Splitting atoms to cool the planet

This week a panel of scientists handed the UN its ideas on ways to wean the world from using coal, oil, and gas. Two ideas led the list: more renewable energy – and more nuclear energy. Other recent studies have reached similar conclusions.

Many environmentalists see wider use of nuclear energy as a non-starter, citing proliferation, terrorism, waste disposal, large and long-term costs and, most of all, lingering doubts about safety. None of these concerns is trivial.

Yet a number of leading environmentalists have jumped ship to back nuclear, saying it's essential to saving the planet from global warming despite potential problems. Something besides wind power, geothermal, solar energy, and conservation is needed to curb carbon use.

At least 30 new nuclear plants are already rising in 13 countries, with nearly half of the new generating capacity in China, India, and Russia. In the US, 15 utilities are planning to build from 26 to 29 new reactors over the next decade or so. Some of these plans are simply "placeholders" for decisions not yet taken. Other plans are more concrete. They all draw on new types of reactors that their makers insist are less complicated, more reliable, and safer than reactors running today. Even if that's true, the success of such plants still depends on the quality of oversight.

In the US, that means ensuring vigilance by the federal Nuclear Regulatory Commission – and tighter scrutiny over the NRC itself. Deregulation in the US has prompted many nuclear-heavy utilities to enhance safety and reliability. A melting reactor is bad for business.

The challenge comes in overseeing the rest. Staff-level NRC officials may call for a plant to be shut down for safety repairs, as happened in 2001 at the Davis-Bessie plant in Ohio. But the utility convinced higher-ups that the problem wasn't dangerous and could be dealt with in a less-costly manner during the next maintenance outage. The delay postponed discovery of a far more serious problem. If the reactor had been allowed to run a few more months, the result could have been an accident worse than Three Mile Island.

Last year, a probe at Detroit Edison's Fermi 2 unit showed that in certifying the emergency backup generators, utility employees had used an outdated standard for a critical component. The result: For 20 years the tests gave no reliable evidence that the generators would work as advertised in an emergency.

The formal mechanisms are in place to catch such errors. But in practice, the desire to avoid imposing financial hardships on a marginally performing utility can trump the best judgment of regulatory staff.

If the reliable utilities have found the economics of failure too expensive to permit poor maintenance and safety records, regulators shouldn't be reluctant to impose the same "incentives" on poor performers.

Nuclear energy's future in a carbon-constrained world is far from certain. Public opinion routinely favors its use by a slight margin. But it's a hold-your-nose vote.

Some of the stumbling blocks to wider use of nuclear energy could be removed if companies make safety as important as the bottom line. The stakes – present and future – are too high to do otherwise.