According to some estimates, Californians face a 99 percent likelihood of experiencing a major earthquake within the next three decades — and odds are good that it'll strike in the vicinity of either Los Angeles or San Francisco. Unfortunately, experts warn that California and the rest of the US are woefully unprepared for such a devastating temblor — even though technology to drastically enhance that preparedness already exists.

That fact was highlighted in late September, when California governor Jerry Brown signed a bill to roll out an earthquake early warning system — called ShakeAlert — across the state. The system, which will offer anywhere from 10 seconds to one minute advanced warning of an impending quake, "will help the public take a number of safety measures to reduce injury and damages," said state spokesperson Greg Renick at the time. "People will be able to take cover ... medical professionals will have a chance to stop procedures, and train operators should have time to slow down and stop the trains."

The only problem? Nobody — not even the state government — is stepping up to pay for it. The legislation explicitly prohibits taxpayer dollars from going towards the system, which is expected to cost $80 million in its first five years, and $11 million a year thereafter to remain operational. "It's been made very clear that the money won't be coming from California, which means it isn't clear where the money will come from at all," says Richard Allen, a seismologist at the University of California, Berkeley, who last week wrote a paper in Nature extolling the importance of financing such a system. "It's clear that we need one. But will it be paid for before the next earthquake, or after?"

Allen, along with collaborators including Caltech and the US Geological Survey, have actually already created a robust prototype of ShakeAlert. The system relies on 400 seismometers, scattered across the state and linked together into what's called the California Integrated Seismic Network. Those sensors detect the very first signs of shaking, and send data back to a processing hub that rapidly generates an alert based on the information being transmitted. That alert includes the magnitude of a coming quake, along with a countdown to the onset of shaking in one's vicinity. Following the quake, the same system "could also give people additional, and very valuable, information," says Thomas Heaton, director of Caltech's Earthquake Engineering Research Laboratory. "Here's what happened, here's where damage occurred, here's which buildings to be worried about."

Several other countries have already seen the benefits of similar systems. In Japan, an alert network established in 2007 offered several seconds of advanced warning to 52 million people prior to a magnitude 9.0 quake that struck in 2011. And last year, sirens rang out in Mexico City around 30 seconds before a magnitude 7.4 earthquake rocked the region. "Japan and Mexico already have these systems. Even Turkey, Taiwan, and Romania have some early warning infrastructure in place," Allen says. "Needless to say, we're quite far behind."

In large part, that's because we simply haven't had a severe earthquake in recent memory. The most recent devastating US earthquake was in 1994, when the Northridge temblor killed 60 people and caused $13 billion in damage. "The sad truth is, people very quickly forget," Allen says. "With earthquakes, it's the thing you could always do next year rather than this year. When resources are limited, there are inevitably competing interests that seem more urgent." Indeed, neither Japan or Mexico installed their early warning systems until shortly after both countries experienced treacherous, deadly quakes. "We can learn from them, or we can repeat that pattern," Allen says. "But I know which option I'd prefer."

While scientists and legislators in California scrounge for the resources to implement ShakeAlert — a process that will entail significant work, including the installation of hundreds more sensors statewide — the rest of the US is likely years off from a nationwide early warning system. Regions across the country have some earthquake sensors that operate under the umbrella of USGS, but "there isn't yet the political will" to upgrade them into a comprehensive alert system, Allen says.

Which means that, for now, the only people in the US who will benefit from earthquake early warning technology are those lucky enough to be testing it. "We could make the system openly available, but it just isn't ready for widespread use yet," Allen says. Right now, ShakeAlert can accurately detect the location and magnitude of minor earthquakes, but it hasn't been tested on simulations of larger temblors. And without additional sensors and more durable equipment, "the system will fail," Allen says. "It may detect some and miss others, and it might not offer a sufficient countdown." Allen also wants to see a public education campaign, focused on using the system and accurately interpreting its alerts, prior to ShakeAlert's widespread distribution.

"Still, that is a dilemma we've discussed amongst ourselves," he says. "We have this vital technology up and running, and we'd really like to be sharing it."