Not In My Backyard

As they say, denial is more than just a river in Egypt.  Apparently, it’s also a fundamental part of nuclear energy planning in the United States.  The recent disaster in Japan has renewed fears about the vulnerability of US nuclear reactors to earthquakes or other disasters.   In at least one case, the reaction from officials has been a not-so-reassuring “don’t worry, it can’t happen here.” 

The plant generating the most attention is the Diablo Canyon nuclear power plant in coastal California, brilliantly built in a seismically active area with two offshore faults within three miles.   According to Chris Kirkham of The Huffington Post, some wise souls raised fears about earthquake hazards back in the 1980’s when the plant was being approved, calling for additional planning to consider this risk.  The federal Nuclear Regulatory Commission (NRC) responded that this was unnecessary because the plant could withstand any earthquake that might occur.   After activists sued the NRC, the U.S. Court of Appeals in Washington DC  weighed in with their not-so-scientific wisdom, and determined that if energy companies had to plan for every gosh darn contingency, nothing would ever get built. 

Officials maintain that the Diablo Canyon plant can withstand a 7.5 magnitude earthquake.  But as Kirkham notes, this response is eerily similar to that of Tokyo Electric Co.  officials who said that they didn’t anticipate an earthquake as strong as the 9.0 March 11 quake.   And so it begs the question:  what if a Big One near the Diablo Canyon plant happens to be higher than the NRC expected, say an 8.0 or 9.0?  What if said earthquake generates a tsunami?

The U.S. Appeals Court decision raises a legitimate question that gets at the heart of risk assessment and management:   How much do we plan for?  We can’t let paranoia dictate everything in life or we’d be paralyzed.  To get around this, planners typically apply a basic formula for risk assessment that goes something like this:  

Probability of an event happening ×  The consequences if it happens  =  How much we need to worry about it.

NRC officials seem to be saying that the probability of an earthquake above 7.5 is so low that -- even though the potential consequences are horrific – we can ignore it.

It seems reasonable to disregard an event with near zero probability of occurring.  What’s unreasonable is the excessive confidence in that probability.  Science writer Joel Achenbach of The Washington Post has written about earthquakes for years, and in the wake of the recent disaster has shared his take on those who maintain they can predict where or when major earthquakes will occur with any accuracy.   See his blog posts (one and two) and his recent Post article.   In short, they can’t, and many scientists, including some at the U.S. Geological Survey (USGS), acknowledge that. 

In the midst of all this, the NRC announced that it is undertaking an analysis of at least 27 of the 104 US nuclear reactors to determine earthquake risk according to E&E Publishing/Greenwire.  The joint analysis by the NRC, USGS and the Electric Power Research Institute will combine the latest data with new modeling techniques and focus on reactors in the East and Midwest, where seismic activity is less understood.

Sounds encouraging.  Except that all of this analysis is still based on probabilities, which are based on what scientists think they know. According to E&E Publishing, USGS and NRC scientists acknowledge that they don’t always know faults exist until an earthquake actually happens.  This was almost the case with Diablo Canyon.  Tom Dickerson of Rolling Stone reported that geologists discovered the second fault near Diablo Canyon – less than a mile away -- only after the plant was already built.  The California Energy Commission has voiced concern about the other unknowns associated with the two nearby faults, including how they might interact together and whether they could extend under the plant.  

Formulas, models and discussions of acceptable risk make perfect sense in the abstract.   But given what we know now, how many of us would approve a Diablo Canyon in our own backyard?   I know I wouldn’t.  This is the barometer that all public officials should use for deciding acceptable risk:  would you want this near your home?  Your children?  How much faith do you really have in formulas?  

Would you bet your life on it?   Then don’t bet someone else’s.