Confessions Of A Nuclear Power Safety Expert

By on June 16, 2011 in News 11

3464115270_3c602de1d8An expert on the safety of nuclear power plants comes to the conclusion that there is simply no such thing as an 100 percent safe nuclear reactor. Via Miller-McCune:

I soon came to the conclusion that neither international cooperation nor technological advancements would guarantee human societies to build and safely run nuclear reactors in all possible conditions on Earth (earthquakes, floods, droughts, tornadoes, wars, terrorism, climate change, tsunamis, pandemics, etc.). I am sadly reminded of this turning point in my life as I listen to the news about the earthquake, tsunami and extremely worrying nuclear crisis in Japan.

When Italy decided in the mid-’70s to add nuclear power to its power portfolio, young mechanical and nuclear engineer Cesare Silvi was among those attracted to the opportunities it presented. His work centered on nuclear safety issues — in particular, what might happen if something unexpected struck a power plant.

Corners he saw cut there eventually soured Silvi on that endeavor. His next position — at the Italian Commission on Nuclear and Alternative Energy Sources, which included work on nuclear disarmament — eventually soured him on nuclear energy itself.

“[If we] continue with nuclear power, there will definitely be worse accidents,” he argued in the wake of Japan’s Fukushima Daiichi disaster. Over the weekend, Italian voters agreed and overwhelming rejected restarting nuclear power in their country.

“Why not consider Three Mile Island, Chernobyl and Fukushima as warnings of greater catastrophes to come and avoid the inevitable by shutting them down, much like changing your diet and/or lifestyle after finding out that your cholesterol or blood pressure is elevated, rather than continuing down the same path until a heart attack or stroke strikes?”

“I was looking at low-possibility events, like a meteor striking the housing of a reactor or a car thrown at it by a tornado. These definitely had a small chance of happening, but the end result would have been horrific.” Plus, he says now, the proliferation of nuclear plants just adds more targets.

“Many laughed at such speculation and planning,” he says, “but then again, how many would have taken seriously a recommendation of extending the height of the seawall at Fukishima another six meters? They would have questioned your sanity, if you had argued that the 10-meter barrier was inadequate.

“Our problem is that we don’t know what will happen on any scale of time. Such uncertainty is OK when dealing with train trips or dinner choices. But it becomes problematic when considering the possible spread of very dangerous material that will stay deadly for hundreds, if not thousands, of years.”

“Human history is full of madness, full of catastrophes. Imagine if we had nuclear reactors when we fought wars in the past. If you try to consider all the events that might happen over the years, you start to ask, ‘What are the benefits of such an effort, especially when you have opportunities to get electricity in many other ways?’”

, , , , , ,

  • emperorreagan

    Not in particular about this article, but I’d say that anyone who believes they can say anything with 100% certainty is probably a madman.

    About this article: Engineers, whether they’re designing a car or a nuclear power plant, pick design conditions.  There are conditions under which every single piece of technology you ever touch may fail.  I think this guy probably has gone a bit overboard looking at worst case scenarios.  If you do that across the board, then you have to abandon technology entirely – for example, you can’t guarantee that a meteor won’t hit a chlorine plant.  It’s also not possible to build a meteor resistant chlorine plant, so instead of risking the chlorine cloud killing people, we’ll abandon on processes that require chlorine.

  • Anonymous

    Wouldn’t the chlorine dissipate fairly quickly if a disaster struck the plant? I imagine there might be danger in the immediate surrounding area for a few hours, maybe a day. Cleaning up and removing that threat would be trivial compared to the mess created by a nuclear meltdown.

  • emperorreagan

    Immediate danger to the surrounding area for chlorine and other industrial chemicals may indeed go away in a few hours or a few days.  However, many of the chemicals we manufacture and use in industry are highly toxic, carcinogenic, cause birth defects, etc. and a problem at that sort of plant can be swept under the rug much more easily than a meltdown, until there are cancer spikes and birth defects a couple of years down the road.  So maybe a meltdown ends up impacting fewer people, depending on the population density of where the reactor is built versus a chemical plant.    

    That’s not really the point I wanted to make though – more that decision making based on fanciful hypotheticals would paralyze the world.  This dude is caught up on 1 in two hundred trillion+ occurrences.  To make a rational decision, you have to weigh all of the outcomes you can conceive of for nuclear power and weigh them against coal, natural gas, and the rest of the array of options for generating power.  Nuclear certainly has the big scary possible negative, but there are tons of factors to consider in a cost-benefit analysis.  His 1 in two hundred trillion+ fantasy of a meteor of significant size to cause a meltdown hitting a plant is going to be statistically irrelevant.     

    By way of example, let’s say I run a lottery.  You have a 1 in 1,000,000 chance of winning my lottery (0.0001%).  You play every day for June, for a total of 30 plays.  The probability that you win once in your 30 plays is 0.003%.  Odds are slightly better with 30 plays, but you’re still not going to bet your retirement in my lottery.  

    The odds of you getting hit by lightning is ~1 in 600,000 or ~0.00017% – the odds of you, your son, or the next 3 generations of sons beyond that getting hit by lightning in one of those lifetimes is 0.00083%.  You’re not going to base too many life decisions on that probability, I’m sure, or tell your kids to hide in the basement their entire lives.

    The odds of a meteor hitting someone’s house are 1 in about 200,000,000,000,000.  No one is going to buy an insurance policy against a meteor strike (5×10^-13% chance for your lifetime).  If you assume the house goes through the same 5 generations, the chance is still basically zero by the time your great-great grandson is living there.    

    The odds that a meteor of size large enough to cause a nuclear meltdown hits that plant are going to be even longer than a meteor hitting your house. It’s basically zero. It isn’t going to factor into any sort of analysis, because there are much more likely things that could cause a meltdown.

  • Anonymous

    “If you do that across the board, then you have to abandon technology entirely”

    No, you would have to abandon applications of technology that require centralized administration and as a result necessarily suppress alternatives. We would be better off without uranium or chlorine. 

  • emperorreagan

    Are you making an anarcho-primitivist argument?

    Because without plants that produce a variety of chemicals, you can throw modern electronics, solar panels, and the like out the window.  Maybe that’s a good argument to make, but to be clear you can’t abandon industry and expect to maintain many of the trappings of modern western life.

  • Anonymous

    I am not, although they have the right idea; the point I’m making is that the current state of industry is not the only way by which ‘technology’ can be applied and controlled; read ‘towards a liberatory technology’ by murray bookchin for an (albeit outdated) idea of what I am talking about. Whenever you have centralized control or administration of technology you have the people who are making the decisions very far removed from most of the consequences of what they are doing; ‘risks’ that informed people would find unacceptable to them may be balanced out by the ‘benefit’, which could just be the profits they are making. Nearly all of what we currently use could be replaced or supplanted by carbon-based materials; for example, polymers produced from carbon contained in ‘waste’ combined with a 3-D printer is an infinitely desirable alternative to polyvinyl chloride, which is produced by huge industry because it is cheap. I do not know if such a solution presently exists, if it does not it is because the PVC guys got there first, like a fungus which secretes toxins that prevent the subsequent growth of other life. 
    My point is that it’s not as simple as ‘chemicals = technology’. To borrow a quote from bookchin, 

    “Is society so “complex” that an advanced industrial civilization stands in contradiction to a decentralized technology for live? My answer to this question is a categorical no. Much of the social “complexity” of our time originates in the paperwork, administration, manipulation and constant wastefulness of capitalist enterprise. The petty bourgeois stands in awe of the bourgeois filing system – the rows of cabinets filled with invoices, accounting books, insurance records, tax forms and the inevitable dossiers. He is spellbound by the “expertise” of industrial managers, engineers, stylemongers, financial manipulators, and the architects of market consent. He is totally mystified by the state – the police, courts, jails, federal offices, secretariats, the whole stinking, sick body of coercion, control and domination. Modern society is incredibly complex, complex even beyond human comprehension, if we grant its premises – property, “production for the sake of production,” competition, capital accumulation, exploitation, finance, centralization, coercion, bureaucracy and the domination of man by man. Linked to every one of these premises are the institutions that actualize it – offices, millions of “personnel,” forms, immense tons of paper, desks, typewriters, telephones, and, of course, rows upon rows of filing cabinets. As in Kafka’s novels, these things are real but strangely dreamlike, indefinable shadows on the social landscape. The economy has a greater reality to it and is easily mastered by the mind and senses, but it too is highly intricate – if we grant that buttons must be styled in a thousand different forms, textiles varied endlessly in kind and pattern to create the illusion of innovation and novelty, bathrooms filled to overflowing with a dazzling variety of pharmaceuticals and lotions, and kitchens cluttered with an endless number of imbecile appliances. If we single out of this odious garbage one or two goods of high quality in the more useful categories and if we eliminate the money economy, the state power, the credit system, the paperwork and the policework required to hold society in an enforced state of want, insecurity and domination, society would not only become reasonably human but also fairly simple.”

  • Anonymous

    No, it’s not overboard, it’s risk assessment. The probability of failure rises to 100% over time, from whatever cause, and the danger from failure is catastrophic. A chlorine release, while very dangerous, will dissipate quickly, whereas a nuclear accident like we see in Fukishima will be deadly for the next few centuries.

  • Anonymous

    Yes, yes & yes, but your use of the meteor example is a straw man argument. A meteor strike is an exceedingly rare occurrence, but if you consider all the different things that could lead to a nuclear disaster, it’s not nearly as unlikely as you make it out to be. There are also plane crashes, accidental or deliberate. Earthquakes & tsunamis, sabotage, warfare, or even operator error. Remember Chernobyl? In the US, we got lucky that 3-Mile Island is still a habitable zone. Cher

  • emperorreagan

    I’m not making the straw man argument – the guy in this article made the straw man argument, talking about low possibility events.  The sum of all low probability events is still going to be just about zero.  He is arguing that low probability events and the inability of humans to essentially predict all possible outcomes is a reason to abandon nuclear power.  In essence, the argument that is being made in this article is that while it’s an extremely low probability event, lightning could hit and kill you on a cloudless day so you should never walk outside.  His argument is flawed.  Nothing is 100% safe.  

    Natural disasters like earthquakes, operator error, and the like are much higher probability events that have to weigh in to any decision.  The odds of improper material handling, leaking nuclear materials, etc. are even higher than the odds of a meltdown (and I would argue a more troublesome feature of nuclear power, because disasters prompt aggressive response, whereas long term pollution is ignored).  The odds of contractors trying to cut corners are high.  

    He also equates raising the sea wall at Fukishima to truly low probability events (meteors, tornadoes hitting a reactor with a car), which is a false comparison.  The odds of a tsunami and/or earthquake affecting the reactor were high.  They picked the criteria for the sea wall height based on something (maybe 1% or 0.5%) which was obviously inadequate.  However, the height of the wave and the strength of the earthquake weren’t truly low probability events like the ones he mentions.

    I’m not making a pro-nuclear argument.  I’m saying that statements like “I soon came to the conclusion that neither international cooperation nor technological advancements would guarantee human societies to build and safely run nuclear reactors in all possible conditions on Earth (earthquakes, floods, droughts, tornadoes, wars, terrorism, climate change, tsunamis, pandemics, etc.)” or “Our problem is that we don’t know what will happen on any scale of time. Such uncertainty is OK when dealing with train trips or dinner choices.” are simply fear mongering.  You can’t account for all low-probability events in anything.  You have to weight possibilities according to their probabilities and some scale of their impact.  

    To discuss electric generation options in general, you have to weigh the entire process – from the impact of mining/refining and initial plant construction all the way through to final generation and the pollutants that may produce and other risk factors.  That’s how you make a rational argument against nuclear energy – not by publishing crap about meteors, terrorists, and tornadoes throwing cars.

          

  • emperorreagan

    The probability of catastrophic failure does not rise to 100% with time unless you speak in infinite time scales.

    A nuclear power plant has a finite life span, after which time it will be decommissioned.  Speaking in terms of infinite time scales isn’t appropriate. So no, the probability of catastrophic failure never even approaches 100%. 

  • emperorreagan

    His argument seems inconsistent to me.  His statement “If we single out of this odious garbage one or two goods of high quality in the more useful categories” is ultimately an argument for central planning.  

    The more you decentralize production, the more variety you would see.  If a hundred people produced buttons at home versus 1 corporation, you would see far more variety in the buttons produced by the hundred people.  

    I don’t think the problem is as simple as centralization versus decentralization.  Some things work well with centralization.  Some things don’t.  I would rather have local tailors, for example, as opposed to Target or Macy’s.  People would indeed then by fewer, higher quality garments.  I don’t think the idea of a 3D printer used for decentralized production is bad, either.  The better variety and quality of produce at a farmer’s market is preferable to the local grocery store’s selection.

    I would rather have centralized production of chlorine, though.  Production of chlorine gas is one of the two primary means to create hydrogen chloride, which is in fact a very important chemical for a variety of technologies.  HCl is an important etching agent for semiconductors, hydrochloric acid is used in pickling steel, and a wide variety of other industries use it.  A centralized and uniform production, removed from heavily populated areas and conducted by people specifically trained to deal with the hazardous processes involved seems much more desirable to me than everyone who requires it to produce it in small batches locally.