I have long been intrigued with the effort to predict a large, potentially devastating earthquake in the Tokai region of Japan and the legislation that established how scientists and public officials were to respond, should precursory seismic activity reach the point of a warning issued by the prime minister. The Large-Scale Earthquake Countermeasures Act of 1978 has been much maligned over the years, as this region between Shizuoka and Aichi prefectures has been the focus of intensive observation and monitoring by scientists, response planning by prefectural and municipal governments, and yet remained stubbornly quiescent for the last 40 years.

The criticism of the Tokai prediction and the legislation that facilitated its being issued has mainly come from scientists who have argued quite forcefully that reliable earthquake predictions are not possible and that a short-term prediction for Tokai is likely to be a failure.

Worse, should a warning be issued, the rather dramatic interventions envisioned by the legislation allowing the prime minister to mobilize the Self-Defense Forces, carry out large-scale evacuations, restrict travel and commerce, and take other actions are, scientists argue, unwarranted given the uncertainties inherent in earthquake prediction. While Tokai has peacefully slumbered, major earthquakes causing massive damage and staggering death tolls have occurred in other regions of the country, including the Hanshin-Awaji (Kobe) Earthquake of 1995 and the Great East Japan Earthquake of 2011.

In all fairness, the late 1970s, when the Countermeasures Act was passed, were a time when there existed some optimism, naive perhaps, that scientific earthquake predictions would soon be routinely made. In 1975, Chinese seismologists were reported to have successfully issued a short-term prediction 36 hours prior to a magnitude 7.3 earthquake in Haicheng Province, saving the lives of thousands of people. Studies were launched testing the reliability of predictions based on methods used by the Chinese as well as others that emerged during this period. Unfortunately, these predictions ended in failure and optimism regarding prediction waned.

Status of quake forecasting

But where are we now in terms of scientific earthquake forecasting? Most of us are familiar with long-term earthquake forecasting. For example, Japanese scientists believe that there is a 70-80 percent chance that a magnitude 8 or 9 earthquake may occur in the Tokai, Tonankai and Nankai region in the next 30 years. Short-term forecasts, for a few days, are also possible according to a 2011 scientific report titled “Operational Earthquake Forecasting: State of Knowledge and Guidelines for Utilization.” According to this report, the distribution of earthquakes is not uniform. Rather, earthquakes cluster in time and location.

Earthquake probabilities are not constant in time; they fluctuate over days, months and years in association with seismicity within a region. Sometimes, following an earthquake swarm, a series of moderate size earthquakes, or other seismic activity, a larger event occurs as it did in Kyushu in 2016 and in Tohoku in 2011. Such foreshock-mainshock sequences are rare, but depending on the region, the probability of a larger mainshock following increased seismic activity can range up to 10 percent over a few days.

Just after the turn of the new century, Japanese scientists became concerned that a Tokai earthquake could be accompanied by simultaneous rupture of Tonankai and Nankai, adjacent portions of the Japan Trench, the subduction zone that lies off the Pacific Coast of Japan. The resulting mega-quake could equal magnitude 9 and generate a huge tsunami. These concerns were later corroborated by the occurrence of the Great East Japan Earthquake and tsunami, which occurred in the northern portions of the Japan Trench, claiming the lives of over 18,000 people. A magnitude 9 farther south would, according to scenarios developed by the Cabinet Office, impact an urban region of 47 million people, cause up to 323,000 fatalities, displace 3 million people and cause a tsunami that would inundate 1.8 times the land area of the 2011 tsunami.

Recent legislation and public policy changes have extended the mandate for intensive seismic monitoring, response planning and hazard mitigation to these additional areas beyond Tokai. But, in response to criticism, the extensive interventions in the 1978 Countermeasures Act have been abandoned. If seismic activity considered precursory occurs, the Meteorological Agency will issue an advisory that a large and potentially destructive earthquake may occur in a matter of days. But in the absence of specific actions called for in the 1978 Countermeasures Act, what will local government and the public be advised to do and how shall they respond to this advisory? At present, these issues are largely undefined.

A viable action plan

Sociological research has shown that an information vacuum in situations of uncertainty and potential danger is a serious problem. The people of Japan and particularly this region know the danger of earthquakes and tsunami, and will want authoritative information on how best to respond to an earthquake advisory. Saying nothing is not a viable option, and having no plan as to what to say and do is poor public policy.

In response to scientists who will emphasize the uncertainty and possibly low probability that any precursory event will be followed by a massive earthquake, we say, there are options short of large-scale evacuations and moving the Self-Defense Forces into the area. Clearly, we will be operating in a low probability environment, that is, the probability that precursory earthquakes or a measured crustal movement in Tokai, Tonankai and Nankai would be followed by a major earthquake may be 10 percent or less over a period of several days, but the consequences of doing nothing could be extremely consequential in lives lost and damage sustained.

In anticipation of issuing an earthquake advisory, the national government should have a well-articulated plan for both ongoing communications with the public, prefectural and municipal governments, the media and private-sector organizations, and have a set of sensible recommended actions for the region. This communication should include everything scientists know, and don’t know about the hazard and well considered recommendations as to how to respond to an advisory. For example, advise people to review life safety actions in an earthquake, an exercise that should include all locational contexts in which people travel daily during the advisory. Remind them that Japan has an earthquake early warning system and that they should be prepared to respond to an alert.

The period of the advisory should also be an opportunity to reduce household and workplace hazards from non-structural building components or unsecured items that could move during an earthquake and cause injury. Those who live in potential tsunami inundation zones must know that following any major earthquake, they must move quickly and without hesitation to higher ground or vertical evacuation structures, and that movement to these locations should be reviewed in advance of the need to evacuate.

Local government organizations can be advised to put emergency response personnel on alert, to activate emergency operations centers and cancel leaves for critical responders. Rather than moving the SDF into the region under an advisory, place them on alert and instruct them to review their emergency response roles. Deploying them to the region in advance of the earthquake only places more people in harm’s way.

The prime minister, in consultation with the Meteorological Agency, should have the power to declare an emergency if an advisory is issued and be provided discretionary power to take measured and reasonable actions consistent with the likelihood of a major earthquake.

Some observers have claimed that low probability advisories will result in panicked behavior and social disruption or that advisories, if issued multiple times with no resulting earthquakes, constitute “crying wolf.” But social science research has demonstrated that fears of panic are vastly overstated and if properly advised, people will respond appropriately. The “cry wolf” notion must be placed in context; research indicates that repeated warnings that are not followed by the anticipated event lead to public non-response only when costly mitigation measures are accompanied by a high rate of false alarms. There is every reason to believe that the Japanese people will respond reasonably and adaptively.

In response to a lecture I gave regarding low probability earthquake advisories, a colleague remarked: “If there were only a 10 percent chance of rain, I wouldn’t carry an umbrella.” My response was that if one decides that rain is unlikely, fails to carry an umbrella and it rains, he or she is likely to get wet. The consequences, though uncomfortable, are not serious and one will have learned a lesson. The failure to adequately plan for a potentially devastating earthquake could be far more consequential, including death, and the dead do not learn lessons that can be applied later.

James D. Goltz is a visiting research professor at Kyoto University’s Disaster Prevention Research Institute.

In a time of both misinformation and too much information, quality journalism is more crucial than ever.
By subscribing, you can help us get the story right.