The probability that a particular reactor core at a U.S. nuclear power plant will be damaged as a result of a blackout ranges from 6.5 in 100,000 to less than one in a million, according to a 2003 analysis by the Nuclear Regulatory Commission.
Compared to many more mundane technologies, such as cars and airplanes, nuclear power facilities are quite safe. For the past half century, we have assumed they are safe enough.
The disaster at Japan’s Fukushima Daiichi nuclear complex demands that we rethink what “safe enough” means.
American scientists have pointed out that U.S. power plants have almost no chance of experiencing the kind of earthquake and tsunami damage that the Japanese plants sustained. While the United States does have some plants located in earthquake-prone regions, the fault lines near these plants are “slip-strike” faults, while the one that triggered the tsunami in Japan was a “thrust” fault. Slip-strike faults lack the potential to cause catastrophic tsunamis, according to Per Peterson, a senior scientist with the Accelerator and Fusion Research Division at the Lawrence Berkeley National Laboratory.
But the earthquake and tsunami that struck Japan exactly three weeks ago merely precipitated the atomic plant’s crisis. What actually destroyed it was the combination of an extended blackout and the failure of its backup power systems. The question we need to consider is not whether similar earthquakes and tsunamis can happen here; it is whether a plant could, for any combination of reasons, simultaneously suffer a failure of its outside electrical supply and its on-site backups.
The answer, frighteningly but clearly, is yes. Such a combination of failures, though unlikely, can certainly happen here.
Nuclear power plants require huge amounts of power to prevent their reactor cores from melting down and releasing radiation. Given the extremely high power demands, battery power can sustain plants only for a very limited time. Of our nation’s 104 nuclear reactors, 93 can be kept cool using batteries for only four hours.
Since the attacks of Sept. 11, 2001, the NRC has required power plant operators to develop ways to keep plants cool longer when they are severed from the grid. But their abilities to function independently remain limited. Furthermore, the types of natural disasters that might knock out external power can also damage backup systems. Richard Denning, a professor of nuclear engineering at Ohio State University, said in the aftermath of the Japanese accident that the steps many American power plants have taken “certainly could have made all the difference in this particular case,” but only “assuming you have stored these things in a place that would not have been swept away by tsunami.” The Japanese plants were equipped with backup diesel generators, in addition to batteries, but they were swamped by the tsunami within an hour after the earthquake knocked the plants off the grid.
Japanese engineers never allowed for the possibility that their nation could suffer a magnitude 9.0 earthquake and a 45-foot-high tsunami until it actually happened. While scientists may believe such events cannot happen here, there is an element of risk – small, but unquantifiable and certainly present – that they are wrong. Another risk factor is that quakes of lesser but still unanticipated magnitude could affect regions like the East and Midwest, which are not prepared for even California-scale temblors.
Apart from earthquakes, there are other natural conditions that occur here that can cause the same sort of two-pronged damage to the external power grid and a plant’s on-site backups.
In the Northeast, ice storms pose this sort of risk. The catastrophic Great Ice Storm of 1998 left some parts of southern Quebec, western New Brunswick and eastern Ontario without power for an entire month. Ten years later, another severe storm turned off lights for 1.25 million people in New Hampshire and surrounding states. The Northeast is home to 26 nuclear reactors, including New Hampshire’s Seabrook Station and the Vermont Yankee Nuclear Power Station.
A severe ice storm probably would not knock out a plant’s diesel generators, if it has them. But it might delay fuel replenishment long enough for the generators to fail. Or an accompanying flood, either from coastal storm surge or from ice jam flooding on a river, could do the same sort of damage to backup systems that Japan’s plants experienced.
Elsewhere, ice storms may not be a threat, but hurricanes are. In addition to causing the sort of destruction that might disable backup generators, Katrina disrupted electricity service for over 1.7 million people. Louisiana has two nuclear power plants, including the Waterford Steam Electric Station, located in Kilona, only 25 miles from New Orleans. Several other power plants stretch along the Gulf Coast, in Texas, Mississippi, Alabama and Florida. Still others dot the entire Atlantic seaboard.
Many areas of the country are potentially subject to windstorms, river flooding, dam failures, wildfires and other disasters that could pose a threat.
What happened in Japan proved that safety measures there were unacceptably weak. Our plants run similar risks, and we ought to acknowledge that our safety measures are equally unacceptable. A 6.5 in 100,000 risk is low, but when it comes to radioactive materials, it is too high, especially given the margin of error in our own ability to estimate such risks.
Nuclear meltdowns are far less likely than airplane crashes or car wrecks, but they are far more devastating, and the consequences stretch across generations. The magnitude of the potential harm requires that if we are going to use nuclear power at all, we have to backups for the backups, and probably backups for those backups as well. The plants must be safe under all conceivable circumstances and, as Japan showed, in some that are heretofore inconceivable.
Natural disasters and diesel generator failures are by no means inconceivable. Japan’s disaster warns us that the risks they pose to nuclear power plants today are more than we ought to be willing to accept.
Leave a Reply