COPALIS BEACH, WASH. - When Japan first issued a "megaquake" warning last week, Washington state seismologist Harold Tobin watched closely.
The advisory came after a 7.1-magnitude earthquake struck the southern island of Kyushu. While the quake caused little major damage - the largest tsunami it generated was said to have reached knee-deep - it was not the biggest concern.
Seismologists had previously worried that the quake would create stresses that could set off a bomb ticking off the coast: Japan's Nankai Trench, arguably the country's most dangerous fault line. The subduction zone could generate tsunami waves 100 feet high, killing nearly a third of a million people, according to Japanese government estimates.
Did the smaller quake mean the "big one" was around the corner? No one could say for sure, but the odds suddenly increased - if only by a few percentage points.
"Exactly what would keep me up at night," Tobin said, if it happened on the U.S. West Coast.
In Japan, the advisory led officials to close beaches, cancel fireworks celebrations and delay trains, as people rushed to stock up on emergency supplies.
In the US, Tobin said, "we don't have such a protocol."
However, there is a similarly dangerous fault line: the Cascadia subduction zone.
An estimated 14,000 people would die in Oregon and Washington if a magnitude 9.0 earthquake on the Cascadia fault and the resulting tsunami struck.
But if a smaller earthquake occurs near Cascadia in Japan, like the one that just occurred in Japan, seismologists on the ground will have to decide whether and how to warn the public.
It's the scenario Tobin has been pondering for years: If he finds evidence that a devastating earthquake is even slightly more likely, what justifies the alarm? If the odds are high that you'd cry wolf - should you?
"You don't want a mass evacuation panic that isn't justified, but you also don't want people to go their own way," Tobin said.
His dilemma is partly a product of this strange time in Tobin's field: Researchers think they have pinpointed the triggers, or precursors, of earthquakes in the world's most seismic regions, but the science is far from settled. And even if the odds of an earthquake could be higher, the odds remain low. That leaves important questions about when to issue a warning.
On a cold summer day in Washington state, Tobin and a dozen other scientists canoed up the Copalis River to a graveyard of cedar trees that were cleared 324 years ago.
A kingfisher chirped and the wind sent shivers through the tall, golden grass. It is a peaceful spot about a mile from the Pacific coast that tells the story of a violent day.
On January 26, 1700, an earthquake on the Cascadia Fault sent the forest crashing more than 3 feet. Shortly afterward, a tsunami perhaps 100 feet high swept through it at 20 or 30 mph.
The scientists visited the forest to see the geological evidence of the Cascadia earthquake in real life. Every now and then they jumped out of their canoes, dug through the mud and pulled out a 300-year-old pine cone as evidence.
Experts know that the quake was at least a magnitude 8.7. That's how powerful the quake must have been to send the wave recorded in Japan around the world.
"Some of the very best written accounts of our 1700s tsunami come from Nankai," said Brian Atwater, a USGS geologist emeritus who led the canoe flotilla. Atwater used those Japanese accounts, along with plants buried in tsunami-sedimented sand and data from the rings of Washington cedar trees, to piece together the story of that tsunami.
Research by USGS geophysicist Danny Brothers suggests that there have likely been at least 30 major earthquakes in the past 14,200 years along parts of the Cascadia subduction zone, which runs along the U.S. West Coast from Northern California to northern Vancouver Island. A major earthquake there can be expected at least once every 450-500 years, on average.
But Cascadia has been silent for years; some scientists say that's because a large part of it is "stuck" and building up tension. If it ruptures, a chunk of the seafloor will shoot forward - perhaps tens of meters or more. The vertical displacement of the seafloor will send a tsunami toward the coast.
"This is going to be the worst natural disaster in our country's history," said Robert Ezelle, director of the Washington state emergency management department.
For seismologists, the big question now is how to predict this future violence. Rapidly developing research suggests that faults like Cascadia and Nankai could be sending out warning signals: a smaller quake called a foreshock, or a subtle groan that only sensors can detect, what scientists call a slow-slip event.
In Tobin's nightmare scenario, the Cascadia fault suddenly lets out that kind of groan. And then - what to do?
If a major earthquake were to strike Cascadia, more than 100,000 people would be injured, according to predictions, assuming the quake strikes when few people are on the beach. The quake would last five minutes. Tsunami waves would lash the coast for 10 hours.
Inland slopes would liquefy, causing roads and bridges to disappear. About 620,000 buildings would be severely damaged or collapse, including an estimated 100 hospitals and 2,000 schools.
"We are not prepared," Ezelle said candidly.
Washington state is warning residents that they will likely have to fend for themselves and the elements for two weeks.
"It's about neighbors taking care of each other," Ezelle said.
A map of the Pacific Ring of Fire, where tectonic plates meet to form subduction zones and volcanoes, makes Ezelle feel particularly uneasy.
"Over the last 50 to 60 years, you'll see that every subduction zone rupture has had a major rupture, with the exception of Cascadia," he said.
Japan lifted its "megaquake" warning on Thursday after no unusual activity was detected at the Nankai Trough.
In a similar situation in New Zealand in 2016, things played out slightly differently.
In November, a 7.8 magnitude earthquake struck Kaikoura on the eastern side of New Zealand's South Island, killing two people and causing more than $1 billion in damage.
A day later, scientists detected a few centimetres of movement near the coast of the North Island via satellite monitoring. Subtle tremors came from the Hikurangi Margin, a subduction zone and the country's largest fault line, which lies directly beneath the capital Wellington.
It was a slow earthquake, the laziness of the seismic world, caused by the Kaikoura quake. Such earthquakes release their energy slowly over weeks or months, and produce no detectable tremors. Scientists first recognized their existence about two decades ago, thanks to advances in GPS technology.
Some scientists, such as Tobin and geophysicist Laura Wallace, think that these slow-slip events sometimes precede large subduction zone quakes. Scientists recorded a slow-slip event in 2011, before the magnitude 9 Tohoku earthquake and tsunami in Japan that killed more than 18,000 people and triggered the Fukushima nuclear disaster. A similar pattern played out in 2014, before a magnitude 8.1 earthquake struck Chile.
Wallace, who worked for New Zealand research institute GNS Science at the time of the 2016 quake, spent her waking hours tracking the quake's every movement, mapping hazards and answering government questions.
"I don't think I've ever felt such an immense burden of responsibility," Wallace said. "I brought my dog to the office because if there was a big earthquake, I didn't want to be separated from my dog."
Wallace and her colleagues found that the odds of a major earthquake were increased by 18 times, and that the risk within a year was 0.6% to 7%. But the major earthquake never came.
"Which of these slow-slip events are actually going to trigger the next big one?" Wallace said. "It's one of the key problems we're trying to understand."
For the Cascadia subduction zone, more data on slow-slip events, better mapping of the fault zone, and better ability to monitor seafloor faults are needed to better understand the warning signals.
Tobin was part of a team that recently mapped the Cascadia subduction zone in the most detail yet. They found that the fault is divided into four sections, all of which can rupture simultaneously or separately in succession. The individual segments can produce an earthquake with a magnitude of 8 or greater.
Meanwhile, researchers are trying to strengthen the offshore monitoring network for Cascadia.
Japan has a sophisticated arsenal of seafloor sensors, but it is "one of the few places that has these instruments," said David Schmidt, a geophysicist at the University of Washington.
The U.S. lags behind in seafloor monitoring, but Schmidt and Tobin are part of a group that received $10.6 million in federal funding to add seismic sensors and seafloor pressure gauges to a fiber-optic cable off the coast of Oregon.
The devices help keep an eye on Cascadia. If the data can help researchers learn what's normal for the fracture, they might be able to tell when it's time to worry.
This article was originally published on NBCNews.com
