March 20, 2011
Tepco Slow On Reactor Decisions After Earthquake

The Wall Street Journal has an excellent article outlining some of the mistakes made in the aftermath of the earthquake and tsunami. The top management of Tokyo Electric Power Company, (Tepco) which operates the Fukushima reactor, was too slow to accept the necessity of drastic measures.

TOKYO—Crucial efforts to tame Japan's crippled nuclear plant were delayed by concerns over damaging valuable power assets and by initial passivity on the part of the government, people familiar with the situation said, offering new insight into the management of the crisis.

Tepco did not want to lose the reactors as productive assets. Therefore Tepco hesitated too long to inject sea water whose salt would corrode the reactors so much as to make them unusable in the future. Tepco also initially had very poor communication with the reactor site due to communications damage from the earthquake and tsunami.

"This disaster is 60% man-made," said one government official.

Workers at the reactor site also made mistakes such as running out of fuel for a pump and setting a valve to a wrong position. Not surprising given the pressure they were working under.

We should not be complacent that most nuclear reactors aren't situated near offshore subduction zones. Nukes have plenty of ways to fail. Even when working at a more sedate pace some of the mistakes made by nuclear plant operators do not inspire confidence. For almost 18 months Diablo Canyon's emergency cooling water valves were broken without the awareness of the plant operators.

At the Diablo Canyon nuclear plant, operators found themselves unable to open the valves that provide emergency cooling water to the reactor core and containment vessel, during a test on October 22, 2009.

A misguided fix of an earlier problem had prevented the emergency valves from opening, the NRC team sent to investigate found.

With nuclear reactors capable of going very wrong very quickly what's needed? Human decision loops that are fast enough. There's an analogy here with legendary fighter pilot John Boyd's OODA loop (Observe, Orient, Decide, and Act). The problem with nukes is that for months, years, even decades nothing might happen that requires an extremely fast decision loop. But suddenly the calamity strikes (earthquake followed by tsunami that knocks out back-up generators and communications) and people accustomed to a far more sedate pace of decision-making can't speed up fast enough. Even worse, the kinds of people who love to make fast decisions and who are good at making fast decisions (e.g. fighter pilots) won't take jobs that require them to spend decades staring at nuclear power plant consoles waiting for something to happen.

One of the challenges is to know when Business As Usual has ended and fast extreme decisions are required. With the Three Mile Island reactor incident there was not so dramatic a starting point as a 9.0 earthquake that lasted for 5 minutes followed by a tsunami. The TMI guys weren't as abruptly shaken into a heightened state. The Fukushima incident did have 2 big wake-up calls before the reactors started overheating. So the Fukushima reactor operators and their managers higher up in Tepco at least had big prods toward getting into faster decision loops. But decades of complacency left them ill-equipped to step it up.

Another need: A bigger tool box for dealing with nuclear reactors gone awry. Off-site from any reactor many tools should be available for very rapid deployment. For example:

  • Portable large electric power generators and water pumps transportable by helicopter or truck.
  • Capability to run power lines over miles in well less than a day. Lost local power? Bring in power from somewhere else in a hurry. It is a solvable problem. Might require choppers with crews to deposit towers and run lines across farm fields.
  • Lots of satellite phones and other communications gear deployable wherever communications have been lost.
  • Robots that can substitute for humans in many more tasks such as for bringing fire hoses to bear, for looking into reactors and other radioactively hot areas of a site, and for patching holes to fix spent fuel cooling pools that spring a leak.
  • Robotic autonomous vehicles, both on the ground and in the air.
  • Rapidly deployable sensor nets. Find out what is going on in areas where sensors have been disabled or destroyed.
  • Better tools for protecting humans from radiation. Lead-coated vehicles, lead-covered small control rooms within reactor complexes, portable HEPA filters for working in areas with contaminated air, lead chairs, lead-lined clothing, and other tools for protecting humans.

Most these tools do not need to be located at every single reactor site. They just have to be transportable to reactor sites within hours of the start of an incident. The deeper need is an acceptance that things will go wrong (take off those Panglossian rose-colored sunglasses) and therefore additional tools and capabilities should be available for deployment. We need a strong commitment by the nuclear power industry and governments to develop the tools to handle each failure. The next nuclear accident should not require heroic workers getting themselves radioactively damaged. The response time and tools available should allow problems to get stopped at much earlier stages and at far less cost in assets and lives.

There are upsides to the development of the sort of tool sets I describe above: The tools would have other uses. For example, robotic firefighting equipment can save human lives in conventional fires and robotic vehicles can work in other types of disaster zones.

Update: Also see an opinion piece by Christopher Stephens in the WSJ about regulatory oversight in the nuclear power industry.

Share |      Randall Parker, 2011 March 20 04:18 PM  Energy Nuclear

AMac said at March 20, 2011 7:06 PM:

These are all good ideas.

In retrospect, a big problem is that the Defense-in-Depth approach to design wasn't taken far enough.

If the reactor is scrammed and the ECCS is knocked out for a protracted time and power is unavailable onsite for days... would you rather contaminate the ocean or the atmosphere with highly-radioactive steam? "Ocean" is the less-bad answer, and it would certainly have been possible to provide for underwater venting, but it seems that both eventualities were considered so terrible and so unlikely that no provisions were made. Similarly with the choice between venting steam inside the building (and risking a hydrogen explosion), or discharging it directly to the atmosphere (with greater prompt dispersal of high-activity radionuclides). In the end, venting to the building without a way to burn off the hydrogen gave them the worst of both worlds, plus leaving engineers to cope with wreckage strewn around and atop the spent fuel pools.

On emergency planning -- complacency and screw-ups are inevitable. I think one of the most important things is to integrate emergency equipment and procedures into everyday activities. Dedicated emergency satellite phones will have dead batteries and emergency diesels will have clogged fuel filters, unless they are regularly used.

It seems to me that a HAM radio onsite would have been very useful, March 11 to 13. If they could transmit, somebody somewhere would listen, and they could have called TEPCO HQ in Tokyo, long distance. "Hey, I've been talking to your operators, did you know that the land lines are cut and the cell networks are down? By the way, all the backup generators were washed away by a giant wave..."

Fat Man said at March 20, 2011 8:44 PM:

I assume that at an appropriate time when this incident has been wrestled to the ground and hogtied, there will be an inquest. We can hope that the Japanese authorities will include disinterested foreign experts from IAEA and USNRC, inter alia, in the inquest. When the inquest has been made and when it publishes its report then we will know what happened and why, and what additional precautions should be taken. Until then there is little we can confidently say, except that the initial reports are seldom correct.

PacRim Jim said at March 20, 2011 9:35 PM:

Why are these reactors not controlled by AI?

Randall Parker said at March 20, 2011 11:05 PM:

PacRim Jim,

Because AI does not exist yet?

morpheus said at March 21, 2011 8:20 AM:

even a 5 year old can come up with ideas to improve nuke plants safety

im sure he will mention robots like in the list

u could design a nuke reactor that wont melt down even if tryd ur best

but that would cost couple extra millions

so like always greed and fallow the money

prime example:

tzunami wall around plant 6.3 meters wawe 7 meters so for a couple 1000 dollar savings on cement

the loss is in the billions not to mention human deaths caused by it

great posts again randall


Chris T said at March 21, 2011 9:56 AM:

Portable large electric power generators and water pumps transportable by helicopter or truck.

And for Christ's sake, make sure you can connect them!!!

Steve Sailer said at March 21, 2011 7:15 PM:

Another thing to keep in mind is that when the Big One hits, nuclear power plant workers and executives will have lots of personal distractions, like surviving themselves, concern for loved ones who might be missing, injured, or dead, finding enough food, mourning, and depression.

I know a guy who was so shook up by the 1994 Northridge quake and its aftershocks, that he jumped in his car and drove to Santa Barbara for four nights. That seems pretty understandable, but his wife and kids didn't much like that he was so freaked out he disappeared without them.

PacRim Jim said at March 21, 2011 7:59 PM:

Strong AI doesn't exist, but AI capable of reacting more quickly than humans certainly does.

The surfaces of the latest USAF planes are controlled by software, because humans are too slow and too limited.

The same comment applies to cheezy sci-fi potboilers, where humans are attempting to combat swarms of alien ships.

Randall Parker said at March 21, 2011 10:26 PM:


So it is really necessary to bring in people not near the disaster to work in important facilities. So, for example, workers from more distant nuclear reactors should have immediately been choppered in to Fukushima.

Lono said at March 22, 2011 7:46 AM:

I blame the Emperor!

No - seriously - I do - what the heck is his job if to not think about the sustainability of the Empire - and he hasn't even really had to consider the dangers of external or internal violent conflict for over 60 years because they are essentially a U.S. protectorate ever since we wiped the floor with them in WWII.

WTF Emperor!??!? - you SUCK!!!!

John said at March 22, 2011 11:12 AM:

Radiation damages solid state electronics faster than human biology in lots of cases, so robots aren't the easy solution everyone is assuming. A few early attempts at Chernobyl had issues when the robot would fail in a critical location becoming a road block, or get locked up with a grappler holding something highly radioactive, etc. placing people in even more danger of having the original problem, plus a non-functional robot in the way.

Just another joe schmoe said at March 22, 2011 11:44 AM:

I am a former navy nuke and a big proponent of nuclear power. I have seen many of my comrades, who I view as some of the smartest and most trustworthy of brethren, enter the civilian workforce and become corrupted. In the navy, we believed in integrity, honesty, and above all, safety. We were constantly checking systems, logging statuses, verifying outputs and inputs, etc. When they get to the civilian workforce, they realize that they must part with some of that dedication, knowing their career is affected not by how safe the plant is, but rather, how much power their shift can produce and for how long. Let someone else worry about safety. Admiral Rickover (father of navy nuclear propulsion) said that the two most dangerous times in a reactor life-cycle are when it is shutdown and when it is in an extended period of steady state. He went on to say that the danger is not in the power plant itself, but rather, the complacency of the operators to believe that steady state means safe. I was a reactor operator on a trident missile submarine. after about three weeks of not showing any change in power level, I too could have fallen into that trap. The navy, though, required us to perform maintenance, and although a pain in the butt when needing sleep, it was a constant reminder that things can and do go wrong and are never known until you check. I sometimes laugh when I see news reports when a reporter goes to a plant and, when discussing safety, the plant owners show off this multimillion dollar training facility. What a joke. Yes, training is important, but it does not and should never substitute the need to get out of the air conditioning and verify EVERYTHING and assume NOTHING. Every person associated with that plant, from the plant manager to the janitor, should take ownership and feel the same responsibility. EVERYONE.

I end my long post with an acknowledgment that there are good, honest, and decent people that work everyday at power plants around the world. they are trying their best. My beef is really with the one or two (typically smaller and less capitalized) power generation companies that cut corners on a regular basis.

Nuclear Physicist said at March 22, 2011 1:15 PM:

With the current generation of reactors that are designed to use passive cooling following an earthquake like the one in Japan, no active pumping of coolant is needed to keep a reactor core cool. People should keep in mind that the Japanese reactors are 50 year old designs that will never be built again. I'm sure that work will be done to improve emergency response planning for potential accidents in the older designs. But, we shouldn't overreact to what is happening in Japan. Earthquakes of this magnitude have a return period of less than 1 in 1000 years. So, they aren't likely to occur again for a very long time. Therefore, the most likely outcome is that we will have replaced all of the older designs with passively cooled reactors before we need to worry about another beyond design basis earthquake. Even with these older reactors, the likely outcome is zero acute fatalities with a small number of excess cancers somewhere in the future and most of these cancers will likely be curable. The most significant loss from the Japanese reactor plant accident is economic rather than a human death toll. Beside that, in an environment where the earthquake/tsunami death figures are now north of 22,000 people, I find the hysteria over the reactor accident to be a bit ridiculous.

Chris T said at March 22, 2011 1:37 PM:

But, we shouldn't overreact to what is happening in Japan. Earthquakes of this magnitude have a return period of less than 1 in 1000 years. So, they aren't likely to occur again for a very long time. Therefore, the most likely outcome is that we will have replaced all of the older designs with passively cooled reactors before we need to worry about another beyond design basis earthquake.

An earthquake along that specific region of the fault is unlikely for quite some time, but the tsunami was ultimately what crippled the plant. Tsunamis can be generated from another fault even at a considerable distance. Thus we still do need to upgrade protection for existing older designs. A better emergency response plan should also be worked out for all existing and future plants based on lessons from this earthquake.

Joe said at March 22, 2011 1:48 PM:

The Japanese reactors are the Mark I containments, and these reactor designs use a water filled suppression pool to quench steam from the reactor, and to provide emergency core cooling. The torus in the Mark I containments needed much analysis and redesign to comply with seismic standards in the US, and I suspect, in Japan too. That being said, I suspect, that the Japanese operators were too slow to utilize the Automatic Depressurization System (ADS).

The ADS rapidly drops the reactor pressure so that the low pressure coolant injection (LPCI) pumps can pump in very large volumes of water to flood the core (LPCI is the mother of all coolant injection). In using the ADS, the dyammic forces of the steam/water suddenly venting to the torus places enormous stresses on the structures. I suspect the Japanese operators were trained to utilize RCIC and HPCI systems to cool and depressurize the reactor as their first choice, and to avoid using ADS unless absolutely necessary, due to concerns with the structural response of the torus.

In the course of holding the reactor with RCIC/HPCI after the intial SCRAM, the tsunami came and knocked ou their diesels, and they probably lost the option to use ADS and flood up with LPCI. Once they lost the option of using ADS and LPCI, I imagine the decision making paralysis played a part in the rest of the story. They may have waited too long to decide to use ADS/LPCI, and the choice became moot. I'll bet we'll see revised "guidance" on use of ADS with Mark I containments.

BTW, steam venting into the reactor building in and of itself was not the source of hydrogen that blew apart the secondary containment shells - the hydrogen was most likely from steam interacting with the zircaloy fuel rod cladding (as a result of the fuel coming uncovered). Since they had no power to spark off excess hydrogen before it got to explosive levels, the result was BOOM.

My two cents - the facts and sequence of events will come out in a few months.

Dyspeptic Curmudgeon said at March 22, 2011 1:52 PM:

What REALLY surprises me, is the lack of 'what next' planning exhibited at the reactors. The probability of the failure modes exhibited was low, but the absolute certainty of what would happen next was never addressed.

So yes, the reactors were designed for an 8.2 quake and a 6.3 meter wave and survived a 9.0 + 7.0 combo. Sort of. But no-one ever bothered, it appears, to think about 'what then'. (Maybe they figured, 'Ok! And then we're dead'!) If the pumps failed for whatever reasons, then there would be heat problems. They knew they could reduce pressure by venting. But did no-one ever consider what the hell would happen if hydrogen built up in the 'containment' building. Of COURSE it would eventually explode. So why wasn't there a proper vent stack with a pilot light at the top, just like on an oil rig? (And condensation plates to trap and return the condensed, radioactively dirty steam.)

Regarding the fuel storage cells: yes, 20 meters in the air is kinda dumb, especially as there was/is a *certainty* of cracks in the concrete from an 8.2+ earthquake. And i don't think that pool has a plastic lining! It should have been inset into the ground. It would then not have needed tons of extra concrete to hold the entire pool 20 meters above grade, only some extra long hoisting cable. In fact, that end of the building would have needed no massive enclosure at all.

And we really do have to ask Homer Simpson, why there is no externally available standpipe to feed water directly to the pool. The present problem is that the damn pool is 20 meters in the air, and there is no direct method of injecting water. Funny that every office building you walk past has an external 'siamese' connection to allow the fire department to *feed water to the sprinkler system*. For a nuclear plant, maybe the end of the standpipe should be a half mile from the building, behind a large berm, and the standpipe should be a semi-rigid continuous plastic pipe, for just this situation. Similarly, there should be electrical connections directly into the plant system, at that remote site, so that generators and water pumps can be brought in (helicopter if required) and connected.

What is surprising in one sense is that these are 40 year old reactors, and NO-ONE seems to have thought about this. It's like the Operating Manual has had the back half ripped out.

I suspect that none of these plant builders ever hired an ex US-Navy reactor handler, since everything I have read about their training leads me to believe that the largest part of their training essentially consists of answering the question 'and what do you do if that doesn't work' or 'what if that is not available? Think fast! Improvise!' while being sprayed with cold water or burnt, or both at once. Someone who had thought about the matter would soon recognize that minor additions at the construction stage would be priceless if the certainty of that result ever arose from the improbable cause.

That is where we are now: the situation was perfectly certain to arise, if all the power went off.....

Fortunately, it is now never likely to happen again. Human beings may exhibit certain forms of wilful stupidity, but we are reasonably good at learning from our mistakes.

Nuclear Physicist said at March 22, 2011 2:33 PM:

In response to some of the above comments:

1. For most existing reactors, a big tsunami that disables reactor safety systems is not something that is possible. The Japanese reactors survived the earthquake. Unfortunately, the Japanese designers didn't anticipate the magnitude of the resultant tsunami, which is what did most of the damage to the safety systems. So, we all learned something new that will help us upgrade reactors that are vulnerable to tsunami damage. But, we shouldn't be too quick to criticize the designers for this error. Hindsight is always 20-20.

2. The Japanese weren't able to use the ADS/LPSI systems because they failed, not for lack of trying.

3. We don't really know if the fuel pool is cracked, but even if it were, placing it at ground level wouldn't ensure that the water couldn't leak out. In fact, the case could be made that an in-ground fuel pool would have experienced more sheering stress from the earthquake. So, placing it 20 meters in the air is not necessarily a bad thing.

4. The Japanese reactor design did include systems for burning off the hydrogen, but they failed to ignite.

5. Disparaging the Japanese designers and operators as being willfully stupid is a bit harsh. Unless you've been personally involved in complex design and stressful post-accident operating environments like those experienced following the Japanese earthquake, you really should show a little more respect.

6. It was ex-navy nukes at TMI shutting off the emergency core cooling systems that lead to the accident at that reactor. So, don't assume that navy nuke knowledge will save you.

M. Simon said at March 22, 2011 3:39 PM:


I didn't need a WSJ article to tell me this. I saw it by day four at the latest and was commenting on it. It was obvious that they were behind the curve. And my guess was their mindset was "preserve assets" rather than "prevent disaster". Which is why I think at this time that commercial entities are not a good match for nuclear power.

M. Simon said at March 22, 2011 4:05 PM:

5. Disparaging the Japanese designers and operators as being willfully stupid is a bit harsh. Unless you've been personally involved in complex design and stressful post-accident operating environments like those experienced following the Japanese earthquake, you really should show a little more respect.

I was involved in an emergency restart (something commercial plants are not allowed) in a war zone. You just follow procedures. I was sent forward to double check that the emergency generator was operating properly. This was despite the fact that we had indications everything was fine. If the restart is not done very carefully you can get in a very bad accident situation. And for this type "accident" you have to restart the reactor in 30 minutes or less. We did fine. Because we followed our training.

6. It was ex-navy nukes at TMI shutting off the emergency core cooling systems that lead to the accident at that reactor. So, don't assume that navy nuke knowledge will save you.

When I first found that out I was so disappointed because the design defect that "caused" TMI was studied in nuc power school ('65 for me). We were trained that at a TMI type plant (B&W) the core water level indicators would read inverse to what was actually happening (not just TMI in fact). And with the light bulb burned out you might not connect to that fact.

I was so sure that Naval Nukes would have done better that I wrote a letter to that effect to Popular Science that got published. I guess some guys were not as interested in their studies as I was. Hard to believe. It is such an interesting subject.

M. Simon said at March 22, 2011 4:15 PM:

The problem with the spent fuels above the reactor is that if the pool fails you can't service the reactor and if the reactor fails you can't service the pool. Dumb design. And no pool liner? Criminal.

Bernie said at March 22, 2011 4:35 PM:

I'm very disappointed in this post. The analysis has too much speculation and a lack of knowledge about how nuclear power plants are operated and the engineering concerns associated with them.

You assume a reason for delays in adding seawater to the core without any discussion about technical considerations that would lead to that decision. The facts of the event are still not clear, however a very basic timeline we know so far is that the plants were shutdown during or immediately after the earthquake. The emergency diesels functioned properly, but at some time later (an hour?) they were disabled by the tsunami. Without power, how do you add water to a pressurized system? There is a decision needed to depressurize and use fire water pumps to add sea water, or to continue using the steam driven high pressure pumps to recirculate water. Imperfect information about how long it would take to get the diesels back on line, doesn't make that an easy decision. It gets worse, the steam powered pumps are recirculating water without any ability to cool it, giving you a limited time that they will be available, however seawater, when it boils away leaves a coating of salt and other particulate on the fuel cells. This is bad because the coating reduces the heat transfer, and restricts the flow of cooling water. With perfect information about the time it would take to restore power its obvious they should have depressurized sooner, but without any information on the initial damage reports I would not question their motives. That is insulting and uncalled for.

Your ideas on decision making could benefit from a little understanding of hoe these plants are operated. The operators spend one week out of every five in a simulator running casualty drills. 20% of the time at work is spent practicing for these events in a simulator that is identical to the main control room. If you knew anything about human performance in crisis situations you would understand that there are far more errors caused by acting in haste, then by slow deliberate actions, and there are far more ways to do something wrong then to get it right. A lot of effort is made to train people to think prior to acting and to have a second person check to make sure no incorrect action is taken.

Please take some time to learn about nuclear operations next time. It's sad because your posts are much better when you know what your talking about.

Joe said at March 22, 2011 5:32 PM:


You say that they "weren't able to use the ADS/LPSI systems because they failed, not for lack of trying."

From the events I've seen, they didn't try ADS/LPSI, as they were still at high pressure when the diesels failed. Maybe I missed something, but I didn't think you could run the LPSI pumps on inverter power, and I've seen no indication they activated the ADS. Not saying you're wrong or anything, just trying to piece together the events.

Joseph Somsel said at March 22, 2011 7:20 PM:

I'm particularly unimpressed with this article and agree with Bernie above. I'm a nuclear engineer with decades of experience in reactor safety and I see little knowledge or wisdom behind this trivial piece.

As to artificial intelligence, I lead the industry research efforts on the subject in the 80's. We did get one good product out of it, an alarm processor which will be standard on new plants. We can have over 10,000 individual alarms points - certain patterns of alarms point to predictable circumstances. However, AI remains DEDUCTIVE and purely logical. When events get the plant this topsey-turvy, INDUCTION is what's needed, along with INVENTION. These are not computer strong-points.

Some of the suggestions are pretty obvious but the core facts are that a HUGE event hit Japan that overwhelmed all social and government capabilites. The nuclear power plants seem solid relative to the rest of the affected areas.

We'll have extensive and intensive post mortums on the events. Let's not rush to judgment. Also, Japanese management and operations differs considerably from American practices.

Randall Parker said at March 23, 2011 12:21 AM:

Nuclear Physicist,

If the reactor designers couldn't have foreseen a tsunami this large in 1970 and still couldn't foresee a tsunami this large in 2010 (since they obviously didn't upgrade the reactors last year for this sort of accident) it raises the question of what else haven't they foreseen? The nuclear power industry tells us how safe nuclear reactors are. Well, under the circumstances a question arises: how can they know?

But was the tsunami risk really not foreseeable on what is called the ring of fire? I do not get that. If they only planned for the sort of geological events they knew about then that's very imprudent. Our geological history knowledge is far from complete and the omissions from our knowledge very likely include more severe events that are as yet undiscovered. Surprises like the Indonesian tsunami teach a lesson: Our knowledge is incomplete and in a bad way it is incomplete.

Imagine the 1811 and 1812 New Madrid quakes hadn't happened yet and nuclear reactor designers decided they didn't have to deisgn to handle such events for nuclear reactors in Missouri or Arkansas. That'd be more understandable (though in serious error). In Japan's case the subduction zone's existence is known. They either shouldn't have put reactors on the east coast of Japan or they should have put in bigger margins of safety due to their limited knowledge.

You seem to imply that post-disaster environments are too stressful to expect high quality decisions out of plant operators (and that seems quite plausible). Is this a newly formed belief? I would think not. Shouldn't the nuclear power industry already have practices in place to handle this problem?


People in the Japanese government made clear that economic and not technical reasons were the cause of the delay for adding seawater. Are they lying or misinformed? If the report is correct then the Prime Minister of Japan overruled Tepco and ordered the seawater. I'm quoting from the rather pro-capitalism Wall Street Journal that doesn't take a knee jerk anti-management stance. I would expect them to get this close to right.

Insulting and uncalled for: Big things went wrong. The stakes are high. I lost sympathy for the Tepco management when they were slow to relay information about what was really going on. I understand well the knee jerk reaction of management to not share bad information with the public or with customers. I've sat in the room and heard heard the internal debates in other industries when the stakes were much lower.

I've also read quotes from nuclear experts advising the US government last week to tell the Japanese government to react much more rapidly and aggressively. Are these guys wrong?

That Japanese management practices differ: Sure. Are you trying to say their management practices cause them to deal less well with a disaster of this sort? If so, then that ought to weigh in the equation when designing and siting reactors.


I get that some of you work in nuclear power and you might be feeling defensive. But the rest of us have to live with what the nuclear power industry does, that it is taking risks we do not understand, that we do not even know how well it understands those risks, and that human mistakes and human incentives to cut costs will add more risks.

I'm reminded of what Richard Feynman found when he investigated the Challenger space shuttle failure: That at each higher level of management the perceived risk of a shuttle failure went way down (an order of magnitude if memory serves). I doubt the nuclear power industry is immune to this tendency in management layers and the poor decisions this tendency causes.

M. Simon said at March 23, 2011 4:30 AM:

I doubt the nuclear power industry is immune to this tendency in management layers and the poor decisions this tendency causes.

The US Navy fixed that to some extent: you couldn't command a Nuke ship (boat) unless you were nuke qualified. Maybe that ought to be industry practice as well. And my experience was that the Command ranks studying nuke power didn't get any breaks re: difficulty of material. Why? Well if you screwed up your turn on the board it would reflect poorly on the Navy.

AMac said at March 23, 2011 6:55 AM:

The height of the tsunami at Fukushima Daiichi was originally reported as 7 meters, compared to a design specification of 6.3 meters. That never sounded right, given the damage. Now the tsunami's height is claimed to have been 10 meters, which seems more plausible. I can't find a primary cite, but here is one of the engineers involved in building the plant, bringing this up.

AMac said at March 23, 2011 7:06 AM:

Nuclear Physicist and Bernie,

I understand that you dislike Randall's post, but do not understand the basis for your criticisms.

Yes, the situation is fluid, and yes, more and better information will be made public, over time. Why is that a compelling argument for remaining silent, or for restraining informed criticism? Would the same reasoning hold for the Libyan situation and other fast-moving current events?

AMac said at March 23, 2011 10:50 AM:

USGS expert on ancient tsunamis Brian Atwater discusses the record of the past few thousand years, and the risks to the Pacific Northwest and other regions.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©