Evaluation of the Impact and Inter-Generation Risk Transfers Related to the Release and Disposal of Radioactive Waste from the Nuclear Fuel Cycle: A Methodological Exercise P. Croüail 1, T. Schneider 1, A. Sugier 2 1 CEPN BP n 48, 92263 Fontenay-aux-Roses Cedex France 2 IPSN BP n 6, 92265 Fontenay-aux-Roses Cedex France ABSTRACT Reflections about the consequences of decisions involving the long term raise various theoretical and complex issues related to the validity of the quantitative assessment of what could be future risks, but also to the ethical position we are adopting towards future generations. In this perspective, decision-making in the field of radioactive waste management, with a view to maintaining present and future radiation exposures as low as reasonably achievable, implies being able to discriminate among alternative options, i.e., being in a position to evaluate the differences in terms of radiological impacts between the options. Because of the complex and multidimensional nature of the distant future consequences of waste management options, their comparison involves expressing these impacts using various aggregated or disaggregated indicators, taking into account the time during which radionuclides remain in the environment and their local, regional, or world-wide dispersion. This paper is an attempt to contribute to the development of such a framework. It is mainly focused on the risk transfer dimension inherent to waste disposal management. Any decision to protect people now against the potential impacts of radioactive releases into the environment leads inevitably to the exposure of current generations and potentially of future generations. In this perspective, one of the key questions related to waste management is to decide on the best compromise between present dilution-dispersion into the environment or concentration in surface or underground disposal sites. The objective of this paper is to illustrate, the relative impact of different waste management options, focusing especially on inter-generational risk transfers. For the sake of the exercise, calculations have been performed for six particular radionuclides and for the current waste management options combining underground disposal and releases as well as for extreme alternative waste management options, which have never been implemented or envisaged as such in reality: i.e. underground disposal of C-14, Kr-85 and I-129 or, total release, either in the sea or in the atmosphere, of Cs-137, Pu-239 and Np-237. As the alternative waste management options analysed are of a one-off nature, it leads to some extremely high exposures such as for the integral release of Cs-137. But between total disposal and total release, many options are available, which have to be evaluated in an ALARA perspective. Based on these evaluations, the indicators characterising the radiological impacts are discussed taking into account the time and space dimensions. Individual and collective doses associated with the present and future generations are presented for the different options and their usefulness for evaluating these options is considered. In this perspective, this kind of time and space presentation provides useful information for helping decisionmaking processes and facilitating the communication with non-specialists. INTRODUCTION Reflections about the consequences of decisions involving the long term raise various theoretical and complex issues related to the validity of the quantitative assessment of what could be future risks, but also to the ethical position we are adopting towards future generations. However, from a practical point of view, a responsible attitude implies using as best we can all available information on the possible consequences of our present actions, even if this information does reflect the lack of knowledge in the assessment of consequences far in the future. In this perspective, decision-making in the field of radioactive waste management with a view to maintaining present and future radiation exposures as low as reasonably achievable implies being able to discriminate among alternative options, i.e., being in a position to evaluate the differential impacts of potential consequences. Because of the complex and multi-dimensional nature of the distant future consequences of waste management options, their comparison involves expressing this impact using various aggregated or disaggregated indicators, taking into account the time during which radionuclides remain in the environment and their local, regional, or world-wide dispersion. Two types of arguments have been raised [1] against the use of the collective dose to assess the impact of 1
waste management options: - the aggregation on huge populations of extremely low individual doses such as those resulting from waste management, which leads to significant collective doses, together with the use of the linear, no threshold dose-effect relationship to assess the corresponding potential health impact, - the existence of increasing uncertainties, especially with time, which weaken the relevance of estimating impacts in the distant future. These criticisms are highly valid, although any responsible decision-making process on waste management options cannot avoid taking into account, for each option, the magnitude of radionuclide releases into the environment, the time during which these radionuclides remain a source of exposure and how widely they are dispersed geographically, i.e., how large is the size of the exposed population. In view of these difficulties, the evaluation of impacts can be performed either qualitatively by describing, in the most detailed way, the various aspects presented above, or partly quantitatively. Quantification must rely on the best current scientific knowledge of the various mechanisms involved in the radionuclides dispersion and exposure of human beings, as well as the best guess as to the behaviour and size of future generations. So far, the use of the individual dose concept remains the most appropriate performance indicator to assess whether any waste management option is likely to put people in danger (deterministic effects) or keep them at a tolerable level of exposure. The collective dose concept allows evaluating whether these options have a significant impact on public health (stochastic effects). The problem is less the accuracy of these concepts than the difficulty of using them in an adequate and comprehensive decision-aiding framework. Any approach relying on the cut-off of collective exposure either in time, space, or individual level of exposure is misleading in terms of understanding and communication with the public. For example a cut-off at 30 µsv for the individual effective dose, will lead to the impossibility of discriminating options of disposal for which dose in normal situation are very lower. Thus, even if health effects can be considered as uncertain at this level, the individual dose concept remains a simple and useful indicator in the decision-making process. This paper presents a preliminary work, mainly focused on the risk transfer dimension inherent to waste disposal management. Decisions to protect people now against the potential impacts of radioactive releases into the environment leads to the exposure of current generations and potentially of future generations. In this perspective, one of the key questions related to waste management is to decide on the best compromise between present dilution-dispersion into the environment or concentration in surface or underground disposal sites. The objective of this paper is to discuss the use of the indicators related to individual exposure, focusing especially on inter-generation risk transfers. For the sake of the sensitivity assessment, the evolution of the indicators has been explored for the current waste management option of the French nuclear fuel cycle, combining underground disposal and releases, as well as for extreme alternative waste management options, which have never been implemented or envisaged as such in reality. THE NUCLEAR FUEL CYCLE WASTE MANAGEMENT SYSTEM The nuclear fuel cycle consists of a succession of stages, which may differ depending on the type of reactor (PWR, GCR, CANDU, etc.) and the various options that exist for fuel management (with or without reprocessing of spent fuel, recycling of fissile matter, etc). Each of the stages produces radioactive waste in liquid, gaseous or solid forms. The methods of managing this radioactive waste can vary considerably from one facility to another, with the choice of a waste management option depending mainly on the characteristics of the waste being considered. The waste produced is composed of a group of radionuclides, each of which having its own particular characteristics, in terms of both chemical and physical properties (radioactive half-life, mode of disintegration, environmental behaviour, chemical form, radiological impact). As a result, the management of radioactive waste necessarily implies specific management for each radionuclide, or family of radionuclides. Figure 1 illustrates schematically [2] the different management methods for waste produced by the nuclear fuel cycle. Most of the waste is submitted to an initial processing phase. This processing can differ considerably depending on the radionuclide concerned: it may be partial elimination as a result of radioactive decay or various physical and chemical processes. 2
Following this processing, a portion of the radioactive waste can be directly released into the environment in the form of liquid or gaseous effluents ( dilute and disperse principle). In this case, the radionuclides are immediately dispersed into the biosphere, the extent of this dispersion differing according to the geographical characteristics of the site. Waste which is not released into the environment is immobilized with a view to disposal ( concentrate and retain principle). This disposal can be envisaged either on the surface for low level waste (LLW) or intermediate level waste (ILW), or underground for high level waste (HLW). The radioactivity contained in the confined waste is held for a period which can stretch from some hundreds to some hundreds of thousands of years, during which time part of this activity is eliminated as a result of radioactive decay. Then, following this retention phase, there is a gradual release into the biosphere of the remaining radionuclides (delayed dilution/dispersion), usually via underground water. Nuclear fuel cycle Treatment Wastes Dilute/Disperse ENVIRONMENT Early or Immediate releases - atmospheric - liquid Concentrate / Retain Disposal - LLW/ILW surface - HLW underground Dilute/Disperse Delayed releases Figure 1 : Dilute-Disperse and Concentrate-Retain Principles Both strategies are not exclusive, and the current management is mixing the dilution and concentration principles for most of the radionuclides (see Table 1), with a very few exception (for example, krypton 85). In order to illustrate the sensitivity of impact indicators associated with the waste management strategy (dilution or concentration), calculations have been performed for six particular radionuclides and for the current waste management options combining underground disposal and releases as well as for extreme alternative waste management options: underground disposal of C-14, Kr-85 and I-129; total release, either in the sea or in the atmosphere, of Cs-137, Pu-239 and Np-237. As these alternative waste management options analysed are of a one-off nature, it leads to some extremely high exposures such as for the integral release of Cs-137. Nevertheless, between total disposal and total release, many options are available, which have to be evaluated in an ALARA perspective. 3
Current management C-14 Kr-85 I-129 Gaseous releases 37.4 % 100 % 1.6 % Liquid releases 12.5 % 0 % 86.7 % Surface disposal 49.9 % 0 % 7.4 % Underground disposal 0.2 % 0 % 4.2 % Cs-137 Pu-239 Np-237 Gaseous releases < 0.0001 % < 0.0001 % 0 % Liquid releases < 0.0001 % 0.01 % 0.05 % Surface disposal < 0.0001 % 5.9 % 0.2 % Underground disposal 99.99 % 94 % 99.7 % These numbers are very typical of the present French nuclear cycle, operating during 60 years, and considering that an underground disposal site in granite is available. Table 1. Distribution of the activities involved in the current management of nuclear cycle RADIOLOGICAL CRITERIA APPLIED TO SCENARIOS IN NORMAL SITUATIONS Different questions arise from Table 1: Is the current management optimised? On what considerations is it based? If another one is envisaged, what are the useful indicators to decide to change? For some of the radionuclides, the answers are evident, for example there are technical difficulties, even impossibilities, to trap, retain or confine a radionuclide due to its physical and chemical properties. Toxicity and radiotoxicity are also worth considering: for example, Table 2a shows that the maximum individual doses from hypothetical let s say unreasonable - cycles, for which all activities of Cs-137, Pu-239 and Np-237 are diluted or dispersed into atmosphere or in the sea (!). In this case, the calculated individual doses are quite obviously unacceptable, from some msv to several Sv per annum Radionuclide Cs-137 Pu-239 Current management (mainly disposal) Hypothetical management Atmosph. Release Hypothetical management Marine releases 1.4 10-3 (sea releases, reprocessing) > 10 5 900 3.6 10-4 (sea releases, reprocessing) 100 2.9 Np-237 4.3 10-5 (atm. releases, reprocessing) 13 9 10-2 Table 2a. Individual doses to the reference group (msv/year) associated with waste management options for a hypothetical cycle However, the comparison of individual doses related to optional management with doses corresponding to the current practices is not always so trivial. Table 2b shows that the potential of individual dose reduction (from C-14, Kr-85 and I-129) is between one and two orders of magnitude when going from the present management to hypothetical ones, that would allow to completely trap these three radionuclides 4
Table 2b. Radionuclide C-14 Kr-85 Current management (mainly disposal) 6 10-3 (sea releases, reprocessing) 1.1 10-2 (atm. releases, reprocessing) Underground disposal (excluding intrusion) 6.6 10-4 4.8 10-2 4.4 10-3 I-129 (surface disposal) Individual doses to the reference group (msv/year) associated with waste management options for a hypothetical cycle 0 The previous tables shows that the maximum annual individual doses related to current waste management remain well below the limit of 1 msv/year set for the members of the public by the ICRP in its Publication 60. It is also below individual source-related dose constraint: from recent discussions in non gouvernemental bodies, it appears that the dose constraint should be about 0.3 msv per year. The same applies for the management option which consists in trapping and storing C-14, Kr-85 and I-129. For the carbon-14, the evolution in time of the maximum individual exposure is illustrated in Figure 2a (current management) and Figure 2b (hypothetical management), together with the populations concerned (local reference group RG- or world population GP-). 1,00E-02 1,00E-03 Total natural dose GP 1,00E-04 Maximum individual dose Sv/y 1,00E-05 1,00E-06 1,00E-07 1,00E-08 1,00E-09 1,00E-10 RG: reference group GP: global population Direct releases RG GP Direct releases Current management of C-14 Surface disposal Underground disposal RG 1 10 100 1000 10000 100000 RG Time Figure 2a. Maximum Individual Dose (msv/year) associated with the current management of carbon-14 5
1,00E-02 1,00E-03 Total natural dose GP 1,00E-04 Maximum individual dose Sv/y 1,00E-05 1,00E-06 1,00E-07 1,00E-08 1,00E-09 Alternative management of C-14 (underground disposal) RG 1,00E-10 1 10 100 1000 10000 100000 RG: reference group GP: global population Time Figure 2b. Maximum Individual Dose (msv/year) associated with a hypothetical management of carbon-14 (delayed release after trapping during the operational phase) Beyond the direct comparison of maximum individual doses, an important parameter is the moment when these doses are received. Here again, the retention of the underground disposal site leads to a considerable postponement of these doses: this is clearly illustrated in comparing Figure 2a with Figure 2b, where the maximum individual dose is delayed by 9000 years (300 human generations) for C-14. This illustrates the risk of a transfer between present and future generations that can arise due to the choice of waste management. DOSES AND RISKS TRANSFERS This question of possible risk transfers is a keypoint of the decision-making process. Whatever the option selected, waste management leads to a radiological impact. Depending on the radionuclide concerned and the management option, the most exposed population group can be either the workers or the public, in the short, medium or long term. With respect to the different options related to the cycle, the transfer of risks between these different categories of population has to be examined, as well as between these different time frames. These risk transfers may occur at three levels: - Between the public and workers: for example, choosing to trap and store a radionuclide which is usually released into the environment would lead to a reduction in public exposure, but would increase the exposure of workers by creating new operations for processing this radionuclide, hence transferring the risk from the public to the workers. - Between generations: as previously pointed out, storing a radionuclide postpones its release into the environment, except for the fraction eliminated through radioactive decay during the retention period. In this case, choosing disposal rather than immediate release implies transferring risks from current to future generations. - Between local and global populations: here again choosing to store and therefore concentrate a radionuclide, rather than disperse it early into the environment, can lead to a transfer of part of the risk from global to local populations (particularly as regards the risk of intrusion in a disposal site, see further). RADIOLOGICAL CRITERIA APPLIED TO HUMAN INTRUSION For the entire nuclear fuel cycle, the doses related to the additional disposal option of the radionuclides reduces the maximum individual doses to the reference group by deferring them. However, this option gives rise 6
to the risk of intrusion within the disposal site, with the possibility for the individuals concerned of receiving higher individual doses. The maximum individual doses linked to intrusion scenarios for an underground disposal site are shown in the following table for current waste management, as well as for the additional options examined. The intrusion scenarios considered are the drilling of a well near the disposal site (one at 610 m from the high-level vitrified waste and the other at 1230 m from the medium-level waste). The doses are expressed for the entire disposal site as defined in the Everest study. Probabilities of intrusion are not considered here. It should also be pointed out that these doses are deferred for a considerable length of time, from several thousands to several millions of years. Table 3. C-14 I-129 Current management Hypothetical management 1.2 10-5 - 5.3 10-4 5.6 10-3 - 0.25 3.8 10-3 - 1.8 0.89-42 Maximum individual doses linked to intrusion scenarios for underground disposal (msv/year) These maximum individual doses are, on the whole, several orders of magnitude above exposures linked to the normal evolution of the disposal site alone. As regards I-129 in particular, doses can reach levels of several msv/year for the site as a whole, both for current waste management and the additional options envisaged. Potential scenarios have to be considered in order to appreciate the possibility of reduction of the likelihood of human intrusion or to limit its consequences (disaggregated dose/likelihood approach). THE USE OF DIFFERENT RISK INDICATORS The individual dose is a very comprehensive indicator, it is accompanied with high uncertainties especially if the assessment focuses on the long-term consequences: the size of the population concerned, the human behaviour (diet, living conditions, hobbies) and actions (economic crisis, wars, voluntary or inadvertent intrusion ), the evolution of the geologic setting, of the biosphere, and engineered barriers over the long term All these dimensions are not so easily predictable. Proof that the disposal system will always satisfy individual dose criteria cannot be absolute because of these inherent uncertainties. The compliance with constraints should not be considered as sufficient to accept an option of disposal, and conversely, a numerical excess of constraints should not necessarily oblige the decision maker to reject an option, but this must be considered as a startingpoint to determine if additional measures would result in improved protection. In conclusion, the quantification provides only a basis for judgements. Thus, the possibility to strengthen a site specific assessment of a disposal using other indicators than the individual dose should be inevitably envisaged. It should be useful, maybe necessary, to produce other indicators on which uncertainty is lower. From this viewpoint, it would be necessary to give an assessment of the expected activity levels in some of the biosphere and geosphere receptors [3], or the residual radiotoxicity of the potential long-term releases. It should allow to compare and discriminate different options of disposal. In the case of the distant future evaluation, all these indicators - including the individual dose - have not to be considered as an accurate, direct or indirect measurement of the health detriment. They are only simple decision-aiding tools, to be used in the comparison of concepts of storage sites or strategies of disposal. The collective dose which is a further input into optimisation of protection is of limited use in the context of the disposal of long-lived radioactive waste. However, consideration of the number of people potentially exposed and the distribution of individual dose in time can be of fruitful help. CONCLUSION This paper is an attempt to contribute to the development of a framework for the evaluation of the impact and inter-generational risks transfers associated with waste management. Any option envisaged either to dilute/disperse or to concentrate and confine must be evaluated taking into account dose transfers between exposed groups in space and time. For this purpose, the individual dose criterion should be considered as an evolutive indicator. 7
Furthermore, optimisation of protection is a judgmental process with social and economic factors being taken into account, assessed doses or risks to individuals being important but not unique - input into this process. Final decisions rely on a compromise between the short-term protection of the population and the risk transferred to future generations, a compromise which is essentially of an ethical nature. REFERENCES [1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Radiological Protection Policy for the Disposal of Radioactive Waste. Publication 77, Annals of the ICRP 27, Supplement 1997, Pergamon Press, 1998, Oxford and New York. [2] TORT V., LOCHARD J., SCHNEIDER T., SUGIER A., Preliminary Evaluation of the impact and intergeneration risk transfers related to the release and disposal of radioactive waste from the nuclear fuel cycle, CEPN Report n 255, December 1997. [3] INTERNATIONAL ATOMIC ENERGY AGENCY, Safety indicators in different times frames for the safety assessment of underground radioactive waste repositories, IAEA Tecdoc 767, Vienna, October 1994. 8