Fukushima: Simulations Become Reality

Kenneth Bergeron, author of Tritium on Ice, on last week's events at the Fukushima Dai-ichi Nuclear Power Station in Japan.

 I don’t know how many thousands of nuclear reactor accident simulations I and my colleagues at Sandia National Laboratories ran in the two decades following the Three Mile Island accident in 1979, but I do know that a great many of them involved the Station Blackout scenario—the unlikely combination of loss of both off-site and emergency electric power. Commercial power reactors are such complex and massive machines that engineers have not found practical (or at least cost-effective) ways to make them “walk-away” safe after they’ve been shut down. It still takes a fair amount of electric power to cool the fuel rods, which continue to generate heat at a level of a few percent (at first) of the reactor’s operating power. Over the subsequent hours and days and weeks the residual heating gradually decreases, but to have no AC power at a nuclear power plant for sustained periods is bad news: core meltdown is in the offing. That is why so much attention has been paid to Station Blackout in computer simulations and probabilistic studies of reactor safety. 

So it was a kind of dreamlike experience for me to follow the progression of events last week at the Fukushima Dai-ichi Nuclear Power Station in northern Japan. From the moment I knew that on-site emergency AC power had failed (from the tsunami, we later learned) alarm bells went off in that dusty part of my mind that had been dedicated to the study of severe reactor accidents a decade and more ago. Within hours, Fukushima became the new focus of my life. There were conference calls and long conversations with good reporters confessing their poor grades in high school science and (I apparently can’t say no) answers to drive-time talk radio questions and interviews with bad reporters who projected their own fantasies onto my words. I learned how misquotes can propagate at the speed of the Web and cannot be retrieved. I did eleven television interviews in four days, and this from New Mexico, not New York! 

Amidst all this talking, there were moments of amazement for me, such as watching blurry web videos of the top blowing off unit 1’s reactor building, one day into the accident. After several rewinds I’m sure I see it: a shock wave signaled by a faint, almost spherical condensation front spiriting upward above the building. I think, hydrogen detonation! I studied hypothetical hydrogen combustion in reactor containments for years, but I never thought I would see this. It takes a lot more hydrogen to produce a detonation than the milder explosion called deflagration. Immediately, I knew how serious the overheating and core damage already was.

Another moment I will never forget was an interview with BBC radio’s World Service. Just before going on the air, the host, a most proper-speaking Englishman, informed me that at the London studio they were watching the Japanese dump water from Chinook helicopters onto the spent fuel storage pools.  With one hand holding the phone, I one-finger googled “NHK feed” and at the moment we went on the air I was watching the same thing they were.  We shared comments on the water bombardments—‘that first one looked good but the next one seemed off the mark, don’t you think?’ This was heard by BBC listeners all over the globe—it could be about a cricket match. A Londoner and a New Mexican 5,000 miles apart comparing the relative water bombing accuracy of some very scared Japanese pilots flying through a deadly radioactive cloud some 6,000 miles east from him and 6,000 miles west from me.  Amazing.