Four lessons from Fukushima: Improving emergency response

It appears that the managers of the Fukushima Daiichi nuclear power plant, taken by surprise, did not know exactly what to do after a massive earthquake and tsunami struck the plant on March 11. Experts in the United States, thousands of miles away, had a duty to provide timely, helpful advice. Both the press and US officials failed. In particular, the US Nuclear Regulatory Commission’s recommendation to stay at least 50 miles away from Fukushima was inappropriate and may have caused unnecessary panic.

Radiation management after an accident must be handled differently than radiation management during normal plant operations. With the benefit of hindsight, we now know that some critical decisions were not made quickly enough during the earliest hours and days of the emergency response. Other decisions were, on the other hand, premature.

Looking back on the events that unfolded in March, we can learn some important lessons about what should have been done differently. Understanding these lessons is important if we are to avoid a similar crisis in the future.

In June the Japanese delivered a 900-page report to the International Atomic Energy Agency (IAEA). In addition, the Japanese media have collected and released a number of memos and e-mails written by officials, including the prime minister, in the crucial days from March 11 to March 16. From these documents and other information there are four things worth noting that went wrong at Fukushima Daiichi:

Emergency reactor cooling did not begin soon enough. It should be firmly drilled into every reactor operator, as it was into me in 1970: Always keep the reactor cool. When all sources of electrical power failed at Fukushima, and coolant water stopped circulating, the plant’s operators should have injected seawater into the reactors by any means possible within an hour. Instead, they delayed doing so, possibly because of confusion between the plant engineers, management at the Tokyo Electric Power Company (Tepco), and Japan’s prime minister.

Cooling the spent fuel pools was less time-sensitive, but by Sunday night (three days into the accident), seawater should have been used there, too. If seawater injection had started early enough, the fuel would not have heated to a level where the cladding of the fuel rods failed, and hydrogen gas would not have been produced. Once the fuel cladding had failed, radioactive materials entered the water and greatly increased the radiation levels just outside the reactor vessel, hindering the work necessary to stabilize the situation.

If emergency cooling had happened in time to prevent partial meltdowns of the reactor fuel, most readers would never have heard the name Fukushima. There may be technical reasons for the delay, but it seems likely that the plant’s operators hesitated because they knew that injecting corrosive seawater would do permanent damage to the reactors.

It is essential that every nuclear plant operator have a plan for how to respond quickly to a catastrophic loss of coolant — not just for a future, probably rare, reactor accident, but also in case of a terrorist attack. This planning must be done in advance, under calm circumstances, to avoid a situation in which short-term political interests prevail over public-health concerns.

Dose limits should have been raised quickly. Before the Three Mile Island accident in 1979, it was generally accepted, based on recommendations by the International Commission on Radiological Protection (ICRP), that nuclear plant workers should not be exposed to radiation doses exceeding 20 rems (0.2 sievert) in a single year. After Three Mile Island, many national authorities reduced the radiation exposure limits for workers to 5 rems per year. And after the Chernobyl accident in 1986, utilities and nuclear laboratories worldwide effectively reduced the limits again, to 1 rem (0.01 sievert), to reduce bureaucratic hassle from above.

It makes sense to limit exposure to 5 rems per year for a nuclear-power worker during ordinary operations; there is no public health reason for a higher limit. But it is necessary to raise this limit while an accident is in progress. That should be done promptly, and probably automatically. It should be acceptable for a worker to receive up to 20 rems of total radiation exposure during an accident, and indeed at Chernobyl 100,000 or more workers got this dose as “liquidators” (cleanup workers). Before 1979, the international consensus was to allow a one-time dose of 80 rems (0.8 sievert) for an astronaut, or for an individual “to save lives” in a rare situation.

At Fukushima some workers were reportedly pulled off the job the day after the tsunami, to avoid exceeding pre-accident radiation dose limits. This probably delayed the technical response to the accident. Japanese regulators later reinstated the pre-1979 ICRP exposure limits of 20 rems for a single accident.

At Chernobyl, the dosimeters in the control room were designed to measure ordinary doses and became saturated when exposed to the much higher levels of radiation emitted during the accident. Dosimeters capable of measuring the higher levels during an accident were inaccessible. Unfortunately, reactor operators at Fukushima Daiichi and elsewhere did not learn from the Chernobyl experience: At Fukushima, the tsunami made dosimeters unavailable.

Rapid evacuation was unnecessary. An extremely important issue in public health management is when to suggest, when to recommend, and when to order an evacuation. Before 1980 the US Nuclear Regulatory Commission asked for an “Emergency Planning Zone” (not an evacuation zone) with a 10-mile diameter around each nuclear power plant. After the Three Mile Island accident, these areas became “evacuation zones” without much discussion.

A distinction must be made between compulsory evacuation and voluntary evacuation. This facet of the response was barely mentioned in the Japanese report to the IAEA. Faced with a nuclear accident, some people will want to leave voluntarily. At Fukushima it would have been appropriate for the government to recognize this and immediately institute procedures (such as creating one-way roads and banning parking along evacuation routes) to avoid the kind of chaos that occurred after the post-Katrina New Orleans flooding. But compulsory evacuation from the Fukushima area was inappropriate.

At Chernobyl, the evacuation from Pripyat (the community closest to the power plant) did not happen until 34 hours after the crisis began, and was delayed an additional three days for more distant communities. Even in retrospect there is no indication that this was too late. Evacuation can be delayed until a large release of cesium is actually occurring. Then evacuation can be done more calmly and deliberately than was the case at Fukushima.

Examples of senseless restrictions on the public abound and persist today. The high-speed train from Tokyo to northern Japan was discontinued for three months to avoid exposing passengers to minuscule doses of radiation. Those who traveled by air instead got a similar dose from the increased cosmic radiation at higher altitudes!

It is not too late to bring some common sense to the situation. Japanese nuclear safety regulators are discussing a proposal to indefinitely prevent people from returning to the area with the highest deposition of radioactive materials, where residents would receive an estimated dose of 1 rem (0.01 sievert) per year. Scientists around the world should stand up and point out that there is no scientific justification for a compulsory restriction.

The primary purpose of evacuation is to avoid any radiation doses of 200 rems or more, which would lead to acute radiation sickness. Exposures in the vicinity of Fukushima were estimated to be one-tenth of that at most. A secondary purpose is to avoid the effects of radioactive cesium 134 and 137 deposited in the air and soil. A delay of a couple of weeks to calmly assess the situation does not have a significant impact on the total radiation doses received by the general public. Evacuating too quickly, however, almost certainly has significant adverse effects on public health. Unfortunately, the decision-makers in Japan did not even consider the desirability of a slower or less-extensive evacuation effort.

In 1975 I saw figures from the World Health Organization reporting a 5 percent increase in cancer rate for anyone dislocated for any reason. The Report of the President’s Commission on the Accident at Three Mile Island stated: “We conclude that the most serious health effect of the accident was severe mental stress, which was short lived. The highest levels of distress were found among those living within five miles of TMI and in families with preschool children.”

The Japanese should have recognized from their previous experience that evacuations of the elderly could result in unexpected deaths. According to recent Japanese broadcasts, significant numbers of the elderly in shelters have died. The calorie intake of some people in shelters may have been below starvation level. Not all had three meals a day. The emotional status of many may also reduce food and liquids intake. These harms can far exceed any benefits that evacuation may bring.

Even experts failed to describe the effects on health correctly. This is, for example, a major weakness of the Japanese report to the IAEA. Exposure to 200 rems (2 sieverts) of radiation within a week causes acute radiation sickness, and the victim usually dies within a month as his innards are destroyed. If he survives, he faces a 10 percent increase in the probability of developing a cancer that might otherwise occur naturally. This probability increases with extended exposure to radionuclides, particularly cesium 134 and 137.

A few workers at Fukushima received a high dose of 20 rems (0.2 sievert). This is one-tenth of the dose that would cause acute radiation sickness, and indeed there is no report that any workers have low white blood cell counts (a medical indicator of acute radiation sickness). Assuming these men received the highest exposure, no one at Fukushima got acute radiation sickness.

It is important to realize that no member of the public got, or is likely to get, an exposure that came close to doubling this calculated natural cancer rate. No one in the public, and probably no plant worker, can say: “My cancer was caused by Fukushima.” This contrasts with more than 15,000 dead bodies and nearly 8,000 people still missing after the earthquake and tsunami.