Tornado Warning Experiment

[Mike Johnston]

The PURPOSE of this experiment is still kind of baffling to me. On one hand, we supposedly want enhanced wording that will more accurately convey the threat - complete with input from social scientists as to the optimal wording to be used - so the public will take the desired protective action. On the other hand, I keep hearing that the new warnings aren't really meant for the general public (even though the warnings can be obtained by the general public from mulitple channels.) Not sure what the NWS is actually trying to accomplish here.

Also, supposedly this experiment would formalize the definitions of terms like "Tornado Emergency" and thus allow verification stats, etc. To quote from a long thread on this site from last year:

"CRH is also going to develop objective verification metrics to ensure that the top two tiers meet the criteria intended."

So, while there were alot of articles out when this experiment came out last year, what about any evaluation of how it actually worked?

 

[Mike Johnston]

Even though the "audience" is consistent with what RDale described, the actual dissemenation of the product will be the same as it was before. From the product description:

"This enhanced product is available through all distribution channels which currently disseminate TOR, SVR, and SVS warning products."

I don't know if the NWS had a defined "audience" for the legacy tornado warning product, but obviously there is an inconsistency between the defined audience for this experimental product and the distribution channels.

Also, it looks like the experiment almost had to be extended because the required 120-day notice (of a new product to customers) wasn't given last year. The original PDD virtually predicted this by saying that a "second iteration" would be necessary probably in the Spring of 2013 because of the lack of advance notice last year.

 

[rdale]

Very few in the public hear a tornado warning and rush to the nearest NWS data storage website to pull up the full text. Some research was presented at AMS, and the experiment was also expanded / extended because the tags are HIGHLY valuable to automated alert systems, so at least that's of value.

 

[Mike Johnston]

Very few in the public hear a tornado warning and rush to the nearest NWS data storage website to pull up the full text. Some research was presented at AMS, and the experiment was also expanded / extended because the tags are HIGHLY valuable to automated alert systems, so at least that's of value.

I don't know about rushing to the NWS data storage website, but the message gets through on TV. I remember the aforementioned Tornado Emergency messages last year in Wichita. My sister, who lives in southeast Wichita, called me at that time and asked "they're saying it's a tornado emergency now, what does that mean?" Thankfully, I could see the storm on radar and let her know the main part of the storm would likely pass just to her east, but to stay in the basement anyway.

At any rate, if the program is truly an experiment, fine. But, hopefully there will be a real, independent evaluation of the performance before this thing is put into universal operational practice. So many times, these things take on a life of their own and it has more to do with political proponents than actual performance. And, if it's a controlled experiment, then supposedly the higher-tier warnings were to have virtually a zero FAR.

 

[rdale]

Correct, that's the info that TV weathercasters decide to / not to pass along. That's why I said that the public is not the intended audience, TV weathercasters that aren't comfortable with severe weather and emergency managers who lack the training are. Then they can relay that publicly if desired.

 

[Nick Hester]

One thing I have noticed NWS has spread to other WFOs outside this test is the new format for all warnings. Giving the bottom tags for a quick read summary. I think this is great. Though the wording for the public via EAS scrolling and Weather Radio does need tweaked. I am sure this be worked out.

 

[rdale]

I'm confused? The news tags are only for Central Region and won't start until April 1. What are you seeing?

 

[Mike Johnston]

I'm confused? The news tags are only for Central Region and won't start until April 1. What are you seeing?

Well, the warning messages TODAY in Kansas and Missouri look like the experimental language, so I'm not sure they're waiting until April 1st.

Tornado warning, effective right now in Missouri:

"BULLETIN - EAS ACTIVATION REQUESTED
TORNADO WARNING
NATIONAL WEATHER SERVICE SPRINGFIELD MO
1059 AM CST TUE JAN 29 2013
THE NATIONAL WEATHER SERVICE IN SPRINGFIELD HAS ISSUED A
* TORNADO WARNING FOR...
NORTHERN CEDAR COUNTY IN SOUTHWEST MISSOURI...
SOUTHEASTERN ST. CLAIR COUNTY IN WEST CENTRAL MISSOURI...
* UNTIL 1115 AM CST
* AT 1057 AM CST...A SEVERE THUNDERSTORM CAPABLE OF PRODUCING A
TORNADO WAS LOCATED OVER NORTHERN CEDAR COUNTY...AND MOVING
NORTHEAST AT 75 MPH.
HAZARD...DEVELOPING TORNADO.
SOURCE...RADAR INDICATED ROTATION.
IMPACT...SIGNIFICANT HOUSE AND BUILDING DAMAGE POSSIBLE. MOBILE
HOMES COMPLETELY DESTROYED IF HIT. SOME TREES UPROOTED OR
SNAPPED. VEHICLES WILL LIKELY BE THROWN BY TORNADIC WINDS.
* LOCATIONS IMPACTED INCLUDE...
CAPLINGER MILLS...CEDAR SPRINGS...COLLINS...FILLEY...GERSTER...
HARPER...ICONIUM...ROSCOE AND VISTA.
THIS WARNING ALSO INCLUDES TRUMAN LAKE.
PRECAUTIONARY/PREPAREDNESS ACTIONS...
TAKE COVER NOW. MOVE TO AN INTERIOR ROOM ON THE LOWEST FLOOR OF A
STURDY BUILDING. AVOID WINDOWS. IF IN A MOBILE HOME...A VEHICLE OR
OUTDOORS...MOVE TO THE CLOSEST SUBSTANTIAL SHELTER AND PROTECT
YOURSELF FROM FLYING DEBRIS.
&&
LAT...LON 3792 9351 3791 9357 3785 9358 3783 9363
3781 9363 3774 9394 3784 9401 3817 9351
TIME...MOT...LOC 1659Z 238DEG 66KT 3785 9387
TORNADO...RADAR INDICATED
HAIL...<.75IN
$$
KARDELL"

Isn't this the new format?

 

[rdale]

That's the IBW format, but that office started last year. They didn't stop. CR region wide isn't starting until April 1.

http://www.crh.noaa.gov/crh/?n=2013_ibw_info

 

[Mike Johnston]

That's the IBW format, but that office started last year. They didn't stop. CR region wide isn't starting until April 1.

http://www.crh.noaa.gov/crh/?n=2013_ibw_info

You're right. Looks like the warnings down in Texas are being issued in the legacy format.

 

[Patrick Marsh]

One clarification, I've been told by friends at the Omaha office that they have voted *not* to participate in the IBW, so it's all of central-region, minus one office (for now).

Also, there are some offices in the southern region that are planning on doing their own "experiment" that is very similar to IBW.

 

[Jerry Prsha]

I had a customer of mine ask me about this new format.. Specifically, the more terse language. She told me that the message was quite scary so she took it seriously.

I found it odd that she must not have taken it seriously before this but it must be working.. Until people get used to it again.

 

[Patrick Marsh]

Did she say what she will do when the new warnings don't pan out? And it is a when as long as the FAR remains above 70% ...

 

[Mike Johnston]

Wow, didn't realize an individual office could vote out of a regional directive. Interesting. I wonder if it was a philosophical difference or a technical issue?

I just hope they actually do a study of the experimental results before rolling this out for good. These things tend to take on a bureaucratic life of their own. Thought UNC (or some group in North Carolina) was supposed to do a study of last year's results, but can't find that anywhere.

At any rate, I do agree that the FAR is the big underlying problem and I'm not sure that any tinkering around with the wording is going to address the real issue.

 

[Patrick Marsh]

I believe that since it constitutes a change in work required it must be approved by the union.

I could be wrong...

...and the lone holdout might yet be forced to join...