Specific Comments from the NEA EGIR General Comments on the Introduction: • A Scope and purpose section should be added • This chapter should express the idea that the RAP concept is a simplification that can be used to address the complexity of environmental RP. • As mentioned in the EGIR General Comments, there is a need to either sketch out what the Framework is, or to reference where this can be found. This information should be provided in the introduction. • There is a need to have a section describing the document’s scope and purpose Paragraph 3 • Although this sentence references population effects, the remainder of the document focuses on individuals, not populations. The ICRP should make it clear up front what the aim of the document is ( protection of individuals or populations). If the objective of the document is not to define at what population level protection should be directed toward, that, too, should be clarified here. So this should be made clear up front. • Early in the document the ICRP should link its recommendations to what countries are already doing, at least noting that the environment is already protected by many national programs. The **highlighted** sentence should be added to this paragraph: (3) The Commission has therefore broadened its scope in order to address the subject of environmental protection. **The aim of this document is not to determine effects on populations.** Its aims now include that of preventing or reducing the frequency of deleterious radiation effects to a level where they would have a negligible impact on the maintenance of biological diversity, the conservation of species, or the health and status of natural habitats, communities, and ecosystems. In doing so, the Commission acknowledges that, compared with human radiological protection, the more detailed objectives and management of environmental protection are often more complex and difficult to articulate. There is no simple or single universal definition of ‘environmental protection’, and the concept differs from country to country, and from one circumstance to another. Paragraph 4 • National approaches will take precident over ICRP recommendations. The “fitness” of an organism for survival may also be considered as an important endpoint of this assessment – is this included within Morbidity – this could be mentioned in this paragraph (4) The Commission believes that its approach to environmental protection should be both commensurate with the overall level of risk, and compatible with other approaches being made to protect the environment from all other human impacts, particularly those arising from similar human activities. **In national contexts where environmental protection criteria already exist, the Commission’s recommendations may be considered as complementary.** And in view of its overall aims, again compared with human radiological protection, it believes that other ways of grouping radiation effects are likely to be more useful for the protection of non-human species, such as those resulting in early mortality, or morbidity, or reduced reproductive success. Paragraph 6: Referring to the **highlighted** phrases: • The document may “provide a vital component of the framework”, but the ICRP should be more clear about what the Framework is. • It seems that the Commission will eventually make some recommendations, but not “for the time being”. It could be stated here that this report does NOT establish the Framework, because this does not seem to be the case. However, the Framework, wherever it is, should be referenced or described. • The Commission may in the future set dose limits What DOES the Commission intend to do with this document?? The place of this document in the Commission’s recommendations is not clear. (6) Such advice and guidance has to be transparent, and have a common basis arising from our knowledge of exposure to radiation and its effects, set within some form of overall framework. This framework should therefore serve as a basis from which national and other bodies could develop, as necessary, more applied and specific numerical approaches to the assessment and management of risks to non-human species under different circumstances, and different exposure situations. Because of the vast complexity of the living environment, and the limited radiobiological and radioecological data bases relating to it, the Commission considers that, by setting out data for a limited number of Reference Animals and Plants, this will **provide a vital component of the framework** to gather and interpret data in order to provide more comprehensive advice in the future. This information is, however, still fragmentary and the Commission does not **for the time being intend** to set specific dose limits for environmental protection. Paragraph 7: • The addition concerning scientific uncertainty is to clearly flag this early in this document • “the application of the basic approach to different exposure situations” should be the first item on this list of further publications as it is central to moving forward usefully. • The **additional** sentence, placing RAPs in the context of population and ecosystem protection, presents a key topic that should be clearly addressed (7) This report, on the concept and use of Reference Animals and Plants, therefore serves as an introduction to the complex subject of environmental protection with regard to radiation. The current level of knowledge is preliminary at best, and includes much uncertainty. It introduces the rationale for selecting the Reference Animal and Plant types, and then gives emphasis to their biology, to basic aspects relevant to their dosimetry, and to a review and application of what is known about relevant irradiation effects for such types. Further publications will address in more detail such aspects as data bases for modelling exposure, refinements to dosimetry, issues such as RBE and radiation weighting factors, the application of the basic approach to different exposure situations, and how the Commission’s approach to environmental protection compares with others commonly used in relation to industrial practice. **The issue of using RAPs in the context of population and ecosystem protection will also be the subject of future publications.** General Comments on Chapter 2 • There is a need to be clear why these RAPs have been selected – that these choices will be sufficient for use in protecting the environment. Paragraph 10 is not currently sufficient. • There is a need to show that the modeled doses to RAPs are actually comparable to “actual doses” that were used as the basis for assessing responses. It is understood that the ICRP approach is, of necessity, a simplification, but some understanding of the level of conservativeness should be provided, and there is a need to better understand the level of uncertainty. • The term Derived Consideration Level uses the word “derived” in a judgmental rather than a mathematical fashion. The term then gives a more scientific flavor to the concept than actually exists. Some other word, referring to the judgmental nature of the criteria, should be chosen. The word derived is usually taken to mean a secondary quantity based on some primary quantity, and this is not the case here. (it should also be noted that DCL can also be confused with Derived Concentration Level, a completely different concept). • The document needs to be clear on whether the Chapter 2.2 criteria are for selecting reference animals and plants, or more generally any organisms needed for environmental impact assessment. • It seems that the RAPs have been selected based on a pragmatic assessment of available data and a manageable number of organisms for practical use in a regulatory context. This should be explicitly stated. Paragraph 9 • Here, and in section 6.2, natural background is mentioned. However it is not make clear how typical values relate to the huge variability in natural background. This could be mentioned here. • It should also be noted that this paragraph suggests that effects should be linked to background dose rates, but the text throughout the document talks about effects and not about background. • It is also suggested that it may not be consistent with publication 103, which does not use natural background as a reference point. • It is not clear from the text that confounding factors have been appropriately accounted for in these assessments. It should be mentioned that this is or is not the case. • Again, there is no balance between data presentation (that would be expected from UNSCEAR) and data assessment. (that would be expected from ICRP). (9) The Commission has now decided to use a similar system of discrete and clearly defined Reference Animals and Plants for assessing radiation effects in non-human organisms (ICRP, 2003b, 2008) based on the concept developed by Pentreath (1998, 1999, 2002 a,b, 2003, 2004, 2005). This approach involves the use of a limited number of different types of animals and plants as a systematic basis for relating exposure to dose, and dose to different categories of effect, that could be interpreted in terms of the normal biology of these particular types of animals and plants in environmental situations. The effects considered to be of relevance were those of early mortality, morbidity, reduced reproductive success, or some form of observable chromosomal damage, irrespective of whether or not they arose from stochastic or non-stochastic dose effect relationships. It was further considered that it would be helpful to decision making if this information was set out in terms of multiples of the natural background dose rates typically experienced by each type of animal or plant, in the form of Derived Consideration Levels. Paragraph 10 • What about the concept of sensitive species? In human protection, the representative person now represents the “most exposed population”, and in that sense the population needing protection in priority. Do the RAPs address this need? It is perhaps not necessary to provide details on this, in that sensitive species will exist in a great variety, and critical species may exist for the environment in question, but it is perhaps worth while to mention that this may need to be taken into account in assessing environmental impacts, and to reflect on how RAPs and sensitive /critical species should be used. (10) The approach therefore acknowledged that one cannot provide a general assessment of the effects of radiation on the environment as a whole. Nevertheless, it was considered that it should be possible to derive, in time, a reasonably complete set of internally related information for a few types of organisms that were typical of the major environments. Thus, by using sets of dosimetric models and environmental geometries relating to such reference animals and plants, with precisely defined biological characteristics and life histories, and applying them to distributions of radionuclides in different environments, one should be able to make a judgement about the probability and severity of the likely effects of the radiation exposure on such organisms. One should then, in turn, be able to make an assessment of the likely consequences either for individuals, or for the relevant population (depending on the environmental management issue being addressed) using these and other environmental data and information, for such types of animals and plants. Paragraph 11 • The text here says that the work can be used for “drawing comparisons with other – and probably more limited – sets of information on other organisms”, however this is poorly explained in the text. • The text states “It is also similar to the concept of assessment and measurement endpoints used in ecological risk assessments frameworks”, but in other approaches to environmental protection, referenced here, the use so Sentinel Species is common, and this text specifically says that RAPs are NOT sentinel species. The aspects of this approach that are similar to other ecological risk assessment frameworks should be made more clear. • It is irrelevant here that this approach is supported by the IUR. If this is kept, other relevant organisations would have to be asked for their views, as such, the last sentence of this paragraph has been deleted: “The need for such a basic and generalised framework for environmental protection had also been strongly supported by the International Union of Radioecology (Strand et al, 2000)” (11) The concept is therefore similar to that used for human radiological protection, in that it is intended to act as a foundation for the making of a number of basic calculations, and to serve as points of reference for drawing comparisons with other – and probably more limited – sets of information on other organisms. Such a basic reference-animals-and-plants approach had been used previously to provide advice at an international level, primarily in order to establish release rate limits to evaluate potential environmental impacts of radionuclide releases into the marine environment (Pentreath and Woodhead, 1988). This was applied by the IAEA to redefine annual release rate limits for the purposes of the London Convention (IAEA, 1988). It is also similar to the concept of assessment and measurement endpoints used in ecological risk assessments frameworks (Suter, 1999), and to the approach recently used in the shape of ‘reference organisms’ (variously described over a range from multi-phylogenetic assemblages, to generalised phylogenetic types, down to individual species) to assess ecological radiation exposures in Arctic and European environmental situations (Brown et al, 2003; Larsson, 2004). Heading 2.2: • Add the word “reference” in this title 2.2 Criteria for choosing different types of reference animals and plants Paragraph 14: • There may be some types of requirements that may be missing in this paragraph: o Geographic variation o Ecosystem sensitivity o Ecosystems that are not exploited by humans (14) Given that the objective is to provide a starting point for the assessment of radiation exposure, of radiation dose, and of possible dose responses, for such an enormous variety of living animals and plants, it is clearly not easy to select a few biological types for the purpose of creating a small reference set. Thus notwithstanding the need for number of scientific criteria for their selection, one of the first consideration is that of what the information is likely to be used for, and under what circumstances. These are anticipated to include the following: Paragraph 15: • It is perhaps implicitly obvious that the protection of plants and animals will be protective of ecosystems, but it seems that this is a necessary assumption to the statements in this paragraph. As such, it could be explicitly stated. This is partly referred to in paragraph 60 (15) **There are several features common to all of these requirements. One is the assumption that the assessment of radiological effects on plants and animals can be related to broad environmental protection objectives. Another** is the need to have a consistent and transparent approach to relating exposure to dose, and then relating dose to what is known about different sorts of effects on different types of animals and plants. But it is appreciated that the application of this type of information **(both the scientific dose/effect relationship and the regulatory response)** may vary substantially at national or regional level. Paragraph 19: • It should be considered to include a criteria on radiosensitivity **or importance to the ecosystem**, e.g. bacteria has been excluded because it is radioresistant. Other organisms may be somewhat radiosensitive and be of importance to the ecosystem under consideration. (19) Collectively, therefore, in selecting a small but practical set of reference animals and plants, the following criteria were used: Section 2.3: • It is not clear how the information in this section relates to the selection of RAPs. • An introductory paragraph explaining that this section is aiming at identifying practical rationale for selecting RAPs would be useful. • This section should discuss how the selection of RAPs (at the Family level) will impact the uncertainty of effects that are then to be protected against, since effects will be observed at the species and sub-species level. Paragraph 22 • In the 6th sentence, replace “…plants and animals..” by “…plants or animals…” (22) The classification of animals and plants is primarily a reflection of their morphological characteristics, plus physiological and biochemical features, and often draws upon what is known or assumed about their evolutionary history. Such approaches are now greatly strengthened by the use of DNA analyses. Animals are grouped into Phyla, on the basis that each Phylum has, more or less, the same ‘body plan’ (such as chordates, or echinoderms, or arthropods) and within each Phylum they are further grouped into Classes, then Orders, then Families (which share ‘typical’ traits and features), and then Genera as the number of features they have in common increases; finally, Genera are divided into species. There is no absolute definition as to what a species actually is, but it is usually taken as a description of individuals that (it is either known or expected) can only produce fertile offspring as a result of mating with similar individuals. In some cases, even further distinctions are made – into sub-species, or into races and varieties. Plants, too, are characterized in relation to features such as anatomy, embryo characteristics, and biochemistry, and are similarly classified except that they are usually grouped into Divisions rather than Phyla. Features that differentiate either animals or plants at the level of Class or Order are often fairly detailed, and may be more a reflection of their evolutionary history than a factor that is relevant to their general biology today. Such groupings are subject to considerable fluctuations and are the subject of academic study and debate. Thus there are no internationally accepted ‘rules’ on classification above Family (or ‘Super Family’) level, and this has therefore been suggested as the most suitable level of generalisation (Pentreath, 2002 b, 2005; Pentreath and Woodhead, 2001) for reference types of animals and plants. Paragraph 24: • This paragraph does not seem relevant, should be deleted. (24) Animals usually have between 12 and 60 pairs (2n) of chromosomes, but there is considerable variation, even within Orders and Families (for example, in the Diptera (flies) 2n varies from 4 to 20; in the Lepidoptera (butterflies and moths) it varies from 14 to 446). The molecular biology of plants is much more variable than that of animals, with more frequent recombination and re-assortment of genes during meiosis. Nuclei, mitochondria, and plastids within plant cells, all have their distinct DNA systems. Polyploidy is common in plants (50% of all flowering plants), usually because a diploid (2n) plant, by irregular division, gives rise to a tetraploid (4n) plant. Then, as a result of pollination, triploid (3n) plants are formed. These are unable to produce gametes compatible with either ‘parent’, and thus the 2n and 4n forms often diverge because of the resultant genetic isolation (Collinson, 1988). Paragraph 25: • This paragraph states that the RAPs were selected on “…the best practical judgment of the ICRP …” It should be further emphasised that these selections were a first approach, based on more practical considerations than on scientific criteria. (25) Because no clear algorithm for the selection of Reference Animals and Plants can be defined, their selection therefore has to be made on best practical judgment of the ICRP, bearing in mind the need to keep the total number low, to try and cover terrestrial, freshwater, and marine environments, and to satisfy the various criteria discussed in this chapter. Section 2.5: • The text in this section does not present the thinking that went into the selection of each RAP, for example, why a bee and not a beetle? Table 1: • The criteria listed in paragraph 19 does not include the criteria listed in Table 1. Such a table, with all relevant criteria, should be made, and/or the criteria in table 1 should be discussed in the text. • Note that there is legislation on flatfish and crabs in US legislation Table 2: • A table showing the distribution of RAPs to different types of climate / ecosystems should also be added to the report. While emissions tend, today, to be mostly from temperate zones, accumulation in temperate, equatorial and arctic zones are considered. • There are some inconsistencies in this table, e.g. ducks, grass and worms may be found in the marine environment. Also, paragraph 41 refers to trout ONLY in the freshwater environment. The table should be corrected and made consistent with the text. Paragraph 29: • “And they are not intended to serve as ‘sentinel’ organisms or species” OK, but how are sentimental organisms used in the “framework”? • “ the Commission does not intend to make such judgments – although that does not preclude others from making them.” Others will do what they need to do – delete the phrase after the hyphen. • “They are, however, considered to be organisms that are ‘typical’ of different environments, in the sense that one might expect to find them there (earthworms in soil, ducks in estuaries, flatfish and crabs in coastal waters, trout in rivers and lakes and so on) and thus that a more detailed understanding of the relationships between dose and effect for such types should prove to be generally helpful as a basis for more detailed comparisons where necessary.” The disclaimers in this phrase seem to leave little room for using RAPs as any kind of indicators at all. Perhaps some further elaboration of this sentence in the context of RAP use could be useful. • “And they are also not intended to represent key links in ecosystem functioning.” OK, but if so, are RAPs then fit for purpose? (29) It is perhaps also worth stressing at this point what Reference Animals and Plants are not intended to be. As indicated previously, they are not necessarily the objects of protection – although they might be in certain situations. And they are not intended to serve as ‘sentinel’ organisms or species, in the sense that it is considered that if such types are protected then other types will also be protected: as will be seen later, there is insufficient information on radiation effects on different types of organisms to enable such an approach to be made, even if it was desired. Nor are these types of organisms the particular ones that the Commission considers should be particularly protected; the Commission does not intend to make such judgments – although that does not preclude others from making them. They are, however, considered to be organisms that are ‘typical’ of different environments, in the sense that one might expect to find them there (earthworms in soil, ducks in estuaries, flatfish and crabs in coastal waters, trout in rivers and lakes and so on) and thus that a more detailed understanding of the relationships between dose and effect for such types should prove to be generally helpful as a basis for more detailed comparisons where necessary. They are also not intended to represent key links in food chains, although some basic ones could be constructed from the set (deer eat grasses; frogs eat worms, fish eat crabs and so on). And they are also not intended to represent key links in ecosystem functioning. Certain components of ecosysytems, such as bacteria, have been deliberately excluded because they are known to have such a high resistance to radiation that they would not serve as useful reference points within the environmental ranges of dose rate that are of interest. Section 2.6: • In this section, the RAPs are sometimes referred to as included because they are a human resource. It should be clear that this is mentioned simply because there is available data. **Moreover, if the intent of the document is to establish protection of RAPs separate and apart from humans, the references to being a human resource should be removed and the consideration as a human resource removed from the criteria in section 2.5** • The biological descriptions of RAPs seems to be inconsistently provided. It is suggested that the “important” biological descriptive material should be identified and provided for all RAPs Paragraph 30: • The last sentence of the paragraph suggests that RAPs can be used as the basis for describing other species, but the text, here or later, does not provide any guidance on this essential ability. (30) It is not possible to provide a comprehensive biological background here to all of the reference types chosen, but some additional general information relating to their biology and ecology is provided in **Appendix A**, together with a more general discussion on populations. This information should be regarded as also being of relevance in the application of any particular Reference Animal or Plant to any specific application. The following is therefore merely a brief introduction and description of each type. It should again be stressed however that, as is the case for Reference Man, they merely represent a hypothetical animal or plant type. Each reference is not a set of data compiled to describe an average or median individual of, nor a description of, a specific species. These are merely ‘typical’ data of particular ‘types’ of animals and plants, but nevertheless precisely defined numerically so that where differences in such values are necessarily made (for example, with regard to size or shape, or in the time span of different stages of a life cycle) there is a known basis for the dose estimation procedure and hence for any variations that others might use to describe any specific animal or plant. Paragraph 41: • The phrase “body of water” in the second sentence needs clarification in this context. (41) The reference salmonid is taken to be a ‘trout’ rather than a ‘salmon’ in order to avoid the complication of migratory effects of the salmon from fresh water to the marine environment. The reference trout is therefore assumed to have the characteristics of a trout that lives its life in the same (‘soft’) body of water, spawning in a stream that runs into that water. The eggs are laid in late autumn, taking 100 days to hatch. It attains maturity at 4 years old and spawns twice (1500 eggs each year) before dying at the age of 6 years. Paragraph 49 • Grammar change: “…but are not…” becomes “…but is not..” in the second sentence. (49) The reference earthworm is assumed to have the characteristics of a member of the Family **Lumbricidae**, which occur naturally in Europe, western Asia, and North America. It is taken to be hermaphrodite but is not self-fertilizing. It produces 25 capsules per year, from which a single hatchling emerges after 50 days. The adult reaches maturity at 1 year and lives for 5 years. Paragraph 52 • Grammar change: “…word-wide…” becomes “…world-wide…” (52) All grasses belong to the same Family, the Poaceae (formerly the Gramineae). Grasses of one form or another are the predominant plants throughout much of the terrestrial environment. They have a world-wide distribution and occur naturally in a wide variety of forms, including reeds and bamboo, as well as the more familiar cereal crops and the dominant plants of natural pasture land. They serve as food for a wide range of herbivorous mammals, including (as herbage) domesticated forms of cattle, sheep, and horses. They are also the basic food crop for humans all over the world. Their biology has therefore been well studied, including their accumulation of a wide range of chemicals. Their life cycle is highly seasonal. Section 2.7: • Although populations may be important biological entities, it is felt that a broader, ecosystem approach is more relevant. This is clearly hard to assess, in view of its complexity, but should at least here be mentioned. What is the role of a population in an ecosystem, and how would this relate to the radiological protection of the ecosystem? • As such, is the concept of population, as presented here, relevant at all to radiological protection? • If populations are an important concept to this report, then there is a need to explain how RAPs can be applied to the protection of populations. • In this context, the literature referenced should distinguish materials addressing individuals and populations. DCLs should also be specified as being useful for protection of individuals or populations • Again here, it is not clear how population protection fits into the overall framework of radiological protection of the environment. Paragraph 60 • Why and how should “These population characteristics…” be “borne in mind”??? (60) These population characteristics need to be borne in mind when considering the potential consequences of any assumed or observed effects of radiation. The geographic area necessary to support populations of these sizes is also of relevance. It is also recommendeded that any future ecosystem modelling of radiation effects should start with such basic assumptions of population parameters, as indicated in Table 3, in order to relate the different categories of the effects of radiation observed in individuals, or groups of individuals, as discussed later in this report, to effects that might be expected at the level of populations, and of communities of different populations. Table 3 • How are these population characteristics to be used in the context of radiological protection of the environment? Why were these characteristics selected? Paragraph 62 • This paragraph demonstrates why it is very important for the Commission to describe the use of RAPs within an overall environmental protection framework. For example, it seems that RAPs could be useful in planning, for any exposure situations, or perhaps in limiting discharges in ongoing, planned situations, but not in terms of after-the-fact response to existing or emergency situations. The Commission’s views should be put forth. (62) With regard to the exposure of humans to a source, the Commission considers it useful to recognise three **types of exposure situations**, which collectively address all conceivable circumstances. These are as follows. (a) **Planned exposure situations**, which involve the planned introduction and operation of sources. This would also include their decommissioning, disposal of associated radioactive waste, and rehabilitation of the previously occupied land in the case of installations. (b) **Emergency exposure situations**, which are unexpected situations that occur during the operation of a planned situation, or from a malicious act, requiring urgent action. (c) **Existing exposure situations**, which are exposure situations that already exist when a decision on control has to be taken, including natural background radiation and residues from past practices that have been operated outside the Commission’s recommendations with regard to human radiation protection, or long-term exposure situations following emergency situations. Thus what the Commission has hitherto called ‘practices’ could be the origin of planned, emergency, and existing exposure situations. Paragraph 64 • The phrase “…and life cycle…” was added to the 10th line of this paragraph because this is felt to be an important factor in determining exposures. (64) Plants and animals may therefore be exposed to ionising radiation in the environment from different sources, and under different types of exposure situations. In all of these, factors affecting the doses received will vary enormously. Radionuclides distributed in the environment will result in the external radiation exposure of the organisms living in, or close to, a contaminated medium, the dose received being the result of the complex and non-linear interactions of various factors including the contamination levels in the environment; the radionuclide-specific decay properties characterised by the radiation type, the energies emitted, and the yield; the geometrical relationship between the source of the radiation and the target; the composition and shielding properties of materials in, and the nature of, the surrounding medium; and the location, size and life cycle of the organism. Internal exposure results from the accumulation of radionuclides into the tissues and organs of organisms, and from radionuclides that may pass through the gut or otherwise temporarily enter cavities within an organism, and will depend on both the physical decay properties and characteristics of the radionuclide and the biological half life of the radionuclide at either a whole body or organ-specific level. Section 3.2 • All the data referred to in this section refers to exposures to individuals, not to populations Paragraph 65 • The ending phrase, “…, although this will vary from country to country.” Was added to account for different national approaches (65) As with human radiation exposure, the basis of estimates of dose received, or that might be received, will vary from one exposure situation to another. Under many planned or existing exposure situations, in which radionuclides are present in the environment, direct measurements can be made of water, sediment, or soil in order to estimate doses from sources external to an organism. It should also usually be possible to obtain direct radionuclide concentration data in order to estimate doses from internal sources. In other situations, however, a different approach is needed, where resort has to be made to modelling approaches, plus their attendant data bases. This is particularly the case with regard to anticipated exposures from the creation of sources in the future, under planned situations, and in relation to potential emergency situations. All of this work would normally form part of an overall ‘Environmental Impact Assessment’, although this will vary from country to country. The modelling approaches might also require different sets of data to relate exposure to dose. Paragraph 66: • The text in this paragraph refers to steady-state and dynamic situations in the second sentence, but then speaks only of steady-state conditions. Emergency situations would clearly be dynamic in nature. (66) Thus the relationships between the concentrations of radionuclides external to, and those contained within, an organism will differ very considerably under the three different types of exposure situations. For convenience, such relationships are usually considered as being either in a ‘steady state equilibrium’, or in some form of dynamic, non-steady state, condition. For the former, the data are often expressed in terms of a concentration ratio (actually a derived quotient) between the organism and the surrounding medium. Thus, for example, for aquatic organisms, the steady state situation may be expressed as a ‘Concentration Factor’ value to represent the ratio of the concentration of the radionuclide in the organism to that of the ambient water (filtered or un-filtered). For terrestrial organisms, this relationship has sometimes been represented by a ‘Transfer Factor’ value, being the quotient of the concentration of the radionuclide in the organism to that of the soil. Many tabulations have been made of such values, particularly for aquatic organisms, but these have usually been formulated for the purpose of estimating pathways leading to human exposure. They therefore often relate to the concentration of a radionuclide within a specific part of the organism – the part that would normally be consumed by humans. Such data are not therefore always of much relevance to the needs of calculating the dose received by the organism itself, particularly with respect to any particular organ or tissue that would be of interest in terms of estimating specific types of radiation effect. Paragraph 69: • The first bullet point refers to external exposure “sources” but does not refer to immersion in a gas cloud. The word “air” has thus been added to account for this. “- external exposure from contaminated soil, sediment, or water or air;” Chapter 4: General Comments • The models listed in this chapter are very simplified and have high uncertainty. • The extent to which we understand these uncertainties, both in the modelling and the assumptions with regard to exposure (homogeneous, infinite, etc.) should be more thoroughly stressed in the chapter, particularly as this would impact on the uncertainty of the overall assessment. Section 4.1: • The consideration of non-uniform of distribution of radionuclides within organisms, and specific accumulation of radionuclides in particular organs could be important to dose assessment, and may require these aspects to be discussed in this section. Paragraph 79: • The medium referred to in Figure 1 should be specified (79) If the organism's dimensions are much smaller than the radiation range in the medium, then the internal dose approximates to zero due to the escape of radiation from the body, and the external dose approximates #. And, in contrast, when the size of the organism is much larger than the radiation range in the medium, the whole body internal dose approaches the value of # , whereas the external one tends to zero ( #). Ranges of the different radiations are compared in **Fig. 1** (Ulanovsky and Pröhl, 2006), which shows ‘continuous slowing-down approximation’ (CSDA) ranges for electrons # (solid line) and α-particles # (dotted line) as well as mean free path for photons (dashed line). All ranges and the mean free path are derived from Hubbel and Seltzer (1995) and Berger (1999) and are given in meters for particles with energies in range from 10 keV to 10 MeV. Paragraph 84 • The assumptions presented in this paragraph seem to suggest that dose assessment results that will presumably be presented later will be fairly case-specific. It would have seemed that this document would be able to relate its “generic consideration levels” somehow directly to measurable quantities (i.e. air, soil, water concentration). This should perhaps be mentioned in this paragraph or this section. (84) With regard to the application of such models to the chosen Reference Animals and Plants, a number of other factors have also had to be taken into consideration. These have included the following: - comparative shapes and sizes; - the need for dosimetry based only on whole body or to include internal organs; - body composition and density; - environmental geometries; - the possible use of the same geometries for more than one case; - allowance for external coverings; - the choice of radionuclides to be included in the calculations; - the appropriate time integral for the dose-rate calculations; and - the potential extrapolation of the models and data to other biota. Paragraph 85 • The reference to DCF’s at the end of the paragraph, Dose Conversion Factors, is not the same as in the annex, where these are referred to as Dose Conversion Coefficients. The terminology in the text should be harmonised. (85) The values required are those that provide radionuclide-specific conversion factors that can relate activity concentration values of a radionuclide, either in an organism or in its surrounding medium (in units of Bq per unit weight) to an absorbed dose-rate (in units of Gy per unit time). This approach had first been used in the context of calculating the potential exposure of organisms in relation to the defining of limits for the disposal of radionuclides into the marine environment (Pentreath and Woodhead, 1988; IAEA, 1988). Apart from the issue of shapes, as discussed below, it is also necessary to consider the relevant useful units. Because some of the Reference Animals and Plants have life spans of less than a year, and life stages that last only a matter of a few days, but for which differences in dose rate of less than a day are of no significance (or cannot, in any case, be taken into account) then the relevant values would appear to **µGy day−1 per Bq kg−1**. These values are referred to here as Dose Conversion Factors (DCFs). Paragraph 91 • It should be made clear that the uncertainty referred to in the last sentence is only a small part of the overall uncertainty in this assessment. (91) The re-scaling method of Ulanovsky and Pröhl (2006) makes it possible to evaluate internal and external exposure DCFs for aquatic organisms, with different ellipsoidal body shapes, over a wide range of body masses (10−6 to 103 kg) for source particle energies from 10keV to 5 MeV, thus covering all of the radionuclides listed in the ICRP Publication 38 (ICRP, 1983). Uncertainties in the approximations depend on the source particle type and energy, and on the organism’s mass. Estimates by Ulanovsky and Pröhl (2006) have shown that the mean absolute coefficient of variation does not exceed 10% for electrons, and 15% for photons. In many practically relevant cases, the uncertainty does not exceed 3% for electrons and is bounded between 5 and 10% for photons. Section 4.5 • This section is very detailed and should all be moved to an appendix Paragraph 94 • The information in Table 5 should be reviewed for their applicability in different countries: e.g. the mass of the deer does not seem to be representative of animals in the United States. For example, if this represents the average of male and female deer weights, than this should be specified Paragraph 97 • The first sentence speaks about internal exposure, but more detail about the internal dosimetry method used would be useful. • The paragraph states that only photon sources were assumed to contribute to external exposure. What about bremsstrahlung? (97) DCFs for internal exposure of the terrestrial animals and plants have been assessed using Equation (1) and a technique suggested in Ulanovsky and Pröhl (2006). External exposure of the terrestrial animals has been evaluated based on results derived by Pröhl et al., (2003) and presented in Taranenko et al. (2004). The external DCFs from FASSET were derived for adult terrestrial animals of different shapes covering range of body masses ranging from 1.7×10−4 g to 550 kg. Only photon sources were assumed to contribute to external exposure of the animals. The external DCFs have been calculated here as a product of free-in-air kerma, #, in a place occupied by the animals’ body and pre-computed dose-to-kerma ratios, # : Paragraph 101 • References to equations 4.1 and 4.2 seem to be incorrect. (101) For the derivation of DCFs for external exposure, a differentiation has been made between aquatic and terrestrial animals and plants. For aquatic organisms, which are immersed in water, there is no substantial difference between the density in water and the organism. The conditions for radiation transport are therefore relatively homogeneous. Under those circumstances, the application of analytical approaches allows estimates with sufficient accuracy. For this case, DCFs are derived in accordance with Equations (4.1) and (4.2). For terrestrial Reference Animals and Plants, however, the estimation of external exposures is more complex. Paragraph 102 • Is DCF in the 6th line referring to DCC?? • The f3 values ARE given in Appendix C – fix the last sentence to reflect this. (102) To enable the use of specific weighting factors to account for the different radiation types for the absorbed dose, the fractions of the absorbed dose that are due to different types of radiation are also given in **Appendix C**. The fractions are presented only for internal exposure because weakly penetrating short-range radiations (α-particles, fission fragments, and low-energy electrons) are likely to be absorbed by the outer protective layers of the body (skin). The percent fractions of the internal DCFs that are due to α-particles and spontaneous fission fragments, f1, as well as to low-energy electrons (E < 10 keV) and β-particles, f2, are given in the Tables provided in Appendix C. The fractions for photons and high-energy electrons (Eβ >10 keV), f3, are not given in the table because they can be easily derived as complementary to the above given: f3 = 100% - f2 –f1. Figure 4: • Check the locations of the testes and liver. They do not appear to be correct Paragraph 124 • The mass used to assess this brown seaweed could be specified (124) The reference brown seaweed is represented by an ellipsoid with dimensions of 50×50×0.5 cm. Exposure dose rates can be considered for various periods of immersion. Table 7 • Flatfish should be assessed against exposure from the sediment (a planar source) as well as from the water. This could also be the case for the tadpole. • The use of “exposure situation” here is NOT the ICRP use of this term. To avoid confusion, the terminology here should be changed • These exposure geometries should be made to match paragraph 69. • The EGIR felt that this table should be better titled: Summary of exposure geometry assumptions EGIR General Comments on Chapter 5 • This chapter is broadly a listing of scientific studies, with little assessment of their relevance to the objective of this report (i.e. effets on populations or individuals, chronic or acute exposures, dose levels received with respect to the dose bands proposed later, and with regard to the endpoint). Instead, this should be an evaluation of this data, supporting the conclusions that follow regarding DCLs. This report is not an UNSCEAR scientific report, but rather an ICRP report assessing existing data and makng protection judgements. • The ERICA programme and the upcoming UNSCEAR report should be referenced, and relevant data should be extracted and referenced in support of the conclusions of this report, Thus, this report should summarise, in a tabular format, the effects for different end points against different RAP dose-rate bands, and this would then lead into Chapter 6. In this context, the ICRP should focus more on lower-dose non-human exposure experiments (in general those used for low-dose human studies), not on high-level external exposure situations, which are less interesting for protection purposes. Extrapolation from high to low dose rate should be more clearly expressed. This summary format would assist in highlighting gaps in knowledge. Note that there are other reference reports than those that are mentioned here that should be included. • Do not give dose rates to too many significant figures (see paragraph 250) • For those dose rates that are listed, there should be an attempt to keep the units consistent. • Several obscure abbreviations are used (e.g. ILB in paragraph 156, LOED in paragraph 164). This type of usage should be limited, or explained in a glossary if they are broadly used. • Only use references from peer-reviewed journals. • There are many inconsistencies and data gaps in the information provided that would need to be resolved before DCLs could be really used. Paragraph 133 • How does DDREF relate to exposure to radiological protection of the environment? Maybe don’t mention this here. • RBE is expressed in paragraph 145 as the topic of a separate document – so it should not be mentioned here. (133) Many studies with mammals have shown that, in general, mutational dose-responses are linear-quadratic for low LET, and tend towards linearity as LET increases. For low LET radiations, reduction in dose-rate always reduces the frequency of induced gene or chromosomal mutations in mammalian somatic and germ cells. The maximum dose-rate reduction factor is usually 3 to 4, but it can be somewhat higher for chromosome aberration induction in human lymphocytes. A reasonably consistent relationship between RBE and LET for mutation induction has also been recorded in mammals, with maximum values for RBE of around 10 to 20 usually being seen in the LET range 70 to 200 keV µm-1. Paragraph 141: • The reference in this paragraph,” (Jung et al. 201).” Should become (Jung et al. 2001). Paragraph 147 • The interpretation referred to in the last sentence is **precisely** what is needed for radiological protection of the environment, so what guidance is provided by the ICRP? (147) But for animals and plants, the objectives, as noted in Section 2, are more variable. The grouping of effects into those that are stochastic, or not, are therefore of less value because it is the biological consequences that are likely to be of interest, and particularly at the population level rather than at the level of the individual. Nevertheless, the effects of radiation take place at the level of the individual, and thus it is useful to consider such effects in terms of how they might effect populations – such as early mortality, reduced reproductive success, some forms of morbidity, or other effects that are ‘scorable’ but the consequences are not known, such as chromosomal damage. These types of effects on individuals are discussed below in relation to the types of Reference Animals and Plants. No attempt is made here to interpret such effects at a ‘population’ level because this would also require a detailed description, evaluation, and interpretation of the dynamics of a ‘model’ population of each type in order for the data to be sensibly interpreted. Paragraph 153 • The Sr-90 soil concentration referred to in this paragraph, 14 Bq/kg in soil, seems to be too low. Check this figure. Section 5.3.3. Morbidity • There is a need to define what is included in morbidity: only cancer, or also fitness for life? Paragraph 204: • The Sr-90 and Cs-137 soil concentration levels post-Chernobyl, quoted in this paragraph as 1 Bq/kg, seem to be far too low. Check this figure. Paragraph 241 • It was felt that the first sentence should be more correctly expressed by replacing “…appears to be a dose threshold of…”with “ …have been no observed malformations at exposures below …” This wording should match what is stated in Publication 103 • “…based on human and animal data…” changed to “…based on animal data…” (241) With regard to effects on the developing foetus, based on animal data, there **have been no observed malformations at exposures below** about 100 mGy for the induction of malformations, and a threshold of about 300 mGy in the most sensitive pre-natal period (8 to 15 weeks post-conception) for any mental retardation effects. It is also generally observed that the effects of acute irradiation on reproductive capacity (fecundity and fertility) are highly dependent on the time of development or age of the animal when they are irradiated. Paragraph 278: • The Sr-90 concentration referred to in the first sentence, 1 Bq/kg, seems to be too low. Check this number. Paragraph 335: • Change “…surprising lack of data.” To “..lack of data.” To whom would this be surprising? • Add the last three sentences to this paragraph to more fully express the uncertainties of experimental data. (335) As stated in the introduction, there are many data to refer to, but little guidance with respect to their reliability, consistency, interpretability, or utility. There are many data on some types of organisms - such as small mammals, some fish, and on pine trees – that have been gathered for various reasons, in various ways, for many years. But for some types of animals, such as birds, there appears to be a lack of data. And again, for some types of both animals and plants, the information covers a wide range of dose rates, and thus both ‘chronic’ and ‘acute’ exposures, but in other cases the data would appear to be largely derived from the latter type of experiment, and thus of limited value in most environmental situations. Reviews of data are also often difficult to use because the original experimental data are usually reported for dose rates averaged over periods of hours, days, or years. Summaries of the data are therefore often arbitrarily organised into such bands of dose rates. They are also often reviewed collectively with respect to their environment, and thus primarily with respect to their pathways of exposure (such as aquatic) rather than with respect to the effects of radiation in relation to their phylogeny, or to some specific feature of their biology. **In addition, it must be noted that data from experimental sources in some cases conflicts or data from only 1 study is available. One must also recognize that the effect of exposure can vary widely depending on the life stage, life cycle, or environmental conditions. The limitations noted should be carefully considered in applying the derived consideration levels.** Paragraph 340: • The first statement of this paragraph is somewhat emblematic of the problem that the EGIR has with this document. The ICRP has not made a convincing case for a particular framework, or for the use of DCLs and RAPs and their place in the framework. (340) Finally, the question arises as to what can usefully be done with such information, allowing fully for its obvious limitations, and what can be done in a positive way? This issue is addressed in the next chapter, but all of the points made in the discussion above need constantly to be borne in mind. Chapter 6 General Comment • DCLs seem to be part of the RP framework, rather than just a part of RAPs. Neither RAPs nor the Framework are clearly or thoroughly explained, and need more work. In this context, and in view of the relative lack of understanding of the overall framework, it seems to be premature to set DCLs. At the very least, it should be noted that the use of DCLs is not clear. Paragraph 342 • An alternative comparator to that mentioned in the first sentence would be to use the “no-observed-effect” threshold. Background levels are a useful comparator, but not the only one. • The reference to “normal background dose rates” should be made more clear as to whether this means “local-normal”, global or regional average, etc. • The use of DCLs in different exposure situations is referred to in line 15 of this paragraph, but no guidance is given in the document as to how this should be done. (342) It has therefore been suggested that the only other useful comparator might be that of the natural background radiation dose rate normally experienced by such animals and plants (Pentreath, 1999, 2002a). Additional doses that were but fractions, or small multiples, of the normal background dose rates might therefore be unlikely to be the cause of any environmental managerial concern, particularly when considered against those multiples of background dose rates that were known to have specific effects; whereas dose rates that were very much higher, and in the region of known or expected effects, would need to be considered further, alongside other environmental information, within any particular environmental management framework. Thus, collectively, all of the derived information relevant to each type of animal and plant could then be simplified into bands of dose rates relevant to their individual background radiation dose rates in order to set out ‘Derived Consideration Levels (DCLs)’. The purpose of the DCLs would be to serve as points of reference at which one should consider what is known about the effects of radiation on particular types of animals or plants alongside other relevant information: such as the type of exposure situation (planned, emergency, or existing); the size of the area affected; the status of the actual populations or ecosystems concerned; the fraction of a population exposed; the particular animals and plants of interest; and the driving environmental management needs required in order to satisfy the legal framework within which any management action was being taken, and so on. The results of such considerations might then well be that actual managerial action, or other decisions would be taken at dose rates in a band higher or lower than the DCL band, but the reasons for so doing would be clearly stated. Section 6.2: Natural Background • The background numbers given in this section seem to be a bit low, particularly with respect to data from FASSET and ERICA. These should be checked. Section 6.3: Identifying Preliminary Derived Consideration Levels • In this section it should be made more clear in this section WHY the values (or ranges of values) have been selected Paragraph 345 • The ICRP states in the second sentence states that the ICRP’s DCLs are based on informed opinion, not on rigorous analysis. But this type of assessment of data is EXACTLY what the ICRP should do in recommending DCLs. (345) So, using these data, together with the incomplete and varied quality and relevance of the known radiation effects data discussed in Chapter 5, the following Tables **(Table 8 to 11)** provide an attempt to summarise all of the data in order to suggest a preliminary set of Derived Consideration Levels. In doing so, it cannot again be more strongly emphasised that the comments in the ‘boxes’ are an extreme over-simplification of existing data, and that the shading of the boxes to indicate the bands of Derived Consideration Levels are based on informed opinion and not on any statistically derived, or rigorously reviewed and defensible, analysis of all the available data. Revisions will no doubt be made on such a basis; in the meantime, this current exercise will hopefully help to stimulate such actions in the future. Nevertheless, one has to start somewhere, and the following tables and associated commentary need to be viewed in this light. Table 8 • Against what are the DCLs in Table 8 compared when making protection judgements? Mean dose, average dose, etc. • Some of the data in the table conflict with the preceding text. For example, in the 10-100 mrad.day range for rats, the text indicates that there were some effects for one study and none for another and for trout, the table indicates mortality at over 100 mrad/day but portions of the text indicate there were no effects at 2030 mrad/day. These differences should be recognized and an explanation given regarding what information was included in the table and how DCls were assigned. This is not reflected in the table. • Chronic and acute information is confounded here. It should be made clear what is what. • Note that simple shading of this table (in the 0.1 – 1 range) is not a very clear “definition” of DCLs (and was not visible in the “black and white” printed version). DCLs and their meaning should be made more clear. For example, DCLs seem to be a range more than a value, this should be expressed in the text. DCLs appear to be assigned in some cases at the “no information level and at the “very low probability level in other cases.A justification for these decisions should be given. • The value of 0.1 mGy/day seems to be a little low for natural background level. How was this number selected? Section 6.4: Matters for Consideration • This section, and the first part of Chapter 7 seem to be more raising questions and issues than providing answers or direction. These should be put together, and clarified Paragraph 353: • It is stated in the 5th line that doses below the DCL are not necessarily safe, but also states that there is no need to do further assessment below DCLs. These two statements should be reconciled. • The 6th line refers to management actions and decisions, but what kind of action? What kind of decision? This is insufficient guidance to make regulatory decisions. If the dose rates are neither safe nor unsafe, how should this advice be used? (353) The Derived Consideration Levels are NOT intended to be regarded as dose limits, or ‘substitute’ values for them. They are zones of dose rates at which, with respect to the Reference Animals or Plants, or types similar to them, a more considered level of evaluation of the situation would be warranted. It does not imply that higher dose rates would be environmentally damaging, nor that lower dose rates were in some way ‘safe’ or non-damaging. But they are dose rates that could be used in any management action or decision-making process, in terms of being starting points from which further, auditable, information could be appended in order to justify or optimise any subsequent action that was taken. Paragraph 354 • This paragraph refers to other factors to consider, including, “The presence, or expected presence, of additional sources of chemicals, or other forms of environmental stress, in the same area.” But no guidance or idea of how to take these into account along with RAP assessments is provided or even hinted at. Some guidance would be very useful. Paragraph 358: • Note that in paragraph 56 a population was described as a group, not a “small group of individuals” as here. Be careful with terminology. The description of a population is rather vague. • In the 6th line it states: “Thus, it is not possible to say with any confidence that measures to protect individual organisms would also, necessarily, protect the population “ This statement reinforces the comment that setting of DCLs is perhaps premature. • The 8th line suggests that: “Population modelling approaches demonstrate that the linkage between radiation effects in the individuals and in the population is very complex, and dependent on factors other than the radiation doses and the dose-response relationships.” It is not clear how to take both radiation effects and other environmental impacts into account in making making regulatory decisions. ICRP should try to discuss this. • Paragraph 358 suggests that the relationship between individual effects and population effects is, at best, weak. This reinforces the suggestion above that the setting of DCLs at this point is premature. (358) Care should also be taken in using such values to make decisions with regard to populations of animals or plants, as opposed to small groups of individuals, for all of the reasons discussed in Chapter 2. Because although it is reasonable to suppose that any impacts at the population level will be a consequence of responses to irradiation that occur in the constituent individuals, there has, as yet, been little analysis of the links between these two end points (Woodhead, 2005). Thus, it is not possible to say with any confidence that measures to protect individual organisms would also, necessarily, protect the population. Population modelling approaches demonstrate that the linkage between radiation effects in the individuals and in the population is very complex, and dependent on factors other than the radiation doses and the dose-response relationships. Future efforts to develop measures to protect the animate environment from the incremental radiation exposures arising from human activities will therefore need to consider both the individual and the population to ensure that the intended objective is achieved. Chapter 7 General Comments • Paragraphs 361 to 367 usefully presents a number of issues, but there is no framework presented for addressing these issues. • In the spirit of extrapolation, this chapter does not provide sufficient information for extrapolating from RAPs to “real” organisms, and the uncertainties associated with this. As a practical example, there is insufficient information or guidance provided to extrapolate between, for example, rat populations and mouse populations who might be found in the same radiologically affected ecosystem. Paragraph 363: • The assumption that radionuclides released from regulated activities or as a result of accidents are difficult or impossible to recover is common to human protection. I see no reason why this should not be applied to environmental protection. This suggests that this framework and these tools can be best applied at the planning stage, rather than after-the-fact as implied here. (363) For the purpose of pollution control, the above protection objectives may, in turn, require the explicit demonstration of: the avoidance or minimisation generally of harm to the environment; or the ability to deal with the environment that is already deemed to have been harmed. Paragraph 364: • The last phrase in this paragraph seems out of place, and it is suggested that it be deleted. (364) And, for the purpose of nature conservation, the above protection objectives may, in turn, require assessments to be made of: the likelihood of harm to individuals of particular species; potential or actual effects on populations of one or more species, in terms of population integrity and viability (this would also apply to environmental exploitation); potential or actual effects on the principal (or majority) components of a specific habitat, or at a specific place; or potential or actual effects at ecosystem level, within a local area or more generally, but without specific reference or preference to any particular faunal or floral type. Delete the following text: There may even be other considerations, as where the mere presence of radionuclides, “contaminating” an area, may be of concern to certain individuals or sectors of the public for ethical, moral, or social reasons (IAEA, 2002 Paragraph 366: • The 11th line states: “Because radiation effects at the population level – or higher – are mediated via effects on individuals of that population, it therefore seems appropriate to focus on radiation effects on the individual for the purpose of developing a framework of radiological assessment that can be generally applied to environmental issues.” This is a KEY concept but was previously (paragraph 358) presented as not supported by data or models. This should be reconciled. (366) The consequences may need to be evaluated with respect to individual animals and plants, depending on the legal framework within which action is being considered, but undoubtedly the major requirement will be the need to make evaluations at the population or ecosystem level. Radiation effects on higher levels of biological organisation (e.g., populations and ecosystems) occur only if individual organisms are affected, and radiation effects’ data have generally been obtained for individuals rather than for higher levels of organisation. In the natural environment the situation can become very complex because of the interactions between each individual and its surrounding ecosystem. The effects can also be modified by the presence of other environmental stressors or by combined effects related to the presence of other pollutants, and by interactions between different trophic levels. Because radiation effects at the population level – or higher – are mediated via effects on individuals of that population, it therefore seems appropriate to focus on radiation effects on the individual for the purpose of developing a framework of radiological assessment that can be generally applied to environmental issues. This approach is consistent with many of the existing assessment methods for non-radiological environmental contaminants. It is also essential in order to consider how effects such as reduced reproductive success can be interpreted in the context of the normal biology of different types of plants and animals. Even the concept of what constitutes a ‘population’ will differ amongst the various ‘types’ of Reference Animals and Plants. Section 7.3: Differences in Radiation Dosimetry • This material is valuable, but much of this date is already published, and much will be published in the upcoming UNSCEAR report. As such, its use here should be reconsidered, and if it is kept it should be put into an appendix. Table 14: The reference in the Table title should be Eckerman and Ryman 1993. Paragraph 382: • This paragraph highlights that most of the data available concerns high doses and dose rates, and supports that more research should be done on low dose and dose-rate effects before setting DCLs. (382) In contrast to dosimetry, it is not currently possible to provide recommendations as to how to perform extrapolations that have general applicability in relation to radiation effects, and thus each case has to be carefully considered on its own merits. Due to the relative paucity of information, the main cases for extrapolations, and challenges for methodological development, include the following. There are clearly issues with regard to extrapolating from high acute doses and dose rates of low LET - and X-rays to lower doses accumulated at lower dose rates. In the radiobiological and radioecological literature, the qualifiers “low-level”, “chronic”, “higher”, “acute” and so on are often used without any definition. But a radiation exposure lasting several days may be effectively “chronic” for a short-lived organism, and yet effectively “acute” for a long-lived organism. Unfortunately, there are very few data that relate directly to the chronic, low-level irradiation conditions of relevance for animals and plants in the wild i.e. exposures at dose rates of 100 - 1000 Gy day-1 over the life span of the organisms, and the response endpoints most commonly assessed after acute, high dose, irradiation are not those that are relevant in such situations. Paragraph 384: • The 7th line states that “Nevertheless, it is necessary to start somewhere,…” However, it may be premature to speak of protection criteria before considering population effects. These are major issues that should be resolved before going into details of protection approaches. (384) And finally there is the issue of extrapolation from effects in the individual organism to possible impacts at the population and community levels. This will also, in many cases, involve the extrapolation from laboratory conditions (where most experimental information originates) to field conditions (where populations interact with the physical environment as well as with other organisms). Interactions at community and ecosystem level can be particularly complex (Brechignac, 2003; Doi, 2004; Hinton and Brechignac, 2004). Nevertheless, it is necessary to start somewhere, and thus developing an understanding of the effects of radiation on a limited number of animals and plants, at the individual level, and exploring the consequences of such effects at **their** population levels, and amongst different populations, will clearly build into a broader understanding against which these wider issues can be assessed. Chapter 8, Paragraph 385: • The last sentence states: “There is no simple or single universal definition of ‘environmental protection’, and the concept differs from country to country, and from one circumstance to another.” But the ICRP should state what IT intends to protect with its framework and recommendations, and should be clear on this point. This should take into account the need for national flexibility. (385) As stated in the Introduction, the Commission has now broadened its scope in order to address the subject of environmental protection. In doing so, it acknowledges that, compared with human radiological protection, the more detailed objectives and management of environmental protection are often more complex and difficult to articulate. There is no simple or single universal definition of ‘environmental protection’, and the concept differs from country to country, and from one circumstance to another. Paragraph 387: • The underlined phrase in line 6 is added to assure national flexibility. (387) Such advice and guidance obviously has to be transparent, and have a common basis arising from our knowledge of exposure to radiation and its effects, set within some form of overall framework. It intended that this framework should therefore serve as a basis from which national and other bodies could develop, as necessary, more applied and specific numerical approaches to the assessment and management of risks to non-human species under different circumstances, and different exposure situations. **Of course, policy decisions on the need for standards would be left to national bodies.** Because of the vast complexity of the living environment, and the limited radiobiological and radioecological data bases relating to it, the Commission considered that, by setting out data for a limited number of Reference Animals and Plants, this would provide a vital component of the framework to gather and interpret data in order to provide more comprehensive advice in the future, whilst acknowledging that this information is still varied and fragmentary. Paragraph 388: • This paragraph does not mention that the ICRP will develop a further document explaining its Framework and addressing the data uncertainties and inconsistencies. . This is essential and should be the FIRST document that the ICRP developes. (388) This report, on the concept and use of Reference Animals and Plants, has therefore merely served as an introduction to the complex subject of environmental protection with regard to radiation. It introduces the rationale for selecting the Reference Animal and Plant types, and then gives emphasis to their biology, and to basic aspects of dosimetry and irradiation effects. Further publications will address in more detail such aspects as data bases for modelling exposure, possible refinements to dosimetry, and address such issues as RBE, and the application of the basic approach to different exposure situations.