Complex technology history in Smart Brevity
Can it be done? I thought I’d give it try.
Acronyms and words
Fukushima Daiichi - not to be confused with Fukushima Daini, a different nearby nuclear plant.
TEPCO - the Tokyo Electric Power Company, the utility that built, owned, and operated Fukushima Daiichi.
IAEA - International Atomic Energy Agency.
The earthquake
What you need to know: Ground-motion sensors tripped and automatically put the three running reactors at Fukushima Daiichi into shut-down mode. The plant withstood the earthquake reasonably well. Everything went “by the book” the first hour.
• The Tohoku-oki earthquake happened on March 11, 2011 at 2:46 p.m. Japan Standard Time. At magnitude 9, it was, and fortunately still is, the most powerful in Japan’s recorded history. The epicenter was approximately 100 km off the coast.
• At that time, three of the six reactor units at Fukushima Daiichi were operating at full power. Units 4, 5, and 6 were for various reasons off-line.
• Ground-motion sensors (accelerometers that measure horizontal ground movement) automatically initiated a reactor shut-down or “scram.” This is a rapid, full insertion of all control rods.
• With a reactor is turned off, cooling water is pumped through to take away “decay heat.” Unit #1, after the scram, was producing around 100 MW of thermal energy.
• A normal cool-down can take days. The key thing for is to keep the cooling water inside the reactor core above the top of the fuel cell.
• Historical records for Japanese earthquakes are excellent and date back to the year 599.
• The Tōkai segment had major earthquakes in 1498 (8.6 Mk), 1605 (7.9 Mk), 1707 (8.6 Ml), 1854 (8.4 Mk), and 1923 (8.3 Ml).
• An insurer might calculate the likelihood of a magnitude 8 earthquake off Fukushima as a 100- or 150-year event.
• The measured peak ground acceleration (PGA) at Fukushima Daiichi slightly exceeded the design basis of the buildings. These were designed to withstand horizontal ground acceleration 0.447g. The measured PGA was 0.561g. For context on PGA levels, a level of 0.50g is “very high, but well-designed buildings can survive if the duration is short.”
• Aside: The Diablo Canyon Nuclear Power Plant in California, the frequent target of anti-nuclear groups because it is near a fault line, is built to withstand horizontal ground acceleration of 0.75g.
• Nuclear plant personnel in Japan train for earthquakes. At Fukushima there had been an earthquake drill the week before.
• After the quake, Fukushima plant personnel inspected the buildings and reactors and did not note any safety compromising damage. The control room operators were able to use their instruments, open and close valves, and were confident everything was working.
• There was about 40 minutes between the earthquake at 2:46 p.m and the first tsunami wave at 3:27 p.m. In that time the reactor cool-down was proceeding “nominally,” to use the NASA word.
• But: In post-event inquiries, politicians critical of TEPCO were keen to suggest that the earthquake, not the tsunami, “might have” damaged critical core cooling system valves and pipes of Unit 1. This is almost certainly not true.
• The motivation for making this claim stems from an oddity of Japanese nuclear power plant regulation. This requires construction permits, but not ongoing operating licenses. After a company builds a plant, it can — from a strictly legal point of view — run it as it sees fit. Seismicity, in particular location over a known fault line, is the only universally accepted, non-controversial basis on which a reactor construction permit can be denied. If a politician wishes to bash TEPCO, claiming it failed to build to earthquake standards is the best and almost only club available.
Failure of the public power grid
What you need to know: The earthquake knocked out Japan’s entire east coast power grid. The blackout extended from Tokyo to northern Honshu. At Fukushima Daiichi, the loss of grid power automatically the emergency diesel generators. The reactor core cool-down system — pumps, valves, and instrumentation — required AC power levels.
• Had the ground-motion sensors not already done so, loss of grid power would have also triggered a reactor scram.
• A contingency plan to obtain electrical power from another regional grid failed because of a interconnect mismatch.
• The safety systems that cool the core down are the Isolation Cooling Condenser (IC) and the Emergency High Pressure Cooling System (HPIC). The IC is a passive pipe loop, but has control valves driven by electric motors. The HPIC is a powerful pump for injecting cooling water.
• These started automatically at 2:52 p.m., six minutes after the quake.
• In a normal shutdown, the IC can be toggled on and off to control the powerful HPIC. One goal is to keep about 15 feet of water above the core.
• At the time the wave hit, the operator had toggled two IC valves closed to slow down the HPIC pump.
• In 20-20 hindsight, this was a consequential “overthinking” error. The IC valves had been designed to open automatically for a reason. They would have been better off left open.
The tsunami
What you need to know: The earthquake was one thing, the tsunami another. Fukushima Daiichi did have a tsunami barrier, but not one able to protect it from a 14 meter wave.
• The first tsunami wave hit at 3:27 p.m. (15:27), about 40 minutes after the offshore earthquake. A second wave hit at 3:35 p.m. (15:35), 8 minutes after the first. There was a third wave. The peak height was at 15:37.
• The tsunami waves topped about 14-15 meters (46 to 49 feet) above mean sea level at Fukushima Daiichi. That number is based on silt lines left on the buildings. A “wave meter” just offshore jammed at its maximum of height of 7.5 meters and was then carried away.
• The tsunami killed some 18,000 people along the coast. The exact death toll can only be estimated because many victims were swept out to sea.
• Aside: People who focus exclusively on the nuclear accident need to examine their moral priorities.
• The adequacy of TEPCO’s tsunami protection at Fukushima Daiichi was, of course, a contentious issue in the post-accident investigations and litigation.
• TEPCO’s best defense is that a 14-meter tsunami is an “act of God” no one could have reasonably foreseen or prepared for.
• There is in northern Japan a singular sort of historical record for tsunamis. Ancient “tsunami stones” can be found in rough semi-circles facing the coast, marking past high-water points. The stones carry inscriptions in antique dialect which roughly translate: “Don’t even think about building a house between here and and the ocean.”
• The pre-Fukushima general standard for tsunami barriers at nuclear plants along the coast was 5.7 meters (18 feet).
• The barriers around Units #1-4 at Fukushima were 10 meters high. The barrier around Units #5 and #6 was 12 meters.
• Plant elevation is as important as barrier wall height. The beach shoreline at Fukushima was lowered significantly during construction to allow delivery of heavy equipment by ocean-going barge. This also allowed the reactor buildings to be anchored on bedrock.
• The shallow, slanted sea floor topology off Fukushima Daiichi is conducive to wave shoaling, which amplifies tsunami waves.
• In 2008, TEPCO engineers made calculations specifically for the Fukushima Daiichi site. That calculations showed the plant needed a barrier 10.3 meters high.
• The tsunami barriers at the three other nuclear plants along the same coast — Onagawa, Tokai and Fukushima Daini — proved adequate to limit damage.
• A professor of seismology at the University of Tokyo, Kunihiko Shimazaki, warned publicly in 2004 that the 5.7 meter figure for potential tsunami wave height was too low. He thought it should be at least doubled.
• Professor Shimazaki makes an appealing Cassandra figure for the media, but if we take his calculation literally, 11.4 meters is still less than the actual wave height of 14 meters.
• In the post-accident litigation, TEPCO’s lawyers had no problem with the word “tsunami,” but objected strenuously to any use of the word “earthquake.” Critics of TEPCO preferred to conflate the two.
• This relates to the oddity in Japanese nuclear regulation the prohibits building above known fault lines. The 2011 tsunami originated from an earthquake offshore, not underneath the power plant.
• In English and American jurisprudence, an “act of God” defense must satisfy the following four criteria:
[an act] which involves no human agency
which is not realistically possible to guard against
which is due directly and exclusively to natural causes and
which could not have been prevented by any amount of foresight, plans, and care.
• Of these, the point at issue is whether TEPCO failed on (ii) — “realistically possible to guard against” — for example, by building a higher barrier.
• Aside: Diablo Canyon sits on an elevated bluff, but nonetheless has a 9.75 meter tsunami barrier.
Takeaway: TEPCO admits to making a mistake on tsunami protection, but it is lazy and a bit facile to say in hindsight “TEPCO should have known.”
Total Loss of electrical power
What you need to know: A serious design error in the Fukushima plant put the emergency generators in basement rooms.
• The topographic elevation of the basement floors was literally below sea level.
• After the tsunami waves, they were filled with seawater.
• The emergency diesels kept running for 4 or 5 minutes after the second wave, then stopped.
• The unprotected fuel tanks for the generators outside were carried away.
• There was a recommendation by the IAEA in the 1990s that emergency generators and fuel supplies be protected from floods and located in independent structures away from main buildings.
• At Fukushima Daiichi, Units 5 and 6, which were off, did have two backup generators located away from the main building and several meters higher in elevation. One of these worked after the tsunami.
• The electrical feed switchboard panels on the buildings were damaged and need be repaired before electrical power could be restored, even if it had been available.
• The violence of tsunami damage to the grounds of the plant made difficult to impossible to truck in portable diesel generators and heavy electrical cables. The roads had been washed away and were blocked by overturned cars and heavy wreckage.
• Seawater from the wave entered the Unit 1 control room at floor level, knocked out the instruments, and damaged backup batteries except those in the ceiling for the emergency lights. The instruments in Unit were lost at 3:50 p.m. The most important of those were the ones that measured the temperature, pressure, and water level in the core.
• Plant workers made heroic and creative efforts to turn the situation around. They attempted to use an ordinary fire pump to inject water into the cooling system, but at only 100 psi it had difficulty working against the higher pressure in the reactor. To get the closed IC valves back open, workers removed 10 batteries from cars in the parking lot and wired them together.
• With sporadic and intermittent battery power for the instruments, the damage control effort was flying blind. The readings they were able to obtain were often contradictory and difficult to interpret.
• The 3 other nuclear plants on the east coast also had flooding from the tsunami. The critical difference was that at least one usable diesel generator survived.
• The design errors TEPCO made in building Fukushima Daiichi can be summed in three heads: elevation of the site; height of the tsunami barrier; and the emergency generator locations.
• Takeaway: Without AC power to run the cool-down equipment, exposure of the reactor cores was inevitable and unpreventable.
Core exposure and hydrogen build-up
What you need to know: Exposure of the reactor cores created hydrogen gas bubbles on the top floor of the buildings.
• There is no simple analog way to monitor the level of water in the reactor core. It must be inferred from instrument pressure readings. The best guess is that cooling water fell below the top of the Unit 1 core fuel cells around 11 p.m. on the first day, 11 March.
• As is well known, the zirconium cladding used to encase uranium fuel rods reacts with steam to produce hydrogen. This hydrogen can collect at some pressure in the structure around the reactor.
• Aside: This also happened at Three Mile Island. There, the hydrogen ignition was less explosive and not did shatter the containment structure.
• Airtight containment is an example of a safety feature that in turn creates a secondary safety problem that can require yet another safety system. Hydrogen “burners” can be installed on reactors to allow a controlled burn of hydrogen before it builds to explosive levels.
The venting decision
What you need to know: The built-up hydrogen in Unit 1 could have been vented to the outside air, but this would have also released radioactive steam. Bureaucratic dithering and political interference delayed venting until it was too late.
• Plant manager Yoshida recommended around 9 p.m. on March 11 that Unit #1 be vented to the outside air. This recommendation seems to have derived from fears about pressure, not hydrogen. Yoshida commented afterward that everyone was so focused on “saving the core” that they not thinking about hydrogen buildup.
• A general IAEA recommendation not implemented at Fukushima was passive hydrogen vent valves. These open without operator intervention at some pre-determined high pressure level. AC power can be used to close the valve, but cannot be required to open it.
• There was no procedure “in the binder” for a depressurization without electrical power. The operators and managers had to make it up.
• The existing emergency plan with Fukushima Prefecture called for an evacuation of all local residents within 2 km of the plant before a radioactive release. The nearest permanent residence was 1 km from the plant boundary, which in turn is some distance from the reactor buildings.
• The 2 km evacuation order went out at 8:50 p.m. on the evening of March 11. The agreement with the local government called for completion of the evacuation be confirmed before venting could begin.
• By the evening of the first day, various powers in Tokyo, including the Prime Minister, Nato Kan, had taken charge of the emergency response at Fukushima.
• At 9:23 p.m. Kan decided to expand the evacuation radius to 3 km. This at least doubled the number of people who who had to be evacuated before venting could take place.
• Kan also wanted the venting delayed until after he could hold a press conference.
• When they were ordered to evacuate, only 20 percent of residents inside the 3 km zone were aware that there was some kind of problem at the nuclear plant. Most learned about it from the proverbial knock on the door.
• The second day, 12 March, Tokyo expanded the radius of the evacuation zone several more times from 3 km to 10 km to 20 km.
• The crisis managers in Tokyo operated with little sense of urgency. They also appeared oblivious to the on-the-ground difficulties and delays created by their expansion of the evacuation radius.
• No Japanese politician wanted to be the one who authorized a release of “radioactivity,” at any level, over a civilian area. Kan eventually did.
• The hydrogen venting had just started for Unit 1 when the top floor of the building exploded at 3:36 p.m. on 12 March. This was a large explosion akin to a bomb going off.
• Flying chunks of concrete from the Unit 1 building hit the other units, knocking out their instruments and outside portable generators.
• Aside: Proponents of the green “hydrogen economy” should take note. In today’s media environment, a single Hindenburg or Fukushima can cancel an industry.
• The explosion blew radioactive debris from the Unit 1 core into the air. After that, the grounds around the plant and between the buildings were radioactively “hot.” This greatly hampered recovery work at the other units. Workers either had to leave the area or put on radiation suits and breathers.
• An interesting counter-factual study would compare the radiological impact of the planned controlled release with the uncontrolled release of the explosion.
The Chain of Other Explosions
• The failure cascade at Unit 3 paralleled that at Unit 1, but took longer. Unit 3 had a hydrogen explosion at 11:01 a.m. on 14 March, the fourth day.
• The Unit 3 explosion knocked out the Unit 2 cooling pump. The Unit 2 core became exposed by that afternoon. Ironically, Unit 2 did not suffer a hydrogen explosion because a piece of flying concrete from Unit 1 had knocked a hole in its building. The hole allowed its hydrogen to vent. Unit 2 did release radioactive material from its core into the air.
• The Unit 4 reactor was off, but to everyone’s surprise its building suffered a hydrogen explosion on 15 March. Unit 4 shared a smokestack with Unit 3. It was later determined that hydrogen from Unit 3 back-flowed through it into the Unit 4 building, collecting on the top floor.
• All three explosions at Fukushima were hydrogen explosions. They were of course not nuclear detonations. “Black smoke” noted by the media coming from Unit #4 was most likely a burning air conditioner on the upper floor. Media reports of “white smoke” were almost certainly stream.
• Assertions made by some prominent US officials that the spent fuel pools at Fukushima were dry and burning were false.
The Fukushima evacuation took more lives than it saved
What you need to know: The response of Japanese officialdom to the Fukushima release was an over-the-top panic fueled by extreme radiophobia. This panic was communicated to the media and public. Officials effectively cried ‘fire’ in a crowed theater. Fukushima is a case study in how not to respond to radiation emergency.
• There have been no deaths of plant workers or nearby residents attributable to the radiation release.
• One worker at the plant, a 55-year-old man, was diagnosed with lung cancer in 2016. He worked there only during the clean-up and wore protective gear. The Japanese government agreed to pay his family compensation.
• The evacuees and plant workers received comprehensive and continuous health screening for years after the accident.
• Ten years after, the UN’s scientific committee on the effects of atomic radiation (UNSCEAR) concluded “no adverse health effects among Fukushima residents have been documented that could be directly attributed to radiation exposure from the accident.”
• Japan’s Health Ministry independently came to the same conclusion.
• 146,520 people were forcibly evacuated from the Fukushima region by government order.
• There are various figures for the death toll from the evacuation itself. Japan’s Reconstruction Agency in March 2013 put the total at 1,383. A survey by Japanese newspaper Mainichi Shimbun that same year put the figure at 1,600. The Japan Times put the figure at 1,656 in 2014.
• There various causes of death in evacuations. These start with people evacuated from hospitals and nursing homes. “When you evacuate a hospital intensive care ward, you cannot take the patients to a high school and expect them to survive,” one doctor put it.
• Less obvious causes include (a) poor conditions in evacuation centers; (b) depression and “exhaustion,” almost universally reported by evacuees and ( c ) suicide, not uncommon in Japan. A 2014 UN report said that the most significant impact of the Fukushima accident was on mental health, not radiation-related health issues.
• In Japan, victims of radiation exposure are stigmatized and socially shunned. The term Hibakusha is a near-pejorative originally applied to survivors of the Hiroshima and Nagasaki bombings. It the last decade it expanded and is now used to refer to former Fukushima residents.
• Fishing and agriculture in Fukushima Prefecture stopped dead in 2011 and have recovered. Other Japanese would not eat Fukushima fish or vegetables.
• For Fukushima refugees, the stigma additionally feeds off their low former job status (fishing, agriculture); dislike for rural migrants coming into cities such as Tokyo; and resentment of government benefits most receive. The diaspora from Fukushima has created an underclass of displaced persons in Japan.
Takeaway: The fear of radiation, not radiation itself, killed people at Fukushima.
Would have, could have
• In hindsight, the original 2 km evacuation zone was about right.
• A “shelter in place” order outside of that would have been appropriate.
• The winds were light and variable. Airborne measurements of ground surface radioactivity trace a plume leading away from the plant towards the north-east.
• Color-enhanced “hot zone” visualizations are captivating but are based on subtle differences. The measuring instruments are extremely sensitive. The unit scale on this one is in microsieverts (µSv) , not millisieverts (mSv). There are 1,000 microsieverts in a milliservert. I will discuss confusing units in the next section.
• There was one estimate that 81% of radioactive precipitation fell over the ocean.
• For an airborne plume, 1950s advice about fallout remains valid: stay indoors, let it fall to the ground; and wash everything, including yourself, afterwards. With low-level radiation and modern instruments, there is time to take others actions if necessary.
• The map of evaluated townships resembles the one above:
The LNT theory had a death grip on Japanese bureaucrats
What you need to know: The disparity between the radiation and evacuation death tolls shows the price of clinging to the flawed linear no-threshold (LNT) theory, which assumes nearly all radiation is harmful, even very low doses.
• Few radiation releases have been as intensely studied as the one at Fukushima.
• Radiation is so confusing there is strong pressure to just go along with whatever the officials and their experts say about it. Nature did not make physics for dummies.
• The key term in the evacuation map above is “annual accumulation.” Why that is objectionable requires a big digression.
Radiation units
• A plethora of units are used to talk about radiation levels. I will skip the historical ones.
• The unit used most often in policy discussions is the sievert, typically the millisievert (mSv). Note “20 mSv” and “50 mSv” in the evacuation map.
• The sievert is a convenience unit for biological dose. Its derivation requires a number of steps.
• The hard physics SI unit for measuring radioactivity is the becquerel (Bq). One becquerel is one nuclear decay per second. This unit is so small that power-of-ten prefixes are almost always needed for it. The petabecquerel (PBq), 10 to the 15th power, is common.
• Radiation in becquerels is almost always given along with the name of the particular element and isotope producing it. Thus “Cesium-134: Approx. 3.5 PBq” was an estimate of the total amount of radioactive Cesium-134 put into the ocean by the Fukushima release.
• That number just says that many Cesium-134 atoms are there. It says nothing about what that radioactivity is doing.
• The next unit, the gray, is a measure of absorbed energy. It is defined as the absorption of one joule of radiation energy per kilogram of matter.
• Make that a kilogram of human tissue and you get close to the sievert.
• The sievert is convenient because it is a single number. That is also its drawback.
• The three types of radiation, alpha, beta and gamma, have very different modes of operation. For example, alpha radiation will not penetrate a piece of paper or the skin, but gamma will go through anything other than lead. The sievert calculation weighs these to get the single number. So, in theory, 1 Sv of alpha radiation will have the same biological effect as 1 Sv of beta radiation.
• The sievert should always be be used with a unit of time to produce a rate.
• Having a single number is convenient, but much is lost in the computation. A sievert reading says nothing about the source of the radiation, how it might enter the body, or what might be done to prevent it.
Radiation sources
• We live in a sea of low-level radiation. It primarily comes from the ground and cosmic rays. The natural background radiation a person is expected to absorb in the US is 3 mSv a year.
• When one radiation level is higher than another, the next natural question is, “Why? Where is it coming from?” A person living in Denver, Colorado can expect to absorb 11.5 mSv per year from cosmic rays because of the higher altitude.
• The Fukushima reactors released into the air Iodine-131; Cesium-134; Cesium-137; and various noble gases. Iodine-131 has a half life of 8 days. The noble gas was primarily xenon-133, which has low biological significance and decays rapidly. The half-life of Cesium-134 is about 2 years and that of Cesium-137 about 30.
• Immediately after the release, the “air dose” rate was of most interest. This is generally taken 1 meter above the ground at various monitoring stations. For example, at Fukushima at 11:05 on March 15, a radiation monitor at the main gate reported 4.5 mSv/hr.
• To interpret this, if you stood under the main gate for an hour your body would likely absorb 4.5 mSv of radiation. If you simply walked through the gate, the exposure would be nothing near that. If you get a chest CT scan you absorb 7 mSv almost at once.
• Air dose readings were affected by the breeze and were very erratic.
• After the dust settled — literally — the aerial surveys used to make the hot zones maps were measuring ground deposition of the two Cesium isotopes.
• Hand-held dosimeters are handy for A/B comparisons, but the bodily absorption algorithms used by these devices is inescapably fuzzy.
Radiation Levels
• The actual measured radiation doses absorbed by people living around Fukushima were low. In a 2015 report “The Fukushima Health Management Survey: Estimation of external doses to residents in Fukushima Prefecture,” Tetsuo Ishikawa and his coauthors found that the individual doses to 423,394 residents during the first four months had the following distribution: 62.0 percent under 1 mSv, 94.0 percent under 2 mSv, and 99.4 percent under 3 mSv.
• The “How much is that?” question must always be asked and is best answered with comparisons. Those levels accumulated in four months are well below background levels for a year in most parts of the world.
• Any honest discussion of radiation levels must reference the baseline of natural background. Saying a level as “elevated” is a rhetorical trick for propaganda.
• The Lawrence Berkeley Lab found “an elevated” level of iodine-131 in rainwater near Fukushima, but Kai Vetter, the head researcher, also pointed out you’d have to consume 140 liters the rainwater to get the equivalent radiation exposure of a cross-country flight.
• Vetter noted the levels of other radioisotopes spiked and then quickly dissipated.
The accumulation assumption
• Bodily absorption is discussed as a rate, sieverts per some unit of time, and often called the dose.
• “Dose” is an important concept. The dose makes the poison.
• A low-level radiation insult to the body is resembles a burn. A sunburn is indeed an insult from solar radiation.
• While a massive burn can be fatal, the body will repair minor damage. Life on earth has always existed in a radioactive sea. It evolved adaptive biologic repair and removal responses to radiation damage.
• Previous radiation standards recognized the repair process. In the 1950s, the dose limit was 1 mSv per day. This implicitly allowed for recovery and healing. Radiation therapies in medicine routinely have healing times built into their protocols.
• Continuing the medical analogy, what matters is the dose profile, not the cumulative dose. Taking one aspirin per day is not the same thing as taking 365 all at once.
• The body has an additional ability to develop tolerances for insults delivered in low doses.
• These can be surprising. Arsenic is an effective treatment for trypanosomiasis (“sleeping sickness”) and syphilis. It was the dominate therapy prior to penicillin. Medical arsenic was considered a wonder drug and had impeccable scientific credentials, having being worked out by German physician Paul Ehrlich. “Arsenic eaters” started with small amounts and got up to large doses that would kill most people outright.
• If your “harm curve” from low-level radiation exposure never goes down—that is, there is no healing—you can simply add up all the sievert readings into a cumulative total. Sometimes this is done by multiplying short-term readings and extrapolating to a year—e.g., some mSv/hr times the number of hours in a year.
• The annual number is now the basis for regulation. For example, in Canada the effective dose limits for a nuclear energy worker is 50 mSv in one year.
• The point is not to quibble over the number, but the underlying assumption that goes into computing it.
• Linear no-threshold (LNT) is sometimes called a theory or a hypothesis. That would make it subject to disproof, to which it is immune. I believe it’s properly called an algorithm, “a set of rules to be followed.”
• The simple accumulation algorithm has the virtue of administrative convenience and simplicity.
• Unfortunately, it is not based on medical science. There is a big missing mapping between accumulated sieverts and actual health risk.
• There have been many studies of groups who receive higher than average long term radiation exposure. These include flight attendants, residents of Denver, and certain geologic areas. None established a statistically significant increase risk of, for example, cancer.
• The accumulation algorithm has been extraordinarily difficult to sunset or dislodge in the regulatory and emergency response world. Bureaucratic inertia, over-caution and sheer mulishness explain most of that. The cost being wrong is not borne by the elite academic or regulator, but by a diffuse and unorganized public.
• Aside: TSA air travel regulations come to mind. The time lost by millions of passengers counts for little. The TSA administrator, on the other hand, knows he or she will be fired and raked over the media coals if there is a single highjacking.
• At Fukushima there was a failure to balance the risks of the large-scale evacuation with the health risks of low-dose radiation exposure.
• That is the polite way of putting it. The other way is that the residents of Fukushima needed saving from the people who were out to save them.
In place of a conclusion
• Fukushima was originally a class 5 incident on the International Nuclear Event Scale (“Accident with wider consequences”). In this it joined Three Mile Island.
• It was Japan itself that wanted it upgraded it to a class 7 — as serious as Chernobyl. It was Japan itself that made it so.