When the National Weather Service issued a tornado warning for northeastern Allen County, Indiana at 5:14 p.m. Aug. 24, 2016, more people in the county got their initial alerts from mobile phones than any other information source, according to an informal, online survey conducted by the publisher of this blog.

Of people who indicated they were anywhere in Allen County at the time of the warning, slightly more than 32 percent said they first learned of the warning via their mobile phones (including “Wireless Emergency Alerts,” alerts from apps, text messages, social media, etc.). Television was the second-most-frequently cited initial warning source, at nearly 21 percent. Outdoor warning sirens, commonly referred to as “tornado sirens,” came in third, at 17 percent. Just under eight percent of respondents credited NOAA Weather Radio as their initial warning source.

The warned storm created a tornado in northeastern Allen County at approximately 5:27, according to a report from the northern Indiana weather forecast office of the National Weather Service. That was about 13 minutes after the NWS issued the warning. The tornado stayed on the ground until approximately 5:39, cutting a five-and-a-quarter-mile path to the northeast and doing damage consistent with the EF-3 rating on the enhanced Fujita scale.

Warned immediately

Nearly 62 percent of the survey’s 167 respondents indicated that they received the warning “immediately.” Another 19 percent said they received the warning with 10 minutes of its issuance, which would still have been a few minutes before the tornado touched down. Nearly 20 percent of respondents did not learn of the warning any sooner than 30 minutes after the NWS issued it, well after the tornado had lifted.

The NWS drew a five-sided polygon that enclosed 144 square miles to indicate the portion of northeastern Allen County to which the tornado warning applied. As a whole, Allen County encompasses approximately 660 square miles, so the warning polygon included less than a fourth of the county’s total area. The tornado’s entire path remained within the warning polygon, so people outside the polygon were not in danger.

Nearly half of all respondents indicated that they knew immediately whether they were within the warning polygon. Another quarter of respondents knew within five minutes whether they were in the warned part of the county. Nearly 18 percent, however, never knew with certainty before the storm passed whether they were in danger.

Best sources for location information

Among respondents who knew immediately whether they were within the warned area, nearly a third received their initial warning via their mobile phones, 24 percent via TV, 11 percent via NOAA Weather Radio and eight percent each via broadcast radio, amateur “ham” radio or outdoor warning sirens.

Among those who never knew with certainty before the storm passed whether they were in the danger area, there was a tie for the top response on how they first learned of the warning; 26 percent each reported TV and outdoor warning siren. A fifth received initial word of the warning from someone they know and 17 percent received it via their mobile phones.

Performance of outdoor warning sirens

Every operating outdoor warning siren in Allen County, including sirens miles southwest of the warning polygon, sounded shortly after the NWS issued the tornado warning.

Slightly more than 59 percent of respondents reported hearing an outdoor warning siren sometime during the hour of the warning, even if it wasn’t their initial warning source. Nearly 41 percent of respondents never heard an outdoor warning siren. The survey did not ask respondents whether they were outdoors at the time of the warning.

Respondents whose initial warning came from outdoor warning sirens were nearly evenly split with regard to their awareness of whether they were actually in the warned area. Slightly more than 24 percent reported knowing immediately, nearly 28 percent reported knowing within five minutes, nearly 21 percent said they knew within 10 minutes and nearly 28 percent indicated that they never knew with certainty before the storm passed whether they were in the warned area.

Of respondents who got their first notifications of the warning from outdoor warning sirens, more than three fourths were within the city limits of Fort Wayne at the time. Seven percent were in the city of New Haven and no more than four percent reported being in any other location within Allen County. The vast majority of the county’s outdoor warning sirens are located within Fort Wayne and New Haven. Approximately 71 percent of Allen County’s population resides in Fort Wayne and probably even more are employed in Fort Wayne.

People close to the warning

Only seven percent of respondents reported that at the time of the warning, they were in the city of Woodburn, the town of Leo-Cedarville or rural northeastern Allen County (in other words, in or near the warning polygon) at the time of the warning. Of those, the initial warning source was more evenly divided, with 25 percent each reporting mobile phone or television and nearly 17 percent each reporting NOAA Weather Radio, outdoor warning siren or amateur “ham” radio.

Only a third of these respondents reported ever hearing an outdoor warning siren, even if it wasn’t their initial warning source. Eight percent never knew with certainty before the storm passed whether they were in the warned area.

Conclusions

Readers should use caution drawing conclusions from these data, because the survey that generated them was informal, not scientific, and the number of respondents fell far short of the number required for a representative sample of people who were in Allen County at the time of the warning.

In terms of improving the tornado warning system, it would appear that steps to increase awareness of warned locations could be helpful. Outdoor warning sirens, of course, do not provide location information. For that reason, it’s surprising that some respondents who reported initially receiving the warning via outdoor warning sirens also reported knowing immediately whether they were in the warned area. Because Allen County activates all of its sirens for every warning, citizens cannot assume that their ability to hear a siren indicates that they are near the warning polygon, but it’s possible that some people do not realize this and that additional public education might be helpful.

Readers might be surprised that a number of respondents who initially received the warning via television were not immediately aware of whether they were in the warned area. If we assume the TV meteorologists who were on the air live at the time described the warned area, it is possible that those respondents initially learned of the warning not from live meteorologists, but from on-screen textual information.

Finally, it appears that a significant portion of people initially learned of the warning via their mobile phones. The time of day might have skewed those results, because many people were likely commuting at the time, and therefore away from TVs and weather radios. Also, commuters would have been more likely to hear outdoor warning sirens than would be people inside workplaces or homes. Still, it’s possible that the Wireless Emergency Alert system that’s enabled by default on all modern smartphones proved itself to be a valuable source of warning information.