Regulators all over the world formally base their decisions about the acceptability of a particular radioactive waste disposal system upon the related performance (safety) assessment. The key element of this assessment is dose evaluation. However, the requirements for the certainty/accuracy/validity of such evaluation are not clearly defined in advance and are a subject of development in the dose evaluation process itself. Therefore, dose evaluation, as well as the associated licensing procedure that builds on compliance assessment, seems to be a less appropriate approach due to the uncertainty involved. An alternative method for assessing human exposure in the framework of long­term safety assessment should be developed. Such a method, integrated with the concept of reasonable assurance (IAEA, 1997), should build on indicators of future exploitation of the environment — therefore a clear link to spatial planning in site selection process where human activities remain the basis for future exposure assessment (Kontic et al., 1999).

Since this approach is more fundamental, direct and transparent than dose or risk assessment, it is expected that it will be more powerful in confidence building among different social groups, i. e. scientists, regulators, the public and politicians. Eventually, it is also expected to be effectively applied in comparative evaluations of various energy options. In addition, certain ethical dilemmas in the licensing process connected to regulatory decision making in the presence of uncertainty and in the context of the disposal of long lived radioactive wastes, could also be reduced if dose or risk are avoided as individual numerical safety indicators.

3.1.2 Scientific treatment of specific uncertainty and predictions related to spent fuel

In this sub-section an analysis of the impact of uncertainty (associated with the quantity of radioactive waste produced by Krsko NPP in its anticipated operational time) on the waste-disposal strategy, particularly the selection of the disposal option, is presented. The dilemma is whether to build a shallow land repository for the LILW and to treat all high — level and long-lived waste separately; or to adopt deep geological disposal as the option for all waste types produced in the country. Tightly connected to these questions is the credibility of the evaluation of health consequences due to radioactive waste disposal. Indicators can, for example, be the dose and risk in the presence of uncertainty associated with the waste characteristics on the one hand, and societal characteristics and human habits in the distant future on the other. The approach and methods applied in the analyses were as follows: [7]

324. Using 1,876 MW as the nominal power of the plant, and 48.7 t of uranium per cycle, one obtains 11,857 MWd/ tU. When this is rounded off, 12,000 MWd/ tU for burn-up and 320 effective days of operation at full power are obtained.

• A twelve-month cycle was assumed (i. e. the cycle lasts 365 days); the operational period is 320 days and the cooling (decay) period between cycles is 45 days (actually used for refuelling and maintenance).

• One batch of fuel consists of 40 elements, containing 16.24 tonnes of uranium, and on average represents one-third of the total amount of fuel in the cycle (there are three different batches in the reactor during operation). Each batch remains in the reactor for the three following cycles — except the first, second, penultimate and last batches. The real situation is more complicated but corresponds roughly to these assumptions.

• Being aware of the differences between these assumptions and the real operational data for Krsko NPP, a screening calculation of the activity of spent fuel for the first 13 cycles was made, for the purpose of further calibrating the model. The results for the model and those based on operational data differ very little — see Table 3 for details.

• The content (mass) of uranium isotopes per fuel batch is given in Table 4.

• The mass of zircaloy (Zr-40) per batch is 4012.5 kg; the mass of oxygen (O-16) is 2183.5 kg.

• The average power per tonne of uranium is 37.5 MW; the average power of the batch is 609 MW.

	Activity after 7th cycle (Bq)	Cooling period (days)	Activity after 13thcycle (Bq)	Cooling period (days)
Source of operational data: NPP Krsko, and Ravnik and Zeleznik, 1990	7,5E+18	45	N. A.	N. A.
Source of operational data: NPP Krsko, and Bozic (1998)	9,2E+18	45	2,5E+19	32
Model prediction	9,9E+18	45	1,2E+19	45
Difference (%), rounded	9-33		208

Note: One should note that these differences are very low, taking into account that the order of magnitude of the values is E+19, and that the cooling periods differ, after the 13-th cycle, between the model and the real data. The latter is important if the activity of spent fuel decreases rapidly during the cooling period; this would mean that the activity drops by a factor of two over a two week period, i. e. in the period between 32 and 45 cooling days, which is the difference between the real data and the model.

Table 3. Comparison of the calculated (model) results and operational data

The calculated changes of activity of activation products (AP), actinides (ACT), fission products (FP) and total activity per fuel batch with time are presented in Figure 4. The
illustration is for model Batch 6; however, the figures are similar for other model batches. Batch 6 goes into the reactor in the fourth cycle (year). At the moment of irradiation, the total activity immediately increases by about eight orders of magnitude. Before that, the activity is constant at a level of 1.9*106 MBq (the activity of approximately 16 tonnes of non — irradiated fuel). During irradiation, this activity rises slightly from 1.23 to 1.28*1014 MBq, while during the cooling period of 45 days it drops by approximately two orders of magnitude. Each batch stays in the reactor for three successive cycles (except the first, the second, the penultimate and the last), whereupon the batch goes into the spent fuel pit for ultimate cooling and decay. It should be noted that the scale of both axes is logarithmic, so that the origins of axes are avoided in the illustrations.

Batch- enrichment (%)	Isotope (kg)
U-234	U-235	U-236	U-238
2.1	2.44	341.04	2.11	15894.25
2.6	3.25	422.24	2.59	15811.91
3.1	3.89	503.44	3.09	15729.58
3.4	4.22	551.67	0.81	15683.29
3.6	4.55	584.64	0.65	15650.49
3.9	5.36	633.36	1.30	15599.82
4.0	5.85	649.60	2.03	15581.96

Table 4. Mass of uranium isotopes in the fuel (per batch)

——— AP ———— ACT………….. ‘FP total Time (years)

Fig. 4. Activity of model batch 6 over a million years

Model results for all the fuel are presented in Figure 5, which shows time changes in total activity. With regard to activity during the first 34 cycles, an almost linear increase can be identified due to the collection of spent fuel in the spent fuel pit — one batch per cycle/year. After the 35th cycle, i. e. at the end of the assumed operation of the plant, all three batches from the reactor will be placed in the spent fuel pit at the same time, which is seen as an noncontinuous increase in activity. Activity then decreases, depending on the radionuclides contained in the spent fuel. Note again that the scale of the axes is logarithmic. The values of total activity and decay heat for all spent fuel at selected time — points are summarised in Table 5.

The model adequately represents the overall operation of the plant. This was proved in the process of calibrating the model, where data for the past thirteen cycles were used for comparison. However, fuel enrichment, as well as other key operational elements in future cycles, may not remain constant, since an upgrade of the plant’s power in parallel with the replacement of the steam generators has been achieved. Extension of the fuel cycles was also adopted/made. This was the reason for the analysis of the changes in the activity and radionuclide inventory of spent fuel, due to different fuel enrichment and the prolonged operation of the plant. The adopted variation in fuel enrichment was 1% above and below the value applied in previous calculations, i. e. 4% of U-235.

Time (years) Fig. 5. Total activity of all the spent fuel as a function of time

With regard to the prolonged operation of the plant, a five-year variation was applied. All the variations were simulated for the period following the replacement of the steam generators, i. e. after the 17th cycle. The differences are presented in Figures 6 and 7, respectively. It is clear that the differences are so small that they can be neglected, since they are not relevant to the overall waste management strategy. Moreover, the conclusion which can be drawn from this result is that no benefit can be expected in terms of improved safety connected with radioactive waste disposal whether Krsko NPP were closed down immediately or operated for almost a further 12, or even 40 years as the new National Energy Programme suggests.

The results of the modelling show that the main contributors to fuel activity during the period approximately 200 years after irradiation are the fission products; after that, actinides will prevail. The total expected activity of the spent fuel after one million years is 4,8*1014 Bq. The main contributors to this activity are the radionuclides of the U — and Np-chains. Residual thermal power is about 1.0*105 W approximately 200 years after irradiation, about 1.0*104 W after 10,000 years, and about 250 W after one million years.

The problem of uncertainty, which can be treated scientifically, is manageable. It was recognised that the basic characteristics of this waste can be accurately predicted, since all the sources of uncertainty are well defined, understandable and therefore controllable. Residual uncertainty does not change the overall picture of the waste, which would mean that the predictions could clearly be used as a basis for policy-making, i. e. creating a strategy for radioactive waste management, decision-making and also for communication with the public.

Time (years)	Total activity (MBq)	Decay heat (W)
1	5.30*1012	5.55*105
2	6.72*1012	7.19*105
3	7.92*1012	8.71 *105
4	8.70*1012	9.54*105
5	9.23*1012	1.01 *106
10	1.09*1013	1.13*106
15	1.21*1013	1.22*106
20	1.31*1013	1.31*106
25	1.40*1013	1.38*106
30	1.48*1013	1.45*106
35	2.69*1013	2.72*106
75	2.63*1012	3.08*105
100	1.46*1012	2.23*105
300	1.08*1011	8.44*104
1,000	4.11*1010	3.52*104
10,000	1.04*1010	8.23*103
100,000	1.28*109	6.82*102
300,000	8.12*108	3.80*102
1,000,000	4.81 *108	2.53*102

Table 5. Total activity and decay heat of all the fuel from the Krsko NPP at selected time- points over a million years

Fig. 7. Influence of extended or shortened operation of the Krsko plant on the activity of actinides in the complete spent fuel (the basic estimate is that the plant will operate 35 cycles)

With regard to confidence building connected to radioactive waste disposal, the recommendation is that prompt, clear and complete informing of all interested parties and the general public should take place. It should be clearly stated that the spent fuel from Krsko NPP, and a part of the decommissioned waste, will remain radioactive above today’s prescribed levels for hundreds, thousands or even a million years from now. Consequently, a strategy built upon waiting for the activity to "disappear" cannot be effective. Doubts and uncertainties regarding safety assessments in a timeframe of a million years should also be revealed. At the same time, efforts should be made to present the concept of reasonable

assurance (IAEA, 1997) as the most reliable method, and as the basis upon which a waste

management strategy can rely.

4. Concluding remarks

Strategic environmental considerations of nuclear power should inevitably cover the

following (however, not restricted to):

• Comparative evaluation of electricity generation technologies; the evaluation should, in addition to topical consideration, thoroughly deal with the ways on how to overcome specifics and details of individual analysis of a certain technology which is usually influenced by the specific characteristics of the site compared to others in its category, the manufacturing and design characteristics, the power, lifetime and the operating conditions. Results are therefore difficult to transfer from a country to another or one generation unit to another, as most major environmental, economic and social impacts, with the exception of e. g., climate change, are heavily site-dependent. Application of proper indicators in such a comprehensive comparative evaluation may be of practical help and guidance;

• The energy system as a whole; the electricity grid and market issues are rarely taken into consideration when making comparative evaluation. Similarly, the issue of increased share of intermittent RES, its impact on the energy system, and consequent need for the adaptation of environmental impact appraoches by taking into account actual share of each generation technology as provided in the energy system;

• Uncertainties; uncertainties and limitations of various methodologies may be acknowledged by the authors of the studies but those are rarely taken into account when results (or only some of them) are used by policymakers. Strategic considerations should provide guidance/recommendations on how to deal with the uncertainties in the decision-making process associated with comparative evaluation of different electricity generation technologies. This is especially relevant when deciding about long-term impacts, e. g. nuclear waste disposal or societal and spatial consequences of climate change.