How a Great Power Falls Apart
Decline Is Invisible From the Inside
Houses swept away in the wake of the tsunami. The following essay is adapted from the author's book, Obama and China's Rise: An Insider's Account of America's Asia Strategy. (Kyodo / Courtesy Reuters)
On March 11, 2011, I was awakened at 1:45 am by a call from the White House Situation Room reporting that an earthquake measuring 8.9 on the Richter scale (subsequently raised to 9.0) had struck northeast Japan, and that tsunami warnings had been issued. At the time I was senior director for East Asian affairs on the National Security Council, and immediately we formed an interagency group to address the situation. First, we mobilized relief efforts, and the U.S. Pacific Command provided indispensable support for remote and isolated towns. Those operations generated few difficult policy decisions.
Within a few days, however, a threat emerged: the possibility of multiple meltdowns at the nuclear reactor site in Fukushima, 160 miles northeast of Tokyo. The Fukushima Daiichi nuclear power plant had six reactors, three of which were online when the earthquake and tsunami struck. The other three were not, but their highly toxic fuel rods were stored in spent-fuel pools on site.
The reactors survived the earthquake, but the tidal wave rose to 46 feet and overwhelmed the electrical power running the reactor, the backup generators, and the grid servicing the whole region. Without power to the reactors and the spent-fuel pools, it was impossible to cool either the active or the spent fuel rods. In the absence of alternative cooling capacity, the water in the reactors and pools would inevitably boil off, leaving the rods exposed and triggering a partial or complete meltdown.
At the outset we struggled to obtain information about what was going on at Fukushima. Some believed that the Japanese were not being entirely forthcoming with what they knew, trying to put an excessively optimistic gloss on the situation. The president's science adviser, John Holdren, was less critical. He pointed out that none of the instruments used to monitor developments inside the reactors were functional, so information was inevitably sketchy. Radiation leaks and hydrogen explosions made on-the-ground assessments difficult to come by. Besides, a small number of Japanese officials were facing multiple horrors of unimaginable scale, and to expect them to provide us with real-time accurate information seemed to me unrealistic. Although subsequent investigations have pointed to significant errors by the Japanese in estimating the threat and in putting out information in a timely way, the core problem was not a desire to cover up but, rather, the sheer difficulty of knowing what was going on at the reactor site.
The Nuclear Regulatory Commission (NRC) and the Department of Energy (DOE) dispatched experts to Tokyo. One day, the challenge was to cool reactor 2. The next day, it was reactor 3. The next day, it was a spent-fuel pool. Seawater was pumped into reactors to cool them. Then it became clear that salt from the seawater was caking the fuel rods, making cooling even more difficult. It appeared that the radioactive steam had to be vented to prevent further hydrogen buildup and explosions, but that in turn heightened the fear of airborne radiation. Radioactive water in the bottom of the reactors was run off into the ocean to make the reactors accessible, but then radiation in the seawater rose to unacceptable levels. Each temporary solution seemed to generate a new problem, and each day we seemed to be facing a new peril.
Our interagency team rapidly grew into the largest one in memory, as the radiation emanating from Fukushima sparked concerns over the safety of international aviation routes, the global food supply, the safety of the ocean, the edibility of fish, and radioactive clouds sweeping across the Pacific Ocean. In short order, agencies across the federal government were involved -- the National Oceanic and Atmospheric Administration, the Federal Drug Administration, the Environmental Protection Agency, the Department of Health and Human Services, the Department of Transportation, the Department of the Interior, the Justice Department, Customs and Border Protection, the Surgeon General, and Veterans Affairs.
It soon became clear that the potential impact on the U.S. homeland was minimal. The team established procedures for vetting food imports and cargo and passengers traveling from Japan to the United States. The EPA expanded its radiation monitoring; it detected slightly elevated levels in several locations, but they were harmless.
The Fukushima disaster was unique, as the very thought of radiation can cause the public to panic in a way that other threats do not. Many Americans do not realize the degree to which they already are, in fact, exposed to radiation in their daily lives. It may be easy to evaluate and explain in a classroom the acceptability of slightly added risk associated with heightened exposure, but it becomes a far more difficult exercise when the audience consists of families with young children, which was the situation we faced in Japan. And, as little could be done to halt the multiple reactor breakdowns in the site's uniquely hostile environment, it was clear that the situation at Fukushima was anything but benign.
We had to decide whether to declare a larger evacuation zone around Fukushima than Japan did. Modeling conducted by the NRC and the DOE indicated that an evacuation zone of 50 miles would be more consistent with U.S. standards than the Japanese zone of 12 miles, so the administration recommended that all U.S. citizens in the 50-mile zone leave. The discrepancy attracted unwelcome attention and subjected the Japanese government to some criticism. Of course it was considerably easier for us to err on the side of caution, since we had almost no Americans in the area and no responsibility to house or take care of them once they departed, whereas the Japanese had several million people there, all of them the government's responsibility if they moved.
Far more challenging, however, was the question of how to deal with the large American population in Tokyo and at the nearby U.S. bases in Yokota and Yokosuka. About 90,000 Americans lived in Tokyo alone, and there was a vital air force and navy presence at the two bases. Recommendations affecting those locations could send metaphorical shock waves throughout Japan.
The U.S. ambassador to Japan, John Roos, was under tremendous pressure from American government families in Tokyo to authorize their departure. The commander of U.S. Forces Japan was under arguably even greater pressure from dependents at the U.S. bases. Without a sound scientific reason, the administration did not want to take such a step. The White House science adviser offered no support for the claim of a radiation threat to Tokyo or the bases. Some foreign embassies in Tokyo had begun drawing down or moving facilities further south to Kyoto. We wanted to be sensitive to the needs and wishes of families but did not wish to trigger a panic.
A number of heated Deputies Committee meetings ensued at which both Ambassador Roos and military commanders strongly urged the government to authorize the departure of dependents. On March 16, Washington decided to authorize the voluntary departure of dependents, in essence telling U.S. official dependents that they were free to depart Japan at U.S. government expense but could not return until the voluntary departure order was terminated. All official personnel were ordered to remain in place. That decision helped to relieve the pressure building in the U.S. official community for more draconian action. The Japanese government reacted with relative equanimity, since the approval of voluntary departure did not imply concern about Tokyo's safety.
But the relief was only temporary. The administration still lacked a solid basis for projecting future risk. Holdren and the DOE were working closely with Lawrence Livermore National Laboratory to try to develop a model for plausible worst-case scenarios, but it was taking time. In the absence of an authoritative model, other hypotheses filled the void and encouraged ill-considered decisions.
The Pentagon's Naval Nuclear Propulsion unit argued that radiation levels could under foreseeable circumstances soon exceed EPA standards at Yokosuka, which would have meant a considerably higher level in Tokyo. Obviously, if that turned out to be true, we would need to issue directives to protect the health of American citizens, regardless of the Japanese assessment of the situation. Such directives, depending on how drastic they were, could either cause panic in Tokyo or a breach in the U.S.-Japan relationship. With the team, I was determined that we not be stampeded into sudden action in the absence of a sound scientific basis, but we understood the urgent necessity of developing a sound analysis as a basis for a decision, whichever way it went.
Late one night in late March I awakened Deputy National Security Adviser Denis McDonough after midnight and described the dilemma. He called Holdren to ask for the best model he could provide based on extreme scenarios at Fukushima so that we could see if our current precautions were adequate. Holdren did so immediately. The levels were utterly at odds with what the Naval Nuclear Propulsion unit had projected. Under plausible extreme scenarios, Holdren's model suggested that radioactivity at Yokosuka would be about five percent of what the Naval Nuclear Propulsion unit had projected. And only one or two percent of the radioactive particulate would be iodine, not most of it, as the Naval Nuclear Propulsion unit feared. Holdren's rapid and careful work had averted a potential slide toward unnecessary and damaging decisions.
But the pressure did not disappear. Because of the unpredictability of the situation at Fukushima, we needed to draw up contingency plans for the evacuation of all Americans from Tokyo and the bases in the event that the situation warranted it. That was normal and proper, although extremely unlikely. But once Pacific Command began planning for a noncombatant evacuation that, in theory, could involve 90,000 people under panicked conditions, the information would inevitably leak.
It leaked quickly. Stories ran in U.S. military media and the Japanese press that suggested that evacuation was a real possibility. I called the chief of naval operations, Admiral Gary Roughead, with whom I had had very good interactions in the past. I told him of my dismay at the way the story was percolating. I said that I was as strongly in favor of protecting American servicemen's health as anyone, but that we needed a scientific basis for decisions. We also could not be casual about the future of the alliance by allowing for a whimsical decision-making process. Roughead understood. Within an hour, he had called in the defense press and made unequivocal statements to the effect that our forces were not going anywhere, and that evacuation was not in the cards.
These daily crises in response to wildly speculative assessments and reports were testing our patience, not to mention our sleep cycles. We needed a firm scientific basis for decisions. Fortunately, Holdren and the DOE were about to produce one.
Working with Lawrence Livermore National Laboratory, Holdren developed a series of models based on plausible worst-case scenarios. They depicted simultaneous meltdowns at one or more reactors and complete drainage of the spent fuel pools at two reactors. The results for such worst-case scenarios, assuming unfavorable wind patterns from the reactor site and a lack of precipitation, suggested that radioactive plumes in excess of EPA standards would not reach within 75 to 100 miles of Tokyo, and that we would have several days' notice before such a contingency could develop. In other words, there was no plausible scenario in which Tokyo, Yokosuka, or Yokota could be subject to dangerous levels of airborne radiation.
Armed with that information, the national security adviser, Tom Donilon, convened a principals committee meeting. Some at the meeting were still uneasy. They wondered to what degree the assessments were based on information, perhaps defective, supplied by the Japanese. Holdren replied that that was not the case; they were based on models for these types of reactors provided by the NRC and by General Electric, the manufacturer. That effectively settled the argument. From then on, there was no serious discussion of evacuation from Tokyo or the bases. Instead, we increased the flow of information to Americans on the ground about the situation and the risks in terms of airborne radiation and food and water contamination.
Although the Fukushima situation was far from stable, at least a scientific model was now in place to assess the risks to Americans. Washington shared the assessment with the Japanese government, as well as readings of ground radiation developed by U.S. experts brought into the theater. The risk of decisions with no basis in fact that could have undermined the alliance had been eliminated.
In crisis situations, drastic options with grave consequences inevitably have to be considered, as they were in this instance. However, the proper test of the process is whether such options are improperly chosen, not whether they come up. In my view, the thorough and serious interagency processes fully drew out the facts, the scientific assessments, the considerable expertise of numerous agencies, and competing views. As a result, American citizens were fully informed and protected, while rash decisions were avoided. In this extremely complex set of challenges, with many unknowables and unknowns, the results vindicated the process.