to the atmosphere surrounding it purely from a pressure standpoint, especially when talking about tornadoes that reach forward translational speeds of over 50 mph. I would love to see papers or studies exploring this idea, that may be a major factor within the sledgehammer effect. I can definitely see it not being a factor in slow movers, but with fast violent monsters like Smithville/Hackleburg/Mayfield the wind/pressure gradient has to be unbelievably tight, winds can go from 30 mph to about 250 mph in roughly 15 seconds in those extreme tornadoes. If it isn’t the pressure, then what is that factor that could possibly come into play?

Also, trench-digging and deep ground scouring. I find it very hard to believe that that is purely wind.

Another thing? Extreme super typhoons/cat 5 hurricanes can have wind gusts approaching or even exceeding 200 mph rather easily, yet they do nowhere near the level of wind damage that even a low-end EF4 tornado does. There’s definitely a vertical component to the winds that cause a greater capability of creating damage like that, but again, I find it hard to believe that’s the only factor.

The best pressure measurement we have now was still Manchester SD 2003. The pressure dropped 100hpa in just 10s. So the wind just complete its acceleration in such short time. That tornado moved at 8.3m/s when approaching the probe which is only typical level in plain. For tornados much faster and stronger than Manchester like Smithville, I would expect a much shorter and stronger pressure fall and wind acceleration than that. I would be very surprise that kinda of pressure gradient and acceleration itself wouldnt attribute if not done the damage mostly.

 

[slenker]

As for calculating a sort of net force, isn’t that how the engineers get their wind speed estimates anyways? So I’m guessing it wouldn’t change overall ratings much. Either way, if wind speeds are not the only major factor causing extreme damage in tornadoes then a net force calculation is far better for measuring tornadic strength than wind speed, and the EF scale should leave windspeed measurements to DOW.

Also, just to put this out there, not saying wind speeds were probably lower. I genuinely think the strongest EF5’s definitely had wind speeds well exceeding 300 mph. Pressure is very difficult to measure within tornadoes so that adds a further layer of complexity to rating these things accurately.

The best pressure measurement we have now was still Manchester SD 2003. The pressure dropped 100hpa in just 10s. So the wind just complete its acceleration in such short time. That tornado moved at 8.3m/s when approaching the probe which is only typical level in plain. For tornados much faster and stronger than Manchester like Smithville, I would expect a much shorter and stronger pressure fall and wind acceleration than that. I would be very surprise that kinda of pressure gradient and acceleration itself wouldnt attribute if not done the damage mostly.

That’s very interesting and only adds to my suspicion about winds being the only thing doing the damage. But again, it may not be the pressure, it could be something else entirely (which I sort of doubt, but it’s impossible to draw conclusions without studies being done on it.)

 

[Sawmaster]

From what I've read, the pressure drop in very strong tornadoes is 100 mbar, maybe up to 200 and occurs over a sufficient time period and houses are sufficiently porous that it doesn't really do anything.

I can't remember the tornado, but there's a well-known pic of window curtains having been drawn through where the roof met the walls; hard to imagine them being blown through from the inside but easy to see that happening with a vacuum pulling them from the outside. IIRC, the measured pressure when winds blow down garage doors and over-pressure the house is 3psi- not much at all but combined with a strong pressure drop outside would be much stronger in effect. I don't think houses are so porous as to have much effect on the suddenness of the pressure change as a slowly applied 3psi would do nearly nothing. And while I'm not privy to the details of the Texas Tech wind testing procedures, I can't see how they could replicate the pressure drop at the core, but measuring the windspeed would be easy to do accurately since you're creating that wind. It's also one of the main reasons I am suspicious of the correlation between actual winds and either Fujita scale's estimation of them. We've seen enough low-level DOW measurements not aligning with the scale to know there's something more happening that we're missing.

I've always felt that the pressure drop at the core varies in strength, with some tornadoes exhibiting a lot more lofting ability than others which had the same estimated winds. Considering only winds, debris should be mostly moving outward downwind, not so much upward, yet we get debris at high altitudes sometimes even with lower-rated tornadoes and some strong tornadoes which exhibit less lofting than you'd expect. Clearly the pressure drop is a significant factor in the amount and type of damage tornadoes do.

 

[A Guy]

Lofting ability is far more likely to be related to the corner flow structure of the tornado, how abruptly the winds turn and vertical compared to horizontal velocity. While that's obviously significantly controlled by the pressure gradient, it's not pressure itself as a proximate cause.

 

[Sawmaster]

Kind of like how there can be more 165 mph EF3'S than 166 mph EF4'S just like you say 200 mph EF4'S and not 201 mph EF5'S. When I ask other NWS offices about this you never get an answer back. I don't care what people think but I am very suspicious they are hiding something.

I don't think they're trying to hide facts- it's simply that they have no plausible explanation which fits the observed facts, and therefore no way to justify their actions. They remain silent to hide their stupidity on this matter.

 

[Sawmaster]

I see a simple google search would have saved me a wall of text. It seems crazy to me that homes are porous enough for that though, especially when a tornado could be approaching at 70 mph.

Just want to point out that with the structure intact, old homes are quite leaky while newer homes are far less so to gain efficiency in heating and cooling them. There are even cases where homes are so air-tight that the inhabitants became ill from a lack of fresh air and a device known as an "air exchanger" was needed to provide enough fresh air. Porosity varies but can be lower than you'd think.

 

[Sawmaster]

Lofting ability is far more likely to be related to the corner flow structure of the tornado, how abruptly the winds turn and vertical compared to horizontal velocity. While that's obviously significantly controlled by the pressure gradient, it's not pressure itself as a proximate cause.

That makes sense. Good to see the deep reasoning in this thread versus the usual nit-picking arguments which often occur with these discussions.

 

[Shakespeare 2016]

I don't think they're trying to hide facts- it's simply that they have no plausible explanation which fits the observed facts, and therefore no way to justify their actions. They remain silent to hide their stupidity on this matter.

Like the original F-SCALE, Fujita used numbers that were rounded to the nearest 0 or 5. One thing I would like to say in regards is that on the F-SCALE is that 260 mph was the very last numberof F4 and 261 mph was the very first number for F5.