Tornado Emergencies

[Mike Smith]

Chad,

You make an valid point.

I was on the air for WKY TV and Radio (now KFOR TV) in OKC on June 8, 1974, the day OKC was actually hit by 4 different tornadoes, so I certainly understand the problem of 'warning fatigue.'

I believe the level of urgency can be conveyed through appropriate wording of the warnings and severe weather statements as well as through the frequent issuance of LSR's.

For clarity, I am all for letting the public know a large tornado is on the ground and moving toward locations XXX and YYY. I am also in favor of using appropriately stronger language when Doppler shows a rotational shear of 200+knots.

That can be done without setting up a de facto product called a "tornado emergency." By doing a TE, we are implying there are "non emergency" tornado warnings, which I believe is potentially dangerous.

 

[rdale]

Maybe instead of issuing tornado emergencies the NWS should give continuous updates stating that a tornado is on the ground and it's current progress as they hear about it from spotters and tack it on radar? Sort of go live on the air?

That's what TV stations do... Given the listenership of NOAA Weather Radio (1-2%) I don't think that is good use of their personnel.

 

[Jarrod Cook]

Maybe instead of issuing tornado emergencies the NWS should give continuous updates stating that a tornado is on the ground and it's current progress as they hear about it from spotters and tack it on radar? Sort of go live on the air? (instead of having the robot talk about it and giving statements that are 10 or 20 minutes old not that I'm criticising the covereage that night on Noaa) I'm sure a situation like Greensburg warrants this, I mean it's a tornado that's over a mile wide heading for a town! Drop everything and cover that live and you won't need a tornado emergency statement! Many people are so accustomed to nothing happening during a "regular" tornado warning that they just keep going about their regular business. Some situations just deserve more attention than others and the warning process needs to be adapted for this. Most people need a confirmation that something bad is happening or will be happening to get their attention and to take cover. If you hear that the neighbors house a mile or 2 away is being hit (just for an example) you are much more likely to take cover than if it's only "radar indicated".

The problem with this is, you're saying if there is a large tornado confirmed on the ground, NWS employees should drop what they're doing and cover that live on NWR, etc. However, they have more ground to cover than just one county/city. In the case of last night, SGF had tornado warnings for multiple counties. You can't let one situation distract you from the rest of the CWA. That goes back to the whole "why is one tornado more important than the others."

Another thing that I thought of that has not been mentioned (that I know of) is this: I think another reason these statements are being used is because the NWS knows how the general public thinks. The general public thinks that they are much safer from a tornado in a large metropolitan area (although we all know that is not the case). This is just another tool for the meteorologists to portray the threat to those who think the large city will protect them from the tornado.

Also, another reason I believe they are using the statements goes back to what Jeff S. said about the FAR...the public, although they may not know the exact numbers, does know that there is usually not a verified touchdown nearby when a tornado warning is issued, so if the NWS thinks they can save more lives using stronger wording, more power to them. There is a lot of psychology in this stuff...

 

[Craig Maire II]

Rdale, where did you get the 1 to 2% listenership of Noaa Weather Radio? I just believe that a large damaging tornado heading right for a populated area is more important than golf ball sized hail and 70mph winds. Which is more likely to kill more people a l.7 mile wide EF5 or large hail and damaging winds? Also, many TV stations do a poor job of handling severe weather coverage.

 

[Mike Smith]

Rdale, where did you get the 1 to 2% listenership of Noaa Weather Radio?

Every scientific, independent study that has been done has come up with a number of 1 to 3% of the public gets its warnings from NWR. This has been covered in previous threads on this board.

It is so well documented that Liz Quetone of NOAA uses that number in her presentations.

 

[Mikel Shively]

SGF did a wonderful job last night in an extremely dangerous and long-lasting situation. We can sit here on a bulletin board all day and debate whether or not they should have issued the TE. But the fact of the matter is, they are only as good as the data they receive and the report came in saying that a trained spotter was tracking a large, damaging tornado into a city of 150,000. They reacted, and for better or worse, decided a Tornado Warning would not sufficiently convey the danger of the situation.

Agreed!

As someone who works in a county EOC and has direct control on the button to turn on all the sirens in my county, a Tornado Emergency has quite a bit of meaning. At that point I would be putting surrounding counties' on stand-by for additional aid. Beyond that, Im not sure really how much more effective a TE is to the general public...the enhanced wording is great, but im sure to the avg. public, it probably doesn't mean a whole lot...they are already supposed to be taking cover... if everyone waited for a TE to be issued, alot more people would be injured and/or killed every year. The TE doesn't trip the NWR's again, so to people with NWR and using that as primary warning would not know about it at all unless they had it on constant-on.

Maybe run it under a different header like a civil danger warning or something similar so it would trip the NWR's again?

 

[Craig Maire II]

Mike Smith wrote:

'Every scientific, independent study that has been done has come up with a number of 1 to 3% of the public gets its warnings from NWR. This has been covered in previous threads on this board.
It is so well documented that Liz Quetone of NOAA uses that number in her presentations."

So since that's the case why should we even have a Noaa weather radio program basically nobody listens to it? We should all just watch the local news to get our info and save public money for more useful projects.

 

[Bob Hartig]

This is an interesting discussion. I'll toss in my two cents, not from the perspective of a met, but from that of a copywriter who works daily with words.

My first observation is, shouldn't the main concern be with protecting the public? The average citizen isn't in on this debate. But chances are, he or she has heard enough tornado warnings to become a bit inured to the term. Maybe it should galvanize people into instant action, but does it? Realistically, warnings issued on a county-wide basis and regularly conditioned with "Doppler-indicated tornado" (translation: no tornado has actually been sighted) have probably already weakened the impact of the term "tornado warning" enough that the unofficial, enhanced "emergency" wording is unlikely to do it further harm. Isn't that really why the term "tornado emergency" got coined in the first place--because, faced with a real and immediate potential for catastrophic loss of life, and unwilling to accept that many people would die because they'd blow off the standard warnings, some forecaster or media personality chose to use language different and strong enough for the crisis at hand? (As I write, I'm thinking of the 1966 Topeka tornado, and of the now-famous words of the radio announcer who alerted listeners to the disaster heading toward them across Burnett's Mound. "For God's sake, take cover!" might not be very professional, but it got the job done.)

You can talk about how things ought to be all you want, but the acid test is how things are. Again, I'm no met. But the very fact that someone coined the term "tornado emergency," and that the term has since been used on a number of occasions, tells me that the standard language isn't strong enough for some situations. Sure, there's room for abuse of the TE term. But unless and until the term "tornado warning" is refined so that it actually means "ACT NOW OR YOU WILL MOST LIKELY DIE!!!"--and effectively communicates this meaning to the public, not to meteorologists--there needs to be room for trained forecasters faced with an exceptional situation to use exceptional wording.

You know what would make for some truly interesting discussion on this issue? I say this only half ironically: invite some of the citizens of Greensburg to share their thoughts.

 

[Mike Smith]

some forecaster or media personality chose to use language different and strong enough for the crisis at hand?

You know what would make for some truly interesting discussion on this issue? I say this only half ironically: invite some of the citizens of Greensburg to share their thoughts.

Actually, the first TE was issued by the Norman office as the May 3, 1999 tornado was approaching Moore and south OKC. In my opinion, it was completely unnecessary as KFOR TV and KWTV both had their helicopters following the tornado giving blow-by-blow reports. Scientific surveys after the tornado (see: http://ams.allenpress.com/perlserv/....1175/1520-0434(2002)017<0577:RTWDTM>2.0.CO;2 ) indicated that 89% of the population got their warning from TV and 3% from NOAA Weather Radio. I have a tape of the KFOR TV coverage and the words, "tornado emergency" are not uttered (unless the tape was edited without being labeled as such).

While I certainly would not be opposed to sitting down with the residents of Greensburg, news coverage does not indicate the "tornado emergency" was critical:

From: http://edition.cnn.com/2007/EDUCATION/05/06/transcript.mon/index.html

[Jeff] FLOCK [CNN] You got some warning?

[Steve] HEWITT [Greensburg City Manager]: Yeah, we had excellent warning. Our sirens went off for 20 minutes plus.

Further, see:
www3.whdh.com:80/news/articles/national/BO51386/

"We had plenty of warning," said Gary Goodheart, whose house was gutted with only a few walls left standing. "If people paid attention to sirens they should have been able to get to a safe place"

I can post at least a dozen references to the sirens and to TV warnings in the press coverage of the Greensburg aftermath and none mention "tornado emergency." There are lots of storm chasers and meteorologists who talk about the TE, but few, if any, residents or officials quoted mention it.

Let me emphasize that I believe the warnings issued by OUN on May 3 and DDC on May 4 were superb.

But, I believe, in both cases, the TE's were unnecessary and played little ultimate role in the successful outcome of those warnings.

We don't talk about how, on May 5, 2007, a "tornado emergency" was issued for Great Bend, KS which completely busted. The FAR for TE's is, unfortunately, nowhere near zero.

Certainly, others can differ. But, I have spent a great deal of time studying and thinking through this issue. It comes down to something I wrote earlier today: If some tornado warnings are "emergencies" then what are the rest? Non-emergencies? I don't think we want to send that message either implicitly or explicitly.

Thanks for considering my point of view.

 

[Brett Adair]

As far as an emergency, I see that as a tornado on the ground headed toward a populated area no matter what size it would be. Tornadoes can kill no matter the EF rating or size/speed. There was a tornado emergency issued on 5/7/2003 in Shelby/Talladega County AL and that was the first one I had ever seen and it was due to a tornado being on the ground and throwing debris.

 

[John Peters]

You may be missing the point of this discussion. I don't believe the "TE" for Greensburg was a good idea even though it verified. This is for several reasons that I will again state:

1. One human life is just as valuable as another. The discussions that this should be done when a tornado is approaching a "large city" implies that those lives are more valuable than rural or small town lives (think about it before you reply to this contention). As a former TV meteorologist, I heard this from rural viewers all the time. This would only reinforce that belief and, in a way, it would be true if the TE is reserved only for 'large cities.'

The point of the TE is to emphasize the fact that a dangerous tornado is approaching a densely populated area. It doesn't imply that the lives of city residents are any more important than rural residents; the main issue is that there is a potential for a great loss of life if the urgency of the warning is not conveyed. Large tornadoes often touch down in rural areas, and don't get as much coverage as those that touch down in populated areas. Does this mean that the people who live in the populated areas are considered more important? I don't think so - The issue is in regards to a potential for a large loss of life.

People in large cities tend to consider themselves immune to tornadoes, from experience. I can't tell you how many times I've had to correct the "tornadoes don't go through cities" notion in casual conversation. I think a TE with a disclaimer such as "meteorologists are tracking an extremely large and dangerous tornado" could be enough to convey the urgency of the situation, and convince people with incorrect notions of tornado immunity to take cover.

2. Once the expectation is set that a "tornado emergency" will be issued when the 'real thing' is approaching a city, then people will begin to ignore tornado warnings.

Most people haven't a clue as to what a tornado emergency is (again from experience, this may be different in more tornado prone areas). Regardless, a small fraction of a percentage of tornado warnings are issued as tornado emergencies, and unless this changes, i don't think they will be received as the administration "crying wolf".

3. We don't have the meteorological skill to do this consistently well as last night demonstrates. My concern is not that SGF issued a TE that didn't work out -- I don't think any TE's should be issued.

If these warnings are based on observations of a dangerous tornado approaching a population center, i don't see where the meteorological skill argument comes into play. You have a large tornado on the ground, a good concept of storm motion, and an intense TVS (maybe debris ball signature on radar); i believe this is enough to issue a TE, based on the fact that this storm has or likely will drop a large tornado.

If the sirens are already going off (they were), if the TV and radio are in full-out coverage (they were), and if we are not instructing people to do anything different, why issue the TE and run the risk of watering down the effectiveness of tornado warnings?

I don't see how adding an additional statement to the warning to convey the severity of the situation could water down the effectiveness of the warning.

"well there is a big mile wide tornado sighted heading for my house, but NWS issued a Tornado Emergency, so i'm not going to worry"
i can't imagine this happening

Mike you bring up some good points, i'm just adding my .02

 

[Bob Hartig]

While I'm in favor of the use of the term "tornado emergency," it should to be reserved for extreme situations where widespread loss of life appears imminent. Otherwise, TE becomes the new "tornado warning," and will, as Mike Smith and others have suggested, render both the official term and itself meaningless.

I think the criteria for a TE is basic: a strong, solid-state tornado that has been confirmed visually by reliable sources will shortly impact, or is already impacting, a community of any size.

A key word is community. While lives in rural areas are as precious as lives in the city, that's not the issue. The issue is the certainty of large-scale impact with the potential for scores of fatalities. A bomb that explodes on the open prairie is just as deadly as one that goes off in a shopping mall, but the second scenario amounts to a catastrophe while the first may have no impact at all.

All this being said, I'm not in favor of another term per se. I'm in favor of whatever works. If the standard tornado warnings really do prompt entire communities to take immediate life-saving action, then there's no need for anything stronger. At present, though, I suspect the effectiveness of such warnings varies considerably from location to location.

 

[Mike Smith]

Bob and John,

Good posts as part of a good discussion.

There is one thing I wish to clarify,

I don't see how adding an additional statement to the warning to convey the severity of the situation could water down the effectiveness of the warning.

I think you misread what I wrote. I didn't write about warning down the warning in effect when the TE is issued, I talked about watering down the effectiveness of future warnings. This is my central concern.

Mike

 

[Greg Stumpf]

This is so far an excellent discussion that have mostly been constructive and well thought out. Let me offer some of my comments.

1. I think one overlooked and important aspect of "enhanced wording" in NWS Tornado Warnings (and follow-up SVSs), or as stated by a media partner, is to alert first responders that a very major event requiring a greater than usual activation might be required. In the 3 May 1999 OKC case, first responders were poised just outside the path of the tornado, ready to move in immediately after the event. When a tornado might have a really large impact, first responders from surrounding communities might also be put on high alert to help out in the event that the community's responders are overwhelmed.

2. The real intent of these "tornado emergencies" is to convey the warning forecaster's sense of very high certainty of an event. As several other posters have indicated, a Tornado Warning is not a "sharp" forecast (i.e., only 0% and 100% certainties) even though it is currently issued as a deterministic forecast. The NWS is really stuck with deterministic warnings in the present system because there is no easy way for the warning forecaster to express their uncertainty of a situation other than using qualifying verbiage. This is a prime reason to consider adding probabilistic information to severe weather warnings. The more reliable information about a tornado event that a warning forecast has, coupled with their knowledge of the meteorological information presented in the data (radar, surface, lightning, near-storm environment, etc), the higher the forecaster's certainty will be of the event. We are testing probabilistic warning techniques at the Hazardous Weather Testbed - Experimental Warning Program (HWT/EWP) in Norman. I could say a lot more here, but in the interest of time, I'll refrain for now. But if interested, I discussed this topic at length on a High Instability show last September; the podcast for that particular show has since been removed from the site, but I have a copy of the excerpt that I can make available on request.

3. We cannot really determine the true effectiveness of any NWS severe weather product, current or future, without thorough multidisciplinary research into the interrelated scientific, sociological, economic, and ergonomic factors. We all need to be cautious about making assumptions on how "the public" (however "the public" is defined) will use these products or react to them without doing careful sociological analysis collaborating with folks outside the meteorology circles. One such attempt to tie these other disciplines with meteorology is the Weather And Society * Integrated Studies (WAS*IS) program that I was fortunate to have been a participant.

As an aside (which might be a candidate for a split topic), I question the cited usage percentages (e.g., 2-3%) for NOAA Weather Radio because I'm not sure if those numbers represent the true reach of NWR. Many folks might actually be getting their warnings from a NWR source via a third party (e.g., school principals using one NWR to warn 100s of folks, radio deejays using one NWR to warn 1,000s or 10,000s of folks, etc). Nonetheless, I feel that there are much better alternatives (besides media and internet) to the present NWR system that are being and should be used/explored.

 

[Danny Neal]

Excellent discussion. I will add my two cents.

I feel by using the term "tornado emergency" symbolizes the last ditch effort for the WCM to establish the severity of the situation. Correct me if I am wrong and in this no way a bash on any of the service offices, but isn't it the job of that said office to get the warning out in the most urgent and lifesaving way as possible? DDC, OUN, and I will throw in SGF as well, went above and beyond by taking that extra step to warn its responsibilities. All the office can do is issue the warning, and I fully understand all sides of this debate and am fascinated by the responses so far.

as Bob mentions: "A key word is community. While lives in rural areas are as precious as lives in the city, that's not the issue. The issue is the certainty of large-scale impact with the potential for scores of fatalities"

While I am an advocate of the use of the term in appropriate situations, I.E Greensburg and OKC where there was actual live footage of a large damaging tornado moving into a densely populated area, I can not help but wonder why through the ten hours of warnings that that particular storm warranted the tornado emergency? I have only seen that one warning mentioning it. Could it have been because the storm 45 minutes before caused significant damage to Strafford which is 5-10 miles to the east of downtown? It all depends on the audience, the May 5th Great Bend, situation did show a tornado vortex signature moving into the city and given the situation of THAT day and the night before with the same storm system, I guess I can see how we could use that terminology for the situation. The problem is that it didn't verify and I am sure we are all glad it didn't verify. People get their information from the media. As that one study showed, 1 to 3 percent get it from NWR, so even with the tornado emergency being out there only 1,000 people in a 100,000 person city will hear it straight from the horses mouth.

To say it is over-used is a stretch.....just by my estimation maybe 20 to 30 times since 1999 have we heard the terms tornado emergency used? Correct me if I am wrong. It is essentially the NWS saying "we see something, or we have something confirmed, this is not a joke." Which leaves open the argument....Well we are under a tornado WARNING not an EMERGENCY. As you can see I am not for or against one or the other, but I can see the justification of such an issue. OKC is the best case scenario, and Great Bend is the worst. OKC with all the media and chaser reports and someone can just turn on the tv and see "oh there is a mile wide wedge in my backyard" couldn't have worked any better.

Bottom line, I think it is the responsibility of the NWS to issue (at risk of sounding childish) "the best possible warning" they can. To say they don't is a fallacy and I am certainly not saying they don't. I think many offices have coined the emergency tab into their lingo as the ultimate warning. I think that now it is being recognized by members in the field that if a tornado does strike a major populated area like Chicago, and LOT issues a warning and not an emergency if they have concrete reliable evidence of a damaging tornado, then the resulting fallout could be damaging. It would seem to me like everything in their power to do was not exercised. I don't know just inputting my scrambled thoughts on the issue. I am prepared to defend any statements I made in a friendly constructive manner