Japan is accustomed to the constant threat of natural disasters, but when a government advisory over the possibility of a megaquake originating in the Nankai Trough was issued in early August, the nation was put on edge.
As people rushed to top up their emergency kits, officials attempted to clearly state what the advisory really meant — it wasn’t a forecast, and the chance of a big quake was only seen as being slightly higher — but at least initially, confusion reigned.
The episode put a renewed spotlight on a long-simmering debate within the scientific community about whether earthquakes can be predicted. Supporters say it’s about reducing risk, but critics see a system that isn’t based on science and one in which the cons outweigh the pros.
The current system for issuing emergency information was established in 2019, evolving from a temporary framework introduced by the Meteorological Agency in 2017, put in place until new disaster prevention strategies were implemented.
The government categorizes earthquake warnings into three levels of urgency.
The lowest, “extra information,” describes the need for further analysis. The second level — which was issued in August — can be literally translated as “megathrust earthquake attention,” and is issued following a large quake in the Nankai Trough area. It warns of the possibility of another large quake and advises people to prepare for possible quakes, such as by securing furniture, confirming evacuation routes, checking emergency supplies and ensuring means of communication with family.
The highest level is a “megathrust earthquake alert,” meaning a magnitude 8 or greater earthquake has occurred at a plate boundary and another could soon follow. The government would then call on the public to review preparedness and urge the evacuation of people needing special care and those in particularly vulnerable areas who may not be able to evacuate if a quake hits.
The system sat dormant for five years until Aug. 8, when a magnitude 7.1 earthquake struck off the coast of Miyazaki Prefecture. News reports tracked the story, which unfolded in a dramatic arc — a big quake, a tsunami advisory, an all clear — and then a notice that seemed to suggest the Big One was coming.
Echoes of doubt
Manabu Hashimoto, a former director at Kyoto University’s Disaster Prevention Research Institute, is among those who have questioned the effectiveness of the advisory system.
He argues that issuing such advisories, especially since they involve such low probabilities of around 0.5% to 0.6% for a Nankai Trough earthquake over a one-week period, could lead to unnecessary disruptions.
Issuing advisories like the one in August might lead to unnecessary social and economic disruptions, Hashimoto said, adding that if people start to receive such warnings regularly, it could increase the risk of complacency.
“We should stay prepared as though a major event could occur at any time,” he said. “In that sense, this advisory represents an ongoing state of vigilance that should continue indefinitely.”
Hashimoto said the government is not effectively providing the information needed for informed decision-making by the public, but he noted that experts also struggle to communicate the uncertainty and low probability involved in the system.
Takeshi Sagiya, a professor at Nagoya University’s Disaster Mitigation Research Center, is another critic of the system, pointing to the initial public confusion surrounding the advisory.
“I’ve been quite skeptical about the effectiveness of this warning system from the beginning,” Sagiya said.
Experts including Sagiya and Hashimoto have also expressed concerns about focusing too much on specific regions, as the government has with the Nankai Trough.
Before the 1995 Great Hanshin Earthquake, the government’s focus for earthquake preparedness was solely on the Tokai region, leading people in the Kansai region to believe that a quake in Kobe was unlikely, Hashimoto said.
“The government is repeating the same mistakes,” he said, emphasizing the need for a more balanced approach that includes preparation for unexpected earthquakes in less-discussed regions.
Sagiya, meanwhile, noted what he sees as problems with the science, expressing concern that disaster preparedness and official measures for quakes rely too heavily on speculative theories lacking sufficient evidence. The current probability estimates for the Nankai Trough earthquake are also heavily based on such speculative models, he added.
In 2013, the probability of a magnitude 8 or greater earthquake occurring throughout the entire Nankai Trough within the next 30 years was estimated to be between 60% and 70%. As they were based on historical earthquake intervals, they are about 10 percentage points lower than today’s 70% to 80%, Sagiya said.
But if the speculative models are put aside, he said, the figure could fall to 30% or lower.
He added that Japan’s approach tends to highlight the highest probability as a means to emphasize the need for preparedness, but that using a range or average would be more appropriate.
Sagiya cautioned that focusing on the estimate of a 70% to 80% probability that an M8 to M9 earthquake will occur in the Nankai Trough within the next 30 years, a disaster that would have an estimated death toll of hundreds of thousands of people, risks “unnecessarily heightening anxiety” and could lead people to overlook other still-vulnerable regions, creating a false sense of security.
Sagiya cited January’s quake off the Noto Peninsula as an exceptionally rare event, thought to occur roughly once every 3,000 years in that region. Despite this low probability, he emphasized that earthquakes can strike anywhere.
Hashimoto echoed Sagiya’s concerns.
“Defining a worst-case scenario scientifically is nearly impossible,” he said, noting that common damage prediction models are more aligned with engineering, even if they involve some scientific elements. “In essence, much of the warning system's work is not purely scientific.”
A social experiment?
Still, some experts see value in the alert system.
Shinichi Sakai, of the University of Tokyo's Earthquake Research Institute, said it's crucial to understand the objective behind the advisory — its primary goal, he said, is to reduce the potential damage from a disaster.
“While we cannot prevent earthquakes, the aim is to enhance preparedness so that when an earthquake occurs, the likelihood of minimizing damage and increasing chances of survival is improved,” he said.
He acknowledged that evaluating the effectiveness of such advisories is challenging since trials can’t be done repeatedly, and their impact cannot be immediately assessed.
Despite this, he added, researchers can learn from how people react, which could give insight into how to refine future communication strategies and assess the overall system.
“Understanding how people interpret and act upon such notices is crucial. We need to consider how to communicate such warnings effectively,” he said.
But Sagiya takes issue with what he sees as a “social experiment.”
"It feels like the entire country is participating in an experiment,” he said. “As if the Meteorological Agency is observing how society reacts when such information is issued,” and turning citizens into “experimental subjects.”
While he admitted this is a harsh characterization, he questioned whether the system is acceptable.
Sakai said that, while the advisory did serve as a kind of experiment, it highlights the need for continuous improvement.
“To better prepare for actual events, it’s crucial to conduct repeated drills and continuously refine the process,” he said. “This iterative approach helps develop more effective strategies for real-life situations.”
He compared this idea to crime prevention, noting that, just as with natural disasters, the best approach is to avoid risky situations and be on guard for potential threats.
“We can’t feel completely at ease — I don’t think we’ll ever reach a point where we can be truly confident,” Sakai said. “But all we can do is make it better.”
Awareness, not prediction
Naoshi Hirata, chairperson of the Meteorological Agency's Nankai Trough Earthquake Assessment Committee, which monitors and provides expert advice on collected data from the Nankai Trough region, emphasized that the August advisory was not a prediction but an effort to raise public awareness about earthquake preparedness.
“The advisory was there to nudge people to review how prepared they are for an earthquake,” he said, adding that the government did not recommend canceling travel plans or avoiding coastal areas. “We did not tell people to evacuate.”
The Center for Integrated Disaster Information Research at the University of Tokyo conducted an online survey from Aug. 9 to 11 to measure behavioral changes following the Nankai Trough advisory. The survey received a total of 9,400 responses from individuals between the ages of 20 and 69.
The changes in behavior following the advisory were limited, according to the survey, with only a small percentage of respondents saying they took specific actions — 19.7% said they checked supplies such as water and food, 9.2% confirmed communication methods with family, and 8.1% ensured measures to prevent furniture from tipping over. Only 2.1% indicated that they had changed their travel or leisure plans.
Hirata said the advisory served to encourage people to check evacuation routes in case a tsunami warning or a top-level alert is issued.
“Many people learned about this system for the first time and were not sure what to do,” he said.
“But data showed that most people acted calmly, and there was very little panic.”
One important achievement through the advisory, Hirata said, citing the study, is that over 80% of people now at least understand that such advisories could be issued.
The University of Tokyo study showed that 83％ of people heard about the advisory.
Hirata acknowledged that, despite the varying predictions, the government focuses on the worst-case scenario to ensure comprehensive measures are taken.
“Quakes are unpredictable, but hazards can be forecasted,” he said.
Hirata compares the situation to flashing your car’s hazard lights on the highway as a way to signal congestion ahead. However, simply having these lights flashing doesn't always prevent accidents — it shows that without proper precautions, accidents could happen. In essence, he said, evaluating hazards, such as the likelihood of an earthquake, is about communicating potential risks so that appropriate measures can be taken.
That’s where the advisory system comes in, and Hirata notes that earthquakes are indeed likely to occur in succession.
Nankai Trough quakes often come in pairs, Hirata said, adding that once a magnitude 8 earthquake occurs, there is roughly a 10% chance that another earthquake of magnitude 8 or greater will follow.
Many people mistakenly believe it’s safe after a big quake, or that one should be expected after a long period of quiet, but Hirata calls this a myth.
He sees the bigger picture as being something far more important than just a matter of probability, and indicated that fear can be a healthy partner for preparedness.
“In the first place, earthquakes are natural phenomena that occur underground, and they only become disasters when society fails to respond appropriately,” Hirata said.
“We need to be fearful in the right way.”
With your current subscription plan you can comment on stories. However, before writing your first comment, please create a display name in the Profile section of your subscriber account page.