Weather Science Does It Again.

[Mike Smith]

I'll give credit to Kansas Governor Brownback who understood yesterday's watches and warnings saved lives. Story here: www.kansascw.com/kscw/news/kwch-governor-brownback-reacts-to-tornado-damage-20120415,0,4067763.story

Given the number of tornadoes that touched down across the state, Brownback said the state was fortunate.“People paid attention to the advance warnings issued by the National Weather Service and our Emergency Management and took them seriously. They did everything they could to protect themselves and their families. These actions may have saved many lives,â€￾ Brownback said.

Maybe, just maybe, the recent Dallas tornado swarm and yesterday's major outbreak are getting the message across that warnings are valuable and should be acted on. And, just maybe, weather science (including storm chasers) are beginning to get a little of the credit.

I have more thoughts on yesterday's storms with a couple of interesting photos at my blog: http://meteorologicalmusings.blogspot.com/2012/04/some-thoughts-about-yesterdays.html

 

[Mike Johnston]

Mike, appreciate your thoughts on this. Seems like it was very obvious, from the SPC's advance outlooks, to the watches and the warnings that folks were well advised on this outbreak. Have heard some anectodal reports from my sister in Wichita about a mobile home park that sustained damage but virtually all the residents were in shelter, and even though 4 buildings at the huge Spirit (formerly Boeing) plants sustained $200 million in damage, their employees were safely sheltered in an underground tunnel complex. So, good job of "warnings" all the way around, perhaps the best ever.

Having said that, I am curious if you have any opinions on the effectiveness of the experimental impact-based warning formats in yesterday's events. I read many of the warning statements, and I'm not convinced that the new format really added much value. At best, it may be just more wording and some of the warnings looked more like a "fill in the blanks" exercise than anything. I think just directly telling the people what they need to know in the text - what, where, when - is sufficient, keeping it brief but descriptive enough to convey the hazard. Would be interested if you have any thoughts on this.

 

[Jeff Duda]

Maybe...or maybe people who live in Kansas are just more aware of how common it will be for tornadoes to occur in their state, especially from April-June. Thus, maybe they're just naturally more aware and better prepared to handle a warning when it comes.

 

[Kelly Sugden]

How do you know if the experiment was effective with a sample size of 1? It is way too early to draw conclusions on the effectiveness of the impact-based warning project. Just shorthand, I haven't heard of any deaths in Kansas, so if it changed the behavioral science of the user base yesterday (as you might have suggested via the mobile home community/Boeing) and caused people to take the warning seriously then I would consider that effective. Bottom line, it is way too early to tell. Social scientists say it takes 4 sources before people take action for a warning. Hopefully with enhanced warning (based off real science - like a debris ball and resultant low cc - physical remote sensing proof of flying debris) then perhaps people will not take a warning so lightheartedly.

 

[Mike Johnston]

Maybe...or maybe people who live in Kansas are just more aware of how common it will be for tornadoes to occur in their state, especially from April-June. Thus, maybe they're just naturally more aware and better prepared to handle a warning when it comes.

Oh, no doubt the weather-wise folks in Kansas have alot to do with their response. I do believe, though, the high risk issued a day in advance got attention and probably altered people's plans for the weekend day - or at least made them think in advance along the lines of "if I will be at x place during the afternoon and evening, what is my plan in case a warning is issued?" It takes both: good science and a common sense populace.

 

[Kelly Sugden]

I'm biased, but Kansans are extremely weather savy/common sense people. I have given spotter talks and the general attitude of the participating crowd is "why would you do something like that" (get absurdly close). Now, it is like that in many states, but as I said, I'm biased to the good folks of SW/SC Kansas.

 

[Mike Johnston]

How do you know if the experiment was effective with a sample size of 1? It is way too early to draw conclusions on the effectiveness of the impact-based warning project. Just shorthand, I haven't heard of any deaths in Kansas, so if it changed the behavioral science of the user base yesterday (as you might have suggested via the mobile home community/Boeing) and caused people to take the warning seriously then I would consider that effective. Bottom line, it is way too early to tell. Social scientists say it takes 4 sources before people take action for a warning. Hopefully with enhanced warning (based off real science - like a debris ball and resultant low cc - physical remote sensing proof of flying debris) then perhaps people will not take a warning so lightheartedly.

Oh, Kelly, I really haven't drawn any conclusions about the experiment and agree it will take research to evaluate. As for the sample size of one day, actually one concern I had about the experiment was that there were too FEW forecast offices involved and therefore we might not have a sufficient sample size to evaluate. My initial observations are just based on a plain reading of the new warning statements in action. Seemed to be a little bit "canned" and nothing really communicated that couldn't be done in the regular text.

 

[Kelly Sugden]

I am not talking about the sample size of the WFO's but the sample size of high end weather events. You are right, it is more canned... because it comes from a formatter. Same bullet format as the WSWs/NPWs/RFWs. Didn't see much of preemptive critic in those products as I do now (not even 24 hours after the event).

 

[Mike Johnston]

I am not talking about the sample size of the WFO's but the sample size of high end weather events. You are right, it is more canned... because it comes from a formatter. Same bullet format as the WSWs/NPWs/RFWs. Didn't see much of preemptive critic in those products as I do now (not even 24 hours after the event).

I just meant that the more WFO's involved, greater liklihood of more events to be sampled.

 

[Mike Smith]

Having said that, I am curious if you have any opinions on the effectiveness of the experimental impact-based warning formats in yesterday's events. I read many of the warning statements, and I'm not convinced that the new format really added much value. At best, it may be just more wording and some of the warnings looked more like a "fill in the blanks" exercise than anything. I think just directly telling the people what they need to know in the text - what, where, when - is sufficient, keeping it brief but descriptive enough to convey the hazard. Would be interested if you have any thoughts on this.

As you know, I'm not a fan of the "impact based" warning experiment and nothing I heard yesterday changed my mind.

What did shock me is that NOAA Weather Radio doesn't tell its listeners that they were under a PDS (as opposed to regular) tornado watch! I have always believed there is value in the PDS tornado watches and omitting that information from NWR is shocking -- at least to me.

 

[Kelly Sugden]

Very little (like 10 percent) of people get their weather information from weather radio. Of course, I recommend a weather radio to anyone. With this day and age... antiquated technology like sirens... this is becoming a less than ideal way to get updated weather information. IMHO, smart phones are the way to go and obviously others think so as I heard that major providers will be sending out weather information (free? not sure). Watched TWC (only to get an an interview) and some woman said she had no warning... when asked if her weather radio was turned on, she said no! Just out of curiosity, I wonder how many weather radios are incorrectly configured. People need to take personal responsibility and make sure they have easy access to warnings. Relying on sirens (which may be sounded too late or fail for some other reason) is not the way to go. NWR transmitters can also fail without information being disseminated correctly. Warnings via internet/smart phones have way better reliability.

 

[Jeff Passner]

Actaully, a survey shows only 2% of people use weather radio for their warnings. It's pretty much a useless technonology... If you live in the Midwest, ask around --- I bet no one even has them anymore. I think in a few years they will be gone.