• Welcome to TalkWeather!
    We see you lurking around TalkWeather! Take the extra step and join us today to view attachments, see less ads and maybe even join the discussion.
    CLICK TO JOIN TALKWEATHER

Enhanced Fujita Ratings Debate Thread

Aaron Rider said:
All I know is that he listed both as F5 in his books.

Don't know as much about Blackwell.

As for Udall specifically, if that wouldn't be EF5 today, then it's a problem with the scale. Extremely high end context, plus homes utterly obliterated with slabs swept well past clean ... that tornado was, to put it mildly, "extremely" violent. There was a blog post somewhere comparing the Udall-Blackwell outbreak to Greensburg 2007. Very, very similar storm systems. The violence of Udall absolutely equalled Greensburg, at a minimum.

Edit: found post https://www.mikesmithenterprisesblog.com/2022/01/meteorological-mystery-newly-discovered.html

View attachment 39024
View attachment 39025

View attachment 39026
Click to expand...
My dad told me he saw the damage from the Udall 1955 tornado and the El Dorado 1958 tornado and he said the Udall tornado was significantly more intense than the El Dorado tornado.
 
  • Like
Reactions: AJS
Aaron Rider said:
My hot take is that there really weren't that many "overrated" F5s in the 40 years between 1950 and 1990. I am pretty confident these weren't:

Vicksburg - 12/5/1953
Belmond - 10/14/1966
Wheelersburg - 4/23/1968 (this one was borderline based on damage pictures)
Spiro - 3/26/1976 (although tossing 134,000 LB coal rail cars is pretty impressive)
Brownwood - 4/19/76 (it did cleanly sweep away homes, though, and shred mesquite trees so F5 doesn't sound crazy)
Broken Bow - 4/2/1982

Then we have more debatable ones:

Both 5/15/1968 tornadoes - Charles City and Oelwein - were rated F5 by Grazulis. I have not researched these. Homes are described as having been, "swept completely away" in both.

The Delhi, LA 1971 tornado is controversial but pictures posted in the significant tornado threads indicate both it and the Cary-Tillatoba, MS tornado were indeed QUITE violent.

5/6/1973 Valley Mills is accepted as an F5 by most major sources (including Grazulis) solely on the testimony of "wind engineers" who were extremely impressed by the strength it demonstrated in blasting pickup trucks through their.

Several of the 1974 Super Outbreak tornadoes would, of course, not be EF5 today but they were all (based, again, on what I've seen over the years in our significant events thread) certainly F5.

Fort Rice, ND (5/29/1953) was originally rated F4 by Grazulis, but he later upgraded it back to F5. "A large church was leveled and its pews were driven 4 FT into the ground. Parts of a car were carried for 1/2 mile." The church was a Catholic one, Immaculate Conception, and while those are often built very solidly, I don't know how solid that one was. Doesn't appear it was rebuilt (which implies the parish got consolidated or something). Truth be told, there's just not a lot of info on this one, at least from what I could find. https://www.tornadotalk.com/fort-rice-nd-f5-tornado-may-29-1953/

Plainfield 1990 is pretty controversial but it did scour the ground extremely and toss a fully-loaded semi over half a mile.

On top of that, we have the fact that several tornadoes in that period were underrated. Several from Palm Sunday '65, of course; Stratton, NE and Bakersfield Valley, TX in 1990; Custer County, NE in 1961; a few others.

I tend to think Buckeye had one of the best standards (because it's simple): cleanly swept house + extreme contextual damage = an extremely violent tornado, e.g. F5/EF5. So my second hot take is, tornado ratings will always be hard to do. But what we need is common sense. Some tornadoes were definitely overrated 1950-1990, but it seems like common sense was reasonably applied. And by straight common sense, we are obviously missing, in my opinion:

2011 - New Wren and Tuscaloosa; Goldsby and Washington
2014 - Vilonia
2015 - Rochelle and Holly Springs
2016 - Chapman
2023 - Matador

Plus *potentially* but most definitely not *certainly* another 9 (Washington IL, Louisville MS, Stanton NE, Pilger x2, Alpena SD, Camp Crook SD, Bassfield MS, Mayfield-Bremen)
Click to expand...
IIRC Stanton and one or two of the Pilger tornadoes pulled off extremely impressive damage that was probably on par with tornadoes like Chapman and Rochelle. You'd have to dig through the Sig-Tor Event thread to find them but I am pretty certain they, with reanalysis, could be completely solid F5s.

For an example, here is some extremely high end damage from one of the Pilger tornadoes (albeit which one it is I'm not sure):
talkweather.com

Significant Tornado Events

@eric11 Thank you for sharing those excellent damage photos. I don't recall ever seeing most of those. I'm also glad you mentioned something about the extreme tree damage near where the Pilger East/Pilger paths intersected. After checking out the DAT tracks, the patch of trees (location circled...
talkweather.com talkweather.com
 
Last edited:
With some of the advanced computer modeling available nowadays, I'm pretty sure it should be possible to construct a statistical model that takes the radar/velocity/height data along with geospatial data of towns/cities/infrastructure along the path and probabilistically estimating the expected (average) damage rating (EF score), along with a confidence range with probabilities of the likelihood that certain thresholds were (or were not) reached. It'd be an ambitious undertaking, but I feel like it's very doable with current technology.

Example:

Lee County AL 2019
Most Likely Damage Rating: EF 4
Likelihood of Meeting Each Damage Threshold:
EF0 - > 99%
EF1 - 97%
EF2 - 78%
EF3 - 61%
EF4 - 52%
EF5 - 17%

Just a thought.
 
wx_guy said:
With some of the advanced computer modeling available nowadays, I'm pretty sure it should be possible to construct a statistical model that takes the radar/velocity/height data along with geospatial data of towns/cities/infrastructure along the path and probabilistically estimating the expected (average) damage rating (EF score), along with a confidence range with probabilities of the likelihood that certain thresholds were (or were not) reached. It'd be an ambitious undertaking, but I feel like it's very doable with current technology.

Example:

Lee County AL 2019
Most Likely Damage Rating: EF 4
Likelihood of Meeting Each Damage Threshold:
EF0 - > 99%
EF1 - 97%
EF2 - 78%
EF3 - 61%
EF4 - 52%
EF5 - 17%

Just a thought.
Click to expand...

Have you seen this study? Seems similar to what you're referring to. Would be interested in your interpretation.

Comparison of Tornado Damage Characteristics to Low-Altitude WSR-88D Radar Observations and Implications for Tornado Intensity Estimation
 
Some interesting tidbits about concrete homes.

Market Growth:
From 2009 to 2019, the concrete-framed housing market saw a substantial increase of 258%, and its market share doubled from 5% to 10%.

Regional Focus:
Concrete masonry framed homes are more common in the South, with such homes making up 17% of all homes completed in the region.

Source

I still have a hunch they've been overlooked. I'm going to do some digging in Greenfield.
 
"Although the Fujita Scale has been in use for 33 years, the limitations of the scale are well known to the users. The primary limitations are a lack of damage indicators, no account of construction quality and variability and no definitive correlation between damage and windspeed. These limitations have led to inconsistent rating of tornadoes and in some cases an over estimate of tornado wind speeds. Thus, there is a need to revisit the concept of the Fujita Scale and to improve and eliminate some of the limitations."

A Recommendation for an ENHANCED FUJITA SCALE
- Wind Science and Engineering Center, Texas Tech University

It's funny this assumption was used as the entire basis for creating the EF Scale, but NOAA couldn't find one single instance of tornadoes being overrated when they converted EF scale ratings to the much higher F scale winds.

Comparison of F-scale DI wind speeds to radar near-surface wind estimates
1744172088741.jpeg

In fact, the F scale still under rates true tornado wind speeds. However, the true wind speed (solid line) is much closer to the rating wind speed (dotted line).

Relationship between radar-derived wind speeds and EF Scale DI wind speeds
1744172372639.jpeg

The EF scale relationship to true wind speeds is WAY further off with massive under-rating in violent tornadoes, and even some over-rating in the weakest tornadoes.

"61 of 194 EF-scale damage ratings (31%) fall two or more ratings below the radar-estimated EF-scale ratings"
"26 of 194 F-scale damage ratings (14%) fall two or more ratings below the radar-estimated F-scale ratings"

- NOAA/OAR/National Severe Storms Laboratory
 
Grand Poo Bah said:
"Although the Fujita Scale has been in use for 33 years, the limitations of the scale are well known to the users. The primary limitations are a lack of damage indicators, no account of construction quality and variability and no definitive correlation between damage and windspeed. These limitations have led to inconsistent rating of tornadoes and in some cases an over estimate of tornado wind speeds. Thus, there is a need to revisit the concept of the Fujita Scale and to improve and eliminate some of the limitations."

A Recommendation for an ENHANCED FUJITA SCALE
- Wind Science and Engineering Center, Texas Tech University

It's funny this assumption was used as the entire basis for creating the EF Scale, but NOAA couldn't find one single instance of tornadoes being overrated when they converted EF scale ratings to the much higher F scale winds.

Comparison of F-scale DI wind speeds to radar near-surface wind estimates
View attachment 39505

In fact, the F scale still under rates true tornado wind speeds. However, the true wind speed (solid line) is much closer to the rating wind speed (dotted line).

Relationship between radar-derived wind speeds and EF Scale DI wind speeds
View attachment 39506

The EF scale relationship to true wind speeds is WAY further off with massive under-rating in violent tornadoes, and even some over-rating in the weakest tornadoes.

"61 of 194 EF-scale damage ratings (31%) fall two or more ratings below the radar-estimated EF-scale ratings"
"26 of 194 F-scale damage ratings (14%) fall two or more ratings below the radar-estimated F-scale ratings"

- NOAA/OAR/National Severe Storms Laboratory
Click to expand...
i also like the 2 random 166+ mph tornadoes that got rated EF0.
they are both a 4 EF rating underrated of what they both should be.
meanwhile zero 2+ EF rating over rated tornadoes exist.
and all the by 1 EF rating over rated tornadoes are are in the EF0-EF2 group.

heck only 2 tornadoes got rated correctly in the EF3-EF5 group.
 
NewFoundWeatherNerd said:
If every tornado from 4/2/25 gets rated EF3-165 all hell will break loose.
Selmer, Lake city, etc
Click to expand...
Eh, to be fair, the only damage I’ve seen from this past outbreak that appears to actually fit EF4 criteria is Bay-Lake City. I’m guessing at least three tornadoes did actually reach violent strength but the incredible lack of damage photos due to the delays in surveying really makes it difficult to know which ones deserve high end ratings.
 
slenker said:
Eh, to be fair, the only damage I’ve seen from this past outbreak that appears to actually fit EF4 criteria is Bay-Lake City. I’m guessing at least three tornadoes did actually reach violent strength but the incredible lack of damage photos due to the delays in surveying really makes it difficult to know which ones deserve high end ratings.
Click to expand...
Agree. That’s the only possibility I see. Haven’t seen anything from selmer yet to indicate.
 
Kds86z said:
If lake city doesn’t get eF4, I will not be pleased. I mean I don’t go online and bash anyone but I would disappointed.
Click to expand...
Just dont go on Twitter for the next three days lol. But on a more interesting note; a friend of mine noticed they removed the "+" tag from a lot of these EF3s, specifically the aforementioned ones (so just "EF3" without official word).

Selmer did slab/sweep at least three homes; im unsure of their construction quality or anchoring situation. I can only guess. But the degree of wind rowing from that damage is noticeable.
 
Selmer was weird because I feel like the contextual damage was very impressive but I didn’t see much on the home damage front as far as violent damage candidates go. But it looked like there was some scouring and maybe even a bit of granulation with it.
NewFoundWeatherNerd said:
Just dont go on Twitter for the next three days lol. But on a more interesting note; a friend of mine noticed they removed the "+" tag from a lot of these EF3s, specifically the aforementioned ones (so just "EF3" without official word).

Selmer did slab/sweep at least three homes; im unsure of their construction quality or anchoring situation. I can only guess. But the degree of wind rowing from that damage is noticeable.
Click to expand...
I haven’t seen the imagery of any slabbed homes in Selmer - would like to see that.
 
slenker said:
Selmer was weird because I feel like the contextual damage was very impressive but I didn’t see much on the home damage front as far as violent damage candidates go. But it looked like there was some scouring and maybe even a bit of granulation with it.

I haven’t seen the imagery of any slabbed homes in Selmer - would like to see that.
Click to expand...


Not sure on construction of these obviously, but this type of wholesale destruction definitely suggests something into the EF4 range vs. EF3 IMO.
 
You must log in or register to reply here.
Back
Top