A number of radiological protection issues which have implications for the establishment of authorized limits for effluent releases have been explored. Results have been provided which would be useful to the authorities responsible for setting these limits.
The issues considered included:
methods for defining critical groups;
source upper bounds;
the application of dose limits;
comparison between discharge of effluents and trapping, immobilisation and disposal of radionuclides in solid form.
Statistical methods suited to the analysis of models of association between spatially defined variables were developed along the following lines:
design modified tests of simple and partial correlations between sets of spatial variables which take into account the existence of spatial autocorrelation;
investigation of the performance of these statistics in terms of type I error and power and compare them to standard tests;
implementation of these tests on French geographical data with particular reference to the link between lung cancer and industrial exposure;
implementation of multiple regressions at a geographical level with different spatial parametrizations of the variance concariance error matrix;
comparison of some French geographical data with the results given by the modified tests obtained from these multiple regressions;
investigation the feasibility of defining nonparametric tests of association based on permutations.
Further the mortality file was recently updated to include mortality rates for specific cancer sites for the period 1984 to 1986 and in the last part of the project this new mortality data will be analysed along with industrial pollution and low dose radiation with the help of the new statistical methods which have been developed during this project.
The approach used for this study was to calculate derived limits for a few radionuclides and environmental materials, based on both a dose limit for a single year and a limit on the annual average dose over an individual's lifetime. These derived limits are in terms of radionuclide concentrations in environmental materials and thus differences in them will reflect the differences which would occur in effluent release limits.
The results of the calculations of derived limits showed that the most restrictive procedure will always be use to a limit of 1 mSv on the effective dose in a single year. However, for the radonuclides and foodstuffs considered, derived limits calculated on this basis do not differ greatly (a factor of 3 at most) from those obtained on the basis of the 50 mSv dose limit to a single organ. It could also be seen that use of a 1 mSv limit on average annual dose over a lifetime will generally be more restrictive than using a 5 mSv limit for a single year, but exceptions occur when dietary intakes and doses per unit intake are higher in childhood than adulthood (eg plutonium-239 in milk). The conclusions drawn from this work were that while it is possible to use a limit on average annual doses over a lifetime as a basis for setting effluent release limits, it is simpler and more conservative to use a limit on the dose in a single year.
Subsequent to this study, the National Radiological Protection Board (NRPB) published revised generalised derived limits for a number of radionuclides, based on a limit of 1 mSv on the effective dose in a single year.
Several general comments can be made about the state of development of the work on optimisation of protection. It is now clear that the as low as reasonably achievable (ALARA) principle is a workable concept and that tools for structuring optimisation problems and aiding decisions. such as the ALARA procedure, exist and have been shown to work in practice. It is also clear that a structured approach can be applied to all levels of radiation protection problems, but that the resources allocated to each problem need to be commensurate with the level of decision involved. For example, in many cases of low level decision, intuitive thinking by experts coupled with ALARA awareness may be all that is necessary. At high levels of investment, an extensive desk top ALARA study involving several man months of effort may be needed. It is also clear that methods for incorporating these ideas into radiation protection programmes (eg ALARA audits and ALARA predictive plans) exist, and again have been demonstrated to work in practice.
What is required now is that these ideas,principles, tools, and methods by actually used widely in radiation protection programmes. The document produced on ALARA is an important step in this direction, but there is additionally a need to provide manuals of good practise, training on the implementation of these methods, and simple computer tools to aid decision making as well as guidance on the application of these methods to more complex problems. It has also been recognised that there is a lack of relevant data on dose saving techniques and associated costs for detailed application of quantitative methods of operations warranting them. Future work is needed to address these specific areas.
The review of the International Commission on Radiological Protection (ICRP) recommendations on critical groups indicated that, although they work well in practice, there were a number of conceptual problems in their formulation and in their application to the establishment of authorized limits. Firstly, ICRP state that their dose limit for members of the public, and hence also any source upper bound, applies to the mean dose in a reasonably homogenous group ie that the dose to be calculated is that to a single, hypothetical person. However, they also state that the critical group would not usually consist of 1 individual, and that only in an extreme case, such as when dealing with conditions well into the future, would it be appropriate to consider single hypothetical person. These recommendations can be seen as contradictory. Secondly, ICRP state that, in calculating doses to critical groups, it is important to select appropriate mean values for factors such as food consumption rates, but that metabolic parameters (and hence dose per unit intakes of radionuclides) should be chosen to be typical of the relevant age group in the normal population rather than extreme values. Again there is an apparent inconsistency. Thirdly, ICRP give criteria for homogeneity in a critical group which vary with proximity of the mean dose to the relevant source upper bound; such criteria are difficult to apply in prospective situations where source upper bounds may not have been finalised. Fourthly, ICRP recommend that the results of a habit survey at a particular point in time should be regarded as an indicator of an underlying distribution, but no guidance is given on how to determine this distribution, or how to make allowance for possible changes in the distribution if the period for which authorised limits are intended to be in force is fairly long.
As a way of resolving some of these problems, it has been suggested that, at least for some major foodstuffs, it would be prefera ble to use a particular percentile of an entire frequency distribution of consumption rates in a population as the mean for a critical group, rather than to define the group and then calculate the mean.
At present, the identification of critical groups and the calculation of the mean doses to these groups relies to a large extent on expert judgement guided by some empirical recommendations. This approach has the disadvantage that varying degrees of conservatism can be introduced when establishing and explaining authorised limits for effluent discharges, and there is thus the possibility that the real degree of protection achieved in each case will not be clear. The eventual aim of the work is to produce practical recommendations to avoid such situations, not by removing the element of judgement in critical group identification and dose calculation, but by making it more explicit and quantitative.
The concept of upper bounds was introduced by International Commission on Radiological Protection (ICRP) in Publication 37. The purpose of the source upper bound is to act as a constraint on optimisation for that source, so the exposure of any individual will remain below the dose limit, even if the individual is exposed to several sourced. Thus source upper bounds have to be set to be some fraction of the dose limit. The methods for establishing authorized limits for effluent discharges in various countries were reviewed, with the aim of determining whether the source upper bound concept is utilised in practice. From published information on methods for establishing authorised limits in (EC) countries, the United States, Japan and Sweden, it appears that the concept of source upper bounds has not been formally incorporated in legislation, regulations or government policies. Nevertheless, most of these countries employ dose standards which are a fraction of their overall dose limits for members of the public when setting authorized discharge limits for nuclear installations. These standards effectively act as source upper bounds, although they are more frequently described as target, or as site or discharge specific limits.
It was concluded from the review of published information that regulatory authorities in most countries find it useful to have a dose standard lower than the overall dose limit for members of the public in mind when setting authorized discharge limits. From the differences in values used it is apparent that, while these standards effectively act as source upper bounds, they have been derived in different ways. This could be one of the reasons why the concept of source upper bounds has not been formally adopted by regulatory authorities. Another possible reason is that some authorities may prefer to use the concept of dose targets which can be exceeded if it is optimum to do so, rather than imposing formal constraints other than the dose limit at the start of the optimisation process, because in principle the former approach allows more flexibility.
The objective of this work was to highlight some of the difficulties which can occur when, in order to fulfil the as low as reasonably achievable (ALARA) requirement, discharge of radioactive material to atmosphere or to the aquatic environment has to be compared to trapping, immobilisation and disposal of the material as a solid. The study focused on those parts of comparisons concerned with radiological impact on the public.
The main conclusion drawn from the work was that there is a need to improve the consistency between methods for calculating the radiological impacts on the public of solid waste disposal and effluent discharges. Differences appear to have risen because, on the whole, those involved in solid waste disposal assessments are not concerned with discharges, and vice versa. Even if no comprehensive discharge/disposal comparisons are made as an input to setting authorised limits for discharges, as new solid waste disposal facilities are developed questions are bound to arise about how their potential radiological impact compares with that of existing nuclear installations. Hence it would be preferable if effluent discharge and solid waste disposal impact assessment methods were consistent with each other (and with accident consequence assessment methods). This seems more important at present than questions such as those about the weight to be attached to long term doses and risks when waste management methods are compared.
PROJECT 1: ISSUES IN ESTABLISHING AUTHORISED LIMITS FOR EFFLUENT RELEASES.
AUTHORISED LIMITS ARE SET FOR EFFLUENT RELEASES BY NATIONAL AUTHORITIES, USUALLY GOVERNMENT MINISTRIES OR DEPARTMENTS. THEY ARE NORMALLY EXPRESSED IN TERMS OF AMOUNTS OF RADIONUCLIDES IN AIRBORNE OR LIQUID DISCHARGES AND COMPLIANCE WITH THE LIMIT IS A LEGAL REQUIREMENT. IN SETTING SUCH LIMITS, ALLOWANCE HAS TO BE MADE FOR A WIDE RANGE OF FACTORS. IN ADDITION TO ENSURING THAT DOSE EQUIVALENT LIMITS ARE NOT EXCEEDED THERE IS THE OTHER PRINCIPAL REQUIREMENT OF THE ICRP SYSTEM OF DOSE LIMITATION TO CONSIDER, NAMELYTO ENSURE THAT ALL EXPOSURES ARE KEPT AS LOW AS REASONABLY ACHIEVABLE (ALARA), ECONOMIC AND SOCIAL FACTORS BEING TAKEN INTO ACCOUNT. THERE ARE ALSO LIKELY TO BE CERTAIN NON-RADIOLOGICAL FACTORS WHICH MAY AFFECT DECISIONS.
PROJECT 2: METHODS FOT THE PRACTICAL IMPLEMENTATION OF THE ALARA PRINCIPLE.
THE OBJECTIVES OF THIS STUDY, TO BE CLOSELY CO-ORDINAD WITH CEPN (FRANCE) ARE:
1. TO DEVELOP A SIMPLE, GENERAL FRAMEWORK BY APPLYING IT IN EXAMPLE STUDIES.
THESE OBJECTIVES WILL BE ACHIEVED BY FIRSTLY REVIEWING THE DIFFICULTIES WHICH HAVE ARISEN IN THE PRATICAL IMPLEMENTATION OF ALARA, AND THEN SUGGESTING AND APPLYING METHODS BY WHICH THESE DIFFICULTIES MAY BE RESOLVED. THE REVIEW AS TO BE CARRIED OUT BY REFERENCE TO STUDIES REPORTED IN THE LITERATURE AND TO WORK IN PROGRESS AT CEPN AND NRPB. THE EXAMPLES CHOSEN WILL CONCERN A RANGE OF PRACTICES INVOLVING RADIATION EXPOSURE, INCLUDING BUT NOT CONFINED TO, THE NUCLEAR INDUSTRY. THEY WILL BE SELECTED SO AS TO ILLUSTRATE THE SCOPE OF ACTUAL PRACTICAL DIFFICULTIES, RATHER THAN THOSE WHICH COUL THEORETICALLY ARISE. IN SUGGESTING METHODS TO RESOLVE THESE DIFFICULTIES AND IN DERIVING THE GENERAL FRAMEWORK, PARTICULAR ATTENTION WILL BE PAID TO INDICATING THE CAPABILITIES OF VARIOUS DECISION-AIDING TECHNIQUES (EG. MULTI-ATTRIBUTE ANALYSIS, COST BENEFIT ANALYSIS) AND TO IDENTIFYING THE AREAS WHERE JUDGEMENTS ARE REQUIRED BY THOSE RESPONSIBLE FOR TAKING ALARA DECISIONS.