Friday, March 24, 2017

The Future of Tornado Warnings: More Precise, More Lead Time, Fewer False Alarms

Jon Erdman
Published: March 24,2017


An overhaul of the nation's weather warning process, including tornado warnings, currently in development aims to provide more precise warnings with increased lead time to help decision makers and the general public respond.
(MORE: Tornado Central)
The Forecasting a Continuum of Environmental Threats (FACET) program at the National Severe Storms Laboratory in Norman, Oklahoma, seeks to provide forecasters a continuously-updating threat grid to supplement the current warning polygons.

How Warnings are Issued Today

The current warning process has changed little since the1960s, with the exception of subtle tweaks including the use of bullet statements in text to more clearly highlight potential impacts (including the use of tornado and flash flood emergencies) and the use of storm polygons instead of whole counties.
(MORE: Your Odds of Being Hit by a Tornado)
You may receive a warning via your favorite weather app, such as The Weather Channel or Weather Underground, a Wireless Emergency Alert, over the radio or from local television.
If you only hear the warning, you will know that your part of a county or city is in a warning and that you should take shelter immediately.
You're probably used to seeing the graphical depiction of a warning as a polygon either on a website or on television.
If your location is in that polygon, take shelter. If not, it's not a threat, at least not yet.
This paradigm of warnings, together with dual-polarization Doppler radar, severe storm research and a more dense network of spotters has led to an average lead time of 13 minutes for tornado warnings.

Challenges With Current Warnings

We've written before about the false alarm challenge with tornado warnings, due to the limitations of Doppler radar detecting rotation near the ground, among other reasons.
Let's put the detection problem aside and consider the case of a confirmed tornado and its accompanying tornado warning as it's issued now.

1. Some areas get more lead time

In the example tornado warning below, the thunderstorm responsible for the tornado is at the lower left part of the red polygon at the time the warning was issued, near the town of Reasnor.
Assuming this tornado had just developed, prompting this warning, the towns of Newburg, Grinnell and Ewart at the far edge of the warned area get a lot of warning lead time. That's not the case for the town of Reasnor.
Secondly, notice the town of Malcom, near the right edge of the map. It's outside the edge of the tornado warning polygon, so they're not warned yet. But nearby Ewart, for example, gets a lot of lead time.

2. Most warned areas aren't struck by the tornado

Tornado warning polygons will always be larger than any actual tornado tracks, thankfully, due to uncertainty in the track of the parent thunderstorm when a warning is issued.
Consider the infamous Tuscaloosa, Alabama, EF4 tornado during the April 27, 2011, super outbreak.
While this warning technically was correct (a tornado was confirmed within the warned area) and it likely saved lives, the fact is, the large majority of the tornado-warned area was not hit by the tornado.
Tornado warning (red polygon) and actual track (black line) of the Tuscaloosa, Alabama, EF4 tornado of April 27, 2011.
(National Severe Storms Laboratory)
Certainly, nobody that was outside of the massive tornado's path will complain about being over-warned, but the question then becomes whether those same folks will take shelter the next time.
This is where the FACETs program looks to improve warnings.

A Tightened, Continuously Updating Warning

As mentioned earlier, FACET seeks to provide a more precise probability threat map of a severe weather event, such as a tornado.
This threat map, known as PHI or probabilistic hazard information, can ingest both conventional current data such as radar, satellite and surface observations, as well as any high-resolution models, and can be adjusted in real-time by the forecaster.
In the case of the Tuscaloosa tornado, notice the PHI map more tightly represents the area of greatest threat, within the larger tornado warning.
Example of what a probability threat map from FACETs would look like for the April 27, 2011, Tuscaloosa, Alabama, tornado. In this case, the dark red and purple contours of the plume indicate the greatest likelihood of a tornado.
(National Severe Storms Laboratory)
Perhaps most importantly, this map could update as frequently as every minute, giving advanced alert downstream of the conventional tornado-warned area.
(MORE: An Amazing High-Resolution Radar Look at a Tornado)
To see what this would look like animating in a real-time situation, skip to the 1:43 mark in this NSSL video. Again, notice the tighter PHI threat areas, moving forward with time.
The benefits of more precise, accurate warnings with increased lead time are immense. But only if the warnings are heeded. That's where the important cog of social and behavioral science comes in.
"We have social scientists working with users such as emergency managers and broadcasters to help determine the best ways to convey this information," said Alan Gerard, deputy division chief of the National Severe Storm Laboratory's Warning Research and Development Division.
FACET envisions the PHI maps and data could be used, for example, to prompt a "tornado threat increasing" alert when the PHI threat plume is pushing toward your area but not yet close enough for a tornado warning, which would provide valuable extra lead time.
Incidentally, the PHI concept isn't simply for tornado warnings.
Gerard said while the current focus is on hazards associated with thunderstorms such as tornadoes, hail, lightning and flash flooding, this concept can also be extended to other hazardous weather, including winter weather.

When Will This Be the Standard?

It will be several years before these probability threat maps accompany standard warnings issued from the National Weather Service.
First up, some aspects of FACETs/PHI will be rolled out at NOAA's Storm Prediction Center and the Weather Prediction Center in 2018, according to Gerard.
Beyond that, implementation in local National Weather Service forecast offices and for private meteorological companies looks to come online in 2021.
However, Gerard said whether these PHI-based warnings will completely replace the current polygon warnings remains uncertain, dependent on National Weather Service requirements.
"A key premise we are operating under in FACETs is 'do no harm' to our current warning system. We will want to make sure there is no confusion before we would implement PHI operationally."
Jonathan Erdman is a senior meteorologist at weather.com and has been an incurable weather geek since a tornado narrowly missed his childhood home in Wisconsin at age 7. Follow him on Facebook and Twitter.

MORE: Strangest Tornado Damage (PHOTOS)



The Weather Company’s primary journalistic mission is to report on breaking weather news, the environment and the importance of science to our lives. This story does not necessarily represent the position of our parent company, IBM.

No comments:

Post a Comment