The Fujita Scale Controversy

[Hannah.Taylor]

More and more I see so many chasers and even some meteorologists upset with the NWS assigned ratings to various tornadoes of significance.

Back in May, a home in Marion, IL was completely swept of its foundation. The tornado received an assigned rating of EF-4.




Of the most controversial tornadoes was the May 2013 El Reno, OK, which was assigned an EF-3 rating despite numerous readings of well over 200 mph but due to its path very limited damage to property and structures. I've actually heard people get angry over the rating of this tornado. Impassioned anger.

Tornado Tour Guide and Chaser Erik Burns posed a fantastic question last month.



Ironically, there hasn't been a single tornado rated as an EF-5 since El Reno in 2013.

Tuscaloosa 2011 and El Reno 2013 definitely should have been EF-5's.

A commenter on Erik Burns post brought up a great example. How is a hurricane a CAT5 over the ocean when it's not even doing any damage on land but a tornado requires everyone to come measure?


WIBW's Chief Meteorologist Jeremy Goodwin posted this a few days ago.




Is the NWS just reluctant to assign EF-5 ratings now? Is it because of insurance purposes? Is it a government mandate to avoid certain FEMA funding activations?

Whatever the reasons, I personally think they need two scales (or just do away with the Enhanced Scale and return to the original Fujita scale).

Tornado Intensity Rating (TIR)
One for actual measured wind speeds at any point during its tenure between touchdown to lifting. I believe the original Fujita Scale ratings would be acceptable here.

Tornado Damage Rating (TDR)
Ratings here would be based on the damage done. Do away with the Enhanced Fujita nonsense. Call it the TDR Scale. Levels 1-5.

Thoughts?

 

[Randy Zipser]

Whatever the reasons, I personally think they need two scales (or just do away with the Enhanced Scale and return to the original Fujita scale).

Tornado Intensity Rating (TIR)
One for actual measured wind speeds at any point during its tenure between touchdown to lifting. I believe the original Fujita Scale ratings would be acceptable here.

Tornado Damage Rating (TDR)
Ratings here would be based on the damage done. Do away with the Enhanced Fujita nonsense. Call it the TDR Scale. Levels 1-5.

Thoughts?

The concept you propose, Hannah, would be too confusing and not practical to implement, IMHO.

The original Fujita Intensity Scale ("F-scale," from 0-5) was based upon very subjective observation from post-event damage surveys, conducted mostly by a local NWS office. If anything, damage estimates before 2007 using the original Fujita Intensity Scale tended to overestimate what was actually observed. By 2007, wind-engineering research studies at several major universities, most notably Texas Tech University, had progressed to the extent that wind/structural engineers were confident that the damages being observed could be explained by lower wind speeds, based upon what they were observing under controlled-conditions in the laboratory. It was these observations that led to the F-scale being replaced by the Enhanced Fujita Intensity Scale ("EF-scale," from 0-5) on February 1, 2007, in the U.S.

The key point is: accurate tornado damage "intensity" ratings cannot (and should not) be based upon very random, capricious field observation by observers who are using a "trained eye" anywhere from untrained to experienced. They can, however, be based upon acceptable engineering practice and hypothesis (based upon a large sample of past case studies), where a common standard is applied in all cases.

Rather than scuttling the current EF-scale altogether, I believe a better approach would be to do a comprehensive review of the current EF-scale windspeed threshholds, and update this scale based upon the current availability of a much greater volume of case-study data, in this case, 18-years worth of additional structural-damage data (2007-2025). For simplicity, the "E" for "Enhanced" should be dropped. Then, the NWS should adopt a new Revised F-Scale ("RF-scale, from 0-5). It could even be further refined using a year number (as there could be additional revisions in future years), such as "25RF-0" or "26RF-5." Should future case-studies confirm that higher windspeeds are warranted, an "RF-6" level could be added to this classification.

BTW, according to Wikipedia, a research study in November, 2022, has indicated that a more-standardized EF-scale is already in development using mobile Doppler radar (not sure how many mobile Dopplers would be necessary to make this work!) and "forensic engineering" (probably like what Tim Marshall has been doing during his career at Haag Engineering).

Lastly, the reason that the "TIR" above would not work is that to be able to measure actual windspeeds (and these would not be at the ground), a mobile Doppler (e.g., DOW) vehicle would have to be available and in stationary position for scanning prior to tornado touchdown--something that is neither affordable, practical nor possible in the event of a widespread tornado outbreak.

Just my 2-cents worth...

 

[Andy Wehrle]

The concept you propose, Hannah, would be too confusing and not practical to implement, IMHO.

The original Fujita Intensity Scale ("F-scale," from 0-5) was based upon very subjective observation from post-event damage surveys, conducted mostly by a local NWS office. If anything, damage estimates before 2011 (I believe it was) using the original Fujita Intensity Scale tended to overestimate what was actually observed. By 2011, wind-engineering research studies at several major universities, most notably Texas Tech University, had progressed to the extent that wind/structural engineers were confident that the damages being observed could be explained by lower wind speeds, based upon what they were observing under controlled-conditions in the laboratory. It was these observations that led to the F-scale being replaced by the Enhanced Fujita Intensity Scale ("EF-scale," from 0-5) in 2011.

The key point is: accurate tornado damage "intensity" ratings cannot (and should not) be based upon very random, capricious field observation by observers who are using a "trained eye" anywhere from untrained to experienced. They can, however, be based upon acceptable engineering practice and hypothesis (based upon a large sample of past case studies), where a common standard is applied in all cases.

Rather than scuttling the current EF-scale altogether, I believe a better approach would be to do a comprehensive review of the current EF-scale windspeed threshholds, and update this scale based upon the current availability of a much greater volume of case-study data, in this case, 14-years worth of additional structural-damage data (2011-2025). For simplicity, the "E" for "Enhanced" should be dropped. Then, the NWS should adopt a new Revised F-Scale ("RF-scale, from 0-5). It could even be further refined using a year number (as there could be additional revisions in future years), such as "25RF-0" or "26RF-5." Should future case-studies confirm that higher windspeeds are warranted, an "RF-6" level could be added to this classification.

Lastly, the reason that the "TIR" above would not work is that to be able to measure actual windspeeds (and these would not be at the ground), a mobile doppler (e.g., DOW) vehicle would have to be available and in position for scanning prior to tornado touchdown--something that is neither affordable, practical nor possible in the event of a widespread tornado outbreak.

Just my 2-cents worth...

EF-scale became operational in February 2007. Greensburg 5/4/07 was the first tornado to be rated EF5.

 

[Randy Zipser]

EF-scale became operational in February 2007. Greensburg 5/4/07 was the first tornado to be rated EF5.

Thank you, Andy. You are absolutely correct, and I have made that correction in my most recent edit, based upon Wikipedia.

 

[Randy Jennings]

Is the NWS just reluctant to assign EF-5 ratings now? Is it because of insurance purposes? Is it a government mandate to avoid certain FEMA funding activations?

The answers to all of these questions is no. These questions comes up all the time, and no one has ever provided any proof that a NWS EF rating makes any difference on a insurance claim and there are no FEMA "funding activations" based on EF ratings. Modern tools, like the Damage Assessment Tool Kit, training, and policy have made NWS assessments more uniform and bigger events almost always have real experts (like Tim Marshall) assist. I would argue that there really isn't necessarily reluctance to rate a tornado higher (although I am sure they are aware that higher ratings will get them higher review).

Anytime ones tries to develop a system to classify something, their are problems. Just ask as group of librarians about the Dewey Decimal system, vs the Library of Congress system, vs the ton of other systems for organizing library books. The EF scale is no different. Yes it has problems - the chief one being is it uses damage as a proxy for wind speed and in order for things to be rated higher the tornado has to hit a really well built structure. Yes, the EF scale needs to be improved (any they are working on it).

But here's my probably not so popular take - does it mater? Most of the time I think the answer is no. Sure higher end tornados cause more damage and loss of life and classification of a weak vs strong tornado can be helpful in scientific research trying to better forecast and communicate severe weather risks, but I'm not sure we get much benefit from most tornado damage surveys.

 

[Andy Wehrle]

I guess the question boils down to...what are we actually trying to accomplish with the tornado rating scale?

@Randy Zipser, you wrote.

"...wind-engineering research studies at several major universities, most notably Texas Tech University, had progressed to the extent that wind/structural engineers were confident that the damages being observed could be explained by lower wind speeds, based upon what they were observing under controlled-conditions in the laboratory."

Are we trying to determine the minimum necessary wind speed to do the damage, or the actual wind speeds the tornado contained (using damage as a proxy in lieu of direct measurements which will be unavailable for the vast majority of tornadoes)?

Last fall, @Mike Smith wrote this blog post citing research indicating something I've suspected for a number of years now, which is that the original Fujita scale is actually a better estimate of the wind speeds in high-end tornadoes, even if that can't be determined from damage alone.

Finally It Is Stated in the Published Research: The Original Fujita Scale Is Superior to Its Replacement

Ted Fujita with me in 1993. Ted visited WeatherData, Inc on a couple of occasions, this was when he presented some initial findings of his ...

www.mikesmithenterprisesblog.com

The way the scale is being applied has also definitely changed over time, the fact that there has been no EF5 in 12 years and counting is a clear indication of that when there used to be about one F5 per year, maybe occasionally skipping a year, in the '80s and '90s.

I don't see any point in getting "angry" about it, but there are some ratings in that time I definitely disagree with.

 

[Randy Zipser]

I guess the question boils down to...what are we actually trying to accomplish with the tornado rating scale?

@Randy Zipser, you wrote.

"...wind-engineering research studies at several major universities, most notably Texas Tech University, had progressed to the extent that wind/structural engineers were confident that the damages being observed could be explained by lower wind speeds, based upon what they were observing under controlled-conditions in the laboratory."

Are we trying to determine the minimum necessary wind speed to do the damage, or the actual wind speeds the tornado contained (using damage as a proxy in lieu of direct measurements which will be unavailable for the vast majority of tornadoes)?

Last fall, @Mike Smith wrote this blog post citing research indicating something I've suspected for a number of years now, which is that the original Fujita scale is actually a better estimate of the wind speeds in high-end tornadoes, even if that can't be determined from damage alone.

Finally It Is Stated in the Published Research: The Original Fujita Scale Is Superior to Its Replacement

Ted Fujita with me in 1993. Ted visited WeatherData, Inc on a couple of occasions, this was when he presented some initial findings of his ...

www.mikesmithenterprisesblog.com

The way the scale is being applied has also definitely changed over time, the fact that there has been no EF5 in 12 years and counting is a clear indication of that when there used to be about one F5 per year, maybe occasionally skipping a year, in the '80s and '90s.

I don't see any point in getting "angry" about it, but there are some ratings in that time I definitely disagree with.

I am familiar with Mike Smith's paper that you cite above.

To answer your question above, the purpose of having a tornado-damage windspeed rating classification system is to determine some reasonable proxy for what range of windspeeds could cause the the pattern of damage(s) that are actually observed, ideally very soon after the event (to prevent human intervention or modification of the original damage configuration). Thus, viewed this way, it's really not about a single number (i.e., either top or bottom of a particular range), rather it's about the range itself that is accurately described by the rating number (0 through 5, in the present EF-scale classification). And, yes, you may be correct in pointing out that EF-5's are becoming more "rare" over time, and therefore the "5" level should be dropped. It all comes down to what point one wants to make when referencing EF-scale rating numbers. For example, in the famous El Reno tornado (rated only as EF-3 because it occurred mostly over open-country and did little structural damage), the DOW data measured a maximum radar-indicated windspeed of 295mph, but what is often cited in the literature for that tornado is the 318mph maximum for the EF-5 range. The current classification scale is too subjective in how it is applied, perhaps even to score some political points for perhaps obtaining more funding for NOAA or the NWS, for example...and, of course, that's not the purpose it was originally created for.

As I stated in Post #2, what really needs to be done now is to review and update the current EF-scale based upon more recent data obtained since the EF-scale was first introduced in 2007. Perhaps this might lead to a new scale with only 4 levels, but with larger intervals between levels, or 6 levels with smaller intervals; what is so magical about only 5 levels anyway? Another suggestion would be to have all future structural-damage surveys conducted by teams of professional structural engineers, rather than meteorlogists employed at NWS local offices who may be inadequately-trained or too-inexperienced to perform this duty with the level of engineering sophistication (and consistency) required from event to event. Or, as I prefer to think about it, after every major tornado event, the NWS should send a "team of Tim Marshalls" to handle the windspeed assessment task, and NOAA should let contracts with vetted private professional engineering firms across the U.S. who specialize in structural-failure situations. I firmly believe that this essential duty is one area where the private sector could do the job more effectively, less expensively, and more accurately than the current NWS does! And, this kind of partnership will also go a long way toward ensuring a consistently accurate windspeed-range rating for each tornado event. Of course, a separate team of meteorologists could also be assigned aerial or ground-truth assessments to get a more detailed picture about the meteorological (non-windspeed) aspects (i.e., one continuous long-track tormado or a discrete family of tornadoes, cyclonic or anticyclonic, multiple vortex, etc.), depending upon the severity of the event. This is just what Dr. Ted Fujita at the University of Chicago did in the heyday of organized tornado research back in the early 1970s, which then led to photogrammetric determination of tornado windspeeds and his rating classification..

So there's no reason to be "angry," but plenty of reason to get "busy" making changes for improvement such as those discussed above...

 

[Randy Zipser]

The answers to all of these questions is no. These questions comes up all the time, and no one has ever provided any proof that a NWS EF rating makes any difference on a insurance claim and there are no FEMA "funding activations" based on EF ratings. Modern tools, like the Damage Assessment Tool Kit, training, and policy have made NWS assessments more uniform and bigger events almost always have real experts (like Tim Marshall) assist. I would argue that there really isn't necessarily reluctance to rate a tornado higher (although I am sure they are aware that higher ratings will get them higher review).

Anytime ones tries to develop a system to classify something, their are problems. Just ask as group of librarians about the Dewey Decimal system, vs the Library of Congress system, vs the ton of other systems for organizing library books. The EF scale is no different. Yes it has problems - the chief one being is it uses damage as a proxy for wind speed and in order for things to be rated higher the tornado has to hit a really well built structure. Yes, the EF scale needs to be improved (any they are working on it).

But here's my probably not so popular take - does it mater? Most of the time I think the answer is no. Sure higher end tornados cause more damage and loss of life and classification of a weak vs strong tornado can be helpful in scientific research trying to better forecast and communicate severe weather risks, but I'm not sure we get much benefit from most tornado damage surveys.

RJ, here's my take on why a tornado damage survey matters (on its most fundamental level). It's much more basic than paying insurance claims or advancing scientific research. As human beings, we are all creatures of curiosity. We also strive to understand our environment, and to the extent we can, sometimes try to control it. This is the universal reason that humanity has evolved the way it has and has invented all the wonderful gadgetry and technology that nowadays we feel we truly cannot get through a single day of our lives without.

Perhaps, the same thing applies to weather. Ever since the first tornado (or waterspout) was observed, going well back into Roman and Greek times, people have always had a fascination with them. In more modern times, people have acquired the knowledge and determined the means to study them and have invented machines (mechanical movie projectors and, later, electronic computers) to analyze them and figure out the "physical" details (like how fast those winds are rotating). This is much the same reason why we always wondered how fast a baseball pitcher throws the ball, until one day, handheld radar guns and televisions were invented to answer that question for us.

So, in the world we live in today, we have the means to know to a very-high degree of accuracy how fast those winds were in that vortex that destroyed my town or flattened my home or took the lives of my family, neighbors, or friends. This ability gives human beings some small measure of control in our daily lives, if not closure, even though we couldn't have done a thing to change the course of such a natural event that is so vastly beyond our actual ability to control.

The reason those tornado windspeeds (derived from a damage survey) matter (and benefit us) most...is to satisfy just plain normal human curiosity! RZ

 

[Boris Konon]

The thing that gets me is that when a tornado is given 190 mph rating, and people complain that it "should have been an EF5." There is zero sensible wx and impact difference between 190 and 200 mph winds. Damage is absolute and complete either way.

From an engineering standpoint, actual wind speeds are of value, but the vast majority of the time, we are
not talking about that, we are talking about how it impacts society from a public POV. The vast majority of
our structures that exist fail well before 190 mph winds, so focus on exactly what the winds were when it crosses
some arbitrary threshold *we* assign is irrelevant for all practical purposes!

EF4 instead of EF5 sounds like a psychological issue (as in ego or shock/awe) rather than physical issue . People get so caught up in the extent of damage, it skews their perception. The enormity of impact makes them think "this HAS to be the 'worst ever', so EF5 is a no-brainer!"

I seem to recall an account, I think from one of the Tornado Project publications, that the Comfrey MN F4 tornado
on 3/29/1998, some locally were 'upset' that is got 'only' and F4 rating, T-shirts were made stating "F4 my a**!"
So you see the social issues involved that can contaminate the science,.

As for mobile radar sample winds, we are dealing w/ wind measurements at least a few hundred feet above the ground. Tornado winds are not uniform vertical in its column. And we have friction close and at ground level and the fact not all tornadoes are the same horizontal and vertical in extent. So you see the issues here. Too many variables to use mobile radar winds aloft for assign a tornado strength.

Now, one can argue it was EF5 strength when you consider the entire tornado column, but what we are concerned about are winds very close and at ground level were we live. Also, some recent data and studies have suggested that winds from say 500 ft to ground level do not always decrease in strength due to friction, and in fact, some cases, can be stronger. However, this further shows that that are too many variables still and still more to learn before we can start using other ways to rate tornadoes in a standardized fashion.

That being said, it is w/o question many tornadoes in our database are stronger than the EF-scale shows, simply b/c they do not always hit something substantial, or do not hit anything substantial when they are at peak intensity. Studies have suggested that a large percentage of tornadoes may have EF2 winds or greater sometime in the existence.

Regarding wind speeds assigned in tornadoes vs. hurricanes, wind measurements are *in situ* in hurricanes, not inferred by damage. Recons are literally flying in winds exceeding 150 mph at times at the 700 mb level, and GPS dropsondes directly measure winds from FL all the way to the surface, giving us a much better idea of actual wind speeds. We also have had the SFMR since 2004, which is not an in situ measurement, but fully samples the surface winds in all four quadrants as the recon makes multiple passes . Combined this with GFS dropsondes, this gives us a solid picture.

 

[Randy Zipser]

From an engineering standpoint, actual wind speeds are of value, but the vast majority of the time, we are
not talking about that, we are talking about how it impacts society from a public POV. The vast majority of
our structures that exist fail well before 190 mph winds, so focus on exactly what the winds were when it crosses
some arbitrary threshold *we* assign is irrelevant for all practical purposes!

Excellent points all, Boris! However, regarding your observation in the quote above, it may not be particularly relevant from point of view of our society-at-large, but it may be very important to a specific individual who happened to be the victim of that tornado. Remember, we humans all seek an understanding, if not some form of closure, to any sudden, traumatic situation in our lives, like being the victim of a tornado (no matter the intensity rating), and having this one small, additional piece of information is one part of that healing/closure process. So, on the scale of the individual, that tornado wind speed number is not really irrelevant, at least in my view. But, it does make for an interesting point of discussion, as we all seek different ways of coping with a situation (like being a tornado or hurricane victim) that we may someday find ourselves in!

 

[RachelDonoghue]

Finally someone here makes a thread for this topic. I've been meaning to give my two cents on this. I find it incredibly difficult to believe that we haven't had a single EF5 tornado in over 12 years. I think the EF scale is flawed in my opinion. I'm just going to splice up my opinions in sections so it's easier to read here.

Regarding The Whole NWS Refusing To Rate EF5's Controversy:

I definitely believe that the NWS is finding any reason not to rate a violent tornado an EF5. (I by no means, wish for an EF5, but I don't think we should withhold a true rating, for the sake of science.) There have been violent tornadoes over the past 12 years that have had at least one indicator that would be EF5 level. Conspiracy has it that the NWS doesn't want to cause panic. The other conspiracy is that it's for insurance reasons and that the insurance companies just don't want to pay out. With global warming and climate change, I think storms are getting stronger and are happening more frequently.

Examples of tornadoes that come to mind here that should've been rated EF5 to me are Mayfield, Kentucky in 2021 and Rolling Fork, Mississippi. Details are hazy for me since I got back into weather after a decade long hiatus lol. Correct me please if I got dates and states wrong.

Regarding The El Reno 2013 Tornado Being An EF5 Instead Of An EF3:

I honestly thought it was accurate that they downgraded the El Reno tornado down to an EF3 later on. The tornado was a beast, and was immensely powerful. I have no doubt in my mind that things would've gone very differently if the tornado had tracked into OKC. It would've likely caused an even greater loss of life unfortunately, and It would've been an EF5 had it gone into OKC. Thankfully that never happened, and the tornado instead stayed over the countryside.

The winds for that tornado were well into the EF5 range, but the scale mostly goes off of damage, and not winds alone. Due to the lack of structures that it hit, that's what gave it an EF3 rating instead.

I think the rating for the El Reno tornado should stay at EF3.

Regarding Hurricanes:

Truthfully, I don't really have much to say here on this point, as I'm not as interested in hurricanes like I am in tornadoes. Hurricanes are cool storms, but I'm just not obsessed with them like I am tornadoes. I need to learn more about hurricanes honestly.

My Overall Opinion On The EF Scale:

I honestly am not a huge fan of the changes they've made to the original EF scale. I feel it just complicates things. Just my opinion. I think the EF scale they implemented in 2007 was fine. Truthfully though, I really did like the original F scale, and I feel that maybe we should bring that old scale back? Or at the very least, go back to the original EF scale of 2007.

The whole adding both a damage and wind rating is cool, but I also feel that it would only make things even more complex than they already are. This is just my opinion though.

Bonus:

I did think of another reason for the lack of EF5's though. What if another reason we just haven't seen any EF5's over the last 12 years is because our building codes have gotten better and structures are strong enough to prevent an EF5 rating? Just a thought. This thought truthfully doesn't make much sense though, as I feel if our buildings were better, that we'd have even less EF4 ratings... I'm just rambling at this point now lol.

 

[J Miller]

My completely unqualified opinion is as follows: I feel like we aren't going to see another EF5 until we get another Joplin scenario. Ideally, we never see another EF5. I think the biggest issue with the scale is the subjectivity, which the EF scale was originally meant to deal with. There were supposed to be hard limits on the minimum for DIs, ie a slabbed house was supposed to be an absolute minimum EF3 165. However, now we're seeing more and more subjectivity brought into ratings. I don't really disagree with these, per se, most of my disagreements are with overrating, not underrating. But there needs to be a reexamination of the wind spreads for various DIs, and IDEALLY an engineering driven investigation into tornado damage in structures. However, I'm more just frustrated with a lot of the limitations of damage based ratings as a whole. I feel like a single "wind threshold" is really a poor metric. We really aren't even sure how tornadoes actually do damage. Traditional wind analysis assumptions get completely invalidated in tornadoes. Some scientists have proposed that the quick shifting wind direction allows for damage to be done at much lower wind magnitudes than a traditional analysis would show. Every tornado is different, and so where the wind is "blowing" from in reference to the structure changes differently with time. Some tornadoes also will have a much stronger vertical component to their wind than others, especially drillbits and in suction vortices. Then you have to take into consideration exposure time to the wind, something that I personally can't even imagine how we'd investigate in a lab. ESSL research showed that structural failure in high winds was nearly instantaneous, but with long "dwell" time, there's a much longer window for an exceptionally strong gust to come along and whipe the floor with affected structures. Debris is also something that's really difficult to characterize, as the phyisics behind large debris lofting and debris penetration is a REALLY complicated physics problem. So much depends on the speed, angle and debris when it impacts the structure, and trying to parametrize the effects of debris is just going to become a nightmare. Now, the question is, what better option do we have? I genuinely have no idea.

 

[J Miller]

The concept you propose, Hannah, would be too confusing and not practical to implement, IMHO.

The original Fujita Intensity Scale ("F-scale," from 0-5) was based upon very subjective observation from post-event damage surveys, conducted mostly by a local NWS office. If anything, damage estimates before 2007 using the original Fujita Intensity Scale tended to overestimate what was actually observed. By 2007, wind-engineering research studies at several major universities, most notably Texas Tech University, had progressed to the extent that wind/structural engineers were confident that the damages being observed could be explained by lower wind speeds, based upon what they were observing under controlled-conditions in the laboratory. It was these observations that led to the F-scale being replaced by the Enhanced Fujita Intensity Scale ("EF-scale," from 0-5) on February 1, 2007, in the U.S.

The key point is: accurate tornado damage "intensity" ratings cannot (and should not) be based upon very random, capricious field observation by observers who are using a "trained eye" anywhere from untrained to experienced. They can, however, be based upon acceptable engineering practice and hypothesis (based upon a large sample of past case studies), where a common standard is applied in all cases.

Rather than scuttling the current EF-scale altogether, I believe a better approach would be to do a comprehensive review of the current EF-scale windspeed threshholds, and update this scale based upon the current availability of a much greater volume of case-study data, in this case, 18-years worth of additional structural-damage data (2007-2025). For simplicity, the "E" for "Enhanced" should be dropped. Then, the NWS should adopt a new Revised F-Scale ("RF-scale, from 0-5). It could even be further refined using a year number (as there could be additional revisions in future years), such as "25RF-0" or "26RF-5." Should future case-studies confirm that higher windspeeds are warranted, an "RF-6" level could be added to this classification.

BTW, according to Wikipedia, a research study in November, 2022, has indicated that a more-standardized EF-scale is already in development using mobile Doppler radar (not sure how many mobile Dopplers would be necessary to make this work!) and "forensic engineering" (probably like what Tim Marshall has been doing during his career at Haag Engineering).

Lastly, the reason that the "TIR" above would not work is that to be able to measure actual windspeeds (and these would not be at the ground), a mobile Doppler (e.g., DOW) vehicle would have to be available and in stationary position for scanning prior to tornado touchdown--something that is neither affordable, practical nor possible in the event of a widespread tornado outbreak.

Just my 2-cents worth...

Another issue with mobile radar is the process to deriving the wind measurements. A single radar can only measure inbound and outbound winds. Deriving tangential winds requires math, with a lot of empirical correction factors. This is why dual doppler analysis is preferred to single doppler when possible, because you can get a more accurate picture of 2d wind velocities. Even so, retrieving the max winds from a dual doppler setup requires some math, with empirical constants. How do we know if these weighing factors are accurate for tornadoes? Especially with the variability of radar measurements (distance, power, attentuation). The last thing to consider is that radar measurements are all finite resolution. Every pixel on radar is going to be "smudged" to an extent. There's points in that area represented in the reading that were probably higher and lower than what is said, however algorithms in the radar are going to average it out because of the finite accuracy.

 

[Randy Zipser]

I did think of another reason for the lack of EF5's though. What if another reason we just haven't seen any EF5's over the last 12 years is because our building codes have gotten better and structures are strong enough to prevent an EF5 rating?

Good point, Rachel. The original impetus behind the development of the EF-scale was to take whatever structural engineers could learn from observing building damage over a long period of time and apply that "new" knowledge toward improving building designs to better withstand winds from tornadoes (or, by extension of that same thought, hurricanes). The Greensburg, KS, EF-5 in May, 2007, was an excellent "first test"case for demonstrating that the above concept really works; the state-of-science for wind-engineer design (at that time) was incorporated into every structure that was rebuilt there--and that was literally the entire community. Greensburg and Joplin were both EF-5s, but in the four years between these events, a lot of structural-design knowledge was learned that was incorporated into the rebuilding of Joplin, but that was unavailable to Greensburg residents.

Fast-forward to 2025. What professional engineers have learned since rebuilding both Greensburg and Joplin has enormously improved high-wind-resistant building design. But a great deal of that "new knowledge" is already known. If we become complacent, we could fall into "a trap" of believing that we no longer need to keep monitoring the design data we already have, or worse, collecting additional data from current and future natural-disaster events.

Just as we have observed (and are measuring) changes in our long-term climate trends nowadays, short-term events such as tornadoes and hurricanes may also be evolving in subtle ways (i.e., getting stronger or more frequent?) over time. One could also speculate that improvements in building design are offsetting a warmer troposphere due to a changing climate, so that the EF-5 level has not really changed much over, say, the past decade; however, if building improvements are really getting so much better, that might also explain why the EF-5 level is now rarely reached, except for in a few extreme events. That's why a rigorous, objective analysis of more recent data is so necessary and we must follow with an open mind where the data take us. In other words, we must look ahead by using new data and technology to improve upon what we already know (the current EF-scale) rather than looking backwards in time when conclusions may have been flawed for lack of both sufficient data and technology (e.g., the F-scale days).

Lastly, it is vitally important for survival of humankind that we use every bit of knowledge in our arsenal to keep aware (or better still, get ahead) of such trends. Technology is never static, and we must make the best use of it and continue to learn from it--so that we can adapt to whatever Nature, our oceans, and our atmosphere have in store for us.

 

[Jamie H]

There are other ratings as well:

International Fujita scale - Wikipedia

en.wikipedia.org

TORRO scale - Wikipedia

en.wikipedia.org