Rating Tornado Intensity Based on Moblie Radar

[Jake Orosi]

So, you're telling me that rating the El Reno tornado EF-5 versus EF-3 is going to directly lead to an improvement in the warning system? I did not say that researching and record-keeping is pointless. It is important. However, if 20 people are killed by a tornado, does it matter if that tornado is rated EF-3 or EF-4? I appreciate the passion on both sides of the argument, but my point is that it seems like a small thing to argue over given the death and destruction. Don't get me wrong, as a meteorologist, the final rating is important to me, but I don't have a strong feeling either way about the use of radar in determining a rating, so I'm not going to get bent out of shape. It's definitely an interesting debate, and exciting, too, because it's demonstrating a great advancement and application in technology. In the area I work, the odds of me ever having a mobile radar measurement are fairly low. Yes, I have Doppler envy. There I said it.

I guess it all depends on your audience. Maybe I'm looking at things too broadly. I tend to always look at things on the large scale. How does the vast majority of my customers, city/state/county officials, media, and the general public view this? Five years from now, the majority of the general public probably won't remember what the El Reno tornado was rated or that mobile Doppler radar measurements were used to upgrade the rating.

I, on the other hand, am interested purely in the scientific data aspect. It doesn't matter to me what or whether the public thinks about EF-ratings. EF-ratings were not designed with the public in mind, but with scientific data collection in mind.

The EF rating is a wind speed classification, not a damage classification. An Doppler-measured EF-5 tornado that has "only done EF-3 damage" because there happened to be nothing in the damage path that would've lasted through any but EF-5 winds, is an EF-5, because the rating isn't about the damage, it is about the wind speed. The damage a tornado causes is usually how we estimate the wind speed; but over time this has caused laypersons to confuse the method with the purpose. And that persisting confusion is the only reason this debate is even occurring.

The debate is in fact reminiscent of the recent brouhaha over the IAU's reclassification of Pluto as a minor planet rather than a true planet. There wasn't a single argument for Pluto's retaining planetary status that wasn't an appeal to emotion or tradition, two things which should never direct the course of scientific progress.

The prioritization of radar measurement in the classification of tornadoes by wind speed is, similarly, only being opposed with appeals to emotion or tradition. The difference is, these people are even mistaken about the nature of the "tradition".

 

[Rob H]

I, on the other hand, am interested purely in the scientific data aspect. It doesn't matter to me what or whether the public thinks about EF-ratings. EF-ratings were not designed with the public in mind, but with scientific data collection in mind.

Actually, they were designed with engineering, and public safety, in mind - as Fujita's work was funded by NRC grants in the early 1970s. Ted wanted to classify storms by strength, but the only option he had was to survey damage, so that's how the scale was designed.

The EF rating is a wind speed classification, not a damage classification. An Doppler-measured EF-5 tornado that has "only done EF-3 damage" because there happened to be nothing in the damage path that would've lasted through any but EF-5 winds, is an EF-5, because the rating isn't about the damage, it is about the wind speed. The damage a tornado causes is usually how we estimate the wind speed; but over time this has caused laypersons to confuse the method with the purpose. And that persisting confusion is the only reason this debate is even occurring.

The F scale was a damage classification with associated and *estimated* wind speeds. The EF scale was introduced to better incorporate different types of structures/materials and degrees of damage. A tornado that does "EF-4 damage" can't be directly compared to a tornado with measured winds in the EF-4 damage range because the EF *estimated* winds haven't been rigorously verified. It's apples to oranges, and that's not a useful scientific comparison.

The debate is in fact reminiscent of the recent brouhaha over the IAU's reclassification of Pluto as a minor planet rather than a true planet. There wasn't a single argument for Pluto's retaining planetary status that wasn't an appeal to emotion or tradition, two things which should never direct the course of scientific progress.

The prioritization of radar measurement in the classification of tornadoes by wind speed is, similarly, only being opposed with appeals to emotion or tradition. The difference is, these people are even mistaken about the nature of the "tradition".

I don't see how emotion factors into the decision to continue using the EF scale the way it was designed, while leaving the possibility open to enhance the scale again to account for measured wind speeds. It's not tradition, it's the science and defined schema for the way it works, and adding new data to it with minimal discussion is decidedly unscientific. People keep throwing out that not incorporating these measurements is setting "scientific progress" back, but no one can provide valid reasoning for how exactly this is repressing new discovery or ideas. The data is all still there, easily accessible to anyone interested.

 

[Jake Orosi]

Actually, they were designed with engineering, and public safety, in mind - as Fujita's work was funded by NRC grants in the early 1970s. Ted wanted to classify storms by strength, but the only option he had was to survey damage, so that's how the scale was designed.



The F scale was a damage classification with associated and *estimated* wind speeds. The EF scale was introduced to better incorporate different types of structures/materials and degrees of damage. A tornado that does "EF-4 damage" can't be directly compared to a tornado with measured winds in the EF-4 damage range because the EF *estimated* winds haven't been rigorously verified. It's apples to oranges, and that's not a useful scientific comparison.

Be that as it may; the fact is that, outside of an ad hoc method for estimating wind speed, "tornado damage" is scientifically useless data. It depends too much on factors that cannot be controlled or even objectively assessed after the fact; even separate buildings within the same vicinity can have been built out of vastly different materials and to different construction standards. This near-arbitrariness in the amount of damage caused by any particular storm may make for an interesting local historical datum; but for research purposes it is of little value beyond face - "One day there was a really bad/really weak tornado in Unluckyville. The End."

Whereas, accurate wind speed data is just as useful as any other measured atmospheric data when it comes to analyzing and modeling conditions and trends in tornadogenesis and storm intensity.

If the EF-scale is to be put to continued use, I fully expect that it will have to be modified - again, and as often as necessary - as actual measured data is able to better inform our estimates of what wind speed is capable of inflicted what kind of damage. Without those numbers, the EF scale is arbitrary and useless to everyone - even engineers.

 

[Rob H]

I completely agree that it will need to be modified in light of scientific advancement, and there's no reason it shouldn't be. If it wasn't clear, my only argument against what they've done in the past is that they're essentially saying "I'm going to assign this tornado a damage rating based on measured winds that match up with an estimated range of winds for a particular damage rating". That's what seems wonky to me. By all means, they should sit down and hash out what mobile doppler means to EF ratings and how the data should be annotated.

I agree with your argument about accurate wind speed data, but El Reno 1, El Reno 2, Rozel, and whatever other storms have had EF rating affected by mobile doppler, are in the extreme minority and will therefore not be compared against other storms with similar EF ratings where no mobile doppler data was taken, so why bother assigning an arbitrary rating? When studying tornadogenesis the sample set won't be "EF-5 tornadoes", it will be "violent tornadoes with measured winds available" which includes a much more relevant data set.

 

[Chris Nuttall]

That's not the official NWS EF-scale document though

I will edit that post to be more specific.

 

[Dan Robinson]

Here are links to quotes during talks at the Storm Chaser Convention on Saturday that will be of interest to some. These are direct Youtube links to the point in the live stream archive where the quotes are spoken:

Dr. Howie Bluestein:
http://www.youtube.com/watch?v=Kzb-AIDCGC4#t=00h39m10s

Dr. Chuck Doswell:
http://www.youtube.com/watch?v=Kzb-AIDCGC4#t=03h02m30s

From what I understand, these stream archives of the convention won't be left up permanently. I'd recommend listening to the full talks soon if you get a chance.

 

[Jeff Duda]

I was happy to see the way Chuck talked through his portion of the talk about the dangers of chasing tornadoes. I think some people (including some on this forum) need a wakeup call about getting close, especially to particularly large/violent tornadoes. If anyone can give the call, it's Chuck.