Radioactive pollutants: Impact on the environnement 9782759801374

Citation preview

RADIOACTIVE POLLUTANTS: Impact on the environment (Based on invited papers at the ECORAD 2001 International Conference)

EDITORS FRANCOIS BRECHIGNAC AND BRENDA]. HOWARD

SCIENCES 7, avenue du Hoggar Parc d'Activites de Courtaboauf 91944 Les Ulis cedex A, France

Book series coordinated by Henri Metivier Books already published Le Tritium - de I'environnement a I'Homme Y. Belot, M. Roy, H., Coordinateurs. Le Radon - de I'environnement a I'Homme H. Metivier, M.-C. Robe, Coordinateurs. Les installations nucleaires et I'environnement - Methode devaluation de I'imp act radio ecologique et dosimetrique L. Foulquier, F. Bretheau, Coordinateurs. Les retombees en France de I'accident de Tchernobyl - Consequences radioecologiques et dosimetriques Auteurs : Ph. Renaud, K. Beaugelin, H. Maubert, Ph. Ledenvic Calliope - Un outil pedagogique en dosimetrie interne (cederom) Auteurs : B. Le Guen, Ph. Berard, P.N. Lirsac, M.L. Perrin, M.-M. Be, J.L. Malarbet, B. Gibert, M. Roy, H. Metivier Le Cesium - de I'environnement a I'Homme D. Robeau, F. Daburon, H. Metivier, Coordinateurs. Publication 84 de la CIPR, grossesse et irradiation medicale

To be published L'Uranium - de I'environnement a I'Homme H. Metivier, Coordinateur. Catastrophes et accidents nucleaires dans I'ex-Union sovietique D. Robeau, Coordinateur. Complete list, see at the end of the book.

ISBN: 2-86883-544-9 Tous droits de traduction, d'adaptation et de reproduction par tous precedes, reserves pour tous pays. La loi du 11 mars 1957 n'autorisant, aux termes des alineas 2 et 3 de 1'article 41, d'une part, que les « copies ou reproductions strictement reservees a 1'usage prive du copiste et non destinees a une utilisation collective », et d'autre part, que les analyses et les courtes citations dans un but d'exemple et d'illustration, «toute representation integrate, ou partielle, faite sans le consentement de 1'auteur ou de ses ayants droit ou ayants cause est illicite» (alinea ler de 1'article 40). Cette representation ou reproduction, par quelque procede que ce soit, constituerait done une contrefagon sanctionnee par les articles 425 et suivants du code penal. © EDP Sciences 2001

Preface Since the early forties, radioecology has, often in an emergency situation, been faced with the need to evaluate the impact that the military or civil use of nuclear energy has had on the environment. Radioecology developed in parallel with other ecological disciplines especially ecotoxicology, in part through the use of tracers. Radioecologists have aimed to understand processes controlling the environmental transfer of radionuclides and to integrate them into predictive models as well as engineering and restoration techniques. Experience of providing radioecological methods to mitigate the effects of accidents has emphasized the importance of the concept of sustainable development. It has also contributed to the recognition of a second key concept, the precautionary principle, and its practical application in the environment. On the threshold of the 21st century, radioecologists have needed to take stock of the situation and to widen their perspectives. In response to this need, IPSN decided to gather a worldwide assembly by organizing the ECORAD 2001 Conference. This book collates a series of invited contributions at this conference which reflect on on-going discussions and provide reviews of the most up-to-date scientific and technical information regarding continental and estuarine environments. Within this context, and further to defining the current state of the art, the papers also identify possible research themes for the future along with scientific and ethical issues which are becoming increasingly important in response to public concern with respect to environmental radioprotection. Continuing previous similar publication achievements, which particularly focused on the marine environment, IPSN has decided quite naturally to edit this document within its Book series dedicated to radiological protection and nuclear safety. This complements a former publication dedicated to Radionuclides in the Oceans, (P. Guegueniat, P. Germain and H. Metivier, Eds., EDP Sciences, Les Ulis, 1996). The focus here is now on continental and estuarine environments which are addressed through four major chapters. Part 1 addresses the general environmental issues, encompassing radioactivity measuring methods, toxicants impact on the

IV

Foreword

environment either in chronicle or accidental situations, and environmental radioprotection. Parts 2 and 3 refer to the state of the art in terrestrial and freshwater aquatic environments, respectively, and part 4 concludes by addressing the important societal and ethical issues. It is of importance to recall that the production of this book, achieved within a very tight schedule, has been made possible by the very deep and scrupulous involvement of a number of international experts and professionals, within and outside IPSN, who enthusiastically dedicated their time to the number of tasks requested. They all deserve our gratitude, with especial acknowledgements to authors, and also to reviewers: Jean Aupiais, Rodolfo Avila, Jean-Claude Barescut, Nick Beresford, Dominique Boust, Philippe Calmon, Jacqueline Gamier-Laplace, Tom Hinton, Christian Hurtgen, Valery Kashparov, Rene Kirchmann, Henri Metivier, Valerie Moulin, Jean-Marc Peres, Gennady Polikarpov, Claire Sahut, Pascal Santucci, Jim Smith, Herve Thebault, Pierre Toulhoat, Christian Vandecasteele, Gabriele Voigt, Dennis Woodhead. The readers should bear in mind that this book only forms the starter of the ECORAD 2001 conference meal. The remaining scientific matter which has been selected as high quality and relevant by the Scientific Committee will be further published as Proceedings of the conference in the Radioprotection Colloquium series as we did in 1997 with part 1 of the RADOC conference (Radionudides in the Oceans -RADOC 96-97, Proceedings Part 1, Inventories, Behaviour and Processes, Octeville, 7-11 October 1996, Radioprotection-colloques, 32, C2, April 1997). Finally, it is with a great pleasure that we want to warmly thank all involved scientists and experts themselves, from students to leaders. They are those who dedicate their time, and often their life, to continuously improving the understanding of our common world to the benefit of humankind. August 2001, The Editors F. Brechignac and B.J. Howard

Summary Foreword: Thoughts on Radioecology by the millennium shift

IX

Introduction 1. The ecological dimensions of ecotoxicological research Introduction: Environmental toxicology or ecotoxicology? 1. Implementation of ecological theories in ecotoxicology 2. Interaction between environmental contaminants and the characteristics of ecosystem: The case study of eutrophication and persistent organic pollutants Conclusion - Perspectives

3 3 4

14 18

Parti General environmental issues 2. Environmental radioactivity measuring methods 1. 2. 3. 4. 5. 6.

History and goals of environmental radioactivity measurements Radionuclide coverage, sensitivity requirements and materials Methods: From counting decays to counting atoms Quality: Not just a matter of taste Environmental radioactivity and the public Future trends and new frontiers

3. Toxicants in the environment: bringing radioecology and ecotoxicology together 1. 2. 3. 4. 5. 6.

Introduction Bottom-up versus top-down From foxes and frogs to fish Disappearance of the fence Additivity of effects, and the RBE Model mobile element and not total element?

27 27 31 39 47 51 54

63 63 64 65 67 69 70

VI

Radioactive pollutants 7. Chemical toxicity of radionuclides 8. Conclusions

72 72

4. Post accident management 1. Introduction 2. Policy for and guidance on intervention 3. Rapid characterisation of contaminated areas 4. Technical options for, and the efficacy and costs of, post-accident management 5. Maintaining competence and issues of sustainability 6. Summary and conclusions

75 75 77 93 94 95 96

5. International advice and experience relevant to chronic

radiation exposure situations in the environment 1. Introduction 2. Major sources of chronic environmental exposures 3. International protection radiation protection guidance relevant to chronic exposures from areas affected by radioactive contamination. . . 4. Decision aiding and decision making 5. International assessments of areas affected by radioactive residues . . . . 6. Overall conclusions

105 105 106 107 115 117 127

6. Delivering a framework for the protection of the environment

from ionising radiation 1. Introduction 2. A strategy for the future 3. The development of the framework - the FASSET and EPIC approaches 4. Dose effect relations for reference organisms - brief summary 5. Conclusions

131 131 133 136 142 143

Part 2 Radionuclides in terrestrial environments 7. Soil as the main compartment for radioactive substances

in terrestrial ecosystems 1 Introduction Sorption and fixation of radionuclides in soil 3. Quantitative characteristics of radionuclides' mobility and potential bioavailability 4 Vertical migration of radionuclides 5. Runoff of radionuclides 9

149 149 150 157 164 167

VII

Summary 8. Impact of micro-organisms on the fate of radionuclides in rhizospheric soils 1. Introduction 2. Pathways of radionuclides in soils 3. Immobilization of radionuclides by micro-organisms 4. Solubilization of radionuclides by micro-organisms 5. Specific role of mycorrhizal fungi 6. Concluding remarks

175 175 176 178 180 181 182

9. 1. 2. 3. 4. 5. 6.

187 187 188 194 196 198 200

Advances in animal radioecology Introduction Transfer coefficients Absorption and bioavailability Generic approaches to quantifying transfer Environmental transfer to animals Conclusions

10. A nutrient-based mechanistic model for predicting the root uptake of radionuclides 1. Introduction 2. Current knowledge 3. Model development 4. Model testing and validation 5. Conclusions

209 210 211 214 223 232

Part3 Radionuclides in aquatic environments

11. 1. 2. 3. 4. 5. 6. 7. 8. 9.

Advanced speciation techniques for radionuclides associated with colloids and particles Introduction Source related speciation of radionuclides Speciation of radionuclides Speciation techniques LMM speciation techniques Solid state speciation analysis of colloids Solid state speciation analysis of radioactive particles Mobility and interactions Conclusions

12. Modelling radionuclides in aquatic systems: evolution, revolution and the future 1. Introduction and background 2. General criteria for predictive models

243 243 244 245 246 247 249 251 254 256

261 261 263

VI11

Radioactive pollutants

3. Basic principles determining the predictive success of ecosystem models 4. Evolution and revolution in predictive modelling : 5. The highest reference r2 and Kd 6. Model structuring 7. General approaches of simplifications 8. The panel "of driving variables 9. Critical model tests 10. Future developments 11. Conclusions 13. 1. 2. 3.

Estuarine contaminants: fate and environmental risks

Introduction A macrotidal estuary Is an estuary a chemical reactor? 4. Chemical hazards for the estuarine ecosystem 5. Conclusion

266 268 276 278 282 286 290 294 297 303 303 304 312 320 324

Part 4 Ethical and environmental issues 14. The Nord-Cotentin Radioecological Group: An original

experience of pluralistic expertise 1 The context .... .... .... 9 Creation and operation of the group . . . . ... 3 The GRNC's work 4 The role of non-institutional members 5. Conclusion: The GRNC, an innovative pluralistic expertise process. . . .

331 332 333 334 336 339

Foreword: Thoughts on Radioecology by the millennium shift A. Aarkrog 1 Radioecology has throughout the last half of the twentieth century developed into a science, not only dealing with nuclear contamination, but also contributing to our understanding of general environmental pollution problems. The major milestones during the past fifty years were the studies of global fallout, the waterborne discharges from nuclear reprocessing and the Chernobyl accident. These events have given us a good understanding of the environmental behaviour of major radio-contaminants such as 90Sr, 131 137 I, Cs and Pu. The present trends for radioecology involves a further development of models, tracer studies, countermeasures, and inclusion of other species than man in radiological protection. A major task will be a continued effort to inform and educate both the general public, including politicians and news media, but also scientists from developing countries. It would also be desirable to see radioecology closer integrated into environmental studies of other pollutants for instance by developing equidosimetric methods and studying possible synergistic and antagonistic effects. Finally an effort should be made to develop radioecology into a more hypothesis-oriented science. In a millennium perspective, we may envisage an impact from the rapidly evolving bio- and computer technologies and we may even see radioecology as an extraterrestrial science.

Introduction The word Radioecology may be translated as housekeeping with radioactive substances. I define radioecology as the scientific discipline, which studies the environmental behaviour of radionuclides comprising their interaction with the bio- geo- atmo- and hydrospheres. The primary

1

Riso National Laboratory, 4000 Roskilde, Denmark.

X

Foreword

purpose of radioecological studies have been - and are still to a large extent - to provide dose assessments for environmentally dispersed radionuclides. The science of radioecology was developed almost simultaneously in the Former Soviet Union (FSU) and in the United States (USA). This occurred by the entrance to the "Nuclear Age" in the late forties and early fifties. The Russian geneticist Timofeev-Ressovsky (1957) organised a radiation biology laboratory near Sverdlovsk (now Ekaterinburg) in the southern Urals. He used the term "radiation biogeocenologie" for his studies of the behaviour of the radionuclides released to the environment in connection with the development of the first Soviet nuclear weapon at "Chelyabinsk60" (later MA YAK) in the late forties. In the USA, Libby initiated the so-called project SUNSHINE in 1953 (Rand Corporation 1953). Its purpose was to critically re-examine the potential hazards from radioactive fallout - in particular 90Sr - that might result from a large scale nuclear test programme including thermonuclear weapons or nuclear war. In the early fifties the US Atomic Energy Commission also initiated other radioecological activities which among others involved well-known radioecologists such as Wolfe, Odum and Aurbach (Nelson and Evans, 1967). In the following I begin with summarising the major milestones in radioecology over the last five decades. Then I consider the state of the art and finally I shall try to see into the future.

Past milestones Global fallout According to the latest UNSCEAR report (2000) 543 atmospheric nuclear test has been carried out. The first test explosion took place in New Mexico (USA) in 1945 and the last in LopNor (China) in 1980. The total fission yield of these tests corresponds to 190 Mt TNT or 741 PBq 90Sr. Of this 115 PBq 90 Sr were deposited locally at the test sites (notably Bikini and Enewetak in the Pacific) and 16 PBq decayed in the stratosphere prior to deposition. Hence the total amount of globally deposited 90Sr from nuclear weapons testing became 610 PBq 90Sr. The corresponding global deposition of 137Cs was 930 PBq. The early nuclear weapons testing with fission weapons (20-100 kilotons TNT range) mainly produced tropospheric fallout, i.e. the debris from the explosions remained below the tropopause and was not dispersed globally, but was deposited around the latitude band of the test site. However, in 1952, the USA and, the year after, the FSU tested their first thermonuclear

A. Aarkrog

XI

devices (megatons range) and fallout from these explosions occurred worldwide. The distribution of the global fallout shows a maximum around 40 °N and minima at the poles and the equator. The deposition of radionuclides in the Southern Hemisphere is about one third of that in the northern. Furthermore, a seasonal variation is evident. In the northern temperate latitudes, the fallout rate in May-June is 3-4 times that observed in November-December. The average dose to a member of the world population from nuclear weapon testing is calculated to 3.49 mSv. Most of this dose is due to 14C and

will be delivered in the future. For the period 1945-1999 the dose is 0.994 mSv and of this about half is from ingestion, with 137Cs, 14C, 90Sr, 131I and 3 H being the main contributors (UNSCEAR, 2000). The calculation of doses has been based on systematic observation of the relations between deposition of a radionuclide (kBq • m -2 ) and its timeintegrated concentrations in diet (Bq • y • kg-1). UNSCEAR (2000) has e.g. found that the transfer coefficient for 137Cs for a world average diet is 8.4 Bq • y • kg-1 per kBq • m -2 . Taking amount of diet and dose-factor into consideration the dose-deposition coefficient becomes: 55 uSv per kBq 137Cs m~2. It is, however, evident that such a coefficient will show variations due to individual food habits and environmental conditions. For the same diet type e.g. cow-milk the transfer coefficient for 137Cs thus vary by an order of magnitude according to environment. In other words some environments may be ten times more sensitive to contamination (of cow-milk) than others; we say the radioecological sensitivity differs by a factor of ten (Aarkrog, 1979). Some population groups received relatively high doses from global fallout. This was the case for some arctic and subarctic populations that herd and breed reindeer and caribou (Liden, 1961; Miettinen et al., 1963; Hanson et al., 1964; Ramsaev et al., 1965). Maximum concentrations of 137Cs in these population groups were more than 50 times higher than the human body levels found in general in the Northern Hemisphere in 1964-1965 (UNSCEAR, 1966). The reason for the high 137Cs levels in reindeer and caribou was the consumption of lichen by these animals.

Nuclear reprocessing From a radiological point of view nuclear reprocessing has so far been the major source to the ingestion dose from the nuclear fuel cycle. Most of this dose has come from authorised discharges to the sea of 137Cs from Sellafield in the UK in the seventies and early eighties (UNSCEAR, 2000). The collective effective dose from Sellafield derived 137Cs (~ 40 PBq) has been calculated to be approximately 4000 man-Sv, corresponding to an individual average dose to the world population (6 x 109) of 0.7 mSv or a

XII

Foreword

transfer coefficient of 100 man-Sv per PBq 137Cs. Local population groups living near the Irish Sea with a high consumption of fish may back in the seventies, have received annual doses in the order of the natural background (Hunt and Jefferies, 1981).

Chernobyl accident Several radioecological lessons were learned after the Chernobyl accident in 1986. We saw the importance of natural and seminatural ecosystems when it comes to intake of radiocaesium with human diet. It became evident that the contamination of mushrooms by radiocaesium was one of the important pathways in such ecosystems. For example, a strong seasonal variation of 137 Cs in roedeer was demonstrated in Sweden (Johanson et al, 1990). This variation was mainly due to consumption of mushrooms in the autumn. In Scandinavia, the lichen-reindeer-human foodchain was another major pathway. This was in agreement with expectations from the global fallout studies mentioned above. Another observation told us that the relative composition of the fallout from a reactor accident may change with the distance from the reactor. It was thus observed that the ratios 90Sr/137Cs and Pu/137Cs decreased significantly with the distance from Chernobyl. The accident furthermore taught us the importance of seasonality. Thus crops in southern Europe showed higher radiocaesium concentrations than crops from northern Europe for the same deposition density of 137Cs. The reason was precocity of southern crops compared to northern. From the Chernobyl accident about 85 PBq 137Cs, 54 PBq 134Cs, 1760 PBq 131I, 10 PBq 90Sr and 0.07 PBq 239/240pu were reieased together with many shorter-lived radionuclides of less radioecological significance. One of the more serious late effects of the Chernobyl accident has been the approximately 1800 thyroid cancer cases reported in the FSU in children and adolescents for the period 1990-1998. A major radioecological task in the future is a reconstruction of the 131I deposition to make an assessment of the individual doses to these population groups. This may be done by measurement of 129I in the affected areas, but 137Cs deposition data may be a usable alternative. Apart from the thyroid doses from 131I the total collective effective dose to the population in the most contaminated areas in the FSU (> 37 kBq 137Cs m-2) has been calculated to be 60 700 man-Sv (UNSCEAR, 2000). This dose is mainly due to 137Cs (36 125 man-Sv from external exposure and 13 207 man-Sv from ingestion of 137Cs). The total deposition of 137Cs over this area in the FSU was 29 PBq. This give us a transfer factor of 1700 man-Sv per PBq 137Cs.

A. Aarkrog

XIII

Summary Three events have been the main contributors to the exposure of the global population from man made radioactivity. • Global fallout from nuclear weapon testing; • Liquid discharges from nuclear reprocessing; • The Chernobyl accident. These events have furthermore been main objects for radioecological studies over the last fifty years. A way to summarise - in a very condensed and subjective way - the outcome of these fifty years of radioecological studies could be to show a table with the transfer coefficients for 137Cs derived from these studies. Whereas the doses from nuclear reprocessing comprise only the marine environment the two other sources deliver their doses through both the terrestrial and marine environments. Had we for global fallout and the Chernobyl accident considered only the terrestrial environment the ingestion dose transfer factors for these two sources would have increased to 3 and 0.5 respectively, because the transfer coefficients between diet concentrations and deposition are less in the marine than in the terrestrial environment and thus "dilute" the transfer coefficients in Table I.

Table I. Collective dose transfer coefficients. >

Source Global fallout Nuclear reprocessing Chernobyl accident

man* Sv perTBq137Cs Total dose Ingestion dose ,

3 0.1 1

1 0.1 0.4

Present situation Introductory remarks Two years ago Murdoch Baxter, the editor of Journal of Environmental Radioactivity asked the members of the Editorial Board to contribute to a series of millennial editorials which should reflect thoughts on problems related to environmental radioactivity by the entrance to the new millennium / century.

XIV

Foreword

These editorials may give a fair general view of the actual state of the art of radioecology. The members of the editorial board are in this context considered representative of the radioecological science as it has developed until now. In the following I shall try to summarise some major trends from these editorials.

Outlook It is a general opinion among many of the authors (Scott, 2000; Voigt, 2000; Ohmomo, 2000; Whicker, 2000; Woodhead, 2001; Holm, 2001) that the present situation for radioecology looks less promising. The reasons for this are several. Radioecology is strongly dependent on the development of the nuclear option for energy production. In most countries, the public opinion has for several years been against the use of nuclear energy and the Chernobyl accident became the deathblow. One may argue that if people are so afraid of radioactivity they should be in favour of a discipline such as radioecology which improve our understanding of the risks and also develops methods to mitigate environmental contamination. But this is apparently not the case. At most radioecology is considered as a necessary evil in particular in the wake of nuclear accidents. Now fifteen years after Chernobyl it seems difficult to obtain the necessary funding for a continued fruitful development of radioecology. (Scott, 2000; Whicker, 2000). The remembrance is short - in particular among politicians. We should, however, not forget to sweep before our own doorstep. One of the lessons learned after Chernobyl was also the very short memory of radioecologists or rather the lack of interest in using old experience by reading the literature. So a new generation may often reinvent the wheel perhaps in order just to get started. It is always a little frustrating for a scientist to learn that somebody was there before he/she arrived!

New technologies But what do we do then? New technology has given radioecology - as any other science - possibilities which we fifty years ago could only dream about. But this also involves a risk. Modern computers can do nearly anything and they have been a tremendous step forward also when we try to analyse and model the behaviour of radionuclides in the environment. We should, however, never forget that any model should be based on reliable measurements if we wish it to mimic real life. An old expression says: garbage in - garbage out. This is also the case even for the most advanced and sophisticated Information Technology.

A. Aarkrog

XV

Instrumentation (solid state detectors and various mass spectrometers) for measuring the often very low concentrations of radionuclides in the environment has become widely accessible during the last decades. Today we can measure even very tiny and, from a health perspective, completely insignificant concentrations of radionuclides. Although we, from a scientific point of view are welcoming this, we should not ignore the reverse of the medal: if an environmental contaminant is measurable then it - for some people - is a proof that our environment is polluted and that we have to get rid of the source of this "pollution". This belief has probably been one of the major reasons for people's worry for radioactivity- which is so easy to detect even in insignificant quantities. "Do not measure simply because it is possible!!" as Scott (2000) puts it.

Models So today we can model and measure the behaviour of radionuclides in our environment and are thus at least in theory well prepared to handle nearly any release of radioactivity. I may here be too optimistic, but it does not seem likely to me that we could envisage a situation where we, due to lack of radioecological knowledge, were unable to protect man adequately against significant exposure from environmental radioactivity. This is of course not the same as saying that such exposures could not occur. We saw it for instance after the Chernobyl accident. The occurrence of thyroid cancer was, however, not due to missing radioecological knowledge, but to socio economic problems in the FSU. I am neither saying that radioecological models do not need improvements. "Appropriate research in radioecology can do much to reduce the uncertainty and increase the credibility of dose assessment models" as Ward Whicker (2000) rightly reminds us.

Tracer studies From what has been said above it is obvious that radioecology would benefit from having a broader perspective than just that connected to the nuclear option. One of the more successful applications of radioecological methods have been tracer studies. The large injections of radionuclides from nuclear weapons testing, reprocessing and the Chernobyl accidents has made it possible to study many atmospheric, aquatic and biological processes. One may, for example, mention the application of 99Tc and 129I - both radionuclides of little radiological significance - in the studies of the transport and dilution of pollutants in the NE-Atlantic including Arctic

XVI

Foreword

waters (Woodhead, 2001). Another example is the use of tracers (137Cs) to estimate feeding rates under natural field conditions (Whicker, 2000). This can be applied at an early stage to detect external stress in e.g. fish populations exposed to chemical pollutants.

Other pollutants In general radioecology would undoubtedly benefit from being recognised as a usable discipline in the studies of environmental pollution problems. The present congress could be an important vehicle for such an improved co-operation and understanding between radio- and chemoecology. In this context Polikarpov (2001) has suggested to make a comparative ecological equidosimetric assessment on the basis of Gy/y and Sv/y for physical, chemical and biological contaminants. Another aspect in this connection is the study of synergistic or antagonistic effects between different pollutants (Voigt, 2000). UNSCEAR (2000) has recently dealt with the combined exposures to radiation and other agents with respect to the induction of stochastic effects at low doses. With the exception of radiation and smoking UNSCEAR concludes that there is little indication from epidemiological data of strong antagonistic and synergistic combined effects. From a radioecological point of view one may ask if we could imagine that the behaviour of radionuclides in an environment would be influenced by the level of environmental contamination e.g. with chemical pollutants. Such studies have been proposed in the FSU.

Other species The so-called "redforest" of radiation damaged trees observed in the nearzone of Chernobyl has perhaps been the inspiration to develop protection standards targeted specifically to plants and animals (Scott, 2000; Whicker, 2000). ICRP (1977, 1991) has assumed that if man is adequately protected from environmental radioactivity then other organisms can be assumed to be adequately protected as well. Questions have being raised whether this assumption is still valid. In case of forest systems, Amiro (2000) concludes that doses less than 1 mGy/d"1 has little effect on forest organisms, furthermore doses above 1 mGy/d - 1 will only occur in highly contaminated areas where human activity is limited. In other words, harm to humans is in this case decisive for the dose limits. On the other hand, it is known that dose rates well below 1 mGy d-1 can produce chromosome damage and Whicker (2000) is asking whether chronic dose rates below 1 mGy d-1 affect the ultimate viability of long-lived organisms.

A. Aarkrog

XVII

When these discussions earlier have come up - e.g. in connection with possible enhanced radiation exposure to deep sea organisms from dumped rad-waste - it has been argued that although individuals may be at risk, the natural selection will ensure that populations are preserved. Nevertheless it has been concluded by IAEA (1999) that there is a serious need to develop defendable primary dose rate criteria for protection of the environment. This will require new knowledge on the transport, dosimetry (Kocher and Trabalka, 2000) and ecological effects of radionuclides in the environment (Alexakhin, 2000).

Countermeasures A major task - also for radioecology - in recent years has been the development of countermeasures (Scott, 2000; Voigt, 2000; Ohmomo, 2000; Wilkins, 2000; Whicker, 2000) to mitigate the effects of environmental contamination from accidents and at dismantled nuclear sites. In this connection the socio-economic effects and ethical aspects have added a new dimension to the obligations of the radioecologist (Polikarpov, 2001). Any method to clear land from radioactive contamination should not only be economic and technical feasible, but should also be accepted and understood by the people affected by such measures.

Information and education This leads me to our educational and informative obligations. Baxter (1999) has mentioned the need for information and education. It is a deplorable fact that although radioecology has been in existence for half a century, we have not succeeded in explaining to the public the implications of radioactive contamination of the environment. Many people still think that amongst all pollutants the radioactive ones are the most hazardous irrespective of their concentrations and occurrence. We are also obliged to involve scientists from the developing countries in our research. A mutual co-operation might be of benefit to both parties. The developed world would gain an opportunity to work in ecosystems, which are "terra incognito" to most radioecologists. The developing countries would for their part gain access to modern techniques and learn how to carry out radioecological studies so that these turned into more than just monitoring exercises.

XVIII

Foreword

Research The majority of radioecological research has so far been based on field observations of environmental radioactivity. Experiments in the laboratory or in well-defined controlled experimental environments are less frequent. Hinton (2000) has - based on an old article in Science (Plat, 1964) - pleaded for a more hypothesis-oriented radioecological research. To quote Plat: "we measure, we define, we compute, we analyse, but we do not exclude", and further we become "method-oriented" rather than "problem-oriented". I agree, and hope with Hinton that more of the radioecological research in the new millennium/century will develop in an experimental direction, where we will test hypothesis rather than just describe what we observe.

Concluding remarks Although many radioecologist may be worried for the future I have in the millennium editorials also noticed a general optimism. Radioecology has a good chance of survival in the next millennium because it is one of the most fascinating environmental sciences, one in which we see a fruitful interaction between the main scientific disciplines: mathematics, physics, chemistry and biology and all the sub-specialities such as meteorology, oceanography, geology, botany, zoology, physiology and statistics. Radioecologists have very different background and this makes the scientific dialog and co-operation challenging and inspiring. Hence, I think it is desirable that radioecology survives and develops, not necessarily because society finds it environmentally required for utilisation of the nuclear option, but because it is a promising and challenging environmental science.

Future trends Crystal ball gazing does not belong to any of the scientific disciplines mentioned above. So what is said in the following may be considered as pure fiction and no references to scientific journals will be provided. Bioand information-technology will in the new millennium influence all sciences dramatically - radioecology being no exception. Gene therapy and "smart" molecules will be developed to prevent many diseases including cancers. Hence, radiation protection will first of all be concentrated on avoidance of deterministic effects and the need for preventing stochastic effects will become less pertinent. This will probably change the whole philosophy of ICRP. The non-threshold concept may be quit and concern for

A. Aarkrog

XIX

low doses will wither. This again will influence the role of radioecology in radiation protection, as our discipline first of all has been dealing with relatively low activity levels. Hence the need for environmental monitoring of nuclear facilities will become less important. But the radioecological science may take significant advantages of the new bio-information technology. Organisms may be genetically engineered to deal specifically with inhibiting toxic substances, including radionuclides, from entering the foodchain. This may mitigate contamination problems in connection with waste disposal and nuclear accidents. We do; in fact already know that certain sorts of crop-species are showing lower root uptake of certain radionuclides than others. "Nanochips" may be used to study the metabolism of radionuclides (and other substances) at the cellular level and may thus give us a more fundamental understanding of essential environmental processes. This again may be applied to environmental modelling. Further development of analytical techniques may make it possible to identify pollutants from specific sources in minute concentrations reducing the need for collecting large samples - you may carry your samples "under your nails"! This could be used for the monitoring of facilities (nuclear, chemical and biological) in nations not complying with international treaties. I see a future for radioecology where our discipline will be integrated in this emerging bio-IT-complex together with all other bio-geo-chemical sciences. But in a millennium perspective we may also see radioecology and other environmental sciences be extended to other planets and environments perhaps even outside our own solar system. Some may here object by arguing that before taking up extra terrestrial radioecology we should be sure that we are familiar with all environments even the most extreme on our own planet. We may thus still have something to learn about behaviour of radionuclides (and other substances) in tropical environments and in the deep sea. The so-called "black smokers" i.e. sulphurous hot springs in the deep ocean may e.g. mimic what might be expected in some extraterrestrial environments. Although nobody can promise us, fusion power may be reality in this century. We may not expect this to have major radioecological impact. But we know that some less well-studied radionuclides will be produced as induced activity in fusion power plants, so some radioecological research has to be done to study the environmental behaviour of such elements. Let me finish these more or less farfetched thoughts by quoting the Danish humorist and philosopher Storm Petersen: "it is difficult to prophesy - in particular about the future" - so don't blame me if I was wrong!

XX

Foreword

References Aarkrog A. (1979) "Environmental studies on radioecological sensitivity and variability with special emphasis on the fallout nuclides 90Sr and 137Cs" D. Sc. thesis, University of Copenhagen, Denmark. Ris0-R-437. Ris0 National Laboratory. Alexakhin R.M. (2000) "Protection of the environment in the perspective of human health", Radiation Protection Dosimetry, 92,183-188. Amiro B.D. (2000) "Editorial. The radiological benefit of forests", /. of Environmental Radioactivity, 49, 7-9. Baxter M. (1999) "Editorial. Future priorities of JER", /. of Environmental Radioactivity, 42,1-5. Hanson W.C. (1966) "Radioecological concentration processes characterizing arctic ecosystems", Aberg B. and Hungate P.P. (Eds.), Radioecological concentration processes (Pergamon Press, Oxford) pp. 183191. Hinton T. (2000) "Editorial. Strong Inference, Science Fairs and Radioecology", /. of Environmental Radioactivity, 51, 277-279. Holm E. (2001) "Editorial. The Swedish nuclear dilemma", /. of Environmental Radioactivity, 52,113-115. Hunt G.J. and Jefferies D.F. (1981) "Collective and individual radiation exposure from discharges of radioactive waste to the Irish Sea", in Impacts of radionuclide releases into the marine environment (International Atomic Energy Agency, Vienna) pp. 535-570. IAEA (1999) "Protection of environment from the effects of ionizing radiation: A report for discussion", IAEA-TECDOC-1091 (International Atomic Energy Agency, Vienna). ICRP (1977) "Recommendations of the International Commission on Radiological Protection", Publication, 26 (Pergamon Press, Oxford). ICRP (1991) "Recommendations of the International Commission on Radiological Protection", Publication, 60 (Pergamon Press, Oxford). Johanson K.J., Bergstrom R., Bothmer S.V. and Karlen G. (1990) "Radiocaesium in wildlife of a forest ecosystem in central Sweden", Desmet G., Nassimbeni P. and Belli M. (Eds.), Transfer of radionuclides in natural and seminatural environments (Elsevier Applied Science, London) pp. 183-193. Kocher D.C. and Trabalka J.R. (2000) "On the application of radiation weighting factor for alpha particles in protection of non-human biota", Health Physics, 79,407-411. Liden K. (1961), "Caesium-137 burdens in Swedish Laplanders and Reindeer", Act Radial, 56, 237.

A. Aarkrog

XXI

Miettinen J.K., Jokelainen A., Roine P., Liden K. and Naversten Y. (1963) "137Cs and potassium in people and diet - study of Finnish Lapps", Ann. Acad. Sci. Fenn., A, II, 120,1-46. Nelson D.J. and Evans F.C. (1967) "Symposium on radioecology", Proceeding of the second National Symposium held at Ann Arbor, Michigan (United State Department of Commerce, Springfield, Virginia). Ohmomo Y. (2000) "Editorial. A brief history, and views on the future of radioecology in Japan", /. of Environmental Radioactivity, 49, 3-5. Plat J. (1964) "Strong Inference", Science, 146, 347-353. Polikarpov G.G. (2001) "Editorial. The future of radioecology: In partnership with chemo-ecology and eco-ethics", /. of Environmental Radioactivity, 53,5-8. Ramsaev P.V., Shamov V.P., Trozkaj M.N., Lebedev O.V. and Ibatullin M.S. (1965) "Indirect assessment of total body burden of 137Cs in people", Medizinskaj Radiologia, 6, 22-28. Rand Corporation (1953) "Report R-251-AEC, Project SUNSHINE" (Santa Monica, California). Scott M. (2000) "Editorial. Some millennium thoughts", /. of Environmental Radioactivity, 48,127-129. Timofeev-Ressovsky N. (1957) "Use of radiations and radiation sources in experimental biogeocoenology", Bot. zh, 42 (2), 161-194 (in Russian). UNSCEAR (1966) "Report of the United Nations Scientific Committee on the Effects of Atomic Radiation" (United Nations, New York). UNSCEAR (2000) "Sources and Effects of Ionizing Radiation", The United Nations Scientific Committee on the Effects of Atomic Radiation (United Nations, New York). Voigt G. (2000) "Editorial. Trends in radioecology", /. of Environmental Radioactivity, 48, 261-263. Whicker F.W. (2000) "Editorial. Radioecology: Relevance to problems of the new millennium", /. of Environmental Radioactivity, 50,173-178. Wilkins B. (2000) "Editorial. The development of countermeasure strategies in agricultural systems", /. of Environmental Radioactivity, 51,153-156. Woodhead D.S. (2001) "Editorial. Whither marine radioactivity studies?", /. of Environmental Radioactivity, 52,1-3.

This page intentionally left blank

INTRODUCTION

This page intentionally left blank

1

The ecological dimensions of ecotoxicological research Th. Caquet1 and F. Ramade 1

Introduction: Environmental toxicology or ecotoxicology? A major confusion occurs usually between "environmental toxicology" and "ecotoxicology", even among the community of specialised scientists. A number of definitions of ecotoxicology have been proposed since the end of the 1960s (see e.g. Ramade, 1977, 1987; Butler, 1978; Moriarty, 1983) and some people hold both terms as synonyms (Landis and Yu, 1995). In this presentation, we refer to ecotoxicology using the basic acceptation of this term (Ramade, 1977): a science which is concerned with the fate and effects of environmental contaminants at large scale of observation, namely the ecosystem level. Contamination pathways of the biosphere (including biota contamination) and biogeochemical cycling of contaminants are therefore within the scope of this approach. It focuses on the effects of contaminants on populations, communities and even biomes. It includes the effects on ecological processes such as biological productivity, nutrient cycling and ecological succession. Therefore, ecotoxicology relies on a holistic approach which is clearly based on the paradigms and uses the methods of ecology. On the opposite, environmental toxicology is a mechanistic and reductionist approach which focuses on studies performed at the individual level and frequently at a lower level of organization (e.g. cellular level; Fig. 1.1). In fact, the levels of biological organization represent a continuum that could be followed in either direction (bottom-up or top-down). Though ecotoxicology is basically an applied science it also generates a number of fundamental problems. Ecotoxicology requires understanding from theoretical chemistry, physiology and ecology to assess the ecological risks arised by environmental contaminants. It can be used both retrospectively (a fortiori or diagnostic approach: are there effects and what is causing them?) and prospectively (a posteriori or prognostic approach: are there likely to be significant effects?; Calow, 1996). 1

Ecology and Zoology Laboratory, CNRS UPRESA 8079, Ecology, Systematics and Evolution, batiment 442, Universite Paris-Sud, 91405 Orsay, France.

4

Th. Caquet and F. Ramade

Figure 1.1. Fields of investigation of environmental toxicology and ecotoxicology (adapted from Munkittrick and McCarty, 1995).

In this paper, we successively discuss how ecological theories may be implemented into the framework of ecotoxicological studies and how the characteristics of ecosystems may modify the fate and effects of environmental contaminants.

1. Implementation of ecological theories in ecotoxicology Studies in ecotoxicology should be performed according to a precise framework which should include recognition of the spatial extent and temporal trends of environmental contamination (Luoma, 1996). Pollutants rank among several other influential physical, chemical or biological variables in many ecosystems. One of the main objectives of ecotoxicological studies is to identify the cases where pollutants play a critical role. Although the effects of environmental contaminants may be detected at various levels of biological organization, they are frequently characterized at the community level (see e.g. Ramade, 1977, 1992). Changes in species composition are likely to alter ecosystem processes through changes in the functional traits of biota (Fig. 1.2). A considerable amount of work has been devoted to evaluation of the effects of environmental contaminants on communities and ecosystems, and numerous descriptors of these effects have been proposed (Tab. I.I). However, ecotoxicological studies are rarely based on ecological theories or models. Main hypotheses concerning the response of an ecosystem to anthropogenic stress have been summarized by Odum (1985) which considered that

1 - The ecological dimensions of ecotoxicological research

5

stress forces an ecosystem into an earlier, less mature successional stage. The consequences on the response of communities to stress are: • the average size of organisms is reduced; • species diversity decreases; • food chains shorten; • redundancy declines; • the efficiency of resource use declines.

Figure 1.2. Linkages between community characteristics (species composition and diversity) and ecosystem processes (including productivity and nutrient cycling; modified after Chapin et al., 1997). Regional processes include trace gas fluxes to the atmosphere and nutrient fluxes from terrestrial to aquatic systems. Community processes include competition and predation. "Ecosystem services" are the benefits derived by humans from ecological processes.

Size-structure of organisms has been used for a long time in limnology, especially for planktonic and pelagic organisms (see e.g. Sheldon et al., 1972; Kerr, 1974; Borgmann, 1987; Sprules et al., 1991; Havens, 1998; Tittel et al, 1998). Some authors have shown that changes in size-frequency distribution could be used to evaluate the effects of environmental perturbations, including those caused by pollutants (Warwick, 1993; Havens, 1994; Hanazato, 1998). Measurements of abundance (density, biomass) and taxonomic diversity are certainly the most frequently used parameters in the assessment of the ecotoxicological effects of environmental contaminants. However, the exposure-diversity relationship is frequently not linear, which makes the understanding of the impact of a given concentration of contaminant on a community difficult (see e.g. Ramade et al., 1984; Ramade, 1997). Additionally, taxonomic identification of organisms to the species level, which is usually required to compute the various diversity indices, is time-consuming. Moreover, these parameters may sometimes fail in detecting the effects of pollutants, especially under chronic exposure to low level of contaminants (Pusey et al, 1994; Woin, 1998).

6

Th.Caquet and F. Ramade Table 1.I. Examples of descriptors which may be used to detect contaminant effects in aquatic ecosystems and predicted response to stressors.

Descriptots

Relative value according to the effect of stressors No effect

Significant effect

Small Big

Big Small High Low

Descriptors of structure

Size of phytoplanktonic organisms Size of zooplanktonic and benthic organisms

Phytoplankton biomass (Bp) Macrozooplajcton biomass (Bmacro)

Low High High

Microzooplakton biomass (Bmicro)

Low

High

Bz/Bpratio

High

Low

B

High

Low

1

Taxonomic diversity

High

Low

Functional diversity1

High

Low

Zooplankton biomass (B z )

macro/Bmicro

ratio

1

Low

High

Genetic diversity

LOW:

High (Low)

Low (High)

Food web complexity

High High

Food web redundancy

High

Low Low Low

Evenness (Dominance)

Food web length

Descriptors of function

High

Photosynthetic assimilation

Ressource use efficiency (RUE) 2

Gross production/Respiration ratio (P/R)3 Gross production/Biomass ratio (P/B)3 Nutrients cycling Rate of organic matter decay 1

Low Low Low Low

High High High

High High

- ,

, LOW

Low

2

Applicable to all categories of organisms; RUE = assimilation rate of carbon by zooplankton/assimilation rate of carbon by phytoplankton; 3 as defined in Odum (1985).

Menge and Sutherland (1987) proposed a conceptual model of marine rocky intertidal community regulation under conditions of disturbance, which is based on several predictions and assumptions (Tab. l.II). According to Menge and Farrell (1989), this model applies to various habitats, including shallow marine benthic zones, streams, shallow lakes, freshwater and terrestrial communities at high elevations, and some deserts. Although environmental contaminants were not quoted among the factors of disturbance identified by Menge and Sutherland (1987), some of the predictions and assumptions of their model may apply to polluted ecosystems. Studying food webs in acid-stressed lakes, Locke (1996) showed that although several assumptions of the Menge-Sutherland model were met, the predictions of the model were only weakly supported.

1 - The ecological dimensions of ecotoxicological research

7

Table l.II. Predictions and assumptions of the Menge-Sutherland model (Menge and Sutherland, 1987). Predictions

Assumptions

In communities in which competing species coexist (e.g. zooplankton), maximum species richness occurs at intermediate level of stress

Dominant species are susceptible to the disturbance

Food chain length declines with increasing stress

Important consumers are omnivorous

Web complexity declines with increasing stress

Stress can directly maintain low diversity

The relative contribution of factors controlling food web structure varies in a predictable manner along a stress gradient. Predation, competition and disturbance each predominate in tarn as controlling factors With increasing stress

Primary productivity is independent of stress until stress is extreme

Consumers are more sensitive to the stress than are their prey Recruitment is independent of local factors and ;therefore does not vary with the disturbance

Applying the intermediate disturbance hypothesis (Connell, 1978; Sousa, 1984) to freshwater stream invertebrate communities, Hildrew and Townsend developed the habitat templet theory which indicates that environmental variability has a major role in structuring these communities (Hildrew and Townsend, 1987; Townsend and Hildrew, 1994). Productivity and functional diversity are closely related to disturbance frequency (Fig. 1.3). This model has been designed within the frame of understanding the effects of physical stress (e.g. flow perturbation) but some data indicate that it may also describe the effects of pollutants in streams (Usseglio-Polatera et al., 1999). The problems associated with the evaluation of the consequences of the effects of environmental contaminants on communities at the level of ecosystems need to be considered in the context of the studies concerning the relationship between the biodiversity of communities and ecosystem function and stability (see e.g. Tilman, 1996, 1997; Tilman et al, 1996; Chapin et al, 1997, 1998; Peterson et al, 1998; Schwartz et al, 2000; Loreau, 2000).

8

Th. Caquet and F. Ramade

"Stability" is a frequently used term in ecology (and sometimes in ecotoxicology) which has been given a variety of definition (see e.g. McCann, 2000).

Figure 1.3. Disturbance-productivity-diversity model for communities of freshwater benthic macroinvertebrates (from Hildrew and Townsend, 1987).

Stability of a system and its response to disturbances is frequently described using the behaviour of a marble on a surface consisting of peaks (instability domains) and valleys (stability domains; Fig. 1.4). Studies on food webs in ecosystems have shown that their functional characteristics may be studied quantitatively through computation of ascendency (Ulanowicz, 1992). The ascendency index, which is both an estimator of the ecosystem "stability" and changes, is sensitive to stressors. It potentially constitutes a useful tool for the estimation of pollutant effects on the development of ecosystems. Ecological systems are frequently metastable, i.e. they can flip from one state to another. The forces which are necessary for such a change can be very different in type and strength but environmental contaminants may have a role in these phenomena. The return time to equilibrium is named "engineering resilience" whereas the amount of disturbance that a system can absorb without changing stability domain defines the "ecological resilience" of the system (Gunderson, 2000). In engineering science, resilience is a static property of a system. In ecology, stability domains (and therefore resilience) are dynamics and variable. The diversity-stability hypothesis states that increased biodiversity leads to greater ecological stability. The probability that some species may survive following an environmental disturbance and thus replace more susceptible competitors should increase with biodiversity of the ecosystem. However, both experimental (see e.g. Tilman, 1996) and theoretical studies (see e.g. May, 1972; Loreau and Behera, 1999) suggest that the relationship between biodiversity and ecosystem properties is not so simple (McCann, 2000).

1 - The ecological dimensions of ecotoxicological research

9

Figure 1.4. Stability states and metastable points of different local and global stability for a system (modified after Gordon and Forman, 1983 in Seitz, 1994). An undisturbed natural system may be analogous to point D, which is the most metastable point and the point which exhibits highest resilience. Following an initial disturbance, a system may be displaced to metastable points B or C. After a further move to metastable point A by an additional disturbance, recovery to points B or C is more likely to occur than return to natural state (point D). Points E, F and G are instable transient states.

One of the main points in ecotoxicological studies is to identify whether the disturbance induced by pollutants is sufficient to move an ecological system from a stability domain to another. There is no simple answer to this question. The final outcome of the introduction of a contaminant in the environment depends on the nature of the compounds, on the existence of other stressors (either natural or of an anthropogenic origin) and on the characteristics of the exposed ecological system. The existence of redundancy within natural communities is of pivotal importance in this context. Functional redundancy is sometimes predicted by theroretical model linking biodiversity and stability of ecosystem function (Fig. 1.5). In a literal sense, "redundancy" means superfluous or unnecessary but it is a necessary and valuable concept in ecology (Naeem, 1998).

10

3Th. Caquet and F. Ramade

Figure 1.5. Main theoretical models describing the relationship between biodiversity (evaluated through species richness) and the stability of an ecological function (productivity, rate of organic matter decay, CO2 fluxes, etc.; after Peterson et al, 1998). The shape of the bell curve gives information on the functional importance of each species. A) Model of MacArthur (1955). The size of functional space is great and all species have the same functional importance. Species may be added without saturation and the stability of the ecological function increases in a continuous way. B) Idyosincratic model of Lawton (1994). The contribution of a species to an ecological function is deeply influenced by its relations with other species. Introduction or disappearance of a species may have major or no detectable consequences, depending on the concerned species and on the species with which it interacts. C) "Rivet-popping" model of Ehrlich and Ehrlich (1981). The size of functional space is limited. Overlap between species increases with the number of species. An ecological function may persist instead of the disappearance (or reduction in abundance) of a limited number of species because other species with similar roles may compensate for this disappearance (or reduction in abundance). This phenomenon is sometimes described as functional redundancy. The increase in stability associated with the introduction of new species diminishes as species richness increases. D) "Drivers and passengers" model of Walker (1992,1995). The functional importance is very different from one species to another. "Drivers" have a major role whereas "passengers" have a minor influence. Addition of "drivers" increases the stability of the system whereas the introduction of "passengers" has no visible consequences. "Drivers" may have an important effect on the physical attributes of the environment ("ecological engineers") or exhibit very strong relationships with other species ("keystone species").

1 - The ecological dimensions of ecotoxicological research

11

Natural ecosystems often contain many species per functional group. Functional redundancy has been identified in aquatic communities exposed to various stressors (de Noyelles and Kettle, 1985; de Noyelles et al., 1982, 1989; Schindler, 1990; Brock et al., 1993; Frost et al, 1995). It is due to differences in sensibility between species toward stress (Klug et al., 2000). Therefore, a pollutant can cause the replacement of one susceptible species by a former competitor species. In some cases, the regulation of trophic function may be maintained by predator flexibility in prey choice (Sheehan, 1984). Additionally, local extinction of species within functional groups is inevitable and frequent, but reservoirs of species from nearby ecosystems generally ensure that functional group or ecosystem failure, if it occurs, is likely to be transient. However, many stressors have a negative impact on biodiversity and therefore on redundancy. This may lead to an increased susceptibility of disturbed communities to further disturbances. When multiple stressors are present, functional redundancy may be critically affected and stability of the ecosystem may be seriously affected. Scale is another central problem in ecology because ecological phenomena cannot be studied at a single scale (Fig. 1.6) and phenomena that happen at one scale can influence observations at another. Similarly, temporal and spatial scaling is very important in ecotoxicology (Fig. 1.7).

Figure 1.6. Examples of the different scales at which ecological studies may be performed on freshwater macroinvertebrates (freshwater snails of the genus Lymnaea in this example).

The higher the level of biological organisation, the longer the time needed to assess pollutants effects. Extrapolation from short-term to long-term effects is therefore very difficult or even impossible (Ramade, 1997). In many cases, recovery of disturbed ecosystem requires several years. "Recovery" is frequently linked to recolonisation by affected species. In this context, various definitions of population recovery have been proposed: • the number of days an affected population's growth rate lags behind an unaffected population (Kareiva et al., 1996); • the time taken for the population to recover to 80% of the control (Thacker and Jepson, 1993);

Th. Caquet and F. Ramade

12

Figure 1.7. Three-dimensional view of the scales of observations in ecotoxicological studies (after Solomon, 1998 in Kedwards, 2000).

Time

Figure 1.8. Changes with time of the abundance of a given species (X) in control ecosystems (solid line) and in a disturbed ecosystem (dashed line; modified after Wiens, 1996). The dotted line represents the natural variability of the abundance of the species in controls. Recovery can be defined as the return of the perturbed population to the window of natural variability.

1 - The ecological dimensions of ecotoxicological research

13

• the time where there is no more significant difference in population density between a control and an affected population (van den Brink et al., 1996); • the return of the perturbed system to the window of natural variability (Wiens, 1996; Fig. 1.8); • the time when the numbers relative to the control return to an arbitrary 0.9 times the number before the chemical was added (Sherratt et al, 1999). All species do not exhibit the same capacities of recovery following a disturbance. Population density in the disturbed ecosystem, stability of its age/stage structure, stochasticity of its dynamics, genetic diversity, natural mortality or isolation/connection with other population of the same species are among the main characteristics of species that may influence population recovery following a disturbance (Kedwards, 2000; Fig. 1.9). Even if pollution stops and environmental conditions become favorable, a species (or a group of species) which has been severely affected may not recover because it has been replaced by one or several tolerant species with the same functional role, or because its food has disappeared. This phenomenon is frequently observed in conservation biology during reintroduction programs. Metapopulation dynamics and population genetics may significantly contribute to a better knowledge of population recovery (Hanski, 1991, 1994; Hanski and Gilpin, 1997) and increase the relevance of ecotoxicological studies.

Figure 1.9. Theoretical relationship between species characteristics and recovery probability in a disturbed ecosystem (the shape of the relationship is arbitrary; based on data of Kedwards, 2000).

14

Th. Caquet and F. Ramade

2. Interaction between environmental contaminants and the characteristics of ecosystem: The case study of eutrophication and persistent organic pollutants Many studies have focused on the effects of environmental contaminants on the structure and function of ecological systems but few have examined the influence of ecosystem characteristics on the nature and magnitude of these effects. Among environmental factors, some are known to significantly influence the fate of contaminants and sometimes their bioavailability and thus their potential effects. This is for example the case of water pH or hardness or of soil and sediment organic matter content. However, although the interaction between nutrients and toxicity has sometimes been studied, little is known about the effects of trophic status on the effects of pollutants. Many aquatic ecosystems are simultaneously exposed to contaminants and eutrophication. Organic enrichment and environmental contaminants may interact in their effects on aquatic biota but there are few studies on these interactions (see e.g. Pratt and Barreiro, 1998). Effects from eutrophication alone may modify how contaminants affect biota in many ways (Hylland et al, 1996): • one of the characteristics of eutrophic systems compared to oligotrophic systems is the increased biomass in the former. Therefore, a given contaminant load could be "diluted" into more individuals and more biomass in an eutrophic system than in an oligotrophic system, resulting in lower contaminant level in individual organisms; • the concentrations of dissolved and particulate organic material in the water column will be higher in more eutrophic systems, causing a decrease in the water-soluble fraction of contaminants (Baker et al., 1985; Evans, 1988); • sedimentation rates are generally higher in more eutrophic systems, resulting in shorter residence times for particle-bound contaminants in the water column (Pavoni et al., 1990; Millard et al, 1993). These interactions between trophic status and contaminant fate and effects may be complicated by the fact that the contaminants may also affect the biological response to eutrophication. Increased food availability such as that found in eutrophic systems is expected to result in enhanced growth and reproduction. Such responses may be inhibited in the presence of environmental contaminants (Weisse, 1991). Interactions between eutrophication and contaminants have been detected for sediment-dwelling (Hylland et al, 1996) and pelagic organisms

1 - The ecological dimensions of ecotoxicological research

15

(Larsson et al., 2000). In the following we have focused on the interactions between persistent organic pollutants (POPs) and pelagic organisms, which constitute more or less complicated food web (Fig. 1.10). Plankton organisms are the first link for pollutant transfer in the pelagic system. Recent studies have shown that the adsorptive capacity of bacteria is an important route for POPs transportation via the microbial part of the food web. This transfer through the microbial loop may enhance the contamination of higher trophic levels (Broman et al, 1996). Because of their high abundance, their small size and relatively fast turnover times, bacteria cells represent the largest particulate biological surface area in natural waters (approximately 0.2 to 4 m2/m3; Larsson et al., 2000). This renders them especially important as an adsorptive matrix for POPs. Their size also makes them less available for "growth-dilution effects" of POPs, since the time to attain equilibrium between the organic matrix of the "bacterial particle" and the surrounding water is shorter than that of larger particles. Bacteria therefore have the potential to take up a larger proportion of POPs from the water than phytoplankton.

Figure 1.10. Schematic view of the food web in pelagic systems (modified from Larsson et al., 2000). The food web comprises a 'classical' part, based on phytoplanktonic heterotrophic production, and a microbial part, based on the use of dissolved organic carbon (DOC). The microbial part is more important in nutrient-poor ecosystems whereas the "classical" part dominates in nutrient-rich systems.

16

Th. Caquet and F. Ramade

The uptake of POPs in bacteria and phytoplankton is governed by a concentration dependent adsorption process from the water. In laboratory systems, the uptake of POPs in plankton has been described as a rapid process following first-order kinetics. The initial phase is a concentration-dependent adsorption of POPs from the water by plankton that reaches an 'apparent' steady state within a few hours. Swackhamer and Skoglund (1993) and Stange and Swackhamer (1994) have shown that the uptake of POPs in phytoplankton initially follows a fast first-order kinetics but that this phase is followed by a slow uptake that continues for weeks. The adsorbed molecules are transferred to the interior of the cell, with a concurrent further uptake from the water which significantly contributes to the total POP concentration detected in the algae. In the field, POPs concentrations in phytoplankton are in agreement with equilibrium partitioning predictions in winter whereas a significant deviation is observed in summer, when plankton growth rate is high. In fact, if the uptake of POPs is faster than algae individual growth rate, growth rate of the algae will not affect concentration reached in the organisms. If the growth rate is faster, the pollutants will dilute in growing biomass and will not reach equilibrium until growth decreases (Fig. 1.11).

Figure 1.11. Schematic view of POPs "biodilution" in a fast-growing phytoplankton community (from Larsson et aL, 2000). In Time-zone 1, phytoplankton growth rate is still slow and concentration of POPs in the plankton quickly reaches an apparent equilibrium. In Time-zone 2, the growth rate of phytoplankton is faster than the uptake kinetics of POPs. The concentration of POPs decreases (= "biodilution"). In Time-zone 3, growth rate decreases and a new equilibrium in POP concentrations is reached.

Phytoplankton growth rate can be high in eutrophic systems and could induce differences compared to oligotrophic aquatic environments. This would result in lower concentrations of POPs in pelagic producers in

1 - The ecological dimensions of ecotoxicological research

17

nutrient-rich waters compared to that of nutrient-poor ecosystems. This phenomenon will be more important during seasons or situations when the plankton exhibits high growth rates (e.g. spring and summer). As phytoplankton is the main food resource for zooplankton, the effects will cascade to higher trophic levels (Larsson et al., 2000). The size of planktonic organisms may also influence the uptake of POPs. Axelman et al. (1997) showed that the partitioning coefficients of PCBs between organic carbon and water (Koc) and between n-octanol and water (Kow) were in good agreement for bacterioplankton (mean size: 0.2-2 m) but not for bigger cells (mean size: 2-25 um) thus suggesting "biodilution" (Fig. 1.12). The larger cells also showed lower partitioning coefficients than bacteria.

Figure 1.12. Relationship between organic carbon/water partitioning coefficients measured in the field (field logKoc) and w-octanol/water partitioning coefficients (logKow), for two plankton size fractions (based on the data of Axelman et al., 1997). Results of field studies indicate that the uptake of many POPs by aquatic biota is lower in eutrophic than in oligotrophic lakes (Taylor et al., 1991; Larsson et al., 1992). Several causes have been identified (Fig. 1.13): • an increased sedimentation in eutrophic compared to oligotrophic lakes, causing a scavenging of POPs from the water, and a further incorporation into the sediment; • a higher lipid content in algae from nutrient-poor ecosystems;

18

Th. Caquet and F. Ramade

• a higher growth rate of fish in eutrophic lakes, causing growth dilution of POPs; • a higher growth rate of phytoplankton in eutrophic lakes, causing "biodilution" of POPs as described previously.

Figure 1.13. Simplified view of the fate of POPs in eutrophic and oligotrophic environments. In eutrophic systems, two processes affect the contamination of the pelagic community: the exchange from the atmosphere and the sedimentation of the phytoplankton biomass. These processes are of smaller magnitude or counteracted in oligotrophic environments (from Larsson et al., 2000).

Conclusion - Perspectives For thirty years, ecotoxicology has frequently been considered as a simple extrapolation of classical toxicology to natural objects. However, biological responses in the natural environment can be much more complex than those observed in the laboratory and linking an ecological change to the influence of a pollutant requires an understanding of what fundamental ecological processes are sensitive to environmental contaminants. The recent developments of theoretical ecology regarding the effects of disturbance on the structure and function of communities and ecosystems offer a promising framework for ecotoxicological studies. In some circumstances, environmental contaminants may be used as tools to test the predictions of these theories, using for example model ecosystems. Furthermore, at a lower biological level of organisation, ecotoxicology like ecology should utilise benefits of new approaches in population biology (e.g. use of DNA microsatellites, studies on metapopulation) and modelling.

1 - The ecological dimensions of ecotoxicological research

19

References Axelman J., Broman D. and Naf C. (1997) "Field measurements of PCB partitioning between water and planktonic organisms - influence of growth, particle size and solute-solvent interactions", Environ. Sci. TechnoL, 31, 665-669. Baker J.E., Eisenrich S.J., Johnson T.C. and Halfman B.M. (1985) "Chlorinated hydrocarbon cycling in the benthic nepheloid layer of Lake Superior", Environ. Sci. TechnoL, 19, 845-861. Borgmann U. (1987) "Models of the slope of, and the biomass flow up, the biomass size spectrum", Can. J. Fish. Aquat. Sci., 44 (suppl. 2), 136-140. Brock T.C.M., Vet J.J.R.M., Kerkhofs M.J.J., Lijzen J., van Zuilekom W.J. and Gijlstra R. (1993) "Fate and effects of the insecticide Dursban 4E in indoor Elodea-dominated and macrophyte-free freshwater model ecosystems: III. Aspects of ecosystem functioning", Arch. Environ. Contam. Toxicol, 25,160-169. Broman D., Naf C., Axelman J., Bandh C., Pettersen H., Johnstone R. and Wallberg P. (1996) "The significance of bacteria in marine waters for the distribution of hydrophobic organic contaminants", Environ. Sci. Technol., 30,1238-1241. Butler G.C. (1978) "Principles of Ecotoxicology" SCOPE 12 (J. Wiley & Sons, New York). Calow P. (1996) "Ecology in ecotoxicology: Some possible "rules of thumb", Baird D.J., Maltby L., Greig-Smith P.W. and Douben P.E.T. (Eds.), ECOtoxicology: Ecological Dimensions (Chapman & Hall, London), pp. 512. Chapin F.S. III, Walker B.H., Hobbs R.J., Hooper D.U., Lawton J.H., Sala O.E. and Tilman D. (1997) "Biotic control over the functioning of ecosystems", Science, 277, 500-504. Chapin F.S., Sala O.E., Burke I.C., Grime J.P., Hooper D.U., Lauenroth W.K., Lombard A., Mooney H.A., Mosier A.R., Naeem S., Pacala S.W., Roy J., Steffen W.L. and Tilman D. (1998) "Ecosystem consequences of changing biodiversity", BioScience, 48,45-52. Connell J.H. (1978) "Diversity in tropical rain forests and coral reefs", Science, 199,1302-1310. deNoyelles F. Jr. and Kettle W.D. (1985) "Experimental ponds for evaluating bioassay predictions", Boyle T.P. (Ed.), Validation and Predictability of Laboratory Methods for Assessing the Fate and Effects of Contaminants in Aquatic Ecosystems. ASTM STP 865 (American Society for Testing and Materials, Philadelphia), pp. 91-103. deNoyelles F., Kettle W.D. and Sinn D.E. (1982) "The responses of plankton communities in experimental ponds to atrazine, the most heavily used pesticide in the United States", Ecology, 63,1285-1293.

20

Th. Caquet and F. Ramade

deNoyelles F. Jr., Kettle W.D., Fromm C.H., Moffett M.F. and Dewey S.L. (1989) "Use of experimental ponds to assess the effects of a pesticide on the aquatic environment", Voshell J.R. (Ed.), Using Mesocosms to Assess the Aquatic Ecological Risk of Pesticides: Theory and Practice. Misc. Publ. Entomol. Soc. Am., 75,41-56. Ehrlich P.R. and Ehrlich A.H. (1981) "Extinction: the Causes and Consequences of the Disappearance of Species" (Random House, New York). Evans H.E. (1988) "The binding of three PCB congeners to dissolved organic carbon in freshwaters", Chemosphere, 17, 2325-2338. Frost T.M., Carpenter S.R., Ives A.R. and Kratz T.K. (1995) "Species compensation and complementarity in ecosystem function", Jones C.G. and Lawton J.H. (Eds.), Linking Species and Ecosystems (Chapman & Hall, New York), pp. 224-239. Gunderson L.H. (2000) "Ecological resilience in theory and application", Ann. Rev. Ecol. Syst., 31,425-439. Hanazato T. (1998) "Response of a zooplankton community to insecticide application in experimental ponds: A review and the implications of the effect of chemicals on the structure and functioning of freshwater communities", Environ. Pollut., 101, 361-373. Hanski I. (1991) "Single species metapopulation dynamics: concepts, models and observations", Biol. J. linn. Soc., 42,17-38. Hanski I. (1994) "A practical model of metapopulation dynamics", /. Anim. Ecol, 63,151-162. Hanski I. and Gilpin M. (1997) "Metapopulation Biology: Ecology, Genetics and Evolution" (Academic Press, San Diego). Havens K.E. (1994) "Experimental perturbation of a freshwater plankton community: a test of hypotheses regarding the effects of stress", Oikos, 69,147-153. Havens K.E. (1998) "Size structure and energetics in a plankton food web", Oikos, 81, 346-358. Hildrew A.G. and Townsend C.R. (1987) "Organization in freshwater benthic communities", Gee J.H.R. and Ciller P.S. (Eds.), Organization of Communities, Past and Present (Blackwell Scientific Publications, Oxford), pp. 347-371. Hylland K., Skold M., Gunnarsson J.S. and Skei J. (1996) "Interactions between eutrophication and contaminants. IV. Effects on sediment-dwelling organisms", Mar. Pollut. Bull, 33, 90-99. Kareiva P., Stark J. and Wennergren U. (1996) "Using demographic theory, community ecology and spatial models to illuminate ecotoxicology", Baird D.J., Maltby L., Greig-Smith P.W. and Douben P.E.T. (Eds.), ECOtoxicology: Ecological Dimensions. Ecotoxicology Series 4 (Chapman & Hall, London), pp. 13-23.

1 - The ecological dimensions of ecotoxicological research

21

Kedwards T. (2000) "Methods of evaluating population recovery", SET AC Technical Workshop: Regulatory Evaluation of Aquatic Microcosm Studies with Pesticides (Brighton, 25-26 May 2000). Kerr S.R. (1974) "Theory of size distribution in ecological communities", /. Fish. Res. Ed Can., 31,1859-1862. Klug J.L., Fischer J.M., Ives A.R. and Dennis B. (2000) "Compensatory dynamics in planktonic community responses to pH perturbations", Ecology, 81, 387-398. Landis W.G. and Yu M.-H. (1995) ''Introduction to Environmental Toxicology. Impacts of Chemicals upon Ecological Systems" (Lewis Publishers, Boca Raton). Larsson P., Okla L., Collvin L. and Meyer G. (1992) "Lake productivity and water chemistry as governors of the uptake of persistent pollutants in fish", Environ. Toxicol. Chem., 26, 346-352. Larsson P., Andersson A., Broman D., Nordback J. and Lundberg E. (2000) "Persistent organic pollutants (POPs) in pelagic systems", Ambio, 29,202209. Lawton J.H. (1994) "What do species do in ecosystems?", Oikos, 71, 367-374. Locke A. (1996) "Applications of the Menge-Sutherland model to acid stressed lake communities", Ecol. Appl, 6, 797-805. Loreau M. (2000) "Biodiversity and ecosystem functioning: recent theoretical advances", Oikos, 91, 3-17. Loreau M. and Behera N. (1999) "Phenotypic diversity and stability of ecosystem processes", Theor. Pop. Biol, 56, 29-47. Luoma S.N. (1996) "The developing framework of marine ecotoxicology: Pollutants as a variable in marine ecosystems?", /. Exp. Mar. Biol. Ecol., 200, 29-55. MacArthur R.M. (1955) "Fluctuations of animal populations and a measure of community stability", Ecology, 36, 533-536. May R.M. (1972) "What is the chance that a large complex system will be stable?", Nature, 237, 413-414. McCann K.S. (2000) "The diversity-stability debate", Nature, 405, 228-233. Menge B.A. and Farrell T.M. (1989) "Community structure and interaction webs in shallow marine hard-bottom communities: Tests of an environmental stress model", Adv. Ecol. Res., 18,189-262. Menge B.A. and Sutherland J.P. (1987) "Community regulation: variation in disturbance, competition and predation in relation to environmental stress and recruitment", Am. Nat., 130, 730-757. Millard E.S., Halfon E., Minns C.K. and Charlton C.C. (1993) "Effect of primary productivity and vertical mixing on PCB dynamics in planktonic model ecosystems", Environ. Toxicol. Chem., 12, 931-946. Moriarty F. (1983) "Ecotoxicology", 2nd Ed. (Academic Press, London).

22

Th.Caquet and F. Ramade

Munkittrick K.R. and McCarty L.S. (1995) "An integrated approach to aquatic ecosystem health: Top-down, bottom-up or middle-out?", J. Aquat. Ecosyst. Health, 4, 77-90. Naeem S. (1998) "Species redundancy and ecosystem reliability", Conserv. Biol, 12, 39-45. Odum E.P. (1985) "Trends expected in stressed ecosystems", BioScience, 35, 419-422. Pavoni B., Calvo C, Sfriso A. and Orio A.A. (1990) "Time trend of PCB concentrations in surface sediments from a hypertrophic, macroalgae populated area of the Lagoon of Venice", Sci. Total Environ., 91, 13-21. Peterson G., Allen C.R. and Holling C.S. (1998) "Ecological resilience, biodiversity, and scale", Ecosystems, 1, 6-18. Pratt J.R. and Barreiro R. (1998) "Influence of trophic status on the toxic effects of a herbicide: A microcosm study", Arch. Environ. Contam. ToxicoL, 35, 404-411. Pusey B.J., Arthington A.H. and McLean J. (1994) "The effects of a pulsed application of chlorpyrifos on macroinvertebrate communities in an outdoor artificial stream system", Ecotoxicol. Environ. Saf., 27,221-250. Ramade F. (1977) "Ecotoxicologie" (Masson, Paris), 214 p. Ramade F. (1987) "Ecotoxicology" (John Wiley), 262 p. Ramade F. (1992) "Precis d'Ecotoxicologie" (Masson, Paris), 312 p. Ramade F. (1997) "Assessment of damages to ecosystems: A major issue in ecotoxicological research", Qual. Insur., 5,199-220. Ramade F., Cosson R., Echaubard M., Le Bras S., Moreteau J.-C. and Thybaud E. (1984) "Detection de la pollution des eaux en milieu agricole", Bull. Ecol, 15, 21-37. Schindler D.W. (1990) "Experimental perturbations of whole lakes as tests of hypotheses concerning ecosystem structure and function", Oikos, 57, 2541. Schwartz M.W., Brigham C.A., Hoekesema J.D., Lyons K.G., Mills M.H. and van Mantgem P.J. (2000) "Linking biodiversity to ecosystem function: Implications for conservation ecology", Oecologia, 122,297-305. Seitz A. (1994) "The concept of ecological stability applied to aquatic ecosystems", Hill I.R., Heimbach F., Leeuwangh P. and Matthiessen P. (Eds.), Freshwater Field Tests for Hazard Assessment of Chemicals (Lewis Publishers, Boca Raton), pp. 3-18. Sheehan P.J. (1984) "Functional changes in the ecosystem", Sheehan P.J. Miller D.R. and Bourdeau P. (Eds.), Effects of Pollutants at the Ecosystem Level. Scope 22 (John Wiley & Sons, Chichester), pp. 101-194. Sheldon R.W., Prakash A. and Sutcliffe W.H. Jr. (1972) "The size distribution of particles in the ocean", Limnol. Oceanogr., 17, 327-340. Sherrat T.N., Roberts G., Williams P., Whitfield M., Biggs J., Shillabeer N. and Maund S.J. (1999) "A life-history approach to predicting the reco-

1 - The ecological dimensions of ecotoxicological research

23

very of aquatic invertebrate populations after exposure to xenobiotic chemicals", Environ. Toxicol. Chem., IS, 2512-2518. Sousa W.P. (1984) "The role of disturbance in natural communities", Ann. Rev. Ecol Syst., 15, 353-391. Sprules W.G., Brandt S.B., Stewart D.J., Munawar M., Jin E.H. and Love J. (1991) "Biomass size spectrum of the Lake Michigan pelagic food web", Mar. Ecol. Prog. Ser., 54,157-170. Stange K. and Swackhamer D.L. (1994) "Factors affecting phytoplankton species-specific differences in accumulation of 40 polychlorinated biphenyls (PCBs)", Environ. Toxicol. Chem., 13,1849-1860. Swackhamer D.L. and Skoglund R.S. (1993) "Bioaccumulation of PCBs by algae: Kinetics versus equilibrium", Environ. Toxicol. Chem., 12, 831-838. Taylor W.D., Carey J.H., Lean L.R.S. and Mc Queen D.J. (1991) "Organochlorine concentrations in plankton of lakes in southern Ontario and their relationship to plankton biomass", Can. J. Fish. Aquat. Sci., 48, 1960-1966. Thacker J.R.M. and Jepson P.C. (1993) "Pesticide risk assessment and nontarget invertebrates: Integrating population depletion, population recovery and experimental design", Bull. Environ. Contam. Toxicol., 51, 523531. Townsend C.R. and Hildrew A.G. (1994) "Species traits in relation to a habitat templet for river systems", Freshwater Biol, 31, 265-275. Tilman D. (1996) "Biodiversity: Population versus ecosystem stability", Ecology, 77, 350-363. Tilman D. (1997) "Biodiversity and ecosystem functioning", Daily G.C. (Ed.), Nature's Services: Societal Dependence on Natural Ecosystems (Island Press, Washington), pp. 93-112. Tilman D., Wedin D. and Knops J. (1996) "Productivity and sustainability influenced by biodiversity in grassland ecosystems", Nature, 379, 718720. Tittel J., Zippel B., Greller W. and Seeger J. (1998) "Relationships between plankton community structure and plankton size distribution in lakes of northern Germany", Limnol. Oceanogr., 43, 1119-1132. Ulanowicz R.E. (1992) "Ecosystem health and trophic web network", Costanza R., Norton B.G. and Haskell B.J. (Eds.), Ecosystem Health (Island Press, Washington), pp. 190-222. Usseglio-Polatera Ph., Thomas S., Beisel J.-N. and Moreteau J.-C. (1999) "Illustration de la valeur indicatrice des caracteristiques biologiques des macroinvertebres d'une communaute benthique a differentes echelles d'observation", Annls Limnol, 35, 71-80. van den Brink P.J., van Wijngaarden R.P.A., Lucassen W.G.H., Brock T.C.M. and Leeuwangh P. (1996) "Effects of the insecticide Dursban® 4E (active ingredient chlorpyrifos) in outdoor experimental ditches: II. Invertebrate

24

7/7. Caquet and F. Ramade

community responses and recovery", Environ. Toxicol Chem., 15, 11431153. Walker B. (1992) "Biological diversity and ecological redundancy", Conserv. BioL, 6,18-23. Walker B. (1995) "Conserving biological diversity through ecosystem resilience", Conserv. BioL, 9, 747-752. Warwick R.M. (1993) "Environmental impact studies on marine communities: Pragmatical considerations", Aust. J. EcoL, 18, 63-80. Weisse T. (1991) "The microbial food web and its sensitivity to eutrophication and contaminant enrichment: A cross-system overview", Int. Rev. ges. HydrobioL, 76, 327-337. Wiens J. (1996) "Coping with variability in environmental impact assessment", Baird D.J., Maltby L., Greig-Smith P.W. and Douben P.E.T. (Eds.), ECOtoxicology: Ecological Dimensions. Ecotoxicology Series 4 (Chapman & Hall, London), pp. 55-70. Woin P. (1998) "Short- and long-term effects of the pyrethroid insecticide fenvalerate on an invertebrate pond community", Ecotoxicol. Environ. Saf., 41,137-156.

PART 1 General environmental issues

This page intentionally left blank

2

Environmental radioactivity measuring methods R. Michel1

Several aspects of environmental radioactivity measurements are surveyed. Starting with the historical development and the manifold goals of the surveillance of environmental radioactivity, the radionuclide coverage needed, the required sensitivities and the materials to be analyzed are discussed. Presently used analytical and measurement methods are briefly reviewed. Emphasis is laid upon new and advanced techniques for both decay measurements and mass spectrometric methods. Because of the importance of environmental radioactivity measurements for regulatory issues, political decision making and for public discussions there is an extreme need for high quality data. Therefore, current issues of quality assurance and control, of international standardization and harmonization are dealt with in some detail. Finally, the relationship between environmental radioactivity measurements and public perception is discussed and possible future developments of environmental radioactivity measurements are outlined.

1. History and goals of environmental radioactivity measurements The history of environmental measurements started with the onset of the nuclear age, in particular with the atmospheric nuclear weapons tests. It was driven by the concern of global radioactive pollution and of human radiation exposure. Compared to other sources, by far the largest amounts of manmade radioactivity were emitted into the environment by the atmospheric nuclear weapons tests. This led to a contamination of atmosphere, hydrosphere, pedosphere and biosphere. The resulting radiation exposure of man reached 8% of the natural exposure in 1963 (UNSCEAR, 1982)2. Only after 1

Zentrum fur Strahlenschutz und Radiookologie, Universitat Hannover, Am Kleinen Felde 30, 30167 Hannover, Germany. 2 References in this paper are exemplary rather than comprehensive. They invite further reading.

28

R. Michel

the Nuclear Test Ban Treaty came into effect did the undesirable situation started to improve. The legacy of the atmospheric explosions declined strongly and presently the resulting individual radiation exposure is less than 10 uSv/a on average (UNSCEAR, 2000). A wealth of information was obtained in the fifties and sixties on environmental radionuclide abundances, radioecological processes and pathways of radionuclides through the environment to man. Systems of environmental radioactivity surveillance were installed in many countries and surveillance measurements, in particular of air and precipitation, became routine. National and international committees evaluated the findings and comprehensive reports exist. Exemplary, as a source of references of the past achievements the series of UNSCEAR reports can be cited; see UNSCEAR (1988) for historical development and UNSCEAR (2000) as the most recent one. With the beginning of the peaceful use of nuclear energy and the installation of nuclear power plants environmental radioactivity measurements acquired a new quality. Because of the enormous radioactive inventories of power reactors the emissions from such installations and the resulting immissions into the environment had to be surveyed even during routine operation. Since such measurements had to be made against the natural radiation background and against the omnipresent fission and activation products of the atmospheric fall-out of the test explosions, the actual status of environmental contamination had to be measured prior to the operation of a nuclear installation followed by a permanent surveillance of both the emissions and the immissions in the surroundings of the plants. Incidents and accidents in the early phase of nuclear power use demonstrated the necessity of nation-wide surveillance of environmental radioactivity. In Europe, the EURATOM Treaty (EURATOM, 1957) made this an international issue and basic guidelines for the protection of the public and of worker against the dangers of ionizing radiation were set (European Commission, 1959). The national authorities accepted the general responsibility to conform with these EURATOM basic radiation protection standards (European Commission, 1959, 1996a), to perform a permanent surveillance of the radioactivity of air, water and soil and to inform the Commission about the results of this surveillance in compliance with articles 35 and 36 of the Treaty (European Commission, 2000b). A multitude of national systems of environmental surveillance was installed and a basis for a world-wide network of such installation was set into effect. Public attitudes towards environmental radioactivity measurements were drastically changed by the Chernobyl accident in 1986. Differing attitudes in dealing with the Chernobyl fall-out in European countries and lack of information from the former USSR demonstrated the necessity of global dissemination of information on environmental radioactivity and of an international harmonization of their evaluation and assessment. Extending the existing surveillance systems, measures were taken in many countries to

2 - Environmental radioactivity measuring methods

29

handle inhomogenous contamination also remote from nuclear installations, e.g. BMU (1986). In the former USSR, large scale environmental radioactivity measurements were necessary in the regions highly contaminated by the Chernobyl accident. Decisions about evacuation of inhabitants and evaluations of the past, present and future radiation exposures had to be made on the basis of these environmental radioactivity measurements which later were internationally evaluated (IAEA, 1991). The consequences of the Chernobyl accident are still followed up and many independent assessments exist, e.g. (IAEA, 1996; NBA, 1995; SSK, 1996), and most recently by UNSCEAR (2000). With the end of the cold war further new tasks of environmental radioactivity measurements became evident which before were frequently kept secret on national levels. The legacy of the Cold War brought clear evidence of large scale contaminations and radioactive remainders; see UNSCEAR (2000) for a survey and references. There were hitherto unknown major nuclear accidents such as the Chelyabinsk events. There was inadequate dumping of nuclear waste on land and in the sea and large contaminated areas due to Uranium mining and milling as e.g. in Ukraine, Kirgisistan, Kazakhstan, France, Germany, and USA. In particular from the military use of nuclear power, a large number of contaminated sites remained and were recognised by the public. These remainders also included in the Western world the vicinities of reprocessing and fabrication plants, e.g. (Wolbarst et al., 1999), and the large-scale contamination of the test grounds of both atmospheric and underground nuclear explosions, e.g. in Kazakhstan (IAEA, 1998a), Bikini (IAEA, 1998b), Muroroa (IAEA, 1998c). In the military context, measurements of environmental radioactivity are also of imminent importance to detect nuclear weapons proliferation using the emissions of test explosions as well as of enrichment and reprocessing plants. These latter measurements need extremely high analytical sensitivity and isotopic accuracy since far-reaching conclusions have to be drawn from faintest traces in a bulk of environmental radioactivity. Also, the question of battlefield surveillance became an additional public issue raising the question whether reliable experimental data can be obtained to describe or exclude enhanced exposure of the public and of soldiers after the use of ammunition containing depleted uranium (Rostker, 2000; Barley et al, 1999). Further tasks for environmental radioactivity measurements in the military field comprised measurements to describe the impact of the loss of nuclear warheads in plane crashes and of radionuclide energy systems of satellites which were set free by their destruction during re-entrance into the Earth's atmosphere. During the more than 4 decades of environmental radioactivity measurements emissions from technologies not related to nuclear power also became issues. They included the routine release of 3H and 226Ra from industry

30

R. Michel

manufacturing fluorescent dials or the inadequate disposal of such dials after World War II. Emissions of radionuclides from nuclear medicine hospitals today are found in municipal waste and waste waters and disposal of exempted radioactive waste in municipal waste storages or waste-burning facilities has to be surveyed. Accidental releases range from undesired melting of highly active nuclear sources in the steel industry, as e.g. in Taiwan (Woushou, 1997), to the dismantling of orphan sources as in the Juarez (Smith, 1984) and Goiana events (IAEA, 1988). Today an increasing number of orphan sources and of illicit trafficking is observed and environmental surveillance has to contribute to detection and recovery of the dangerous material (UNSCEAR, 1982,1988; Lubenau and Yusko, 1998). Surveillance of natural radionuclides in the environment did not have high priority over many years compared to that of man-made radioactivity. There is, however, an increasing interest in such measurements since enhanced exposure to natural radioactivity is receiving the same legal weight as any other radiation exposure. The recent EURATOM basic safety standards (European Commission, 1996a) explicitly state that enhanced exposure to natural radioactivity and radiation is also an issue of radiation protection. In this context the surveillance of technologically enhanced naturally occurring materials, called TENORM or NORM, becomes important. The historic development of the scopes of environmental radioactivity measurements can be summarized as a path through four different categories of measurements. It started from the surveillance of man-made radionuclides in the general territory of a country, went to that of man-made radionuclides at particular sites in the vicinity of (nuclear and other technical) installations, proceeded to the surveillance of natural radionuclides at particular sites and today ends with measurements of natural radioactivity in the general territories of countries. Measurements of environmental radioactivity in all these four categories have multiple goals • to provide a reliable experimental database to estimate human radiation exposure; • to prove the compliance of practices with radiation protection standards and legal exposure limits as well with international non-proliferation agreements; • to record changes in environmental radioactivity and to establish a scientific basis on which human impact can be estimated with the goal to attain a sustainable development; • to understand the behavior of radionuclides in the environment and their pathways to man; • to exploit the potential of radionuclides in the environment as natural and man-made tracers of environmental processes and, last not least; • to give, beyond any doubt, evidence to the public about all facts of environmental radioactivity.

2 - Environmental radioactivity measuring methods

31

This wide variety of different goals needs various types of measurements, an extreme range of analytical techniques and methods with individual sensitivity requirements which will be discussed below.

2. Radionuclide coverage, sensitivity requirements and materials A wide variety of radionuclides has to be analyzed in environmental materials. They cover fission products, transuranium nuclides, activation products and the naturally occurring radionuclides. Though a few hundred different radionuclides are produced by the fission process, just a limited number of radionuclides with sufficiently long half-lives are of radiological significance and have to be covered by measurements of fall-out from atmospheric explosions and in surveillance of test sites and nuclear installations. These are the short-lived radionuclides 89Sr, 95 Zr, 99Mo, 103Ru, 131I, and 133Xe, the medium-lived ones such as 90Sr, 85Kr, 106 Ru, 110mAg, 137Cs, 134Cs, 124Sb, 125Sb, 154Eu, and the long-lived nuclides "Tc, 129I, 135Cs. Of the actinides which are produced in nuclear chain reactors as well as in nuclear explosions, 236U, 238,239/240pu/ 241 Am/ 243£m have to be covered. Activation of structural and shielding materials of fission reactors produces 54Mn, 57Co, 58Co, 60Co, 110mAg and activation of water and ambient air 3H and 14C. In future fusion reactors, 3H and a variety of activation products will be the relevant radionuclides. With respect to the medium- and long-term storage of radioactive waste and also with respect to the disposal of exempted radioactive waste, the surveillance of medium- and long-lived radionuclides will be of importance. Considering radioactive waste, further activation products from accelerators, in particular from spallation neutron sources and possible future accelerator-driven facilities for transmutation of radioactive waste (Bowman et al, 1992) and for energy amplification (Carminati et al., 1993) have to be considered. According to our present knowledge, this adds the long-lived radionuclides 10Be, 26A1, 36C1, 41Ca 44Ti, 53Mn, 55Fe, 60Fe to the list of relevant radionuclides with respect to long-term disposal of radioactive waste. According to the actual human radiation exposure of large populations, today the naturally occurring radionuclides and the natural and man-made variability of the exposure caused by these nuclides becomes of increasing interest. 222Rn and 220Rn and their decay products make up the bulk of natural radiation exposure and comprehensive measuring programs have been performed and are going on all over the world, see e.g. UNSCEAR (2000) for references. But, because of the limited space of this article with this issue. With respect to external and ingestive internal exposure to natural radionuclides from the geosphere, only the nuclides of the 232Th, 235U, and

32

R. Michel

238

U decay chains are relevant. They are of importance because of geologic anomalies, such as the famous monazite areas in India and Brazil or the 226 Ra anomalies in Iran, which may lead to exposure for up to tens of mSv per year and might reach extremes up to 100 mSv/a. Further, they are the radionuclides which occur in technically enhanced naturally occurring radioactive materials which are problematic for the exposure of workers as well as of members of the public. Such materials occur in coal, gas and oil industries, in mining and milling of a variety of metals as well as in the exhaust gas of power plants burning fossil fuels. The use of naturally or technically enhanced radioactive materials as building and construction materials needs radioactivity measurements of the relevant radionuclides. The EURATOM basic safety standards (European Commission, 1996a) made surveillance of these isotopes and of the exposure to such materials a legal requirement. Whilst external exposure can be easily determined since only the "/-emitting radionuclides are relevant, for the description of the internal exposure one has to consider quite a number of radionuclides which have to be individually analyzed since radioactive equilibrium in nutrients cannot be anticipated. Since, moreover, the dose coefficients (ICRP, 1996) of the radionuclides of the natural decay chains show large differences, one needs a reliable estimate of the internal exposure requiring measurements of the individual abundances in foodstuff and drinking water of 238U, 234U, 230Th, 226Ra, 238 210pb, 210p0 from thg

U decay chain, 235U, 231pa, 227Ac, 227Th, and 223Ra

235

from the U decay chain and 232Th, 228Ra, 228Th and 224Ra. For only a few of these nuclides radioactive equilibrium can be presumed in the hydro- and biosphere so that measurements of short-lived daughters is not needed to derive reliable data on the exposure. Such nuclides are e.g.227Th, 223Ra and 224 Ra. As long as enriched or depleted Uranium can be excluded also the separate measurement of 235U can be avoided. If this is, however, not the case, the almost precision and isotopic accuracy is needed to determine all relevant uranium isotopes, namely 234U, 235U, 236U and 238U e.g. for non-proliferation control. 40 K adds only in rare cases to the external exposure of workers by NORM materials from KC1 and NaCl mining. Internal exposure due to 40K is homeostatically controlled depending on the physiological behavior of the essential trace element potassium rather than on its intake. From the other natural radionuclides in the geosphere only 87Rb adds marginally to the human radiation exposure. But, there are no radiological needs for measurements of this radionuclide. The same is true for all other natural primordial radionuclides. From the cosmogenic radionuclides, only 7Be and 14C are of some, but minor, importance with respect to radiation exposure. Due to their relatively minor variability with variations of the galactic cosmic ray intensity over the solar 11-years cycle, their measurement is only of interest to

2 - Environmental radioactivity measuring methods

33

establish natural baseline data. The anthropogenic changes of 14C will be discussed later. Though technical or medical applications of radioactivity do not add further nuclides, these applications are also targets of environmental surveillance. Radionuclides such as 131I from nuclear-medical therapy, 232Th from welding electrodes, 60Co, 137Cs and 226Ra from radionuclide sources, 235U, 236 U, 238U, 239Pu, and 241Am from the technical and military use of depleted uranium which even may come from reprocessed nuclear fuel and may consequently contain some impurities of transuranium elements (Rostker, 2000), and 3H and 226Ra from luminescent dials have to be taken into account. Actual investigations or surveillance programs usually cover purposedependent subsets of all these radionuclides. Due to the multitude of purposes, however, for all these radionuclides measuring methods have to be available which have to satisfy adequate and reasonable sensitivity requirements. When defining the sensitivity requirements of environmental radioactivity measurements one has to distinguish two extremes of purpose of such measurements, namely measurements which allow determination of the exact present abundance of a nuclide in a particular environmental material, on the one hand, and measurements which have the goal to determine only those concentrations which have a defined radiological significance, on the other. These extremes can be categorized as scientific and surveillance measurements respectively. Measurements to be performed for the purpose of non-proliferation control will always fall into the category of scientific measurements. The radiological significance of an environmental radionuclide concentration is usually expressed in terms of an expected exposure due to this nuclide and the requirements are defined as required detection limits derived from radioecological models which connect the expected exposure with the concentration of the radionuclide in question. In the European reports on environmental radioactivity in the European Community, e.g. (European Commission, 1996b), the required detection limits are related to reporting levels for the surveillance of air, surface water, drinking water, milk and mixed total diet. These reporting levels RL are defined as

with DL: RF: EDO. CF:

annual dose limit, taken in accordance with ICRP (1991) to be 1 mSv; reduction factor of the dose limit, taken to be 1000; effective dose coefficient according to ICRP (1996); annual consumption per person.

Using the thus defined reporting levels, man-made radionuclides are surveyed at an environmental level which causes a risk of stochastic effects as low as 5 x 10~8 a"1 (European Commission, 1996b).

34

P. Michel

The sensitivity requirements of measurements of the actually existing radionuclide concentrations are, at first glance, self-defined. However, they are reasonable only if naturally occurring radionuclides are to be measured because of their omnipresence. The actual concentrations of man-made radionuclides presently are so low in some environmental materials that it needs the most advantaged measuring methods to detect them. Radionuclide concentrations in air can be taken as an example here where the activity concentrations of 60Co, 90Sr, 131I and 137Cs are nearly as low as 10~7 Bq/m3 and about equal to those of cosmogenic 22Na. Naturally occurring radionuclides in air such as 40K, 210Pb and 7Be are two, three and four orders of magnitude more abundant in air than 22Na respectively. Concentrations of 239Pu and 240Pu are even two orders of magnitude lower than those of 60Co, 90Sr, 131 I and 137Cs. In view of such low concentrations, measurements of manmade radionuclides at their actual levels appear not to be reasonable except for scientific purposes. Circumstances, however, which justify performing such measurements also in the course of special surveillance measurements will become evident when dealing with the public opinion about radioactivity in the environment below. They are self-evident in the case of non-proliferation control. The extreme range of radionuclide concentrations to be covered by environmental radioactivity measurements becomes even more evident if one considers that the activity concentrations in air during the Chernobyl accident were between 1 and 100 Bq/m3 in Western Europe and about equal to the activity concentrations in air of Radon decay products in Uranium-rich granite areas such as in the Massif Central in France. Considering further that the Chernobyl accident did not cause a radiologically relevant enhancement of the radiation exposure in Western Europe and that measurements of environmental radioactivity also have to be applicable to accidental situations, where decisions about sheltering, evacuation and performance of thyroid blocking by administration of stable iodine have to be made, the orders of magnitude to be covered is further increased. The cited European report (European Commission, 1996b) and the organization of environmental radioactivity measurements according to the obligations of articles 35 and 36 of the EURATOM Treaty exactly distinguish the two extreme views of environmental radioactivity measurements by the organization of different measuring networks, namely a dense and a sparse network. The dense network spans over entire Europe to give sufficient information to detect and to quantify any significant radiation exposures due to man-made radionuclides on a local scale. The sparse network divides the present European Community in only 32 regions for each of which just one location is estimated to be sufficient to perform in-depth measurements of the actual concentrations of some selected radionuclides to describe longterm trends of environmental radioactivity. Reporting levels according to equation (1) are used and defined for the surveillance of environmental radioactivity in the European Community

2 - Environmental radioactivity measuring methods

35

only for artificial radionuclides. The requirements of the EU basic standards (European Commission, 1996a) with respect to exposure to natural radioactivity will cause the necessity to define analogous required detection limits also for natural radionuclides. Examples for this already exist, for instance, in Germany where required detection limits are defined for the surveillance of the WISMUT remainders and for the remediation measures as a consequence of Uranium mining and milling (BMU, 1997,1999). There are, however, some general problems with the definition of required detection limits to be internationally applicable. First of all, the term detection limit has to be identically defined in the different countries. There is, however, no definition of the detection limit given in the cited EU Report (European Commission, 1996b) and looking e.g. at the EU Drinking Water Guideline (European Commission, 1998, 2000a) even contradictory definitions are given. We will discuss this point below when dealing with quality of measurements. Second, the required detection limits depend on the radioecological model used to connect radionuclide concentrations with dose. Presently used models, however, contain different degrees of conservative assumptions and thus make it difficult, if not impossible, to compare the required detection limits defined by different authorities. Third, the required detection limits depend on assumed consumption habits which differ from one country to another and which may contain, again, differing conservative assumptions. To overcome these difficulties in the international context will be a painstaking, but necessary work for the future. Moreover, there remains a historical distinction in perceiving the exposure to manmade and natural radionuclides, namely to use different measures for them. I shall give an example for the need of harmonization. In Germany, the basis of the required detection limits for surveillance of man-made radionuclides according to the Act on the Precautionary Protection of the Population against Radiation Exposure (BMU, 1986) and to article 35 of the EURATOM Treaty uses a guideline value of the annual effective dose of 1 (iSv which is in accordance with an annual dose limit DL of 1 mSv and a reduction factor RF of 1000 according to equation (1). For the surveillance of natural radionuclides from the WISMUT remainders (BMU, 1997), and for the surveillance of nuclear installations according to the Radiation Protection Ordinance (BMU, 1995a, 1995b) guideline values of the annual effective dose of 10 uSv are applied which are equivalent to a reduction factor RF of only 100. Given the amount of work and the costs induced by required detection limits, the harmonization of rules is essential for the surveillance of both man-made and natural radionuclides. An additional complication arises from the fact that in decommissioning of nuclear installation and exemption of contaminated or activated materials the expected radiation exposures of members of the public should not exceed a few time 10 [oSv per year (UNSCEAR, 2000). Surveillance of such activities and of their consequences for environmental radioactivity measurements will need even more stringent sensitivity requirements.

36

R. Michel

As a final point of this chapter, the materials to be analyzed and the frequency by which these materials have to be surveyed shall be shortly discussed. As wide the variety of purposes of environmental radioactivity is, as differing are the requirements for materials to be analyzed, sample numbers and sampling frequencies. Given the costs induced in each actual investigation or surveillance program these requirements have to be scrutinized. If the ultimate purpose is the determination of radiation exposure there are different ways to address this problem. The main distinction of approaches is whether they make use of reference persons or of really existing ones representing a critical group. Whilst the approach using reference persons has been successfully applied in planning of practices and installations it appears to be hardly feasible to determine the actual radiation exposure arising from existing installations realistically by it. Even, if the results of environmental radioactivity measurements of a wide variety of foodstuffs is used as input to respective model calculations using e.g. models as ECOSYS (Muller and Prohl, 1993) only exposure averages of normally behaving persons can be determined with some confidence. The problem rests then with non-normal behavior of certain individuals and with their lifestyle and consumption habits. Even if one excludes extremes or pathological behaviour large uncertainties remain for the determination of individual exposure. If averages of the exposure are sufficient, the use of reference persons is adequate. The cited EU report (European Commission, 1996b) makes use of this reference-person concept with a minimum coverage of materials to be analyzed, namely air, drinking water and mixed total diet. Another approach in the context of the exposure of a reference person is the use of aggregated transfer factors in a fall-out situation based on the analysis of soil and the determination of deposition densities. Thus, by a minimum of materials analyzed, the exposure can be estimated. In this case, the radioecological model used to determine the aggregated transfer factors is decisive and again the extent to which conservative assumptions are included. Individual and realistic exposures can only be determined using the critical group concept evaluating the actual exposure of actually existing people. To determine the effective annual dose E(t1) for a year t1, this requires to quantify its contributions due to external exposure Eext and due to ingestion and inhalation of radionuclides, E- and Einh. According to ICRP 60 (ICRP, 1991) or the EURATOM Basic Safety Standard (European Commission, 1996a) such an evaluation has to be performed for the different tissues T taking into account the age dependence of exposure and of dose factors

Frequently, it is convenient to distinguish for the ingestion exposure pathway, consumption of drinking water and of nutrients so that the annual

2 - Environmental radioactivity measuring methods

37

effective dose for a tissue is given by with ET(t1): ET ext(t1):

annual effective dose in tissue or organ T during the year t1 in Sv. annual effective dose in tissue or organ T due to external exposure to airborne radionuclides and to radionuclides in or deposited on the ground during the year t1 in Sv; ET, inh (t1): annual effective dose in tissue or organ T durch Inhalation during the year t1 in Sv; ET, ing, w (t 1 ): annual effective dose in tissue or organ T due to consumption of drinking water during the year t1 in Sv; ET, ing, n (t 1 ): annual effective dose in tissue or organ T due to consumption of nutrients during the year t1 in Sv. The external exposure due to Y-radiation is given by

with Hx(t) ambient y dose rate as a function of time t for the actual locations. Analogously, the exposure of the skin due to (3-immersion of airborne radioactivity can be evaluated. The internal exposure due to inhalation is given by

with r: Ar(fg):

radionuclide in question; activity concentration of the radionuclide r in air during a year t0 in Bq • nT3; V(t$): age dependent breathing rate in m3 • s"1;

&inh r T a8e dependent dose factor for tissue or organ T due to inhalation of a radionuclide r in Sv • Bq"1. according to ICRP (1996). The internal exposure due to consumption of drinking water is given by

with C^(f0): time dependent activity concentration in drinking water of a radionuclide r during the year £Q in Bq • I"1; w U (to): age dependent consumption of drinking water in 1; g- r T: age dependent dose factor for tissue or organ T due to ingestion of a radionuclide r in Sv • Bq"1 according to ICRP (1996).

38

R. Michel The internal exposure due to consumption of nutrients is given by

with Cr(to): time dependent activity concentration in a nutrient n of a radionuclide r during the year tQ in Bq • kg"1; n U (t0): age dependent consumption of a nutrient n in kg. For an determination of the dose as realistic as possible, integrals over the time differentials of the dose factors are used in equations (5—7) to account properly for the build-up of the exposure due to long-lived radionuclides with long biological half-lives as they occur e.g. in the natural decay series. In the latter cases a conventional calculation of internal exposure, e.g. for ingestion in equation (8) on the basis of 50- or 70-years dose factors g{ r T according to ICRP (1996), would strongly overestimate the dose committed between years t0 and t^.

For all radionuclides with short physical or biological half-lives, however, equation (8) also holds for a realistic assessment of the committed dose and for them also equations (5) and (6) can be simplified accordingly. The determination according to equations (2-7) of an actual exposure of an individual member of the public is extremely costly and time-consuming. The situation becomes even worse if the distribution of effective doses in a population as well as the expectation value of such distributions shall be determined by the approach described here. Therefore, it is mainly used in the context of the critical group concept. Here, the recent investigation of exposure from emissions from La Hague (Nord-Cotentin Radioecology Group, 1999, 2000) can be taken as an outstanding recent example. The difficulty is, however, to identify the group at risk and it may only be possible by chance to find it. For an assessment of human exposure, environmental radioactivity measurements have to provide a complete and reliable data-set of the concentrations of natural and man-made radionuclides in air, water, soil and nutrients. If extremely low exposures are excluded on the basis of a de minimis concept certain radionuclide concentrations can be excluded as not relevant. There are no de minimis if environmental radioactivity measurements related to proliferation control are to be considered. Such measurements have to distinguish faintest traces of elemental and/or isotopic patterns hidden under the bulk of environmental radioactivity. For them, methodologies

2 - Environmental radioactivity measuring methods

39

for ultra-trace analysis at the pg- and fg-levels have to be applied which comprise the use of clean rooms and exclusively highest-purity reagents for sample preparations and chemistry. Even then, they can only be successful if the almost blank and contamination control is performed. It is, however, a particular advantage for such measurements that unambiguous isotopic signatures of uranium and transuranium elements allow to distinguish the relevant components within the ubiquitous environmental radioactivity. Comparable analogues from other fields of sciences or applications can be found in geo- and cosmochemistry where analyses by isotopic dilution are common and mass spectrometric analysis of long-lived radionuclides and of actinides in radioactive disequilibrium are used for dating purposes. Just in rare cases, e.g. if claims of particular exposures to transuranium radionuclides emitted from nuclear installations have to be investigated against the background of nuclear weapons fall-out, such analyses are also necessary in the context of the determination of human radiation exposure.

3. Methods: From counting decays to counting atoms During five decades of environmental radioactivity measurements an outstanding evolution of measuring methods took place. Decay counting evolved from single channel counting over coincidence techniques with just a few counting channels to more or less sophisticated a-, (3-, and y-spectrometry with systems coupled to multi-channel analyzers. In parallel, the evolution of mass spectrometric methods made it possible to measure the low isotopic abundances of radionuclides in the presence of stable isotopes or isobars many orders of magnitude more abundant than the radionuclides themselves. Also general trends of technology development contributed to the improvement of methods as did the increasing computerization and miniaturization which made it feasible to handle huge amounts of data and to use the most sophisticated methods for their evaluation. The evolution of analytical- and radio-chemistry strongly influenced the methodology of environmental radioactivity measurement. Use of clean rooms, high-purity reagents, strict contamination and blank control and impressive advances of chromatographic methods have to be mentioned here. Here, the old and today partially obsolete methods will just shortly be mentioned. Rather, the present state of the art will be discussed in this chapter emphasizing some outstanding developments. Lacking more refined methods, the onset of decay counting made use of simple measurands such as gross alpha and gross beta activity concentrations in the various environmental materials to be analyzed. Large volume counters were developed with strong low-activity passive shielding and with active suppression of the cosmic ray background by coincidence and

40

R. Michel

anti-coincidence shields. These low-level counters developed into various directions by the improvement of electronics in particular of coincidence techniques. Measurands such as residual beta activity concentration allowed coarse discrimination of radiation from artificial radionuclides against that from natural radionuclides from the decay series. Measurements of gross beta activity (as 90Sr-equivalent) or gross alpha activity (as Pu-equivalent) are still frequently performed though no longer state-of-the-art. Today, many of these techniques and their respective measurands are obsolete. The only justification of their application is in measurements which simply are intended to exclude the presence of any relevant a- or (3-activity. Even if they are used for such purposes as prescribed for instance in the annex of the new drinking water guideline (European Commission, 1998, 2000a) more advanced nuclide-specific methods have to be used if certain limits which are considered as normal or harmless are exceeded. Having in mind the main goal to quantify the human exposure by measurements of environmental radioactivity and considering the enormous differences in dose coefficients for different a- or (3-decaying radionuclides, measurands such as gross alpha or gross beta or residual beta activity concentrations do not satisfy any need of an adequate dose assessment. Today, nuclide specific data are needed in any case and they can be obtained from decay measurements either by single or few channel counting measurements after chemical separation of a particular radionuclide or by spectrometric measurements. Thus, counting measurements are still relevant for the analysis of certain radionuclides such as 3H, 14C, and 90Sr though also in these cases other methods are strongly competing and even superior. For the measurement of (3-decaying radionuclides the development of liquid scintillation spectrometric methods was the break-through to nuclide-specific measurements. Liquid scintillation methods evolved from single- or few-channel counting systems into multi-channel spectrometric ones and find widespread application. This applies also to the analysis of a-emitting radionuclides. Alpha liquid scintillation spectrometry with (3/y discrimination has become a quick, reliable and low cost method with high sensitivity (~ 0.1 mBq). For a survey on recent developments of liquid scintillation spectrometry see Cook et al. (1996). For a- and y-spectrometry it was the development of solid state semiconductor detectors made of silicon or germanium which opened the field of nuclide-specific measurements, a-spectrometry with Si detectors still needs highly sophisticated chemical separation schemes to extract the radionuclides from the matrix and to prepare thin samples to avoid self-absorption and strong peak tailing due to stopping of the a-particles in the sample. In contrast, y-spectrometry opened the field of non-destructive measurements of a wide range of relevant radionuclides. Just for completeness it should be mentioned that for y-radiation the evolution from counting to spectrometry passed the phase of solid state scintillation detectors having high counting efficiency but poor energy resolution.

2 - Environmental radioactivity measuring methods

41

Though the resolution of plastic or Nal(Tl) scintillation detectors is much worse than that of semiconductor detectors, they have maintained importance for certain applications such as measurements of ambient dose rates, distinction of natural and artificial radionuclides in field measurements and in spectrometric systems where they serve in combination with semi-conductor detectors as active shields or for the suppression of the Compton background. Also coincidence techniques which were successful as low-level counters using y-y-coincidence or triple-y-coincidence techniques (Herpers et a/., 1969) developed into highly sophisticated y-y-coincidence spectrometry, e.g. (Perkins et al., 1970), which can be regarded as one extreme of lowlevel y-spectrometry and which today are used in quite a number of laboratories. The development of y-spectrometric methods was strongly influenced by three aspects. First, the availability of intrinsically pure germanium allowed improvements of energy resolution and the construction of large volume germanium detectors which reached the same efficiency as Nal(Tl) detectors and of dedicated detectors optimized for the measurement of low-energy Xor y-radiation or of high-energy y-radiation. The improvements of detector size and of specialized design is, however, accompanied by complications in the evaluation of y-spectra which may need corrections for systematic sum coincidences and complicated methods of efficiency calibration. For a general discussion of related problems see, for instance, Debertin and Helmer (1988). Second, the improvement of detector shielding and of the reduction of natural radioactivity of the detector itself and the surrounding components. The improvement of detector shielding included the use of low-activity lead shields, inner lining of the lead shields by other metals to reduce cosmic ray neutrons and to shield the radiation from their capture y-rays and from the y-rays of 210Pb inside the lead shield. Methods of reducing y-background from Radon decay products inside the shield range from the reduction of the free volume inside the shield by e.g. paraffin, to float the interior by absolutely filtered air or by high-purity nitrogen. Considerable improvements were made over the years due to the development of material sciences in particular with respect to the production of high-purity material which allowed to select and use low-activity components only and to construct special low-level detectors. Third, the application of active shields and of electronic components which make use of coincidence and anti-coincidence techniques allowed reduction of the Compton background and, to a certain degree, also of the cosmic ray background. The presently final and decisive step of low-level yspectrometry is, however, the combination of all these methods described above and to go with the laboratory into the deep underground. This is carried out to reduce the hadronic component of the cosmic radiation and to avoid both the continuum of cosmic ray background and the 74Ge(n,n'y) and 72 Ge(n,n'y) background. Compared to an unshielded detector at sea level,

42

R. Michel

the background of a y-spectrometer can be reduced by three orders of magnitude and more in a deep-underground laboratory by application of these combined methods. For a detailed discussion of these aspects see for instance Heusser, (1995). A contrary approach of measuring environmental y-emitting radionuclides is to use no shielding at all and to directly assess the abundances of radionuclides by field measurements via in situ y-spectrometry. With these technological developments, the state of the art of radioactivity measurements can be described as follows. For all y-emitting radionuclides yspectrometry with Ge-detector is the method of choice. For the analysis by yspectrometry of radionuclides from the natural decay chains, however, a careful selection of y-energies used for evaluation and the consideration of and correction for interfering y-lines is important, e.g. (Schkade et at, 1999). Moreover, it is essential to attain radioactive equilibrium in the samples to be measured since many radionuclides are determined via the y-rays of their decay products. This holds for 238U, 226Ra, 227Ac, and 228Th which are determined via the y-rays of 234Th or 234mPa, by 214Pb or 214Bi, by 228Ac, and by 224Ra, 212 Pb or 208T1, respectively. Further, measurements with different detectors (high- and low-energy) are partially needed, e.g. to obtain reliable results for 210 Pb. For the low-energy y-radiation of this nuclide the determination of the counting efficiency in large volume samples is difficult and may need an additional measurement with a small sample by a low-energy detector. For radionuclides emitting exclusively a- or p-radiation non-destructive methods cannot be applied and chemical or physical-chemical separation from the matrix are needed combined with skilful procedures for the preparation of the radionuclide samples to be measured. I will discuss the required separation techniques only in some special cases. Otherwise, I just mention some references from a broad literature of evaluated separation schemes available and prescribed by national authorities for the respective analyses (AFNOR, 2000; AKU, 1999; BMU, 1998; DoE, 1979). These evaluated schemes mostly make use of well established chemical techniques and usually recognize new analytical developments only at late stages of maturity. New developments and techniques such as modern chromatographic methods as e.g. capillary electrophoresis, ionic chromatography, and HPLC are scarcely found in these schemes today. Just becoming state-of-the-art in scientific investigations of environmental radioactivity and in proliferation control, modern chromatographic technique provide a wealth of possible improvements in particular with respect to automation of separations and to suppression of interferences. Independent of the application of modern or old-fashioned chemical techniques, radiochemical separations remain the prerequisites for the determination of a- and p-decaying radionuclides by consequent counting or spectrometric methods. For 14C, the analyses generally comprise the separation as CO2 followed by measurement in gas-filled counters, by liquid scintillation counting or spectrometry or by accelerator mass spectrometry which will be discussed

2 - Environmental radioactivity measuring methods

43

below in some detail. For 85Kr counting techniques using gas-filled counting tubes into which the Krypton extracted from the air is included is still the method of choice. For 3H, 89Sr, 90Sr, and "Tc, liquid scintillation counting or spectrometry is the method of choice after separation or enrichment though counting measurements with gas-filled detectors still remain a practicable option. A particular problem arises in case of the analysis of 90Sr (and 89Sr). Here, all counting techniques have the disadvantage that after separation the build-up of the 90Sr decay product 90Y (T 1/2 = 64.1 h) has to be waited for. The delay of more than one week between chemical separation and counting is not acceptable in the case of a nuclear accident where the decision about the restriction of consumption of certain foodstuffs depends on information about the fall-out of 90Sr. Considerable efforts have been undertaken to develop fast methods for the determination of 90Sr in case of nuclear emergencies when it is also required to deal with the counting-interfering presence of 89Sr (T1/2 = 50.5 d). To accelerate the chemical separation of 90Sr advantage has been taken from the development of special extraction reagents such as crown-ethers which, however, do not decrease the waiting time between separation and counting (Tait and Wiechen, 1997). Except for the mass spectrometric methods discussed below, liquid scintillation B-spectrometry can be applied to solve this remaining problem. |3-spectrometry allows to distinguish 90Sr, 90Y and 89Sr so that within less than two days after sampling a decision measurement can be made as to what degree consumption decisions are needed (Filfi et al, 1998). p-decaying radionuclides of the natural decay series generally do not pose a problem since they all can be analyzed by y-spectrometry of their decay products though in some cases chemical separations have to precede the measurements. Counting and spectrometric methods, even in combination with chemical separations, come to their limits if long-lived radionuclides with halflives exceeding 1000 years have to be measured. In principle, it would be much more advantageous to count the atoms of those radionuclides instead of waiting for their decay. Thus, mass spectrometric methods are the obvious choice. However, conventional mass spectrometry was and still is not able to detect the spurious amounts of long-lived radionuclides in the presence of stable isotopes or isobars which might be more abundant by more than 10 orders of magnitude. In addition, conventional mass spectrometry has to struggle with molecular interferences which may mimic the presence of a radionuclide. Therefore, conventional mass spectrometry in form of classical ICP-MS has been applied successfully to environmental radioactivity measurements only in case of uranium and thorium where the problem is not the isotopic analysis but the analytical blank. Only recently, new developments in mass spectrometry have opened up a new world of environmental radioactivity measurements. Extreme sensitivity

44

R. Michel

and suppression of isotopic and isobaric interferences were successfully realized by accelerator mass spectrometry (AMS), by multi-collector magnetic sector field ICP-MS and by laser resonance ionisation mass spectrometry (RIMS). In case of AMS and ICP-MS, environmental radioactivity measurements have widely profited from analytical needs of geo- and cosmochemistry and still today the analytical capabilities used in these fields of sciences are not yet fully exploited for environmental radioactivity measurements. In accelerator mass spectrometry (AMS), the problem of isobaric, isotopic and molecular interferences can be avoided for long-lived radionuclides such as 10Be, 14C, 26A1, 36C1,41Ca, 53Mn, 59Ni, 60Fe, 90Sr, 99Tc, 129I, and 236 U. Also for AMS, the radionuclide in question has to be chemically separated with or without addition of a stable isotopic carrier and a sample for measurement has to be prepared. The AMS measurement is then performed with a nuclear physics accelerator mostly a tandem accelerator with a few MV terminal voltage. From the sample, negative molecular or atomic ions are produced in a Cs sputter ion source. The extracted ions are analyzed by magnetic and electric deflection and injected into the accelerator where they are accelerated towards the HV terminal. At the terminal a thin gas target or a thin stripper foil transforms the negative ions into positive ones and, at the same time, destroys any molecules in the beam. The positive ions are accelerated again to ground potential. When leaving the accelerator a suitable charge state of the ions is selected and by beam-forming electric and magnetic analyzers isobaric interferences are suppressed. The high energies of the ions now allow to make use of detection methods of nuclear physics which distinguish Z and A of the detected particles. Applying time-of-flight methods and AE-E detectors isotopic interferences are suppressed and isotopic ratios of the radionuclide and a neighboring isotope of down to 10~15 can be measured. For a detailed review and recent achievements see Finkel and Suter (1993) and the proceedings of the last AMS conference in Kutschera et al. (2000). Presently, AMS measurements are performed at a considerable number of AMS facilities all around the world. The disadvantage of AMS using huge accelerator systems decreased during recent years and dedicated smaller machines have been constructed. Just recently, an extremely small AMS machine has been developed (Suter et al., 2000) so that a future of AMS as a widespread laboratory method comes into view. AMS was successfully applied to the measurement of long-lived cosmogenic nuclides in terrestrial and extraterrestrial matter; see Michel (1999) for a review. With respect to the analysis of man-made radionuclides in the environment, its capabilities are just beginning to be fully exploited. For the radionuclides 14C, 36C1, and 129I, AMS offers the possibility to investigate the long-term impact of the peaceful and military use of nuclear fission and fusion. In the case of 14C, the activities of modern carbon can also be analyzed by low-level counting methods and the question which method is

2 - Environmental radioactivity measuring methods

45

advantageous is decided by the time needed for analysis and how low 14 C/C ratios are to be covered. In using 14C as an environmental tracer, lowlevel counting quickly comes to its limits and AMS is the method of choice. It should be emphasized that AMS is also the method of choice for all the long-lived radionuclides produced as activation products at accelerators, spallation neutron sources and possible future accelerator-driven devices. In this context, also recent and ongoing measurements of long-lived activation products in materials from Hiroshima and Nagasaki have to be mentioned which are measurable by AMS with the purpose to resolve still existing discrepancies in the evaluation of the contributions of neutrons and y-radiation to the exposure resulting from the explosions of nuclear weapons at Hiroshima and Nagasaki (Riihm et ol., 1998). To cover both natural and man-made abundances of 36C1 and 129I only AMS is capable to cover the entire range of isotopic ratios observed in nature. 36C1 showed an impressive bomb peak up to three orders of magnitude above normal, which, by now, has nearly decayed to normal ratios, the bomb produced 36C1 being diluted by marine Cl and washed out of the atmosphere (Synal et al, 1990, 1997). Also the activities of 14C which in the beginning of the sixties went up to 200 percent of the activity of modern prenuclear carbon has nearly returned to normal values, the biospheric response becoming normal on a time scale of 15 years (Levin et al., 1997; Nakamura et al, 1987). The situation is more complicated for the long-lived radionuclide 129I (T-t/2 = 15.7 Ma). Since the middle of our century the environmental levels of 129 I with 129I/127I ratios of ~10~12 have been dramatically changed as a consequence of the civil and military use of nuclear fission; see Schmidt et al. (1998) for a review. From the radioecological point of view, our knowledge about 129I in the environment is still marginal. In 1962, radiochemical neutron activation analysis (RNAA) became available as a first analytical method to determine 129I in environmental samples (Studier et al., 1962). Manifold analyses demonstrated the extreme changes as a consequence of atmospheric weapon tests and even more pronounced due to releases from reprocessing plants. But, as discussed by Schmidt et al. (1998) in detail, RNAA is only capable of measuring I29i/127l ratios above 10~10 and the natural abundances and their transition to modern high contamination levels could not be quantified. Only by accelerator mass spectrometry did it become possible to determine all 129I/127I ratios occurring in nature. Moreover, many of the early analyses suffered from methodological deficits such as missing blank analyses and quality control. Although the number of 129I investigations increased dramatically in recent years, the radioecology of 129I is still insufficiently known. We therefore started a systematic investigation to establish reliable analytical for 129I and 127I analysis in various environmental materials and to close some gaps in our knowledge about the environmental abundances of 129I, its radioecological behavior and to exploit the potential of 129I for the retrospective dosimetry of 131I exposure

46

R. Michel

after a nuclear accident such as the Chernobyl accident protocols (Schmidt et al, 1998; Szidat et al, 1999,2000a, 2000b). The results demonstrate that the natural abundances of 129I have been sustainably changed and that there is a continuing increase of anthropogenic 129 I in the pedosphere and the terrestrial biosphere in Europe. Though the environmental 129I levels are not of radiological concern at present, the future development should be carefully monitored. Generally, 129I, together with 14C and 36C1, can be regarded as excellent tools to quantify the longterm impact of nuclear power on the environment. It has to be mentioned that in situations with high 129I releases simpler methods as AMS and RNAA can be sufficiently sensitive to determine environmental 129I abundances. So, Frechou et al. (2000) demonstrated that direct y-X spectrometric measurements are possible e.g. in the vicinity of the La Hague reprocessing plant. These authors also developed new biospheric reference materials for the determination of 129I and validated it using three independent methods direct y-X spectrometric measurements, RNAA and AMS. Tait et al. (2000) showed that fast chemical separations with subsequent y-spectrometric measurements by low-energy Ge-detectors can be also sufficiently sensitive to reach the required detection limits for the surveillance of radioactive waste depositories in case of accidental releases. The second new-frontier mass spectrometric method for environmental radioactivity measurements is ICP-MS. It has been successfully used for many years for trace element analysis. Also for ICP-MS, radiochemical separations have to be applied to avoid elemental interferences and a careful control respectively determination of the chemical yield is necessary. The sensitivity is in the fg-range, but molecular and isobaric and isotopic interferences remained crucial. Recently, however, ICP-MS acquired a new quality which probably will be of great impact on environmental radioactivity measurements. With multiple collector ICP magnetic sector field mass spectrometry (Halliday et al., 1995) the problems of classical ICP-MS was widely overcome and at the same time the sensitivity was still significantly improved. In geo- and cosmochemistry, this allowed to exploit radioactive disequilibria for dating purposes with an unforeseen accuracy (Halliday et al., 1998). Disequilbrium dating using the uranium and thorium decay series, e.g. (Stirling et al., 2000), are closely related to environmental radioactivity measurements and the potential of the new technique for environmental radioactivity measurements is striking. High precision actinide isotopic analyses and measurements of transuranium elements, as e.g. plutonium in materials with extreme Pu/U ratios below 10~14, are essential for migration studies and for proliferation control. Measurements by the new ICP-MS techniques will also strongly facilitate comprehensive surveillance of long-lived radionuclides from the natural decay series in water and foodstuffs with the purpose of a better understanding of the pathways of the heavy radioelements to man and of the resulting radiation exposure. The new ICP-MS technique has, moreover, a large potential for automation by direct coupling of the analyti-

2 - Environmental radioactivity measuring methods

47

cal detection system with the radiochemical separation using capillary electrophoresis, ionic chromatography, HPLC. In view of the trend to use mass spectrometric methods in environmental radioactivity measurements, resonance ionisation mass spectrometry (RIMS) has to be mentioned as a third outstanding development which becomes increasingly important. For this method the carrier-free element is chemically separated and measured by a RIMS set-up in which a time-of-flight mass spectrometer is fed by ions from a laser ion source. The suppression of undesired particles in the mass spectrometer is achieved by ionizing only atoms of the desired element respectively isotope by multiple-resonant laser excitation and ionisation. The new method has an extremely high element sensitivity and is well suited for trace analysis of long-lived radionuclides in environmental, biological and technical samples. It has the same superiority compared to counting measurements as discussed above for AMS and ICPMS. RIMS has been successfully used to analyze 41Ca, 89Sr, 90Sr, 99Tc, and Puisotopes at environmental abundances and it is applicable also for the analysis of Cm, Bk and Cf (Wendt et al, 1997a, 1997b, 1999; Erdmann et al., 1997; Eichler et al., 1997; Miiller et al, 2000; Passler et al, 1997). A special advantage of RIMS plutonium isotopes is that the isotopic composition of all isotopes can be determined with an unprecedented accuracy with detection limits down to 106 atoms per sample.

4. Quality: Not just a matter of taste It cannot be taken for granted that every measurement result has high quality and is reliable. A permanent struggle for precision and accuracy is characteristic for any experimental research. But, erroneous, low-quality, false or even falsified data do exist and wrong interpretation of even reliable results adds to confusion in scientific and public discussions. In view of the severe impact of environmental radioactivity data on the public opinion and on society in general, proof of quality and reliability is a most important issue and a basic requirement for measurements of environmental radioactivity. This requires a sophisticated system of quality assurance and control which shall shortly be discussed here. A first requirement is that the purpose of a measurement and thereby the measurand are unambiguously defined. There is a difference whether the measurand is the activity concentration of a particular sample or whether it is the activity concentration of a particular environmental material. Whilst in the first case the measurand can be assumed to be well defined in the second case the question whether a sample or a set of sample is representative for the material in question is a difficult problem and usually the measurand in this case is not well defined.

48

R. Michel

The quality of a measurement starts with sampling strategy, sampling methods and their documentation. The choice of the measurand and the sampling strategy are influenced by the purpose of the measurements, see Gilbert (1987) for some discussion of the general problems of sampling. It can even be the case that it is not possible to define the measurand exactly and that only distributions of results of measurements can be used to attack the problem in question. As an example, the frequently observed log-normal distribution of activity concentrations in environmental materials must not necessarily be a consequence of unreliable measurements. It represents in the optimum case of reliable measurements of individual samples the distribution of the true values of the measurands in the samples of an environmental compartment. This effect is e.g. caused by input and output probabilities for a compartment which are linearly normal distributed. On the other hand, such a distribution can be caused by insufficient control of analytical blanks and by systematic influences on the measurements which are not properly corrected for. Therefore, to judge about the question whether a distribution of results is an experimental artifact or whether it represents the natural conditions can only be decided by a thorough investigation and determination of the uncertainties of the measurement. With the ISO Guide for the Expression of Uncertainty in Measurement (ISO, 1995) an internationally accepted unified approach to the determination of experimental uncertainties does exist. In contrast to still frequent practice it requires to take into consideration all sources of uncertainty. In particular, it distinguishes between Type A and Type B uncertainties. Type A uncertainties are those which can be evaluated by the experimentalist by repeated or counting measurements and which can be expressed as standard deviations. Type B uncertainties are those which cannot be evaluated by the experimentalist themself but which can be obtained them only from other sources. Type B uncertainties are, for instance, reported in literature or stated in certificates, e.g. for calibration standards and nuclear decay data used. While Type A uncertainties can be evaluated by conventional statistics, Type B ones cannot. Using, however, Bayesian statistics also Type B uncertainties can be expressed as standard deviations and can be propagated to combined standard uncertainties associated with a measurement result as the best estimate of the true value of the measurand (Weise and Woger, 1993, 1999). In this case, also a probability distribution for the true value of the measurand can be given which makes use of any existing information and which can be improved if further information emerges. It is a conditio sine qua non that for any measurement an analysis of the experimental uncertainties has to be performed. A result without a statement of the combined standard uncertainty associated with it and with a list of all considered sources of uncertainties is no result and worthless. Considering, however, the determination of the human internal radiation exposure there are still some unresolved issues since there are no uncertainties of dose factors available. Here further investigations are urgently needed.

2 - Environmental radioactivity measuring methods

49

If experimental uncertainties are known according to the ISO Guide also the problem of an unambiguous definition of decision thresholds, detection limits and limits of the confidence interval (named characteristic limits in short) can then be made (Weise, 1998). In an upcoming ISO standard (ISO, 2000a) the determination of characteristic limits will be internationally laid down. A national German standard on this issue is already published (DIN, 2000). The approach to the determination of characteristic limits is general as the approach for the evaluation of uncertainties. It does not only apply to radioactivity measurements but can easily be extended to any analytical measurement where a measurand of quantity in a sample has to be measured against a background or blank of the general environment (Michel, 2000). The determination of a net concentration of a chemical element against the analytical blank is an example here. Generally, the determination of characteristic limits according to the cited standards is straight forward and only in some cases as in y-spectrometry do some more sophisticated mathematical methods have to be applied (Weise and Michel, 1995). The commitment to a unified approach for the determination of characteristic limits is also of another practical importance. Since required detection limits are used in many countries to specify requirements for methods used for environmental radioactivity measurements, an unambiguous definition of detection limits is necessary. The reporting levels used by the European Union (European Commission, 1996b) in practice have the character of required detection limits since results of measurements are discarded for which the detection limit is not below the reporting level. There is, however, some confusion that the respective report uses the term detection limit but it does not specify how the detection limits have to be calculated and does not give the probabilities of the errors of 1st and 2nd kind on which the detection limit has to be based. It becomes even worse if one takes the example of the Annex of the EU Drinking Water Guideline (European Commission, 1998, 2000b) where "required detection limits" are given for radioactivity measurements. Again there is no meaningful specification of the definition of detection limits and with regard to other analytical investigations of the water even contradictory statements are made. Also in the general scientific literature widely inadequate use is made of characteristic limits and insufficient information is frequently given about the conditions under which these limits were determined. Therefore, a standardized use of characteristic limits is urgently needed as a part of setting unified requirement for measurements and for an unambiguous report of such limits for a measurement. With the upcoming ISO 11929-7 (ISO, 2000a) or with DIN 25482-10 (DIN, 2000) such a standardized approach is possible and a requirement for any measurement of environmental radioactivity. An experimental and analytical problem generally arises from the fact that environmental radioactivity has always to be measured against an environmental background and/or an analytical blank. While it is standard

50

R. Michel

practice to correct for counting or spectrometric background in the measurements themselves, the interference by analytical blanks is frequently still neglected. One can make the statement that any measurement claiming an enhanced abundance of a radionuclide or element in a sample is worthless if no information on the analytical and/or environmental blank is given. Though the use of clean rooms and high-purity reagents may considerably reduce analytical blanks the demand is with the experimentalist to demonstrate and to report in a traceable form the actual blanks. An impressive example for blank problems is a recent round robin exercise for the determination of 129I in environmental materials (Roberts et al, 1997, 2000). While the measurements of 129I in readily prepared Agl samples by AMS proved to agree within reasonable limits, the results for the entire analysis of 129I in environmental samples scattered by orders of magnitude. This disagreement could be pinpointed to result from differences in the analytical blanks of the various laboratories. For 129I and 36C1 differences of orders of magnitude in the analytical blanks can easily occur, see e.g. (Schmidt et al., 1998) and only repeated and consequent blank analyses can provide a basis for accurate results. A general guarantee for accuracy can, however, not be given. To demonstrate accuracy and reliability, repeated analyses of reference materials and frequent participation in round robin exercises or interlaboratory comparisons are essential. Large efforts have been undertaken to establish reference materials. However, in spite of the intensive work done for instance by the IAEA and many national or other international institutions, for many materials reference materials are still lacking and the further development of new materials can only be highly encouraged. In view of the general globalization process, which affects also the discussions of environmental radioactivity, interlaboratory exercises on an international level have to be preferred to national ones. It is, however, important that the same standardized and evaluated methods are used for the analyses of samples from such exercises and for normal ones of the daily work to assure that the same claim of accuracy can be made for all results obtained by a laboratory. In this context, quality assurance and control on the basis of national and preferably of international standards becomes essential. Only by use of strict analytical protocols and completely traceable performance of measurements it is possible to guarantee quality. Evaluated procedures for measurements of environmental radioactivity do exist in many countries (AFNOR, 2000; AKU, 1999; BMU, 1998; DoE, 1979), but only a few of them are standardized on a national or international level. Thus, ISO TC 85 SC 2 has set the goal of establishing such standards in the future and work has begun to develop respective standards starting with the measurements of radionuclides in soil (ISO, 2000b). This standard will not only deal with the radiochemical and analytical aspects of soils, but also with sampling strategies and methods which just recently have been investigated by international collaborations (Anonymous, 2001). It will still take some time until internationally accepted standardized procedures will be

2 - Environmental radioactivity measuring methods

51

available. But, such standards are urgently needed to qualify and certify laboratories performing measurements of environmental radioactivity. It is a general trend that certification of laboratories will be required in the future if measurements are performed on a commercial basis or if legally required measurements have to be performed. Peer review of scientific investigations and publications has strictly to investigate whether adequate measures of quality assurance and control have been taken if no certification for a research group exists. But, even with all possible measures of quality assurance and control, the struggle for accuracy and reliability will never end. As a further point with respect to quality it has to be emphasized that any judgment about anomalous conditions, i.e. enhanced radioactivity in the environment, requires as a basis detailed knowledge about what is normal. This knowledge is not always available. As an example, the existing data about naturally occurring radionuclides in many cases are strongly biased due to the fact that samples were taken more frequently in regions where geologically enhanced radioactivity was expected. There is a considerable lack of data in geologic low-activity areas for nearly all environmental materials. This lack of knowledge and the consequent bias makes reasonable judgement about NORM and its influences on the human radiation exposure severely difficult. Finally, the documentation of the results is an important facet of the quality of a measurement. It is, moreover, a basic requirement for traceability of a result. It has to be as complete as possible describing the sampling aspects, the analytical methods, the measures of quality assurance and control used, the standards on which the result is based and all information including the nuclear data used for evaluation. It has to state the result and its combined standard uncertainty, the characteristic limits and the way they are computed including the assumed probabilities of the errors of the 1ST and 2nd kind and the probability considered for calculating the confidence limits. The results of blank analyses and of reference materials also have to be included into the documentation. Given the enormous number of measurements of environmental radioactivity all over the world, adequate documentation becomes a major problem if just printing media are considered because of the tremendous amount of information needed to judge about the quality of a measurement result. Here, new methods of electronic documentation and publication are urgently needed to inform the scientific world as well as the general public.

5. Environmental radioactivity and the public In the public perception, radioactivity, in general, and radioactivity in the environment, in particular, is directly connected with the risk of stochastic effects of ionizing radiation and discussed with emotion and belief rather

52

R. Michel

than rationally. Without discussing the reasons for the irrational public perceptions of risk and radioactivity (Pretre, 1992, 2000), an attempt must be undertaken to rationalize the way societies treat these issues. The complexity of this problem is that it contains a sequence of questions which only partially can be solved by the methods of natural sciences. It starts with the question: What can we know about the abundances of radionuclides in the different environmental materials and what is the resulting radiation exposure? It continues with: What is the risk associated with this radiation exposure? and it ends with the questions: Is the risk acceptable for society, for the groups at risk and for the individual? and What shall society and the individual do? The questions for the abundances of radioactivity, for the resulting exposure and for the associated risk can be solved by methods of natural sciences. The questions for the acceptability of risks and for the decisions about what has to be done cannot be solved by these methods. These questions have to be decided by society taking into account ethical, political, psychological, social and, last not least, economical issues and are beyond the topic of this paper. When dealing with these complex issues in the public, measurements of environmental radioactivity provide the fundamental data, on which the evaluation of the resulting radiation exposure is based. These data must be of outstanding quality and beyond any doubt because, otherwise, there is no scientific basis for any public discussion. The fact, that the determination of the risks associated with an exposure is not finally settled adds already enough confusion to the public discussion. These still exists lack of knowledge about the induction and mechanisms of stochastic effects and the fact that the risks of radiation exposure have to be evaluated on the basis of incomplete information as e.g. the future frequency of cancer in the still living survivors of the Hiroshima and Nagasaki explosions (Nagataki, 2000). To improve the present unsatisfactory situation it will not be sufficient to keep the knowledge of the abundances of radionuclides in the environment and of the resulting exposure in the scientific community. All existing information must be broadly disseminated and has to be freely and easily available in forms which are understandable for the individual target groups which cover, besides scientists, governments, regulators bodies, politicians and decision makers, media, environmental activists, political parties and the silent majority of the population. The present habits of publishing only in scientific journals and of preparing excellent national and international reports on these issues are not enough. They must be accompanied by availability of all required information in the electronic media with complete transparency. To this end a hierarchy of informative systems has to be established nationally and internationally which can be realized by the use of the internet. The hierarchical structure will result from the degree of in-depth information expected by the respective user group. It must range from a level of general information to the possibility of accessing the real data. Such a system has to take special care of the needs of children in schools and students in academic education.

2 - Environmental radioactivity measuring methods

53

While present days adults are widely influenced by opinions and beliefs resulting from the threads of the cold war, more rational perception of radioactivity and radiation exposure will widely rely whether respective information can be disseminated to young people. In this context, some developments have to be mentioned which I consider to be unfavorable. One is the use of reporting levels as a cut-off criterion for the publication of data as e.g. in the reports of the EU (European Commission, 1996b) where existing measurement results are omitted if they are below the reporting levels. For a member of the public such a procedure can only appear acceptable after a relatively complex consideration taking into account the rational basis of the reporting levels. In this case, the publication of the actual data and in case of results below the detection limits the statement of the detection limits is preferable. In public discussions of an actual problem we also have to accept that measurements of the really existing radionuclide concentrations will be required even if those concentrations are beyond any radiological relevance. It has, however, to be made clear to the public, that the timeliness of respective results cannot be always guaranteed since such measurements occur on time-scales which may by long compared to those of discussions in a society. The invention of such in-depth measurements in the coarse network used by the EU in surveillance of environmental radioactivity (European Commission, 1996b) is a good example for what has to be done if an actual issue is in the public discussion. Moreover, the results obtained by the Nord-Cotentin Radioecology Group (1999, 2000) in the discussion of the exposure and the induced risk from the emissions from the La Hague reprocessing plant are an outstanding example here. Another problem arises from the way actual exposures are evaluated and presented to the public. The evaluation of the human exposure is the ultimate goal of environmental radioactivity measurement. The EURATOM basic safety guidelines (European Commission, 1996a) require the surveillance of human radiation exposure, in general, and the detailed evaluation for groups with elevated exposure. They further demand that exposure has to be determined as realistically as possible. There is, however, a widespread confusion due to the fact that conservative approaches are frequently used to model exposures the results of which may exceed a realistic exposure estimate by orders of magnitude. Whilst these conservative approaches are fully acceptable and justified for planning purposes they must not be used to estimate the actual exposure in a given situation where measured data exist and more reliable and realistic dose estimates can be made. In future information policy clear distinctions between these case of exposure estimates have to be made. Also the use of individual and collective doses poses severe problems for the public. While individual doses resulting from a practice or a particular exposure situation can be directly compared with the average natural exposure or with dose limits, collective doses are not easily understandable, in

54

R. Michel

particular, if the integration times exceed by far human lifetime. Collective doses surely are required and meaningful in surveillance, planning and optimization of practices. However, the misuse of collective as a basis to calculate expected cases of stochastic diseases in situations where the individual exposure is comparable to or even lower than the average natural exposure must be abandoned and international agreement about cut-off criteria for the calculation of collective dose are needed. Finally, an important issue for public discussion is the consistent treatment of exposures due to natural and man-made radioactivity. There is a discrepancy between the perception of the large variability of the natural radiation exposure, on the one hand, and of the usually very small contributions of man-made radionuclides, on the other. Even in accidental situations or large scale contaminations such discrepant judgements can be observed. On the basis of the requirements of e.g. the EURATOM basic safety standards (European Commission, 1996a) and of an increasing awareness of the scientific and regulatory communities it will be an important issue for future considerations to struggle for a consistent system of radiation protection when dealing with natural and man-made radioactivity.

6. Future trends and new frontiers Summarizing, future trends of measuring environmental radioactivity can be recognized with respect to (i) the development of the measurements themselves and of the analytical practices, (ii) their fields of applications and (iii) the dissemination of the results. In parallel to the further refinement of existing methods of nuclear radiation spectrometry, the future developments will comprise increasing applications of mass spectrometric measurements The analytical techniques will be further improved by especially tailored chemical reagents for extraction of radionuclides from environmental materials and by the development of automated chemical systems for separation of a large number of radionuclides from one sample. Finally, laboratory practice will be dominated in the future by systems of quality assurance and control. Certification of laboratories and complete traceability of the results will be state-of-the-art without which no measurements will be acceptable. The purposes of environmental radioactivity measurements will improve from documentation of man-made changes of the environment to measurements dedicated to the achievement of a sustainable development. This requires • surveillance of (further minimized) emissions from nuclear installation with the goal to attain negligible emissions; • guidance of measures for remediation of contaminated territories and sites and for a safe disposal of radioactive waste, and

2 - Environmental radioactivity measuring methods

55

• quantification of the long-term impact of man on the radiation exposure of man and other parts of the biosphere in addition to the evaluation of the short-term impact of man. Moreover, there will be an increasing importance of measuring natural radioactivity and potentially enhanced exposures to these nuclides from non-nuclear technologies. Finally, measurements of environmental radioactivity will be increasingly used to exploit the information about environmental processes in which long-lived radionuclides act as tracers and document both natural and man-made environmental change. Last but not least, the future will see a widely changed attitude with respect to the dissemination of the results and the information of the public which besides quality of the measurement results is the key issue for a rational perception of environmental radioactivity in society.

References AFNOR (2000) "French Standards in connection with measurements of environmental radioactivity; air: NF N° 60-312, 60-760 to 60-770, 13466; water: NF N° 60-802; soil: NF N° 60790-1 to 60790-8; bioindicators: NF N° 60780-1 to 60780-8", AFNOR, Paris. AKU (1999) "Recommendations for the surveillance of environmental radioactivity" (in German), Loseblattsammlung des Arbeitskreise Umweltuberwachung (AKU) des Fachverbandes fur Strahlenschutz e.V., November 1999, Winter M., Narrog J., Kukl W. and Vilgis M. (Eds.), ISSN 1013-4506. Anonymous (2001) "Comparative Evaluation of European Methods for Sampling and Sample Preparation of Soils", Special issue of The Science of the Total Environment, 264, N° 1 and 2. Beer J., Blinov A., Finkel R.C., Hofmann H.J., Lehmann B., Oeschger H., Sigg A., Schwander J., Staffelbach T., Stauffer B., Suter M. and Wolfli W. (1990) "Use of 10Be in polar ice to trace the 11-year cycle of solar activity", Nature, 347,164-166. Beutier D., Dellero N. and Pradel J. (2000) "Comparing the dose contributions from man made and from natural radioactivity", IRPA-10, 10th Int. Conf. of the International Radiation Protection Association, Hiroshima, May 14-19, 2000, P-4b-239. Bittner S. and Edelhauser H.G. (1999) "Duties of the Federal Ministry of Environment, Nature Conservation and Nuclear Safety in the Event of a Nuclear Accident", Kerntechnik, 64, N°. 3,118-123. BMU (1986) "The Federal Minister for the Environment, Nature Conservation and Nuclear Safety, Act on the precautonary protection of the popu-

56

R. Michel

lation against radiation exposure" (Precautionary Radiological Protection Act- StrVG) vom 19. Dez. 1986, BGBL. 1986, Part I, p. 2610. BMU (1995a) "The Federal Minister for the Environment, Nature Conservation and Nuclear Safety", Directive of 11 April 1995 on monitoring of environmental radioactivity referred to in StrVG, part II: Measuring programme in accidental situations, GMBI. 46, No 14 p. 261. BMU (1995b) "The Federal Minister for the Environment, Nature Conservation and Nuclear Safety", General administrative provision of 27 September 1995 for the Integrated Measuring and Information System referred to in StrVG (AW-IMIS), Bundesanzeiger 47, No 200 a, p. 3. BMU (1997) "The Federal Minister for the Environment, Nature Conservation and Nuclear Safety", Directive of 27 November 1997 on monitoring radioactive effluents and environmental radioactivity caused by mining activities (REI Bergbau), BMU letter RSII 7-15013/5-. BMU (1998) "The Federal Minister for the Environment, Nature Conservation and Nuclear Safety, Measuring instructions for the monitoring of environmental radioactivity and of radioactive effluents from nuclear plants" (in German), Gustav Fischer Verlag, Stuttgart, Jena, Liibeck, Ulm. BMU (1999) "The Federal Minister for the Environment, Nature Conservation and Nuclear Safety, Basis for the assessment of radiation exposure due to environmental radioactivity caused by mining activities" (Assessments - Bergbau), version 30 July 1999; Basis for the assessment of radiation exposure due to radon and its short-lived decay products inhalation caused by mining activities (Assessments - Bergbau: Part Radon), version 30 July 1999. Bowman C.D., E.D. Arthur, P.W. Lisowski, G.P. Lawrence, R.J. Jensen, J.L. Anderson, B. Blind, M. Cappiello, J.W. Davidson, T.R. England, L.N. Engel, R.C. Haight, H.G. Highes, J.R. Ireland, R.A. Krakowski, R.J. LaBauve, B.C. Letellier, R.T. Perry, G.J. Russell, K.P. Staudhammer, G. Versamis and W.B. Wilson (1992) "Nuclear energy generation and waste transmutation using an accelerator-driven intense thermal neutron source", Nucl. Instr. Meth. Phys. Res. A, 320, 336-367. Carminati F., R. Klapisch, J.P. Revol, Ch. Roche, J.A. Rubio and C. Rubbia (1993) "An Energy Amplifier for Cleaner and Inexhaustible Nuclear Energy Production Driven by a Particle Beam Accelerator", CERN/AT/93-47(ET). Cook G.D., Harkness D.D., Mackenzie A.B., Miller B.F. and Scott E.M. (1996) "Proc. Int. Conf. on Advances in Liquid Scintillation Sprectrometry 1994", Glasgow, Scotland, August 8-12,1994, Radiocarbon, Univ. of Arizona, Tucson, USA. Debertin K. and Helmer R.G. (1988) "Gamma- and X-spectrometry with semiconductor detectors", North Holland, Amsterdam, Oxford, New York, Tokyo.

2 - Environmental radioactivity measuring methods

57

DIN (2000) DIN 25482-10, Nachweisgrenze und Erkennungsgrenze bei Kernstrahlungsmessungen; Allgemeine Anwendungen, Beuth Verlag Berlin. DoE (1979) "EML Procedure Manual" (previously HASL-Manual), Harley J.H. (Ed.), Environmental Measurement Laboratory, U.S. DoE, New York, August 1979. Eichler B., Hiibener S., Erdmann N., Eberhardt K., Herrmann G., Kohler S., Trautmann N., Passler G. and Urban F.-J. (1997) "An atomic beam source for actinide elements: Concept and realization", Radiochimica Acta, 79, 221-233. Erdmann N., Herrmann G., Huber G., Kohler S., Kratz J.V., Mansel A., Nunnemann M., Passler G., Trautmann N., Turchin A. and Waldeck A. (1997) "Resonnance ionization mass spectroscopy for trace determination of plutonium in environmental samples", Fresenius J. Anal, Chem., 359, 378-381. EURATOM (1957) "EURATOM Treaty establishing the European Atomic Energy Community", Rome. European Commission (1959) "EURATOM Council Directives of 2 February 1959 laying down the basic standards for the protection of the health of workers and the general public against the dangers arising from ionizing radiations", OJ No 11, 20. 2.1959, p. 221/59. European Commission (1996a) "96/29/EURATOM Council Directive of 13 May 1996 laying down basic safety standards for the health protection of the general public and workers against the dangers of ionizing radiation", OJ L-159 of 29/06/96 page 1. European Commission (1996b) "Environmental Radioactivity in the European Community", 1987 - 1990, 2nd Edition, EUR 15699 3A, Luxembourg, 1996. European Commission (1998) "98/83/EC, Council Directive of 3 November 1998 on the quality of water intended for human consumption", OJ L-330 of 5/12/98 page 32/54. European Commission (2000a) "Draft modification to OJ L-330 of 5/12/98 page 46/54" (Version 21/03/00 of Draft Annexes II and III of the Council Directive 98/83/EC). European Commission (2000b) "2000/473/EURATOM Commission Recommendation of 8 June 2000 on the application of Article 36 of the Euratom Treaty concerning the monitoring of the levels of radioactivity in the environment for the purpose of assessing the exposure of the population as a whole", OJ L-191 of 27/07/2000 page 37. Filfi M., J. Handl, R. Michel, V. P. Slavov and V.V. Borschtschenko (1998) "A Fast Method for the Determination of Strontium-89 and Strontium-90 in Environmental Samples and ist Application to the Analysis of Strontium-90 in Ukrainian Soils", Radiochimica Acta, 83, 81-92. Finkel R.C. and Suter M. (1993) "AMS in the earth sciences: Techniques and applications", Adv. Anal. Chem., 1,1-114.

58

R. Michel

Frechou C., Calmet D., Bouisset P., Piccot D., Gaudry A., Yiou F. and Raisbeck G. (2000) "129I and 129I/127I ratio determination in environmental biological samples by RNAA, AMS and direct y-X spectrometric measurements", 5"1 International Conf. on Methods and Applications of Radioanalytical Chemistry, Kailua-Kona, 9-14 April 2000, Log N°. 279. Gilbert R.O. (1987) "Statistical methods for environmental pollution monitoring", Van Norstrand Reinhold Company, New York. Halliday A.N., Lee D.-C, Christensen J.N., Walder A.J., Freedman P.A., Jones C.E., Hall C.M., Yi W. and Teagle D. (1995) "Recent developments in inductively coupled plasma magnetic sector multiple collector mass spectrometry", Intl. J. Mass Spec. Ion Proc., 146/147, 21-33. Halliday A.N., Lee D.-C., Christensen J.N., Rehkamper M., Yi W., Luo X., Hall C.M., Ballentine C.J., Pettke T. and Stirling C. (1998) "Allpications of multiple collector-ICPMS to cosmochemistry, geochemistry and paleoceanography", Geochim. Cosmochim. Ada, 62, 919-940. Harley N.H., Foulkes B.C., Hilborne L.H., Hudson A. and Ross Anthony C. (1999) "A review of the scientific literature as it pertains to the Gulf war illnesses", National Defense Research Institute, Rand Corporation. Herpers U., Herr W. and Woelfle R. (1969) "Evaluation of Mn-53(n,y) Activation, Al-26 and special trace elements in meteorites by gamma-coincidence techniques", P. Millman (Ed.), Meteorite Research, D. Reidel, Dordrecht, Holland, 387-396. Heusser G. (1995) "Low-radioactivity background techniques", Ann. Rev. Nud. Part. Sci., 45, 543-590. IAEA (1988) "The radiological accident in Goiana", STI/PUB/815, IAEA, Vienna. IAEA (1991) "The international Chernobyl Project", three volumes, IAEA, Vienna. IAEA (1996) "One decade after Chernobyl: Summing up the consequences of the Accident", Pore, of an Int. Conf. of the EC, IAEA, WHO, Vienna, 8-12 April 1996, IAEA, Vienna, 555 pages. IAEA (1998a) "Radiological assessment at the Semipalatinsk Test Site, Kazakhstan: Preliminary assessment and recommendations for further study", IAEA, Vienna, 43 p. IAEA (1998b) "Radiological conditions at Bikini Atoll: Prospects for resettlement", IAEA, Vienna, 67 p. IAEA (1998c) "The radiological situation at the atolls of Mururoa and Fangataufa: Main Report", Summary Report and Technical Report in six volumes, IAEA, Vienna. ICRP (1991) "1990 Recommendations of the International Commission on Radiological Protection", ICRP Publication 60, Pergamon Press, New York.

2 - Environmental radioactivity measuring methods

59

ICRP (1996) "Publication 72, Age-dependent Doses to Members of the Public: Part 5 Compilation of Ingestion and Inhalation Dose Coefficients", Annals of the ICRP, 26, N° 1, Elsevier, Oxford. ISO (1995) "ISO Guide to the Expression of Uncertainty in Measurement", ISO International Organization for Standardization, Geneva 1993, corrected reprint Geneva. ISO (2000a) "ISO/DIS 11929-7, Determination of the Detection Limit and Decision Threshold for Ionising Radiation Measurements - Part 7: Fundamentals and General Applications", ISO/TC85/SC2/WG17 Geneva. ISO (2000b) "ISO/WD 18589 Measurements of radioactivity in the environment - soil - Parts 1-8", ISO/TC85/SC2/WG17 Geneva. Kutschera W., Golser R., Priller A. and Strohmeier B. (2000) "Accelerator Mass Spectrometry", Proceedings of the Eighth Int. Conf. on Accelerator Mass Spectrometry, Vienna, Austria, 6-10 September 2000, Nucl. Instr. Meth. Phys. Res. B, 172,1-977. Levin I. and Kromer B. (1997) "Twenty years of atmospheric 14CO2 observations at Schauinsland Station", Germany, Radiocarbon, 39, 205. Lubenau J.O. and Yusko J.G., "Radioactive materials in recycled metals - an update", Health Physics, 74, 293-299. Michel R. (1999) "Long-Lived Radionuclides as Tracers in Terrestrial and Extraterrestrial Matter", Radiochim. Ada, 87, 47-73. Michel R. (2000) "Quality Assurance of Nuclear Analytical Techniques Based on Bayesian Characteristic Limits", Proc. MTAA-10, J. Radioanal. Chem., 245,137-144. Muller H. and Prohl G. (1993) "ECOSYS-87: A dynamic model for assessing radiological consequences of nuclear accidents", Health Physics, 64, 232252. Muller P., Blaum K., Bushaw B.A., Diel S., Geppert Ch., Nahler A., Nortershauser W., Trautmann N. and Wendt K. (2000) "Trace detection of 41Ca in nuclear reactor concrete by diode-laser-based resonance ionization mass spectromtry", Radiochimica Acta, 88,487-493. Nagataki S. (2000) "Activities and future plans of Radiatiation Effects Research Foundation", 10th Congress of IRPA, Hiroshima, May 14-19, 2000, Proceedings CD: Invited Lecture L-2, IRPA. Nakamura T., Nakai N. and Oshishi S. (1987) "Applications of environmental C-14 measured by AMS as a carbon tracer", Nucl. Inst. Meth. Phys. Res. B, 29, 355-360. NEA (1995) "Chernobyl, Ten years on - Radiological and health impact. An assessment by the OECD NEA Committee on Radiation Protection and Health" NEA/OECD Paris. Nord-Cotentin Radioecology Group (1999) "Detailed final reports: Vol. 1 Inventory of radioactive releases from nuclear facilities; Vol. 2 Critical review of measurements in the environment; Vol. 3: Transfer models for

60

R. Michel

radionuclides through the environment; Vol. 4: Estimate of doses and associated leukemia risk", IPSN, Fontenay aux Roses, France. Nord-Cotentin Radioecology Group (2000) "Estimation of exposure levels to ionizing radiation and associated risks of leukemia for populations in the Nord-Cotentin", Summary Report, IPSN, Fontenay-aux-Roses, France. Passler G., Erdmann G., Hasse H.-U., Herrmann G., Huber G., Kohler S., Kratz J.V., Mansel A., Nunnemann M., Trautmann N. and Waldeck (1997) "Application of laser mass spectrometry for trace analysis of plutonium and technetium", Kerntechnik, 62, 85-90. Perkins R.W., Rancitelli L.A., Cooper J.A., Kaye J.H. and Wogman N.A. (1970) "Cosmogenic and primordial radionuclide measurements in Apollo 11 lunar samples by nondestructive analysis", Proc. of the Apllo 11 Lunar Sci.Conf., 2,1455-1469. Pretre S.B. (1992) "Psycho-Soziologische Muster der Technik-Betroffenheit", atomwirtschaft, November 1992, 518-522. Pretre S.B. (2000) "Communication with the public: Radiation, risk and perception in context", 10th Congress of IRPA, Hiroshima, May 14-19, 2000, Proceedings CD: T-15-1, P-10-171, IRPA. Roberts M.L., Caffee M.W. and Proctor I.D. (1997) "129I interlaboratory comparison", Nucl. Instr. Meth. Phys. Res. B, 123, 367-370. Roberts M.L. and Caffee M.W. (2000) //129I interlaboratory comparison: Phase II results", Kutschera W., Golser R., Priller A. and Strohmeier B. (Eds.), Accelerator Mass Spectrometry, Proceedings of the Eighth Int. Conf. on Accelerator Mass Spectrometry, Vienna, Austria, 6-10 September 2000, Nucl. Instr. Meth. Phys. Res. B, 172, 388-394. Rostker B. (2000) "Environmental Exposure Report, Depleted Uranium in the Gulf (II)", Interim Report of Dec. 13, 2000, US Department of Defense, Washington. Ruhm, W., Kellerer, A.M., Korschinek, G., Faestermann, T., Knie, K., Rugel, G., Kato, K. and Nolte, E. (1998) "The dosimetry system DS86 and the neutron discrepancy in Hiroshima - historical review, present status, and future options", Radiat. Environ. Biophys., 37, 293-310. Schkade U.-K., Beyermann M., Hartmann M., Naumann M., Seehafer M., Ullmann W., Will W. and Winkelmann I. (1999) "Verfahren zur Bestimmung von natiirlichen Radionukliden in der Umwelt zur Erfiillung der Richtlinie zur Emissions- und Immissionsiiberwachung bei bergbaulichen Tatigkeiten" (REI Bergbau), Report ST-IB-2, BfS, Berlin. Schmidt A., Schnabel Ch., Handl J., Jakob D., Michel R., Synal H.-A., Wagner M.J.M. and Suter M. (1998) "On the Analysis of 1-129 and 1-127 in Environmental Materials by Accel-erator Mass Spectrometry and Ion Chromatography", Science Total Environment, 223,131-156. Smith R.J. (1984) "Juarez: An unprecedented radiation accident", Science, 223,1152-1154.

2 - Environmental radioactivity measuring methods

61

SSK (1996) "Ten years after Chernobyl; An Assessment", A. Bayer, A. Kaul and Chr. Reiners (Eds.), Gustav Fischer Verlag, Stuttgart. Stirling C.H., Lee D.-C, Christensen J.N. and Halliday A.N. (2000) "Highprecision in situ 238U- 234U- 230Th isotopic analysis using laser ablation multiple-collector ICPMS", Geochim. Cosmochim. Ada, 64, 3737-3750. Studier M.H., Postmus C. jr., Mech J., Walters R.R. and Sloth E.N. (1962) "The use of 129I as an isotopic tracer and its determination along with normal 127 I by neutron activation analysis", J. Inorg. Nucl. Chem., 24, 755-761. Suter M., Jacob S.W.A. and Synal H.-A. (2000) "Tandem AMS at sub-MeV energies - Status and prospects", Kutschera W., Golser R., Priller A. and Strohmeier B. (Eds.), Accelerator Mass Spectrometry, Proceedings of the Eighth Int. Conf. on Accelerator Mass Spectrometry, Vienna, Austria, 610 September 2000, Nucl. Instr. Meth. Phys. Res. B, 172,144-151. Synal H.-A., Beer J., Bonani G., Suter M. and Wolfli W. (1990) "Atmospheric transport of bomb-produced 36C1", Nucl. Instr. Meth. Phys. Res. B, 52,483488. Synal H.-A., Beer, Schotterer U., Suter M. and Thompson L. (1997) "Bomb produced 36C1 in ice core samples from mountain glaciers", in Proc. of a Workshop on Glaciers in the Alps: Climate and Environmental Analysis, Wengen, 21-23 October, 1996, p. 99. Synal H.-A., Bonani G., Dobeli M., Ender R., Gartenmann P., Kubik P., Schnabel C. and Suter M. (1997) "Status report of the PSI/ETH AMS facility", Nucl. Instr. Meth. Phys. Res. B, 123, 62-68. Szidat S., Schmidt A., Handl J., Jakob D., Michel R., Synal H.-A. and Suter M. (1999) "Analysis of Iodine-129 in Environmental Materials: Quality Assurance and Applications", Proc. MTAA-10, /. Radioanal. Chem., 244, 45-50. Szidat S., Schmidt A., Botsch W., Michel R., Synal H.-A., Suter M., Lopez-Guiterrez J.-M. and Stade W. (2000a) "Radioecology of 129I: Results from recent and older European samples Analysis of 129I in environmental matrices: sample preparation and quality control", Kutschera W., Golser R., Priller A. and Strohmeier B. (Eds.), Accelerator Mass Spectrometry, Proceedings of the Eighth Int. Conf. on Accelerator Mass Spectrometry, Vienna, Austria, 6-10 September 2000, Nucl. Instr. Meth. Phys. Res. B, 172,699-710. Szidat S., Schmidt A., Handl J., Jakob D., Michel R., Synal H.-A., Schnabel Ch., Suter M. and Lopez-Gutierrez J.M. (2000b) "RNAA and AMS of Iodine-129 in Environmental Materials - Comparison of Analytical Methods and Quality Assurance", Kerntechnik, 65,160-167. Tait D. and Wiechen A. (1997) "A fast method for the determination of 90Sr in milk by solid phase extraction with Cryptand 222 on cation exchange resin", Kerntechnik, 62, 96. Tait D., Haase G. and Wiechen A. (2000) "A fast method for monitoring 1-129 in milk in the surveillance of radioactive waste repositories in the event of accidental releases", Kerntechnik, 65,168-171.

62

R. Michel

UNSCEAR (1982) "United Nations Scientific Committee on the Effects of Atomic Radiation, Ionizing Radiation: Sources and biological effects", 1982 Report to the General Assembly, United Nations, New York. UNSCEAR (1988) "United Nations Scientific Committee on the Effects of Atomic Radiation, Sources, Effects and Risks of Ionizing radiation", 1988 Report to the General Assembly, United Nations, New York. UNSCEAR (2000) "United Nations Scientific Committee on the Effects of Atomic Radiation, Sources and Effects of Ionizing Radiation", 2000 Report to the General Assembly, United Nations, New York. Weise K. (1998) "Bayesian-statistical decision threshold, detection limit and confidence interval in nuclear radiation measurement", Kerntechnik, 63, 214-224. Weise K. and Michel R. (1995) "Erkennungsgrenzen, Nachweisgrenzen und Vertrauensbereich in der allgemeinen Kernstrahlungs-Spektrometrie", Kerntechnik, 60,189-196. Weise K. and Woger W. (1993) "A Bayesian theory of measurement uncertainty", Meas. Sci. Technol., 4,1-11. Weise K. and Woger W. (1999) "MeGunsicherheit und Mefidatenauswertung", Wiley-VCH, Berlin. Wendt K., Bhowmick G.K., Bushaw B.A., Herrmann G., Kratz J.V., Lantzsch J., Miiller P., Nortershauser W., Otten W.-W., Schwalbach R., Seibert U.A., Trautmann N. and Waldeck A. (1997a) "Rapid trace analysis of 89>90Sr in environmental samples by collinear laser resonnance mass spectrometry", Radiochimica Ada, 79,183-190. Wendt K., Blaum K., Bushaw B.A., Juston F., Nortershauser W., Nunnemann M., Trautmann N. and Wiche B. (1997b) "Ultratrace analysis of calcium with high isotopic selectivity by diodelaser resonance ionisation mass spectrometry", Fresenius J. Analy. Chem., 359, 363-363. Wendt K., Blaum K., Bushaw B.A., Griming C, Horn R., Huber G., Kratz J.V., Kunz P., Miiller P., Nortershauser W., Nunnemann M., Passler A., Schmitt A., Trautmann N. and Waldeck A. (1999) "Recent developments in and applications of resonance ionization mass spectrometry", Fresenius J. Anal. Chem., 364, 471-477. Wollbarst A.B., Blom P.F., Chan D., Cherry R.N. Jr., Doehnert M., Fauver D., Hull H.B., MacKinney J.A., Mauro J., Richardson A.C.B. and Zaragoza L. (1999) "Sites in the United States contaminated with radioactivity", Health Physics, 77, 247-260. Woushou P. et al. (1997) "Co-60 contamination in recycled steel resulting in elevated civilian radiation doses: causes and challenges", Health Physics, 73, 465-472.

3

Toxicants in the environment: bringing radioecology and ecotoxicology together S.C. Sheppard 1

Radioecology and ecotoxicology are both well-established disciplines, and both deal with the fate and effects of contaminants in the environment. Until the last few years, there was not a strong need for interaction between these disciplines. However, as the nuclear industries and regulators place greater emphasis on investigating the protection of non-human biota, the opportunity for synergy between the disciplines has increased. Radioecology is very strong in understanding the fate of radionuclides, in dealing with the additivity of doses, and in understanding physiological effects of radiation. Ecotoxicology has particular strengths in identifying the level of biological organization to protect and in extrapolation among species, toxicity endpoints and contaminants. The synergy of these disciplines will result in a very powerful assessment capability. This paper explores the opportunities for synergy, illustrated with a few specific examples drawn from recent radionuclide ecological risk assessments.

1. Introduction Radioecology as a discipline has taught us a great deal about the transport and behaviour of radionuclides in the environment. A large part of radioecology has been devoted to understanding how humans may be exposed to radionuclides released into the environment. The food chain models are an outcome of this work. However, there has always been a branch of radioecology where the effects of radiation on other organisms were studied. One of the original objectives of this research was to understand the ecological consequences of nuclear weapons. There was also a major emphasis on beneficial effects such as mutation breeding and hormesis. In the past five to ten years, there has been a new branch of radioecology developing. In it, the objective is to protect non-human biota from the effects of radionuclides. This work is re-examining the older paradigm that "if humans are protected 1

ECOMatters Inc., Pinawa, Manitoba, ROE 1LO, Canada.

64

S.C. Sheppard

then all other organisms are also protected". Radioecology has enjoyed a strong international network, supported and guided by international consensus organizations such as IAEA, ICRP and UNSCEAR. Ecotoxicology was begun and has evolved in a different way than radioecology. From its beginning, the objective has been to protect nonhuman biota, with early emphasis on aquatic habitats and man-made chemicals. It has since been applied to almost all environments and contaminants, and the toxicological aspects now include biokinetics similar to that developed for human toxicology. Because ecotoxicology was not anthropocentric, it developed the logic to define what should we protect in the environment and how this should be done. For both radioecology and ecotoxicology, there is a strong research requirement to provide concepts and data. However, overlying both is a need for predictive capability. The potential impacts of contaminants must be estimated in order to assess the impacts of human activities, and to achieve an optimal cost/benefit situation. This paper will emphasize the assessment roles of radioecology and ecotoxicology, with the full recognition that these roles could not advance without the underlying research. The commonalties and differences between traditional ecotoxicology and radioecology will be explored and illustrated with results from recent ecological risk assessments involving radionuclides.

2. Bottom-up versus top-down The history of the two disciplines has resulted in different approaches. The steps made to protect non-human biota from radionuclide contamination could be described as "top-down". Starting from a strong position in the protection of humans from radionuclides, a common first statement was to consider that if humans are protected, then all other species will be protected. The strength of this argument came from the concepts: 1) that human individuals are protected, but only populations of non-human biota need be protected; and 2) it is probable that humans may be the organism most sensitive to radiation. In a practical sense, this statement may well be true, but the consensus now is that it must be proven, perhaps on a case-by-case basis. The next step in the top-down development led to publication of international consensus dose-limits for all non-human biota, and then for classes of non-human biota. These remain a preferred benchmark in many programs worldwide. In contrast, the basic approach in ecotoxicological assessment tends to be bottom-up (e.g., Suter, 1993). After defining a very specific environmental feature to protect, an "assessment endpoint" is defined. The feature might be the population of clams in a particular stretch of river, and the assessment endpoint the reproductive capacity of this population. Almost inevitably,

3 - Toxicants in the environment

65

there are no data for the effects of the contaminant of concern on the chosen assessment endpoint. This gap is bridged, for assessment purposes, by defining a 'measurement endpoint' which is an organism, endpoint and exposure characteristic that can be used as a surrogate for the assessment endpoint. To assess impacts on the riverine clam, there may be information for a marine mollusk exposed in the laboratory. Surrogates are not perfect, so adjustment factors are applied to ensure that the measured effects data will not underestimate the toxicity to the assessment endpoint in the field. Clearly, the top-down and bottom-up approaches can yield the same results. They can be based on the same data. Perhaps the key difference is in how they are perceived by the regulators, the public and the environmental non-government organizations. The top-down approach carries the perception that details may be overlooked to achieve a sweeping recommendation. It has aspects of an opaque "trust us, we are experts" position. The bottom-up approach can convey the perception that details and site-specific issues are considered. The assumptions may be more transparent. The bottom-up approach appeals to an ecologically democratic perspective, where humans are considered as only part of the ecosystem. It may be the only approach when it becomes important to protect valued non-human individuals, such as endangered species and domestic pets. I believe that radioecological assessments will have to adopt the approaches common in ecotoxicology.

3. From foxes and frogs to fish An immediate problem in assessment of the impacts of radionuclides on nonhuman biota is the lack of data. There is an urgent need for bioaccumulation and effects data for all classes of organisms. In fact, the need will never be filled: there are far too many endpoints, species, contaminants and exposure characteristics. More importantly, from the assessment perspective, is to have means to extrapolate among endpoints, species, contaminants and settings. Radioecology has an exceptionally strong foundation in extrapolation among radioactive contaminants. The additivity of radiation dose estimates, accounting for relative biological effectiveness and other weighting factors, is unique among environmental contaminants. For extrapolation among endpoints, there are crude relationships that can be used (perhaps this will never be an elegant science). For example: • effect concentrations for reproductive endpoints are usually lower than for lethality endpoints; • ecosystem function endpoints are often less sensitive that single-species endpoints, because of ecological redundancy; • bioaccumulation is often related to the concentration of mobile chemical species, as opposed to total element concentrations;

66

S.C. Sheppard

• bioaccumulation is related to body mass, so that allometric equations can be used to extrapolate among species; • in plants, radiation sensitivity is related to chromosome volume. The use of allometric equations is common in ecotoxicology for mammals and birds. Figure 3.1 from Macdonald (1996) shows for caesium (Cs) the loglog relationship between the ingestion-to-tissue transfer factor (F) and animal body mass (BM). The data fit the allometric equation (P < 0.001, 2 = 0.78): F = 8.89 (BM)-°-73.

Figure 3.1. Relationship of feed-intake to mammal-flesh transfer factor (F) for caesium to body mass (BM) of the mammal (Macdonald, 1996). This is described by the allometric equation: F = 8.89 (BM)~°-73.

This can be used to estimate F for any mammal that is chosen as the assessment endpoint. Interestingly, Macdonald (1996) developed an allometric equation to estimate F for iodine (I), and it had a very similar exponent term. An interesting hypothesis is that the exponent term varies little among elements (radionuclides), and so allometric equations to predict F for all elements requires estimation only of the value of the multiplicand. This approach was used (Sheppard, 2000) in the models for reactor derived release limits where game were part of the diet of the critical group. Another extrapolation technique is analogous to the QSAR approach in ecotoxicology. The Quantitative Structure/Activity Relationships (QSAR) is an extrapolation tool used in ecotoxicology (and elsewhere) that takes what is known about an organic compound and estimates what impact it may have, in the absence of specific data. Boethling and Mackay (2000) provide an up-to-date book of such extrapolation methods. When the concept is transferred to inorganic contaminants, the structure of the compound may not be a useful variable, but the partitioning among environmental compartments is useful. In effect, the concept becomes a multiple regression equation to predict the unknown parameter from other parameters that are known.

3 - Toxicants in the environment

67

To derive F values for muskrat (Ondatra zibethicus), Zach et al. (1998) used stable element tissue concentration data for muskrat liver and F values for domestic animals to develop a multiple regression, that was then used to extrapolate F values for other elements. Specifically, in developing the regression, the (log) liver/feed concentration ratio observed in muskrat for 31 stable elements was the dependent variable. After stepwise regression, it was determined the best predictors were the (log) F values for beef, poultry and their interaction (product). The regression equation was then used to predict the concentration ratio for other elements in muskrat. These were converted to F values using estimated daily intakes by muskrat. The results of the allometric and multiple regression approaches are compared in Figure 3.2. Clearly, extrapolation is not desirable, and different methods will give different results, but the requirement for the information is urgent. It is appropriate for radioecology to explore the extrapolation tools used in ecotoxicology, and to expand the capabilities in this area.

Figure 3.2. Comparison of ingestion-to-tissue transfer factors (F in d/kg) for muskrat (Ondatra zibethicus) extrapolated across 24 elements using allometric equations (Sheppard, 2000) versus multiple regression based on stable element analogues (Zach et al., 1998).

4. Disappearance of the fence The previous two sections described issues dealing with the biotic endpoints to be assessed. A related issue is where the organisms to be protected may reside. When the objective in radioecology was to protect humans, the environment was divided into a worker (transient) environment and a public (resident) environment. Conceptually, these are often separated by the fence that borders the facility being assessed. The workers spend some hours each day inside the fence, and the public lives continuously outside the fence. Fences or other restraints do not control non-human biota. Birds may roost

68

S.C. Sheppard

and nest on the roofs of reactor halls. Small mammals and earthworms may burrow into contaminated earth. Fish may preferentially visit cooling water outfalls. An example of the degree of encroachment by biota was observed in a recent investigation of the ecological effects of in-situ abandonment of contaminated river sediments (Sheppard and Helbrecht, 2001). The site is a nuclear research facility. During characterization of the outfall, divers swam into the 1.2-m-diameter process water outfall pipe and were met by a school of White Suckers (Catostomas commersoni). The fish were probably attracted there by a food source, and were subject to undiluted effluent from the nuclear research laboratory. This illustrates that any access point may be encroached. Clearly, it is necessary to assess all organisms in all the contaminated environments they are able to access. The decision about assessment endpoints will imply what organisms and settings are to be assessed. Nonetheless, it is now much more important to be able to model the transfer of contaminants to biota in near-field settings. This has interesting ecological implications. For mobile organisms, what portion of the home range is contaminated? How does the habitat quality in the near-field affect exposure? Is the near-field exposure still a chronic exposure, or does it take on aspects of acute exposure? An illustration (Tab. 3.1) comes from a recent ecological effects review done for the Darlington fourreactor site (ESG International and ECOMatters Inc. 2001). The specific activity of 14C in air was computed at 50, 150, 300 and 1000 m from the release point, based on dispersion models for near- and far-field. At 150- to 1000-m distances, there is ecological habitat and it was assumed the organisms could be at isotopic equilibrium with the air, and the incremental dose above background (Tab. 3.1) is calculated accordingly. Of course, these positions are much closer to the source than are used for critical-group human exposure pathways. However, non-human biota can reside even closer. For illustration in Table 3.1, radiological dose was also computed for organisms at 50 m from the atmospheric release points. This is in the constructed area, and is arguably ecologically "affected" already, regardless of potential radionuclide effects. Also, there are few organisms that can reside here, but isotopic equilibrium could apply for epiphytic plants and invertebrates consuming those plants at this position. The dose estimates are high enough that effects are possible. As the assessment endpoint comes closer to point source contaminant releases, it is inevitable that exposure concentrations will exceed effect concentrations for individual organisms. Ecological principals become very important, because the assessment endpoints are usually populations. How many individuals of a population can be impacted before the population is impacted? This is excellent opportunity to integrate knowledge of the effects of habitat quality and dispersal of organisms into assessments. It may not be adequate to dismiss the issue because the near-field areas are "affected" habitat by the constructed facilities and therefore not ecological.

69

3 - Toxicants in the environment

Table 3.1. Estimated and observed specific activities of 14C in biota tissues near a four-reactor complex, starting at 50 m from the atmospheric emission source.

Distance from Concentration • Source of data source (m) in air(Bq/m^

0

monitoring

50

near-field model :

150

far-field model

300

far-field model

1000

far-field model

between 1000 asa observed in andthefeoce small maiturijai tissues

1.20 x 10-2 1.20 X

1

l0 1

2.30 x 10-1

1.00 x 10

-2

2.10 x 10

Specific activity, - Speciffe ;activityi not including Incremental with background background dose (Gy/a) (Bq/kg C) (Bq/kg C) -

4

4

467 x 100

1.44 x 10

3

1.69 x l0

8.95 x 10-2

6.25 X 102

8.75 x l02

3.89 x 10-2

2

2

8.17 x 10-3

7.50 x 10

3

7.53 x 10

1.31 x 10

3.81 x 10

**

3,50 x l02

Note: Internal dose conversion factor from Zach and Sheppard (1992), assumes all radiation deposited in the tissue.

5. Additivity of effects, and the RBE Ecotoxicology has struggled to deal with the interaction among contaminants and among the effects of contaminants. There are almost no singlecontaminant situations, and other stressors such as thermal effects need to be included. There is a generality that stressors that have different physiological modes of action will interact less than stressors that have similar modes of action. Ionizing radiation has modes of action that are fairly dissimilar to the modes of action of many other contaminants, and this may simplify what is otherwise a very complex issue. The additivity of radiation dose has already been described as a major strength coming from radioecology. However, it is still subject to debate. In particular, the relative biological effectiveness (RBE) is being re-examined. The RBE is used to adjust for the greater damage that results from alpha particle emissions compared to photon emissions. It is essential to note that this is not simply a function of the emission characteristics, for example, the linear energy transfer. It varies with effect endpoint (Fig. 3.3). The underlying experiments essentially match the effect on some endpoint of alpha exposure compared to the photon exposure giving the same effect. The results vary with endpoint (as well as with dosing methods and the reference photon energies). RBE values for alpha of 40 have been proposed in Canada (Bird et al., 2000), and there are values much higher. Setting these values has important implications because it may again open the question of the RBE to be used for human dose assessment.

S.C. Sheppard

70

Grouped by type of endpoint Figure 3.3. Variation in Relative Biological Efficiency (RBE) of alpha emitters as reported in the literature and summarized by Bird et al. (2000). The RBE are grouped into three endpoint types: cell mortality on the left, genetic effects in the middle, and sub-lethal effects on the right. In general, there is less uncertainty and lower RBE values for cell-mortality endpoints, and more uncertainty and apparently higher RBE values for the more ecologically relevant sub-lethal physiological endpoints. Bird et al. opted for an RBE value of 40 for non-human biota.

There is an alternative to using RBE, and that is to follow ecotoxicological practices and define different effect levels for the different types of radiation. For example, it would be possible to define an effect level for alpha exposure, and another for photon exposure. I believe there is a general consensus to not do this, but instead to persist with the RBE concept and develop new RBE values for ecological endpoints. This is clearly a role for the international consensus organizations. The RBE is essential to maintain the additivity of radiation doses, a very desirable situation.

6.

Model mobile element and not total element?

In both radioecology and ecotoxicology, there is recognition and considerable research on the affect of bioavailability on exposure. Much of the internal dose to an organism is the result of absorption of the bioavailable radionuclide in the environment. Plants are a good example, where plant uptake

3 - Toxicants in the environment

71

is from soil solution, and only the mobile and bioavailable fraction of the elements in the soil are in soil solution and have the potential to be absorbed. It is for this reason that some probabilistic assessments invoke a negative correlation between the plant/soil concentration ratio and the soil solid/liquid partition coefficient, Kd (Sheppard and Sheppard, 1989). The less sorbed the element, the more likely it is taken up by plants. A recent assessment of heavy metal emissions from the mining industry (Sheppard et al., 1999) converted the entire exposure and effects characterization to a free-ion basis. This involved both geochemical speciation and sorption models applied to the estimated environmental concentrations and to the literature information on effects. An interesting note made in this study was that modeling the accumulation of soluble element concentrations in soil is much simpler than modeling the accumulation of total element concentrations. This results from the assumption that only the soluble element is available for migration in the soil. With this assumption and using a linear sorption model such as Kd, the steady state soluble element concentration in soil is the same for all elements (Fig. 3.4). This is because the concentrations that accumulate in the soil depend on the loss by leaching of the element in pore water solution. The steady state concentration is therefore established by the dissolved element, and the amount on the solids is only the support needed to maintain that pore water concentration. Of course, the time to reach steady state, and the total element concentrations, vary a great deal among elements, depending on the Kd values. The simplicity of this lies in the assumption of the linear sorption Kd model, but given that many assessments make that assumption, there are an important and practical implications of this observation.

Figure 3.4. Estimated concentration of soluble element in soil (mass per m3 soil pore water) with time under a continuous unit deposition (mass per m2 per annum). The elements vary because of different solid/liquid partition coefficients (Kd).

72

S.C. Sheppard

7. Chemical toxicity of radionuclides Despite the strong basis for the additivity of radiation dose, there is another effect mechanism to consider. Radionuclides with very long half-lives have such low specific activities that they may be more toxic chemically than radiologically. This may well be true for 129I for non-human biota (Sheppard and Evenden, 1995). In this paper, the level of stable I in soil found to be chemically toxic, if it were present purely as 129I, would not have posed a radiation hazard to non-human biota. It is commonly assumed that natural uranium (U) is more toxic chemically than radiologically. This excludes the impacts of some of the U decay chain radionuclides. Uranium is especially toxic to renal function (Leggett, 1989; Paternain et al., 1989). On the issue of chemical toxicity of U, ecotoxicology and radioecology are on exactly the same footing because radiation is not a factor. An interesting example of the role of U ecotoxicity has emerged from the setting of environmental quality guidelines for airborne releases (Tab. 3.II). The air quality guidelines were set to protect humans and the environment based on the potential toxicity of U to plants. Airborne U was not found to impact humans or non-human biota. The build-up of U in soil resulting from deposition from the atmosphere was also not found to impact humans. However, the concentrations in soil could exceed the concentrations known to be chemically toxic to plants. Table 3.H. Proposed air quality standards for uranium in Ontario (Fleming et al., 2000), an example of an environmental cleanliness standard for a radionuclide being set based on an ecological rather than human health endpoint.

ladpoint

Exposure pathways

Human health

inhalation and ingestion by critical group, intake limit of 0.6 micro grams/kg/d (Health Canada, 1999) deposition to soil over 50 years,, phytotoxic limit of 300 mg/kg soil (Sheppard etal.,1992)

Phytotoxicity

Potential standard (micro grams/m3 air, 24-hour average) 0.48

0.075

8. Conclusions Radioecology and ecotoxicology are important disciplines, both dealing with the effects of contaminants on the environment. From their traditional strengths, they approach the protection of the environment from radionu-

3 - Toxicants in the environment

73

elides in different ways. Inevitably, the two approaches will merge, and the outcome will benefit from the strengths of both disciplines. The present direction seems to lead to a bottom-up approach to identify the environmental components to be protected, an ecotoxicological method. It will benefit from the vast amount of radioecological information on radionuclide transfers and additivity of radiation exposures.

Acknowledgements The author gained considerable insight into these issues through discussions with Drs. Glen Bird, Pat Doyle, Nava Garisto, Don Hart, Colin Macdonald and Patsy Thompson. Funding for the various projects used as examples came from Atomic Energy of Canada Limited, Environment Canada, and Ontario Power Generation.

References Bird G.A., Thompson P.A., Macdonald C, Sheppard M.I. and Sheppard S.C. (2000) "Supporting document for the Priority Substances List assessment of releases of radionuclides from nuclear facilities (impacts on nonhuman biota)", Environment Canada, Hull, Quebec. Boethling R.S. and Mackay D. (2000) "Handbook of Property Estimation Methods for Chemicals", Environmental and Health Sciences (Lewis Publishers, Boca Raton). ESG International Inc. and ECOMatters Inc. (2001) "Darlington Nuclear Generating Station ecological effects review", Report G0544 (draft) for Ontario Power Generation (Toronto). Fleming S., Marsh M., Wagenaar A. and Bloxam R. (2000) "Ambient Air Quality Standards for Uranium", Working Draft (Ontario Ministry of the Environment, Toronto). Leggett R.W. (1989) "The behavior and chemical toxicity of U in the kidney: a reassessment", Health Physics, 57, 365-383. Macdonald C.R. (1996) "Ingestion rates and radionuclide transfer in birds and mammals on the Canadian Shield", Atomic Energy of Canada Limited Technical Report, TR-722, COG-95-551. Paternain J.L., Domingo J.L., Ortega A. and Llobet J.M. (1989) "The effects of uranium on reproduction, gestation and postnatal survival in mice", Ecotoxicology and Environmental Safety, 17, 291-296. Sheppard S.C. (2000) personal communication, taken from contributions to July 2000 draft of "Guidance for Calculation of Derived Release Limits for Radionuclides in Airborne and Liquid Effluents from Ontario Power

74

S.C. Sheppard

Generation Nuclear Facilities" (Ontario Power Generation, Toronto). Sheppard S.C., Evenden W.G. and Anderson A.J. (1992) "Multiple assays of uranium toxicity in soil", Environmental Toxicology and Water Quality, 7, 275-294. Sheppard S.C. and Evenden W.G. (1995) "Toxicity of soil iodine to terrestrial biota, with implications for 129I", /. Environ. Radioactivity, 27,99-116. Sheppard S.C. and Helbrecht R.A. (2001) "Feasibility of in-situ abandonment of contaminated Winnipeg River sediments at the Whiteshell Laboratories process water discharge: An evaluation of human health and ecological risks", IAEA Technical Document (submitted). Sheppard S.C. and Sheppard M.I. (1989) "Impact of correlations on stochastic estimates of soil contamination and plant uptake", Health Physics, 57, 653-657. Sheppard S.C., Sheppard M.I. and Bird G.A. (1999) "Critical load modelling: Cd, Cu, Ni, Pb, Zn and As emitted by smelters and refineries", Report for Environment Canada addressing CEPA PSL2 assessments (Ottawa). Suter G.W. (1993) "Ecological Risk Assessment" (Lewis Publishers, Boca Raton). Zach R. and Sheppard S.C. (1992) "The food-chain and dose submodel, CALDOS, for the assessment of Canada's nuclear fuel waste management concept. AECL report", AECL-10165, COG-91-195. Zach R., Rowat J.H., Dolinar G.M., Sheppard S.C. and Killey R.W.D. (1998) "Ecological risk assessment for the proposed IRUS low level waste disposal facility at AECL's Chalk River Laboratories", Atomic Energy of Canada Limited Report, TR-7.

4

Post accident management G.N. Kelly1,2

Effective post-accident management depends first and foremost on sound policy and practical means for its implementation. The lack of, or relatively poorly developed, policy or guidance on post-accident management was largely responsible for the disparate response across Europe to the Chernobyl accident and for the difficulties in reaching agreement on matters of international importance (e.g., trade in contaminated foodstuffs). Much has been learned from the post Chernobyl experience, both in terms of policy development and its practical implementation. However, recent experiences with the management of BSE (Bovine Spongiform Encephalopathy) and GMO (Genetically Modified Organims), in particular, exemplify the difficulties of decision making in complex social and political environments. This paper reviews the progress that has been made since the Chernobyl accident in the development of policy for post-accident management and how it can be implemented. Consensus on policy in this area does, however, remain elusive and the remaining difficulties are addressed. Insufficient attention has been given in the past to the social and political aspects of post-accident management and some of the more recent and highly promising developments are discussed. Rapid and reliable characterisation of contaminated areas is important for coherent and effective post-accident management and the advances that have been made in this area are described. Sources of information on the technical options for managing contaminated environments, including their efficacy, cost, practicability, etc., are summarised.

1. Introduction The Chernobyl accident had a profound effect on emergency preparedness and post-accident management world-wide and, in particular, in Europe. 1

The views expressed are those of the author and not necessarily those of the European Commission 2 European Commission, DG Research, Brussels, Belgium.

76

G.N. Kelly

Deficiencies in arrangements for dealing with such a large accident, at both national and international levels (e.g., in world trade in foodstuffs), led to many problems of a practical and political nature. Many lessons were learnt and considerable resources have since been committed to improve emergency preparedness and post-accident management in order to avoid similar problems in future. Improvements have been made at national, regional and international levels and have been diverse in nature. Some of the more notable at an international level are the convention on early notification (IAEA, 1986), limits for the contamination of foodstuffs in international trade (FAO/WHO, 1991) and broad agreement on the principles of intervention (IAEA, 1994), albeit less so on their practical interpretation. At a regional level, many bi- and multi-lateral agreements or regulations have been brought into force for the timely exchange of information (e.g., Council Decision, 1987a) and the efficacy of such arrangements is increasingly being demonstrated in regional exercises. At a national level, the improvements have been diverse, ranging from the installation of extensive networks of gamma monitors to provide early warning of an accident to more robust and efficient arrangements between the many organisations with a role or responsibility in an emergency and the longer term post-accident management. Some fifteen years after Chernobyl, it is timely to reflect on what has been achieved and whether there remains a need for further improvement. Particular attention is given to the adequacy of policy for intervention and the extent to which arrangements for its practical implementation are sustainable given the rarity of nuclear accidents. The respective roles of technical and broader social and political considerations in shaping policy are addressed, in particular in the context of the marked differences between current international guidance and how accidentally contaminated areas are being managed in practice. Insufficient attention has been given in the past to the social and political aspects of post-accident management. Some of the issues that need to be given greater attention in future, for the development of sound and practicable policy, are addressed. Consideration is finally given to the technical options for managing contaminated environments and a number of key references that provide information and guidance on the efficacy, cost, practicability, waste management implications, etc., of the different options are identified. Consideration is limited here to post-accident management issues (i.e., excluding urgent measures taken during the accident or which are likely to be of short duration). While there is no universally agreed definition of what constitutes "post accident", it is assumed here to start when the release of radioactive material from the affected installation has ceased and to end when no further measures are being taken to mitigate the consequences of residual environmental contamination.

4 - Post accident management

77

2. Policy for and guidance on intervention A sound policy, and robust planning for its practical implementation, are essential for effective post-accident management. Prior to the Chernobyl accident there was little or no clear guidance, at national or international levels, for post-accident intervention in response to very large accidents (i.e., where the geographical extent and/or duration of the intervention or postaccident management measure (e.g., food restrictions, relocation) could be large). Policy, therefore, had to be developed on a relatively ad hoc basis and there was little opportunity for considered reflection. One consequence was that intervention was often less than optimal, with some measures being taken that were probably not fully justified and vice versa. For example, debate still continues on the merits or otherwise of the decision in the Former Soviet Union to relocate permanently hundreds of thousands of people from settlements contaminated as a result of the Chernobyl accident. The lack of broadly agreed guidance was also largely responsible for the varied response (in terms of the post-accident measures taken) across Europe. This caused some loss of confidence and trust among the affected populations in the actions taken by their authorities and contributed to anxiety that could, otherwise, have been avoided or at least reduced. Major improvements, especially in Europe, have been made to emergency preparedness and post-accident management since the Chernobyl accident. Notwithstanding these improvements, managing the aftermath of an accident will always be difficult. This is inevitable given the social and political context of any accident (e.g., public concern and anxiety, political reaction, recriminations over its cause, culpability), particularly those having a nuclear origin. These difficulties would escalate in the post accident phase (i.e., in the weeks and months and, in extreme cases, years following an accident) when the measures taken to protect those affected would be exposed to the full gaze of the public, the media and political scrutiny. These difficulties would arise irrespective of how well conceived and justified the intervention measures were and means for addressing them should form part of effective post-accident management. In the following sections, the development (both pre- and post-Chernobyl) and current status of international policy and guidance on post-accident management is described and is contrasted with actual practice in the management of accidentally contaminated areas. The reasons for the differences between international guidance and practice are identified, together with suggestions as to how this gap may be closed in future.

2.1. International Guidance Following the Chernobyl accident, much effort at an international level was directed towards developing a sound and broadly agreed policy for postaccident management and guidance on its implementation. This process has

78

G.N. Kelly

been lengthy and not without difficulty and remains unfinished. While broad agreement has been reached on the principles3 underlying decisions on intervention, important differences remain on how these principles are/have been interpreted and/or applied in practice. The absence of, or limitations in, policy and/or guidance on post-accident management, at national and/or international levels, prior to the Chernobyl accident is at first sight surprising. The potential for large accidents was well known from numerous risk assessments carried out in the 1970s and 1980s (e.g., USNRC 1975; Kelly et al, 1982) and the particular problem of caesium contamination had long been recognised. Indeed, in these risk assessments, plausible assumptions had to be made on the post-accident measures that the relevant authorities might be expected to make in order to protect the population from the long term effects of radioactive material deposited in the environment. Substantial research had also been carried out on caesium decontamination by a number of organisations by the late 1970s, in particular of urban surfaces, with seminal work by the Riso laboratory in Denmark (Gjorup, 1982). In principle, therefore, the technical basis existed in the late 1970s/early 1980s for establishing policy for post-accident management and how it should be implemented in practice. The reasons why such policy and/or quantitative guidance was not developed or implemented in most countries is a matter for speculation. The following may have been influential: the absence of any broadly agreed international policy or guidance; the lack of a perceived need (i.e., the possibility of nuclear accidents - in particular those with other than a local impact - being judged to be exceedingly remote); the inherent difficulty in developing broadly applicable quantitative guidance on post-accident measures, such as relocation, where the size of an accident (or the extent of the contamination) could be expected to have a major influence on the level (of dose or contamination) at which it might be implemented (Kelly et al., 1983); and a generally held, but possibly mistaken, view that guidance on such measures would be better left to the aftermath of an accident when, in principle, time would be available to both consider and take due account of the prevailing circumstances. Intervention levels (or emergency reference levels as they are known in the United Kingdom) were first established in the late 1950s, following the Windscale accident, by the UK Medical Research Council. Similar concepts and levels for intervention following an accident were subsequently adopted by many countries and incorporated into their national guidance or regulations. Prior to the 1980s, little attempt was made to achieve any broader 3

Decisions on intervention are based on three simple principles: prevent serious deterministic health effects; the intervention should be justified, in the sense that it should achieve more good than harm; and the level at which the intervention is introduced (and at which it is later withdrawn) should be optimised, so that it will produce a maximum net benefit.

4 - Post accident management

79

international consensus on the choice of levels for different types of intervention. The accident at Three Mile Island in the USA in 1979 provided, however, a timely stimulus for review and improvement of emergency arrangements in many countries, including the development of more broadly agreed guidance at an international level. In the early 1980s the European Commission published guidance (EC, 1982) from its Article 31 Group of Experts on intervention following nuclear accidents. This guidance contains numerical intervention levels for the more urgent countermeasures, in particular sheltering, evacuation and the issue of stable iodine. Quantitative guidance was not, however, proposed for intervention in respect of contaminated foodstuffs nor for external doses that may be received over the long term from residual ground contamination. In these cases, it was judged that more time would be available and account could be taken of monitoring results, etc., in deciding on the nature and extent of intervention. Nonetheless, it was recommended that intervention levels for such measures should be developed by those responsible nationally and that they should be included in emergency plans for specific sites. This guidance from the Article 31 Group proved to be most useful in promoting a common understanding of the principles underlying intervention and in achieving a measure of uniformity in the levels and policy being adopted in Member States of the European Union (EU). A succession of similar publications followed from various international organisations, in particular the International Commission on Radiological Protection (ICRP, 1984) and the International Atomic Energy Agency (IAEA, 1985). In addition to intervention in the early phase of an accident, these publications also provided quantitative guidance on intervention in the intermediate phase, in particular for the control of foodstuffs and relocation of people from contaminated areas. The content of all these documents, in particular the principles on which intervention should be based and the quantitative levels proposed, were broadly the same, even identical in some cases. Each of these documents was, however, prepared largely by the same group of experts and the unanimity achieved should not necessarily be taken as representative of that among the broader international community. The guidance contained in these documents was well conceived and proved to be of practical value in establishing intervention policy in some countries. With the benefit of hindsight, the guidance could be criticised in some areas. However, its existence was of much value in the immediate aftermath of the Chernobyl accident. Given the apparent unanimity of view expressed in this international guidance, the large differences in national responses to the Chernobyl accident, in terms of post-accident measures taken, may seem surprising. Indeed, in the absence of the Chernobyl accident, the international guidance would probably have remained largely unchanged for a substantial period as, conceptually, it was well founded. Four factors contributed to this unanimity (with regard to the principles and the quantitative guidance on

80

G.N. Kelly

levels for intervention) not being reflected in practice. Firstly, few countries had, prior to the Chernobyl accident, formally applied the principles to the establishment of (or revision of existing) intervention levels particular to their own circumstances. Secondly, the international guidance had largely been conceived in the context of relatively small (design basis) accidents where the extent of intervention would be limited in both distance and time; the question of how to deal with very large accidents, having an impact over large areas (even trans-boundary), had not been addressed explicitly. Thirdly, for each type of intervention, a range (with the bounds of the range differing by a factor of ten) of values was recommended requiring judgement on the part of the user to determine the level appropriate to his/her particular circumstances. Fourthly, the guidance contained some ambiguities in the quantities used and how they were to be applied in practice. Following the Chernobyl accident there was much activity internationally in developing revised guidance on intervention. Soon after the accident the EC introduced a Regulation (Council Regulation, 1987b) controlling the content of radionuclides in foodstuffs that could be imported into the EC from third countries. This Regulation, in an amended form, still remains in effect. A further Regulation (Council Regulation, 1987c) was subsequently introduced to control the content of radionuclides in foodstuffs that could be marketed following any future accident in the Union. This Regulation contains specific limits on the radionuclide content of different food types but contains provision for changes in these levels in light of the circumstances prevailing at the time of an accident and for intervening changes in scientific knowledge. The World Health Organisation issued guidance on the control of foodstuffs (WHO, 1988) and the Food and Agriculture Organisation together with World Health Organisation (FAO/WHO, 1991) subsequently developed guidance on the levels of radionuclides in foodstuffs that could move freely in international trade. The IAEA also revised its guidance in light of the Chernobyl experience. Interim, informal, revision to its guidance (IAEA, 1988) was issued in 1987 and this was followed by a more substantive revision in 1990 (IAEA, 1991). Further revision followed in (IAEA, 1994) and this remains the Agency's current guidance on the subject, albeit supplemented by its Basic Safety Standards (IAEA, 1996). ICRP, after a long process, amended its guidance in (ICRP, 1993), bringing it fully into accord with its revised system of protection set out in (ICRP, 1990) where an important distinction is drawn between the control of practices and intervention. The European Commission published guidance, developed by its Article 31 Group of Experts, on relocation and return of people after a nuclear accident (EC, 1993); this guidance is broadly comparable with that in (ICRP, 1993). Further guidance on intervention can be found in (ICRP, 1999) where a unified approach is proposed for the protection of the public in situations of prolonged exposure, irrespective of whether the source is of a natural or artificial origin.

81

4 - Post accident management

It is evident from the large number of international publications that much effort has been expended over the last one to two decades on the principles of intervention and guidance on their application. It has also been a time of continual change - not so much in the principles which have largely remained unaltered - but in the guidance on intervention levels, both numerically and in the quantities in which they are expressed. For those actively participating in the development and/or revision of the policy and/or the guidance, it has been a fascinating and engaging process. For those who were interested solely in the practical use of the guidance or were affected by its implementation (i.e., those living in, or consuming foods from, a contaminated area), the continuing change can only have been a source of frustration, confusion and concern. The extent to which the quantitative guidance offered by just one of the international organisations has changed over the period is illustrated in Table 4.1. Caution should be exercised in attempting to make rigorous comparisons or draw rigorous conclusions from the Table. Some of the quantities are not identical and many have various qualifications attached to them in terms of how they are to be applied and/or transformed for specific, as opposed to, generic use. What is clear, however, from Table 4.1 is the scale and nature of change that has occurred over a relatively short period (less than a decade). It is not surprising, therefore, that much confusion/concern still reigns over this issue.

Table 4.1. The development of quantitative generic guidance by IAEA on post-accident intervention. Intervention

Safety Series 72 (IABA, 1985)

Safety Series 72 (rev) CIAE&, 1991)

Safety Series 109 (IAEA, 1994)

Withdrawal of foodstuffs

5-50 mSv in the first year

About one to a few tens of mSv in a year

Values expressed in terms of activity concentration of different nuclide groups in food1

Relocation

50 - 500 mSv in the first year

About a few to a few hundred mSv in a year

30 mSv in the first month for Temporary Relocation2 1 Sv in a lifetime for Permanent Relocation

1 For example, 1 kBq/kg for 137Cs and 131I - the corresponding annual dose would, typically, be about a few mSv assuming that the whole diet was contaminated at these levels. Lower concentrations would be appropriate for less costly interventions than withdrawal of food, i.e., replacement fodder for animals, soil treatment to reduce uptake of radionuclides, etc. 2 Return when the dose in a subsequent month falls below 10 mSv provided the lifetime dose does not exceed ISv.

82

G.N. Kelly

The generic intervention levels for temporary and permanent relocation in (IAEA, 1994 - para. 436) were derived on the basis of radiation protection principles alone; broader issues of a social, ethical, psychological, political or cultural nature were deliberately excluded. Simple cost-benefit analyses were used in the derivation of intervention levels. For each intervention, a balance was made between the cost of the intervention and the reduction in risk from the resulting radiation dose averted (assuming a cost per unit radiation dose). A range of "optimised" levels was established for each intervention (the range reflecting uncertainties in the inputs to the cost-benefit analysis) and the generic intervention level assigned a value towards the middle of this range. The potential importance of other "non-radiation protection" factors and the site- and accident-specific situation is, however, fully recognised in (IAEA, 1994), as is the need for them to be taken into account by the relevant national authorities in establishing intervention levels appropriate to their particular circumstances. The same basis was used to determine "optimised" intervention levels for foodstuffs. However, other considerations influenced the selection of the generic intervention levels. Achieving consistency and simplicity in application and, in particular, compatibility with the levels in (FAO/WHO, 1991) were the driving factors rather than selecting values towards the middle of the range of the "optimised" levels, as was done for other types of intervention. Several factors influenced this choice: the advantages that would accrue in terms of maintaining trust and confidence in the authorities by using internationally recognised values; the impracticality of having levels for national use that differed from those in international trade; and avoiding potential anomalies that may occur between neighbouring states. More recently, ICRP has issued recommendations on the protection of the public in situations of prolonged radiation exposure (ICRP, 1999); some aspects concern intervention in respect of long term exposure from accidental contamination of the environment. The recommendations provide, for the first time, a largely unified and consistent approach for protection from all sources of prolonged radiation exposure, whatever their origin (e.g., regulated practices, natural sources, radioactive residues from past regulated or unregulated practices, accidents). As such, they have much to commend them and have the potential to provide a valuable framework for a more consistent approach to the control of such sources in future. The length of the report and complexity of some of its arguments and recommendations may, however, hinder its rapid translation and/or adoption into practice. The report reinforces the important distinction made in (ICRP, 1990) between practices and intervention and sets out the principles of the system of radiation protection that should be applied to each. Justification of intervention is "to be assessed using a decision-aiding process requiring a positive balance of all the relevant long-term attributes related to radiological protection (in addition to the avertable doses, both individual and collective, other attributes include the following: the expected reduction in anxiety caused by the situation, the reassurance to be provided by the intervention, the social

4 - Post accident management

83

cost, harm and disruption that may be caused by the implementation of the protective actions). The results of the decision-aiding process should be used as an input into a decision-making process which may encompass other considerations and may involve relevant stakeholders". Optimisation of protective actions is recommended to be performed following the general approach to optimisation of protection recommended by ICRP in the context of practices, with the optimum scale, extent and duration of the protective action being selected from the justified options. National authorities (and international authorities as appropriate) are recommended to pre-determine specific reference levels (such as intervention levels) for particular prolonged exposure situations amenable to intervention; the scale, duration and extent of intervention should be determined on a case by case basis following the recommended principles of justification and optimisation. Generic reference levels (see Tab. 4.II), expressed in terms of the existing annual dose (from all sources), are also proposed for use as guidance, where necessary, in the above process. Several important caveats are placed on the use of the generic reference levels; the reader is referred to the original text for their detailed nature and content. Subject to these caveats, the report considers that "an existing annual dose approaching about 10 mSv may be used as a generic reference level below which intervention is not likely to be justifiable for some prolonged exposure situations. Below this level, protective actions to reduce a dominant component of the existing annual dose are still optional and might be justifiable. In such cases, action levels specific to the particular component can be established on the basis of appropriate fractions of the recommended generic reference level. Above the level below which intervention is not likely to be justifiable, interTable 4.II. Quantitative recommendations for generic reference levels for intervention (ICRP, 1999). Cpncefrt

Qwanttfj

Generic reference level for interventions almost always justifiable (above which intervention should almost always be considered justifiable)

Existing annual dose (summation of all [prolonged] annual doses attributable to all sources of prolonged exposure in a given location)

Generic reference level for interventions not likely to be justifiable (below which intervention is optional but not likely to be justifiable and above which it may be necessary)

Existing annual dose (summation of all [prolonged] annual doses attributable to all sources of prolonged exposure in a given location)

Vitae t»S¥fy)

=s 100

=s 10

84

G.N. Kelly

vention may possibly be necessary and should be justified on a case by case basis. An existing annual dose rising towards 100 mSv will almost always justify intervention and may be used as a generic reference level for establishing protective actions under nearly all conceivable circumstances". The report makes one further important qualification on intervention in respect of radioactive residues. "Where the origins of the situation are traceable and where those who produced the residues can still be made retrospectively liable for the protective actions, national authorities may consider applying a specific restriction to the individual doses attributable to the residues, constraining the resulting doses to levels below those resulting from the optimisation process. For this purpose, additional protective actions may be required from those who created the situation. Such specific dose restrictions, however, may be higher than the dose constraints or dose limits applied to practices". The recommendations are based on sound and clear principles and a practical and flexible framework has been developed to guide their proper application. This flexibility provides much scope and freedom for interpretation, in particular in the justification and optimisation processes - so much so that some might argue that any outcome is possible subject to the relative weights given to the various inputs to these processes. In some cases the guidance is sufficiently equivocal to help bridge the gap between earlier international guidance (e.g., IAEA, 1994) on post accident intervention and what is actually happening in practice in contaminated areas (see Sects. 2.3 and 2.4); whether this was intentional or accidental is not clear - whatever, it is no bad thing. Two detailed aspects of the guidance are, however, matters of concern. These are the distinction drawn between "decision aiding" and "decision making" and the implied role of the radiation protection specialist in providing advice/judgements on intervention in complex areas such as post accident management. Those drafting the report see themselves as "providers of decision aiding recommendations mainly based on scientific considerations on radiological protection; the outcome of their advice is expected to serve as an input to the final (usually wider) decision making process which may include other societal concerns and considerations. The decision making process may include the participation of relevant stakeholders rather than radiological protection specialists alone". This distinction raises three important concerns. Firstly, there is an implication, albeit tacit, that the input from the radiation protection community (because it is mainly based on scientific considerations) has a different status to other inputs. Secondly, compartmentalisation of the issue into the so called "radiological protection" and "other" factors will hinder broad agreement being reached on complex issues such as post accident management, in particular, it will encourage polarisation of the interested parties. Thirdly, decision aiding (as well as decision making) processes, if they are to be effective, must be inclusive, i.e., include from the outset all

4 - Post accident management

85

considerations seen as relevant by the various parties to the decision. Formulating advice on the basis of a sub-set of the relevant considerations will be seen by many as an attempt to contrive or constrain the outcome of the decision making process. All of these aspects militate against fair process which is a pre-requisite for broad agreement on complex issues that may affect society at large. A further concern is the scope of the attributes claimed to be related to radiation protection and which, by implication, have been taken into account (mainly on the basis of scientific considerations) in developing the recommendations and guidance in the report. These attributes include the expected reduction in anxiety, re-assurance provided by intervention, social harm and disruption caused by intervention. By including these attributes in the decision aiding recommendations in the report, there is an implication that judgements on these matters are solely, or at least largely, the preserve of the radiation protection specialist. Many would dispute this on two grounds: firstly, that others are far better placed to provide input to, and exercise judgements on, such matters; and, secondly, the level of competence and expertise within the radiation protection community on such matters. These issues are symptomatic of a deeper problem that has emerged within the radiation protection community. There has been a tendency for it to try and take the leading role in every situation where radiation protection has had a role to play, even when radiation protection issues are but one of many considerations and rarely the most important. This is discussed further in Section 2.3. Further reflection is needed within the radiation protection community as to its role in more complex societal issues such as post accident management where it is but one of many constituents in the process. Recent trends in the risk governance field (TRUSTNET, 2000) may provide some useful insights in such a reflection.

2.2.

Practice in the Former Soviet Union and Comparison with International Guidance

It is instructive to contrast the international guidance with practice in the Former Soviet Union (FSU) where the impact of the Chernobyl accident, and the need for intervention, were greatest. Prior to the break-up of the FSU, a criterion of 350 mSv was introduced as the basis for making decisions on the further relocation4 of people from contaminated settlements. The dose to be compared with this criterion was the lifetime dose from the accident (taken as that arising over a 70 year period). The choice of 350 mSv was based on a

4

I.e., in addition to those who had previously been evacuated and re-settled in the immediate aftermath (days and weeks) of the accident.

86

G.N. Kelly

judgement that no medical consequences would be detectable in a population exposed at or below this level; it is also the product of 70 years and the then current dose limit (5 mSv/a) for members of the public. Those expected to receive doses in excess of this level were to be permanently relocated. In settlements where the expected dose was less than this level, all further intervention was to cease (to minimise further stress and disruption) and life "returned to normal". This criterion met with strong resistance not wholly unconnected with the dissolution of the FSU. Further information on this dose concept and its limitations can be found in (IAC, 1991). In response to this resistance, new policies were developed in each of the three countries for the long term management of contaminated settlements (Balonov, 1999). In 1991, Russia adopted a new law (RSFSR, 1991) which required the obligatory resettlement of people living in areas where the average annual dose exceeded 5 mSv (or the average level of contamination of 137 Cs exceeded 1.4 MBq m~2). A new concept was subsequently developed in 1995 by the Russian National Commission on Radiation Protection (NCRP, 1995) with the following features: in settlements where the estimated annual average dose is less than 1 mSv, no restrictions were to be applied; where the dose is in the range of 1 to 5 mSv (the zone of radiation control) the population should be monitored and countermeasures taken on the basis of the optimisation principle; where the dose is above 5 mSv monitoring and countermeasures should also be performed and assistance given to those who voluntarily wish to resettle. This concept was taken up in a new law (RF, 1996) in which an annual average dose of 1 mSv has a central role. Firstly, it is the level above which optimised countermeasures are required and, secondly, it is the level below which people can be returned to settlements from which they were relocated. Broadly similar trends in policy and practice are evident in Belarus and Ukraine, but there are differences in detail. Major differences are evident in the standards applied in Russia (and elsewhere in the FSU) and the international guidance. Obligatory or voluntary resettlement for average annual doses in excess of 5 mSv are much lower than the relocation criterion of 30 mSv per month for temporary relocation and 1 Sv for permanent relocation proposed in (IAEA, 1994). Likewise, an average annual dose of 1 mSv for return to resettled areas is much lower than that of 10 mSv in a month in (IAEA, 1994). These differences reflect very different views on the long term management of contaminated areas. The international guidance clearly distinguishes between practices and intervention and is adamant that the dose limits for controlling the former are not, per se, appropriate for the controlling the latter which need to be assessed independently. In Russia, and elsewhere in the FSU, it has been judged that, in the longer term after an accident, it is reasonable for the population to demand the application of the same standards of protection as the rest of the population. This position is not unique. In a recent analysis of case studies of contaminated sites, (Hedemann Jensen, 2000) concluded that in several cases (e.g., Goiania accident, Rongelap Island contaminated by

4 - Post accident management

87

nuclear weapons tests, Chernobyl accident) the dose limit or lower was used as the criterion for restoration of contaminated areas. Two contrasting views can be taken over the apparent contradictions between the quantitative international guidance and experience and practice with post accident management. Firstly, the principles of intervention are being mis-applied, in particular because of the use of dose limits for practices for interventions. Secondly, that the practice is fully in accord with the principles of intervention, in particular the inclusion of other social, cultural, ethical, etc., factors (which are excluded from the quantitative generic levels derived internationally) fully justifies intervention at levels far below the generic guidance. The standards applied to the control of foodstuffs are also more restrictive than those developed for the free movement of foodstuffs in international trade (FAO/WHO, 1991). In Russia, the permissible levels for caesium-137 in milk and meat are 370 and 600 Bq/kg, respectively (compared with values in (FAO/WHO, 1991) of 1000 Bq/kg). In Ukraine and Belarus, there is evidence of a trend towards even lower levels of the order of 100 Bq/kg in some circumstances. This again represents a desire to improve the level of protection as time progresses. Whether such measures are optimal, given the socio-economic situation in the countries involved, is open to question. However, they presumably reflect a perceived social or political need.

2.3.

Reconciliation of theory and practice possible ways ahead

The principles underlying intervention, including the important distinction between practices and intervention, are broadly accepted within the radiation protection profession. Some, however, have reservations on the detailed interpretation of these principles, in particular in the derivation of generic intervention levels. These principles and generic guidance are, however, less readily accepted in the wider community. This results from the difficulties the public (and/or their political representatives) have in understanding or appreciating the use of different standards before and after an accident, i.e., for practices and intervention. While the logic of ICRP in making a clear distinction between practices and intervention is considered by the radiation protection profession to be largely impeccable, it is evident that such a distinction is untenable for others. There is a further, perhaps more important, problem with the generic guidance that has so far been developed. This concerns the somewhat artificial separation of the factors that have been taken into account in deriving generic quantitative guidance; moreover, on occasions there has also been a lack of consistency and clarity in what has and has not been included in such processes. The two most important principles underlying intervention can,

88

G.N. Kelly

in simple language, be expressed as follows: intervention should only be undertaken when it does more good than harm and the level at which it is introduced should achieve the most good. These are very simple and laudable objectives with which it would be difficult to disagree; putting them into practice, however, in a consistent and broadly acceptable manner presents more difficulty. It is clear that social, cultural, ethical and political values will have a major influence on what constitutes good and harm and the respective balance or trade-off between them. Significant differences can thus be expected in the application of these principles in countries with different historical backgrounds, value systems and economic status. Moreover, sound and broadly acceptable judgements on such matters would require inputs from a wide spectrum of society and could never be the preserve of one or other narrow technical community. In addition to providing guidance on the principles underlying intervention, international organisations have judged that guidance on their implementation is also needed and, in particular, the provision of quantitative generic intervention levels. It would be difficult not to agree with this judgement; some smaller countries may not have the resources to carry out such tasks themselves and there is much to be gained, in terms of public confidence and trust, in promoting the use of common intervention levels, at least in neighbouring countries. Establishing quantitative generic guidance has, however, proved difficult for the reasons set out above. Indeed it has only proved possible by artificially constraining the problem, in particular limiting consideration to a sub-set of the relevant issues (and not necessarily the most important) that need to be addressed when making judgements on whether intervention would achieve "more good than harm and the most good". The generic intervention levels in (IAEA, 1994) were, in principle, derived taking into account only the so-called "radiation protection" factors5, i.e., the cost of the intervention and the dose averted by it. This constraint is clearly stated in the guidance and exhortations are made to those responsible nationally for setting intervention levels to take other important factors into account. Despite these caveats and their obvious importance, they are often overlooked for various reasons. In situations where taking account of these caveats (i.e., the inclusion of all other relevant factors) would probably result in significant modifications to the generic levels of intervention, the advisability of issuing guidance on such levels is highly questionable. Those who directly use the generic levels, without further detailed consideration (and there will be many that will act in this way) as to their appropriateness in their circumstances, are unlikely to achieve the underlying objectives of intervention. 5

Some may consider that the scope of the "radiation protection factors" has been overly constrained, in particular considering that one of the central tenets of radiation protection is "to reduce doses to as low as reasonably achievable, social and economic factors taken into account".

4 - Post accident management

89

The approach followed in (IAEA, 1994) in developing generic intervention levels is not wholly consistent. In principle, in the derivation of generic levels only the so called "radiation protection" factors were to be taken into account. In practice, other considerations influenced the choice of generic levels for some interventions. For foodstuff restrictions, there was an overriding wish for the generic intervention levels to be compatible with the levels in (FAO/WHO, 1991) for the free movement of foodstuffs in international trade (see Sect. 2.2). The advantages accruing from the use of internationally agreed values fully justified this departure from the declared approach. However, unless clearly recognised, such departures are a source of possible confusion and/or error when generic levels are being transformed for specific use. Differences are also evident in the interpretations of (IAEA, 1994 and ICRP, 1999) regarding the so-called "radiation protection factors" that each has included in their respective derivations of generic intervention levels and generic reference levels. The former limited consideration to the costs and doses averted from intervention whereas the latter also included the following: reduction in anxiety, reassurance, social cost, harm and disruption of the intervention. The derived levels in the respective guidance are not, therefore, strictly comparable, having been derived on different bases. This further hinders a clear understanding of the changes (and their implications) that have occurred in this field over the years. It would be unreasonable, however, to be overly critical of the approaches followed in (IAEA, 1994 or ICRP, 1999). This issue is complex, as evidenced by the considerable time and effort devoted to it over the past decade or so; moreover, achieving broad agreement on levels for some interventions is somewhat of an intractable problem. The importance and usefulness of gaining wide acceptance of intervention levels cannot be over-stated and international agencies are to be encouraged in this pursuit. In its absence, differences between countries could undermine national policies and become the source of much public and political concern and disruption following an accident. However, it would be counter-productive to seek or claim broad agreement where it is illusory (i.e., possible only by addressing a part (and probably not the most important) of the problem). Given the reservations expressed above on the generic levels in (IAEA, 1994 and ICRP, 1999) for post accident management6, it is useful to explore other alternatives for improving guidance on the application of the principles. If quantitative generic guidance is to be of practical value for use regionally or internationally, it is clear that account must be taken of all relevant factors and of the views of all interested parties (or stakeholders in modern parlance) in its derivation. Achieving such agreement will be difficult but not impossible, in particular if there is a shared recognition of the 6

The same reservations do not exist, or are much less important, for urgent countermeasures taken in the early phase of an accident.

90

G.N. Kelly

need and a commitment to resolve the issue. Agreement at a regional level may be easier than internationally due to greater commonality in social, ethical, political values, etc. Achieving such agreement, however is far beyond the scope or remit of the radiation protection community and would need to be undertaken by those ultimately responsible for post accident management policy and its implementation7. Historically, there has been a tendency for the radiation protection community to take the leading role in every situation where radiation protection has had a role to play. Indeed, this is not surprising given the manner in which the optimisation principle in the system of protection has been framed, i.e., a requirement to include social and economic factors when judging whether exposures have been reduced to as low as reasonably achievable. This has encouraged the view that "societal aspects should be integrated into radiation protection decisions" rather than that "radiation protection aspects should be integrated into societal decisions" (Pretre, 1998). Some continue to believe that these issues are largely a matter for the radiation protection community alone; the more enlightened recognise that decisions on complex issues, such as post-accident management, require much broader consideration with inputs from many constituencies, of which the radiation protection community is but one and rarely the most important. Moreover, the ability of the radiation protection community alone to exercise social, ethical, cultural, economic and other judgements on behalf of society is clearly limited and some may even question its legitimacy. Further progress in this area will require the radiation protection community to recognise its more limited role in such matters, i.e., to provide technical input on matters where it has recognised expertise and competence as part of a wider decision process. Once this more limited role is recognised, many of the concerns and problems of the past decade or so over post accident management may subside and perhaps disappear. Broad agreement at an international level on post accident management remains an important goal in terms of enhancing public confidence and trust in any measures taken. Such agreement, however, must be achieved between those ultimately responsible for taking decisions at a national level and not by one narrow faction among the many who will have an input on such matters. Persuading those who own this problem to allocate their time to address and resolve it may, however, prove difficult given other commitments and pressures. Experience has already been gained by a few countries in the development of policy for post accident management and its implementation; this could form a basis or platform in a search for broader agreement. 7

The development of the (Council Regulation, 1987b) on the content of radionuclides in foodstuffs that could be imported into the EC from third countries is a good example of this wider process. The levels recommended by the radiation protection experts were reduced by a factor of four to reflect uncertainties and perceived disagreements between different technical constituencies.

4 - Post accident management

91

2.4. Important factors to be addressed in post-accident management The European Commission has, within its research programmes, funded many projects over the past decade on post-accident management. Part of this research focused on the less technical issues (e.g., social, political, psychological, cultural) which can have an important, sometimes overriding, influence on post-accident management policy and its implementation. Two areas where ongoing research may have important implications for policy are described.

2.4.1. Decentralised approach and importance of local involvement/commitment One of the important lessons learnt rapidly after the Chernobyl accident was that proper account must be taken of social and psychological factors in planning and implementing an effective programme of countermeasures in response to a nuclear accident. Failure to do so, as was evident in the Former Soviet Union, may generate much public anxiety and opposition and make poor use of resources that are often limited in the aftermath of an accident. The integration of such factors into decision making is not, however, without difficulty, both conceptual and practical. A number of research projects have been funded by the Commission which analysed, in some depth, the role of social and psychological factors in the planning and implementation of countermeasures (EC, 1992; French et al, 1992; Lochard et al, 1996; Heriard Dubreuil et al, 1999; Allen et al, 2000). Much of this research was of a high academic standard, provided many valuable insights into the underlying issues and contributed to the development of decision support systems designed to incorporate social and psychological aspects. Exploitation of the research has, however, been less effective in terms of improving the conditions of those continuing to live in contaminated settlements, one of the prime drivers for the research. This situation has changed more recently with the implementation of, and success achieved by, the ETHOS project (Heriard Dubreuil et al., 1999). The ETHOS project was launched in 1996 and carried out, as part of the RODOS project (Ehrhardt, 2000), under the auspices of the Commission's 4th Framework Programme. It was concerned with the development and application of a novel approach for improving living conditions in contaminated settlements. Previous research had identified two main impediments to improving the situation in the affected settlements: firstly, the development of a dependency culture among those affected and, secondly, highly centralised planning and implementation of remedial measures. To overcome these deficiencies, a decentralised approach was developed and tested in the

92

G.N. Kelly

village of Olmany in Belarus. This approach has proved particularly successful. It has resulted in the population taking far greater responsibility for its own actions, in reduced levels of radio-caesium in children in the village, in the production of less contaminated agricultural produce and in economic improvements through the marketing of foodstuffs that previously exceeded contamination limits. The approach is currently being extended to a number of settlements in the region of Stolyn and, with support from the TACIS programme, will be disseminated more widely in Belarus, Russia and Ukraine. The success of the approach has been recognised by the Belarussian authorities at local, regional and national levels where there is a now a commitment to apply it more widely. This project exemplifies what can be achieved through the active involvement and commitment of local communities in decisions affecting their welfare. Without such involvement, experience has shown that it is very difficult, if not impossible, to achieve tangible improvements in living conditions in areas affected by long term contamination. There are important lessons to be learnt from this experience, in particular in planning for the management of similar situations in the European Union, should they ever occur (whether from radioactive or other persistent contaminants). Some would contest this view, claiming that the ETHOS experience is unique to the Former Soviet Union and cannot be transferred to the European Union where conditions (e.g., social, political and economic) are different. While such differences must be acknowledged, they do not in any way detract from, or negate, the overall findings of the project, i.e., the importance of actively involving local communities in the long term management of their contaminated settlements. To discount these findings, for what are largely superficial reasons, would be to the detriment of effective planning for postaccident management in the European Union.

2.4.2. Stakeholder involvement and consumer resistance Notwithstanding the sound basis of international guidance on levels of contamination in foodstuffs (i.e., in the (Council Regulation, 1987c) and in (FAO/WHO, 1991)), difficulties may be encountered in their implementation in the event of any future accident. Much experience has, unfortunately, been gained in the last few years in Europe of public reaction to "contamination" of foodstuffs by various agents (e.g., BSE, benzene in water, dioxins in animal feed, genetically modified products, etc.). It is evident that, irrespective of the actual health risk, there will be much resistance to the consumption of food perceived as "contaminated", especially where "uncontaminated" sources are readily available. This aspect needs more detailed consideration and contingency arrangements made in the event that public reaction and/or commercial considerations necessitate the adoption of much lower, i.e., more restrictive, standards.

4 - Post accident management

93

As in many other fields, there is increasing recognition of the need and importance of involving all major stakeholders in the development of post accident management policy and practice. It is evident that this matter has for too long been the preserve of the radiation protection community; this wider involvement can only result in the establishment of more robust and practicable approaches. An initiative in this direction has recently been taken in the UK (Nisbet et al., 2001) in the context of agricultural countermeasures. An Agricultural and Food Countermeasures Working Group was established in 1997 with participation, at senior level, from a wide range of stakeholders (e.g., local and central government, farmers, food and drink industries, food retailers, consumer organisations, environmental groups, etc.). The Group has provided an effective forum for: establishing lines of communication between organisations who in the event of an emergency would be involved in decisions on intervention and/or would be affected by their implementation; disseminating relevant information on agricultural countermeasures; assessing the practicability of various countermeasure options and their implications for the food industry. In addition, the members would provide the core of any group that would be convened in the event of an accident to provide considered input to decisions on countermeasures. Building on the success of this initiative, the European Commission, under the auspices of its 5th Framework Programme, is funding a project (FARMING, 2000) in which similar groups are being established in several other countries and networked. This will provide a European dimension to many of the issues and will also promote more common approaches in the event of any future accident. Ideally, this network should, in time, become self-sustaining. As such, it could provide a useful source of input to the Commission in the event of any future accident affecting Europe (in particular in the context of reviewing the maximum permitted levels in foodstuffs as foreseen in the (Council Regulation, 1997c)).

3. Rapid characterisation of contaminated areas The rapid establishment of where radio-active material is deposited is important for effective post accident management and public reassurance. Monitoring after an accident is an intensive activity and, as its spatial resolution is refined, areas with enhanced contamination relative to their immediate surroundings will be found. Such findings are inevitable and understandable for the technical community. They have, however, proved to be a source of concern to those living in contaminated areas, often contributing to a loss of confidence and trust in the authorities. Rapid and reliable characterisation of contaminated areas is the only way to overcome or minimise this problem.

94

G.N. Kelly

Land based monitoring systems have a very limited capacity for fulfilling these needs, in particular for large accidents given the extent of the area to be monitored. Airborne gamma monitoring (on fixed wing aircraft or helicopters) has the potential to fulfil this role but, prior to the Chernobyl accident, the European capability was very limited. This situation has since changed and several European countries have now developed the capability to deploy measurement teams within hours of notification of an accident to measure the levels and pattern of deposited radioactive material. The overall European capability is a powerful resource for use in any future accident. However, to ensure that it can be deployed effectively and in concert, more needs to be done in terms of inter-calibration of the various teams and to better evaluate their performance when measuring mixed fresh fission products that would generally be encountered in practice. Most of the experience of the European airborne gamma teams is with measuring caesium radionuclides from Chernobyl or natural radioactivity. In this context, the Commission has, under the auspices of its 4th and 5th Framework Programmes, funded a network on airborne gamma monitoring. This network is currently preparing a major inter-comparison exercise (including air- and land-borne systems) to be held in Scotland in 2001 (ECCOMAGS, 2000). This inter-comparison will demonstrate the capability of the European teams to act jointly and produce a composite map of a contaminated area in real time. In addition to providing public reassurance, rapid characterisation or mapping of contaminated areas will enable resources to be more effectively utilised in the immediate post-accident management period.

4. Technical options for, and the efficacy and costs of, post-accident management Following the Chernobyl accident considerable resources were allocated to the development, investigation and application of options to reduce the consequences of environmental contamination. Much experience had been gained in the Former Soviet Union in managing the consequences of the Mayak accident and this was put to good use following Chernobyl. The types of measure that can be taken are numerous and range from the very simple with low costs (e.g., street cleaning) to those with major socio-economic consequences (e.g., relocation of settlements). It is far beyond the scope of this paper to summarise or do justice to all that has been done in this field (where the Commission has been particularly active through both its research and assistance programmes) in the past decade or so. Those with a particular interest in this topic are referred to the following publications (Hubert et al., 1996; Roed et al, 1995; Brown et al, 1996; Strand et al., 1997; Anderson, 1996; Woodman et al, 1997; EC, 1996; EC, 2000; IAEA, 2001). This list of reference is far from exhaustive and is

4 - Post accident management

95

intended solely to point those interested in a direction where they can find out more on the technical options for the post accident management of rural and urban environments. Each of these publications contains extensive referencing. Most of the referenced reports either review what has been done or distil and compile the available knowledge and information for the purposes of assessing the efficacy, cost, waste generation, practicability, etc., of different options. Such compilations provide a useful basis for choosing which post accident measures should be implemented. Given the extensive work that has been carried out in this area over the past two decades (and earlier in some cases), the technical information now available is largely adequate for the purposes of making sound choices on which options to use for post-accident management. Improvements in knowledge could doubtless be made through further research; it is unlikely, however, that these would significantly enhance the efficacy of existing remedial measures or choices between them. The general adequacy of the existing information or knowledge base is not, in itself, a guarantee of effective post accident management. For this to be achieved, this knowledge needs to be translated into post accident management plans with due attention given to logistical issues, waste management, public reaction, consumer resistance, etc. Particular attention needs to be given to the less tangible issues such as public reaction and acceptance, etc.; otherwise, the best prepared plans (from a technical viewpoint) may fail in practice.

5. Maintaining competence and issues of sustainability With a few reservations, the knowledge and competence in, and arrangements made for, the longer term management of any future accident are broadly adequate. Whether they will remain so in future is more questionable. Fifteen years have now passed since the Chernobyl accident and the subject of post-accident management is no longer at the forefront of political debate nor public attention (at least not in the West). In the immediate aftermath of the accident the scientific and technical resources allocated to this issue were increased dramatically, in particular in Europe. Effort in this area probably peaked in the early 1990s and has experienced a slow decline since. This decline can be expected to continue as the memory of Chernobyl fades even more and the safety of nuclear installations is further increased. New reactor designs aim to limit the consequences (in terms of the need for intervention) of any accident to within the site boundary. This has led some to claim that less effort need be allocated to emergency response and post accident management in future. While such arguments are not without substance, a proper balance must be maintained. Preparedness to respond to and manage the long term impact of accidents must remain an essential part

96

G.N. Kelly

of an in depth and precautionary approach to nuclear safety (i.e., prevention, mitigation and response). It is essential for confidence building, no matter how low the chance of an accident is thought to be. Maintaining preparedness for responding to accidents is, however, resource intensive and the continuing need to exercise arrangements is demanding, especially on the time of senior people. Pressures will inevitably arise for arrangements that are less resource intensive. A major challenge for the emergency response and post accident management community in the coming years will, therefore, be to maintain or improve current arrangements in a climate of declining resources. This can be achieved given sufficient foresight and political will, but it will not be easy. There will be a need for vigilance and complacency must be avoided. A further difficulty will be to maintain sufficient knowledge and competence in post accident management and ensure that it can be effectively accessed or retrieved when needed, possibly far in the future. Nuclear accidents are rare events and should, in principle, be even rarer in future as lessons are learnt from past mistakes and safety improvements made. The commercial generation of electricity from nuclear energy began almost five decades ago. In this period there have been only two major accidents with commercial reactors, firstly at Three Mile Island in 1979 and Chernobyl in 1986. Earlier, in the exploitation of nuclear fission for defence purposes, major accidents also occurred at installations at Windscale and Mayak in 1957. It is impossible to predict when, or indeed if, a future accident may occur. It could, however, be many decades from now. By then the institutional memory, acquired by those who were actively engaged in the research, development, planning and/or implementation of post accident management following the Chernobyl accident, will no longer be available. Some loss of knowledge and competence is inevitable over time, in particular if it is not further developed or practically exercised. Minimising this loss and ensuring that an adequate level of competence, expertise and planning will be available to meet any future need are major challenges for those responsible for emergency and/or post accident management. Achieving this will not be easy, in particular in the context of other demands and a declining interest in post accident management. This issue is, however, too important to be ignored - failure to give it due attention could have major social and economic consequences in the event of any future accident.

6. Summary and conclusions International policy and guidance on post accident management. Much has been achieved since the Chernobyl accident in terms of policy for post-accident management, how it should be implemented and international agreement on the principles for intervention. Less agreement, however, exists on

4 - Post accident management

97

the interpretation or application of the principles when making decisions on (or establishing levels for) intervention. Major differences are evident between the generic intervention levels developed internationally and what is being done in practice in the post-accident management of contaminated areas, not only around Chernobyl but also elsewhere. Contrary to the international guidance, dose limits (or, in some cases, lower levels) are often being used as the boundary for decisions on the need or otherwise for intervention and/or the return to "normal living". Many, if not most, in the radiation protection community claim that this is a mistaken use of the system of dose limitation as dose limits for practices are not intended for interventions; furthermore, that through such action, public resources are being wasted in that they could have been allocated to better effect elsewhere in society. Others might argue that such practice is fully in accord with the principles of intervention, in particular resulting from due weight having been given to all of the factors (technical, social, political, ethical, economic, etc.) relevant to such decisions. From a narrow technical viewpoint, it is easy to share the views of many in the radiation protection community. However, the premise or logic on which they are based is false and is a result of taking too narrow a view of the issue. The more technical radiation protection considerations (i.e., the cost of and risk averted by intervention) are but one of the many inputs into decisions on complex issues such as post accident management and rarely are they the most important. Radiation protection experts are not alone in being concerned over the decreasing influence that technical considerations have on decision making. This is now common place in many controversial areas of science and technology where the expert view is no longer pre-dominant (or at least less influential than hitherto) and must be balanced with broader societal concerns. To claim or suggest that decisions are wrong, or less than optimal, because technical arguments no longer hold sway is a negation of the democratic decision making process. The radiation protection community has for too long tried to take the leading role in every situation where radiation protection has had a role to play. Indeed, this is not surprising given the way in which the optimisation principle, perhaps the main pillar of the system of dose limitation, has been framed. A requirement "to reduce doses to as low as reasonably achievable, social and economic factors taken into account" encourages the view that societal aspects should be integrated into radiation protection decisions. The more enlightened have long recognised that, at least for more complex issues such as post accident management, the opposite is true - radiation protection aspects should be integrated into societal decisions. Once this is fully recognised, many of the current difficulties and disagreements over intervention levels for post accident management will vanish. The pursuit of broad agreement on intervention within the narrow confines of the radiation protection community will be seen as an illusory goal. Broad agreement at an international level on post accident management remains, however, an

98

G.N. Kelly

important goal in terms of enhancing public confidence and trust in the measures taken. However, such agreement must be achieved between those ultimately responsible for such decisions nationally and not by one narrow faction among the many who will have an input on such matters. The radiation protection community may be better advised to restrict itself to seeking agreement and providing input in those areas where it has recognised competence and expertise, for example on the principles for intervention and on the health effects of radiation. Others should be left to weigh this input along with those from other relevant disciplines or communities and interest groups. The pursuit by the radiation protection community alone of broadly agreed generic intervention levels derived from consideration of but a few of the relevant issues, is mistaken. This is best exemplified by post accident management practice where the intervention levels adopted bear little relation to the generic levels. Indeed, given the large disparity between practice and theory, there is a risk that the broader guidance on the principles of intervention, which is well founded, may become discredited. This must be avoided. Post accident management planning and practice. Considerable knowledge and expertise has been acquired over the past decade or so on post accident management. This knowledge is judged to be largely sufficient for the development of sound policy on post accident management and for its effective implementation. Further research and development would doubtless improve knowledge but, in general, its impact on policy or practice would not be large given the relative maturity of this topic (at least the more technical aspects). The general adequacy of the existing information or knowledge base is not, in itself, a guarantee of effective post accident management. For this to be achieved, this knowledge needs to be translated into post accident management plans with due attention given to logistical issues, waste management, public reaction and acceptance, consumer resistance, exercising the arrangements at different levels, etc. Particular attention needs to be given to the less tangible issues such as public reaction and acceptance, etc.; otherwise, the best prepared plans (from a technical viewpoint) may fail to achieve their objectives in practice. More needs to be done in this area to increase confidence in the adequacy and robustness of existing arrangements; in particular, it is evident that the arrangements for, and exercising of, post accident management are far less mature than those for emergency response. There is one important exception to the generalisation about the need for further research and development. This concerns the broader societal aspects and how they can be better factored into the formulation of policy for post accident management and its subsequent implementation. Two key issues have been highlighted in recent research. Firstly, the importance of actively involving local communities in, and getting their commitment to, decisions and post accident measures affecting their welfare - without such involve-

4 - Post accident management

99

ment and commitment, experience has shown that it is very difficult, if not impossible, to achieve tangible improvements in living conditions in areas affected by long term contamination. Secondly, the need for and importance of broader participation (or in current parlance, stakeholder involvement) in the development of post accident management policy and practice - without the involvement of all interested parties the policy is destined to fail or, at best, be less effective than it would otherwise have been. There are important lessons here for those responsible for the development of policy for post accident management (whether from radioactive or other persistent contaminants) and its implementation. One of the more important challenges for the future will be to ensure that sufficient knowledge and competence is maintained for the management of any future accident. Nuclear accidents are relatively rare events and will probably be even rarer in future; several decades may elapse before another accident (if ever) occurs. By then the institutional memory, acquired by those who were actively engaged in the research, development, planning and/or implementation of post accident management following the Chernobyl accident, will no longer be available. Responding effectively to this challenge will not be easy, in particular in the context of other demands and a declining political interest in this topic. The importance of this issue should not, however, be under-estimated. It behoves those responsible to give the matter serious attention and take steps to ensure the sustainability of the competence and expertise necessary for effective post accident management.

References Allen P. et ol. (1999) "Social and Psychological Aspects of Radiation Protection after Accidents", Final Report of an EC Funded Project (SPARPA) (European Institute of Health and Medical Sciences, Risk and Decision Making Research Group, University of Surrey, UK). Anderson K. (1996) "Evaluation of Early Phase Nuclear Accident Clean-up Procedures for Nordic Residential Areas", Nordic Nuclear Safety Research, NKS (96) 18, Roskilde. Balonov M.I., Anisimova L.I. and Perminova G.S. (1999) "Strategy for Population Protection and Area Rehabilitation in Russia in the Remote Period after the Chernobyl Accident", /. Radiol. Prot., 19 (3), 261-269. Brown J. et al. (1996) "Review of Decontamination and Clean up Techniques for Use in the UK following Accidental Releases of Radioactivity to the Environment" (NRPB-R288, Chilton). Council Decision (1987a) "87/600/EURATOM, on Community Arrangements for the Early Notification of Information in the Event of a Radiological Emergency", Official Journal of the European Communities, OJ L371/76 of 30/12/87.

100

G.N.Kelly

Council Regulation (1987b) "(EURATOM) - Laying down conditions governing imports of agricultural products originating in third countries following the accident at the Chernobyl nuclear power station (87/3955/EURATOM, 89/4003/EURATOM, 90/737/EURATOM, 2000/616/EC; 2000/1609/EC; 2000/1627/EQ", Official Journal of the European Communities (L-371 of 30/12/87, p. 14; L-382 of 30/12/89, p. 4; L-82 of 29/3/90 p. 1; L-75 of 24/03/2000 p. 1; L-185 of 25/07/2000 p. 27; and L187 of 26/07/2000 p. 7). Council Regulation (1987c) "(EURATOM) - Laying down Maximum Permitted Levels of Radioactive Contamination of Food Stuffs and of Feeding Stuffs following a Nuclear Accident or any other Case of Radiological Emergency (87/3954/EURATOM, 89/2218/EURATOM, 89/944/EURATOM, 90/770/EURATOM)", Official Journal of the European Communities (L-371 of 30/12/87, p. 11; L-211 of 22/7/89, p. 1; L-101 of 13/4/89 p. 17 and L-211 of 27/07/89 p. 1). ECCOMAGS (2000) "European Commission, 5th Framework Programme, European Calibration and Co-ordination of Mobile and Airborne Gamma Spectrometry", Project co-ordinated by D. Sanderson (SURRC, East Kilbride, UK). EC (1982) "Radiological Protection Criteria for Controlling Doses to the Public in the Event of Accidental Releases of Radioactive Material - A Guide on Emergency Reference Levels of Dose from a Group of Experts convened under Article 31 of the Euratom Treaty" (European Commission, Luxembourg). EC (1992) "International Chernobyl Project - Input from the Commission of the European Communities to the Evaluation of the Relocation Policy adopted by the Former Soviet Union", EUR 14543 EN (European Commission, Luxembourg). EC (1993) "Radiological Protection Principles for Relocation and Return of People in the Event of Accidental Releases of Radioactive Material, Radiation Protection - 64", Doc XI-027/93 (European Commission, Luxembourg). EC (1996) "The Radiological Consequences of the Chernobyl Accident", Proc. of the First International Conference, Minsk, 18-22 March 1996, European Commission, EUR 16544 EN (European Commission, Luxembourg). EC (2000) "Proceedings of the Workshop on Restoration Strategies for Contaminated Territories resulting from the Chernobyl Accident", Brussels, 29-30 June, 1998, EUR 18193 EN (European Commission, Luxembourg). Ehrhardt J. and Weis A. (2000) "RODOS: Decision Support for Off-Site Nuclear Emergency Management in Europe", EUR 19144 EN (European Commission, Luxembourg). FARMING (2000) "European Commission, 5th Framework Programme, Food and Agriculture Restoration Management Involving Networked Groups", Project co-ordinated by A. Nisbet (NRPB, Chilton).

4 - Post accident management

101

FAO/WHO (1991) "Food and Agriculture Organisation of the United Nations/World Health Organisation, Codex Alimentarius", General Requirements, Section 6.1, Guideline Levels for Radionuclides in Foods following Accidental Nuclear Contamination (Joint FAO/WHO Food Standards Programme, Rome). French G.S., Kelly G.N. and Morrey M. (1992) "Decision Conferencing and the International Chernobyl Project", /. Rodiol. Prof., 12 (1), 17-28. Gjorup H.L. et al. (1982) "Radioactive Contamination of Danish Territory after Core-Melt Accidents at the Barseback Power Plant" (Riso-R-462, Roskilde). Hedemann Jensen P. (2000) "An Analysis of Case Studies - Contaminated Facilities and Sites", Proc. of International Symposium on Restoration of Environments with Radioactive Residues, Arlington, USA, 29 Nov - 3 Dec, 1999 (IAEA, Vienna). Heriard Dubreuil G. et al. (1999) "Chernobyl Post-accident Management: The ETHOS Project", Health Physics, 77, 361-372. Hubert P. et al. (1996) "Experimental Collaboration Project No 4 - Strategies for Decontamination", EUR 16530 EN, European Commission, Luxembourg). IAC (1991) "The International Chernobyl Project, Technical Report - Assessment of the Radiological Consequences and Evaluation of Protective Measures", Report by an International Advisory Committee (IAEA, Vienna). IAEA (1985) "International Atomic Energy Agency, Radiation Protection Principles for Sources not under Control: Their Application to Accidents", Safety Series No. 72 (Vienna). IAEA (1986) "International Atomic Energy Agency", Convention on Early Notification of a Nuclear Accident (Vienna). IAEA (1988) "International Atomic Energy Agency, Revised Guidance on the Principles for Establishing Intervention Levels for the Protection of the Public in the Event of a Nuclear Accident or Radiological Emergency", IAEA-TECDOC-473 (Vienna). IAEA (1991) "International Atomic Energy Agency, Radiation Protection Principles for Sources not under Control: Their Application to Accidents", Safety Series No. 72 (Rev) (Vienna). IAEA (1994) "International Atomic Energy Agency, Intervention Criteria in a Nuclear or Radiation Emergency", Safety Series No. 109 (Vienna). IAEA (1996) "International Atomic Energy Agency, International Basic Safety Standards for Protection against Ionising Radiation and for the Safety of Radiation Sources", Safety Series No. 115 (Vienna). IAEA (2001) "International Atomic Energy Agency, Guide on Decontamination of Rural Settlements in the Remote Period after Radioactive Contamination with Long-Lived Radionuclides", TECDOC (to be published).

102

G.N. Kelly

ICRP (1984) "International Commission on Radiological Protection, Protection of the Public in the Event of Major Radiation Accidents: Principles for Planning", Publication 40 (Pergamon Press, Oxford, New York). ICRP (1990) "International Commission on Radiological Protection, 1990 Recommendations of the International Commission on Radiological Protection", Publication 60, Annals of ICRP, Vol. 21, No 1-3 (Pergamon Press, Oxford, New York). ICRP (1993) "International Commission on Radiological Protection, Principles for Intervention for Protection of the Public in a Radiological Emergency", Publication 63 (Pergamon Press, Oxford, New York). ICRP (1999) "International Commission on Radiological Protection, Protection of the Public in Situations of Prolonged Radiation Exposure", Publication 82, Annals of ICRP, 29 (1-2) (Elsevier, Oxford). Kelly G.N. et al. (1982) "An Assessment of the Radiological Consequences of Releases from Degraded Core Accidents for the Sizewell PWR", NRPBR137 (Chilton, London, HMSO). Kelly G.N. et al. (1983) "The Influence of Countermeasures on the Predicted Consequences of Degraded Core Accidents for the Sizewell PWR", NRPB-R163 (Chilton, London, HMSO). Lochard J. and Belyaev S. (1996) "Decision Aiding System for the Management of Post-accidental Situations", Joint Study Project No 2, EUR 16534 EN (Office for Official Publications of the European Communities, Luxembourg). Nisbet A. (2001) "Stakeholder Pre-Involvement in the Post Accident Management of Rural Areas - A Government Perspective", IN Proc of the Second Villigen Workshop: Better Integration of Radiation Protection in Modern Society (NEA/OECD, Paris). Pretre S. (1998) "Decision Making in Abnormal Radiological Situations", IN Proc of Workshop on The Societal Aspects of Decision Making in Complex Radiological Situations, pp 9-19, Villigen, 13-15 January, 1998 (NEA/OECD, Paris). RF (1996) "Law of the Russian Federation "On Radiation Safety of the Population" of 15 May 1991". RNCRP (1995) "Concept of Radiation, Medical and Social Protection and Rehabilitation of Population of the Russian Federation Subjected to Accidental Exposure" (Russian National Commission on Radiation Protection.) Roed J. et al. (1995) "Practical Means for Decontamination Nine Years after a Nuclear Accident", Riso-R-828(EN), Roskilde. RSFSR (1991) "Law of the RSFSR "On Social Protection of Citizens Affected by the Radiation due to the Chernobyl NPP" of 15 May 1991". Strand P. et al. (1997) "Reclamation of Contaminated Urban and Rural Environments following a Severe Nuclear Accident", Nordic Nuclear Safety Research, NKS (97) 18, Roskilde.

4 - Post accident management

103

TRUSTNET (2000) "The Trustnet Framework: A New Perspective on Risk Governance", EUR 19136 EN (European Commission, Luxembourg). USNRC (1975) "US Nuclear Regulatory Commission, Reactor Safety Study, An Assessment of the Accident Risks in US Commercial Power Plants", WASH-1400 (NUREG-75/14). Washington DC. WHO (1988) "World Health Organisation, Derived Intervention Levels for Radionuclides in Food - Guidelines for Application after Widespread Radioactive Contamination from a Major Radiation Accident" (Geneva). Woodman R.F. et al. (1997) "Options for the Management of Foodstuffs Contaminated as a Result of a Nuclear Accident", NRPB-R295 (Chilton).

This page intentionally left blank

5

International advice and experience relevant to chronic radiation exposure

situations in the environment G. Linsley1

The paper starts by briefly summarizing the major sources of prolonged exposure to ionizing radiation resulting from residual radioactivity in the environment. The rehabilitation of areas affected by such residues is under consideration in many countries. Recognizing that there was a lack of radiation protection guidance for aiding decisions in this context, the international organizations have addressed the situation. Their decision aiding guidance is described and its role in decision making is discussed. Finally, examples are given of some internationally organized assessments of areas affected by residues from nuclear weapons testing, including descriptions of environmental measurements, radionuclide transfer modeling and exposure scenario development and of the results of the assessments and their implications for remediation planning.

1. Introduction Chronic or prolonged radiation exposure situations are those in which the exposure of the affected members of the public is constant or decreasing slowly over the years. Typical prolonged exposures are those delivered by "natural" sources such as cosmic radiation and primordial radionuclide decay chains. Some "artificial" sources may also deliver prolonged exposures, for example, long-lived radioactive residues from human activities. Particular interest in prolonged exposures has arisen recently because of recognition of the need to deal with the problem of environments affected by long lived radioactive residues, for example, those affected by radioactive residues from accidents, nuclear weapons testing and from past practices involving the use of radioactive materials. In parallel with these developments 1

International Atomic Energy Agency, P.O. Box 100,1400 Vienna, Austria.

106

G. Linsley

international radiological protection guidance for aiding decision making in this context has been published. Several international meetings have been organized in recent years on the subject of environmental restoration. They include the Workshop on Relative Effectiveness of Agricultural Countermeasure Techniques, Brussels, October 1991 (Howard and Desmet, 1993), the Symposium on Remediation and Restoration of Radioactive - Contaminated Sites in Europe, Antwerp, October 1993 (EC, 1994), the Symposium on Environmental Impact of Radioactive Releases, Vienna, May 1995 (IAEA, 1995), the Conference on the Radiological Consequences of the Chernobyl Accident, Minsk, March 1996 (Karaaglou et al., 1996), the Workshop on Restoration Strategies for Contaminated Territories resulting from the Chernobyl Accident, Brussels, 1998 (Cecille, 2000), the Symposium on the Restoration of Environments with Radioactive Residues, Arlington, December 1999 (IAEA, 2000) and the Conference on Radiation Legacy of the 20th Century: Environmental Restoration, Moscow in October 2000 (MINATOM, 2001). It is evident from the proceedings of these meetings that experience is still being acquired on the subject of environmental restoration, on relevant assessment methods, on decision making processes, on remediation techniques and technologies and procedures for involving affected members of the public. This paper focuses mainly on the radiological protection guidance relevant to environmental restoration and on international experience of assessments carried out in this context. It draws mainly on documents of the ICRP and of the IAEA.

2. Major sources of chronic environmental exposures A wide diversity of types and locations of radioactive residues in the environment have originated from numerous practices and events involving the use of radiation sources. The testing of nuclear weapons in the atmosphere allowed large amounts of radioactive materials to be released and dispersed throughout the world. Also, in the production of weapons, local residues of waste radioactive material were left in the environment surrounding the installations where the weapons were produced and tested. The nuclear fuel cycle associated with the generation of electrical energy in nuclear power reactors has also caused radioactive materials to be released to the environment. Within the nuclear fuel cycle the amounts of releases are greatest from uranium mining and milling operations and in the fuel reprocessing stage. Significant releases have occurred in accidents, at nuclear reactors such as Windscale and Chernobyl, and at waste storage sites such as Kyshtym. The re-entry of satellites into the atmosphere has on a few occasions caused releases of radionuclides to the environment. Industrial

5 - International advice and experience

107

and medical practices in which radiation sources are used generally do not result in significant releases to the environment, although the misuse of a medical therapy source at Goiania, Brazil caused widespread contamination. The historic production and use of radium for medical and industrial purposes has left contaminated sites in many countries, although usually the areas affected are small in size. For perspective, the largest source of prolonged or chronic exposure is due to naturally occurring radionuclides in the environment. The average dose to the world's population from natural radiation sources is estimated to be 2.2 mSv, of which 40% is external and 60% internal exposure. The highest component of the annual effective dose (1.0 mSv) comes from inhalation of radon and thoron and their short lived decay products. Of next importance is the dose from external irradiation (0.9 mSv), arising approximately equally from cosmic radiation and terrestrial sources. Less significant contributions to dose (0.3 mSv) come from ingestion of 40K, 210Pb, 210Po, and a few other terrestrial radionuclides. Variations about the mean values by factors of 5 to 10 are not unusual for many of the radionuclides. The greatest variation occurs for indoor radon concentrations, which span over four orders of magnitude. Much of this naturally derived exposure is unavoidable and only in a few situations, for example, in cases of high indoor radon concentrations is remediation attempted. In addition to normal background exposure, enhanced exposure to natural radiation occurs from coal and other fossil fuel burning, mineral processing, some consumer products and the use of fertilizers. Some materials with relatively higher concentrations of radionuclides are sometimes used in building materials, and wastes from mine tailings have been used as backfill in building construction. All of these circumstances can create sites of contamination and enhanced exposures to individuals.

3. International protection radiation protection guidance relevant to chronic exposures from areas affected by radioactive contamination Recognizing that there was a general lack of guidance on protection from ionizing radiation in the case of protracted or chronic exposures and prompted by the clear need for such guidance for aiding decision making in the particular case of the rehabilitation of areas affected by residual deposits of radioactive materials from past activities, the IAEA started a project in 1993 to address the problem. A small working group was established and met on several occasions over a three year period. The outcome of the working group

108

G. Linsley

was an IAEA Technical Document titled "Application of radiation protection principles to the cleanup of contaminated areas" (IAEA, 1997). Later, a Task Group of the International Commission on Radiological Protection (ICRP) worked from 1997 until 1999 on the development of protection criteria for prolonged public exposure to ionizing radiation. The report of the Task Group was approved for publication by the ICRP and issued with the title "Protection of the Public in Situations of Prolonged Radiation Exposure" as ICRP publication 82 in 2000 (ICRP, 2000). The following paragraphs are based upon the previously mentioned documents.

3.1. Practices and intervention Prolonged exposures are delivered by long lived natural and artificial radionuclides (and their short lived progeny) which are present in the human habitat either as a permanent natural feature or as radioactive residues. Radioactive residues may remain after the termination of regulated activities introduced by society; these activities are termed practices. Also, they may have been left by unregulated past activities and events. Exposure to natural sources and to radioactive residues already existing (de facto) in the human habitat may be subject to protective actions implemented through a process termed intervention. Some prolonged exposures to natural sources and almost all prolonged exposures to radioactive residues are controllable. Prolonged exposures that are essentially uncontrollable, or not amenable to control, are generally excluded from the scope of regulations on radiological protection. Relevant radioactive residues include those from the operation and decommissioning of practices, or from activities that were conducted either outside any control, or under radiological protection requirements less stringent than those applying today, or from accidents that released long lived radionuclides to the environment.

3.2. Quantities for use in assessing prolonged exposures The relevant quantity to be used in the assessment of prolonged exposure situations caused by radioactive residues is the annual effective dose, referred to simply as annual dose in this paper, attributable to the exposure. A subsidiary quantity used in the context of the ICRP publication is the summation of the annual doses caused by all the existing sources of prolonged exposure in a given human habitat; this quantity is termed the existing annual dose. The prolonged annual dose that is added to the existing annual dose as a result of a practice is termed the additional annual dose, and the

109

5 - International advice and experience

prolonged annual dose that is, or may be, removed from the existing annual dose by intervention is termed the averted annual dose.

3.3. Radioactive residues from practices The principles of the System of Protection for practices are the justification of the practice, the optimization of radiological protection, with regard to any source within the practice, and the limitation of individual doses attributable all practices under control. These principles should be applied prospectively at the planning stage of any practice expected to produce radioactive residues able to deliver prolonged exposures. They are applicable to the design, operation and decommissioning of the practice and its radiation sources (Fig. 5.1).

Existing Annual Dose

Pre-practice existing annual dose

Introduction, operation and decommissioning of a beneficial practice

Additional prolonged annual dose attributable to the practice (it excludes transitory doses)

Post-practice existing annual dose

Time

Figure 5.1. Simplified schematic presentation of the existing annual dose over the time when a new beneficial practice is operated and decommissioned.

3.3.1. Justification of a practice and optimization of protection The justification of a practice producing radioactive waste delivering prolonged exposure requires that all relevant long term factors should be considered prior to the adoption of the practice. Important factors are those

110

G. Linsley

related to the long lived radioactive residues that are expected to be discharged into the environment or to remain in the human habitat after the decommissioning of the practice. These factors include the prolonged additional annual doses, both individual and collective, that are attributable to the discharges and residues. With regard to any source in a justified practice that delivers prolonged exposure, the optimization of protection requires the selection of the optimum protection option under the prevailing social and economic circumstances, taking into account the relevant long term factors. The optimum option may be selected using the optimization techniques recommended by the ICRP. Under certain conditions, sources used in justified practices can be exempted from the regulatory requirements if the individual additional and annual doses attributable to the source are below around 0.01 mSv, because protection can be assumed to be optimized.

3.3.2. Individual dose restrictions The assessment of justification and optimization may introduce individual inequities, which may be important, when radioactive residues delivering prolonged exposures are involved. These are due to the wide spatial and temporal distribution of prolonged exposures, and which can, for example, affect future generations. In order to limit these inequities, stringent individual dose restrictions are applied, as dose constraints to single sources within the dose limits for all practices under control. The ICRP recommends that a maximum value of dose constraint, in the optimization of radiological protection for a single source, of no more than about 0.3 mSv is appropriate. Where an operating source or practice gives rise to a prolonged component of exposure, appropriate source related dose assessment methodologies should be used to ensure compliance with the established dose constraint, taking account of any reasonably conceivable combination and build up of exposure. If, in a particular situation, such verification of compliance is not feasible, it may be prudent to apply a dose constraint of the order of 0.1 mSv to the prolonged component of the individual dose. However, in order to avoid biasing the optimization of protection, this advice should be applied with extreme care and flexibility. The ICRP recommends that the sum of the prolonged exposures, e.g. as a result of the accumulation of radioactive residues from continuing practices, and the transitory exposures from all regulated practices should be restricted by a dose of 1 mSv in a year. The aim should be to prevent individual annual doses attributable to all current and predictable future practices exceeding the dose limit of 1 mSv.

111

5 - International advice and experience

3.4.

Interventions

The principles of the System of Protection for intervention are the justification of intervention and the optimization of the protective actions. These principles should be applied to any de facto exposure situations involving radioactive residues that deliver controllable prolonged exposure (Fig. 5.2). Existing annual dose Averted annual dose by the intervention

Pre-intervention existing annual dose Post-intervention existing annual dose

Intervention

Time

Figure 5.2. Simplified schematic presentation of the existing annual dose over the time when intervention is taken.

3.4.1. Justification and optimization of interventions The justification of intervention in prolonged exposure situations involving radioactive residues requires a positive balance of all relevant long term attributes. In addition to the avertable annual doses, both individual and collective, other attributes include the following: the expected reduction in the anxiety caused by the situation, the reassurance to be provided by the intervention, and the social cost to be incurred by implementing the protective actions. The assessment of justification should be based on radiological protection considerations, but the results of the assessment should be used as an input into a wider decision making process which may include other considerations, and may involve relevant stakeholders and search for their informed consent. Figure 5.3 shows that all key attributes can be considered disadvantageous if there is no intervention. Intervention may reduce some of the disadvantageous attributes associated with non-intervention (e.g. doses and anxiety), or even eliminate some (e.g. political pressure), while introducing new attributes of both disadvantageous nature (e.g. the cost of intervention, and some social disruption) and advantageous nature (e.g. reassurance). The attributes cost of intervention, social disruption and reassurance are now shown in the

112

G. Linsley

left side of Figure 5.3 because their value is assumed nil without intervention. The attribute political pressure is also assumed to diminish to nil after intervention and therefore is not shown in the right side of the figure. From the justified intervention options, the optimum protective actions(s) (including form, scale and duration) should be selected, following the general approach to the optimization of protection recommended by the ICRP in the context of practices. For some prolonged exposure situations involving radioactive residues, the restricted use of the human habitat can be the outcome of optimization.

Figure 5.3. A schematic view of how the principle of justification of intervention is applied.

3.4.2. Specific reference levels In situations of radioactive residues, national authorities and relevant international organization should establish predetermined specific reference levels which can be conveniently expressed in terms of the averted annual dose, or a related subsidiary quantity. The use of predetermined specific reference levels can facilitate timely decisions on interventions and the effective deployment of resources, however, an improper use may lead to inconsistencies with the principles of justification and optimization.

3.4.3. Generic reference levels The use of generic reference levels, expressed in terms of existing annual doses, may also be useful. They are particularly convenient when intervention is being considered in some generic prolonged exposure situations, such as

5 - International advice and experience

113

exposures to those radioactive residues that are a legacy from the distant past. These levels, however, should be used with extreme caution. Often, some controllable component of the existing annual dose is clearly dominant and the use of the generic reference levels should not prevent the implementation of specific protective actions for reducing these dominant components. These protective actions should be decided upon a case-by-case basis, following the requirements of the System of Protection for interventions. A low level of existing annual dose does not necessarily justify the System of Protection not being applied to any of its components; conversely, a high level of existing annual dose does not necessarily require intervention. With these provisos, it is considered that an existing annual dose lower than about 10 mSv seems to be generally tolerable and may be used as a generic intervention level below which intervention is not likely to be justified. However, below this level, protective actions to reduce a dominant component of the existing annual dose are still optional and might be justified. Action levels specific to particular components can be established on the basis of appropriate fractions of the recommended generic reference level. Above a level of existing annual dose of around 10 mSv, intervention may possibly be necessary and should be justified on a case-by-case basis. Situations in which the annual equivalent dose thresholds for deterministic effects in relevant organs could be exceeded will almost always require intervention. In establishing this requirement, uncertainties in the current estimates of deterministic effects from protracted exposures should be prudently taken into account. An existing annual dose rising towards 100 mSv will almost always justify intervention and may be used as a generic reference level for establishing protective actions under nearly any conceivable circumstance (Fig. 5.4).

Figure 5.4. Schematic representation of the generic reference levels.

114

G. Linsley

In choosing the generic reference levels the following was taken into account: (i) the source-related action level for radon recommended by the ICRP and IAEA corresponds to an annual dose of 3-10 mSv for simple remedial measures (for more severe measures, such as the permanent removal of people from their homes, the action level could be an order of magnitude higher); (ii) when assessing generic guidance on relocation, the ICRP has shown, in an example of generic optimization, that relocation was optimized at a dose (including transitory exposures) of about 10 mSv per month and recommended a level of 5-15 mSv per month as a range of optimized values for relocation, and (iii) the intervention level recommended by the ICRP (and established in international standards) for permanent resettlement following a nuclear accident is 1000 mSv in a lifetime, which would correspond to an average annual dose of about 15-20 mSv.

3.5.

Application of the guidance to radioactive residues

The recommended dose constraints should be applied prospectively to the prolonged exposure from the radioactive residues expected to remain in the human habitat after the discontinuation of a practice, for instance, at the site of a decommissioned installation. In principle, this dose constraint may be expected to be no higher than the dose constraint applied to the operational phase of the practice. However, there is not necessarily a presumed equality between the dose constraint applied before the discontinuation of a practice and that applied afterwards. If the operational dose constraint was very low, maintaining it in the post-decommissioning phase could introduce an unreasonable restriction. For radioactive residues from other past human activities and events that were not regulated as practices, the need for protective actions should be determined on a case-by-case basis, following the recommended intervention principles of justification and optimization of the protective actions, rather than through pre-selected individual dose restrictions. If necessary, the recommended generic reference levels of existing annual dose may be used as guidance. However, in cases where the origins of the situation are traceable, and those who produced the residues can still be made retrospectively liable for the protective actions, national authorities may consider applying a specific restriction to the individual doses attributable to the residues. This would require additional protective actions to be taken by those who created the situation. Residues that are deemed not to require protective actions should be the subject of further restrictions. After an accident that has produced radioactive residues in the environment, intervention can be permanent or may need to be discontinued at some stage. The simplest basis for justifying the discontinuation of interven-

5 - International advice and experience

115

tion is to confirm that the exposures have decreased to the action levels that would have prompted the intervention. It is important to recall, however, that the protective actions taken would have been intended to produce a substantial reduction in the exposure remaining after the accident: it is not necessarily sufficient to make marginal improvements aimed at reducing the exposures to values just below the action level. If such a reduction in exposure is not feasible, the generic reference level of existing annual dose below which intervention is not likely to be justifiable could provide a basis for discontinuing intervention. In any case, after intervention has been discontinued, the remaining residual existing annual dose should not influence the normal living conditions in the affected area (including decisions about the introduction of new practices), even if the dose is higher than that existing in the area before the accident.

3.6. Summary of recommendations for situations involving radioactive residues The quantitative recommendations provided are summarized in Table 5.1. The information is presented in a simplified form and the table does not include any reference to specific intervention and action levels of averted annual dose nor to collective doses in general. In its upper part, the table presents the quantitative recommendations on individual additional annual dose; the lower part shows the quantitative recommendations on individual existing annual dose. Therefore, in these two parts, the dose ranges are expressed in different quantities and cannot be compared.

4. Decision aiding and decision making The two international meetings cited earlier IAEA (2000; 2001) reviewed national experiences of the remediation of areas affected by radioactive residues. The situations considered included some which arose within the civil nuclear industry, including sites affected by uranium mining and milling, nuclear fuel fabrication and processing and nuclear power plant decommissioning; they included others from the military sector, including nuclear weapons production and testing, and others resulting from nuclear and radiation accidents; finally they included some from the decommissioning of non-nuclear facilities in which radionuclides had been employed. Both meetings included sessions in which case studies involving the assessment and remediation of areas affected radioactive contamination were presented and then analyzed. Many of the situations presented would, in terms of ICRP definitions, be identified as "intervention situations" but

116

G. Linsley Table 5.1. Summary of ICRP recommendations for situations involving radioactive residues. Concept

Quantity

Value [mSv]

Exemption for practices Criterion for deriving exemption levels for sources within practices, for which protection is optimum and which are part of a justified practice

Additional annual dose summation of all annual doses, transitory and prolonged, attributable to a source within a practice

Dose constraint for practices Applicable to the individual dose from a source within a practice; to be used for the optimization of protection of the source

Additional annual dose summation of all annual doses, transitory and prolonged, attributable to a source within a practice

=£0.3

Dose limit for practices Applicable to the individual dose contributed by all relevant practices

Aggregation additional annual dose summation of all annual doses, transitory and prolonged, attributable to all relevant practices

1

Lower generic reference level below which intervention is optional but not likely to be justifiable and above which intervention may be necessary

Existing annual dose summation of all prolonged annual doses attributable to all sources of prolonged exposure in a given location

^10

Upper generic reference level above which intervention should be considered almost always justifiable

Existing annual dose summation of all prolonged annual doses attributable to all sources of prolonged exposure in a given location

=£100

-0.01

were treated by national authorities as "practice situations". The dose criteria being applied as "targets" for restoration were usually 1 mSv/year or some smaller fraction of it. The basis for choosing these criteria was usually the dose limit for members of the public. The reasons for this are various; in some cases the ICRP concept of intervention is not recognized in national legislation, with the implication that all situations are effectively treated as practices. In others, the criteria had often been chosen taking into account the attitude of the public for whom the dose limit for members of the public is a

5 - International advice and experience

117

well-known and established reference value (even though it is designed for different purposes). In some areas affected by the residues from an accidental release of radioactive materials it was said that the public would not accept dose standards less than those being applied in so-called "normal" areas unaffected by radioactive residues. In some cases, the cost implications of applying dose criteria of 1 mSv/year or less as a basis for restoration action would be enormous because of the extent of the affected areas. Therefore, the only feasible strategy would be to exclude public access from such areas. Thus it became clear from the presentations and discussions at these conferences that decision making on environmental restoration involves many factors, not only those concerned with radiation risk but also consideration of economic and social factors and of public opinion and local politics. Radiation protection guidance is simply one of the factors; it is a decision aiding component, one element in the overall decision making process. The President of the Arlington Symposium, Dr. C.B. Meinhold, summarized the situation as follows, "It would seem that the international radiation protection community, IAEA and ICRP, for example, should continue to provide clear advice based on excellent science and sound professional judgement as their decision aiding contribution to the broader issue of decision making".

5. International assessments of areas affected by radioactive residues Since the first nuclear weapons were used in 1945 more than 2400 nuclear weapons experiments have taken place worldwide. The weapons test sites include: Algeria (Reggane and In-Ekker); Australia (Monte Bello, Emu and Maralinga); China (Lop Nor); Kazakhstan (Semipalatinsk and other sites); Russian Federation (Novaya Zemlya, Totsk, and Kapustin Yar); USA (Nevada and Amchitka, Alaska); various locations in the Pacific and Atlantic Oceans including the Marshall Islands, Maiden, Christmas and Johnston Islands, as well as the sites in India and Pakistan where testing was recently done. The IAEA, in the context of its statutory obligation to establish standards of safety for the protection of health and to provide for their application at the request of a State, has been involved in assisting countries to assess the radiological situations at several of these nuclear weapons test sites. In addition, the IAEA has organized an international assessment of the radiological impact of the dumping of radioactive waste in the Arctic Seas (IAEA, 1998d).

118

5.1.

G. Linsley

Semipalatinsk, Kazakhstan

In 1993, the Government of Kazakhstan informed the IAEA of its concern about the radiological situation in Semipalatinsk, where nuclear-weapons testing was carried out from 1949 to 1989. It requested assistance, and a preliminary radiological evaluation of Semipalatinsk was subsequently done (IAEA, 1998a). During the period 1949-1989 the former Soviet Union conducted about 460 nuclear weapons tests within the Semipalatinsk test site. They included explosions that were conducted on the surface or in the atmosphere. Five of these surface tests were not successful and resulted in the dispersion of plutonium in the environment. Starting in 1961, more than 300 test explosions were conducted underground. Thirteen of the underground tests resulted in release of radioactive gases to the atmosphere. The only on-site inhabitants during the testing program were in the town of Kurchatov whose purpose was to service the site, and in the small settlements of Akzhar and Moldari along the northern edge of the site. Recently there has been a limited amount of resettlement within the area, mostly by semi-nomadic farmers and herders. The bulk of the local population is in settlements just outside the site border. The total population of these settlements is estimated to be 30 000 to 40 000 people. Based on information collected during the missions and subsequent research, there is sufficient evidence to indicate that most of the area has little or no residual radioactivity directly attributable to nuclear tests in Kazakhstan. There are, however, areas that have elevated residual radioactivity levels within the test site where the surface tests were performed and where underground tests vented to the atmosphere. Preliminary surveys of these areas indicated that the contamination is relatively localized. Due to the limited amount of survey data that was collected during the missions, the existence of actinide residues from the failed nuclear tests could not be corroborated. Currently there are no restrictions of access to the nuclear test site and limited reoccupation has already begun. An assessment of the exposure of persons who, on a daily basis, visit the areas where the surface tests and vented underground explosions has been undertaken. The findings of this assessment indicate annual exposures in the region of 10 mSv, predominantly due to external exposure. If these areas were permanently settled in the future, estimated exposures could be up to 140 mSv per year. This annual exposure is above the ICRP generic reference level of 10 mSv/y at which intervention is expected to be undertaken. Remedial action is, therefore, considered necessary for these localized elevated areas. However, due to budgetary and other constraints, the most appropriate remedial action at this time may be to restrict access to these areas. The measurements made by the IAEA experts corroborate, to a reasonable degree, the extensive surveys carried out by different organizations

5 - International advice and experience

119

from Kazakhstan and the former Soviet Union. The combined results are considered sufficient to form the basis of a preliminary assessment of the radiological situation of the area around the Semipalatinsk test site. The one exception to the above conclusion is the drinking water supply. While samples of drinking water taken during the missions showed no elevated levels of artificial radionuclides, sampling was not comprehensive. As such it is difficult to draw general conclusions about the entire water supply. In addition the results do not provide any guarantee about the future security of the water supply. The external radiation dose rates and soil activity outside the test site are the same, or close to, typical levels in other regions and countries where no nuclear weapons testing had been carried out. Some areas show small increases but these are not significant in terms of the exposure to the local population. In summary, although the IAEA preliminary study could offer reasonable assurances of safety to the permanent resident population of the region, it found high levels of radioactive residues in some areas of the site itself. These findings do not take into account the potential radiological consequences of underground testing at Semipalatinsk, which the IAEA study did not assess.

5.2.

Bikini A toll, Marshall Islands

In 1994, the Government of the Republic of the Marshall Islands - a Pacific Ocean archipelago of around thirty atolls and a few reef islands - requested assistance from the IAEA. The request was to conduct an independent international review of the radiological conditions at Bikini Atoll and to consider and recommend strategies for the eventual rehabitation of the Atoll by the Bikinians (IAEA, 1998b). Bikini Atoll is located 850 kilometers northwest of Majuro on the northern fringe of the Marshall Islands and is composed of more than 23 islands and islets. Four islands (Bikini, Eneu, Nam and Enidrik) account for over 70% of the land area. Bikini and Eneu are the only islands of the atoll that have had a permanent population. In 1946, Bikini Atoll was the first site in the Marshall Islands used for nuclear-weapon testing by the United States. In 1948, Enewetak Atoll, a neighboring atoll, replaced Bikini Atoll as the test site. In 1954, Bikini Atoll was reactivated as a test site until the US terminated nuclear-weapon testing in the Marshall Islands in 1958. Prior to the first nuclear test in 1946, the 167 Bikinians living on Bikini Island were evacuated to Rongerik Atoll, about 200 kilometers to the east. It was intended that they should reside there until an unspecified future date when the testing would be completed. The Bikinians remained on Rongerik Atoll for a period of two years. In 1948, they were moved briefly to Kwajalein Atoll and later in the same year to Kili, a small reef.

120

G. Linsley

Problems arose, however. They included the fact that Kill has no lagoon, no protective reef and no fishing grounds. The small beach is frequently subject to high waves. The Bikinians saw the move to Kili as a temporary relocation and were reluctant to change from being fishermen to farmers. By the time nuclear weapons testing in the Marshall Islands was terminated in July 1958, sixteen nuclear weapon tests had been conducted on Bikini Atoll over a 12-year period. All of those tests were surface or atmospheric tests. They were conducted in or over the Atoll lagoon, thereby dispersing the explosion's effects over all of the islands of the Atoll. The history of the radiological assessments and the movement of the local population is very important in understanding the overall concerns. In August 1968 - following a number of radiological surveys that had been carried out since 1958 to assess the impact of the US program of nuclearweapon testing - it was announced that Bikini Atoll was safe for habitation and approved for resettlement. The atoll was cleared of debris and fruit trees were replanted. A further radiological survey of Bikini Atoll was carried out in 1970. Eventually, 139 Bikinians resettled on the Atoll. However, the Bikini people remained unconvinced of the safety of the Atoll, and in 1975 they initiated a lawsuit against the US Government to terminate the resettlement effort until a satisfactory and comprehensive radiological survey had been carried out. Additional radiological data were collected for evaluation in 1975, 1976, and 1978. In September 1978 it was decided to relocate the 139 Bikinians who had returned to Bikini Atoll back to Kili Island and to Ejit Island at Majuro Atoll. After a second relocation, a new radiological survey, sponsored by the United States, was performed. This survey consisted of using detectors mounted in helicopters to plot contours of external gamma dose rates. Also, samples of vegetation, marine foods, animals and soil were collected and analyzed. Revised radiation dose evaluations were published in 1980 and 1982. They indicated that - should the Bikinians decide to resettle their island - the terrestrial food chain would be the most significant exposure pathway. This dose assessment was most recently updated in 1995 on the basis of a continuing measurement program at the Atoll. Following the US survey, the Government of the Republic of the Marshall Islands commissioned a separate radiological assessment. By this means, Bikini Atoll, as well as all other atolls in the republic, were to be monitored for radioactive residues. A Scientific Advisory Panel of wellknown and respected scientists provided oversight. Laboratory quality control programs were implemented to ensure that the surveys could provide accurate measurements and reproducible data. In general, the study confirmed the findings of earlier measurement programs. The findings of the survey were published and a report on Bikini Atoll was released in February 1995.

5 - International advice and experience

121

In August 1995, six months after the survey report was issued, the Nitejela (Parliament) of the Marshall Islands considered the survey findings but it did not accept them.

Request for an International Review In 1994, the Government of the Marshall Islands requested the IAEA to conduct an independent international review of the radiological conditions at Bikini Atoll, and to consider and recommend strategies for the resettlement of the atoll. The IAEA responded to this request by convening an Advisory Group, which met in December 1995. The Group was convened under the framework of an IAEA technical co-operation project. There were three main objectives of the international review: • to assess the radiological conditions on Bikini taking into account the information submitted by the republic's Government; • to ascertain whether any corroboration of the available information on the current radiological conditions at the Atoll is needed; and • to determine whether any remedial actions for the purpose of radiation protection were required and, if so, the form, scale and duration of such an intervention. The international review took into account all of the available data from the Marshall Islands' survey, as well as of a large number of other assessments made by scientists from around the world.

IAEA survey of Bikini Atoll In May 1997, the IAEA sent an environmental monitoring team to Bikini Atoll to perform a limited program of environmental measurements and sampling. Measurements were made of the absorbed dose rate on the air and of the concentration of the most radiologically significant radionuclides in representative samples of soil and foodstuffs. The purpose of this survey was to validate the data that had been obtained by previous surveys. Measurements taken during the survey were generally in good agreement with previously reported values.

Conclusions Based on its review, the IAEA Advisory Group determined that no further corroboration of the measurements and assessments of the radiological conditions at Bikini Atoll is necessary. The data that have been collected are of sufficient quality to allow an appropriate evaluation to be performed. The

122

G. Linsley

limited IAEA monitoring of the area provided a good quality assurance verification of the previously collected data. It was recommended that Bikini Island should not be permanently resettled under the present radiological conditions. This recommendation was based on the assumption that persons resettling on the island would consume a diet of entirely locally produced food. It was concluded that if a diet of this type were permitted, it could lead to an annual effective dose of about 15 mSv. This level was judged to require intervention of some type for radiation protection purposes. There are a number of remedial actions that could be performed that could lead to permanent rehabitation of the island. These include the periodic application of potassium-based fertilizer where edible crops are grown or the removal of the topsoil from the island. It is generally felt that the most reasonable approach would be to use potassium fertilizer. Since most of the radioactivity in the plants is due to the uptake of the radioactive cesium, the potassium would replace this element, thereby reducing the overall exposure to the population. This has been verified by limited scale testing of the technique. The scraping and removal of the topsoil would cause serious environmental harm to the island and have social consequences.

5.3.

Mururoa and Fangataula, French Polynesia

In August 1995, France became the first nuclear weapons State to ask the IAEA to evaluate a nuclear test site, namely the Atolls of Mururoa and Fangataufa in French Polynesia. France had conducted 193 nuclear experiments at these atolls. Following the French request, the IAEA organized the study which has been finalized and published by the IAEA in eight volumes (IAEA, 1998c). Mururoa Atoll has been populated only occasionally in the past, and there is no evidence that Fangataufa has ever been inhabited. The lack of a water supply, and the vulnerability of the atolls to the sea, make it difficult for people to live there. However, for the purposes of the International Study on the Radiological Situation at the Atolls of Mururoa and Fangataufa, the existence of a hypothetical population resident on Mururoa was assumed in order to determine potential radiation doses. The estimation of doses to more distant communities was also necessary in order to establish the significance of any releases of radioactive materials. The assessment takes account of dispersion of radionuclides both from the underground and atmospheric tests, and from accelerated releases of material due to disruptive events of natural or human origin, such as a landslide, or to changes in climatic conditions.

5 - International advice and experience

123

The assessment was concerned with the present and future radiation doses due to residues at Mururoa and Fangataufa Atolls. Doses received in the past as a result of the fallout at the time of the French atmospheric testing were not evaluated in the Study.

Categories of exposure The Study estimated dose rates due to exposure to the residual radionuclides from the French nuclear tests for critical groups of people, for both present and future conditions. Estimates were made for the following categories of groups: • present exposure for hypothetical inhabitants of Mururoa and Fangataufa atolls; • present exposure for inhabitants of Tureia Atoll, the nearest inhabited atoll in the region; • future exposure for inhabitants of the region and any inhabitants of Mururoa and Fangataufa atolls as a consequence of the residual radioactive material now present in the environment and that part of the radioactive material contained underground which will migrate to the accessible environment in the future; • potential exposure of inhabitants of the region and hypothetical inhabitants of Mururoa and Fangataufa atolls as a consequence of postulated disruptive events.

Pathways of exposure The major contribution to radiation exposure was assessed to be via the ingestion pathway. Realistic diets were used for populations such as those dwelling on Tureia Atoll, the nearest inhabited atoll, and for hypothetical inhabitants of Mururoa and Fangataufa atolls. For hypothetical populations elsewhere, high consumption rates, in particular for seafood, were assumed to ensure that upper limit estimates of dose rates were obtained. Where possible, the concentrations of radionuclides in foodstuffs were obtained by direct measurement. If direct measurement was not possible, as for hypothetical populations, concentrations in foodstuffs were estimated from measurements of soil concentrations.

Present doses at the atolls A population permanently resident on the atolls with a diet of local produce and seafood from the lagoons would not generally receive a radiation dose

124

G. Linsley

attributable to the residual radioactive material exceeding 0.01 mSv per year. This is equivalent to a very small fraction (less than one part in 200) of the total dose that such a resident population would unavoidably receive from natural radiation sources. (Tab. 5.II). The dose estimates are based on measured levels of man-made radionuclides in the environments of the atolls, which will include contributions from global fallout (i.e. fallout from all atmospheric nuclear weapon testing). Except in the case of seafoods, it is not possible to determine those contributions to the measured environmental levels, and therefore those fractions of the estimated doses that are due to global fallout. The present average annual dose within the entire 20° to 30° southern latitude band of Mururoa Atoll due to global fallout is estimated to be of the order of 0.002 to 0.003 mSv. Comparisonof ofpredicted predictedadditional additionalradiation radiationdoses dosesatat Mururoa, Mururoa, Table 5.II. 5.II. Comparison Fangataufa and Tureia Tureia atolls atolls with natural background. Source of dose

Dose fcnSv per year)

Global natural background doses • Typical range • Maximum • Average

I1tto o l10 O -100 -100 2.4 2.4

Mururoa and Fangataufa atolls • Dose due to natural background radiation

1.4to 3 1.4

Estimated current additional doses from remaining residual radioactive material at Mururoa and Fangataufa atolls • Maximum at Tureia Atoll • Average at Mururoa and Fangataufa atolls • Maximum at Kilo-Empereur region of Fangataufa Atoll