Over-Alerting: Episode 5

FALSE ALERTS

So far, relevance has focused on alerts that require action 🏃‍♀️‍➡️ (they are urgent, severe, and certain), alerts that you can actually do something about 😴 (as in, not told to shelter inside when it's the middle of the night), and alerts that are geographically specific 🌎 (you're in the area of the threat and the threat is near you). Today, we'll be talking about that one historical event that is seared into the minds around the world. THE FALSE ALERT. 

False alerts can result from many things - for instance, uncontrollable system malfunctions, such as in the January 2025 L.A. wildfires 🔥 when a message got "stuck" and echoed through the county for hours and hours afterwards. 

Another type of false alert is when a test goes awry, such as in 2020, when ShakeAlert issued a false alert to Ridgecrest, CA 🏜️ - the same location that experienced significant shaking the year before. 

Another type of false alert is the kind that is summed up in one word: "Hawaii." 🌴 This one references the event in 2018 when a WEA was issued for an incoming ICBM 🎇 and not canceled until 38 minutes later ⏰ . 

False alerts are clear examples of over-alerting. It's the alert that should not go out. It's also one of the reasons that alerting authorities are so reluctant to "push the button." And rightly so; no one wants to be remembered for the alert that resulted in calls to the FCC or congressional hearings. 

But we have tools to address each of these situations. 

For the first instance above - including the location and the time in that message would have dramatically reduced confusion and uncertainty about who the message was meant for and when action was being recommended. 

For the second instance above - USGS had already developed messages for EXACTLY this circumstance and quickly issued a retraction, letting people know they were investigating the issue. 

In the third instance above - well, that one stays in our collective memories because there was no plan to get a cancellation message out immediately, but having one in hand would have averted the 38 minutes of terror that many people experienced. 

If you find these summaries interesting, you should take a look at the complete paper written by Jeannette Sutton and Michele M Wood. There is a LOT more detail, including quotes, in the full paper. 

You can find the open-access version here: Sutton, J. & Wood, M. Opting Out.