9.4 Cookie-Cutter Weather Warnings: The Discontinuities and 'holes' of Weather Warnings

Friday, 20 June 2014: 9:30 AM
Alpine Ballroom (Resort at Squaw Creek)
Matthew Cappucci, STUDENT, Plymouth, MA

Weather, as all meteorologists know, is quite extraordinary; seemingly simple ingredients such as air, moisture, and heat can combine together to form an endless array of spectacular meteorological phenomena, oftentimes catching the public off guard. If weather is so unique, however, one must wonder why issuing ‘cookie cutter', computer generated weather statements is “the norm”.

The 122 NWS WFOs scattered about the country are instrumental in the forecast and warning of severe weather, issuing over 100 short-range products during severe weather events to keep the public informed, as well as safe. While timely, specific warnings such as these have undoubtedly attributed to thousands of lives saved in the past decade, the definitive criteria required for an event to qualify the issuance of a particular weather warning limits effectiveness.

On February 11, 2003, for example, the National Weather Service office in Lincoln, Illinois was forced to make a tough decision as to which product to issue during a multi-state thundersnow squall line event producing hurricane force winds. The question, however, soon became which product to issue. Due to the short-fused call-to-action of this particular event, as well as the presence of isolated “thundersnow”, WFO Lincoln opted to issue 5 severe thunderstorm warnings, a practice that would normally be considered atypical. Because of the lack of inclusive products, however, the severe thunderstorm warning served as a last resort for forecasters, who recognized the importance of communicating the damaging wind potential to the public. Because of the boundaries placed on normal forecast products, however, events such as these pose a formidable challenge to forecasters attempting to warn of a particular threat.

More noteworthy, however, was the issuance of eleven tornado warnings during hurricane Katrina in 2005 - for areas where tornadoes were not even forecasted to occur! On most occasions, the tornado warning is considered the most significant of all severe weather warnings, and a great pressure is placed on any forecaster who is prompted to issue such a product. As the eyewall of hurricane Katrina moved onshore, wind gusts of 100-120 miles per hour spread across a large portion of the New Orleans metropolitan area, leading to the issuance of tornado warnings for “extreme winds”. While not, in any respect, an actual tornado, the 100+ mile per hour winds led to the issuance of nearly 20 tornado warnings from both the National Weather Service in Baton Rouge, as well as Jackson, MS. The confusion resulting from these tornado warnings eventually resulted in the creation of a new product by the NWS, dubbed the “extreme wind warning”; the EWW may only be issued when “winds of 115 miles per hour are expected to overspread a WFO's CWA within an hour...”. More information pertaining to the Extreme Wind Warning is available online at http://products.weather.gov/PDD/EWW.pdf.

An ideal example of how templated, “cookie-cutter warnings” have failed a community has been the May 2011 tornado that struck Joplin, MS. While tornado warnings had been issued a full 21 minutes in advance, the “cookie- cutter” nature of the warning failed to accurately convey the menacing danger facing the public. The “cookie cutter” warning generation program, furthermore, failed to accurately report the correct movement and location of the EF-5 tornado, which was clearly resolved on radar; this catastrophe could have easily been avoided had forecasters intervened and relied less on the “cookie cutter” nature of today's severe weather warnings; even a single word or sentence can make the difference between a resident ignoring warnings and a resident seeking shelter. While manually writing every tornado or severe thunderstorm warning would be far too time consuming, a simple phrase could help individualize the warning while taking mere seconds to include. In the case of the Joplin tornado, the warnings were not taken seriously until the phrase “this storm is moving into the city of Joplin” was added.

This presentation will be comprised of three sub topics. The first topic of discussion will be the 'holes' of weather warnings, particularly the occasional unusual yet significant weather event that does not meet the requirements of any weather warnings. The second discussion topic will focus on certain events, such as Hurricanes Katrina and Sandy, when traditional weather warnings did not go far enough in alerting the general pubic as to the formidable threats posed by the storms, requiring forecasters to go 'above and beyond'. Finally, the warning systems of other countries will be examined, and potential improvements and changes suggested. The closing will also address the roles of broadcast meteorologists in communicating weather warnings effectively.

On occasion, forecasters must go ‘above and beyond' to convey a potential danger in the most accurate way possible. Regardless of the technicalities, forecasters must be given a greater freedom in issuing severe weather warnings, which must be modified in order to be more inclusive of severe weather events. Many other countries have implemented a simple categorical system of severe weather warnings, issuing instead purple, red, yellow, and green alerts (equivalents of emergencies, warnings, watches, and statements/advisories, respectively). While the degree of severity is conveyed through each of these warning thresholds, each ‘level' of severe weather product may be implemented for a myriad of weather events, regardless of the specifics behind them. One must ask themselves if this may be a more effective, and less ambiguous, means of communicating a threat to the general public.

Please visit the following links for more information regarding “extreme” weather warnings: http://mesonet.agron.iastate.edu/vtec/#2005-O-NEW-KJAN-TO-W-0136/USCOMP-N0R-200508291940 http://www.alabamawx.com/?p=33310

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner