NOAA: Better Methods for Tornado Predictability - WBOY - Clarksburg, Morgantown: News, Sports, Weather

NOAA: Better Methods for Tornado Predictability

Posted: Updated:

Researchers develop method to better predict severity of tornado outbreaks

April 1, 2013

Contact: Keli Pirtle, 405-325-6933


Weather forecasters would love to be able to predict the severity of tornado outbreaks, giving those in their path potentially life-saving information.


Now, using new experimental high-resolution forecast models, researchers have developed a method to help forecasters better predict the severity of tornado outbreaks. A report on this work by scientists at NOAA's National Severe Storms Laboratory and the NOAA Cooperative Institute for Mesoscale Meteorological Studies at the University of Oklahoma is available online and will be included in the April 2013 print edition of the journal of Weather and Forecasting.

Researchers found that the amount of rotation in the rapidly rising air within simulated storms – a measure known as updraft helicity -- was strongly related to the track-length of observed tornadoes in previous severe weather events.  

These results show the potential for forecasters to use this measure to make very reliable predictions of the magnitude of spring season tornado outbreaks. Researchers also found that they could predict with a high degree of certainty, the most destructive tornado outbreaks, which typically occur only once every few years.

Although the next generation of weather prediction models cannot provide exact predictions of thunderstorm timing and location, these models can directly simulate the characteristics of individual storms that produce severe weather while still covering large areas of the globe. However, predicting storms over large areas while simultaneously zooming in on potential tornado-spawning portions of a storm remains challenging.

Fortunately, researchers believe that models can reveal important clues about the type of severe weather storms can spawn in localized areas. The problem is similar to using Doppler radar observations to predict whether a storm is producing a tornado or not. Doppler radars don't have sufficient resolution to detect tornadoes. However, if a "hook echo" feature is present in the radar data along with a strong signal for rotation, forecasters recognize that there is a good chance the storm is producing a tornado.  

In a similar way, although the next generation of forecast models will not have sufficient resolution to simulate tornadoes, they can skillfully predict the general characteristics of tornado-producing storms. Thus, when the forecast model predicts structures typically seen with observed tornadoes, this can indicate to forecasters that tornadic storms are likely.  

In the study, researchers used forecast examples from three major tornado outbreaks that occurred during 2011, a year with one of the most destructive tornado seasons on record. For each of the three cases – the "Super Outbreak" of tornadoes centered over Mississippi, Alabama, and Tennessee on April 27,  outbreaks centered over North Carolina on April 16, and Oklahoma and Kansas outbreaks on May 24 – the technique produced very reliable predictions of cumulative tornado path lengths.  

The work conducted for this study emerged from recent NOAA Hazardous Weather Test bed Spring forecasting experiments. The NOAA Storm Prediction Center and National Severe Storms Laboratory organize these annual experiments to test new concepts and technologies for improving the prediction of hazardous weather and to accelerate the transfer of promising new tools from research to operations. Work is ongoing to find innovative ways to address the challenge of extracting useful information from high-resolution forecasts of thunderstorms to contribute to further advances in severe weather forecasting. 

Powered by Frankly