Draft document: Recommendations
Submitted by H.P.Kapteinat, VGB
Commenting on behalf of the organisation

Comment to ICRP 2006, draft September 08, 2006, by VGB VGB is a technical association of power plant operators with international membership of utilities and other organisations. Experts from a VGB Working Panel on Practical Radiation Protection drafted the following comments which should be considered in developing future radiation protection rules and regulations to ensure that they are as practice related as possible. General In its draft ICRP 2006 the ICRP considers in para. 12, that “the Commission’s extensive review of the vast body of literature on the health effects of ionising radiation has not indicated that any fundamental changes are needed to the system of radiological protection” and adds, that “the revised recommendations should not be interpreted as suggesting any changes to radiological protection regulations that are appropriately based on its previous recommendation in ICRP 60 and subsequent policy guidance.” Taking into account this statement, it would have to be expected that the new recommendations only try to give guidance on those parts, where former formulations have led to misunderstandings, where inconsistencies have been identified or where more information seems to be needed. Especially those central topics like optimisation and the use of dose constraints could be identified in this respect. However, we had to notice that some paragraphs might give rise to new misunderstandings, variations in the wording occur and it is unclear whether these variations have only linguistic background. On several occasions the formulation of the text remains in general and more philosophical terms, while objective and quantitative criteria are missing. In chapter 11.1.1 it is stated correctly, that radiological arguments are only one part of a decision and many other aspects, which are often of qualitative nature, like e. g. societal and political arguments, interests of stakeholders etc., have influence and have to be taken into account to get more reliable and sustainable solutions. But having this in mind it is indispensable to use objective criteria in radiation protection which are based on quantitative considerations. Only in this case they would be a profound and reliable basis for decisions which go beyond the scope of radiation protection. Unfortunately we have the impression that ICRP tries to integrate more and more such non-radiological conditions for a decision making process into the system of radiation protection. As a consequence, the recommendations lack clarity and precision, and a multitude of interpretations become possible. Furthermore, it gets more and more difficult to balance the efforts and costs for dose reduction against the benefits of this reduction. It can often be observed that the public and many stakeholders exaggerate the risks of radiation exposure to an extent, which leaves no room for an adequate solution based on objective risk criteria. If decisions are based on such grounds, they will necessarily be unbalanced and will lead to high efforts in manpower, techniques and last but not least costs which are not justified by radio protection arguments.. If it is, however, accepted that the basis for decisions include “weak” parameters like public opinion, stakeholder contribution etc., this fact needs to be clearly stated along with the decision itself. In any case, such decisions can never serve as a commonly accepted basis for decisions which are optimised with respect to radiation protection. In this context we regret that ICRP defines no lower “cut off” criteria for the dose in the optimisation process. Such a lower cut off criterion could be in the order of 10 µSv/a. It is theoretically clear that optimisation could lead to a dose lower than the value defining the scope of radiation protection (as defined in para. 226) but it is highly improbable that this will be the result of a process without undue efforts to reduce doses. Chapter 5.8 Optimisation of protection (incl. dose constraints) The wording concerning optimisation and dose constraints used by ICRP often cannot be clearly interpreted by professionals in radiation protection. Both topics are interrelated, but the ICRP recommendation is not very clear on this issue. On the one hand, dose constraints belong to the optimisation, while on the other hand, dose constraints are the starting point for optimisation. Though addressed at many places in the document, the topic of dose constraints is never clearly discussed. For example, no recommendations on how to derive dose constraints are given. We understand that they are not the result of optimisation and not simply a pre-defined fraction of the dose limits. While it is outlined what dose constraints are not, no statement on what they are is made. The reader gets the impression that they might be a result of simple arbitration without any objective criteria If, however, the dose constraints fulfil such a central role in radiation protection, it would be highly desirable that the ICRP makes much clearer statements on their definition and setting. Proposal: Give more guidance for the process of defining and setting constraints Constraints The objective of setting constraints, e. g. as formulated in para. 210, remains unclear and gives rise for arbitrary selections. There is no distinct separation from the optimisation. We understand that dose constraints have the objective – to assure compliance with dose limits in case of more than one source giving rise to exposure to one person, – to prevent too high individual doses even as the result of an optimisation process. It might be questionable that one needs dose constraints at all to achieve these goals. The objectives above have been adhered to ever since radiation protection limits exist, already at times when no formal dose constraints have been used, without compromising safety. We understand that dose constraints shall neither substitute nor preclude optimisation. Therefore, they should not be derived from already optimised practices as expressed in para. 210 with reference to “good practice elsewhere”, which implies that already optimised radiation protection solutions “elsewhere” exist. This is contradictory to the demand to optimise below the constraints. So, the reference to “good practices” should be deleted. Although much is said in para. 197 it will not be sufficiently clear that dose constraints are not intended to take the place of new and lower limits. Dose constraints can be a helpful tool while planning any practise and they also can be used to find out retrospectively, if there was any essential deviation from the plan. But this is a very crucial point: If dose constraints take the role of a dose limit, then non-compliance can have significant consequences for any use of radiation sources. In such a case, the actual limits would de facto be reduced to an arbitrary fraction of the current limits. However, this would be a substantial misinterpretation of the role of dose constraints. The risk estimates on which the dose limits have been based have not changed. The risk estimates define the dose limits, not the dose constraints. Therefore, any change of the limits will be inappropriate. Proposal: extend para. 197 in the following way:“…section 5.8.1 below. They are intended neither as a form of retrospective nor prospective dose limitation,…” The new table 4 on page 61 presents a system of dose constraints changing from “maximum dose constraints” in the 2005 draft of the ICRP recommendation to “bands of projected effective dose”. We explicitly miss a minimum value of a constraint in the band “under 1 mSv/a”. We propose to introduce a value of 0.1 mSv/a. Proposal: Introduce a value of 0.1 mSv/a as minimum value for a constraint in the band below 1 mSv/a. Optimisation (below constraints) The objective of optimisation is defined in different paragraphs of the document in different ways, making it difficult to implement optimisation in practice. The formulations in - para. 30 : “level of protection should be the best, maximising the margin of good over harm” - para. 197: “appropriate level of protection” - para. 221: “questioning whether the best has been done under the prevailing circumstances”, do not describe the process precisely and consistently, but evoke the questions: What means “the best”? Does it mean the “lowest dose”? Proposal: The process of optimisation has to be described more precisely and consistently by taking into account para. 227, which states: “Optimisation is not minimisation; the best option is not necessarily the one with the lowest dose.” para. 226 contains the statement, that a lower bound for optimisation cannot be defined. However, it should be clearly stated in this context that doses below some 10 µSv/a are insignificant and trivial and therefore need not be regarded when performing the optimisation process Proposal para.226: We recommend a lower boundary of some 10 µSv/a for the optimisation process. Results with doses below this value should automatically be considered as optimised. Optimisation below this dose range is meaningless as is already stated in the IAEA document “Principles for the Exemption of Radiation Sources and Practices from Regulatory Control” Safety, Series 89 (1989), the fundamental document on the trivial dose ranges. In the ICRP document, optimisation is defined as a process with attributes as follows: – ongoing, cyclical (para. 196), – interactive, continuous (para. 221), – interactive and iterative (para. 197). This may indeed be the case for some complex cases. For most applications in everyday practices, interactive, cyclical or iterative elements in the optimisation process would result in a totally undue burden. Having achieved a certain level of exposure a practice should be regarded as optimised and only in longer periods or after unusual events or after some significant new experience has been gained, the radiation protection situation should be re-evaluated checking whether the process is still optimised. Proposal: para. 196: extend the first bullet point “Evaluation of the exposure situation in cases where the circumstances relevant to the optimisation process have changed significantly.” Quantitative methods are referred to only in para. 194 with the short hint to ICRP 37 and ICRP 55 which are described as still valid. It would be highly desirable that the reference to quantitative methods be more directly in the text of the new recommendation. Proposal para. 194: Extend “…remain valid. The decision aiding techniques are still essential to find the optimised radiation protection solution in an objective manner; this includes methods for quantitative optimisation such as cost-benefit analyses.” Para. 46, exemption Although it is explicitly stated that the interpretation of triviality as the 10 µSv/a concept is only one part of the full meaning of this concept, the reader is left with the feeling that this is the only possibility to interpret triviality, as is explicitly stated in Annex I, No. 3 (a) of the EURATOM BSS. It should be noted that the “Annals of the ICRP – The Scope of Radiological Protection Regulations – 02/258/05 spring 2006 version” give a much clearer picture in this regard. In Section 3, “Dichotomous Control”, the ambivalent attitude of society towards natural and man-made exposure situations is outlined and recognised as the basis for defining different levels for exemption in these two situations. In para. 49 of ICRP’s “The Scope of Radiological Protection Regulations” it is explicitly stated that: “The 10 µSv in a year criterion has been widely used for exemption of artificial sources and its acceptance for this purpose is well recognised. For exposure from naturally occurring radioactive materials, the criterion might well be established in the order of 1 mSv in a year.” This means that the concept of triviality should be better explained in Draft Recommendation 02/276/06 with respect to both types of exposure situations. It should also be noted that the concept of 10 µSv in a year could be better clarified in this context in order not to leave the reader with ambiguous interpretations as to its application. Again, this is better explained in ICRP’s “The Scope of Radiological Protection Regulations” where para. 41 states that: “This value was intended to refer to an average individual annual dose in the critical group. Moreover, the unavoidable statistical distribution of doses in any situation of exemption would imply that some individuals may be exposed for higher doses than ‘some tens of microsieverts per year’, within a distribution that may include extremes that conceivably could be orders of magnitude higher than the average.” This is an important clarification of the 10 µSv concept and should also be included in Draft Recommendation 02/276/06. Proposal: Explain and clarify the concept of triviality and the 10 µSv concept by expressing the understanding of ICRP’s “The Scope of Radiological Protection Regulations” also in the Draft Recommendation 02/276/06 Para. 318 risk constraints The background of risk constraints is not made clear enough in the ICRP’s document to implement them in practice. The generic risk of 2 x 10-4 for occupational exposure does not precisely say what risk is meant, i.e. a fatal risk per year, per lifetime, or per incident. Exposure is defined in the ICRP recommendation as a risk. Taking this into account a risk of 2 x 10-4 would correspond to a dose of 4 mSv (single exposure). Does that mean that a dose of 4 mSv is the acceptable dose for an incident? Or should this be multiplied by the probability of this incident? There is obviously a need for clarification. Proposal: give a definition of the term “risk” used in para. 318 Chapter 10 protection of the environment It is difficult to understand, why an enlargement of the radiation protection system from a situation where the focus lies on the protection of man to the environment is really needed. The ICRP document states in para. 20 that there might be situations where there are no human beings and the environment needs protection. It would be logical that on the other hand, in areas which are densely occupied by man, as is the case at least for Middle Europe, such an enlargement of radiation protection is unnecessary. Such a statement would be helpful and it should be included in chapter 10. Proposal: include a statement that for densely populated regions the proof of a sufficient protection of humans is sufficient to guarantee also the protection of the environment The significance of the determination of doses for reference plants and animals is not adequately explained. While it is scientifically possible to perform a dose evaluation for non-human biota, dose criteria for the interpretation of the evaluation results are needed. Those dose criteria have not yet been developed in a consistent and comprehensive way for world-wide application. There is a lack of those considerations in chapter 10. Giving recommendations in the way ICRP does in this respect gives the impression that a ship starts and nobody knows where the journey will go to. Para. 389 and chapter 4.6 (para. 149) In the paragraphs and chapters on the assessment of doses ICRP addresses very well the uncertainties which are necessarily involved in every model and the necessity to derive “best estimates” of model parameter values. Unfortunately the following fact for the practise of radiological protection is not adequately acknowledged by ICRP The exposure of occupationally exposed persons can be very precisely determined by individual measurement (personal dosimetry). Thus the records will show real doses which can be compared with respective constraints or limits to evaluate the system of radiological protection For the exposure of members of the public there is basically a different situation. Doses will normally not be measured but calculated on the basis of models which in general are conservative, i.e. tend to overestimate the real doses. While comparing these calculated doses with e.g. set dose constraints for members of the public they often are regarded, as if they would be measured real doses, and their conservative and statistical character is not taken into account. In consequence that leads to unjustified exaggerations of protection measures. In this context we miss a clarification that calculated estimates of exposure cannot be treated in the same way as measured exposures, and that it is not justified to take any upper value of a calculated probability distribution of doses as an adequate value for such comparisons. Proposal: Clarify that calculated estimates of exposure should not be treated in the same way as measured exposures and give some guidance how to treat these calculated estimates.