Draft document: Recommendations
Submitted by Dr H H Landfermann, BMU - Federal Ministry of the Environment, Nature Conservation and Nuclear Safety of Germany
Commenting on behalf of the organisation

SUMMARY The new ICRP recommendation of 5 June 2006 is expressly welcomed. It is based on previous radiation protection principles and attempts to update these in light of the practical experience and new scientific findings obtained in the meantime. Nevertheless there are a number of comments, suggestions and discussion points which are briefly summarised in the following. For details, please refer to the comprehensive opinion in the Annex at the end of this summary. 1. The recommendation is based on additional documents (foundation documents, additional building blocks). These documents contain fundamental statements which cannot be found in full in the recommendation, but which are very important. For this reason the key statements in these papers should be repeated in the recommendation itself. 2. Aspects concerning the transposition into national law should be addressed with caution and restricted to scientific requirements and basic conceptual issues. 3. In practice, an essential element of radiation protection is compliance with the ban on dilution. This should be mentioned and once again expressly confirmed. 4. The clearance/exemption values should be based on 0.01 (artificial radionuclides) and 1 mSv/a (natural radionuclides outside the fuel cycle) respectively. All considerations on exclusion, exemption and clearance should be dealt with very cautiously and restricted to consistent and practice-oriented basic statements on concept, as otherwise national legislation rights are affected (see in particular opinion on the document Scope of Radiation Protection Regulations in the Attachment) (paragraph 47). 5. The appropriate area of application of the collective dose should be described in as much detail as possible (paragraphs 145-148 and 229-231). 6. The principle of justification should be explained more, especially in comparing situations in which exposure becomes higher (planned activities) or lower (existing exposures), including the impacts on individuals and society. More detailed explanations are needed especially with regard to the description of the benefit of medical research. Indication of a reference dose would be desirable (paragraphs 185-190, 247-250, 267-268 and 345-347). 7. The dose constraints are a special chapter. This should be consistently described and explained and made consistent with regard to natural exposure. Further explanations are required here especially for the areas of medico-legal exposures and therapy patients (paragraphs 202-210 and 219). 8. The area of natural exposures, insofar as these can be influenced, should be dealt with as far as possible in a similar way to artificial exposure. Deviations should remain limited to vital exceptions. A general demarcation of 1Bq/g for the exemption of natural uranium or thorium from radiation protection is too high. The explanations and derivation for this value are not convincing. The significance or role of the 40 Bq/m³ radon value is also unclear (paragraphs 291-294 and 340 – 342). 9. In the restriction of radon exposure in homes and the workplace it is desirable to have a recommendation based on assessments of the risk. Corresponding statements should also be made for thoron (paragraphs 296 – 301 and 304). 10. There should be better and more complete explanation of the potential exposure (318 and 319). 11. The new protection concept for the environment should be extended. The protection target should be named based on the overall description of the environmental pollution and consequences.The possibility of using appropriate indicators to recognise increased environmental pollution at an early stage, and if necessary abating it, should be pointed out (paragraphs 355-357). 12. Additional building block on the optimisation Optimisation should use simple and consistent decision-making processes in which decision-making is transparent and comprehensible and which concentrate on the main parameters. Practical examples would be helpful here. The doses used as a basis should be realistically determined. Data on this would be helpful. Equally, the collective dose should be defined with a view to a useful area of application. A very detailed dose matrix is probably not very helpful in practical decision-making, at any rate further methodological guidelines are needed for dealing with such a matrix. As a matter of principle the same consideration should be given to protecting future generations as to the protection of the present generation. 13. Additional building block on the representative individual At the beginning there should be a brief description of the new concept, especially how it differs from the previous one and the advantages connected with this. In particular more data should be given on living habits (planned and existing situation). The age-averaging approach should be explained and justified in more detail in terms of content and concept with regard to possible suitable use cases. The advantages and disadvantages should be discussed and the limits of such an age-averaging clearly indicated. There should be a discussion of possible implementation problems in practice with the concept of dose commitment following incorporations. Finally, the application of potential exposure should be restricted to situations for which the probability is not too low. ANNEX Comments on the Draft Recommendation of the ICRP on Radiological Protection The new Recommendation of the ICRP on Radiation Protection is highly appreciated. It carries forward the well established and proven principles of radiation protection taking into account developments in radiation protection methodology and, to the extent already scientifically confirmed, new expertise on radiation effects. It ensures and facilitates continuity and stability of the radiation protection systems on high scientific and technical standards and with close link to practicability. However, there are also some areas for further discussion and possible improvements: General comments: The new Recommendation refers to two foundation documents in Annex A and B and to three additional building blocks on the representative exposed person, on the optimisation and on the scope of radiation protection. It is not totally clear to what extent these papers will be incorporated into the new Recommendation. There are aspects and figures of these documents that are repeated in the main document, and others that are not. Important information, e. g. on the representative exposed person or on the scope of regulations are found in the respective building blocks but are not or not completely repeated in the main document. Therefore it is not clear whether these aspects are still valid and therefore to be considered as a part of the main recommendation. It should be ensured that the main statements of these additional papers as well as the outlined qualitative and quantitative criteria and requirements are reflected in the main document in a way that it can be used as a self-contained document. As a fundamental principle of radiation protection, the dilution of radioactive material aiming at meeting dose limits or other criteria is generally forbidden. This principle is reflected in most of the radiation protection legislations in the world. The new Recommendation does not place emphases on this matter. If the ICRP still holds to this aspect it should be mentioned in the recommendation. Comments to the paragraphs and additional building blocks in detail: I. Paragraph (47) under 2.4. Exclusion and Exemption Paragraph (47) refers to the detailed guidance on exemption and exclusion provided by the ICRP’s document “Scope of Radiation Protection Regulations”. Comments to this document are not presented here, but can be found in the Attachment. II. Paragraphs (145) to (148) under 4.5.7. Collective dose and paragraphs (229) to (231) under 5.8.7. Application of optimisation and constraints The limitations on the use of the collective dose discussed in the paragraphs (146) to (148) and repeated in (229) are important and need to be taken account of in practice. But the detailed statement on this matter in paragraph (146) that “In the case of low individual dose which are small fractions of the radiation dose received from natural sources and may involve wide geographical areas and/or long time scales, the use of collective dose for risk estimates is not a reasonable procedure as it both aggregates too much information for the decision making process and combines several sources of uncertainty” appears to require further clarification: Guidance on the qualitative parameters (small fraction of dose received from natural sources, wide geographical areas, long time scales) would be very desirable. As such these parameters should not entirely be left to the interpretation and judgment of the reader. In particular, options to define a meaningful lower cut-off E1 in Equation 4.12 in paragraph (148) should be discussed. For example, the de-minimis dose level of same 10 Sv/a or a fraction of it could be used in many situations as E1 level. The de-minimis dose level, by its definition, demarcates the range of trivial dose or risks, providing consistence with approaches taken in other areas of radiation protection. This also would cover the problem of simultaneous exposure to several sources. Support is also needed to chose time frame parameters and criteria for the geographic dimensions to be considered (this could be done, for example, applying the lower dose cut-off). Further, it should be discussed whether the reasons given (aggregation of too much information and combination of several sources of uncertainty) are actually the real reasons for not using collective doses in the above mentioned cases. Aggregating information and combining different sources of uncertainty is something which often is necessary within the decision-making process and, as such, not something inappropriate or generally avoidable. On the contrary, in many cases it may be misleading to limit the decision making to a few well known facts and exclude those where knowledge is limited or less confirmed. A more appropriate reasoning would be that summing up of trivial doses over many people and/or many generations may lead to very high collective doses, whilst the individual risks involved are actually trivial or practically zero due to the uncertainties. In such situation, the collective dose reflects the actual increase of the collective risk only on a weak basis. Uncertainties may play a role (in particular over long time scales) in limiting the meaning of collective dose estimates. However, this depends substantially on the situation considered (for example, uncertainties for waste management facilities on the surface or near to it are, over the same time scale, usually orders of magnitude higher than those in assessments of deep geologic facilities). Therefore, it should be mentioned that – to the extent that uncertainties restrict meaningful time frames for collective dose assessments - this restriction will depend on the situation considered and time frames should be chosen accordingly. For situations with exposures occurring over large populations, large geographical areas and/or long time periods it is proposed in paragraph (229) that - instead of the total collective dose - the collectives doses of population groups with homogeneous characteristics should be calculated and included into the optimisation process. Of course, such analysis may have the advantages of giving a more distinct picture of the exposure situation. But finally this approach will often not facilitate but rather complicate the decision-making processes because available alternative options may lead to different effects on the various dose groups in terms of degree and direction. As a consequence, decision-making becomes more difficult and a group-specific weighting of effects would be required. Here it might be difficult to decide to which group – depending on the particular option - more or less attentions should be paid. In paragraph (231) it is proposed to give more weight to medium and high doses and to doses in the near future than to small doses and doses in the remote future. This is a possible strategy but in practice difficult to handle, when aiming at transparent and reproducible decisions. In addition, this qualification gives rise to some objections. Taking into account the LNT-Hypothesis, in many cases it might be more meaningful to reduce a large number of small exposures (above trivial doses) than paying too much attention to a smaller number of higher exposures. In addition to this, exposures of future generations should be considered as important as those of current individuals based on ethical considerations. In some cases (disposal of radioactive waste, groundwater contaminations by mining activities) only future exposures may be relevant. Here and in similar situations it might be even better to compare options on the basis of total collective doses deliberately aggregating information rather than splitting up into various details which may even increase the level of uncertainty. III. Paragraphs (185) to (187) under 5.7. The principles of protection, (188) to (190) under 5.7. 1. Justification in situations involving occupational and public exposure, paragraphs (247) to (250) under 6.1. Justification of radiological procedures, paragraphs (267) and (268) under 6.4.1. Volunteers for research and paragraphs (345) to (347) under 9.4. Justification (emergency situations and existing situations) The protection of human health and the environment from the harmful effects of ionising radiation will continue to be based on the three proven principles of justification, limitation and optimisation. The principle of justification of human actions leading to changes in radiation exposure conditions is very fundamental and therefore - in general and with respect to the different exposure situations – needs more guidance. It is suggested to incorporate the following considerations into the Recommendation: As a matter of principle (with few exceptions like radiation therapy), radiation exposures of humans and the environment are undesired due to the risks they entail and therefore should be avoided or decreased. Consequently, human activities which are accompanied by a change in radiation exposure must satisfy the principle of justification if they are to be carried out nevertheless. The principle of justification is applied to human activities which lead to an increase in exposure as well as to interventions (emergency and existing situations) aimed at removing sources or decreasing existing radiation exposure. The principle of justification is always applied irrespective of the level of exposure. Activities which lead to an increased exposure therefore must be justified, notably also those below (officially) specified dose limits and reference values. The principle of justification is based on the consideration of scientific-technical, economic and social aspects and is applied – in the broadest sense – by weighting the benefits and harm connected with the human activity which leads to a change in exposure. In this complex evaluation, a human activity is only justified if the benefits, opportunities and advantages outweigh the harm, risks and disadvantages. This must also take into account the consequences for the environment. These very comprehensive and general statements should be explained more concretely and differentiations should be made with regard to practical application in the respective sectors. As a matter of principle (see above), a distinction must be made between two areas: A) situations in which additional exposure occurs and B) situations in which a decrease of an existing exposure is desired, i.e. intervention situations (e.g. accidents or remediation of contaminated areas). With regard to A), it should be particularly noted that the balancing of benefits and harms should also take into account radiation-free alternatives and their respective risks and opportunities. With regard to B), it should be noted that the benefits of decreasing an existing exposure must be weighted against the harm resulting from an intervention activity, including the costs. This can also mean that the do-nothing option is the outcome of the evaluation, i.e. that no intervention measures take place and the continued existence of exposure is accepted. In the case of very low exposures justification for A) in general will be based on the low risk incurred. In contrast, for higher exposures justification is rather unlikely. For B) the opposite applies, i.e. for higher exposures (e.g. in the field of high dose, in particular when deterministic effects could occur) interventions are generally justified. To facilitate the practical application, intervention thresholds can be laid down. For very low exposures a justification of interventions often no longer exists due to the low risk. For each area of application the benefit in its specific form and the party which benefits must be identified. A corresponding procedure takes place to determine the harm. Often the benefit or harm relates to specific individuals concerned. Sometimes , however, these relate to society as a whole. Weighting the benefits quantitatively against the harm is only possible after the benefit or harm has been quantitatively identified and methods for making the evaluation parameters comparable have been established This is sometimes very difficult. For this reason, in a number of cases the principle of justification is applied qualitatively rather than quantitatively. Applying the principle of justification is not always a one-time procedure. The process must be repeated where necessary if new relevant findings become available. Existing exposure situations: The above general principle of justification, which is based on the net benefit derived from the consideration of benefit and detriment/disadvantage/cost components, occasionally poses problems in practice, in particular with regard to the rehabilitation of contaminated sites. Firstly, extensive (costly) studies (measurements) and assessments may be necessary in order to define – quantitatively, if appropriate – the different relevant components for the consideration process and to compare them. In a phase in which it is not yet clear whether interventions are required, such great effort seems inappropriate. Secondly, the consideration process often involves a high degree of uncertainty and is partly influenced by subjective standards. The determination of the achievable reduction in radiation exposure by various means (individually and collectively, now and in the future) and the comparison with the costs associated with each option is very complex, and the results strongly depend on the assessment methodology and the underlying assumptions. If further components (e. g. qualitative attributes) have to be taken into account as well, the assessment even becomes more complicated and the vagueness and uncertainty of the results increases. Thirdly, intervention usually means that a private person/enterprise is responsible for bearing the costs (infringement of fundamental rights). Such a step requires clear, simple and legally certain decision-making criteria. The results of the aforementioned consideration process do not constitute such an instrument. Against this background, the use of intervention levels in the sense of “generic justification criteria” may be more feasible as a tool of justification, only requiring the determination whether the actual values exceed this level. Of cause such “generic justification criteria” must be combined with the possibility of not taking intervention action in the case that the intervention is clearly disproportionate. On the other hand, this approach should not rule out the possibility of taking measures for situations where the intervention levels are not exceeded, if exposure reduction can be achieved by simple means. Radiological procedures and medical research: Radiological procedures are exposure situations where it usually difficult to quantify or weight the benefit of exposure against the disadvantages, in particular the detriment of exposure. This is even more problematic for medical research activities involving radiation (volunteers for research). Sometimes high exposures to individual test persons are justified by misusing the benefit-argument (great benefit to society if findings will apply in the future). Some practical guidance for the implementation of the justification principle in this context would be appreciated, as well as greater detail concerning the justification of radiological procedures. For the volunteers in research, at least maximum exposure values should be outlined, above which justification is not or only under exceptional circumstances possible. An orientation for such values could be derived, for example, from the occupational exposure limits. IV. Paragraphs (202) to (209) under 5.8.2. Factors influencing the choice of source-related constraints, paragraph (210) under 5.8.3. Selection of a dose constraint and paragraph (219) under 5.8.6. Dose constraints in public exposures The 3-band-system for the definition of adequate dose constraints in paragraphs (204) to (206) and summarised in Table 4 appears in general very reasonable. According to paragraph (210) the selection of a constraint for a specific situation or for a certain category of situations should be done on the basis of a generic optimisation of the protection. Further guidance is given in paragraphs (211) to (220). But this reasonable system for defining constraints is, however, to some extent abandoned or at least ambiguous as far as exposures to natural sources are concerned. Firstly, in paragraph (206) the general statement that exposure situations involving abnormally high levels of natural background radiation may be in the third band (20-100 mSv/a) appears to be not convincing. Obviously, radon in dwellings belongs to the second band (see table 4) because it is an exposure situation where the individuals receive a direct benefit (living in houses). Beyond that, depending on the practical examples to be treated under “abnormally high levels of natural background radiation”, it is to be questioned whether not all such exposures would rather match the second band (1-20 mSv/a). For example, living in an area of high natural background exposure (from soil or due to the elevation) is linked to the benefit of having the social and economical basis of living this particular place. An assignment to the third band does not seem to be appropriate. Secondly, it does not appear to be reasonable to treat all existing exposure situations (exposures to the public) according to the third dose band as proposed in paragraph (219). The definition of the different dose bands for setting constraints refers to the question whether the individual receives a benefit (which is in most existing exposure situations not the case) and whether the reduction of exposure would involve unreasonable efforts or severe disruptions. The latter is also reflected in the principle of generic optimisation in setting the constraints. These principles are not fully met by generally assigning existing exposure situations to the dose band of 20 to 100 mSv. In most situations, the generic optimisation would - based on the available control options - lead to choosing much lower constraints. Practically, a general assignment to the fist dose band would be appropriate and feasible with only few exceptions. Only in very rare cases constraints in the third region might result. The German uranium mining remediation project shows that a general dose constraint of 1 mSv per annum can be appropriate. Practical remediation options which are also economically feasible exist which allow to reduce doses below this constraint. Similar approaches have been used in several other countries for existing exposure situations. In such situations, there is no reason to place a substantial health burden on people (with no benefit) by accepting annual doses of 20 mSv or more, while having feasible options available to further reduce the exposure. In principle, the fact that the exposure situation in question is not a planned, but an existing one, does not justify such high dose constraint. Rather, such existing exposure situations should – where possible – not be handled different from situations where the individuals who happen to live near a new site (nuclear reactor, disposal or reuse of NORM) are protected by a dose constraint below 1 mSv/a. Using the first dose band and a principle dose constraint of 1 mSv per annum (with the possibility to allow for exceptions to prevent disproportional efforts or other severe disadvantages) would also be consistent with the optimisation principle for existing situations stated in paragraph (59) of the Additional Building Block “The Optimisation of Radiation Protection – Broadening the Process” stating the following: “In case of existing situations, including those arising from emergencies and past activities, the optimisation aim at the progressive reduction of individual doses towards the level that are applicable for normal situations”. The third band should be reserved for immediate actions in emergency situations only. Even long term clean-up activities with respect to emergency consequences should not exclusively be assigned to the third band but – where appropriate – also to lower values. V. Paragraph (267) under 6.4.1. Volunteers for research (269) under 6.4.2. Medico-legal exposure These paragraphs do not fit into the section 6.4.The optimisation of protection for patient comforters and carers. They should rather be separated. Insurances under medico-legal exposure are only one example for this specific category, another one is customs check to find drugs or immigrants. Under Medico-legal exposure it should not be said that public constraints are not applicable and therefore the authority should use higher values similar to those for carers and helpers of a few mSv. Rather the formulation should qualify that, where inevitable, higher values than public constraints can apply but not higher than those of carers and helpers. VI. Paragraph (276) under 6.5. Release of patients after therapy with unsealed radionuclides Following the basic principle of radiation protection it would be desirable to add to this paragraph that the decision whether and how long a patient should be hospitalised after a therapy, e. g. with high activities of radiopharmaceuticals, should be based on an optimisation process taking into account the minimisation of the exposure of other individuals to the extent reasonable. VII. Paragraphs (291) to (294) under 7.3. Controllability of natural sources and paragraphs (340) to (342)] under 9.2. Existing situations As stated in the Recommendations, it is a historical fact that natural radiation has not been dealt with comprehensively in the past radiation protection strategies although it is by far the largest contributor to human exposure. The main focus was put on the radiation protection in nuclear industries and the application of artificial radionuclides. The reasons are seen in the different perception of radiations risks caused by natural and man-made exposures and the accordingly following prioritisation. Undoubtedly, such different levels of concern of the public and some times of the decision makers exist regardless of the facts that harmfulness of radiation does not depend on its origin in nature or activity of man and that there is no clear borderline between these two categories (e.g. “man-made” enhancement of concentrations of natural radionuclides in industrial by-products and residues as well as in the environment). Another important aspect is that amenability and controllability of the artificial and natural radiation different sources may differ due to the ubiquitous existence of natural radionuclides and radiation. However, the societal perception of radiation and risks is without doubt a factor not to be neglected in the decision making processes, but this does not constitute a health or radiation protection principle. It is not meaningful and without scientific basis to establish radiation protection approaches depending on the public perception aspects or on stakeholders’ wishes. Therefore it appears questionable to incorporate this way of thinking at length into the Recommendations (e. g. as it is done for existing situations in the paragraphs (340) to (342)) and – by doing this – manifest and qualify the perception aspect as a primary radiation protection principle, especially when no further – scientific or technical - criteria are mentioned. Of course, there are many differences between man-made and natural sources and these lead very often too different levels of controllability and, as a consequence, to different dose constraints. But for the recognition of these differences it is not required to revert to the common public perception and different levels of societal concern for natural and man-made sources (stakeholders). Instead of basing the radiation protection elements of this important area (naturally occurring radionuclides) with its comparatively high exposures on beliefs, it should at least be attempted to choose a basis which is as much as possible focussing on scientific and technical arguments and balancing these facts with practicability and finally (but not as a primary factor) with other reasonable aspects like stakeholders’ wishes. The exclusion level of 1000 Bq/kg for the heads of the uranium and thorium series as quoted in (294) and as before mentioned in the ICRP – Recommendation (ICRP, 2006) “The Scope of Radiation Protection Regulations” (also referred to in paragraph (47)) is from a health protection point of view not the adequate starting point for the above mentioned attempt. Although this value probably reverts to the exemption levels published by the IAEA in RS-G-1.7, it should be noted that this value has not been derived on a risk basis and that, therefore, this publication contains provisions for the regulatory authorities to define lower levels if needed, especially for building materials. According to the German experience, lower criteria are required, for example, even for unmodified materials containing elevated levels of naturally occurring radionuclides if these are used in the construction of houses or disposed of in large amounts. For example, depending on the site and material parameters a waste rock dump of about 10 ha with a average concentration of 1000 Bq/kg of the U – decay chain in equilibrium can give rise to exposures of the public of several mSv/a, mainly via the groundwater pathway. Taking into account that these exposures are high and in principle amenable to control and, as a consequence, practically avertable to a great extent, the value of 1000 Bq/kg would conflict with the statement of the Recommendation in paragraph (280) under 7. Exposure to natural sources stating that principles for exclusion and exemption of natural sources are that individual risks are insignificant and radiological protection is optimised. Therefore, this value can neither serve as a general reference point for exclusion and exemption not for the design of radiation protection systems in general. Moreover, the argument which is often referred to that this concentration is of the order of the high range of those prevailing in “normal” soil or rock in the world does not convince. First of all, the most extreme situations in the world should not be chosen as benchmarks to establish radiation protection concepts of individual countries with frequently totally different general exposure situations in terms of prevailing background. Secondly, the proven concept of using the exposure of the affected individuals as an essential criterion for human health protection (dose concept) should not be abandoned in favour of concentration values which are not linked to the exposure. Rather, the variation of the natural exposure in a representative area or number of countries could be used as a reasonable scale to be considered, e.g. stating for Europe that an additional exposure below a value between 0.3 and 1 mSv/a caused by exposure situation which are not warranted to be controlled may be acceptable. Exposures above these values are still well amenable to control measures without undue burden on industries and regulators – as experience in Germany shows. For convenience the so derived additional acceptable dose of e.g. 1 mSv/a could then be converted into concentration values. Using realistic scenarios as applied in Germany, 1 mSv/a corresponds to a lower boundary of 0.2 Bq/g for the primordial radionuclides of the U - and Th - series which is at the same time equal to a very high percentile of radionuclide concentrations in “normal” soil and rock in Europe. For naturally occurring radionuclides in the nuclear fuel cycle it may even be possible to apply the 10 Sv/a – criterion as commonly used for artificial radionuclides. The above discussion is not aiming at introducing the German radiation protection approach for natural exposures into the ICRP’s recommendation, but it is recommending to replace or at least substantially amend the concept of public perception and of criteria which are difficult to justify by more objective and sound criteria. In paragraph (294), a radon exclusion level from radiation protection of 40 Bq/m3 is proposed by referring to the variation of residential radon concentrations between different regions (representing the global mean indoor radon concentration). The meaning of this value and its practical applications needs clarification. Does it apply to indoor or outdoor radon or is it a general exclusion level for radon concentration in ambient air? What are the circumstances and the purpose for its application? What is the practical benefit of this value? Depending on the answers to these questions, an explanation is required on how the value of 40 Bq/m3 is justified (unimportant exposure, no amenability to control, control unwarranted …). VIII. Paragraphs (296) to (301) under 7.4. Constraints for radon in dwellings and workplaces Paragraph (296) refers to the new studies and available information on radon risk of people at home. These estimates explicitly show that in the range where currently action levels for dwellings are recommended (between 200 and 600 Bq/m3) and even below this range, the risks to people is substantially increased. Although national authorities may establish their own radon constraints (as with other sources) including the application of optimisation and even taking into account the controllability of radon in homes and economical factors, the above epidemiological findings imply that a radon constraint of 600 Bq/m3 for homes as given by the Recommendation might be too high, especially for new buildings. IX. Paragraph (304) under 7.4. Constraints for radon in dwellings and workplaces For radon-220 (thoron) the current conversion convention suggested by the ICRP’s publication 65 is not applicable to thoron decay products. It would be appreciated if the ICRP could close this gap. X. Paragraphs (318) and (319) under 8.3. Assessment of potential exposures For clarification it should be pointed out in paragraph (318) that the risk constraints given there are annual risk values. Furthermore, paragraph (319) should be extended to the fact that the concept of risk constraints (unconditional probability of incurring a health effect, given by the combination of the probabilities that a radiation dose will be incurred and the probability of harm by this dose) is not only limited by the uncertainties of predicting very unlikely events but also when a very low probability of an event is combined with a high dose if the event happens. As an example, in paragraph (97) of the Additional Building Block “Assessing Dose of the Representative Individual for the Purpose of Radiation Protection of the Public” the potential exposure on an area contaminated with sparsely distributed hot particles (Atoll of Mururoa) is mentioned. In contrast to a uniform contamination, the probability that an individual would be exposed by incorporating a hot particle is very low. But the dose incurred if incorporation happens may be very high (possibly even leading to localized deterministic effects). Similarly to the considerations related to the uncertainties, it should be proposed in paragraph (319) that in the above situation the unconditional probability of incurring a health effect may be misleading and that it is recommended to treat probabilities of receiving a dose and probabilities for health effects as separate quantities in the decision-making process, especially with respect to the optimisation of protection. In general, the problem should be discussed in more detail in the document and it should be made clear that the approach of the unconditional probability of incurring a health effect will often but not always be adequate. XI. Paragraphs (355) to (357) under 10.2. Reference Animals and Plants The protection of the environment receives growing attention in radiation protection. Whereas until now radiation protection has considered the environment only with regard to the transfer of radioactivity through it, basically looking at the exposure of humans caused afterwards, it is now substantially broadening with respect to effects on flora and fauna, irrespective of any human connection with it. In addition to the still valid assumption that the protection of the environment is ensured indirectly by the control standards for the public and that in general species are not put at risk as long as the public is protected, it is now proposed to seek ways to demonstrate, directly and explicitly, that the environment is being protected sufficiently. In order to achieve this, the Recommendation focuses on the development of a broader scientific understanding of the consequences of exposure to non-human species (radiation impact on biota by radionuclides entering the environment) and proposes – assisted by the concept of Reference Person – the development of a set of Reference Animals and Plants. This Reference Animal and Plant approach - complementing the human protection system - shall be the basis for drawing broader conclusions on radiation consequences for other types of organisms and shall serve as a trigger for detailed investigation where warranted. Aiming at sustainable protection of the environmental from radiation and other hazards (on a more global perspective) the approach outlined above alone does not seem to be conceptually sufficient. A comprehensive concept pursuing the principle of sustainability more stringent should not only look at criteria which are based on the extent to which the environment and reference species can bear the burden of pollution (in the worst case allowing for exploiting the “environmental capacity” to the maximum possible dose limit for reference species in remote places where humans are not present), but should also taking into account the fundamental criterion of contaminating the environment only to the extent unavoidable. In order to accomplish this idea it would in addition to the above concept be preferable to arrange for identifying changes in the environment at a very early stage by means of monitoring individual physical indicators (e. g. crypton-85 in the environmental media, radionuclides in the oceans etc.). Such an indicator system of monitoring negative changes in the environment would be capable of detecting cumulative pollution by various sources in the long run and would work cost-effective and efficiently with respect to an early detection and effective countermeasures to reduce emission by the sources. XII. Glossary of key terms and concepts The glossary as it is now covers the radiation protection terminology and methodology in general. Certainly it is beneficial to have such a compilation, but it should be thought about limiting the glossary to the terms and facts used in the Recommendation. XIII. Additional Building Block “The Optimisation of Protection – Broadening the Process” The principle of optimisation of protection – as a central element of radiation protection - is further substantiated by the Additional Building Block “The Optimisation of Protection – Broadening the Process”. The following comments refer to this paper. XIII.1. List of attributes to be used in the optimisation process The table on Page 19 provides a non-exhaustive list of representative attributes to be used in the optimisation process to select the best protection option. These attributes are described further in the text (in particular in 5.Exposure Distribution). Although it is appreciated that the decision-making process should consider all relevant facets of the situation under consideration, there is a number of questions with regard to the practical application: One of the aims of the Recommendations is to simplify the system of protection. Accordingly, for example, only three age groups instead of seven and further-on gender-averaged data instead of gender-specific data shall be used for estimating annual doses of the representative individual in prospective assessments, e. g. in comparing calculated doses with dose limits. These simplifications are abandoned with respect to the optimisation process although they here (for optimisation) seem to be even more adequate (addressing doses in the tolerable dose range below the applicable constraints or limits) than with respect to the evaluation of dose limits. For consistency reasons and since the optimisation process usually deals already with many different factors, any simplification possible and meaningful should apply. The decision-making process should lead to a clear preference of a certain protection option and the reasons for this decision should transparent and defendable. With the number of attributes to be considered the process becomes more and more difficult and the quality of decision-making does not always improve with the number of attributes taken into account (such as age and gender, see above). Only attributes relevant for the situation under consideration should be involved, while attributes of little or non significance reduce transparency and ability to communicate the results and should therefore be abandoned. The number of attributes actually to be considered should be reduced to the necessary minimum. It would be beneficial to have practical guidance in how to identify the important attributes. Some of the attributes in the optimisation process are quantifiable; others are not and therefore need qualitative judgement. Especially stakeholders’ wishes can be important qualitative attributes to be included. Combining quantitative and qualitative attributes in a comprehensive optimisation process very often represents substantial problems and may lead to some arbitrariness of the result of the decision-making being influenced by different perspectives and judgments of involved persons. Even in cases where only quantifiable attributes are relevant, the decision-making requires that attributes of different kind, e. g. exposure and costs, are put into a rational relationship reflecting their relative importance and “weight” in the decision-making process. Practical guidance, perhaps by using examples, would be very helpful in this respect. XIII.2. Dose assessment in the optimisation process (also paragraphs (228/229) of main recommendation) Usually the assumptions for dose assessments with respect to the investigation of compliance with dose limits are conservative. This is also reflected in the Concept of the Representative Person. The Additional Building Block “Assessing Dose of the Representative Individual for the Purpose of Radiation Protection of the Public” defines the habits of such a reference person (lifestyle, food consumption etc.) by reflecting the 95 % percentile of the actual distribution of the population in question. This approach entails substantial conservatism with regard to actual doses received by the majority of the population. This is adequate for compliance assessments with dose limits because these are defined as the margin of the unacceptable dose range (under normal conditions). For optimisation considerations, however, this degree of conservatism is not required and could lead to unnecessary protection efforts (including high costs and possibly other types of detriment) if the exposures which are being prevented by the protective measure to a large extent are only hypothetical (very low probability). Therefore, dose assessments which are used in the optimisation process should be more realistic. The problem should be discussed and practical advice should be given on adequate conservatism. Paragraph (68) advises against the use of the collective dose for large populations, large geographic areas and large periods of time. Certainly these aspects represent important limitations of the collective dose concept. However, for practical application a judgment of how large is too large with respect to these parameters is crucial. The only quantitative guidance can be derived from Figure 7 in Annex A, providing weight factors for future exposures. These factors suggest that exposures over an order of magnitude of 1000 years should be fully taken account of in the decision-making process (weighing factor equal to 1). This implies that such exposures also could be seen as meaningful contributors to the collective doses. It would be warranted if corresponding indications of the meaning of ‘large’ were discussed in the main Recommendation and if some quantitative criteria or practical approaches were given, to the extent possible. Furthermore, guidance would be helpful how to define subgroups of homogeneous characteristics (see paragraph (70)). This guidance should take into account that not every possible distinction of subgroups is important for the decision-making. On the contrary, it should be considered that too many subgroups may complicate the decision-making process (see above). Paragraph (69) states that the use of collective doses may mask the inherent uncertainties attached to the dose assessments. This may be the case for assessment carried out over very long timescales and very large areas. But this is not necessarily the case for population groups confined in space and time. For such cases, the assessment of collective doses may even be less uncertain than the assessment of individual doses for the representative individual, at least, when realistic assumptions for habits are used (see above). The reason lies in the fact that extreme behaviours, which may determine doses assessments for the representative individual, will be averaged out in doses assessments for larger population groups. Furthermore, it is possible to estimate uncertainties in collective dose assessments, for example by performing probabilistic modelling. The use of confidence bands for collective doses is actually often easier than determining percentiles of an individual dose assessment. The reasons are a better statistical basis and the fact that the confidence bands are mainly determined by more or less average habits instead of extreme habits in the case of representative individuals. XIII.3. Dose matrix Figure 3 shows a complex matrix representing dose information about the relevant groups of individuals in space and time. In addition to these dose attributes, the decision-making requires to include further aspects, e. g. costs and stakeholders’ wishes. All these attributes are to be considered adequately and according to their relative importance. It is without doubt that such a multi-dimensional matrix as shown in Figure 3 can help to present results and the reasoning behind proposed decisions. However, it does not solve the basic (and almost always present) dilemma of decision-making, which is that in non-trivial situations usually many different attributes with different units of measurements (or even without, for qualitative factors) and usually conflicting preferences with regard to the protection options under considerations have to be compared to each other. Further disentangling the dose components into population groups with more or less homogeneous characteristics (e.g. different villages and different time periods, gender, ages and living habits) would substantially complicate the decision-making process. The draft document does not give any hints of how to actually approach this basic challenge of decision-making and how to handle the great variety of information. Reference is only made to earlier ICRP Recommendations presenting several approaches for decision-aiding techniques and it is stated that the Commission still believes that these can provide useful input into the decision-making process by providing decision aid. However, the presentation of the dose matrix in the draft document seems to raise the complexity of the decision-making process. A large number of attributes for the decision-making is mentioned (see above comments) and no indication is given how to find an optimal balance between these factors in practice. In complex decision-making situations it seems difficult to imagine that the use of such a matrix system actually would facilitate the decision-making process. XIII.4. Protection of future generations (also paragraph (233) of main recommendation) Figure 6 in Annex A2 and the corresponding discussion leave the impression that the assignment of relative weights to doses received at different time scales is a quite arbitrary process, which could have very different (even opposite) results for different situations depending, for example, on the stakeholders’ consideration. This seems to be in contradiction with other basic radiation protection principles. One of these fundamental principles (in particular relevant for the disposal of radioactive waste) is that future generations should be subject to the same level of protection as current generations. This would suggest equal weights for doses received in different time frames. It seems to be in contradiction with this principle if the option for decision-makers in a specific situation (possibly based on interaction with their particular stakeholders) exists to more or less arbitrarily abandon this principle and either assigning higher or lower weights to the protection of future generations. It is questionable to actually leave the above fundamental principle of the protection of future generations in favour of the discretion of specific decision-makers and stakeholders. Obviously, uncertainties restrict long term assessments and limit meaningful time periods for dose assessments. The type and extent of uncertainties depends on the concrete situation dealt with, which will have to be considered by the decision-makers and stakeholders. To limit assessment time frames based on such considerations will be necessary in many cases. This, however, does not constitute just a preference of people involved in the decision-making, because technical arguments and guidance for these can be (and should be) given. Independently of how the discussed issue will be dealt with, the right part of Figure 6 contradicts with Figure 7. Furthermore the statement on Page 47 that beyond time frames of a few generations predicted doses should not play a major part in the decision-making process appears to go too far. This would mean that for the disposal of radioactive waste exposure assessments in most cases would not influence the decision-making, because from most disposal facilities significant doses are only predicted in the longer term beyond a few generations. Disregarding possible exposures from disposal facilities in total would not be consistent with the approach taken to radioactive waste disposals in most countries and it would also contradict ICRP Recommendations on this subject. XIV. Additional Building Block “Assessing Dose of the Representative Individual for the Purpose of Radiation Protection of the Public” XIV.1. Characteristics of the representative individual In defining habit data for the representative individual, no distinction is made whether they apply to planned practices or, for example, to existing exposures due to past activities (intervention). But conservative elements (such as using the 95th percentile discussed under 3.3.Characteristics of the representative individual) may be only appropriate for planned practices, but not necessarily for intervention situations in which dose assessments often are based on more realistic assumptions in order to avoid disproportions. Similarly, more realistic assumptions may have to be used for exposure assessments in the area of NORM industries as compared to assessments performed for exposures within the nuclear fuel cycle. Another aspect which may determine the nature of assumptions is the purpose of the assessment: Demonstrating that a dose limit is met requires sufficiently conservative assumptions to be able to prove that chances of exceeding the limit for any person are very small. Using a dose assessment, on the other hand, to demonstrate compliance with a dose constraint or within the process of optimization may be performed using less conservative assumptions because individual persons with extreme habits are of less concern in these cases. It should be considered to extend the document to the above discussion and – aiming at more flexibility - to allow for different types of assumptions to be used for the representative individual depending on the scope and purpose of a dose assessment. XIV.2. Representative Individual versus critical group concept The fundamental difference between the concept of the representative individual and the critical group concept does not become really apparent. There are some discussions about this, but they are partially more confusing than helpful. For example, the discussion in paragraph (65) on the weakness of the critical group concept implying a detailed knowledge of local habits and on the advantage of the representative individual concept allowing to use national or regional information about the 95th percentile is not very convincing. In many countries (such as Germany) food consumption rates are used for assessing compliance with dose limits which are based on such percentiles of national data. Their applicability for local critical groups (e.g. living in the vicinity of a reactor) is seen as a given, so that the problem discussed does not (and did not) arise. On the other hand, if local habits do affect exposures, they will require consideration independent of which concept (representative individual or critical group) is used. It would be beneficial to add a discussion to the document which clearly identifies the changes in concept which the Commission has in mind with formulating the new representative individual concept as well as the advantages associated to it. XIV.3. Averaging doses over 5 years The Recommendation states that averaging doses over a 5-years-timespan can be appropriate in the case of prospective dose assessments (e. g. when designing a radiation protection concept for a particular practise or facility). Since dose limits apply to the sum of external and internal dose, averaging the intakes over a period of five years may be appropriate to provide some flexibility. On the other hand it is recommended to incorporate all available details (detailed age dose coefficients, no averaging of intake, gender-specific data etc.) in the case of retrospective dose assessments. According to this Additional Building Block (paragraph 35) retrospective dose assessments take place e. g. when demonstrating compliance with dose constraints (dose limits). It might be problematic to apply different assessment methods when designing and licensing a practice or facility (prospective) and when verifying compliance with the limits and constraints set (retrospective) because contradictions are very likely. In order to avoid such contradictions it should be clearly stated that no different dose assessment methods should apply throughout the whole process of designing, implementing and verifying radiation protection irrespective of whether the different steps are prospective or retrospective. Under 3.4 Age-specific dose coefficients as well as under 5.9 Dose limits of the main Recommendation, e. g. for flexibility reasons, a simplification of the age groups concept by allowing for averaging doses over 5 years is proposed. Paragraph (79), however, seems to contradict this approach because it appears to say that the foetus or the breast-fed infant should be used in compliance assessments if their estimated doses are higher than those for the age group of 0 to <6 years. Using this approach, there would be no longer a difference between the new averaging concept and the current detailed age groups concept in the case of infants. Since doses to the breast-fed infants often dominate dose assessments, it should be clarified whether the Commission believes that averaging exposures over five years is adequate and, if so, the discussion in paragraph (79) should be revised accordingly. Related to the previous comment, it should be discussed whether it is actually appropriate to base dose assessments for the simplified age categories on dose coefficients for specific age groups (Table 2). It should be taken into account that dose assessments are not only determined by the dose coefficients, but also by many other assumptions which change, in particular, over the first years of life. Apart from factors like food consumption rates there are assumptions which can give rise to additional exposure pathways depending on the age of the child. For example, it will only be reasonable to assume that children play on their own on contaminated areas (e.g. accessible mine waste dumps) after a certain age (e.g. above two years). Such aspects would be lost if assessments were based only on one year old infants. A better approach to avoid such problems would be to base the assessment of the age category of 0 to <6 years (and, for consistency, for the others as well) on the average of the doses separately estimated for the first five single ages groups. Using this as a general rule, it still could be permitted to use the simplification of estimating doses only for single years as being representative for an age category if circumstances permit (i.e. in particular if no substantial changes between exposures at different ages within this age category are expected to be present). XIV.4. Potential exposure As discussed above with regard to the main Recommendation (X. Paragraphs (318) and (319) under 8.3.Assessment of potential exposures) the proposed use of the unconditional probability of incurring a health effect under 4.2. Situations of potential exposure should be limited to situations where it is adequate. It should not apply in cases of very low probabilities of exposures and associated high doses if the exposure happens (see example in paragraph (97) and discussion above). Applicability should be discussed in greater detail and further guidance should be given to this respect.