BRATTLEBORO — Aside from the actual meltdowns themselves, two different kinds of nuclear accidents are just as bad.
One is an explosion ejecting spent fuel, typically at a reprocessing plant. The other is a spent fuel fire, such as in the spent fuel pool at Fukushima Daiichi in Japan. We could lump them together and call them “dangerous fuel accidents.”
Each of our worldwide 432 commercial nuclear reactors adds one reactor year for every year it runs, so the various nuclear reactors in the world have accumulated about 8,000 reactor years.
According to the figures from the nuclear industry and the Nuclear Regulatory Commission (NRC), we should expect a dangerous fuel accident in a commercial facility once every 20,000 to 50,000 reactor years, or about once in every 10,000 reactor years for the old boiling-water reactors like Vermont Yankee.
This translates to one meltdown every 50 to 100 years, worldwide.
With 8,000 reactor years accumulated so far, we should have had no more than one dangerous fuel accident, at most. Instead, we have had 10.
I am not counting the partial meltdowns in the Sodium Reactor Experiment (SRE) reactor in 1959 or Fermi-1 in 1966. I am also not counting the explosion and total meltdown in Lucens in 1966. All of these involved designs that were earlier than those in current use.
The ones I am counting include those at Saint-Laurent (one in 1969 and the other in 1980), Bohunice in 1977, Three Mile Island in 1979, Chernobyl in 1986, Tomsk in 1993, and three meltdowns and a spent-fuel fire at Fukushima Daiichi in 2011.
The predictions on safety have been proven wrong. They are not just a little wrong. The accidents are coming at a rate of one dangerous fuel accident every 800 reactor years, more than 20 to 50 times faster than they should.
There are reasons for this situation.
It is partly because the planners use a concept called a “design basis accident” or a “maximum credible accident” as the basis for their plant designs. They imagine what they think can happen, and they plan for that. Everything else is “beyond design basis” and not credible.
The problem is, in the real world, stuff happens. In fact, every one of the accidents listed was “beyond design basis.” Every one was credible only in retrospect.
The safety projections of the nuclear industry are being overwhelmed by one simple design flaw: they fail to take reality into account.
We can see the result of the thinking of the nuclear industry especially clearly at Fukushima Daiichi.
Common sense based on reality would seem to dictate that a nuclear power plant in the most earthquake- and tsunami-prone place on earth should be built to withstand the worst earthquake on record, 9.5 on the Richter scale, and hold back the tallest open-ocean tsunami waves ever to hit Japan, a bit over 30 meters (more than 90 feet).
The earthquake of 9.0 that happened, and the 14-meter-high wave that followed, would have been withstood easily even if that standard had been used.
Instead, the industry chose to build the plant to specifications based on the worst events recorded over a 60-year period where the plant was constructed, on one small beach in Honshu. These were an earthquake of 7.6 on the Richter scale and waves of 4.7 meters.
The nuclear industry and the NRC are not driven by common sense. And, in mathematical terms, the data set they were using was not statistically significant because it represented only 60 years of history at that beach, so they failed even from a purely scientific point of view.
What they actually are driven by seems to be a question of what they can afford to do.
They could have hardened the plants against an earthquake of 9.5 and built them to withstand a 30-meter wave. But they did not. Even if they had been told to prepare for a 9.0 earthquake and 14 meter tsunami, they doubtless would have objected to the cost. And with captive regulators, they would have carried the point.
* * *
You might have thought the recent events in Japan would make the question of safety perfectly clear. Instead, they have only brought a new message from the nuclear industry, intended to give us comfort.
According to the industry's latest pronouncements, radiation is not nearly as dangerous as anyone thought. They are saying we can live with it safely, and let them continue making money as though nothing had ever gone wrong.
It reminds me very much of the old Stanley Kubrick movie, Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb. But that was a comedy; this is real.
The NRC has decided to do a two-pronged investigation to study the safety of the U.S. nuclear industry. Unfortunately, this investigation will be meaningless unless it addresses the inadequacies of analysis based on the maximum credible accident.
Nevertheless, predictably, they will refuse to take real-world conditions into account and continue using the old criterion, despite its clear failure to predict anything at all correctly. And we will need to experience an event sufficiently painful - much more painful than what has happened in Japan - before we get change on this issue.
The real world has a habit of catching up with people who live in fantasy. There is an alternative, which is to require the nuclear industry and the NRC to face reality, but they won't do it on their own.