California Earthquake False Alarm Offers a Lesson for Crisis Communicators

Article

An earthquake false alarm nearly four years ago can teach disaster and emergency planners – and any crisis communicators, including insurers – a thing or two about communicating to the public during and after a crisis.

In May 2019, an alert message was sent to the residents in Ridgecrest, California, warning them of a possible earthquake. It turned out to be a false alarm, which had a negative impact on people living in an active earthquake area that was struck by a series of quakes 10 months earlier.

The false alert, which came from the ShakeAlert early warning system that operates along the West Coast, was caused by a mislocated system test message. The alert was followed by a cancellation message: “USGS ShakeAlert message cancelled. Investigating. If you protected yourself, well done.”

The false alarm was bad enough, especially for those who had previously gone through the recent quakes. However, it was the second message that a team of social scientists found fascinating enough to study.

They sought to find out what people who saw the message understood about the threat and their safety, and what actions they believed they should take as a result.

The researchers conducted individual interviews with 40 Ridgecrest residents who got the alerts, as well conducting a series of focus groups in Southern California. The goal was to learn what information people most need after an emergency.

The study showed people were confused about content in the follow-up message that referred to investigatory actions – regardless of whether they experienced the quakes nearly a year earlier – and that those who felt the earlier quakes had a largely negative response to content that complemented them if they took action to protect themselves in response to the initial earthquake early warning.

In a nutshell, people who got the post-alert message found it to be largely ineffective.

People were confused about just what was being “canceled” in the follow-up message. They were unclear whether it was the message that was canceled, or if the earthquake had already occurred or if it was not going to occur. That it was “canceled” also left it unclear whether people should be worried about earthquakes later on, and what they should do to be safe going forward throughout the day.

Recipients in the study were also confused by the word “investigating.” Was it a statement that something went wrong? And if so, what went wrong? Some study participants surmised that the investigation might focus on why the earthquake did not happen, others guessed it was focused on why the message was sent, and others figured it pertained to what consequences the sender of the false message faced.

So, what did those people who received the message want instead?

They simply wanted an explicit statement about their safety, the study showed. The researchers found that what people need most following such an emergency or lack thereof is information that affirms their current level of safety so they can return to normal.

That wasn’t the first time a false alarm raised alarms and then drew scrutiny for the follow-up messaging.

A warning of an incoming missile was sent to cellphones in Hawaii in 2018 reading: “BALLISTIC MISSILE THREAT INBOUND TO HAWAII. SEEK IMMEDIATE SHELTER. THIS IS NOT A DRILL.” That follow-up message was criticized more for its latent timing. It took nearly 40 minutes for a second mobile alert to state there was no missile.

It appears some people, beside the study researchers, are thinking about the aftermath of an errant alert. The earthquake message study notes that the United States Geological Survey has created post-alert messages in advance, while other organizations have also been encouraged to develop these messages ahead of time.

Jeannette Sutton, a University of Albany professor and the corresponding author of the study, has a bottom-line message for those organizations developing these messages.

“They just need to know that they’re safe,” she said.

Sutton says communicators need to consider what people need to know and what they should be doing – not just before and during, but after an event like an earthquake, hurricane, or a wildfire.

“In general, we can do a better job when an event is over,” Sutton said. “How do people know the threat has passed and it’s safe for people to come out, or return, or that it’s safe for them? That would be a takeaway for all hazards.”

Such advice isn’t just important for people’s peace of mind and maintaining their trust in the alerting organization, but it is essential considering that in the absence of good, clear information, people tend to turn to less credible sources to find things out as soon as possible.

That, unfortunately, for many people means going on the misinformation swamp of social media to find out from others what has happened and what to do.

“If you don’t tell people it’s safe, they’re left to do their own information seeking to figure it out,” Sutton said.

She described what is known as convergence behavior, in which people tend to converge at the scene of a disaster, or when they have a need for information or help and they are driven by informational convergence to a source.

Broad crisis communications lessons can be gleaned from the study by Sutton and her coauthors, fellow University of Albany researchers Savanah Crouch and Nicholas Waugh, and Michele M. Wood, at the Department of Public Health at California State University, Fullerton.

The study, which was published in the International Journal of Disaster Risk Reduction, also showed that people had issues with the original earthquake alert message, “Earthquake! Expect shaking. Drop, Cover, Hold On. Protect yourself now.”

One person interviewed told researchers: “It didn’t say where it was expected, whether it was [expected], what the timeframe would be, whether it was an immediate problem. I don’t know how localized it was.”

Sutton said the technology and the nature of the alerts may continue to leave some people unclear about what they are supposed to do.

ShakeAlert, for example, can only provide a warning for a general area, and the timing of the quake is not exact. Additionally, the messages must be short so people can read the text quickly and take appropriate and urgent action, she explained.

“Until we have a system with better technology, we’re going to have some barriers to fight against,” she said.

The study’s take-home message for crisis communicators, including insurers reaching out to insureds during and after a disaster to offer information and services is: consider providing information people need clearly, and consider having a framework for these messages already in place.

Sutton suggested an idea communication may take the form of a simple text message with a phone number of who to call along with a link to basic, necessary information compiled on one webpage, so people can save that on their phone and carry it around until it’s needed.

In contrast, she has seen websites for emergency information that require people to go through several webpages to get to the information they need.

She’s also heard the frustration from people who have gone through disasters who had difficulty finding the information and help that they needed right away.

“Communicating clearly and giving people quick access to the information points that they need would be tremendously useful,” she said. “That’s an ongoing crisis communication issue, helping people to navigate afterwards.”