RWMAC's Advice to Ministers on the Radioactive Waste Implications of Reprocessing

7. DISCHARGES AND THE RADIATION DOSES ASSOCIATED WITH THEM

7.1 Methodology

Discharges arising from the various combined reprocessing scenarios (see section 4.2) were assessed in terms of total activity, critical group dose and 500-year world collective dose (note that, in this context, dose is a measure of the radiation received). Both liquid and aerial discharges were considered.

The activity estimates given in this report are expressed in terms of becquerels, Bq. One Bq is equivalent to one nuclear disintegration per second. A terabecquerel, TBq, which is also referred to, is equivalent to 10¹² Bq. Radiation doses are expressed in terms of sieverts, Sv, which are a measure of the radiation received. There are prescribed limits to the doses that are received by members of the public and workers from man-made radiation, that are referred to later in this section. Also referred to are millisieverts and microsieverts which are respectively a thousandth and a millionth of a sievert.

The analyses were undertaken at RWMAC’s request by BNFL, who pointed out that estimation of the figures is inevitably judgmental due to the complex interaction of reprocessing and treatment plants on the Sellafield site. The means of analysis themselves, although well-established and accepted, also inevitably involve needs for assumption and approximation, and thus at least some element of uncertainty. Some of BNFL’s assessment models were updated as a result of RWMAC comments for the purpose of the analyses. Nevertheless, RWMAC believes that the discharge activity and dose figures given in this section should be viewed as indicative rather than definitive.

The levels of activity, critical group dose and collective dose associated with reprocessing were forecast by BNFL for "representative" Magnox and THORP annual throughputs. The outcome of such analyses is set out in Annex 7. The representative throughputs used were 1,200 tHM/yr for Magnox reprocessing and 1,000 tHM/yr from THORP reprocessing, which are considerably in excess of the actual throughput rates achieved in recent years (see section 6.3). This means that the discharge and dose estimates may effectively be considered as upper boundary estimates. Alternatively, this could also mean that the timescales modelled by BNFL could give rise to reduced timescales over which Magnox reprocessing discharges and doses might extend. In practice, activity outputs will not be related linearly to throughput rates, although such an assumption could be used to give some indication of the potential effect of variation in such rates.

In compiling the Annex 7 figures, BNFL estimated the ongoing discharge levels from Sellafield site plants under three headings: Magnox, THORP and Other. The first two covered plants directly concerned with ongoing Magnox and THORP reprocessing. The Other category includes waste management and treatment plants (such as the Waste Vitrification Plant, Solvent Treatment Plant, Site Ion Exchange Plant, and the Miscellaneous Beta Gamma Waste Store), the Calder Hall reactors, and site research and analytical facilities. Discharges from plant in the Other category will only be associated to a limited extent with ongoing reprocessing activities although they may also deal with the treatment of some historic wastes that have arisen from past reprocessing. Models of radionuclide transfer through the environment, uptake and dose to humans were then used to estimate dose to the critical group and collective dose from the discharge information.

In respect of the Other category plant, BNFL stated that they did not believe that there were any significant consequential issues regarding liquid discharges from ongoing reprocessing activities. For aerial discharges the situation was different, but some broad judgements could be made. Here they believed that about 20 microsieverts per year of the Other plants’ critical group dose arises from the storage and processing of waste liquors, mainly HAL. This would primarily come from carbon-14 and iodine-129. Again using broad judgement, about half the discharges from other plants would be consequential upon the processing of backlog wastes and half from ongoing arisings under the RWMAC’s Combined Reference scenario. The ratio would change for other scenarios but broad scaling would provide an indicative picture.

It should be noted that the critical groups identified for the Annex 7 analysis of liquid and aerial discharges and doses are different. The critical group for aerial discharges is comprised of infants living close to the Sellafield site. The critical group for liquid discharges are high rate consumers of fish and shellfish in West Cumbria. There is only a small amount of overlap or "crossover" of dose between the two groups, of approximately 15 microsieverts per year.

7.2 Discharges activity

The indicative discharge data in Annex 7 show that, in terms of activity, by far and away the biggest contribution to liquid discharges, for both Magnox and THORP reprocessing, comes from tritium. For aerial discharges the releases assessed in terms of activity are dominated, for both Magnox and THORP by krypton-85. But these predominant contributors to discharge activity, are not the most important contributors to critical group dose.

For liquid discharges, the biggest contributions to the critical group dose come from technetium-99 and ruthenium-106 released as a result of Magnox reprocessing. The largest contributions to critical group dose from liquid discharges from THORP reprocessing come from cobalt-60, ruthenium-106 and caesium-137, but these are significantly less than the Magnox reprocessing liquid discharge critical group dose. For aerial discharges, the greatest contributions to critical group dose come from tritium, krypton-85 and iodine-129 in the case of Magnox reprocessing and iodine-129, and, to a lesser extent, krypton-85 and carbon-14 in the case of THORP reprocessing. For aerial critical group dose, the contributions from THORP, notably from carbon-14, are greater than those from Magnox reprocessing.

For collective dose the most important contributors are once again different. Here, the greatest contributions to collective dose from liquid discharges are overwhelmingly due to carbon-14 both for Magnox and THORP reprocessing (the Magnox contribution being substantially the higher). For aerial discharges, krypton-85 is the greatest contributor to collective dose, both for Magnox and THORP reprocessing, with carbon-14 and iodine-129 also contributing.

The key point arising from this analysis is that the criteria selected for assessment will affect judgement concerning the radionuclides that are most important. Those contributing most to discharge activity are not the most important in terms of dose. Various factors account for this difference: radioactive half-life, transfer of the radionuclide through the environment, rate of uptake into the human body and the doses delivered to the organs of the body.

Figure 2 sets out the estimated discharge activity for the three combined scenarios considered by RWMAC. That is the Combined Early Termination scenario (M3 and T5), the Combined Reference scenario (M2 and T2) and the Combined Extended scenario (M1 and T1, see Section 4.2). It should be noted that the data relate to all discharges from Sellafield except those from the Calder reactors and future decommissioning included under the "Other" plant heading, which are clearly divorced from ongoing reprocessing activities. More is said of the Figure 2 data in section 7.5 in the context of the UK’s OSPAR commitments.

7.3 Doses from discharges

The critical group and collective doses estimated by BNFL for future operations, as requested by RWMAC, are summarised in Annex 7. As was acknowledged in section 7.1, these estimates can only be indicative for several reasons. First, they relate to greater throughputs than have actually been achieved in recent years. Second, and understandably, they are based on current critical groups but future critical groups may be unpredictably different if habits and populations change. Finally, any mathematical modelling technique, however well-developed and accepted as those used for dose estimation are, still inevitably introduces at least some uncertainty.

(a) Indicative activity from liquid discharges

(b) Indicative activity from aerial discharges

Figure 2. Indicative assessment of discharge activity from reprocessing operations

It should also be noted that the estimated doses are attributable entirely to discharges for the year in question and do not include a component from radionuclides previously discharged to the environment. Consequently, they cannot be compared directly with dose estimates based on environmental measurements, which include a component from historically deposited radioactivity in the environment. These include figures quoted in BNFL’s Annual Report on Discharges and Monitoring of the Environment¹⁵ and the Radioactivity in Food and the Environment (RIFE) report²⁷ published jointly by the Ministry of Agriculture Fisheries and Food and the Scottish Environment Protection Agency.

The figures quoted in Annex 7 indicate the contributions to critical group dose for ongoing reprocessing activities for the representative throughputs assumed. For example, looking at part (a) of the table for liquid discharges, under the critical group dose heading one can see that the dose estimated for ongoing Magnox reprocessing is estimated to be 32 microsieverts per year. Bearing in mind what was said in section 7.1, Other plant activities would be expected to add little to this liquid discharge figure. The ongoing Magnox reprocessing contribution to aerial doses can be seen from part (b) of the Annex 7 table to be 6 microsieverts per year. However, again bearing in mind what was said in section 7.1, there is a need to make some allowance, probably of the order 5 microsieverts per year, for the contribution from Other plant operations. Summarising, it may be said that, in broad terms, ongoing Magnox reprocessing would contribute about 30 microsieverts per year to the liquid discharge critical group and about 10 microsieverts per year to the aerial discharges critical group dose.

Similarly, the corresponding indicative doses for ongoing THORP reprocessing are estimated to be about five microsieverts per year for the liquid discharge critical group (i.e. significantly less than the Magnox contribution) and about 30 microsieverts per year, including a contribution from Other processes, for the aerial discharge critical group dose (i.e. significantly greater than the Magnox contribution).

There are a number of supplementary observations that can be made concerning those dose estimates. First, to reiterate the point made in section 7.1, the contributions to the doses to the liquid and aerial discharge critical groups will not be additive because the groups are different and there is relatively little overlap in the doses received via these two discharge pathways.

Second, as was pointed out earlier, these projected doses, estimated for future reprocessing activities, are in addition to the doses derived from radioactivity in the environment as a result of reprocessing and its associated discharges in the past. Doses from historic concentrations of radionuclides in the environment from reprocessing have been estimated to be about three times that from ongoing reprocessing activities¹⁰. This mainly arises from historic actinide discharges (i.e. plutonium and americium).

Third, despite focussing on the indicative doses attributable to Magnox and THORP reprocessing per se, sight must not be lost of the contribution from Other ongoing processes on the Sellafield site. These processes are estimated to add about 11 microsieverts per year to the liquid discharge critical group dose (i.e. about one-third) and about 50 microsieverts per year for the aerial discharge critical group dose (i.e. almost double that from Magnox and THORP operations combined). These contributions bring the total liquid and aerial critical group dose from all ongoing activities at the Sellafield site to about 50 and 80 microsieverts per year respectively for the reprocessing throughputs considered. The biggest component of the Other facilities contribution to aerial discharge critical group dose, about 20 microsieverts, comes from the Calder Hall reactors which will close by 2006-2008 (see Table 1).

Fourth, the doses estimated can be given perspective by the average annual dose to members of the UK public from natural background sources of 2,240 microsieverts per year (within a range of 1,000-100,000 microsieverts per year)²⁸, the prescribed dose limit to a member of the public from man-made radiation of 1,000 microsieverts (1 millisievert) per year and the prescribed dose constraint for a single site (e.g. Sellafield) of 500 microsieverts per year³.

Also shown in Annex 7, are indicative world 500-year collective doses for ongoing processes at Sellafield. The estimates derived are about 110 man-sieverts attributable to Magnox and of the order 95 man-sieverts for THORP. These would be supplemented by about 130 man-sieverts from Other ongoing practices. Thus until Other processes operate at much less than their current level, the doses associated with their discharges will contribute significantly to the site total and will need to be managed within the overall discharge strategy alongside those from reprocessing. For comparative purposes, the UK and world collective doses from all sources of natural radiation are 130,000 man-sieverts and 13,000,000 man-sieverts respectively²³.

While such comparisons with background radiation doses may be made, it is equally acknowledged that the view of individual members of the public might be influenced by other factors e.g. perception of the possibility of accidents, a personal dread of radiation and a different weighting of voluntary and involuntary risk.

Figures 3 and 4 indicate how the critical group and collective doses attributable to ongoing Magnox and THORP reprocessing activities would change over time for the three scenarios considered by RWMAC. Again, the contribution of the Calder reactors and future decommissioning have not been included (see section 7.2). More is said of the Figure 3 estimates in section 7.5 in the context of the UK’s OSPAR commitments.

a) Indicative dose to critical group from liquid discharges

b) Indicative dose to critical group from aerial discharges

Figure 3. Indicative assessment of critical group doses from reprocessing operations

(a) Indicative collective dose from liquid discharges

(b) Indicative collective dose from aerial discharges

Figure 4. Indicative assessment of World (500 year) collective doses from reprocessing operations

7.4 Radiation doses to workers

It is inevitable that any nuclear activity, including reprocessing, will give rise to radiation doses to workers. For this RWMAC study, it has not been possible to estimate future worker doses likely to be associated with the scenarios considered. However, BNFL’s Sellafield, Chapelcross and Drigg Radiation Exposure Review²⁹, gives some relevant worker exposure data for reprocessing activities which are indicative of their general levels.

Magnox reprocessing

Over the five years up to and including 1998, BNFL estimate there has been a trend of general improvement in worker dose, with year on year variations due to production programmes and shutdowns. 1997 included a nine-month shut down, which involved considerable engineering work by Magnox reprocessing personnel. Overall in 1998 there was a decrease in the collective and average whole body dose from 1997 and whilst some of the decreases could be subject to reduced workload, there was also thought to be greater worker awareness of the requirement for minimising radiation exposure. The average dose in 1998 was of the order of 1.9 millisieverts (1,900 microsieverts) per year²⁰. This compares to the current legal dose limit for radiation workers in the UK of 20 millisieverts (20,000 microsieverts) per year and the average dose to members of the UK population from natural background radiation of 2.2 millisieverts (2,200 microsieverts) per year. The number of individuals at the higher end of the dose distribution continued to decrease. For the first time, no individuals received a dose of greater than 10 millisieverts (10,000 microsieverts) per year.

THORP

The majority of assessed doses with BNFL’s THORP Group are very low, with 94 per cent being below 2 millisieverts (2,000 microsieverts) per year and the average dose for THORP Group being of the order of 0.6 millisieverts (600 microsieverts) per year²⁹. Overall doses for 1998 were similar to 1997, but this was considered significant given that THORP throughput increased from 624 tHM in 1997 to 764 tHM in 1998.

Collective dose

The collective radiation exposure of workers attributed by BNFL to reprocessing between 1960 and 1998 is shown in Figure 5. This shows that the dominant effect overall is a reduction in worker dose, presumably due to a range of improvements in plant and practices²⁹.

Figure 5. Estimates of collective worker doses from reprocessing

7.5 Implications of the findings for OSPAR Sintra agreement implementation

At the 1998 Ministerial meeting of the Oslo and Paris (OSPAR) Commission, contracting parties to the 1992 Convention for the Protection of the Marine Environment of the North East Atlantic, including the UK, agreed an OSPAR plan for radioactive substances. The OSPAR objective is to prevent pollution of the maritime area defined under the Convention from ionising radiation, through progressive and substantial reductions of discharges, emissions and losses of radioactive substances. The ultimate aim is to achieve concentrations in the environment near background levels for naturally occurring radioactive substances and close to zero for artificial substances.

In June 2000, the UK Government issued a consultation document on a proposed UK Strategy for Radioactive Discharges 2001-2020² setting out how it intended to meet its OSPAR obligations.

The proposed strategy is intended to set a framework for radioactive discharges from UK installations over the 2001-2020 period. The consultation document² indicates that the scope of the UK strategy encompasses radioactive discharges from licensed nuclear sites, defence activities and other nuclear and non-nuclear sources of radioactive discharges. It covers both liquid and aerial discharges, although it is assumed that in general liquid discharges will have the largest and most measurable effects on the marine environment.

The declared aims for the proposed UK strategy are:

• progressive and substantial reductions in radioactive discharges from the UK as a whole and from each of the main sectors responsible for such discharges
• progressive reduction of human exposure to ionising radiation resulting from radioactive discharges, so that no member of the general public in the UK will be exposed to a dose of more than 0.02 millisieverts (20 microsieverts) per year, as a result of authorised radioactive discharges made from 2020 onwards
• progressive reductions in concentrations of radionuclides in the marine environment resulting from radioactive discharges, such that by 2020 they add close to zero to historic levels

More specific proposals made in respect of discharges from spent fuel reprocessing are that:

• by 2006, but without prejudice, to the Environment Agency’s review, (of Sellafield), technetium-99 discharges from reprocessing are assumed to be reduced to below 10 TBq per year
• by 2020, total alpha/beta liquid discharges from reprocessing (excluding tritium) are expected to be reduced to less than 30 TBq per year

From the viewpoint of the current study, the findings set out in Annex 7 and Figures 2 and 3 may be considered in light of these strategy proposals. This must take account of the scheduled closure of the Calder Hall reactors in the period 2006-8 and also B205 by around 2012 (given that, in the latter case, the necessary reprocessing rates can be achieved). It must also be emphasised that the data presented in figures 2 and 3 assume current levels of abatement technology, a subject that will be returned to later in this section.

Given no improvement in abatement technology, and the discharge levels and critical group doses set out in Figures 2 and 3, it is difficult to see how the Combined Extended Scenario (M1T1), with prolonged reprocessing, could be seen to comply with the UK’s OSPAR objectives. The figures, along with those in Annex 7, do however, show that the Combined Reference scenario (M2T2) could be seen to comply. The figures show a progressive and substantial reduction in discharges, and that the removal of the Magnox reprocessing contribution to liquid discharge critical group dose, will allow the latter to be reduced below 0.02 millisieverts (20 microsieverts) by 2020 given the limited contribution of Other activities on the Sellafield site to liquid discharge critical group dose (see Annex 7).

The estimates in Annex 7 and Figures 2 and 3 do not include the additional 1,500 tHM to be generated under BNFL’s revised Magnox business plan (see section 6.3). Further, they are based on assumptions of high reprocessing throughput and closure of B205 around 2012. If lower throughputs are achieved in practice (see section 6.3), reprocessing facility closures will be delayed with corresponding impacts on future discharge and dose projections, that could conflict with stated OSPAR strategy objectives. Additional analyses would be necessary to investigate such effects comprehensively and RWMAC have suggested to BNFL that such analyses be undertaken as an adjunct to those provided for the work reported here. Also, Annex 7 shows that taking into account Other plant contributions, even allowing for closure of the Calder reactors, it may be difficult to achieve a 0.02 millisievert (20 microsievert) level for aerial discharge critical group dose, if THORP was still to be operating by 2020 without improved abatement.

Unsurprisingly, the Combined Early Termination scenario could be said to provide the clearest demonstration of compliance with the UK’s proposed OSPAR objectives.

Achievement of OSPAR objectives could, however, be facilitated by improvements in abatement technology. A relatively recent BNFL review of the options for this is set out in Table 5 of the interim report of its national stakeholder dialogue Discharges Working Group¹⁰.

The specific objective of reducing technetium-99 discharges from reprocessing to 10TBq/per year by 2006 will depend critically on the ability to develop technetium abatement technology, given that the revised BNFL business plan foresees Magnox reprocessing continuing until around 2012. From information provided to RWMAC, the main problem may not be extraction of the technetium-99 and other radionuclides but rather incorporation into a form of solid waste that would be deemed suitable for eventual underground disposal, if this were to be the option eventually chosen for long-term management of the waste, and impacts on worker doses.

However it is clear that, given the Table 7(a) estimate of 63 TBq/yr for technetium-99 liquid discharges, the task of reducing such discharges to a level of 10 TBq per year is likely to be difficult. Magnox reprocessing rates need to be kept up to allow closure of B205 by around 2012 (see section 6.3). If abatement technology cannot be found to take technetium-99 discharges down to the desired level, the only option for reduction left would appear to be to retain greater amounts of liquid waste on site until a suitable solution could be found. This would clearly be undesirable. RWMAC therefore believes that this is an issue that needs to be kept under extremely close review.

RWMAC notes that every change of operation on a nuclear site will have some effect on worker dose, and to some extent any action taken to reduce off-site discharges and dose (prolonged tank storage of liquors, additional treatment stages etc) may tend to increase worker dose as the off-site dose reduces. The OSPAR Sintra agreement, whilst acknowledging the need to pay particular attention to the safety of workers in nuclear installation, requires the UK to make substantial reductions in discharge levels by 2020. Hence if reprocessing is to continue at Sellafield and the OSPAR Sintra objective is to be met there could, potentially, be some impacts on worker does that should also be assessed more fully.

In particular, whilst use of abatement technology in order to achieve targets is to be encouraged per se, RWMAC notes that in all cases there will be complex interplay of several relevant factors. Notably, any reductions in discharges must be balanced against:

• any safety or environmental problems arising from the management of any resulting solid wastes
• any increases in worker doses, arising from either the abatement technology itself, or from the downstream management of any resulting solid wastes
• the cost, not only of the abatement itself, but also of the downstream management of any resulting solid wastes

This means that the application of guiding principles such as ALARA, which is already difficult because of the component of subjectivity, is rendered even more complex in the consideration of abatement technology use.

In addition all of the above discussion takes no account of cost-benefit considerations which will have to be suitably fed in to the decision-making process.

[ Previous ] [ Contents ] [ Next ]