1.3
The History (and Future) of Tornado Warning Dissemination in the United States
Timothy A. Coleman, Univ. of Alabama, Huntsville, AL
After the success of Fawbush and Miller in forecasting a tornado at Tinker AFB in 1948, and the United States Weather Bureau's lifting of the ban of the word “tornado” in its public forecasts in 1950, public “tornado warnings” began. The U. S. Weather Bureau, in cooperation with other federal, state, and local government agencies, and the news media, found numerous ways to disseminate the imminent threat of a tornado to the public.
Early on, outdoor warning sirens, originally designed to warn of air raids and/or nuclear attack, were also adapted for use in broadcasting tornado warnings. A long single note, or “alert” tone, was emitted from these sirens to indicate tornado warnings, while an oscillating frequency, or “wail”, was used to indicate enemy attack. The Emergency Broadcast System (EBS) replaced the CONELRAD system in 1963, primarily as a system for warning the public of enemy attack. However, some local governments later began to use the EBS for disseminating tornado warnings through television and radio stations. The NOAA Weather Wire Service (NWWS), initially a teletype-based system and now a satellite system, is still the primary direct link between the National Weather Service and the broadcast media.
NOAA Weather Radio (NWR) began as a small network of 50-60 radio stations in the 1950s and 1960s, primarily in coastal areas and serving the marine and aviation communities. After the Super Outbreak of tornadoes across the central and eastern United States in April 1974, congress appropriated the money to expand the network to cover 70% of the country's population with an additional 330 weather radio stations. A 1975 White House Policy Statement declared NWR the official direct government source of information on weather hazards to the public. The broadcasts were 24 hours a day, 7 days a week, and originated from local National Weather Service offices. Specially equipped receivers would activate upon the broadcast of a 1050 Hz tone when a tornado warning was issued, allowing the NWR receiver to alert people even when it was in standby mode, eliminating the need for a person to listen to the broadcast constantly during a period of severe weather. After the Palm Sunday Tornado outbreak in 1994, NWR was expanded further, and currently reaches over 95% of the population of the United States. Specific Area Message Encoding (SAME) was also introduced in the 1990s, allowing the type and specific county of a warning to be transmitted through an additional set of tones. SAME-equipped receivers only activate when a tornado (or other hazard) impacts the county (or counties) at or near the location of the receiver, as programmed by the user. This greatly reduced the number of times a receiver would alert people at a given location, helping to alleviate some of the “cry wolf” syndrome.
Recently, the National Weather Service has introduced “storm-based” warnings, that are issued for the specific area a tornado may affect, not regarding county boundaries. While, in theory, this new system of warnings should even further reduce the area warned for each storm, improving the false alarm ratio even more, outdoor warning sirens and NWR have not yet adjusted to the new technology. In most counties, if any portion of that county is under a tornado warning, all sirens across the county are sounded. This potentially results in more siren activations than previously occurred, since multiple warnings may be in effect for a given county at any time, resulting in multiple siren activations. The same problem exists for NWR, since even SAME technology is only activated on a county basis. Some technologies are being developed that will incorporate the exact area of the storm-based warning into the warning system, allowing only the outdoor sirens and NWR receivers (or some other electronic device) affected by the warning to be activated.
Session 1, Environmental History and the History of Atmospheric Sciences
Wednesday, 20 January 2010, 8:30 AM-10:30 PM, B203
Previous paper