More than 20 years before the disaster at the Fukushima No. 1 nuclear power plant, a scientist was issuing warnings that a tsunami of an unanticipated scale could knock out a nuclear reactor.
More than 20 years before the disaster at the Fukushima No. 1 nuclear power plant, a scientist was issuing warnings that a tsunami of an unanticipated scale could knock out a nuclear reactor.
The scientist's article ran in a trade journal for electrical engineers in Japanese. For his efforts, the utilities basically blacklisted the author.
"One thing that is often overlooked is functional disorder due to flooding," said the article, which appeared in a 1988 issue of Electric Power Civil Engineering. "The possibility of flooding cannot be ruled out totally even if calculations show that a given location will not be flooded."
Nobuo Shuto, a professor emeritus of tsunami engineering at Tohoku University, wrote the article upon request.
Tsunami pose a particular threat to nuclear plants, since it is difficult for numerical simulations to set a clear upper limit on their height. Shuto, now 77, pointed out that a tsunami that struck without warning could impact electrical and water intake systems, and called for safety measures to be taken even if there are no known precedents for one.
Shuto thought he was just writing frankly about the limitations of research and the need for preparedness.
"Tsunami cannot be blocked just with physical structures," Shuto said. "I simply said that measures should be taken to protect the vulnerable parts of nuclear reactors, but I later learned from an acquaintance that power utilities thought of me as a persona non grata."
Anti-nuclear movements were on the rise following the 1986 disaster at the Chernobyl nuclear plant in what was then the Soviet Union. But the government and power utilities had built up a "safety myth" of nuclear power generation and were pressing ahead with plans to build more nuclear reactors.
Tsunami were hardly on the radar screens of power utilities and the government when they began building nuclear reactors in Japan in the 1960s.
A regulatory guide on the review of safety design for nuclear power reactors, which the government put in place in 1970, did cite tsunami, along with earthquakes and strong winds, as a potential threat. But applications for permits to build nuclear reactors, which were filed at the time, only made perfunctory notes of tsunami, such as "there are no precedents" or "there is no danger of damage." The applicants sometimes didn't even bother to mention the threat of tsunami.
When Tokyo Electric Power Co. (TEPCO) in 1966 applied for a permit to build the Fukushima No. 1 nuclear power plant, it only cited the tsunami height of 3.122 meters, which was measured locally when a giant earthquake in Chile sent tidal waves across the Pacific in 1960, which left 142 people dead or missing in Japan.
The plate tectonics theory, which forms the basis of contemporary seismology, was yet to be established at the time. Numerical methods to simulate tsunami heights were also underdeveloped, so only the largest documented tsunami in history were taken into consideration.
The Sanriku coast of the Tohoku region was struck by the Meiji Sanriku tsunami of 1896, which claimed the lives of about 22,000 people, and the Showa Sanriku tsunami of 1933, which killed more than 3,000. But there was no record of a major tsunami hitting the coast of Fukushima Prefecture.
A major push for safety measures came following an earthquake in 1993 that struck southwest of Hokkaido. A towering tsunami that hit Okushiri island overwhelmed coastal levees and destroyed villages, leaving 198 people dead or missing on the island alone.
"That was the moment when I came to be aware of tsunami for the first time," said a former TEPCO executive whose expertise is in nuclear power. "I had never given them a thought until that time. But even after the event, few discussions were made within the company."
By that time, numerical methods were available to simulate tsunami and to predict their heights on the basis of seismic source and topography data. In a 1994 report to the government, TEPCO cited a simulated tsunami height of 3.5 meters at the Fukushima No. 1 nuclear plant, far short of its ground elevation of 10 meters.
The government, in the meantime, during discussions of disaster prevention measures to be taken on regional levels, said that consideration should be given to scientifically plausible maximum tsunami heights instead of the largest known tsunami heights. The utilities were called upon to shift their approaches for safeguards accordingly.
Those concerns resulted in the "Tsunami Assessment Method for Nuclear Power Plants in Japan," a guideline released in 2002 by the Japan Society of Civil Engineers. Shuto, as well as civil engineers at power utilities, helped draw up the guideline, which said that tsunami simulations should be repeated under different sets of conditions and that the most impactive values should be adopted.
The projections, thus derived, were about double the maximum known tsunami heights and were therefore thought to contain some leeway for safety.
The tsunami height projections for all nuclear plants were revised. The projection for the Fukushima No. 1 plant was raised to 5.7 meters.
As it turned out, however, the new projections had no margin of safety, because they failed to foresee the giant earthquake that was to spawn an even larger tsunami in 2011.
Meanwhile, the government's Headquarters for Earthquake Research Promotion in 2002 said that a tsunami earthquake was just as likely to strike off Fukushima Prefecture as it was off the Sanriku coast. More facts about the Jogan tsunami of 869, which ravaged the current-day Miyagi and Fukushima prefectures, also gradually came to be known in subsequent years.
But tsunami events, whose scale and source remain undefined, do not easily lend themselves to simulations. TEPCO did project that there could be tsunami in excess of 10 meters, enough to flood the grounds of the Fukushima No. 1 plant, but postponed taking safeguards on the argument that the estimate contained a number of assumptions.
The utility assigned more importance to elucidating tsunami scientifically and studying methods of their numerical modeling.
"We had started discussions on countermeasures," said one TEPCO engineer. "I feel so bad about it."
It was half a year before the Great East Japan Earthquake hit that one TEPCO division began studying safety measures against potential flooding.
The 2002 JSCE guideline only presented tsunami simulation methods and did not mention how to prepare for tsunami of an unanticipated scale. And experts in the fields of equipment and its operation had few opportunities to communicate with civil engineers and tsunami experts.
"We should have discussed more deeply what would happen if (a tsunami beyond anticipation) were to hit," said Fumihiko Imamura, a 50-year-old professor of tsunami engineering at Tohoku University, who helped compile the guideline. "I did share, with tsunami engineers at power utilities, both expertise and the perception of what are the major challenges, but we did not share them with the whole community."
The government's regulatory guide on anti-seismic design of nuclear reactors was revised in 2006. The amended guide expressly said that tsunami safety should be considered. It also mentioned "residual risk," or the potential of an accident occurring beyond all expectations.
The government was obligated to review the safety of all nuclear reactors in Japan, according to the revised regulatory guide.
The year after the guideline was amended, the Niigata Chuetsu-oki earthquake shook TEPCO's Kashiwazaki-Kariwa nuclear power plant with a ground motion exceeding the anticipated maximum.
The government therefore prioritized the review of anti-seismic safety and postponed the consideration of tsunami.
Safety reviews for the Fukushima No. 1 nuclear plant were concluded in 2009, but TEPCO never produced an appraisal of its tsunami preparedness. The government did not press the utility to do so.
It was evident that tsunami beyond the probable maximum could occur, albeit at a low probability, and that a loss of power supply and cooling functions could lead to core meltdowns. But concerned parties lacked a sense of urgency and didn't believe it could really happen. The Fukushima No. 1 nuclear plant thus remained vulnerable to tsunami when time ran out on that fateful March 11.