Tornado Warning Experiment

[Jeff Duda]

It looks like the specific text has been altered from last year's experiment, so he probably wasn't misinforming. However, it already seems to be more complicated to me than it needs to be. It's just a matter of tags for whatever processor reads the text of the warning to use when making a graphic or sounding a particular alert tone. Hell, they'll probably make it so you can filter out certain types of tornado warnings.

Off the topic, but when will "law enforcement" stop being an acceptable confirmation of a tornado? What is the accuracy of law enforcement reported tornadoes?

 

[rdale]

I don't think there are 5 categories, so wording change isn't an issue. Regarding LE tornadoes, I don't understand... Are you saying the NWS uses a LE report as a confirmed tornado for Storm Data purposes? I don't think that is accurate.

 

[Mike Johnston]

Well, as TWC presented it graphically, the 5 levels of warning (paraphrasing a bit):

1. Severe thunderstorm with possibility of a tornado.

2. Radar indicated tornado.

3. Observed tornado.

4. Considerable damage threat tornado.

5. Catastropic damage threat tornado.

That's the way TWC presented it - not saying it's official, but the NWS central region director they interviewed seemed to agree.

Re: law enforcement reports, it's just an example of an "observed tornado" and doesn't have anything to do with whether it's confirmed.

 

[Jeff Duda]

I don't think there are 5 categories, so wording change isn't an issue. Regarding LE tornadoes, I don't understand... Are you saying the NWS uses a LE report as a confirmed tornado for Storm Data purposes? I don't think that is accurate.

Using this page as my reference:
-Second example down from the top: the example warning reads "SOURCE...LAW ENFORCEMENT CONFIRMED TORNADO". I was half joking about that, given how many "sheriffnadoes" are reported. However, a colleague of mine and currently a forecaster at a central region office told me an anecdote about a caller reporting a tornado at night when they were really just looking at a grain silo in the distance. Even though the caller was familiar with the area, an inanimate object still fooled that person into being very convinced (because he called it in twice) he was looking at a tornado. I don't remember for sure if the caller was law enforcement in this anecdote, but for the purposes of my argument, I'll just say he was.

-If you look at the table on that page, there are 5 rows, although two of them correspond to severe thunderstorm warnings. However, I think having a second type of severe thunderstorm warning that includes mention of a possible tornado is not a good idea. I always thought that the possibility of a tornado was implicit within EVERY severe thunderstorm warning, whether that text was included or not. Look at the fourth example down from the top. What if you were to only see the text at the very bottom of the warning? Would you immediately know it is just a severe thunderstorm warning?

 

[Mike Johnston]

I think possibly we're confusing the terms "confirmed" and "verified" here. Confirmed seems to be a term sometimes used in warning text, as in "this tornado has been confirmed by trained spotters", etc. Verified has to do with whether the report of a tornado is subsequently verified for purposes of matriculating from the preliminary local storm report to the final storm report database. At any rate, under the new warning regime, they're using the term "observed" and Jeff is correct that reports by law enforcement are given the same validity as trained spotter reports; I think that is what he was questioning. Correct me if I'm wrong, Jeff.

 

[Chris_Spannagle]

Using this page as my reference:
However, a colleague of mine and currently a forecaster at a central region office told me an anecdote about a caller reporting a tornado at night when they were really just looking at a grain silo in the distance. Even though the caller was familiar with the area, an inanimate object still fooled that person into being very convinced (because he called it in twice) he was looking at a tornado. I don't remember for sure if the caller was law enforcement in this anecdote, but for the purposes of my argument, I'll just say he was.

Jeff, this story is a true story that happened to a co-worker of mine at WDTB when she was working in the Tulsa WFO many years ago. This is a story she tells at our workshops for NWS interns each year. The call was from a local EM.

 

[rdale]

I always thought that the possibility of a tornado was implicit within EVERY severe thunderstorm warning,

Oh no... Many severe storms are capable of nothing more than 1" hail and 58.1mph winds. The tornado threat is not great in all. But many times you'll see a blanket tornado warning issued for 4-8+ counties just in case a EF0 pops up. Allowing it to be a SVR with "possible tornado" means the outcome is the same, but the blanket isn't needed sounding hundreds of tornado sirens for a tree knocked down.