Demo Now
Buy Now
Menu
Demo Now
Buy Now
blog_header

Helping Business Weather the Storm

Storm-Based Tornado Warnings Reduce False Alarms and Hopefully Will Eventually Lead to Better Public Response

by Mike Eilts, on Apr 17, 2012 12:53:34 PM

After experiencing a tornado in my hometown of Norman, Oklahoma, and hearing all of the stories of many people’s experiences during the event, I am very worried about the public’s general apathy toward tornado warnings and the resulting lack of taking any action.

What worries me? Many things, but my basic premise is that people do not respond because they perceive that most NWS tornado warnings are false alarms, i.e.,” I have lived through many tornado warnings without taking any action so why should I worry this time?”

I believe this lackadaisical response to tornado warnings is mainly caused by the perceived very high false alarm rate for tornado warnings, and there is something we can do about it now! The National Weather Service does a credible job of warning for tornadoes. Average lead-time between the issuance of a tornado warning has gone from zero minutes to 14 minutes in the past two decades mainly due to the use of the NEXRAD Doppler radar network and its capability to observe circulations in storms. But the false alarm ratio is still ~75%, meaning that 3 out of 4 tornado warnings do not subsequently have a tornado in them.

A second reason for the perceived high false alarm rate is the way that media relay information about storm warnings and how local emergency management activate sirens when tornado warnings occur.

The National Weather Service has been issuing their warnings (tornado, severe thunderstorm and flash flood) as “storm-based” warnings for many years now. That means that they do not issue warnings based on political boundaries (like cities or counties). Rather, the warnings are issued as polygons which precisely outline the area that the expert meteorological believes is in the hazardous path. Yet, many communities continue to set off their sirens for the entire city or county.

Also, warning graphics shown by local television meteorologists often show whole counties rather than the actual polygon warnings. And warnings provided by many “weather alerting” services alert for whole counties too.

So, can you imagine if 75% of tornado warnings are false alarms and if you also hear sirens, see graphics or get alerts every time a warning is anywhere in your county, (I get about 3 warnings from a county-warning based service for every warning that actually covers my location) then the total “perceived” false alarm ratio is nearly 90%. This is even a fundamental issue with traditional weather radios, which offer only county-based granularity even when equipped with the S.A.M.E. decoding technology.

We have a weather alerting app, called iMap Weather Radio, that alerts only if your location is within a NWS storm-based polygon. We have over 250,000 users of this app, and we have many testimonials about how we have saved people’s lives during tornados. However, our biggest complaint is: “I did not get an alert for the tornado/severe thunderstorm warning in my area even though the sirens were going off and the television meteorologist was showing I was in the warning.” We actually had to build a tool so that we could investigate each of these complaints……and guess what? EVERY single complaint we have received we have shown that the location of the person was actually NOT in the warning polygon even though they were getting alerts from multiple credible sources!

Two steps can be taken now to help address this issue. First, the technology exists to control siren networks in a way that only those within a warning polygon are activated. There should be an emphasis by federal, state, and municipal authorities to implement the use of such technologies. Second, media outlets needs to cease using county-based displays of warnings. These issues greatly accentuate the perceived false alarm rates that ultimately lead to lackadaisical response from the public!

Topics:CEO Corner

Comments