Tornado "confirmed" by dual-pol radar signature

Bob Hartig said:
Today is the first I've heard of anything like it. Dick McGowan has a thread going about this on FB, and I'm surprised to see how much consensus there is that the Correlation Coefficient furnishes conclusive evidence of a tornado even without ground proof. The thinking is that the CC can discern debris, and a tornado is the only thing that can loft debris to the height of a radar scan. That makes sense, though I hate to admit it. I'm already not crazy about the phrase "Doppler-indicated tornado," and something in me just balks at the thought of "Doppler-confirmed tornado," or whatever it will be called.
Click to expand...

Honestly, though, a Tornadic Debris Signature is more than just Correlation Coefficient. The NWS has a "general" criteria for calling something a TDS. Typically you look for a CC value of 0.8 or less, but you also have to look at ZDR... typically you want a local minimum at about 0 in the same area of the observed CC value. The other two qualifiers are a strong SRV couplet and abnormally high reflectivity... all in the same spot. Now, these are not NWS-wide criteria... just suggestions from their organization wide training. So, they did not simply just confirm the tornado off only CC. I understand the hesitancy to rely on a TDS to confirm a tornado, but honestly... if you're experienced reading radar, and you have strong signals in each of the 4 categories stated above, its almost impossible for the signature to be created by anything other than lofted debris. And the only thing that is going to cause lofted debris that significant is a tornado.

Edit: Sorry, I did not mean to rehash much of what Jeff already stated... somehow I missed your original reply in the thread Jeff.
 
I get that, Zack. I'm not questioning the reliability of the CC when used as described in conjunction with BR and SRV. I'm just emoting. I haven't yet used the new GR3v2 with dual-pol products in the field; that will change this spring, but right now I'm not familiar with it firsthand. So while I can see that it's a huge step forward in terms of the reliability of warnings, I'll maintain a reserved stance about basing LSRs on radar products alone without ground proof. I'm open to being shown otherwise, but I'll need some convincing.
 
Last edited by a moderator:
I guess maybe the Lake Charles office actually did perform a damage survey. Hard to tell by the exact language, but I doubt they'd know all that detail without having gone to the site:

PUBLIC INFORMATION STATEMENT...CORRECTED
NATIONAL WEATHER SERVICE LAKE CHARLES LA
648 PM CST FRI FEB 22 2013

...NWS STORM SURVEY FOR STORM DAMAGE THAT OCCURED THURSDAY
FEBRUARY 21 2013...

...A SUPERCELL MOVED ACROSS VERNON PARISH AND INTO NATCHITOCHES
PARISH THAT PRODUCED AN EF-1 TORNADO. BELOW IS THE PRELIMINARY
SUMMARY OF THE TORNADO...


TORNADO 1: KURTHWOOD LA 4 E

RATING: EF-1
ESTIMATED PEAK WIND: 100 MPH
PATH LENGTH /STATUTE/: 2.5 MILES
PATH WIDTH /MAXIMUM/: 50 YARDS
FATALITIES: 0
INJURIES: 0

START DATE: 02/21/13
START TIME: 03:00 PM CST
START LOCATION: 3.6 MILES ESE OF KURTHWOOD/VERNON/LA
START LAT/LON: 31.325/-93.111

END DATE: 02/21/13
END TIME: 03:10 PM CST
END LOCATION: 4.3 MILES NE OF KURTHWOOD/VERNON/LA
END_LAT/LON: 31.365/-93.105

THE TORNADO TOUCHED DOWN IN A PINE FOREST SOUTH OF LA 465 AND
SNAPPED NUMEROUS TREES ALONG WITH UPROOTING A FEW LARGE PINE
TREES. THE DAMAGE PATH STARTED BETWEEN KURTHWOOD AND
HUTTON...THEN MOVED NORTH OUT OF VERNON PARISH AND INTO
NATCHITOCHES PARISH.
Click to expand...
 
It doesn't seem all that ambiguous to me; the report regards a specific "damage path" and describes some of the damage. The only two possibilities are that ground truth was obtained, or that the dimensions and characteristics of the "damage path" were completely made up. While the decision to "confirm" a tornado based solely on radar findings can be debated, it's hardly likely that the WFO would've invented a precise damage path and uprooted trees for their report.

For what it's worth, "Doppler-indicated tornado" doesn't bother me so much. I realize that its primary purpose is to motivate those unconcerned about a "thunderstorm capable of producing a tornado" into taking life-saving action. It's essentially the WFO's way of saying SERIOUSLY GUYS THERE'S REALLY A TORNADO WE CAN SEE IT FROM HERE GET IN YOUR BASEMENT ALREADY while still sounding professional and/or government-agency-like.
 
Jake Orosi said:
It's essentially the WFO's way of saying SERIOUSLY GUYS THERE'S REALLY A TORNADO WE CAN SEE IT FROM HERE GET IN YOUR BASEMENT ALREADY while still sounding professional and/or government-agency-like.
Click to expand...

No, that's not the case at all. "Doppler-indicated tornado" is simply a sign that the rotation and other evidence just crossed a threshold to "okay, this might really be a tornado." That's not the same as a TDS being discussed here.
 
I don't see what the problem is with leveraging technology to detect a "probable or possible" tornado ... especially in areas and days where few chasers and spotters are out, and especially in the wee hours of the morning when all are asleep and darkness makes visual confirmation virtually improbable in unpopulated areas. If there are winds or twisters intense enough to create a definite [debris] signature I think the NWS would be derelict NOT to issue a warning. You can bring up the issue about "crying wolf" too many times, but the dangers and issues of not warning trumps that, every time. The point is, something is creating havoc and showing up on radar. Tornado or micro-burst or whatever, it's presence and direction of travel and potential impact areas needs to be conveyed. Just my 2 cents.
 
I don't think anyone here is arguing that the weather service cried wolf in this scenario. As far as issuing a warning based on detection, this was a fairly clear-cut case: there were several indicators known to be associated with tornadoes appearing on radar at the same time in the same place. I would say confidence in issuing a warning was reasonably pretty high. However, it was the issuance of an LSR confirming that a tornado had occurred using only the radar signature as evidence that I thought was bold.
 
STexan said:
I don't see what the problem is with leveraging technology to detect a "probable or possible" tornado
Click to expand...

There's no problem at all with it. It's what the bulk of tornado warnings are, it has been that way for a long time, and it's a very good thing. I wouldn't own radar software if I didn't recognize its immense advantages. Since I'm the guy who said he's not crazy about the term "Doppler-warned tornado," let me say plainly that the benefits of Doppler-based warnings far outweigh any potential cry-wolf effect. No one here would ever say otherwise.

It's the term "Doppler-warned tornado," not the technology, that annoys me. But that has just been personal finickiness on my part. A tornado by definition requires surface rotation, and since such activity lies below the level of Doppler scans, all that Doppler has been able to verify till now has been the presence of above-ground rotation. To call that a tornado is not accurate. But such language is designed for the general public, not storm chasers, and there is no better solution. Moreover, the arrival of dual-pol seems to have significantly increased the degree of certainty for Doppler-based warnings.

In Michigan, where I live, the bulk of our tornado warnings are associated not with distinct supercells but with linear MCSs with transient circulations that rarely produce. They could produce, though, given the right conditions, and occasionally they do; and moreover, sometimes longer-lived mesocyclones are embedded in the line. Because a tornado, however brief, could conceivably pop up almost anywhere along a QLCS, large sections get tornado warned. (A couple years ago, I saw a long, skinny warning box that stretched from Kalamazoo north to Grand Rapids on an east-moving line.) As a storm chaser, I don't find such warnings very meaningful. But again, they're intended for the general public, not me, and I hope the public takes them with complete seriousness. After all, what better options do warning meteorologists have? None that I'm aware.

LSRs are a different matter, but there's no need for me to rehash what has already been said.
 
Last edited by a moderator:
With a good debris signature collocated with a good rotational velocity couplet and good thunderstorm reflectivity, what are the odds that this is indeed a tornado? Very high, in my opinion. Up into the high 90% range. I think that's sufficient for language like that. You won't be able to pin down the exact start and end points, nor the exact strength, but its almost certain than there was a tornado.
 
Sorry, I'm a little late to the party on this one. We had a discussion about this at WDTB after I noticed this LSR. Basically a case like this is treated like a spotter report and is acceptable. However, the LSR is not an official report and will not make it into Storm Data unless there is some sort of ground truth for the event to even be considered for Storm Data. Storm Data is the only official source for verifying events by the NWS.

As far as the warning process goes, myself and a former co-worker are keeping track of every warning statement that contains wording of a tornado confirmed by a dual-pol debris signature. A majority of the time the text in the warning will have something along the lines of "Radar CONFIRMS a tornado or a radar CONFIRMED tornado.

Bob, at the Training Branch, we take issue with the long, large tornado warning polygons as well and discourage the use of them in all of our training.
 
Thanks for chiming in! Your last line is critical... It should not just be a training issue. The software should simply reject it unless the forecaster clicks a lot of "red flag" icons reminding him many many times what a bad idea that is.
 
Bob Hartig said:
A couple years ago, I saw a long, skinny warning box that stretched from Kalamazoo north to Grand Rapids on an east-moving line.
Click to expand...

I don't mean to digress from the topic at hand but I'm pretty certain this is that warning you're talking about. Sorry my knowledge of pulling up past data is limited. I'm mobile or I would have pulled the shape file and posted an image. I was in Indiana at the time and was shocked to see a large portion of the squall tornado warned back in my home area. I believe there may have been additional polygons that accompanied this one overlapping the same time.

http://mesonet.agron.iastate.edu/wx/afos/p.php?pil=TORGRR&e=201010261445

EDIT: Here's an image...
102610TORGRR.jpg

That's over a 2400 square mile tornado warning!
 
Last edited by a moderator:
That was an interesting day... We had some good :) meeting between NWS and EM after that.

But it's been great for my business!
 
You must log in or register to reply here.
Back
Top