Recently Paul Douglas wrote a column at the Huffington Post regarding the need for different classifications of tornado warnings to cut down on false alarms. He proposed "Tornado Alert" and "Tornado Emergency" based on the threat level of the current weather situation. Under his system, a Tornado Alert would be issued when environmental conditions are favorable for tornado development. When a confirmed tornado is on the ground as reported by spotters, storm chasers, law enforcement, or National Weather Service Doppler radar, a Tornado Emergency would be issued.
The National Weather Service is experimenting with a similar system across portions of the Midwest.
As of April 2, 2012, three levels of tornado warnings are being tested to heighten awareness of the severity of the storm. This is referred to as an “Impact Based Warning” system.
- An "ordinary" tornado warning
- "Particularly dangerous situation”
- "Tornado Emergency”
Example of a NWS tornado warning with the enhanced wording:
THIS IS A PARTICULARLY DANGEROUS SITUATION.
SOURCE... EMERGENCY MANAGEMENT CONFIRMED LARGE AND DESTRUCTIVE TORNADO
IMPACT... COMPLETE DESTRUCTION OF ENTIRE NEIGHBORHOODS IS LIKELY. MANY WELL-BUILT HOMES AND BUSINESSES WILL BE COMPLETELY SWEPT FROM THEIR FOUNDATIONS. DEBRIS WILL BLOCK MOST ROADWAYS. MASS DEVASTATIONS IS HIGHLY LIKELY, MAKING THE AREA UNRECOGNIZABLE TO SURVIVORS. TORNADO MAY BE UN-SURVIVABLE IF SHELTER IS NOT SOUGHT BELOW GROUND LEVEL.
The warning will also be tagged with the appropriate tornado threat level at the end of the warning text to illustrate the magnitude of the event.
I believe that a system similar to the what the National Weather Service is testing, with a few tweaks, would be most effective at communicating the severity of tornadic events. With a tornado warning false alarm rate of 76 percent, according to the National Severe Storms Lab in Norman, Oklahoma, the public is already skeptical about heeding an alert. This alarmingly high statistic is why I understand Douglas’ call for revamping the tornado warning system.
Here are the issues I have with multiple tiers of tornado warnings:
- It will be more terminology for the general public to understand. Sure any meteorologist or weather enthusiast will know the difference, but we need to put ourselves in the shoes of the public. I prefer to keep things simple (the KISS principle). Added complexity to a situation that requires quick decision-making and action could be the difference between life and death. We still have confusion over weather watches and warnings, which have been around for almost 50 years. A recent study conducted at the University of Oklahoma tested the weather knowledge of 762 Texans, Oklahomans, and Californians. It found 36 percent of respondents did not know the difference between a severe weather watch and severe weather warning. That is too high of a number. Adding layers of different tornado warnings on top of that would be ineffective in my opinion.
- New dual-polarization radar currently being installed across the country will able to detect non-weather echoes better. This means debris from tornadoes will be more visible on radar. This technological upgrade will help cut into the false alarm rate over time with increased meteorological skill. I would like to see how dual-pol performs before going through a major change of the tornado warning system.
- Timing. Strong rotation on radar (“a tight velocity couplet”) is a pretty good indicator that a tornado may occur at any time. Under Douglas’ scenario, this would be a Tornado Alert, which may not be taken seriously by the public due to the low-end nature of the tornado alert. Images from National Weather Service radar are not instantaneous – scans are several minutes apart. With a quick moving storm, a tornado could drop out of the sky, and cause considerable damage before the message can be relied to the National Weather Service for upgrading the threat level to a Tornado Emergency. By this time, it is already too late. Minutes earlier, the radar was already providing a pretty good clue that something was up.
How would I change the tornado warning process for saving lives? First of all, I would stay with just the tornado warning. What needs to be improved upon is empathizing the reasoning for the warning. I would use the tornado tags of the experimental warning program as part of the warning announcement: “Tornado warning – radar indicated a possible tornado…”. “Tornado warning – a tornado as been observed in/at [location XYZ]”. This kind of communication will allow the public to make a more informed decision on seeking shelter. Ultimately, citizens are responsible for paying attention to what is going on around them, and their own safety. I’m also in favor of the enhanced wording such as “COMPLETE DESTRUCTION” and “UN-SURVIVABLE” in the warning text to get the point across. At times, this language is necessary to convey the type of impact the storm will have in an area: “Tornado warning – catastrophic damage from a tornado was reported 10 miles west of [location ABC]. Complete destruction of [location ABC] is possible.” Would that not be an immediate call to action?
Science is not at the point yet where it can determine which storms will produce tornadoes. Research is conducted every year to find this answer. What we can do until then is better communicate the weather situation – using words that match the impact, and telling the audience why the warning was issued.