You are here

No radiation dose too low to cause any effects

Nuclear Monitor Issue: 

D(December 19, 1997) uring a recent IAEA/WHO conference the argument was heard that small doses of radiation are not harmful, but in fact beneficial. Another claim was that there is a threshold below which radiation is safe. However, research since the discovery of radiation, now more than 100 years ago, show that the risks are still underestimated.

(483/4.4804) WISE Amsterdam -The debate on the risks of low-level radiation is still going on. During a conference, held November 17-21 in Sevilla, Spain, organized by the IAEA and the World Health Organization (WHO), a vocal, pro-nuclear minority argued that the radiation protection "establishment" is suppressing data that shows a small amount of radiation is not harmful but beneficial. They don't accept the linear non-threshold model that says there is no threshold dose below which radiation is harmless. The non- threshold theory is the central issue in the debate, whose outcome has huge economic, political and social consequences. This minority succeeded in setting the tone and subject of the conference. They argued that there was ample proof for the theory of "adaptive response", which says that radiation doses have positive health impacts. In fact, there is no proof for this theory, so it is rather religion than science.

The International Commission for Radiation Protection (ICRP), a self-appointed, voluntary commission that makes recommendations on radiation limits, accepted the linear non- threshold in 1991. The theory was further supported by the latest data from the continuing study of Japanese A-bomb survivors. These show significantly higher risks with doses of around 50 milliSievert (mSv). The survivor population could be divided into three groups that received doses of no more than 20, 50 or 100 mSv. There were also excess cancers in the 20 mSv group. Women were shown to be at twice the risk as men at the same age. Children under 10 were at the greatest risk. This new result is another blow for the US Health Physics Society, which claims there is no proof of negative effects from radiation doses below 100 mSv (in most countries the permitted annual dose for workers in the nuclear industry is 20 mSv - in the US, 50 mSv - and for the public 1 mSv). An opposite, "super-linear" theory was suggested by the Japanese A-bomb researchers. It says that the incidence of some cancers increases more steeply at low-level radiation doses. The scientific explanation is that radiation damage to DNA is likely to cause double-strand DNA breaks, which are more difficult to repair than single-strand damage caused by chemicals. A misrepair of the DNA may cause the cell to become a cancer cell. Another assumption of the pro-nuclear minority is also under attack. It argues that because emissions from nuclear installations are far below natural background radiation doses, there is no reason to worry. This argument is based on the assumption that natural radiation is safe, which is not. It is calculated that background radiation in the UK, US and other countries cause some 4%-5% of all cancer deaths. The ICRP also does not use background radiation as a criterion for acceptable doses.

In 1990 the ICRP recommended in its Publication 60 to lower the radiation dose limit. For workers it was lowered from 50 to 20 mSv a year and for the public from 5 to 1 mSv annually. These limits are now adopted by the European Union (EU) countries, even by France and the UK, the WHO and the IAEA. Current US standards for workers are maintained at the old 50 mSv, but for the public the new ICRP standard of 1 mSv was adopted. Many "third world" countries and Russia did not adopt the ICRP 1990 limits.

Global Collective Dose
The concept of Global Collective Dose tries to calculate the damage of radiation doses to a population, from the operation of a nuclear facility over the lifetime of the radionuclides it releases. Collective dose estimates are very uncertain, because size and behavior of future populations are unknown and unforeseen changes in habits of humans and in radiation damage assumptions make any prognosis of death rates unreliable. Still, the concept is useful in comparing risks of different nuclear facilities, as long as one is aware of the uncertainties of the estimates. The term person-Sievert is used to measure the population dose, which is the sum of individual doses in a defined population. For example, 50 persons who each received 20 milliSievert, collectively received 1 person-Sievert. The global collective dose from radionuclide releases from the British reprocessing plants at Sellafield is estimated at about 4,100 person-Sievert per year. This is very much compared to the global collective dose from a nuclear reactor, estimated at 45 person-Sievert per GigaWatt-year (the annual maximum production of a 1000-MW plant). This estimate is based on information in a 1993 report from the UN Scientific Committee on Effects of Atomic Radiation (UNSCEAR). Based on present assumptions of radiation doses, all atmospheric atomic bomb tests caused a global collective dose of 30 million person- Sievert. Worldwide nuclear power production in 1989: 400,000 person-Sievert. Ten years of reprocessing at Sellafield: 40,000 person-Sievert. The ICRP also adopted the use of collective doses. Other pro- nuclear organizations, like the US Health Physics Society, are very much opposed to the measurement of collective doses, as is the US Nuclear Waste Advisory Committee. Their problem is that as knowledge of radiation effects continues to grow, it becomes more and more clear that those effects must be taken more seriously than ever thought. Both environmental agencies and activists can use this concept, despite its uncertainty, in their assessments of both civil and military nuclear operations and in criticizing Environmental Impact Statements. They can demand that global collective doses for example from uranium mines are calculated over the lifetime of uranium tailings (some 730,000 years) and that lifetime monetary costs should be calculated for all nuclear discharges. Or whether anti-nuclear activists and nuclear industry like it or not, global collective doses will play an increasing part in decision making and licensing of nuclear facilities.

The value of cancer deaths
In a capitalistic society it is normal to calculate the costs of death. What's the value of a human life, how much money are private enterprises prepared to pay to prevent one cancer death? The calculation of the risk of dying from cancer caused by a certain amount of radiation depends on the assumption of a risk factor per Sievert. The US Nuclear Regulatory Commission (NRC) assumed a value of US$1 million per person-Sievert in 1991 to determine whether it was economical to reduce radiation doses for workers. The US nuclear industry assumed a much lower value of US$0.15 million. This is another reason why the nuclear industry is not glad about the concept of collective doses: it can be used to calculate other costs related to nuclear energy, which tend to become higher each decade. Over the past 100 years, both the limit of allowed radiation doses and the risk factor per unit of radiation dose have been tightened as more knowledge about radiation effects was acquired. New data and theories such as the super-linear relationship between radiation and risk will probably make them more strict in the future. This will raise the environmental costs of existing and proposed nuclear facilities and help to make them even more uneconomic.


  • Bulletin of the Atomic Scientists, November/December 1997
  • Nucleonics Week, 13 November 1997

Contact: Institute for Energy and Environmental Research IEER, 6935 Laurel Avenue, Takoma Park MD 20912, USA
Tel: +1-301-270-5500; Fax: +1-301-270-3029