Guest Post by Elaine Hirsch. Elaine is kind of a jack-of-all-interests, from education to technology to public policy, so she is currently working as a writer for various education-related sites.
Black Swan Theory, as explained by Nassim Nicholas Taleb in his 2007/2010 book, The Black Swan, describes an event which is a disproportionally-rare occurrence, is unpredictable, but has a high-impact when it does occur. According to Taleb, Black Swan events include the September 11 attacks, the rise of the Internet, World War I and the development of the personal computer. As a result, the event’s non-predictability causes behavioral/psychological changes within people, especially ones who adhere to the scientific method for identifying events. Statistically speaking, these outliers pay a disproportionate role in public opinion and public policy.
Critics of nuclear energy point to the destructive capabilities of failed reactors and long-lasting effects of radioactive energy as reasons of pessimism. According to USA Today, the Union of Concerned Scientists cited “serious safety problems” that plague U.S. Nuclear plants as a main reason for halting nuclear energy programs.
Proponents of nuclear power plants point to the safety measures already in place and attempts to increase the safety of nuclear energy. Safety systems at nuclear power plants include the reactor protection system (RPS), essential service water system (ESWS), emergency cooling system (ECS), emergency electrical systems (ECS), containment systems, standby gas treatment and ventilation and radiation protection. All of these systems work to immediately stop the nuclear reaction in case of an emergency. The RPS terminates the nuclear reaction, stopping the production of heat, so that other systems can remove decay heat from the reactor’s core. Not all heat removal systems exist in all nuclear reactors. Every nuclear reactor has some combination of systems to remove decay heat from the core.
In evaluating whether Black Swan Theory contributes to the anti-nuclear sentiment, one must consider whether such events were isolated outliers or part of a larger trend in nuclear energy.
The Fukushima Daiichi disaster was unpredictable. This accident occurred after Japan suffered the worst earthquake in recorded history, which resulted in the malfunction of the nuclear reactor Most nuclear experts would have agreed, before the fact, that if such an impact occurred in the location where it did occur, a nuclear accident would result. Of course, even the best technologies at our disposal today cannot predict all of the effects that earthquakes can cause. Variables such as urban layouts vary across the world, and in these cases, only hindsight is 20/20.
.jpg){kind=spacer}
The Mihama and Oak Harbor incidents were results of human error, not the inherent danger of nuclear energy. The Mihama Nuclear Power Plant accident was a steam, not a nuclear, accident. The burst pipe had not been replaced or even checked for corrosion in 27 years. In Oak Harbor, Ohio, corrosion on a control rod could have, and should have, been both predicted and addressed long before the point at which the plant had to be shut down for two years.
The Statistics
Britain’s Health and Safety Executive has stated that there is a “million to one chance of an accident at a nuclear power plant killing people living nearby,” and that “such a level would equate to the average annual risk of dying in a traffic accident.”
A report from the Nuclear Energy Agency of the Organization for Economic Co-Operation and Development (OECD) states, “For OECD countries, frequency-consequence curves show that the risk of a nuclear accident with more than 100 latent fatalities is a factor of ten or more lower than the risk of an accident with 100 immediate fatalities from coal, oil, natural gas or hydro energy chains, and almost a factor of one thousand lower than the risk from liquefied petroleum gas (LPG).”
As the statistics show, well-managed nuclear programs pose less risk than other programs which we view as “safe”. The best way to change public opinion about the likelihood of nuclear accidents is to make more information available not only about the safety procedures in nuclear reactors, but also about the true causes of recent nuclear accidents—failure to arrest corrosion in two cases and an overwhelming natural disaster in the third case. Only by doing so will humans be able to driven by reliable statistics instead of the black shroud of the black swan.
—————
*Editor: These deaths were from causes unrelated to radioactivity — a collapsing crane and two tsunami-related drownings at the plant. See details here.
Note: Related to the points made in Elaine’s post, the IAEA is currently undertaking a significant review of Japan’s reactor fleet:
The preliminary report from the IAEA team to the Japanese government said NISA and nuclear operators had “promptly addressed” emergency safety measures after the accident at Fukushima Daiichi. It contained a range of recommendations to NISA to ensure thorough and lasting improvements in safety are made… The ‘secondary assessment’ of Japan’s program will be an overall evaluation that relates closely to the stress tests carried out in the European Union and elsewhere. However, the IAEA said the secondary assessment is to “inform whether to continue or halt operations,” implying that permanent closure of nuclear sites is a possibility. Many improvements to safety, site preparation and emergency management have been recommended in other countries, but none have so far suggested the permanent shutdown of a nuclear facility.
Brave New Climate 
30 replies on “Black Swan theory and the anti-nuclear sentiment”
you say “The Fukushima Daiichi disaster was unpredictable”
but this is a well known bias in human thinking (can’t recall but i think it’s called availability bias)
It’s only unpredictable until it happens – at which point it becomes predictable.
A familiar example is where people who know a stretch of road is dangerous (by having witnessed dangerous behaviour or near misses etc) complain about it to their council only to be ignored until a fatal accident occurs.
The council will say that it has been unpredictable – when perhaps the locals had been right and it was predictable.
Also this – if it can be predicted (even by people like our locals who obviously, until the fatal accident, were basically ignorant fools with no clue about the world) then it’s not a Black Swan event.
To be a Black Swan event not even the predictions of silly doomers would have existed – it was completely out of the blue and a surprise to everyone.
Purposefully confusing a Nuclear Facility accident with Black Swan events is to pretend that the many many thousands of people who have over the years warned about the dangers of nuclear accidents and the various prior accidents like Three Mile Island and Chernobyl
never happened
pop
ps, i am not anti nuclear (nor am i pro-nuclear)
LikeLike
Elaine: A well thought-out presentation. I could quibble over details (for example, it was not a corroded control rod at Davis Besse) but would rather point out a broader issue raised by your post.
Careful analysis of these four events, in addition to hundreds of others in unrelated fields, shows that the root cause of failures include, and likely are dominated, by human failure at one level or the other. Little fools make little mistakes — but these are generally corrected by the institutional structures put in place to oversee operations. Big fools, on the other hand, occasionally make big mistakes — and these are not so easy to correct through institutional oversight procedures.
Senior management is known to be at least partly to blame for the results at Chernobyl, Fukushima-Daiichi, and Davis Besse. I don’t know the details of Mihama. However, I have reviewed investigations of a wide range of major-consequence events from Space Shuttle crashes to the failure of earthen dams. Root causes of all of these events were dominated by faulty decision-making at senior levels.
So, how can we reduce the frequency of these unhappy events? In the case of nuclear fission energy we are, through experience, coming around to the tentative conclusion that the WORST consequences of a major nuclear plant failure are predominantly financial and social — and NOT the massive death-dealing catastrophes favoured by modern alarmists. Fukushima Daiichi, Chernobyl (with its much larger consequences), and Three Mile Island Unit 2 are excellent illustrations of the weight of this hypothesis.
Financial and social consequences surely can be reduced through good stewardship by senior management and by wise oversight by national and international institutions. We cannot sustain modern societies without large supplies of safe and economical energy. Nuclear fission power can do this job, for millennia.
The future is up to us humans — don’t blame it on the machines.
LikeLike
Elaine Hirsch (should) read Charles Perrow on Normal Accidents.
The lessons of Fukushima from the Bulletin of Atomic Scientists:
LikeLike
Just one little correction, the crane death was at the Fukushima Daini plant, not Fukushima Dai-ichi….
LikeLike
Mike: Charles Perrow made, in a different way, exactly the same point as did Nassim Taleb.
The important point is that massive fatal consequences that might follow from rare-event failure sequences in nuclear power plants are apocryphal, in the sense of “invented”, or “mythical”.
LikeLike
The Black Swan was reviewed rather recently in the Notices of the American Mathematical Society and quite a negative review it was; Taleb might have a modest point to make but he certainly did not demonstrate it in that book.
The US NRC uses probablistic risk assessment for the purposes of estimating NPP safety. According to the chapter “Understanding Risk” by Professor Cohen (Pittsburgh)
http://www.phyast.pitt.edu/~blc/book/chapter8.html
the LLE risk from NPPs in the US is less than the LLE risk from eating peanut butter according the NRC and somewhat great than the risk of eating peanut butter according to UCS. In any case, the risk of accidental electrocution is multiple times higher and yet still heavily dominated, several orders of magnitude, by the LLE risk of Being Poor.
In my life I have known several who have died prematurely from various causes. The second leading cause in my private sample is vehicular homicide. The first is cancer. Those interested in reducing LLE may wish to direct their energies towards leading causes.
LikeLike
Mike: Charles Perrow: Amazon says:
Got that? Sociologist, not nuclear engineer and not physicist. It is innumerate humanitological hubris to agree with a sociologist on the subject of nuclear power. Charles Perrow doesn’t know any more about nuclear power than you (seem to)
Human operators are irrelevant to accidents that are impossible in Generation 3 and 4 reactors. Human operators are irrelevant to accidents that are prevented by the action of electronic control systems in the reactors we have.
HUGH GUSTERSON, who wrote the Bulletin of Atomic Scientists article you reference, is an anthropologist, not nuclear engineer and not physicist.
We Could have replaced our Generation 2 reactors with Gen 3.5 by now if (anti-nuclear protagonists) hadn’t blocked the road. In fact, the control systems have been updated. So far, there have been ZERO casualties caused by American nuclear power plants; an amazing record. So far, there have been 1 Million deaths per year worldwide, from C-O-A-L. Over 200,000 Americans have been killed by COAL.
Something else you (may not know about): Natural Background Radiation. Look it up. Did know that there is natural background radiation ? How else would we date Egyptian mummies with the radioactive carbon they ate thousands of years ago? Of course it is not possible to be exposed to less radiation than the natural background where you live. Here are some natural background readings:
Guarapari, Brazil: 3700 millirem/year
Tamil Nadu, India: 5300 millirem/year
Ramsar, Iran: 8900 to 13200 millirem/year
Denver, Colorado 1000 millirem/year
A not entirely natural reading:
Chernobyl: 490 millirem/year
Some background reading:
http://en.wikipedia.org/wiki/Background_radiation
http://www.unscear.org/unscear/en/publications/2000_1.html
And COAL contains URANIUM and all of Uranium’s decay chain. The truth is, all natural rocks contain most natural elements. Coal is a rock. The average concentration of uranium in coal is 1 or 2 parts per million. Illinois coal contains up to 103 parts per million uranium. Coal also contains the radioactive decay products of uranium. A 1000 million watt coal fired power plant burns 4 million tons of coal each year. If you multiply 4 million tons by 1 part per million, you get 4 tons of uranium. Most of that is U238. About .7% is U235. 4 tons = 8000 pounds. 8000 pounds times .7% = 56 pounds of U235. An average 1 billion watt coal fired power plant puts out 56 to 112 pounds of U235 every year. There are only 2 places the uranium can go: Up the stack or into the cinders and ash. Coal contains so much uranium and thorium that we could get all of the uranium we need from coal cinders and ash.
http://www.ornl.gov/info/ornlreview/rev26-34/text/colmain.html
Calculate your annual radiation dose:
http://www.ans.org/pi/resources/dosechart/
MODERATOR
Your comment has been edited to comply with BNC Comments Policy which requests commenters to “Play the ball not the man”. Sarcasm is discouraged.
LikeLike
Mike’s quotation of Charles Perrow essentially addresses a strawman. “The perfectly safe reactor” is “technoscientific hubris” that does not exist in the industry. The aim is to make failure as unlikely as possible – not impossible – and then make the consequences as limited as possible – not zero. But any “failure” needs to be understood as a continuous measurable outcome, rather than a yes/no point, in this context – how far did the reactors, containment and countermeasures fail and how much did the consequences expand? It is simply shallow philosophy to think otherwise.
I should add that this framework allows learning and improvement from very minor “failures” indeed – a desirable outcome in anyone’s book, I’d think.
And Dan alludes to this too – but I thought it bore repeating.
LikeLike
@ David B B : And may I add , always wear a seatbelt when getting in a car. And wear a helmet on a bike , also make sure to apply sunscreen when working in the sun for longish periods.
LikeLike
unclepete — I put on the seatbelt after I’m sitting down after having entered the vehicle. :-)
LikeLike
Humans take risks equals the Fukushima event. The proposed Jervis Bay reactor was located at close to sea level in a known tsunami zone. There is a big price to pay for persistent stupidity in the case of nuclear rectors.
LikeLike
None of the individual circumstances could be considered a ‘Black Swan’, but I’d argue that Elaine is correct about the synergy of events. Think about another example given – WWI. In retrospect it is difficult to fathom that people could not envisage the effect of mechanised warfare and entrenched positions overseen by automatic gunfire. But they did not see, based on 19th C experience, how these (and other factors) would combine. Come WWII, the situation was totally different, because so much had been learned from the WWI Black Swan.
I would draw a similar conclusion about Fukushima. Almost all of the individual causes and consequences were foreseen in one way or another, but the not the combination (and feedbacks). Poor planning and oversight were clearly also factors, as Dan and Joffan explain, but the Black Swan nature of this event is quite defendable.
LikeLike
The Nuclear power plants @ Fukushima not only survived the earthquake, they started to shut down.Not one death or any injury thus far has been due to that event. All injuries were due to the tsunami. Please see my blog for more details.
LikeLike
“As a result, the event’s non-predictability causes behavioral/psychological changes within people, especially ones who adhere to the scientific method for identifying events.”
(Deleted pejorative) you have no clue of what scientific method is, because what you claim is that those who adhere to it are prone to form an opinion based on magical thinking. Actually it is quite the opposite: the lack of knowledge in probability and statistics makes easier it (deleted pejorative) to overrate the anecdotical events.
LikeLike
Mike:
While it may be true that reactors can be tightly-coupled complex systems, this is not illustrated in the Fukushima accident. It is well understood that if you deny cooling to a recently shutdown reactor for a protracted period, fuel damage and eventually melting will occur; that result is imminently predictable. We don’t need an accurate, detailed representation of the thermal-hydraulics to know this.
On the other hand, I think that TMI and Chernobyl could be given as examples to make Perrow’s point. In TMI, the operators misinterpreted some readings and didn’t recognize that those readings implied that the PORVs were open. The Chernobyl accident resulted from a power excursion due to a feedback effect in an unusual operating regime that wasn’t understood by the operator. Both accidents proceeded in part because of complex effects that were not understood by the operators. I don’t understand why Perrow thinks that this is true of Fukushima.
Barry and Elaine:
It’s not clear to me what aspect of the Fukushima accident qualifies it as a black swan event. As I pointed out above, the risk of fuel damage and melting from extended station blackout events is well known. The earthquake and tsunami can be regarded as rare events (or even very rare events), but neither was unpredictable in any sense. The consequences of inundating the EDGs were predictable.
I suspect that if someone had hypothetically described the accident progression to a Fukushima Daiichi engineer prior to the accident, he would have said that an earthquake and tsunami that will exceed the design basis is unlikely, but if it caused the plant to be in a lengthy station blackout, fuel damage was likely and the longer the plant was in that state the more likely severe and complete melting becomes. The error in his analysis, if there is one, would be in estimating the likelihood of the initiating event, not predicting the consequences.
LikeLike
IMHO ‘Black Swan’ is an inappropriate term to apply to the nuclear power industry.
All reactors have a non-zero probabilistic risk analysis done in the design phase.
One Gen II nuclear reactor catastrophically failing in 10,000 reactor years is not significantly outside the design risk probabilities.
The only thing that was ‘unpredictable’ was which of the 400 operating nuclear plants would be subject to a series of events that would result in catastrophic failure.
In the US residential natural gas fires were responsible for an average of 43 deaths per year in the period of 2000-2004.
Click to access GasHomesFactSheet.pdf
All sources of energy carry a non-zero risk of causing harm.
It would only be a ‘Black Swan’ is we fooled ourselves into believing the risks were zero.
LikeLike
The cause of the Fukushima accident was prolonged station blackout. If there had been mobile diesel generators on flatbed trucks stationed nearby but away from the tsunami zone, the disaster could have been avoided. There was a lack of planning; the analysis is quite simple, and the short-term solution is straightforward. Of course the long-term solution is to replace aging reactors with modern designs.
LikeLike
Barry:
“I would draw a similar conclusion about Fukushima. Almost all of the individual causes and consequences were foreseen in one way or another, but the not the combination (and feedbacks).”
Really? Was the height of the tsunami foreseen? This was a nuke plant operating on the most tsunami-prone coast in the world – they should (and could) have built the thing to run reliably underwater for a prolonged period of time.
LikeLike
The Black Swan idea may be applicable but I don’t think its adequate to explain the disproportionate size of the nuclear fear. I’d put that down to excellent advertising and great imagery from the anti-nuke camp. I sucked me in for more decades than I care to admit.
I came around a corner on my bicycle last week to find a car stopped in the middle of the road and a woman flagging me down. She was terrified (not paralysed with fear, but very very scared). There was a spider in her car. A non-venomous harmless, largish spider. For some strange reason, telling her the spider was harmless had no effect (she knew that anyway). This I think comes far closer to capturing what is happening with nuclear power that the black swan concept.
LikeLike
When I think of the “Black Swan” phenomenon, I think of Thomas Kuhn’s The Structure of Scientific Revolutions in which the underlying scientific paradigm is fundamentally shifted, such as the Copernican Revolution, the emergence of quantum physics, etc.
Why do we happily drive alongside a fully laden LPG tanker which shares the road with idiots, or sit in an A380, of which nearly half the take-off weight is jet fuel, yet worry so much about the most regulated industry in the world? Geoff’s arachnophobia example clearly isn’t a “Black Swan” event – why is a deep distrust and fear of nuclear any different?
LikeLike
Geoff Russell, @ 2 February 2012 at 8:09 AM
Spot on. On this I agree with you 100%.
LikeLike
If you are a hammer, everything looks like a nail. If you are a sociologist, every technological accident looks like a completely human one. Sadly, I very much doubt that Charles Perrow understands how a nuclear plant works, he has no in depth knowledge of nuclear technology. His ordeal is disappointingly predictably wrong.
The fact is, that Fukushima simply had inadequate design basis. 6 meter tsunami protection in a region that fairly routinely gets >10 meter tsunamis.
So in the case of Fukushima I’d suggest that the engineers are correct. Had Fukushima used 20 meter sea walls or if it was simply built in the upland, it would have been fine (exemplified by the fact that the equipment worked fine even after the earthquake – the flooding by the tsunami did it in).
LikeLike
Geoff Russell, on 2 February 2012 at 8:09 AM said:
I’d put that down to excellent advertising and great imagery from the anti-nuke camp
There was also some great imagery from the US Department of Defense and their Soviet counterparts.
IMHO Anti-nuclear sentiment can be explained by the phenomenon that some people are perfectly content to drive a motorcycle without any safety gear but fear flying in an airplane.
The airplane is orders of magnitudes safer then the motorcycle, but a passenger in an airplane has absolutely no control over the sequence of events.
LikeLike
I tend to agree with those arguing against the Black Swan event as applied to nuclear. All accidents are preventable. Especially Fukushima where there were reports given to TEPCO over 3 decades that a huge earth quake could occur and, the sea wall was inadequate.
If we believe that it was a Black Swan event, then there is nothing we could of done. I don’t buy this for a New York second. The sea wall should of been built to *maximum* height, not minimum height. The diesel fuel tanks should of been located behind the plant, not right on the intake structure making them vulnerable to tidal wave.
If there was hubris, it was not in the tech, it was in it’s deployment, Black Swan or no Black Swan.
Secondly, politically, that is people’s fear driven by the internet and by anti-nukes poising as experts: A Black Swan event can’t be explained to make it seem that context (nuclear energy) is still safe. It doesn’t matter about the facts, only the truth, the former being objective, the latter being subjective but not necessarily false.
The fear generated by Fukushima is simply impossible to deal with except within our own pro-energy, pro-development, climate change activist community. This is a very small community. We can only try, Black Sw. or not.
LikeLike
David Walters
Especially Fukushima where there were reports given to TEPCO over 3 decades that a huge earth quake could occur and, the sea wall was inadequate.
The fact that something ‘could occur’ and the probability of that something occurring being substantially high enough to warrant action are two different discussions.
A few decades ago a meteor crashed thru the roof of the house across the street from my mother.
Obviously, in hindsight for that family the choice not to have a ‘meteor proof’ roof was a poor one. To my knowledge, no building code in the world requires ‘meteor proof roofs’.
My point is that in ‘all things human’ we set a standard of ‘acceptable risk’.
LikeLike
I agree with the thrust of the post. I have a few comments:
I’d suggest explaining that an accident was human error doesn’t help to reduce public fear. Their concern is ‘in that case, we’ll always have nuclear accidents because there are always humans involved’.
“likelihood of nuclear accidents” – I suggest the emphasis should be on providing more information about the consequence rather than the likelihood. I think people are most misinformed about the consequence rather than the likelihood (see expansion of this point below).
“More information” – definitely. This is the key. But it needs to be broadcast and encouraged by government, and backed by clearly supportive government policy, to give it credibility.
“about the true causes of recent nuclear accidents” – I suggest that approach may not get the desired result. The response is – ‘there will always be another unforseen reason for an accident’. Explaining the causes reinforces the (correct) view that systems aren’t perfect, humans make mistakes, they always will, so there will always be accidents, no matter how rare. What I suggest we need is a balanced, objective assessment and explanation of the consequences. For example, I’d like to know if the exclusion zones around Chernobyl and Fukushima are really justified on the basis of projected latent consequences. How does the consequences of the contamination in these two areas compare with the consequences of chemical contamination in many other parts of the world (e.g. Ruhr Valley, Sydney). We live in all these contaminated areas and we live with the consequences. So is the decision to maintain these two exclusions zones supported on an objective basis?
This raises the question: ‘why is the author comparing latent fatalities for nuclear with immediate fatalities for fossil fuels?’ Why isn’t the comparison of like with like? I understand why the author has done this, but many people wont. I’d suggest presenting both sets of figures (immediate and latent) for both nuclear and for the fossil fuel chains (and for renewables). And of course, the figures should be for generating electricity only; they should not include the fatalities related to the other uses of fossil fuels.
It is not fair or balanced to exclude the crane and drowning deaths at Fukushima from the accidents in the accidents attributable to nuclear generation, because we include the same sorts of accidents for solar, wind, coal, gas etc. The fatalities for wind are not due to the wind blowing (i.e. fuel related). They are due to the fatalities in the mining, processing, manufacturing, fabrication, construction, maintenance, decommissioning of the plants and the transport of materials between all the steps. Likewise for solar. For fossil fuels many of the deaths are in construction, O&M and steam failures too. We should not separate off the fatalities that are not caused by radiation. These must be included. The various ways of generating electricity must be compared on a properly comparable basis.
LikeLike
No, it’s not the same. A 1 in a 100 year incident/aprox (for large earthquake/tsunami) is not the same as your example. That’s the point. If you said “ah, so, Indian Point Power Plant (the one 50 miles up the Hudson River from NY) MIGHT get hit by a tsunami” then yes, your comparison would hold.
Fukushima and the entire west coast of Japan has a much higher, a serious “non-black swan” chance of something bad happening like it did, than either the Indian Point or your house mishaps occurring.
It seems if I were to follow your view, then nothing is learned from Fukushima. I think there is a LOT to learn and it’s needs to be implemented, such as relocation of aux. diesel storage plant, their hardening against quakes; hardening of reactor cooling pump booter pumps and circulating water pumps; better temporary power solutions, etc etc.
If this were really a “Black Swan” event, we wouldn’t be having this conversation. It would be like the meteor hit on your house.
LikeLike
David Walters — The fundamantal problem at Fukushima was that (almost) all seismologists thought, before the event, that the Tohoku fault was not capable of generating more than a moment magnitude 8.0 earthquake. They underestimated by a factor of 32 in energy released.
MODERATOR
Please supply a supporting link to your statement. Thank you.
LikeLike
David Walters, on 4 February 2012 at 8:21 AM said:
Fukushima and the entire west coast of Japan has a much higher, a serious “non-black swan” chance of something bad happening like it did
Twenty thousand people lost there lives in what some are calling a ‘high probability’ event in Fukushima.
Given the number of casualties caused by the Tsunami it would appear that the risk the operators of the nuclear power plant perceived in relation to Tsunami’s was similar to the risk perceived by the broader local community.
Otherwise they would not only have had a high seawall protecting the nuclear power plant but they would also have had a seawall to protect the community as a whole.
In the community of Wethersfield, CT meteorites crashing thru roofs are not ‘black swan’ events. There have been 13 meteorites known to hit a structure in the US in the last 100 years. 2 of those occured in Wethersfield, CT about 10 years apart.
Click to access Cassidy.pdf
Nobody in Wethersfield, CT has had their roof reinforced to protect against meteorites even though Wethersfield, CT is a ‘meteor magnet’.
LikeLike
Ms. Moderator — As with the Richter scale, an increase of one step on this logarithmic scale corresponds to a 10^(1.5) ≈ 32 times increase in the amount of energy released, …
http://en.wikipedia.org/wiki/Moment_magnitude_scale
and the Tohoku-Oki earthquake was moment magnitude 9.0 with quite unexpected characteristics:
http://www.sciencedaily.com/releases/2011/05/110519141622.htm
In particular, seismologists seem to have expected a moment magnitude 8.0 (or just possibly a bit more depending upon the siesmologist):
http://www.latimes.com/news/nationworld/world/la-sci-japan-earthquake-20110310,0,2161671.story
Occurring 373 km (232 mi) northeast of Tokyo, the Tōhoku earthquake came as a surprise to seismologists, since the Japan Trench was known for creating large quakes, but was not expected to generate quakes above an 8.0 magnitude.[309][310]
http://en.wikipedia.org/wiki/2011_T%C5%8Dhoku_earthquake_and_tsunami
MODERATOR
Thank you DBB for your most comprehensive references:)
LikeLike