The tornado that struck downtown Atlanta, Georgia on March 14 came perilously close to being a "worst case scenario" - packed stadium, downtown traffic, expensive real estate, little warning. City residents were incredibly lucky to escape without a single casualty and with only about $250 million in damages. The Atlanta tornado, together with other recent urban tornadoes, demonstrates the need to take a different approach to disseminating severe weather warnings in urban areas. After all, it's only a myth that tornadoes stop at city lines.
A major issue that the Atlanta tornado made especially clear is that the proliferation of communications technologies and lack of tornado sirens in urban areas means that it's even more challenging today than ever before to warn urban populations of rapidly changing and violent weather conditions. But in addition to the challenges, there are also significant opportunities for broadening the reach of weather warnings.
Atlanta residents had little if any warning that a tornado was about to strike on Friday evening March 14th. For the thousands of basketball fans who were sitting in the stands at the Georgia Dome, there was no inkling that a tornado would soon sideswipe the stadium with winds stronger than a Category 1 hurricane. According to the Atlanta Journal Constitution, there were very few tornado sirens that were activated in most areas of Fulton County. The main reason for the lack of sirens, the paper stated, was the absence of funds for maintaining extensive siren systems. However, it's doubtful that sirens would have done much good in a loud urban environment with offices, convention centers and sports stadiums. In that environment tornado sirens might blend in with other emergency sirens, such as police cars and ambulances.
In fact, once the tornado warning was issued for Atlanta, the city's first responders stepped up to the plate with their own improvised siren system. Kelvin Cochran, Atlanta's fire chief, told a fire and rescue publication that the warning prompted the city to order fire engines to leave their stations with their sirens and air horns blaring in order to warn those within earshot of the tornado danger.
"That was essentially the only warning residents had," Cochran said.
In addition to Atlanta, many of the cities that have been struck recently by tornadoes haven't had tornado siren systems. For example, authorities in New York City and Salt Lake City lacked sirens to activate when tornadoes struck those locations in the recent past.
According to Roger Edwards of the Storm Prediction Center, budget constraints and misconceptions about tornado risks are the main reasons why siren systems are not more prevalent.
"In conversations with emergency managers and spotter coordinators," Edwards wrote, "I have found that the two most common reasons for a lack of sirens are low budgets and the perception that tornadoes cannot happen in an area. The latter is false; and the former is a matter of fiscal priorities."
The National Weather Service doesn't require tornado sirens for tornado-prone areas, and instead leaves it up to local officials to decide whether or not to install siren systems. What the NWS does call for is for people to own and operate NOAA Weather Radios, which sound an alert whenever a severe weather warning is issued.
NWS spokesman Ron Trumbla told the Atlanta Journal Constitution that for the weather agency, warning dissemination is all about sirens and NOAA Weather Radio.
"We see the sirens as a good outdoor warning system," Trumbla said. "We see the NOAA weather radios as the ideal indoor warning system."
However, in a city NOAA weather radios may not be so ideal. For one thing, they're out of step with the communications landscape of urban areas. Instead of sirens and NOAA Weather Radios, efforts to improve the readiness of urban areas to respond to a tornado threat should instead focus on developing better "push" technologies to get warnings out via cellular phone networks and online services.
In short, NOAA needs to help move weather warning dissemination away from a core reliance on NOAA Weather Radio and into the Web 2.0 world.
People use a wide variety of communications technologies to get their weather information while they're at work, mainly through the Internet. For example, it's now possible for users of PDAs to get an email alert through the cooperation of the associated network that the user is on. People can already subscribe to services such as The Weather Channel's Notify! program to get warnings on their cell phone or PDA. Technology has even advanced to the point where it's possible to transmit basic weather information via umbrella. But such programs are limited, are not led by the NWS, and are often fee-based, which provides a disincentive to some potential users who might benefit from severe weather information.
NOAA's reliance on NOAA Weather Radio doesn't take into account the need to warn urban residents during the day or during a commute time when they can be more easily reached via online push technologies. Take Washington as an example. How many legislative assistants on Capitol Hill, lobbyists on K Street, or diplomats in Foggy Bottom have a NOAA Weather Radio on standby at their desk? My guess is very few. Most urbanites don't listen to NOAA Weather Radio at work, in their car, or on the Metro. Nor do the majority of them have a portable NOAA Weather Radio on standby at a Wizards, Nationals or Redskins game. The radios are probably more commonly used at home rather than at work or in transit.
One potential drawback to a push system that was made clear to me after trying out an early version of a warning dissemination program via text messaging, is the high rate of false alarms. It's likely that if the NWS were to move towards more push technologies that individuals would receive far more false alarms than verified tornado warnings, and this could backfire and actually increase complacency as warnings are consigned to junk mail status. Hopefully the benefit of receiving a timely, accurate warning will outweigh the cost of nuisance false alarms, but it's not clear that would be the case for everyone.
Local emergency management officials and sports stadium owners/operators are likely candidates to lead the charge for creating more disaster-resistant urban communities, and for recognizing the distinct nature of how urbanites communicate compared to those in rural areas. Stadium security is a gaping hole in tornado risk management and general emergency management in this country, and this was made more evident by the Atlanta tornado that managed to frighten but not injure those in attendance at the college basketball tournament game at the Georgia Dome.
In addition, the Atlanta tornado and other recent urban tornadoes should be enough to convince management officials and the general public that the notion that tornadoes don't strike cities is just a myth. If downtown high rises and skyscrapers deterred tornadoes, why hasn't every small town in Kansas, Oklahoma, Nebraska and Texas scrambled to build a skyscraper for tornado protection instead of another storm shelter?
The truth is that tornadoes don't respect manmade structures, nor are they deterred by most natural boundaries, such as rivers, hills, lakes, etc. As Kevin Myatt of the Roanoke Times put it in a recent column, "Tornadoes sometimes even skip over trailer parks."
The scenario of a large tornado striking a major city was simulated in a study by the North Texas Council of Governments and the National Weather Service. The study transposed the major tornado event of May 3, 1999 that affected the Oklahoma City area onto the Dallas-Ft. Worth Metroplex. The result: If the May 3, 1999 outbreak had occurred over North Central Texas, the potential damages could have exceeded $3 billion, with a loss of life in the hundreds or thousands.
And closer to home, the Weather Channel devoted an episode of their scare tactic of a TV show "It Could Happen Tomorrow" to a D.C. tornado scenario. The real-life version of that scenario almost played out with the deadly College Park, Maryland tornado of 2001. That event featured a brief and weak tornado touchdown near the Pentagon.
According to Roger Edwards' online tornado FAQ, since 1997 there have been seven tornado strikes on central business districts of major population centers in the United States, with more if the definition of an urban area is widened to include surrounding areas. Two of those tornadoes, one in Nashville and another in Little Rock, ranked as F-3 twisters, while the others were weaker.
In summary, in order to avoid a major loss of life when the next tornado strikes an American city, NOAA officials should work with tornado prone cities and telecommunications providers to reshape the warnings communication landscape for urban areas. This would be a far more preferable and productive path to take than simply relying on NOAA Weather Radio.