"TORNADO EMERGENCY"

[Eric Baker]

Mike, have you researched every TE ever disseminated, I haven't? Again, are May 8, 2003 and May 9, 2003 Failures. No one died as a tornado tracked across Moore, and no one died as a tornado tracked from west OKC into northwest OKC. What is to say that TE's are a failure.

 

[Justin Wiley]

Mike, have you researched every TE ever disseminated, I haven't? Again, are May 8, 2003 and May 9, 2003 Failures. No one died as a tornado tracked across Moore, and no one died as a tornado tracked from west OKC into northwest OKC. What is to say that TE's are a failure.

But a person living in Arkansas could have heard about the tornado emergencies those the last couple of days in Oklahoma and then have a tornado warning issued for their county the next day, and not took it as seriously and could have gotten injured or killed.

It's hard to prove, but it's possible.

 

[Eric Baker]

#1. The public doesn't know the difference, they just know the sirens are going off and the TV programming is weather bulletins. Our friends, who are in touch with weather lingo, might.

Sorry Mike, but not all people have a TV that get local stations, especially in rural areas. And a lot of people have weather radios, which as you know reads the text by computer voice. (the radio doesn't actually read, the computer at the nws office reads the warning, and sometimes a meteorologist will come on to clarify)

Anyway, we all agree that the warning system, esp for tornados, is flawed. There needs to be some continuity in warning texts, and that unity needs to be carried throughout all NWS WFO offices. With a 78 almost 79% false alarm rate, the public is going to has some doubts about the warning system.

My dad goes outside and watches for something to happen. My old next door neighbors sat on their roof. I'm not the only one who has witnessed it.

So, what is there to do. Well the polygons are pretty good. Then the sirens Mike was talking about. In OKC if the is a Tornado Warning, all of the sirens go off in the entire city, and sometimes most of the county. OKC is Huge. A more localized warning approach with sirens may also help in bigger metropolitan areas.

As for the issuing of Tornado Warnings, there seems to be two types.
Type 1: Doppler Indicated (severe thunderstorm capable of producing a tornado.
Type 3: Confirmed Tornado (trained storm spotters or off duty NWS meteorologists report a tornado, or sheriffnados)

Should there then be a differentiation between the two?

 

[Scott Weberpal]

c) My multiple challenges to demonstrate a positive (i.e., people who heard the TE and went to shelter specifically because they heard it) have gone unanswered.

This should be proof enough.

Mike

No, that proof is not enough. How can you ask someone to scientifically prove anything when the sample size is so small? If you're talking about only a few verifications, there's absolutely no way to make a conclusion on either side. There is also no proof that issuing tornado emergencies will "lessen" the effect of tornado warnings.

 

[Mike Smith]

Sorry Mike, but not all people have a TV that get local stations, especially in rural areas. And a lot of people have weather radios, which as you know reads the text by computer voice. (the radio doesn't actually read, the computer at the nws office reads the warning, and sometimes a meteorologist will come on to clarify)

Anyway, we all agree that the warning system, esp for tornados, is flawed. There needs to be some continuity in warning texts, and that unity needs to be carried throughout all NWS WFO offices. With a 78 almost 79% false alarm rate, the public is going to has some doubts about the warning system.

My dad goes outside and watches for something to happen. My old next door neighbors sat on their roof. I'm not the only one who has witnessed it.

So, what is there to do. Well the polygons are pretty good. Then the sirens Mike was talking about. In OKC if the is a Tornado Warning, all of the sirens go off in the entire city, and sometimes most of the county. OKC is Huge. A more localized warning approach with sirens may also help in bigger metropolitan areas.

As for the issuing of Tornado Warnings, there seems to be two types.
Type 1: Doppler Indicated (severe thunderstorm capable of producing a tornado.
Type 3: Confirmed Tornado (trained storm spotters or off duty NWS meteorologists report a tornado, or sheriffnados)

Should there then be a differentiation between the two?

Eric, I hardly know where to start. I believe you may be extrapolating your family and friends to the general population which is natural but not likely to yield a true picture. So, here are some quick comments...

Rural televison: DirecTV and Dish Network? Just about everyone I know who is out of range of the direct broadcast signal gets local stations that way.

Weather radio? 3% of population receives their warnings from them (based on two peer-reviewed studies).

I do not agree the warning system is flawed. The definition of "flawed" is "an imperfection or weakness and especially one that detracts from the whole or hinders effectiveness." A system that has cut tornado deaths by 90% is not "flawed." Imperfect yes, but not "flawed."

There are too many sub-categories of warning-related messages now. The NWS made a positive step last winter by cutting back the number of winter storm-related messages. The last thing we need is more messages related to tornadoes and convective situations, which is one of the reasons I object to TE's.

Eric, it seems we are getting far away from the topic of "tornado emergencies." You and I have difference of opinion, which is fine.

Best wishes,
Mike

 

[Mike Smith]

No, that proof is not enough. How can you ask someone to scientifically prove anything when the sample size is so small? If you're talking about only a few verifications, there's absolutely no way to make a conclusion on either side.

Please let me refer you to the AMS's excellent edition of Weather and Forecasting devoted to the May 3, 1999 tornado outbreak where the term "tornado emergency" was first put into existence. It is available here: http://ams.allenpress.com/perlserv/?request=get-toc&issn=1520-0434&volume=17&issue=3

There are two papers specifically about the warnings and how people were injured. I just reviewed it again (it has been several years) and, unless I am missing it, there is no evidence the TE made a difference.

The sample size of TE's is not particularly small, but other than that I agree with the gist of what you are saying which is given the ongoing dispute over the effectiveness of TE's why not do some scientific research on it? As I have previously stated, I have never seen a study that proves the positive but it may be appropriate to do a study to know for sure.

There is also no proof that issuing tornado emergencies will "lessen" the effect of tornado warnings.

You are correct there is no "proof" but there is an analogous situation. Since the Department of Homeland Security started its Advisory System the threat level has never been below "elevated." It is currently "elevated" for the public-at-large and, for years, has been "high" for airlines.

Yet, nothing has happened -- there have been zero successful attacks (thank goodness!) during this period of time, so it has a 100% false alarm rate.

Let me ask you, "Do you go check the Advisory Level before you get on an airplane?" Do you know a single person who goes to the web site and checks the threat level? I suspect the answer is "no" because of the 100% FAR and resulting loss of credibility.

If you agree with this analogy then it is clear that too high a FAR can hurt the effectiveness of a warning system.

By putting out "tornado emergencies" we are implying that ordinary tornado warnings are not "emergencies" (even though we do not have the scientific skill to predict very short term tornado behavior). And, because so many TE's have busted, we risk conditioning people not to pay attention to TE's and (by extension) not to pay attention to the warnings (after all, they are now a less important message than an "emergency").

I don't want to see this happen but fear it may have already occurred in a couple of localities.

 

[Scott Weberpal]

I would also like to clarify (because it probably isn't clear) and I'm not an advocate for or against tornado emergencies, though it may seem I'm a quite for them. I'm simply trying to play devils advocate in a situation where I don't believe anyone can conclude that they do or do not help. I simply don't have a problem with them being used unless it is found they are detrimental to the overall warning process.

 

[Eric Baker]

Mike, your 90% decrease doesn't sit well with me. Not only is it advancements in the warning system, it is advancements in medicine, more hospitals, quicker response time, safe rooms, and many other factors which have led to less deaths. You cant just compare deaths with population increase. Plus you have to take things into consideration, like where the deaths occurred and compare it to specific population increase. As far as friends, family, and neighbors, I have only chased for the past two years. May 24, 2008 was a field day for locals. There were people sitting in their front yards watching the storm, while part of the meso was directly overhead. You may not have seen things like this, but I for sure have. And not just on that single occasion.

I have learned more about Tornado warnings and statistics about tornados in the past two days than I ever thought I would know LOL.
I hope I am not bringing any animosity to the table, just good opinionated fun.

 

[Danny Neal]

"However, 75% of all tornadoes cause only F0 or F1
damage (Fig. 4). They are quite weak storms. During
the F5 tornado of May 3, 1999, the NWS Forecast
Office in Norman issued a warning headlined “Tornado
Emergency” to indicate the presence of an extreme
event. The local emergency managers and media all
credit this innovative wording as helping to save lives
during this storm. Would it be scientifically possible and
sociologically advisable for the NWS to work on
developing a new category of warning to indicate that
there is a high likelihood of a strong or violent tornado
occurring in the very near future?"

http://www.spc.noaa.gov/publications/mccarthy/tor30yrs.pdf

Interesting quote from McCarthy's paper a few years ago.

 

[Jeff House]

Medicine and medical care have little do to with the decrease in tornado fatalities. We have the best health care system in the world, but it is not why we have fewer tornado deaths. It is all about knowledge, and we know knowledge is power. People know the safety rules better, but a huge majority of the improvement is better warnings and lead-time. Knowing a dangerous storm is coming is huge knowledge power. Statistical studies backed by science prove the warning system does work.

We do need to work on false alarms. TEs do not help because they either confuse the public or get them playing meteorologist when they need to be taking shelter. Tornado warnings require quick action promoted by clear messaging. A little info in the warning text is great but kill TEs until we get the science and FA rate down. How is that for a compromise amigos?

Regardless of the TE debate for the public, let us all stay safe and have good luck chasing!

 

[Shane Adams]

Finally, in the two plus years since Greensburg, we have not had a single TE that has been an unqualified success. On the other hand, we have had numerous busts as recently as last night. Shouldn't we be learning something from this?

Yes, what we've learned is used correctly, there cannot be a failed Tornado Emergency.....because it is not a warning, it's an enhancement of the warning. A large violent tornado must be ongoing inside a community current-time or imminent before the TE is used. Therefore, applied as it was on 5-3-99 (date of creation) and 5-4-07, it can never fail.

Wow, this thread just won't die (thanks to folks like me

 

[Ben McMillan]

Sorry Shane, I’m going to keep this thread going a little more.

We are beginning to see some National Weather Service offices, in this case NWS Des Moines, take up more of an official position in the emergency debate.

Yesterday, March 12, DMX released a media advisory calling Tornado Emergencies a high impact call to action.

Their policy states...

"A Tornado Emergency means that significant, widespread damage with a high likelihood of numerous fatalities is expected to continue with a strong and violent tornado."

Full release- http://www.crh.noaa.gov/images/dmx/Media%20Advisory%202010-01.pdf

I think they did a good job of putting into words guidelines for a complex and controversial topic.

Have any of you seen these types of policies coming from your, or other, local offices? It will be interesting to see how long it takes to implement a standard for this nationally.

 

[rdale]

The NWS Directives were supposed to be rewritten last year regarding the TE policy... My guess is that a draft is floating around there for (hopefully) implementation nationwide this year.

 

[Randall Marable]

The NWS Directives were supposed to be rewritten last year regarding the TE policy... My guess is that a draft is floating around there for (hopefully) implementation nationwide this year.

A little off-topic but I guess the same should be done when local WFO's issue statements for 'Snow Emergencies' (ie. Baltimore/Washington) and 'Hurricane Emergencies' (ie. Mobile/Penscola). If these terms are going to be used, then all local offices should be on the same page while everyone concerned will understand the threat accordingly. (I could possibly see some issues with such directives with respect to 'Snow Emergencies' since a heavy snow/accumulation event is all relative to a particular region. So in essence, a 'Snow Emergency' for one NWS CWA may not necessarily qualify for another area. Thus, possibly opening the door for more confusion with respect to issued 'emergency' statements.)

The other issues surrounding 'Tornado Emergency' statements have already been discussed and debated extensively throughout this thread. I would just recommend the NWS and it's WFO's reserve these types of 'emergency' statements to extremely high impact, highly probable events. But even when it comes to short-term, local weather forecasting, this may not always be feasible. Just food for thought.

 

[Kory Hartman]

Yesterday, March 12, DMX released a media advisory calling Tornado Emergencies a high impact call to action.

Their policy states...

"A Tornado Emergency means that significant, widespread damage with a high likelihood of numerous fatalities is expected to continue with a strong and violent tornado."

I think they did a great job explaining what it is and when it is used. Nearly all of the time the wording will be used inside of a Severe Weather Statement, so everyone's concern about a "new" type of Warning is unfounded.

Now, the next question, how many people actually see/hear the SWS information during an event?