Expanding warning times for tornadoes....

[Rob Gardner]

This thread is in no way intended to knock the science aspect of storm chasing. I think it is a vital and integral part of meteorology.

That being said there is one particular coin phrase that has me curious and that is the one of "if we can expand tornado warnings to be an hour in advance or more we could save lives..."

Now while I can see how having warnings more in advanced can have the potential to save lives I also wonder how having warnings 60-90 minutes out could be detrimental...

For example

1. With longer warning times how many people may try to leave their homes to evacuate or to go to a loved ones house

2. How many people would get used to the extended time and then not take a warning seriously when it was issued with only minutes to spare

3. Would more notice just lead to more apathy? As it is I think people already don't take tornado warnings as seriously as they should.

Tornado watches give people hours of warning that tornadoes are possible and even with long track warning when long lived tornadoes have been on the ground there are still people caught unprepared or simply unaware.

To me I think that tornado watches should have people in a state of alert and preparedness and warnings should give enough time for people to act on that preparedness and take their place of shelter. Where science would appear to save the most lives would be in education and accuracy. There will always be a margin of error built into tornado warnings because the consequences of crying wolf are minimal compared to the consequences of not issuing a warning for a tornado, but the more accurate science gets so that the amount of wolf crying decreases, coupled with education can hopefully lower apathy and then in turn save lives.

If anything what amazed me this chase season was the number of people I saw in tornado warnings who were aware of the warning but were unconcerned and even worse unprepared.....

 

[rdale]

To calm your concerns - we're not even in the ballpark of having 1hr lead times. I don't hear that phrase used "realistically" in the warning arena.

Actually I thought this thread was going to be on how offices are blowing the 45 minute regulation on the maximum duration out of the water quite frequently this season. Thinking that people are going to sit in a shelter for 80 minutes is just ludicrous, yet some offices issue them regularly in excess of an hour. Great for verification odds, but horrible practice.

 

[Verne Carlson]

Add to that list #4

4) with extended warning times people will stay longer at home packing up their belongings, pictures, documents, pets, etc.

I chuckle when scientists say that extended warning times will save lives, it may actually have the opposite effect.

 

[Greg Stumpf]

Extending warning lead times can be useful, but the warnings must include time of arrival and time of departure information for all points downstream of the hazard. This information would alleviate the concerns raised by Rob, Rob, and Verne.

When scientists talk about extending lead times, what is really happening is that forecasts for hazards will become less uncertain at longer forecast periods owing to advances in the science. We can arbitrarily extend warning durations today, but the cost would be larger false alarm areas/times (even though there is a greater likelihood to verify the warning polygon *area* by a single *point* severe weather observation). For example, in the future, a warning for an hazard 60 minutes upstream will have a higher certainty than a 60-minute warning today.

 

[Chris Nuttall]

This thread is in no way intended to knock the science aspect of storm chasing. I think it is a vital and integral part of meteorology.

That being said there is one particular coin phrase that has me curious and that is the one of "if we can expand tornado warnings to be an hour in advance or more we could save lives..."

Now while I can see how having warnings more in advanced can have the potential to save lives I also wonder how having warnings 60-90 minutes out could be detrimental...

For example

1. With longer warning times how many people may try to leave their homes to evacuate or to go to a loved ones house

2. How many people would get used to the extended time and then not take a warning seriously when it was issued with only minutes to spare

3. Would more notice just lead to more apathy? As it is I think people already don't take tornado warnings as seriously as they should.

There is where the societal impact field of study is starting to take off. These studies are just really beginning to start. It will be interesting to see what they find in the next few years. For those that want to know more, I suggest taking a look at WAS*IS.

 

[Jason Hobbs]

Honestly, upon hearing sirens as it is, people will do one of three things:
1. Take shelter.
2. Ignore it.
3. Run outside.

Camp Ripley, MN is an excellent example of number 3. Back in July, we kept falling under watches and severe weather warnings but Ripley's response is to set off the sirens for all of it. On the upside, it gives good warning to the Soldiers out on the ranges/field exercises that the bad line of crap they've been seeing rolling down the horizon is just as bad as they think it is, but on the downside, those of us who are, by trade, Fobbits and generally tend to stay in the main part of the camp (Ie, us aviation folk, who already had scrambled to get the aircraft out of the area the day prior.) hear the sirens as a call to roll out and look at the storm.

I know there have to be plenty of people who think these sirens are basically "camera warnings" and will run out to take pictures accordingly. With an hour of lead-time, you're going to have some crazies getting into their cars and driving INTO the danger-area.

 

[J Tyler]

My opinion is, the biggest problem we face today is the warning not getting TO the public in rural areas. Small town radio stations are now all automated and offers NO severe weather coverage. Rural populations don't typically carry "smart phones". The major market big city television station won't risk turning off their primary audience by cutting in for a storm warning in a rural county. A lot of the smaller towns don't have a warning siren system (or they are in poor condition).

The exception to all this is Oklahoma City, where Gary England sends his flying monkeys out to all corners of the state and does wall to wall coverage.

What I don't understand is people who don't spend $30 and buy a NOAA all hazards radio. When I'm out spotting/chasing in small towns, we ALWAYS have people driving up to us asking what is happening. I ask if they have a weather radio, and so far I have NEVER had a person say "yes". Incredible.

So, my point is, we can increase warning time to 6 hours, but if people don't get the message, it won't matter.

 

[Melissa Jablonski]

IWX does a really good job of downstream warning. One thing I have noticed though is a lot of people don't know what towns are 30 miles Southwest or North West of them. Usually tornadoes that come here come from those directions. The forcasters over in North Webster Will re- issue a warning rather than run a warning too long. They also keep updating the storm location on the weather radio and other weather sources. I just thought I'd give credit where it is due.

 

[ahaberlie]

Sirens, at least around here, are met with tons of apathy. Some people don't even know what they are for, most do not take shelter. Unfortunately, I think the only thing that will change this is an actual tornado plowing through downtown madison.

Not sure how you fix this; if the NWS decides not to tornado warn a storm and it becomes tornadic, that is less time that the public can prepare for it. If the NWS tornado warns a storm, and it doesn't become tornadic, the next time, people may not heed the warning because they have lost confidence in the system, and god forbid they inconvenience themselves for half an hour or so.

I keep telling people that in 1984 an F5 destroyed the town of Barneveld just 30 miles to our west, so even a dangerous outlier tornado is definitely possible around here. In fact, I think about 6% of the recorded F5's since 1950 have occurred in Wisconsin.

 

[rdale]

A great paper just came out from JHSEM showing how little value sirens have these days. But most studies show that there aren't enough false alarm TORs in one county to have that much of a negative impact.

 

[Andrew Burnett]

A couple thoughts.

I would like to see more resource put into real time notifications. Text messages of Tor warnings, e-mails, and such. One easy place to go to sign up for these things and an easy way for local athorities to promote them. Something tied to the most recent real time technology. I would also like to see some sort of spotter network "soft" warning capablity. Basically if you have over x number of people reporting a tornado, that area is instantly warned.

The municiple system of activating storm sirens needs revamped. At least in smaller towns alot of times the sirens don't go off until the storm is well past. I remember one day watching a local storm on TV.net it went warned And I took my laptop outside and watched the storm. There was not much to see, but the muni sirens did not go off until the storm was well past town. If a tornado had actually touched down sirens would have not done a thing.

 

[rdale]

I would like to see more resource put into real time notifications.

That requires WAY too much resources that the NWS doesn't have. Private sector already does a great job with it. Anyone can sign up at weather.com for free

I would also like to see some sort of spotter network "soft" warning capablity. Basically if you have over x number of people reporting a tornado, that area is instantly warned.

I'll throw a HELL-no on that one. The warning decision still needs to have human interference. If you want to sound sirens based on that, so be it, but no warning.

At least in smaller towns alot of times the sirens don't go off until the storm is well past.

Take that up with your town. That has nothing to do with NWS or the warning process. There are some VERY-low cost systems out there which tell EMAs which sirens to sound and which to keep quiet in a VERY timely manner.

 

[ChristianTerry]

Even with expanded times the main problem is not that people get the warning, but how they respond to the danger. Plenty of people don't take warnings seriously because of past warnings that proved false. And for those that do take them seriously, knowing what to do if one strikes and not panicking.

 

[rdale]

Plenty of people don't take warnings seriously because of past warnings that proved false.

Every bit of research says that is not the case...

And for those that do take them seriously, knowing what to do if one strikes and not panicking.

I've not heard of people going into panic ahead of a tornado and forgetting to go to the basement - is that a common issue.

 

[Tony Cook]

Every bit of research says that is not the case...

What research are you referring to? Section 3 of this AMS paper by Barnes et al. deals with the topic:
http://www.evegruntfest.com/pdfs/61-Barnesetal07.pdf

However, they do not discuss the public perception of False Alarms for tornadoes specifically. They instead use Earthquakes, Hurricanes, Dam Breaks, etc., where at least some concrete research exists to support their thesis.

I find it hard to believe that a high false alarm rate (~0.75- 1 in 4 tornado warnings verify) has a negligible impact on the general public's perception of tornado warnings. Add in the media blitz effect, where every network goes wall-to-wall and preempts programming (see: Mad Criminal Minds Viewer)
http://www.wikio.com/video/angry-weather-caller-3328538

But most studies show that there aren't enough false alarm TORs in one county to have that much of a negative impact.

With media outlets, you are not dealing with a single county having false alarm TORs. You are dealing with the all the counties covered by that network/outlet. This can result in tens of warnings in a single event.

My feeling is that the general public does to some extent become desensitized to repeated tornado warnings that do net verify. Granted, my opinion is largely based on anecdotal evidence, and on personal conversations with large numbers of friends, family... i.e. the general public. I don't see anything in the Barnes paper to convince me otherwise.

Is there any other research that applies specifically to tornado warnings?

TonyC