The sudden onset of almost summer-like weather earlier this week (and this weekend), along with the weather that produced 63 tornadoes to the south but snow here March 2, and this being National Severe Weather Preparedness Week bring this item to mind, from meteorologist Mike Smith:

Starting April 1, in the geographic areas served by the National Weather Service offices in Kansas City, Wichita, Topeka, Kansas City, Springfield (MO), and St. Louis there will be multi-tiered severe thunderstorm and tornado warnings.

The changes which I will describe below spring from the high death toll from U.S. tornadoes in general, and the Joplin tornado in particular, in 2011.

The Joplin tornado May 22, 2011, an EF5 on the Enhanced Fujita scale of tornado damage, killed 160, injured more than 990, and caused an estimated $2.2 billion in damage.

To review for non-weather-geek readers:  A tornado warning (three of which were issued for the Ripon area in 2011, a personal record since Dane County in 1984, the year of the killer Barneveld tornado) is issued when a tornado or funnel cloud is sighted by trained tornado spotters, a tornado is indicated by weather radar, or weather radar detects a severe thunderstorm capable of producing a tornado (or STCOPATs as I call them).

To expand upon that, the Weather Service in a few Kansas and Missouri offices will be able to issue two augmented warnings:

• A Particularly Dangerous Situation tornado warning. (Similar to a PDS Tornado Watch, issued when “long-lived, strong and violent tornadoes are possible,” such as EF4 or EF5 tornadoes.  There is also a PDS Severe Thunderstorm Watch (for winds around 90 mph or 1.5-inch hail) and Flash Flood Watch, but they have never been issued in this state.)
• A Tornado Emergency, which is now issued for large tornadoes heading toward populated areas. Twelve Tornado Emergencies were issued during the March 2 outbreak.

What’s the problem here? Says Smith, author of the fascinating Warnings: The True Story of How Science Tamed the Weather:

Unfortunately, they are going to allow a sentence to be added to severe thunderstorm warnings that states, “A tornado is possible.” What do you or a school principal do with that? Go halfway down the basement stairs?

Given the political pressure the National Weather Service seems to be under at the moment, I forecast that many severe thunderstorm warnings will contain that unfortunate sentence and the “overwarning” problem, which we know causes complacency, will get measurably worse. …

The first problem is that the science does not exist to do this! We have no skill at short-term tornado strength forecasting. None.

Second, who is going to be able to keep straight whether a “tornado emergency” is better or worse than a “particularly dangerous situation”?

Third, even if #1 and #2 were not issues, what do you want the public to do differently?! Since we meteorologists want everyone to take shelter during a tornado warning, the two “tornado warnings on steroids” are superfluous. …

This isn’t just my opinion. Dr. Laura Myers, a social scientist at Mississippi State University, wrote yesterday,

My conclusion: It would seem that more detail and more warning levels would help, but I think it just leads to confusion and [warning] fatigue.

When a tornado is bearing down, people need to act and act quickly. Having to think through warning types is counterproductive.

Two comments explain both the Weather Service’s rationale, and why they may be wrong:

I suspect that the experiment in question is driven by a conviction that lack of an enhanced warning or tornado emergency message was a reason, or perhaps THE reason, so many died in Joplin, therefore, an ironclad policy on issuing tornado emergencies will prevent that from ever happening again. If that is so, it would be a classic case of “not seeing the forest for the trees.”

If warning fatigue and lack of visibility of the actual tornado were primary contributing factors to the Joplin death toll, then all the super duper enhanced warning language in the world probably wouldn’t have made much difference.

Yeah, but …

No, it’s exactly the opposite. … The reason most people didn’t react to the warning until it was too late was because they have had two dozen tornado warnings in the past 3 years, and none resulted in a tornado.

And to expand:

The general consensus of the public at large is that most warnings are for some place else. The town where I live has had approximately 10 tornado warning over the last 6 years, out of those 10 only 2 were of any threat to my house. By threat I mean the my town was in the path of the storm. Now all of the warnings were valid for the areas issued. However only 20% even included my area. There are places within my county where over the past 6 years that no warnings were valid, but they were warned just the same. The invalid warning messages being sent are desensitizing the warning message.

There are many competent Sociological, and Psychological studies on warnings, and experts in the field who could help in this matter. Almost all would state that a focused, accurate, and direct warning will work better than a complex, wordy, inaccurate, and irrelevant one. This does not take a PhD in Sociology to understand, as it is a common daily occurrence.

The people in Joplin were in their cars, away from home and out and about because the warnings had become meaningless. If you read the response from the victims, almost all to a person stated that they would not take cover till they saw the storm. Why was that? They had been trained to do so by sitting in their shelters for a storm that was 20 mi to the NE moving away from them. They wasted time sheltering from a storm that was no threat to them. So at that fateful day they did not heed the warnings, used the past training as given by the system, and many died because of it. Also, this was an extremely violent storm, one difficult to survive even with the best of shelters.

That last sentence is the meteorological equivalent of the law-school phrase that good cases make bad law. EF5 tornadoes have wind speeds beyond 200 mph. (Wisconsin has had three — in Colfax June 4, 1958, with 20 killed; in Barneveld June 7, 1984, with seven killed; and in Oakfield July 18, 1996, with none killed. Five other Wisconsin tornadoes before the Fujita scale was created, including the New Richmond tornado of 1899 that killed 117, are estimated to have been  EF5s as well.)

My contention for a few years has been that the STCOPAT warnings are not helpful because they lead to more tornado warnings without actual tornadoes, which lead to ignored tornado warnings. A tornado did actually occur during the third tornado warning last year, which was issued while we were at the Ripon library. My wife, who went through the same tornado spotter training as I did, was driving into Ripon at the time, and didn’t see anything that looked like a tornado at the same time a tornado was causing damage to a farm outside Ripon. The 2010 tornado season started east of Green Lake, and before that in June 2004 a tornado sucked a couple out of their basement near Markesan, killing the man and severely injuring the woman. And that’s been it in 13 years of living in Ripon, which has certainly had more than five tornado warnings in that time.

Wisconsin doesn’t have as many tornadoes as the main parts of Tornado Alley, but we have enough that cause enough damage to make improving how the National Weather Service warns about tornadoes important. If terminology can be improved, that should save lives and prevent injuries. But accuracy is more important. And the Weather Service still issues too many warnings for tornadoes that don’t occur (at least as far as those in the warned area consider) that makes expanding the number of tornado warnings seem like change instead of  proress.