Tornadoes that should have been rated F5/EF5

Oct 14, 2015
27
4
1
Oakville, Ontario, Canada
I thought I'd put together a list of tornadoes that probably should have been rated F5 or EF5, mostly due either to extreme recorded winds or to inconsistent or overly conservative ratings. I'm almost certainly missing some, so feel free to add to the list:

NOTE: List is ongoing, will add more tornadoes later

Mayflower, Arkasnsas - April 27, 2014
El Reno, Oklahoma - May 31, 2013 - based on extreme peak winds
Bennington, Kansas - May 28, 2013 - based on extreme peak winds
Goldsby, Oklahoma - May 24, 2011
Blanchard, Oklahoma - May 24, 2011
Tuscaloosa, Alabama - April 27, 2011
Dunlap, Tennessee - April 27, 2011
Bowdle, South Dakota - May 22, 2010 - based on extreme peak winds
Mulhall, Oklahoma - May 3, 1999
Loyal Valley, Texas - May 11, 1999
Kellerville, Texas - June 8, 1995
Pampa, Texas - June 8, 1995
Red Rock, Oklahoma - April 26, 1991 - based on extreme peak winds
Bakersfield Valley, Texas - June 1, 1990
Daullatpur-Saturia, Manikganj District - April 26, 1989**
Saragosa, Texas - May 22, 1987

**I know that the area was basically slums and wilderness, but apparently the tornado debarked trees and demolished houses within a 6 square kilometer radius of the funnel - this thing should have been rated F5 for sure.
 
Feb 12, 2013
56
1
6
The thing is scientists give the rating based on damage alone because they need a constant way to give those ratings. Would be nice if they would allow to give measured wind speeds a say in what to give a tornado in its rating.

The Mayflower tornado was not given an EF-5 rating because it was found out the house that was reduced to a concrete slab was not secured with nuts and washers. Plus the cars close to the house where not moved and thus it was given a lesser intensity rating.

The El Reno tornado was mostly over land that had little to no structure. Thus it was just given that EF 3 rating.

Maybe we should have a different scale based on measured wind speed and radar intensity? Not to replace the EF scale but something to work with it. A scale that gives a tornado a damage POTENTIAL rating.
 
  • Like
Reactions: Greg McLaughlin
Aug 9, 2008
84
6
6
Pueblo, CO
I'll give a nod to the Saragosa, TX tornado. Not sure too many of today's chasers have been to Saragosa - I have - but there are not very many sturdy structures in that area and I'm sure they weren't too much more sturdy in 1987. A slow-moving strong tornado like that in a more populated area could have likely inflicted EF-5 damage upon further survey and keep in mind we know a hell of a lot more in 2015 than we did 28 years ago in that regard.

Rob - I'd love to see that once finished. I'm sure we all will at the same time, but hopefully they will include real-time data in their final conclusions/revisions as you listed.
 
Dec 13, 2003
612
65
11
64
La Plata, Maryland
La Plata, Maryland - April 28 2002. The tornado wiped out 65% of the town. One house was complety leveled and left nothing but the foundation, but was found to have been not anchored down properly. First it was given an F-4 rating than raised to F-5 But later it was officially rated F-4 even though some sites still list it as an F-5.
 
Dec 13, 2003
612
65
11
64
La Plata, Maryland
Winds were believed to have reached F-5 briefly and the National Weather Service initially rated the tornado an F5 but, subsequent surveys of the damage by structural engineers ( including Tim Marshall) believed the destruction was "more consistent " with F4 damage even though wind speeds may have exceeded 260.. at one point, but yeah I guess you're right. Given the old F scale, by damage I'm sure they got it right. I remember there was quite the back & forth over this for some time though!
 
Jan 12, 2015
51
50
11
Have any info on the Dunlap tornado? I've heard case after case as to why Tuscaloosa should be rated EF5 but I've heard very little about Dunlap from 4/27/11.
 

Shane Adams

Is this thread based on the desire to have measured winds be a part of the rating process? If so, I was unaware that many tornadoes had been sampled. Or, is this just disagreement with the assessment of the professionals who are the world's leading experts at rating tornado damage?
 

Jeff Duda

EF6+, PhD
Staff member
Supporter
Oct 7, 2008
3,544
2,619
21
Denver, CO
www.meteor.iastate.edu
I'm curious as to what the purpose of this thread is. The aim of the science of using the (enhanced) Fujita scale to rate tornadoes is to estimate the wind speeds in the tornado by sampling damage produced by the tornado. The obvious issues with the EF scale include 1) there need to be sufficiently strong DIs as to bound the damage produced by the tornado, and 2) the tornado must actually impact DIs. Barring the few recent tornadoes in which mobile Doppler radar measurements indicated EF5+ winds within the tornado, why do you think there exists a list of "should be" EF5s? Are you saying you think NWS personnel who rate tornadoes intentionally reduce the final released EF-scale rating? I know there has been some controversy lately with the policy of incorporating mobile radar observations into EF-scale ratings, but to my knowledge, the database of tornado ratings has been rather consistent over time. Keep in mind that the official measurement for observed wind speeds in tornadoes is the maximum 3-second gust at 10 m AGL. While no one would reasonably argue against the fact that there were winds well in excess of 200 mph in the 31 May 2013 El Reno tornado, those measurements were not taken at 10 m AGL; rather they were obtained higher up. So really, the EF3 rating that tornado received is really the best anyone can do with the current state of the science.

Perhaps instead of referring to these tornadoes as "should have been EF5s" we could call them tornadoes that likely had EF5 wind speeds near the surface.
 
Last edited:
Mar 2, 2015
17
20
0
Woodbury, MN
I think the EF scale is just fine the way it is. Jeff hit the nail on the head with this one. That being said, any time a violent tornado goes through a populated area it's tragic no matter what it's rated. It's something that shouldn't even be debated when lives are lost. EF-0 or EF-5. WHO CARES? Some people give chasers crap for seeing a birdfart, but sometimes those birdfart tors can cause some serious damage, and loss of life. Sometimes there's huge wedges that don't touch a damn thing. It's all the same. A tornado is a tornado.
 

rdale

EF5
Mar 1, 2004
7,504
1,077
21
52
Lansing, MI
skywatch.org
Well, let's not go that far. There is a BIG difference between EF0 and EF5. Those who research tornadoes and develop new warning methodology to keep people alerted care. Does 4 versus 5 matter? No. But 0 to 5? Yep. Bigtime.
 

Tim Marquis

Jeff,

Although I agree that I dont see what this post accomplishes. Its like any number of penalties that should or should not have been called in a football game, you can't retroactively go back and decide. I will say I disagree with the following "...some controversy lately with the policy of incorporating mobile radar observations into EF-scale ratings, but to my knowledge, the database of tornado ratings has been rather consistent over time." The rating system hasn't been consistent over time. In fact its probably one of the least consistent data sets we have. Take for example the Kellerville, TX tornado on June 8, 1995. Did you know that V1 found a house that not only was swept clean but 1/2 the foundation was lifted by the tornado? The house was "likely missed by the NWS survey team." Hence the tornado got rated an EF4 vs an EF5. You tell me, if a tornado can physically lift 1/2 a concrete slab and later lift pavement off a Farm to market road if that constitutes a 10 on the DI scale.

I think the work to incorporate more modern technology to estimate ground level wind speed should be included when the data sets are available to complement the traditional rating scale. We shouldn't view having observations as a detriment to the rating network for the sake of consistency (ie Rosell May 18,2013 being EF4 vs EF2, El reno May 31,2013 being rated EF5 vs EF3 due to mobile radar data). That's like saying I'll stick with a Apple I for my working computer to stay consistent with my work.
 
Feb 22, 2015
155
210
11
Norman, OK
The Mayflower tornado was not given an EF-5 rating because it was found out the house that was reduced to a concrete slab was not secured with nuts and washers. Plus the cars close to the house where not moved and thus it was given a lesser intensity rating.
There were numerous structures in Vilonia that were missing from the DAT survey (at least looking at the overhead shots along the track). In addition, there was a home that was properly anchored there that was completely swept away, but they chose not to upgrade due to tertiary evidence, which was questionable at best (and they seemed to go to some length to avoid the EF5 rating). I've had this discussion several times in different places and the conclusion reached by the majority every time was there is a lot of inconsistency between different NWS offices and how the EF-scale is used (and that Vilonia was indeed an EF5). Some digging was also done and found that the person "in charge" of rating the tornado (J. Robinson) had a rather unrealistic view of how 1/2 family residences should be treated in the highest DI level.

For the record, aside from the radar windspeeds (which is a whole different issue), the only tornadoes that have been listed here that probably could/should/etc. have been F/EF5 are Vilonia, Kellerville and the 5/24/11 ones. Adding on to Jeff's final sentence there, we can probably say that most of these tornadoes listed were at EF5 strength at some point, but that's an entirely different thing than having the damage/evidence left behind to prove it.
 
Oct 14, 2015
27
4
1
Oakville, Ontario, Canada
Is this thread based on the desire to have measured winds be a part of the rating process? If so, I was unaware that many tornadoes had been sampled. Or, is this just disagreement with the assessment of the professionals who are the world's leading experts at rating tornado damage?
I'm not just taking any violent tornado and saying "it should have been (E)F5. Every tornado I've listed was based either on it containing 200+ mph winds, or based on its rating being disputed by at least a few experts. I'm not asking for a revision of the EF scale or a re-rating of any of the tornadoes so much as I'm listing tornadoes which were likely of F5/EF5 intensity at some point but never confirmed or given an official rating.
 
Jan 14, 2011
2,941
2,759
21
St. Louis
stormhighway.com
I know we have a separate thread on El Reno, but from what I understood, the consensus within the science/research community was virtually unanimous on EF-5, with principal figures like Bluestein and Doswell asserting there was "no question" it was EF-5. (See their Chasercon talks for those quotes) Has this opinion changed within the research community?
 
Oct 10, 2004
1,264
370
11
37
Madison, WI
IMHO, Vilonia/Mayflower '14 was the most blatantly underrated violent tornado of the EF-scale era. As Andy B. notes, they used "tertiary" evidence to justify rating textbook EF5 damage below EF5.

Tuscaloosa '11 was very borderline, and a good case could be made for either rating. Personally I'd have probably gone with EF5 just in deference to the fact that the tornado did high-end EF4 damage in both metro areas it impacted, and appeared to intensify based on radar while in the rural areas in between. Then again, that's why I'm not a meteorologist nor a wind engineer.

I've heard a good case made that if there should have been a fifth EF5 on 4/27/11, it should have been the Ringgold, GA storm which would have made it the first and only E/F5 in Georgia history.
 
Feb 22, 2015
155
210
11
Norman, OK
I've heard a good case made that if there should have been a fifth EF5 on 4/27/11, it should have been the Ringgold, GA storm which would have made it the first and only E/F5 in Georgia history.
A lot of the upgrade talk for that tornado was based on this photo of the Cherokee Valley Rd area...

060fde24d4dfe000c05a71a31c4a81fe.jpg

It was found that most of the homes in this area were not properly anchored, so EF5 was not given, although this is still obviously extreme damage. FFC does tend to be a conservative office in general, there was a tornado in Polk/Floyd Counties on 3/15/08 that was looked to be a strong candidate for EF4, but was rated EF3.
 
Oct 14, 2015
27
4
1
Oakville, Ontario, Canada
It's the lack of consistency between research centers that causes this problem. What one center calls a high-end EF3, another calls a low-end EF4. Whether to go with liberal or conservative damage ratings is a debate best left to the experts.