Quotes of the Day

An accident at a nuclear power plant is a little bit like driving your car over a cliff. The odds of it actually occurring are not high — what with all the guard rails and traffic laws and simple common sense keeping you safe. But if it does happen, you're in very, very deep trouble.

Of all the varieties of risk humans assess poorly, it's that kind — the improbable but disastrous — that flummoxes us the most. Even during the peak of the AIDS crisis, the odds of anyone outside of the high risk pools actually contracting the virus were comparatively low, but we reacted with hysteria and confusion and bias all the same. The same was true of mad cow disease (remember that?), and the same continues to be true of real but still manageable emergencies like the H1N1 flu epidemic of 2009.

The George W. Bush administration famously enunciated what it called "the 1% doctrine," the idea that even a 1% risk of an attack by terrorists or weapons of mass destruction would be treated as a 100% certainty, simply cause of the enormous potential cost of inaction. But there's a price for rash, wongheaded action too — in lives, treasure, global credibility — and we paid it after the rush to war in Iraq in 2003.

The same fallible balancing of risks and probabilities is going on again, as countries around the world watch events unfolding in Fukushima and wonder just what the danger is of a similar disaster striking their own nuclear power plants and just what they should do to prevent it. In the U.S., with 104 active reactors — the most in the world by a large margin — the question is particularly pressing.

The truth of the nuclear equation is, in many ways, less scary than the headlines would make it seem. The reactors we have online provide 20% of the power we consume — no small thing — and while we have not built any new plants since the Three Mile Island disaster in 1979, we haven't had any accidents even remotely like it either. Could the airline industry or the space shuttle program boast the same kind of safety record? The relative safety of America's nuke plants is a credit to strict rules and good oversight — both mostly supplied by the Nuclear Regulatory Commission (NRC) — so credit should be given where it's due.

But so should criticism. The NRC's response to Fukushima has been brisk and candid concerning events on the ground half a world away. It was NRC chief Gregory Jaczko who was the first to reveal that the coolant pool in reactor 4 had likely lost all its water — a very dangerous development and one the Japanese themselves had been dissembling about for a while. But the same American watchdogs have been more dilatory about stateside reactors. The NRC is set to vote on a 90-day review of the health of America's entire nuclear grid, which is better than not conducting such a review. But it also has the feel of mere bureaucratic due diligence about it — as NRC spokespeople themselves have made clear.

Bill Borchardt, the commission's director for operations made a fair enough point when he remarked to the Associated Press that he and other regulators have "asked ourselves the question every single day: 'Should we take regulatory action based upon the latest information?'"

What he meant by that is should they take action based on the latest turns in the Japan accident in particular — and the smart answer to that kind of question is generally, no. Senate Minority Leader Mitch McConnell made a similar argument when he observed that the middle of an environmental emergency is never a good time to make long term energy policy. But that doesn't mean it isn't a very good time to examine some of your bedrock assumptions. And here the NRC is falling down.

Borchardt recently — and alarmingly — told the Associated press, "I'm 100% confident in the review that we've done and we continue to do every single day that we have a sufficient basis to...conclude that the U.S. plants continue to operate safely." Hundred percent doctrines are as foolhardy as the 1% kind, and the NRC's own track record makes that clear.

As TIME's Eben Harrell reported on Time.com, the Union of Concerned Scientists (UCS) has issued a report of what it calls 14 nuclear near-misses in 2010 alone, all of which should theoretically have been prevented by the NRC. Flooding in a Joliet, Ill. plant; a roof leak leading to electrical shorts in an Annapolis, Md. plant; stuck valves in a Diablo Canyon, Calif. plant which went undetected for 18 months, were all among the commission's white knuckle moments.

It's the California plants, of course, that cause the most worry, since the state's sweeping coastline and extreme tectonic instability so closely mirror Japan's. The state legislature is raising the pressure on utility companies to delay relicensing of any nuclear power plants until new, three-dimensional seismic mapping can be conducted of any relevant faults, including those off coast that could cause tsunamis. The plant operators themselves are resisting, which is certainly to be expected, but post-Fukushima, their arguments pack a lot less punch. Operators of the San Onofre plant near San Diego, for example, point to the towering 30-ft. seawall that should surely keep their plant safe from any monster wave they're likely to get. That's a fair enough assumption, except for the fact that it's the same one the Fukushima designers made — until the very moment a 33-ft. tsunami overtopped their wall, washed out their back-up generators and set the entire long drama in motion.

"It's time to revisit the safety of these plants in light of what we have learned from Japan," Calif. State Senator Ellen Corbett understatedly told the AP. Corbett points to the 2-3% annual risk of a major quake in California and a 46% chance of one that exceeds 7.6 magnitude occurring sometime over the next 30 years.

Ultimately, that kind of cold number crunching has to be part of any long-term safety plan. But as the NRC begins its 90-day review — with periodic updates required at 30 and 60 days — they ought to keep in mind that common sense, cool heads and an utter absence of hubris must be equally big parts of the mix. Disasters go from being remote to real in an eyeblink. The key is to have the seawalls in place — both literal and regulatory — before the inevitable surprises happen.

• Jeffrey Kluger
• After Japan, nuclear safety rests on balancing fear and hubris