Fourteen. That is how many U.S. wildfires in the past ten years produced documented complaints, investigations, or official findings that evacuation warnings were never sent, sent too late, or never reached the people in danger. Fourteen communities told, after the smoke cleared, that the alert system failed them.
The fires are scattered across the map; Tennessee, California, Oregon, Colorado, Texas, Nevada, Hawaii. The death tolls differ. The failure pattern does not.
It looks like this. Officials rely on opt-in alert systems that most residents never join. CodeRED in Paradise. Nixle and SoCoAlert in Sonoma County. CodeRED again in Klamath River. Enrollment typically runs 15 to 40 percent, and "enrolled" does not mean "warned." In Paradise, fewer than 40 percent of residents had signed up before the Camp Fire; of those, only about a third answered the alert call live. In Sonoma County during the 2017 Tubbs Fire, the first emergency message went to just 213 phone numbers in an area with more than 13,000 residents. The rest were unreachable by design.
It looks like this. Officials sometimes choose not to warn at all. Sonoma County deliberately withheld Wireless Emergency Alerts during the Tubbs Fire, fearing a mass alert would cause panic; a state review later condemned the "duplication, inconsistency and confusion" of the whole notification effort. Butte County made the same choice during the Camp Fire, which killed 85 people. Boulder County made it again during the 2021 Marshall Fire. And in Maui, as Lahaina burned, county officials decided not to sound any of 80 outdoor warning sirens, while the cellphone alert that did go out arrived hours late. One hundred and two people died.
It looks like this. The infrastructure that warnings ride on dies with the disaster. Cell towers burn or lose power mid-evacuation; the Paradise sheriff's office lost its own communications as the town fled. Sirens and broadcast alert systems go untested for years, then fail silently. In Washoe County, Nevada, this past August, emergency managers believed they were sending EAS broadcast alerts during the Hawk Fire; the alerts never aired, and the county only learned that when a reporter checked the station logs. In Texas, residents of Fritch said the town's sirens sounded too late as the Smokehouse Creek fire approached. In Oregon, Jackson County never issued a broadcast alert via TV or radio as the Almeda fire tore through Talent and Phoenix. In Woolsey's path, officials sent no mass-notification alerts at all, relying on Twitter posts instead.
And it looks like this. Nobody checks whether the warning reached the people most likely to die. After-action reviews keep finding the same equity gaps; seniors living alone, disabled residents, Spanish-speaking households, and poorer communities miss the alerts most. In Washoe County, the family of the Hawk Fire's sole victim told reporters they never received an evacuation alert; Spanish-language barriers were part of the story. In Pacific Palisades, some residents closest to the fire's start got their first alert 40 to 90 minutes after neighboring homes were already burning, because the alert process was so tangled across agencies that no one could say who was responsible for sending it.
These were not surprises. Every one of these failures was foreseeable, and every one of them is fixable. Opt-in systems could be replaced or supplemented with universal ones. Sirens could be tested instead of assumed. Backup communications could be planned rather than improvised. And broadcast radio — the one channel that keeps transmitting when cell towers go dark — could be given the prominent role it has earned. After the Camp Fire destroyed Paradise, the town installed 21 new sirens and an AM radio station. Paradise learned.
Serious preparation has a shape, and it is not complicated. Multiple channels carrying the same message, so that the failure of any one does not silence the warning. Systems tested before the fire, not discovered broken during it. Universal alerts that reach everyone in the danger zone without requiring a prior signup, a charged phone, and English fluency. And broadcast radio — especially AM, which rides out power failures and reaches every car on the road — treated as a primary warning channel rather than an afterthought. During the Palisades fire, the volunteer-built smartphone app Watch Duty relayed evacuation information faster than the official notification systems; a volunteer app should not be outperforming the government. None of this is exotic technology. It is the difference between planning for the world as we wish it were and planning for the world as it is.
The accountability record so far is thin. After Maui, the county emergency management director resigned. After Gatlinburg, a federal judge stripped the National Park Service of its immunity in a lawsuit brought by victims' families. But mostly the pattern is reviews, findings, and hundreds of recommendations — and then the next fire, the next county, the same failures. Recommendations are not accountability.
Which brings us to the harder question. Two official reviews in this decade have tied warning failures directly to deaths. In Gatlinburg, where the county admitted it never sent a mobile alert during the 2016 fires that killed 14 people, an independent review concluded that insufficient warning "cost lives." And in the 724-page Palisades and Eaton report released this month, investigators found that delayed alerts "contributed to 18 of the 19" Eaton deaths. Not bad luck. Not an act of God. A system that officials chose not to build, not to test, and not to use.
The fires will keep coming. The failure to prepare for them is a choice. At what point do we begin to hold public officials responsible for failing to learn from past failures?
No comments:
Post a Comment