We Must Take a Very Hard Look at Nuclear Power

After two weeks of the continuing saga of a dangerous situation at Japan's Fukushima Dai-ichi nuclear power station, and just after things began to look a tad more optimistic, today there is a sobering new development, that there may have been a breach of the reactor core with much greater release of radiation than anyone thought.

So I feel it's time to comment on nuclear power. In an era when we hear, ad nauseam, about "reducing America's dependence on foreign oil," and when we also have grown less tolerant of the problems with burning fossil fuels like coal to generate electricity, nuclear power has, to some minds, been a power source that should not be overlooked.

However, a lot of people are afraid of nuclear power. The world has seen some chilling instances of what can happen to atomic power plants, in Chernobyl and Three Mile Island (Pennsylvania).

At one point it was argued that a standardized reactor design, such as they use in France, would simplify the design of new nuclear reactors and thus make them safer. So to speak, we would be building a "perfected" nuclear plant.

However, even barring earthquakes—and how foolish does it now seem to build nuclear plants in an earthquake-prone country like Japan?—there will always be mechanical failures and operator errors.

To draw an analogy from flying: We've got very good and well-tested planes, these days. Occasionally a plane still crashes, because of some unforeseen equipment problem, pilot error, or both.

I recently revisited the story of what happened at the Three Mile Island nuclear power plant, near Harrisburg, Pennsylvania, in 1979. That plant had three lines of defense in the form of pumps for cooling water. Yet the operators did not believe the alarms that were going off—warning lights and sirens that signaled a loss of reactor cooling. Even worse and more incredible, they shut off the third line of backup pumps. (I wonder if anyone knows why they did that.)

So, you can build the most fail-safe, fool-proof, and idiot-proof reactor possible. You can train your operators to death. Yet there is no guarantee that they will do the right thing at all times. Human beings are always going to be the wild card, and human beings are fallible.

So you can reduce the statistical odds that something terrible will happen, but you can't get those odds down to zero. And the consequences of what can go wrong with a nuclear plant are very, very bad because radiation is very nasty stuff. It might be time to conclude that we don't want to take the risks.

Copyright © 2011 by Richard Stein