The False Alarm Conundrum
By Ben Yelin, CHHS Research Assistant
On Friday June 1st, 2012, the mid-Atlantic was bombarded with a cluster of severe, and frankly, frightening thunderstorms and tornadoes, which caused 2 injuries and property damage across the Baltimore area. The National Weather Service issued 23 tornado warnings. Turns out, 12 tornadoes were confirmed to touch down in the Washington/Baltimore metropolitan area. The National Weather Service was understandably liberal in issuing tornado warnings, based on the belief of many emergency managers that there should be an abundance of caution responding to potential risks. There is, of course, an equally legitimate concern about issuing too many storm warnings to the point that the public becomes complacent when severe weather alerts are issued. Because of the conflict between having too few and too many severe weather warnings, the National Weather Service and emergency managers must find better ways to effectively communicate risks to the general population.
Media critic Dave Hughes, who runs DCRTV.com, was particularly displeased with what he determined was an “overhyping” of the June 1st storm. The Washington Post’s weather blog, the Capital Weather Gang, wrote a post in which they quoted Mr. Hughes as saying:
“Oh that nutty Washington Post. The paper’s website goes completely over the top with hype for Friday’s storms – tornadoes, hail, high winds, dark clouds, deadly lightning, flash floods. But Saturday’s print Post doesn’t even put the big weather story which caused almost no damage in the metro area, on the front page. Does the DC area media go completely nuts when spring storms threaten?”
The Capital Weather Gang notes that Mr. Hughes raised some valid concerns, and suggests that the National Weather Service is too liberal in issuing severe thunderstorm and tornado warnings. The blog, in a later post, linked to a recent study that lends some credence to Mr. Hughes’ perspective. In 2009, researchers from the journal Weather, Climate and Society found a statistically significant correlation between the number of tornado warnings in a given area and the number of injuries from tornadoes. This may suggest that in areas with more warnings, residents become complacent and do not take warnings as seriously as they should.
There is an equally valid concern that a more conservative approach to issuing tornado warnings would not properly notify the public of potential weather dangers. Tornadoes are unpredictable by nature; one could veer slightly off path and strike an area that was not under a warning. With no warning, residents would not take shelter, and would be exposed not only to the tornado itself, but to the damage of falling debris and other post-tornado hazards—all raising the potential for more significant injuries or fatalities. The conflict between issuing adequate warnings and the potential for “overhyping storms” presents the question of whether authorities should err on the side of caution and issue more warnings, or whether they should instead reserve tornado warnings only when there is a very specific threat.
This, however, is a false dichotomy. The goal of emergency management officials should be to develop an innovative warning system that balances the concern for false alarms with the need to err on the side of caution. The National Oceanic and Atmospheric Administration issued a number of wise recommendations in its report in response to the Joplin, Missouri tornado in May, 2011. (This report was also referenced in the later piece by the Capital Weather Gang). One of NOAA’s key recommendations is to conduct social science research to figure out the types of warnings that will prompt residents to respond. Additionally, the report recommended a tiered warning system that reserved a non-routine warning mechanism for the most extreme events. NOAA also recommends that warnings be more impact-based rather than phenomenon-based for clarity on risk assessment. For example, instead of noting the presence of large funnel clouds and dark green skies, warnings should instead talk about the potential for injury and destruction of property if proper preparedness measures are not taken.
These are only some of the recommendations of the report, but they provide an innovative set of solutions to the dilemma between issuing too many false alarms and issuing too few warnings. Hopefully, officials in the Baltimore/Washington region will consider these recommendations for future severe weather threats.