Weather and Society Watch
Tornado Warning Communication and Emergency Manager Decision-Making
by Cedar League*, Brenda Philips**, Ellen J. Bass***
According to the Storm Prediction Center 1 , 2011 is tied as the second deadliest year on record for tornado deaths with 552 direct tornado fatalities. The May 22 Joplin, Mo., tornado itself accounted for 158 deaths and over 1000 injuries. 2011's 12 natural disasters accounted for $1 billion or more in damages each 2 .
Despite these losses, one positive outcome has been the countless examples of how these events have brought citizens closer, working together to rebuild their homes, their communities, and their lives. The weather enterprise has also come together to address the impacts of these events, and there is a timely opportunity to “strike while the iron is hot” to raise public awareness about severe weather threats, warnings, preparedness, protective action, and response.
The weather enterprise refers to the National Weather Service (NWS), emergency managers (EMs) and storm spotters, the media, and the private weather sector. An integrated warning system describes how these groups (ideally) work together to forecast, detect, and communicate severe weather threats to the public 3 .
Emergency managers are an important link between the NWS, the media, and the public. EMs are responsible for communicating severe weather information and tornado warnings to the public on a regional, city, or neighborhood scale. This information is communicated through a variety of methods, such as activating sirens, notifying schools, hospitals and other critical services, activating the emergency alert system, cable television interrupt, social media, or a mass notification system.
Yet, communicating the uncertainty of if, when and where a tornado may occur is a challenge for EMs (as it is for the entire weather enterprise). P rior research indicates that social, environmental, and technological forces shape the organizational decision-making processes of emergency managers 4 . Emergency managers are also diverse in terms of their background, training, and access to resources; thus, there is no one size fits all approach to EM operations and decision-making 5 .
The Center for Collaborative Adaptive Sensing of the Atmosphere (CASA) 6 is conducting research that highlights the critical role EMs play in an integrated warning system while also informing the development of a network of low-power, high resolution, X-band radars. CASA's new technology has the potential to increase lead-time for tornado warnings through earlier detection, reduce false alarms, and reduce uncertainty in end-user decision-making. But i n order to make advances in new technology most effective, a better understanding of end-users' needs and decision-making processes is needed, as well as the behavioural response to severe weather events.
Since 2007, CASA's End User Integration team, comprised of systems engineers, geographers, a resource economist, sociologists, a political scientist and meteorologists, has used a mixed methods approach including interviews, surveys, focus groups, observations, and part-task experiments with Oklahoma and Texas emergency managers. The goal of these studies includes identifying: 1) the types of weather warning systems used to disseminate warnings to the public, including the advantages and disadvantages of those systems; 2) EM warning practices including subregional warnings and preferred lead times for warnings; and 3) the use of weather products and storm spotters in EM decision-making.
CASA's End User Integration team recently conducted the “Tornado Warning and Technology Survey” with 125 Oklahoma EMs and 157 Texas EMs. Furthermore, focus groups were conducted with 55 Oklahoma EMs and 10 Texas emergency officials to uncover qualitative information and more in-depth detail about their warning and decision-making process during tornado events. Below are select findings from this research, including quotes taken from Oklahoma EMs during focus group sessions 7 .
Sirens (or outdoor warning systems) were by far the most prevalent warning system used by EMs. However, EMs noted several downsides to siren systems. Sirens are not meant to be heard indoors; however, some EMs sense that the public expects to hear them regardless of their location. Sirens also evoke confusion among the public: “ Our sirens have the capability where they can alert for tornadoes… high winds, chemical spills… they have different tones. But in our area, we do not do that because of the confusion that it can emit from the sirens… We sound them for tornadoes, not high winds.”
Emergency managers rely on multiple warning methods to reach different sectors of the public and to ensure redundancy in case a single notification method fails. They actively encourage purchasing of weather radios: “ We really promote those because that gets to people regardless of whether they are within earshot of the outdoor warning siren or not .” While EMs from smaller jurisdictions use mass notification systems to disseminate tornado warnings, those in larger jurisdictions find these systems too limiting with respect to call rate, and it would take far too long for all of the messages to be sent.
With the NWS's move to storm-based or polygon warnings, the CASA team investigated EMs' subregional warning capabilities. While less than half had the capability, even fewer were actually using it. One risk averse EM noted, “ If we sound the sirens, it's for everybody .” Other EMs said they are simply too busy to make a decision about warning a smaller area of the public. Some EMs are also concerned about over-warning the public, which they believe prevents the public from taking protective actions.\
Timing of Warnings
As some EMs are risk averse, while others are concerned about over-warning, the issue of EM tornado warning dissemination is critical. EMs do not always activate the sirens following an official NWS warning and will sometimes activate even if an official NWS warning is not in effect. As one EM noted, “ No, I do not send out warnings automatically just because I'm included in the warning area… I consider warning areas, but I also want to make sure there's an imminent threat because we found that if you put out too many warnings, people become complacent, and also if you put out a warning too early, then they don't react in the way that we want them to .” Another stated “ If I can see something, if I've got eyes on something… we will always err on the side of caution .”
Warning Lead Time
The team investigated the EM's ideal lead time to issue a tornado warning. There was some consensus for twenty minutes: “ Twenty minutes would be really nice… We can do what we need to do. If you're not going to do what you need to do in 20 minutes, forget it. ” One EM wanted more than 20 minutes lead time in order to prepare for an event; however, he did not want to give the public any more than 20 minutes, as he believed they would go out and do something careless.
The majority of Emergency Managers rely on storm spotters as their “eyes on the ground” to verify severe weather. Spotters are typically public works or SkyWarn volunteers. While spotters are a critical component of EM decision-making, one EM notes, “ One of the problems you find with spotters is nightfall. Once it becomes nightfall or it's rain-wrapped, they can't see it. ”
Discussion and Future Work
To reduce future losses, the weather enterprise can learn from the tornado events of 2011, working collaboratively across multiple disciplines to look for a way to move forward. There have already been several notable efforts in 2011 to address the societal and physical impacts of severe weather, and what can be improved upon in the future (e.g. NWS Service Assessments for the Joplin Tornado and the Historic Tornadoes of April 2011, Integrated Warning Team meetings [summer 2011], Weather and Society*Integrated Studies Workshop [August 2011], National Weather Association Town Hall Meeting for Alabama survivors [October 2011], and the Weather Ready Nation Conference [December 2011]).
Communicating a tornado warning to the public is not as easy as just “pushing a button.” Furthermore, new technology may not automatically translate into better warnings and warning decisions. A better understanding of the warning process and of public response to tornado events is needed to reduce negative impacts of natural hazard events. It is for this reason that CASA is incorporating social science research into the design of the radar network to make sure the new radar system meets the needs of the user.
For emergency managers, a real benefit of the CASA network is its ability to provide neighborhood-scale information. For example, on May 24, 2011, an emergency manager in Newcastle, Okla., was able to detect a shift in the direction of a tornado headed for his jurisdiction using CASA radar, allowing enough time for officials to relocate first responders and keep them safe from impact. This is just one example of how high resolution, low level data can benefit end-user decision making for very localized events.
CASA is now in the process of deploying a five-year, $10 million project in collaboration with the North Central Texas Council of Governments in the Dallas-Fort Worth area beginning February 2012 to establish the nation's only Urban Weather Demonstration System. Our vision is to create a national model for weather observation from a technological, operational, and socioeconomic benefits perspective, where multiple stakeholders jointly fund the operation of radar observation networks. We will continue our research with the emergency management community, with the goal of reducing losses to severe weather and flooding events.
* Cedar League (firstname.lastname@example.org) holds a master of arts in geography from the University of Colorado, Colorado Springs (UCCS). She is currently a research assistant with the Center for Collaborative Adaptive Sensing of Atmosphere at UCCS and will be presenting her research on Jan. 24 at 2:00 p.m. in room 243 at the 7th Symposium on Policy and Socio-Economic Research.
** Brenda Philips (email@example.com) is a resource economist at the University of Massachusetts, Amherst. She is director of the End User Integration Thrust for the Center for Collaborative Adaptive Sensing of Atmosphere. She holds a master of business administration from Yale and is a working towards her doctorate in Resource Economics at the University of Massachusetts, Amherst. To learn more about CASA's user-centered research, please visit presentation TJ25.5 on Jan. 26 at 4:30 PM in room 242, and poster 648 on Jan. 25 in Hall E.
*** Ellen J. Bass (firstname.lastname@example.org) is an associate professor of Systems and Information Engineering at the University of Virginia. She holds a doctorate from Georgia Institute of Technology in Systems Engineering. To learn more about EM decision making, see Bass et al. 2009. Incorporating emergency management needs in the development of weather radar networks. Journal of Emergency Management, 7(1): 45-52
This work was supported in part by the Engineering Research Centers Program of the National Science Foundation under NSF Award Number 0313747. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Science Foundation.
 Storm Prediction Center, 2011. Available from: www.spc.noaa.gov/climo/torn/fataltorn.html (Accessed 19 Dec. 2011)
 National Oceanic and Atmospheric Administration, 2011. Available from: http://www.noaa.gov/extreme2011/ (Accessed 19 Dec. 2011).
 Doswell C.A. III, Moller A.R., Brooks H.E. 1999. Storm spotting and public awareness since the first tornado forecasts of 1948. Weather and Forecasting , 14: 544-557.
 Donner W. 2008. Decision making as community adaptation: a case study of emergency managers in Oklahoma. Disasters , 32: 292-302.
 Bass E.J., Baumgart L.A., Philips B., et al. 2009. Incorporating emergency management needs in the development of weather radar networks. Journal of Emergency Management, 7(1): 45-52 .
CASA is a National Science Foundation Engineering Research Center that has developed a new paradigm for weather warning and forecast based on small, densely spaced networks of x-Band radars. See McLaughlin, D. et al. 2009. Short-wavelength technology and the potential for distributed networks of small radar systems. Bulletin of the American Meteorological Society , 90(12): 1797-1817.
 These results were published in the following article: League C.E., Díaz W., Philips B., et al. 2010. Emergency manager decision-making and tornado warning communication. Meteorological Applications, 17:163–172.