Editor’s note: Any day now, the Bush administration is expected to officially recommend building a high-level radioactive waste repository at Yucca Mountain. Facing long odds in Congress, state officials in Nevada are hoping that the specter of transporting nuclear waste on the nation’s highways and railroads will persuade other states to keep the waste where it is. But, as Jon Christensen writes, the main reason why the nation shouldn’t gamble on Yucca Mountain is the mountain itself.
We’re on our way to Yucca Mountain. And there are some things I must tell you before we get there. That is, before the Bush administration and Congress decide once and for all to entomb the nuclear age’s most deadly legacy in the inauspicious desert landscape alongside Highway 95 about 100 miles northwest of Las Vegas. The most important thing to remember is this: It’s not about Yucca Mountain. And yet it is.
The other thing to keep in mind is that there is not just one Yucca Mountain. There are three. There is Yucca Mountain, the place, a heap of ash and rubble that was blasted from a nearby volcano some 12 million years ago and cemented together and eroded over eons into the shape of a wave breaking westward across a desert sea. There is little love lost on this Yucca Mountain, even by Nevadans like me, who cultivate a taste for such unworldly landscapes. If I didn’t have to come to Yucca Mountain, I wouldn’t. And neither would you.
But we do. And I have, over and over again, to try to understand this place and what we are doing here, because there is another Yucca Mountain. And this Yucca Mountain is the political answer to the question of what to do with highly radioactive spent fuel from nuclear power plants, weapons facilities and research reactors around the country. This Yucca Mountain offers salvation for the nuclear industry—and for politicians from states that don’t want the waste stockpiled within their borders.
Then there is Yucca Mountain, the computer model. This Yucca Mountain is the most difficult to see, let alone understand. In fact, it doesn’t really even exist. It is nothing more than the output of a computer program called a Monte Carlo simulation that calculates the probable risks of burying nuclear waste in an idealized version of the real mountain. This ghost of Yucca Mountain embodies both the technocratic hubris and the gambler’s faith in the odds that have taken us to this strange place.
A barren ridge on the border of the atomic-bomb-blasted Nevada Test Site, Yucca Mountain may look like the perfect place to bury the most long-lasting toxins ever made. But under the surface, as the Energy Department’s own exhaustive research has shown, Yucca Mountain has serious flaws. The Energy Department hopes to get around these problems with the computer model, which shows that, chances are, no real harm will be done, if you believe the model.
As D-Day for Yucca Mountain nears, these three different versions of the same place are coming together. Energy secretary Spencer Abraham has vowed to formally recommend Yucca Mountain before the end of February. Nevada governor Kenny Guinn will then formally object, sending the decision to Congress where a simple majority of both the House and the Senate will be all that is needed to override the state’s pro-forma veto.
Nevada’s odds have never looked good in Congress. And they don’t look much better these days, despite the fact that Harry Reid, a Nevada Democrat, became the majority whip when the Democrats gained control of the Senate last year. In a dispirited long-distance call from Washington, Reid told me that he had picked up one or two votes on top of the 34 votes he marshaled against building a temporary repository in Nevada in 2000. That was enough to prevent the Senate from overriding a Clinton veto but nowhere near enough to prevent a majority of senators, who want nuclear waste out of their states, from sending it to Yucca Mountain.
The unfortunate truth is that Congress doesn’t really care much about Yucca Mountain, the actual place. “In fact, the technical suitability of the site is less of a concern to Congress than the broader issue of whether the nuclear waste problem can be solved at an affordable price in both financial and political terms,” noted a cover letter to a draft Energy Department recommendation that was leaked to the media last year. The Energy Department quickly disavowed the letter, blaming a contractor for the inadvertently telling wording. Of course, when politicians talk about Yucca Mountain, they don’t say they want a decision based on money and politics. No. They want a decision based on “sound science.”
But it is science that has revealed Yucca Mountain’s flaws. Over more than 20 years of poking and prodding, this otherwise unremarkable place has become one of the most intensely studied pieces of real estate in the world, at a cost of close to $6 billion so far (and $58 billion if it is built). Any way you look at it, it’s not a pretty picture. There are still some very troubling unknowns. The mountain is crisscrossed by earthquake faults. There are dormant volcanoes nearby. But despite the fact that this is a desert, the main concern turns out to be water.
Only four to 10 inches of rain falls on Yucca Mountain each year, barely enough to keep a sparse covering of grass and creosote bush alive. Originally, the volcanic ash that makes up the mountain was thought to be so tightly compressed that what little water there is would not flow through the layers of rock. Nuclear waste could be isolated from the environment by the rock itself. A five-mile tunnel that was drilled into Yucca Mountain for scientific studies appears bone dry. But, as scientists have dug into the mountain, they have found that the rock is riddled with fractures.
On average, they found a fracture every couple of inches. And they found water moving through the fractures. They thought the water was moving slowly. But then they found Chlorine 36, an isotope left by atmospheric bomb testing in the 1950s, in a sample of water taken at the repository level, 800 feet below the surface. That meant rainwater might have percolated down to the area where nuclear waste will be stored in just 50 years and in another 50 years or so could reach the aquifer 1,000 feet lower. At the nearby Nevada Test Site, scientists found that plutonium from underground bomb tests had hitched a ride on microscopic specks of clay suspended in groundwater and had traveled nearly a mile in 30 years, much faster than they had expected.
Add all of that up, and Yucca Mountain does not look like such a great place to bury something that must be kept isolated from the environment for 10,000 years (and hopefully much longer, because the half-lives of some of the most potent elements, such as plutonium, are much longer; but that is the time period mandated by law). In fact, the Energy Department has essentially conceded as much. It now asserts that waste canisters made from a high-tech alloy called C-22 will last at least that long, even though the metal has been tested for only a few years.
This is where the computer model of Yucca Mountain comes back into the picture. Officially it is called the “total system performance assessment.” The analysis uses a Monte Carlo simulation, a technique commonly used by scientists, businesses and industries to model the probability of various outcomes in a complex situation with many different inputs, each of which has a wide range of probabilities of occurring.
Take the probability that water will drip through cracks in the mountain and onto the waste canisters; mix that with the probability that the canisters will corrode; add to that the probability that water will carry the contamination to the aquifer below; and finally factor in the probability that a family living in nearby Amargosa Valley will drink and irrigate crops with that water. Incorporate the probability of a volcanic eruption and another Ice Age making this a much wetter place, and then throw in the probability that some future prospector—let’s call him the “unluckiest man in the world,” as Energy Department scientists do—will decide to drill or dig at the site.
After sampling all of these variables many times, like drawing cards for hundreds of poker hands, the Monte Carlo simulation spits out a probability curve. Lo and behold, the result: the prediction that radiation will not leak from the site for at least 10,000 years (remember, the canisters are predicted to last that long; if they don’t all bets are off, and everything will happen that much quicker).
But if the canisters last that long, the radiation will have diminished. And any contamination that leaks from the repository will be partially absorbed in the rock and diluted in the water under Yucca Mountain. So the dose to a hypothetical family in Amargosa Valley won’t rise above 15 millirems, the maximum off-site dosage allowed by the Environmental Protection Agency, for hundreds of thousands of years, according to the model.
“In some sense, it is science fiction to project out 300,000 years,” Abraham Van Luik, the official in charge of the modeling, once commented while showing me around Yucca Mountain. “It gets more and more difficult to defend your assumptions as you move into the future.”
But, he hastened to add, “our modeling is over-conservative.” One thing is certain, Van Luik asserted: “Absolutely no one is going to get hurt by this repository for hundreds of thousands of years.”
The problem is that “absolutely” is not something that can be said about a model based on probabilities. There are very few certainties about this Yucca Mountain. The Energy Department likes to say that the model has revealed “no show stoppers” at Yucca Mountain—no single factor that would disqualify the site. But because of the uncertainties inherent in the model, the Energy Department is probably incapable of even recognizing a “show stopper,” let alone determining whether the site will be safe, says Rodney Ewing, a University of Michigan scientist who specializes in nuclear waste management and served on a peer review panel that evaluated the Yucca Mountain computer model in 1998.
“The uncertainty in these analyses was so large as to make them unusable,” says Ewing, who published a scathing article in the journal Science a year later, criticizing the department’s reliance on the computer model, rather than the actual characteristics of the mountain, to determine whether the site was appropriate. “One should not expect greater success with such a prediction than we have in other fields,” he says, “such as predicting which presidential candidate gets the electoral votes from Florida.”
In fact, says Ewing, we have a lot more experience predicting political results than we do predicting how a complicated combination of geology and engineering will perform over thousands of years.
“If an airplane were built in this way,” he says, “that is, smaller versions of the plane hadn’t been test- flown, but you were assured that good and competent engineers and scientists had modeled the plane’s ability to fly, would you fly on the first airplane based on these analyses?”
Yucca Mountain will not only be the first high-level radioactive waste repository in the United States. It will undoubtedly be the only one. And the only thing we know for sure is that we cannot know for sure whether it will be safe.
And if that makes you nervous, just try to remember this: It’s not about Yucca Mountain.
And yet it is.