The dangers from nuclear power in light of Fukushima

This is a joint post, by Chris Goodall of and Mark Lynas ( We make no apologies for length, as these issues can really only be properly addressed in detail. An abridged version of this article was published in the TODAY newspaper of Singapore on April 6 2011. How risky is nuclear power? As the Fukushima nuclear crisis continues in Japan, many people and governments are turning away from nuclear power in the belief that it is uniquely dangerous to human health and the environment. The German government has reversed its policy of allowing the oldest nuclear plants to stay open and Italy has reportedly abandoned its efforts to develop new power stations. Beijing has stopped approving applications for nuclear reactors until the consequences of Fukushima become clear, potentially affecting up to 100 planned new stations. The mood towards the nuclear industry is antagonistic and suspicious around the world. We think this reaction is short-sighted and largely irrational.

For all its problems, nuclear power is the most reliable form of low carbon electricity. It remains the only viable source of low-carbon baseload power available to industrialised economies, and is therefore responsible for avoiding more than a billion tonnes of CO2 emissions per year. In addition to these unarguable climate benefits, we believe that nuclear power is much safer than its opponents claim. Despite the hyperbolic nature of some of the media coverage, even substantial radiation leaks such as at Fukushima are likely to cause very little or no illness or death. No power source is completely safe, but compared to coal, still the major fuel for electricity generation around the world, nuclear is relatively benign. About 3,000 people lost their lives mining coal in China alone last year. Many times that number died as a result of the atmospheric pollution arising from the burning of coal in power stations.

Although much journalism of the last few weeks has provided careful assessment of the true dangers of nuclear accidents, we thought it would be helpful to pull together the results of scientific studies on the damage caused by nuclear radiation to human health. Our aim is allow readers to put some perspective on the radiation risks of nuclear power, particularly after accidents, and to appreciate the context of the oft-quoted units of ‘millisieverts’, ‘bequerels’ and other measurements. This is a complicated story, because not all radiation is the same – a crucial factor is the timescale of exposure. There is a big difference between the expected impacts of exposure to huge amounts in a very short period, large doses over several weeks, and long-running or chronic exposure.

We examine these three scenarios in turn. The results seem to be quite clear to us: accidents and leaks from nuclear power stations are not likely to cause substantial numbers of illness or deaths, even under exceptional circumstances such as are currently being experienced after the combined earthquake and tsunami disaster at Fukushima. This is an important conclusion given the potential for nuclear power to continue to mitigate global warming, which presents vastly greater risks on a global scale. We are not advocating slackness or complacency, just suggesting that a rational and balanced assessment of the risks of radiation is a good idea. To hastily abandon or delay nuclear power because of radiation risks from accidents such as that at Fukushima is poor policy-making.

Some background

All of us are exposed to radiation every day of our lives. Very little of this comes from nuclear power or nuclear weapons. Other sources are far more important. One example: potassium is a vital chemical for carrying electrical signals around our bodies but a rare, naturally occurring, isotope, potassium 40, is radioactive. The tiny amount inside us produces 4,000 decays of individual nuclei every second. This internal nuclear fission of potassium atoms and from a radioactive natural isotope of carbon is responsible for about 10% of the annual dose received by someone in the UK.

More important sources are the radon gas produced in granite rocks, cosmic radiation and doses from medical equipment. By contrast, and despite the attention we pay to them, nuclear power stations and nuclear weapons are responsible for much less than half of one percent of the radiation typically absorbed by people in the UK. The same rough percentages apply to other countries operating nuclear reactor fleets.

The average background radiation across the UK is about 2.7 millisieverts (mSv) a year. (A ‘millisievert’ is a measure of radiation exposure to the body and is therefore a useful unit to directly compare the radiation received from different sources). People in Cornwall, where there is far more radioactive radon around because of local geology, experience more radiation than in other areas. Their dose may be as high as 10 mSv, almost four times as much as the UK average. In fact nuclear power plants could not be built in the granite areas of the county because the natural background radiation at the boundary of the power station would be higher than is allowed under the strict rules governing the operation of nuclear plants. Cornish radiation isn’t that unusual: parts of Iran, India and Australia have even more natural nuclear fission than Cornwall.

So our first point is that nuclear power is an almost trivial source of radiation, dwarfed by natural variations in other sources of radiation. The second is that exposure to radiation in the UK is tending to rise, but certainly not because of nuclear power or leaks from other nuclear operations. Instead it comes from the increased use of radiation in diagnostic equipment used by health care professionals. One scan in a CT machine will add about 10mSv to a person’s annual exposure – 3 million Britons went through this process last year. Per head of population, the number is even higher in the US.

These are two important basic numbers to help us assess just how dangerous nuclear power is: 2.7 mSv a year for the average natural background radiation received by the typical person in the UK and 10 mSv for a single CT scan. We will use these numbers to compare the radiation effect of nuclear power and to assess the importance of the very rare but severe accidents at nuclear power plants.

The impact of exposure to very high levels of radiation over a few hours

a) Chernobyl workers (1)

The fire and explosion at Chernobyl in 1986 was the world’s most severe accident at a civil nuclear power plant. It is the only such event which is known to have killed workers from the effect of radiation. About six hundred people were involved in work on the site of the power plant during the first day after the accident, of which 237 were thought to be at risk of acute radiation syndrome (ARS) because of their degree of exposure. 134 individuals developed symptoms of ARS and 28 died as a result. The deaths were generally due to the skin and lung problems, compounded by bone marrow failure. All but one of people killed received a dose of radiation above 4,000 mSv, with one of the deaths occurring after a dose of about 3,000 mSv.

The implication of this is that ARS will usually only kill someone who has experienced the impact of over 4,000 mSv. Indeed, many workers at Chernobyl actually received doses above 5,000 mSv and survived. By comparison, the workers engaged in the repair at Fukushima are being carefully monitored to ensure their total exposure does not go above 250 mSv, less than a tenth of the minimum level at which an ARS victim died at Chernobyl. As at 23rd March, 17 workers had received more than 100 mSv of radiation, forty times the yearly radiation received by the typical UK resident and equivalent to ten CT scans. It has been reported that two workers received radiation burns to the legs after exposure in contaminated water to 170 millisieverts per hour doses in Unit 3 on 24 March (2). To date this remains the only known health impact suffered by Fukushima workers.

But what of the longer term dangers to Chernobyl workers who suffered massive radiation exposures? Of those who survived acute radiation syndrome, 19 out of the 106 died between 1987 and 2006. These deaths included 5 cancers. 87 people were still alive in 2006; 9 of them had been diagnosed with various cancers including cases of leukaemia. The problem with using these statistics to draw definitive conclusions is that the numbers of workers affected by extremely high levels of radiation in the Chernobyl emergency are not large enough to give robust data on the long-term impact across wider groups. But the 20 year survival rate of the workers exposed to the greatest radiation – 82% – and the unremarkable percentage either dead of cancer or living with it – 14% in total, within ‘normal’ bounds – suggests that the human body is usually able to recover from even extremely high doses delivered in a short period of time. (This comment is not intended to diminish the severity of the effects of ARS: many of the survivors have suffered from cataracts, sexual dysfunction, skin problems and other chronic illnesses.)

Fourteen healthy children were borne to ARS survivors in the first five years after the accident. There is no evidence of genetic damage passed to future generations.

b) Chernobyl’s wider early impacts

Several hundred thousand workers were involved in the aftermath of the accident (the so-called ‘recovery operation workers’ or ‘liquidators’). These people’s average total dose was about 117 mSv in the period 1986-2005, of which we can assume the large part was experienced in the first months after the accident or at the time the sarcophagus was being placed over the reactor core a couple of years later. The exposures in this group ranged from 10 to 1,000 mSv. The UN Committee on Chernobyl comments that ‘apart from indications of an increase in leukaemia and cataracts among those who received higher doses, there is no evidence of health effects that can be attributed to radiation exposure’. The suggestion here is that the overall impacts on cancer rates among the people with lower doses – but which are still very much higher than would normally be experienced in the UK – is limited.

This conclusion has been attacked by some groups. In particular, Greenpeace published a report entitled The Chernobyl Catastrophe: Consequences on Human Health in 2006 that estimated a figure for total deaths resulting from the disaster that was many times greater than official estimates. Nevertheless, most scientific reports, including all the many official reports into the accident, have concluded that the long-term effects of radiation on the recovery workers, as opposed to the much smaller numbers working inside the plant immediately after the explosion, have been very limited.

After the 1986 Chernobyl disaster large numbers of people in surrounding populations were exposed to the radioactive isotope iodine 131, largely through consuming milk and other farm products. The human body takes up iodine and stores it in the thyroid. Radioactive iodine accumulates in this small area of the body and gives the thyroid gland disproportionate exposure. The effective dose of radiation to the thyroid among some people in the areas affected by Chernobyl fallout ranged up to 3,000-4,000 mSv.(3)The concentration of radioactive iodine in the thyroid has produced large numbers of cases – probably about 7,000 by 2005 – of thyroid cancer among the millions of people in the affected areas. (4) These cases are highly concentrated among people aged less than 18 at the time of the disaster and the impact on adults appears to be very much less or even negligible.(5) The risk of getting thyroid cancer among the most affected group is continuing to rise even now. The implication is clear: severe doses of radiation twenty five years ago produced damage that is still causing cancer today.

Thyroid cancer is treatable and death rates are low. The number of people who had died of thyroid cancer in the affected areas by 2005 was 15. (6) We have been unable to find a scientific assessment of how many people are likely to die in the future from thyroid cancer in the Chernobyl region but the effective treatment for this disease may mean that relatively few of those affected will die. The incidence of thyroid cancer after Chernobyl could have been very substantially reduced if the authorities had acted to provide the local populations with iodine tablets. The effect of taking these tablets is to flood the thyroid gland with normal iodine, reducing the uptake of iodine 131 and thus cutting the dose of radioactivity. Of those countries closest to the nuclear power plant, only Poland seems to have widely distributed iodine, although this is a well understood and simple way of reducing thyroid cancer risk from radioactivity. Second, the authorities could have banned the sale of milk, which is the medium through which most iodine 131 enters the human body and which is why young children appear to have been most severely affected.

It is notable that the authorities around Fukushima are taking an extremely precautionary approach to iodine 131 exposures in the surrounding populations, both in rejecting milk and distributing iodine tablets. Given the experience of Chernobyl, this seems sensible, even though the real risks of exposure and developing cancer as a result are very much lower.

c) Fukuryu Maru fishing boat

In the 1950s and early 1960s nuclear weapons powers like the US, Britain, China and Russia carried out above-ground explosions of atomic bombs in remote areas. (In 1963 these tests provided about 5% of the radiation dose experienced by people in the UK, over five times the impact of Chernobyl, which added less than 1% to the total dose for the average person in 1986, the year of the explosion). One of these tests was in 1954 at Bikini Atoll, one of the Marshall Islands in the mid Pacific. The device turned out to be much more powerful than expected by the US scientists running the experiment, with an explosive power of about one thousand times the 1945 bombs over Japan. As a result the fallout extended well beyond the exclusion zone established by the US, and a Japanese fishing boat was caught in the aftermath of the explosion.

The 23 individuals on this boat received huge doses of radiation – probably averaging between 4,000 and 6,000 mSv. The fishermen suffered severe radiation burns within hours and decided to return to their home port in Japan. Upon their arrival two weeks later their symptoms were recognised to be caused by radiation and they were treated for ARS. Unfortunately, one of the treatments the fishermen received was blood transfusions using blood which was infected with the hepatitis C virus. One of the crew members died a few months after the explosion from liver disease, which may have resulted from hepatitis as much as from acute radiation syndrome. The other fishermen also suffered disease from the hepatitis C in the transfusions and many of them died of liver problems. This experience complicates any medical conclusions that might be drawn about the immediate or long-term impacts of severe radiation exposure.

As at February 2011, it is reported that of the 22 crew members who survived ARS, nine are still alive 57 years later.(7)The average age of these survivors is over 80. These individuals all seem to have had major health problems during their lives, but the cause may well be the transfusions rather than the radiation. Once again, the main implication of the Fukuryu Maru event is that even huge doses of intense radioactivity can cause surprisingly few fatalities.

d) Hiroshima and Nagasaki

The survivors of the atomic bomb blasts were exposed to high but varying levels of radiation. The death rates of nearly 90,000 survivors have been painstakingly studied and compared with people from other cities, so are a valuable source of information from a horrific real-world experiment. Most survivors endured an exposure of less than 100mSv and for these people there is no statistically significant increase in cancer risk. One study shows, for example, that the number of deaths from solid cancers among those who received less than 100 mSv was 7,647, compared to 7,595 that might have been expected based on the experience of populations in other Japanese cities. (8) The increment of 52 deaths is less than 1% above the expected level, and the result is statistically meaningful because it involves a relatively large group.

Above 200 mSv of total exposure, the effect of the radiation becomes a little more obvious but it is not until the dose was greater than 1,000 mSv that a major increase in cancers occurs. Over 2,000 mSv, the risk of a survivor of the bombs dying from a solid cancer is approximately twice the level of risk in non-affected cities. But, even at this very high dose, the number of people dying from solid cancers was 18% of all bomb survivors, to which should be added the 3% of people dying from leukaemia. Compare this, for example, to the UK, where about a quarter of all today’s deaths are from cancer, presumably because of other factors.(9) So it is fair to say that even severely irradiated Japanese atomic bomb survivors appear to be at less risk of developing cancer than normal British people.

e) The effects on soldiers exposed to radiation at tests of nuclear bombs

US and UK research has shown that soldiers experiencing radiation in the aftermath of tests of nuclear bombs, such as at the ‘Smokey’ test in Nevada in 1957 have not had higher than expected incidence of cancer. Although this group seems to have experienced more leukaemia than would have been predicted, the number of other cancers has been lower. The overall death rate from cancer is not higher than in a control group.(10)

Severe exposure over longer time periods

In the previous section we looked at single catastrophic events that caused high doses of radiation, showing that only very high doses, perhaps ten or hundred times the yearly amount received from background sources, substantially affect the risk of future cancers. The same is true of less intense individual events that are repeated many times over a period, even though these events may add up to very high levels of total exposure.

a) Radiotherapies for cancer

Highly targeted bursts of radiation are used to kill cancer cells in radiotherapy. As a result the patient receives very large total doses of radiation over the period – perhaps a month – of the treatment. The amount of radiation received may be as much as 30,000 mSv, many times that sufficient for a fatal dose. This amount does not cause acute radiation sickness because the patient is given time to recover between the doses, allowing damaged non-cancerous cells to recover, and because much of the power is directed at specific internal sites in the body, where the radiation does indeed cause cell death. (That after all is the point of radiotherapy – to kill the cancerous cells in the patient’s tumour.) Some of the radiation reaches other healthy parts of the body and does seem to cause small increases in the likelihood of development of another cancer. But, as the American Cancer Society says, ‘overall, radiation therapy alone does not appear to be a very strong cause of second cancers’.(11) For this reason, radiation overall cures many more cancers than it causes in today’s populations.

b) Workers manufacturing luminous dials for watches

A classic study by Rowland et al in 1978 investigated the incidence of bone cancer among workers painting luminous dials on watches with radioactive paint before the second world war. (12) Workers ingesting more than 10 gray, a measure equivalent to more than 100,000 mSv, had very high incidence of bone cancer. Those taking in less than 10 gray had no cases of bone cancer at all. In his book Radiation and Reason, Oxford University Professor Wade Allison comments that this is ‘a most significant result’ because it shows a clear demarcation between the level of longer-term exposure that seems to cause obviously enhanced cancer risk and that which does not.(13) The threshold – 10 gray – is a level never likely to be now experienced by anyone as a result of nuclear power. It is far greater, for example, than the exposure of any workers fighting the fire at Chernobyl.

The impact of chronic enhanced background radiation

Thus far we have tried to show that only very high levels of radiation, such as are very rarely ever encountered, will tend to produce statistically significant increases in cancer and other diseases. The last category of exposure is to very long-lived elevated levels of exposure. With the exception of thyroid cancer, and high levels of radon gas if the victim also smokes (see below), raised levels of radiation appear to have a small effect on the likelihood of cancer or other diseases. In fact, some people say that small increases in the total amount of radiation received per year have no impact whatsoever on illness rates, or that some dosages of elevated radiation can even be beneficial.

The standard way of viewing the impact of radiation on human health is called the ‘linear, no threshold model’ or LNT. LNT assumes that increased rates of cancer seen in populations such as the atomic bomb survivors can help us predict the degree of cancer arising from radiation at much lower levels of radiation. The theory says that there is a straight line relationship: simply put, if a 1000 mSv dose gives 10% of people cancer, then a 100 mSv total exposure will induce the disease in 1% of the population. With this model of the relationship between radiation and cancer, all incremental doses are bad, from whatever base level, because they add to risk. But the evidence from many studies is that it is difficult to show any unfavourable effect from elevated levels of exposure. For example, people living at higher altitudes in a country generally get more background radiation than those at sea level because of greater cosmic ray density. However we could find no study that showed that these people experience more cancers or other radiation-related diseases. In Ramsar, Iran, naturally high background radiation delivers a hefty dose of 260 millisieverts per year to local residents, a hundred times higher than 2.7 mSv/yr experienced by the average UK citizen, and also ten times higher than doses normally permitted to workers in nuclear power stations. However, there is no observed increase in cancer in this or any other area where levels of background radiation are up to two orders of magnitude higher than normally observed. (14)

The LNT model is controversial because it is based on statistical assumptions (which reflect a very precautionary approach) rather than observed biological effects of radiation – it would predict higher rates of radiation-induced cancer in Ramsar where radon levels are exceptionally high despite no evidence of these occurring in reality. It has been criticised because the body can repair most DNA damage caused by radiation, and cells have mechanisms that perform this healing role on a constant basis. An analogy would be blood loss: whilst losing half a litre of blood (such as a blood donor might) causes no health impacts whatsoever, losing 5 litres of blood would be fatal. In this case clearly there is a threshold for harm, so a ‘linear no threshold’ assumption is biologically incorrect.

There is one important exception, however, to the rule that increased background radiation presents no additional health problems. In many parts of the world, particularly those with granite rocks close to the surface, radon gas represents the most important source of natural exposure to radiation. Radon is a short-lived radioactive element that arises from the decay of fissile uranium. As we said above, for the UK population as a whole the average total absorption of radiation is about 2.7 mSv per year but many people in Cornwall receive much more, largely from the pooling of the gas in their homes and workplaces.

Studies have suggested that this increase has a very small effect on the incidence of most cancers and other illnesses, although the research is not yet definitive about the precise relationship between radon gas exposure and rates of cancer. However, radon does have an observed effect on lung cancer occurrence, particularly among smokers, and this effect increases with the typical densities of radon in the home. In homes with the highest radon levels, the chance of a smoker getting lung cancer rises from about 10% to about 16%, according to one study.(15)

The US National Cancer Institute concludes: “Although the association between radon exposure and smoking is not well understood, exposure to the combination of radon gas and cigarette smoke creates a greater risk for lung cancer than either factor alone. The majority of radon-related cancer deaths occur among smokers.” (16)

a) The impact of living near a nuclear power plant

Several studies have shown ‘clusters’ of solid cancers and of leukaemia around nuclear installations in the UK and other countries, although the vast majority show no relationship between the two. (17) In particular, the incidence of childhood leukaemia appears to be marginally higher than the national average in some areas close to nuclear sites and at some locations the rate of such cancers appears to rise with closeness to the site. (This suggests a risk that is related to the dose experienced by the child, and thus in line with LNT theory).

This is a worrying finding and much research has tried to find out why the chance of cancer appears to be slightly higher in these places. But the issue is this: why should there be an increased risk of cancer around nuclear sites when the aggregate level of radiation exposure is so low compared, for example, to parts of Cornwall? Similarly, why do we not see higher incidences of childhood cancers around large coal-fired power stations, which emit far higher levels of radiation than nuclear sites as a result of the radioactive material contained in the coal being dispersed from the chimneys? And, as a separate point, why have some of the rates of higher-than-expected cancer fallen at some sites when radiation levels have remained approximately constant?

Scientists working on this issue have no convincing explanation for the higher rate of childhood cancers in these clusters. But many experts now believe that what is known as ‘population mixing’ may be responsible for the observed increase. Mixing occurs when a new population, such as those recruited to construct or operate a nuclear power station, arrives in the area. This may, one theory goes, cause unusual infections in the area and the end result of some of these infections may be childhood leukaemia.

To repeat: the clusters of cancer around some nuclear sites for some periods of time appear to suggest a worrying relationship between nuclear power stations and cancer. But the relatively low levels of radiation at these places, compared to around coal-fired power stations or areas with high natural background radiation, makes it extremely difficult to see how radioactivity could cause the higher levels of cancer.

b) Workers in defence industries exposed to radiation

Oxford’s Professor Wade Allison reports on a survey of a huge number (174,541) workers employed by the Ministry of Defence and other research establishments.(17) This study found that the workers received an average of 24.9 mSv above background radiation, spread over a number of years. But even though this amount is small when expressed as a figure per year, the large number of people in the study should enable us to see whether there is any effect on cancer incidence of low levels of incremental exposure. (Any increase will be much more statistically significant than any additional cancers in smaller groups.) In fact the survey found that the workers suffered from substantially less cancer than would be expected, even after correcting for factors such as age and social class. (The mortality rate for all cancers was between 81% and 84% of the level expected). This suggests that the increased radiation they experienced delivered no additional cancer risk at all.

More on Fukushima

How dangerous are the levels immediately next to the Fukushima boundary fence? The power plant operator TEPCO issues data every day from measurements taken at one of the gates to the plant. (18) On March 27th, about two weeks after the accident, the level had fallen to about 0.13 mSv an hour – and was continuing to decline at a consistent rate. (In the course of writing this article, the number rose to about 0.17 mSv an hour but then started to decline again.) If someone stood at that point for a year, he or she would receive about 1.1 Sv. This is a very high level – about 400 times background level in the UK – but would not necessarily have fatal effects. Professor Allison argues in a post on the BBC News web site that a figure of 100 mSv a month, or 1.2 Sv a year, would be a good level to set as the maximum exposure for human beings before real risk was incurred. (19)

Radiation intensities obey the inverse square rule: as we move away from the source of radiation, the level of radiation will decrease by the square of the distance. (This excludes the impact of fallout from an explosion or of radiation carried in plumes of steam). Thus a reading obtained 2km away from the plant will be one hundredth of the level at 200m distance. In other words if the plant were in the UK, with its average background dose of around 2.7 mSv per year, and the monitoring at the Fukushima gate is 200 metres from the source of the radiation, then the level of incremental radiation would be no greater today than the background level at a distance of 4 kilometres from the plant. Since much of the radiation emanating from Fukushima is iodine 131, which has a half life of 8 days, the level of contamination of the surrounding area will continue to fall rapidly.

The effect on water supplies in Tokyo and elsewhere

The authorities in Tokyo recommended in mid-March that infants were not provided with tap water after levels of radiation rose to higher than usual levels. The peak level reached at a Tokyo water supply plant was 210 becquerels per litre and this prompted the decision – anxious parents were provided with bottled water instead. (A becquerel is a measure of the number of nuclear fissions, not a measurement of the dose of radiation absorbed). Children will be most susceptible to the effects but an infant drinking this water for a year will absorb the equivalent of about 0.8 mSv of radiation, or less than a third of normal absorption by an adult in the UK. (20)

There are significant divergences between different country approaches to radiation in water. The European limit for radiation in public water supplies is set at 1,000 becquerels per litre, nearly five times that declared ‘unsafe’ for infants by the Tokyo authorities. (21) In one study carried out by the British Geological Survey in Tavistock, Devon, private water supplies were found to contain as much as 6,500 becquerels per litre, and no ill effects have been reported.(22)

Although this is not directly stated, we can assume that the large majority of this radioactivity in British water is derived from the decay of radon. This means that in the UK, the level is likely to remain at a roughly consistent level year after year. But in Japan the radiation is more likely to be from the decay of iodine 131, which has a very short half life. So the radiation in Japanese tap water will quickly fall, and already appears to be doing so. Thus the risk of any radiation damage, even for very young children, from drinking tap water in Tokyo is not just small but infinitesimal.


Overall the average UK person ets approximately 0.2% of his or her radiation exposure from the fallout from nuclear plants (and from nuclear accidents) and less than 0.1% from nuclear waste disposal. This compares to about 15% from medical imaging and other medicinal exposures and about 10% from the natural decay of potassium 40 and carbon 14 in the body. Naturally-occurring radon is many hundreds of times more important as a source of radiation than nuclear power stations and nuclear fallout. Even for those who believe in a direct linear relationship between radiation levels and the number of cancer deaths, the effect on mortality of normal operation of nuclear power stations would be impossible to discern statistically and in our opinion is likely to be non-existent.

It can only be in the event of a serious accident that we have any reason to be really concerned about nuclear power. We have tried to show in this article that even when such accidents occur the effects may be much less extensive than many people imagine, particularly given the constant media coverage devoted to Fukushima. Chernobyl killed 28 people in the immediate aftermath of the disaster. All these people had experienced huge doses of radiation in a short period. Mortality since the accident among the most heavily dosed workers has not been exceptionally high. And many studies after Chernobyl have suggested that – with the exception of the thyroid variant – cancer rates have only increased very marginally even among those exposed to high doses of radiation after the accident.

While reported rates of other, non-cancer, illnesses may have risen, researchers seem to think that much of this rise is due to the impact of other factors, such as the need to evacuate from the area, increased smoking, drinking and other risky behaviours, or even the wider effect of the breakup of the Soviet Union soon after the accident. There is substantial evidence, as the UN reports on Chernobyl attest, that the psychological impacts of fear of radiation far outweigh the actual biological impacts of radiation. Thus, misinformation about exaggerated dangers of radiation is actually likely to be harmful to large numbers of people – a point which should be borne in mind by anti-nuclear campaigners. This appears certainly to have been the case after Chernobyl and Three Mile Island (in the latter case the radiation released was negligible, but the political fallout immense).

We hope that a more rational sense of risk and an appreciation of what we have learned from past experience will prevent the repeat of this experience after Fukushima. It is important to appreciate that whilst radiation levels at the boundary fence are still high, they are dropping sharply. Even today, March 28th, the radiation exposure of a person a few kilometres from the plant (in the precautionary exclusion zone) is likely to be lower than experienced by many people living in Cornwall or other places with high radon density. Similarly, the peak levels of radiation in the water supply have constantly been well below levels regarded as safe in other parts of the world.

No technology is completely safe, and we don’t wish to argue that nuclear power is any different. But its dangers must be weighed against the costs of continuing to operate fossil fuel plants. Just down the road from us is Didcot A power station, a large coal-burning plant with poor pollution control and therefore with substantial effects on local air quality, as well as more substantial emissions of radiation than from any UK nuclear power station and a Co2 output of about 8 million tonnes a year. We offer a view that Didcot has caused far more deaths from respiratory diseases than all the deaths ever associated with nuclear energy in the UK, and that coal power is a far more legitimate target of environmental protest than nuclear.

Chris Goodall and Mark Lynas, 29th March 2011.

(With many thanks to Professor Wade Allison for his help on the research for this article. All errors are ours.)

1 Much of the data in this section is taken from Sources and Effects of Ionizing Radiation, United Nations Scientific Committee on the Effects of Atomic Radiation, 2008 report to the General Assembly, published February 2011.


3 Prof. Wade Allison, Radiation and Reason, page 100

4 Taken from Sources and Effects of Ionizing Radiation, United Nations Scientific Committee on the Effects of Atomic Radiation, 2008 report to the General Assembly, published 2011

5 Sources and Effects of Ionizing Radiation, United Nations Scientific Committee on the Effects of Atomic Radiation, p 19

6 Sources and Effects of Ionizing Radiation, United Nations Scientific Committee on the Effects of Atomic Radiation, 2008 report to the General Assembly, published 2011, page 15

7 Interview report,

8 Preston Dale et al. (2004) Effect of Recent Changes in Atomic Bomb Survivor Dosimetry on Cancer Mortality Risk Estimates, Radiation Research.

9 National Statistics UK

10 American Cancer Society,

11 American Cancer Society 1

2 Rowland et al, 1978: Dose-response relationships for female radium dial workers, Radiation Research, 76, 2, 368-383

13 Wade Allison, Radiation and Reason, 2009

14 M. Ghiassi-nejad et al., 2002: Very high background radiation areas of Ramsar, Iran: preliminary biological studies, Health Physics, 82, 1, 87–93

15 British Medical Journal

16 American Cancer Society,

17 Details of some of these studies are discussed in a 2005 report on the Committee on Medical Aspects of Radiation in the Environment available at

18 Wade Allison, Radiation and Reason, page 127


20 Professor Richard Wakeford of the Dalton Nuclear Institute, quoted on the BBC News web site at

21 This limit is what is called an ‘action level’. That is, the authorities expect something to be done when higher levels are observed

22 Neil M. MacPhail, A radon survey of Ministry of Defence occupied premises in Her Majesty’s Dockyard, Devonport, unpublished MSc dissertation, University of Surrey 2010.