Chapter 3: Can humans stop a crisis? ‘Hubris of wisdom’

japanese 日本語版

Photo: The damaged main clerical building at the Fukushima No. 1 nuclear power plant after the Great East Japan Earthquake (Provided by Tokyo Electric Power Co.)

It might have been a preconceived notion that we held.
It might have been a preconceived notion that we held.

At 2:46 p.m. on March 11, 2011, a massive earthquake struck the Pacific Ocean floor far to the east of Japan. At that time, Masao Yoshida, the plant manager of the Fukushima No. 1 nuclear power plant, was going over documents in his manager's office on the second floor of the main clerical building located almost in the center of the plant site. The shaking gradually gained in intensity and objects fell from the shelves; a TV set toppled over. Yoshida grabbed on to his desk and tried to dive under it, but couldn't. He left the manager's office and entered a general work office where the ceiling panel had fallen off. Documents from the shelves were scattered over the floor and white dust floated in the air. That was the beginning of a catastrophic disaster in which power sources necessary for cooling the nuclear reactors were all lost. The accident was of a severity never imagined in the manual of accident management measures compiled to deal with such an event.

Question: Regarding the loss of power sources, it appears that the situation was one in which no consideration was given to the possibility that a neighboring facility from where electricity could be shared might also be flattened at the same time. As the plant manager, what was your thinking with regard to preparations for accident management? Did you ever consider the possibility of a situation in which several plants would all malfunction simultaneously, or did you never think that could happen? Or, was it possible that you never even thought about such a possibility?

Yoshida: In a word, while I knew there were discussions during the design stage, within the design talk, while it may be improper to now say the accepted notion, there was as I just said a fundamental way of thinking of placing priority on internal factors. While I was never deeply involved in that matter from the time I joined the company, in other words, I did understand that such thinking was what was involved in thinking about nuclear plant design. For the latest incident, upon having moved to the operating side, if I were asked if I had thought about such an event occurring simultaneously from the operating side, I would have to say, unfortunately, that I never thought about that until March 11.

Q: From the standpoint of using this as a lesson for the future, what do you think led to such a way of thinking?

Yoshida: For one thing, in the sense of simultaneously malfunctioning, there was a simultaneous breakdown at the Kashiwazaki (-Kariwa nuclear power plant) when the (Niigata) Chuetsu-oki Earthquake struck (in 2007). While there was a simultaneous breakdown, for us, the plant stopped operations and there was serious damage, but in a word, all the reactors stopped operating in a safe manner. Those in charge of safety would say that regardless of what the next step is, if the operations stopped safely that is all that matters. From that standpoint, one can say that all the reactors stopped properly despite the huge earthquake that hit. Moreover, inspections after the fact found that the safety equipment was largely undamaged even though the quake greatly exceeded the one estimated in designing the plant. In other words, even though the quake struck at once and was powerful enough to stop all the plants, it only went so far as stopping operations. It did not go beyond that since the earthquake did not cause, such as in the latest accident, the loss of all cooling sources. The earthquake that was several times larger than the seismic motion used in the design process in a sense demonstrated that Japanese design was, after all, correct. So, conversely, that line of thinking emerged as a result.

Photo: Masao Yoshida, left, who was in charge of TEPCO's nuclear asset management department at the time, explains on April 6, 2007, about more cover-ups by the utility of problems at a nuclear power plant. (Tomomi Miyazaki)

Immediately after the earthquake, Yoshida issued an instruction to group managers to check on the safety of group members. At about 3 p.m. on March 11, 2011, Yoshida gathered in the emergency disaster response room on the second floor of the quake-proof control center building along with the heads of other units as well as those in charge of the sections responsible for power generation and maintenance.

The first report that was submitted to that group was that reactor scram had operated safely for the No. 1 reactor, No. 2 reactor and No. 3 reactor, which were all in operation at the time of the quake. A scram is the emergency shutdown of the reactors through the insertion of control rods into the nuclear fuel to restrain nuclear fission reactions. The control rods are designed for automatic insertion in the event of a major quake. The report was that the function had operated successfully. Nuclear fission reactions had stopped, meaning the reactors had moved out of a state of criticality in which nuclear fission reactions occur continuously.

The reactors had completed the first procedural step in a normal manner.


On the other hand, it appeared that external power sources to the nuclear plant had been cut off. A report also came in that emergency diesel generators had begun operating. It was later learned that the loss of power sources was due to the toppling of power transmission towers by the quake.

However, at that moment the only thing plant officials knew was that there was some form of power source since the emergency generators were working. Later, towering tsunami hit the Fukushima No. 1 nuclear plant. Because the quake-proof control center building was located on a hill away from the coast, Yoshida was unable to immediately grasp what had happened. However, ominous reports continued to pour in.

“The emergency diesel generator at the No. 1 reactor has stopped and all power sources have been lost.”

“The same thing has happened at the No. 3 reactor.”

“This time it is the No. 2 reactor.”

Piecing together those bits of information led to the likely realization that the entire nuclear plant had been inundated by the tsunami.

It finally dawned on Yoshida that he was confronting an unimaginable situation because events showed it was possible for all power sources at the plant to be lost.

The officials at TEPCO who compiled the accident management measures made so many detailed assumptions that it led to the false sense that the steps covered all possible events and conditions. However, almost all of the measures that were put together could only work effectively with the availability of a power source. In a nutshell: the major precondition of the accident management measures was that all power sources would never be lost.

Yoshida said, "To be honest, I was just devastated."

That was because he was expected to come up with steps to deal with the accident that were not contained in any manual for accident management.

Photo: The containment vessel of the No. 1 reactor at the Fukushima No. 1 nuclear plant as photographed in autumn 1970

Question: While it seemed there was the precondition that in the end you would receive some source of power, was the forecast ever made that such sources of power could also become unusable?

Yoshida: I believe it was an issue of probabilities. Speaking in extreme terms, this becomes a discussion based within the limits of our past experience. In other words, internationally there are about 400 or 500 nuclear reactors, right? Excluding experimental reactors, commercial reactors have been in operation since about the mid-1960s or so. If we talk in terms of reactor-years, if the 400 reactors have been in operation on average for 20 years, the global experience will be about 8,000 reactor-years of actual operations. While various problems have been experienced in that time, what we have been talking about regarding the latest accident in which all power sources and alternative sources were lost, has never occurred. While that may have been a preconceived notion that we held, it also conversely served to give us confidence. (omission)

Q: Looking back on what you went through, do you, first of all, now believe that alternative water pumping using fire trucks as a water source should have been clearly written in the accident management manual?

Yoshida: I believe so from the standpoint of now. However, as for even myself, what I thought might be a gamble was using the fire protection lines because we did not know until the very end if it would work or not. We would have had to reduce core pressure to a level that would allow the fire protection system to inject water. That would mean, in turn, that the water level also had to be reduced. We then ended up pumping in water using the fire protection system. However, if the fire protection lines had been cut off somewhere because of the earthquake, we would not have been able to pump in any water at all.

Q: Did you hold the expectation under the assumption that the fire protection system within the buildings were probably all right?

Yoshida: Of course, I did have such expectations. After the earthquake struck the Kashiwazaki plant, I inspected everything there, such as the fire protection lines at the turbine building and the reactor building. While there was some deformation, there was no case of the lines being cut through. So, in a sense, realistically speaking, I did think that the lines would likely hold up, but I could not be absolutely sure. I had no way of knowing what happened inside the buildings after that earthquake struck. In the end, it was a gamble of whether the fire protection lines were sound or not. But, since that was the only option, my feeling was that I wanted to use that to pump in water.

Photo: A police officer gives instructions to evacuating residents in Tomioka, Fukushima Prefecture, on March 12, 2011. (Toru Nakata)

Yoshida was placing his bets on the fire protection lines that are installed throughout the reactors. The accident management measures included the use of diesel-driven fire pumps to inject water into the reactors using the fire protection lines.

However, Yoshida was also concerned that the diesel-driven fire pumps would not be able to pump in water if pressure within the core was high because the pressure of the fire pumps was weak. For that reason, he also thought up a measure to inject water by connecting fire trucks that can eject water at higher pressure to the fire protection system. He also made the decision to cool the reactors using seawater rather than restrict the water used to fresh water.

There was the possibility the fire protection lines could be damaged by the quake because their anti-quake resistance was lower than other important equipment found around the reactors.

However, Yoshida felt the fire protection system could be used. In part, that was because he inspected the damage to the Kashiwazaki-Kariwa nuclear plant in Niigata Prefecture after the Chuetsu-oki earthquake struck in July 2007. At that time, Yoshida was the head of TEPCO's nuclear asset management department.

After the Chuetsu-oki earthquake, the number of fire prevention tanks was increased at the nation's nuclear plants because of the criticism that was raised over the delay in putting out the fire at the electrical transformer at the Kashiwazaki-Kariwa plant. However, Yoshida felt those tanks alone would not be enough, so he decided at an early stage to use seawater.

Audio: Plant manager Yoshida urges a switch in pumping water into the No. 3 reactor from the diesel-driven fire pump to fire trucks.

Photo: Workers use a crane to remove nuclear fuel rods from the storage pool at the No. 4 reactor of the Fukushima No. 1 nuclear power plant at 1:27 p.m. on March 18, 2011. (Eiji Hori)

Question: During the summer questioning, you said it was difficult to gain the cooperation of companies for using alternative methods of water pumping even though the normal task of those companies was in extinguishing fires at the plant because there had never been talk of having those companies become involved in finding alternative water sources. You also said something about writing down the possibility beforehand that using fire trucks as an alternative water source could lead to delays in the start of such operations. Could you elaborate?

Yoshida: From the standpoint of now, I believe that is what I would have said, but moving back to the previous stage, it becomes a question of whether those individuals in charge of thinking up various accident management measures were actually thinking about such alternatives. I want to say it is my belief that they gave no thought to such matters. Those involved in accident management will come up with various complaints after the fact. I have absolutely no trust in those at TEPCO involved in safety. (omission)

Q: So you are saying that while they understood in their mind that there was the possibility of using seawater, they were never seriously considering using seawater in the end as the last step.

Yoshida: They did not have such thinking. If they had thought about that possibility, they should have designed another line that could be used to pump up seawater. Because we first used as a water source seawater brought in by the tsunami and which had accumulated in the valve pit of the No. 3 reactor and we also implemented other measures that we thought up at the plant site. No one involved in designing accident management measures beforehand had thought about such possibilities.
From my perspective, they were only considering the form such measures should take. As for my part, while I did not make the ultimate decisions, I did operate the plant based on those decided measures and was serving as plant manager so I do feel great embarrassment that I did not think about such matters. However, if I were asked if those who thought up the various measures at the beginning based on various assumptions had made that consideration with the resolve to do whatever it takes, I would have to say I feel there was no one who thought in that way.

Photo: TEPCO officials bow their heads while kneeling to residents who were forced to evacuate because of the accident at the Fukushima No. 1 nuclear power plant in a central district community hall in Kashiwazaki, Niigata Prefecture on March 31, 2011. (Koji Shimizu)

Two ideas thought up by Yoshida was the use of fire trucks and seawater to cool the reactors. However, reading the flow of the questioning of Yoshida leads to the suspicion that the government panel did not view those two methods in a favorable light.

It appears the panel thought the best method would have been to follow the accident management measures decided on beforehand and if that was not possible, it meant there were deficiencies in those measures.

However, Yoshida did not appear to give that much thought, even though he was the individual who had to actually confront the unfolding disaster.

Yoshida was asked if he opened up the accident management manual and used it as a reference. He said he never referred to it or even opened it up.

He explained how ineffective measures thought up by people beforehand can be.

Yoshida also explained that nuclear plants in Japan were designed with priority placed on internal factors leading to malfunctions. He went on to explain that no thought was given to malfunctions occurring simultaneously at a number of plants due to external factors, such as tsunami, tornado, a plane crash or an act of terrorism.

There was an opportunity to revise such thinking after a number of reactors experienced malfunctions simultaneously at the Kashiwazaki-Kariwa nuclear plant in Niigata Prefecture that was hit by the Chuetsu-oki earthquake in 2007. However, because all the reactors were successfully cooled, Yoshida argued it only provided evidence that previous measures had been correct.

He also said that people can only make assumptions within the limits of their past experience. He explained that it is only possible to say after the fact that everything would have gone smoothly if proper assumptions had been made and measures based on that were all written down.

While those points mentioned by Yoshida can contribute to future discussions, it is difficult to pick up on those points from the report compiled by the Investigation Committee on the Accident at the Fukushima Nuclear Power Stations.

To be followed by Epilogue