The March 11, 2011 earthquake and tsunami caused massive destruction in Japan. Six nuclear reactors at the Fukushima Dai'ichi power plant became a significant cause for concern in the aftermath as the plant lost power and reactor cores started melting down. Both the earthquake and tsunami were much larger than predicted and caught everyone, including nuclear power plant designers, by surprise. How could the geologists who predicted these events have been so wrong?
The reactors had been designed to withstand an earthquake of magnitude 7.9. After all, in the entire 20th century, the maximum earthquake experienced in the region was magnitude 7.8. The Tohoku earthquake on March 11 was magnitude 9.0, releasing 45 times more energy than the reactor was designed to withstand. The tsunami wave height was also significantly underestimated. The Fukushima plant was designed to withstand a tsunami of 5.7 m height. Recent estimates put the Tohoku tsunami wave height at 14 m.
In the nuclear field, policy makers often demand predictions about Earth behavior: What's the largest earthquake that will occur at the site of a nuclear power plant? Will the nearby fault move? Is this a reasonably safe location to site a geologic repository for high- level radioactive waste? The advantage of these predictions is that they provide simple and straightforward parameters for policy decision-making. The disadvantage, as we now know, is that they can't be made with the necessary accuracy.
Engineers tell us that they can build structures, including nuclear power plants, to withstand earthquakes, but they need to know, with some degree of precision, the maximum earthquake likely to occur. This is a reasonable request from an engineering point of view: Engineers make predictions for a living. They design technologies, such as nuclear power plants, that operate on the scale of decades, and make successful technologies, by building prototypes, testing, and fixing the failures.
Geology is a historical science; it explains the past. When geologists make predictions, they cannot do so with much accuracy or precision. They can tell you that there will be an earthquake on a fault, such as the San Andreas, sometime in the future, and that, based on historical and other evidence, it might be large, but they cannot say when the earthquake will happen (even to the century) nor how large it will be (certainly not to such an exact number as 7.9).
The Earth systems that geologists study--the subduction trench off the east coast of Honshu Island, Japan, for instance, the one responsible for the Tohoku quake--are complex. Multiple faults make up the area, some exposed, some not; strain is built up unevenly on these faults, and releases unevenly, too. These Earth systems operate at the scale of thousands to millions of years. A few decades are almost irrelevant to most Earth systems. Over such time scales, no testing of predictions can be done. The more complex the system, the longer the time scale, and the larger the area under investigation, the greater the opportunity for failure of a precise prediction.
Why these failures? Geologic knowledge is incomplete. Geologists simply do not know all the processes that operate in a given system, nor all the boundary conditions of the processes, and often not even all the basic parameters. For instance, the recent 6.3 magnitude earthquake in Christchurch, New Zealand, occurred on a previously unknown fault that did not break the surface. Geologists are learning all the time, adding to the current level of knowledge. But it should come as no surprise, then, that when asked for a precise value--the maximum possible earthquake, for example--geologists too often lack a reliable capability to provide this information.
Does this mean geologic information is irrelevant? Not at all. On the contrary, it is terribly important. One must understand, to the best of one's ability, the geologic conditions at different sites for nuclear power plants or geologic repositories. There are clearly locations that are not suitable for such facilities. Locating a nuclear power plant atop an active fault is probably not a good idea, for instance. Sites can be judged in a comparative sense: One site may pose less geologic hazard than the other. What geology cannot provide is precision: Predicting the maximum earthquake to hit the Fukushima area as a magnitude 7.9, a prediction given to two significant figures, was clearly unreasonable, as we know with hindsight. Geology can tell you where it is dangerous to build; it cannot guarantee safety.
The lesson of Fukushima is to carefully consider the geologic setting before you build--and moreover, before you build six reactors in close proximity to one another. The nuclear crisis at Fukushima will hopefully cause a reexamination of all nuclear power plants in seismically active regions--and create pause in those countries that are considering building new plants in seismically active areas. This crisis may inspire less reliance on the specificity and precision of geologic predictions, but more on the substance. And we can apply this type of knowledge to geologic repositories for nuclear waste, for which there is now a clear and urgent need.
This essay is the second in an Inside NOVA series presenting different viewpoints on Japan's nuclear crisis and its impact on the future of nuclear energy. Read other articles in this series.
For more on Japan's devastating earthquake and tsunami, watch the original one-hour documentary Japan's Killer Quake, which premieres Wednesday, March 30 at 9pm on most PBS stations and will be streaming online after that date. NOVA will also investigate the future of alternative and renewable energy, including nuclear, in Power Surge, premiering Wednesday April 20 at 9 pm. Please check your local listings to confirm when these programs will air near you.
March 30, 2011 5:21 AM
Considering the potential for disaster in a nuclear power plant they should be built to withstand the largest earthquake possible no exceptions. I know that cost has to be considered in the planning of a power plant. So what is the cost of this one? Environmental damage will with out doubt be in the billions. More than enough to have built the reactors to withstand any eventuality. So was the design tempered by an acceptable risk? Probably and the only reason to do that is profit. That is the Achilles Heel of the nuclear industry. To do all the things necessary to make nuclear reactors safe would be cost prohibitive. In an industry where no one knows what to do and by definition has no plan to deal with meltdowns or large releases of radiation. There can be no compromise in their design.
Nuclear power is clean efficient power as long as nothing goes wrong. When it does however, and it does, it is arguably the most destructive polluting energy source we have. I believe reactors can be built that are safe. Do I believe they will be, no. Cost is a factor, more disturbing to me I don't believe the experts are as confident as they claim to be. I heard the comment just the other day "who could have foreseen this disaster"? I don't know but someone should have. If we are not at the point in our acquisition of knowledge that we can, then we have no business building nuclear reactors. Until we can prove we have the knowledge to deal with nuclear disasters we have no business operating the ones we do have. Someone needs to take the initiative and bury these now before we all wish we had. The only conceivable reason I can think of not to is some nuclear engineer or utility somewhere really wants this to work, regardless of cost. I'm hoping they will stand back and reevaluate their motivation. Typhoon season is right around the corner and your reactors are no longer built to withstand them.
March 30, 2011 11:02 PM
Dear Mr. Miller,
I would like to remind you to give Dr. Macfarlane the attention her professionally written article deserves. If you had read the article, your statement, "they[Nuclear Power Plants] should be built to withstand the largest earthquake possible no exceptions", would have been answered. As Dr. Macfarlane clearly stated, at the time the Fukushima Dai-ichi reactors were built, the largest recorded earthquake was 7.8. The plants were built to withstand 7.9. In terms of pure energy released, the difference between a 7.8 earthquake and a 7.9 is approximately 1 megaton of TNT. Few power plants of other energy sources would take into consideration such margins of safety. In addition, in the case of the Fukushima plant, the reactors were shut down and brought to a safe condition immediately after the earthquake. The source of the current series of crises is the tsunami that resulted.
As for your doubt about expert confidence, what engineers face at Fukushima are very different from the problems engineers faced at TMI or Chernobyl. In those reactors, engineers and plant personnel began with lack of knowledge of what the problem was. In Fukushima, engineers know exactly what the problems are, but face tremendous logistical and technical hurdles to overcome. To put it in lay-man's terms, it is the difference between a doctor poking around to figure out what is wrong with his patient, versus a doctor who knows what needs to be done but lack the tools and facilities to do so. This is worsened by the fact that this reactor is of an old design. Much like all equipment and facilities with old designs, known issues and solutions can't possibly be completely addressed. Had governments and utilities around the world been more avid in upgrading and building next generation plants, we would be able to avoid such a crisis. For example, the Westinghouse AP1000 has a completely passive safety cooling system, which can keep the reactor safe for three days without any electricity, giving operators the time to avoid this crisis. This is the same in GE's ESBWR design. Yet as of today, no AP1000 or ESBWR have broken ground in Japan or the US. In contrast, China has began construction on the first AP1000s, and if said AP1000s were in Fukushima instead of those 40 year old BWRs, we would be receiving very different news out of Japan. My favorite explanation to those who don't know anything about nuclear engineering is this simple question, would you feel more safe getting into a well maintained car designed 40 years ago or a brand new car designed last year?
Of course, as a member of the public one is entitled to demand the best safety possible from any energy source. However, if we were to accept a no-risk attitude towards technology, then we won't be able to use any technology. All technology, from nuclear reactors to automobiles, have inherent risks. Some might argue there is no such risk with solar or wind. This is simply not true. Few people who envisioned a world covered by solar panels or dotted with wind turbines took into consideration the amount of rare earth elements that would go into said devices. If you don't know what environmental consequences (some might call them disasters) of extracting rare earths, just search for "rare earth mining". Solar and wind advocates also forget to mention the square mileage needed to feasibly supply the energy intensive cities humanity has built in the 20th century. For solar plants this land can't support life, and will threaten the local ecosystems. For wind turbines, the local avian and insect populations are also effected. Society must conduct a realistic risk-benefit analysis of all its energy resources, as we are all entitled and responsible to make an fully informed decision upon which energy resources we would like the majority of our energy comes from, and which we would prefer to avoid. Such a decision will be difficult and nowhere near idealistic, but it will dictate how humanity will survive in the coming century.
Sincerely,
-Raymond
March 30, 2011 11:26 PM
I totally agree. Accidents happen. We have to plan for them and the worst imaginable. It doesn't take much for a nuclear plant and its waste to cause vast destruction. It's too costly of a venture.
This is one place where the free-market gets it right. Private enterprise no longer participates in nuclear investments because the obvious and always rising risks and costs make it a lose, lose.
It's frightening to hear that nuclear energy is safe and viable as it is neither. It hasn't been safe and viable in the past and it still is NOT. I agree we "have no business...operating the nuclear plants we currently have".
Obama is by far my favorite president in my lifetime. But so many others, and unfortunately he as well, got it wrong when they said nuclear power is an important and viable option in our future. It isn't, NOT even if our “current thinking” has included 5 or 10 or 20 levels of redundancy in a human effort to “prevent nuclear disaster”.
March 31, 2011 11:31 PM
Dear Raymond, I in no way meant any disrespect to Dr. Macfarlane. The article she wrote was exceptional and I agree I should have noted that in my reply to it. I am truly sorry I neglected to say so at that time. Dr. Macfarlane I hope you will accept my apology.
I take no exception to Dr. macfarlane's article and I am well aware of the consequences of the other forms of energy we are obliged to utilize in the maintenance of our civilization. Global Warming, Coal Ash, Oil Spills even Wood Burning Stoves have consequences. But it seems, correct me if I'm in error, when it comes to nuclear disasters the doctor may know what needs to be done but never has the logistics or tools he needs to do it. It does not matter to me if the vehicle I ride in is vintage or new as long as the driver knows how to operate it, it has good tires and a tool box. In layman's terms "I'm good to go". So where is the tool box, why is it taking so long to use it? I'm sorry Sir, but your wrecked vintage car is still killing things I'm hoping we are not just waiting for it to run out of fuel. Please forgive me if I have doubts about new cars when last I looked they were involved in the deaths of 36000 per/yr.in the U.S. alone. I also find this unacceptable.
I have grown to love the pollution tinted red sunsets of my youth but I guess I could get used to Violet if I have too, do I have too?
April 3, 2011 11:30 PM
How about requiring a large safety zone around all nuclear power plants and using that space to install supplementary concentrating solar power plant which could keep the batteries charged in the event of a power outage, and in other times provide additional electricity to the power grid. I was thinking that the solar should be on high towers or even in a ring sort of like Stonehenge but higher. That way any Tsunami would flow through without damaging the solar installation.
I was also thinking that maybe the Tsunami could be diverted around critical installations by using a half-moat facing the water's edge, and maybe a series of angled wall sections that would break up the water into smaller sections so that the force might be less. (I am not sure if this would work, it's just a suggestion.) The half-moat would direct the water to the sides of any protected area. There might need to be another one in back to protect when the water recedes. Access over the moat would be via bridges designed to let water flow through underneath.