Institute for Energy and Environmental Research
Proposed standards are thus all based on the assumption that some radioactivity from a repository will reach the environment. Given this assumption, there are several possible approaches to developing standards.
One approach would be to establish upper limits on the "acceptable" health risks for both individuals and the population as a whole.13 An acceptable health risk can be translated into a radiation dose limit, which can be put together with assumptions about environmental transport of radioactivity in order to establish repository performance standards. The performance standards can then be used to establish selection guidelines to evaluate the suitability of potential repository sites. This results in what is sometimes called "health-based" (or "risk-based") standards.
Another approach results in "technology-based" standards. This approach starts by considering what is possible, based on the current state of scientific knowledge, the best currently available technology, and a reasonably good geological site. Technology- based performance standards can then be developed which are intended to ensure that a good site is chosen and that the technology used to construct a repository is appropriate. They are oriented towards assuring that compliance with the standards can be measured and achieved. This can be done independent of the health risks that would arise from the standards that result, which might be either large or small.
We believe that the ideal route to environmental protection would rely on the first approach, which takes the health of human beings as its primary standard. The second approach essentially relies on circular reasoning: standards are set so as to ensure that the standards can be met. When taken to its extreme, this represents an attitude that environmental contamination is an inevitable result of technological progress, that human beings have no choice but to live with the consequences, and that environmental regulation is limited simply to a kind of "sweeping up" after the mess is made.
The first approach, however, as an across-the-board model, provides a much more complete conception of the possibilities for environmental preservation; comprehensively applied, it would extend health- and environmentally-based criteria to decision-making about which technologies are selected in the first place to fill a given societal need, and how manufacturing and production processes are designed.
Unfortunately, as we have discussed elsewhere, little thought went into the consequences during the years when nuclear technology was under development, and we are now confronted with a situation where the waste has been produced. Under such constrained circumstances, it is arguable that the second technology-based approach is as good an approach as can be taken. This is, in fact, the approach which the EPA took as its primary one in developing its standards.
Unfortunately, both of these approaches were essentially abandoned by federal law. The 1982 Nuclear Waste Policy Act asked the DOE to submit a list of potentially acceptable repository sites within six months of promulgation of the law. It did not require the EPA, however, to come up with final standards for acceptable dose limits -- logically the first step -- until six months after the DOE's initial repository list was submitted. Further, the NRC was to establish technical performance standards (theoretically consistent with the EPA standards) for the repository about the same time that the EPA standards were supposed to be issued.
The way the process has worked in practice is even worse than provided for in the provisions of the 1982 waste policy law. The EPA's final standards were delayed, and therefore were not issued until 1985 over two years after the final NRC standards.14 Further, the EPA standards were challenged in court by several states and environmental groups on the grounds that, among other things, they were too lax, and violated provisions of the Safe Drinking Water Act. The court agreed that the standards violated the Safe Drinking Water Act, and vacated them, requiring the EPA either to bring the standards into compliance with existing law, or explain why the inconsistency exists.
The EPA is now in the process of considering the issues raised by the court which vacated the standards, and plans to issue new standards in 1992. Because the new standards are likely to be similar in many respects to the 1985 standards, we will review the vacated EPA standards below along with the NRC regulations.
1985 EPA Standards
The EPA environmental standards for waste repositories15. set out long-term containment requirements that limit releases of radioactivity to the environment for 10,000 years. These are expressed in terms of radionuclide release limits for the rate of release of individual nuclides in the waste to the environment.
Corresponding to the radionuclide release limits is an overall population dose limit extending to 10,000 years; individual doses are limited for only 1,000 years. The overall population dose from all radioactivity escaping from the repository may not cause more than 1,000 premature cancer deaths over the 10,000-year period (or, one extra fatal cancer, on average, every 10 years).16 The individual dose to any member of the public from radioactivity escaping from the repository is limited to no more than 25 millirems per year. 17 After the first 1,000 years, only the overall population dose limit remains, and there is no cap on individual doses.
There are also groundwater protection requirements which limit the concentrations of radioactivity from the repository in groundwater to such that persons who draw all of their drinking water from that source of groundwater would receive not more than four millirems per year. Like the individual dose limits, these standards apply only for a period of 1,000 years after disposal.
As pointed out in a 1983 report by a panel of the National Academy of Sciences (NAS) (which criticized the draft EPA standards two years before the final ones were issued), 18 as well as by the EPA itself, 19 the 1,000- and 10,000-year time frames are short, given the long-term risk posed by some radionuclides. The NAS panel said it was possible for groundwater to become contaminated long after 10,000 years, and that the population dose received after this time could actually increase beyond the EPA limits, even if these limits were not exceeded within the first 10,000 years. The panel pointed out that some long-lived radionuclides would be capable of delivering very high doses from groundwater for many tens of thousands of years. 20 For example, according to one set of assumptions about radionuclide release from reprocessing waste made by the National Academy of Sciences, significant doses from the radionuclide lead-210 could peak shortly before 100,000 years, and continue for long afterwards. The possible doses from cesium-135 peak at almost one million years. 21
Another problem with the 1985 EPA standards is that after 1,000 years, only overall population dose is limited; there is no limit after this time on the maximum dose an individual might receive. 22 This should be of particular concern given that the NAS panel found that under certain conditions radionuclides from a repository could deliver annual individual doses from neptunium-237 of up to 10,000 rems -- a lethal dose. 23 According to the1985 EPA standards, this could happen without any violation of the standards, as long as the total number of people exposed was within the population limit.
In its study, the NAS panel proposed a slightly different basis for standards. Rather than limiting overall population exposure and individual exposures for only 1,000 years, the Panel advocated an individual dose-based standard for all future times:
[T]he most meaningful and useful form of the criterion is the annual or lifetime radiation dose to an individual exposed at some future time to radionuclides released to the environment from a geologic repository. We have adopted as our criterion an annual dose of 10 -4 Sv [10 millirem] to an individual, averaged over his lifetime, calculated at all future times. 24
Missing from this recommendation, however, is the concept of a population dose limit. With the addition of a population dose limit, such an approach is a sound one, because it would place an upper limit to the amount of health risk to any individual from radioactive waste for all time, and it would also limit the total number of deaths as well.
The EPA is now moving further in this direction with its new draft standards, which retain the population dose limit, extend the individual dose limit to the entire 10,000-year time-frame, and may also lower the individual dose limit to 10 millirems. 25 Whether these changes make it into the actual final rule remains to be seen.
However, it appears highly unlikely that the EPA will promulgate standards which address the criticisms pertaining to the shortness of the overall 10,000-year time frame. 26 The basic reason for this is that the current state of knowledge about geological processes and risk assessment simply does not allow confidence in predictions or control of the situation beyond this time frame. As it is, 10,000 years is probably pushing the limits of what is possible, given the current knowledge base.
This is an illustration of the fundamental problems posed by the whole nuclear waste dilemma. Nuclear waste will present a significant hazard for far longer than the 10,000-year regulatory time-horizon currently envisioned. Rather than address this defect, let alone that it was not even considered before proceeding with nuclear technology in the first place, current thinking simply accepts such shortcomings -- indeed, even accepts dose-levels which may actually rise significantly after the 10,000 years, according to the NAS and the EPA -- because we cannot currently do better.
NRC Standards
The Nuclear Regulatory Commission promulgated its high-level waste repository performance standards in 1983.27 These standards require:
Thus, it can be seen that for a number of aspects of repository performance, federal standards as they currently exist are inadequate, and provide little assurance -- especially over the long-term -- that the ¬environment or future public health will be protected for the length of time indicated by the characteristics of the radioactive materials.
Return to Publications Main Page
Return to IEER Home Page
Institute for Energy and Environmental Research
Comments to Outreach Coordinator, ieer@ieer.org
Takoma Park, Maryland, USA
Last Updated October, 1996
11. NAS 1983, p. 11.
12. EPA 1985, p. 38066.
13. It is important here to understand the concepts of individual and population risks. An individual risk is often expressed in terms of the chance that an individual might get a fatal cancer (for example, a typical risk limit for individuals in environmental regulation is one in one million). A population risk is often expressed in number of expected deaths. Although the risk to individuals is important to consider, the risk to the population as a whole is just as meaningful from a public policy perspective. This is because the total population risk is what indicates how many people will die as a result of a given risk. Thus, population risk provides a basis for arguing that a "small" individual risk of death imposed on a large number of people is morally worse than the same-sized individual risk imposed on a small number of people. The risk to each exposed individual is the same in both cases, but in the first case, a larger number of people will die. Conversely, even a small overall population risk, if borne primarily by a small number of people, can result in huge risks to each of the few individuals exposed. For example, if an environmental risk sufficient to kill one person is spread evenly among 10 people, each will have a one in 10 risk of dying. Spread among one million people, each individual will have a one in a million risk. In each case, the population risk is the same (one death), but we would generally consider the risks to the arbitrarily exposed individuals to be unacceptably high in the former case. A humane public health and environmental policy will minimize both types of risk, individual and population.
14. EPA standards were published on 19 September 1985. The NRC standards were published on 21 June 1983.
15. The EPA standards were incorporated into federal regulations at 40 CFR 191 (EPA 1985).
16. EPA 1985, p. 38069.
17. EPA 1985, p. 38068. This is the same as the dose limit allowed to individual members of the public from the nuclear power fuel cycle by EPA standards at 40 CFR Part 190.
18. NAS 1983.
19. EPA 1985, p. 38076.
20. NAS 1983, pp. 227-228.
21. NAS 1983, Figure 9-1, p. 256.
22. Although, as we note below, the new EPA standards may extend the dose limit to 10,000 years.
23. NAS 1983, p. 225.
24. NAS 1983, p. 212.
25. EPA 1991, pp. 18, 48.
26. As discussed in EPA 1989, and as is clear from the proposed new rule, EPA 1991.
27. At 10 CFR Part 60.
28. This implies that the release of radioactivity from the repository is controlled by the NRC standards for about 100,000 years (after this time, 100,000 parts out of 100,000 i.e. all the radionuclides are allowed to have leaked).