Severe WX Severe Weather Threat 3/14-3/16

[Aaron Rider]

Without a doubt EF-5 contextuals (or damage, if you count foundation damage as a DI)

If I'm understanding the tweet correctly, the debris on that foundation was cleaned up post-tornado, not swept away by the storm. Is that correct?

For whatever it's worth, ChatGPT's opinion of these pictures is, "The damage seen here suggests extreme winds, but not necessarily the highest-end EF5 behavior." And "Unless there’s additional evidence (such as a well-anchored home being fully swept clean with no remaining debris), I’d say the Diaz tornado was correctly rated EF4. It’s possible some localized EF5 wind speeds occurred, but the overall damage doesn’t fully support an EF5 designation."

But ChatGPT hasn't seen pictures of the cars the tornado apparently mangled. Well, like I said, it definitely flirted with EF5. Buckeye's right, though, this isn't a case to get outraged over in my opinion.

 

[AJS]

If I'm understanding the tweet correctly, the debris on that foundation was cleaned up post-tornado, not swept away by the storm. Is that correct?

For whatever it's worth, ChatGPT's opinion of these pictures is, "The damage seen here suggests extreme winds, but not necessarily the highest-end EF5 behavior." And "Unless there’s additional evidence (such as a well-anchored home being fully swept clean with no remaining debris), I’d say the Diaz tornado was correctly rated EF4. It’s possible some localized EF5 wind speeds occurred, but the overall damage doesn’t fully support an EF5 designation."

But ChatGPT hasn't seen pictures of the cars the tornado apparently mangled. Well, like I said, it definitely flirted with EF5. Buckeye's right, though, this isn't a case to get outraged over in my opinion.

I agree. It’s up to LZK at this point though. I wouldn’t necessarily disagree if it was rated EF5. Damage seems to be noticeably more intense than a lot of our higher end tornadoes in the last two years.

 

[Aaron Rider]

I agree. It’s up to LZK at this point though. I wouldn’t necessarily disagree if it was rated EF5. Damage seems to be noticeably more intense than a lot of our higher end tornadoes in the last two years.

Honestly, though, I didn't show it the pictures of that airplane and a few other things. I just saw those in the thread.

One thought that just went through my head was, we hear about "low-end" EF4 or "high-end." Certainly if you were going to argue for the existence of "low-end" EF5, well, you get my point...

But, yes, ColdFront is right about not letting the ratings game get to us.

 

[tornado examiner]

(Boo)

 

[AJS]

Honestly, though, I didn't show it the pictures of that airplane and a few other things. I just saw those in the thread.

One thought that just went through my head was, we hear about "low-end" EF4 or "high-end." Certainly if you were going to argue for the existence of "low-end" EF5, well, you get my point...

But, yes, ColdFront is right about not letting the ratings game get to us.

I agree. I am very pleased with LZK and their surveying. They have regained a lot of my trust and have proven to me they’re dedicated to giving out fair and warranted ratings again!

I really applaud them for their work this outbreak. They have taken the time to realize the tornadoes that touched down had violent damage indicators and properly rated them as such.

 

[slenker]

Honestly, though, I didn't show it the pictures of that airplane and a few other things. I just saw those in the thread.

One thought that just went through my head was, we hear about "low-end" EF4 or "high-end." Certainly if you were going to argue for the existence of "low-end" EF5, well, you get my point...

Regardless of ChatGPT's input, I would agree with this being rated either 190 EF4 or a baseline >200 EF5. Most definitely fits the criteria of a tornado almost perfectly straddling the line to me. I still would like to see just a bit more before saying it should receive the rating, like any ground scouring or extreme tree/forest damage.

 

[AJS]

Regardless of ChatGPT's input, I would agree with this being rated either 190 EF4 or a baseline >200 EF5. Most definitely fits the criteria of a tornado almost perfectly straddling the line to me. I still would like to see just a bit more before saying it should receive the rating, like any ground scouring or extreme tree/forest damage.

The ground in the background looks to be scoured from what I can see. Can’t tell for sure though.

 

[andyhb]

Would like to see an aerial survey of this path since it seems there was other EF4 damage along the path per the description from LZK.

 

[Maxis_s]

If I'm understanding the tweet correctly, the debris on that foundation was cleaned up post-tornado, not swept away by the storm. Is that correct?

For whatever it's worth, ChatGPT's opinion of these pictures is, "The damage seen here suggests extreme winds, but not necessarily the highest-end EF5 behavior." And "Unless there’s additional evidence (such as a well-anchored home being fully swept clean with no remaining debris), I’d say the Diaz tornado was correctly rated EF4. It’s possible some localized EF5 wind speeds occurred, but the overall damage doesn’t fully support an EF5 designation."

But ChatGPT hasn't seen pictures of the cars the tornado apparently mangled. Well, like I said, it definitely flirted with EF5. Buckeye's right, though, this isn't a case to get outraged over in my opinion.

ChatGPT isn't good with recognizing tornado damage from images (or recognizing most images at all). It probably doesn't recognize the cracked/uplifted part of the foundation.

 

[AJS]

Would like to see an aerial survey of this path since it seems there was other EF4 damage along the path per the description from LZK.

They are certainly taking their time so far with it. They’re doing a great job.

 

[Kds86z]

I just can't believe this outbreak happened. Feels like a dream (or nightmare).

I also can't believe that we had a near EF5 tornado in MARCH, in the first real significant threat of the year. I just can't wrap my head around that right now.

Same… I don’t know how we beat this outbreak this year…

 

[Aaron Rider]

ChatGPT isn't good with recognizing tornado damage from images (or recognizing most images at all). It probably doesn't recognize the cracked/uplifted part of the foundation.

Oh, it did, it just said it had no idea what the quality of the foundation was.

Make no mistake, it has its limitations. I actually always ask it to argue with me because I find it has a tendency to back up our innate biases. Anyway, I thought it was interesting.

 

[slenker]

Make no mistake, it has its limitations. I actually always ask it to argue with me because I find it has a tendency to back up our innate biases. Anyway, I thought it was interesting.

Little side comment, but I have found this to absolutely be true when I ask it questions involving my physics/astronomy studies. I've input questions that tell you what the answer you should get is, and it will literally just do wrong math to get to that answer and say, "see? My way was right." Lol. I always make sure to tell it it did it wrong and that seems to sometimes work.

 

[AJS]

Gonna be interesting to see a flyover of this damage path like Andy said. Wonder if there are more areas of intense damage they’ve just discovered.

 

[tornado examiner]

Gonna be interesting to see a flyover of this damage path like Andy said. Wonder if there are more areas of intense damage they’ve just discovered.

It’s this area of damage. EF4 damage was found in this area

 

[AJS]

It’s this area of damage. EF4 damage was found in this area

After seeing every bit of damage from this tornado so far, I am confident in saying it is a lot more intense than some of the recent higher end tornadoes we’ve seen such as Rolling Fork and Greenfield.

 

[AJS]

It’s this area of damage. EF4 damage was found in this area

Any more damage photos from this area?

 

[Grand Poo Bah]

Without a doubt EF-5 contextuals (or damage, if you count foundation damage as a DI)

This is incredible, and I don't disagree with anything anyone is saying, but from a purely objective standpoint i'm seeing a lot of bulldozer tracks and evidence of clean up in these pics. I wouldn't doubt the foundation damage was even done by a bulldozer. It's very painful for me to play devil's advocate here, because i'm the biggest critic of the EF scale there is, so I would love to be proven wrong on this.

 

[AJS]

This is incredible, and I don't disagree with anything anyone is saying, but from a purely objective standpoint i'm seeing a lot of bulldozer tracks and evidence of clean up in these pics. I wouldn't doubt the foundation damage was even done by a bulldozer. It's very painful for me to play devil's advocate here, because i'm the biggest critic of the EF scale there is, so I would love to be proven wrong on this.

I know it’s hard to tell, but I can’t see any bulldozer tracks in the grass/land in the far background. I could be wrong though.

 

[MMWX]

If I'm understanding the tweet correctly, the debris on that foundation was cleaned up post-tornado, not swept away by the storm. Is that correct?

For whatever it's worth, ChatGPT's opinion of these pictures is, "The damage seen here suggests extreme winds, but not necessarily the highest-end EF5 behavior." And "Unless there’s additional evidence (such as a well-anchored home being fully swept clean with no remaining debris), I’d say the Diaz tornado was correctly rated EF4. It’s possible some localized EF5 wind speeds occurred, but the overall damage doesn’t fully support an EF5 designation."

But ChatGPT hasn't seen pictures of the cars the tornado apparently mangled. Well, like I said, it definitely flirted with EF5. Buckeye's right, though, this isn't a case to get outraged over in my opinion.

ChatGPT is fun to play around with, but in it's current state it cannot accurately assess tornado damage.