NUCLEAR SAFETY

INFORM ATION CENTRE

FOR REFERENCE PURPOSES PLEASE DO NOT REMOVE THIS COPY RADIATION AND SOCIETY: COMPREHENDING RADIATION RISK The following States are Members of the International Atomic Energy Agency:

AFGHANISTAN HOLY SEE PERU ALBANIA HUNGARY PHILIPPINES ALGERIA ICELAND POLAND ARGENTINA INDIA PORTUGAL ARMENIA INDONESIA QATAR AUSTRALIA IRAN, ISLAMIC REPUBLIC OF ROMANIA AUSTRIA IRAQ RUSSIAN FEDERATION BANGLADESH IRELAND SAUDI ARABIA BELARUS ISRAEL SENEGAL BELGIUM ITALY SIERRA LEONE BOLIVIA JAMAICA SINGAPORE BOSNIA AND JAPAN SLOVAKIA HERZEGOVINA JORDAN SLOVENIA BRAZIL KAZAKHSTAN SOUTH AFRICA BULGARIA KENYA SPAIN CAMBODIA KOREA, REPUBLIC OF SRI LANKA CAMEROON KUWAIT SUDAN CANADA LEBANON SWEDEN CHILE LIBERIA SWITZERLAND CHINA LIBYAN ARAB JAMAHIRIYA SYRIAN ARAB REPUBLIC COLOMBIA LIECHTENSTEIN THAILAND COSTA RICA LITHUANIA THE FORMER YUGOSLAV COTE D'IVOIRE LUXEMBOURG REPUBLIC OF MACEDONIA CROATIA MADAGASCAR TUNISIA CUBA MALAYSIA TURKEY CYPRUS MALI UGANDA CZECH REPUBLIC MARSHALL ISLANDS UKRAINE DENMARK MAURITIUS UNIT¿D ARAB EMIRATES DOMINICAN REPUBLIC MEXICO UNITED KINGDOM OF ECUADOR MONACO GREAT BRITAIN AND EGYPT MONGOLIA NORTHERN IRELAND EL SALVADOR MOROCCO UNITED REPUBLIC ESTONIA MYANMAR OF TANZANIA ETHIOPIA NAMIBIA UNITED STATES OF AMERICA FINLAND NETHERLANDS URUGUAY FRANCE NEW ZEALAND UZBEKISTAN GABON NICARAGUA VENEZUELA GEORGIA NIGER VIET NAM GERMANY NIGERIA YEMEN GHANA NORWAY YUGOSLAVIA GREECE PAKISTAN ZAIRE GUATEMALA PANAMA ZAMBIA HAITI PARAGUAY ZIMBABWE The Agency’s Statute was approved on 23 October 1956 by the Conference on the Statute o f the IAEA held at United Nations Headquarters, New York; it entered into force on 29 July 1957. The Head­ quarters of the Agency are situated in Vienna. Its principal objective is “ to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world” . © IA E A , 1996 Permission to reproduce or translate the information contained in this publication may be obtained by writing to the International Atomic Energy Agency, Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria.

Printed by the IAEA in Austria August 1996 STI/PUB/959 PROCEEDINGS SERIES

RADIATION AND SOCIETY: COMPREHENDING RADIATION RISK

PROCEEDINGS OF AN INTERNATIONAL CONFERENCE ON RADIATION AND SOCIETY: COM PREHENDING RADIATION RISK ORGANIZED BY THE INTERNATIONAL ATOMIC ENERGY AGENCY AND HELD IN PARIS, 24-28 OCTOBER 1994

V O L U M E 2

INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1996 VIC Library Cataloguing in Publication Data

International Conference on Radiation and Society : Comprehending Radiation Risk (1994 : Paris, France) Radiation and society : comprehending radiation risk : proceedings of an International Conference on Radiation and Society : Comprehending Radia­ tion Risk / organized by the International Atomic Energy Agency and held in Paris, 24-28 October 1994. — Vienna : The Agency, 1994. 3 v. ; 24 cm. — (Proceedings series, ISSN 0074-1884) STI/PUB/959 ISBN 92-0-103096-7 (v.2) Includes bibliographical references.

1. Radiation— Risk assessment— Congresses. I. International Atom ic Energy Agency. II. Title. III. Series: Proceedings series (International Atomic Energy Agency).

VICL 96-00153 FOREWORD

An understanding of risk is essential for making optimum choices and for gaining acceptance of the decisions made. Comprehending radiation risk is a challenge for society. In order to explore the subject and to bring about a better understanding of the risks associated with exposure to ionizing radiation, an IA E A international conference on Radiation and Society: Comprehending Radiation Risk, was held from 24 to 28 October 1994. Upon invitation by the Government of France, the conference was held in Paris. It was hosted by the Institut de protection et de sûreté nucléaire (IPSN) and convened at the new conference facilities in the Carrousel du Louvre. This conference was the first major international meeting devoted to the comprehension of radiation risk and the interest was significant. The format of the conference, with ample opportunity for discussion, attracted a large audience of over 400 persons from 50 countries and 8 international organizations, a mixture of techni­ cal specialists, social scientists, media professionals, and policy and decision makers, including high level officials from national health authorities and regulatory b o d ie s . The plenary programme consisted of three distinct elements: three ‘technical days’, a ‘media day’, and a ‘decision makers day’. Prior to the conference, 12 back­ ground papers had been prepared to serve as a basis for the technical discussions. In parallel to the plenary sessions, scientific posters were exhibited. The proceedings of the conference are published in three volumes. Volume 1 (issued in July 1994) contains the ten background papers prepared for the conference by Riskkollegiet (Swedish Risk Academy). Volume 2 contains the two background papers prepared for the conference by IP SN , and the text of the poster presentations. Volum e 3 contains the speeches, the ten introductory papers, summaries of the tech­ nical discussion sessions, the keynote paper on uncertainties in the health impact of environmental pollutants and summaries of the discussions in various forums. EDITORIAL NOTE

The Proceedings have been edited by the editorial staff of the IAEA to the extent considered necessary for the reader's assistance. The views expressed remain, however, the responsibility of the named authors or participants. In addition, the views are not necessarily those of the governments o f the nominating Member States or of the nominating organizations. Although great care has been taken to maintain the accuracy of information contained in this publication, neither the IAEA nor its Member States assume any responsibility for consequences which may arise from its use. The use of particular designations of countries or territories does not imply any judge­ ment by the publisher, the IAEA, as to the legal status of such countries or territories, of their authorities and institutions or of the delimitation of their boundaries. The mention of names of specific companies or products (whether or not indicated as registered) does not imply any intention to infringe proprietary rights, nor should it be construed as an endorsement or recommendation on the part of the IAEA. The authors are responsible for having obtained the necessary permission for the IAEA to reproduce, translate or use material from sources already protected by copyrights. Material prepared by authors who are in contractual relation with governments is copyrighted by the IAEA, as publisher, only to the extent permitted by the appropriate national regulations. CONTENTS

BACKGROUND PAPERS PREPARED FOR THE CONFERENCE

B Y I P S N

Management of radiation risk IAEA-CN-54/1B ...... 3 P. Hubert Impact of radiation on the environment (IAEA-CN-54/2B) ...... 41 D. Robeau

POSTER PRESENTATIONS

ASSESSMENT OF RADIATION EXPOSURE LEVELS

(Technical Session 1)

Annual effective radiation doses from natural sources in Moldavia ( I A E A - C N - 5 4 / 3 4 P ) ...... 51 O. Iacob, E. Botezatu, C. Crecea, C. Diaconescu, L. Clain A W orkshop on Harmonization of East-W est Radioactive Pollutant Measurements, Standardization of Techniques and Considerations of Socioeconomic Factors: Report on the Commission of European Communities Workshop held in Budapest, Hungary, A u g u s t 1 9 9 4 ( I A E A - C N -5 4 / 4 1P ) ...... 5 6 B.R. Orton, D.A. Vorsatz Mid-term radioecological and radiobiological consequences of Chernobyl fallout in an alpine environment (IAEA-CN-54/45P) ...... 6 2 H. Lettner, W. Hofmann, J. Pohl-Riiling, F. Steinhäusler Apport de nouveaux dosimètres électroniques pour la dosimétrie opérationnelle en médecine nucléaire (IAEA-CN-54/49P) ... 68 B. Aubert, A. Lamon, C. Parmentier La protection des intervenants de la maintenance des c e n tra le s n u c lé a ir e s fr a n ç a is e s ( I A E A - C N - 5 4 / 5 2 P ) ...... 7 3 R. Dollo Occupational exposure in Japan from 1976 to 1992 (IAEA-CN-54/66P) ... 76 T. Yamaguchi, K. Kawai A study of the radioecological situation in regions adjoining the 30 km zone around the Chernobyl nuclear power plant ( I A E A - C N - 5 4 / 7 1P ) ...... 83 M.D. Bondarkov, I.N. Vishnevsky, N.P. Donets, V.A. Zheltonozhsky, A.A. Sotnikov Evaluation of radiation burdens on the basis of spectrometric a n d r a d io m e tric d ata ( I A E A - C N - 5 4 / 7 2 P ) ...... 8 6 M.D. Bondarkov, I.N. Vishnevsky, N.P. Donets, V.A. Zheltonozhsky, A.A. Sotnikov Are human teeth an indicator of 90Sr contamination? ( I A E A - C N - 5 4 / 8 1 P ) ...... 9 0 E. Botezatu, L. Birleanu Les observations permanentes de la radioactivité de l’IPSN : Evolution du césium 137 dans les aérosols atmosphériques entre 1958 et 1994 (IAEA-CN-54/89P) ...... 9 3 D. Calmet, P. Bouisset, G. Kerleau, J. Allenou Transfer of radioactive caesium to hunters and their families by food chains in forest ecosystems (IAEA-CN-54/92P) ...... 9 8 G. Àgren, R. Bergman, B.-M. Drottz-Sjöberg, A. Enander, R. Falk, K.J. Johansson, L. Johansson S p a n is h N a t io n a l D o s im e t r y B a n k ( I A E A - C N - 5 4 / 9 4 P ) ...... 10 6 J. Muñoz Qualities and quantities in radiation protection (IAEA-CN -54/96P) ...... 10 9 D.T.Y. Chen Occupational exposure to radon progeny (IAEA-CN-54/100P) ...... 115 W. Kraus, W. Röhnsch, J. Schwedt, W. Ullmann In situ measurements of density of 137Cs contamination of the forest system in Ukraine (IAEA-CN-54/109P) ...... 12 0 V. Poiarkov, A. Nazarov, H. Shames

ASSESSMENT OF RADIATION HEALTH EFFECTS

(Technical Session 2)

Radioactive ‘hot particles’: Are they still a problem? (IAEA-CN-54/1P) .. 129 M.W. Charles, P.J. Darley Cancer risk for United States nuclear workers, 1970-2000 ( I A E A - C N - 5 4 / 1 2 P ) ...... 13 4 R.E. Lapp Prenatal X ray exposure, cosmic radiation and unwarranted p r e g n a n c y te rm in a tio n s ( I A E A - C N - 5 4 / 2 6 P ) ...... 137 Y. Kalish Radiation risk facts versus perceptions (IAEA-CN-54/32P) ...... 1 40 R.E. Lapp Data needs for analysis of characteristics of the recent increase of childhood thyroid cancer incidence in Belarus ( I A E A - C N - 5 4 / 4 2 P ) ...... 143 T. Abelin, F. Gurtner, M. Egger, J.I. Averkin, A.E. Okeanov Lifetim e risk due to selected causes of Indonesians’ death (IAEA-CN-54/105P) ...... 147 E. Hiswara

IMPACT OF RADIATION ON THE ENVIRONMENT

(Technical Session 3)

Technical information background for decision making in case of nuclear accidents (IAEA-CN-54/19P) ...... 153 P. Zombori, A. Andrási, J. Urbán, I. Németh Impact of natural radioactivity from coal fired power plants ( I A E A - C N - 5 4 / 2 1P ) ...... 158 J. Kovac, N. Novosel Le contrôle de la radioactivité à proximité des centrales é le c tr o n u c lé a ire s fra n ç a is e s ( I A E A - C N - 5 4 / 5 4 P ) ...... 161 A. Lecorre S e v e r e ra d ia tio n a c c id e n ts a n d th e e n v iro n m e n t ( I A E A - C N - 5 4 / 6 9 P ) ...... 163 R.M. Alexakhin Surveillance de l’impact des rejets marins de l’usine de re tra ite m e n t d e L a H a g u e ( I A E A - C N - 5 4 / 9 0 P ) ...... 16 6 S. Le Bar Radioactive effluents from spanish light water nuclear power plants: T h e in te rn a tio n a l situ a tio n ( I A E A - C N - 5 4 / 9 5 P ) ...... 17 0 M.J. Barahona, L.M. Ramos Action levels for radon exposures caused by radon exhalation from relics of uranium mining and m illing (IAEA-CN-54/98P) ...... 175 W. Kraus, M. Kümmel, S. Przyborowski, W. Röhnsch Soil-wood l37Cs transmission coefficient for Ukrainian c o n ife r o u s fo re sts ( I A E A - C N - 5 4 / 1 0 8 P ) ...... 18 0 V. Poiarkov, A. Nazarov, A. Zagrebin, N. Davidov Assessment of radiation exposure from discharges of uranium ore concentrate processing and nuclear fuel reprocessing facilities ( I A E A - C N - 5 4 / 1 1 7 P ) ...... 18 4 D.J. Assinder, S.M. Mudge, G.S. Bourne

PERCEPTION OF RADIATION RISK (Technical Session 4)

Perception of natural, medical, and ‘artificial’ radiation exposures ( I A E A - C N - 5 4 / 4 P ) ...... 191 K. Becker N u c le a r d e m o c r a c y ( I A E A - C N - 5 4 / 9 P ) ...... 195 E. Tóth P e o p le a n d r is k s ( I A E A - C N - 5 4 / 1 0 P ) ...... 19 9 G. Marx Ionizing radiation: On why present terminology causes confusion ( I A E A - C N - 5 4 / 1 3 P ) ...... 2 0 5 J. Christensen Promoting comprehension and understanding of radiation p ro te c tio n ( I A E A - C N - 5 4 / 1 4 P ) ...... 2 1 0 M.J. Gaines Importance of public perception (IAEA-CN-54/23P) ...... 2 1 7 J. Read L ’ im p o rta n c e d e la p e rc e p tio n d u p u b lic ( I A E A - C N - 5 4 / 2 3 P ) ...... 2 2 0 J.A. Read Comprehending radiation risk in Lithuania (IAEA-CN-54/24P) ...... 2 2 3 D. Sidiskiene, T. Nedveckaite Comment mieux informer le public de la réalité des ra y o n n e m e n ts io n is a n ts ( I A E A - C N - 5 4 / 5 1 P ) ...... 2 2 9 J. P. Chaussade Nucléaire et information des professions de santé (IAEA-CN-54/53P) ...... 2 3 1 C. Gallin-Martel, C. Vrousos, H. Kolodie, H. Pons, M. Durr Psychological dimensions of risk perception and trade-off ( I A E A - C N - 5 4 / 6 4 P ) ...... 2 3 4 Y. Tanaka Important aspects of risk perception concerning the u r a n iu m in d u s tr y ( I A E A - C N - 5 4 / 7 5 P ) ...... 2 3 9 L. H. S. Veiga, E. C. S. Amaral, G. Blaylock Analysis with fuzzy measure theory of public attitude towards th e u s e o f n u c le a r e n e r g y ( I A E A - C N - 5 4 / 8 4 P ) ...... 2 4 4 Y. Nishiwaki, J. Benedikt, K. Fujimoto, C. Preyssl Perception des risques nucléaires et information (IAEA-CN-54/91P) ...... 2 5 0 J. Brenot, S. Bonnefous, P. Hubert Misrepresentation of radiation risk: The main approach in the anti-nuclear campaign in Russia (IAEA-CN-54/93P) ...... 2 5 7 G.A. Kaurov Risk perceptions of Indonesian students vs. US and Hong Kong students ( I A E A - C N - 5 4 / 1 0 4 P ) ...... 2 6 0 E. Hiswara Radioecological education and perception of radiation risk in B e la r u s ( I A E A - C N - 5 4 / 1 1 4 P ) ...... 2 6 4 A.I. Stavrov Comprehending radiation risks by assessment of incidents and their causes in licensed activities (IAEA-CN-54/115P) ...... 2 6 6 R.J. Catlin Perception of radiation risks associated with the disposal of radioactive transuranic waste in New Mexico (IAEA-CN-54/116P) ...... 2 7 3 R.H. Neill Perception and interpretations of radiation risk in the population: A p r e lim in a r y re p o rt ( I A E A - C N - 5 4 / 1 2 0 P ) ...... 2 7 6 A. Tonessen, J.B. Reitan, P. Strand, R. Waldahl, L. Weisceth, B. Màrdberg Aversion to radiation: An ethical perspective (IAEA-CN-54/121P) ...... 2 8 3 D.H. Oughton

M ANAGING RADIATION RISK (Technical Session 5)

Regulation of practice involving radiation sources in th e S lo v a k R e p u b lic ( I A E A - C N - 5 4 / 3 9 P ) ...... 2 9 1 P. Vrabcek Changing the occupational dose limit for pregnant workers: Public consultation and consensus seeking (IAEA-CN-54/48P) ...... 2 9 4 C. Pomroy Actions entreprises par Electricité de France en Bélarus: Une solidarité entre électriciens (IAEA-CN-54/50P) ...... 3 0 0 J. Lallemand, M. Bertin, D. Hubert Les objectifs de la radioprotection à Electricité de France: L e L i v r e B la n c d e la r a d io p ro te c tio n ( I A E A - C N - 5 4 / 5 9 P ) ...... 3 0 3 Ph. Rollin Reduction of workers’ exposure and dose allocation (IAEA-CN-54/65P) .. 306 J. Onodera, T. Nishizono Les mesures radon réalisées par l’IPSN durant la réhabilitation du site de stockage de résidus de traitement de minerai d’uranium d u B o u c h e t , e n r é g io n p a ris ie n n e ( I A E A - C N - 5 4 / 8 5 P ) ...... 3 1 0 V. Labed, M.C. Robé, A. Beneito, J.M. Maurel, P. Richon La gestion du risque radiologique en exploitation dans l’usine de retraitement de La Hague (IAEA-CN-54/86P) ...... 3 1 5 J. Kalimbadjian Strategy of making public exposure decisions, taking into account the radiation incidents experience of Minatom of Russia ( I A E A - C N - 5 4 / 1 0 1 P ) ...... 32 1 V.A. Gubanov Management of radiation risk: A guide to demonstrating A L A R P fo r a n e x is t in g p la n t ( I A E A - C N - 5 4 / 1 2 5 P ) ...... 3 2 7 J. Telfer, S. Mortin THE NUCLEAR W EAPONS LEGACY (Case Study 1)

Cancer risk among nuclear bomb survivors (IAEA-CN-54/67P) ...... 3 3 7 Y. Shimizu, K. Mabuchi, I. Shigematsu Dose response of radiation induced cancer at low dose level a m o n g n u c le a r b o m b s u r v iv o r s ( I A E A - C N - 5 4 / 6 8 P ) ...... 3 4 2 S. Fujita, Y. Shimizu, K. Mabuchi, I. Shigematsu Dose assessment, radioecology and community interaction at former nuclear test sites (IAEA-CN-54/127P) ...... 3 4 6 W.L. Robison

CANCER AND LEUKAEM IA CLUSTERS (Case Study 2)

Risk of childhood leukaemia from the radiation exposure of fathers before conception (IAEA-CN-54/25P) ...... 3 5 3 R. Wakeford, W.J. Anderton Cluster of childhood leukaemias near the German boiling water reactor at Krümmel: Ways of elucidation (IAEA-CN-54/46P) ...... 3 5 7 I. Schmitz-Feuerhake, B. Dannheim, I. Grell-Büchtmann, A. Heimers, W. Hofmann, B. Oberheitmann, H. Schröder, H. Ziggel Cluster of childhood leukaemia in Elbmarsch, Germany ( I A E A - C N - 5 4 / 8 3 P ) ...... 3 6 1 D. Harder

RADON IN HOM ES (Case Study 3)

The optimization of communication and decision making in ra d o n p o lic ie s ( I A E A - C N - 5 4 / 3 P ) ...... 3 6 9 G.X. Eggermont, A.J. Poffijn Radon programme in the Czech Republic (IAEA-CN-54/20P) ...... 3 7 6 E. Tylová Radon in Latvia’s dwellings (IAEA-CN-54/31P) ...... 3 7 9 M. Dambis An example of elevated indoor radon concentration in a k in d e r g a rte n in S lo v e n ia ( I A E A - C N - 5 4 / 3 6 P ) ...... 3 8 3 J. Vaupotic, I. Kobal, P. Jovanovic, J. Planinió Uncertainties in exposures to radon progeny: Their impact on the evaluation of risk estimates (IAEA-CN-54/38P) ...... 3 8 8 P. Duport, R. Janica Mesure du radon dans les habitations en France (IAEA-CN-54/87P) ...... 3 9 4 M.H. El Jammal, J.P. Gambard, J. Carmes Radon et risque de cancer: Etudes epidémiologiques après e x p o s itio n p r o fe s s io n n e lle o u d o m e s tiq u e ( I A E A - C N - 5 4 / 8 8 P ) ...... 3 9 7 M. Tirmarche Countermeasures against excessively enhanced levels of radon in homes in mining areas (IAEA-CN-54/99P) ...... 3 9 9 R. Czarwinski, W. Kraus, R. Lehmann, W. Röhnsch Study of indoor radon concentration in Poland (IAEA-CN-54/126P) ...... 4 0 5 K. Mamont-Cieêla, J. Jagielak, S. Rosiiíski, A. Sosiñska, M. Bysiek, J. Henschke

RADIOACTIVE W ASTE DISPOSAL AND THE ENVIRONMENT

(Case Study 4)

Overcoming the risk perception gap: A socially acceptable a p p ro a c h to w a ste m a n a g e m e n t ( I A E A - C N - 5 4 / 5 P ) ...... 4 1 3 E.R. Frech, M.A. Greber Radioactive waste disposal and the environment in Latvia ( I A E A - C N - 5 4 / 3 7 P ) ...... 4 1 8 A. Salmins T r a n s m u ta tio n o f n u c le a r w a ste : S a fe ty a sp e cts ( I A E A - C N - 5 4 / 8 0 P ) ...... 4 2 2 T. Thedéen

CHERNOBYL HEALTH EFFECTS (Case Study 5)

Internal dosimetry, biological dosimetry, and biomedical studies in children from the former U SSR exposed to the consequences of the Chernobyl accident (IAEA-CN-54/22P) ...... 4 2 7 F. Mauro, R. D e Vita, L. Padovani, M. Spand, G. Tarroni, S. Bazzarri, P. Metalli, G. Serlupi Crescenzi Cancer and non-cancer diseases of the thyroid gland and their dose dependence in children and adolescents affected as a result of the Chernobyl accident (IAEA-CN-54/27P) ...... 4 3 2 A.F. Tsryb, E.M. Parshkov, V.K. Ivanov Dose-morbidity dependence of Chernobyl accident emergency w o rk e r s ( I A E A - C N - 5 4 / 2 8 P ) ...... 4 3 6 V.K. Ivanov, E.M. Rastopchin, A.F. Tsyb Delayed radiation effects of the accident at the Chernobyl nuclear power plant on the exposed population in the R u s s ia n F e d e r a t io n ( I A E A - C N - 5 4 / 2 9 P ) ...... 4 3 9 A.I. Gorsky, V.K. Ivanov, A.F. Tsyb Reconstruction of the absorbed external doses to the population living in areas of the Russian Federation contaminated as a result of the Chernobyl accident (IAEA-CN-54/30P) ...... 444 V.A. Pitkevich, V.K. Ivanov, A.F. Tsyb, V.M. Sheshakov, A.V. Golubenkov, R.V. Borodin, К S. Kosykh Consequences of the Chernobyl accident in Georgia (IAEA-CN-54/33P) .. 449 D. Mandzhgaladze, M. Shavdiya Concentrations of the most important radionuclides in milk in Slovenia after the Chernobyl nuclear accident (IAEA-CN-54/35P) ... 451 P. Jovanovic, M. Kanduc Computer solution for a cancer epidemiology research department in the Republic of Belarus (IAEA-CN-54/43P) ...... 453 M.P. Günter, J.P. Bleuer, T. Abelin, J. Averkin, I. Prudyvus, S. Rubtsov BACKGROUND PAPERS

Prepared for the Conference by IPSN

IAEA-CN-54/1B

MANAGEMENT OF RADIATION RISK

P. HUBERT Institut de Protection et de Sûreté Nucléaire, Fontenay-aux-Roses, France

1. INTRODUCTION

The need to control the risk from ionizing radiation can be tracked back to the eve of the twentieth century. However, as knowledge improved and practices expanded, the approaches to this control have evolved. No longer is the mere respect of some forms of exposure limits or safety related standards sufficient. Rather, it is widely admitted that there is a need for managing radiation risk, both by balancing the advantages and disadvantages of enhancing protection and by setting up a proper organization that allows handling of the risk. This paper describes the multiple aspects of radiation risk management and points out the main related issues.

2. RISK MANAGEMENT PRINCIPLES AND SYSTEMS

2.1. Managing radiation risk among other risks

Publication 26 of the International Commission on Radiological Protection (ICRP) [1] is generally considered a landmark in the development of risk manage­ ment in the field of protection, although there were elements in previous ICRP docu­ ments, as well as in non-radiological areas. Indeed, this document explicitly states that the limitation of exposures is not the unique principle of protection, but that a practice entailing exposures should be justified and exposures kept ‘as low as reasonably achievable’ (ALARA), economic and social considerations being taken into account. At that time, for most people, this ‘ALARA principle’ generally meant that an operator had to perform a cost-benefit analysis when dimensioning protection or safety equipment [2]. Hence a great weight was given to the need to find monetary values for the detriment associated with exposures. At the present stage of develop­ ment, such an economic optimization is not viewed as the alpha and omega of risk management; rather, it is the tip of an iceberg [3, 4]. A broad set of management principles, assessment tools, organization of work and controls must be implemented that will be addressed thereafter. ICRP approaches have inspired national regulations on radiation protection in Europe, and similar provisions have been made in the USA (e.g. Ref. [5]) based on

3 4 BACKGROUND PAPERS similar principles [6]. The idea that zero risk is not achievable and that there should be some balance between risks and the cost of safety was also applied to accident risk management, although there is some reluctance in establishing it as a regulatory requirement, for reasons that will be discussed further. The US National Regulatory Commission and the UK Health and Safety Executive have elaborated much on the subject (e.g. Ref. [7] for assessment and Ref. [8] for policy making), and the Inter­ national Atomic Energy Agency is making syntheses in this field [9]. It would be misleading to consider risk management as an approach specific to radiation risk. On the contrary, it has been adopted in numerous areas, either in regulations or in regulatory principles or explanatory material from authorities. Some countries have promoted risk management principles that are a common framework for the management of risks of every nature, both accidental and arising from routine exposures, including even domestic risks as in the Netherlands [10, 11], or biological hazards and highway hazards as in a Federal regulation in Switzerland [12]. In other countries, the scope of regulation is not ‘universal’ , either because the regulated field is restrained or because the use of risk management deci­ sion aiding techniques is only a recommended input in assessment of a safety case, such as for the general industrial risk in the UK [13] or traffic accidents in France, where cost-benefit approaches are recommended [14] or economic assessment of land use planning was performed [15]. In the USA the need to use decision aiding tools for managing risks was recognized in the late 1970s [16]. In dealing with risk, three rationales can be distinguished. In a quite synthetic review of the US Federal acts dealing with risk, the 30 major acts have been classi­ fied as ‘health only’ , ‘technology only’ or ‘balancing’ statutes [17]. Regulations under the ‘health only’ statutes are those in which health criteria should be the only ones considered. Setting up of zero risk limits, or even negligible risk limits (as the ICRP did in 1958 [18]), pertains to this logic. Examples are portions of the Clean Air Act and the well known Delaney clause (1958) to the venerable Federal Food, Drug and Cosmetic Act (1906), which bans carcinogenic products. ‘Technology only’ statutes are those which make reference to the best available technology (e.g. portions of the Clean Air Act, Clean Water Act, and Occupational Safety and Health Act). Even in such cases the practicability, the costs of implementation and the balance with other benefits have been used in practice, either by agencies for priority ranking or in adjudication by courts when not by the US Congress itself. Congress played a great role when the ban on saccharin by the Food and Drug Administration, in accordance with the Delaney clause, was rescinded. This debate was given impor­ tant publicity, and it contributed to public understanding of the problem of balancing risks and benefits. In any case, most of the regulations (75% of them according to Ref. [17]) are of the ‘balancing’ type, in that they require that advantages and disad­ vantages of risk reduction be balanced. Indeed, they cover a wide variety of pro­ visions, from the Atomic Energy Act to the Consumer Protection Safety Act. Incentives for using risk management approaches in establishing regulations are also BACKGROUND PAPERS 5 linked to executive orders. In 1981, an executive order from the Reagan Administra­ tion required quite a formal cost-benefit analysis of proposed new rules. The above considerations emphasize the fact that requirements and incentives for risk management policies are not specific to radiation risk. It is a matter of debate to decide upon the need for integrating approaches. There is a good consensus that it would be desirable to have a common framework, and this is also a direct conse­ quence of the ALARA philosophy; if money is to be spent for the most effcient pro­ tection measures, unified protection policies are a necessity. However, those views are usually considered as unrealistic, and risk management is still ‘sectored’.

2.2. Basic steps in defining risk management policies

In spite of the above quoted divisions, a common scheme applies to radiation risk and other risks, to releases and accidental situations, and to industrial, environ­ mental or consumer risk. On demand of the Environmental Protection Agency, and for general purposes, the US National Academy of Sciences has identified the main steps (see Fig. 1). Approaches specific to radiation risk are very similar (e.g. Ref. [20]). A first series of steps encompasses what can be called ‘risk assessment’ . The basic approaches are usually quite close for all substances that are suspected to be carcinogenic. In particular, a common hypothesis is the absence of threshold in dose. The comparatively important amount of knowledge accumulated about ioniz­ ing radiation should not be an excuse for overlooking the early steps. First, the theoretical knowledge must be regularly revisited; second, in the identification of exposures, completeness should never be taken for granted. For example, it was only in the 1980s that radon was recognized as the primary source of radiation exposure in the general population. The process may stop there, with the definition of a reference dose for toxic products or a virtual safe dose for carcinogenic agents. In risk management, further steps are required (see the right hand part of Fig. 1), in order to balance risks and benefits for alternative strategies or technical solutions: evaluation of detrimental impacts, identification of alternatives, implementation of a decision aiding technique, choice, etc. Provisions for the assess­ ment of the actual degree of application and effectiveness of the proposed actions, as well as more general return from experience, were somewhat neglected in the first formalized risk management schemes. They do not appear in Fig. 1. They would be considered now as key points. In this scheme the participant responsible for deciding and implementing the risk management policy is not defined. It can be either a regulatory agency or a corporate manager. Indeed, the actual risk management system involves many participants, each of them being responsible for the risk policy at his level. In the previous description of the US regulatory framework, the need for administrations or agencies to apply decision aiding methods in defining their regulations was high- 6 BACKGROUND PAPERS

FIG. 1. Main steps in defining a risk management policy (from Ref. [19]).

lighted. In ICRP-60, the stress is rather put on the actions of the operators. But in both cases, optimization or other techniques for risk management are universal requirements. It is not possible to discuss risk management further without admitting the fact that it has a double meaning. First, risk management can be understood as a deci­ sional process for achieving an optimum risk level by applying rational decision aid­ ing techniques, among which cost-benefit analysis (CBA) is the most widely used. Second, risk management is an operational system in which various participants have responsibilities, which can be called the risk management system. The diffculty is not to find powerful risk management theories; it lies in the search for the fit between such theories and the actual risk management system. Coming back to the history of risk management for radiation risk, ICRP-26 was not the first recommendation stating that the use of CBA would allow a better allocation of resources for protec­ tion, but it marked a real innovative step in its attempt to embody risk management theory in the actual system. Many of the developments of ICRP-60 can be viewed as improvements in this direction. BACKGROUND PAPERS 7

2.3. Methods for decision making

Decision makers must respect some forms of limits, based for example on acceptability considerations, and they must, within this framework, apply a rational method for balancing advantages and disadvantages of further risk reduction. ‘Optimization’ has become a generic term in radiation protection, although it refers to CBA, which is only one of the decision aiding categories of methods. The basic idea of CBA, which can be tracked back to the middle of the last century, is to apply an ‘efficiency criterion’ in the selection of the best option. The efficiency is merely based on comparison of the total costs and benefits, aggregated at the level of society. A generalized application of CBA in radiation risk would result in a maxi­ mum level of protection for a given amount of expense by ensuring that the money is spent on the most efficient options. For practical applications in the field of radia­ tion risk it is almost the unique approach. However, a category of methods grouped together under the name of ‘decision theory’ must be mentioned. In such approaches, the decision derives from the theory of games [21]. Under a certain number of assumptions, a decision maker is able to set up a ‘utility function’ for the valuation of the consequences of various alterna­ tives. This theory is designed for coping with uncertainty, and the best decision is the one that maximizes the expectation of utility [22]. A classical application has been made for the problem of siting nuclear facilities [23]. Such approaches do not require monetary valuation of risks, and they are well fitted for choices with a great uncertainty or for dealing with probabilistic events [24]. An important weakness is that decision relies on the preferences of an individual decision maker. Therefore, different decision makers can arrive at different solutions, which has prevented use of this method for radiation risk, a field in which consistency is highly desirable. However, it may become more interesting as the need to deal with uncertain events increases. One can think here of the question of ‘potential risks’ as raised by ICRP-60 or the IAEA, for which CBA is quite diffcult to implement, in particular when events that are both rare and uncertain are considered [9, 13, 25-29]. There is a third category of approaches, known as ‘social choice theory’, where the goal is to define a rational synthesis of the preferences of all the individuals who can be affected by the decisions. Developments in this direction are quite academic, but it is necessary to quote such approaches, because the problem they address cannot be forgotten. Classical optimization or more sophisticated CBA or decision analyses fail to take into account the views of people who are actually or potentially exposed to ionizing radiation. Because of this weakness, the implementa­ tion of risk management has never been fully accepted by some groups of people. CBA, sometimes called optimization or ALARA analysis, implies that the global costs and benefits associated with a given option are quantified. There are obviously some inherent limitations. First, assessment of impacts must be feasible, the difficulty and uncertainty of which are a well known problem, as well as assess- 8 BACKGROUND PAPERS

TABLE I. LIMITATIONS OF COST-BENEFIT THEORIES AND POSSIBLE RESPONSES FOR RADIATION RISK

Limitations Responses

Value of impacts for non-market commodities Cost of man-sievert Multi-dimensionality of impacts is not Guidance on long term effects ... considered Guidance on ‘accidental risks’ (‘potential risks’) Distribution of costs and benefits is not Definition of responsible participants considered Design of a compensation system Rules for constraining inequities

ment of the costs, which may also be quite difficult to do in complex systems. Second, a monetary value should be assigned to the impacts. The most usual impact being a probability of cancer death, no market price can be used and the possible methods do not give fully consistent results. In the field of radiation risk, impacts can be expressed as man-sievert, i.e. as collective dose, and comparisons can be made on this basis. However, the level of individual dose may be of importance so that individual doses should be valuated according to their levels, and there might be a need to aggregate other factors, leading to the need to weight long term versus short term impact, local versus worldwide impacts, certain exposures versus accidental exposures, or social impacts (e.g. the fact of being resettled) versus health impacts. Valuation systems or multi-attribute approaches are often used, but there is a need for ensuring consistency between analyses. Third, the efficiency criterion is a global criterion. It does not take into account the distribution of costs and benefits. Economists call it a potential Pareto solution, because it improves the over­ all wealth of the society but some individuals may suffer a net loss. A compensation system must be set up. In other words, an ALARA solution may be unfair, and it may infringe upon equity principles. Such limitations to cost-benefit approaches are generic, and they were identified a long time ago. Responses, or attempts at responses, have been established in the field of radiation risk (see Table I). Table I is not exhaustive, but it exemplifies how some of the problems that are dealt with in radiation risk recommendations are actually rooted in the generic weak­ nesses of optimization approaches. The system of ‘constraints’ is recommended by ICRP-60. The principle is to assign boundaries for individual doses in performing an optimization, so as to avoid that relatively high figures result from the optimiza­ tion process. The use of ‘constraints’ on individual doses has many rationales. Here it may be seen as a surrogate to the compensation system that economists recommend BACKGROUND PAPERS 9 to limit inequities when it is possible to assign a clear monetary value to losses. In other areas (e.g. water pollution and sometimes air pollution), a compensation sys­ tem has been set up by assigning monetary values to pollutants and by organizing a market for ‘rights to pollute’ . It must be recognized that these problems were not an obstacle to optimization in the past 15 years. In particular, the economic valuation of the impact was not necessary in many instances where the problem was only to reduce doses at the best cost and not to find an absolute ‘optimum’. However, the above quoted issues arise as optimization practices develop. It is remarkable that the issues raised by the limitations of cost-benefit theory are among the most important issues debated now when dealing with radiation risk.

2.4. Participants and responsibilities

The risk management system encompasses the participants in charge of protec­ tion at various levels, the procedures that are applied, and the related organization. Risk management principles may be explicitly defined or may not exist, but the risk management system is always present, whatever opinion one may have about its effi­ ciency. As said above, the risk management system must be built up in a consistent way, should one wish to apply principles such as optimization. In the past 10 years, there has been an adaptation of management practices among the operators, and fac­ tors such as the importance of work management have been highlighted [30]. As a corollary, the risk management principles should be different when the actual system is different; the principles that were applied in the nuclear industry cannot be applied in the same way to other industries [31]. The participants are numerous. Considering only those who have a direct responsibility in the system, excluding those in charge of expertise, inspection or emergency response, a quite hierarchical system appears. In this system every responsible person and body defines rules and constraints for the lower level. At the top of this structure are international organizations which elaborate recommenda­ tions (e.g. ICRP, IAEA). National authorities are invited to implement these recom­ mendations, and, doing so, they elaborate more precise prescriptions and adapt the basic principles to national organizations and national laws and regulations. Opera­ tors must comply with such rules, but there are several levels of decisions which must be distinguished. Corporate managers must organize the management of their companies and possibly set objectives related to protection and safety. They also have an important responsibility when setting budget priorities and controlling expenditures and when allocating manpower to safety and protection tasks. In most countries, it is however the plant operator who bears the legal responsibility in case of incident. He also implements protection. According to their roles, contractors have a variable influence on the organization of work and safety, but there are always possibilities for actions and they must define their protection policies. Important con­ tractors may be responsible for the design of specific protections for an operation 10 BACKGROUND PAPERS or for work planning, while very small contractors have the same means of action as individual workers. Finally, the individual worker has some control over his own dose. The main roles and responsibilities of these various participants are shown in Table II.

TABLE II. MAIN PARTICIPANTS IN RISK MANAGEMENT SYSTEMS AND THE ROLES THEY CAN HAVE

Technical Modifying Participants constraints and Tools and expertise behaviours restrictions

International Fundamental limits Promotion of Standardization of organizations optimization units Standardization of assessment tools (e.g. dose response) National Derived limits Responsibilities, Research and authorities Standards liability and expertise body Compulsory compensation Dissemination of qualification of Reference values of information operators human life National training Testing and programme inspection Provisions for emergency response Corporate Corporate constraint Strategic goals in ALARA structure managers Quality and safety protection and ‘Intelligent’ criteria reference values dosimetry Compulsory use of Call for tender Return from ‘approved’ material policy experience and contractors Training policy ALARA tools Design criteria Weight of safety Training centres and protection in management Plant Implementation of Training practices Analytic dosimetry operators protection shielding Weight of protection Contractors Dose credits per in usual decisions Workers operation Internal procedures Work management preparedness BACKGROUND PAPERS 11

Basically, a participant can follow three approaches for diminishing the risk. He can act on the technical system (e.g. limits, protections, safety measures, redun­ dancy, alarms); this is the classical approach to protection. He can also modify the behaviour of other participants, very often those who depend on him in the hierar­ chy. In particular, a national authority must assign responsibilities, either in specific regulations for radiation risk or in general regulations and laws. Legal responsibility and liability are quite often defined by law. Corporate managers must also define responsibilities as regards protection and safety, obviously within the framework of national rules. Responsibility for and possibly achievements in protection are also a matter for specific clauses in calls for tenders and contracts with contractors. Together with the definition of responsibilities, the involvement of national authori­ ties and high level management in the development of a safety culture is another way to modify behaviour. Last, monetary incentives and guidelines do also strongly influence the behaviour of participants. Enforcement of liability rules and financial penalties, or stopping of operations in case of anomaly, is a powerful incentive, but assigning guidelines for the cost of man-sievert is also an efficient approach. The third approach is to elaborate assessment tools and expertise. International organiza­ tions play a modest role in defining responsibilities, but they are major contributors to the definition of radiological units by the International Commission on Radiation Units and Measurements (ICRU). The ICRP also has a fundamental influence in proposing a series of assessment tools, validated by the international scientific com-

FIG. 2. Achievement of ALARA for steam generator replacement (from Ref. [36]). 12 BACKGROUND PAPERS munity. The dose-response relationship is the most obvious, but the models for com­ puting internal doses after intakes are also essential. International and national agreements on tools and assessment are necessary. Almost all participants have to decide on protection options (see Table II), so they must optimize whatever choice they make. In doing so, they must act within the framework imposed on them by higher level rules. Therefore, risk management can be seen as a cascade of nested optimizations, the higher level defining constraints (in the general sense) for the lower one. It is necessary to give more information on the way risk management principles and systems are set up in the field of radiation risk. However, it is interesting to quote some data to show that the above provisions proved to be efficient. The dose limit for workers was 50 mSv/a, but only a handful of workers received such a dose in countries with a nuclear industry, and most countries see such cases as incidents. The average dose is rather 2 mSv in the nuclear industry (e.g. Refs [32-34]). As regards public exposures, the authorizations for effluent discharges always target doses lower than the limit to the public [35]. Doses of tens of microsieverts are usually estimated in a conservative manner to a reference group. Within plants, optimization has been applied to many operations. A well known achievement is steam generator replacement, where doses have been cut by a factor of almost 10 [36] (see Fig. 2).

3. DEFINING RISK MANAGEMENT PRINCIPLES FOR RADIATION RISK

3.1. The ICRP three basic principles

In 1991, the ICRP proposed an elaborate definition of the three principles that were proposed in its previous recommendations. The three principles can be summed up as follows: — Justification of a practice: No practice should be undertaken unless benefits offset the radiation detriments. — Optimization of protection: Exposures should be kept as low as reasonably achievable, economic and social factors being taken into account. — Individual doses and risk limits: Individual doses should be kept below some limits (for risk, no limit is proposed in the recommendations). The precise definitions contain more elements (see Ref. [4], Section 112). In particular, two new concepts are introduced in these principles, whose practical implementation still requires elaboration: — Potential exposures (exposures due to accidents or incidents in which there is ‘a potential for exposure but no certainty’): From a management point of view, BACKGROUND PAPERS 13

such exposures must not be confused with those that really occurred after an accident (intervention situation). They are associated with a priori assessment. When the magnitude of such exposures is such that deterministic effects such as death may occur, it is recommended to use the concept of risk rather than dose. — Constraints: In addition to limits, the implementation of optimization should be such that ‘constraints’ on individual dose, or risk, are not neglected. It has been shown before why such provisions are desirable when applying cost-benefit analysis, because of limitations of the method. The three principles mentioned above can be considered as the basis for radia­ tion risk management, and all international recommendations (e.g. Ref. [25]) refer to them. However, this broad consensus does not mean that all the problems are solved — far from it. First, some elements still need to be clarified (e.g. the above quoted new concepts). At present, different organizations and individuals derive different interpretations. Second, the whole set of principles applies to ‘practices’, i.e. human activities which increase the exposures or the number of individuals exposed. The ICRP states that the third principle (limits) must not be applied to inter­ vention, neither to post-accident intervention nor to intervention against natural exposures such as to radon. Rather than limits, procedures and optimization schemes to choose the level of intervention are proposed [37, 38]. Return from experience shows that these principles were efficient for the management of exposures associated with practices during normal operations. An important characteristic of this system is that two rationales apply together and are interfaced in risk management. First, individual dose limits are derived from consideration of acceptability; more precisely, they are defined as being just below ‘unacceptable’. This is the application of a ‘health only’ rationale which does not involve economic consideration or technical feasibility. Second, exposures should be reduced further, and risks and benefits must be ‘balanced’. Note that for interven­ tion, the system is conceptually simple because ‘intervention limits’ are all based on the balance of risks and benefits.

3.2. Acceptability and optimization

The acceptability of risk has been a matter of social debate for the last 20 years, and research in the field of risk perception has accompanied the development of this debate. However, in the meanwhile, acceptability concepts have become a practical tool in risk management. In the 1970s it was admitted that risk management policies, or, as it was said at this time, safety policies, could not realistically aim at a ‘zero risk’ level. Instead, it was admitted that they should target an ‘acceptable level of risk’. 14 BACKGROUND PAPERS

This approach has been used by the ICRP since 1977 [1] in setting up limits for the exposure of workers and the public to ionizing radiation. After it was made sure that dose limits would not be associated with deterministic effects, the search for acceptability was based on the assumption that this carcinogenic risk did not admit a threshold, and an assessment tool was provided, i.e. a dose-response rela­ tionship that allowed association of a given level of dose with the corresponding level of risk to exposed individuals. This approach has been applied to other carcinogens, based on similar hypotheses. Concerning the risk of catastrophic events at hazardous installations, a similar evolution has taken place and it is now generally admitted that such an accident is possible, if highly improbable. Once again, the management of risk must target an acceptable level for the probability or the consequences of accidents, or even both. This approach to risk management, in which ‘acceptability’ plays a great role, is described below. The ambiguity of the risk concept, whose meaning wavers between the proba­ bility and the consequence or even the nature of a hazard, need not be emphasized. However, the phrasing of questions about risk is usually clear enough. The concept of acceptability is subject to more developments. Tolerability, negligible risk, de minimis level, and ALARA level are some of the notions that have been introduced in order to discuss the question of acceptability. The description made by the UK Health and Safety Executive is very representative of the use of acceptability con­ cepts in risk management (Fig. 3) [39]. In spite of differences in vocabulary, other countries advocate a similar approach (e.g. Switzerland [12] and the Netherlands [11]), as does the ICRP. The common features are: — the distinction among three areas for the level of risk (not tolerable; acceptable under certain conditions, usually under the condition that further risk reduction would be too expensive for its achievement; not worthwhile to deal with); BACKGROUND PAPERS 15

— the association of practical procedures with the above levels (e.g. regulatory prescriptions, optimization processes, and exclusion from the domain of appli­ cation of regulations).

Therefore, the two rationales of acceptability and optimization are not conflict­ ing, but they are embodied in the risk management process. Indeed, the risk manage­ ment approach is characterized not only by the use of various levels of risk but also by the link between these levels and associated procedures. This scheme remains quite theoretical, unless two series of tools are available. First, tools for the establishment of risk criteria must be available. In corporate risk management, such criteria can be quite numerous (e.g. loss of produc­ tion, loss of credibility, material damages). For the authorities, human losses are very often the decisive criterion. These criteria can be used to define limits (e.g. dose limits, risk limits, safety goals) or may be the variables that enter into the optimiza­ tion process (e.g. expected number of deaths), or both. For example, in classical radiation protection situations, the collective dose is optimized and the individual dose is kept below the limit. The individual risk (e.g. probability of dying when living for 1 year in the vicinity of a hazardous facility, or when working 1 year in given conditions) cor­ responds to a first series of criteria. As regards radiation risk, the classical criterion is the individual effective dose, which is linearly related to the probability of radia­ tion detriment. The societal risk, sometimes called collective risk or group risk, refers to another series of criteria measuring the impact on society from an activity. Criteria are numerous; especially when dealing with accident situations, one may wish to consider the probability of a severe accident or the expected number of deaths from potential accidents or any combination of probabilities and consequences of major accidents. For instance, car driving corresponds to an individual risk of dying of 2 x 10“4 per year for the average French individual, and to a societal risk of 10 000 deaths. This example illustrates the balance between individual and collective risk in a simple accident situation. In radiation risk the collective dose is the classical criterion for societal risk. In spite of many limitations, it is widely applied in optimization. Second, risk assessment tools must be available that allow assessment of the level of risk and also establishment of a clear link between the achievement of a given performance for the risk level and the implementation of a given safety or protection option. In the example above mentioned of traffic accidents, body count easily allows risk assessment, without any discussion about uncertainty, but it does not allow risk management, because there is no modelling for the effectiveness of road safety mea­ sures. As regards radiation risk, the dose-response relationship allows linking of dose with a risk criterion (e.g. lifetime risk of dying from a cancer due to the exposure), and computation of a shielding may associate design with this dose. The risk criterion can therefore be related to an operational control. In case of major 16 BACKGROUND PAPERS

hazards, probabilistic safety assessment allows the derivation of risk estimates, but it is often quite difficult to go back from the satisfaction of a given criterion (e.g. number of deaths in an accident whose probability is 10~6) to the practical measures used (e.g. steel thickness of a containment vessel). So the very same theoretical approaches fit to the management of normal oper­ ation and severe accident risks. But with respect to the availability of management tools, this symmetry does not hold. Both on the criteria and on the assessment tools for severe accidents there is still a lack of agreement, so that many people do not think it possible to relate compliance with criteria back to the actual safety measures used.

3.3. Issues related to individual risk

Dose, risk and detriment

Effective dose is a usual criterion in dealing with individual risk, and primary limits are expressed as effective dose. It must be made clear that the effective dose is not a simple measure of exposure, but that it is a real criterion for risk manage­ ment. Indeed, effective dose is a measure of risk (for a fatal cancer the probability of attributable death is proportional to the dose), and, even more, it is a valuation of risk. The effective dose is based on the concept of radiological detriment. The var­ ious components of radiological detriment — probability of cancer death, probability of suffering a non-fatal cancer, probability of diseases in offspring due to genetic effects — have been compared, and weighting factors have been applied in defining a ‘detriment’ . When using effective dose, one is also using this valuation system. For example, non-fatal cancers are weighted according to the lethality fraction, and even for fatal cancers the relative loss of life expectancy is taken into consideration. The average loss of life expectancy being 15 years for all fatal cancers, a weight of 2 is applied to cancers for which the loss is 30 years. Therefore, it must be kept in mind that the effective dose is not a neutral criterion and that value judgements are em­ bodied in the ‘dose’.

Choice of a limit

Occupational exposure to carcinogens is an area in which the acceptability approach was developed quite early. In 1977 the ICRP provided the assessment tool (the dose-response relationship for ionizing radiation), the criterion (probability of death due to 1 year of exposure, computed on the basis of a lifetime risk estimation of cancer death) and the acceptable level (about 6 x 10"4). This approach has been applied to other carcinogens and also in other circumstances. An overall limit for cumulative hazard to workers has been set up in the Netherlands [11]. In that same country and in Switzerland [12], an individual limit for people living near hazardous installations has been proposed. BACKGROUND PAPERS 17

TABLE III. GOALS FOR INDIVIDUAL RISK IN OFFICIAL DOCUMENTS (PROBABILITY OF DEATH FROM 1 YEAR OF ACTIVITY)3

Workers Limit (yearly risk) Public

Intolerable: HSE and CIPR =» 1/1000 1/10 000 Intolerable (HSE) 5/100 000 <= Intolerable (ICRP) 1/100 000 Intolerable when de facto situation (Netherlands) 1/1 000 000 Broadly acceptable (HSE); intolerable (Netherlands) 1/10 000 000 Broadly acceptable when de facto situation (Netherlands) 1/100 000 000 Broadly acceptable (Netherlands) Food additives (USA)

a From Refs [4, 11, 13, 40].

Numerical levels associated with acceptability are more seldom put forward. Implicit analyses can be made. For example, looking at the significance of threshold limit values (TLVs) for occupational exposure to carcinogens (radiation, acryloni- trile, asbestos, benzene) and given the availability of quantified dose-response rela­ tionships, the target levels of risk were found to range in 1988 between lO-3 and 10~5 [31]. The figure of 10‘4 is a reference because it corresponds to the occupa­ tional fatality rate in a country such as France. When more goals are contemplated, the range of figures becomes very broad (see Table III). One must distinguish goals applying to workers and goals applying to the public, intolerable and ‘broadly acceptable’ cases, new facilities, and de facto situations. Also, the cumulative risk from all hazards can be addressed (10~5 for a member of the public in the Netherlands) or the risk from one source (10-6 in the same country). In addition, some goals are straightforwardly used to derive practical limits (e.g. TLVs for ionizing radiation) whereas others are instead ‘desirable targets’. For example, the objective of 10“6 for a whole life of exposure used by many US administrations does not result in the banning of all activities leading to such a level of risk. It can only be said that usually no action is taken below this level, 18 BACKGROUND PAPERS but substances leading to much higher risks (up to 10 3) may also not be regulated [40]. For radiation risk, the limit is defined as a dose, but it is discussed on the basis of the associated mortality. It is associated with an unacceptable level, and it is strictly implemented. Besides, optimization has kept average actual doses at a much lower level. All these elements must be considered when comparing various levels (see Table III). The probability of deaths is most often the unique criterion for deciding on acceptability, but other criteria are sometimes looked at. The ICRP in its last recom­ mendations proposes a multi-attribute approach in choosing the limits. It compares the impact of choosing different annual limits on the various attributes (see Table IV ). This remains a quite specific approach because it was not applied to other carcinogens, and because actual decision making does not make use of all these attributes. For example, although it is quite a sensible criterion for addressing individual risk, and in spite of the availability of quantified estimates, the loss of life expectancy is not really used for discussing occupational limits of exposure outside the radiation field; its use by the ICRP is quite modest, and its use in optimization by national authorities or the operators is almost nil.

TABLE IV. USE OF A MULTI-ATTRIBUTE APPROACH IN DEFINING AN OCCUPATIONAL LIMIT OF EXPOSURE: ATTRIBUTES OR DETRIMENT DUE TO EXPOSURE OF THE WORKING POPULATION3

Annual effective dose (mSv) 10 20 30 50 50 (1977 data) Approximate lifetime dose (Sv) 0.5 1.0 1.4 2.4 2.4

Probability of attributable death (%) 1.8 3.6 5.3 8.6 2.9

Weighted contribution from non-fatal 0.4 0.7 1.1 1.7 — cancer (%) Weighted contribution from hereditary 0.4 0.7 1.1 1.7 1.2 effects (%) Aggregated detriment (%) 2.5 5 7.5 12 — Time lost due to an attributable death 13 13 13 13 10-15 given that it occurs (years) Mean loss of life expectancy at age 0.2 0.5 0.7 1.1 0.3-0.5 18 years (years)

a From Ref. [4]. BACKGROUND PAPERS 19

Looking at the references to which criteria are compared, the picture is quite clear as regards occupational risk. There is some consistency between choices in radiation risk and other sectors, and the subject has been addressed in many instances. However, the rationale for all these quantitative goals is not always made explicit. Comparisons with other activities, or with risks considered as ‘acceptable’ or with natural risks, are common. The reference to occupational fatality rates is a classical approach (10~4 on average, 10~3 in some ‘risky’ activities). Other figures can serve as references, such as the ‘natural’ mortality rate in countries with a good health record. A criterion can be the proportional increase of the rate at given ages (given for Sweden in Fig. 4). The minimum (1 to 5 year old girls) is about 10“5.

FIG. 4. Change in the age and sex specific mortality rate for all causes (with reference to the Swedish population in 1986) after an exposure of 50 mSv per year from Ref. [4]). The change is shown for the additive and. relative projection model. 2 0 BACKGROUND PAPERS

Considering risk to the public, there is much less available material for discus­ sion of the bases for the choices that have been made. General comparisons with risks due to other exposures in normal life have been attempted. But such compari­ sons are much less convincing than comparisons of occupational risks, because the sources of risk are really not comparable. Very often the limit to the public is defined as a fraction of the limit to workers.

Negligible levels

Negligible levels are those below which it is not worthwhile to manage the risk. In the radiation field there was not much use of this concept before it was intended to define exemption levels. Exemption levels are levels of activity content or volumic activity below which there is no need to apply the system of protection. The Basic Safety Standards propose the use of exemption levels and provide quantitative figures for radionuclides [25], and the European Union intends to do so. These figures were based on a target of 10 /xSv per year and per individual [41], and they are derived from the use of various exposure scenarios [42]. So exemption is based on an individual risk criterion. It is about of 5 X 10“7 per year. This figure can be com­ pared to the 10“6 per lifetime below which US regulators estimate there is no need to regulate. It must be said that some consideration should be given to societal risk by considering the collective dose, and ICRP-60 stresses this point. However, this last consideration does not seem to haved played a great role so far.

Dose and risk constraints

The issue of dose and risk constraints has led to major discussions and analyses, and it would be ambitious to make an exhaustive synthesis. ICRP-60 has proposed the use of “ restrictions on the doses to individuals (dose constraint), or the risk to individuals ... (risk constraint)” so as to limit the inequity that can be associated with optimization. This applies to both public and occupational exposures, and is generally felt to be a sensible improvement (see Section 2.3), but interpreta­ tions may be divergent. Several organizations (e.g. the IAEA [43]) and working groups (e.g. OECD/NEA Group of Experts on Dose Constraint) are dealing with this question and there is some literature on the subject (e.g. Refs [44-47]). For the ICRP, constraints bear on an individual risk criterion and are applicable within the process of optimization. Technically speaking, the optimization process is already constrained by the need to comply with the individual dose limit, but it was felt useful to add a lower ‘constraint’ because the limit of exposure is just on the borderline of ‘unacceptable’ . Since constraints must be applied within the framework of the optimization process, a first series of questions addresses the field of application of optimization. In the view of some people, any responsible person, at whatever level, whatever the BACKGROUND PAPERS 21 stage of development (design, operation, etc.), must optimize his decisions. He has therefore to respect constraints. The responsible person can be an authority when defining levels for authorized discharges. He can be a subcontractor when implementing training programmes or allocating the workforce to different tasks. He can also be the operator of a plant when designing an operation. This view is consis­ tent with a global view of risk management such as in the USA wherever agencies must ‘optimize’ their policies. Other people focus on the last point, i.e. the require­ ment for an operator to perform optimization. Somehow the ICRP paved the way for this discussion, stating that constraints are to be applied “ In relation to any partic­ ular source within a practice” , but it also says that this must apply to a “ class of occupation” which “ should be defined in fairly broad terms” . Actually some of the diffculties in enforcing the concept of constraint arise from the fact that optimization is still a recent practice and the various levels of appli­ cation and the respective roles of participants are not fully clarified. Among the issues raised is that constraints may apply to the annual individual dose or to the individual dose associated with a given operation. It is necessary to identify who defines the constraints (national or local authorities, or operators), on which bases (the reference to ‘good practices’ makes for consensus, but this requires qualification of homogeneous practices), and how they are enforced. Because con­ straints are used prospectively at the design or planning stage — the stage at which optimization is performed — it is diffcult to derive conclusions from the actual doses after the task is performed. Unless constraints are purposely violated at the design stage, it is impossible to say that the operator did not comply with the regulation. Finally, a great number of restrictions on individual doses already apply. One can quote the ‘dose credit’ that an operator establishes for an operation, ‘dose tar­ gets’ that corporate managers can set up, ‘investigation levels’ that the authority may establish, the ‘reference levels’ or ‘guidance levels’ that are proposed for diagnostic examinations, and the dose limitations associated with discharge limits from nuclear installations [35, 43, 48, 49]. So there is a need to establish the links between such quantities and the ‘constraints’ . Whatever the eventual characteristics of a system for the application of con­ straints, this point has raised considerable interest and all the work that is going on will lead to a much better understanding of the actual practices of optimization.

Individual risk and potential exposures

The question of potential exposures also raised important discussions and it has been addressed in many documents [25, 26, 29]. Looking at the particular aspect of individual risk, there is less controversy. The criterion is the mathematical expecta­ tion of the detriment. For workers who can receive doses in small incidents, doses can be above the annual limit but still remain in the range of stochastic effects. Thus the mathematical expectation of the dose can be used, and there are not many objec­ 22 BACKGROUND PAPERS tions to adding it to the annual dose (possibly after discarding the dose and dose rate effect factor). For higher exposures, there is a probability of immediate death associated with the dose. This can be multiplied by the probability for a given worker to be the victim of an accident, and the resulting quantity can be compared with a risk limit. Two problems have been noted. It would be necessary to define the detri­ ment for non-lethal deterministic effects (e.g. cataract, permanent sterility, amputa­ tion). Also, a probability of 10~3 per year of dying in a radiation accident, although consistent with the dose limit, is quite unacceptable to workers and managers. Possi­ bly the fact that immediate deaths are at issue diminishes acceptability. Computation of this risk for workers is felt to be a possible task. In many cases, simple accident sequences contribute to the risk. For the public, the question is far more complex because exposures are usually the consequence of a very severe acci­ dent following a safety failure at a nuclear installation. Assessment requires the implementation of a full probabilistic safety analysis of the installation.

Optimization of individual doses and intervention levels

In all the previous issues, the individual risk (actual risk or dose) was not the variable which is subject to optimization. It was a rather a constraint or a limit. However, the ICRP stated that individual doses should be optimized as such. Within the framework of practices, there are no real examples of such an optimization. However, individual doses are sometimes taken into consideration. In some instances the operator gives some weight to the individual doses, the usual approach being to optimize a variable which mixes individual and collective dose. A utility function for the radiological impact is set up that is a weighted collective dose; the higher the individual dose, the more it weights [50]. Incidentally this weighting system allows reduction of inequities in cost-benefit analysis and it is an alternative to the use of constraints. Individual doses must be the result of optimization in another circumstance. When the situation is not a practice but an intervention, the ICRP recommends that ‘action levels’ or ‘intervention levels’ should be the result of optimization. In such cases the decision is to be taken by the national authorities who are thus in charge of optimization.

3.4. Issues related to societal risk

Collective dose

Collective dose is a widely used criterion for the societal impact of exposures. The collective dose is quite seldom used as a limit; it is rather used within the optimi­ zation process as the criterion for the valuation of the impact. It still raises some problems. The monetary value of man-sievert is an example. Corporate management BACKGROUND PAPERS 23 may wish to choose a different value than the national average, or countries in which a market economy is not well established may have difficulties in defining a reference value, as is the case when discussing remedial actions after the Chernobyl accident. A second example is the need to look at the heterogeneity of individual exposures. This point was mentioned above. Finally, when integrated on a long period of time, as must be the case with waste, or on a large area, collective dose is sometimes felt to be an inadequate criterion.

Safety goals and potential risk

Regarding severe accidents, it is generally felt that societal risk criteria become important. Even criteria very similar to individual risk criteria are actually societal risk criteria. For example, the ‘probability that an accident causes a dose above 150 mSv, or one death’ has no significance except at the group level. Criteria, assessment tools and management tools are still being discussed. In the field of radiation risk, two approaches to management of such situations are now mixed. One is the approach of probabilistic safety goals or safety criteria (e.g. Refs [9, 27, 28, 51]), in which safety is likely to be designed to directly meet criteria thanks to probabilistic safety assessment. The other is derived from radiation protec­ tion [26] and should be extended to safety within the framework of potential exposures. Application of the latter concept was proposed in the Basic Safety Standards [25]. Once again it is still impossible to synthesize all discussions here; only some general issues in risk management are exemplified. No explicit goals for the criteria have been formulated in the case of a widespread population risk associated with routine exposures, but for major acci­ dents a great deal of work has been devoted to the question. It is generally admitted that the mathematical expectation is not a very satisfactory criterion and that the probability and the consequences of accidents must be handled together. A represen­ tation of the risk called ‘risk profile’ or ‘Farmer curve’ or ‘CCDF’ (complementary cumulative distribution function) was popularized in the 1975 ‘Rassmussen Report’ on the probabilistic safety assessment of nuclear power plants [52] and it has been applied to many other cases (Fig. 5). In such an approach, the probabilities and consequences of all accident scenarios are computed, and ranked in such a way that the curve displays the proba­ bility that a given number of victims will die in an accident, on a yearly basis. This kind of display is not quite intuitive, but it is now of general use, and acceptability has been discussed within this framework in many instances (the IAEA [9] for nuclear industry, the UK [8], the Netherlands [11] and Switzerland [12] for general purposes). This having been done, it is possible to define an ‘intolerable risk’ area, a ‘negligible risk’ area, and an area in which optimization must be implemented, so that the general principles for managing risk can apply (see Fig. 6). 2 4 BACKGROUND PAPERS

FIG. 5. Representation of risk in a probability consequence mapping (from Ref. [53]). ‘DRS’ is the Deutsche Risikstudie.

This approach requires even more value judgements than for individual risk. A problem is associated with the fact that a greater weight is commonly put on ‘cata­ strophic accidents’ than on more usual accidents. The Dutch administration is the most explicit on this subject [11]. It defines acceptability criteria for accidents with a potential for 10 deaths (a probability of 10'5 for intolerable and 10~7 for negligi­ ble) and a power law for the ‘aversion to catastrophes’. The power 2 is proposed; i.e. should the consequences be n times more important, the probability must be n2 lower. There is a reluctance to give an important role to this kind of safety objective, because the assessment process is not yet fully mastered. Because of the lack of BACKGROUND PAPERS 25

g 1 .OOE-03 r >. g 1.00E-04 ■ ç § 1.00E-05 ■ 8 ~ 1.00E-06 ■ фС ©® 1.00Е-07 ■ £ $ 1.00Е-08 ■ £ S<0 1.00Е-09 ■ о £ 1.00Е-10 ■ _i______i______i______i 1 10 100 1000 10 000 Number of deaths in an accident

FIG. 6. Acceptability scheme of the Netherlands. knowledge, some degree of conservatism is introduced when demonstrating compli­ ance with a given target. As a consequence, it is difficult to choose the ‘optimal’ option because this degree of conservatism is essentially unknown. It is also diffcult to go back from the ‘probabilistic safety goals’ to practical measures and to an opera­ tional control system. Nevertheless, this approach is felt useful because it helps in clarifying the safety policies that are followed and because it promotes a better knowledge of the installations, failure modes and consequences of accidents.

Choice o f criteria

It has been noted that traditional risk indices, such as the expected number of deaths, are not adequate for decision making on major hazards. Two issues are associated with these questions. First, mortality indicators do not give due account of the variety of consequences of accidents: reversible and nonreversible health effects, soil or water pollution and other environmental impacts, financial costs, etc. There is a need to better characterize the possible impacts and criteria [54]. Second, for major hazards, the results of numerous probabilistic safety analyses (e.g. Canvey Island and other UK studies [8], the Rijnmond area [55], pressurized water reactors [7], and hazardous material transportation [56]) have shown that the expected num­ ber of deaths is often very low, and generally out of range with the magnitude of the problems. It has been mentioned that some institutions have concluded that poten­ tial accidents which may result in a great number of human casualties should be given an ‘extra weight’ in order to take due account of their ‘catastrophic aspect’ , but few analyses have allowed the grounding of such systems. However, surveys have shown that this factor is important. 2 6 BACKGROUND PAPERS

Elicitation of decision makers’ preferences, through social science methodolo­ gies using interviews and questionnaires, has already been applied to controversial issues implying strong value judgements (e.g. siting energy facilities [23]). A survey aiming at doing so in the perspective of introducing risk management in community planning, with a special concern for hazardous material transportation, was per­ formed among decision makers from the area of the city of Lyon [24]. The results cannot be considered as representative of every area, or of every problem, but it is interesting to show how the concepts put forward in the previous paragraphs can operate. The answers on ‘acceptable probabilities’ have been used to define the ‘aversion factor’ for the energy facility based on the number of expected deaths. The aversion appeared to be important. The power factor was close to 2 as in the Dutch approach [10], which is generally considered to be on the high side of ‘aversion to catastrophic events’ . Evaluation of risk at the societal level is also of importance for non-accidental risk. It has been mentioned that the collective dose should come into play when deal­ ing with exemption or clearance levels. In the definition of national policies on radon or on medical exposure, societal impacts are given an important weight. In those cases a great variety of criteria are envisaged, dealing with health, economic impact, acceptability and so on.

4. RISK MANAGEMENT SYSTEMS

4.1. Management by operators

Practices and radiological impacts

Activities involving actual or potential exposure to ionizing radiation can be split into three different branches: the nuclear industry, the health sector, and other activities which include industrial radiography and a growing sector, the industrial applications of accelerators. Obviously there are few common points among the operators in the various sectors. The nuclear industry is managed by big industrial companies, other industries can be very small and very new, and the health sector is not an industry. Besides, what can be called ‘safety and protection culture’ and risk management practices have seen an uneven development of the various activities. As regards occupational doses, they are shared among these activities (see Fig. 7). Within the European Union, the nuclear industry is the most important source of exposure. Attention should be paid to developing industrial activities, but they do not yet contribute significantly to the total exposure. As regards public doses, the potential exposures are mainly a matter of concern for the nuclear industry, although accidents with other sources must not be neglected. BACKGROUND PAPERS 27

Nuclear Fallout

Total industry

Occupational doses Public doses

FIG. 7. Occupational doses according to branch of activity (European Union [33]) and public doses according to source (world average [57]).

Looking at actual exposures, the greatest share comes from natural exposures whose management is not the responsibility of operators, but medical exposures are also important (about 1 mSv per person and per year, with sensible differences according to country).

Evolution of management approaches

New features and new concepts, in addition to the reduction in dose limits, challenge the actual radiation protection and global management of companies. The last recommendations of ICRP-60 put the stress on optimization and other opera­ tional procedures, and they also contain many features which are an incentive for a growing involvement of management in radiation protection. This move goes along with the general tendency in risk regulations. The ability of licensees and operators to cope with these new requirements has been largely discussed. However, in many instances, a strong regulatory background has not been necessary to incite operators and licensees to develop effcient radiation protection programmes. The companies’ ethics, the pressure of public opinion, and workforce concerns, together with the heavy dosimetric impact of some operations, have already resulted in significant efforts and measures. As a matter of fact, the management of radiation protection has experienced a significant change in the last decades. The actual implementation of ALARA approaches and the setting up of active radiation protection programmes at nuclear sites and for activities involving radiation exposure are typical examples of such evo­ lution. As quoted above, the evolution of the implementation of ALARA has resulted in putting forward the managerial aspects and especially the questions associated 28 BACKGROUND PAPERS with management of the workforce (e.g. see Ref. [58] or [30]). At present, what can be called ‘ALARA organization’, ‘radiation protection organization’ and ‘radiation risk management’ all refer to the same thing. In this development, one first step was implementation of ALARA approaches for specific operations that result in high doses. The example of steam generator replacement has already been mentioned, and similar well defined maintenance oper­ ations, such as shot peening, were the occasion of a successful implementation of ALARA [36]. As regards more complex operations (e.g. management of outages), and the global doses, the evolution is less satisfactory. The records of many compa­ nies (e.g. Cogéma and CEA [59], Sellafield [60]) also show global reductions in doses, both individual and collective. However, at the begining of the 1990s, it was rather questionable to assume that there is a general decreasing trend at the interna­ tional level. In particular, the total doses in nuclear power plants did not show any more significant reductions in many countries [61-64]. This resulted in the develop­ ment of new and more ambitious actions, and the involvement of corporate manage­ ment increased.

Organization of ALARA

The major characteristic of optimization is that the status and achievements of optimization or ALARA programmes in companies are evolving very rapidly. What was 15 years ago a matter of research or, at best, of case studies, what was 5 years ago applied only to pilot projects or to very sensitive operations, is now, in some companies, part of the standard procedure for radiation protection management. A common pattern in ALARA approaches in the various companies and indus­ trial sectors can be delineated. Infrastructures, tools, reports, procedures, and refer­ ence figures have been set up in many areas. It is reasonable to use as a reference an organization close to the one described in many instances (e.g. by NRPB, BNFL, EDF, Framatome, Cogéma, CEPN; see Refs [65-68]). Developments take place in three areas: the setting up of a management system called the ‘ALARA structure’ , a broad policy of motivation associated with the development of a safety culture, and the promotion of suitable tools. Some examples of such elements are provided in Table V. Among the tools, databases on the doses associated with various tasks are essential, together with the possibility of real time follow-up. The development of electronic operational dosimetry has met this requirement. The definition of company goals and criteria also plays a role both for practical purposes and in the motivation of the workforce and managers. A goal might be a maximum individual annual dose, as is often the case for design studies (e.g. 5 mSv in the Cogéma Melox plant, or 10 mSv in AGRs reactors and at the Sizewell В power plant).More specific goals are connected to a given task and may be set up directly by the project group. Other goals are set up at the corporate level (e.g. total collec­ tive dose for the company). BACKGROUND PAPERS 29

TABLE V. COMPONENTS OF CORPORATE MANAGEMENT OF

RADIATION RISK

ALARA structure ALARA committee ALARA co-ordinator Project group Working groups on generic problems

ALARA tools Dosimetric databases and operational monitoring Return from experience Tools for analytical dose previews and follow-up ALARA reviews of jobs Contractor approval Training Value of man-sievert Motivation Setup of goals and targets Promotion of safety culture Protection made a company priority

In parallel with goals, criteria may be proposed, or reference figures; among them is the monetary value for man-sievert. With respect to the organization, EDF (the French utility) constitutes an interesting example, because it is a big company, so that important responsibilities exist at both the corporate level and the plant level. A system of ALARA committees and working groups was set up (see Fig. 8). A first characteristic of this system is to allow the simultaneous involvement of corporate and plant management, and to establish horizontal links among plant managers so that experience can be shared promptly. The second aspect deals with the preparation of projects, the follow-up of the activity and the a posteriori review of achievements. Once again it involves the ‘ALARA committee’, which is the core of the system, and which is close enough to plant or company management to take strategic decisions and to select objectives. Such an involvement has been said to be decisive in 95% of ALARA actions [69], and the lack of such involvement has been outlined by British authorities [70]. Another structure is the ‘project group’, whose members are in charge of the preparation and debriefing of the project. An ALARA co-ordinator is then in charge of the implementation of ALARA during the operation itself. Note that the ALARA approaches for the design stage, for specific operations, and for reduction of routine exposures are quite different. The above description rather relates to a given project such as a steam generator replacement. Apparently, more experience has been gained in design and in planning specific intervention than in optimizing routine operations. 3 0 BACKGROUND PAPERS

FIG. 8. ALARA management system in the French utility company EDF (from Ref. [67]).

Some perspectives and issues

First it must be mentioned that the management system described above may not be applicable to all cases. Even in important companies, alternatives can be found, but in small companies or in the health sector this very structured system may fail, and another type of management may be more efficient. Among the problems, the need for more exhaustive databases on dosimetry is also often underlined. The question of the monetary value of the collective dose is more and more felt to be a critical issue as optimization develops. In the past 10 years it was not such an important issue, because ALARA was not fully developed, and cost effectiveness studies were performed that did not require valuation of the collec­ tive dose, but there is now an evolution. Figures can be provided by experts at the national level (e.g. Ref. [71]), or they may be defined by company managers. Con­ sistency is required for many reasons. For example, it does occur that this figure is explicit in contracts between operators and subcontractors. Corporate values are usually higher than national references, but they are quite variable and may range between US $150 000 and $700 000 per man-sievert. When a weighting factor for comparatively high doses comes into play, they may be much higher. A figure of US $2.7 million has been proposed for doses between 30 and 50 mSv [67]. BACKGROUND PAPERS 31

The inclusion of contractors in the management system is another goal of many policies. Indeed, in the French nuclear industry the collective dose to contractors is more than 60% of the total, and they represent 85% of the workers above 20 mSv [32]. Training, involvment of the contractors in planning, approval of companies, and clauses in contracts are among the approaches that are being undertaken. In all these areas, methods are available. More fundamental developments or at least investigations are required to take into account the need to deal with potential risk, to improve decommissioning operations, and to deal better with work manage­ ment. In this last case, qualification of working conditions and types of organizations and assessment of effciency have not been fully achieved.

4.2. Management at the national and international levels

Participants and elements of the system

The risk management system at the national and international levels can be termed the national or international infrastructure for protection and safety. All the above descriptions have shown that the national or international authorities are not the only participants, and the elaboration of regulatory requirements is far from being the only goal (see Section 2.4). The following components can be listed: — a set of legislation and regulations, not all solely designed to address radiation protection; — a regulatory authority empowered to review and inspect activities, usually comprising a central body and field inspectors; — an institution regrouping experts for the assessment of radiation risk and pro­ tection and safety measures, and possibly for the definition of risk policies; — facilities for the dissemination of information and for training; — facilities for the logistics of radiation protection (e.g. dose registries, ambient measurement networks, reference laboratories, emergency response teams). The roles of authorities, expert institutes and operators must be carefully sepa­ rated, and they must be independent, but other components are designed to be shared. Among the functions to be fulfilled are: — control of the regulated practices (e.g. regulations and requirements, inspec­ tions, monitoring, assessment, enforcement actions); — direct protection of the population when this is not the responsibility of opera­ tors or when a situation is out of their control; — qualification of laboratories, institutions, scientific tools, methods and prac­ tices related to radiation protection; 32 BACKGROUND PAPERS

— development of knowledge on radiation protection (e.g. compiling databases, performing studies and research, setting research goals in other areas, promot­ ing training programmes and academic work); — assessment of the global performance of radiation related activities; — establishment of national objectives and information of the public.

New environment

Most national systems were designed in the late 1960s or early 1970s, and they now must contend with quite a different environment. The sources of radiation risk are not the same, and the organization of the operators has dramatically changed. Nuclear power production has become a large industry, medical practices have spread out and are more sophisticated, and optimization has structured the risk management of many operators. Internationalization of both the activities of and the approaches to risk management do not allow a country to have its own policy. Within the European Union a legal framework ties national authorities, but public opinion is even a stronger tie because it is more and more determined at an international level. Indeed, another evolution is that it is no longer accepted that decisions refer­ ring to risk are not fully explicit. Authorities on one side, experts on the other are bound to communicate their strategies or to make the risk understandable. And last the evolution of international recommendations requires new actions from authorities and necessitates a broader expertise. This new environment raises many questions. Some of these are highlighted here.

Harmonization of sectorial approaches

It has been said that the risk management by operators was quite heterogeneous according to sector (nuclear industry, health, other industry). Unfortunately, the regulatory provisions, the competent authorities and even the expert institutions are also different, with only a few exceptions. The efforts of the National Radiation Pro­ tection Board in the UK, whose statistics encompass all sectors [33] and which pro­ motes optimization in all sectors, are worth noting (e.g. Ref. [72]). Consistency among policies is desirable. Should the national system be more homogeneous, efforts could be more rationally distributed.

Development of international expertise

International expertise is not new for radiation risk. For three quarters of a cen­ tury, standardization, development of models and development of assessment tools have been constant. But the need for an increase in international expertise remains. Assessment of the safety of reactors in eastern Europe, requirements for consistency BACKGROUND PAPERS 33 in the analyses on two sides of a border, and the necessity to have quality assurance for complex risk assessment have resulted in a considerable increase in the collabora­ tion among institutes for expertise.

Involvement in operational protection

An involvement of the national system is essential. It has been shown that once radiation protection programmes are set up, important and cost effective reductions in exposures result. However, the commitment to setting up these programmes is still lacking in many areas. It is believed that the commitment of authorities should be made clear in order to support the development of corporate policies. The issue of databases dealing with the doses associated with various tasks and activities and with radiation protection techniques (including managerial aspects) has been raised at the corporate level. When there is no agreement among operators, national or even international action is obviously a necessity for the institution of such programmes, the most important of them being the ISOE project of NEA [61]. Some databases, such as this one, are already well advanced, but all national infrastructures are not yet fully involved. In addition, in contrast to the nuclear indus­ try, other branches concerned with radiation protection are much less organized and the incentive has to come from outside. Such databases are a tool for corporate or plant managers implementing ALARA, and for an authority assessing achievement in radiation risk or setting priorities. They are also a key tool when authorities must define or discuss constraints. Assessment of the good performance of the global radiation protection opera­ tional system is a difficult task. However, it is not out of reach. The issue is of course already dealt with in many ways, mostly on the basis of dose records. As regards the operational aspect itself, thére are at present requirements, but the ‘softer’ part of the management (e.g. detailed procedures, unwritten behaviours, responsibility of health physicists) is not often addressed. The question of the monetary value for man-sievert was pointed out. Even if the authorities do not wish to define reference figures, the national infrastructure must have the ability to cope with this question, if only for its own purpose.

Optimization at the national level

With respect to the general aspects of risk management, in some countries (e.g. the USA) the authorities may have to prove that their policies are optimized. The need for optimization at the national level is now clearer for radiation risk, and it calls for broadening the expertise in this direction. The ICRP calls openly for the use of optimization at the national level in dealing with intervention [4]. A first case is the definition of intervention level after an accident [37]. A second case is the defi­ nition of policies for radon in dwellings, and more precisely for the definition of 34 BACKGROUND PAPERS action levels [38]. In many countries decisions have been taken, but formal decision analysis or optimization has played a substantial role in only a few cases. There is still a need for the national risk managment system to adapt in order to effectively support these management approaches. Lastly, but a more controversial issue, it has been said that the authorities must apply optimization when defining levels for dis­ charge authorization.

Regulatory requirements

The authorities may choose to put forward requirements or to base their poli­ cies on inducements. This is a difficult choice when addressing corporate manage­ ment [73, 74]. It was not discussed in this document because it depends heavily on the national regulatory framework and risk culture. Rather, the focus was on the material necessary for a knowledgeable choice. Altogether, no spectacular actions must be envisaged, but a greater involvement of the authorities and associated infras­ tructures is necessary. It is believed that this effort will result in both better assess­ ment of practices and better practices.

5. CONCLUSION

Risk management in the field of radiation risk has now quite a long history. It started in the late 1970s with the promotion of optimization (the ALARA prin­ ciple). At that time, it was mostly viewed as the development of rational decision aiding methods for choosing options when zero risk levels were not achievable. In the last 20 years, the focus has shifted from theory to practice, and so has the mean­ ing of ‘risk management’ . It has been shown that risk management is a complex concept. On one side it refers to the principles that are followed when trying to control risk, and social values come into play. On the other side it is associated with the practical behaviour of the participants. In this respect, radiation risk management is quite similar to the management of other risks and the main difficulty lies in setting up systems that fit with the principles, and also in revising the principles in order to face the evolution of the practical management of the risk. The ALARA approach, as it is applied by operators in the nuclear industry, does not refer any more only to the use of cost-benefit approaches but also to the appropriate organization of the work implemented in order to comply with the princi­ ple. Development of ALARA textbooks and implementation of ‘ALARA organiza­ tion’ by corporate managers are now well engaged and have been achieved by many operators. BACKGROUND PAPERS 35

This fact has been recognized by many national authorities and international bodies, which have started to adapt their principles and practices in order to be more effective with respect to these aspects. The new ICRP recommendations contain many developments on possible means both to correct the inherent weaknesses of the optimization approaches (cf. constraints or potential risk) and to assess that management of operators is effi­ cient as regards risk (references to ‘good practices’). International discussions show that there is a new challenge for the authorities. After having ensured that limits are not infringed, after having promoted risk management, the authorities must now address the essential issue of the proper means for assessing and guiding this management.

REFERENCES

[1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP, ICRP Publication 26, Pergamon Press, Oxford and New York (1977). [2] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP, ICRP Publication 37, Pergamon Press, Oxford and New York (1983). [3] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP: Optimization in Decision Making in Radiological Protection, ICRP Publication 55, Pergamon Press, Oxford and New York (1989). [4] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP, ICRP Publication 60, Pergamon Press, Oxford and New York (1991). [5] NATIONAL RESEARCH COUNCIL, Optimization of Public and Occupational Radia­ tion Protection at Nuclear Power Plants, NUREG-CR 3665, Washington, DC (1984). [6] NATIONAL ACADEMY OF SCIENCES, Considerations of Health Benefit Cost Analysis for Activities Involving Ionizing Radiations and Alternatives, EPA 520/4-88-003, US Environmental Protection Agency, Washington, DC (1977). [7] NUCLEAR REGULATORY COMMISSION, Reactors Risk Reference Document: Main Report, Office of Nuclear Regulatory Research, Rep. NUREG 1150, Washing­ ton, DC (1987). [8] HEALTH AND SAFETY EXECUTIVE, The Tolerability of Risk from Nuclear Power Stations, 1992 Revn, HMSO, London (1992). [9] INTERNATIONAL ATOMIC ENERGY AGENCY, Status, Experience and Future Prospects for the Development of Probabilistic Safety Criteria, IAEA TD 524, IAEA, Vienna (1989). [10] VAN KUUEN, C.J., “ Risk management in the Netherlands: A quantitative approach” (Proc. Two Meetings Held at IIASA: “ Technological Risk in Modem Society” and “ Safe Technological Systems” ), IIASA, Laxenburg, Austria (1988). 36 BACKGROUND PAPERS

[11] DIRECTORATE GENERAL FOR ENVIRONMENTAL PROTECTION, Premises for Risk Management: Risks Limits in the Context of Environmental Dangers, Berkeley, С A (1989). [12] CONSEIL FEDERAL SUISSE, Ordonnance sur la protection contre les accidents majeurs (OPAM), Conseil Fédéral Suisse (1991). [13] HEALTH AND SAFETY EXECUTIVE, Quantified Risk Assessment: Its Input to Decision Making, HMSO, London (1989). [14] MINISTERE DE L'URBANISME, DU LOGEMENT ET DES TRANSPORTS, Direction des Routes: Instruction relative aux méthodes d’évaluation des investisse­ ments routiers en rase campagne, Note Dr. Setra, Paris (1986). [15] ROCARD, P., SMETS, H., Evaluation socio-économique des mesures de maîtrise de l’urbanisation au voisinage des installations dangereuses, Préventique, Grenoble (1990). [16] NATIONAL RESEARCH COUNCIL, Decision Making in the Environmental Protec­ tion Agency, Committee on Environmental Decision Making, National Academy of Sciences, Washington DC (1977). [17] MERKHOFER, M.W., Decision Science and Social Risk Management, Reidel, Dordrecht (1987). [18] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP, ICRP Publication 15, Pergamon Press, Oxford and New York (1958). [19] NATIONAL RESEARCH COUNCIL, Risk Assessment in the Federal Government: Managing the Process, Committee on the Institutional Means for Assessment of Risks to Public Health, National Academy Press, Washington, DC (1983). [20] STOCKELL, P.J., CROFT, J.R., LOCHARD, J., LOMBARD, J., ALARA, From Theory Towards Practice, CEC Rep. EUR 13796 (1991). [21] VON NEUMANN, J., MORGENSTERN, О., Theory of Games and Economic Behaviour, Princeton University Press, Princeton, NJ (1944). [22] PRATT, J.W., RAIFFA, H., et al., The foundations of decision under uncertainty: An elementary exposition, JASA 59 (1987) 353-375. [23] KEENEY, R.L., Siting Energy Facilities, Academic Press, New York (1980). [24] HUBERT, P., BARNY, M.H., MOATTI, J.P., Elicitation of decision makers prefer­ ences for management of major hazards, Risk Anal. 11 2 (1991). [25] UNITED NATIONS, F AO, IAEA, ILO, NEA, РАНО, WHO International Basic Safety Standards for Protection Against Ionizing Radiation and for the Safety of Radia­ tion Sources, Vienna (1994). [26] INTERNATIONAL ATOMIC ENERGY AGENCY, Extension of the Principles of the Radiation Protection to Sources of Potential Exposure, Safety Series No. 104, IAEA, Vienna (1990). [27] INTERNATIONAL ATOMIC ENERGY AGENCY, The Role of Probabilistic Safety Assessment and Probabilistic Safety Criteria in Nuclear Power Plant Safety, Safety Series No. 106, IAEA, Vienna (1992). [28] INTERNATIONAL ATOMIC ENERGY AGENCY, The Safety of Nuclear Installa­ tions, Safety Series No. 110, IAEA, Vienna (1993). BACKGROUND PAPERS 37

[29] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP: Protection from Potential Exposure; A Conceptual Frame­ work, ICRP Publication 64, Pergamon Press, Oxford and New York (1993). [30] ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT, “ Work management to reduce occupational doses” (Proc. NEA Workshop, 1992), OECD, Paris (1993). [31] HUBERT, P., Comparison of the Methodologies for Risk Management: Applied Com­ parison of Optimisation to Nuclear and Non Nuclear Activities, Final Contract Rep. EEC-DGXII B 16-0207F, CEPN Rep. No. 171, Paris (1990). [32] HUBERT, P., PAGES, P., “ Occupational radiation exposure in the French nuclear industry: Impact of 1990’s ICRP recommendations” (Portamouth 94 Proc.), Nuclear Technology Publishing (1994) 411-414. [33] HUGHES, J.S., O’RIORDAN, M.C., Radiation Exposure of the UK Population, NRPB-R263, National Radiological Protection Board, Chilton, Didcot (1993). [34] McDONNELL, C.E., CROFT, J.R., ATKINSON, R.H., HYDE, M.L., Assessment of Occupational Exposures to Ionizing Radiation of Workers in the Member States of the European Communities, NRPB- M445, Contract Rep., National Radiological Pro­ tection Board (1993). [35] COULON, R., GANZ, J., LUYKX, F., RMOS, L., SERRO, R., WESTEN, J., “ The application of the optimization principle to public protection from effluent discharges” (4th European Seminar on Radiation Protection Optimisation, Luxembourg, 1993). [36] VESSIERES, G., BERARD, S., LEFAURE, C.D., “ DOSIANA: A software package for maintenance operations dose management at PWRs” , Occupational Radiation Pro­ tection (Proc. Int. Conf. of the British Nuclear Energy Society, Guemesey, 1991). [37] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP: Principles for Intervention for Protection of the Public in a Radiological Emergency, ICRP Publication 63, Pergamon Press, Oxford and New York (1993). [38] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP: Protection Against Radon-222 at Home and at Work, Recom­ mendations of the ICRP, ICRP Publication 65, Pergamon Press, Oxford and New York (1993). [39] HEALTH AND SAFETY EXECUTIVE, The Tolerability of Risk from Nuclear Power Stations, HMSO, London (1992). [40] TRAVIS, C.C., CROUCH, E.A.C, KLEMA, E.D., Cancer risk management, Envi­ ron. Sei. Technol. 21 (1987) 415. [41] INTERNATIONAL ATOMIC ENERGY AGENCY, Principles for the Exemption of Radiation Sources and Practices from Regulatory Control, Safety Series No. 89, IAEA, Vienna (1988). [42] HARVEY, M., MOBBS, S., COOPER, J., CHAPUIS, A.M., SUGIER, A., SCHNEIDER, T., LOCHARD, J., JANSSENS, A., Principles and Methods for Estab­ lishing Concentrations and Quantities (Exemption Values) Below Which Reporting Is Not Required in the European Directive, Commission of the European Communities, XI-028/93 (1993). 38 BACKGROUND PAPERS

[43] INTERNATIONAL ATOMIC ENERGY AGENCY, Establishment of Source Related Dose Constraints for Members of the Public, IAEA TECDOC 664, IAEA, Vienna (1992). [44] LOCHARD, J., SCHNEIDER, T., SUGIER, A., HUMBERT, P., Le concept de con­ trainte de dose, Rep. CEPN, Fontenay-aux-Roses (1994). [45] SUGIER, A., CANCIO, D., FRITELLI, L., HUBERT, P., “ Source related dose con­ straints” (Proc. Int. Conf. on Harmonization in Radiation Protection, Taormina, 1993). [46] JAMMET, H., SUGIER, A., “ Application of the ICRP recommendations: Use of con­ straints and levels in concrete cases” (Proc. IRP A Congr., Montreal, 1992) 1047-1052. [47] DUNSTER, H.J., The use of constraints by ICRP, J. Radiol. Prot. 12 4219 (1992) 224. [48] INTERNATIONAL ATOMIC ENERGY AGENCY, Principles for the Establishment of Upper Bounds to Doses to Individuals from Global and Regional Sources, Safety Series No. 92, IAEA, Vienna (1989). [49] COMMISSION OF THE EUROPEAN COMMUNITIES, Methods used for fixing dis­ charge limits for radioactive effluents from nuclear installations in the Member States, Radiat. Prot. No. 42 (1988). [50] LEFAURE, C., LOCHARD, J., SCHNEIDER, T., SCHEIBER, C., Propositions pour un système de valeurs monétaires de référence pour l’homme Sievert, CEPN Rep. 193, Fontenay-aux-Roses (1993). [51] INTERNATIONAL ATOMIC ENERGY AGENCY, Probabilistic Safety Assessment, Safety Series No. 75-INSAG-6, IAEA, Vienna (1992). [52] RASMUSSEN, N., Reactor Safety and Assessment of Accident Risks in US Commer­ cial Nuclear Power Plants, WASH 1400 (1975). [53] CHADWICK, “ Comparative environmental and health effects of different energy sys­ tems for electricity generation“ (Senior Expert Symp. on Electricity and the Environ­ ment, Helsinki, 1991), Key Issues Paper No. 3, International Expert Group 3. [54] SCHNEIDER, T., LOCHARD, J., Réflexions sur l’acceptabilité sociale et les consé­ quences économiques d’un accident nucléaire, CEPN Rep. 191 (1992). [55] DIENST CENTRA AL MILIEBEHEER RUNMONDS, Risk Analysis of Six Poten­ tially Hazardous Industrial Objects in the Rijnmond Area: A Pilot Study, Rep. to the Rijnmond Public Authority, Reidel (1982). [56] HUBERT, P., PAGES, P., Risk management for hazardous material transportation: A local study in Lyon, Risk Anal. 9 4 (1989). [57] UNITED NATIONS, Sources, Effects and Risks of Ionizing Radiation, United Nations Scientific Committee on the Effects of Atomic Radiation, 1988 Report to the General Assembly, UN, New York (1988). [58] BROOKHAVEN NATIONAL LABORATORY, Third International Workshop on Implementation of ALARA at Nuclear Power Plants, May 8-11, 1994, Long Island, NY, US NRC and ALARA Centre, Brookhaven National Laboratory (1994). [59] INSTITUT DE PROTECTION ET DE SURETE NUCLEAIRE, Statistique annuelle d’exposition externe des travailleurs du groupe CEA, Rep. IPSN/DPHD/SEGR, Fontenay-aux-Roses (1992). BACKGROUND PAPERS 39

[60] ANDERSON, R.C., COATES, R., ‘ ‘Dose reduction and the application of the ALARP principle to occupational exposures in the nuclear fuel reprocessing plant in Sellafield in Cumbria” , Occupational Radiation Protection (Proc. Int. Conf. of the British Nuclear Energy Society, Guemesey, 1991). [61] ILARI, O., VIKTORSSON, C., “ An international contribution to occupational dose control in a nuclear power plant” , Occupational Radiation Protection (Proc. Int. Conf. of the British Nuclear Energy Society, Guemesey, 1991). [62] BENEDITTINI, M., TABARE, M., Expositions professionnelles dans les réacteurs à eau pressurisée: Comparaison international de quelques indicateurs globaux entre 1975 et 1989, Rep. CEPN R-178, Fontenay-aux-Roses (1990). [63] PFEFFER, W., “ Achievements of radiation protection in nuclear power plants: Key factors relevant for dose reduction” (Proc. OECD-NEA Workshop on Radiation Pro­ tection Towards the Turn of the Century, Paris, 1993). [64] AEN, ISOE Nuclear Power Plant Occupational Exposure in OECD Countries: 1969 to 1991, AEN/NEA, Paris (1993). [65] COATES, R., LEFAURE, C., SCHIEBER, C., “ The role of work management in occupational dose control” (Proc. OECD-NEA Workshop, Paris, 1992). [66] LEFAURE, C., CROFT, J.R., “ Elements for designing ALARA programmes for the maintenance and routine operations of nuclear facilities” , Occupational Radiation Pro­ tection (Proc. Int. Conf. of the British Nuclear Energy Society, Guemesey, 1991). [67] LEFAURE, C., CROFT, J., PFEFFER, W., ZEEVAERT, T., ‘‘ALARA in European nuclear installations” (Third Int. Workshop on Implementation of ALARA at Nuclear Power Plants, Hauppauge, NY, 1994). [68] STRICKER, L., ROLLIN, P., DOLLO, R., “ Electricité de France ALARA policy” (Third Int. Workshop on Implementation of ALARA at Nuclear Power Plants, Haup­ pauge, NY, 1994). [69] DIONNE, B.J., BAUME, J.W., Occupational Dose Reduction and ALARA at Nuclear Power Plants: Study on High Dose Jobs, Radwaste Handling and ALARA Incentives, NUREG/GR 4254, PNL 51888 (1985). [70] ROBINSON, I.F., TURTON, D., “ Regulatory experience with ALARP investigation report at some UK nuclear sites” , Occupational Radiation Protection (Proc. Int. Conf. of the British Nuclear Energy Society, Guemesey, 1991). [71] ROBB, J.D., WRIXON, A.D., Revised Estimates of the Monetary Value of Collective Dose, Rep. NRPB-M157, Chilton (1988). [72] NATIONAL RADIATION PROTECTION BOARD AND THE ROYAL COLLEGE OF RADIOLOGY, Patient Dose Reduction in Diagnostic Radiology, NRPB, Chilton, Didcot (1990). [73] BINES, W.P., BEAVER, P.F., “ Experience with the 1985 UK ionizing radiation regulation: The regulators viewpoint” , Occupational Radiation Protection (Proc. Int. Conf. of the British Nuclear Energy Society, Guemesey, 1991). [74] HUBERT, P., “ The regulatory assessment of new risk management practices” (Proc. OECD/NEA Workshop on Radiation Protection Towards the Turn of the Century, Paris, 1993).

IAEA-CN-54/2B

IMPACT OF RADIATION ON THE ENVIRONMENT

D. ROBE AU Institut de Protection et de Sûreté Nucléaire, Fontenay-aux-Roses, France

1. INTRODUCTION

The entire ecosystem is irradiated by natural and artificial radionuclides. This radioactivity is distributed heterogeneously all over the Earth and moves from one component to another, from a plant or animal species to another, transported by fluids. Understanding of environmental damage and environmental transfer is the objective of radioecology, the study of the impact of radiation on the environment. In particular, radioecology contributes to the study of the movements of air, surface water, and underground water, as well as the transfer of radioactive and stable ele­ ments in soil, plants and animals, and the food chain. In this background report are described: — the origins of natural and artificial radioactivity; — the contribution of radioactivity observations to the study of transfer processes in fresh water, ocean and the atmosphere; — the radiosensitivities of plants and animals, especially those species studied in the contaminated Kyshtym and Chernobyl areas.

2. ORIGIN OF THE NATURAL RADIOACTIVITY IN THE ENVIRONMENT

2.1. Natural essential radionuclides

The Sun and its planets were born from interstellar gas produced during the explosion of a supernova. The solar system was formed approximately 4.5 x 109 years ago. The Earth, essentially formed of heavy nuclei, contains rela­ tively little hydrogen and helium. The disintegration of the radionuclides heats the core of the Earth. These unstable nuclei are the natural essential radionuclides. Only radionuclides whose half-life is greater than 0.3 x 109 years still contribute significantly to natural radiation, because their half-lives are the same order of magnitude as their age, e.g. 40K (half-life 1.3 x 109 years), 238U (half-life 4.5 x 109 years) and 232Th (half-life 1.3 x 109 years). Radionuclides with a rela­ tively shorter half-life such as 235U (half-life 0.7 x 109 years) are rare, because the

41 42 BACKGROUND PAPERS amounts of these have decreased. Those with a very long half-life such as 147Sm (half-life 100 x 109 years) do not have negligible radioactivity but always have low activities per unit mass. 40K, 238U and 232Th are present in rocks. Mainly, their presence contributes to external radiation. 40K, present in the entire biosphere, is the ten thousandth part of the potassium in nature.

2.2. Natural secondary radionuclides

The natural secondary radionuclides released continuously by disintegration of the natural essential radionuclides have a relatively short half-life. Of these, 243U (half-life 0.25 X 106 years) has the longest half-life. Except for the radon isotopes, and their short half-life disintegration products, traces of these radionuclides are present in aerosols produced by resuspension of dust. 222Rn, produced by disintegration of 226Rn, and 220Rn (thoron), produced by disintegration of 224Rn, and their short half-life disintegration products are impor­ tant sources of radioactivity. These gases emanate from rocks, move by underground water, and are accumulated in cavities and dispersed in air. At equilibrium, average concentrations are assessed at 2 Bq/m3 in open air and 20 Bq/m3 indoors for 222Rn and at 0.04 and 0.4 Bq/m3 for thoron.

2.3. Induced natural radioactivity

Nuclear reactions between certain nuclei and natural radiation produce induced natural radionuclides. For example, neutrons produced in uranium ore by spontaneous fission (reactions in 238U with a half-life of 8 X 1015 years) can yield fission products, 239Pu and other transuranium elements by capture reactions. Traces of these elements are found in ore. The most important source of induced natural radioactivity is reactions between nitrogen, oxygen, argon and cosmic radia­ tion. About 20 such radionuclides have been identified. Most important are 14C, 3H (tritium), 22Na and 7Be. This inventory concerns only natural sources not modified by human activities such as use of potassium hydroxide or natural phosphates, extraction of uranium ore, burning of coal in electrical power plants, spreading of phosphated fertilizer, and use of gypsum in construction.

3. ORIGIN OF THE ARTIFICIAL RADIOACTIVITY IN THE ENVIRONMENT

The main sources of artificial contamination of the environment are:

— gaseous and liquid releases from nuclear power plants, nuclear fuel reprocess­ ing plants, nuclear research centres, etc.; BACKGROUND PAPERS 43

— fallout from atmospheric nuclear weapons tests; — contamination due to major nuclear accidents (Chernobyl, Kyshtym, Windscale). Atmospheric nuclear explosions were the most important source of radioactive releases: 100 times more radioactive aerosols, 10 000 times more tritium, and 3 times more 85Kr were released into the atmosphere than by the Chernobyl nuclear power plant accident. The Chernobyl disaster produced 10 times more radioactive gases and 30 times more radioactive aerosols and liquid effluents than the routine releases of the nuclear installations in the world. Fission and especially fusion nuclear bomb tests released 160 x 106 TBq of tritium (1 Tbq = 1012 Bq) into the atmosphere between 1950 and 1970. The inven­ tory of tritium in oceans in 1972 reached 60 x 106 TBq. In 1980, the tritium accumulated in the environment was assessed at 40 x 106 TBq. It represents the greatest part of the world inventory, but is decreasing. Tritium appears in nuclear fuel by ternary fission and by activation reactions with various elements such as lithium, boron and deuterium. In fuel reprocessing, a great deal of tritium is released. In the world, at the end of 1989, nuclear power plants produced more than 300 GW of electrical energy, with a yearly tritium production of approximately 150 X 103 TBq. Nuclear weapons tests produced 14C from atmospheric nitrogen. The quantity produced up to 1970 was assessed at 200 x 103 TBq. This artificial production reached 70% of the initial natural inventory. The activity has decreased since 1965, and absorption by oceans and the biosphere is greater than production. Nuclear power plants and nuclear fuel reprocessing plants release 14C into the environment. Pressurized water reactors release approximately 200 TBq a year. Radioactive aerosols in the environment come from nuclear weapons tests and the nuclear industry. Unlike noble gases, tritium, and ,4C, aerosols are not dis­ tributed evenly.

4. TRANSFER OF THE RADIOACTIVITY IN THE ENVIRONMENT

The average radioactivity due to tritium in surface water is 0.5 Bq/L. This radioactivity induces an average 0.35 Bq/kg contamination of plant species and a 1 pGy/h internal irradiation of plant tissue. Radioactivity due to 14C in the biosphere is 230 Bq per kilogram of carbon. This activity induces a 30-50 Bq/kg contamination in plants and a 1 nGy/h internal irradiation of plant tissue. The aver­ age radioactivity of 222Rn in underground water is 200 Bq/L and induces a 1 ^Gy/h internal irradiation. The maximum is on the order of 8000 Bq/L. The mean radioactivity in plant species due to lead and 210Po is on the order of 10 and 8 Bq/kg, respectively. 4 4 BACKGROUND PAPERS

Because the radioactivity is not homogeneous, its transfer between the compo­ nents of the environment, between plant and animal species, is of interest in radio­ ecology. Researches on the behaviour of the radionuclides are based on the observation of contamination.

4.1. Transfer of radioactivity in water

Observations of the contamination of the components of the aquatic ecosystem (sediments, plants, water, fish, etc.) permit study of the fixing mechanisms of the major radionuclides such as strontium, caesium and ruthenium — in particular in the bryophytes, such as mosses, good bio-indicators of radioactive contamination. Such studies in situ are done to determine the radioecological parameters characteristic of the uptake of radionuclides, e.g. the biotic and abiotic factors. These researches have been developed by the Institut de Protection de Sûreté Nucléaire (IPSN) for the past 10 years.

4.2. Transfer of radioactivity in the sea

For 12 years, the laboratory of marine radioecology of the IPSN has studied the transfer of radioactivity in the English Channel and the North Sea coming from releases of nuclear fuel reprocessing plants in La Hague (France) and Sellafield (UK). Every year, two campaigns are organized to analyse water dispersion, using the soluble radionuclides as tracers. Follow-up of the contamination is linked to the identification of the water dispersion. These radioecological studies contribute to the understanding of current flow and permit validation of hydrodynamic models devel­ oped in France, Denmark and Germany. These models allow forecasting of radioac­ tive or chemical pollution in the English Channel and the North Sea. They could be used for assessment of the consequences of an accidental pollution. Studies of sediments and particles suspended in water are based on observa­ tions of their contamination. Radionuclides are accumulated by sedimentation and by absorption on suspended matter. Their importance as secondary sources of radioac­ tivity is not negligible, because a balance generally is established between water and sediment. When radioactivity in water decreases, sediment releases a part of the fixed radioactivity. This phenomenon has been observed in the Irish Sea after the Sellafield plant had considerably reduced its releases. Three characteristic radionu­ clides of the releases of the La Hague and Sellafield plants are being studied: 125Sb (half-life 2.73 years), 137Cs (half-life 30.17 years) and 134Cs (half-life 2.06 years).

4.3. Transfer of the radioactivity in the atmosphere

To observe the contamination of the atmosphere, the ISPN manages six mea­ surement stations on the French mainland and two marine stations. Analysis of the BACKGROUND PAPERS 45 measurements allows a better understanding of atmospheric aerosols. Thus, during atmospheric tests of nuclear weapons 137Cs was retained in the stratosphere. Fallout from the stratosphere and seasonal air movement explain the annual variation of the radioactivity: a maximum in the summer, a minimum in the winter. These observa­ tions allow determination of the decrease rate of caesium in the stratosphere. The half-life of 137Cs in air was 1 year from 1950 to 1985. During the Chernobyl accident, the troposphere was loaded with 137Cs. The deposition of aerosols and the decrease of the caesium in the troposphere reservoir were important. In 1986, during the months following the accident, the half-life of 137Cs in air was 8 days. The deposited caesium is resuspended in air. Observations show that the annual variation of the resuspended radioactive particles is opposed to that observed when the radioactivity comes from the stratosphere: a minimum in the summer and a maximum in the winter. Since 1986, the half-life of the 137Cs in air has been 3 years.

5. RADIOSENSITIVITY OF PLANT AND ANIM AL SPECIES

5.1. Plants

The lethal dose (LD50, the dose resulting in 50% mortality) of plant species varies from 10 to 1000 Gy. The most radioresistant plants are mosses, lichens and unicellular species. The most radiosensitive are woody species, whose LD50 varies between 10 and 200 Gy. The resinous woody species are especially radiosensitive. The radiosensitivity of crops is expressed as reduction of yield (YD50 is the dose causing a 50% reduction in expected yield). For cereals, the YD50 varies between 5 and 50 Gy. Rice is an exceptionally radioresistant cereal; its YD50 is 140 Gy. For leafy vegetables, the YD50 varies between 2 and 60 Gy. Root vegetables are less radiosensitive, their YD50 varying between 5 and 100 Gy. The YD50 of herbs and pastures is from 150 to 200 Gy. Unfavourable growing conditions, such as moisture deficiency, can increase radiosensitivity by a factor of 4. The radiosensitivity of a plant increases with age. Irradiation of tissue and moisture deficiency increase the stress on a plant, which can induce desiccation and possibly death. Stress can be detected early by study of vari­ ous substances such as enzyme systems: peroxidase or superoxide dismutase. Trees are radiosensitive. This radiosensitivity is stronger in autumn than in spring, the period of growth. Below the LD50 (50-200 Gy), trees have difficulties in reproduction and reduction of photosynthesis. Morphological anomalies are observed in chronic irradiations of several years, on the order of 1 mGy/h. Metabolic modifications in herbs (diminution of growth, disappearance of the more sensitive species) for dose rates between 1 and 10 mGy/h are observed. The same kind of metabolic modifications are observed in lichens at 1 Gy/h dose rate. 4 6 BACKGROUND PAPERS

5.2. Animals

The LD50 for small mammals is between 6 and 16 Gy. It varies from 1.5 to 2.5 Gy for large mammals. The LD50 for birds is on the order of 10 Gy. Irradiation of immature birds to 50% of the LD50 is observed to reduce size at maturity by about 10%. The LD50 for reptiles and amphibians varies from 2 to 22 Gy. That of inver­ tebrates is approximately 600 Gy. For viable invertebrate young the LD50 is about 100 Gy, for larvae it is about 20 Gy. The LD50 for fishes varies from 10 to 100 Gy. The LD50 for fish larvae and embryos is some tenths of a Gy.

6. EXPERIENCES FROM THE KYSHTYM AND CHERNOBYL ACCIDENTS

6.1. The accident at Kyshtym (1957)

Among the plant species, the main damage from the Kyshtym accident was to forest. The dose delivered to forest during the summer and autumn following the accident was estimated at about 40 Gy. The following spring, a high mortality of pines was observed. These pines died at doses greater than 20 Gy. Morphological anomalies were observed at doses greater than 5 Gy (desiccation, loss of needles, absence of new shoots, reduction of the size of cones, reduction of the quantity of seeds) during the 2 or so years following the accident. Birches, less radiosensitive than pines, bore up to the irradiation and the observed LD50 was estimated at about 200 Gy. Sublethal effects occurred for doses greater than 50 Gy (delay of budding). The forest seemed returned to normal 10 years after the accident. However, the litter, trees and new shoots were still contaminated to a level limiting normal exploitation of the forest. In areas where seeds received doses greater than 20 Gy, corresponding to 90Sг deposition of 1 GBq/m2, germinations were not observed. In areas where seeds received doses on the order of 6 Gy (300 MBq/m2), germinations and growth were observed. Recovery of the forest was normal in less than 100 MBq/m2 contaminated zones. The fauna was touched by the contamination. Rodents have been particularly studied, during 40 generations. These rodents lived in a zone where 90Sr contami­ nation was 44 MBq/m2. The annual dose to the skeleton of these animals was esti­ mated at 3 Gy in 1962 and 0.2 Gy in 1980. During this period, cellular aberrations of the red bone marrow and very weak reproduction were observed. However, rodents survived and seemed to bear up to the detriment caused by the irradiation. Cattle grazing under the plume during the accident were contaminated by 90Sr deposition of 1 GBq/m2, receiving a dose estimated at 50 Gy to the gastrointestinal BACKGROUND PAPERS 47 system and more than 2 Gy to the skeleton. Most of these cattle were dead at tens of days after the accident. Observations on fish, which were irradiated to a substantial dose rate (some mGy/h) after the accident, do not show deleterious effects.

6.2. The accident at Chernobyl (1986)

The main damage to plant species caused by the Chernobyl accident was to forest. Five to six hundred hectares of forest was irradiated at from 80 to 100 Gy. A second zone (300 ha) was irradiated at from 8 to 10 Gy and a third zone (12 000 ha) at about 3 Gy. Radiological observations in the Chernobyl area were equivalent to those in the Kyshtym area. A large part of the trees in the first zone died. Studies led by the IPSN in this zone have shown that for 137Cs and 90Sr deposits on the ground, which were almost the same, there is in pine wood approxi­ mately 10 times more strontium than caesium. These studies established that the maximum activity in wood is on the order of 150 kBq/kg. The peripheral part of the trunk is the most contaminated because the radionuclides coming from roots are dis­ tributed vertically by the vascular system situated underneath the bark. These radio­ nuclides are then distributed to the heart wood. The preponderant role of the roots in timber contamination can be seen in trees which have sprouted since the accident and whose levels of radioactivity are equivalent to those of adult trees.

7. CONCLUSIONS

Studies led by the IPSN have shown that, for both populations and the environ­ ment, global exposure to natural radioactivity is more important than exposure to artificial radioactivity. These exposures are important for plant and animal species in particular conditions, but the flora and the fauna generally have a strong radi­ oresistance. Studies in radioecology provide material for reflection with respect to radiation protection. They aim especially at explanation of fundamental processes of transfer in the environment at both macroscopic and microscopic levels. The improvements in knowledge about deposition from the troposphere and resuspension of aerosols from soil, passage of aerosols from the stratosphere to the troposphere, water dispersion in the English Channel and the North Sea, water-sediment transfer on the sea floor, and fixing mechanisms of elements in mosses are typical examples of results from the rigorous observation of the radioactivity in the environment.

Technical Session 1

ASSESSMENT OF RADIATION EXPOSURE LEVELS

IAEA-CN-54/34P

ANNUAL EFFECTIVE RADIATION DOSES FROM NATURAL SOURCES IN MOLDAVIA

O. IACOB, E. BOTEZATU, C. GRECEA, C. DIACONESCU, L. CLAIN Institute of Public Health and Medical Research, Iasi, Romania

1. INTRODUCTION

Radiation from natural sources represents the most important contributor to the total radiation exposure of populations and includes cosmic rays and gamma terres­ trial radiation from radionuclides (40K, 238U and 232Th series) in the Earth’s crust and in building materials as external radiation sources, as well as naturally occurring radionuclides taken into the human body through inhalation (222Rn and 220Rn progeny) and ingestion (40K, 238U — 226Ra, 210Pb — 210Po, 232Th) as internal radi­ ation sources. The purpose of this study was to evaluate the extent of natural radiation exposure in order to keep the overall level of exposure under control in accordance with requirements of Romanian legislation for radiological protection people and the environment.

2. METHODS

Our methods to estimate exposures were specific to each type of radiation source and exposure pathway. All the dosimetric coefficients used in our estimates are those adopted by UNSCEAR 1993 [1]. Population radiation exposure is expressed in terms of effec­ tive dose, the new basic dosimetric quantity recommended by the ICRP [2] for radio­ logical protection. The effective doses from cosmic rays were calculated for both directly and indirectly (neutron) ionizing components by use of analytical expressions developed for the general relationship between annual effective dose and altitude. The effective doses due to terrestrial gamma irradiation were derived from average activity concentrations of 40K, 228U and 232Th in soil, determined by multi­ channel gamma spectrometry.

51 52 POSTER PRESENTATIONS

Internal irradiation by ingestion of food was evaluated from the natural radio­ active content of diet determined by gamma spectrometric measurements and radio­ chemical methods. Internal irradiation through inhalation was estimated from 222Rn and 220Rn progeny concentrations measured in typical Moldavian dwellings, using the active method of suction of air through a filter and counting of the deposited activity with a ZnS alpha scintillation counter [3]. Radiation induced risk estimates were made from annual collective doses by application of a risk factor of 5 X 10-2/Sv for fatal cancers for a population of all ages and both sexes, as recommended by the ICRP [2].

3. RESULTS AND DISCUSSION

The results regarding external irradiation from cosmic rays are listed in Table I as population weighted averages. The annual value due to the ionizing component is about 252 /xSv and that from the neutron component is 40 /¿Sv, for a total of 292 fiS\. Table II shows our results regarding the average activity concentration of natural radionuclides in soil, the absorbed dose rate in outdoor and indoor air, and the corresponding population weighted average annual effective doses. The average outdoor and indoor terrestrial absorbed dose rates in air from terrestrial gamma irradiation are 53 and 74 nGy/h, respectively. The overall average annual effective dose resulting from external terrestrial gamma exposure is 434 ¡iSv, with the most important contribution from 232Th. Table III shows the annual activity intake and the corresponding annual effec­ tive doses. The average dose of 223 /¿Sv is due especially to 40K (170 ¿tSv), which is metabolically controlled in the body, and to 210Pb — 210Po (46.64 ftSv).

TABLE I. EXTERNAL IRRADIATION DUE TO COSMIC RAYS

Annual effective dose Cosmic radiation (MSv)

Ionizing component 252 Neutron component 40

Total 292 POSTER PRESENTATIONS 53

TABLE II. EXTERNAL TERRESTRIAL GAMMA IRRADIATION

Activity concentration Absorbed dose rate Annual effective dose Radionuclide in soil in air (^Sv) or decay (Bq/kg) (nGy/h) series Average Range Outdoors Indoors Outdoors Indoors

«K 450 250-1100 18.6 26.0 29 123 238U 31 8-56 14.4 20.0 23 95 232Th 32 11-65 20.0 28.0 32 132

Total 84 350

Total 53 74 434

TABLE III. INTERNAL IRRADIATION DUE TO INGESTION OF NATURAL RADIONUCLIDES

Annual activity intake Radionuclides (Bq) Annual effective dose G*Sv) Average Range

40K 31 700 23 400-35 400 170 238u 6.6 4.2-8.3 0.83 226Ra 16.4 13.4-20.7 3.28 2l0Pb 39.4 29.0-45.0 40.0 2l0Po 33.2 24.8-40.0 6.64 232Th 2.3 1.4-4.7 1.77 Total (rounded) 223

Table IV includes our results regarding the most important natural radiation sources, inhaled 222Rn and 220Rn progeny. The population weighted average values of the equilibrium equivalent concentration (EEC) of radon are 19.0 Bq/m3 indoors and 3.6 Bq/m3 outdoors, with the corresponding annual effective doses estimated at 1250 fiSv and 80 /xSv, respectively, for a total of 1330 /¿Sv. The contribution of inhaled radon progeny to internal alpha irradiation is smaller: the average annual effective dose is about 210 /¿Sv. 54 POSTER PRESENTATIONS

TABLE IV. INTERNAL ALPHA IRRADIATION FROM INHALATION OF 222Rn AND 220Rn SHORT LIVED DECAY PRODUCTS

Average activity Annual effective dose Source of exposure Location concentration (EEC) 0*Sv) (Bq/m3)

Indoors 19.0 1250 Rn progeny Detached house 32.5 2140 Block of flats 8.7 570 Outdoors 3.6 80 Total 1330

Indoors 0.9 190 Rn progeny Detached house 1.3 270 Block of flats 0.6 130 Outdoors 0.2 20 Total 210

TABLE V. AVERAGE ANNUAL EFFECTIVE DOSES FROM NATURAL RADIATION BACKGROUND

Annual effective doses Source Per capita Collective % (mSv) (man • Sv)

Cosmic radiation 292 1 518 11.7 Terrestrial gamma radiation 434 2 257 17.4 Ingestion 223 1 160 9.0 222Rn progeny inhalation 1330 6 916 53.5 220Rn progeny inhalation 210 1 092 8.4

Total (rounded) 2500 13 000 100 POSTER PRESENTATIONS 55

4. SUMMARY AND CONCLUSIONS

Table V summarizes the annual effective dose both per capita and collective arising from natural radiation exposure of the Moldavian population. The average annual effective dose from the natural background of Moldavia is about 2500 /¿Sv (292 /xSv cosmic irradiation, 434 /¿Sv terrestrial gamma irradia­ tion, 223 p S \ ingestion of natural radionuclides, 1330 ¿¿Sv and 210 /tSv inhalation of 222Rn and 220Rn short lived decay products), with the most important contribu­ tion from 222Rn progeny (53.5%). The corresponding collective effective dose of about 13 000 man-Sv may be responsible for 650 potential fatal cancers annually.

REFERENCES

[1] UNITED NATIONS, Sources and Effects of Ionizing Radiation (Report to the General Assembly with Scientific Annexes), Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), UN, New York (1993) 33-91. [2] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990 Recommendations of the International Commission on Radiological Protection, ICRP Publication No. 60, Pergamon Press, Oxford and New York (1991). [3] IACOB, O., BOTEZATU, E., “ Human exposures to indoor airborne short-lived 222Rn and 220Rn daughters” (Proc. Int. Conf. Indoor Climate of Buildings, Bratislava, 1992), University of Bratislava (1992) 159-165. 56 POSTER PRESENTATIONS

IAEA-CN-54/41P

A WORKSHOP ON HARMONIZATION OF EAST-WEST RADIOACTIVE POLLUTANT MEASUREMENTS, STANDARDIZATION OF TECHNIQUES AND CONSIDERATIONS OF SOCIOECONOMIC FACTORS: Report on the Commission of European Communities Workshop held in Budapest, Hungary, August 1994

B.R. ORTON Radiation Protection Office, Brunei University, Uxbridge, United Kingdom

D.A. VORSATZ Lawrence Berkeley Laboratory, Energy and Environment Division, Berkeley, California, United States of America

1. INTRODUCTION

The objective of this successful workshop was to bring together scientists from the Czech Republic, Slovakia, Hungary and Romania with those from England, Poland, Germany and Sweden to discuss harmonization of measurements on radon, radioactivity in river water, and the networks used to measure and distribute infor­ mation on ambient radioactivity. The frameworks for Government regulations which set safe levels for members of the public and workers in all these areas were com­ pared. The justification for monitoring the radiation exposure of the public at large was given by demonstrating the degree of harmonization that exists and its economic and social value.

2. RISK AND RADON

The basic equation defining risk (R) is R = PC, where P is the probability of occurrence and С is the severity of occurrence (P = 1, certainty; С = 1, death). A low risk can be set at 1/106 and is called 1 microrisk. To obtain a general idea of the risks run by the public, 1 microrisk is equivalent to travelling 65 km by car or 2500 km by air or drinking half a litre of wine. Radioactive exposure to a dose equivalent of 1 mSv can lead to 50 microrisks of cancer [1], and such exposures can POSTER PRESENTATIONS 57 be received by members of the public in their homes. This perceived risk has stimu­ lated many investigations of radon in the home as it represents a large proportion of the radiation exposure of the public. There exists a measure of harmonization in the radon levels which require remedial action. For example, in Sweden [2], the Czech Republic [3], Slovakia [4] and England [5] the level is set at 200 Bq/m3, in Germany at 250 Bq/m3 [6]. There are no levels available for Hungary [7] or Roma­ nia [8]. Different methods were used for testing the radon levels (i.e. etch track detectors [2, 4-6], membrane filtering [8] and electret [3, 5]). C liff [5] reported on radon levels in the UK. He showed that there were about 20 000 homes which exceeded the UK action level of 200 Bq/m3. For these houses remedial action was recommended. The most effective method was found to be the building of a radon ‘sump’ into which radon gas collects and is then vented to the atmosphere without entering the house. However, the effectiveness and cost of mitigation steps were highly variable, and had to be checked by further measurements. The presence of radon in the home has been extensively used for the teaching of radioactivity con­ cepts in schools [9, 10]. It has been strongly argued by Tóth [9] that by using the knowledge of the children the inhabitants of houses with high radon levels were edu­ cated in the problems of ionizing radiation and risk. They could then make rational choices about the priorities for mitigation.

3. WATER

Liquid radioactive waste is mainly disposed of into drains and then into rivers. It can enter the domestic supply when river water is the source of household water. Newstead [11] outlined the monitoring of rivers and streams in the UK. The fre­ quency of testing was increased in 1985 because of public pressure, and it was shown that the total alpha and beta activities in UK rivers were all below the World Health Organization’s guideline values of 0.1 and 1.0 Bq/L. The only significant dose of 5 /xSv to the public was due to the 250 mBq/L of 137Cs recorded after the Chernobyl accident. Measurements of waters of the Danube from Hungary [12] and Romania [13] all showed large increases following Chernobyl. In contrast, samples taken from the Danube near the Hungarian nuclear power plant at Paks always have very low levels of activity [12]. Monitoring of the Danube at Budapest [14] has shown 6 mBq/L of 125I and 131I from hospitals which discharge quantities of these isotopes into the drains. Danube water measurements have been made at Bratislava, Slovakia, but are not yet fully established [15].

4. NETWORKS AND POST-CHERNOBYL RESULTS

Tatara [16] outlined the Slovak radiation monitoring network (SURMS) which links the experts and specialized laboratories of five institutions of the Slovak Repub­ lic. This network ensures that routine radioactive measurements are made over the 58 POSTER PRESENTATIONS whole of Slovakia. A further network of on-line counters at 26 locations forms the Slovak component of the International Radiation Information System (IRIS), based in Germany. The system extends to the Czech Republic and was described by Bucina [17]. The Hungarian information system [18] includes many of the facilities found in IRIS but also has provision for estimates of dose to members of the public. Such estimates are of great importance in a nuclear emergency, such as Chernobyl. Mócsy [19] gave estimates of dose in the range between 14 and 283 mGy for the Cluj region of Romania as a result of radioactive caesium from the Chernobyl cloud. Radioactive caesium fell on many countries in Europe and was taken up in the food chain, includ­ ing milk products. The fallout missed countries in the southern hemisphere such as Venezuela. However, careful gamma ray spectroscopy has shown 137Cs present in powdered milk products for sale in that country [20].

5. REGULATORY ORGANIZATIONS

Marshall [21] described the regulatory organization for the control of ionizing radiation that exists in the UK. It incorporates certain basic requirements of fun­ damental importance in making a workable system. The practical limits of exposure are the same as those recommended by the International Commission on Radiological Protection (ICRP). The controls imposed fit the legal system and are practical, enforceable and transparent to the user. To be certain that anomalies do not arise, exemptions for certain classes of materials and operations are incorporated in the legislation. In Slovakia [16], Romania [8] and Hungary [18] the respective Ministries of Health and Hygiene follow recommendations of the ICRP, but detailed legislation has yet to be completed. The Bulgarian approach is to provide access to information on radioactivity under an environmental law [22].

6. ENVIRONMENTAL ECONOMICS

Szlávik [23] considered the economics of one of the sources of environmental radioactivity, nuclear power. He showed that the traditional free market mechanisms cannot be applied because they are short term, whereas the detrimental effects of radiation are long term, because of the problems of waste storage and the latent period for development of cancer. For sustainable growth to be maintained, long term effects must be a feature of economic and social practices. Kobjakov [24] stressed the dilemma existing in Hungary, where the development of the free market has produced a call for western consumerism coupled with international obligations to improve the environment. Only 10% of Hungarians would give priority to environmental protection over economic development, compared with 21% in the EEC. It was pointed out that environmentalists require a ‘market’ to produce funds in the same way that business requires a market for its goods. POSTER PRESENTATIONS 59

A high level of radioactivity in the environment can result in deaths. Death through cancer, particularly of children, is most unacceptable to the public. Marin [25] described the determination of the value of a ‘statistical life’. The reasoning was that if a group of people will pay £n to reduce their risk of death by a low probability of x-1, then a group of x people would expect a reduction of 1 death per year. Together they will pay £xn, the price of a life. For 1992 monetary values and atti­ tudes, the statistical life is worth £2-3 million, with the cost of reduction of 100 microrisks at between £200 and £300.

7. PUBLIC INVOLVEMENT AND PERCEPTION OF RISK

Marshall [21] described the period of consultation before the Thermal Oxide Reprocessing Plant (THORP) at Sellafield in Cumbria, UK, was started. Great care was taken to consult the public through local meetings and individual answers to objectors’ letters. Persson [26] turned from the particular to the general problem of the communication of risk of radiation to the general public. The guidelines presented included the compassionate acceptance of the public as a legitimate part­ ner, together with that of the media. Risk must be put into perspective by comparison with everyday risk, as listed by Marx [1]. In Sweden, after the Chernobyl accident, this procedure was followed and resulted in a lower number of induced abortions in Sweden compared with other European countries. However, ‘radiophobia’ is an important social psychological factor. Measurements of public perception of risk and radiophobia have been undertaken by Drottz-Sjöberg [27] in both Sweden and the Commonwealth of Independent States (CIS). People in the CIS most greatly affected by the Chernobyl accident showed greatest suspicion and lack of confidence in professional experts. Generally, the lower the knowledge base, the higher is the per­ ception of the risk from radiation exposure both now and in the future.

8. CONCLUSIONS [28]

Measurements of environmental radioactivity have a valued place in the pro­ tection of the public in southeastern Europe. There exists a strong group of scientists who have been conducting these measurements. The degree of harmonization was greatest in measurements on radon and through networks. Less harmonization exists in water measurements because techniques and organizational structures are still being developed in a number of countries. It is strongly recommended that there should be organized a ‘Danube Watch’, based in Budapest, to network measurements conducted of the water of this river which is so vital to the people who live along its banks. All radioactive pollution measurements, together with mitigation of radon in the home, can be justified by the value of a statistical life saved. However, in the 6 0 POSTER PRESENTATIONS present state of public opinion, unmodified by education, the criteria must be the value of a perceived risk to a statistical life. The way that this perception of risk works can be formulated in a simple concept we can call the ‘Budapest körut’ or ‘ring’. Public perception of the dangers of radioactivity drives Governments to finance scientists to make measurements of radioactivity, the scientists must justify the costs of measurements by the potential lives saved and this information on lives saved must be fed back to the public to ensure continued support. Like the Budapest körnt, this ring leads to bridges to other topics.

ACKNOWLEDGEMENTS

Our great thanks are due George Marx, Sándor Tarjan and János Szlávik, who contributed much time and effort to make our workshop function smoothly. We also acknowledge the financial support received from Gammadata Mätteknik (Sweden), Silena (Austria) and EG&G (Austria), who provided funds for additional people to attend the workshop. Further support was provided by Eötvös University, Budapest, Hungary, and Brunei University, London, UK.

CONTRIBUTORS

[1] G. MARX, Department of Atomic Physics, Eötvös University, Budapest, Hungary. [2] L. SAMUELSSON, Linköping University, Sweden. [3] J. THOMAS, National Institute for Public Health, Prague, Czech Republic. [4] NIKODEMOVA, Institute of Preventive and Clinical Medicine, Bratslava, Slovakia. [5] K.D. CLIFF, National Radiological Protection Board, Chilton, UK. [6] R. CZARWINSKI, Federal Office for Radiation Protection, Berlin, Germany. [7] I. NIKL, National Research Institute for Radiobiology and Radiohygiene, Budapest, Hungary. [8] C. MILU, Institute of Hygiene and Public Health, Bucarest, Romania. [9] E. TÓTH, Department of Atomic Physics, Eötvös University, Budapest, Hungary. [10] A. ROSS, Track Analysis Group, Department of Physics, Bristol University, UK. [11] S. NEWSTEAD, Her Majesty’s Inspectorate of Pollution, Lancaster, UK. [12] M. IVÓ, Environmental Control Institute, Water Laboratory, Baja, Hungary. [13] S. SONOC, Environmental Radiation Laboratory, Bucharest-Afumati, Romania. [14] S. TARJAN, Head of Radiological Section, Food Testing Institute, Budapest, Hungary. [15] Z. KORENOVÁ, Water Research Institute, Bratislava, Slovakia. [16] M. TATARA, Institute of Preventive and Clinical Medicine, Bratislava, Slovakia. [17] I. BUCINA, Institute of Public Health and Hygiene, Prague, Czech Republic. [18] B. KANYÁR, National Research Institute for Radiobiology and Radiohygiene, Budapest, Hungary. [19] I. MÓCSY, Institute of Public Health and Medical Resarch, Cluj, Romania. POSTER PRESENTATIONS 61

[20] L. SAJO BOHUS, Universidad Simon Bolivar, Caracas, Venezuela. [21] J.D. MARSHALL, Her Majesty’s Inspectorate of Pollution, Lancaster, UK. [22] G. PENCHEV, Institute of Legal Science, Sofia, Bulgaria. [23] J. SZLÁVIK, Head of Department of Environmental Economics, Technical University, Budapest, Hungary. [24] Z. KOBJAKOV, Department of Economics, Eötvös University, Budapest, Hungary. [25] A. MARIN, London School of Economics, UK. [26] L. PERSSON, Division of Nuclear Safety, IAEA, Vienna. [27] B.-M. DROTTZ-SJÖBERG, Centre for Risk Research, Stockholm, Sweden. [28] B.R. ORTON, Radiation Protection Officer, Brunei University, Uxbridge, UK. 6 2 POSTER PRESENTATIONS

IAEA-CN-54/45P

MID-TERM RADIOECOLOGICAL AND RADIOBIOLOGICAL CONSEQUENCES OF CHERNOBYL FALLOUT IN AN ALPINE ENVIRONMENT

H. LETTNER, W. HOFMANN, J. POHL-RÜLING, F. STEINHÄUSLER Institute of Physics and Biophysics, University of Salzburg, Salzburg, Austria

1. INTRODUCTION

Parts of the Eastern and Central Alps in Austria (provinces of Salzburg and Upper Austria) and in Germ any (southern Bavaria) are am ong the regions which received the highest surface deposition of radioactive fallout in western Europe due to the nuclear accident at Chernobyl in April 1986. The contamination with 137Cs varied between 10 and 80 kBq/m 2 at the time of deposition. The contam ination was very inhomogeneous, mostly because of different geographical situations and m eteorological conditions at the time of passage of the radioactive cloud. The m ost contam inated regions are situated at the northern flank of the A lp s and the northern parts o f the H ohe Tauern region because o f the higher precipitation rates at the w ind­ w ard position in these geographical regions. Also, the nuclide transfer factors and nuclide binding efficiency vary considerably for the different soil types. Though m ost of the agricultural land is characterized by low transfer factors, resulting in m inor contam ination of the food produced in these areas, in som e parts of the country at higher altitudes the bioavailability of radionuclides is m uch higher. A s these areas are intensively used for agricultural production during the sum m er, this can result in considerable contamination of local food products. Furthermore, whole body activity concentration of w orkers was also increased tem porarily because of the con­ sum ption of predom inantly local food products.

2 . M E T H O D S

Since 1986 continous measurements of the radionuclide contamination in environmental samples including soil, vegetation and food products have been carried out. For the assessm ent of nuclide contamination, the first survey of the Province of Salzburg w as made im m ediately after the fallout [1] by gam m a dose rate m easurements, followed by further detailed surveys in 1988 and 1993 [2, 3] using POSTER PRESENTATIONS 6 3 soil sample analysis, gam m a dose rate measurements and in situ gam m a spectro­ m etry. In addition to these m ethods, the use of biom onitors w as tested for determ ina­ tion of the amount of nuclide deposition [4]. For detection of the radiobiological effects, the structural aberrations in lym phocytes o f the peripheral blood were inves­ tigated in the period from 1987 to 1992 [5]. In 1987, blood sam ples from 16 persons from urban areas and in 1991-1992 from nine persons who live and w ork in Alpine areas with high transfer factors were analysed for chrom osom al aberrations.

3. PRESENT SITUATION

The ground contamination with the long lived nuclides 134C s and 137C s of between 10 and 80 kBq/m 2 (Fig. 1, date of reference to = 1 M ay 1986) from the Chernobyl fallout resulted — am ong other effects — in a significant increase of gam m a exposure with m axim um values of up to 3 piSv/h as com pared to 0.1 ¿¿Sv/h for the pre-Chernobyl conditions. Because of m igration into deeper soil layers and radioactive decay since 1986, the exposure has decreased significantly. In 1993 the rem aining contribution to the gam m a exposure which comes solely from caesium nuclides was determined to be 14.2 nSv/h per 10 kBq/m 2 of 137Cs, plus the alm ost negligible amount of 0.4 kBq/m 2 from 134Cs. In 1993, on average the rem aining increase of the environmental gam m a exposure was approximately 30% above the pre-1986 values. Contrary to this increase of environm ental radiation, the average nuclide contam ination of agricultural products was rather low, because of the low availability of I37C s for the local vegetation in the majority of the agricultural regions, m ainly at lower altitudes. How ever, in areas with high 137C s transfer fac­ tors according to specific soil and clim atological conditions, the locally produced food could still show significant levels of contam ination. Such an area is, for exam ­ ple, the Alpine region of the H ohe Tauern, used extensively for agricultural purposes during the sum m er. In 1993 the contam ination of food products from these areas was still up to 3 0 % o f the caesium levels m easured in the initial period im m ediately after the fallout deposition in 1986. The depth distribution of the caesium nuclides in intensively used soil from Alpine regions differs significantly from that in intensively used soil characteristic of flat land and valley floors. W hile the depth distribution in the latter soil category is characterized by m igration into deeper soil layers, w ith the m axim um activity below 4 cm , the depth distribution in Alpine soils show s typically m inor m igration into deeper soil layers (Fig. 2). In these soils the m axim um activity is in the upper 2-4 cm soil layer. Below this m axim um , the activity concentration decreases exponentially with depth, determined by a half-depth of around 1.5 cm. The difference of the soil depth distribution for the tw o soil categories corresponds to different transfer factors, w hich are up to 2 orders of m agnitude higher in Alpine soils than in soils from areas at low er elevations above sea level. This phenom enon is m ost pronounced in soils with underlying silicic bedrock [2]. In upland areas with 6 4 POSTER PRESENTATIONS

О 50 ЮО km

FIG. 1. Radioactive contamination in the Province of Salzburg and the adjacent parts of Upper Austria following the nuclear accident at Chernobyl. Date of reference: 1 May 1986.

underlying calcareous bedrock, the transfer factors are only slightly higher than those found in intensively used flat land. The high transfer factors are causally related to high contamination of the agricultural products of the fallout areas affected: A s an exam ple, the tem poral changes in the caesium contam ination of m ilk over the past years in an Alpine region are show n in Fig. 3. This agricultural produc­ tion region is situated 1600 m above sea level with a surface contam ination of about 60 kBq/m 2 137Cs in 1993. The contamination was still up to 200 Bq/kg in m ilk (which is above the official Austrian intervention level). The ‘half-life’ of the decrease of caesium contamination in m ilk amounts to 4.4 years, which is typical POSTER PRESENTATIONS 6 5

Soil depth (cm)

FIG. 2. Soil depth distribution of l37Cs in two different typical soil types. (1) Soil from intensively used agricultural land. (2) Soil from intensively used Alpine regions at altitudes above 1500 m.

Y e a r

FIG. 3. Average contamination in milk produced in contaminated Alpine region between 1986 and 1994 during the summer. For comparison, the radioactive decay curve of ,37Cs is shown (dotted line). 6 6 POSTER PRESENTATIONS for this region. A s the food consum ption of the w orkers consists to a large extent of local products, this resulted in a significant increase of the w hole body concentra­ tion. In 1991 and 1992 this population group was investigated by m easuring whole body contamination and by analysis of chrom osom al aberration. The whole body contamination varied over a range from 10 to 140 Bq/kg of 137Cs. The low dose levels attributed to these whole body contam inations are com parable to the dose levels found in the urban area of Salzburg in 1987, when chrom osom al aberration of 16 volunteers w as investigated. In the 1987 study, an increased aberration rate at dose levels of about 3 0% above the pre-Chernobyl environm ental dose level was found [3]. A t dose levels up to 6 0 % above the pre-Chernobyl dose level, the chrom o­ som e aberration rate appeared to decrease again. In two cases, with chrom osom e aberration data available also for the period prior to the fallout deposition for com parison, the number of dicentrics increased by a factor of 4 and the sum of chrom osom al aberrations by a factor of 7 in the year 1987. Subsequently the aberra­ tion rates decreased again with decreasing environm ental doses to a level in 1990 about twofold above the pre-Chernobyl value.

4. CONCLUSIONS

Because of specific soil and agricultural conditions in Alpine regions, high transfer factors in agricultural areas m ust be considered, resulting in long term con­ tam ination of the local food products, partially still exceeding national intervention levels 8 years after the initial fallout deposition. The resulting low dose irradiation found after the Chernobyl fallout con­ tributed to a significant increase of chrom osom al aberrations in peripheral blood cells above pre-Chernobyl levels.

REFERENCES

[1] STEINHÄUSLER, F., HOFMANN, W., DASCHIL, F., REUBEL, В., Chernobyl and its radiological and socio-economic consequences for the Province of Salzburg, Austria, Environ. Int. 14 (1988) 91. [2] LETTNER, H., “ Post-Chernobyl distribution of the 137Cs concentration in soil and environmental samples in montainous and plain areas of the Province of Salzburg, Austria” (Proc. Int. Symp. on Environmental Contamination Following a Major Nuclear Accident), Rep. No. LAEA-SM-306/60, IAEA, Vienna (1990) 193-203. [3] LETTNER, H., BOSSEW, P., HUBMER, A.K., Kontamination durch radioaktiven Fallout im Bundesland Salzburg und angrenzenden Teilen Oberösterreichs, Report of the Austrian Federal Ministry of the Environment, Vienna (1994). POSTER PRESENTATIONS 6 7

HOFMANN, W., ATTARPOUR, N., LETTNER, H., TÜRK, R., Cesium-137 con­ centrations in lichens before and after the Chernobyl accident, Health Phys. 64 (1993) 70. POHL-RÜLING, J., HAAS, O., BROGGER, A., OBE, G., LETTNER, H., DASCHIL, F., ATZMÜLLER, C., LLOYD, D., KUBIAK, R., NATARAJAN, A.T., The effect of lymphocyte chromosomes of additional radiation burden due to fallout in Salzburg/Austria from the Chernobyl accident, Mutat. Res. 262 (1991) 209. 6 8 POSTER PRESENTATIONS

IAEA-CN-54/49P

APPORT DE NOUVEAUX DOSIMETRES ELECTRONIQUES POUR LA DOSIM ETRIE OPERATIONNELLE EN M EDECINE NUCLEAIRE

B. AUBERT, A. LAM O N Service de physique, Institut Gustave-Roussy

C. PARMENTIER Service de médecine nucléaire, Institut Gustave-Roussy

Villejuif, France

1. INTRODUCTION

L ’optim isation de la radioprotection des travailleurs passe tout d ’abord par une connaissance aussi précise que possible des expositions en fonction des actes ou des tâches effectués. Pour cela, une analyse détaillée des postes et des conditions de travail est indispensable. Pour m ener à bien cette analyse, il convient de procéder, en complément de la dosimétrie individuelle réglementaire, à une dosimétrie individuelle opérationnelle afin de connaître de manière précise et détaillée les niveaux d ’exposition auxquels le travailleur est soum is. Jusqu’à présent, les dosim ètres disponibles pour une telle dosim étrie (stylo- dosim ètre et dosim ètre individuel à base de com pteur G eiger-M üller ou de détecteur solide) présentaient des caractéristiques telles que la sensibilité, la fiabilité, la fragi­ lité, le coût, la consom m ation, souvent peu adaptées à une utilisation en m ilieu hospitalier. G râce aux récents progrès en m atière de m iniaturisation et d ’intégration, de nouveaux dosim ètres électroniques ont fait leur apparition avec des caractéris­ tiques techniques qui les rendent tout à fait attractifs pour ce type d ’utilisation [1]. N ous avons évalué ces dosim ètres d ’un point de vue physique et dans le cadre des différents postes de travail concernant les activités diagnostiques et thérapeu­ tiques en m édecine nucléaire.

2. MATERIEL ET METHODE

Les dosim ètres utilisés1, qui contiennent un petit détecteur au silicium associé à une électronique intégrée, se présentent sous un format réduit (dim ension d ’une

1 Dosimètre Dosicard développé par la société Nomatek. POSTER PRESENTATIONS 6 9 carte de crédit). Ils bénéficient de toutes les fonctionnalités attendues d’un tel appareil pour l’application envisagée (lecture directe, dose, débit de dose, alarm es sonore et visuelle). D e plus, l’électronique intégrée permet la gestion des inform a­ tions dosim étriques (intégration de la dose par m ois, par an, etc.) et les échanges avec un systèm e d ’exploitation externe. L ’étude physique a été effectuée à l’aide de sources de technétium 99m, d’iode 131 et de césium 137 afin de couvrir un large dom aine en énergie (de 140 à 662 keV). Nous nous som m es plus particulièrement intéressés à:

— la distribution des réponses des 10 dosim ètres dont nous disposions, — la réponse en fonction de l’énergie, — la reproductibilité à court terme (quelques heures) et long terme (quelques j o u r s ) , — l’influence de l’activité (ou du débit), — l’influence de m ilieu diffusant derrière le dosimètre.

Cette étude a pu être menée grâce à une fonctionnalité de l’électronique associée; il s’agissait de la possibilité d ’obtenir le résultat des 100 dernières m esures de dose effectuées selon un temps de comptage program m able de 6 à 250 min. Les postes analysés concernaient la préparation, le fractionnem ent et l’injec­ tion des radiopharm aceutiques, la réalisation des exam ens scintigraphiques, l’adm i­ nistration des activités thérapeutiques d ’iode 131 et les soins aux patients hospitalisés pour radiothérapie métabolique.

TABLEAU I. REPONSE (MOYENNE ET ECART- TYPE) DES DOSIMETRES EN FONCTION DE L’ENERGIE

Débit mesuré en /nSv/h pour 1 GBq à 1 m Radionucléide

Moyenne Ecart-type

Technétium 99m 17,9 1,8 Iode 131 74,9 2,6

Césium 137 105,2 3,5 7 0 POSTER PRESENTATIONS

3. RESULTATS

3.1. Etude physique

Le tableau I présente pour les trois radionucléides sélectionnés la valeur m oyenne de la réponse des 10 dosim ètres et Fécart-type associé. Les résultats ont été normalisés pour une source de 1 G Bq située à 1 mètre, les sources et les dosim ètres étant placés dans l’air. La reproductibilité à court terme a été testée en exposant les 10 dosim ètres pendant 6 h à une source de technétium 99m , d ’activité initiale 58 G Bq. L ’évolution de la m esure du débit de dose, par tranche de 6 m in, a perm is d ’ajuster la variation des m esures par une exponentielle et de la com parer à la période réelle. La m oyenne des périodes estimées est de 362 m in avec un écart-type de 7 m in (période réelle 3 6 0 m i n ) . La linéarité en fonction du débit a été testée au m oyen d ’une source de tech­ nétium 99m d ’activité initiale 22,4 G Bq, pendant 38 h soit plus de 6 périodes (varia­ tion du débit d ’un facteur 80). La figure 1 présente la variation de la réponse de l’un des dosim ètres et l’estimation de la période par ajustement d ’une exponentielle. Chaque point de m esure représentant l’intégration sur 10 m in, on observe pour les tem ps élevés des fluctuations liées aux faibles valeurs de débit m esurées ( < 10 piSv).

Tem ps (m in)

FIG. 1. Variation en fonction du temps, c ’est-à-dire en fonction du débit, de la réponse d ’un dosimètre à une source de technétium 99m. POSTER PRESENTATIONS 71

TABLEAU II. DOSE EQUIVALENTE DELIVREE AU PERSONNEL POUR LES POSTES LES PLUS EXPOSES DANS LES APPLICATIONS DE MEDECINE NUCLEAIRE

H mesurée Gamme de valeurs H estimée Poste OxSv/j) OiSv/j) (mSv/an)

Injection scintigraphie osseuse 39,7 12,8-109 8,7

Injection scintigraphie thyroïdienne 4,4 * 0 -1 4 ,3 1,0 Manipulateur médecine nucléaire 5,0 0 ,9 -8 ,1 1,1 Technicien laboratoire chaud 7,0 = 0 -6 1 ,0 1,5

Infirmière hospitalisation 5,2 = 0 -4 7 ,8 1Д

3.2. Etude des postes de travail

Le tableau II présente, pour un certain nombre de postes couvrant les différentes applications diagnostiques et thérapeutiques, la valeur m oyenne de la dose équivalente par jour de travail, la gam m e de ces valeurs sur la période d ’obser­ vation (de 3 à 9 sem aines) et la dose équivalente annuelle extrapolée sur la base de 220 jours de travail par an. Ces dosim ètres permettent égalem ent de connaître la valeur de dépassem ent de seuils, en dose ou débit de dose, préalablement choisis, ainsi que la durée de ces dépassements. Il est à noter que seuls les postes d’injection pour scintigraphie osseuse et de préparation des radiopharm aceutiques (technicien au laboratoire chaud) ont présenté des dépassem ents du seuil de débit réglé à 25 /xSv/h. Les valeurs m axi­ m ales relevées étaient respectivem ent de 1416 /xSv/h (pour une durée de 8 s) et de 6 2 4 fxSv/h (pour une durée de 125 s).

4. DISCUSSION ET CONCLUSION

L ’évaluation des nouveaux dosimètres électroniques à laquelle nous avons procédé en m ilieu hospitalier, et plus particulièrem ent dans le cadre de l’utilisation des rayonnem ents ionisants en médecine nucléaire, est à ce jour largement satis­ faisante. En effet, l’étude physique a montré une excellente sensibilité (= 1 /¿Sv) largem ent supérieure à celle des film s dosim ètres (> 2 0 /xSv), une reproductibilité à court et m oyen terme et une réponse en fonction du débit tout à fait com patible avec une application de radioprotection. Leur facilité d ’utilisation ainsi que leur 7 2 POSTER PRESENTATIONS

robustesse ont égalem ent été appréciées. La dispersion des réponses des différents dosimètres, de 3 à 10% selon l’énergie, est également acceptable pour la radioprotection. Parm i les résultats concernant l’exposition du personnel, on peut relever que le poste le plus exposé, c ’est-à-dire celui des injections pour scintigraphie osseuse (« 8 0 0 MBq/patient), conduit à une exposition annuelle proche de 10 m Sv, soit la moitié des dernières limites recommandées par la Com m ission internationale de protection radiologique (CIPR) [2]. Ce facteur 2 ne constitue pas une garantie suffisante de non-dépassem ent des limites, aussi l’analyse fine que nous permet ce dosim ètre sera très utile pour étudier les m oyens et les m éthodes pour réduire l’expo­ sition du personnel en axant nos efforts sur les phases les plus critiques. La lecture directe de la dose ou du débit de dose est égalem ent un point très im portant pour la form ation et l’inform ation du personnel. En conclusion, ces dosim ètres électroniques offrent un potentiel tout à fait intéressant pour une pratique de la radioprotection visant à appliquer au m ieux les recom m andations de la C IP R et en particulier le principe d ’optimisation.

REFERENCES

[1] LACOSTE, F., LUCAS, M., Le système Dosicard, Radioprotection, 28 (1993) 77-82. [2] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION (ICRP), 1990 Recommendations of the International Commission on Radiological Protection (ICRP Publication 60, Oxford, Pergamon Press (1991)). POSTER PRESENTATIONS 7 3

IAEA-CN-54/52P

LA PROTECTION DES INTERVENANTS DE LA MAINTENANCE DES CENTRALES NUCLEAIRES FRANÇAISES

R . D O L L O Exploitation du parc nucléaire, Electricité de France, Paris, France

INTRODUCTION

A ux 18 ООО agents d ’Electricité de France (E D F ) dédiés à l’exploitation et à la maintenance des centrales nucléaires s’ajoutent environ 26 000 intervenants extérieurs affectés uniquem ent aux travaux de maintenance en période d ’arrêt pour rechargem ent de combustible. Chaque année, les arrêts de tranche représentent quelque 14 millions d ’heures de travail sur un temps très court (en dehors de l’hiver, période de forte consom mation). Depuis le début des années 90, E D F a engagé une politique de partenariat avec les entreprises prestataires pour progresser dans la qualité des interventions et répondre aux critères de professionnalism e, de sûreté et de sécurité. La politique E D F de réduction de dose se traduit en particulier par l’am élio­ ration de la qualité du suivi dosim étrique de tous les intervenants E D F ou extérieurs qui doivent bénéficier du m êm e niveau de protection. Plusieurs instrum ents ont été m is en place qui concernent: la form ation des intervenants, l’attribution d ’un carnet d ’accès aux installations et l’am élioration des m esures de suivi dosim étrique.

FORMATION DES INTERVENANTS

E D F a engagé un vaste program m e de sensibilisation et de form ation «As low as reasonably achievable» ou A L A R A de tous les acteurs. Il s’agit là d ’une prem ière étape en vue de m odifier la culture des acteurs de l’électronucléaire. La qualité de form ation est essentielle. En 1990 a été créé le Com ité français de certification des entreprises pour la form ation et le suivi dosim étrique du person­ nel travaillant sous rayonnem ents ionisants (C EFR I). Le C E F R I a été créé en plein accord avec les exploitants (E D F , la Com pagnie générale des m atières nucléaires ou Cogém a, le Com m issariat à l’énergie atomique ou C EA , les Arm ées), le Ministère de la santé, celui du travail et de l’industrie, et les autorités en radioprotection. Cet organism e délivre des agrém ents après audit: 7 4 POSTER PRESENTATIONS

— aux organism es de formation, — aux entreprises de travail temporaire mettant à disposition du personnel, — aux entreprises employant du personnel travaillant dans les installations nucléaires.

ATTRIBUTION D ’UN CARNET D ’ACCES

Pour contrôler efficacement les exigences vis-à-vis des intervenants, E D F a m is en place un carnet d ’accès rendu obligatoire pour intervenir sur ses installations. Ce carnet com porte, outre l’identification de la personne, ses qualifications professionnelles, et sa form ation en m atière de sûreté, de sécurité et de radioprotec­ tion. Enfin, un volet est consacré à la dosim étrie opérationnelle. Ce carnet permet également d ’inclure la carte individuelle de suivi médical délivrée par le Service central de protection contre les rayonnements ionisants (SC PR I) et gérée par le m édecin du travail de l’employeur.

AM ELIORATION DES MESURES DE SUIVI DOSIMETRIQUE

Pour optim iser les résultats en radioprotection, E D F a développé depuis une quinzaine d ’années une dosim étrie opérationnelle comprenant:

— des dosim ètres individuels à lecture directe et instantanée, — une gestion locale des doses (par centrale nucléaire).

Chaque intervenant a connaissance en temps réel de la dose reçue lors d ’une intervention donnée sur un site donné et peut agir en conséquence (anom alie, respect des prévisions).

D o s i n a t :

Pour connaître les doses cum ulées sur plusieurs périodes et plusieurs sites, en 1992, E D F a m is en place Dosinat (Systèm e de dosim étrie nationale), systèm e infor­ m atique de consolidation nationale par rapprochem ent des fichiers locaux. Il perm et de suivre en tem ps réel la dosim étrie opérationnelle de tous les inter­ venants (agents E D F , prestataires ou personnel d ’entreprises de travail tem poraire) dans les centrales nucléaires qui sont toutes connectées. Il contient actuellement les données de 47 000 personnes, dont 18 000 agents E D F et 29 000 prestataires appartenant à 1200 entreprises extérieures. POSTER PRESENTATIONS 7 5

140 г

120 55 (D g 100 z 80 o 3 c r

ъ 60 о 45 (D g- ф —t и - Dose collective EDF + intervenants extérieurs (D- о 40 Otu О (D -Objectifs C 20 - Nombre de réacteurs

35 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995

FIG. 1. Dose collective et nombre de réacteurs 1986-1995.

D o s i m o :

Les difficultés rencontrées dans le suivi des travailleurs m obiles ont conforté l’opportunité d ’un suivi ne se lim itant pas aux seules centrales nucléaires (Rapport Office parlementaire — M . B IR R A U X , Directive Euratom 90/644). Après accord de la Com m ission nationale inform atique et libertés, E D F et les autres exploitants français du cycle nucléaire (C E A , Cogém a, les Arm ées) ont décidé la m ise en œ uvre d ’un systèm e de collecte élargi à tous leurs sites, en collaboration avec le Groupe intersyndical de l’industrie nucléaire (G IIN ). Dès 1995, Dosim o (Systèm e de dosim étrie opérationnelle) permettra de connaître le cum ul des doses reçues chez chacun des exploitants.

BILAN

On note un infléchissement de la tendance en 1992-1993 (Fig. 1). Entre 1983 et 1989, la dose m oyenne par tranche et par année fluctuait entre 1,8 et 2 h-Sv. En 1990, elle fut de 2,35 h -Sv et, en 1991, de 2,44 h-Sv, avec notam ­ ment un nom bre important de visites décennales. La prom otion d’A L A R A à partir de 1992 a perm is de com mencer à inverser cette tendance: 2,36 h-Sv en 1992 et 2,04 h-Sv en 1993. La dosimétrie a baissé de près de 15% en 1993. Cette dim inution concerne essentiellem ent les intervenants extérieurs. L ’objectif qu’E D F s’est fixé, de 1,6 h-Sv en 1995, reste toutefois très a m b i t i e u x . 7 6 POSTER PRESENTATIONS

IAEA-CN-54/66P

OCCUPATIONAL EXPOSURE IN JAPAN FRO M 1976 TO 1992

T. YAMAGUCHI, K. KAW AI Departm ent of Health Physics, Japan Atom ic Energy Research Institute (JAERI), Tokai-mura, Naka-gun, Ibaraki, Japan

1. INTRODUCTION

This paper describes the present status of occupational radiation exposure of m onitored w orkers in the categories of nuclear pow er generation, m edicine, indus­ try, industrial radiography, and research and education by review of the dose equiva­ lents for the period from 1976 to 1992, and the impact for the future introduction of new dose lim its based on the IC R P 1990 recommendation.

2. ANNUAL MEAN DOSE AND COLLECTIVE DOSE

Figure 1 show s the trends in the num ber of m onitored radiation w orkers in the occupational categories. The number of radiation workers and annual dose data related to nuclear power generation are m ostly obtained from the Annual Report of the Japan Nuclear Safety Com m ission (JN SC ) [1]. In the rem aining categories, about 9 0 % of these data is provided by statistical data from film badge service com panies [ 2 , 3 ] . Figure 2 shows trends in the collective dose and mean annual dose in these categories. The results of investigation of occupational exposures for about 330 000 radiation w orkers in 1992 show an individual mean dose of 0.4 m Sv and an annual collective dose of 135 m an-Sv. The num ber of radiation w orkers has increased at an average annual rate of 7 % , and the num ber of nuclear pow er plants and the num ­ ber of facilities using sources of ionizing radiation have increased steadily [4]. N ot­ w ithstanding such increases in the num ber of w orkers and facilities, the annual m ean dose in 1992 decreased to 34% of the highest in 1978, and the collective dose decreased to 6 7% of the highest in 1981. The annual mean dose in nuclear power generation — especially for boiling water reactors (B W Rs) — reached a peak in 1978 but has decreased subsequently. N u m b e r of w o r k e r s 0 1 x 3 0 1 x 2 x10b 0 1 x 1 5 0 1 x 4 I. . Trends theFIG.innumber 1. radiationof workers by category industry.of 1976 Nuclear power generation o i t a r e n e g e r n i e c i w d o e p M r a □ e l c u y N h ■ p a r g o i d a r l a i r t s u d n I □ y r t s u d n II Research and education n o i t a c u d e d n a h c r a e s e I R ONS N IO T A T N E S E R P R E T S O P 2 9 9 1 7 7 7 8 POSTER PRESENTATIONS

у И ' "j*-S-B -B -B —B-B--Эг:-^--В—В--В

1976 1980 1984 1988 1992

Y e a r

FIG. 2. Trends in the mean annual dose and collective dose by category of industry. POSTER PRESENTATIONS 7 9

This reduction im plies that effective dose reduction m easures such as the follow ing have been implemented [5-7]:

(a) quality assurance of fuel and materials (adaptation of corrosion resistant m aterials, low cobalt content m aterials, etc.); (b) feedwater quality control (oxygen control, Ni/Fe control, purification of con­ densed water, etc.); (c) development of automatic devices (C R D remote exchangers, fuel assembly exchangers, decontam ination m achines, etc.); (d) reinforcem ent of radiation shielding at the workplace (on the surface of pum ps and pipes in the reactor containm ent vessel, etc.); (e) education and promotion of the A L A R A (as low as reasonably achievable) philosophy through radiation w ork, etc.

In industrial radiography, the annual mean dose has been decreased by stan­ dardization of gam m a irradiation devices, regulation of education for w orkers and institution of a system for reporting exposure dose to competent authorities [8]. The annual m ean dose in the other categories has decreased gradually. The decrease in the rate of collective dose in nuclear pow er generation is the largest in all categories. The collective doses in the categories of industrial radiography and research and edu­ cation also show a trend towards decrease. How ever, the collective dose in industry does not show a distinct decrease. For the category of m edicine, the collective dose has been on the increase from 1989. Recent increasing use of X ray diagnostic equip­ m ent and increase in the num ber o f patients m ay contribute to this rise. A n d a change in 1989 of the dose evaluation method, based on revised regulations reflecting IC R P -26, for w orkers w earing a lead apron has possibly influenced this rise [9-11]. The num ber (with percentage) of m onitored w orkers exceeding the dose lim it of 50 m Sv was 26 (0.008%) in all categories in 1992. M any of these workers were in the category of medicine. The number of radiation workers exceeding 20 m Sv decreased from 1145 (0.62%) in 1980 to 296 (0.09%) in 1992.

3. DOSE DISTRIBUTION OF W ORKERS

The hybrid log normal (H LN ) model [12], developed at the Japan Atom ic Energy Research Institute, w as applied to interpret the dose distributions of w orkers subject to radiation protection adm inistration. W e used the H L N m odel to investigate the characteristics o f individual dose distributions by plotting the individual doses on hybrid scale probability paper. Figure 3 show s individual dose distributions in each category in 1992. In nuclear pow er generation, the plotted dose points fit the H L N m odel line. This m eans that the exposure to w orkers w as appropriately controlled under the dose level of 15-30 m Sv, and there were no workers exceeding 30 m Sv in 1992. In other categories, the plotted dose points gradually deviate from the H L N 8 0 POSTER PRESENTATIONS m odel line to the higher dose side above the range of 20 m Sv. T his suggests that there are w orkers exposed to higher radiation risk w ho m ay be under w eak adm inistrative control for radiation protection. But the num ber of such high radiation risk w orkers is sm all and they seem to be special technicians. M ost of them are in the category of medicine.

Annual dose (mSv)

FIG. 3. Individual dose distributions in 1992 by category of industry. POSTER PRESENTATIONS 81

TABLE I. NUMBER OF RADIATION W ORKERS W ITH DOSES EXCEED IN G 20 mSv/a IN 1990 AN D 1992

Number of radiation workers exceeding 20 mSv/a Categories

1990 1992

Nuclear power generation 212 20 Industry 39 23 Industrial radiography 30 17

Medicine 296 234 Research and education 6 2

All categories 583 (0.2% )a 296 (0.09%)a a Percentage of radiation workers exceeding 20 mSv/a with respect to total workers in all categories.

4. CONSIDERATION OF A NEW DOSE LIMIT

The 1990 IC R P recommended annual dose limit is 20 m Sv averaged over 5 years. Further dose reduction was considered in com pliance with this recom m en­ dation in each category, which resulted in substantial decrease of the num ber of workers exceeding 20 m Sv in 1992 com pared with the num ber of workers in 1990, as shown in Table I. A long term effort to reduce personal exposure was implemented in the category of nuclear pow er generation, and efficient reduction of the num ber of w ork­ ers incurring high exposure was achieved as m entioned above. Com pliance with the dose limit of 20 m Sv averaged over 5 years m ay be possible in this field. O n the other hand, it w ill be difficult to reduce the num ber of specialists incurring high exposure in the m edical field. T o reduce such exposures, e.g. to m edical doctors per­ form ing heart catheterization and angiography, it m ay be necessary to increase the num ber of specialists, im prove and periodically maintain X ray diagnostic equip­ ment, and reinforce the shielding around devices. 8 2 POSTER PRESENTATIONS

REFERENCES

[1] JA P A N N U C L E A R S A F E T Y C O M M IS S IO N , Annual Report of Nuclear Safety (1979-1992). [2] C H I YO D A S A F E T Y A P P L IA N C E C O . Ltd, Film Badge News (1976-1992). [3] N A G A S E L A N D A U E R Ltd, Nagase Film Badge News (1976-1992). [4] JA P A N R A D IO IS O T O P E A S S O C IA T IO N , Statistics on the Use of Radiation in Japan 1993 (1993). [5] O H B A , T., et al., Reduction of radiation exposure and radioactive wastes on Kashiwazaki Kariwa Unit 1, Thermal Nucl. Power 39 8 (1988). [6] SASAKI, F., Recent decreasing trend of personnel exposure in nuclear power plants in Japan and analysis of major factors contributing its reduction, Thermal Nucl. Power 42 2 (1991). [7] S U T O H , Y ., et al., “ Operation experiences of Onagawa Unit 1 for reduction of radia­ tion sources” , Japan Atomic Industrial Forum (Int. Conf. on Water Chemistry in Nuclear Power Plants, 1988) 255-260. [8] T A IG A , H ., Control of radioisotopes in industrial radiography, Isotope News, No. 447 (1991). [9] M A R U Y A M A , T ., Collective Effective Dose Equivalents from Medical and Occupa­ tional Exposure in Japan, NIRS-M-82 (1992) 75-84. [10] K U S A N O , S., Digital radiography — Present status and future expansion, J. Natl. Def. Med. Coll. 18 49 (1993) 292-301. [11] K IK U C H I, T ., “ Radiation protection of worker—assurance of 20 mSv” , Radiological Protection in Medicine News Letter No. 4 (1992) 12-17. [12] K U M A Z A W A ,S ., et al., W hy do we need dose distribution model? Radiation Protec­ tion Dosimetry 36 (1991) 269-273. POSTER PRESENTATIONS 8 3

IAEA-CN-54/71P

A STUDY OF THE RADIOECOLOGICAL SITUATION IN REGIONS ADJOINING THE 30 km ZONE AROUND THE CHERNOBYL NUCLEAR POW ER PLANT

M.D. BO NDAR’KOV, I.N. VISHNEVSKIJ, N.P. DONETS, V.A. ZHELTONOZHSKD, A.A. SOTNIKOV Nuclear Research Institute, Ukrainian Academ y of Sciences, Kiev, Ukraine

A s a result of the accident at the Chernobyl nuclear pow er plant, a considerable portion of Ukrainian territory was radioactively contaminated. This included the Chernigov Region, which was the subject of our m ost extensive survey. This region aroused particular interest because of the patchy nature of its contam ination. A survey was carried out in the Priluki, Bakhm ach, Borzna, Varva, Srebnoe, Ichnya and Talalaevka districts (southern districts), as well as a follow -up survey in the Chernigov, Kozelsk, Koryukovka and Semenovka districts of the Chernigov Region directly adjoining the 30 km zone. The radioecological investigations m ainly i n v o l v e d :

(1) m easurem ent o f the exposure rate in the territory of settlements in the districts s t u d i e d ; (2) measurement of the exposure rate and beta contamination on the prem ises of public buildings in settlements in the districts studied; (3) soil sam pling and laboratory testing of specim ens for radionuclide content; (4) m easurem ent of radiation burdens in a num ber of the settlements located within the strictly controlled area.

The content of the gam m a em itting radionuclides 137Cs, 134Cs, 144Ce, 106Ru, 40K , 228Ra and 226Ra as well as the 90Sr concentration of the soil samples was determined. For each settlement in the districts studied, the m inim um , m axim um , average and 9 0 % quantile values of the density of surface contamination by the long lived radionuclides 137C s and 90Sr were calculated and entered on maps. The measurements made in the southern districts showed that the average 137C s contam ination w as 0.1-0.1 Ci/km 2. Statistical analysis of all the data showed that in these districts, as a rule, A (90Sr) = (0.4-0.8) A (137Cs). This ratio roughly corresponds to the global fallout level. Such a level of con­ tam ination corresponds to that of ‘clean’ districts, so that it m ay be concluded that in these localities an effect of the Chernobyl accident is barely discernible. 8 4 POSTER PRESENTATIONS

TABLE I. GENERALIZED RESULTS OF MEASUREM ENTS OF THE DENSITY OF 137Cs AND 90Sr SOIL SURFACE CONTAMINATION (Ci/km2) IN THE CHERNIGOV REGION

No. of Minimum Maximum Average 90% quantile Region samples value of ,37Cs value of 137Cs value of ,37Cs value

Southern districts

Bakhmach 419 0.02 0.47 0.09 0.15 Borzna 349 0.02 0.69 0.11 0.19 Varva 236 0.02 0.30 0.06 0.09 Ichnya 459 0.01 0.50 0.06 0.11 Priluki 683 0.01 0.54 0.07 0.13 Srebnoe 233 0.02 0.19 0.07 0.14 Talalaevka 230 0.01 0.17 0.05 0.07

Northern districts

Kozelsk 2007 0.01 13.90 0.51 0.96 Koryukovka 1738 0.01 21.80 1.32 5.20 Semenovka 1661 0.02 10.50 1.48 3.78 Chernigov 2100 0.01 46.60 2.05 4.80

No. of Minimum Maximum Average 90% quantile Region samples value of 90Sr value of 90Sr value of 90Sr value

Southern districts

Bakhmach 419 0.01 0.07 0.03 0.04 Borzna 349 0.01 0.18 0.03 0.05 Varva 236 0.01 0.08 0.03 0.05 Ichnya 459 0.01 0.25 0.04 0.06 Priluki 683 0.01 0.28 0.04 0.07 Srebnoe 233 0.01 0.10 0.04 0.07 Talalaevka 230 0.01 0.06 0.02 0.04

Northern districts

Kozelsk 870 0.01 1.47 0.51 0.24 Koryukovka 531 0.01 0.96 0.07 0.11 Semenovka 686 0.01 1.12 0.09 0.19 Chernigov 1027 0.01 8.04 0.20 0.45 POSTER PRESENTATIONS 8 5

Overall analysis of the data obtained in the Chernigov, Kozelsk, Koryukovka and Semenovka districts shows that the contamination in these districts has certain untypical characteristics. First, in many places 90Sr activity is commensurate with 137Cs activity, so that in classifying the settlements it has to be borne in mind that the norms for 90Sr are lower by a factor of 3-5 than those for 137Cs. Thus, the settlements where 137Cs contamination is 0.4-0.6 Ci/km 2 can be listed with the settlements belonging to the fourth category for ^ Sr activity. Second, in a number of villages the systematic presence of 144Ce was discovered. Investigations carried out on samples from the 30 and 60 km zones around the Chernobyl nuclear power plant show that the presence of 144Ce indicates the existence of transuranic isotopes. If estimates are performed up to the present time, taking into account that 241Am accumulation takes place according to the chain 241Pu(Ti/2 = 13 years) — 241 Am ( T j /2 = 433 years), it is found that in some samples the activity of transuranic elements can be of the order of 0.04 Ci/km 2. The gamma spectra in certain samples also indicate clearly the presence of ‘hot particles’ (minute particles from fuel ele­ ments). This conclusion was reached by com paring these gamma spectra with spectra obtained from samples taken in the 30 km zone. It is also noteworthy that wide scatter of the 90Sr activity level (from 0.05 to 0.5 of 137Cs activity) is to be observed in these presumed ‘hot particles’. This clearly indicates that a proportion of the hot particles has already been dissolved by organic acids and that the active migration of transuranic isotopes is possible in this locality. In order to eliminate any possible systematic errors, frequency distributions for the m ain settlements and entire districts were determined with a personal computer. These show reasonable agreement in the shape of the distribution between 137C s and 90Sr (see Table I). This inspires a measure of confidence in the reliability of the data obtained. 8 6 POSTER PRESENTATIONS

IAEA-CN-54/72P

EVALUATION OF RADIATION BURDENS ON THE BASIS OF SPECTROMETRIC AND RADIOM ETRIC DATA

M.D. BO NDAR’KOV, I.N. VISHNEVSKIJ, N.P. DONETS, V.A. ZHELTONOZHSKD, A.A. SOTNIKOV Nuclear Research Institute, Ukrainian Academ y of Sciences, Kiev, Ukraine

A s a result of the accident at the Chernobyl nuclear pow er plant, a large part of Ukraine was radioactively contaminated. The spectrometric, radiometric and radiochem ical m ethods now available m ake it possible to determine with reasonable accuracy the density of soil surface contam ination w ith radionuclides for the purpose of evaluating the radiation situation in the affected areas. The im pact of contam ina­ tion of a region on the hum an body can be determined by evaluating the radiation burdens on the basis of the experim ental data for the density o f soil surface contam i­ nation with radionuclides. O ver the past few years, the Nuclear Research Institute of the Ukrainian Academ y of Sciences has obtained a large am ount of data on the density of surface contam ination with radionuclides in districts of the Chernigov Region as a result of the Chernobyl accident. The density of surface contam ination with gam m a emitting radionuclides was determined by means of gam m a spectrometry, while the density of 90Sr contamination was determined by radiochemical analysis. The radiation burdens of the populations of districts of the Chernigov Region were evaluated from the average densities of surface contam ination for each district surveyed. Table I shows the average surface contamination densities for the districts surveyed and the evaluated external radiation doses. For com parison, the external doses due to 137C s and 40K are given separately. The internal dose can be deter­ m ined on the basis of the concept of biological rem oval of radionuclides from the hum an body, using a single compartment model. A s is well known, the radionuclide accumulation factor of a critical organ depends on the radionuclide’s effective biological half-life and the proportion of its intake into the organ concerned. In its m ovem ent from one part of the biosystem to another, radiocaesium behaves like potassium. Caesium is completely absorbed by the gastrointestinal tract and the absorption coefficient for it is 1. A feature of caesium behaviour is its fairly even distribution throughout the body. A s a result, 137C s irradiates all the critical organs, f2 = 1. POSTER PRESENTATIONS 8 7

TABLE I. RESULTS OF DETERMINING RADIONUCLIDE CONTAMINATION DENSITY AND THE DOSE EQUIVALENTS DUE TO EXTERNAL EXPOSURE FOR THE DISTRICTS OF THE CHERNIGOV REGION SURVEYED

,37Cs 90Sr 40K DCX' 137Cs Dex. 4°K D istrict (av.) (av.) (av.) (m rem /a) (m rem /a) (C i/km 2) (C i/km 2) GiCi/kg)

Bakhmach 0.09 0.03 11.0 1.2 13.2 Borzna 0.11 0.03 11.1 1.4 13.3 V arva 0.06 0.03 10.0 0.8 12.0 Ichnya 0.06 0.04 10.4 0.8 12.4 K ozelsk 0.51 0.11 11.8 6.6 14.1 K oryukovka 1.32 0.07 11.4 17.1 13.7 Priluki 0.07 0.04 12.5 1.0 15.0 Sem enovka 1.48 0.09 9.2 19.3 11.0 Srebnoe 0.07 0.04 12.0 1.0 14.5 Talalaevka 0.03 0.02 11.8 0.6 14.2 C hernigov 2.05 0.18 8.2 24.6 9.9

137Cs metabolism in the human body can be described by the sum of two exponents with an effective biological half-life of T[ = 2 days and T 2 = 110 days (for an adult). Strontium belongs to the alkaline earth elements, and is m ainly deposited in bone tissue. The absorption coefficient into the blood after its entry into the gastro­ intestinal tract is 0.3, and the proportion o f its content in bone tissue in relation to the overall quantity in the body is f = 0.97. The effective biological half-life of 90Sr is 18 years (T 1/2 = 28 years), so that in the event of chronic 90Sr intake by the hum an body no equilibrium is established and its concentration in the bone tissues increases. The consequence of 90Sr incorporation in hum an bone tissue is irradia­ tion of critical organs such as red bone m arrow and bone surface tissue. Red bone m arrow receives 2 0 % of the total radiation dose of the bone tissues, and the bone surface tissues 24% . In order to take account of the contribution of the radiation doses of the red bone m arrow and bone surface tissues to the total effective dose equivalent in man, weighting coefficients of 0.12 and 0.3 respectively are introduced. 8 8 POSTER PRESENTATIONS

TABLE II. RESULTS OF DETERMINING RADIONUCLIDE CONTAMINATION DENSITY AND THE DOSE EQUIVALENTS DUE TO INTERNAL EXPOSURE FOR THE DISTRICTS OF THE CHERNIGOV REGION SURVEYED

Di« for Dint for D int for nu eff a D eff for D istrict 137Cs 40K 90Sr (total) 4°K (m rem /a) (m rem /a) (m rem /a) (m rem /a) (m rem /a)

Bakhmach 2.24 19.9 1.02 3.6 33.1 Borzna 2.69 20.1 1.28 4.3 33.4 V arva 1.45 18.2 1.27 2.4 30.2 Ichnya 1.54 18.8 1.40 2.6 31.2 Kozelsk 12.60 21.3 4.20 19.5 35.4 Koryukovka 32.90 20.7 2.30 50.1 34.4 Priluki 1.82 22.7 1.61 3.3 37.7 Sem enovka 37.00 16.6 3.20 56.5 27.6 Srebnoe 1.86 21.9 1.53 3.1 36.4 Talalaevka 1.14 21.4 0.82 1.9 35.6 Chernigov 47.20 14.9 6.32 72.4 24.8

a Total effective dose equivalent due to exposure from radionuclides of ‘Chernobyl’ origin.

The results of evaluation of the radiation burdens of the adult population for 1992 are given in Table II. For com parison, the internal radiation doses and the total effective dose equivalents are given separately for radionuclides of ‘Chernobyl’ origin and for 40K . It is to be noted that the total effective dose equivalent of the population from ‘Chernobyl’ radionuclides is m ade up of the external radiation dose ( — 30% ), the internal radiation dose from 90Sr (~ 10% ) and the internal radiation dose from 137C s ( — 60% ). A ll the calculations were perform ed without taking into account the effect of the transuranic alpha emitters. The inclusion of contam ination by the transuranic elem ents w ill of course increase the calculated dose, but not by m ore than a few per cent for districts w ith low contam ination levels and w ith the existing ratio o f contam i­ nation by transuranic elements and 137Cs. However, in districts with high con­ tamination levels (Kozelsk, Koryukovka, Sem enovka and Chernigov), the presence of 144Ce has been observed, with a contamination density of 0.03-0.08 Ci/km 2. POSTER PRESENTATIONS 8 9

Evaluation with the correlation coefficient (144Ce/239Pu) * 3 show s that in these districts a 239Pu contamination density of up to 0.02 Ci/km 2 m ay be expected. A conservative estim ate of the annual effective dose equivalent due to internal exposure from inhaled plutonium for the sixth year after the Chernobyl accident w ould then be 11.7 mrem/a. Such an estimate already m akes a significant contribution to the total dose. M oreover, given the fact that 241 A m accum ulation occurs according to the chain 241Pu (T 1/2 = 13 years) — 241 A m (T 1/2 = 433 years) the contribution of the dose from the transuranic elements w ill grow substantially on account of the 241 Am , w hich accumulates in the bone tissues (45% ) and in the hum an liver (45% ) with a biological half-life of 140 years. This also illustrates the need to carry out direct m easurem ents of the content of transuranic elements in the hum an body. In certain districts with a 137C s surface contam ination density of 1-5 Ci/km 2, direct measurements were performed of the 137Cs content in the human body using a whole body counter. The data obtained were used to determine the annual effective dose equivalents due to internal exposure, these being in the range 20-150 mrem/a. This fits well with the calculations described above. Com parison of the direct m easurem ents with the calculated potential annual dose equivalents for the adult population show s a steady correlation between them. M oreover, statistical analysis show s that there is a 9 5% probability that these two statistical samples describe the sam e effective dose equivalent. O n the basis of all the foregoing it m ay be concluded that the m ethod proposed for determ ining the internal and external radiation doses can be used, as a conserva­ tive approach at least, for determining the radiation burdens experienced by the population. It should also be noted that the radiation burdens resulting from ‘Chernobyl’ radionuclides are becoming comparable with background exposure levels (due to 40K), where contamination is greater than 1 Ci/km 2. 9 0 POSTER PRESENTATIONS

IAEA-CN-54/81P

ARE HUM AN TEETH AN INDICATOR OF 90Sr CONTAMINATION?

E. BOTEZATU Institute of Public Health and Medical Research

L. BÎRLEANU University of Medicine and Pharmacy

Iasi, Rom ania

1. INTRODUCTION

During the Chernobyl accident m any fission products were released into the atmosphere, such as 90Sr [1]. Given the osteotropism of 90Sr, we have been interested in its accum ulation in hum an teeth.

2. MATERIALS AND METHODS

In order to evaluate the 90Sr content in hum an teeth, 120 deciduous and per­ m anent tooth sam ples were analysed. A m inim um of 50 teeth per sample for seven age groups were collected from Bacau, Galati, Iasi, Neam t and Suceava counties by dental clinics. Incisors, m olars and canines were analysed, without separation of the crow n from the root, taking into account the birthdate only and not the tooth type. Calcium content was determined by the oxalate precipitation method. The 90Sr content was analysed by the im proved Patti-Jeanm aire method [2] using fum ­ ing nitric acid for separation. 90Sr was determined by its progeny product 90Y , beta counted in the oxalate form. The 90Sr to Ca ratios were evaluated.

3. RESULTS AND DISCUSSION

The 90Sr concentrations in deciduous teeth of young children born during the post-Chernobyl period (1986-1987) amounted to 10.8-330 mBq/g Ca and were 10-600 tim es higher than those obtained for perm anent or deciduous teeth of all the other age groups (Table I). A s far as the latter are concerned, som e sam ples had undetectable 90Sr activity and the maximum values were in the range of 0.5-32.4 mBq/g Ca. POSTER PRESENTATIONS 9 1

TABLE I. ^Sr LEVELS IN HU M AN TEETH (mBq/g Ca)

B ir t h d a t e G a la t i S u c e a v a B a c a u N e a m t I a s i A v e r a g e

1 9 4 0 <0.2-0.5 0 . 5 - 1 . 2 0 . 6 - 1 . 6 4 . 1 - 8 . 5 2 . 1 - 8 . 5 2 . 8 ± 2 . 4 1 9 4 1 - 1 9 5 0 < 0 . 2 - 0 . 5 0 . 2 - 0 . 7 4 . 5 - 9 . 9 < 0 . 2 - 1 . 0 0 . 2 - 1 . 0 2 . 5 ± 2 . 1 1 9 5 1 - 1 9 6 0 0 . 5 - 8 . 0 0 . 7 - 4 . 5 4 . 4 - 6 . 6 2 1 - 3 2 . 5 1 1 . 0 - 2 9 . 0 1 3 . 6 + 9 . 8 1 9 6 1 - 1 9 7 0 1 2 - 2 3 . 5 1 . 1 - 1 3 . 5 1 . 4 - 1 2 6 . 2 5 - 7 . 6 4 . 0 - 8 . 9 8 . 9 + 4 . 6 1 9 7 1 - 1 9 8 0 0 . 5 - 1 1 . 2 6 . 4 - 1 2 . 7 2 . 9 - 5 . 8 0 . 5 - 3 . 2 1 6 . 2 - 3 2 . 4 7 . 7 ± 4 . 5 1 9 8 1 - 1 9 8 6 0 . 5 - 8 . 0 0 . 5 - 4 . 0 0 . 7 - 1 . 5 0 . 5 - 6 . 3 0 . 5 - 2 . 0 2 . 5 ± 2 . 0 1 9 8 6 - 1 9 8 7 1 0 . 8 - 4 8 . 0 3 4 . 7 - 9 9 . 5 3 1 - 1 3 6 1 9 . 2 - 3 3 0 1 1 4 - 3 2 8 1 5 4 ± 9 3

The average of 2.5 ± 2.0 for children 5-10 years of age (born before M ay 1986) rose to 154 + 93 mBq/g Ca for children 5-7 years of age (born after the Chernobyl accident) (Table I). These results obviously show that a 90Sr contam ination of hum ans occurred as a consequence of the Chernobyl accident. A drawback to our study is that we have not recorded such data as inform ation on the m other’s residence location during pregnancy, the child’s residence location during its first year, durations and particu­ lars of breast feeding or artificial feeding, and other inform ation regarding the teeth (type, with and without caries). Table I shows that the average values of 90Sr content in hum an teeth rose from 2.5 ± 2.1 for those born before 1950 to 13.6 ± 9.8 for those born in the period 1951-1960, then decreased to 8.9 ± 4.6 and 7.7 + 4.5 mBq/g Ca for individuals bom during 1961-1970 and 1971-1980, respectively. The variations about the m ean are typical of such survey measurements. The existence o f the higher values for young adults is interpreted as reflecting that these individuals were children during the period of greatest 90Sr deposition from radioactive fallout after nuclear w eapons testing. These results are in agreem ent w ith the data already reported in the literature [3]. The highest values were recorded for Iasi and Neam t counties and the lowest for Galati county, w hich corresponds to dietary 90Sr intake. A s an intriguing result, the values obtained for Suceava county are not the highest, as expected, this county being considered as the m ost contaminated in Rom ania by the Chernobyl accident [4]. A possible explanation is a higher Ca content of sam ples collected from this c o u n t y . 9 2 POSTER PRESENTATIONS

4. CONCLUSIONS

Follow ing the Chernobyl accident, there was contamination of hum ans with 90Sr, and the deciduous teeth reflect the peak, depending on prenatal and postnatal accum ulation of 90Sr. Hum an teeth definitely constitute a powerful indicator of environm ental and hum an contam ination with 90Sr.

REFERENCES

[1] INTERNATIONAL ATOMIC ENERGY AGENCY, Summary Report on the Post- Accident Review Meeting on the Chernobyl Accident, Safety Series No. 75-INSAG-1, IA EA , Vienna (1986). [2] PATTI, F., JE A N M A IR E , L., Methode simplifiée de dosage du Strontium 90 sur des quantités importantes de cendres d’os, CEA-R2998, Fontenay-aux-Roses (1966). [3] C U A , T.F., 90Sr from Worldwide Fallout in Teeth and Bones (1992) 171 pp. [4] B O T E Z A T U , E.I., IA C O B , O ., Estimates of radiation doses due to ingested radiocae- sium and strontium-90 to the Romanian population: (the second and third years after the Chernobyl accident, Environment 3 3 (1992) 47 (in Romanian). ' POSTER PRESENTATIONS 9 3

IAEA-CN-S4/89P

LES OBSERVATOIRES PERMANENTS DE LA RADIOACTIVITE D E L ’I P S N Evolution du césium 137 dans les aérosols atmosphériques entre 1958 et 1994

D. CALM ET, P. BOUISSET, G. KERLAU, J. ALLENO U Laboratoire de m esure de la radioactivité de l’environnement, Institut de protection et de sûreté nucléaire, Orsay, France

1. INTRODUCTION

La prem ière opération de collecte systém atique à long term e et à grande échelle d ’échantillons à des fins de m esure de la radioactivité a été m ise en place en France par le Com m issariat à l’énergie atomique (C EA ) dans les années 60. Deux obser­ vatoires, l’un concernant l’atm osphère avec la collecte des aérosols atm osphériques et des précipitations, l’autre les indicateurs alim entaires, ont été m is en œ uvre en 1958. Plus tard, ces prem iers observatoires ont été complétés d’un observatoire m arin qui couvre les façades atlantique et méditerranéenne. L ’ensemble de ces observatoires perm anents de la radioactivité a perm is de détecter et de préciser les caractéristiques des ém issions de nucléides radioactifs dans l’environnement, de déterm iner leur dispersion et de suivre leur distribution spatio-tem porelle à l’échelle du territoire national. Les résultats obtenus m ensuellem ent offrent une im age réactu­ alisée périodiquem ent de la situation radiologique à l’échelle régionale, hors site nucléaire, permettant d ’apprécier les risques sanitaires éventuels. A la suite de l’accident de Tchernobyl, les observatoires de l’Institut de protec­ tion et de sûreté nucléaire (IP SN ) ont été réorganisés à la dem ande du M inistère de l’industrie et font l’objet d ’une publication inform atisée sur un serveur M initel acces­ sible au public. Ils fonctionnent indépendam m ent des program m es de surveillance de la radioactivité m is en place par les industriels exploitant des installations nucléaires de base (C EA, Com pagnie générale des matières nucléaires ou Cogém a, Agence nationale pour la gestion des déchets radioactifs ou Andra, Electricité de France, etc.) ou des réseaux de contrôle gérés par l’Office de protection contre les rayonne­ ments ionisants (O PRI) pour le compte du Ministère de la santé. 9 4 POSTER PRESENTATIONS

FIG. 1. Stations de prélèvements de l ’observatoire atmosphérique.

2. MATERIELS ET METHODES

En ce qui concerne plus précisément l’observatoire de la radioactivité atmosphérique, il couvre les deux hémisphères où sont prélevés des aérosols atm osphériques et des eaux de précipitations. Les aérosols atm osphériques sont collectés dans huit stations de prélèvement (Fig. 1). Réparties pour six d ’entre elles sur le territoire m étropolitain, elles sont représentatives des principaux climats: Fiers et Bordeaux pour le clim at océanique, POSTER PRESENTATIONS 9 5

Orsay, Dijon et Tilly pour le clim at continental, et La Seyne-sur-M er pour le clim at méditerranéen. D eux autres stations sont situées dans l’hém isphère Sud, à Papeete (Tahiti) et à Saint-Denis-de-la-Réunion. Les aérosols atm osphériques sont prélevés par filtration de grands volum es d’air pendant 10 nuits consécutives (environ 100 000 m 3). La radioactivité volumique pour les radionucléides artificiels, inférieure au m icrobecquerel par m 3, est alors m esurable par spectrom étrie gam m a. Les eaux de précipitations sont collectées mensuellement à l’aide de pluviom ètres de grande surface de collecte (1 m 2) pour recueillir plusieurs dizaines de litres d ’eau. A u laboratoire, les eaux sont filtrées puis subissent une évaporation par chauffage (80 °C) jusqu’à l’obtention d’un volum e de 14 ml. La radioactivité volum ique pour les radionucléides artificiels, de l’ordre de la centaine de m icro­ becquerel par litre, est alors mesurable.

3. RESULT ATS-DISCUSSION

L ’analyse des données de radioactivité obtenues depuis 1958 sur les aérosols perm et d ’apprécier l’évolution de la contam ination atm osphérique du territoire fran­ çais. D eux nucléides radioactifs sont d ’un intérêt particulier: le césium 137 (Fig. 2) et le béryllium 7. Le césium 137 provient des essais aériens d’armes nucléaires et le béryl­ lium 7 d’un m écanism e naturel de spallation dans la haute stratosphère. Tous les deux se retrouvent initialement «stockés» dans le réservoir stratosphérique. Leur diffusion à partir de ce réservoir et le mouvement saisonnier des masses d’air expliquent la variation annuelle de leur radioactivité: un m axim um l’été, un m ini­ m um l’hiver, ce que montre la figure 2 jusqu’en 1984 pour le césium 137 dans les aérosols, et jusqu’à nos jours pour le béryllium 7. La vitesse m oyenne avec laquelle décroît la radioactivité des aérosols est égale entre deux injections. La période effec­ tive de décroissance est de 1 an (la période radioactive du césium 137 est de 30 ans). Les m esures de césium 137 réalisées sur les précipitations qui participent au les­ sivage des couches atm osphériques confirm ent cette période apparente. A la fin de l’année 85, avant l’accident de Tchernobyl survenu le 26 avril 1986, le niveau en césium 137 est à son point le plus bas, à la lim ite du m esurable. L ’acci­ dent de Tchernobyl s’accom pagne d ’une contam ination très intense et très brève de la troposphère. Ceci explique la décroissance très rapide de la radioactivité du césium 137 dans les aérosols (en 1986, la période effective de décroissance est alors de 8 jours). Q u ’il soit relâché lentement par le réservoir stratosphérique ou plus rapidem ent par le réservoir troposphérique com m e après l’accident de Tchernobyl, finalement le césium 137 se trouve déposé sur le sol. D e là, il est régulièrem ent rem is en suspen­ sion dans les aérosols atm osphériques. L ’après-Tchernobyl nous m ontre que la vari­ ation atm osphérique, lorsque le réservoir est le sol, apparaît en exacte opposition 6 9

1 0 0 0 ONS N IO T A T N E S E R P R E T S O P

FIG. 2. Evolution mensuelle du césium 137 dans les aérosols prélevés en France de 1958 à 1994. POSTER PRESENTATIONS 9 7 avec celle liée à une origine stratosphérique; on observe cette fois un m axim um l’hiver, un m inim um l’été. Pareil phénomène était visible dès l’hiver 1984 car le césium 137 déposé au sol pendant 40 ans était devenu prépondérant sur celui du réservoir stratosphérique non réalim enté depuis plus de 10 ans. Entre 1987 et 1993, on constate un allongem ent de la période effective de décroissance qui atteint 3 ans. 9 8 POSTER PRESENTATIONS

IAEA-CN-S4/92P

TRANSFER OF RADIOACTIVE CAESIUM TO HUNTERS AND THEIR FAM ILIES BY FOOD CHAINS IN FOREST ECOSYSTEM S

G. ÀGREN, R. BERGM AN National Defence Research Establishment, U m e â

B.-M. DROTTZ-SJÖBERG Centre for R isk Research, S t o c k h o l m

A . E N A N D E R National Defence Research Establishment, K a r l s t a d

R . F A L K Sw edish Radiation Protection Institute, S t o c k h o l m

K.J. JOHANSSON Sw edish University of Agricultural Sciences, U p p s a l a

L. JOHANSSON Radiation Physics Department, U m e â

S w e d e n

1. INTRODUCTION

Hunters and their fam ilies living in the boreal zone in Sweden are to a high degree living on products from the forest ecosystem. Earlier investigations have show n that the average cum ulative intake of radiocaesium of the Sw edish population from this type of food is as high as or even higher than that of products from agri­ cultural ecosystem s [1]. Groups with high consum ption of forest products, such as hunters, can receive internal doses considerably higher than those of the average population. Studies on the population of northern Sweden [2] have show n that the transfer of 137C s from ground to m an is m uch m ore effective in areas with relatively POSTER PRESENTATIONS 9 9 low deposition levels than in areas w ith high levels. T his could be an effect of infor­ m ation and m ay reflect different attitudes tow ards 137C s contam inated food products in areas with high and low deposition levels, or that countermeasures have been effective. Sim ilar reactions can be expected am ong hunters in areas w ith high deposi­ tion levels, where the activity in gam e meat often is too high for it to be sold. The aim of this study was to assess the average whole body content of 137C s in hunters living in areas with different deposition levels and to obtain a deeper understanding of the relation am ong whole body content of radiocaesium , specific food consum ption, and perception of the consequences of the Chernobyl accident. This paper presents the first results from this study.

2 . M E T H O D S

Measurem ents were performed in spring 1994, in two areas in a region in northern Sw eden and in three areas in a region in the m iddle part of Sw eden (Fig. 1). The areas in both these regions were chosen to constitute areas with different deposi­ tion levels [3] (Table I) within the boreal zone. The equipment used for the whole body measurements was installed in a standard container transported by a truck. Placem ent of the container at a central location in the respective regions studied m eant that the travel distance for the people m easured could be reduced considerably. This was important in order to m inim ize the refusal frequency. A total of 340 individuals were invited to participate; some were later excluded from the sample and the remaining sample was of 317 individuals. The counter is equipped with a H PG e detector affixed above a ‘cradle’ built from lead, serving, together with a collim ator, as a background shield. This con­ struction gives a background shield in all dow nw ard directions of at least 7 cm. The m easuring geom etry can be described as a m ix com bining the features of chair and arc geometries. This geometry has the advantage of an almost constant distance between the m easured person and the detector in its field of view between the head and knees. The m inim um detectable activity is estimated to be 150 Bq of 137C s for a 1000 s measurement, which was the adopted measurement time. This figure is, however, to som e degree dependent on the deposition level outside the container. To provide complementary information on the whole body measurements, each individual w as asked to fill in a questionnaire. Data from both m en and w om en were gathered for investigating the relationship between whole body content of radiocaesium and specific food consumption. The questionnaire also asked about current life conditions, interests and leisure activities, health status, food purchases and private produce. The respondents were also asked about various concerns in relation to the Chernobyl accident, e.g. w orry about health consequences and per­ ceptions of the consequences of the accident. They rated ordinary life hazards, as POSTER PRESENTATIONS

FIG. 1. Map of Sweden indicating the location of the measurement areas.

TABLE I. DEPOSITION LEVELS OF 137Cs IN CHOSEN AREAS

D eposition A rea County (kB q/m 2)

A näset Västerbotten 7 N ordm aling Västerbotten 45 By Västmanland 10 H arbo Uppland 40 Gävle Gästrikland 80 POSTER PRESENTATIONS 1 01 w ell as risks related to ionizing radiation. The respondents m ade rather precise esti­ mates of consumption per month of different kinds of meat, fish, berries, m ushroom s, and beverages such as m ilk and w ild berry juices. Questions were asked about food preparation, i.e. who usually prepared the meals and whether food preparations had changed because of concern about caesium content. The design of the questionnaire study included, apart from the measurement groups, a control group of Swedish hunters. This group was drawn from a random sample of registered hunters from the county of Västerbotten. Both the m an and the w om an in the selected household were invited to respond to a questionnaire, in accordance with the procedure used in the measurement groups. M em bers of the measurement groups were also requested to report in a stan­ dardized diary their specific food intake, as estim ates of w hat they actually had eaten and in what amount, during 1 week.

3. RESULTS AND CONCLUSIONS

The measured average values for the whole body content of 137C s for males and females, together with the observed m axim a, for each geographical region are sum marized in Table II. Observed body concentrations for men and wom en were tested (t tests) and were found to differ significantly (p < 0.05) in all areas except one (the village By). This difference in whole body concentration can possibly be explained by the fact that the effective biological half-life is shorter for w om en than for men. Another explanation could be a lower 137C s intake for wom en than for m e n . O f the m easurem ent group, 161 individuals (51 %) submitted a diary providing a detailed account of their specific food intake during 1 week. O f these, 135 (84% ) indicated that the w eek w as typical for their usual diet. Individuals sharing a house­ hold can be presum ed to consum e food from the same sources, prepared generally in the same way. The data from the diaries supported this. The food consum ption pattern within households was very sim ilar, although wom en indicated on average a som ewhat higher vegetable consum ption and low er meat consum ption. The 137C s whole body contents for m en and wom en sharing households were also found to be positively correlated (correlation 0.53, p < 0.001, controlling for area deposition l e v e l ) . Investigation of individual intake of different foods and 137C s whole body content revealed a consistent pattern of positive relationships for all m easurem ent regions only as regards consum ption of m oose meat. Intake of m oose meat is the single dietary factor m ost clearly related to whole body content. Table III sum m a­ rizes m ean self-rated intake of m oose meat and correlation coefficients with 137C s whole body content for m en and w om en in each region. The low and non-significant correlations found for one area (the village Harbo) can possibly be attributed to a 1 0 2 POSTER PRESENTATIONS

TABLE П. OBSERVED W HOLE BODY CONTENT OF 137Cs FOR THE DIFFERENT REGIONS STUDIED

Whole body content of ,37Cs (kBq) 06X П

Mean + SD Observed maximum

By M 37 0.7 ± 0.7 2.5 F 25 0.4 ± 0.4 1.3

Anäset M 31 0.6 ± 0.3 1.5 F 29 0.3 ± 0.2 0.7

Gävle M 19 4.7 ± 2.7 11.8 F 17 1.9 + 0.7 2.9

Harbo M 25 3.1 + 1.5 6.3 F 18 1.7 ± 0.7 3.1

Nordmaling M 37 3.5 ± 1.9 7.7 © OO F 23 1.6 ± 3.3

TABLE III. M EAN SELF-RATED INTAKE OF MOOSE M EAT AND CORRELATION COEFFICIENTS W ITH W HOLE BODY CONTENT OF I37Cs FOR M EN AND W OMEN IN EACH REGION

Moose meat Correlation Area Sex (kg/month) moose meat/whole body mean + SD content of 137Cs

By M 1.3 ± © 00 0.43a F 1.1 ± 0.7 0.52a

Anäset M 1.6 ± 1.2 0.44a F 1.4 ± 1.2 0.5a Ö OO Gävle M 1.6 + 0.53a F 1.3 ± 0.7 0.72b Harbo M 1.7 ± 1.2 0.17c F 1.1 ± 0.6 0.2C Nordmaling M 1.9 ± 2.0 0.52b F 1.4 ± 1.3 0.36a

a p < 0.05. b p < 0.01. c Not significant. POSTER PRESENTATIONS 1 0 3 som ew hat different dietary pattern am ong these respondents. Both the questionnaire and the diary data indicate a m arkedly higher consum ption of both deer meat and m ushroom s in this area. Respondents in the Harbo area were also those w ho in the questionnaire gave the highest rating of personal importance of m ushroom picking as an activity. W hile this dietary pattern m ight be expected to result in a raised w hole body content level, this m ay be offset by behavioural factors. The respondents from Harbo indicated to a significantly greater extent that they took precautions in the preparation of the food in order to reduce possible radioactivity. Thus the results seem to indicate the im portance of exam ining both dietary and behavioural habits in different regions m ore closely.

TABLE IV. CORRELATIONS BETW EEN W HOLE BODY CONTENT OF l37Cs AND DEPOSITION LEVEL FOR DIFFERENT BACKGROUND VARIABLES

137Cs (Bq) Group R P N

Mean SD

Total group 0.63 0.0005 261 1780 1842

Men 0.72 0.0005 149 2308 2177 Women 0.74 0.0005 111 1087 868 Married 0.62 0.0005 218 1797 1887 Single 0.82 0.0005 21 1495 1659 Have no children < 12 years 0.63 0.0005 243 1771 1864 Have children < 1 2 years 0.65 0.003 18 1903 1554 Live in city 0.71 0.0005 31 1749 1537 Live close to city 0.80 0.0005 27 1474 1101 Live in town 0.59 0.0005 35 2115 1894

Live in the countryside 0.65 0.0005 142 1773 2060 Work indoors 0.58 0.0005 125 1800 1899 Work both indoors/outdoors 0.63 0.0005 40 1709 1888

Work outdoors 0.74 0.0005 39 2400 2209 Believe that the home area 0.51 0.0005 134 2352 2102 is contaminated

Do not know if the home 0.75 0.0005 63 1127 1191 area is contaminated

Believe that the home area 0.70 0.0005 39 895 1108 is not contaminated 1 0 4 POSTER PRESENTATIONS

TABLE V. PEARSON PAIRW ISE CORRELATIONS WITH PREDICTORS ENTERED AS INDEPENDENT VARIABLES IN A STEPW ISE REGRESSION ANALYSIS W ITH W HOLE BODY CONTENT OF 137Cs AS THE DEPENDENT VARIABLE

Variable Beta weight p value R2 N

DEP 0.502 < 0 .0 0 0 5 SEX - 0 .3 1 9 < 0 .0 0 0 5 M EAT1 0.137 0.002 M EA T2 0.148 0.001 BLUE 0.207 < 0 .0 0 0 5 EXAM 0.119 0.006 PERCEPT 0.044 0.274 0.603 300

FIG. 2. Observed ratios between whole body content and average deposition level o f 137Cs in the respective areas studied. POSTER PRESENTATIONS 1 0 5

Table IV shows the correlations between whole body content of 137C s and deposition level for background variables in the questionnaire. Furtherm ore, the m ost prom ising variables, based on Pearson pairwise correlations, were entered as independent variables in a stepwise regression analysis with whole body content of 137C s as the dependent variable. The results (Table V ) showed that the deposition level in the hom e area (D EP), m ale or female (SE X ), how often the respondents ate m oose meat (M EA T 1) and estimated consum ption of m oose meat (M EA T 2), amount of frozen bilberries in the refrigerator (B LU E ), and the extent to w hich the respon­ dent had sent food products for exam ination for radioactive content (E X A M ) gave significant contributions to the model. The perceived risk of injury due to consum p­ tion of game (PERC EPT) also entered into the model, but with a nonsignificant contribution. The model explained 60% of the variance (R2 = 0.603) in 137Cs whole body content in the measurement group. In Fig. 2 the observed ratio between w hole body content and average deposi­ tion is plotted against the deposition level in the respective areas. Even though there are great uncertainties, this figure show s that the ratio decreases as the average depo­ sition level increases. T his effect is consistent but not so large as in studies on general populations [2, 4] where this ratio decreases by a factor of 4 when the deposition level increases by a factor of 10. This m ay reflect that hunters are a group unlikely to reject eating gam e even if the activity is high, or alternatively that they are well inform ed and estimate the risks from eating gam e to be low.

ACKNOWLEDGEMENTS

The authors w ould like to thank K. Lidström at the National Defence Research Establishment, Um eà, Sweden, for invaluable assistance during the measurement phase of the investigation. This project is financed by the Sw edish Radiation Protec­ tion Institute.

REFERENCES

[1] BERGMAN, R., NYLÉN, T., PALO, T., LIDSTRÖM, K., “The behaviour of radio­ active caesium in a boreal forest ecosystem” , Swedish Defence Research Establish­ ment, F O A report A 40066-4.3, F O A N B C Defence, Umeá, Sweden (1991). [2] J O H A N S S O N , L., À G R E N , G ., 137Cs in the population of northern Sweden, Radiat. Prot. Dosim. 55 (1994) 131. [3] E D V A R S O N , K . , “ Fallout over Sweden from the Chernobyl accident, The Chernobyl Fallout in Sweden: Results from a Research Programme on Environmental Radiology (M O B E R G , L., Ed.), Swedish Radiation Protection Institute (1991) 47-65. [4] F A L K , R., E K L U N D , G ., G IER TZ, H ., Ö S T E R G R E N , I., “ Cesium in the Swedish population after Chernobyl: Internal radiation, whole-body counting” , The Chernobyl Fallout in Sweden: Results from a Research Programme on Environmental Radiology (M O B E R G , L., Ed.), Swedish Radiation Protection Institute (1991) 547-577. 1 0 6 POSTER PRESENTATIONS

IAEA-CN-54/94P

SPAM SH NATIONAL DOSIMETRY BANK

J . M U Ñ O Z Consejo de Seguridad Nuclear (CSN), Madrid, Spain

1. INTRODUCTION

The National Dosim etry Bank (B D N ) was designed to be a useful instrument for the protection of exposed workers. O n the basis of individual doses, in conjunc­ tion w ith the type o f facility w here they were received and the type of w ork involved, it is possible to m onitor and control the individual conditions of an exposed worker. In addition to this prim ary objective, the B D N ’s structure and utilities are such that it can be used for applications such as determ ining the suitability of the w orking con­ ditions in various areas of ionizing radiation applications, evaluating exposure trends and the m ost affected areas, and supplying statistical data that can be used for legal s t u d i e s .

2. STRUCTURE

In order to be able to achieve the planned objectives, w hich go beyond what w ould be involved in m onitoring com pliance with the regulatory dose lim its, the data entered into the B D N are of three quite distinct types:

(1) data for identifying the exposed workers; (2) data for characterizing the received doses; (3) data for characterizing the circumstances, location and type of work.

Data o f the first type are fixed, since they are entered into the B D N once only, at the time when the exposed w orker is registered. Data of the second type are entered on a m onthly basis so that the B D N is con­ sistent with the period for w hich the dosim eters are used in practice. Data of the third type are also entered on a m onthly basis, the objective being to link each dosim eter reading as precisely as possible to the circum stances of the e x p o s u r e . A three level structure has been developed:

— Level 1 corresponds to the ‘generic branch’ com prising facilities involving occupationally exposed workers. POSTER PRESENTATIONS 1 0 7

— Level 2 corresponds to the ‘specific branch’ com prising facilities involving occupationally exposed workers, so that for each generic branch there are a num ber of specific branches. — Level 3 corresponds to the ‘type of w ork’ of an occupationally exposed person within each of the specific paths.

In accordance with this structure, each item of dosim etric data included in the B D N is assigned a further classification in the form generic branch-specific branch- type of w ork, w hich enables the B D N to perform statistical studies on the follow ing:

— general areas of w ork (nuclear power plants, industrial radiation facilities, e t c . ) ; — specific areas of w ork (radiodiagnosis, gam m a radiography, etc.); — specific groups of people (physicians, supervisors, etc.).

3. DATA ORGANIZATION AND MANAGEMENT

A s there is no single centralized laboratory in Spain carrying out dosim eter readings, access to inform ation is dependent on the flow of inform ation that exists between the dosim etry services and dosim eter users. Consequently all inform ation is sent by the dosim etry services. The m ost im portant processes involved in the m anagem ent of the database are:

(1) loading of data; (2) verification of the management status of the database; (3) performance of calculation processes; (4) modification of controlled data; (5) selective retrieval of information. The B D N com prises a series of m odules that m ake it possible to carry out these p r o c e s s e s .

3.1. Input of measurem ent data

The loading process can be perform ed either interactively or by m eans of m ag­ netic media. The latter is the m ethod used for the system atic loading o f the inform a­ tion sent by the dosim etry services. The data are validated by m eans of this process prior to final input into the database and corrections are m ade to any data containing e r r o r s .

3.2. Maintenance

The m ain function of the maintenance m odule is to m ake it possible to carry out various transaction m odification and consultation operations on data considered permanent, for the purpose of com pleting or m odifying the data sent by magnetic m e d i a . 1 0 8 POSTER PRESENTATIONS

3.3. M onitoring of overdoses and emergencies

The function of the m onitoring is to identify and m onitor individual cases where the legally established dose lim its or previously established dose values m ay be exceeded.

3.4. Official reports

The reports m odule m akes available, in a predetermined format, dose reports that are used in the C S N ’s internal management procedures or supplied to outside organizations.

3.5. Consultation and listing

The consultation and listing m odule m akes it possible to carry out selective retrieval of the inform ation contained in the database. It provides access to all the data, im posing the restrictions considered necessary for each process, and enables the results to be printed.

3.6. Statistics

The statistics m odule is used for the calculation processes involved in carrying out statistical studies of exposure trends. They are perform ed for different groups, areas and types of w ork, and m ake it possible to obtain the follow ing inform ation:

— range distribution of persons monitored by dosimeter; — number of persons monitored; — num ber of persons with insignificant doses; — average individual dose; — percentage of persons with doses not in excess of established limits.

B y m onitoring these parameters, it is possible to identify situations that are significantly different from the grand average and where action is necessary in order to achieve the A L A R A (as low as reasonably achievable) objective.

4. FUTURE ACTIVITY

The B D N was conceived as a dynam ic instrument for use in the operational protection of outside w orkers in accordance with Directive 90/641/Euratom. Consideration m ay be given to the input of additional data in order to carry out epidem iological studies. POSTER PRESENTATIONS 1 0 9

IAEA-CN-54/96P

QUALITAS AND QUANTITAS IN RADIATION PROTECTION

D.T.Y. CHEN Baden, Switzerland

1. INTRODUCTION

Radiation protection (RP) is an aspect of what is know n as health physics (H P). In this paper, I shall assum e that the task of R P is to optim ize and to lim it radiation burdens, and above all to justify harm s against benefits derived through the introduc­ tion of radiation to society. H P is supposed to deliver necessary data for this difficult and delicate task. R P has been praised for playing a leading role in protecting society against harm ful environm ental agents, since it is the first science to have established quantitative protection standards based on H P. Fundam ental for H P to be a quantita­ tive science is successful establishm ent of proper units of measurement. Physics is know n as the m odel for quantitative disciplines. Difficulties arise when physics is com bined with qualitative m easurem ents in biology. In this paper, I shall describe the difficulties encountered in the developm ent of notions for radiation units and look for factors that m ay influence their developm ent in the future. I firm ly believe that the present recom m endations from R P are m isleading. I w ill develop theses in this respect, and suggest rem edial steps.

2. PAST EFFORTS TO CONCEIVE PROPER NOTIONS FOR RADIATION UNITS

Developm ent of radiation units in the early period of H P w as dictated by the im m ediate injuries confronted then, and is characterized by uncleam ess. ‘Erythem a dose’ (about 600 Röntgen) w as introduced as an effect dose. W ith X rays of different energy and w ith skin as a biodosim eter, the erythema dose could not be determined exactly. ‘U nit skin dose’ w as then conceived. It w as defined in term s of its genera­ tion and is thus an exposure. U nit skin dose w as replaced by ‘Röntgen’ defined with radium instead of X ray apparatus to im prove exactness. ‘Röntgen’ w as defined and redefined: as charge or energy per volum e or m ass, using air or tissue as the m aterial of interaction. The later notions of Röntgen are absorption doses. This reflects difficulties experienced in extending the notion of Röntgen not only for X rays but also for gam m a and other radiation. This also reflects uneasiness due to a dis­ crepancy between requirem ents for exactitude and necessity: ‘absorption doses’ were 1 1 0 POSTER PRESENTATIONS defined, as they can be determ ined with acceptable certitude in physics, but ‘effect doses’ to relate burdens to biological end points are needed in HP. At the end of W orld W ar II, after intense preoccupation with R P during the ingenious and in­ genuous developm ent of the nuclear bom b, Cantril and Parker [1] introduced ‘rep’, Röntgen equivalent physics, an absorption dose, and ‘rem ’, Röntgen equivalent m an, an effect dose. Röntgen was originally defined as exposure. It seems that, at that time, notions on exposure, absorption doses and effect doses were not clear. Eleven years later, the International Com m ission on Radiation Units and Measurem ents (IC R U ) [2] cleared up the difference between exposure dose and relative biological effectiveness dose (RBE). Another 15 years was needed for exposure to be taken as such, as field intensity, and not as dose [3]. R B E dose is an effect dose. The term ‘equivalent’ w as used to im ply the provisional character of the dose notions rep and rem. R B E dose w as thought to be the proper dose, the realization of the equivalent notion of rem. R B E dose was introduced to enable quantification of burden in m ixed radiation fields:

RBE dose = E R (RBE) r • D R w h e r e (RBE) r is the weighting factor for the absorption dose D R from radiation R. In 1959, the International Com m ission on Radiological Protection (IC R P ) [4] m ade com prehensive recommendations as an accom panying measure to the antici­ pated introduction of peaceful uses of nuclear energy: R B E dose was used exclu­ sively for external and internal (through incorporation) irradiation and for biological end points in general. How ever, the concept of R B E was not realized in practice, and in 1977 the IC R P [5] introduced a quality factor (Q F) to substitute for it. The new R B E dose, the ‘Q F dose’, was called ‘dose equivalent’. For the second time, the term ‘equivalent’ w as em ployed to indicate the provisional character of a dose notion. In Germ an speaking countries, this provisional character is not realized, and dose equivalent is called equivalent dose (Äquivalentdosis). In 1991 the IC R P [6] endorsed this m isunderstanding in its 1990 recommendations, and changed ‘dose equivalent’ to ‘equivalent dose’ with a m inor change in the num erical values of the factor (Q F)R. The name (Q F)R was at the same time changed to W R. So a proviso- rium w as changed into a definitivum , and essential provisos were dropped. The m ost important one, in m y opinion, is the assum ption that W R is the same for somatic and genetic effects. The situation in neglecting the difference between som atic and genetic effects w as worsened through the introduction of a new dose notion, effective dose equiva­ lent, and the situation was made im possible through its exclusive use in the 1990 IC R P recom m endations. Effective dose equivalent is conceived to convert dose due to partial body irradiation to a w hole body dose. This conversion enables sum m ation of som atic burdens under all conditions, and this effective dose equivalent is sup­ posed to be a com m on unit for radiological burdens in all cases, also for burdens POSTER PRESENTATIONS 1 11 to society as a whole. ‘Effective dose equivalent’ w as at once renam ed to ‘effective equivalent dose’, and sim plified to ‘effective dose’. It is obtained through extending the weighting factor W R by another weighting factor W T for tissue T :

effective dose = E R E x W R • W T ■ D R T

The apparent sim ilarity of this effective dose to the form er R B E dose is strik­ ing. W x’s are to be deduced from radiation sensitivity of the respective tissues. In practice they are estim ated from their cancer liabilities, from one of the som atic inju­ ries. Genetic injuries are not form ally taken care of. They are supposed to be introduced indirectly through W x for gonads. Accordingly, genetic effects are included only figuratively in the notion of effective dose as ‘detriments to individuals’. N ow , the situation for a society is not the sam e as that for an individual. W hile the age of an individual is counted in years, the life of a society spans generations. Thus the im portance o f genetic detrim ents is different for the health of an individual than for that of a society. In the definition of the effective dose, som atic detrim ents only are considered, or m ixed figuratively with genetic ones. This om ission or ‘m ixing’ disables the effective dose from accounting for genetic load to a society. E.E. Pochin of the IC R P [7] pointed this out as follows: “ the m ore precise the deri­ vation of an effective dose is m ade to appear, the m ore m isleading it m ay be in som e situations: already the genetic effect of exposure to radon and its daughter products in the hom e could be misinterpreted as being about 25 % (the W x for the gonads) of its 1.3 m Sv per year effective dose contribution to background radiation rate, instead of in fact contributing about zero gonad dose” . Lim itation of the genetic load to society is no longer explicitly m entioned in the 1990 IC R P recommendations, as it is in the previous recommendations. This perhaps is an expression of an incom petence of the IC R P in this respect. This infer­ ence of mine is supported by a quotation from S. Abraham son of the IC R P [8]: “ unlike som atic effects, our decisions impact future generations. They alone know if we decided w isely” .

3. RADIATION PROTECTION SITUATION AT PRESENT

In H P, the present situation in the evolution of notions of radiation units is m arked by the hesitation of the IC R U to endorse the concept of dose equivalent as equivalent dose [9] and effective dose as a substitute for R B E dose [10]. I shall characterize the present situation in R P with several personal experiences in this field. Because of the successful introduction of peaceful uses of nuclear energy and of radiation in nuclear m edicine in developed countries, the IC R P recom m endations are w elcom e support as practical guidelines in these new ly created fields. In Germ an 1 1 2 POSTER PRESENTATIONS speaking countries, they not only serve as guidelines but also are incorporated in law and ordinance. R B E dose of the 1959 IC R P recommendations was implementated 4 years later in the 1963 Sw iss R P ordinance as effect dose (W irkungsdosis), and dose equivalent of the 1977 IC R P recom m endations, 1 year previously, in the 1976 Sw iss ordinance as equivalent dose (Äquivalentdosis). Effective dose, which the IC R P form ally introduced in its 1991 recomm endations, found its way into official docum ents in Switzerland in 1984 [11]. Since 1984,1 have intervened and protested against its use as ‘the’ unit, without success. Replies from officials are alw ays: “ The genetic effects are included in the notion of effective dose” or “ This is the interna­ tional state of know ledge at present” . In October 1993,1 attended the International Sym posium on Biological Effects of Low Level Exposures to Radiation and Related Agents, ISB E L L E S 93, in Changchun, China. Key persons from the IC R P as well as influential people from the nuclear industry and nuclear m edicine took part. The main theme was: “ How low is low enough?” M any papers were devoted to the ‘conditioning effect’ of sm all pre-doses in radiation therapy. In the plenary discus­ sion, I pointed out that radiation therapy is of a quite different nature than R P , and that the question of “ H ow low is low enough?” in the sense of horm esis cannot be answered so long as the problem of “H ow high m ay a de m inim is dose be?” in the sense of doses of negligible significance is not solved in RP.

4. PERSPECTIVE ON THE FUTURE

The radiation unit has experienced a period of storm y developm ent during the last two generations. The present situation is unsatisfactory. W hether this problem , of fundamental importance to HP, may be granted a longer time for peaceful developm ent depends on m any factors. One important factor is surely the pressure of necessity. This, in turn, depends on the intensity of efforts to introduce peaceful uses of nuclear energy to the developing countries, now that their growth seem s to be im paired in the developed countries. Early confusion in the notions of radiation units was dictated by urgent needs due to immediate injuries and diseases. Let us hope that societies w ill not have to experience the sam e fate in genetic injury as m any individuals have in somatic injury. A n important difference is the relatedness of genom es: som atic injuries to individuals do not affect each other, whereas genetic injuries to societies do. Another factor, surely as im portant as the first one, is the conscientiousness of organizations, especially of the R P organizations: whether we w ill continue to pretend to know enough already about H P, w hile w e actually do not. In the first Sem inar on Interfaces in Nuclear Safety and Public Health, organized by the Nuclear Energy Agency of the O E C D in Paris in 1985 for delegates of the regula­ tory com m unity of R P organizations from m em ber countries, there were voices in the above m entioned sense: “ we pretend we know the dose-effect relationship” . In the second sem inar in 1990, questions in H P were left behind, and problem s in R P POSTER PRESENTATIONS 1 1 3 were tackled: W hat is the m eaning of the terms ‘risk’ and ‘optim izing’? H ow can we explain these terms, which we ourselves are not clear about, to the general public? In this IA E A conference, the exclusive club in the N E A sem inar is expanded to include general m em bers from the R P scientific com munity. I have the sincere belief that this extension from occupationals to vocationals w ill help to restore due respect to the profession of R P as a whole. It seem s that after the N E A sem inars and this IA E A conference, the next step w ould be involvem ent of non-professionals and interested m em bers of the public.

5. THESES AND PROPOSALS FOR DISCUSSION

The follow ing theses are set up that should lead to m y conclusion: that R P m isses its task to protect society from undue radiation injury, if effective dose is used as the com m on unit for radiological burden in general, both for individuals and for a s o c i e t y :

— Thesis 1 : Radiation has been proved to cause genetic injuries in addition to som atic ones. — Thesis 2: W e are unable to estimate the amount of injury due to a certain genetic load on a population: direct hum an data require observations in the span of generations, and indirect estimation m ethods (doubling dose and the direct method from m ouse data) have no sound scientific basis. — Thesis 3: For due protection of a society, the prevailing genetic load to the population m ust be know n as exactly as possible in order to be able to account for the accum ulated im pact on the later life of a society, even if we have no dose-effect estimation now. — Thesis 4: Effective dose cannot tell the am ount of genetically effective com po­ nent in a given radiation load: this com ponent m ay vary by orders of m agnitude (0.0... to 500%).

Now to m y proposals. In Germ an and Sw iss RP, a question circulates: “D o you intend to protect for or from radiation?” The answ er of the Inform ation Officer of the Society for Radiation Protection reads: “ In case of doubt, in favour of the profession” . In this sense, I suggest:

— Proposal 1: T o protect the reputation of our profession, if it is not already too late, we should refrain from opening professional discussion to include non­ professionals, and w ait until w e can be satisfied with what w e have to com m u­ nicate. I, personally, shall be very unhappy if we have to com municate now. — Proposal 2: To im prove the quality of what we have to com m unicate, effective dose should be used only to account for doses to occupationals, and a ‘m ore suitable’ dose notion should be used as soon as possible to account for genetic 1 1 4 POSTER PRESENTATIONS

load to populations, for instance, a form al introduction of the old notion of genetically significant dose. — Proposal 3: The R P community should, together with the nuclear industry, exert all its influence to restore the disaster regions arround Chernobyl and Celjabinsk-65 to norm al conditions, so that we m ay have, somehow, clean clothes in w hich to stand before the public: there is only one hum an genome!

REFERENCES

[1] C A N T R IL , S .T ., P A R K E R , H .M ., The Tolerance Dose, Rep. U S A E C M DDC- 1100, Technical Information Division, Oak Ridge, T N (1945). [2] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASURE­ M E N T S , Report of the ICR U 1956, National Bureau of Standards Handbook 78, Government Printing Office, Washington, D C (1957). [3] INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASURE­ MENTS, ICRU Report No. 20 (1971). [4] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP, Publication 1, Pergamon Press, Oxford and New York (1959). [5] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the ICRP, Publication 26, Pergamon Press, Oxford and New York (1977). [6] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990 Recommendations of the ICRP, Publication 60, Pergamon Press, Oxford and New York (1991). [7] P O C H IN , E .E ., Radiation Protection: Past and Future Needs (4th Int. Symp. Radiation Protection: Theory and Practice, Malvin (1989). [8] A B R A H A M S O N , S., Risk estimates past, present and future, Health Phys. 59 1 (1990) 99. [9] A L L IS Y , A ., JEN N IN G S, W .A ., et al., Quantities and units for use in radiation protection draft report, ICR U News (December 1991) 7. [10] ROSSI, H .H ., There is hardly one sievert, ICR U News (December 1991). [11] EIDGENÖSSISCHE KOMMISSION ZUR ÜBERWACHUNG DER RADIO­ A K T IV IT Ä T , Bericht der Eidg. Kommission zur Überwachung der Radioaktivität für das Jahr 1982 zuhanden des Bundesrates, 26 (1984). POSTER PRESENTATIONS 1 1 5

IAEA-CN-54/100P

OCCUPATIONAL EXPOSURE TO RADON PROGENY

W. KRAUS, W. RÖHNSCH, J. SCHW EDT, W. U LLM ANN Federal Office for Radiation Protection, Berlin, Germ any

1. RADIATION PROTECTION SURVEILLANCE IN THE EASTERN PART O F G E R M A N Y

In the eastern part of Germ any the occupational exposure of particular groups of w orkers to radon and its progeny has been regularly m onitored since the 1970s. This follows from the inclusion of their w ork places in the radiation protection system. This system applies not only to uranium ore m ining and m illing but also, according to legal regulations, to conventional ore and m ineral underground (non­ uranium ) m ining, shaft construction, securing of closed m ines against mechanical dam age, opening of caves and m ines to public exhibition, and use of water treatment and other facilities, e.g. scientific institutes and radon spas.

2. ESTIMATION OF EXPOSURE

The individual exposure E pot is determined by m easuring the potential alpha energy concentration Cpot of short lived radon progeny at representative w ork places i and m ultiplying it by the w orking tim e t¡ spent at that or a com parable w ork p l a c e :

Epot — Cpot,i t¡ i Cpot is m easured by m ine radiometers, i.e. battery powered devices of sm all m ass. A n automatic m easurem ent and evaluation program m e is integrated. The alpha radiation of the radon progeny separated on a filter (air flow rate 2-3 L/min, sam pling tim e 5 m in) is m easured w ith a sem iconductor detector at fixed time inter­ vals (M arkov method) [1]. Additionally, active devices for the m easurem ent of the integral exposure

Epot Cpot dt were introduced in the individual m onitoring of the cleanup staff of the uranium industry in 1991 [2]. 1 1 6 POSTER PRESENTATIONS

TABLE I. NUM BER OF MONITORED W ORKERS W ITH EXPOSURE DUE TO RADON PROGENY (IN THOUSANDS)

1976- 1975 1989 1990 1991 1992 1993 1988

Uranium mining 14.3 15.4 15.1 12.6 Cleanup operations 5.2 3.3 2.6 Uranium milling 3.4 3.2 2.8 2.2 Non-uranium industry 3.3 8.1 5.8 3.7 2.0 1.1 0.8

According to the legal regulations, the annual limit of exposure is 8 X 1010 M eV-h/m 3 = 12.8 mJ-h/m3 or about 4 W LM .

3. RESULTS OF SURVEILLANCE

The num ber of monitored workers is shown in Table I. W hile the num ber of workers in the uranium industry was approxim ately constant from the early 1970s to the late 1980s, in the non-uranium industry the num ber was increased m arkedly in the m id 1970s because of the successive inclusion of several other types of w ork places in the radiation protection system . A continuous reduction of staff in m ost of the facilities with exposure to radon progeny has been seen since 1989. Uranium m ining and m illing were discontinued in 1990. Cleanup units have been treating the w ork areas of the form er uranium m ining and m illing plants since 1991. Figures 1 and 2 show the average individual exposure to radon progeny in various industries. The data for uranium m ining and m illing have already been published [3]. The decrease of the average annual exposure show s the effectiveness of technical and organizational radiation protection m easures introduced since 1975. These m ainly consist of im proved ventilation and sealing off of radon sources.. Figures 3 and 4 give an overview of the fraction of w orkers with annual exposures above 1.2 W L M , i.e. three tenths of the limit.

4. CONCLUSIONS

A s in other countries, the occupational individual and collective exposure to radon progeny in eastern Germ any is in m any cases significantly higher than the occupational exposure incurred from artificial radiation sources [3, 4]. W ith the POSTER PRESENTATIONS 1 1 7

Average annual exposure (WLM)

1975 1980 1985 1990

ED Non-uranium industry £3 Uranium milling И Uranium mining H Cleanup units

FIG . 1. A verage annual exposure due to inhalation o f radon progeny in various practices.

Average annual exposure (WLM)

1975 1980 1985 1990

К Non-uranium mining 0Mining damage removal О Other facilities Ш Shaft construction S S h o w caves and mines □ Water treatment

FIG . 2. A verage annual exposure due to inhalation o f radon progeny in various practices

outside the uranium industry. 1 1 8 POSTER PRESENTATIONS

Fraction (%)

1975 1980 1985 1990

ED Non-uranium industry ^U ranium milling Ш Uranium mining Cleanup units

FIG. 3. Persons with annual exposure >1.2 WLM (due to inhalation of radon progeny) in various practices.

Fraction (°/o)

1975 1980 1985 1990

E3 Non-uranium mining Й Mining damage removal О Other facilities dShaft construction SlShow caves and mines DWater treatment

FIG. 4. Persons with annual exposure >1.2 WLM (due to inhalation of radon progeny) in various practices outside the uranium industry. POSTER PRESENTATIONS 1 1 9 latest dose conversion coefficient of ICRP-65, 0.5 W L M corresponds to 2.5 m Sv [5]. If one compares the exposures shown in Figs 1 and 2 with the mean annual individual occupational exposure of 1.87 m Sv due to artificial radiation sources for G erm any in 1992 [6], it seem s that there is still an im balance in our efforts at devel­ oping individual m onitoring of exposure to artificial and natural radiation sources for w ork places.

REFERENCES

[1] S E R D Y U O K O V A , A .S ., et al., Isotopy Radona, Atomisdat, Moscow (1975) 186 pp (in Russian). [2] U L L M A N N , W ., et al., Jahresbericht 1993, Bundesamt für Strahlenschutz, Salzgitter (1994) 166-169. [3] U N IT E D N A T IO N S , Sources and Effects of Ionizing Radiation, U N S C E A R 1993 Report, Annex D , United Nations, New York (1993) 447. [4] S C H W E D T , J., et al., Jahresbericht 1991, Bundesamt für Strahlenschutz, Salzgitter (1992) 129-131. [5] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Protec­ tion Against Radon-222 at Home and at Work, ICRP Publication 65, Pergamon Press, Oxford and New York (1994) 24. [6] BUNDESMINISTERIUM FÜR UMWELT, NATURSCHUTZ UND REAKTOR­ SIC HERH EIT, Bericht der Bundesregierung an den deutschen Bundestag über Umweltradioaktivität und Strahlenbelastung im Jahre 1992, Bonner Universitäts- Buchdruckerei (1994) 10. 1 2 0 POSTER PRESENTATIONS

IAEA-CN-54/109P

IN SITU M EASU REM ENTS OF DENSITY OF 137Cs CONTAMINATION OF THE FOREST SYSTEM IN UKRAINE

V. POIARKOV, A. NAZAROV, H. SH AM ES1 Ukrainian Radiation Training Centre, Kiev, Ukraine

1. INTRODUCTION

The Chernobyl accident resulted in significant contam ination of U krainian ter­ ritory by various radionuclides. The activity of the m ain dose form ing radionuclide released into the environment, 137Cs, was estimated to be approximately 3 . 1 X 1016 Bq. The density of surface contam ination of different natural system s by 137C s is the characteristic used for assessm ent and prediction of the radioecologi­ cal situation. Standard m ethods for the determ ination of the density of surface contam ination are based on sam pling of the superficial layer of soil, gam m a spectroscopic m easure­ m ents of radionuclides present in these sam ples, and subsequent recalculation of the density of surface contamination. In practice, such methods are very difficult. Because of the high non-uniform ity of fallout, num erous sam ples of soil are required to reach suitably accurate assessm ent of the density of surface contamination. A s has been show n in several studies, the density of soil contam ination can be obtained through in situ m easurem ents of the gam m a radiation fields over the area investigated. This technique has the advantage of averaging out any large inhom ogeneities in the horizontal distribution of fallout, and therefore assessm ent of the density of contamination will be more representative and result in higher accuracy and reduction of expenses for such investigation. For validation of the in situ method and study of m igration processes, the inventory of 137C s (activity per unit area) w as determined through a com bination of in situ spectrom etry using a N al(T l) detector and of soil sam ple analyses of num erous Ukrainian forest sites for the past 3 years.

1 IA E A Fellow, Tajura Nuclear Research Center, Libyan Arab Jamahiriya. POSTER PRESENTATIONS 1 21

TABLE I. CHARACTERISTICS OF SAM PLING SITES

Age of Place of sampling, direction and External Forest Soil Site forest distance from Chernobyl dose (/iR/h) composition type (a)

1 Rudnja Dimer region, block 15, 30-50 10p + b 50-60 sand lot 4, south 45 km

2 Rudnja Dimer region, block 15, 60-70 9p + lb 50-60 sand lot 15, south 45 km

3 Kotovka Polesje region, block 14, 200-230 10p 60-70 sand lot 25, west 45 km

4 Kotovka Polesje region, block 14, 100-130 10p 50-60 sand lot 25, west 45 km

5 Pakul Chernigov region, block 34, 70-80 9p + lb 60-70 sand lot 1, northeast 70 km

6 Pakul Chernigov region, block 40, 40-50 9p + lb 60-70 sand lot 8, northeast 70 km 7 Bretskoe Karukovka region, 80-100 8p + 2b + о 50-60 sand block 35, lot 1, northeast 130 km

2 . M E T H O D S

2.1. Sites and sam pling

A description is given in Table I of som e of the 20 experimental sites on the western, southern and northern deposition traces established as a network with the aim of studying m igration of radionuclides in the forest ecosystem . These sites are characteristic of Ukrainian coniferous forests with sandy soils. Representative sam ples of soil and forest litter were taken on flat sites (circle with a radius of 5 m ) in a forest w ith typical vegetation for a given locality in term s of type, age and plant density. A n additional criterion w as that external gam m a radia­ tion did not vary by m ore than 2 0 % within a site. Five or ten sam pling points were selected at each experim ental site, and sam ples of forest litter and com plex sam ples of soil were taken layer by layer at depths of 0-2, 2-5, 5-10, 10-15 and 15-20 cm at each point. Sam ples were considered to be representative if the parameters of interest were close to the m ean values characteristic for a given locality. 1 2 2

Each of the sam ples of forest litter consisted of five cores 14 cm in diameter. A com plex sam ple of each soil layer for each sam pling point w as obtained by averag­ ing ten core samples of 6.2 cm diameter. The density of surface contam ination was m easured for each sam pling point.

2.2. In situ measurements

The theory and methods of in situ gam m a spectrometry have been described fully in num erous w orks [1] and consist of the follow ing: A detector placed in the field at som e height above the ground responds to gam m a rays that are being emitted from the soil. The photopeak seen at 662 keV in a field spectrum results from the full absorption of 137C s gam m a rays that have not undergone any scattering on the path from soil to detector. The count rate of the 137C s photopeak can ultim ately be related to the level of activity in the ground. In situ spectra were measured by m eans of a portable m ultichannel analyser (SN IP, Italy) with a N al(Tl) detector (63 by 63 m m size, resolution ~ 8 % for energy 662 keV). The detector, on a tripod, w as placed at 0.7 m above the ground both in the centre and at the outskirts of the sam pling site. The tim e o f spectra m easurem ent did not exceed 10 m in. After measurements, the spectra were recorded on a floppy disk for analysis in laboratory conditions.

Depth (cm)

FIG. J. Percentage depth distribution of 137Cs for coniferous forests. TABLE II. STATISTICAL ANALYSIS RESULTS OF SAMPLING AND IN SITU M EASUREM ENTS ONS N IO T A T N E S E R P R E T S O P 3 2 1 124 POSTER PRESENTATIONS

3. RESULTS

The bulk of the 137Cs activity is retained in the top 15 cm of soil profiles. The litter of the coniferous forest contains 40-60% of the total 137Cs activity, and the surface layer (litter and 5 cm of soil) contains 90-95% (Fig. 1). Values of 137Cs distribution at depth h in soil can be fitted by an exponential dependence (see also Ref. [2]) with a deviation of less than 15%:

A = Aq exp [-ah] where A is the cumulative inventory of 137Cs down to depth h, A0 is the total inven­ tory of the surface layer, a is the reciprocal of the relaxation length in cm-1, and h is the linear depth. Statistical analysis (Table II) shows that the density of surface contamination significantly varies from point to point within each site (square < 100 m2), and the overall precision, representing the sum of sampling and analytical errors, lies between 15 and 50%. Such high inhomogeneity of the radioactive contamination is evidently due to the changing surface profile and the air streams interacting with it during deposition [3]. High fallout nonuniformity therefore is considered to be the major factor affecting accuracy in evaluating the density of contamination and radionuclide trans­ mission [4, 5]. The error associated with gamma spectrometric measurements of caesium did not exceed 15%. The full absorption peak count rate variations of the spectra measured at each control site did not exceed 15% and suggest that in situ measurements of the gamma radiation field produce more representative results than the sampling method, because any large inhomogeneities in the horizontal distribution of fallout are averaged out. The correlation coefficient between the average density of contamina­ tion and the count rate of the full absorption peak of 137Cs is 0.99, and the calibra­ tion dependence can be fitted by the following equation (Fig. 2):

A [Ci/km2] = ICs к + b

where A is the average density of contamination (1 Ci/km2 = 37 kBq/m2), I Cs is the count rate in the full absorption peak of 137Cs from in situ spectra, and к is the calibration coefficient. There are three peaks of Cs isotopes in the in situ spectra: 661.6 keV from 137Cs and 604.7 and 795.8 keV from 134Cs. Peaks at 604.7 and 661.6 keV are not resolved. But count rate correlation between these peaks is still quite stable (as shown by measurement) and changes with time are low. This fact facilitates spectra treat­ ment and allows determination of the density of surface contamination through the 137Cs full absorption peak without count rate subtraction from 604.7 keV. POSTER PRESENTATIONS 125

A (Ci/km2)

SNIP. CR (x 100) Count rate (1/s)

FIG . 2. C alibration dependence fo r field m easurem ents: plot o f SN IP. A v.s. S N I P . C R .

The following calibration coefficients were obtained for the measuring system based on SNIP and the 63 by 63 mm Nal(Tl) detector: Slope к = (1.17 ± 0.04) 10~2 Intercept b = (0.68 ± 0.40)

Correlation coefficient = 0.996 126 POSTER PRESENTATIONS

REFERENCES

[1] M IL L E R , K .M ., H ELF ER , I.K., Calibration factors for Ge detectors used for field

spectrometry, Health Phys. 55 (1988) 15.

[2] M IL L E R , K.M ., KU IPER, J.L., H ELFER, I.K., 137Cs fallout depth distribution in

forest versus field sites: Im plication for external gam m a dose rate, J. Environ. Radioac­

tivity 12 (1990) 23.

[3] L IV E N S, F.R., Q U A R M B Y , C., Sources of variation in environmental radiochemical

analysis, Envirom ent International 15 (1988) 271.

[4] P A R K IN SO N , J.A., H O R R ILL, D.A., An assessment of variation due to laboratory

and field conditions in the m easurem ent o f radionuclides, Nucl. Inst. M eth. Phis. Res.

223 (1983) 598.

[5] POIARKOV, V.A., NAZARO V, N.A., KALETNIK, N.N., Post-Chernobyl radi­

om onitoring of Ukrainian forest ecosystem s, J. Environ. Radioactivity (accepted for

publication). Technical Session 2 ASSESSMENT OF RADIATION HEALTH EFFECTS

IAEA-CN-54/1P

RADIOACTIVE ‘HOT PARTICLES’: ARE THEY STILL A PROBLEM ?

M .W . CHARLES, P.J. DARLEY School of Physics and Space Research, Birmingham University, Birmingham, United Kingdom

The evaluation of the hazard posed by very small radioactive sources (diameter < 1 mm) has become popularly known as the ‘hot particle’ problem. The ‘problem’ arises both from the difficulty of calculating or measuring the highly spatially non- uniform dose from hot particles and from the difficulty of evaluating the potential biological response. Hot particles are ubiquitous in all nuclear reactor environments but appear to be of particular practical concern in ageing water reactor (PWR and BWR) plants. In recent years the authors have co-ordinated an extensive collabora­ tive dosimetry and radiobiological study to develop improved dosimetry techniques for the measurement of radiation dose from hot particles and to relate exposure to early and late stochastic and deterministic effects, particularly in the skin. The con­ tinuing concern regarding the carcinogenic effects of hot particles is discussed in the light of the extensive literature which now exists for hot particle exposures of the lung and on the skin.

1. ORIGIN OF THE HOT PARTICLE PROBLEM

Concern regarding the carcinogenic effects of highly localized radiation exposures was focused strongly by the work of Geesaman [1] in 1968 and Dean and Langham [2] in 1969. In both cases an evaluation of hot particle carcinogenesis was made using the data of Albert et al. [3] for electron irradiation of rat skin. The induc­ tion of hair follicle tumours showed a power law dose response ( ~ D 4) which was interpreted by Geesaman in terms of a pseudo-threshold of about 20 Gy. The peak ratio of tumours to atrophied hair follicles was about 1:2000. Geesaman generalized this finding to infer that in any organ, if the local tissue dose was in excess of a threshold of about 20 Gy, then the probability of a tumour was about 10-3 to 10'4. Dean and Langham used the mathematical dose response, rather than a threshold concept, to predict an enhanced carcinogenic effect of 239Pu particles in the lung compared to uniform irradiation. They indicated the limitations of their conclusions because of the use of rat skin data for predicting the response of human lung. Tamplin and Cochran [4] extended these arguments to the case of plutonium in the

129 130 POSTER PRESENTATIONS

lung to predict that if the plutonium lung burden was composed of particles capable of delivering the 20 Gy threshold then the lung cancer risk would be very high at plutonium levels considered by the ICRP to be acceptable. Plutonium hot particles were considered to be 115 000 times more carcinogenic than calculations based on ICRP risk figures and the evaluation of mean organ dose would suggest. There are a number of severe problems with this hypothesis: (1) Hair follicle tumours are rare in other animals and in man. (2) Few tissues other than skin contain structures similar to hair follicles. (3) The correlation in Albert’s work between hair follicle atrophy and tumour for­ mation was subsequently shown to be fortuitous. Experiments with rat hair in the growing phase show tumour induction at doses below that which produced follicle atrophy [5]. (4) When the dose response is highly non-linear for a uniform exposure, then some enhancement may be expected for high dose hot particle exposures. Highly non-linear dose responses for carcinogenesis are rare, however, particularly for humans.

2. RESPONSE TO THE PROBLEM

These conjectural arguments were reviewed by a number of workers [6-10]. Their conclusions supported the continued use of mean organ dose as a predictor of cancer risk. In 1980 the ICRP [11] reviewed the biological effects of inhaled radio­ nuclides. The only indication of any hot particle effect was a 2-3 times greater risk for inhaled insoluble compared to soluble alpha emitters. The ICRP noted, however, that results of experiments designed specifically to detect hot particle effects had been negative. Coggle et al. [12] and Lambert et al. [13] have failed to find evidence for a hot particle effect in the lung in a series of experiments using gridded X ray fields and plutonium aerosols in mice. In our experimental programme we have concen­ trated on the effects of hot particles on the skin because of some concerns in the nuclear industry regarding skin contamination, because the origins of the Geesaman arguments were for skin, and because the dosimetry and biology are more easily con­ trolled than for lung exposures. The results of our programme have been reviewed extensively elsewhere [14-16] and have provided a basis for ICRP guidance on the limitation of hot particle exposures [14]. The ICRP concluded that the end point of practical concern following an acute hot particle exposure is skin ulceration. They used pig skin data to propose a threshold dose of 1 Sv over an area of 1 cm2 at a depth of about 100-150 /xm which would prevent the occurrence of even superficial transient ulceration. The skin cancer mortality risk from such an exposure was con­ sidered to be negligible (< 10~7) on the basis of a review of human data on skin POSTER PRESENTATIONS 131 cancer risks. Extensive experiments [15] on mouse skin with small radioactive sources (2 mm diameter 170Tm) were considered to provide strong evidence that spatially non-uniform exposure is less carcinogenic than uniform exposure for the same mean dose. There are now a range of radiobiological data for skin and lung, for stochastic and deterministic effects, which should allay fears regarding the risks of general non-uniform exposures in all tissues. However, in the aftermath of the Chernobyl accident, concern has again been raised regarding the enhanced carcinogenicity of hot particles.

3. SOME RECENT STUDIES

Lang et al. [17] have reported that hot particles implanted under the skin induced epidermal tumours in excess of expected calculated values. This was said to support a significant hot particle effect. The mechanism was thought to be sustained mitotic activity in the chronic wound which develops around the radiation source (said to be consistent with a role for cell proliferation in multistage carcinogenesis). In fact the cancer incidence following a hot particle exposure of mouse skin was compared with a figure derived for uniform exposure in man! From use of a more appropriate figure for uniform irradiation in the mouse [16] there would appear to be no significant hot particle effect. Two squamous cell carcinomas were seen in 32 irradiated skin sites exposed to C e/144Pr hot particles to a total exposure of 5 x 1010 beta particles (corresponding to a skin surface dose averaged over 1 cm2 of 25 Gy). Lower energy beta particles were said to produce a dose approximately 10 times higher. For SAS mice this would produce a skin cancer incidence of about 10% for uniform irradiation and 3% for particle exposures [16]. In humans the inci­ dence would be about 0.1% (based on an average incidence risk figure of 10% Gy“1 for uniform skin exposure) [14]. The observed incidence of two skin cancers in 32 sites (6%) therefore appears to support a reduced carcinogenic effect of hot par­ ticles, as previously reported for SAS mice [16], rather than an enhancement as claimed. In a study of malignant transformation of mouse fibroblasts by uranium aerosols released from Chernobyl, Servomaa and Rytomaa [18] considered that Chernobyl particles were ‘orders of magnitude’ more effective at inducing transfor­ mation in mouse fibroblast C3H ЮТ'/г cells than uniform gamma irradiation. This was based on the observation that transformed foci developed in every culture for hot particles, with zero incidence in controls. The gamma radiation induced transfor­ mation frequency was 2.5 x 10~4 G y'1 per surviving cell. No dosimetry data were given. However, since typical exposures were with particles of 300-1200 Bq activity for 6-10 weeks, the average dose can be calculated to be about 0.5-3 Gy over an area of 1 cm2. On the basis of the gamma radiation induced transformation fre­ quency and typical cell densities, the incidence of one transformed focus per culture 132 POSTER PRESENTATIONS is not unexpected and is not indicative of a significant hot particle enhancement factor as claimed. Likhtarev et al. [19] have also considered the in vitro effects of Chernobyl hot particles in a detailed and careful comparison of transformation induction following uniform and non-uniform exposure of C3H 10TЧг with S r/^Y , Chernobyl hot par­ ticles and gamma radiations. A small hot particle effect of between 1.5 and 5 at mean doses of 0.01-1.0 Gy was observed. While this effect is not large, it is of interest and we are repeating the experiment in order to provide a link with our extensive in vivo mouse skin carcinogenesis data [16].

4. SUMMARY

Hot particles, contrary to some widely publicized claims, have not been shown to be more carcinogenic than the same activity uniformly distributed in an organ. The majority of studies specifically designed to test the hot particle hypothesis in fact find a reduced effect for non-uniform exposures. In a limited number of studies which provide indications of an enhancement, the effect is small and may be the result of biological and dosimetric uncertainties and small deviations from linearity in the dose response at low doses.

REFERENCES

[1] G E E SA M A N , D.P., An Analysis of the Carcinogenic Risk from an Insoluble Alpha

Em itting Aerosol Deposited in Deep Respiratory Tissue, Rep. UCRL-50387 and

Addendum , University of California, Berkeley, C A (1968).

[2] D EA N , P.N., L A N G H A M , W .H ., Tum origenicity of small highly radioactive parti­

cles, Health Phys. 16 (1969) 79-84.

[3] A LBERT , R.E., BU RN S, F.J., H E IM B A C K , R.D., Skin damage and tumor formation

from grid and sieve patterns of electron and beta radiation in the rat, Rad. Res. 30

(1967) 525-540.

[4] T A M P L IN , A.R., C O C H R A N , T.B., The Inadequacy of Existing Radiation Standards

Related to Internal Exposure of M an to Insoluble Particles of Plutonium and Other

Alpha Em itting Hot Particles, National Resources Defense Council, W ashington, D C

( 1 9 7 4 ) .

[5] BU RN S, F.J., SIN CLA IR, I.P., ALBERT, R.E., V A N D ER LA A N , М ., Tumor induc­

tion and hair follicle dam age for different electron penetrations in rat skin, Rad. Res.

67 (1976) 474-481.

[6] UNITED KINGDO M M ED ICAL RESEARCH COUNCIL, The Toxicity of Pluto­

nium, H M SO , London (1975).

[7] BAIR, W .J., R IC H M O N D , C.R., W A CH O LZ, B.W ., A Radiological Assessment of

the Spatial Distribution of Radiation Dose from Inhaled Plutonium , Rep. W A SH -1320,

United States Atom ic Energy Com m ission, W ashington, D C (1974). POSTER PRESENTATIONS 133

[8] D A G LE, G.E., SA N D ER S, C.L., Radionuclide injury to the lung, Environ. Health

Perspect. 55 (1984) 129-137.

[9] FEIN EN DEG EN , L.E., HUG, O., JACOBI, W .E., O BERH AUSEN, E., Zur Toxi­

zität heißer Partikel, insbesondere Plutonium , G . Fisher Verlag, Stuttgart (1986).

[10] B U R K A R T , W ., Radiation biology of the lung, special Decem ber issue, Sei. Total

Environ. 89 (1989).

[11] INTERNATIONAL CO M M ISSIO N ON RADIOLOGICAL PROTECTION, Biologi­

cal Effects of Inhaled Radionuclides, Publication 31, Pergam on Press, O xford and N ew

York (1980).

[12] C O G G LE, J.E., PEEL, D .M ., T A R LIN G , J.D., Lung tumour induction in mice after

uniform and non-uniform external thoracic X-irradiation, Int. J. Rad. Biol. 48 (1985)

9 5 - 1 0 6 .

[13] LAM BERT, B.E., PHIPPS, M.L., LINDOP, P.J., BLACK, A., M OO RES, S.R.,

“ Induction of lung tum ours in m ice follow ing the inhalation of 239P u 0 2” , in Proc.

3rd Int. Sym p. of the Society for Radiological Protection, Inverness, SRP, Reading

( 1 9 8 2 ) 3 7 0 - 3 7 5 .

[14] INTERNATIONAL CO M M ISSIO N ON RADIO LO GICAL PROTECTION, The Bio­

logical Basis for D ose Lim itation in the Skin, Publication 59, Pergam on Press, O xford

and New York (1992).

[15] C H A R LE S, M .W ., A general consideration of the choice of dose lim its, averaging

areas and w eighting factors for the skin in the light o f revised skin cancer risk figures

and experim ental data on non-stochastic effects, Int. J. Rad. Biol. 57 (1990) 841-858.

[16] C H A R LES, M .W ., W ILLIA M S, J.P., CO G G LE, J E., Skin carcinogenesis following

uniform and non-uniform beta irradiation, Health Phys. 5 (1988) 399-406.

[17] LANG, S., KOSM A, V.M., SERVO M AA, K., RUUSKANEN, J., RYTO M AA, T.,

Tum our induction in m ouse epiderm al cells irradiated by hot particles, Int. J. Radiat.

Biol. 63 (1993) 375-381.

[18] SE R V O M A A , K., R Y T O M A A , T., Malignant transformation of mouse fibroblasts by

uranium aerosols released from Chernobyl, Frontiers in Radiation Biology (R IK L IS,

E., Ed.), V C H , W einheim (1990) 589-594.

[19] LIKH TAREV, I.A, REPIN, V.S., BO NDARENKO , O.A., N ECH AEV, S.Yu., “Do

Chernobyl hot particles represent a public health hazard?” , paper presented at IA E A

Research Co-ordination M eeting on the Radiobiological Im pact of H ot Beta Particles

from the Chernobyl Fallout: Risk Assessm ent, Bulgaria, 1993. 134 POSTER PRESENTATIONS

IAEA-CN-54-12P

CANCER RISK FOR UNITED STATES NUCLEAR W ORKERS, 1970-2000

R.E. LAPP LAPP Inc., Alexandria, Virginia, United States of America

The potential cancer incidence attributable to ionizing radiation is assessed for nuclear power workers in the USA exposed during the 1970-2000 time period. This potential health detriment is estimated in terms of excess cancer cases based on risk data for Japanese nuclear bombing survivors provided by the Radiation Effects Research Foundation (RERF). Lifetime exposure data for 1 230 000 workers are estimated using Nuclear Regulatory Commission (NRC) data [1]. Data for dose distribution and for collective career dose are given in Table I. In 1970 the average worker exposure was 0.01 Sv/a. By the year 2000 this average should drop to 0.002 Sv over the 30 year span. The lifetime occupational dose will average 0.009 Sv for all ever-monitored workers. For workers with measurable dose the career dose will average about 0.015 Sv. The occupational dose may be compared to that accumulated by off-site exposure to natural background radiation and to medical radiation. The latter may be 10-30 rem for most US citizens over a lifetime. Thus the off-site radiation exposure of a nuclear worker greatly exceeds his or her on-site exposure. Only one worker in every 340 exceeds a career dose of 0.2 Sv. RERF data for 79 972 nuclear bomb survivors [2] show their collective exposure to be about the same as that for the USA’s workforce from 1970 to 2000. However, nine-tenths of the collective dose for the nuclear bombing survivors was accumulated above 0.20 Sv. The epidemiology for survivors with less than 0.2 Sv is non-robust. RERF cancer risk below 0.2 Sv has to be extrapolated from the high dose domain. Models of dose-response have to be constructed for estimating low level radiation risk. The nature of these risk models may include that of zero risk below a threshold dose. Transfer of cancer risk from the Japanese experience to occupational exposure involves adjustment for the difference between a sharp flash of radiation and pro­ tracted dose. A dose rate effectiveness factor (DREF) of 2.5 is assumed. RERF epidemiology diagnosed 8613 primary cancers for 1958-1987. Comparison of exposed and non-exposed survivors identified an excess of 503 radiation induced POSTER PRESENTATIONS 135

TABLE I. DATA FOR DOSE DISTRIBUTION

Lifetim e dose interval Collective dose Num ber of workers ( S v ) ( m a n • S v )

0 4 7 0 0 0 0 0

0 - 0 . 0 0 1 3 1 0 0 0 0 1 0 0

0 . 0 0 1 - 0 . 0 1 2 3 0 0 0 0 1 0 0 0

0 . 0 1 - 0 . 0 5 1 7 0 0 0 0 4 7 5 0

0 . 0 5 - 0 . 1 3 5 0 0 0 2 5 0 0

0 . 1 - 0 . 2 1 5 0 0 0 2 0 0 0

0 . 2 - 0 . 5 3 5 0 0 1 1 1 0

0 . 5 + 1 0 0 6 0

1 1 5 0 0 A l l 1 2 3 0 0 0 0 ( r o u n d e d )

cancers. A DREF of 2.5 reduces this excess to 200 excess cancers. This leads to the expression of 200 radiation cancers through to the year 2020 for the USA’s nuclear workforce exposed from 1970 to 2000. The RERF results exclude as radiogenic cancers of the oral cavity, oesophagus, rectum, gall bladder, uterus, prostate, pancreas and central nervous system. Cancer of the thyroid gland is not indicated for adult exposures. By the year 2020, the USA’s nuclear workforce will express over 250 000 cancer cases. The potential 200 radiation attributed cancers represent a less than 0.1 % risk. Given our lack of knowledge of low dose effects, the risk could be zero. In Hiroshima and Nagasaki, stomach cancer accounted for 85 radiation attributed cases. These represented only 6.5% of the 1305 cancers in the exposed cohort. The 93.5% represent incidence related to the Japanese diet. In response to an act of the United States Congress (Public Law 97), the National Institutes of Health (NIH) published [3] a 355 page set of radioepidemiolog- ical tables relating radiation exposure to the probability that a diagnosed cancer was caused by the radiation. This 1985 publication presented statistical data for 13 cancer types (males and females) for various times of exposure and diagnosis. For example, the tables show that a male worker exposed to 0.3 Sv at age 35 and diagnosed at age 50 would have a 2% probability of radiation causation. The United States Congress held public hearings to investigate the application of the NIH tables to compensating ‘downwinders’, i.e. citizens in Utah exposed to radioactive fallout from atmospheric testing of nuclear weapons. In my testimony [4] 136 POSTER PRESENTATIONS

I estimated that the probability that the most exposed ‘downwinders’ in Utah would have radiation induced cancers due to fallout was less than 1 %. My analysis of the most recent data on fallout [5] shows an average exposure for Utah of less than 0.003 Sv. At this low level of exposure there was no evidential basis for compensat­ ing the ‘downwinders’. Nonetheless, the Congress passed a Radiation Compensation Act authorizing awards up to US $100 000 for cancer cases attributed to fallout. No dose determina­ tion was required. To date US $141 million has been awarded to some 2000 claimants. (Clearly, no health physicists have been elected to Congress.) Employees at the Nevada test site filed 216 lawsuits claiming various cancers attributed to occupational exposure. Claimants alleged whole body doses ranging from 0.05 to 0.39 Sv. A Federal District Court in Nevada tried six cases. In the summer of 1994 the Court ruled [6] in favour of the defendant, in this case the US Government. The judge took into account such factors as latency, dose-response and causation other than radiation. One case involved a worker with 40 years of employment and a dose of 0.39 Sv. He was a long term heavy smoker who died of lung cancer. Using an updated risk for lung cancer, the NIH tables yield a probability of causation of 3.5% for radiation attribution. Thus, even for relatively high occupa­ tional exposure, radiation risk is very small. As an occupational risk, ionizing radia­ tion should be ranked as a mild carcinogen. Below 0.30 Sv the risk is speculative.

REFERENCES

[1] RA D D ITZ, C.T., H A G EM EY ER , Occupational Radiation Exposure at Commercial

Nuclear Power Reactors, Nuclear Regulatory Comm ission Rep. NUREG-0713,

U SN RC, W ashington, D C (1992).

[2] RADIATIO N EFFECTS RESEARCH FOUNDATION, Cancer incidence in atomic

bom b survivors, 1958-1987, Rad. Res. 137 Suppl. (1994) 1-112.

[3] N A TIO N A L IN STITU TES OF H EALTH , Report of the N IH Ad Hoc W orking Group

to Develop Radioepidemiological Tables, N IH Publication No. 85-2748, NIH ,

Bethesda, M D (1985).

[4] LA PP, R.E., Assessm ent of Nuclear Risk, Testim ony before the United States Sente

Public W orks Com m ittee Hearing on the N IH Radiological Tables, 12 June 1985.

[5] ST E V E N S, W ., et al., Final Report: A Case Control Study of Leukem ia Deaths in Utah

and Exposures to Radioactive Fallout from the Nevada Test Site (1952-?), National

Cancer Institute Contract #N01-CO-23917, National Cancer Institute, Bethesda, M D .

[6] Federal Supplement 1994 W L 413236 (D, Nev.). POSTER PRESENTATIONS 137

IAEA-CN-54/26P

PRENATAL X RAY EXPOSURE, COSM IC RADIATION AND UNW ARRANTED PREGNANCY TERM INATIONS

Y. KALISH Department of Medical Physics, Beilinson Medical Center, Petah-Tikva, Israel

1. INTRODUCTION

Over the last 40 years, foetal exposure to X rays has been the subject of numerous studies. Lack of clear information gave rise to unjustified panic among the public in this matter [1-3] . The aim of this paper is to set the record straight by sum­ marizing the information accumulated to date and comparing the amount of radiation to which the embryo-foetus is exposed from X rays with that from cosmic and other natural sources of radiation. Possible embryonic-foetal damage due to irradiation may be classified into two principal types. The first is teratogenesis, or the creation of birth defects; this may occur when exposure to radiation takes place in the first 12 weeks of pregnancy, when the embryo is in the stage of organogenesis [4]. The second type is carcinogen­ esis, or malignancy producing effects; this is liable to occur when exposure to radia­ tion takes place in the second and third trimesters of pregnancy and is manifested in the first decade of the child’s life. The existing information on embryonic damage due to irradiation in the first trimester of pregnancy is based on studies conducted with laboratory animals, follow-up studies of individuals exposed to the nuclear bomb explosions in Japan and various statistical analyses. It appears that the probability that embryonic-foetal irradiation at a level of 10 mSv (1 rem) will cause developmental harm or cancer during the childhood years does not exceed 1:1000. This figure is negligible when compared to the incidence of birth defects (4-6%) among the general population. Table I summarizes the levels of radiation to which the embryo-foetus is exposed during diagnostic radiology. An important fact should be mentioned here: in only one in every 1000 X ray examinations of pregnant women is the level of radiation to which the foetus is exposed equal to or greater than 10 mSv (1 rem). In laboratory studies on animals, researchers found that they had to use high levels of radiation before any effect could be noted. As long as the level remained below 150 mSv (15 rem), no differences were observed between the control and the irradiated group. A follow-up study on women exposed to radiation at levels between 100 and 190 mSv (10 and 19 rem) during or after the nuclear explosions in Japan 138 POSTER PRESENTATIONS

TABLE I. ESTIMATED AVERAGE DOSE TO FOETUS PER RADIOGRAPHIC EXAMINATION

Exam ination m S v m r e m

D e n t a l 0 . 0 0 0 6 ( 0 . 0 6 )

H e a d up to 0.005 (up to 0.5)

Cervical spine up to 0.005 (up to 0.5)

Extrem ities up to 0.005 (up to 0.5)

S h o u l d e r 0 . 0 0 5 ( 0 . 5 )

C h e s t 0 . 0 0 5 ( 0 . 5 )

Thoracic spine 0 . 1 1 (ID

Upper gastrointestinal series 1 . 7 ( 1 7 0 )

P e l v i s 2 . 1 ( 2 1 0 )

A b d o m e n 2 . 2 ( 2 2 0 )

IVP 5 . 9 ( 5 9 0 )

Barium enema 9 . 0 ( 9 0 0 )

P e l v i m e t r y 1 2 . 7 ( 1 2 7 0 )

showed that babies in Hiroshima were born with relatively small head circumfer­ ences; however this fact was not observed in Nagasaki. It is possible that malnutri­ tion and multiple diseases, which were more prevalent in Hiroshima than in Nagasaki, played an important role in this finding. It is therefore commonly accepted, as stated in a report of the US National Council on Radiation Protection [4], that radiation levels of up to 50 mSv (5 rem) — levels only very rarely used in diagnostic radiology [3] — present no real danger to the embryo-foetus, and the advantages gained by clinical diagnosis far outweigh the negligible risk of embryonic damage.

2. NATURAL RADIATION

To gain the proper perspective, we should consider natural — including cosmic — radiation. We are exposed to cosmic radiation even when we are far away from X ray institutes and hospitals. In the US Rocky Mountain area (in the states of Colorado, New Mexico, and Utah), where the natural uranium content of the soil is quite high and the attenuation of cosmic radiation by the atmosphere is diminished by the high altitudes, the annual level of radiation exposure per person exceeds that of other areas in the country by about 1 mSv (100 mrem). Millions of people (obvi­ ously including pregnant women) live in Colorado, yet the incidence of cancer there is some 35% below the national average [2]. POSTER PRESENTATIONS 139

We may then conclude that any foetus being carried by a woman living in Colorado will receive, during the 9 months of gestation, a surplus irradiation of 0.75 mSv (75 mrem). A routine chest X ray of a pregnant woman exposes the foetus to a dose of 0.005 mSv (0.5 mrem). In other words, a foetus of a pregnant Colorado resident is exposed to surplus radiation equal to that from 150 chest X rays. Let us now examine the situation regarding dental X rays, which many preg­ nant women refuse, thereby neglecting their health for fear of damage to the foetus. If we assume a radiation level of 0.0006 mSv (0.06 mrem) per dental X ray (see Table I), some 1250 dental X rays would be required to equal the amount of surplus radiation absorbed by any foetus being carried by a woman living in the Rocky Mountains. Would anyone even think of putting up notices throughout that area, warning women to refrain from conceiving, or if already pregnant, to move away? In some regions in India and in other countries, background radiation is extremely high — up to 13 mSv (1300 mrem) per year; nevertheless, studies conducted in these areas indicate no increase in morbidity. As previously indicated, findings show that the irradiation of 1000 pregnant women with 10 mSv (1 rem) radiation will, in all probability, result in the birth of, at the most, one harmed baby. If we assume that the probability of harm to one foetus irradiated with 10 mSv (1 rem) is equal to that of harm to 2000 foetuses each irradi­ ated with 0.005 mSv (0.5 mrem), we find that a chest X ray examination of 2 million pregnant women will result .in the birth of, at the most, one harmed baby, whereas the number of babies with birth defects born to the same 2 million women, irrespec­ tive of radiation, will be at least 80 000. We have discussed dental and chest X rays. As may be seen in Table I, a similar situation exists regarding other X ray examinations outside the abdominal region. In the case of abdominal X rays, where the foetus is directly exposed to radia­ tion, the level of foetal exposure is obviously higher than during dental or chest X rays. However, even then, levels are rarely above 50 mSv (5 rem) and the risk remains slight. It may therefore be deduced that only very rarely, if at all, will the level of foetal irradiation in diagnostic radiography justify termination of pregnancy.

REFERENCES

[1] M O L E , R.H ., Radiation effects on prenatal development and their radiological sig­

nificance — review article, Brit. J. Radiol. 52 (1979) 89.

[2] C O H EN , L .B ., Before It’s Too Late, Plenum Publishing Corp., New York and London

( 1 9 8 3 ) .

[3] BAKER, M .L., V A N D ER G R IFT , J.F., D A LR Y M PLE, G.V., Fetal exposure in diag­

nostic radiology, Health Phys. 37 (1979) 237.

[4] NATIONAL COUNCIL ON RADIATION PROTECTION, Medical Radiation

Exposure of Pregnant and Potentially Pregnant W om en: Recommendations of the

National Council on Radiation Protection and Measurem ents (N C R P Rep. No. 54),

N CRP, W ashington, D C (1977). 140 POSTER PRESENTATIONS

IAEA-SM-326/32P

RADIATION RISK FACTS VERSUS PERCEPTIONS

R.E. LAPP LAPP Inc., Alexandria, Virginia United States of America

Prior to the nuclear bombing of Hiroshima in 1945, there was very little fear of ionizing radiation among the public. In the 1950s the USA began atmospheric testing of nuclear weapons and the Atomic Energy Commission (AEC) assured the ‘downwinders’ in Utah that there was no danger from the tests. The AEC set an annual limit of 0.039 Sv for the downwinders. Four decades later the down­ winders in Utah and uranium miners who developed lung cancer were awarded US $141 million in Federal compensation for 1940 malignancies. The fear of radiation was real but compensation was a matter of ‘radiation politics’. A comprehensive epidemiological study [1] of downwind states found no excess of cancers that could be attributed to radioactive fallout from nuclear tests. Fallout cases were 250 or more miles downwind of the weapons tests. Employees at the nuclear test site were much closer to the tests and exposed to higher levels of fallout. They filed 216 claims for malignancies (see Ref. [2]), attributed to doses of 0.049-0.39 Sv. In a landmark decision rendered in July 1994 [3], the Federal District Court of Nevada ruled against the plaintiffs. The Court considered the type of cancer, its latency, the radiation dose and other causation. On one hand, the Federal Government passed out public funds to downwinders with no evidence of radiation dosage, and on the other, the Federal Court denied compensation for cases involving much higher doses. The explanation for this dis­ jointed policy is that the United States Congress avoided dealing with facts about dose-response. Congressmen sought to reward their constituents who identified their malignancies as caused by test fallout. Elsewhere [4], I detail the facts about fallout and cancer litigation where the imputed exposures are in the low dose domain. The latter is generally considered to be exposures of less than 0.3 Sv. Downwinder doses in Utah averaged about 0.003 Sv. (The average dose for those exposed at Hiroshima and Nagasaki was 0.3 Sv, or 100 times more than in the Utah fallout.) What is the likelihood that low doses of ionizing radiation cause cancer? The answer is that there are no statistically sound epidemiological data relating low doses to cancer excess in an exposed population. The Radiation Effects Research Foundation (RERF) in Japan has carried out extensive investigations of some 80 000 nuclear bomb survivors, beginning in 1950. The RERF database [5] defines POSTER PRESENTATIONS 141 our knowledge of dose-response. Unfortunately, below 0.3 Sv, the RERF data are infirm and estimates of this risk are based on retropolation of high to low dose data. This means that low dose risks are speculative. In fact, the risks may be zero. RERF research data show an excess of some 500 malignant neoplasms among Japanese exposed in 1945 to a mean dose of 0.3 Sv. Their collective exposure, i.e. number of persons multiplied by their mean dose, exceeded 12 000 man-Sv. This yields a risk indicator of one cancer per 25 man-Sv. In the USA the most extensive epidemiological study was devoted to the follow-up of 32 643 workers at the Hanford Nuclear Production Plant in the state of Washington. Their average dose was 0.026 Sv, and their collective dose was 854 man-Sv. Thus the Hanford collective dose was only 7% as large as that for Hiroshima and Nagasaki. This disparity in collective dose meant that the Hanford epidemiological study could not challenge the RERF risk data unless low doses of radiation delivered over a period of years rather than as a sharp flash were more biologically effective than the nuclear bombing radiation. Animal studies indicate that protracted low doses may be two to ten times less biologically effective than a single exposure. A reduction factor of 2.5 will be assumed. This changes our risk indicator of 25 man • Sv/excess cancer to 62.5 man-Sv/cancer. Applying this risk factor of the Hanford collective dose of 845 man • Sv yields an estimate of 13 excess cancers to be expected for the Hanford workers. In 1976, A. Stewart was purported to have found excess cancer risk among Hanford workers. She was the mother of a controversy extended to this day [6]. She predicted [7] ultimate expression of 200 excess cancer deaths among the workers. By adjusting for incidence, not mortality, and allowing for latency, some 300 cancer cases would be attributed to occupational radiation. That is 2300% more than our estimate of 13 cases. In the most recent epidemiological study of Hanford workers, that of E. Gilbert [8], observed and expected cancer deaths at Hanford are as follows:

Dose interval 0.01-0.049 0.05-0.09 0.10-0.19 0.20 + (Sv)

Observed cancer deaths 316 55 43 40 Expected cancer deaths 311 56 42 43

Clearly, the statistics blur any assignment of risk, but it is significant that no dose trend is apparent. If cancer induction is proportional to dose, one would expect this to be apparent for high doses, and it is not. The Hiroshima data display a very pronounced dose effect above 0.3 Sv, leaving no doubt as to radiogenesis. The weight of scientific evidence is very much opposed to Stewart’s contentions, but 142 POSTER PRESENTATIONS newspapers do not feature this in their coverage. Stewart provides ‘big numbers’ of radiation casualties, and this gets media attention. In consequence, the public is led to believe that even small doses of radiation are very harmful. Professional societies are reluctant to deal with controversy, so that exaggerations of radiation risk are propagated in the absence of their refutation. An inordinate media overplay of radia­ tion risk leads to public misperception of this risk. Federal agencies fail to publish rational risk assessments for public understanding. Instead, the National Academy of Sciences publishes [9] voluminous reports unsuited for public consumption. The cleanup of US nuclear weapon production facilities is now estimated to cost up to US $1 trillion. This remediation is based on a ‘cleaner than clean’ philosophy, not on the rational assessment of radiation risk. The United States Environmental Protection Agency (EPA) would spend huge sums to reduce radon levels inside buildings to ‘safe’ levels. Yet, radon risk in homes is speculative. We have witnessed in our time the gross distortion of ionizing risk as a poten­ tial public health hazard. ‘Radiation’ and ‘cancer’ have become the twin fear words of the twentieth century.

REFERENCES

[1] M A C H A D O , S.G., Cancer mortality and radioactive fallout in southwestern Utah,

Am . J. Epidem iol. 125 (1987) 44-46.

[2] LA PP, R.E., “ Cancer for U.S. nuclear workers 1970-2000” , Paper P12, Proc. Conf.

on Radiation and Society: Com prehending Radiation Risk, Paris, October 1994, IA E A ,

Vienna (1996).

[3] Federal Supplement, W L413236 (D, Nev.).

[4] FO STER, K.R., et al., “ Phantom risk — scientific inference and the law ” , Chap­

ter 13, The Fallout Controversy (LA PP, R.E., Ed.) (1993).

[5] RADIATIO N EFFECTS RESEARCH FOUNDATION, Cancer incidence in atomic

bom b survivors, Rad. Res. 137 (1994) 1-112 Supp.

[6] K N EA LE, G.W ., et al., Reanalysis of Hanford data, Am . J. Ind. Med. 23 (1993)

3 7 1 - 5 8 2 .

[7] W A L D , M ., Pioneer in radiation sees risk even in small doses, The New York Tim es

( 1 9 9 3 ) 1 .

[8] G ILB E R T , E.S., et al., Updated analyses of combined mortality data for workers at

the Hanford site, Oak Ridge National Laboratory and Rocky Flats W eapons Plant,

Rad. Res. 136 (1993) 408-421.

[9] NATIONAL RESEARCH COUNCIL BIOLOGICAL EFFECTS OF IONIZING

R A D IA T IO N C O M M IT T EE, Health Risks of Radon and Other Internally Deposited

Alpha-Em itters, B E IR IV (1988) 1-602; Health Effects of Exposure to Low Levels of

Ionizing Radiation, B E IR V (1990) 1-421, National Research Council, W ashington,

D C (Note: B EIR IV is being revised as B EIR V I and B EIR V as B EIR VII.) POSTER PRESENTATIONS 143

IAEA-CN-54/42P

DATA NEEDS FOR ANALYSIS OF CHARACTERISTICS OF THE RECENT INCREASE OF CHILDHOOD THYROID CANCER INCIDENCE I N B E L A R U S

T. ABELIN, F. GURTNER, M. EGGER Department of Social and Preventive Medicine, University of Berne, Berne, Switzerland

J.I. AVERKIN State Research Institute of Oncology and Medical Radiology, Lesnoy/Minsk, Belarus

A.E. OKEANOV Belarus Centre for Medical Technology, Minsk, Belarus

1. INTRODUCTION

An increase in diagnoses of childhood thyroid cancer starting in 1990 has been reported for the Republic of Belarus [1,2]. However, there is debate on whether this increase is real and attributable to radiation released following the Chernobyl nuclear accident, or rather an artefact due to incorrect histological diagnosis, more complete case reporting and mass screening of children after the accident [3-5]. Tumour characteristics and the geographical distribution of cases have recently been described [6, 7]. These findings have strengthened the case for a real increase in the incidence and for a causal relationship with the Chernobyl accident. The pur­ pose of this presentation is to expand on these analyses within the limits of available information, and to discuss the type of data still needed to fully clarify the epidemio­ logical nature of this situation.

2. METHODS

Analyses are based on detailed information on the cases of childhood thyroid cancer reported to the Belarus Tumour Registry from 1976 to 1991. Less detailed data were available for cases diagnosed in 1992. Verification of histological diag- 144 POSTER PRESENTATIONS noses by Swiss expert pathologists was performed for all cases through 1991. Inci­ dence rates were based on population figures stratified by age, sex and district of residence at diagnosis. Data on size and clinical staging of cancers were based on surgeons’ notes as reported to the tumour registry and observations from the mor­ phological examinations by pathologists. Incidence rates were computed by both age at diagnosis and age at presumed exposure in 1986.

3. RESULTS

From 1976 to 1985, the incidence of thyroid cancer up to age 15 as reported to the Belarus Tumour Registry was 0.041 per 100 000 person-years (95% Cl: 0.019-0.078). This is similar to international rates, which range from 0.072 (0.045-0.110), as pooled for nine eastern European tumour registries, to 0.194, the average rate observed both in four Nordic and in ten US registries. Observed rates for Belarus increased more than 50 fold to 2.34 (1.77-3.04) in 1991 and to 2.55 (1.94-3.28) in 1992. The highest rates for 1990 and 1991 were observed in the three districts situ­ ated closest to the Chernobyl accident site: 13.08 cases per 100 000 person-years, or six cases. All of these cases were of advanced tumours of at least 15 mm in diameter and/or métastasés. The next highest rate was in the city of Gomel with 8.39 cases per 100 000 person-years, or 20 cases, 65% advanced, and in the remaining rural areas of Gomel oblast and the neighbouring districts of Brest oblast. The total rate for Gomel oblast was 6.10, or 48 cases, 73% of them advanced disease. Overall, tumour diameters measured up to 10 mm in 25.0% of cases, 10.1-15 mm in 26.4%, 15.1-30 mm in 26.4% and over 30 mm in 11.1%. Another 11.1% were between 10.1 and 40 mm, but without precise indication of size. This compares to a review of seven autopsy studies of occult thyroid carcinoma [8], where, of 286 occult cancers found, 96.2% were below 10 mm in diameter and none above 15 mm. Only one out of nine children diagnosed in the 10 year period preceding the accident was less than 10 years old, whereas 57% of 76 cases diagnosed in 1990 and 1991 were of children 0-4 years of age at the time of the accident.

4. DISCUSSION

Comparison of Belarus pre-accident rates with rates observed in other tumour registries around the world shows that previous underreporting cannot explain the reported increase of childhood thyroid cancer. Likewise, based on comparisons of observed tumour sizes with those found in occult carcinoma studies, a spurious increase due to detection of occult carcinoma by active case finding is improbable. POSTER PRESENTATIONS 145

The geographical pattern and age distribution support a causal association with the nuclear accident and suggest a minimal latency period of about 4 years. Detailed analysis of information available from the Belarus Tumour Registry has thus allowed addressing of many of the issues raised after the first reports. A number of questions, however, remain. For the cases diagnosed in 1992 and 1993, data are needed (i) on the mode of diagnosis including type of symptoms in self­ referred cases and type and circumstances of examination in screening of detected cases, and (ii) tumour stage at diagnosis and subsequent clinical course. Further­ more, information is required (iii) on migration patterns following the accident. For example, the unexpectedly high incidence rate in the city of Gomel could be linked to migration from heavily contaminated areas. A case-control study aiming to deter­ mine the dose-risk relationship has been concluded and should be published in the near future; however, (iv) the behavioural factors determining the risk of thyroid cancer are still ill defined. Behaviours that may have led to increased uptake of radioiodine during the days following the accident include playing outdoors and con­ sumption of leafy vegetables and milk. Similarly, the role of pre-existent iodine defi­ ciency and supplementation following the accident remains unknown. Therefore, data on (v) the pre-accident prevalence of iodine deficiency and post-accident iodine supplementation in different regions of Belarus should be obtained. Finally, informa­ tion is needed on (vi) the quality and completeness of registration within the Belarus Tumour Registry. The envisaged epidemiological study designs include matched case-control studies with both hospital and community controls, retrospective surveys based on cases and school classes, and quality control studies based on the patient charts in oncological and endocrinological dispensaries. Prospective registration should minimally include routine verification of histopathological diagnoses and informa­ tion on place of residence at the time of the Chernobyl accident. The registry should allow adequate monitoring of cancer incidence and detailed epidemiological analyses in the future.

ACKNOWLEDGEMENTS

Thanks are due I.M . Drobyshevskaya, Minister of Health of the Republic of Belarus, and K. V. Kasakov, former Minister of Health, for supporting international collaboration in Chernobyl related health research.

REFERENCES

[1] KAZAKO V, V.S., D EM ID CH IK, E.P., ASTAKH O VA, L.N., Thyroid cancer after

Chernobyl, Nature 359 (1992) 21. 146 POSTER PRESENTATIONS

[2] B A V E R ST O C K , K., et al., Thyroid cancer after Chernobyl, Nature 359 (1992) 21.

[3] B ER A L, V., R EEV E S, G., Childhood thyroid cancer in Belarus, Nature 359 (1992)

6 8 0 .

[4] SH IG EM A T SU , I., T H IESSEN , J.W ., Childhood thyroid cancer in Belarus, Nature

359 (1992) 681.

[5] RO N, E., LU BIN , J., SCH N EID ER, A.B., Thyroid cancer incidence, Nature 360

(1992) 113.

[6] F U R M A N T C H U K , A.W ., et al., Pathomorphological findings in thyroid cancers of

children from the Republic of Belarus: A study of 86 cases occurring between 1986

(‘post-Chernobyl’) and 1991, Histopathology 21 (1992) 401.

[7] A B E L IN , T ., et al., Thyroid cancer in Belarus post-Chernobyl: Im proved detection or

increased incidence?, Soz. Prävm ed. 39 (1994) 189.

[8] H ARACH , H.R., FRAN SSILA, K.O., W ASEN IU S, V.-М., Occult papillary carci­

nom a of the thyroid. A ‘norm al’ finding in Finland: A system atic autopsy study, Cancer

56 (1985) 531.

[9] A B ELIN , T., EG G ER , M ., R U C H T I, C., Thyroid cancer in Belarus post-Chernobyl,

B M J 308 (1994) (in press). POSTER PRESENTATIONS 147

IAEA-CN-54/105P

UFETEM E RISK DUE TO SELECTED CAUSES OF INDONESIANS’ DEATH

E. HISWARA Centre for Standardization and Radiation Safety Research, National Atomic Energy Agency, Jakarta, Indonesia

1. INTRODUCTION

A man or a woman could die from one of a number of causes. Such a cause could be an activity that he or she was involved in or something that could not be avoided. In either case, lifetime risk of death can be estimated based on the mortality rate for each cause, or the dose rate in case of radiation exposure. In general, lifetime risk of death is defined as the possibility of death faced by someone in his or her entire life. To estimate how big the risk is from a particular cause, comparison can be made to risks from other causes. In this study, the method described by Greenhalgh [1] was used. The maximum lifetime of Indonesians was assumed to be 90 years, and the causes compared for their risks were smoking ten cigarettes per day from the age of 20, accidents on the road, accidents at work, death in infancy, disease, natural disaster, radon exposure and background radiation.

Age (years) FIG. 1. Life expectancy in Indonesia. 148 POSTER PRESENTATIONS

2. M ETH OD

To calculate lifetime risk, the concept of life expectancy was used. Life expec­ tancy (x) is the mean length of additional life beyond age x of all people alive at age x. Data on life expectancy for Indonesians were taken from reports published by the Indonesian Central Bureau of Statistics (BPS) [2]. By assuming that all Indone­ sians will be dead at age 90, and that the relationship between life expectancy and age is linear, Fig. 1 can be constructed. The risk factor given by the ICRP [3] was used to calculate lifetime risk related to radiation exposure. Table I summarizes the data used to calculate the lifetime risk for Indonesians for these eight causes.

TABLE I. DATA USED IN THIS WORK

N o . C a u s e D a t a R e m a r k

1 Sm oking ten cigarettes a day 9.32 X 10'3 Death rate 1986

2 Accidents on the road 6.07 x 10'5 Death rate 1990

3 Accidents at w ork 2.69 X 10 6 Death rate 1990

4 Death in infancy 0 . 0 3 9 M ale death rate 1990

0 . 0 3 2 Fem ale death rate 1990

5 A ll diseases 0 . 0 4 2 4 Death rate 1986

6 Natural disaster 2.29 X 10‘6 Death rate 1990

7 Radon exposure 0.42 W LM a-1 Exposure rate with

ventilation

5 . 1 4 W L M a -1 Exposure rate without

ventilation

8 Background radiation 5.26 X 10"4 Sv a '1 Exposure rate in

I n d o n e s i a

3. RESULTS

The results of calculation of lifetime risk from each cause considered are given in Table II. All but risk from background radiation are given for each sex. All risks calculated were averaged over the whole population. For accidents at work, however, it was possible to average only over workers, since only they are at risk. POSTER PRESENTATIONS 149

TABLE П. LIFETIME RISK DUE TO SELECTED CAUSES OF INDONESIANS’ DEATH

N o . C a u s e S e x Lifetim e risk

1 Sm oking ten cigarettes a day M 0 . 4 1 3

F 0 . 4 3 7

2 Accidents on the road M 2.84 X 10"3

F 3.01 x 10~3

3 Accidents at w ork M 1 . 1 9 x H T *

F 1.26 x 10‘4

4 Death in infancy M 2 . 2 0

F 1 . 9 1

5 A ll diseases M 2 . 3 9

F 2 . 5 3

6 Natural disaster M 1.29 x 10-4

F 1.36 x 10“4

7 Radon exposure M a 7.34 x 10-3

F a 9.79 x 10'3

M b 0 . 0 9

F b 0 . 1 2

С 8 Background radiation 2.36 x 10'3

a W ith ventilation. b W ithout ventilation. c For both sexes.

In general, as can be seen from Table II, death from disease gave the highest lifetime risk, followed by death in infancy, smoking ten cigarettes a day, radon exposure, accidents on the road, background radiation, natural disaster and accidents at work.

4. CONCLUSION

This paper presents calculations of lifetime risk due to eight causes of death for Indonesians. The result shows that death caused by disease gave the highest risk, while accidents at work gave the smallest risk. 150 POSTER PRESENTATIONS

REFERENCES

[1] G R E E N H A LG H , J.R., Reduced Life Expectancy Due to Sm oking, Accidents, M alig­

nancies and Exposure to Radon Daughters, N RPB-M 95, N RPB, Chilton (1983).

[2] CEN TRA L BU REAU OF STATISTICS, Welfare Indicators 1992, ISBN 979-402-

862-2, B PS, Jakarta (1993) (in Indonesian).

[3] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990

Recom m endations of the International Com m ission on Radiological Protection, Publi­

cation 60, Pergam on Press, Oxford and New York (1991). Technical Session 3 IMPACT OF RADIATION ON THE ENVIRONMENT

IAEA-CN-54/19P

TECHNICAL INFORM ATION BACKGROUND FOR DECISION M AKING IN CASE OF NUCLEAR ACCIDENTS

P. ZOMBORI, A. ANDRÁSI, J. URBÁN, I. NÉMETH KFKI Atomic Energy Research Institute (AERI), Budapest, Hungary

1. INTRODUCTION

Measures taken after major nuclear accidents can affect the living conditions of many people and require a considerable amount of human and material resources. Therefore, decisions in such cases must be based on thorough consideration of all aspects, including the most comprehensive knowledge of the radiological situation in the environment. All possible means for the assessment of the temporal and spatial variation, isotopic composition and dose contribution of the radionuclide contamina­ tion should be utilized together to support the decisions best suited for the given situ­ ation. The major technological systems providing information for the decision makers (ordered on the post-accident time scale according to response time and applicability) are as follows:

Early phase Intermediate phase Late phase

Early warning systems, Mobile radiological Network of analytical

telemetric networks laboratories laboratories

Mobile laboratories are considered to be the fastest, most flexible and most complex units suited for rapid direct measurements and for sampling. These systems combine the advantages of both types of network (early warning stations and analyti­ cal laboratories), and provide fast response and a wide range of analytical capabilities even in remote areas where radiological data can otherwise not be acquired. After the Chernobyl accident the Health Physics Department of AERI estab­ lished a mobile radiological laboratory for a nuclear emergency [1]. The instruments and methodology applied in this system and the means of data preparation for the decision makers are discussed in this paper.

153 154 POSTER PRESENTATIONS

2. M ETH ODS

The selection of the four wheel drive VW Transporter van as a base vehicle was a compromise between demands and possibilities: it is large enough to house the most important measuring instruments and sampling devices and to provide working area for the staff, but is compact and small enough to be easily operable by laboratory personnel. The instrumentation of the van makes a number of essential radiological mea­ surements possible which are detailed in the following sections.

2.1. Route monitoring

An automatic route monitoring system collects data on the temporal and spatial variation of the dose rate in contaminated areas. A Geiger-Müller tube based dose rate meter, fixed on the front windshield of the van, and a global positioning system (GPS) receiver device using satellite positioning signals are connected through two serial communication ports to the laptop personal computer (PC) of the mobile unit. The data (date, time, geographical co-ordinates, altitude, speed and dose rate) are continuously collected (with selectable collection time and alarm level), displayed and stored automatically during travelling. The four wheel drive vehicle can also per­ form dose rate scanning on rough terrain and in remote areas.

2.2. In situ gamma spectrometric measurement of the fallout radionuclide concentration

The radionuclide composition and the concentration of the fallout contamina­ tion are investigated by the method of in situ gamma spectrometry. A downward looking HpGe detector is positioned in the middle of a preferably flat area and the gamma spectrometric measurements (typically for 1000-2000 s) are used to deter­ mine the concentration of the gamma emitting contaminants distributed over the given area. A new method has been developed to estimate the depth distribution pro­ file of the fallout activity [2]. In situ gamma spectrometry is perhaps the most powerful method of the mobile laboratory. Its isotope selectivity, wide measuring range (detection of 137Cs con­ tamination from 102 to 107 Bq/m2), and the amount of information obtainable with a single spectrometric measurement make the method valuable and essential in the event of a nuclear emergency. POSTER PRESENTATIONS 155

2.3. Gam m a spectrometric determination of contamination in food and environmental samples

Environmental contamination is usually investigated by the laboratory analysis of samples collected in the affected area. Transport of samples to a central laboratory may lead to substantial delay, so proper tools and instrumentation were provided to make the sampling, preparation and measurement possible on the site. Soil, water, vegetation and food samples can be analysed for gamma emitting radionuclide con­ tamination with the shielded HpGe gamma spectrometer built into the mobile unit.

2.4. High precision dose rate measurements

The external dose rate is determined with high precision meters covering a wide range of potential environmental levels (10-5 to 103 mSv/h). A high pressure ionization chamber (RSS-111) is used for measurement of normal and elevated environmental levels while a portable Geiger-Müller tube based detector (SSM1) is better suited for fast surveying on highly contaminated areas. This latter device is used for automatic route monitoring purposes as mentioned earlier.

2.5. A ir activity concentration measurements

A combined three stage filter sucking air from the external atmosphere is used for the assessment of the air activity concentration in the contaminated area. A plastic fibre filter traps aerosol particulates passing through the filter system. This part is measured by gamma spectrometry to determine the isotopic composition of the air­ borne (aerosol) contamination. The second and third stages of the sandwich filter are intended for the sampling of elementary and organic iodine not attached to the aero­ sol particles. These filters (a thin paper disc impregnated with CuOS and activated charcoal granules of 25 cm3 volume) are then measured with a plastic beta counter. The sampling device is battery operated and can be run while travelling.

2.6. Personal dosimetry

Continuous personal dosimetric monitoring of the operating staff is provided by the use of an electronic personal dose meter with direct LCD readout of collected dose and preset dose rate alarm level. In addition, solid state dosimeters (TLD bulbs) are used for personal dosimetry. The collected dose can be evaluated immediately after irradiation with the special portable TL reader PILLE. Enough TLD bulbs are available for distribution to people living or staying longer in the contaminated area, to assess the individual and population doses. 156 POSTER PRESENTATIONS

3. DATA PREPARATION FOR DECISION MAKING

The basis of a decision about an important measure to be taken is the reliable measurement of environmental radiation dose rates and radionuclide concentrations. A single measurement, in general, is not enough for the judgement; a large number of data must be collected to have an overview of the situation and a reliable basis for predictions. The proper treatment of the data sets collected by the mobile radio­ logical laboratory helps decision makers to find the solution best suited for the situa­ tion. The management of large databases, evaluation of complex measurements, compilation of all acquired information, easy access to the individual measurement results and algorithmic improvement of the estimates of all important radiological parameters are provided as computer software tools installed on the laptop PC of the mobile unit. The final results are then communicated to the evaluation centre via radiotelephone.

3.1. Geographical distribution of dose rates (ERSCAN 3)

The results of dose rate measurements obtained during route monitoring are stored in files, 500 sets in one. They can be displayed on a graphic screen, presenting the geographical distribution of the values colour coded on a map and the time sequence of the measured data. The specially designed software ERSCAN3 provides easy access to each individual set of numerical data while showing all the information graphically on one screen. Consecutive sets of data can be loaded by increasing or decreasing the serial number of the data file.

3.2. Evaluation of gamma spectrometric measurements (G ’Peak Worksheet)

A program system developed for gamma spectrum analysis (G’Peak Work­ sheet) is used to evaluate in situ and sample gamma spectrometric measurements by the on-board computer of the mobile unit. Thus the results of this high complexity method (isotopic composition and concentration of the deposited activity on/in the soil and in environmental and food samples) are available shortly after arrival in the contaminated area.

3.3. Compilation of environmental radiological data (ERIDAN U S)

The spatial distribution of the dose rate and in situ gamma spectrometric mea­ surements can be displayed on a PC screen in a way similar to that described for route monitoring. The software ERIDANUS helps in getting information on the con­ tamination and radiation levels measured so far and presenting the data on a map. POSTER PRESENTATIONS 157

3.4. Calculation of optimum estimates for the environmental radiological parameters (OPTIM EST)

Decisions after an accident are often made on the basis of transport calculations using emission and meteorological data. Because of the uncertainty of model parameters and limited knowledge of the initial conditions, the predictions can vary over orders of magnitude. A combination of the model predictions and the correction for the emission data measured by the mobile laboratory will result in a better estima­ tion of the actual radiological situation at the investigated site. The computer code OPTIMEST was developed for the complex treatment of the collected data by the on-board computer of the mobile unit. The derived optimum estimates of the radio­ logical data can then be transmitted by radiotelephone to the evaluation centre.

4. CONCLUSIONS

Based on the instruments and methods integrated in the mobile laboratory, the staff operating in the contaminated area can provide all relevant radiological data that are required for decisions on the necessary measures to be taken. The mobile units can survey a large area within a relatively short period of time and transmit informa­ tion to the evaluation centre. Besides the technical devices, the unit operated by the Health Physics Department of AERI is equipped with software tools which make data reduction and analysis possible on the site, thus providing better support for the deci­ sion making officials. The methods and procedures are regularly tested in international intercompari­ son exercises which are organized each year in the region (central Europe) [3]. The mobile unit was used to make a countrywide survey of environmental radioactivity in Hungary in 1992, with special regard to contamination deposited after the Chernobyl accident.

REFERENCES

[1] A N D R Á SI, A., et al., “ M obile monitoring unit for assessing environmental radioac­

tive contam ination” , Proc. IR P A 8 Congress, Montreal, 1992, IR P A , W ashington, D C

(1992) 428-431.

[2] Z O M B O R I, P., et al., A New Method for the Determ ination of Radionuclide Distribu­

tion in the Soil by In Situ Gam m a Ray Spectrometry, Rep. KFKI-1992-20/K, K F K I,

Budapest (1992).

[3] A N D R Á SI, A., et al., Report on a W orkshop on M obile Laboratories for Monitoring

Environm ental Radiation, Rep. K F K I-1991-41/К, K F K I, Budapest (1991). 158 POSTER PRESENTATIONS

IAEA-CN-54/21P

IM PACT OF NATURAL RADIOACTIVITY FROM COAL FIRED POW ER PLANTS

J. KOVAC Institute for Medical Research and Occupational Health, University of Zagreb

N. NOVOSEL Ministry of the Economy Zagreb, Croatia

1. INTRODUCTION

Electrical power requirements will necessitate doubling the present generating capacity in Croatia in the future. As a result, environmental discharges associated with the coal power industry will considerably increase. One of the main sources of uranium and thorium that have been detected and are not connected with the nuclear industry has been found to be energy production using fossil fuels [1, 2]. By coal burning (in coal fired plants at about 1700°C) the activity originating from uranium and thorium is redistributed from underground (where the impact on humanity is nil) and liberated into the environment. Most of the radioactive substances are concen­ trated in the ash and slag, which are heavy and drop to the bottom of a furnace. Lighter fly ash, however, is carried up the chimney and into the atmosphere and irradiates people and contaminates food crops [3]. Also, 222Rn escapes into the atmosphere during incineration, while the non-gaseous members of the uranium decay series remain in the ash and slag. The bottom ash and slag are usually deposited in a waste pile, from where some activity may leach into aquifers or be dispersed by wind. For this reason, extensive investigations have been under way for several years in the coal fired power plant (CFPP) Plomin in Croatia, which uses an anthracite coal with a higher than usual uranium content and normal thorium content.

2. METHODS

Radiation surveys of the working area inside the CFPP were made with a Victoreen Thyac in beta-gamma survey meter with a detector held 1 m above the surface. Later on, a network of thermoluminescent dosimeters was organized. Air samples for determination of total alpha activity were collected on Schneider-Poelman filters with the aid of an air pump. At the same sampling stations POSTER PRESENTATIONS 159 the fallout samples were collected. One litre from the monthly sample was evapo­ rated to dryness for determination of total alpha activity, using a surface barrier Si detector and 1 К analysing system [4, 5]. To obtain the monthly mean concentrations of gamma emitters in surface air, the airborne particles were collected on glass fibre filters (with a nominal collection efficiency of 99.9% for 0.3 /um particles) by means of a high volume air sampler at a flow rate of 2000 m3 in 24 h. The activities of natural radionuclides were mea­ sured by use of a Ge(Li) detector joined to a 4 К channel analysing system and con­ nected on-line with the computer [4, 5]. At the beginning of our investigations, the working level (WL) was measured using the method of Holmgren [6]. Later we found much more convenient the field method developed by Scott [7]. All the necessary corrections were included in the calculations, including the time spent at each place by each worker.

3. RESULTS AND DISCUSSION

The external level of natural radiation around working places at the time of measurement varied from 13 to 39 n C -kg ^-lT 1, with occasional higher values. The highest value obtained on one occasion under the ash hopper was 129 nC •kg-1 -h”1. The background of the surrounding countryside showed fluctua­ tions between 1 and 2 nC-kg"1 -h”1. A group of five different work places were chosen for WL measurements inside the CFPP. The WL values (WLMs) have shown great variations between two measurements depending on the activity of the coal and combustion products present at the time of measurements in the CFPP. Places on site with good ventilation had 0.85-2.55 WLM (calculated on the basis of 170 working hours per month). The highest data were at the fresh waste pile and beside the bottom ash (13.6 W LM). The work places listed according to decreasing contamination were: fresh waste pile (13.6 W LM ); below the ash hopper (10.2 W LM); coal conveyor belt (2.6 W LM); steam generator building (1.0 W LM ); under the stack (0.9 W LM ). WL data from the office building (500 m distance from the plant) and from 10 km southwest of the CFPP did not differ (0.5 W LM ). The sampling stations were in the direction of the most frequent plume flow. Monthly mean values of total alpha activities in surface air showed fluctuations through the year, depending on meteorological parameters. The lowest values were in summer (less than 0.003 mBq/m3, with a maximum of 2.4 mBq/m3). The maxi­ mum value was obtained in winter (24.4 mBq/m3) at a location 1.5 km from the CFPP, in the direction southeast by east. At the same location monthly fallout samples were collected. The maximum value of total alpha activity was measured in the September sample (27.4 Bq/m3), which was in direct connection with the amount of rain. 160 POSTER PRESENTATIONS

Specific activities of 238U, 226Ra and 40K in surface air samples were 4.04 mBq/m3, 0.28 mBq/m3 and 0.62 mBq/m3, respectively. The mean values for the northern hemisphere are 0.026 mBq/m3 for 238U, 0.003 mBq/m3 for 226Ra and 30.8 mBq/m3 for 40K [8]. It is obvious that the activity levels for uranium and radium around the CFPP exceed these values by factors of 150 and 100. More recently some of the measurements were repeated, and the results have shown decreased contamination. The WL data on site are unrestricted and under no occupancy time limit. The reason for this probably is use of coal with low concentra­ tion of uranium and progeny. The old waste pile has been covered with an 0.5 m thick level of soil and grass has been sown, which is another reason for decreased contamination in the surrounding area.

4. CONCLUSION

At present it seems that covering the waste pile with soil was a good solution for reducing radioactive contamination in the environment and for keeping the possi­ ble risk to the surrounding population under permanent control. However, it is necessary to control the activity of coal used for combustion, slag and ashes deposited on new waste piles. WL measurements should be conducted at least once a year at critical working places.

REFERENCES

[1] E ISE N B U D , M ., PET R O W , H .G., Radioactivity in the atmospheric effluents of power

plants that use fossil fuel, Science 144 (1964) 288.

[2] B E C K , H .L., et al., “ Perturbations on the natural radiation environment due to the

utilization of coal as an energy source” , Natural Radiation Environment Ш ,

CONF-780422, 2 (1980) 1521.

[3] B A U M A N , A., et al., Technologically enhanced natural radioactivity in a coal fired

power plant, Proc. Int. Sym p. Bom bay, 1982 (V O H R A , K.G., Ed.), W iley Eastern

Lim ited, New Delhi (1982) 401.

[4] ENVIRONMENTAL MEASUREMENTS LABORATORY, HASL Procedures

M anual (H A R LE Y , J.H., Ed.), EM L-300, United States Atom ic Energy Com m ission,

New York (1971).

[5] IN TERN A TIO N A L ATO M IC EN ERG Y A G EN C Y , Measurement of Radionuclides in

Food and the Environm ent, Technical Reports Series No. 295, IA E A , Vienna (1989).

[6] H O LM G R EN , R.M ., W orking levels of radon daughters in air determined from

measurements of RaB + RaC, Health Phys. 27 (1974) 141.

[7] SC O T T , A .G ., A field method for measurement of radon daughters in air, Health Phys.

41 (1981) 403.

[8] U N IT E D N A T IO N S, Ionizing Radiation Sources and Biological Effects (Report to the

General Assem bly), United Nations Scientific Com m ittee on the Effects of Atom ic

Radiation (U N SCEA R), UN, New York (1982). POSTER PRESENTATIONS 161

IAEA-CN-54/54P

LE CONTROLE DE LA RADIOACTIVITE A PROXIM ITE DES CENTRALES ELECTRONUCLEAIRES FRANÇAISES

A. LECORRE Département sécurité, radioprotection, environnement, Electricité de France, Paris, France

PRESENTATION DES INSTALLATIONS

Electricité de France (EDF) produit 75% de son électricité par ses centrales électronucléaires à eau sous pression (REP). Ces centrales forment un parc composé de 34 unités de 900 MW et 20 unités de 1300 MW , dont la première a été couplée au réseau en 1977 et la dernière en 1993. Trois autres unités de 1400 MW sont en construction.

LES CONTROLES REGLEMENTAIRES DANS L’ENVIRONNEMENT

Les autorisations et modalités de rejets radioactifs gazeux et liquides ainsi que les contrôles associés dans l ’environnement sont définis par des arrêtés interministériels. Ces contrôles dans l’environnement sont effectués au quotidien par l’exploitant selon un programme et des modalités définis par le Service central de protection contre les rayonnements ionisants (SCPRI), qui est l ’autorité de contrôle de l’Etat français. Le SCPRI vérifie la validité des résultats et les compare avec ses propres échantillons. En outre, cet organisme effectue la surveillance de la radioactivité sur l’ensemble du territoire français (en particulier par le réseau Teleray). Autour de chaque centrale sont contrôlés le rayonnement gamma ambiant (journalier) en 8 points autour du site dans un rayon de 5 km, les aérosols dans l’air (journalier) en 4 points dans un rayon de 1 km, l ’eau de pluie, les eaux souterraines (mensuel), les eaux de surface (à chaque rejet radioactif liquide), le lait et les végétaux (mensuel) en 2 points dans la région proche du site. Chaque centrale dispose de moyens importants pour effectuer ces contrôles (un laboratoire extérieur au site, deux véhicules spécialisés et une équipe de trois chimistes). 162 POSTER PRESENTATIONS

LES ETUDES COMPLEMENTAIRES

Autour de chaque site nucléaire, EDF fait réaliser par l’Institut de protection de sûreté nucléaire (IPSN) un suivi radioécologique dans les milieux terrestre, aquatique et marin. Le suivi est fait annuellement par spectrométrie gamma sur des indicateurs judicieusement choisis (sédiments, algues, poissons, céréales, mousses, etc.). De plus, tous les dix ans, une étude approfondie est faite pour chaque site. Elle permet, par comparaison avec le «point zéro initial» réalisé avant le démarrage de la centrale, d’évaluer l’impact radioécologique dans l’environnement de l ’installation.

LA COMMUNICATION DES RESULTATS

Cette communication se fait à plusieurs niveaux: au niveau régional par les sites de production auprès des élus et des habitants locaux, aux niveaux national et international par les Services centraux d’EDF et par l ’IPSN. Différents moyens sont utilisés: réunions locales, publications régionales, consultation par minitel, conférences et congrès scientifiques. POSTER PRESENTATIONS 163

IAEA-CN-54/69P

SEVERE RADIATION ACCIDENTS AND THE ENVIRONM ENT

R.M. ALEXAKHIN Russian Institute of Agricultural Radiology and Agroecology, Obninsk, Russian Federation

In severe radiation accidents with releases of radionuclides into the environ­ ment, high enough radiation doses are formed to potentially cause direct radiation injury of natural ecosystems. The dose fields characterizing the irradiation of plants, animals and humans in radioactive contamination of the environment are highly het­ erogeneous, and irradiation of natural objects per se has features such as non- equidosal effects. In other words, humans and various objects in the natural environ­ ment can receive different absorbed doses for an equal density of radioactive fallout. In this case, maximal doses in different representatives of nature (e.g. coniferous trees and some groups of animal populations such as those inhabiting forest litter or bottom sediments) can be 10-100 times higher than those in humans. Natural and artificial ecosystems and their main components are characterized by great differ­ ences in radiosensitivity. The total effect of these two factors — namely, heter­ ogeneity of the dose field and non-equidosal pattern of irradiation, on the one hand, and differences in radiosensitivity, on the other — predetermines considerable differ­ ences in the degree of radiation injuries in ecosystems of various kinds. In the radia­ tion accidents in the southern Urals (1957) and at Chernobyl (1986), coniferous forests proved to be in fact the only natural ecosystems where radiation injury manifestations were notable at the ecosystem level in large areas. Evident symptoms of radiation injury, up to lethality, were found in small areas of deciduous forests, as well as in mammals. Mesofauna of forest litters and benthic organisms inhabiting the water ecosystems proved to be radioecologically hot sites. The minimal injurious dose of chronic irradiation at the levels of ecosystems and populations of living organisms is 5 Sv/a. In representatives of animal populations in the zone of the Chernobyl accident, radiation injury of the thyroid caused by accumulation of iodine radionuclides proved to be of considerable pathological significance. For the processes of radiation effects in accidental radioactive contamination of lands, two (or three) periods are characteristic — the phase of acute radiation injury with pos­ sible manifestations of radiation injury is followed by the phase of post-radiation recovery [1]. Radionuclides transferred to the environment are included into trophic chains of migration leading to man. Accumulation of radionuclides in nature and in agricul- 164 POSTER PRESENTATIONS tural products can pose limitations on human economic activities; therefore, studies on the behaviour of radionuclides in human habitats and definition of the parameters of radionuclide transfer via food chains are of primary interest. Several features can be distinguished that characterize radionuclide transfer in food chains after release of radioactive substances into the environment in radiation accidents. The first post­ accident stage (during the first year) is characterized by particularly intensive con­ tamination of plants and animals. This is because radionuclides transferred to the environment are novel ingredients, and their distinguishing feature is increased biomobility. Further on, with ‘ageing’ of radionuclides, their migration ability is, as a rule, considerably diminished. With respect to the formation of internal radiation doses, it is important that radionuclide content in agricultural products (milk, meat, plant products, etc.) is abruptly diminished during the second and subsequent years after the accident. For the first 2-3 (and subsequent) years after the Chernobyl accident, the ecological 137Cs half-lives (the periods of twofold decrease of 137Cs concentration) in milk, meat, potatoes and cereal grains ranged from 1.5 to 3 years (T 1/2 137Cs is 30 years) [2, 3]. In migration of radionuclides, so-called ecologically hot trophic chains are formed, i.e. the elements of the chain where radionuclide transport proceeds the most intensively. Such ecologically hot trophic chains include the meadow soil-plant systems, where many radionuclides (e.g. 90Sr and 137Cs) are particularly highly mobile. In radioactive contamination of the environment, so-called critical objects are of importance (agricultural products in the first place) that contribute the most to the internal dose to the population. In the regions affected by the accidents in the southern Urals and at Chernobyl, milk proved to be the critical product, ‘responsi­ ble’ for formation of more than 70% of the internal irradiation dose [1,4]. The most important factors affecting the biological mobility of radionuclides in the environ­ ment include parameters such as their form. In the Chernobyl accident, radionuclide deposition occurred in the form of large particles of non-destroyed fuel (uranium dioxide matrix) in the near zone (5-10 km from the reactor). In the process of destruction of these particles, radionuclides contained in them became more and more available for assimilation by plants. Radionuclide availability for inclusion into terrestrial food chains depends on the characteristics of the radionuclides and the properties of the environment (soil, climate, etc.). In the area affected by the Chernobyl accident, the internal dose to the population of regions with heavy fertile soils (chernozem) amounted to only 10% of the total dose, whereas in the regions with light soddy-podzolic soils this value amounted to 90% [4]. Based on the regularities of behaviour of radionuclides in agricultural food chains, a complex of countermeasures can be realized that gives the possibility of 30-50% decrease of the total radiation dose in populations in contaminated areas [1, 3, 5]. POSTER PRESENTATIONS 165

REFERENCES

[1] A LEX A K H IN , R.M ., Radioecological lessons of Chernobyl, Radiobiologiya 33 1

( 1 9 9 3 ) 3.

[2] A L E X A K H IN , R.M ., Countermeasures in agricultural production as an effective

m eans of m itigating the radioecological consequences after Chernobyl nuclear pow er

plant accident, Science of Total Environm ent 137 (1993) 9.

[3] NIKIPELOV, B.V., ROM ANOV, G.N., BULDAKOV, L.A., BABAYEV, N.S.,

K H O LIN A , Yu.B., M IK E R IN , Ye.I., Radiation accident in the South Urals, Atom -

naya Energiya 67 2 (1989) 74.

[4] B A LO N O V , M .I., (Ed.), Manual on the Radiation Situation and Doses Received by

Populations o f Regions of the Russian Federation Subjected to Radioactive Contam ina­

tion caused by the Chernobyl N P P Accident, Ariadna-Arcadia, St. Petersburg (1993).

[5] KO RN EYEV, N. A ., POVALYAYEV, A.P., A LEXA K H IN , R.M., PANTELEYEV, L.I.,

RATNIKOV, A.N., KRUGLOV, S.V., SANZHAROVA, N.I., ISAMOV, N.N.,

SIR O T K IN , A .N ., The agricultural production sphere — radiological consequences of

the accident at the Chernobyl N P P and basic countermeasures, Atom naya Energiya 65

2 (1988) 129. 166 POSTER PRESENTATIONS

IAEA-CN-54/90P

SURVEILLANCE DE L ’IM PACT DES REJETS MARINS DE L ’USINE DE RETRAITEM ENT DE LA HAGUE

S. LE BAR Service de prévention et de radioprotection, Compagnie générale des matières nucléaires, La Hague, France

1. LES PRINCIPES DIRECTEURS

Toute activité industrielle est susceptible d’avoir un impact sur son environne­ ment. Pour minimiser cet impact en matière de déchets, de rejets ou de dosimétrie des personnes, la politique permanente de la Compagnie générale des matières nucléaires (Cogéma) s’inscrit dans la logique du principe «As low as reasonably achievable» (ALARA) qui a guidé tant les principes de conception que les modes d’exploitation des usines. Ainsi, à La Hague, la maîtrise de l ’impact des rejets marins est assurée par la mise en place d’un système complet d’études préalables, de réglementation, de prévention, de surveillance et de contrôles. Les équipes en charge de ces activités sont indépendantes des équipes d’exploitation, et relèvent directement de la Direction de l’établissement.

2. LE CONTEXTE REGLEMENTAIRE: L’ETUDE D’IMPACT ET LES AUTORISATIONS

2.1. L ’étude d’impact

En amont de la fixation des niveaux de rejets autorisés, l’étude d’impact a fourni un référentiel permettant de définir le cadre réglementaire. L ’ensemble des données concernant la population, la courantologie, la radioécologie et les activités industrielles ont été utilisées pour la réalisation de l ’étude d’impact. L ’étude des voies de transferts de la radioactivité vers la chaîne alimentaire ainsi que l ’étude des consommations alimentaires ont permis d’évaluer l’impact sanitaire maximum des rejets en mer sur différents types de populations (urbaine, rurale, côtière, pêcheurs). La dose maximale susceptible d’être reçue par la population la plus exposée (familles de pêcheurs artisans) a été calculée pour des rejets égaux à 100% des limites d’autorisation annuelle. Dans ces conditions pénalisantes, la dose reçue ne dépasserait pas 1 % de la limite maximale admissible pour la population (5 mSv), soit une valeur inférieure aux fluctuations de la radioactivité naturelle du Cap de La Hague. POSTER PRESENTATIONS 167

2.2. Les autorisations de rejets

Le cadre réglementaire des conditions de rejets du site de La Hague est défini dans l’arrêté du 22 octobre 1980 et le décret du 28 mars 1984. Les limites fixées par les arrêtés d’autorisation publiées au Journal officiel du 10 avril 1984 sont les suivantes pour les rejets liquides: tritium (37 000 TBq), radioéléments autres que le tritium (1700 TBq), strontium 90 + césium 137 (220 TBq), radioéléments émetteurs alpha (1,7 TBq). Les textes définissent les conditions de rejet, le contrôle de ces rejets, l’organi­ sation de la surveillance de l ’environnement sous la tutelle de l ’Office de protection contre les rayonnements ionisants (OPRI), le rôle du laboratoire d’analyse, la perma­ nence des personnes qualifiées en radioprotection, etc.

3. REJETS EFFECTIFS EN MER

Le point de rejet est situé à 1700 mètres de la côte. Les rejets s’effectuent par une conduite, contrôlée dans sa totalité au moins une fois par an. Aucun rejet n’est effectué sans une analyse préalable (activités volumiques alpha et bêta, tritium, spectrométrie gamma, pH, quantité de matières en suspension, activités volumiques du strontium et du plutonium, dosage de l’uranium) d’un échan­ tillon représentatif de la totalité du volume rejeté et après homogénéisation. La décision de rejet est donnée, par écrit, par l’ingénieur radioprotection au vu des résultats d’analyses. Les activités rejetées en mer en 1993 ont été, en pour­ centage de l’autorisation annuelle: — tritium: 13,9% — émetteurs alpha: 5,9% — émetteurs bêta (hors tritium): 4,3% — ^Sr + 137Cs: 13,1%.

4. SURVEILLANCE ET CONTROLE DE L’ENVIRONNEMENT DE L’USINE DE LA HAGUE

4.1. Surveillance de l’impact des rejets marins

La surveillance de l ’environnement marin est effectuée par le Service de prévention et de radioprotection de l ’établissement de La Hague et complétée par les études effectuées par le Laboratoire d’études de radioécologie de la façade atlantique (LERFA) de l ’Institut de protection et de sûreté nucléaire (IPSN) et le laboratoire du Groupe d’études atomiques (GEA) de la Marine nationale. Le contrôle de cette surveillance est effectué par Г OPRI. 168 POSTER PRESENTATIONS

Cette surveillance consiste à prélever des échantillons représentatifs dans l ’environnement, à les analyser en laboratoire et à établir les bilans nécessaires à l'interprétation des résultats. Cette surveillance et ces études s’exercent à trois niveaux: — Le champ proche (200 km de côte): les analyses portent sur l’eau, les sables et sédiments, les algues (excellents bio-indicateurs grâce à leur faculté de fixer et concentrer des radionucléides), la végétation de bord de mer, ainsi que les mollusques et coquillages. — Le Champ moyen (en haute mer de Granville jusqu’en baie de Seine) où l’eau, le sable, les sédiments et les poissons péchés près des côtes sont prélevés et analysés. — Le champ élargi (jusqu’en mer du Nord) où le laboratoire de radioécologie marine de l’IPSN étudie in situ la dispersion des radioéléments rejetés en mer au travers d’analyses d’eaux et de sédiments. Des études en cours, effectuées dans le cadre des programmes européens MAST ou «Marine science and technology» vont ainsi permettre de dresser une car­ tographie complète des transferts de radioactivité dans les écosystèmes.

5. ANALYSES ET RESULTATS

Au total, 17 000 échantillons sont prélevés et 50 000 analyses sont effectuées chaque année au titre de la surveillance de l ’environnement. En particulier pour les rejets marins, 1000 échantillons faisant l’objet de 8000 analyses sont prélevés avant décision de rejet en mer; 1000 autres échantillons faisant l’objet de 2200 analyses sont prélevés pour la surveillance de l ’environnement marin. Pour effectuer sa mission de contrôle des rejets et de surveillance de l ’environ­ nement, le Service de prévention et de radioprotection de l’établissement de La Hague dispose: — d’une équipe de prélèvements d’échantillons. — D ’un laboratoire environnement effectuant les analyses sur les échantillons recueillis. — D ’un laboratoire de moyenne activité effectuant les analyses sur les effluents avant rejet en mer. Ces laboratoires sont équipés de deux salles de préparation chimique permettant d’effectuer les analyses spécifiques: broyage, étuvage, distillation tritium, dosage uranium, dosage du potassium par absorption atomique, préparation plutonium par séparation sur résine, dosage du stron­ tium, évaporation des liquides pour comptage alpha, bêta, électrodéposition pour spectrométrie alpha, etc. Ils participent à différents programmes d’inter- comparaison avec des laboratoires tels que ceux du Laboratoire de métrologie POSTER PRESENTATIONS 169

et des rayonnements ionisants (LM RI) et de l’Agence internationale de l’énergie atomique (AIEA). — D ’une équipe d’exploitation des résultats. Cette équipe constituée de 6 techni­ ciens et d’un ingénieur établit l ’ensemble des bilans, vérifie leur cohérence, demande les analyses complémentaires et prépare la diffusion des résultats.

L ’établissement Cogéma de La Hague effectue une large diffusion de ces résul­ tats: soit sous forme de documents écrits et diffusés mensuellement au niveau local et régional, soit au travers du réseau téléphonique national Magnuc de l’IPSN qui est accessible à tous en permanence par le Minitel, soit enfin par le réseau des bornes de communication de Cogéma installées dans les restaurants d’entreprises de l’éta­ blissement, au centre d’information du public de l’usine de La Hague et à l ’espace communication de la ville de Cherbourg.

CONCLUSION

L ’importance et la cohérence du programme de surveillance du milieu, les moyens techniques et humains mis en œuvre, une politique de transparence et de diffusion systématique des résultats, comparés et complétés par des études et des mesures d’autres laboratoires constituent des outils qui permettent de démontrer l’impact négligeable du fonctionnement de l’usine de La Hague sur son environne­ ment marin. Dix ans de mesure en mer ont mis en évidence une diminution continue de la radioactivité artificielle dans l ’ensemble des eaux de la Manche et de la mer du Nord. Cette diminution est d’autant plus significative qu’elle est intervenue dans la période de montée en puissance industrielle des installations. En effet, tandis que le tonnage retraité annuellement est passé de 250 tonnes en 1984 à près de 1000 tonnes en 1993, les rejets ont été diminués d’un facteur 5 à 10 pour les émetteurs alpha et bêta-gamma hors tritium. Aujourd’hui, la radioactivité due aux rejets de La Hague ne dépasse pas le millième de la radioactivité naturelle de l’eau de mer. 170 POSTER PRESENTATIONS

IAEA-CN-54/95P

RADIOACTIVE EFFLUENTS FROM SPANISH LIGHT W ATER NUCLEAR POW ER PLANTS: THE INTERNATIONAL SITUATION

M.J. BARAHONA, L.M. RAMOS Consejo de Seguridad Nuclear (CSN), Madrid, Spain

1. INTRODUCTION

The Spanish light water nuclear power plants are the following: José Cabrera (1968), Santa Maria de Garoña (1971), Almaraz 1 and 2 (1981/1983), Aseó 1 and 2 (1983/1985), Cofrentes (1984), Vandellós 2 (1987) and Trillo (1988). The time span considered for each of the radionuclide groups under study ends in 1992 in all cases, but begins on different dates depending on the startup date of each of the plants and the data available. The pattern of emission of radioactive effluents (Fig. 1) was analysed by: — identifying the most significant deviations from the mean trend shown by the values considered; — comparing the Spanish nuclear power plants (NPPs) with those of other Euro­ pean Community countries (for the period 1977-1991) and the USA (for the period 1978-1988). The following parameters were calculated for each of the radionuclide groups considered: — annual normalized activity of each NPP: quotient of the annual activity and annual gross power output for each NPP; — mean normalized activity of each NPP: arithmetic mean of the annual normal­ ized activity values for each NPP; — total normalized activity of pressurized water reactors (PWRs) and boiling water reactors (BWRs): quotient of the annual activity of all the PWR and BWR plants combined and their annual gross power output; — mean total normalized activity of PWRs and BWRs: arithmetic mean of the annual normalized activity values for PWR and BWR plants. Analysis of the pattern of effluent emission by the Spanish NPPs reveals certain isolated increases and other more or less prolonged trends which can be attributed to different incidents and situations occurring at the various power plants. POSTER PRESENTATIONS 171

Liquid effluents of PWR plants Liquid effluents of BWR plants Total excluding tritium Total excluding tritium U3E«09| L0E<09|

UC*OJ tee«03|

« * • 0 1 1 L0E

LOE *00; t-OE-OO ___ _ EEC EFC ^ ж Spain ч *^4 LOE *01 \ 4 , f. U6A , T • -j * ■ "«-w tce-02 Spain x ' / у 106*021 LOE-03 t.OE-03 \ LOE-O« w e -04

106-06 uoe-oe

L0E *06

Years Years

Gaseous effluents of PWR plants Gaseous effluents of 8WR plants Noble gases Noble gases

10E«06| 10E

l «* oo¡ изе«ов| S p a in EEC ME«04| 106*041 \ ^ \ < A LflE«ttj LOE 4M i x * v ______L0E«H U E O i S p , i B LŒ«01 j L0E<1

10E<00, V0&00|

U 6-01 106-01

106*01 LOE-02

U E 4 I 106*09* •o « П Г4 N П «0 Y e a r s Years

Gaseous effluents of PWR plants Gaseous effluents of BWR plants 1 3 1 , 1 3 1 ,

10E«00| 10c«00¡

LOE-01 LOE *01

loe-o ji LOE *02 цел

lœ -o ii L06-06

L0C *04 i LOE-04

LOE-06 LOE-Об S p a ™ ^ ~ \ ¿ *

tœ -oe v 4 ^ - LOE-Об

LOE «07 LOE-07

LOE-06 LOE-06

LOE*00* LOE *00 •o ю Years Y e a r s

FIG. 1. Effluents (activity in G B q/G W 'h) o f N PPs in Spain, other European Com m unity

countries and the USA. 172 POSTER PRESENTATIONS

2. LIQUID EFFLUENTS

2.1. Total activity excluding tritium

Total normalized activity excluding tritium at the Spanish PWR plants fluc­ tuates around 6.6E —3 GBq/GW-h, some values being quite steady. In the BWR plants a downward trend is observed with values between 1E+0 GBq/GW-h and 7E—5 GBq/GW-h (although the range is narrower after discounting the contribution of the first years of operation of the Santa Maria de Garoña plant prior to the safety review conducted there). In Spain and the European Community the total normalized activity, excluding tritium, at the PWR plants is almost always below that of the BWR plants. In the USA, however, the two types of plant show similar values. For both the PWR and BWR plants, the Spanish values are generally lower than those of the USA and European Community, although the PWR values are very close to those of the USA.

2.2. Tritium

The total normalized activity of tritium at the Spanish light water reactor (LWR) plants shows a steady trend, with values within the range of 1-4 GBq/GW-h in the case of the PWR plants and 0.01-0.3 GBq/GW-h for the BWR plants. In Spain, the European Community and the USA the normalized activity of tritium for the PWR plants is greater, for 10 years at least, than that for the BWR plants. At both the PWR and BWR plants the Spanish values are below those recorded for the European Community and the USA, although they are very close to the latter, especially in the case of the PWRs.

3. GASEOUS EFFLUENTS

3.1. Noble gases

The total normalized activity of noble gases at the Spanish LWR plants has a more or less steady trend if, in the case of the BWR plants, the contribution of the first years of operation of the Santa Maria de Garoña plant is discounted. In the case of the PWR plants, the total normalized activity of noble gases fluc­ tuates around 36 GBq/GW-h, while for the BWR plants it fluctuates around 67 GBq/GW -h. If all the years of operation of the Santa Maria de Garoña plant are taken into account, the total normalized activity of noble gases at the BWR plants fluctuates around 6 TBq/GW-h. POSTER PRESENTATIONS 173

In both Spain and the USA the total normalized activity of noble gases at PWR plants is lower than that recorded for the BWR plants. However, in the European Community the PWR values are slightly higher. For the PWR plants the Spanish values are similar to those of the USA and lower than those of the European Community. The BWR values are all similar, with the US values slightly higher than the others.

3.2. Iodine-131

The total normalized activity of 131I at the Spanish PWR plants shows a slightly upward trend for the past 6 years, with values fluctuating around 44 kBq/GW-h. At the BWR plants the total normalized activity of 131I shows pronounced fluctuations, with values between 4 kBq/GW-h and 0.3 MBq/GW-h. In Spain and the European Community the normalized activity of 13II at the PWR plants is lower but still similar to that for the BWR plants. However, in the USA the values for the PWR plants are appreciably lower than those for the BWR plants. The normalized activity values for 131I in the USA are higher than those in the European Community at both the PWR and BWR plants, the Spanish values falling between the two.

3.3. Particles

The total particulate activity at the Spanish LWR plants shows a slight down­ ward trend, most of the values being between 4 kBq/GW-h and 6 MBq/GW-h. In Spain and the European Community the normalized particulate activity at the PWR plants is similar to that recorded for the BWR plants. In the USA the PWR values are slightly lower than those for the BWR plants. The total normalized particulate activity in Spain is similar to that for the European Community and lower than that for the USA, especially in the case of the BWR plants.

3.4. Tritium

The total tritium activity at the Spanish LWR plants has a more or less steady trend, with values between 0.04 and 2.5 GBq/GW -h in the case of the PWR plants, and between 0.02 and 0.13 GBq/GW-h for the BWR plants. In Spain, the European Community and the USA, the normalized tritium activity at the PWR plants is similar to that recorded for the BWR plants. The Spanish values for the BWR plants are similar to those of the European Community and below those of the USA. 174 POSTER PRESENTATIONS

At the PWR plants, between 1975 and 1982, the Spanish values are similar to those recorded for the USA, although occasionally a little higher, and are apprecia­ bly higher than those of the European Community. However, between 1983 and 1992 the Spanish values fall below those of the USA and, while remaining higher than the European Community values, come very close to them. POSTER PRESENTATIONS 175

IAEA-CN-54/98P

ACTION LEVELS FOR RADON EXPOSURES CAUSED BY RADON EXHALATION FROM RELICS OF URANIUM M INING AND M ILLING

W. KRAUS, M. KÜMMEL, S. PRZYBOROWSKI, W. RÖHNSCH Federal Office for Radiation Protection, Berlin, Germany

1. INTRODUCTION

Ore and other raw material mining and processing lasting for centuries, above all the intensive uranium ore mining and milling after World War П, have left a great number of relics such as waste rock deposits and tailings ponds in the affected regions of Saxony and Thuringia. In many cases these relics contain radionuclides of the uranium decay chain in above average concentration. This situation arose during a time when regulations for radiological protection did not exist, were not at the latest standard or were not observed. In addition, the radiological situation in the mining districts is very complex in view of the natural geological and topographic conditions which influence the exposure level. Moreover, the districts are densely populated and intensively used for industrial and agricultural purposes. To clarify the situation and to meet the public concern about adverse health effects, a Federal investigative project has been started [1]. The relevance of the radiation exposure of the public owing to the relics is assessed based on recommen­ dations of the German Commission on Radiological Protection [2]. The Commission classified the situation as of a ‘pre-existing’ type in which interventions must be con­ sidered to decrease the radiation exposure of the public to an acceptable level, if required. Remedial measures should be considered if the effective dose owing to the mining relics exceeds the reference level of 1 mSv/a. Independent of this level, a reference level of 250 Bq/m3 for the indoor radon concentration should be observed. One of the pathways investigated and assessed is exposure to radon exhaled by the relics. Whereas the radon emission from waste rock piles and the exposure caused thereby is generally negligible, the situation as to tailings and certain other residues needs careful consideration. In this, two peculiarities must be taken into account: — The natural outdoor radon concentration in the mining regions is relatively high for geological reasons and shows a considerable range. 176 POSTER PRESENTATIONS

— The largest part of the exposure caused by outdoor radon occurs within houses, because air exchange transforms outdoor to indoor radon and people spend more time indoors than outdoors.

For these reasons, evaluation of radon exposure in the context of overall source related exposure involves special difficulties and needs specific treatment.

2. NATURAL OUTDOOR RADON CONCENTRATION

In order to measure the long term normal geogenic outdoor radon concentra­ tion and the contributions from mining relics, intensive investigations have been carried out since 1983 in certain districts of Saxony and Thuringia. About 480 mea­ suring points were established on or adjacent to mining sites or affected grounds as well as at increasing distances from them. Diffusion chambers were used equipped with solid state nuclear track detectors. The exposure time of the detectors was about half a year (spring-summer and autumn-winter). The investigations led to the conclusion that for distances greater than 500 m from the sites the outdoor radon concentration is not measurably enhanced by radon exhalation from mining relics. Based on this, the natural geogenic radon concentra­ tion range was determined for different mining regions. Results are shown in Table I [3]. The natural outdoor radon concentration corresponds approximately to a log normal distribution (Fig. 1) and reaches at least 80 Bq/m3. This level must be con-

TABLE I. OUTDOOR RADON CONCENTRATION IN THE ENVIRONMENT OF RELICS OF URANIUM MINING AND MILLING IN SAXONY, GERMANY

N o . o f Radon concentration (Bq/m 3) D i s t a n c e R e g i o n measurement t o r e l i c s p o i n t s S t a n d a r d R a n g e A v e r a g e d e v i a t i o n

> 5 0 0 m G o t t e s b e r g 5 1 4 - 6 9 4 0 1 7

J.-G.-Stadt 2 8 8 - 8 0 3 0 1 3

L e n g e n f e l d 2 1 1 0 - 8 5 3 4 1 7

M a r i e n b e r g 2 1 1 0 - 6 1 2 6 1 3

< 5 0 0 m G o t t e s b e r g 3 1 5 - 7 1 3 4 1 6

J.-G.-Stadt 3 0 1 0 - 1 0 0 0 7 8 1 4 0

L e n g e n f e l d 8 1 0 - 1 3 0 5 1 3 1

M a r i e n b e r g 1 4 1 0 - 5 1 2 8 1 4 POSTER PRESENTATIONS 177

Logarithm of radon concentration (Bq/m 3)

FIG . 1. C um ulative frequency o f background outdoor radon concentration.

sidered as the upper end of the ‘normal range’ of the natural outdoor radon concen­ tration in these regions. Against this level, contributions of mining relics must be differentiated and evaluated.

3. INDOOR EFFECT OF OUTDOOR RADON

Indoor-outdoor air exchange influences the indoor radon concentration of houses, in the equilibrium case adding an outdoor contribution to the concentration stemming from ‘internal’ sources, e.g. from underground below the building: 178 POSTER PRESENTATIONS

where C¡ is the indoor concentration in a room, Cq is the outdoor concentration, E is the average exhalation rate per surface unit of the room, S is the exhaling surface, V is the room volume, and к is the air exchange rate. Therefore, as a conservative approximation one can assume that an additional contribution by a mining relic to the outdoor radon concentration corresponds to an additional indoor radon concentration in houses built nearby. The main component of human radon exposures results from staying within the residential building since the average duration of indoor stay is supposed to be 80% of the day. That is why outdoor radon concentrations should be evaluated as if a source related additional component contributes to the indoor radon exposure. The radon exposure caused by the relevant mining relic can then be dealt with as if completely incurred all day under indoor conditions. This is certainly a conser­ vative approach since the outdoor equilibrium factor of radon is low near the radon source. Under these conditions, and making use of the dose conversion convention proposed by ICRP-65 [4], the average outdoor radon concentration corresponding to an effective dose of 1 mSv/a amounts to about 50 Bq/m3.

4. CONCLUSIONS

Because the effect of the outdoor radon from mining relics occurs in houses, the logical conclusion is to treat it — from the radiological point of view — in the same manner as the indoor radon exposure caused by other geogenic or technogenic sources. For indoor radon the German Commission on Radiological Protection has defined a ‘normal range’ of concentration up to 250 Bq/m3 [5] derived from the measured frequency distribution in German houses. Below this level, no remedial measures are considered necessary. From the same basic idea, i.e. orientation to the natural range of variation, the outdoor radon must be radiologically evaluated by set­ ting the upper end of the regional natural range as a reference level, thus separating the ‘no action’ from the ‘action’ range. But a general problem remains with respect to outdoor radon emitted from mining relics: there is somebody responsible for this particular radon source and for the resulting exposure, and consequently for remedial measures at the source, if required. That makes it necessary to define a source related contribution to the outdoor concentration at the sites of houses which is acceptable from the radiological point of view and above which remedial measures must be considered. Taking into account all the aspects described, the following conclusions can be drawn with respect to radon exhaled from mining relics in the mining regions of Saxony and Thuringia: POSTER PRESENTATIONS 179

(1) A reference level of 80 Bq/m3 is recommended for outdoor radon in these regions. If the measured concentrations are below this level, no remedial actions are required with respect to the radon pathway. If the long term outdoor radon concentrations exceed this level, an attempt should be made to determine the source related contribution to the outdoor concentration at the residential site using site specific modelling of radon dispersion. (2) An action level of 50 Bq/m3 is recommended for the contribution of mining relics to the outdoor radon concentration near existing houses, potential build­ ing sites or other locations of permanent human occupation. This value makes sense for two reasons: — The average natural outdoor radon concentration is approximately 30 Bq/m3, and 50 Bq/m3 corresponds to the difference from 80 Bq/m3, the upper end of the normal geogenic variation. Since above this range source related considerations exist, on average without such considera­ tions nobody should be exposed to more than 50 Bq/m3, i.e. to more than 1 mSv/a, by the mining relic. — A concentration of 50 Bq/m3 corresponds to only 20% of the reference level for indoor radon concentration in Germany. A lower intervention level for the source related component seems to be irrelevant. In the way described, i.e. considering the radon exposure caused by mining relics in terms of the resulting outdoor concentration as well as of the emitting source, a useful procedure for a reversible evaluation of both the source and the impact is available.

REFERENCES

[1] RÖHNSCH,W ., ETTENHUBER, E., “Relics of mining and milling in eastern

Germ any and their radiological consequences” (Proc. Int. Sym p. on Rem ediation and

Restoration of Radioactive-Contam inated Sites in Europe, 1993, Antwerp, Belgium ),

EC Doc. XI-5027/94 (1994) 119.

[2] GERM AN COM M ISSION ON RADIOLOGICAL PROTECTION, Strahlenschutz­

grundsätze für die Verw ahrung, Nutzung oder Freigabe von kontaminierten M ateri­

alien, Gebäuden, Flächen oder Halden aus dem Uranerzbergbau, Veröffentlichungen

der Strahlenschutzkom m ission, Vol. 23, Gustav Fischer Verlag, Stuttgart (1992).

[3] DUSHE, C., K Ü M M EL, M., Untersuchungen zur Bestimmung des natürlichen

Rn-222-Freilufitpegels in den Bergbaugebieten Sachsens, Thüringens und Sachsen-

Anhalts, Bundesam t für Strahlenschutz, Jahresbericht 1993.

[4] INTERNATIONAL COM M ISSION ON RADIOLOGICAL PROTECTION, Protec­

tion against radon-222 at hom e and at w ork, IC R P Publication 65, Pergam on Press,

Oxford and New York (1993).

[5] GERM AN COM M ISSION OF RADIOLOGICAL PROTECTION, Strahlenschutz­

grundsätze zur Begrenzung der Strahlenexposition durch Radon und seine Zerfalls­

produkte in Gebäuden, Em pfehlungen der Strahlenschutzkom m ission, Bundesanzeiger

No. 155 (1994) 8766. 180 POSTER PRESENTATIONS

IAEA-CN-54/108P

SOIL-W OOD 137Cs TRANSM ISSION CO EFFICIENT FOR UKRAINIAN CONIFEROUS FORESTS

V. POIARKOV, A. NAZAROV, A. ZAGREBIN, N. KALETNIK Ukrainian Radiation Training Centre

N. DA VIDOV Forest Ministry of Ukraine Kiev, Ukraine

1. INTRODUCTION

After the Chernobyl accident, nearly 7 tons of irradiated reactor fuel was released into the environment. The total activity of discharged radionuclides was about 2 X 1018 Bq. A large part of the highly contaminated area is forest, with 137Cs contamina­ tion of 1-5 Ci/km2 (1 Ci/km2 = 37 kBq/m2). The main part of the activity in the forest is still in the forest litter and top layer of soil [1, 2]. Because of radionuclide migration, all elements of the forest ecosystem are also contaminated. In many cases the contamination level of berries, mushrooms, wood and other forest products could be higher than the permissible level. It is particularly important to be able to predict the level of contamination of wood in the contaminated area. Many factors influence the uptake coefficients of radionuclides from soil to wood. But it is complicated to obtain in each case sufficient information for assess­ ment of all these factors. For routine purposes it is more useful to obtain an average value of the uptake coefficients for a large number of samples. To obtain such coeffi­ cients for soil where the density of contamination is 1-5 Ci/km2 was the main aim of this research.

2. SAMPLING AND ACTIVITY MEASUREMENTS

Sampling technique depends on the aims of the monitoring. In this study the major interest was sampling of soil and wood (or other types of vegetation). The soil samples were collected for the following main reasons:

— assessment of radionuclide storage in the contaminated area; — prediction of wood and vegetation contamination. POSTER PRESENTATIONS 181

To meet the first objective, three different techniques of sampling were used: for forest, for pastures or other nondisturbed sites, and for cultivated fields. In all cases it is very important to determine total radionuclide activity in the area. The depth of sampling must be greater than the expected depth of fallout penetration (20 cm for forest). For forest, samples of soil and litter were separately collected, because of differences in the following treatment of the samples. Representative samples of soil and forest litter were taken on flat sites of a forest with typical vegetation for a given site in terms of type, age and plant density. An additional criterion was that external gamma radiation did not vary by more than 20% within a site. For prediction of radionuclide transfer in wood, soil samples were taken layer by layer at depths 0-5, 2-5, 5-10, 5-15 and 15-20 cm at each point. Samples were considered to be representative if the parameters of interest were close to the mean values characteristic for a given locality.

1— I— I— I— i— I— ГН — I I I I j I I I I I I I j I— I— I— ¡“

0 .2

с ® 0.15 cr 2 LL

0.1

0.05

0 _L_i__ i_i__L_i_____ i_ j ______I_i______i______i______I______i______i______i______I______i______i______i______I______i______i______i______I______-1 3 7 11 15 19 23 Transmission coefficient (m2/kg) (« 1E-3)

FIG. 1. Frequency histogram of transmission coefficients. 182 POSTER PRESENTATIONS

Each of the samples of forest litter consisted of five cores 14 cm in diameter. A complex sample of each soil layer for each sampling point was obtained by averaging ten core samples of 6.2 cm diameter. The samples were placed in plastic bags after sampling. Also, at each site, samples of wood were collected from a typical tree for a given locality. These samples were selected from different heights of the tree (approximately 0.5, 3 and 6 m) and then separately measured. All samples were air dried at 40-50°С in the laboratory for a few days. In the laboratory all samples were dried at 100°C. The vegetation samples were milled and placed in vials of 1000 ml volume. The soil samples were crushed and homogenized manually and then sieved to less than 2 mm size, and subsamples were taken for radiometric analysis. All samples were counted using Ge(Li) detectors with the aim of obtaining the content of caesium radionuclides measured over a time of 1-3 hours. Specific mass activity of soil and forest litter was then recalculated as density of sur­ face contamination for the investigated area. The density of surface contamination (Qs) is determined by the following formula:

Qs = % — where Qm is the mass specific activity of the soil or forest litter, S is the square of sampling, and M is the mass of the sample collected for each layer.

3. RESULTS

Figure 1 shows the frequency distribution of the transmission coefficient (TC) values of I37Cs from soil to wood. This distribution can be fitted by a log normal dependence:

TC = — Q s where Qm is the specific activity of wood. There is no significant correlation between TC and type of soil. The main result of this research is that TC varies widely for areas with a den­ sity of contamination of 37-200 Bq/m2. Minimum and maximum TC values are 0.0005 and 0.020 m2/kg, respectively. POSTER PRESENTATIONS 183

REFERENCES

[1] M ILLE R , K.M ., K U IPER , J.L., H ELFER , I.K., 137Cs fallout depth distribution in

forest versus field sites: Im plication for external gam m a dose rate, J. Environ. Radioac­

tivity 12 (1990) 23.

[2] POIARKOV, V.A., N AZARO V, N.A., KALETN IK, N.N., Post-Chernobyl radio­

m onitoring of Ukrainian forest ecosystem s, J. Environ. Radioactivity (accepted for

publication). 184 POSTER PRESENTATIONS

IAEA-CN-54/117P

ASSESSM ENT OF RADIATION EXPOSURE FROM DISCHARGES OF URANIUM ORE CONCENTRATE PROCESSING AND NUCLEAR FUEL REPROCESSING FACILITIES

D.J. ASSINDER, S.M. MUDGE, G.S. BOURNE School of Ocean Sciences, University of Wales (Bangor), Menai Bridge, Gwynedd, United Kingdom

1. INTRODUCTION

British Nuclear Fuels pic (BNFL) operates a uranium ore concentrate process­ ing plant at Springfields, Lancashire, UK. The company is authorized to discharge naturally occurring beta and gamma emitting radionuclides into the nearby Ribble estuary that arise from the purification of uranium ore concentrates. The principal radionuclides identified are 234U, 238U, 228Th, 230Th, 232Th and their decay products, 234Th and 234mPa (from 234Th decay). There are small amounts of artifi­ cial radionuclides which are discharged from the utilization of reprocessed uranium supplied by BNFL, Sellafield, principally 237Np and "Tc. Radionuclides derived from Sellafield discharges, 80 km to the north, are also found in the Ribble estuary via transport through the Irish Sea, principally 137Cs, 238Pu, 239>240Pu and 241Am. The School,of Ocean Sciences, University of Wales (Bangor), operates a well equipped radiochemical laboratory examining many aspects of natural and artificial environmental radioactivity in the marine and terrestrial environments. As part of a radiological assessment of the Ribble estuary for Her Majesty’s Inspectorate of Pollution concentrating mainly on external and inhalation dose and leisure occupancy, beta dose equivalent and gamma air kerma rates were measured and surface sediment samples were taken at many sites during periods of both discharge and plant shutdown. Full results and methodology of this and previous studies of the Ribble River are presented elsewhere [1-5]. This paper presents a brief overview of the results obtained and discusses the local public perception that Springfields dis­ charges are predominantly responsible for doses around the Ribble estuary, whereas the major contributor is Sellafield derived radionuclides. POSTER PRESENTATIONS 185

TABLE I. CONTRIBUTION TO THE BETA DOSE EQUIVALENT AND GAMMA AIR KERMA RATES FROM SPRINGFIELDS, SELLAFIELD AND THE NATURAL BACKGROUND3

E x p o s u r e P e n w o r t h a m S a v i c k L o n g t o n L y t h a m

Gam ma nGy/h (%)

B a c k g r o u n d 6 4 ( 3 9 ) 6 4 ( 3 5 ) 6 4 ( 3 1 ) 6 4 ( 6 8 )

Springfields 7 0 ( 4 3 ) 8 0 ( 4 3 ) 1 0 ( 5 ) 1 0 ( 1 1 )

S e l l a f i e l d 3 0 ( 1 8 ) 4 0 ( 2 2 ) 1 3 0 ( 6 4 ) 2 0 ( 2 1 )

T o t a l d o s e 1 6 4 ( 1 0 0 ) 1 8 4 ( 1 0 0 ) 2 0 4 ( 1 0 0 ) 9 4 ( 1 0 0 )

Beta /iSv/h (% )

B a c k g r o u n d 0 . 1 6 ( 2 ) 0 . 1 6 ( 1 ) 0 . 1 6 ( 2 9 ) 0 . 1 6 ( 1 5 )

Springfields 7 ( 9 5 ) 1 0 ( 9 6 ) 0 . 1 ( 1 8 ) 0 . 6 ( 5 7 )

S e l l a f i e l d 0 . 2 ( 3 ) 0 . 3 ( 3 ) 0 . 3 ( 5 3 ) 0 . 3 ( 2 8 )

T o t a l d o s e 7 . 3 6 ( 1 0 0 ) 10.46 (100) 0 . 5 6 ( 1 0 0 ) 1 . 0 6 ( 1 0 0 )

a Values were assigned to each source by observing changes in the dose rates between

sam pling visits.

TABLE П. PUBLIC RADIATION EXPOSURE ABOVE NATURAL BACKGROUND FOR EXTERNAL AND INHALATION PATHWAYS FOR LEISURE ACTIVITIES AROUND THE RIBBLE ESTUARY

G a m m a B e t a I n h a l a t i o n T o t a l U s e r s S i t e ( / i S v / a ) ( f i S v / a ) ( f i S v / a ) ( f i S v / a )

A n g l e r s P e n w o r t h a m 1 4 . 1 5 . 9 0 0 . 8 2 0 . 8

W ildfow lers B a n k s 1 5 . 3 1 . 7 0 . 3 1 7 . 3

L o n g t o n 3 4 . 1 1 . 7 0 . 4 3 6 . 2

L y t h a m 3 0 . 9 3 . 9 0 . 4 3 5 . 2

W a l k e r s P e n w o r t h a m 8 . 1 2 . 7 2 1 2 . 8

L y t h a m 6 0 . 6 1 0 2 . 7 7 3 . 3

S a v i c k 6 . 5 0 . 7 0 . 1 7 . 3 186 POSTER PRESENTATIONS

2. SURFACE SEDIMENT ACTIVITY CONCENTRATIONS

Surface sediment activity concentrations were measured by gamma and alpha s p e c tro m e try fo r 137C s , 238P u , 239,240P u a n d 241A m ( S e lla fie ld d e riv e d ) a n d 228T h , 230T h , 232T h , 234T h , 234mP a , 234U a n d 238U ( S p r in g fie ld s d e riv e d ). A c t iv it y c o n c e n ­ trations for Sellafield derived and Springfields derived radionuclides apart from 234Th and 234mPa showed significant relationships to the sediment grain size (% < 6 3 /xm) with higher levels in finer grained sediments at upstream sites in the estuary. For 234Th and 234mPa, the relationship with grain size was obscured by the effect of the short half-life for 234Th, but the highest activity concentrations were also found at upstream sites with fine grained sediments and were more likely to show variability at any given grain size. Repeated surveys showed that activity con­ centrations fluctuated over relatively short periods of time because of sediment redis­ tribution throughout the estuary by river runoff variations and tidal cycle effects.

3. BETA DOSE EQUIVALENT AND GAMMA AIR KERMA RATES

Maximum total dose rates arising from radionuclides discharged from B N FL Sellafield and Springfields were 204 nGy/h in air for gamma emissions at 1 m above the sediments and 19 /¿Sv/h to skin for beta emissions at 30 cm above the sediments. The maximum gamma air kerma rate was principally (64% ) due to Sellafield derived radionuclides and was located on the established salt marshes. The maximum beta dose equivalent rate was almost entirely (96% ) due to Springfields derived radionu­ clides and was associated with fine grained sediments in the upper reaches of the estuary (Table I). Using occupancy data derived from habit surveys, the maximum annual whole body dose for leisure activities was calculated as 73 цSv/a for people walking on the intertidal sediments. This was half the dose received by the critical group of houseboat dwellers assessed annually by the M inistry of Agriculture, Fish­ eries and Food (U K). For people walking in Penwortham Park and fishing along the banks of the Ribble estuary, the doses were less than 21 /xSv/a (Table П).

4. PERCEPTION OF RISK

O f the total dose above natural background (Table II), the majority was derived from the gamma emitting radionuclides of Sellafield origin. Even though the surface sediment activity concentrations for 234mPa can exceed 1 M Bq/kg, the maximum beta dose equivalent component of the total dose is 16.5 % . It is, therefore, the artifi­ cial radionuclides discharged from Sellafield into the marine environment 80 km dis­ tant that produce the largest increase in public dose in the Ribble estuary. This is in contrast to the local public perception, enhanced by newspaper and television reports, that it is the Springfields plant which is mostly responsible. POSTER PRESENTATIONS 187

REFERENCES

[1] ASSINDER, D.J., YAMAMOTO, M., KIM, C.K., SEKI, R., TAKAKU, Y., Y A M A U C H I , Y ., K O M U R A , K ., U E N O , K ., Radioisotopes of thirteen elements in intertidal coastal and estuarine sediments in the Irish Sea, J. Radioanal. Nuclear Chem. Articles 170 (1993) 333. [2] AS S IN D E R , D.J., M U D G E , S.M ., B O U R N E , G.S, Radiological assessment of the Ribble estuary. I. Distribution of radionuclides in surface sediments, J. Environ. Radio­ activity (in preparation). [3] M U D G E , S .M ., B O U R N E , G .S., A S S IN D E R , D.J., Radiological assessment of the Ribble estuary. II. Beta dose equivalent and gamma air kerma rates and doses to critical groups, J. Environ. Radioactivity (in preparation). [4] B O U R N E , G .S., A SSIN D ER , D.J., M U D G E , S .M , Radiological assessment of the Ribble estuary. III. Redistribution of radionuclides associated with fine-grained sedi­ ments, J. Environ. Radioactivity (in preparation). [5] MUDGE, S.M., BURGESS, P., ASSINDER, D.J., BOURNE, G.S., Measurement of beta dose equivalent and gamma air kerma rates in the Ribble estuary, J. Radiol. Protec­ tion (in preparation).

Technical Session 4 PERCEPTION OF RADIATION RISK

IAEA-CN-54/4P

PERCEPTION OF NATURAL, M EDICAL, AND ‘ARTIFICIAL’ RADIATION EXPOSURES

K . B E C K E R ISO/TC 85 ‘Nuclear Energy’, c/o DIN , Berlin, Germany

For somebody who has been involved for 35 years in various continents and various responsibilities in radiation protection, it remains difficult to understand why such incredibly different weighting factors are attached by the media, and consequently by many politicians and non-technical organizations as well as by the general public, to the risks associated with the three main sources of population exposures, namely:

— ‘natural’ radiation, amounting to about 75% of the total collective radiation dose during the past 50 years [1-3], with very large local fluctuations with or without modifications by human activities (e.g. up to 1:106 in ambient radon concentrations, apparently without detrimental health effects for anybody but very heavy smokers in very high radon environments); — diagnostic or therapeutic medical exposures amounting to about 24% of the collective radiation dose, which could easily be substantially reduced in many cases by simple means, without any loss of benefit; and — the very small additional ‘artificial’ component (mostly due to early nuclear weapons tests, the nuclear fuel cycle, the Chernobyl accident and industrial uses of radiation) amounting to a total of less than 1 % of the above.

The imbalance in public perception and its consequences may be illustrated with a few examples:

(1) The global population dose (collective effective dose equivalent commitment) caused annually by the complete nuclear fuel cycle amounts to less than 3000 man • Sv. This compares to 110 000 man • Sv (or about 40 times as much) due to the use of coal, and 300 000 m an-Sv (or 100 times as much) due to the phosphate fertilizer industry worldwide (see U N SC EA R data in Ref. [4]). The total dose caused by the Chernobyl accident compares with about four times this dose caused since 1986 by the phosphate industry.

This paper reflects the personal opinion of the author, and not of any organization with which he is associated. Valuable comments from numerous colleagues are gratefully acknowledged.

191 192 POSTER PRESENTATIONS

(2) Depending on the ingestion path, many natural and artificial non-radioactive compounds are by a factor of 104 to 1010 more toxic for humans than the ‘ultrapoison’ plutonium, which has recently again been described by many media and politicians as the most dangerous material on earth. In fact, perfluor buten (resulting from burning Teflon), ricin from plants, as well as toxins from fish and snakes and bacteria have an L D 50 which is much smaller than that of plutonium [5], and its frequently quoted long half-life is obviously far exceeded by the infinite half-life of stable elements and compounds. (3) The average additional dose to workers in the nuclear industry (about 2 mSv/a) equals almost exactly that of airline crews [6]. (4) W hile the authorities of one large country propose a limit for radon in public water supplies of 10 Bq/L, in popular health spas of various other countries hundreds of thousands of patients drink and bathe in water, to their obvious benefit [7], which contains up to 12 000 Bq/L. The currently proposed maxi­ mum permissible radon level in the work place is about 1000 Bq/m3 in air, but the 24 h average in a hall of the public water treatment plant of H of in Bavaria is 730 000 Bq/m3 [8]. (5) It has been well established that the death toll from the Chernobyl accident is currently fewer than 50 persons, with the possibility of a few hundred addi­ tional future thyroid cancers in the Gomel area. Nevertheless, claim s of very high casualities and strange health effects — most recently of trisomy 21 in children bom in West Berlin after the accident [9] — are published and seriously discussed by many scientists without any regard for dose and radio­ biological considerations. (6) Recently, an Asian country refused a gift from the European Communities of 385 t of a valuable high vitamin wheat/soybean compound because it “ was a clear case of the disposal of contaminated food in a poor country” , despite certificates that its activity was below 50 Bq/kg. Instead of feeding the popula­ tion, it was shipped from Chittagong to Singapore [10]. (7) After the Chernobyl accident, Germany spent about US $40 million on the ‘decontamination’ and disposal of a cattle feed additive containing less activity than many commercial fertilizers. The plant which had been developed and built for this purpose was, because nobody wanted it even as a gift, láter sold as scrap for about 1 % of its construction costs. (8) ‘Exemption levels’ currently recommended by the ICRP, the Basic Safety Standards, etc. amount to about 0.4% of natural exposures. This value appears to be extremely low compared with the fluctuations in the natural exposure levels, and in view of the potential unnecessary expenses of billons of dollars, e.g. in decommisioning nuclear facilities in Germany, if such recommended values are actually applied.

Many more such examples of the waste of monetary and manpower resources due to an irrational phobia, in particular of ‘artificial’ radiation, could be given (e.g. POSTER PRESENTATIONS 193 see Ref. [11]). Such behaviour may be seen as one of the many meaningless luxuries which only a few rich countries are able to afford, but it is unfortunately spreading, and obviously affecting attitudes of developing countries also. For example, at least one developing country is planning a sophisticated country-wide nuclear accident monitoring network at a cost which would be of substantially higher value in other areas of its public health system. After all, in more than half of the w orld’s countries even the basic rudiments of radiation protection are lacking, and could be introduced with a fraction of the funds that are unnecessarily spent in some of the more affluent countries [12]. And most of the recent accidents involving radiation sources have occurred in developing countries. There has been much speculation about the reasons for this unbalanced percep­ tion and these reactions, e.g. the basic psychological problems of less educated per­ sons with adequate risk evaluation [13], or changing patterns in the risk acceptance of society [14]. In the author’s opinion, the main reasons for the widespread lack of a more realistic understanding of radiation risks can be found both on the side of the radiation protection community and among the non-expert opinion makers and the public. Examples of the first category are:

— overconservative, often politically motivated recommendations and regulations for radiation protection (such as no-threshold assumptions and overestimates of radon risks, where rather arbitrary ‘intervention levels’ without sound radioepidemiological basis may affect not only the local real estate markets, but whole national economies, as in deep gold mining in South Africa); — economic interests of the ‘radiation protection industry’, such as instrument manufacturers, monitoring services, and radiation research instititions, result­ ing in a strong overemphasis on radiation compared to other risks.

Into the other category fall:

— overrepresentation in the media of a small number of ideologically motivated anti-nuclear activists, frequently lacking scientific background and integrity (self-appointed experts such as ‘critical’ or ‘concerned’ scientists, etc.); — among journalists and editors, a lack of scientific education for a sufficiently qualified judgement and frequently a tendency towards unqualified sensa­ tionalism (mostly for commercial reasons); and — on the part of the general public, a preference for sensational and frightening stories over substantial and serious information, combined with a very scepti­ cal attitude towards science and technology, considered more fashionable among ‘intellectuals’ than understanding of facts. In some countries such atti­ tudes are also supported by many teachers, ‘progressive’ parts of the clergy, and politicians.

However, with some patience and personal courage, radiation protection specialists may be able to correct this unfortunate situation to a large degree. In the 194 POSTER PRESENTATIONS meantime, one could follow the conclusion of some Swedish experts in this field [15]: “ It is recommended that persons who appear unduly fearful of radiation are handled with care and are recommended professional medical treatment” .

REFERENCES

[1] U N IT E D N A T IO N S , Sources and Effects of Ionizing Radiation (Report to the General Assembly), Scientific Committee on the Effects of Atomic Radiation (U N SC EA R ), Publication No. E 94.IX.2, UN, New York (1993). [2] K A U L , A . , K R A U S , W . , SCHM ITT- HANNIG, A . , Exposure of the public from man- made and natural sources of radiation, Kerntechnik 59 3 (1994) 98-104. [3] G O N Z A L E S , A.J., Global levels of radiation exposure: latest international findings, IA E A Bull. 4 (1993) 42-51. [4] S T E IN H Ä U S L E R , F., “ Technologically enhanced natural radiation and the signi­ ficance of related risks” , High Levels of National Radiation (Proc. Conf. Ramsar, Iran 1990), A E O I, Iran (1993) 163-175. [5] S T O L L , W ., B E C K E R , K ., Ultrapoison Plutonium? Atomtechnische Wissenschaft (April 1989) 170-177 (in German). [6] NATIONAL RADIOLOGICAL PROTECTION BOARD, Radiation at Work: Low Specific Activity Scale in the Oil Industry, U K N R P B , Chilton, Didcot (1991). [7] PRATZEL, H.G., LEGLER, B., AURAND, K., BAUMANN, K., FRANKE, T., Wirksamkeitsnachweis von Radonbädem im Rahmen einer kurort-medizinischen Behandlung des zervikalen Schmerzsyndroms bei degenerativen HWS-Veränderungen. Eine prospektive, randomisierte Doppelblindstudie, Phys. Rehab. Kur. Med. 3 (1993) 76-82 (in German). [8] B E C K E R , D .E ., R E IC H E LT , A ., RIEPL, S., Professional Radiation Exposure in a Water Treatment Plant, T Ü V Bayern/Sachsen (1992) (in German). [9] SPERLIN G, K ., Brit. Med. J. 309 (1994); “ Verwirrende Erbschäden” , Frankfurter Allgemeine Zeitung (3 August 1994) p. N1. [10] German Bundestag, parliamentary question by I. Irmer and answer by H. Schäfer (21 June 1994). [11] B E C K E R , K ., The floating world of personnel monitoring, Radiat. Prot. Dosim. 51 (1994) 235-238. [12] B E C K E R , K ., Radiation protection in developing countries, Radiat. Prot. Dosim. 37 (1991) 217-219. [13] S T O L L , W . , Radioactivity: subjective or objective risk? Atomtechnische Wissenschaft (March 1992) 160 (in German). [14] PRÊ TRE, S., Atom, Symbols and Society: Intellectual Infection or Risk Conscious­ ness? Forum Energy and Medicine, ISBN No. 3-9520289-2-4 (1993) (in German). [15] DROTZ- SJÖBERG, В .-M., PER SSON , L., Public reaction to radiation: Fear, anxiety or phobia? Health Phys. 64 (1993) 223-231. POSTER PRESENTATIONS 195

IAEA-CN-54/9P

NUCLEAR DEM OCRACY

E . T Ó T H Lauder Yavne School, Budapest, Hungary

Like many other countries all over the world, Hungary is now learning to be a democracy. One year after the free elections (1992), two old women complained to the mayor of a northeastern Hungarian village: they had difficulties with breathing in their bedrooms. Their home was flooded with carbon dioxide of natural origin com­ ing from the soil. The mayor, elected by an 80% majority of the village, declared an emergency situation in the two houses, asked for the help of the Member of Parliament representing the region, and alerted both the Civil Defence and the Government Health Office. W ithin a month the Government gave 7 million Hungar­ ian forints (U S $100 000) to the village for survey and mitigation operations from the National Catastrophe Fund. An Operating Committee was created to co-ordinate the work of 25 different institutes. Research was done in a small part of the village, and the foundations of the two houses were reconstructed in 9 months at a cost equal to the value of the houses themselves. A happy ending, isn’t it? But where is the democracy in this story? Breathing problems due to carbon dioxide can be recog­ nized even by dictators, and if they have enough money available to solve the problem in this way they can even show spectacular results. People’s gratitude will be guaranteed. In this village not only carbon dioxide but also radon enters the houses. Radon is a radioactive inert gas. It is present everywhere all over the world in various con­ centrations. You can get a radiation dose from radon as well as from any other natural or artificial radioactive sources. In many countries there are laws setting a limit for radon concentration in homes; this limit varies from 150 to 400 Bq/m3 from country to country. Experts agree that living in a house having a concentration 400 Bq/m3 through a year is equivalent to receiving a 20 mSv dose. A dose of 20 m Sv is about the highest dose per year a worker received in the last 10 years in the Hungarian nuclear power station. The problem can be solved without any lesson in democracy in a country where there is a law setting an upper lim it for radon in buildings, and offering financial sup­ port for the survey and for mitigation. (Whether the law was made in a democratic way or not is another question.) But in Hungary the law speaks only about artificial radioactivity. Unfortunately, human biochemistry does not know whether an ioniz­ ing particle originated in an artificial or a natural radioactive source. What should 196 POSTER PRESENTATIONS be done in our small northeastern Hungarian village when we find dozens of houses with radon concentration above not 400 Bq/m3 but 1000 Bq/m3? The first question is how to identify the high radon level houses. In a totalitar­ ian regime it is easy to do. You should convince one person, the leader of the com­ munity, that the survey is important; he will order measurement for the whole village, he w ill even find financial support for that. O r he may not. But in the latter case you know that there is nothing to do. In a democratic society it is much more complicated. First of all you have to teach some nuclear physics to the elected leaders of the village on a level where they understand enough to convince themselves of the necessity for exploration of the radon levels in the village. The leaders are not physi­ cists, they are farmers, miners or workers in the nearby brick factory. They are much more interested in unemployment problems, they are much more interested in the condition of the roads or in the new telephone network of the village, and, of course, in local gossip and in their own prestige. This is understandable: the next election is approaching. M ost of the leaders are old enough to have forgetten all the physics learned in school, and anyway no nuclear physics had been taught to them. In our village we were lucky: at least the mayor — originally an electrician — was ready and able to learn about radon. The other members of the village council smiled at his efforts, so he needed continuous m oral support and/or pressure from us to find and to implement possible solutions. The occurrence of radon in the village was discovered by physicists. The radon problem in living rooms is a new one in Hungary. Previously no one bothered about it. W e had greater problems. Physicists working in this field now became excited. But they are also human beings, they have to survive, they have to buy more modern equipment and provide scholarships for graduate students. They live also in the market. They asked for financial compensation for their survey. When the Govern­ ment’s aid was over, they withdrew from the village. In those days no one knew the percentage of the houses heavily affected by radon. It is a question whether a physi­ cist has a moral obligation to continue such a work without compensation and to con­ vince the local or national authorities about the necessity of this work, or whether he has to wait for a contract from the authorities. In the case of our Hungarian village the physicists who started the work declared that the whole village had to be sur­ veyed, and they waited for a financial contract from the mayor. But the Hungarian state budget is in deficit. In Hungarian high schools we have been teaching a month of nuclear physics since 1981. W e explain radioactivity as the ‘cooling’ of nuclei, with as many experi­ ments as possible using simple Geiger-Müller counters. At the beginning we measured the activity of artificial sources made for schools. From 1986 we used the Chernobyl fallout as a source at hand. A few years later we had to find a new source of activity. W e collected dust particles from the air with a vacuum cleaner and measured the activity of the radon progeny. At the end of 1980s a school network POSTER PRESENTATIONS 197 was created in Hungary to measure radon in the air. Students and their teacher learned the technique for measuring radon. They found differences between different buildings, between different towns, between different seasons, but never found very high concentrations of activity in living rooms. They learnt from books that in some countries the radon activity is unacceptably high, they learnt the risk of radon exposure, they read about how the houses have been protected in other countries. The radon work for these students and their teachers was active and interesting but without real responsibility. Then on 15 March 1992 a phone call came from that small northeastern Hungarian village ... Some people thought that perhaps school teachers and their students could also do some useful work. At long last we had found a place where there is high radon activity in real living rooms, not only as described in books. At first it was not easy to enter the houses. Before our arrival at the village a crowd of white collar people invaded it, and they were so busy that they had no time to inform the people about their work, about radon, about the measured data. Radon was a mystery. When we arrived at the village I asked the local physics teacher for permission to speak to the pupils about the nature of radon. I offered simple charcoal absorbers which the children could take home so that after a week they could find out the radon level in their own rooms. The pupils went home, they informed their parents and asked them to let them bring home the charcoal absorbers. M y students from Budapest (aged 13-15) helped me. They had to take a very serious exam after a course of nuclear physics, and only then they were allowed to join the radon expedition to the village. For half a year we used charcoal and the continuously monitoring ATM OS ionization chamber plus air pump to measure radon activity concentration in 20 min (made by Gammadata in Uppsala). The charcoal absorbers were evaluated by a G e(Li) gamma detector at Eötvös University, Budapest. W e discovered that higher radon concentra­ tions occurred not only in the places where people from the research institutes worked before but also elsewhere in the village. Together with the students we decided to make an extended exploration of the village. So far, we have measured the radon activity concentration using alpha track detectors in 50% of the houses. The bulk of the work of the evaluation of the detectors was done by students and their teachers of the Lauder Yavne School in Budapest. Local students helped in communication with the local people. The Nuclear Physics Department of Eötvös University checked our work continously with more complex methods. A ll the work was done with youthful enthusiasm and interest, without any financial support. Now we have thousands of measurements of the spatial and seasonal distribution of radon activity concentration. What should we do with these data? There are several possible ways of reducing the radon level in a living room. The simplest way is to ventilate the room, especially in the evening before you go to bed. In the case of high activity this is not enough. In such cases some reconstruc­ tion of the building is needed: the method w ill depend on the structure of the house. To change living habits (e.g. m oving to a less affected room or ventilating regularly 198 POSTER PRESENTATIONS

in the evening) or to change the structure of the building needs the co-operation of the inhabitants and it needs money. How can one obtain the co-operation of the peo­ ple, who never attended high school? How can we create money for mitigation in a poor evolving democracy where there is no Big Brother to watch and help us? These have become the key questions. W e — students and their teachers — decided to give the data to the house owners directly, and simultaneously we explained the data. This means that we had to teach people in each house with high activity how to ventilate their rooms. W e also discussed whether we should give the data to the Operating Committee. They might have been ready to start the mitigation operations or they might say: no money, no action. It has turned out that the old roots of autocratic habits prevented the Committee from solving the problem in co-operation with the citizens. More than a year later they are still discussing the ‘ultimate solution’ for all the houses. Together with the students we have rejected their method. W e believe more in d e m o c r a c y . Our recipe is very simple indeed. W e try to teach the citizens as well as the leaders of the village how to live with radon. W hen we have measured a home we give the data to the citizen. In the most affected houses we introduce the inhabitants to measurement techniques: we teach the people how to read the data directly from continuously operating equipment. In this way they experience personally the benefi­ cial effect of ventilation. W e explain the meaning of the results and try to answer all their questions. W e try to find a solution for reduction of the radon level by think­ ing together with the inhabitants. W e also share our doubts with them. So the people feel they are involved in solving the radon problem. They feel their opinion is impor­ tant, they have to make decisions. They learn democracy: decision making based upon a shared understanding of information. I believe, because my young students believe, that our effort will not only decrease the radon levels but also increase the democratic spirit in this village. POSTER PRESENTATIONS 199

IAEA-CN-54/10P

PEOPLE AND RISKS

G . M A R X Physics Department, Eötvös University, Budapest, Hungary

On 6 August 1995 the public will recall the tragic bombing of Hiroshim a and Nagasaki. Environmental activists may equate Hiroshima, Nagasaki, Windscale, TM I, Chernobyl and all the nuclear plants. It is now our moral duty to educate the public on the concepts of acceptable and unacceptable risks. This will become a precondition for democratic decision making in the 21st century. According to our experiences in Hungary, quantitative argumentation is not alien to young people. W e offered in-service training to Hungarian physics teachers, distributing Geiger-M üller counters interfaced to personal computers, descending to coal and uranium mines, visiting coal fired and nuclear power plants. Sm all groups of Hungarian teachers paid visits to the TM I and Chernobyl power stations a few years after the accidents, carry­ ing radiation detectors in hand. (Furthermore, study trips to C ER N and to IA E A ’s laboratories in Seibersdorf were offered.) By noticing the huge interest of their students, the teachers were ready to learn and teach risk assessment. This has the consequence that the majority of Hungarian high school students prefer now the nuclear option among energy alternatives. Here the lessons offered to teachers and their pupils w ill be reviewed.

1. ACCEPTABLE AND UNACCEPTABLE RISKS

When you cross a road, look to the right at first, then look to the left at the middle of the road — we used to tell our pupils, without adding: — look up as well, to see whether an aeroplane is falling on your head. — The latter event also has a finite probability! If two people die of such accidents in one year in the U SA , there is a 1 % chance that somebody w ill die from such a danger among 1 m illion people in the next year. By everyday experience, common sense judges such a risk practically zero: not worth bothering about. For a simple discussion, let us introduce

Supported by OTKA-Project T7603. 2 0 0 POSTER PRESENTATIONS

the concept of microrisk as a risk that may kill one from among 1 million people exposed. International experience indicates that 1 microrisk is incurred as follows:

— in travelling 2500 km by train, — in flying 2000 km by plane, — in driving a car for 65 km. — in bicycling for 12 k m , — in riding a motorcycle for 3 km, — in smoking a cigarette, — in living 2 weeks with a smoker, — in drinking a bottle of wine, — in living with the weather (the risk of being killed by lightning in the coming 3 y e a rs), — in sleeping in a brick house for a week, — in breathing in a polluted city like Budapest for a day.

I have already taken such risks! One may conclude that people consider a few microrisks affordable. According to California law nobody may be exposed to an effect that may cause cancer, without being informed of this danger. But what does non-zero risk mean? According to the legal praxis, an exposure above 10 microrisks must not be caused without advance warning. This is why a warning is printed on every packet of cigarettes:

SMOKING MAY BE HARMFUL FOR YOUR HEALTH

2. CHEM ICAL RISKS

Comparing the mortality statistics of smokers and non-smokers, the W orld Health Organization estimates that about 29 000 Hungarians die every year from the effects of smoking. (Other types of suicides kill only 5000 Hungarians per year.) About 27 billion cigarettes are sold, indicating 1 cigarette ~ 1 m icrorisk, and an average smoker consumes 8000 cigarettes per year! A child in a family of smoking parents inhales involuntarily the equivalent of 20 cigarettes per year, corresponding ‘only’ to 20 m icrorisks per year. But this exceeds the legal lim it permitted to to be caused to uninformed citizens! At least one third of parents smoke. Simple m ultipli­ cation may give you the number of young victims for each year. The output of the world tobacco industry is 5 x 1012 cigarettes per year, thus the number of its casualties makes a seven digit number. In most developed countries smoking is declining. Tobacco companies look for new markets in eastern Europe and the Thirld W orld. Until 1990 advertising cigarettes was forbidden in Hungary. Now you see everywhere the posters indicating that by smoking a certain type of cigarette, you may become as macho as the cowboy shown on the picture. Because of aggressive POSTER PRESENTATIONS 201 advertising, cigarette consumption is increasing in eastern Europe and in southern Europe as well. Possibly smoking (and not industrial pollution) explains the elevated cancer risks in these regions. In Hungary, chemical air pollution is estimated to be the cause of 10% of deaths (about 15 000 victims) and smoking 20% (nearly 30 000 victims) each year. According to the W orld Health Organization (1994), “ smoking in the largest lethal epidemic in eastern Europe and the Third W orld” . It is difficult to make quantitative assessments in the case of low level risks. The Environmental Protection Agency (U SA) has adopted simple proportionality between dose and risk (without threshold). The slope of this straight line is 0.1 microrisk/mg of arsenic. Coal and iron ore contain a lot of arsenic; the global arsenic emission is above 50 million kg per year. In Prague, the contamination of the air reached the value of 70 ng/m3 during some winters. The lung cancer risk of workers in a factory with 80 ng arsenic/m3 in the air is 10 microrisks per year. There are some country wells in Hungary with an arsenic concentration exceeding 0.05 m g/L. Drinking their water has been forbidden because the associated risk exceeds 10 microrisks per year.

Lead is a common chemical poison (about 0.3 m icrorisk/mg), originating e.g. from lead pipes. British authorities forbid levels of lead in drinking water above 0.05 m g/L, but to eliminate lead pipes would cost m illions, thus in old buildings the concentration may be 10 or 100 times higher. Children’s bodies are hungry for Ca, and may mistake Pb for Ca and incorporate Pb into protein, making the enzyme unable to perform its duties. Thus lead may cause hyperactivity and a drop in intelli­ gence. The W orld Health Organization recommends that the lead concentration in blood should not exceed 0.2 m g/L, but it exceeds 10 m g/L in the blood of European city children. — The largest intake of lead is via the air. Leaded gasoline was invented by Thomas M idley for General Motors in 1926, but now 95% of the gaso­ line is unleaded in the USA. In Germany it is 63%, in Hungary only 30%. The tolerated concentration in air is 0.3 /¿g/m3, but — because of the use of leaded gaso­ line — in Budapest it may exceed 1 /¿g/m3. (Here the lead content of the blood occasionally exceeds 100 m g/L.) Unfortunately, lead does not decay physically. From the blood it is biologically eliminated with a half-life of 25 days, from tissues in 40 days, from bones and teeth only in 25 years.

3. RADIATION

Pale skinned northern people like to enjoy sunshine. (Recall the sun tanned blond movie stars wearing bikinis in Hollywood movies!) In 1928 Thomas Midley invented freon to be used in refrigerators made by General Motors, just because of the stability of the molecule. These man-made freon molecules are durable enough to diffuse up to the stratosphere, where the hard ultraviolet rays of the sun break them up, and the liberated C l and F atoms catalyse the decay of the ozone shield. Soft ultraviolet photons cross the broken shield and may cause skin cancer. In the 2 02 POSTER PRESENTATIONS

U SA more than 100 000 skin cancer cases are registered every year. They have doubled in 20 years and quadrupled in 40 years even in Europe. This is possibly due to the northerners’ habit of spending their summer holidays in the Mediterranean or possibly due to the increase in the man-made ozone hole. According to the Environ­ mental Protection Agency (U SA), 1% thinning of the ozone layer may increase the ultraviolet radiation by 2%, c a u s in g 4% increase in skin cancer for the pale skinned population (tens of thousands worldwide). In Hungary, the risk of lethal skin cancer has increased from 20 to 50 microrisks per year within the last decade. People are most afraid of ionizing radiation due to radioactivity. Fortunately, among different sources of risks, this is the easiest to measure. (It is a common experiment in Hungarian schools.) So we can find out how much radiation reaches us. Pupils calculate that about 8000 radioactive decays occur within our bodies between two heartbeats and 15 000 ionizing particles strike our bodies from outside each second — this during m illions of years of human evolution. In X ray diagnosis, 100 billion hard photons pass through us, so radiation cannot be so dangerous. The Hungarian Schoolbook describes the unit of radioactive dose: 1 m Sv (millisievert) means 0.001 joule absorbed ionizing radiation per kilogram of body mass. (In the case of rays consisting of heavier particles, it is adjusted to their stronger biological effects.) The Schoolbook contains a sheet for pupils to calculate their own radiation d o se .

Cosmic radiation at sea level 0.3 mSv/a Activity in the body and environment 0.5 mSv/a Radon inhaled outdoors 0.2 mSv/a

All natural sources (rounded) 1.00 mSv/a

Civilization has changed our life styles. W olves and smallpox have been eradi­ cated, but other risks were created. If you live in a house, add:

Ground floor dwelling 0 .5 m S v / a Light concrete house (9 mg U/kg) 1 .8 m S v / a Brick house (3.5 mg U/kg) 0 .7 m S v / a Wooden house (0 mg U/kg) 0 .2 m S v / a Radon excess in the house (rounded off) 1 m S v / a

“M AKE A NEST IN A TREE FOR YOURSELF! You may spare 1 mSv.”

Air flight for each 2500 km 0 .0 1 m S v Wristwatch with luminous dial 0 .0 2 m S v / a Black and white TV , viewed 1 hour per day 0 .0 1 m S v / a Color TV , viewed 1 hour per day 0 .0 2 m S v / a Medical X rays, average 0.5 mSv/exposure

Technological load (Hungarian average) 1 m S v / a POSTER PRESENTATIONS 203

The average load on the Hungarian citizen is about 3 mSv/a. There were zones in Hiroshima and Nagasaki where people survived but received radiation doses of 100 m Sv. Their medical history and the causes of their death were tracked carefully, and compared with those of the Japanese population living elsewhere. By subtracting the normal mortality, and by assuming a linear proportionality between dose and risk, the radiation risk has been found to be 50 m icrorisks per m Sv (Version A). This is used in official estimations. Each mSv would mean the same risk as

— smoking three packets of cigarettes, — drinking a glass of wine daily for a year, — being X rayed for kidney metabolism.

Bernard Cohen (University of Pittsburgh) intended to decipher the risks of low doses em pirically, by com paring the lung cancer statistics of different counties in the U SA with the measured radon activity concentrations in these counties. For doses below 5 mSv/a he obtained a drop of cancer risk instead of the expected rise! This surprising conclusion of a threshold was confirmed by Chinese and Japanese statis­ tics; it is not contradicted by the more elaborate Swedish evaluation. Low radiation levels may activate the immune defence of the body, as vaccinations do (Version B).

4. COLLECTIVE RISKS OF COMFORT

There are over 5000 m illion people living on Earth, exposed to natural radia­ tion (1 mSv/a), living in houses (1 mSv/a), expected to be X rayed, to insulate houses, to fly for business or holiday (1 mSv/a). The product of population number and personal dose gives the collective dose:

Natural radioactivity 10 000 million man-mSv/a Watches with luminous dials 2 million man-mSv/a Phosphate fertilizer industry 300 million man-mSv/a Coal fired industry, heating 100 million man-mSv/a Nuclear industry, public 1 million man-mSv/a Nuclear industry, workers 2 million man-mSv/a From atmospheric tests (1994) 50 m illion man • mSv/a

Nuclear bombs were tested in the air mainly in the 1950s and 1960s, since when the leftover activity has gradually decayed. But some singular events have contributed to ‘radiophobia’: 2 0 4 POSTER PRESENTATIONS

W indscale reactor accident 6 million man-mSv Harrisburg reactor accident (Three M ile Isand) 0.05 million man-mSv Chernobyl reactor accident 6 0 0 million man-mSv Hiroshima bomb 1 .5 million man-mSv 60 megatonne hydrogen bomb test 1 0 0 0 million man-mSv A ll atmospheric bomb tests 3 0 0 0 0 m illio n m a n • m S v 10 years of nuclear industry 3 0 million man-mSv 10 years of coal fired industry 1 0 0 0 million man-mSv

If our student is interested in estimating the worldwide number of victims, (s)he has to multiply the empirical doses by 50 per m illion m Sv (Version A ) or by — 0 (Version B). For a nuclear bomb test or for coal burning (s)he w ill obtain a six or seven digit number, comparable with the number of victims of the wars in the second half of the 20th century in the Far East, in the oil rich G ulf area or in Europe. By comparing this risk with other (more advertised) risks, the young people may make their own rational decisions themselves. POSTER PRESENTATIONS 205

IAEA-CN-54/13P

IONIZING RADIATION: ON W HY PRESENT TERM INOLOGY CAUSES CONFUSION

J. CHRISTENSEN Stockholm, Sweden

Use of argot and violation of epistemological principles impair comprehension of statements about ionizing radiation. Abolishment of ‘becquerel’, ‘gray’ and ‘sievert’ is one suitable measure for urgently needed improvement. One of the causes of the anxiety and fear evoked by news about the use of radi­ ation or about accidents involving radiation is the term inology used by the authorities addressing the public. Radiological terminology is so strange and alienating to the general public that the result could hardly be other than fear. In what follows, I shall try to show not only that radiological jargon is unnecessary but also that it has to a major extent been established through neglect or misunderstanding of words and of the epistemological principles of measurement.

1. UNITS OF RADIATION DOSE

The word ‘dose’ is derived from the Greek dido (I give) and dosi (given). It means roughly what has been given or ingested, best known to the public from medi­ cal prescriptions. Now, one formidable stumbling block for the average citizen’s (say, M r. A ci’s) comprehension of statements about doses of ionizing radiation is that by dose the radiologist (say: M r. Radiot) designates not what for instance the person, M r. A ci, has received, but what the tissue irradiated has received at the point of irradiation per unit mass. Thus to receive (and absorb) some dose at an X ray examination of an aching tooth is something very different from receiving the same dose to the whole body. This property of the radiation dose concept is bound to m is­ lead M r. A ci again and again, because M r. Radiot conceals it cunningly in his argot. Not only is his use of dose at variance with usage in M r. A ci’s language, he also makes the shrewd trick of renaming the standard unit. Consistently, pretending good intentions, M r. Radiot — seemingly without blushing — declares about the dose that “ to avoid confusion it has been given the special name ...” ! [1].

By ‘the absorbed radiation dose at a point in a body’ is designated the amount of radiation the body has been given and has absorbed, expressed as the energy of the absorbed radiation per unit of mass at that point [2]. The SI units of energy and mass are joule (J) and kilogram (kg), the SI derived unit of radiation dose is J/kg. 2 0 6 POSTER PRESENTATIONS

This unit name, which clearly shows that the dose is in fact a specific dose, i.e. per unit of mass, is rarely used by M r. Radiot. If he uses SI units at all, he practically always substitutes the special name ‘gray’ for J/kg.

2. UNIT OF ACTIVITY

Assume that the activity of some mass is 117 decays per second (briefly: 117/s). A professional radiologist (let us stick to naming him ‘M r. Radiot’) does not describe such a simple situation in this simple way— which would be almost directly comprehensible to M r. A ci. No! Again M r. Radiot takes to the trick of special nam­ ing; he says instead: “ The activity of the mass is 117 becquerel” or he uses another unit, namely ‘curie’.

3. DURATION OF RADIONUCLIDES, LIFE AND DEATH

The decay of a radionuclide is not an utterance of life: a radionuclide is not alive. Nonetheless one may see nuclides of long duration designated long lived or long living and their halving time half-life, etc. Life is a fundamental phenomenon. To use its name not only as an occasional metaphor but systematically as the name of lifeless things and items is a corruption of language in Austin’s sense [3]. (A quo­ tation from Ref. [3], paper No. 4, p. 83: “ Misnaming is not a trivial or laughing matter: If I misname I shall mislead others, and I shall also misunderstand informa­ tion given by others to m e.” ) In the short run this may be of minor importance to scientists in the field, although they often work with radionuclides in living organ­ isms. To politicians and to the general public it is rather the contrary — especially when such actions or decisions (e.g. about national defence and energy supply) are to be taken on the basis of a comprehension of effects of inert matter on living tissue, i.e. where the distinction between not living and living is essential.

4. UNNECESSARY AND ILLEGITIM ATE ARGOT

Incontestably, to those working within certain branches of science and trades it would be inconvenient or even practically impossible not to use a language of their own, an argot. Can the units, words and expressions used by professionals in the field of ionizing radiation be considered to belong to such a legitimate argot? W hat is gained by substituting becquerel (Bq) for decays/second (or /s) hardly balances the disadvantages. By using the unit name ‘sievert’, one indicates that the POSTER PRESENTATIONS 207 value in question is not the actual energy absorption. However, several different quantities are designated by sievert, i.e. local equivalent dose, equivalent whole body dose and effective equivalent dose. Thus, if aiming at a definite expression, we can­ not avoid defining in words which one of the quantities we refer to. Consequently, we might as well right from the beginning have used words and the unit name J/kg, which has the advantage of being commonly known from declarations of nutritional value on food packages, from heats of combustion, etc. W ithin thermodynamics a number of functions having energy dimension are used: enthalpy, internal energy, Gibbs free energy, Helmholtz free energy, exergy, anergy. The conceptual difference between any pair of these functions is at least as large as that between the concepts of dose and equivalent dose (or dose equivalent). Nonetheless — and in contrast to the doses — the same unit is used for all the func­ tions (in the SI system: joule). Scarcely anybody today would in earnest suggest, in M r. Radiot’s manner, that joule be replaced by specific names, one for each of the fu n c tio n s . To accept such a naming praxis would by analogy make it permissible to replace pint (or litre) by specific names according to whether the liquid were beer, water or m ilk, and names of units of power by different names for electrical power and heat, e.g. M W (e) and M W (h), and pound (or kg) by different names for amounts of apples, pears or potatoes. But such a praxis is contrary to what is taught today in elementary schools. In non-elementary terms, the praxis may rightly be designated a reactionary, epistemological contamination. The contamination is that the concepts of quantity and of unit of measurement are intermingled. And what is reactionary in this abusive praxis is that it implies a small but manifest step backwards in the development of human knowledge. Until a few centuries ago our knowledge of Nature was with minor exceptions chaotic and unstructured (see for instance Pliny the Elder’s Naturalis Historia, which in excerpts was what in the major part of Europe was offered as the sole education in the science of Nature in the higher schools until towards the middle of the 19th century). Follow ­ ing Galileo Galilei’s exhortative motto, to measure and make measurable, we have to an astounding extent succeeded in establishing order and structure where chaos reigned. The law of gravity, the laws of inertia, the laws of thermodynamics, Newton’s laws, M axw ell’s formulations of the laws of electromagnetism: all of these are equations of quantities, and an essential element in the thus established order is the conception of a quantity as the product of a (measured) number and a unit (of measurement). If, when applying the law of energy conservation to, say, a heat engine generating electricity in the steady state, one uses different units for electrical and for thermal power, the resulting equation cannot be solved; one w ill have to add im plicitly or explicitly an equation such as this:

1 MW(e) = 1 MW(h) 2 0 8 POSTER PRESENTATIONS

The same is true when making an energy balance for an irradiated volume of a body, if some doses are expressed in sievert and some in gray. Tw o equations will have to be added explicitly or implicitly:

1 sievert = 1 gray = 1 J/kg

Alternatively, one would have to assign some sort of dimension to factors that are now and have hitherto been considered to be dimensionless — such as thermal effi­ ciency, quality factor, R BE. To use specific units in this way is thence not only a conceptual contamination but a regressive step in the development of human knowledge. Expressed in more tangible terms, replacing J/kg by two other names has several disadvantages:

(1) That we are led to forget that it is dose per kilogram. (2) That the connection with energy and the active awareness of the (negligible) heating which the ionizing irradiation (usually) gives rise to are forgotten.

(3) That the relation is lost with other types of radiation — in particular with radar and microwaves, where joule and watt are exclusively used and where permis­ sible exposure limits are expressed in W /kg, W /m2.

Neither can becquerel, besides hertz for per second, withstand the criticism here above, although it has been accepted by the International Bureau of W eights and Measures [4] (by resolution 6 of the 13e Conférence Générale des Poids et Mesures: “ Activity (of a radioactive source) 1 per second, s"1” [5]). And sticking to per second improves comprehension, in that it has important psychological and peda­ gogical advantages, prim arily that the uniform ity of essential traits of the phenomena remain present in the mind; e.g. that an average decay rate of 20 per second is the same as the vibration rate of a very deep organ pipe (approximately the E flat of the sub-contra-octave), 50 or 60 per second is the alternating rate of the alternate current in our electricity supply, which may be observed by moving a hand swiftly under the light of a glim lamp, and so on. Such facts and insights tend to enhance the aver­ age citizen’s comprehension of, and diminish the puzzling and scaring effect of, news and information about ionizing radiation.

5. CONCLUSION

Several of the terms, units and unit names used within radiophysics and in related fields must be considered illegitimate as language for communication, both externally and within the profession. W hen radiophysicists, radiobiologists and the POSTER PRESENTATIONS 2 09 authorities address the general public, politicians and other decision makers, they o u g h t:

(1) Never to use becquerel, gray or sievert. (2) To use instead, when needed, decays/s and J/kg and units, names and abbrevia­ tions overtly derived from or related to these.

REFERENCES

[1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, A Com­ pilation of the Major Concepts and Quantities in Use by ICRP, Publication 42, Perga­ mon Press, Oxford and New York (1984) 1. [2] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990 Recommendations of the ICRP, Publication 60, Pergamon Press, Oxford and New York (1991) 1-3. [3] A U S T IN , J.L., Philosophical Papers, 2nd edn, Oxford University Press, Oxford (1970). [4] BUREAU INTERNATIONAL DES POIDS ET MESURES, Le Système International d’Unités (SI), 3e édition, Sèvres (1977) 10. [5] The International System of Units, National Physical Laboratory, London (1973) 35. 2 1 0 POSTER PRESENTATIONS

IAEA-CN-54/14P

PROM OTING COM PREHENSION AND UNDERSTANDING OF RADIATION PROTECTION

M.J. GAINES National Radiological Protection Board (NRPB), Chilton, Didcot, Oxfordshire, United Kingdom

A radiation protection organization needs to explain its work to the public and certain professional and political groups. The way the National Radiological Protec­ tion Board (N RPB) does this is described against a background of situations specific to the organization and its work.

1. O N E A I M

The aim is to explain radiation protection in a way that is comprehensible and believable.

2. TWO DIFFICULTIES

Firstly, radiation protection is difficult to comprehend. Although it has a body of knowledge and language all its own, it also has inputs from biology, physics, che­ m istry, epidemiology, statistics, genetics, law, public administration, etc. Secondly, it involves political judgements which inevitably run counter to somebody’s values.

3. THREE PROBLEMS OF THE MEDIA

The lack of space and time, and the effects of competition, are well known. Space is always at a premium, whether in a newspaper or in broadcasting. Journalists are always operating to deadlines which seem to get tighter as the technology devel­ ops. Competition can have unfortunate effects, e.g. in making all newspapers cover roughly the same subjects. POSTER PRESENTATIONS 211

4. FOUR SOLUTIONS

Develop relationships

At N RPB we develop relationships, particularly with science correspondents and other specialists: we invite journalists to visit us; we mail press releases, advisory publications, etc.; we brief the press on important topics; we organize press conferences on major items; we help with numerous enquiries, which has to be done quickly but accurately; we organize feedback from journalists. But the needs of the media have to be understood; e.g. N R PB has a variety of contacts with the B B C (see b o x ).

BBC and NRPB

Type of programme N R PB role

News (radio and TV) With science correspondents — a good understanding

Consumer programmes Constructive relationships Features/documentaries N RPB informative but prudent

Some characteristics

World service Story of global interest National programmes Story of national interest Regional T V Story of strong regional interest Local radio Story of good local interest

B ut also:

Reporters travelling to Russia Radiation protection advice N IR standards for broadcasting installations Advice to engineers

Communicate on their terms

Express yourself clearly in simple language. Clarify the meaning of technical terms. Explain any assumptions or extrapolations, and distinguish these from factual information; e.g. the linear relationship is an assumption (with some scientific sup­ port), not a statement of fact. For radio, keep statements short and simple. It is not a suitable medium for long, complicated statements. For television, the main infor­ mation carrier is the picture. 2 12 POSTER PRESENTATIONS

Train spokesmen to perform in these media

Training for broadcast media is particularly important.

Get the science right

Get the science right and carry as many other specialists as possible with you so that when journalists phone round they hear good, accurate accounts that support your statements.

5. FIVE FRUSTRATIONS

NRPB press officers are often on strong ground, e.g. if dealing with NRPB views on IC R P Publication 60, since N RPB has published its advice, or on medical radiology, since N RPB has worked with the professional bodies to reduce radiation exposure. On discharges from nuclear installations, N RPB staff can often draw on an expert view based on our own extensive studies. However, there are the following frustrations:

Transparency and science

N RPB press officers have to stay within the remit of N RPB, which is not part of the nuclear industry and not part of the Governm ent’s authorization and regulatory system. As the representatives of a scientific body they have to demonstrate that everything is ‘clear and above board’ and that there is open-mindedness. For instance, although N RPB advice is something that we must explain and defend, it is still necessary to make clear that N RPB is receptive to scientific evidence that might eventually cause that advice to be changed.

Global scientific village

O f course, such evidence does not necessarily come conveniently presented in the British Medical Journal, The Lancet o r Nature. Indeed, the classic situation is a scientific paper, published in a far away country, which is well publicized and news of which reaches the media in Britain ahead of the text of the paper; they contact N RPB for our views. Inevitably, it takes time to get a copy of the paper and even more for N RPB scientists to study it. But this is what has to be done before a scien­ tific body can comment directly on the paper, and therefore press officers can only supply background information, e.g. pointing to other research on the same subject which has not necessarily come to the same conclusion. They are, of course, helped if N RPB has convened a committee of distinguished specialists in the subject who have already published advice, so that their credibility can be added to that of N R PB. POSTER PRESENTATIONS 213

Scary anecdotes

On other occasions, intensive media interest focuses on information that is bas­ ically anecdotal (e.g. mobile phones causing brain cancer). It is easier for press officers here because there is no scientific paper to obtain and assess and they can point to studies which have found no risks from mobile phones. However, they have to do this with a gentle touch if a major study of this subject is under way, the out­ come of which they must not prejudge.

M ajor effort, m inor result

Radon, for example, is very often relatively easy to deal with. It is also desirable to keep the subject before the public, particularly in high radon dose areas. However, as much effort can be put into helping a small local or regional newspaper as a national publication.

Law yers in the public cause

Public debate can be generated through a high profile lawyer acting on behalf of clients, even though the risk is clearly low, if it exists at all (e.g. the cancer risk from power lines). The lawyer knows how to get his story across in the media, which puts everyone, including N RPB, apparently on the defensive in an area where its track record is good. It also causes concern among the public — not surprisingly when you consider how many people live near power lines, substations, etc. — and they flood N RPB with enquiries.

6. SIX RESPONSES OF LAYM EN

Generally, for the public to respond there has to be an interest, such as the fol­ lowing: a perceived radiation protection problem for themselves (e.g. a visit to Kiev); involvement in a cause where radiation is part of the debate (e.g. in anti- nuclear groups or as veterans of weapons tests); involvement in a cause where radia­ tion protection is not central but is brought in (e.g. with power lines, where people may be against them because they spoil the view); a constitutional role in radiation protection (e.g. elected representatives of local authorities, acting on behalf of a citizen); medical treatment for self or fam ily or friends; teachers and students, gener­ ally with educational projects in mind. 2 1 4 POSTER PRESENTATIONS

7. SEVEN GROUPS THAT REALLY MATTER

For N RPB, groups that really matter are as follows:

Specialists in radiation protection and the related sciences. These are particularly important because they, more than anybody, understand what we are doing.

Politicians and Governm ent officials. N RPB acts as a focal point for many organi­ zations, including Government departments, and has ongoing relations with officials and intermittent contact with politicians. This involves many face to face meetings, often between individual N RPB staff and individual officials.

Lo cal authorities. N R PB has co-operated with environmental health officers in local Government on matters such as radon and radiation monitoring. W e work closely with their professional body. W e also have less regular, but nevertheless important, dealings with other officials.

M edical. We work closely with the professional bodies for radiologists, radi­ ographers and hospital scientists. This again is an ongoing professional relationship fortified by regular participation in exhibitions at annual conferences, scientific papers and more general articles in the literature, and a helpful response to enquiries.

Pressure groups. We have constructive relationships, e.g. with Friends of the E a r t h .

Schools. It is necessary to tailor the information to the needs of the schools and to work with teachers in dealing with the practical problems of implementing the national curriculum, e.g. producing wall charts ‘by teachers, for teachers’.

Trade unions and employers; architects, surveyors, estate agents, building engineers, etc. (on radon); consumer organizations; veterans of the weapons tests; clients for services. These are grouped together simply because direct involve­ ment is on a much smaller scale, although on certain occasions they make us very busy (e.g. dealings with the veterans of weapons tests become very significant whenever we publish a new study of their health statistics).

It is important, when considering the results of media coverage, especially bad coverage, to remember that all organizations such as N RPB w ill have had dealings with a variety of groups such as these who will have already formed an opinion, preferably a good one. POSTER PRESENTATIONS 215

8. EIGHT TECHNIQUES OF COMMUNICATION

For NRPB, the techniques of communication are as follows:

One to one encounters.

Conferences, symposia, sem inars. Papers presented at these events are very much the medium with which scientists feel most comfortable.

Special events such as open days and general symposia. Open days are one of the most effective ways of communication with any group of people — the general pub­ lic, in particular, can be greatly influenced by visits to installations.

Professional journals and technical publications.

G eneral publications. Radiation protection has to be made accessible to lay people.

Visual m edia. Examples are videos, TV, exhibitions and well illustrated publica­ tions — yes, a good illustration is worth a thousand words.

Exhibitions. Like conferences, these bring people with similar interests face to face.

The m edia. They play a key role in making radiation protection accessible to lay people and making your organization’s good name familiar to a wide audience.

9. NINE PERCEPTIONS

Public debate on scientific subjects proceeds at two levels, among the specialists and among the public; the public debate is greatly influenced by, and generally follows, the debate among the specialists. Scientific credibility, reputation, whatever you call it, is at a premium; it must underpin all pronouncements. Commu­ nication skills are important; generally, they do not just happen — they have to be worked at. One must employ public relations professionals and give them the resources they need. Encourage scientific staff to help them. Think visually — visual media can be very powerful. Timing is also very important. Familiarity should lead to favourablility, but you have to work at it. Get all of this right and you will have more than comprehension — you will have understanding in the sense of sympathy and tolerance.

10. 1 0 6 G L A N C E S

NRPB has published over a million broadsheets in its ‘At-a-Glance’ Series, which is intended for various groups of people with little or no knowledge of radia- 2 1 6 POSTER PRESENTATIONS tion protection. The series is underpinned by scientific work but the jargon of radia­ tion protection is avoided. At an International Public Relations meeting in January 1992, organized by the European Nuclear Society, over one hundred public relations professionals from all over the world voted the Series the winner of their Award of Excellence for Print M a te ria ls. Spinoffs from the main series have included a schools wall chart, slide sets and the ‘Radiation at W ork’ Series (for workers). POSTER PRESENTATIONS 2 17

IAEA-CN-54/23P

IM PORTANCE OF PUBLIC PERCEPTION

J . R E A D Transport Canada, Ottawa, Ontario, Canada

‘Public opinion’, ‘public expectation’ and ‘public perception’ are three phrases used in representing views of the public. The first, ‘public opinion’, is most commonly used in a neutral sense. That the public has an opinion is not by itself either positive or negative. The phrase ‘public expectation’ can also be neutral. However, it is sometimes used in a negative context to suggest that expectations are too high, and sometimes in a positive context, although less frequently, to suggest that what is expected is a reasonable part of the total of what is actually wanted. The third phrase, ‘public perception’, is often used in a pejorative sense to mean unsubstantiated or non-scientifically based beliefs. In turn there is the implication that these beliefs must be filtered out of any technical argument. There may be valid reasons to do this when comparing one scientific idea with another, but, as the following paragraphs show, this is not a correct view when setting public policy. In recent meetings reviewing the transport of plutonium by sea, the delegate from Iceland expressed his concern about the possible effect on the fishery in Iceland of a release of plutonium in Icelandic waters. His view that no one would buy Icelan­ dic fish as a result was dismissed by another delegate as simply a poor understanding of how plutonium would behave in a release, especially with respect to fish. This dismissal of Iceland’s comment was immediately supported by another delegate who observed that there were no scientific facts which favoured the view of the delegate from Iceland. The concern was labelled as public perception, with the clear implica­ tion that it was not worthy of serious attention. However, a release of plutonium in Icelandic waters could lead to a significant decline in the purchase of Icelandic fish. Consumers will choose based on their beliefs, which follow from their perceptions. For Iceland, should such a release occur today, public perception would be critically important. It is interesting to observe that a similar situation arose after the Chernobyl accident. People believed the contamination affected most of Europe and for several years thereafter it was very difficult to sell strawberry jam from Poland in Canada. As a final example of the impact and hence importance of public perception, recall that Sweden is terminating its nuclear power programme. Being a democracy, it will, in the end, do what the majority of its citizens believe should be done. Such beliefs are influenced by public perception. 2 18 POSTER PRESENTATIONS

In all democratic systems, elected officials are influenced by their electorate, and if the public perception on an issue is clear, one should not be surprised if the country’s Government adopts the majority point of view. (Some observers have gone so far as to say that more important than public perception is the politician’s perception of public perception.) The debate on nuclear power and its subsequent shaping of public perception has not been a fair debate. The nuclear age was embraced on behalf of the general public by scientists. Information on actual or potential problems, as well as uncer­ tainties, was not provided to the public in some cases, or explanations were offered by scientists whose only focus was on success. Lacking were knowledgeable critics and opportunities to review policy. In the absence of apparent openness on the part of industry and Government, and in light of the image of being omniscient cultivated by those working in the area, it is not surprising that some people who wanted to hear more than just conclusions from the experts would begin to suspect that the full story is not being told because there are problems or side effects which need to remain hidden. A risk study was recently conducted in Canada with respect to the movement of low level radioactive waste. At no point in the first version of the study was the word ‘radioactive’ used. In a different situation a Government official spoke of the controlled release of water from a US reactor site into the Great Lakes. The presenta­ tion was very complete, but again at no time was the word ‘radioactive’ used. In each case I observed the same response by different people . For the first I was told “ But you know that all that dirt is really radioactive” . In the second someone said to me “ The waste water is actually radioactive” . From each person there was a clear mes­ sage to be suspicious because the true story was not being told. There was no longer an automatic acceptance of conclusions offered by experts. A more visible manifestation of mistrust and also of a dislike of being manipu­ lated was the negative response in the USA to the Japanese film in which Pluto Man drinks a glass of water containing plutonium to show how harmless plutonium can be. Pluto Man is seen as suggesting to children that plutonium is so harmless you can even drink it. From before Shakespeare penned his line “ Me thinks he doth protest too m uch” , a one sided argument has had the potential to raise suspicion and lead people to a conclusion opposite to the one presented. A debate in which the knowledgeable people will discuss only the positive will not be successful over the long period. By praising the positive aspects of nuclear power while downplaying the nega­ tive, if addressed at all, both industry and Government have contributed to a negative public perception — “ They’re holding things back, hiding things, which means there’s something wrong” . The debate on nuclear power doesn’t need a new set of ‘pat’ [ready] answers. The provision of conclusions by experts, no matter how simply or cleverly POSTER PRESENTATIONS 2 19 expressed, will no longer satisfy people who want to understand radiation and its possible effects in different situations so that they may form their own conclusions. Indeed, a failure to have this discussion will no longer result in a neutral public which will continue to trust its experts to decide for them, but will lead to the conclusion that the experts either are hiding negative consequences or are not really experts. Public perception is formed in many ways. An approach must be found to allow the public to learn about radiation and its possible effects so that they may form their own conclusions.

The Chernobyl accident was serious. It has had a profound impact on people’s views, illustrated by the strong reaction in western Russia to a recent exercise in Finland simulating a radiation emergency. The debate on nuclear energy must include the good and the bad, and as well allow for a discussion of unknowns. The people will decide with or without this knowledge. They will no longer accept the delegation of decisions in these matters to experts. They may choose no nuclear power as in Sweden, or support nuclear power as in France, but the choice will be theirs in the end and it should be made based on information which goes beyond experts’ conclusions. 220 POSTER PRESENTATIONS

IAEA-CN-54/23P

L ’IM PORTANCE DE LA PERCEPTION DU PUBLIC

J .A . R E A D Transports Canada, Ottawa, Canada

Les expressions «opinion publique», «attente publique» et «perceptions du public» expriment toutes la même chose. Premièrement, l’expression «opinion publique» est utilisée le plus souvent dans son sens neutre. Que le public ait une opinion ne veut en rien indiquer qu’il s’agit d’une chose positive ou négative. L ’«attente du public» peut aussi être neutre. Cepen­ dant, cette expression est souvent utilisée dans un contexte négatif dans le sens que les attentes sont trop élevées, mais elle est aussi employée dans un contexte positif — bien que plus rarement — dans le sens que les attentes sont raisonnables vu ce que l’on attend en totalité. Finalement, «perception du public» est souvent utilisée au sens péjoratif, à savoir des croyances non fondées ou non scientifiques. En même temps, cela implique que ces croyances soient dépouillées de tout argument tech­ nique. Il existe de bonnes raisons pour les éliminer surtout quand on compare deux idées scientifiques; toutefois, comme il est démontré dans le paragraphe qui suit, ceci n ’est pas une façon correcte d’agir lorsque l’on promulgue une politique publique. Récemment, lors d’une réunion sur le transport du plutonium par mer, le délégué de l’Islande a manifesté son inquiétude quant à l’impact possible des déversements de plutonium sur l’industrie de la pêche dans son pays. Selon son point de vue, plus personne n’achèterait de poisson islandais. Point de vue qui fut rapide­ ment rejeté par un autre délégué qui expliqua qu’il s’agissait là d’une mauvaise compréhension du comportement du plutonium dans l’eau, surtout dans les poissons. Ce dernier argument fut aussitôt soutenu par un autre délégué qui fit remarquer qu’il n’existait aucune preuve scientifique permettant de valider le point de vue du délégué de l’Islande. L’inquiétude exprimée fut donc reléguée au rang de «perception du public», avec la nette implication qu’elle ne méritait aucune attention sérieuse. Il n’en reste pas moins qu’un déversement de plutonium dans les eaux islan­ daises pourrait mener à un déclin important de la demande de poisson de cette région. Les consommateurs choisissent selon ce qu’ils croient et ces croyances proviennent de leurs perceptions. Pour l’Islande, une telle perception aurait des conséquences graves si un tel déversement se produisait. Il est intéressant de constater qu’une situation semblable s’est produite après l’accident de Tchernobyl. Tout le monde pensait que la contamination avait touché toute l’Europe et, pendant plusieurs années, la confiture de fraises en provenance de Pologne s’est vendue très difficilement au Canada. POSTER PRESENTATIONS 221

Comme dernier exemple illustrant l’impact et donc l’importance de la percep­ tion du public, il faut se rappeler que la Suède abroge son programme d’énergie nucléaire. Somme toute, étant une démocratie, elle fait ce que la majorité de ses citoyens pensent être la meilleure chose à faire. De tels jugements sont influencés par la perception du public. Dans tous les systèmes démocratiques, les élus sont influencés par leur élec­ torat. Si l’opinion du public est claire sur un sujet, il n’est pas surprenant de constater que le gouvernement adoptera le point de vue de la majorité. Certains vont même plus loin en affirmant que ce qui est plus important que la perception du public est la perception que se fait le politicien de la perception du public. Le débat sur la question du nucléaire et la perception du public n’est pas un débat complet. Les scientifiques sont entrés dans l’ère atomique au nom du grand public. Dans certains cas, le public n’a jamais eu d’information sur les problèmes actuels ou possibles, ni même sur les incertitudes; lorsqu’il a reçu des explications, celles-ci ont été fournies par des scientifiques simplement intéressés par le succès. L ’occasion de revoir la politique ou de disposer même de critiques fondées ne s’est pas présentée. Ainsi, il ne faut pas s’étonner, devant l’absence d’une certaine ouverture de la part de l’industrie et du gouvernement et l’apparente omniscience pronée par ceux qui travaillaient dans le milieu, que ceux qui ne voulaient pas se limiter à la connais­ sance des conclusions des spécialistes ont commencé à soupçonner que, si on ne dévoilait pas tout, c’est qu’il y avait des problèmes et des impacts qu’il fallait cacher. Au Canada, on a récemment effectué une étude sur le risque du transport des déchets radioactifs de faible activité. Dans la première version de l’étude, aucune mention du terme «radioactif» n’apparaît. Dans un contexte différent, un représentant du gouvernement a parlé du déversement contrôlé d’eau provenant d’un réacteur américain vers les Grands Lacs. La présentation, bien que complète, n’a fait, là encore, aucune mention du terme «radioactif». Dans les deux cas, j ’ai pu observer la même réaction de la part de différentes personnes: «mais, vous savez, toute la boue est, en réalité, radioactive», ou bien, «l’eau usée est, en réalité, radioactive». Chaque personne ainsi lançait le message qu’il fallait se méfier car toute la vérité n’était pas dite. D ’emblée, l’acceptation des conclusions offertes par les spécialistes était re je té e . Une manifestation visible de la méfiance et aussi du mépris d’être manipulé s’est fait sentir par la réaction négative aux Etats-Unis au court métrage japonais dans lequel M. Pluto boit un verre d’eau contenant du plutonium pour montrer que le plutonium ne présente pas de danger. En fait, ce que M . Pluto suggère aux enfants, c’est que le plutonium est tellement inoffensif qu’on peut même en boire. Comme l’écrit Shakespeare: «Me thinks he doth protest too much» (Je pense qu’il proteste trop), un débat ne mettant en lumière qu’un aspect de la question accroît le doute et amène les gens à croire à la conclusion contraire. Un débat au cours duquel on ne présente que le côté positif de la question ne mènera nulle part 2 22 POSTER PRESENTATIONS

à long terme. Promouvoir les côtés positifs de l’énergie nucléaire et minimiser ses aspects négatifs, si toutefois ces derniers sont abordés, a fait en sorte que l’industrie et le gouvernement ont contribué à la perception négative du public: «On ne nous dit pas tous les faits, on nous les cache, donc, il faut se méfier». Le débat sur l’énergie nucléaire n’a pas besoin de réponse «toute prête». Les gens ne se satisfont plus des conclusions fournies par les spécialistes, peu importe qu’elles soient simples et intelligemment exprimées. Ils veulent comprendre les rayonnements et leurs effets possibles dans diverses situations pour pouvoir former leurs propres conclusions. En effet, empêcher cette discussion ne mènera plus à une réaction neutre de la part du public, qui autrefois faisait confiance aux spécialistes, mais plutôt à la conclusion que les spécialistes cachent les conséquences négatives et qu’ils ne sont pas de vrais spécialistes. La perception du public se forme de diverses façons. Il faut trouver une solution pour permettre à ce dernier d’en savoir davantage sur les rayonnements et sur leurs effets possibles, de façon qu’il puisse tirer ses propres conclusions. L’accident de Tchernobyl était grave. Le point de vue des gens en fut profondément affecté, comme le démontre la forte réaction à l’Ouest de la Russie, lors d’un récent exercice de simulation d’urgence nucléaire en Finlande. Il faut inclure, dans le débat sur la question de l’énergie nucléaire, les bons et les mauvais côtés et aussi permettre des discussions sur les faits inconnus. C ’est le public qui décide, qu’il ait ou non tous les éléments. Il n’acceptera plus de déléguer les déci­ sions aux spécialistes. Peu importe si le public refuse l’énergie nucléaire, comme en Suède, ou l’accepte comme en France, il n ’en reste pas moins que le choix lui revient en fin de compte. Ce choix devrait se faire en connaissance de cause au-delà des conclusions prônées par les spécialistes. POSTER PRESENTATIONS 2 23

IAEA-CN-54/24P

COM PREHENDING RADIATION RISK IN LITHUANIA

D. SIDISKIENE Radiological Laboratory, National Hygiene Centre

T. NEDVECKAITE Radiation Safety Laboratory, Institute of Physics

Vilnius, Lithuania

1. INTRODUCTION

The consequences of the Chernobyl accident and the operation of the Ignalina nuclear power plant (INPP) in eastern Lithuania are the two main contributors to emergency exposure that may be held responsible for radiation risk in Lithuania. The National Hygiene Centre at the Ministry of Health, as well as other research institu­ tions, carries out radiation control and assesses its possible influence on human health. The results of these investigations are presented herein.

FIG. 1. Mean 1311 activity in milk in different areas of Lithuania after the Chernobyl acci­ dent: A, up to 370 Bq/L; B, from 370 Bq/L to 1000 Bq/L; C, over 1000 Bq/L. 224 POSTER PRESENTATIONS

2. CONSEQUENCES OF THE CHERNOBYL ACCIDENT IN LITHUANIA

A radiation cloud crossed Lithuania during the first hours after the Chernobyl accident and drifted towards the Scandinavian countries. Among other factors, exter­ nal exposure rates varied over a broad range and peaked at 150 mSv/h and more. Numerous activity concentration measurements in foodstuffs, agricultural products and environmental samples have been made in the National and Regional Hygiene and Agricultural Centres and specialized research institutes. During the very first days the greatest danger was associated with hot par­ ticles [1] and radioiodine, especially in the southeastern part of Lithuania (Fig. 1). The proportion of gaseous 13II species of airborne iodine identified during that period of time [2, 3] exceeded 2-4 times the aerosol fraction (Fig. 2). The radioio-

6 0 , -

Date

FIG. 2. 1311 activity in air (Vilnius): A, aerosol fraction; B, molecular iodine; C, methyl iodide. POSTER PRESENTATIONS 225

D a t e

FIG . 3. Time dependent approximation o f 1311 activity in milk in different areas of Lithuania (according to Fig. 1).

TABLE I. THYROID EQUIVALENT DOSE OUTPUT VARIABLES IN LITHUANIA (ACCORDING TO FIG. 1) AFTER THE CHERNOBYL ACCIDENT

Deterministic Thyroid equivalent dose output thyroid variables (mSv) Age Area equivalent ______dose w „ w , 95th , _ „ Mean SD Median Mode (mSv) percentile

Infant A 21 22 10 20 16 39 В 61 64 28 59 49 115 С 158 164 70 150 127 280 Adult A 2.7 2.8 1.0 2.6 2.3 4 В 7.9 8.3 3.1 7.8 6.8 11 С 20 21 7.5 20 17.7 28 2 26 POSTER PRESENTATIONS dine activity in milk consumed by inhabitants of Lithuania varied over a broad range and decreased with an effective half-life ranging from 4.2 + 0.6 days to 5.2 ± 0.9 days (Fig. 3). Thyroid examinations by dosimetric teams [4] were not available in Lithuania for reasons beyond the control of experimenters. Because of this, thyroid equivalent doses were calculated using the modified ICRP three compartment cyclic model. Additionally iodine deficiency was measured. However, as a rule there are signifi­ cant uncertainties in metabolic parameter values and errors in environmental mea­ surements. That is why the stochastic method was applied for dose assessment. Thyroid dose evaluations in three areas of Lithuania with different contamination levels are presented in Table I. W e must keep in mind that the period of latency of thyroid diseases as a conse­ quence of the influence of ionizing radiation varies between 5 and 30 years. At the present time in Belarus the incidence of malignant thyroid diseases has risen,

965 1970 1975 1980 1985 1990

B.

0 . 0 1

0.001 1965 1970 1975 1980 1985 1990 Date

FIG . 4. Dynamics of radiocaesium (A) and radiostrontium (B) activity in the main foodstuffs categories from 1965 to 1993 in Lithuania. POSTER PRESENTATIONS 227 especially among children [5]. The first thyroid gland investigations revealed a rather high frequency of this pathology among the Lithuanian population, especially among secondary school pupils, up to 60% [6]. These investigations will be continued according WHO recommendations. Besides radioiodine, increased contamination of the environment and food­ stuffs by radiocaesium and radiostrontium was detected after the Chernobyl accident. W hile the level of radiostrontium in soil actually did not change significantly, the level of radiocaesium increased on the average by 4.5 times. The effect of the Chernobyl accident was evident in foodstuffs consumed in Lithuania (Fig. 4), although estimation of the internal dose from radiocaesium indi­ cates that it did not exceed 0.05 mSv during the first year after the accident. It is interesting to note that the activity level in drinking water was not influenced by the Chernobyl accident. This is explained by the fact that surface water is not used for consumption in Lithuania.

3. PRELIMINARY RESULTS OF THE DOSE ASSESSMENT AND HEALTH EFFECTS PROGRAMME CARRIED OUT IN THE INPP AREA

A second source of potential radiation risk in Lithuania is the INPP. Two of its functioning reactors are of the same type as in Chernobyl, but 1.5 times more powerful. This plant is listed among the most dangerous in Europe. There have been many minor breakdowns at the plant. W ithin a 30 km radius, among other investiga­ tions, regular measurements of external exposure and contamination of foodstuffs, drinking water, grass and raw milk are conducted to check the possible influence of the INPP on human health. Human health conditions, especially with respect to thyroid diseases, in the area of the INPP were compared with those in other regions of Lithuania (Table П).

TABLE П. PREVALENCE OF THYROID GLAND PATHOLOGY IN THE INPP AREA AND BACKGROUND REGIONS OF LITHUANIA, 1993

Abnormal Persons FT4 TTH Region thyroid examined (pmol/L) (mlU/L) (%)

Visaginas (5 km from INPP) 299 38 11.8 ± 0.37 2.11 + 0.1 Zarasai (19 km from INPP) 146 61.5 12.8 ± 0.45 2.02 ± 0.15 Kupiskis (background) 372 40.5 1.3 ± 0.09 8.5 ± 0.28 228 POSTER PRESENTATIONS

Palpation, ultrasound and determination of TTH, FT 4 and some antibodies were used for assessment. The data were statistically reliable (p < 0.01). From these preliminary data it may be inferred that the present situation in the region of possible influence of the INPP does not markedly differ from that in other regions of Lithuania.

REFERENCES

[1] LUJANAS, V., MASTAUSKAS, A., LUJANIENE, G., SPIRKAUSKAITE, N.. Development of radiation in Lithuania, J. Envir. Radioactivity 23 3 (1994) 249. [2] STYRA, B., NEDVECKAITE, T., FILISTOWICH, W., Iodine Isotopes and Radia­ tion Safety, Gidrometeoizdat, St. Petersburg (1992) (in Russian). [3] NEDVECKAITE, T., FILISTOWICH, W., Determination of gaseous and particulate l29I in atmospheric air by neutron activation analysis, J. Radioanal. Nucl. Chem. 174 1 (1993) 43. [4] INTERNATIONAL ATOMIC ENERGY AGENCY, The International Chernobyl Project (An Overview), IAEA, Vienna (1988). [5] ASTACHOVA, L.N., LEONOVA, L.L., DROZD, V.N., etal., “ Assay of work done in the Endocrinological Department Clinic at the Radiation Medicine Institute of Belorussia during 1989-1990” (Proc. Resp. Conf. Minsk, 1991), Minsk (1991) (in Russian). [6] SIDLAUSKAS, D., DANYS, J., KRASAUSKIENE, A., et al., Investigation of Thyroid Diseases in Lithuania in Relation to Radiation Doses from Chernobyl, Ecologi­ cal Series, US Environmental Protection Agency, Washington, DC (in press). POSTER PRESENTATIONS 2 2 9

IAEA-CN-54/51P

COMMENT M IEUX INFORM ER LE PUBLIC DE LA REALITE DES RAYONNEM ENTS IONISANTS

J.P. CHAUSSADE Direction de la communication, Electricité de France, Paris, France

1. UNE MECONNAISSANCE TOTALE PAR LE PUBLIC

Il est intéressant de mesurer, pour les rayonnements ionisants et leurs risques, l’écart entre la perception par le public et la réalité. Le public ignore que la radioactivité est depuis toujours aussi un phénomène naturel. Lorsque nous donnons des chiffres qui comparent des doses liées à telle acti­ vité industrielle ou médicale à celles reçues naturellement, nous sommes accusés de vouloir cacher la réalité. Tout ce qui est artificiel est suspect sinon dangereux. Quarante-quatre pour cent du public en décembre 1993 (contre 48% en décembre 1992) considère que la radioactivité artificielle est plus dangereuse que la radioactivité naturelle. L ’accident de Tchernobyl a apporté une image très catastrophique de tout ce qui touche à la radioactivité. Désormais, celle-ci signifie pour la santé: effets géné­ tiques, cancers et leucémies. Moins de la moitié de la population associe le danger à l’importance de la radioactivité. Quarante-sept pour cent du public (45% en 1992) pense que, lorsque la radioactivité est très faible, ce n’est pas dangereux. Les écarts sont importants selon les groupes sociaux. Le pourcentage est de 37% pour les jeunes de 18 à 24 ans (44% en 1992) et de 57% (52% en 1992) pour les personnes de 50 à 64 ans. Il est de 42% pour les personnes sans diplôme (42% également en 1992) contre 64% pour celles de formation supérieure (54% en 1992). Cette inquiétude est largement exploitée par les opposants au nucléaire qui occupent le terrain vaste de la désinformation. Il est très facile de faire peur avec 50 Bq de radioactivité dans 1 kg de champignons secs!

2. REDONNER AUX ACTEURS LEUR CREDIBILITE

Face à cette inquiétude du public, il est nécessaire de pratiquer la même politique de communication que pour la sûreté, à savoir la politique de transparence. C ’est le seul moyen pour que les acteurs soient crédibles. 2 30 POSTER PRESENTATIONS

Plutôt que de réagir par rapport aux propos alarmistes, plutôt que d’être en position défensive, les acteurs doivent avoir une position plus positive, plus active. Ils doivent profiter de toutes les occasions pour fournir des informations en donnant les valeurs mesurées et sans chercher à rassurer à tout prix. Au-delà des données brutes, il faut pouvoir fournir des éléments pour apprécier l’importance d’un incident et ses conséquences sanitaires. Pour cela, il est bon de donner des valeurs suivies dans le temps et dans l’espace. Indépendamment des acteurs eux-mêmes, les porte-parole seront à rechercher à l’extérieur des entreprises, dans des laboratoires proches des universités par exemple, pour éviter de laisser le monopole de l’information indépendante à des opposants au nucléaire.

3. TENTATIVE DE REPONSE A L’ATTENTE DU PUBLIC: LUTTER CONTRE LA PEUR D’INFORMER

Tous les publics ne sont pas concernés de la même façon par les rayonnements ionisants, et il y a lieu de privilégier dans ce domaine ceux qui habitent à proximité d’une centrale nucléaire. Ceux-là doivent pouvoir disposer à tout moment des mesures de radioactivité dans l’environnement et de toute information relative à la vie d’une centrale nucléaire, et en particulier aux incidents. La politique de transparence doit être appliquée de façon rigoureuse. Tout doit être présenté et expliqué. C’est la peur d’informer qui engendre l’inquiétude puis l ’o p p o sitio n . N ’ayons pas le Becquerel honteux! POSTER PRESENTATIONS 231

IAEA-CN-54/53P

NUCLEAIRE ET INFORM ATION DES PROFESSIONS DE SANTE

C. GALLIN-MARTEL, C. VROUSOS1, H. KOLODIE1, H. PONS, M. DURR 2 Service général de médecine du travail, Electricité de France, Paris, France

L ’information des professionnels de santé dans le domaine des effets médicaux et sanitaires des rayonnements ionisants et de l’utilisation industrielle de l’énergie nucléaire reste trop souvent insuffisante et limitée aux spécialistes. Or de la qualité de cette information dépendent pourtant en grande partie l’acceptation par la popula­ tion d’un voisinage à risque potentiel et, bien entendu, ses réactions en cas d’incident ou d’accident. En effet, des différentes enquêtes réalisées, il apparaît qu’en situation normale les professionnels de santé constituent des référents essentiels pour le public et qu’ils sont des relais d’information crédibles. En situation de crise, le champ de l’information rationnelle sera altéré par les rumeurs et la désinformation dont les conséquences en terme de santé publique peuvent être caractérisées par l’émergence de comportements inadaptés à l’échelle individuelle et collective. Dans cette situa­ tion, les médecins et pharmaciens, dont l’autorité morale et la compétence technique sont reconnues, auront un rôle essentiel de conseiller en santé publique à remplir auprès de la population. Le corps médical perçoit à la fois le rôle que la population est susceptible de lui demander de remplir dans ces circonstances et son manque, souvent très impor­ tant, d’informations adaptées. Interrogés sur les effets médicaux des rayonnements ionisants, les professionnels de santé déclarent être non ou mal informés à près de 90%. Le même pourcentage souhaite être informé sur la conduite à tenir face à un irradié ou un contaminé et sur les mesures sanitaires de protection. Cette information doit transiter par l’intermédiaire de documents écris à 58% , de séances d’Enseigne- ment postuniversitaire (EPU) à 55% , ou de colloques à 46%. Quant à la formation initiale des médecins, il faut rappeler les résultats de l’enquête réalisée par questionnaire en septembre 1992, auprès de 508 facultés de médecine de 24 Etats membres de l’OCDE. Plus de la moitié de ces facultés n’ont pas un enseignement organisé de radiopathologie et de radioécologie durant le cursus des études médicales. Sur les établissements qui déclarent enseigner ces matières, seuls 18% exigent un contrôle de connaissances.

1 Service de cancérologie radiothérapie, Centre hospitalier régional universitaire, Grenoble. 2 Production transport, Paris. 2 3 2 POSTER PRESENTATIONS

Face à cette attente de formation et d’information des professionnels de santé, l’Electricité de France (EDF) a décidé de réaliser une information objective, validée scientifiquement, crédible et adaptée aux lecteurs. C’est dans ce but que des spécialistes hospitalo-universitaires, des représentants des ordres et des organisations syndicales respectives, des responsables de la formation continue et des grandes administrations impliquées sont associés à la réalisation des documents publiés au plan national et aux séances d’enseignement postuniversitaire organisées au niveau loco-régional. Les principales actions nationales réalisées concernent:

1. MEDECINS

La brochure «Médecins et Risque Nucléaire» constitue un des éléments de l’opération Isère Département pilote dans la prévention des risques majeurs initiée en 1987-1988 par le M inistre de l’environnement. Trois niveaux d’information sont proposés: le premier précise la conduite à tenir face aux diverses modalités d’acci­ dents radiologiques; le deuxième développe les aspects sanitaires; le troisième pré­ cise la place du médecin de famille dans l’organisation générale des secours. En 1992, la quatrième édition actualisée et restructurée a été publiée. Sa diffusion auprès de l’ensemble des médecins généralistes et des médecins spécialistes concernés a été réalisée au plan national fin juin 1992. Au total, plus de 100 000 exemplaires ont été diffusés. La revue «Médecins et rayonnements ionisants» publie des articles concernant la radiopathologie, la radioécologie et l’utilisation médicale des rayonnements sous l’égide d’un comité scientifique national présidé par le professeur M. Tubiana. Sa diffusion est de près de 40 000 exemplaires. Sa périodicité est de trois numéros par an. Une évaluation de la perception de cette revue par le corps médical a été réalisée début 1993. Sur 500 réponses reçues: 96% souhaitent continuer à la recevoir, 74% la conservent, 17 % la transmettent; la qualité de sa réalisation est jugée agréable par 66% des répondants. Sept numéros ont été publiés depuis septembre 1991. De très nombreux articles sont publiés dans les revues de médecins généralistes ou dans les revues spécialisées sous l’égide des docteurs M. Bertin et J. Lallemand du Comité de radioprotection d’EDF. Des réunions scientifiques de haut niveau faisant le point sur un thème précis sont organisées chaque année sous leur responsabilité.

2. PHARMACIENS

Deux brochures d’information sont en cours de réalisation en étroite collabora­ tion avec l’ensemble de la profession et les structures compétentes (Service central POSTER PRESENTATIONS 2 33 de protection contre les rayonnements ionisants ou SCPRI et Institut de protection de sûreté nucléaire ou IPSN): «Pharmacien et nucléaire, informations pratiques» (8 à 12 pages) dont la diffusion concernera tous les pharmaciens en France; et «Pharmacien et nucléaire, conduite pratique en cas d ’accident» (60 pages), qui sera adressée aux facultés de pharmacie, aux pharmaciens hospitaliers et sapeurs- p o m p ie rs. Leur parution est prévue en novembre 1994.

3. VETERINAIRES

Une série d’articles a été publiée sous forme de suppléments à la «Dépêche vétérinaire» sous la responsabilité de M. Michon, membre de l’Académie vétérinaire: Effets des irradiations aiguës sur les animaux domestiques (n° 17 du 09/02/91); Les pollutions radioactives: conséquences pour le cheptel (n° 22 du 26/10/91); Pollutions radioactives de denrées alimentaires d’origine animale: prévi­ sions quantitatives (n° 25 du 25/04/92); Les normes de protection radiologique — l'interprétation sanitaire de la pollution radioactive des denrées alimentaires (n° 33 du 19/06/93); Organisation de la sûreté nucléaire — maîtrise des conséquences d’un accident nucléaire (n° 37 du 19/03/94).

4. INFIRMIERES

En partenariat avec la Société française de biophysique et médecine nucléaire, EDF a mis en place un stand d’information lors des Salons infirmiers européens de 1993 et 1994 qui se sont déroulés à Paris. Une très importante documentation a été distribuée et des contacts multiples se sont établis avec la profession. Un logiciel réalisé par le professeur J.C. Artus, en collaboration avec la Société française d’énergie nucléaire, a connu un grand succès. Une brochure spécifique d’infor­ mation est prévue. Malgré cet effort en faveur de l’information des personnels de santé, il reste encore beaucoup de tabous à dissiper. L ’actualité nous confirme que des informa­ tions plus ou moins alarmistes diffusées par les médias méritent d’être reprises sous la plume de personnalités compétentes et d’être placées dans leur juste contexte. Cet effort d’information doit être poursuivi activement pour que le corps médi­ cal puisse remplir avec compétence et indépendance son rôle de conseiller en santé p u b liq u e . 2 3 4 POSTER PRESENTATIONS

IAEA-CN-54/64P

PSYCHOLOGICAL DIMENSIONS OF RISK PERCEPTION AND TRADE-OFF

Y . T A N A K A Gakushuin University, Tokyo, Japan

1. INTRODUCTION

Scientific measurement of risk is often found to contradict public reactions to risk. Scientific risk assessment and risk perception may be considered as being dic­ tated by different kinds of rules and subject to different systems of measurement. A recent UNSCEAR report [1] on perception of risks from radiation and other sources lists 19 ‘dimensions’ of perceived risk, such as ‘media attention’ and ‘benefit’. The purpose of this paper is to examine (a) the differences between psychological risk

(Risk) + 2 x x Nuclear weapon Cancer — — x / X x Leukaemia

Plutonium x

x Automobile Radiotherapy

Helicopter x x -2 + 2 (Non-benefit) (Benefit)

Radiation related laboratory x | Japanese reprocessing plant x x Uranium

Sea trip x x x X ray

Japanese nuclear powered ship 7' x Pimple x Japanese enrichment plant Nuclear reactorx

Japanese nuclear power station x

-2 (Safety)

FIG. 1. Risk-benefit perception in nuclear experts (N = 545, 1981). POSTER PRESENTATIONS 2 35

(Risk) + 2 Poison W a r x Nuclear weapon C a n c e r ;% < î Stroke x 'Fire Earthquake Leukaemia x — Flood Japanese nuclear power station Racing car T y p h o o n - r I Sky diving \ x Radiation related laboratory Volcano eruption-' Rope dancing x Plutonium x x Nuclear reactor Irregular pulse" x Uranium Japanese nuclear powered ship x x Automobile I Japanese enrichment plant x x Radiotherapy Helicopterx x -2 + 2 Japanese reprocessing plant (Non-benefit) (Benefit)

x Pimple x Sea trip

-2 (Safety)

FIG. 2. Risk-benefit perception in the lay public (N = 474, 1983).

(Risk)

/ Cardiac surgery X ray / у Automobiles Anti-cancer drugs / ^ S m o k in g Sleeping pills v 4c ^ Nuclear power stations Insecticides ^ ¡ T 44 Radiation treatment tor cancer Antihypertensive drugs v \ \ ^ Irradiation of vegetables Antidepressants Food additives ^ T ra v e l by air (Media's 0 (Media's risk boosting) risk boosting) (W e a k ) 4 Antibiotics (Strong) Alcoholic beverages^

Acupuncture ^ эЦ Vitam in pills Herb medicines

-2 -2 0 (Safety)

FIG. 3. Risk perception and media effects in the lay public (N = 630, 1988). 2 3 6 POSTER PRESENTATIONS perception and scientific risk assessment and (b) the dimensionality of risk percep­ tion, both on the basis of the results obtained from recent studies conducted by the present author [2-5].

2. PSYCHOLOGICAL MEASUREMENT OF RISK PERCEPTION

C.E. Osgood has offered a reliable psychological measuring instrument he terms the ‘semantic differential’ in order to measure the ‘feeling tone’ of various events. This method can be used to measure both the intensity and the directions of fear and other affective responses which might be aroused towards various cognitive events. First, by using the semantic differential, ‘feeling tones’ towards various risk related cognitive events were measured and allocated in a quadrant defined by ‘benefit’ and ‘safety’. Figures 1 and 2 display perceptual differences between nuclear experts and the lay public. W hile nuclear experts tend to perceive radiation related events as ‘beneficial’ and ‘safe’, the lay public tend to perceive them as ‘beneficial’ but ‘risky’. Most differences are found to occur along the risk-safety dimension. If the intensity of perceived risk is greater than that of perceived benefit, people would avoid such events. It is also suggested that the more the risk related events draw the attention of the mass media, the more the lay public tends to perceive them as risky. Figure 3 clearly shows a correlational trend between perceived risk and media e ffe c ts.

TABLE I. SCALE FACTOR ANALYSIS: OPINION LEADERS (N = 1722, 1992)

Nuclear power generation (NPG)

I. Benefit factor

Beneficial for mankind 0.87 Beneficial for myself 0.87

II. Safety factor

Safe 0.85 N ot uneasy 0.81 POSTER PRESENTATIONS 2 3 7

TABLE П. SCALE FACTOR ANALYSIS: LAY PUBLIC (N = 1511, 1992)

Nuclear power generation (NPG)

I. Benefit factor

Feel N P G necessary 0.70 Feel N P G beneficial 0.74

II. Safety factor

Feel N P G safe 0.75 Feel N P G easy 0.76

T A B L E III. M U L T I P L E R E G R E S S I O N ANALYSIS: OPINION LEADERS (N = 1722, 1992)

R(NPG ACCEPTANCE) = 0.85а,Ь

ß(NPG SAFETY) = 0.38

ß(NPG BENEFIT) = 0.67

a p < 0.001. b Formula for multiple regression:

R(NPG ACCEPTANCE) = ß(NPG SAFETY) ^(NPG SAFETY) + ^(NPG BENEFIT) ^(NPG BENEFIT)-

TABLE IV. MULTIPLE REGRESSION ANALYSIS: LAY PUBLIC (N = 1722, 1992)

R(NPG ACCEPTANCE) = 0.80а Ь

ß(NPG SAFETY) = 0.36

0(NPG BENEFIT) = 0.54

3 p < 0.001. b Formula for multiple regression:

R(NPG ACCEPTANCE) = ß(NPG SAFETY) ^(NPG SAFETY) + ß(NPG BENEFIT) X(NpQ BENEFIT)- 2 3 8 POSTER PRESENTATIONS

3. PSYCHOLOGICAL DIMENSIONS OF RISK PERCEPTION

W hen submitted to a statistical operation called factor analysis, different scales may be grouped into orthogonal factors or psychological dimensions, based on their correlational characteristics. Tables I and П show the highest loading scales of the first two factors. The factor analytic results indicate that ‘benefit’ and ‘safety’ are the most salient factors in judging nuclear power generation, and similarly so in both opinion leaders and the lay public. By use of a statistical test called multiple regres­ sion analysis, it is also possible to examine whether acceptance as the dependent vari­ able can be predicted by a number of independent variables. Tables III and IV suggest that the acceptability of nuclear power generation can be significantly predicted by ‘benefit’ and ‘safety’ and that ‘benefit’ is clearly more important than ‘safety’, again similarly in both opinion leaders and the lay public.

REFERENCES

[1] UNITED NATIONS, Perception of Risks from Radiation and Other Sources, Scientific Committee on Effects of Atomic Radiation (UNSCEAR), Rep. A/AC.82/R.537, UN, New York (1993) 7-8. [2] TANAKA, Y., “ Sociology of nuclear power” , Denryoku Simposha, Tokyo (1985) (in Japanese). [3] TANAKA, Y., “ Chernobyl syndrome” , Denryoku Simposha, Tokyo (1989) (in Japanese). [4] TANAKA, Y., “ Risk perception: Analyzing images and fears” , Improving Drug Safety - A Joint Responsibility (DINKEL, R., HORISBERGER, B., TOLO, K., Eds), Springer-Verlag, New York (1991). [5] TANAKA, Y., “ Subjective culture and public acceptance of advanced science and technology” (Proc. 34th Annual Convention of the Japan Soc. Psychol. Assoc., Tokyo, Japan, 1993) 282-285. POSTER PRESENTATIONS 239

IAEA-CN-54/75P

IM PORTANT ASPECTS OF RISK PERCEPTION CONCERNING THE URANIUM INDUSTRY

L.H.S. VEIGA, E.C.S. AMARAL Instituto de Radioproteçâo e Dosimetría, Comissâo Nacional de Energía Nuclear, Rio de Janeiro, Brazil

G. BLAYLOCK Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States of America

1. INTRODUCTION

Usually in the uranium industry most of the attention has been focused on radionuclide contamination and radiation risk to health rather than on other risks, such as those arising from non-radioactive pollutants. This is mainly because of the public perception of greater risk from radiation as far as nuclear facilities are c o n c e rn e d . In Brazil, the Poços de Caldas plateau presents an opportunity for a risk evalua­ tion. This very well known region of high natural radioactivity has been used over the last two decades for intensive agricultural production as well as for the operation of a uranium mining and milling facility. M oreover, different risk perceptions exist between the operation of nuclear facilities and enhanced exposure to natural radia­ tion. These aspects are of prime concern, since they are basic to quantifying and comparing the additional acceptable risks to be considered in a decision making p ro c e ss . The agricultural activities led to an enhancement in the natural radiation exposure of the local rural population, mainly as a result of external gamma irradia­ tion and inhalation of radon and thoron progenies from the ingestion of local food products and inhalation of resuspended dust of minor radiological importance [ 1]. The mean increase in the estimated radiation risk for a prematurely induced cancer w a s a b o u t 2% for a population with the age distribution of those living on the plateau and staying there for the rest of their lives.

Regarding the impact of the uranium industry, a significant increase of radio­ nuclide concentration was observed only in surface waters of nearby rivers. Problems involved in comparison of radiation risks from both these types of activi­ ties, application of regulations, and considerations about their social and economic 2 4 0 POSTER PRESENTATIONS effects have been well described by Amaral et al. [2, 3]. However, the effect of non­ radioactive elements in surface waters from the uranium industry was not considered in previous work. This paper evaluates and discusses the contribution of radioactive and non­ radioactive pollutants present in the liquid effluent from the uranium industry to the radiation and toxicological risks of a local critical group. It also intends to call the attention of the public and regulatory agencies to important aspects of risk perception and to the potential necessity of an integrated risk assessment for licensing and control of nuclear facilities.

2 . M E T H O D S

The impact of the uranium industry was assessed for a critical group composed of adults who use the local rivers for water supply, livestock watering, fishing, vegetable irrigation, swimming and sunbathing. Multiplicative models were used to estimate the transfer under equilibrium conditions of pollutants from the river water to the different compartments (vegetables, fish, meat, milk and sediment). The potential exposure was calculated for this critical group who ingest local food products and are also exposed to gamma irradiation as a result of water immersion and sediment exposure. Among all the possible pollutants discharged into the aquatic environment by the uranium industry, 238U , 234U , 230T h , 226R a , 210P b , 210P o , M n , F “, A l, U nat, F e and Ba were characterized as the most important to be investigated [3-5]. Data on river water concentration of these pollutants reflecting the impact of mining and milling activities over the last 10 years were used in this calculation. For the radionuclides, carcinogenic radiation risk was estimated by the product of lifetime radiation exposure (units of committed effective dose equivalent) and the carcinogenic risk conversion factor for public exposure of 6.0E-02 per sievert [6]. For non-radioactive pollutants, the potential adverse effects to human health were evaluated by the hazard quotient (HQ), which is the ratio of the intake of the contaminant to the reference dose, RfD [7, 8]. A hazard index (HI) for multiple exposure pathways was calculated by summing all hazard quotients for each element and pathway. Al and Fe were excluded from this evaluation, since there is no reference dose for these elements in the literature [8].

3. RESULTS AND CONCLUSION

Results show that the estimated committed effective dose due to the uranium industry was 1.0E-1 mSv/a for all pathways of exposure, with about 90% of this value resulting from ingestion of water and local agricultural products (Table I). This POSTER PRESENTATIONS 241

TABLE I. ESTIMATED COMMITTED EFFECTIVE DOSE TO THE LOCAL CRITICAL GROUP DUE TO RADIOACTIVE POLLUTANTS

Effective dose Pathway (mSv/a)

Water 4.3E-2 Vegetables 4.7E-2 Meat 7.5E-3 Milk 2.2E-3 Fish 3.0E-3 External exposure 3.6E-4

Total 1.0E-1

T A B L E II. E S T I M A T E D H A Z A R D I N D E X T O T H E LOCAL CRITICAL GROUP DUE TO NON-RADIOACTIVE POLLUTANTS

Pathway Hazard index

Water 6.9 Vegetables 1.2 Meat 0.068 Milk 0.0087 Fish 0.081

Total 8.3

estimated dose corresponds to an excess lifetime carcinogenic risk of 3.0E-4 (50 years lifetime exposure, adult). Since the dose limit according to radiation pro­ tection standards for the public [6] corresponds to an excess lifetime carcinogenic risk of 3.0E-3, the estimated radiation risk of 3.0E-4, one order of magnitude lower, may be seen as of low radiological importance. Therefore, regulatory action to reduce this risk would be unwarranted, once the estimated risk has already been o p tim ize d . 242 POSTER PRESENTATIONS

M oreover, the estimated additional exposure due to the uranium industry is low compared to the natural radiation exposure in the region, as mentioned before. However, the risk arising from this natural source is not perceived by the public. Table II shows the estimated hazard index for the non-radioactive pollutants. A hazard index of 8.7 was estimated for all pathways, water and vegetable ingestion being the greatest contributors, HI = 7.3 and 1.2, respectively. That the hazard index is greater than 1.0 is of concern with respect to potential toxic human health e ffe c ts [8]. Among the non-radioactive pollutants, manganese is the one that raised this index above unity. It is present in uranium ore and is also added in the uranium oxidizing process at the facility. Manganese release into the environment may be seen as of prime concern for regulatory control, even though the risk from this source is not perceived by the public, either. These data show that in this kind of facility the risk from non-radioactive pollu­ tants may be of greater concern than that from radioactive pollutants, to which most of the attention has been directed. This can be a real problem in many countries, where control of radioactive and non-radioactive pollutants in effluents is performed by different regulatory agencies. The results from this work point out the need for an integrated human health risk assessment, in which all pollutants of possible con­ cern with respect to the protection of human health are considered. The different scientific databases and the reliability of the different risk evaluations are also impor­ tant dimensions to be considered.

REFERENCES

[1] AMARAL, E.C.S., ROCHEDO, E.R.R., PARETZKE, H.G., PENNA FRANCA, E., The radiological impact of agricultural activities in area of high natural radioactivity, Radiat. Protection Dosimetry 45 (1992) 289. [2] AMARAL, E.C.S., Modificaçâo da exposiçâo à Radiaçâo Natural devido a Atividades Agrícolas e Industriáis numa área de Radioatividade Natural Elevada no Brasil, PhD thesis, Federal University of Rio de Janeiro (1992) 130 pp. [3] AMARAL, E.C.S., PARETZKE, H.G., PENNA FRANCA, E., “ Agricultural versus uranium industry regarding the exposure to natural radiation” (Proc. Symp. on Nuclear Energy and the Environment, Rio de Janeiro, July 1993). [4] VEIGA, L.H.S., AMARAL, E.C.S, BIDONE, E.D., FERNANDES, H.M., “ Radio­ active and nonradioactive pollutants as contributors to the human health at mining areas” (Proc. Int. Conf. of Heavy Metals in the Environment, Toronto, Canada, September 1993). [5] PRADO, V.C.S., FERNANDES, H.M., VASCONCELLOS, L.M.H, REIS, Y.G., GODOY, J.M. and ROSA, L.P., “ Impact assessment over surface waters in Poços de Caldas uranium mine and milling site, Brazil” (Proc. Int. Conf. of Heavy Metals in the Environment, Toronto, Canada, September 1993). POSTER PRESENTATIONS 243

[6] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Recom­ mendations of the International Commission on Radiological Protection, Publica­ tion 60, Pergamon Press, Oxford and New York (1991). [7] UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, Risk Assessment Guidance for Superfund, Vol. I of Human Health Evaluation Manual, OSWER Direc­ tive 9285.7-01a, EPA Office of Emergency and Remedial Response, Washington, DC (1989). [8] UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, Integrated Risk Information System (IRIS database), EPA Office of Research and Development, Washington, DC, 1993. 2 44 POSTER PRESENTATIONS

IAEA-CN-54/84P

ANALYSIS W ITH FUZZY M EASURE THEORY OF PUBLIC ATTITUDE TOW ARDS THE USE OF NUCLEAR ENERGY

Y. NISHIWAKI, J.BENEDIKT University of Vienna

K. FUJIMOTO International Atomic Energy Agency

Vienna, Austria

C. PREYSSL European Space Agency, Noordwijk, Netherlands

1. INTRODUCTION

This paper is concerned with applying fuzzy measures and fuzzy integrals to analyse public attitude towards the use of nuclear enrgy. A questionnaire was set up and data were collected in universities. Pro groups and con groups recognize risks to a different degree, but both know about the risk. Pro groups consider that benefit or economic progress is an important aspect associated with the use of nuclear energy, while con groups do not dare to consider it important as compared with the risk. The analysis was made in co-operation with Tokyo Institute of Technology, Yokohama and Tokyo (T. Onisawa, M. Sugeno, Y. Harima), and Kinki University, Osaka (H. Kawai, H. Morishima, T. Koga) in Japan.

2. ATTITUDE MODEL

It is important to identify the structure of public acceptance or rejection when new technologies are developed and implemented. The structure of attitudes should include the essential attributes and their interrelation. In such a structural analysis the attitudes need to be decomposed into meaningful attributes by a suitable model. Fishbein et al. have proposed an attitude model as follows. This model has been used in the Joint IAEA/IIASA Risk Assessment Project of the International Atomic POSTER PRESENTATIONS 245

Energy Agency. The attitude A 0 towards an object or an event is assumed to be expressed by the summation of (e¡ x b¡):

П A 0 = e¡ x bj ( 1)

i = l where e¡ is the subject’s evaluation of the attribute i, b¡ is the strength of his belief in the attribute i about the object and n is the number of his salient beliefs about the object [1-4]. However the data obtained in this type of study may be more or less subjective, i.e. fuzzy, and the following problems may be pointed out: (1) A person does not always have an additive measure such as probability to evaluate fuzzy objects. (2) The attributes of an object in his evaluation process are not always independent of each other.

In either case a linear model such as Eq. (1) may not be applicable since it is based on the assumptions of additivity and independence. Sugeno has proposed the concepts of fuzzy measures and fuzzy integrals and applied these concepts to model human subjectivity. The concept ‘fuzzy m easure’ is considered as a generalized probability measure, which is interpreted as a subjective measure to evaluate fuzzy objects. It is not necessary to assume subjective and/or objective independence and additivity in a fuzzy integral model. Using fuzzy integrals, we can build a subjective evaluation model of a fuzzy object associated with various attributes.

3. FUZZY MEASURES

A fuzzy measure is an extended probability measure in one sense, which has, in general, only monotonicity without additivity. Let X be a universal set and (B be a Borel field. Then a set function g defined on (B with the following properties is called a fuzzy measure.

(i) g (ф) = 0, g(X) = 1. (ii) If А, В € (В and А С В, then g(A) < g(B). (iii) If F n € (B for 1 < n < oo and the sequence {Fn} is monotonie (in the sense of inclusion), then limn g(F„) = g(limn_ 0o Fn).

The triplet (X, (B, g) is called a fuzzy measure space, and g is called a fuzzy measure of (X, (B). 2 4 6 POSTER PRESENTATIONS

For applications, it is enough to consider the finite case. Let К be a finite set К = {sj, s2, ..., sn} and P(K) be a class of all the subsets of K. Then a fuzzy measure g of (K, P(K)) is characterized by the first two properties since the third one implies continuity. In particular, g({s¡}) for a subset with a single element s, is called a fuzzy density like a probability density. We denote g 1 = g({s;}). The fuzzy measure in this paper may be interpreted as the grade of subjective importance, which is non-additive and which one attaches to an attribute or attributes of an object in his evaluation process.

4. FUZZY INTEGRAL MODEL [5-8]

Let h be a measurable function from X to [0, 1]. Then the fuzzy integral of h over A with respect to g is defined as

/ a h (x ) » g = su p [а Л g(A П F„)] (2)

a € [ 0 ,l ] where Fa = {x| h(x) > a} and Л stands for minimum. A is the domain of the fuzzy integral which is omitted if A is X. Now let us see how to calculate a fuzzy integral. For simplicity, consider a fuzzy measure g of (K, P(K)) where К is a finite set previously defined. Let h:K — [0, 1] and assume without loss of generality that h(sj) > h(s2) 2: ... > h(sn). Renumber the elements of K, if not. Then we have

/h(s) ° g = V [h(s¡) Л g(K¡)] (3) 1 = 1 where K¡ = {s1( s2, ..., s¡} and V stands for maximum. Let A be an object and К = {sj...... s„} be a set of its attributes. By hA : К — [0, 1], we denote the characteristic function of the object A which can be given in two ways: (a) objectively from the physical properties of the attributes such as the characteristic function of a geometrical pattern and (b) subjectively by an evaluator, according to his judgement. Let g be a fuzzy measure to express the grade of subjective importance of the attributes in the overall evaluation of the object. Then the overall evaluation of A is given by

E a = / h A(s) » g (4 ) POSTER PRESENTATIONS 247

There are two cases in applying a fuzzy integral to evaluation problems. (i) We have r objeets similar to each other with the same attributes and model a person’s subjective evaluation process. So we have

E; = /hj(s)° g, 1 < i < r (5)

where i is the number of an object, h¡ is the characteristic function of the i-th object and g is a person’s subjective measure. This evaluation problem has been studied in Refs [5-7]. (ii) We have only one object and model the evaluation process, on an average, of m persons. In this case we also have

Ej = / h j ( s ) ° g , 1 < j < m ( 6 )

where j is the number of a person, hj is the subjective characteristic function of the object given by the j-th person, and g is the mean fuzzy measure among persons in the sense that g minimizes a performance index, e.g. Eq. (7). This is the case considered in this paper.

In short, we have many objects or many evaluators. In either case, let E* be a subjective evaluation of the j-th object or that of the unique object by the j-th person. The parameter g in the model is identified so as to minimize the perfor­ mance index

(7)

5. STRUCTURE IDENTIFICATION

Now let us discuss how to identify the structure of a model. There is no general method for structure identification. Suppose that we can list all the possible attributes of an object. D e fin e

g({Sj, Sj}) - (g1 + gj) i * j m j = \ g ¡ л g j ( 8 )

o , i = J

(9 ) 248

Щ = Y t тУ/(п “ ^ (1°) i = l

/¿¡j > - 1 is the degree of overlapping of g between the i-th and the j-th attributes. In the above if > 0, then g is super-additive for s¡ and sJ; i.e. g({s¡, sj}) > g 1 + gJ. This means that s¡ and s¡ are important since the grade of subjective impor­ tance becomes very large if s¡ and Sj are joined together. It can be considered that s¡ su p p o rts Sj. I f ¡¿¡j < 0, then g is sub-additive. It can be considered that s¡ casts a doubt on s¡ in the evaluation of an object when both are evaluated. As an extreme case, suppose that s¡ and s¡ are the same attribute. In such a case we would have

g({Si, Sj}) = g ¡ V gj

The grade of subjective importance of {s,, Sj} does not increase. From this follows

Mij — — 1 • my is th e n o rm a liz e d /¿¡j. T h e n щ is the value related to the mean degree of overlapping between the j-th attribute and other attributes. In order to indicate greater contribution of m¡j to щ when |m¡j| is large and to reflect the signs of m¡j on i?j, E q . (1 0 ) is a ssu m e d . W e c all щ 6 (-1 , 1) an overlapping coefficient. If 0 < 7?j < 1, then the j-th attribute does not overlap with others on an average, i.e. this attribute is relevant to evaluate an object.

N o w d e fin e

fj = 1 + i,j(l - g j), - 1 < Vj < 0 (11)

W e may eliminate those attributes which may be considered irrelevant in a model by referring to the value of |j, i.e. by taking account of both importance gj and overlapping rjj. £j is called a necessity coefficient expressing the degree of necessity of the j-th attribute in the structure of a model. Note that £j = 0 if ijj = —1, i.e. complete overlapping, and if gj = 0, i.e. no importance. The attribute Sj may be eliminated if the following inequality is satisfied:

------y ------< 0.7 (12) m ax j,i)¡ < о ч

where s¡ is not eliminated if ^ > 0. In other words, in case r/j > 0, s¡ may be regarded as no overlapping on an average and therefore this attribute is considered relevant and will not be eliminated. In c a se i}j < 0, Eq. (11), which takes gJ into consideration, will be used. However, a threshold 0.7 is introduced in this case so that the necessity coefficient of the rele- POSTER PRESENTATIONS 2 49 vant attribute may not have too widely different values. And, if the grade of subjec­ tive importance for sk, a fuzzy density gk, is very small among others, then we may also drop the к-th attribute sk [ 9 - 1 1 ] .

6. CONCLUSIONS

(1) Pro groups seem to consider in common that ‘economic progress’ or ‘technical and economic progress’ is an important aspect of the use of nuclear energy, although they may also recognize risks such as ‘potential for threat’. (2) Con groups are similar in that they have risk considerations such as ‘nega­ tive impacts of large scale technology’. Two attributes belong to this aspect: (a) ‘it exposes people to hazards which they cannot influence by any actions of their own’; and (b) ‘it leads to accidents which affect a large number of people at the same tim e’. They also seem to consider that the economic or fringe benefit aspect is a relevant factor but not an important one for the use of nuclear energy as compared with the risk .

REFERENCES

[1] FISHBEIN, M ., An investigation of the relationships between beliefs about an object and the attitude toward that object, Human Relat. 16 (1963) 233. [2] FISHBEIN, M ., AJZEN, I., Belief, Attitude, Intention and Behaviour: An Introduction to Theory and Research, Addison-Wesley, Reading, M A (1975). [3] NIEHAUS, F., SW ATON, E., “ Risk assessment and public acceptance on nuclear power, Progetto Vese” , Valutazione Effetti Ambientali e Socioeconomici dei Sistemi Energetici, ENEA, Rome (1981). [4] SW ATON, E., RENN, О., Attitude towards nuclear power: A comparison between three countries, W P-84-11, IIASA (1984). [5] ISH II, K ., SUGENO, M ., A model of human evaluation process using fuzzy measures, Int. J. Man-Machine Stud. 22 (1985) 19. [6] SUGENO, М ., Theory of fuzzy integrals and its applications, thesis, Tokyo Institute of Technology (1974). [7] SUGENO, М ., “Fuzzy measures and fuzzy integrals: A survey” , Fuzzy Automata and Decision Processes (GUPTA, M .M ., SARIDIS, G .N., GAINES, B.R., Eds), North- Holland, Amsterdam (1977) 89-102. [8] W IERZCHON, S.T., An algorithm for identification of fuzzy measures, Fuzzy Sets Systems 9 (1983) 69. [9] ONISAWA, T., SUGENO, М ., NISHIW AKI, Y., KAW AI, H., HARIMA, Y., Fuzzy measure analysis of public attitude towards the use of nuclear energy, Fuzzy Sets Sys­ tems 20 (1986) 259. [10] N ISH IW A K I, Y ., et al., “ Risk assessment of atmospheric contamination due to com­ bustion of fossil-fuels in Japan and possible application of fuzzy set” , (10th Regional Confr. IRPA and SFRP Annual Congr. 1982, Avignon, France, 1982). [11] NISHIW AKI, Y., KAW AI, H., MORISHIMA, H., ITAKURA, T., KOBAYASHI, S., TERANO, T., SUGENO, M ., Accidents and Human Factors (Application of Fuzzy Theory), Jülich, Germany (1984). 2 5 0 POSTER PRESENTATIONS

IAEA-CN-54/91P

PERCEPTION DES RISQUES NUCLEAIRES ET INFORM ATION

J. BRENOT, S. BONNEFOUS, P. HUBERT Institut de protection et de sûreté nucléaire (IPSN), F-92265 Fontenay-aux-Roses Cedex, F ra n c e

INTRODUCTION

La prise en compte dans la gestion des risques technologiques de leur percep­ tion par le public constitue un fait relativement récent. C’est au début des années 1970 que le caractère irréaliste de la recherche du «risque nul» a été reconnu. La notion de risque «acceptable» a été introduite et, dès lors, des divergences d’ap­ préciation entre les gestionnaires et le public se sont clairement manifestées. Pour une activité ou une situation à risques, les techniciens et gestionnaires quantifient des probabilités et des conséquences, quand les individus du public s’appuient sur tout un ensemble de critères qualitatifs pour traduire leurs perceptions. L ’étude des per­ ceptions et de leurs liens avec l’acceptabilité s’est imposée, et les résultats qui ont été obtenus sont utilisables pour l’information et la communication.

TYPES D’ETUDES

Les travaux sur les risques perçus, initiés par P. Slovic aux Etats-Unis vers 1965 [1], se poursuivent toujours. Ceci s’explique en effet par la multiplicité des situations à risques, la variété des groupes d’individus concernés et la diversité des contextes sociaux. En France, l’Institut de protection et de sûreté nucléaire (IPSN) conduit depuis 1977 des études relatives à la perception des risques, auxquelles participe la Com­ mission de Communautés européennes. Les travaux menés sont de type conceptuel (notions de risque, d’acceptabilité) ou s’appuient sur des enquêtes d’opinion. A cet effet, un baromètre de l’opinion sur la perception des risques et de la sécurité a été mis en place en 1988: il donne lieu en moyenne à deux enquêtes par an représenta­ tives de la population française. En 1991, l’Observatoire de l’opinion sur les risques et la sécurité a été créé à l’initiative de l’IPSN: il permet de réunir trois fois l’an des ingénieurs, chercheurs, experts, responsables de l’administration et des gestion­ naires qui s’intéressent à la perception par le public des risques technologiques et naturels et aux actions engagées pour les maîtriser. POSTER PRESENTATIONS 251

La plupart des résultats ci-dessous mentionnés ont été observés en France; ils sont de portée plus générale car très semblables à ceux obtenus dans de nombreux pays d’Europe et en Amérique du Nord.

HIERARCHIE DES RISQUES PERÇUS

Pour un large éventail de situations, au sein desquelles se trouvent les activités nucléaires, la hiérarchie obtenue en terme de risque perçu est assez stable au cours du temps. Son utilisation permet de relativiser certains domaines et d’en identifier d’autres qui préoccupent durablement le grand public. De plus, d’un pays à l’autre, les hiérarchies obtenues sont généralement proches. C’est vis-à-vis des déchets radioactifs que le risque perçu est le plus élevé; viennent ensuite à un degré moindre les centrales nucléaires, l’irradiation des produits alimentaires et, enfin, à un très fai­ ble niveau, les rayons X et la radioactivité naturelle. On peut ainsi s’attendre à un public plus ou moins réceptif selon le thème traité. A ce titre, informer sur le radon représente une gageure.

RISQUES PERÇUS ET RISQUES OBJECTIFS

La hiérarchie des situations en terme de risque perçu est relativement proche de celle obtenue à partir d’évaluations quantitatives (avec des données statistiques ou par modélisation) souvent identifiées aux risques «objectifs». Certains résultats de ce type ont été obtenus en France en 1986 avec des quantifications s’appuyant sur des taux de mortalité; d’autres viennent de la comparaison des perceptions du public avec celles des experts en sécurité dont les avis sont considérés comme se rapprochant de l’objectivité. Ceci conduit à rechercher des mécanismes plus sophistiqués que la sim­ ple «mésinformation» pour expliquer les différences d’attitude face au risque. Informer en postulant que le public est incohérent est une erreur. Les dimensions qui interviennent dans la perception des risques sont issues d’études qui ont un double objectif, d’une part une meilleure compréhension de la perception, et d’autre part une modélisation des attitudes et comportements du public face aux activités à risques. On a pu recenser jusqu’à 19 dimensions (Tableau 1) qui sont en relation étroite avec l’acceptabilité du risque, attitude importante à connaître dans l’optique d’une gestion [2]. Certaines dimensions concernent l’individu face au risque (de 1 à 7), d’autres la nature du risque (de 8 à 15), d’autres enfin la gestion sociale de celui-ci (de 16 à 19). Ces dimensions ne sont pas toutes pertinentes pour chaque activité; ainsi, le caractère immédiat du risque ne s’applique pas aux déchets radioactifs, au radon ou aux rayons X. Elles n’influencent pas toutes la perception du risque dans le même sens. Dans le cas des activités nucléaires par exemple, on a souvent observé qu’une 252 POSTER PRESENTATIONS

TABLEAU I. DIMENSIONS DE LA PERCEPTION

1. La familiarité 2. La compréhension 3. L ’incertain 4 . L ’acceptation tacite 5. L’implication personnelle 6. Le contrôle personnel 7. La valeur morale 8. Le potentiel catastrophique 9 . L ’existence d’un historique d’accidents 10. Le caractère immédiat 11. La réversibilité 12. L ’a p p ré h e n sio n 13. Les conséquences sur les enfants 14. Les conséquences sur les générations futures 15. L ’identification des victimes 16. L ’éq u ité 17. Le bénéfice attendu 18. La confiance envers les institutions 19. La médiatisation

importante médiatisation peut améliorer la compréhension tout en augmentant le risque perçu. A l’évidence, ces dimensions ne sont pas indépendantes; en effet, vis-à- vis d’une activité à potentiel catastrophique élevé, l’individu à la fois appréhende et ne dispose d’aucun moyen de contrôle personnel. Comme d’un risque à l’autre ce ne sont pas les mêmes dimensions qui interviennent dans la perception, il ne peut être question de développer une information stéréotypée.

VARIABILITE DES PERCEPTIONS

Pour une même situation, le risque perçu peut varier fortement d’un groupe d’individus à un autre. A été mise en évidence l’importance de facteurs psycholo­ giques (comme l’anxiété), physiologiques (tels que l’âge et le sexe), cognitifs (con­ naissance et expérience), sociaux (traduisant l’appartenance ou l’insertion), et plus généralement culturels (modèles de société). POSTER PRESENTATIONS 253

Axe 2

La drogue •

■ L e SID A

■ L'alcoolisme

■ Le tabagisme

Les accidents de travail ^

i La pollution atmosphérique Les accidents domestiques.

Le terrorisme •

Axe 1 Les déchets ménagers Les accidents de la route Les déchets chimiques » La pollution de l’eau ■ Les catastrophes naturelles ■ La pollution atmosphérique

Les raffineries de pétrole ■ Les centrales nucléaires ■ pollution automobile / . L'insécurité dans les villes " Le transport des mat. dangereuses. • Les installations chimiques

FIG. 1. Espaces de perception.

ESPACES DE PERCEPTION

Les situations peuvent être évaluées en terme de risque perçu pour l’individu ou pour la société, ou encore selon l’urgence à prendre des mesures de sécurité. Quelle que soit la façon d’évaluer, les situations se rangent en grandes catégories homogènes qu’il est possible de mettre en évidence par des techniques statistiques [3]. Ainsi, une analyse en composantes principales des réponses du public concer­ nant vingt situations évaluées sur le danger qu’elles inspirent fait apparaître quatre familles de risques (Fig.l). De façon générale, les activités industrielles ne sont pas jugées par le public de la même façon que les activités de loisirs, les activités domes­ tiques, les modes de transport ou les pratiques médicales. En conséquence, il faut éviter de comparer des situations qui n’appartiennent pas à la même catégorie. Par exemple, le radiodiagnostic et le risque radon ne peuvent être comparés aux installa­ tions nucléaires et, dans ces domaines, les politiques d’information seront très différentes quant à la nature des contenus et moyens utilisés, des responsables et des cibles visés. 2 5 4 POSTER PRESENTATIONS

Com pétence (% Oui)

FIG. 2. Plan compétence: crédibilité.

CREDIBILITE DES INFORMATIONS

La médiatisation d’une situation dangereuse, la capacité de la contrôler au plan individuel et la confiance accordée aux autorités dans un tel cas sont des points intimement liés. Nombreuses sont les études qui traitent de la compétence et de la crédibilité des intervenants dans la gestion des risques. Pour l’industrie et l’énergie nucléaires, l’image des multiples intervenants est évaluée à l’aide de jugements por­ tant sur la compétence technique et la crédibilité des informations qu’ils diffusent (Fig. 2). En France, les principaux intervenants institutionnels ont plutôt une image de compétence technique quand les associations de consommateurs, écologistes et médecins ont plutôt une image d’information crédible [4]. Ces images, fortement ancrées dans l’esprit du public, conditionnent en partie l’accueil réservé aux informa­ tions données et posent donc la question du choix du meilleur canal pour transmettre ces informations. Dans le cas des rayons X, l’individu du public n’en perçoit pas le POSTER PRESENTATIONS 2 55 risque. D’autre part, il s’en remet au corps médical pour ses expositions. Par ailleurs, les médecins ont bonne image dans le domaine radiologique. Dès lors, on peut douter de l’intérêt d’informer le public sur ce risque. Ne vaut-il pas mieux chercher une diminution des expositions et donc orienter l’effort vers les praticiens?

CONCLUSIONS

La perception des risques dus aux rayonnements et son corollaire l’acceptabi­ lité des activités nucléaires sont dans tous les pays reconnus comme étant d’une grande importance. Leur perception peut être analysée sur deux plans:

— au plan individuel: les individus inquiets et angoissés rejettent les activités nucléaires car ils refusent tout risque supplémentaire; il y a aussi ceux qui, ayant des connaissances techniques ou vivant à proximité d’une installation nucléaire, expriment des jugements en général plus favorables aux activités nucléaires; enfin, il y a tous les autres individus, c’est-à-dire la majorité, pour qui l’attitude la plus courante relative à une installation nucléaire est de dire «pas dans mon jardin», attitude à laquelle sont confrontés tous les projets d’installations industrielles importantes. Dans le cas des installations nucléaires, cette attitude s’explique par le fait que les risques apparaissent sou­ vent mal connus et inquiétants, qu’il est impossible d’exercer un contrôle per­ sonnel, qu’il faut faire confiance, et que la situation n’est généralement pas équitable car les bénéfices sont nationalement répartis quand les risques sont localement subis. — au plan sociétal: comme membres d’une communauté, les individus se sentent investis d’une responsabilité vis-à-vis des générations futures, et leurs percep­ tions des activités nucléaires sont fortement influencées par les acquis culturels, les positions politiques ou idéologiques.

Une politique d’information et de communication doit, pour être efficace, œuvrer sur les deux plans et tenir compte des dimensions de la perception. Elle peut s’appuyer pour cela sur des enquêtes nationales ou spécifiques qui fourniront des indications sur les dimensions qui interviennent dans l’acceptation ou le rejet d’une activité à risques. Connaître les points de vue et en tenir compte concourent à la recherche d’un consensus sur les risques nucléaires.

BIBLIOGRAPHIE

[1] SLOVIC, P., FISCHHOFF, B., LICHTENSTEIN, S., in Society Risk Assessment: How Safe is Safe Enough? (Schwing, Albers, Eds), Plenum Press, New York (1980) 1 8 1 -2 1 4 . 2 5 6 POSTER PRESENTATIONS

[2] COVELLO, V .T., Environmental Impact Assessment, Technology Assessment, and Risk Analysis (V.T. Covello et al., Eds), NATO ASI Series G 4, Springer Verlag, Berlin (1985) 1-14. [3] BRENOT, J., Perception of radiation risks, Proc. Conf. on Radiation Effects and Pro­ tection, Mito, Ibaraki, Japan, 1992, 329-341. [4] BONNEFOUS, S., POMMIER, S., BRENOT, J., Perception des risques et de la sécurité, Résultats du sondage de mai 1994, Note SEGR/LSEES 94/42, juin 1994. POSTER PRESENTATIONS 2 57

IAEA-CN-54/93P

M ISREPRESENTATION OF RADIATION RISK: TH E M AIN APPROACH IN THE ANTI-NUCLEAR CAM PAIGN IN RUSSIA

G.A. KAUROV Department of Information and Public Relations, Ministry of the Russian Federation for Atomic Energy, Moscow, Russian Federation

Documents from the United Nations Conference on the Environment and Development in Rio de Janeiro cite global warming, depletion of the ozone layer, acid rain and heavy metal accumulation in the ground as being due to modes of production and consumption in the developed parts of the world which need to be changed. Key to this are measures to reduce the use of non-renewable fossil fuels. As we approach the end of the twentieth century, man has actually developed means of meeting energy demands without the need to burn fossil fuels, the most important of these being nuclear power. However, a move away from power generation based on fossil fuels would not suit the all-powerful oil monopolies, as it would affect their vast profits. Consequently, they have conducted an uncompromising campaign against nuclear power. In this campaign they have exploited every opportunity to alarm the public about radiation risk, distorting the true picture. The campaign against nuclear power is particularly insidious in that it is being conducted with the aid of ecological organizations and movements which by their very nature should really be in favour of developing nuclear power. In Russia these are Greenpeace, the Socioecological Union, Doctors of the World for Nuclear Disarmament, etc., numbering about 40 movements in all. From analysis of data on the socioecological and medical consequences of the three biggest nuclear accidents in Russia (Kyshtym in 1957, Chernobyl in 1986 and Tomsk-7 in 1993), it is clear that the main approach in the campaign against nuclear power is skillful disinformation of the public as to the radiation effects of accidents. It is disturbing and ruthless in character and very cleverly directed and co-ordinated. The aim of this disinformation is to alarm the public and cause it to reject nuclear technology. The public is kept as far as possible ignorant of the fact that with appropriate technical and organizational measures it is possible to eliminate or minimize the radiation risk from any accident at a nuclear power plant or other nuclear facility. This disinformation has led to a sharp fall in the comprehension of nuclear matters by the Russian people which makes it easy for the anti-nuclear campaigners to manipulate public opinion. The public is kept largely ignorant about the ability of living beings to adapt to appreciable radiation effects and the fact that the human 2 58 POSTER PRESENTATIONS

TABLE I. LIST OF NUCLEAR EVENTS ACCORDING TO DATA OF THE CLINIC OF THE BIOPHYSICS INSTITUTE OF THE RUSSIAN MINISTRY OF HEALTH 3

P e rio d 1 9 5 0 -1 9 5 9 1 9 6 0 -1 9 6 9 1 9 7 0 -1 9 7 9 1 9 8 0 -1 9 9 2 T o ta l

N u m b e r o f 19 2 9 4 6 3 8 132 events

N u m b e r o f 7 9 156 158 4 8 2 (3 8 4 )b 875 people affected

N u m b e rs 6 5 6 7 6 9 2 1 8 (1 3 4 )b 4 1 9 diagnosed as (o f w h o m suffering from 6 5 d ie d ) acute radiation sickness or local radiation injury

a From A.K. Guskova. b The figures in parentheses refer to the accident at the Chernobyl nuclear power plant in 1 9 8 6 .

body, including that of the most zealous ‘Green’, is to a large extent formed under the effect of radiation fields. The mass media fail to report on the radiation back­ ground and the way it fluctuates in different parts of the world. An important role in the whipping up of radiation hysteria is played by the notion of a thresholdless linear dose-risk relationship, which is subscribed to by many scientists despite being in evident contradiction to the well known fact that the biological effect is dependent on the dose rate. Following the lead of the anti-nuclear campaigners, the mass media keep silent about the fact that the nuclear industry has a better accident record than any other major branch of industry. Table I shows the number of nuclear events based on data supplied by the Clinic of the Biophysics Institute, directed by A.K. Guskova. The data cover approximately 80% of the events occurring in the former USSR over a 40 year period. In that time, 419 people were injured in radiation accidents, 65 of whom died, whereas several hundred thousand people were killed or injured in accidents in the coal, gas and oil industries. By concealing from the public the lack of any proper evidence to support the non-threshold hypothesis, the anti-nuclear campaigners have succeeded in brain­ washing many millions of people in Russia and other countries of the Commonwealth POSTER PRESENTATIONS 2 5 9 of Independent States, mercilessly alarming them with tales of supposedly inescapa­ ble afflictions caused by even minor doses of radiation. Particularly disturbing are their prophecies with respect to children living in areas with increased radiation backgrounds. This all leads to many people’s losing confidence in the future, giving rise to psychological disorders, drunkenness and an increase in the suicide rate. Radiation psychosis in Russia is fostered by the numerous social organizations supposedly established to reduce the negative effects of radiation accidents (the Chernobyl Union, Chernobyl Aid, etc.). W ithout the periodic stirring up of radiation problems, these organizations would be deprived of financial support, from both within Russia and abroad. Many leaders of such organizations conduct their some­ times rather dubious business by exploiting the nuclear illiteracy of the people and Government of the country. Russian scientists and specialists try to counteract this harmful trend for the country despite often encountering aggressive resistance from the anti-nuclear factions. They are subjected to unfounded criticism and sometimes open abuse. Moreover, the ‘Greens’, headed by academician Yablokov, rudely dis­ parage without any substantiation the results of the brilliantly executed study by foreign specialists (WHO mission in 1989 and the International Chernobyl Project) sincerely trying to assist the unfortunate people of our country. All of this, coupled with the clear need to develop nuclear power in Russia, has made it necessary for the Government to set up a system in the country to tell the truth about nuclear affairs to a bewildered population overcome by fears about radiation risks. By decision of the Russian Government, an Interdepartmental Coun­ cil has been set up consisting of the deputy heads of 14 ministries and departments with the task of co-ordinating public relations work in the country. A public information centre has been set up in Moscow along with regional centres in six regions of Russia. Information centres and information task forces have been set up at all nuclear power plants and facilities of the Ministry of the Russian Federation for Atomic Energy. The basic principles for public relations work are maximum possible openness, timeliness and truthfulness of information plus a vigorous approach. The aim of the work is to implement the decisions taken by the President of Russia, the Government and a number of regional administrations to expand the energy supply in the future on the basis of nuclear power. 2 6 0 POSTER PRESENTATIONS

IAEA-CN-54/104P

RISK PERCEPTIONS OF INDONESIAN STUDENTS VS. US AND HONG KONG STUDENTS

E . H IS W A R A Centre for Standardization and Radiation Safety Research, National Atomic Energy Agency, Jakarta, Indonesia

1. INTRODUCTION

A psychometric paradigm has been widely used for studying perceived risk [1, 2]. This strategy uses psychophysical scaling and multivariate analysis tech­ niques that provide quantitative representations of risk perception. In these tech­ niques, people make judgements on the current and desired riskiness of diverse hazards and the desired level of regulation for each hazard. These judgements are then related to several characteristics that have been shown to account for their per­ ceptions, including both qualitative aspects (e.g. voluntariness, dread, knowledge) and quantitative ones (e.g. the number of deaths). The objective of the present survey was to study the risk perceptions among Indonesians, and to compare them with results of previous studies of the USA and Hong Kong, places with very different socioeconomic and cultural characteristics.

2 . M E T H O D

A two part questionnaire was administered to 64 Indonesian students in 1993. The questionnaire was a voluntary take-home assignment. The students were mostly those attending the University of Indonesia, except 20 who were high school stu d en ts. The questionnaire originally developed by Slovic et al. [2] was shortened con­ siderably, and hazard items unfamiliar to Indonesians, such as mushroom hunting and sunbathing, were eliminated.

3. R E S U L T S

Table I shows mean perceived risk ratings of Indonesian students, together with the ratings of 175 US students from a study in 1979 [2] and of 65 Hong Kong students from a study in 1985 [3]. POSTER PRESENTATIONS 261

TABLE I. COMPARISON OF PERCEIVED RISK RATINGS OF INDONESIAN VS. US AND HONG KONG STUDENTS

M e a n 3 R a n k H a z a rd

In d o n e sia USA HK In d o n e sia USA HK 19 9 3 1 9 7 9 1985 199 3 19 7 9 1 9 8 5

Nuclear weapons 89 7 8 7 8 1 1 4 AIDS 87 — 2 — — W a r fa r e 85 78 8 0 3 2 2 Heroin or opium 83 6 3 7 9 4 9 3 Alcoholic beverages 81 57 4 7 5 10 14 S m o k in g 7 6 6 8 6 9 6 8 6 F ire w o rk s 6 4 31 32 7 2 0 2 6 Mountain climbing 5 6 — 8 — Nuclear electrical power 55 72 6 8 9 6 7 C a ffe in e 51 4 0 4 5 10 21 17 Space exploration 5 0 25 3 8 11 2 6 2 2 Automobiles 4 9 — — 12 — — Radiation therapy 4 9 53 4 8 13 13 13 Pregnancy/childbirth 4 8 3 0 3 6 14 2 2 2 4 Food preservatives 4 8 4 2 5 3 15 16 11 Motor vehicles 4 7 55 63 16 11 8 Commercial aviation 4 6 31 4 2 17 19 18 S u rg e ry 4 5 — 18 — — Fire fighting 43 — 19 - — C osm etics 4 2 20 2 8 2 0 2 9 2 8 Chemical fertilizers 41 55 4 6 21 12 16 R a ilro a d s 3 8 2 9 3 9 2 2 2 4 21 Birth control pills 3 7 51 41 2 3 14 19 Non-nuclear electrical 3 2 2 6 38 2 4 25 23 p o w e r F is h in g 3 2 — — 2 5 — — Prescription drugs 3 0 — — 2 6 — — D a m s 2 9 — — 2 7 — — Home appliances 2 7 — 28 — B rid g e s 26 — 29 — B icyc le s 21 24 3 4 3 0 27 25

Overall meanb 5 4 4 6 5 0

a Scale ranged from 0 (not risky) to 100 (extremely risky). b Only to those hazards applied for students from the three countries. 262 POSTER PRESENTATIONS

■ Cosmetics

Birth control pills • " Food jreservatives ■ Non- nuclear electrical power Nuclear electrical power ■ ■ Bicycles • AIDS ■ Fishing • S ■noking • > Alcoholic beverages Automobiles ■ Heroin/opium » ■ Firew3rks Nuclear weapons •

■ Warfare

i i i t -3-2-10 1 2 3 Factor 2. Dread risk

FIG. 1. Location o f 15 hazards within the two factor space for Indonesian students.

Overall, Indonesian students’ ratings were not significantly higher than either US or Hong Kong students’ ratings, with an overall mean of 54 vs. 46 and 50 across 20 hazards. Indonesians rated 14 and 12 hazards higher and six and eight lower than US and Hong Kong students did, respectively. In the case of rank order, Indonesians rated twelve and eight hazards riskier and six and ten hazards less risky than US and Hong Kong students did, respectively. Two hazards, however, were rated the same for both Indonesian and US students (nuclear weapons and radiation therapy) and Indonesian and Hong Kong students (smoking and radiation therapy). It was also found that six hazards (i.e. fireworks, space exploration, alcoholic beverages, cosmetics, pregnancy/childbirth and nuclear weapons) were rated sub­ stantially (±10) riskier by Indonesians than by both US and Hong Kong students. In addition, heroin or opium, commercial aviation and caffeine were rated substan­ tially riskier by Indonesian than by US students. In the case of factor structure, the Indonesian students’ factor structure did not vary from that of both US and Hong Kong students (Fig. 1). US students were higher than Indonesians on dread risk only for birth control pills and nuclear electrical POSTER PRESENTATIONS 2 63 power, also on unknown risk only for nuclear electrical power, while Hong Kong students were higher on dread risk for cosmetics and non-nuclear electrical power. Indonesians were higher than Hong Kong students on dread risk for heroin or opium.

4. CONCLUSION

The findings from this comparative analysis show that the risk perceptions of Indonesian, US and Hong Kong students were not extremely different. Although there were a number of exceptions, most hazards were perceived in similar patterns. This is perhaps due to the fact that cultures around the world are gradually converg­ ing toward a common end point. W ith the rapid growth in information technology, people with dissimilar cultures may receive the same information, ideas, views, etc. This, in turn, will drive them to have similar perceptions of risks.

REFERENCES

[1] FISCHHOFF, В., SLOVIC, P., LICHTENSTEIN, S., READ, S., COMBS, B., How safe is safe enough?, A psychometric study of attitudes toward technological risks and benefits, Policy Sei. 8 (1978) 127. [2] SLOVIC, P., FISCHHOFF, В., LICHTENSTEIN, S., “Facts and fears: understand­ ing perceived risk” , Societal Risk Assessment: How Safe is Safe Enough? (SCHW ING, R., ALBERS, W .A ., Jr., Eds), Plenum Press, New York (1980). [3] KEOW N, C .F., Risk perceptions of Hong Kongese vs. Americans, Risk Analysis 9 3 (1989) 401. 264 POSTER PRESENTATIONS

IAEA-CN-54/114P

RADIOECOLOGICAL EDUCATION AND PERCEPTION OF RADIATION RISK IN BELARUS

A.I. STAVROV National Training, Research and Information Centre for Radiation Protection, Power Engineering and Radioecological Education, Minsk, Belarus

The energy crisis in Belarus, caused by the practically complete absence of its own fuel resources (only 10.5% of requirements are covered by the production of about 2 million tons of oil, 0.5 billion cubic metres of gas, a small quantity of peat, etc.), has resurrected interest in nuclear power engineering. The question of whether to build a nuclear power plant (NPP) has been put on the agenda of the day in condi­ tions where almost a quarter of the territory of Belarus is contaminated by radiation, the population has endured serious consequences from the Chernobyl NPP accident, and the national economy faces considerable difficulties in overcoming these conse­ quences. At the same time, the solution to the problem of power supply in Belarus will influence the whole area of economic reforms and consequently the fulfilment of the State programme for overcoming the consequences of the catastrophe. The perceptions of a considerable part of the population of Belarus concerning the possi­ bility of construction of a NPP have formed through the prism of this catastrophe and are often based on incorrect representation of the degree of radiation risk from the Chernobyl accident as well as from operation of a future NPP. In this connection, a specific necessity is radioecological education of the population and management bodies of Belarus. Radioecological literacy of all sectors of society is a guarantee of adequate perception of real radiation risk, which will permit effective solution of the problems of rehabilitation of radioactively contaminated areas and provide answers to the questions related to development of nuclear power engineering. Thus, correct understanding by the public of radiological risk is not just a guarantee of support for the idea of development of nuclear power engineering, but first is a basis for ade­ quate understanding of the consequences of catastrophe and formation of correspond­ ing models for behaviour in conditions of radioactive contamination of the environment. This in turn will guarantee to a considerable extent the radiation pro­ tection of citizens living in the regions that have suffered from the Chernobyl acci­ dent, so that they can overcome psychological stress and thereby improve their state of health. For the realization of these tasks, in January 1993 the National Training, Research and Information Centre for Radiation Protection, Power Engineering and POSTER PRESENTATIONS 265

Radioecological Education (BelTRIC) was founded. Its main activities are the fol­ lowing: radioecological education of specialists who are working to overcome the consequences of the Chernobyl accident in Belarus, and provision to the public and to management bodies of information about radiation protection, radioecology, con­ sequences of the Chernobyl accident, and issues related to nuclear power engineer­ ing. The experience of the Centre has shown:

(1) The main objective of the Centre should be informational and educational activity directed toward the radioecological education of specialists in the national infrastructure (the network of radiation control organizations, scien­ tific institutes, etc.) oriented towards fulfilment of the State programme for overcoming the consequences of the Chernobyl accident. (2) These specialists can become the staff bases for the overall radioecological education of the population, because the citizens of Belarus trust them to a great extent. This will lead to formation of correct representation of radiation risk that will increase the effectiveness of work on overcoming the conse­ quences of the Chernobyl accident. (3) Radioecological education of the members of the mass media in the Republic is of great importance. A large part of the population has a notable trust in newspaper articles and radio and TV programmes, which permits effective use of the media to provide the population of Belarus with radioecological k n o w le d g e . (4) Wide use of world experience in providing radiation protection is of specific importance. As a result of the complicated situation in Belarus, the population is more inclined to trust foreign scientists and specialists. Effective information exchange is necessary first of all in the sphere of radioecological education and training of specialists. Realization of BelTRIC’s approaches mentioned above has demonstrated noticeable results. Over 3000 specialists trained in radioecol­ ogy at the Centre strongly influence the formation of correct representation of radiation risk by the various sectors of the population and management bodies of Belarus. The practice of organizing uniformly composed groups of specialists (radiation control specialists, representatives of local administra­ tion, ministry workers, journalists, etc.) into specific education programmes taking into account their professional orientation is recognized as quite effec­ tive. Such a practice has attracted to the training process highly qualified scien­ tists and specialists in the field of radiation protection. In turn, correct choice of groups of students permits the formation of a correct understanding of the radiation risks connected with overcoming the consequences of the Chernobyl accident and developing nuclear power engineering in Belarus. 2 6 6 POSTER PRESENTATIONS

IAEA-CN-54/115P

COM PREHENDING RADIATION RISKS BY ASSESSM ENT OF INCIDENTS AND THEIR CAUSES IN LICENSED ACTIVITIES

R.J. CATLIN University of Texas - Houston Health Science Center, Houston, Texas, United States of America

1. INTRODUCTION

Perception of radiation risks can be enhanced by careful assessment of the types and causes of incidents that involve sources of radioactivity in licensed activi­ ties over time. Data for this present assessment are taken from reports [1-9]

Incident type

Overexposure 144 Badge overexposure 76 High bioassay Misadministration 51 Dose irregularity Dispensing irreg. Contamination - RM Lost rad. material Abandoned source Stolen rad. material Unauth. RM disposal Unauth. RM shipment Lost/stolen X ray Source disconnect Leaking source Equip, malfunction Damaged gauge Transport accident Exp. assess, error Laser exp./injury Fire - rad. material RM in scrap Other

0 20 40 60 80 100 120 140 160 Number of incidents (total 509)

FIG. 1. Radiation incident experience, State of Texas, USA: third quarter 1992-second quarter 1994. POSTER PRESENTATIONS 267 published quarterly by the Bureau of Radiation Control (BRC) of the State of Texas Department of Health, USA, which provide individual details on the identity of the licensee or registrant, type of work activity, and the nature, cause and validity of the incident. W hile intended to develop perspectives of risks to licensed users and the general public from incidents, BRC summaries are restricted to identification of

TABLE I. RADIATION INCIDENT EXPERIENCE, STATE OF TEXAS, USA: THIRD QUARTER 1992-SECOND QUARTER 1994

In c id e n t % o f to tal T y p e N u m b e r

Overexposure 199 2 8 .3 Badge overexposure 7 6 1 4 .9 High bioassay 3 0 .6 Misadministration 51 1 0 .0 Dose irregularity 14 2 .8

Dispensing irregularity 10 2 .0 Contamination — radioactive material 14 2 .8

Lost radioactive material 2 8 5 .5 Abandoned source 10 2 .0 Stolen radioactive material 17 3 .3 Unauthorized radioactive material disposal 18 3 .5 Lost/stolen X ray machine 4 0 .8 Unauthorized radioactive material shipment 4 0 .8 Source disconnect 2 5 4 .9 Leaking source 21 4 .1 Equipment malfunction 15 2 .9 Damaged gauge 10 2 .0 Radioactive material transport accident 7 1 .4 Exposure assessment error 1 0 .2 Laser exposure/injury 3 0 .6

Radioactive material fire 7 1 .4 Radioactive material in scrap 8 1 .6 O th e r 19 3 .7

T o ta l 5 0 9 1 0 0 .0 2 68 POSTER PRESENTATIONS hospitals and radiographers where radiation overexposures have occurred. These quarterly reports are important to risk comprehension because the incidents involve both workers and members of the general public, especially exposures and risks of exposure from medical activities. This assessment is focused on repetitive patterns of incident types and their causes, in order to facilitate better understanding of risk severity and potential, and the planning needed for risk minimization and avoidance.

2 . M E T H O D

Some 509 radiation incidents involving licensed users and registrants were filed with the Texas BRC for the period April 1992 through June 1994, as shown in Fig. 1. During the same period, the BRC received about 200 complaints, of which 50% were found to be regulatory violations, and the other 50% to be without merit. This assessment reviews only the reported incident experience. All incidents were categorized by the BRC as to type of incident, name of licensee or registrant, and location in Texas. Categories of incidents adopted by the BRC were used primarily in this assessment, as given in Table I, and unique descriptions were converted to the more general BRC category. In case of multiple causes, the predominant cause was selected for categorization and secondary factors were not addressed further.

3. RESULTS AND CONCLUSIONS

3.1. Results

The 509 radiation incidents reported by Texas licensees and registrants included 144 cases of overexposure of personnel, of which 101 resulted from errors made by workers, three were due to equipment failure, and 40 were of undetermined origin, as shown in Fig. 2. In 42 cases, overexposure was caused by overload of health care workers from excessive medical service schedules. Surveys and other exposure control procedures were not followed in 30 instances. In 23 cases, radiation dosimeters came into close proximity to radiation sources under circumstances where the overexposure of the wearer could not be ruled out. Overexposures of radiation dosimeters not involving exposure to personnel occurred in 76 incidents, as shown in Fig.l. The principal cause, as shown in Fig. 3, was the inadvertent proximity of the dosimeter to sources of radiation that occurred in 38 cases. Deliberate tampering with dosimeters occurred 11 times. In four instances, the dosimeters were misassembled or otherwise damaged. Dosimeters were exposed to radiation three times while worn by workers during medical examinations or treatment. Causes of dosimeter overexposure were undeter­ mined in nine instances. POSTER PRESENTATIONS 2 69

Causes

■ Worker error | 1101

Med. proc. overload 42

Procedures ignored 30

Badge near source 23

Other

I Equipment failure ■ 3

Undetermined 1 1 20 40 60 80 1 0 0 1 2 0 Number of incidents (total 144)

FIG. 2. Overexposure (personnel) incidents, State o f Texas, USA: licensees and registrants, third quarter 1992-second quarter 1994.

Causes Badge near source

Dosimeter tampering

Dosim. misassembly

Dosimeter damaged

Dose from med. proc.

Incorrect readout

Undetermined _L I 0 1 0 2 0 3 0 4 0 Number of incidents (total 76)

FIG. 3. Badge overexposure incidents, State of Texas, USA: licensees and registrants, third quarter 1992-second quarter 1994. 2 7 0 POSTER PRESENTATIONS

Causes Wrong pharmaceut.

Wrong patient

Incorrect procedure

Incorrect dosage

Biol, interference

Other _L 5 10 15 2 0 25 Number of incidents (total 51)

FIG. 4. Radiopharmaceutical misadministration incidents, State o f Texas, USA: licensees and registrants, third quarter 1992-second quarter 1994.

Causes Wrong pharmaceut. [

Incorrect dosage

Incorrect procedure

Displaced implant

Loss of tag

Other

6 8 10 12 14 Number of incidents (total 24)

FIG. 5. Radiopharmaceutical dose or dispensing irregularity incidents, State o f Texas, USA: licensees and registrants, third quarter 1992-second quarter 1994. POSTER PRESENTATIONS 271

The misadministration of radiopharmaceuticals to patients occurred in 51 cases, including use of the wrong pharmaceutical for 24 patients, administration to the wrong patient 11 times, and use of incorrect procedures in 11 instances, as shown in Fig. 4. In two cases, the dosage was incorrect. There were one case of bio­ logical interference with medication and two other less defined cases. In addition to the misadministration cases, there were 24 incidents that involved dose irregularities or dispensing irregularities, as shown in Fig. 5, in which risk to patients was low or absent. Half of these incidents, 12 in number, involved preparation of the wrong pharmaceutical. Other cases, smaller in number but of equal potential, involved incorrect dosage or procedure. In four cases, there was loss of the radioactive tag, and in one case the patient displaced an implanted source to an unwanted location. Other principal incidents of a repetitive nature, as given in Table I, included lost and stolen radioactive material (45 incidents), disconnection of radiography sources from their cables or housings (25 incidents), leaking sources (21 instances), and equipment malfunction (15 incidents). In ten cases, radioactive sources were abandoned in underground boreholes. Of the total of 509 incidents, 55% occurred

Number of incidents 60

50

40

30

2 0

1 0

0 23412341 23

I 92 I 93 I 94 I Calendar quarter/year

Incident type

Ш Overexposure Ц Misadministration Щ Loss of rad. mat’l I I Badge overexposure Dose/dispens. irreg.

FIG. 6. Radiation incident experience, State o f Texas, USA: licensees and registrants, prin­ cipal incident distribution by calendar quarter. 272 POSTER PRESENTATIONS at medical facilities, about 42% occurred in industry (about 40% of these involved industrial radiographers), and the remaining 3% took place at universities.

3.2. Conclusions

Review of the data in Figs 1-5 shows that worker error is the chief contributor to radiation exposures and to medical incidents. Some of the personnel overexposure cases had serious potential for high or severe doses; most occurred as technical exposures near the levels of regulatory dose limits. On the other hand, many of the medical incidents have serious potential for risk to the general public, for both the intended patient and others unintended. Further analyses are needed to provide understanding of the significance of risk from such incidents and means for its avoidance by regulators, workers and the public. For example, as shown in Fig. 6, the principal types of incidents appear randomly distributed over time, indicating that regulatory authorities and users of sources of ionizing radiation are not learning from their mistakes or those of others. The public also needs to become aware of the possi­ ble public health effects of the increasing usage of radioactive materials and sources in industrial and medical activities. However, these 509 incidents generally represent a small fraction of the activities in the many successful programmes for the use of sources of ionizing radiation. Regulatory authorities need to assist their licensees, registrants and the general public in the understanding of risks from these incidents, and in ways to minimize their occurrence.

REFERENCES

[1] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Second Quarter 1992, Austin, Texas (1992). [2] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Third Quarter 1992, Austin, Texas (1992). [3] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Fourth Quarter 1992, Austin, Texas (1992). [4] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, First Quarter 1993, Austin, Texas (1993). [5] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Second Quarter 1993, Austin, Texas (1993). [6] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Third Quarter 1993, Austin, Texas (1993). [7] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Fourth Quarter 1993, Austin, Texas (1993). [8] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, First Quarter 1994, Austin, Texas (1994). [9] TEXAS BUREAU OF RADIATION CONTROL, Incident Summary, Second Quarter 1994, Austin, Texas (1994). POSTER PRESENTATIONS 273

IAEA-CN-54/116P

PERCEPTION OF RADIATION RISKS ASSOCIATED W ITH THE DISPOSAL OF RADIOACTIVE TRANSURANIC W ASTE IN NEW M EXICO

R .H . N E IL L New Mexico Environmental Evaluation Group, Albuquerque, New Mexico, United States of America

1. INTRODUCTION

The Waste Isolation Pilot Plant (W IPP) is a planned geological repository for the permanent disposal of 178 000 m 3 of transuranic (TRU) waste resulting from the defence programmes of the United States of America. The repository is located in southeastern New Mexico at a depth of 653 m in the lower part of a 600 m thick Permian age (225 million years old) bedded rock formation known as the Salado fo rm a tio n . Potential radiation exposure from transportation, operational emplacement and long term disposal can be calculated on the basis of routine operations or accidental releases. Since there is a considerable amount of experience on the transportation and management of radioactive materials, there is relatively little concern in comparison to that about predictions of the behaviour of radioactive waste over 10 000 y e a rs after disposal. In 1978, the State of New M exico had a number of concerns about the radiation risk from the proposed radioactive waste repository. The US Department of Energy agreed to fund an independent technical oversight group called the Environmental Evaluation Group (EEG) to evaluate the impact on public health and the environ­ ment. This independent review, frequently quite critical, has done much to assure the policymakers in both the New Mexico Executive Branch and the Legislature, as well as the public, that a technical oversight agency, totally independent of the Federal government, is evaluating the impact of WIPP on the public health and sa fety .

2. RISK PERCEPTION ISSUES

Specific recommendations for an oversight group such as EEG to help reduce perceived risk are as follows:

(a) Unbiased work: It is essential that such work be unbiased in its analyses and be neither a proponent nor an opponent in its outlook. 2 7 4 POSTER PRESENTATIONS

(b) Scientific peer review: The work must be subjected to scientific peer review, because the validity of our conclusions is dependent upon professional and public scrutiny. (c) Publication of results: Viability is a function of visibility and the EEG has pub­ lished over 55 major reports. Presentations are made at legislative hearings, Congressional hearings, and other state, national and international professional conferences. (d ) Openness: Direct availability and accessibility to the press and the public are e sse n tia l.

Factors affecting public perception and acceptability include the following:

(a) Period of isolation: A unique aspect to public acceptability of the disposal of radioactive waste is assurance that the unwanted radioactive residuals will not return to the environment and pose a risk to people at some distant time after disposal. It is difficult to convince people that our ability to predict over such time spans warrants such confidence. However, the salt beds where these wastes are to be entombed for 10 000 years were precipitated 225 million years ago, and this fact together with the plans for multiple natural and engineered barriers helps mitigate such concerns. (b ) Absolute risks: The US Environmental Protection Agency has established limits on potential radiation risks to future generations of 1000 latent cancer fatalities over 10 000 years. While an increment in societal risk of 0.1 death per year from cancer is vanishingly small in the incidence of cancer in the population, it can be of considerable concern as an acceptable risk to the individual. (c) Relative risks: The public generally rejects as irrelevant any comparisons using relative risks such as cigarette smoking and auto accidents. While such com­ parisons may be intended only to provide a perspective, they are frequently viewed as justifications for the exposure. (d ) Changes in mission: Changes in the expected size of the expected inventory of waste, the inclusion and then deletion of high level waste, experiments deemed necessary and then discarded have contributed to public apprehension about whether technical and other authorities have a clear sense of direction. (e) Changes in regulatory authorities: Licensing authority for the radioactive waste shipping container was given to the US Nuclear Regulatory Commission (NRC) in 1989, for the safe management and disposal of TRU waste to the US Environmental Protection Agency (EPA) in 1992, and for mine safety to the US M ine Safety and Health Administration (MSHA) in 1992. It is essential that there be a separation of responsibilities between the agency charged with the disposal of the waste and the agency having the regulatory authority to estab­ lish the standards for acceptable risk and to approve the application. POSTER PRESENTATIONS 2 75

(f) Scheduling: It is essential that the public believes that schedules for the disposal of radioactive waste are based on prudent scientific analyses of technical and socioeconomic issues of concern and not on arbitrary dates to begin disposal. The public must be convinced that all avenues of differing technical concerns are being addressed and adverse views fully considered. (g) Technical complexity: The units used to describe radiation are complex. Trans­ uranic waste includes the very long lived radioactive actinides whose activity exceeds 3.7 x 10 6 Bq/kg waste. While this does not convey much informa­ tion to the public, it may not be any clearer to say that each gram of waste has 222 000 radioactive atoms disintegrating each minute for thousands of years.

3 . S U M M A R Y

There are many highly significant and complex factors that bear on public con­ cern. Ionizing radiation is an extraordinarily effective tool that society is unwilling to abandon, whether for medical diagnostic and therapeutic applications, energy con­ version systems for electrical power, or fundamental research. The need is not to educate the public, but to create a climate in which the public has confidence that our scientific and technical processes are open and structured, and that they fully address conflicting and adverse points of view. It is hoped that, in this way, differ­ ences between the actual risks of environmental radiation exposure and the perceived risks will converge. 2 7 6 POSTER PRESENTATIONS

IAEA-CN-54/120P

PERCEPTION AND INTERPRETATIONS OF RADIATION RISK IN THE POPULATION: A PRELIM INARY REPORT

A. T0NESSEN, J.B. REIT AN, P. STRAND Norwegian Radiation Protection Authority

R.WALDAHL, L. WEISÆTH University of Oslo,

Oslo, Norway

B. MÁRDBERG National Defence Research Establishment, Stockholm, Sweden

1. INTRODUCTION

Ionizing radiation has a tremendous potential for awakening strong reactions in the population. The aim of this study was to describe the diversity of ways radia­ tion risk is perceived and interpreted in the Norwegian population. At this exploratory level, it was chosen to do a representative opinion poll of Norwegians. The poll was done in June 1993 by personal interview with 1005 respondents and was conducted by the Norsk Gallup opinion poll institute. The ques­ tions asked in the interview covered aspects such as respondents’ attention to the issue, their knowledge, and their confidence in various information sources. Some results from the survey were presented at the Fourth International Kongsvoll Sympo­ sium [1]. This paper will focus on other parts of the survey, mainly on gender differ­ ences and probabilities of different reactions as a consequence of a nuclear e m e rg e n c y . In short, the main findings presented earlier [1, 2] are that for 65 % of the pub­ lic the issue of ionizing radiation is very important and 80% report feeling poorly prepared to face a nuclear emergency. Seven years after Chernobyl, when asked how strongly they would react to another accident, the response is on the same scale as if their neighbourhood had been exposed to a serious pollution situation. From L. W eisæth’s previous research we were able to look into the trend from 1986 to 1993. There seems to be a trend of sensitization rather than adaptation in the population. POSTER PRESENTATIONS 2 77

This is more apparent when we also look into figures from an opinion poll done in June 1994 [3]. The answer to a question about the public’s expectations of another accident is more related to how important the respondents consider the issue than to how concerned they are on a daily basis with the issue.

2 . R E S U L T S

As in previous research [4-6] done on the issue of public reactions to radio­ activity, we find clear gender differences in our data. The male respondents evaluate themselves as follows with respect to the female respondents:

— to a greater extent to follow media coverage of the issue; — to a greater degree to understand the media coverage; — to know better where to get information in case of a nuclear emergency; — to be better informed about the issue; — to have better knowledge about how to protect themselves in case of ‘fallout’ ; — to be better prepared to deal with a nuclear emergency.

‘Radiation is more dangerous than I used to believe’

Education level

- -Д- - Male mean —о — Female mean

FIG. 1. Interaction effect of gender and educational level with respect to the trend in how dangerous the public believes radiation is. 278 POSTER PRESENTATIONS

Female respondents report that they:

— are more afraid of being exposed to ionizing radiation from the surroundings; — are more afraid of being exposed to ionizing radiation from stored nuclear w e ap o n s; — are more afraid of leukaemia as a harmful consequence of ionizing radiation; — are more afraid of being exposed to ionizing radiation from nuclear power p lan ts; — receive a higher share of their dose level from radioactive fallout.

If educational level is used as an indicator of knowledge, the hypothesis that more knowledge will lead to less concern may not be appropriate for 50% of the population. Figure 1 shows that females with higher education agree more often with the statement that “ ionizing radiation is more dangerous than I used to believe” while their male counterparts to a greater extent express disagreement with the statement. It is also a gender effect (but not in interaction with educational level) that female respondents express more disagreement with the statement “ Chernobyl meant little for my view on the danger associated with ionizing radiation” . Besides the effects of gender, there is a general tendency that older individuals are more concerned with the issue of radioactive contamination and nuclear acci­ dents. Also, the left wing side of the political arena is more concerned about the issue than the right.

2.1. Reactions in case of radioactive contamination due to a nuclear accident

The respondents were presented ten different ways of reacting in case of a situation with radioactive contamination due to a nuclear accident in the district where the respondents live. Each coping alternative was presented one by one, and the respondents were asked to judge how likely each reaction was. They were asked to choose one of the answer categories: very likely, fairly likely, not very likely and don’t know. Table I gives the ‘judged likelihood’ for ten possible reactions to radioactive contamination due to a nuclear accident. In this way of screening probabilities, it is of course a problem that the ten alternatives given may not really cover important response patterns for the respondent. W ith this important limitation, of the ten listed alternatives the three most likely reactions are to get hold of all possible information, to follow advice given by authorities, and to find out how health personnel and experts deal with the situation. The high likelihood that one will seek to find out ‘experts’ behaviour’ reminds us how very important ‘role behaviour’ is. It is not only what advice the experts give but their response in fact to the situation. POSTER PRESENTATIONS 2 7 9

TABLE I. FREQUENCY DISTRIBUTIONS IN PER CENT OF THE

TEN REACTION ALTERNATIVES

Estimated likelihood for the reaction Presentation Reaction alternative number3 Very Fairly Not very Do not likely likely likely know

1 “ Do nothing particular” 11.8 22.2 56.9 9.1 10 “ Leave the place” 11.5 17.4 56.4 14.6 2 “ Wait and see” 24.5 45.2 25.7 4.7 3 “ Find out how the neighbours cope” 34.9 38.7 22.0 4.4 9 “ Get iodine tablets” 36.9 23.2 15.0 24.9 7 “ Keep oneself indoors” 52.0 32.4 9.3 6.3 8 “ Be careful about food and water” 50.9 36.1 5.9 7.1 4 “ Find out how experts cope” 68.5 24.3 4.6 2.7 6 “ Follow authorities’ advice” 77.5 19.6 0.9 2.0 5 “ Get all possible information” 79.0 17.7 1.5 1.8

a The order in which the coping alternatives were presented.

— — — — — — — — — — — — — — — — — — — — — — — — — € )0 Nothing particular

B Leave the place

A "Wait and see"

y "Do as neighbours"

▼ ------V n «t inrilnA tablet»

______^ Keep oneself Indoors S -.г.г.Г.Г.Г.Г.Г...... - r - ______...... “ ' ■ < ^ Careful food/water

ф ■■■ — 11 ————— "Do as the experts"

я Follow authority ad. шшт t Most likely reaction ♦ Get all poss.lnform.

FIG. 2. Probability often response alternatives in case o f a nuclear accident that affects the respondents ’ district with radioactive contamination. 2 8 0 POSTER PRESENTATIONS

A principal component factor analysis with varimax rotation groups the alter­ natives into three components1: factor 1, alternatives 6, 4, 5 and 3 in a ‘gather information’ coping strategy; factor 2, alternatives 8, 7, 9 and 10 in an action oriented coping strategy; and factor 3; alternatives 1 and 2 in a ‘wait and see’ s tra te g y . The empirical grouping of alternatives does give meaningful groups, but this three factor solution explains only 55% of the total variance.

2.2. Gender differences in likelihood of different reactions

In Fig. 2 the gradient of the line shows the gender differences; an inclining line indicates a more likely male reaction and a declining line indicates a more likely female reaction. There are three gender differences that are significant2 a t th e 0.05 level as a main effect in a one way ANOVA: to find out how the neighbours cope with the situation, to get hold of iodine tablets and to keep oneself indoors. Females respond that they more likely will try to find out what the neighbours do and to keep themselves indoors, while getting hold of iodine tablets is indicated to be more likely by male respondents.

2.3. Educational level

W hen we split the respondents into three groups according to their educational level, there are significant3 (0.05) main effects on four of the ten response alterna­ tives. The four reactions where there is a main effect of educational level are (effect of higher educational level in parentheses):

— “ do nothing particular” (less likely with higher education); — “ find out how health personnel and other experts deal with the situation” (more likely); — “ get hold of all possible information” (more likely); — “ follow as much as possible the advice given by authorities” (more likely).

1 The alternatives are identified by presentation number, and the order in which each alternative is mentioned for a factor indicates that this alternative has the highest factor lo a d in g . 2 O f course, significance level is problematic because respondents were sampled to represent the Norwegian population, and not sampled in the two groups separately. 3 When we consider significance level we must keep in mind the large number of Ns in the data and that statistical tests such as x 2 are inflated by sample size. POSTER PRESENTATIONS 281

2.4. Interaction effect of gender and educational level

There are interaction effects (as also shown in Fig. 1) of gender and educa­ tional level when it comes to the response alternatives “ follow as much as possible the advice given by authorities” and “ get hold of iodine tablets” . Female respon­ dents with higher education answer that they are more inclined to follow authorities’ advice than their male counterparts, and higher educated male respondents answer that a demand for iodine tablets will be more likely among them in case of radioactive contamination.

3. DISCUSSION

A fundamental question in this line of research is how to explain the clear gender differences in the data. W hat part of the gender differences could be reason­ ably attributed to, for instance, the data sampling situation? The personal interview situation is a face to face dialogue. Because of sex stereotypies in our culture, it may be harder for male than for female respondents to admit their lack of knowledge, their fears, etc. Put another way, what part of the observed gender difference is due to a data gathering problem and what part is more directly connected to the issue of radioactivity? W hen it comes to the observed gender differences on information related ques­ tions in the survey, there is a consistent pattern. W hatever is the chicken or the egg, it is not unreasonable that when one understands very little of the media coverage (as more often reported by females) then there is not so much point in giving one’s attention to the media coverage. W ith less knowledge about what to do in a fallout situation, or about where to get information, it is not unreasonable to feel less pre­ pared to face an emergency scenario. W hen it comes to informing the public, it may seem like 50% of the target group (the population) is neglected or overlooked. To look further into the observed gender differences will be a main objective for the project group. The project group will also try to detect patterns of variation between individuals by using latent profile analysis (LPA) [7] on the survey data. The LPA method postulates groups defined by the way respondents have answered the survey. The groups constructed from this analysis will be further studied in order to see if there are some commonalities, e.g. liKgroup demographics.

4. CONCLUSIONS

If we look at the literature in this field, some experts take as their point of departure that the population apparently is ‘wrong’ or misinformed about the risk magnitude [8]. W hen the majority of the population is more afraid of ionizing radia- 282 POSTER PRESENTATIONS tion, which has an estimated ‘objectively computed’ lower risk than, say, smoking, then one has nearly ‘proved’ the population to be wrong. O f course, it is important to calculate risk parameters, but when it comes to understanding reactions in the population there are many important missing links. For instance, we know from our survey data that the respondents’ risk estimate is not most important in predicting how concerned the respondents say they are with the issue. Our starting point must be genuine respect for the diversity of ways people per­ ceive and interpret radiation risk. W hen it comes to value systems, it is obvious that no one has any means to construct a private truth about reality that has more legitimacy or is more true than others’, but how our values influence our thoughts and feelings about radiation is easily missed. No ‘private world experience’ is more or less valid because of congruence or no congruence with a risk estimate at the inter- subjective level. It is surely unethical to try to manipulate — to make the public believe, for instance, that nuclear power plants are safe just because it would be convenient from a governmental point of view. Such conduct would be close to an elite or expert rule system, but in our democracy it should be the other way around, i.e. the government represents the people, so that if the public is concerned, so should the government be. Experts can and will give advice about risk for accidents and what consequences are likely if some catastrophe happens, but the values assigned to those consequences must be decided by the public.

REFERENCES

[1] T0NNESSEN, A., REIT AN, J.B., STRAND, P., W ALDAHL, R., WEISÆ TH.L., “ Interpretation of radiation risk in the Norwegian population: A national survey in 1993” (Proc. Kongsvoll Symp.), TAPIR, Trondheim, Norway (in press). [2] NORWEGIAN RADIATION PROTECTION AUTHORITY, Strâlevem nytt 6-94 (Newsletter 6/94); Engstelse for radioaktiv forurensning og atomulykker (1994). [3] M Æ RLIE, M ., Perception of risk from electromagnetic fields (in preparation). [4] W EISÆ TH, L., “ Psychosocial reactions in Norway to nuclear fallout from the Chernobyl disaster” , Communities at Risk: Collective Responses to Technological Hazards (COUCH, S.R., KROLL SM ITH, J.S., Eds), Peter Lang Publishing, New York (1991) 53-80. [5] DROTTZ, B.M ., SJ0BERG, L., Opplevelser i samband med Tjemobylolyckan: En intervjuundersokning om oro, strâlningsrisker och kärnkraft, Forskningsrapport 1-86, Psykologisk Metod AB. [6] BAUM , A ., GATCHEL, R.J., SHAEFFER, M .A ., Emotional, behavioral and psycho­ logical effects of chronic stress at Three M ile Island, J. Consult. Clin. Psychol. 51 (1983) 565. [7] M ÁRDBERG, B., LPA2: A Fortran V computer program for Green’s solution of latent class analysis applied to latent profile analysis, Educ. Psychol. Meas. 35 (1975) 163. [8] COHEN, B., Nuclear journalism lies, damned lies and news reports, Policy Rev. 26 1 (1983) 70. POSTER PRESENTATIONS 283

IAEA-CN-54/121P

AVERSION TO RADIATION: AN ETHICAL PERSPECTIVE

D.H. OUGHTON Isotope Laboratory, Laboratory for Analytical Chemistry, Agricultural University of Norway, A s , N o rw a y

1. INTRODUCTION

Evidence of an irrational, exaggerated fear of radiation in the general public is often said to be provided by people’s expressing a greater aversion to the low prob­ ability of death arising from radiation exposure than to the higher probability of death from other actions (e.g. transport risks) [1]. The radiation health detriment is a mathematical representation of radiation risks, defined as the product of the probabil­ ity and the severity of physical harm, wherein harm is represented by death, non- fatal cancer and hereditary effects. In an appendix to its 1990 recommendations, the ICRP acknowledged the distinction between a multi-attribute description of risk and the mathematical representation, but, for radiological protection purposes, the radia­ tion health detriment was assumed to be the primary factor in the risk evaluation [2]. There is an increasing awareness that the concept of ‘risk’ cannot be represented solely by a probability of physical harm. Many factors can influence a person’s attitude towards radiation and other hazards, including for example percep­ tion of the benefits and the degree of individual consent and control associated with a risky action [3,4]. Therefore, it seems important to distinguish between aversions to radiation that are based on irrational grounds (e.g. due to media sensationalism or a misunderstanding of the probabilities of harm) and aversions based on what could be deemed rational grounds. Information on factors influencing risk perception will improve our understanding of people’s attitude to radiation and could also play a significant role in matters of policy and decision making, especially in the manage­ ment of radiation risks. One might argue that certain psychological, social and ethical values can all give rise to rational aversions to particular types of risk (i.e. reactions that are consistent and attributable to specific factors). Hence this paper will present and discuss some ethical issues that might be taken into consideration when assessing radiation risks. 2 8 4 POSTER PRESENTATIONS

2. A RATIONAL CHOICE

According to the Bayesian theory of rational behaviour under uncertainty, a rational choice is that which maximizes the net expected benefit, or minimizes the net loss, to the individual. In technical terms, the rational choice or action is that which maximizes the person’s expected utility, where utility is equated with happi­ ness, well-being or preference satisfaction [5]. Expected utility is calculated by summing the products of probabilities and utilities (consequences) over all possible outcomes of the possible actions. A rational person should prefer a risky action only if the expected utility of that action is greater than the utility of competing alternative actions. Hence, any decision to accept or reject a risk will be dependent on:

(a) the initial situation of the individual; (b) the probability and degree of harm; (c) the probability and degree of benefit; and (d) the available alternatives.

Hence, it does not necessarily follow that people are behaving irrationally, or not conforming to Bayesian rationality, if they accept a high probability of harm in certain situations but reject a lower probability of harm under other circumstances. If the perceived benefit from driving a motor car is much greater than the perceived benefit from using nuclear power, the individual is not being irrational in showing greater aversion to the risks from radiation than to the risks from a motoring a cc id e n t.

3. ETHICAL VALUES

The Bayesian theory of rational choice has close links with the ethical doctrine of utilitarianism, whereby the best action, decision or policy is that which maximizes net utility for society: ‘the greatest good for the greatest number’. In more general terms, and with obvious similarities to the philosophy of radiological protection, the action or policy can be justified if a risk cost-benefit analysis shows a net (monetary) benefit for society. Utilitarianism is one variant of consequentialist ethics, wherein an action is judged in terms of its consequences alone. Other ethical doctrines, however, base judgements not on the consequences of the action, but on the action itself. Opponents of utilitarianism claim that it is flawed because it would permit morally reprehensible actions such as murder or suppression of information if the happiness of the masses outweighed the harm to the few. Many philosophers have stressed that a decision theoretic evaluation of rational action, both for an individual and for society, need not be based only on Bayesian calculations of expected utility: the evaluation can take account of other factors, including ethics [6- 8]. Similarly, the radiation health detriment is not the only factor POSTER PRESENTATIONS 285 influencing radiation risk assessment; a number of psychological and ethical values can influence both an individual’s perception of risk and political decisions regarding the management of risks in a society. In many cases, a decision based on expected utility reaches the same conclusions as other doctrines, e.g. implementation of countermeasures after an accident. Ethical dilemmas arise when the action judged best under utilitarian rules comes in conflict with what is judged best under the rules of other ethical doctrines. The nuclear power issue is a classic example of a conflict over the moral worth of an action (e.g. production of electricity) versus the moral reprehensibility of a possible consequence of that action (e.g. induction of cancers by radiation).

3.1. Liberty: freedom of choice

Moral philosophers, including supporters of the Bayesian theory, acknowledge that the freedom of the individual to accept or reject a risk has ethical and psychologi­ cal significance [4-6]. One can argue that an individual who rejects a forced or involuntary risk of death is not being irrational if he gives free and informed consent to an identical risk of death under different circumstances. People tend to show a consistent and reproducible enhanced aversion to forced, as compared to voluntary, risk. For an individual to give free consent, the existence of applicable alternatives is a necessary condition. It has been suggested that freedom of choice has a utility value in itself which can be included in the calculation [5]. Hence, with regard to the exposure to radiation risks, voluntary workers in the nuclear industry have a different ethical status compared to the general public. Although this is reflected in dose limits, the ethical differences are rarely highlighted.

3.2. Equality: distribution of risks and benefits

If radiation risks devolve on persons who are either unwilling or unable to reap the benefits, inequities in the distribution of risk and benefit can arise among differ­ ent groups of people, namely across age groups, income brackets, regions, nations, and generations. Radiation risks from nuclear power might be said to give rise to such inequities — present day benefits at the expense of risks to future generations being probably the most contentious [6]. Inequity could thereby have some effect on risk perception. Compensation for exposure to the risk is a possible solution, but difficult in practice. Decision theories based on maximizing utility alone are, in principle, indifferent to the distribution of utility among individuals. Other ethical doctrines place value on the distribution of risk and benefits. 2 8 6 POSTER PRESENTATIONS

3.3. The naturalistic fallacy

Comparison of man-made radiation risks with the risks from naturally occur­ ring background radiation is useful for putting the risks into perspective. However, a number of risk assessors make logical and ethical errors when they attempt to justify the acceptability of anthropogenic risks in terms of the risks from natural background radiation. In ethics, this error is termed ‘the naturalistic fallacy’, the mistake of confusing judgements of goodness or moral worth with descriptions of factual properties, or the logical error of deriving an ‘ought’ from an ‘is’ [9, 10]. One cannot derive normative ‘ought’ statements on man-made radiation risks (e.g. An annual dose of 10 цSv represents an acceptable risk...) from empirical ‘is’ obser­ vations of the background radiation risk (... because the background radiation dose is much higher). To give an analogy, one cannot say that it is acceptable for me to push one old lady down the stairs because 1000 old ladies fall down the stairs each year as a result of the natural frailties of old age.

3.4. Responsibility and blame

The importance of intention in judging actions has a particular significance for radiation risks. The culpability of an agent, or an industry, is not only assessed by the consequences of its actions (i.e. number of radiation induced cancers): many other factors can be taken into account, including how the harm was brought about by the agent or whether alternative courses of action existed. Ethics and law clearly distinguish between an act of omission (not preventing a harm) and an act of commis­ sion (being directly responsible for bringing about a harm) [11]. According to this principle, cancers caused by the nuclear power industry might be deemed more morally reprehensible than cancers from not preventing exposure to naturally occur­ ring radiation. Also, intended harms are thought by some ethicists to be sometimes harder to justify than foreseen harms [11]. An example of the former would be cancers caused by radiation experiments carried out on people in the 1940s and 1950s, whereas concerns arising from the use of nuclear power belong to the latter category of foreseen but not intended harms.

4. CONCLUSIONS

Failure to include ethical, social and psychological factors in evaluation of radiation risks is certain to provoke public-expert conflicts. If radiological protection authorities pay due respect to the following points, they not only will be showing greater respect for ethics but also should help to promote public negotiation towards a more rational approach to risk assessment and management: POSTER PRESENTATIONS 2 8 7

— Risk perception is not only governed by probabilities of harm; rational aver­ sion to risk can include ethical, social and psychological factors. — Decision theory is not limited to consequences alone; there are both ethical and practical grounds for paying attention to issues such as consent, equality and responsibility within radiation risk management. — Increased voluntariness and a greater degree of free informed consent from the public can be achieved by involving individuals and the community in the deci­ sion making process. — Care should be taken when comparing natural and anthropogenic radiation e x p o su re s .

Radiation doses cannot be justified by comparison to natural background ra d ia tio n .

ACKNOWLEDGEMENTS

The author would like to express her gratitude to the Norwegian Ethics Programme for a research grant and to thank D. Fellesdal, professor of philosophy at Stanford University and the University of Oslo, and colleagues at the Isotope Laboratory for helpful comments on a draft of this paper.

REFERENCES

[1] STARR, C., W HIPPLE, C., Risks of risk decisions, Science 208 (1980) 1116. [2] ICRP, 1990 Recommendations of the International Commission on Radiological Pro­ tection, ICRP Publication 60, Annals of the ICRP, Vol. 21, Pergamon Press, Oxford (1 9 9 1 ). [3] FIRSCHOFF, B., SLOVIC, P., LICHTENSTEIN, S., “Facts and fears” , Societal Risk Assessment (SCHW ING, R., ALBERS, W ., Eds), Plenum Press, New York and London (1980). [4] SLOVIC, P., Perception of risk, Science 236 (1987) 280. [5] HARSANYI, J., “ Morality and the theory of rational behaviour” , Utilitarianism and Beyond (SEN, A., W ILLIAM S, B., Eds), Cambridge University Press, Cambridge (1982) 39-62. [6] SCHRADER-FRECHETTE, K.S., Risk and Rationality, University of California Press (1991). [7] NO ZICK, R., The Nature of Rationality, Princeton University Press, Princeton, NJ (1 9 9 2 ). [8] F0LLESDAL, D ., Rationality and Irrationality, Epistemología 9 (1983) 5. [9] MOORE, G.E., Principia Ethica, Cambridge (1903). [10] H U M E, D ., A Treatise of Human Nature, 2nd ed., Oxford University Press, Oxford (1978), original text 1739-1740. [11] FOO T, P., Virtues and Vices and Other Essays in Moral Philosophy, Basil Blackwell, Oxford (1978).

Technical Session 5 MANAGING RADIATION RISK

IAEA-CN-S4/39P

REGULATION OF PRACTICE INVOLVING RADIATION SOURCES IN THE SLOVAK REPUBLIC

P . V R A B C E K Nuclear Regulatory Authority of the Slovak Republic, Bratislava, Slovak Republic

1. ANALYSES OF CURRENT LEGISLATION

The legislation of the Slovak Republic in the field of nuclear safety is that enacted in the former Czechoslovakia. The Slovak Constitution of 1992 kept in force the following acts, regulations and decisions related to practices involving radiation sources as well as nuclear power plants (NPPs):

— Act 455/1991 regarding commercial trade; — Act 238/1991 on waste; — Regulation 65/1972 of the Slovak Ministry of Health providing for protection of health against ionizing radiation; — Regulation 560/1991 of the Federal Ministry of Foreign Trade on licensing of the import of goods and services; — Decision of 25 June 1981 by the Czechoslovakian Atomic Energy Commission (CAEC) on testing of equipment for transport and loading of radioactive m a te ria l.

There are no deficiencies in these regulations from the formal point of view. The problem is the discrepancy between the real state of affairs in Slovakia and this legislation. For example, some institutions which provide services for the territory of Slovakia formerly in the Czechoslovakian Federation remain in the Czech Republic. So, the Institute for Research, Development and Use of Radioisotopes in Prague has provided tests of radiation sources for Slovakia and has collected Slovak­ ian NPP radwaste for the central (Federal) disposal plant, which now lies in the Czech Republic, also. Act 2/1993, which was passed by the Slovak Parliament, outlines the responsi­ bilities of the Nuclear Regulatory Authority of the Slovak Republic (NRA SR). One of the competencies of the NRA SR is the licensing of production and import of radi­ ation sources. The substantial requirements for obtaining a license are: qualification of responsible workers, authorization of the work place, certification of measurement instruments and transport packing sets, and documented proof of radwaste handling procedures.

291 2 92 POSTER PRESENTATIONS

The first two items are assigned to the competence of the public health authori­ ties. The public health service has been developed to a high level in Slovakia. The regional public health offices are, paradoxically, under the jurisdiction of the Ministry of the Interior. The Ministry of the Economy as the successor of the Federal Ministry of For­ eign Trade can grant a permit for the import of radiation sources only on the basis of a license having been granted by the NRA SR. For production or repair of radia­ tion sources, there is a like procedure. The regional trade offices, which are under the jurisdiction of the Ministry of the Interior, apply to the NRA SR for a license for these concessions. They can issue a concession only after a positive decision by the nuclear regulatory body. Complications for the licensee occur in procuring support documents for the decision of the NRA SR, when the licensee must find services abroad because of lack of institutions in Slovakia, as mentioned above. So, for example, there is not a solu­ tion to the problem of complex metrological assurance of the physical quantities of ionizing radiation. The Slovak M etrological Institute in Bratislava has done an analy­ sis for the NRA SR [1], which shows that the current structure of ionizing radiation metrology cannot meet the demands of licensees. There is no State testing laboratory for transport packing sets and for closed radiation sources in the Slovak Republic. A State radwaste management project has been developed by the Ministry of the Economy, and the Government has defined the important measures for realizing it [2]. Unfortunately, realization of the project is not possible until 1995 because of lack of a central disposal plant. Temporarily, the producers of radwaste will use the Czech installation for permanent disposal. However, this procedure is complicated by the fact that such export/import must be approved by the environmental ministries of both countries.

2. PROPOSALS FOR PROBLEM SOLUTION

The problems described in our analysis could be solved in two ways:

— short term temporary measures within the framework of the current legislation; — creation of new legislation with application of the experience of the highly developed countries.

Short term measures are intended to prevent escape of radwaste into the environment and to protect people against the risks associated with exposure to ioniz­ ing radiation. Therefore, the NRA SR requires from a license applicant strong evi­ dence of proper management of radwaste resulting from the radiation sources used. A co-operation agreement between the Ministry of Economy and the NRA SR could help to accelerate realization of the radwaste management project. All importers or producers of radiation sources must register any delivery with the NRA SR. This system enables very simple tracking of the condition and the POSTER PRESENTATIONS 2 93 movement of the radiation sources by means of inspections performed by the NRA SR. The Ministry of Health licenses the import of equipment containing radiation sources. Therefore, it is necessary to reach agreement with this ministry concerning the central register for the entire Republic. The policy of the NRA SR on future legislation is based on the analysis of the current legislative situation, which has revealed a number of areas of weakness and some areas not covered at all. It has been decided by the Government to elaborate a new nuclear act [2], which regulates radiation protection and the safety of radiation sources. The special regulations in this area should be prepared in co-operation between the Ministry of Health and the NRA SR. International experience based on current scientific knowledge would be used to the maximum extent in defining these regulations.

REFERENCES

[1] GÁBRIS, F ., ZEM A N , J., BELAÑ, J., Analysis of the Current State of the Metrologi­ cal Assurance of the Ionizing Radiation in the Slovak Republic, Rep. SM Ú , Bratislava (1 9 9 3 ). [2] Resolution of the Government of Slovak Republic of 8 March 1994, No. 190. 2 9 4 POSTER PRESENTATIONS

IAEA-CN-54/48P

CHANGING THE OCCUPATIONAL DOSE LIM IT FOR PREGNANT W ORKERS: PUBLIC CONSULTATION AND CONSENSUS SEEKING

C . P O M R O Y Atomic Energy Control Board, Ottawa, Ontario, Canada

1. INTRODUCTION

In July 1991 the Atomic Energy Control Board (AECB), the nuclear regulatory authority in Canada, published a consultative document (C-122) containing proposals to change the current Canadian radiation dose limits to those recommended in ICRP-60 [1]. W e received a number of written responses, many of which expressed concern about the proposed reduction of the limit for pregnant workers from the current Canadian limit of 10 mSv to essentially 1 mSv during the pregnancy. The reason for the concern was that it would be difficult to achieve such a reduction in dose and even more difficult to prove compliance even if the reduction were achieved, particularly in the case of internal doses. M ost of the responses came from female staff in nuclear medicine departments. There was concern that employers would deal with this problem by removing workers from radiation work if they became pregnant. In large departments this could be done by reassignment, but in smaller institutions it could mean layoff, in order to have the resources to hire a replacement. There was a strong feeling that this process could eventually lead to a reluctance to hire women for nuclear medicine and other radiation work.

2. MEETINGS

In view of this concern, we met with representatives of organizations such as the Canadian Association of Medical Radiation Technologists, the Canadian Organi­ zation of M edical Physicists, the Canadian Association of Radiopharmacists and the Canadian Nurses Association. W hile the concerns were reiterated at this meeting, no suggestions for a solution were forthcoming. We therefore decided to hold a series of public meetings with the female radiation workers themselves, to give them the opportunity to suggest ways of resolving the problem. These public meetings were held in seven major cities where nuclear medicine technologists, medical physicists and other medical radiation workers are concen­ trated. At some meetings female workers from power utilities and research establish- POSTER PRESENTATIONS 295 ments were present. Representatives of the Canadian Labour Congress, the Canadian Union of Public Employees and the Canadian and Manitoban Human Rights Com­ missions also attended one meeting. We held an additional meeting at Key Lake uranium mine in northern Saskatchewan. This was at the request of the uranium mining industry, which employs a significant number of women. The total attendance at all the meetings was about 350 people. The meetings took the form of an introduction explaining the proposed new limit and the resulting concerns, followed by a technical presentation describing radiation risks to the foetus. This presentation was prepared by a consultant to the AECB, D.K. Myers, and was based on his report to the AECB on the same s u b je c t [2].

3. TECHNICAL PRESENTATION

Considerable effort was made to present the technical information on risk in a form that could be readily understood by the participants. Bar charts were used wherever possible to compare radiation risk from the current 10 mSv limit and a 2 mSv limit based on the ICRP-60 recommendations. The risk figures included the latest data for childhood cancer, cancer in later adult life, birth defects and genetic defects. In all cases these risks were presented in terms of cases per 10 000 live births, and the uncertainty in the risk estimates was emphasized. The natural rates of the various effects were also presented in the same way. The presentation was tested on a group of AECB staff, both technical and non­ technical, and modified according to their comments to make it as comprehensible as possible. At the meetings a generous time allowance was made for questions on the risk figures and how they were obtained. Discussion with some of the participants during the lunch interval indicated that the material had been generally well under­ stood. This was helped by the fact that the audience was made up largely of technolo­ gists, who have a relatively high level of education. We then presented a summary of the written comments on our original proposal. Representatives of professional associations, employers and trade unions were given the opportunity to make presentations, giving further details of their c o n c e rn s.

4. DISCUSSION GROUPS

Following the presentations, lunch was provided. This gave an opportunity for the participants and AECB staff to talk informally. After lunch we asked the par­ ticipants to work in groups of about ten people and to prepare their suggestions for 2 9 6 POSTER PRESENTATIONS the formulation of a regulation. A very informal attempt was made to divide up the audiences into groups which each included different viewpoints, e.g. technologists, unions and employers, so that differences of opinion could be debated within each group. The number of groups at each meeting was between two and six, for a total of 29 groups. Each group was given about an hour and a half to debate the issues and to formulate its position. Blank sheets for the overhead projector were provided and the groups were asked to write down their conclusions in point form. Each group was also asked to appoint a spokesperson to report its conclusions to the meeting as a whole. During this stage the other groups commented on each group’s conclusions.

5. GROUP REPORTS

The summary reports of each group showed considerable similarities, not only within each meeting but between the meetings, so that a definite consensus emerged. This provided a firm basis for us to revise our original proposal. The points of agree­ ment between the groups are listed below, with dissenting opinions noted.

(a) The philosophy of setting the dose limit for the foetus the same as for the public is acceptable, but the ICRP recommendation on the public dose limit is too low. Not all participants were opposed to the ICRP limit, and in particular the trade union representatives strongly supported it. (b) The risks, as described in the technical presentation, do not justify such a dras­ tic change in the limit (some groups felt that no change was justified). (c) The strongest suggestions for the actual value of the proposed limit were for a compromise between 2 mSv and 10 mSv to the foetus, but several groups saw no reason to change from the current limit of 10 mSv.

6. NEW PROPOSAL

We have now prepared a new proposal for a dose limit of 4 mSv to the foetus. Although the limit must be set by reference to the risks to the foetus, doses to the foetus cannot be measured with any accuracy, particularly from internal contamina­ tion of the mother. Therefore, foetal doses must be controlled by control of doses to the mother. This was the apparent intention of the ICRP in specifying a dose limit to the surface of the m other’s abdomen and a special limit on intake. Our proposal is more specific and will limit the dose to the mother by means of a ‘combining for­ mula’ that takes both external and internal dose into account:

— (m S v ) 4------ï— (Bq) < 1 4 0 .2 A L I where D e is the external dose and I is the intake o f radioactive material. POSTER PRESENTATIONS 297

7. FOLLOW-UP

All participants were sent a report of the meeting and the summary of the com­ ments of the groups. The new proposal was also sent to all participants, to let them know that their participation had been instrumental in the development of AECB policy. As part of the Canadian regulatory process, there will be another opportunity for comment when the complete package of new radiation protection regulations, which include most of the recommendations of ICRP-60 except the limit for pregnant workers, is published. This was the first time that the AECB had held a series of public meetings to address a specific regulatory issue. The exercise was judged to be successful since we received very useful suggestions which also represented the consensus of a large number of people directly concerned with the issue.

8. PROPOSALS OF OTHER COUNTRIES

For comparison with the consensus of the meetings, the proposals of other countries, namely the UK and the USA, in response to the ICRP-60 recommenda­ tions are discussed below. In the UK, the National Radiological Protection Board (NRPB) has stated that the exposure of pregnant workers “ should be as low as reasonably achievable [ALARA] and such as to make it unlikely that the dose to the foetus will exceed 1 mSv during the remainder of the pregnancy” . The UK Nuclear Installations Inspectorate (the UK regulatory authority), in applying the NRPB recommendations, does not intend to set a special limit for pregnant workers, but instead will emphasize the application of ALARA to the working environment, so that it is very unlikely that a dose of 1 mSv to the foetus will be exceeded. W hile in most instances such an approach will give the foetus adequate protec­ tion, the absence of a numerical limit would mean that in cases where a dose of more than 1 mSv has occurred and it is thought necessary to prosecute the licensee, the prosecution would depend on the interpretation of ALARA rather than non- compliance with a numerical limit. In the USA, the National Council on Radiation and Measurements (NCRP) has recently recommended, in its Publication No. 116, a foetal limit of 0.5 mSv per month. This is equivalent to about 4 mSv during the remainder of a pregnancy following its timely declaration. The US National Regulatory Commission (NRC) regulation, made before the NCRP recommendation appeared, is a limit to the foetus of 5 mSv during the entire pregnancy, received at a fairly uniform rate. The NRC intends to adopt the NCRP recommendation in the next round of changes to its regulations. 2 98 POSTER PRESENTATIONS

9. THEORETICAL POSTSCRIPT

After the meetings were over, we came across an article written by a former Canadian Deputy Minister (head of a Government department) entitled “ Getting Consent of the Governed” [3]. This discussed the problems of obtaining consensus on issues put before the public for consultation. The Canadian regulatory process requires that the public be given ample opportunity to comment on any proposed regulations. The AECB provides such consultation by publishing consultative docu­ ments, on which individuals or institutions provide their comments. This is what initiated the public meetings, as described above. The former Deputy M inister’s point is that when asked to comment, factions with vested interests not only present opposing viewpoints but also tend to exaggerate them to make maximum impact. For the Government itself, or its officials, to com­ promise between extreme viewpoints is to please few and antagonize many. The solution to this dilemma is to shift the point of compromise, by making the various

FIG. 1. Schemes for developing regulatory policy by presentation of conflicting viewpoints and by compromise. POSTER PRESENTATIONS 2 99 factions develop their own compromise, which the Government undertakes to imple­ ment (Fig. 1). All interest groups are informed that failure to develop such a com­ promise would leave the Government free to act unilaterally. Thus those who want to influence the outcome have an interest in working towards a compromise. This is in fact what happened when we divided the participants at our meetings into groups, each including as many different viewpoints as possible. The groups certainly reached a consensus which made the revision of the policy a straightfor­ ward matter.

REFERENCES

[1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990 Re­ commendations of the ICRP, Publication 60, Pergamon Press, Oxford and New York (1 9 9 1 ). [2] MYERS, D .K ., “ Comments on the ICRP-60 Rationale for Dose Limits for the Preg­ nant W orker” , INFO-0421, Atomic Energy Control Board (1992). [3] KROEGER, A ., Getting Consent of the Governed, Institute for Research on Public Policy, Policy Options 13 (10 December 1992). 3 00 POSTER PRESENTATIONS

IAEA-CN-54/50P

ACTIONS ENTREPRISES PAR ELECTRICITE DE FRANCE EN BELARUS: UNE SOLIDARITE ENTRE ELECTRICIENS

J. LALLEMAND, M. BERTIN, D. HUBERT Comité de radioprotection, Electricité de France, Paris, France

L’accident de Tchernobyl a entraîné une très importante contamination en Ukraine, autour de la centrale accidentée, mais également en Bélarus, république voisine de 12 millions d’habitants. Cette contamination a été secondaire aux rejets d’iode et de césium radioactifs. Son importance n’est pas seulement corrélée à la distance par rapport à Tchernobyl. Un bilan long et méticuleux publié par l’Agence internationale de l’énergie atomique (AIEA) a permis d’en préciser, fin 1990, la géographie en «taches de léopard». La méconnaissance des niveaux de pollution radioactive dans les jours et semaines qui ont suivi l’accident n’a pas conduit à l’application de mesures prophylactiques et a entraîné une irradiation, à des niveaux variables, de la population biélorusse. Les enfants, dont la thyroïde est particulièrement radiosensible, ont reçu les doses les plus élevées. De nombreuses initiatives nationales et internationales d’aide et d’assistance ont été prises dès 1986 dans un double but humanitaire et scientifique. Elles concernent essentiellement l’Ukraine et manquent cruellement de coordination. C ’est dans ce contexte et dans un souci de solidarité entre électriciens que les directions de l’Electricité de France (EDF) ont décidé d’établir, en 1992, un plan d’aide et de collaboration avec les autorités biélorusses confrontées à la difficile gestion des conséquences sanitaires de l’accident de Tchernobyl sur leur territoire. C ’est ainsi qu’un contrat a été signé en février 1993 par EDF d’une part et par le Gouvernement biélorusse d’autre part (Ministères de la santé et de l’énergie, Comité gouvernemental d’étude des conséquences de l’accident de Tchernobyl et Académie des sciences). Ce contrat est actuellement établi pour une durée de trois ans et porte sur une enveloppe globale de 10 millions de francs français. Il prévoit: a) pour la participation française: POSTER PRESENTATIONS 301

EDF a choisi d’inscrire cette collaboration dans le cadre d’une action concrète. Elle comporte deux volets:

i) La fourniture de matériel dosimétrique et technique comprenant1 :

— un spectromètre gamma qui a été installé à l’Institut de l’énergie, — une soixantaine de dosimètres d’ambiance et 600 dosimètres individuels qui ont été fournis au Ministère de l’énergie, — 80 balises de surveillance de l’environnement qui ont été mises à la disposition des autorités biélorusses. ii) La fourniture de matériel médical1 autour de laquelle s’articule l’essentiel de ce programme. Dans ce domaine, il s’agit surtout de répondre de façon rapide et utile aux attentes prioritaires des médecins confrontés à l’augmentation importante des cancers de la thyroïde chez l’enfant.

C ’est ainsi qu’une gamma-caméra Sopha-Médical est arrivée à M insk (Institut des radiations) en décembre 1993. Cet équipement est indispensable au diagnostic, au suivi et au traitement du cancer de la thyroïde. Jusqu’à présent, il n’y avait pas d ’installation de ce type en Bélarus. Pour en assurer une utilisation optimale, deux médecins ont été formés au Centre hospitalier universitaire (CHU) de Créteil. Rappelons, en effet, qu’en Bélarus le problème médical essentiel, suite à l’accident de Tchernobyl, concerne la pathologie thyroïdienne et, en particulier, l’augmentation incontestable de l’incidence des cancers de la thyroïde chez l’enfant (environ 250 cancers en excès depuis 1990). Ceci peut s’expliquer par l’importance des rejets d’iodes radioactifs lors de l’accident, la fixation élective de ces radioélé­ ments sur la thyroïde et la radiosensibilité particulière de cet organe chez l’enfant. b) pour la participation biélorusse:

Elle s’inscrit, quant à elle, essentiellement autour d’une information en retour portant sur les divers aspects de cette catastrophe, en particulier et surtout sur le suivi sanitaire et médical des populations irradiées du fait de cet accident. Il apparaît tout à fait fondamental de tirer le maximum d’enseignements de cette catastrophe, en particulier sur le plan médical. C ’est là l’une des justifications de cette action.

1 II s’agit là des actions entreprises dans le cadre de la première année de collaboration. 3 02 POSTER PRESENTATIONS

BIBLIOGRAPHIE

BERTIN, M ., LALLEM AND, J., Augmentation des cancers de la thyroïde de l’enfant en Bélarus, Ann. Endocrinol. 53 (1992) 173-177.

BERTIN, M ., LALLEM AND, J., HUBERT, D., Tchernobyl, près de 8 ans après, Lyon Pharmaceutique, 45 1 (1994) 9-15.

COM ITE DE RADIOPROTECTION, Irradiation par l’iode radioactif, EDF, Paris (1992) 6 8 pp.

DEMIDCHIK, E.P., OKEANOV, A.E., REBEKO, V.Y., et al., Current trends in prophylaxis, early diagnosis and curative treatment of the thyroid cancer (Communication personnelle).

KAZAKOV, V.S., DEMIDCHIK, E.P., ASTAKHOVA, L.N., Thyroid cancer after Chernobyl. Nature 359 (1992) 21.

OLEYNIK, V.A ., «Thyroid cancer in children of Ukraine before and after Chernobyl accident» (Proc. Nagasaki Symp. on Chernobyl, Nagasaki, 1993).

PARMENTIER, C., ROBEAU, D., SCHLUMBERGER, M ., AUBERT, B., NENOT, J.C., PARM ENTIER, N ., L ’après-Tchernobyl: des cancers en excès, La Recherche 24 (1993) 7 3 8 -7 4 0 .

W ILLIAMS, E.D., PINCHEERA, A., KARAOGLOU, A., CHADWICK, K.H., Thyroid cancer in children living near Chernobyl, European Communities, EUR 15248, Luxembourg (1 9 9 3 ). POSTER PRESENTATIONS 303

IAEA-CN-54/59P

LES OBJECTIFS DE LA RADIOPROTECTION A ELECTRICITE DE FRANCE: LE LIVRE BLANC DE LA RADIOPROTECTION

P h . R O L L IN Comité de radioprotection, Electricité de France, Paris, France

Dans le climat passionnel qui a suivi l’accident de Tchernobyl, tout ce qui touche au nucléaire a plus ou moins été remis en question par un public de plus en plus préoccupé par son environnement. De nouvelles exigences ont ainsi vu le jour. La Commission internationale de protection radiologique (CIPR) a émis en décembre 1990 de nouvelles recommandations plus restrictives. Par ailleurs, une certaine dégradation dans les résultats dosimétriques est apparue dès 1989 dans les centrales nucléaires françaises. Tout cela a conduit l’Electricité de France (EDF) à une analyse critique de la situation et à définir des orientations dans tous les domaines de la radio­ protection; c’est l’objet du Livre Blanc, paru à l’été 1993, qui en aborde les diffé­ rents aspects:

— fondements médicaux et biologiques, — protection des travailleurs, — protection du public et de l’environnement, — information et formation, — situations d’urgence.

On peut en retenir les grandes lignes ci-après: a) L’application des techniques d’optimisation («As low as reasonably achiev­ able» ou ALARA) dans l’organisation du travail est un point clé de la réduction des doses; des structures nationales et régionales ont été mises en place à cet effet; des objectifs ambitieux (réduction des doses collectives annuelles de 35 % entre 1992 et 1995) ont été affichés. La réduction des doses est effectivement a m o rc é e . b) Des efforts ont également été engagés, en vue de réduire les doses, par d’autres m é th o d e s:

— Conception des installations: étude de nouveaux matériaux, meilleur agence­ ment des matériels et des locaux pour réduire les temps d’exposition en période d’entretien. 304 POSTER PRESENTATIONS

— Etude de moyens robotiques. — Etude de nouvelles techniques de décontamination, etc. c) De nombreuses dispositions ont été prises pour amener la protection des travailleurs d’entreprises extérieures au même niveau que celle des travailleurs d’EDF. Ceci a notamment conduit:

— A développer un système EDF de suivi dosimétrique national, appelé Dosinat, utilisant les données de la dosimétrie individuelle journalière obtenue au moyen de dosimètres électroniques; ce système informatique permet de suivre les travailleurs migrants passant de site en site, en conservant la trace de leur dosimétrie, auparavant difficile, voire impossible à connaître (ceci en partie à cause du secret médical attaché à la dosimétrie légale par films). — A développer les relations contractuelles avec les entreprises. — A formuler des exigences en matière de formation tant à la radioprotection qu’au travail en milieu nucléaire, formalisées par la création d’un «carnet d’accès», en relation avec le Groupement intersyndical de l’industrie nucléaire (GIIN).

En ce qui concerne le domaine de l’environnement:

— Les rejets radioactifs des centrales, déjà très faibles, seront ramenés au niveau des unités de production les plus performantes dans ce domaine. A noter qu’aller plus loin ne présente aucun intérêt au plan sanitaire, et n’aurait qu’un caractère purement démagogique. — Les quantités de déchets radioactifs sont progressivement réduites par la mise au point de techniques de compactage, d’incinération, de recyclage. — Des études sont engagées pour l’élimination, ou au moins la réduction de volume, des déchets à vie longue.

L’amélioration d’ensemble de la radioprotection passe par une sensibilisation générale du personnel («culture de radioprotection») et aussi par une redéfinition des «métiers de la radioprotection» et leur revalorisation. L ’information du public est très importante. EDF s’engage à fournir à tous une information claire, fondée sur la transparence et crédible parce que loyale. Mais bien informer est toujours difficile en raison de la diversité des publics; c’est ainsi qu’EDF a entrepris et poursuivra une politique d’information adaptée aux différents publics qui peuvent être des vecteurs privilégiés de transmission de l’information, n o tam m en t:

— les professions de santé, — les médias, — les enseignants et les élèves, — les élus. POSTER PRESENTATIONS 305

Le fonds documentaire à l’usage des différentes catégories de public va être prochainement entièrement renouvelé. Enfin, une des conclusions du Livre Blanc est que la collaboration avec les Pouvoirs publics doit être développée. Bien sûr, chacun doit rester dans son rôle, mais les moyens à mettre en œuvre sont très importants et chacun doit pouvoir y contribuer. Ceci porte aussi bien sur la réalisation d’une grande base de données (confidentielles) nationale que sur l’amélioration des structures de suivi médical des entreprises, actuellement trop dispersées. 3 0 6 POSTER PRESENTATIONS

IAEA-CN-54/65P

REDUCTION OF W ORKERS’ EXPOSURE AND DOSE ALLOCATION

J. ONODERA, T. NISHIZONO Japan Atomic Energy Research Institute (JAERI), Tokai-mura, Naka-gun, Ibaraki, Japan

1. INTRODUCTION

There are two issues in planning radiological protection measures. One is the issue of efficiency, i.e. how to reduce the sum of the cost of detrimental exposure and the cost of protection measures. The other is that of equity, i.e. how to allocate dose among workers in selected protection measures. This is the issue related to individual dose. Efficiency is usually considered in the optimization of protection measures, while dose allocation is often neglected in the planning stage. Practically, however, individual as well as collective dose should sometimes be taken into account in evaluating and selecting the protection measures, because of the existence of various constraints such as daily individual dose limit. In such a case, cost-benefit analysis, which is the typical method for optimization of radiological protection, often cannot be applied in selecting the protection measures. This study examines two issues: (a) the benefit of dose reduction measures excluding collective dose reduction and (b) the distribution of individual dose among workers. Examples are taken from actual cases of dose reduction measures during the current reactor dismantling work at the Japan power demonstration reactor (JPDR) at the Japan Atomic Energy Research Institute (JAERI) since 1986 [1].

2. BENEFITS OF DOSE REDUCTION MEASURES

This section explores benefits of protection measures other than the reduction of collective dose. The case is the installation of shielding around the lower part of the reactor pressure vessel (RPV). Installation of the shielding reduced the average dose rate in the working area to about one third. The attained collective dose reduc­ tion was 4 x 10 '2 man-Sv; the cost-effectiveness ratio of the installation was cal­ culated at 8 x 1 0 5 $/man-Sv. When compared with the value used in ICRP-55 [2], 2 X 1 0 4 $/man-Sv, this ratio indicates an overprotected situation according to cost-benefit analysis. POSTER PRESENTATIONS 307

However, in application of cost-benefit analysis to welfare economics, there is an idea of risk premium dealing with uncertainties in cost and benefit [3]. This idea may possibly be applied to the cost-benefit analysis of radiological protection, and more protection costs would be allocated to a unit collective dose. This paper will not evaluate the risk premium itself, yet we should consider this if more protec­ tion costs are to be allocated based on the benefits of protection measures.

2.1. Enhancem ent of w orkers’ consciousness by installation of the shielding

Protection measures based on hardware, e.g. shielding installation, are impor­ tant for dose reduction. Additionally, the consciousness of the workers regarding dose reduction is an important factor. The introduction of dose reduction measures based on hardware makes the workers feel safe and assured, which in turn makes their work efficient, leading to further dose reduction. For instance, imagine a situa­ tion when management fails to implement effective exposure reduction measures and expects workers to protect themselves, e.g. by shortening work time on their own initiative in an area where the highest level of exposure was anticipated in the RPV dismantling. W orkers would become sceptical about management’s attitude, and this would adversely affect the workers’ psychological condition. Therefore, in this case, the protection measure using hardware, which itself reduced the average dose rate to one third, resulted in the enhancement of workers’ consciousness of dose reduc­ tion. For Japanese nuclear power stations, it has also been pointed out that the enhancement of workers’ consciousness of dose reduction is one of the important factors of radiation exposure reduction [4, 5].

2.2. Dose reduction of skilled workers

A worker skilled in non-destructive inspection worked on the inspection of welded pipes. This worker’s dose was 3.9 mSv for the 3 month working period, equivalent to about 15 mSv per year, which is considerably high. The saved dose of this worker through the above mentioned shielding was calculated as 3 mSv. The number of workers with specific skills, such as welding and various inspections, is limited; therefore, changing workers for the purpose of individual dose restriction is not always easy. Accordingly, for workers working in several nuclear facilities in a year, adherence to the originally planned dose level at each facility is important. In this case, the installation of the shielding was effective in reducing the individual dose, which made the whole non-destructive.inspection possible considering the limited number of workers available. The fact that skilled workers are necessarily highly exposed would be a con­ straint when introducing the dose limit reduction recommended by ICRP-60 [6]. From the viewpoint of dose reduction of skilled workers, it is also important to train skilled workers systematically. 308 POSTER PRESENTATIONS

2.3. Effect of dose rate reduction

The dose rate in the working area was reduced to approximately one third by installing the shielding. This means that the possible working time within the speci­ fied dose per day (usually 0.2-0.5 mSv) increased. As a result, workers’ mental stress from working hour control was reduced, which probably led to a decrease in the incidence of human error.

3. ALLOCATION OF DOSE AMONG WORKERS

Various categories of workers were engaged in the RPV dismantling. Each w orker’s task was different from others’ according to his own status or skill, result­ ing in different dose levels. Yet this allocation of dose can be controlled, to some extent, by the supervisor. In general, without any control, the dose distribution of workers is expected to show a log normal distribution. In the case of JPDR dis­ mantling, however, the distribution among workers with a high level of dose did not fit to log normal. Figure 1 shows log normal probability plots of cumulative dose distribution of workers from November 1989 to M arch 1990. Differences among the workers with high dose tended to become smaller with time. This was the result of

и 2 Ц

сз > 1 43 t ,1 § о a д 0 > ъ а 1 n § I 3 и

0 . 1 0 . 2 0 .5 1 2 1 0

Individual dose (mSv)

FIG. 1. Individual dose distribution of workers. POSTER PRESENTATIONS 309 the management for equalization of the dose level among several workers with high dose in order to avoid a situation where only a limited number of workers would receive an exceptional dose. Dose equalization generally decreases work efficiency, and is not good for the purpose of collective dose reduction. However, from a management perspective, it is preferable to keep the dose differentiation among workers narrow. Besides, the psychological state of workers is important. It is not desirable to have a situation where a worker feels that only he receives a high dose. Accordingly, the sense of equity in risk distribution through equalization of dose is useful in enhancing co-operation, particularly among the group of workers with a relatively high level of exposure.

4. CONCLUSION

This study examined the benefits and dose allocation of radiological protection measures. Cases were taken from actual experience at the JAERI, and the main focus was on human aspects such as the enhancement of workers’ consciousness and the sense of equity in risk distribution. There may be some aspects of the evaluation results which apply only to Japanese radiological control methods based on co­ operative labour relations. However, the conclusion is that protection measures should be based on due consideration of the psychological state of the workers.

ACKNOWLEDGEMENT

This study was performed by the JAERI under a contract from the Science and Technology Agency, Japan.

REFERENCES

[1] ISHIKAW A, M ., et al., JPDR decommissioning program — plan and experience, Nucl. Eng. Des. 122 (1990) 357. [2] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Optimi­ zation and Decision-Making in Radiological Protection, ICRP Publication 55, Perga­ mon Press, Oxford and New York (1988). [3] STIG LITZ, J.E., Economics of the Public Sector, 2nd edn, Norton, New York (1988). [4] SHIRAISHI, M ., DANDA, H ., Efforts on exposure reduction at Shimane Power Plant, Thermal Nucl. Power 40 4 (1989) 407 (in Japanese). [5] SASAKI, F., Recent decreasing trend of personnel exposure in nuclear power plants in Japan and analysis of major factors contributing to its reduction, Thermal Nucl. Power 42 2 (1991) 157 (in Japanese). [6] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990 Recommendation of the International Commission on Radiological Protection, ICRP Publication 60, Pergamon Press, Oxford and New York (1991). 310 POSTER PRESENTATIONS

IAEA-CN-54/85P

LES MESURES RADON REALISEES PAR L’IPSN DURANT LA REHABILITATION DU SITE DE STOCKAGE DE RESIDUS DE TRAITEMENT DE MINERAI D’URANIUM DU BOUCHET, EN REGION PARISIENNE

V. LABED, M.C. ROBE, A. BENEITO, J.M. MAUREL, P. RICHON Institut de protection et de sûreté nucléaire, Département de protection de l’environnement et des installations, Service d’études et de recherches en aérocontamination et en confinement, CEA, Centre d’études nucléaires de Saclay, Gif-sur-Yvette, France

INTRODUCTION

L ’exploitation de certains minerais comme l ’uranium conduit à la production de grandes quantités de matériaux stériles appelés résidus, où sont présents des radioéléments dont l’activité spécifique est faible mais de période radioactive longue. Ces résidus sont bien évidemment d’origine naturelle mais nécessitent que des précautions soient prises dans leur gestion afin de ne pas augmenter de manière significative le risque pour les populations présentes et futures. Plusieurs possibilités de gestion de ces résidus existent depuis le site d’extraction jusqu’au site de traite­ ment du minerai. Nous présentons ci-après un exemple de réhabilitation de site et les méthodes de mesure utilisées pour suivre l’efficacité de la couverture mise en place.

DESCRIPTIF DU SITE

Entre 1949 et 1971, le Commissariat à l ’énergie atomique (CEA) a exploité en région parisienne à 40 km de Paris une installation de traitement de minerais d’uranium au Bouchet. Le terrain sur lequel était située l’usine a été démantelé dans les années 70. Cependant, un terrain annexe, extérieur à l’installation, ayant servi jusqu’en 1956 au dépôt de résidus de traitement de minerais d’uranium et, jusqu’en 1971, comme bassin de décantation des boues contenues dans les effluents de l ’usine, est resté sous la surveillance du CEA. Ainsi, le fonctionnement de l ’usine du Bouchet a conduit à la constitution sur ce terrain d’une superficie inférieure à 0,02 km2 d’un dépôt d’environ 20 000 tonnes de résidus. POSTER PRESENTATIONS 3 1 1

TABLEAU I. MESURES D’EAP DANS L’ENVIRONNEMENT DU SITE DU BOUCHET AVANT REHABILITATION

EAP 222Rn (nJ/m3)

Année % LAI station 2 2 Station 1 1 au vent Station 2 sous le vent du dépôt du dépôt

1990 38 391 124

1991 33 333 105

1992 30 299 94

1 Station caractéristique du bruit de fond régional. 2 Calcul effectué en considérant une personne du public séjournant 8760 heures, soit une année à la station 2.

Les matériaux radioactifs présents sur ce terrain proviennent donc exclusive­ ment du traitement de minerais; ce sont des produits que l’on trouve dans la nature (uranium, thorium, radium et leurs descendants comme le radon). Les différentes mesures effectuées dès 1990 sur les dépôts de résidus radifères ont montré que les nuisances potentielles sont essentiellement liées à la présence du radon. Des mesures intégrées sur des périodes mensuelles de concentration en éner­ gie alpha potentielle (EAP) due aux descendants à vie courte du radon réalisées sous le vent du dépôt et à quelques dizaines de mètres de celui-ci, ont mis en évidence une augmentation de l’exposition du public (tableau I). Dans le cadre des industries extractives, la réglementation nationale et interna­ tionale fixent des «Limites annuelles d’incorporation» ou LAI en valeur ajoutée à l ’exposition naturelle pour le public inhalant de l’air contenant du radon 222, à 2 mJ/an. Le 27 mai 1991, un arrêté préfectoral a prescrit au CEA non seulement de mesurer la concentration en descendants à vie courte du radon en 5 points autour du site mais également d’effectuer une cartographie des émissions de radon en vue de déterminer avec précision la surface à réhabiliter. Les mesures de flux d’émission radon [1] à l’interface sol-atmosphère ont eu lieu au points suivants:

— aux nœuds d’un maillage de 16 m x 16 m sur toute la surface des dépôts; — aux nœuds d’un maillage 2 m x 2 m sur les zones hors clôture, lorsque les mesures de débits d’impulsion gamma réalisées à un mètre au-dessus du sol conduisent à des valeurs supérieures à 500 chocs/seconde. 3 1 2 POSTER PRESENTATIONS

g* Zones situées en dehors de la clôture où les flux sont supérieurs à: 5 x 105 atome • rrr2 ■ s-1

Bassin de N

Echelle 16 cm I— |

FIG. 1. Le site du Bouchet.

L ’exploration du site a conduit à réaliser environ 350 mesures dont 189 en dehors de la clôture. A l ’intérieur de la clôture, les flux varient de 3 x 103 à 8,6 x 107 atome-m"2-s"1. A l’extérieur de la clôture, les mesures de flux ont permis de définir trois zones de flux supérieur à 5 x 105 atome-m"2-s '1 ainsi qu’une zone de flux supérieur à 4 x 105 atome-m'2-s"1. Toute zone ayant un flux d’émission supérieur à 5 x 105 atome-m"2-s"1 a été incluse dans le périmètre de réhabilitation du site (Fig. 1). Rappelons que le flux moyen à la surface de la terre est de 104 atome-m"2-s“1. En France, dans les massifs granitiques, les flux peuvent atteindre 105 atome-m"2-s '1, voire 106 à 107 atome-m“2-s-1, plus particulièrement sur des indices uranifères [2]; par contre, ils sont compris entre 5 x 103 et 2 x 104 atome-m"2-s '1 dans le Bassin parisien [2]. Un second arrêté préfectoral (3 août 1992) a fixé les conditions de réhabilita­ tion en vue de réduire l ’impact radiologique induit par ces dépôts, en limitant les concentrations moyennes annuelles en EAP pour les descendants à vie courte du radon 222 et du radon 220 à proximité desdits dépôts. Les travaux de réhabilitation commencés au quatrième trimestre 1992 ont consisté: — à nettoyer et à débroussailler le terrain, — à l ’aplanir et à le compacter puis à mettre en place un treillis métallique et une membrane géotextile pour séparer les résidus de traitement de la couche d’argile, POSTER PRESENTATIONS 3 1 3

— à mettre en place et compacter une couverture d’argile de plusieurs dizaines de centimètres d’épaisseur, — à déposer une couche drainante constituée de sable et de graviers pour main­ tenir la couche d’argile humide puis un tapis de terre végétale et de gazon.

Au cours des travaux, une rapide cartographie des flux d’émission a été faite après la phase d’aplanissement et de compactage du terrain. Les points de mesure ont été choisis de manière aléatoire. Sur l ’ancien bassin de décantation, on a noté une nette diminution des flux (< 5 x 105 atome-m'2-s '1) qui est probablement due: — au compactage du terrain, — à l’apport de matériaux de remblai (ancienne digue, ...) entraînant un effet d’autoconfmement sur le bassin.

Après la mise en place de la couche d’argile, des mesures de flux d’émission ont été réalisées:

— aux nœuds des mailles de 32 m X 32 m sur la base du maillage établi initiale­ ment en 1991 sur toute la surface du site, — aux points déjà cartographiés après la phase d’aplanissement et de compactage du terrain.

Sur l’ensemble du site, les flux sont inférieurs à 5 x 105 atome-m'2-s“1, excepté en un point représentatif d’une superficie inférieure à 1 m2. Pour 68% des mesures, les flux sont inférieurs à 104 atome-m’2-s '1. Les distributions en fréquence des flux mesurés en 1991 avant toute réhabilita­ tion et de ceux mesurés après mise en place de la couche d’argile mettent en évidence l ’effet de confinement apporté par les réaménagements.

Avant réhabilitation Après réhabilitation

t 10 100 1000 10000 100000 1 10 100 1000 10000 100000

FIG. 2. Distribution en fréquence des flux d ’émission radon (en 104 atome■ m~2-s~'). 3 1 4 POSTER PRESENTATIONS

Limíte de concentration 800 P moyenne annuelle en - EAP 222 au droit du dépôt en valeur ajoutée au bruit de fond régional 700 \- 280nJ/m3

- 6 0 0

¿.500 CM CM CM Après c 4 0 0 *oo réhabilitation (Q a.Œ 3 0 0 Ш 200

100 Bruit de fond régional № III.’ . o ■ i n i m 11 i m u n п « n * I 23 4S67I9I0I112I23 4 36 7 I9I0III2I 2 3 4 î 6 7l9IOIII2123 4ï67t9IOlll2l234i6 1990 1991 1992 1993 1994 Mois

FIG. 3. Concentration en EAP des descendants à vie courte du Rn.

Depuis le début des travaux de réhabilitation en mai 1993, à la station 2, la valeur moyenne d’EAP (Fig. 2) sur un an est inférieure à 30 nJ/m3, soit une réduc­ tion d’un facteur 10 environ par rapport aux années précédentes (Fig. 3). Les valeurs d’EAP montrent que les objectifs fixés par l’arrêté sont atteints.

REFERENCES

[1] LABED, V., WITSCHGER, O., ROBE, M.C. SANCHEZ, B., «Radon 222 emission flux and soil-atmosphere interface: comparative analysis of different measurement techniques», First International Workshop on Indoor Remedial Action, Rimini, 1993, à paraître. [2] R O B E , M .C ., R A N N O U , A ., L E B R O N E C , J., Radon measurement in the environ­ ment in France, Rad. Prot. Dosim. 45, 1/4 (1992) 455-457. POSTER PRESENTATIONS 3 1 5

IAEA-CN-54/86P

LA GESTION DU RISQUE RADIOLOGIQUE EN EXPLOITATION DANS L’USINE DE RETRAITEMENT DE LA HAGUE

J. KALIMBADJIAN Service de prévention et de radioprotection, Compagnie générale des matières nucléaires, La Hague, France

Le retraitement des combustibles usés satisfait un double objectif: — séparer la matière énergétique encore valorisable, uranium et plutonium, des produits de fission, — conditionner les produits de fission de façon sûre dans une matrice stable à long terme. A cette activité sont associés des risques radiologiques bien identifiés, dont la prévention et le contrôle constituent une partie essentielle de l ’expérience industrielle des usines de retraitement.

1. LES RISQUES RADIOLOGIQUES

1.1. Nature des risques

La conception des ateliers, les modes d’exploitation, la radioprotection et la surveillance médicale du personnel de La Hague prennent en compte les deux natures de risques radiologiques: — le risque d’irradiation externe, principalement associé aux rayonnements bêta et gamma des produits de fission, — le risque de contamination, lié de façon prépondérante aux émetteurs alpha et principalement au plutonium.

L ’importance relative des deux risques varie d’une étape à l’autre du retraite­ ment en fonction de la nature, de la forme physico-chimique et de l’activité spéci­ fique des substances radioactives manipulées dans les différents ateliers. En termes quantitatifs, dans les objectifs d’optimisation globale de la dose reçue par chaque individu, irradiation externe et contamination interne s’ajoutent. Qualitativement, les conséquences potentielles de ces deux types d’exposition doivent être différienciées. 3 1 6 POSTER PRESENTATIONS

1.2. Principe directeur

Le principe fondamental retenu est celui de la non-contamination permanente des ateliers, y compris lors des interventions. La mise en œuvre de ce principe se traduit par l’importance accordée aux actions de prévention. On peut citer notamment: — avant le démarrage de toute installation, les tests vérifiant l’efficacité des barrières de confinement (filtres et ventilation principalement); — en fonctionnement normal, le contrôle radiologique continu ou périodique des matériels, des locaux, des effluents, à un niveau suffisamment fin pour déceler le plus tôt possible toute dérive ou anomalie de fonctionnement; — lors d’une intervention pour maintenance, l’analyse préalable très détaillée de l ’état radiologique des équipements concernés, la définition précise du mode opératoire, puis la surveillance stricte des conditions d’exécution, notamment quant au respect des prescriptions en matière de tenue, de protection respiratoire, et quant au contrôle radiologique des lieux en matière de contami­ nation atmosphérique et surfacique.

2. COMPOSANTES PRINCIPALES DE LA MAITRISE DES RISQUES

La maîtrise des risques radiologiques repose sur trois grands types de principes:

2.1. Les principes de conception

Un objectif de conception de dimensionnement avait été fixé pour les ateliers de La Hague mis en service depuis 1989: parvenir à un nombre quasiment nul d’agents dont la dose équivalente reçue dépasse 5 mS/an en conditions normales d’exploitation. Pour atteindre cet objectif, on a conçu les usines en s’appuyant sur les principes suivants: — confinement statique (principe des trois barrières de confinement), — confinement dynamique (ventilation), — dimensionnement des protections biologiques, — dispositifs adaptés d’intervention pour la maintenance (conception des Enceintes mobiles d’évacuation du matériel ou EMEM). Les débits de dose mesurés sur les postes de travail en conditions normales sont, en conséquence, très faibles: (0,15 /xGy/h). Le nombre de postes de travail permanent en zone contrôlée est pratiquement nul compte tenu de la centralisation du contrôle-commande et de la conduite à distance. POSTER PRESENTATIONS 3 1 7

2.2. Les principes d’organisation

Ils établissent les relations entre le Service de prévention et radioprotection (SPR) et les responsables d’exploitation. Il faut souligner particulièrement:

— L ’indépendance hiérarchique et la maîtrise décisionnelle et opérationnelle du SPR qui ne rend compte qu’au directeur de l’établissement. — Simultanément, la présence permanente d’agents de radioprotection parmi les équipes d’exploitation dans chacun des ateliers. — L ’extension des missions du SPR, qui comprennent à la fois la prévention des risques, la surveillance radiologique des zones de travail et de l’environne­ ment, le suivi dosimétrique du personnel et les actions de formation; 300 per­ sonnes sont affectées à ces différentes tâches.

2.3. Les principes d’exploitation

Ils définissent précisément les procédures d’intervention pour la maintenance et les travaux neufs, et les rôles respectifs des trois partenaires: l’exploitant, le service de maintenance intervenant, le service de radioprotection. Une intervention dans un atelier exige une décision de la direction, sur la base d’un «Dossier d’inter­ vention en milieu radiologique», qui doit rassembler les trois partenaires dans un système interactif:

— l ’exploitant doit justifier son intervention de façon à éviter les doses inutiles, — la radioprotection doit fixer les contraintes radiologiques individuelles et collectives pour limiter les doses, — l’intervenant doit mettre en œuvre les meilleures méthodes et principes de travail (préparation — outillage — protections) pour optimiser l’intervention. La gestion du risque radiologique en exploitation illustre ainsi l ’application effective du principe ALARA ou «As low as reasonably achievable», par la mise en place du dialogue entre les trois partenaires, arbitré par la direction si nécessaire.

3. APPLICATION: LES OBJECTIFS FIXES

Au-delà du respect de la contrainte de 5 mSv/an en conditions normales, les objectifs de progrès actuellement poursuivis sont: — En priorité, la réduction des doses les plus élevées, liées aux interventions: faire en sorte qu’aucun agent ne reçoive une dose équivalente ou supérieure à 10 mSv/an. — En continuité avec l’effort passé, la diminution du bilan dosimétrique global. 3 1 8 POSTER PRESENTATIONS

FIG. 1. Doses moyennes annuelles par agent: Cogéma + entreprises.

û.со э + см Û_ ч Э ? «ф "D *1 S Ф с О) с .о

FIG. 2. Evolution du bilan dosimétrique et de la production des usines UP2 et UP3.

A. RESULTATS

Les évolutions observées des doses individuelles et des doses collectives en exploitation sont présentées sur les figures 1 et 2. On peut faire les observations suivantes: — De 1987 à 1993, la dose moyenne par agent exposé a diminué de façon régulière jusqu’à être divisée par 2 (Fig. 1). POSTER PRESENTATIONS 3 1 9

— Dans le même temps, la production passait de 425 à 954 t/an, de sorte que l’impact radiologique professionnel du retraitement, mesuré en dose collec­ tive rapportée à la production d’électricité nucléaire, est passé de 0,8 à 0,16 homme-Sv par GWe-an; la part strictement liée à l ’exploitation des ateliers de retraitement (UP2 et UP3) représente environ la moitié de ce total, soit 1,9 homme-Sv en 1993 (Fig. 2). — Entre 1987 et 1993, le nombre d’agents recevant plus de 10 mSv/an a été pratiquement réduit d’un facteur 10. Au total, la dose collective professionnelle strictement liée aux activités de retraitement de l ’usine de La Hague (1,9 homme-Sv en 1993) représente à peine le dixième de la dose engendrée par la radioactivité naturelle (21 homme-Sv sur la population de 9000 personnes surveillées en dosimétrie sur le site).

5. COMMUNICATION

La démarche ALARA repose sur la motivation et le comportement de chacun des acteurs. Pour cela, l ’établissement veille à ce que les travailleurs aient connaissance: — en temps réel, de l ’état radiologique des ateliers dans lesquels ils interviennent; — régulièrement, de leur bilan dosimétrique personnel.

La diffusion des résultats est assurée aux organismes responsables concernés, dans le cadre strict du respect de la confidentialité.

6. CONCLUSION

La gestion du risque radiologique met directement en jeu le comportement des différents acteurs; elle se traduit par une véritable éthique concrète et pratique en situation d’exploitation industrielle, dont on peut souligner ici quelques composantes: — L ’entreprise applique les principes d’égalité, en assurant la même surveillance sanitaire pour ses agents et pour les intervenants des entreprises externes, et d’équité en recherchant en priorité la réduction des doses individuelles les plus élevées. — Elle applique aussi le principe de précaution, en maintenant comme objectif la diminution du bilan dosimétrique global, alors qu’il ne représente déjà qu’une dose collective marginale comparé à l ’irradiation naturelle sur la population des personnes employées sur le site. 3 2 0 POSTER PRESENTATIONS

— L ’application du principe ALARA passe par la séparation des rôles et «l’équilibre décisionnel» entre le service exploitant, le service de maintenance et le service de radioprotection; cet équilibre institutionnel tient en particulier au fait que les agents de radioprotection sont physiquement présents 24 heures sur 24 et bien intégrés dans les ateliers de production. — La démarche ALARA repose, en outre, sur la motivation et l ’implication forte de chaque individu; elle se développe dans un climat psychologique favorable, et le niveau de confiance de chacun, à son poste de travail, joue un rôle déter­ minant. L ’établissement de La Hague accorde une importance particulière aux facteurs nécessaires à cette confiance, tels que l ’exigence de non-contamination absolue dans les ateliers, l’information des travailleurs sur leur situation dosimétrique, ou encore les actions de formation en radioprotection. Réciproquement, c’est bien par le niveau de confiance et de sérénité du person­ nel sur son lieu de travail que se mesure le succès d’une politique de gestion du risque; à cet égard, les résultats obtenus sur le site de La Hague sont très satisfaisants. POSTER PRESENTATIONS 3 2 1

IAEA-CN-54/101P

STRATEGY OF MAKING PUBLIC EXPOSURE DECISIONS, TAKING INTO ACCOUNT THE RADIATION INCIDENTS EXPERIENCE OF MINATOM OF RUSSIA

V.A. GUBANOV Committee for Safety, Ecology and Emergency Situations of Minatom of Russia, Moscow, Russian Federation

Before analysing the applicable laws and regulations of the Russian Federation which stipulate the intervention criteria and ensure public protection in case of radia­ tion accidents, it is necessary to address the history of this problem and to trace the evolution of the concept of safe residence in contaminated territories and variation of intervention levels, using the Chernobyl accident as an example. One of the lessons of this accident is that the general public had not perceived (in the opinion of the scientists) the rather optimal intervention criteria to be used as guidelines and procedures and kept on requiring elaboration of more stringent criteria. During the first post-accident years, laws and regulations were based on the concept of temporary dose limits. In 1986, relying on the public protection require­ ments of the document SP-AES-79 for the case of a design basis accident at a nuclear power plant, the National Radiation Protection Commission of the USSR (NRPC) recommended temporary dose limits which were approved by the Ministry of Health of the USSR (Table I).

TABLE I. TEMPORARY RADIATION DOSE LIMITS

Temporary dose limit Year (rem)

Until the end of 1986 10

1987 3

1988 2.5

To lift the restrictions on public residence in the contaminated territories, it was necessary to devise a new criterion of safe residence. In 1989 the NRPC recom­ mended, as a criterion, a 35 rem total lifetime dose. However, this criterion has never been put into effect. 3 2 2 POSTER PRESENTATIONS

On 8 April 1991 the Cabinet of Ministers of the USSR approved a concept of public residence in the territories which were radioactively contaminated after the Chernobyl accident. The concept stipulated that protective measures (remedies) should be taken if an additional radiation dose exceeded 1 mSv (0.1 rem) a year above natural and maninduced background. Taking into consideration the actual situation in the contaminated areas, the concept allowed decisions to be made on public residence depending on the levels of 137Cs, 90Sr and 239’240Pu contamination. The USSR law ‘On the social protection of residents who were impaired by the Chernobyl disaster’ was adopted on the basis of the above concept in May 1991. During the same period, a similar law was passed in Russia, which was subsequently revised and amended; the latest revision of that law was approved in 1993. The law stipulates that any relevant decision should be made taking guidance from the dose criteria and taking into account the nature of a residence territory. According to the level of radionuclide contamination, all territories are classified into three zones:

(a) alienation zone: 30 km radius zone; (b) evacuation zone: 137Cs > 15 Ci/km2, 90Sr > 3 Ci/km2, 239’240Pu >0.1 Ci/km2, averaged annual dose > 5.0 mSv (0.5 rem); (c) residence zone with a right for evacuation: I37Cs 5-15 Ci/km2, average annual dose > 1 mSv (0.5 rem). Thus, contamination standards applicable to the later stages of an accident are determined by the law. As regards the earlier and intermediate stages, in May 1990 the Chief State Physician of the USSR approved ‘The criteria for making decisions on public protection measures in case of a nuclear reactor accident’ as a guide for taking basic public protection actions against radiation exposure; these criteria are still in force. The criteria include evacuation of the population at an early stage of an accident if the estimated doses 10 days after the accident range from 50 to 500 mSv for adults and from 10 to 50 mSv for children and pregnant women; they also include population resettlement at the intermediate stage if the estimated doses for the first year range from 50 to 500 mSv. Analysis of the post-Chernobyl changes in the criteria leads to two conclusions: (1) After the Chernobyl disaster and up to now, all intervention criteria have been continuously made more and more stringent (intervention levels were lowered). The same applies to the permissible radionuclide content in foodstuffs. (2) In spite of the fact that there are criteria for the early and intermediate stages of an accident, in practice (as the Tomsk radiation incident showed in 1993) the most stringent (lower) intervention criteria, stipulated in the Chernobyl law for the later stages of an accident, are applied. POSTER PRESENTATIONS 3 2 3

On 4 March 1994 the National Radiation Protection Commission of the Russian Federation approved an elaborated concept of radiation, medical, and social protection and rehabilitation of the population of the Russian Federation subjected to radiation exposure from a nuclear emergency. In compliance with the above con­ cept, all decisions on intervention are to be made taking into account the exposure (radiation burden). However, the general public was opposed to the changes in this Chernobyl law, as well as to the application of dose values alone, because it will deprive a part of the population of certain social benefits. Therefore, it can be emphasized once again that, in practice, political circumstances always introduce corrections towards establishment of lower dose limits and other criteria. The accident on 6 April 1993 in Tomsk at the radiochemical plant of the Siberian Chemical Combine is another illustration of the abovementioned. Although the radioactive trail generally did not affect the inhabited areas, and in places where contamination did occur the dose rates were much lower than the intervention levels set forth for the early stage of an accident, the local administration considered a pos­ sible evacuation of the population. The following localities were in the contaminated zone: settlement Georgievka (16 km away from the accident site, 30 inhabitants, gamma dose rates ranging from 18 to 45 /¿R/h) and settlement Chernaya Rechka (34 km away from the accident site, radiation level 12-50 ¿¿R/h). In other words, there are standards but in real life the decision makers, trying to avoid criticism that the measures undertaken are not sufficient, agree to inap­ propriate actions. In addition, the regional authorities requested over 10 billion roubles for the Tomsk accident remedies and they did receive the money, while other financial resources had to be used for the reconstruction of production facilities. All the above money was spent for improving the regional infrastructure. And now I would like to give a more detailed description of the accident and make some conclusions: At 9:08 a.m. (Moscow time) a service tank containing uranium solution after recovery of plutonium and fission products exploded at the reprocessing plant of the Siberian Chemical Combine. The cause of the explosion was an exothermic chemical reaction between organic solution precipitates and nitric acid. As a result of the explosion, the tank ruptured; the cover of the pit which housed the tank was damaged, and the outside wall of the shop building was des­ troyed. The subsequent short circuit of electrical wiring in the building led to the roof catching fire which was quickly extinguished by the firefighters and personnel. Tank and pit cover damage resulted in a release of the reaction gases and aerosols which, in turn, led to radioactive contamination of the shop rooms, the building roof, part of the surrounding plant area and some off-site areas (approximately 35 km2). The source term contained 95Zr, 95Nb, 103Ru and 106Ru radionuclides (Table II). The total length of the radioactive trail where the radioactivity level exceeded 60 /¿R/h was 15 km. The radiation doses received by the personnel and firemen while fighting the fire did not top a few mSv (excluding three persons who received over 6 mSv), 324 POSTER PRESENTATIONS

TABLE II. RADIONUCLIDE COMPOSITIONS OF PLUME

Beta and gamma emitters

Half-life Isotope % activity (d)

Ruthenium-106 368 35 Ruthenium-103 39 2 Zirconium-95 64 20 Niobium-95 35 42 Niobium-94 104 1 Caesium-137 30 1 Chromium-51 28 1 Arsenic-125 2.8 1

Alpha emitters

Half-life Isotope % activity (a)

Plutonium-239 2.4 x 104 79 Uranium-234 2.4 X 105 12 Uranium-238 4.5 x 109 9 Uranium-235 7 x 108 None

i.e. less than the annual permissible radiation dose for personnel (50 mSv). There were no injuries after the explosion and fire. In compliance with the International Nuclear Event Scale developed by the IAEA, this incident corresponds to level 3, ‘serious incident’ without overexposure of personnel. More details about the earlier stage of the accident: At 12:30 p.m. a conclusion was made that the radiation environment in Tomsk-7 had not changed. By that time, the principal direction and magnitude of the radioactive trail on site and off site had been estimated. At 12:10 p.m. the Tomsk-7 municipal radio broadcasting committee transmit­ ted the first public information about the accident at the Siberian Chemical Combine. According to the estimates, released radioactivity totalled about 40 Ci. Total radioactivity of the solution inside the tank was around 550 Ci. Thus the radioactive discharge was less than 10% of the initial volume. POSTER PRESENTATIONS 3 2 5

As of 14 April, the concentration of 239Pu radioactivity in the samples taken from the radioactive trail in the surveillance zone did not exceed 0.008 Ci/km2. On 7 April aerial survey and airborne gamma measurements were performed to determine the contaminated territories. On 7 April an ad hoc commission was set up and left by air to the site of the accident. At the same time, the Emergency Situation Control Centre started its operation in Minatom. On the initiative of the Ministry of the Russian Federation for Foreign Affairs and Minatom, a group of IAEA experts was invited to visit the accident site; the group confirmed the correctness and reliability of the findings of independent com­ missions set up by the Russian institutions. The Siberian Chemical Combine completed the accident recovery in August. The accident pinpointed the necessity of setting up automatic radiation monitoring systems around such facilities. Otherwise, the promptness of radiation environment assessment, especially offsite assessment, deteriorates. At present, automatic radia­ tion monitoring systems are being set up at all Minatom installations with nuclear reactors or high level storage facilities; such systems will have a capability of on-line data transmission to the Minatom Emergency Situation Control Centre. Minatom continuously enhances its preparedness for potential emergencies, and undertakes efforts to improve the emergency prevention and response system of the nuclear industry. The Minatom system is a subsystem of the Russian Radiation Hazard Emergency Response System. To ensure continuous preparedness, the Emergency Situation Control Centre with a master station was established and is in operation in Minatom. Its basic tasks are as follows: — assurance of on-line acquisition and transfer of accident and emergency information; — maintenance of on-line communication (including connection of databases) with experts at research, design and development institutions of the nuclear industry who provide scientific support to and exercise supervision of especially hazardous production facilities; — elaboration of recommendations on how to implement emergency localization and recovery actions; — real time acquisition and presentation of information about radiation and chem­ ical environments at the enterprises of the nuclear industry, other ministries, and agencies as well as programme information support of the Industry Com­ mission for Emergency Situations (ICES).

To ensure prompt execution of emergency response and recovery at the nuclear industry facilities of Minatom, five Emergency-Technical Centres (ETCs) were set up. The ETCs of Minatom are the divisions of constant preparedness oriented towards carrying out relevant specialized operations during transport accidents and incidents, and operations in radioactively hazardous environments in the territories 3 2 6 POSTER PRESENTATIONS where they are located. In addition, by a decision of Minatom the ETCs can be involved in other operations in radiation environments. The ETCs are located at the following nuclear industry enterprises, taking into account the specific nature of the latter: VN IIEF (Arzamas-16) and VN IITF (Chelyabinsk-70), ETCs for nuclear weapons emergencies; Novo-Voronezh NPP, an ETC for emergency response and recovery at the NPP and research reactors; Siberian Chemical Combine (Tomsk-7), an ETC for emergencies at the nuclear fuel cycle enterprises; and the V.G. Khlopin Radium Institute (St. Petersburg), an ETC for accident and emergency prediction. Assurance of the preparedness of the nuclear industry for the early stage of an accident is a priority target of all efforts aimed at improving the available emergency response system of Minatom. A document entitled Criteria for Making Decisions on Public Protection in Case of a Radiation Accident is currently being elaborated in Minatom. This docu­ ment will include types and scenarios of different accidents, taking into consideration recent international recommendations, as well as a digital computer program for forecasting environmental contamination and a list of optimal intervention measures as applied to a specific radioactively hazardous facility of the nuclear industry. POSTER PRESENTATIONS 3 2 7

IAEA-CN-54/125P

MANAGEMENT OF RADIATION RISK: A GUIDE TO DEMONSTRATING ALARP FOR AN EXISTING PLANT

J. TELFER Scottish Nuclear Limited, Peel Park, East Kilbride

S. MORTIN Nuclear Electric pic, Gloucester

United Kingdom

1. INTRODUCTION

All industrial activity carries risks to both the public and workers from acci­ dents. Risk management is therefore a key part of managing industrial processes. A key requirement of UK legislation is that the operator must do all that is reasonably practicable to reduce the risks from the operation of his plant, and must demonstrate that he has done so to the regulator. This imposes a continuous requirement to review the operation and maintenance of the plant to ensure the application of the principle of as low as reasonably practicable (ALARP) risks. In addition, a particular require­ ment for a nuclear plant is periodic safety reviews of the safety case for the plant against modern standards. In order to ensure that decisions are made in a consistent manner and can be made transparent to the regulator, Scottish Nuclear Limited (SNL) and Nuclear Elec­ tric pic (NE) have developed guidelines and supporting methodology for the assess­ ment of risk. This guidance in our Safety Review Guidebook (SRG) for gas cooled reactors facilitates the demonstration that all that is reasonably practicable has been undertaken to reduce accident risks, or, alternatively, identifies shortfalls requiring attention. The SRG contains three basic elements: (a) the nuclear safety principles (NSPs) against which the plant should be assessed; (b) guidance on the methods to be used for assessing against those principles, including the detailed application of ALARP; (c) guidance on the specification of design changes where these are identified as necessary. This paper concentrates on the NSPs and the methodology for the application of ALARP, as these are the key tools in risk management. 3 2 8 POSTER PRESENTATIONS

2. NUCLEAR SAFETY PRINCIPLES

The NSPs are high level principles for the review and are largely consistent with the corresponding provisions of the revised N11 Safety Assessment Principles [1], which are the regulating authority’s equivalent document. These basic principles cover the general requirements for safety reviews, deterministic engineering princi­ ples and probabilistic principles associated with limiting doses to the public and to workers, and guidance on the methods for supporting safety claims. The deterministic engineering principles address the requirements for redun­ dancy (the single failure principle), diversity, and protection against internal and external hazards and the need for emergency procedures and accident management guidelines. The probabilistic principles associated with limiting doses to the public take advantage of the approach set down in the UK’s Health and Safety Executive’s docu­ ment The Tolerability of Risk [2]. Figure 1 is a graphical representation of the approach. There are three regions. — If the level of risk is low and below a specified level known as the ‘broadly acceptable region’ , then there is no need for detailed work to demonstrate that the risks are ALARP.

Unacceptable save in exceptional circumstances Just tolerable

Broadly

No need for acceptable detailed justification of ALARP Neglible risk

FIG. 1. Tolerability of risk. POSTER PRESENTATIONS 3 2 9

— If the risk is high, above the level of tolerability, then the risk is considered unacceptable and cannot be justified except in extraordinary circumstances. — If the risk is between these two levels, i.e. above the broadly acceptable region but below the upper boundary of tolerability, detailed work needs to be presented to demonstrate that the risks are ALARP. That is, safety improve­ ments should be implemented or it should be shown that the cost of reducing the risk is disproportionate to the safety benefit. The NSPs present these two levels in the form of a dose band-frequency stair­ case as shown in Fig. 2, the doses being effective doses. The ‘broadly acceptable staircase’ has been structured to be equivalent to a 10-7/reactor year individual risk of fatality arising from potential accidents. This is consistent with The Tolerability o f Risk [2], which sets a value of 10"6/year for all risks from a site. The factor of 10 reduction in the value given in the NSPs is to take account of sites involving multi­ ple reactors and the risks from normal operation. It is considered that meeting the

Dose/frequency

1E-1 CTJ Q)>N Ф CL 1E-2

o> O’Э Ф 1E-3 TS Ф о BSL Ъ ф 1Е-4 Q. Is нО 1Е-5

BSO 1E-6

0.1 1 10 1 00 1 000 10 000 Maximum effective dose (mSv)

FIG. 2. Risk staircase from SRG. BSO , broadly acceptable staircase; BSL, upper boundary o f toler­ ability staircase. 3 3 0 POSTER PRESENTATIONS above numerical targets, together with satisfying the deterministic engineering principles in the NSPs, will, in combination with good engineering and radiological practices, provide sufficient confidence that the risks are ALARP without the need for specific detailed ALARP analysis. The staircase for the upper boundary of tolerability is set at 2 orders of magni­ tude higher than the broadly acceptable staircase.

3. ALARP GUIDANCE

Clearly, in reaching decisions on whether an ALARP position has been reached, a number of factors need to be addressed and these may be qualitative or quantitative in nature. The SRG stresses that when considering whether the risks

FIG. 3. Schematic representation of the decision process. POSTER PRESENTATIONS 3 3 1 from an existing plant are ALARP or options for plant improvement to reach an ALARP position need to be presented, broad alignment of both qualitative and quantitative factors is necessary. The overall framework for the approach is shown in Fig. 3. Although not dis­ cussed in this paper, Fig. 3 indicates that in addition to the health and safety con­ siderations which influence ALARP, there are other factors which may result in our companies’ taking an overall decision that goes beyond ALARP requirements. Qualitative factors include assessment against the engineering principles, feasibility, and complexity.

4. QUANTITATIVE APPROACH TO RADIATION RISK

Quantitative assessment involves the balancing of costs of potential enhance­ ments against the benefits in terms of the risk reduction. The cost side of potential enhancement includes, as one would expect:

— design, procurement and installation costs; — maintenance and test costs; — loss of generation costs; — decommissioning costs. The benefit side is less tangible, being related to the aversion of loss to society were an accident to occur. Costs in this case derive from: — societal health effects; — agricultural restrictions; — evacuation and relocation of the public.

The societal effects have been translated into a set of probabilistic conse­ quences using the CONDOR code [3] developed by NE and others. The NRPB/CEC code COCO-1 [4] has in turn been used to provide monetary valuations to the expec­ tation values. The benefit side is weighted by the frequency of the potential release event and the remaining life of the plant. No discounting of these costs has been undertaken, making the monetary valuations somewhat pessimistic. The cost-benefit ratio (CBR) is then calculated as:

cost of potential enhancement CBR = ------valuation of the risk averted by the enhancement

The use of the assessed CBR in the decision process is shown in Fig. 4. If the CBR is less than 0.1, the potential enhancement is strongly indicated as being 3 3 2 POSTER PRESENTATIONS

X is typically 10

Area requiring и %> s , -J <л ^ w w f v closer examination of pessimisms and uncertainties in calculating cost and rfétermînéd by the detriments spoifiç$ o f «helase (toterabilíty oí risk) X is typically 1

Modification normally indicated X is typically 0.1

cost of the modification X = ------summed detriment saved

FIG. 4. Representation of quantitative ALAKP con­ siderations for accidents.

reasonably practicable. If the CBR is between 0.1 and 1, the modification is still indi­ cated as reasonably practicable but the weight attached to the assessment should take into account the uncertainties and conservatism in the analysis. If the CBR is between 1 and 10, the cost of the modification is disproportionate to the value of the reduced risk, but may still be regarded as reasonably practicable depending on the level of risk. If the CBR is greater than 10, the modification is strongly indicated as not reasonably practicable. It is clear from the above that the quantitative analysis is not intended to be a precise marker of whether the risks are already ALARP or whether that enhance­ ment should be implemented. The results of the quantitative analysis should be taken together with those of the qualitative analysis, to come to an overall judgement on the reasonable practicability of reducing risks.

5. RADIATION RISK MANAGEMENT

The principles and methods described above provide a systematic, consistent and transparent approach to assessing radiation risks ro individuals and society POSTER PRESENTATIONS 3 3 3 within the tolerability of risk framework. The standards do allow for a plant to be operated above the broadly acceptable level of risk provided the regulator is satisfied that the residual risk is ALARP. For review of plant safety, the SRG methods pro­ vide a clear indication of the acceptability of the risks presented by the plant and a framework for ALARP assessment including monetary valuation. SN and NE have therefore a clear basis for deciding whether the risk presented by the plant is ALARP or whether identified plant improvement options should be implemented. The SRG should also give confidence to the UK regulator that SN and NE have adopted a sys­ tematic and transparent risk management approach to the demonstration of ALARP.

REFERENCES

[1] H E A L T H A N D S A F E T Y E X E C U T IV E , Safety Assessment Principles for Nuclear Plants, H S E (1992). [2] H E A L T H A N D S A F E T Y E X E C U T IV E , The Tolerability of Risk from Nuclear Power Stations, H S E (revised 1992). [3] N U C L E A R ELEC T R IC PLC, C O N D O R : a Radiological Assessment Code, Nuclear Electric TD/RPB/REP/OO10 (1990). [4] H A Y W O O D , S., et al., COCO-1: Model for Assessing the Cost of Offsite Conse­ quences of Accidental Releases of Radioactivity, NRPB-R243 (1991).

Case Study 1 THE NUCLEAR WEAPONS LEGACY

IAEA-CN-S4/67P

CANCER RISK AMONG NUCLEAR BOMB SURVIVORS

Y. SHIMIZU, K. MABUCHI, I. SHIGEMATSU Radiation Effects Research Foundation (RERF), Hiroshima, Japan

1. INTRODUCTION

The Radiation Effects Research Foundation (RERF) and its predecessor, the Atomic Bomb Casualty Commission (ABCC), have conducted mortality surveillance on a cohort of 120 000 nuclear bomb survivors and non-exposed controls (Life Span Study (LSS) sample) since 1950. Periodic analyses of the mortality data have been published, with the most recent results covering the period 1950-1985 [1]. Mortality data ascertained through the nationwide family registers are virtually complete and without bias with respect to radiation dose. This is one of the most important features of this follow-up study. However, the accuracy of death certificate based diagnoses is variable for different diseases. In order to ascertain cancer incidence cases in a systematic manner, population based tumour registries were established in 1958. A comprehensive analysis of the LSS cancer incidence data for 1958-1987 based on tumour registries has been published recently [2-5]. Cancer incidence data offer more accurate diagnostic information and better insight for non-fatal cancers, e.g. thyroid and skin cancers, but case ascertainment may not be complete and is affected by migration of study subjects from the registry’s catchment area. Thus, mortality and incidence data are complementary and they are both essential for risk assess­ ment. The present report describes the cancer risks reported in the studies for mortal­ ity and incidence data.

2. SITE SPECIFIC CANCER RISK

Besides the well known increase of leukaemia risk (except for chronic lymphoid leukaemia and adult T cell leukaemia), there has been demonstrated increased risk of solid cancers (Table I). Dose response is non-linear with an upward curvature for leukaemia but remarkably linear for solid cancers. Significantly elevated risk has been found for cancers of the stomach, colon, lung, breast, ovary and thyroid. No significantly increased risk has been observed for chronic lymphoid leukaemia, adult T cell leukaemia, and cancers of the rectum, gallbladder, pancreas, uterus and bone. The results for cancers of the salivary glands, esophagus, liver, urinary bladder, skin (non-melanoma), and nervous system and for malignant

3 3 7 3 3 8 POSTER PRESENTATIONS

TABLE I. CURRENT EVIDENCE ON CANCER RISK FROM RERF STUDIES OF NUCLEAR BOMB SURVIVORS3

Strong Weak None

Statistically significant results Borderline statistical No statistically significant in one or more studies. significance or effect observed. This Questions about potential inconsistent results. More may reflect true lack of biases are largely resolved. studies may be needed. effect or result from Risk clearly related to inadequate sample size. amount of exposure.

Leukaemia (except chronic Cancer of Chronic lymphoid lymphoid leukaemia and Salivary glands leukaemia adult T cell leukaemia) Esophagus Adult T cell leukaemia Liver Cancer of Cancer of Skin (non-melanoma) Stomach Rectum Urinary bladder Colon Gallbladder Nervous system Lung Pancreas Multiple myeloma Breast (female) Uterus Malignant lymphoma Ovary Bone Thyroid

a Source: Radiation Effects Research Foundation: a brief description.

lymphoma and multiple myeloma are less clear. Results for these sites are either of borderline statistical significance or inconsistent. More studies may be needed. However, the confidence limits on the risks for the different sites are all rather wide and statistically there is no evidence that the risk of radiation induced cancer differs by site except for leukaemia.

3. TEMPORAL PATTERN OF CANCER RISK

The pattern of appearance over time of radiation induced cancers differs between leukaemia and cancers other than leukaemia. Radiation induced leukaemia occurred 2-3 years after exposure, and both relative and absolute risks reached their peaks within 6-8 years after exposure and have decreased with time. This temporal pattern differs markedly by age at the time of bombing (ATB) and type of leukaemia. It has been shown that the younger the age ATB, the greater was the risk of leukae­ mia in the early years following the bombing, and the more rapid was the decline POSTER PRESENTATIONS 3 3 9

Excess relative risk (males)

Attained age (years)

Excess absolute risk (males)

Attained age (years)

FIG. 1. Plots o f the fitted excess relative risks (upper panel) and absolute risks (lower panel) for all solid tumours in males as a function of attained age [3]. 3 4 0 POSTER PRESENTATIONS

TABLE II. PROPORTION ALIVE AMONG NUCLEAR BOMB SURVIVORS BY AGE AT THE TIME OF BOMBING (ATB): LSS SAMPLE

A g e A T B % alive in 1990

< 1 0 9 4 . 1

1 0 - 1 9 8 6 . 3

2 0 - 2 9 7 7 . 4

3 0 - 3 9 5 1 . 4

4 0 - 4 9 1 5 . 7

< 5 0 0 . 7

T o t a l 5 6 . 4

in the risk thereafter. Acute forms of leukaemia are primarily responsible for these trends. Despite the decrease in risk with time, current data do not indicate that excess leukaemia risk ever disappears completely. For cancers other than leukaemia, radiation induced cancer begins to appear much later and the absolute risk (excess cases per 104 PYSv) increases proportion­ ately with the background cancer rates, which increase with advancing age, and the excess relative risk does not change over time for specific age ATB cohorts except for the < 10 and 10-19 ATB cohorts (Fig. 1). For those exposed at young ages, a very high excess relative risk is found in the early years and this is followed by a decline. It should be noted, however, that the majority of persons exposed at ages <20 years are still alive (Table II). Therefore, further follow-up is essential since the temporal pattern of radiation risk in this group is still not completely known and its future trend will affect risk assessment.

4. EFFECT OF AGE ATB AND SEX ON CANCER RISK

For the same attained age, both the excess relative risk and the absolute risk (excess cases) of cancers other than leukaemia (Fig. 1) are higher for younger age ATB subcohorts than for older ones, suggesting that sensitivity to radiation carcino­ genesis may differ by age ATB. Although for cancers other than leukaemia, the abso­ lute risk does not differ significantly by sex, excess relative risk is higher for females than for males. For cancers of the esophagus and lung, the difference in excess rela­ tive risk by sex is statistically significant. The higher excess relative risk among POSTER PRESENTATIONS 3 4 1 females for cancers other than leukaemia at least in part reflects the fact that the back­ ground rate is higher for males than for females. These findings underscore the importance of allowing for age ATB and sex effects in risk assessment.

REFERENCES

[1] S H IM IZ U , Y ., K A T O , H ., S C H U L L , W .J., Studies of the mortality of A-bomb sur­ vivors. 9. Mortality, 1950-1985. П. Cancer mortality based on the recently revised doses (D S86), Radiat. Res. 121 (1990) 120. [2] MABUCHI, K., SODA, M., RON, E., TOKUNAGA, M., OCHIKUBO, S., SUGIMOTO, S., IKEDA, T., TERASAKI, M., PRESTON D.L., THOMPSON, D.E., Cancer incidence in atomic bomb survivors. I. Use of the tumour registries in Hiroshima and Nagasaki for incidence studies, Radiat. Res. 137 (1994) SI. [3] THOMPSON, D.E., MABUCHI, K., RON, E., SODA, M., TOKUNAGA, M., OCHIKUBO, S., SUGIMOTO, S., IKEDA, T., TERASAKI, M., IZUMI, S., PR E ST O N , D .L ., Cancer incidence in atomic bomb survivors. II. Solid tumors, 1958-1987, Radiat. Res. 137 (1994) S17. [4] PRESTON, D.L., KUSUMI, S., TOMONAGA, M., IZUMI, S., RON, E., KURAMOTO, A., KAMADA, N., DOHY, H., MATSUO, T., NONAKA, H., THOM PSON, D.E., SODA, M., MABUCHI, K., Cancer incidence in atomic bomb survivors. III. Leukemia, lymphoma and multiple myeloma, 1950-1987, Radiat. Res. 137 (1994) S68. [5] RON, E., PRESTON, D.L., MABUCHI, K., THOMPSON, D.E., SODA, M., Cancer incidence in atomic bomb survivors. IV. Comparison of cancer incidence and mortality, Radiat. Res. 137 (1994) S98. 3 4 2 POSTER PRESENTATIONS

IAEA-CN-54/68P

DOSE RESPONSE OF RADIATION INDUCED CANCER AT LOW DOSE LEVEL AMONG NUCLEAR BOMB SURVIVORS

S. FUJITA, Y. SHIMIZU, K. MABUCHI, I. SHIGEMATSU Radiation Effects Research Foundation (RERF), Hiroshima, Japan

1. INTRODUCTION

There are few epidemiological studies available to evaluate the shape of the dose-response function of radiation induced cancer at low dose level. This issue is important as the even much lower dose level is of interest in radiation protection. The Atomic Bomb Casualty Commission (ABCC) and its successor, the Radiation Effects Research Foundation (RERF), have studied the survivors of the 1945 nuclear bombings of Hiroshima and Nagasaki since 1950. The nuclear bomb survivors were exposed to a broad range of whole body radiation, and many received low doses. Indeed, among about 87 000 survivors in the Life Span Study (LSS) cohort who were assigned DS86 dose estimates, 90% received less than 0.5 Sv. Excess risk of cancer after exposure to low doses of radiation has been evalu­ ated in two ways: by direct analyses of low dose exposed survivors and by extrapola­ tion from high dose studies. Analysis of the dose response for cancer mortality and other indices of radiation damage within the low dose range of T65DR was con­ ducted in an attempt to detect the phenomenon of radiation hormesis, if present, among nuclear bomb survivors, and failed to suggest the existence of radiation hormesis [1]. The availability of updated cancer mortality data for 1950-1985 [2] (5936 solid tumour cases and 202 leukaemia cases) and cancer incidence data for 1958-1987 [3, 4] (8613 solid tumour cases and 231 leukaemia cases) and of the revised dosimetry system (DS86) prompted extensive analyses of dose response within the low dose range among survivors in the LSS cohort [5]. Also, analyses of the nonlinearity of dose-response curves for cancer mortality [6] and for cancer incidence [7] were performed for low dose extrapolation. In this paper, we present a summary of these reported studies on dose response of radiation induced cancer at low dose level among nuclear bomb survivors.

2. RESULTS AND DISCUSSION

For the whole range of exposure doses, including high doses, the cancer mortality data for 1950-1985 [2] and the cancer incidence data for 1958-1987 [3, 4] POSTER PRESENTATIONS 3 4 3

Bone marrow dose (Sv) Colon dose (Sv)

(a) (b)

FIG. 1. Excess relative risk estimates and 95% confidence intervals for organ dose groups in the dose range < 4.0 Sv o f (a) mortality from leukaemia and (b) mortality from all cancers except leukaemia. Fitted dose-response curves are plotted by solid line for the linear model and by dotted line for the linear-quadratic model, respectively.

in the LSS cohort were used to examine the dose response of radiation induced cancer for leukaemia and all cancers except leukaemia as a group (solid tumours) separately, as it is known that the responses are different. The results were similar for both the mortality and the incidence data. There was an apparent non-linearity in dose response for leukaemia, while the dose response for solid tumours was remarkably linear. In Fig. 1, estimates of excess relative risk are shown for six dose groups in the dose range lower than 4.0 Sv. While a linear-quadratic model is consis­ tent with the data for leukaemia, the quadratic term in a linear-quadratic model is not significantly different from zero and the linear model is rather consistent with the data for solid tumours, although the linear-quadratic model cannot be excluded. With the data limited to the low dose range lower than 0.5 Sv, estimates of relative risk for five dose groups 0.01-0.02, 0.02-0.05, 0.05-0.10, and 0.10-0.20 Sv compared with the 0 Sv group are shown in Fig. 2. Relative risks up to 0.2 Sv for both leukaemia and solid tumours are not significantly different from unity [5]. Statistically significant risks are presently seen only above 0.2 Sv. This result, however, does not mean that there is no excess risk below 0.2 Sv, because of the scatter of the data points in the low dose range. Published analyses do not adequately allow for sex and age at exposure effects on the risk. Allowing for these effects increases low dose leukaemia risk estimates. For solid tumours, these dose groups all have positive nominal risk estimates and the slope of the dose response 3 4 4 POSTER PRESENTATIONS

ю

ур\л 0Л|гв|эу ю о

о S' сл

с о оо

о г -AV о сэ О о о о ур\1 ал1№|вц

S' СЛ low low dose range except (<0.5 leukaemia Sv), and of (c) (a) incidence mortality from of thyroid leukaemia, cancer. (b) mortality from all cancers FIG. 2. Relative risk estimates and 95% confidence intervals for organ dose groups in the

ф с о со

>ÍS|j елцерц POSTER PRESENTATIONS 3 4 5 for doses lower than 0.5 Sv does not differ significantly from the slope for the dose range lower than 4 Sv. For some specific cancer sites, point estimates of the excess risk are negative in some low dose categories, although not significantly so. The cancer mortality [6] and incidence [7] data were also examined to estimate a low dose extrapolation factor (LDEF), which is the factor by which linear risk esti­ mates from these data should be divided to give appropriate estimates of risk at low dose level. It is assumed here that the risk at low dose level is estimated by the linear term in a linear-quadratic model, which is consistent with the data. This factor is equivalent to the dose and dose rate effectiveness factor (DDREF). Considerations were given to the uncertainty of dose estimates and limitation of data for linearity of dose response. The estimated LDEF for leukaemia is about 2 (mortality 2, inci­ dence 2.5), which is significantly greater than 1. The upper 90% confidence limit is 8.4 for incidence data analysis. For solid tumours the estimated LDEF is about 1 (mortality 1.2, incidence 1) with small variation, and the 95% upper confidence limit is about 2 for incidence data analysis.

REFERENCES

[1] KATO, H., SCHULL, W.J., AWA, A., AKIYAMA, M., OTAKE, M., Dose- response analyses among atomic bomb survivors exposed to low-level radiation, Health Phys. 52 (1987) 645. [2] S H IM IZ U , Y ., K A T O , H ., S C H U L L , W .J., Studies of the mortality of A-bomb survivors. 9. Mortality, 1950-1985. П. Cancer mortality based on the recently revised doses (D S86), Radiat. Res. 121 (1990) 120. [3] THOMPSON, D.E., MABUCHI, K., RON, E., SODA, M., TOKUNAGA, M., OCHIKUBO, S., SUGIMOTO, S., IKEDA, T., TERASAKI, M., IZUMI, S., P R E ST O N , D .L ., Cancer incidence in atomic bomb survivors. Part II: Solid tumours, 1958-1987, Radiat. Res. 137 (1994) S17. [4] PRESTON, D.L., KUSUMI, S., TOMONAGA, M., IZUMI, S., RON, E., KURAMOTO, A., KAMADA, N., DOHY, H., MATSUO, T., NONAKA, H., THOMPSON, D.E., SODA, M ., MABUCHI, K., Cancer incidence in atomic bomb survivors. Ш . Leukaemia, lymphoma and multiple myeloma, 1950-1987, Radiat. Res. 137 (1994) S68. [5] SHIMIZU, Y., KATO, H., SCHULL, W.J., MABUCHI, K., “ Dose-response ana­ lysis among atomic-bomb survivors exposed to low-level radiation” , Low Dose Irradia­ tion and Biological Defense Mechanisms (S U G A H A R A , T ., S A G A N , L .A ., A O Y A M A , T ., Eds), Elsevier, Amsterdam (1992) 71-74. [6] PIERCE, D .A ., V A E T H , M ., The shape of the cancer mortality dose-response curve for atomic bomb survivors, Radiat. Res. 126 (1991) 36. [7] V A E T H , M ., P R E ST O N , D .L ., M A B U C H I, K ., “ The shape of the cancer incidence dose-response curve for the A-bomb survivors” , Low Dose Irradiation and Biological Defense Mechanisms (S U G A H A R A , T ., S A G A N , L .A ., A O Y A M A , T ., Eds), Elsevier, Amsterdam (1992) 75-79. 3 4 6 POSTER PRESENTATIONS

IAEA-CN-154/127P

DOSE ASSESSMENT, RADIOECOLOGY AND COMMUNITY INTERACTION AT FORMER NUCLEAR TEST SITES1

W.L. ROBISON Lawrence Livermore National Laboratory, Livermore, California, United States of America

1. INTRODUCTION

The United States of America conducted a nuclear testing programme at Bikini and Enewetak atolls in the Marshall Islands from 1946 through 1958. A total of 66 nuclear devices were tested — 23 at Bikini Atoll (total yield 77 megatons) and 43 at Enewetak Atoll (total yield 33 megatons). This resulted in contamination of many of the islands at each atoll. The BRAVO test (yield 15 megatons) on 1 March 1954 contaminated several atolls to the east of Bikini Atoll, some of which were inhabited. We have conducted an experimental, monitoring, and dose assessment programme at atolls in the northern Marshall Islands for the past 20 years. The goals have been to: (1) determine the radiological conditions at the atolls; (2) provide dose assessments for resettlement options and alternate living patterns; (3) develop and evaluate remedial measures to reduce the dose to people re- inhabiting the atolls; (4) discuss our results with each of the communities and Government officials of the Marshall Islands to help them understand the data as a basis for resettlement decisions. The remaining radionuclides at the atolls that contribute any significant dose are 137Cs, 90Sr, 239 +240Pu and 241Am.

2. FIELD EXPERIMENTS

We have evaluated all potential exposure pathways: terrestrial food chain, external gamma, marine food chain, inhalation, and cistern and groundwater. Our

1 Work performed under the auspices of the US Department of Energy at Lawrence Livermore National Laboratory under contract W-7405-Eng-48. POSTER PRESENTATIONS 3 4 7

TABLE I. ESTIMATED 30 YEAR EFFECTIVE INTEGRAL DOSE IN cSv FOR THREE CONTAMINATED ATOLLS IN THE MARSHALL ISLANDS WHEN IMPORTED FOODS ARE AVAILABLE

Bikini Enjebi Rongelap Island3 Islandb Island0

External 0.91 0.5 0.24

Ingestion

137Cs 8.9 3.0 0.33 90Sr 0.10 0.10 0.0087 239 + 240^ 0.011 0.0014 0.0013 241Am 0.0067 0.0014 0.0014

Inhalation 239 + 240py 0.013 0.053 0.0029 24’A m 0.0074 0.022 0.0019

Total 10 3.6 0.59

“ Bikini Atoll. b Enewetak Atoll. c Rongelap Atoll.

TABLE II. ESTIMATED 30 YEAR EFFECTIVE INTEGRAL DOSE IN cSv FOR THREE CONTAMINATED ATOLLS FOR THE VARIOUS EXPOSURE PATHWAYS

Bikini Enjebi Rongelap Pathway Island“ Island b Islandc

Terrestrial food 9.0 3.0 0.34 External gamma 0.91 0.5 0.24 Marine food 0.0049 0.0021 0.0016 Cistern and ground water 0.016 0.0045 0.00051 Inhalation 0.021 0.077 0.0048

Total 10 3.6 0.59

a Bikini Atoll. b Enewetak Atoll. c Rongelap Atoll. 3 4 8 POSTER PRESENTATIONS dose assessments indicate that most of the potential dose comes from 137Cs via the ingestion pathway, as shown in Table I. Caesium-137 also contributes to the total dose via the external gamma pathway. The other radionuclides contribute a small portion of the total estimated dose. The contribution of each pathway is shown in Table II. The terrestrial food chain is the most significant contributor to the total dose, accounting for about 90% of the 30 year effective integral dose at Bikini Atoll. Consequently, the diet, particularly the percentage intake of local foods, becomes very important, and more detailed discussions and validation of diet models are given in Refs [1, 2]. The external gamma pathway is next in significance and the other pathways contribute little to the total estimated dose. More detail on the methodology to evaluate each of the pathways can be found in Refs [1-3]; these references also include an uncertainty analysis for the estimated doses.

3. REMEDIAL MEASURES

The role of 137Cs and the terrestrial foods in the total estimated dose led to an experimental programme to evaluate remedial measures that might remove 137Cs from the soil or block the uptake into food crops. Many types of remedial measures were evaluated, including saltwater leaching, vegetation cropping, clay and zeolite soil amendments, and excavation of the top 40 cm of the soil column. The excavation of the surface soil is very effective in reducing the radionuclide inventory and, conse­ quently, any potential dose. However, the environmental impact of removing the coconut groves, other types of food trees, and the surface soil containing organic matter that has taken centuries to develop is enormous. One of the more effective and environmentally sound methods found was the application of potassium (K) to the surface soil. The К can be applied as KCl or as a full fertilizer including nitrogen (N) and phosphorus (P), and natural rainfall transports the К to the root zone. This also happens to be the easiest method to implement of all those we evaluated. The results of one major field experiment involving coconut trees is shown in Fig. 1. Coconuts were selected for the initial experiments because they are a major food source in the Marshall Islands. Various amounts of К and K + NP were applied to coconut trees in a 12 acre site [4]. The initial К treatment consisted of four equal applications over the first 9 months. As can be seen, reduction in the 137Cs concen­ tration of more than 20 fold has been achieved. This provides a significant reduction in the total estimated dose such that all islands could be inhabited and the people would receive annual doses in the range of worldwide background doses. Similar results have been found for other important food crops such as breadfruit, Pandanus fruit, papaya, banana and vegetables. This option for reducing the dose to returning populations does not require removal of the soil and vegetation, and leaves the island soil and vegetation system intact. POSTER PRESENTATIONS 349

FIG. l. Bikini Island coconut l37Cs-K experiment. KIOOO indicates that potassium (K) was applied at 1000 kg/ha, K2000 at 2000 kg/ha. Nitrogen and phosphorus (NP) were applied at N = 530 kg/ha and P = 230 kg/ha. KO is the control. The K20001 treatment (marked with X ) indicates the start of a slow increase in the uptake of 137Cs about 4 years after the initial application of K; this row of trees received only the first application, i.e. 1000 kg/ha in February 1985.

4. COMMUNICATION AND INTERACTION

An equally important part of our programme has been the repeated interaction with the atoll communities. We have had many meetings with the communities at our field experimental sites at their atolls, at their current residential locations, and at our laboratory in Livermore, California. We have had many hours of discussions at each of these meetings about our dose assessment results, our experimental data and resettlement options. Over the years, these meetings have led to a continuing interac­ tion between the scientists and the people. The same topics and issues are discussed at each meeting, but also new topics or questions arise at each meeting as the people develop a greater understanding of the radiation related issues. The repetition has proved to be very important. The increased interaction between the scientists and the people has over time led to a development of a level of trust in the scientists by the communities. It has also led to an expansion of com­ munity interest in and understanding of the radiological issues. The trust by the 3 5 0 POSTER PRESENTATIONS people in our data and reports is very important, because these data and communica­ tions are the basis from which the people can make an informed decision about the resettlement options at the atolls.

REFERENCES

[1] ROBISON, W.L., CONRADO, C.L., BOGEN, K.T., An Updated Dose Assessment for Roneglap Island, Lawrence Livermore National Laboratory, Livermore, C A , UCRL-LR-107036 (July 1994). [2] INTERNATIONAL ATOMIC ENERGY AGENCY, “A dose assessment for a US nuclear test site — Bikini Atoll” , Assessing the Radiological Input of Past Nuclear Activities and Events, IAEA-TECDOC-775, Vienna (1994) 11-24. [3] ROBISON, W .L., CONRADO, C.L., PHILLIPS, W .A., Enjebi Island Dose Assess­ ment, Lawrence Livermore National Laboratory, Livermore, C A , UCRL-53805 (July 1987). [4] R O B IS O N , W .L ., S T O N E , E .L ., The effect of potassium on the uptake of 137Cs in food crops grown on coral soils: Coconut at Bikini Atoll, Health Phys. 62 (1992) 496. Case Study 2 CANCER AND LEUKAEMIA CLUSTERS

IAEA-CN-54/25P

RISK OF CHILDHOOD LEUKAEMIA FROM THE RADIATION EXPOSURE OF FATHERS BEFORE CONCEPTION

R. WAKEFORD, W.J. ANDERTON British Nuclear Fuels pic, Risley, Warrington, United Kingdom

1. INTRODUCTION

In November 1983, a television documentary claimed to have found an excess of childhood leukaemia cases in the coastal village of Seascale near the Sellafield nuclear complex in West Cumbria, England. This excess was confirmed by an independent scientific inquiry which consequently recommended that further detailed investigations be carried out. One recommendation was that a case-control study of childhood leukaemia and lymphoma in West Cumbria should be conducted. The initial results of this case-control study were reported in February 1990 by Gardner et al. [1]. The study examined affected children bom, and subsequently diagnosed during 1950-1985, in West Cumbria. The most notable finding was a positive statistical association between relatively high radiation doses measured on film badges worn by men employed at Sellafield before the conception of their chil­ dren and the incidence of leukaemia among these children. The authors suggested that this association was sufficient to account for the excess cases in Seascale. The UK Committee on Medical Aspects of Radiation in the Environment (COMARE) [2] were “ cautious in our interpretation” because the association was “ novel” and “ based upon very small numbers” . COMARE stressed that “ these findings require confirmation from other studies” . However, coverage of this study by the media did not reflect the scientific uncertainty in interpretation of the association. Radiation workers and the general public were left with the impression that this epidemiologi­ cal study had established a definite link between occupational levels of radiation exposure and leukaemia in children, generating considerable concern. In addition, the findings became the principal basis of lengthy legal action in the English High Court. Since the publication of these findings much additional scientific information has become available which has enabled the association to be assessed more reliably.

3 5 3 3 5 4 POSTER PRESENTATIONS

2. SUBSEQUENT FINDINGS

Reports of excess childhood leukaemia cases around the nuclear installations at Dounreay, Caithness, and at Aldermaston and Burghfield, West Berkshire, were published in the 1980s. These reports were difficult to interpret, but the cases have been the subject of detailed case-control studies [3 ,4]. These studies are small scale and of limited statistical power, but they have demonstrated that the excess cases can­ not be accounted for by paternal employment in the nuclear industry prior to conception. Two larger case-control studies conducted in Ontario and in Scotland failed to find evidence in support of paternal preconceptional irradiation (ppi) increasing the risk of childhood leukaemia [5, 6]. In addition, close examination of leukaemia incidence among the Japanese children born to survivors of the nuclear bombings of Hiroshima and Nagasaki did not reveal a dose related risk [7]. Further detailed studies of West Cumbria have also been published. Parker et al. [8] linked the fathers of every child born in Cumbria during 1950-1989 to men employed at Sellafield. Only 8% of births to fathers employed at Sellafield prior to conception of children were to residents of Seascale, and these births were associated with just 7% of the collective dose of ppi. This small fraction of the putative excess risk of childhood leukaemia contrasts strongly with the distribution of excess cases of childhood leukaemia among children of the Sellafield work force: the excess cases are effectively confined to Seascale, whereas the collective dose is not. The UK Health and Safety Executive (HSE) have comprehensively inves­ tigated cancer among the children of the male Sellafield work force [9]. They found that the association between leukaemia and paternal preconceptional dose was restricted to those born in Seascale. No association with dose due to internally deposited radionuclides was reported, and no association was found between ppi and other forms of childhood cancer. Little et al. [10] have compared the results from the HSE study with those for Hiroshima and Nagasaki, Ontario and Scotland. Not only is the association between childhood leukaemia and ppi confined to Seascale births, but the Seascale association is statistically incompatible (p < 0.01) with all the other data sets considered, including children of the Sellafield work force born in West Cumbria outside Seascale. Little et al. [10] investigated the possibility that the confinement of the associa­ tion to Seascale might be due to synergy between ppi and some ‘Seascale co-factor’ . However, they found that such an interaction would have to be implausibly strong. Synergy was also examined by Wakeford et al. [11], who showed that the radiation induced mutation rate required to initiate such a leukaemogenic process would have to be incredibly high — some 80 times greater than the rate for all dominant heredi­ tary effects in the first generation. Kinlen [12] has demonstrated that the excess of cases diagnosed while the chil­ dren were resident in Seascale occurs both among those born in the village and POSTER PRESENTATIONS 3 5 5 among those bom elsewhere. Moreover, the excess of affected children born outside Seascale cannot be accounted for by ppi. Therefore, even if ppi were to be a cause of childhood leukaemia, this factor would not be sufficient to explain the excess cases which have occurred in Seascale. Most of the scientific evidence outlined above was available to the judge in the aforementioned legal action. He concluded ‘decisively’ that ppi was not a material contributory cause of the Seascale excess. Similarly, an authoritative review of the scientific evidence by Doll et al. [13] concluded, “ In our opinion, the hypothesis that irradiation of the testes causes any detectable risk of leukaemia in subsequent off­ spring cannot be sustained’ ’. It seems most likely that the fathers of affected children born in Seascale were associated with high cumulative film badge readings by chance. Unfortunately, given the extensive coverage by the media of the original association reported by Gardner et al., media reporting of subsequent ‘negative’ findings has been patchy at best and at times somewhat grudging. Members of the public can be forgiven for a lack of appreciation of the strength of evidence against a causal interpretation of this association.

3. CONCLUSIONS

The example of the statistical association between childhood leukaemia and paternal preconceptional irradiation illustrates the great care with which novel associations reported in epidemiological studies need to be interpreted. Non- scientific journalists cannot be expected to appreciate the distinction between associa­ tion and causation unless they are appropriately informed. If sufficient care is not taken by scientists in communicating novel epidemiological results to journalists, then alarming over-interpretations of statistical associations will continue to occur in the media. Unfortunately, ‘positive’ findings are more attractive to journalists than ‘negative’ findings, and the sparse media coverage of results published after the original report of Gardner et al. illustrates just how difficult it is to redress the balance of public perception once the initial period of excitement has passed.

REFERENCES

[1] GARDNER, M.J., SNEE, M.P., HALL, A.J., POWELL, C.A., DOWNES, S., T E R R E L L , J.D., Results of case-control study of leukaemia and lymphoma among young people near Sellafield nuclear plant in West Cumbria, BM J 300 (1990) 423. [2] FREEMAN, R.N., Radiation, Commons Hansard 170 (1990) 430. [3] URQUHART, J.D., BLACK, R.J., MUIRHEAD, M.J., SHARP, L., MAXWELL, M., E D E N , O.B., A D A M S JON ES, D ., Case-control study of leukaemia and non- Hodgkin’s lymphoma in children in Caithness near the Dounreay nuclear installation, BMJ 302 (1991) 687. 3 5 6 POSTER PRESENTATIONS

[4] ROMAN, E., WATSON, A., BERAL, V., BUCKLE, S., BULL, D., BAKER, K., R Y D E R , H ., B A R T O N , C ., Case-control study of leukaemia and non-Hodgkin’s lym­ phoma among children aged 0-4 years living in West Berkshire and North Hampshire health districts, B M J 306 (1993) 615. [5] McLAUGHLIN, J.R., KING, W.D., ANDERSON, T.W., CLARKE, E.A., A S H M O R E , J.P., Paternal radiation exposure and leukaemia in offspring: The Ontario case-control study, B M J 307 (1993) 959. [6] K IN L E N , L.J., C L A R K E , K ., B A L K W IL L , A ., Paternal preconceptional radiation exposure in the nuclear industry and leukaemia and non-Hodgkin’s lymphoma in young people in Scotland, BM J 306 (1993) 1153. [7] YOSHIMOTO, Y., NEEL, J.V., SCHULL, W.J., KATO, H., SODA, M., ETO, R., M A B U C H I, K ., Malignant tumors during the first 2 decades of life in the offspring of atomic bomb survivors, Am . J. Hum. Genet. 46 (1990) 1041. [8] PARKER, L., CRAFT, A.W., SMITH, J., DICKINSON, H., WAKEFORD, R., BINKS, K., M cELVENNY, D., SCOTT, L., SLOVAK, A., Geographical distribution of preconceptional radiation doses to fathers employed at the Sellafield nuclear installa­ tion, West Cumbria, BM J 307 (1993) 966. [9] H E A L T H A N D S A F E T Y E X E C U T IV E , H SE Investigation of Leukaemia and Other Cancers in the Children of Male Workers at Sellafield, H S E Books, Sudbury, U K (1993). [10] LITTLE, M.P., W AKEFORD, R., CHARLES, M .W ., A comparison of the risks of leukaemia in the offspring of the Sellafield workforce born in Seascale and those born elsewhere in West Cumbria with the risks in the offspring of the Ontario and Scottish workforces and the Japanese bomb survivors, J. Radiol. Prot. 14 (1994) 187. [11] WAKEFORD, R., TAWN, E.J., McELVENNY, D.M., BINKS, K., SCOTT, L.E., P A R K E R , L., The Seascale childhood leukaemia cases — the mutation rates implied by paternal preconceptional radiation doses, J. Radiol. Prot. 14 (1994) 17. [12] K IN L E N , L.J., Can paternal preconceptional radiation account for the increase of leukaemia and non-Hodgkin’s lymphoma in Seascale? BM J 306 (1993) 1718. [13] D O L L , R., E V A N S , H.J., D A R B Y , S.C ., Paternal exposure not to blame, Nature 367 (1994) 678. POSTER PRESENTATIONS 3 5 7

IAEA-CN-54/46P

CLUSTER OF CHILDHOOD LEUKAEMIAS NEAR THE GERMAN BOILING WATER REACTOR AT KRÜMMEL: WAYS OF ELUCIDATION

I. SCHMITZ-FEUERHAKE, B. DANNHEIM, I. GRELL-BÜCHTMANN, A. HEIMERS, W. HOFFMANN, B. OBERHEITMANN, H. SCHRÖDER, H. ZIGGEL Department of Physics, University of Bremen, Bremen, Germany

1. INTRODUCTION

Between November 1989 and May 1991, five cases of leukaemia in children under 15 years of age were diagnosed in a number of small villages situated opposite the Krümmel boiling water reactor (1300 MW(e)) on the river Elbe, about 35 km southeast of the city of Hamburg in northern Germany. Additionally there was one case of juvenile leukaemia and one of aplastic anaemia in a child. Only 0.1 case of childhood leukaemia would have been expected within this period (Fig. 1).

1. INTRODUCTION

According to the Sellafield experience, clusters are investigated in the follow­ ing phases: — Phase I. ‘Suspicion’: An excess of cancer in children is suspected in the population. — Phase П. ‘Collection of data and arguments’ : Television teams start question­ ing inhabitants of the suspicious region and ‘experts’ . — Phase Ш. ‘Public pressure on politicians’: Continuous discussions on televi­ sion and in magazines take place. — Phase IV . ‘Forcing of official reactions’: A committee of inquiry is set up. — Phase V. ‘Lasting uncertainty’ : The committee confirms the cases, but denies that the exposure was sufficient.

In recent years, two leukaemia clusters in German children were treated exactly in this way [1, 2]. In 1991, an exceptional increase was discovered in the region of Elbmarsch near the Krümmel plant (BWRK). It was evidently caused by 3 5 8 POSTER PRESENTATIONS

FIG. 1. Map of the Elbmarsch region. Residences of the leukaemia cases and year of diagnosis.

the reactor because there was not only a local correspondence (Fig. 1) but also the most probable incubation period of about 5 years, related to the onset of power production in 1984. Furthermore, the magnitude of the effect was self-explanatory (7.4 fold elevation relating to 10 years of observation) because no other environmen­ tal agent than radioactivity could generate such an increase. The Federal Minister of the Environment and Reactor Safety had declared on television that a viral infection was the probable explanation (phase V of investiga­ tion). An inquiry commission had been founded by the governments of Nieder­ sachsen and Schleswig-Holstein, which are involved because the diseases were observed in the first area and the plant is surveyed by the second. The hypothesis of viral infection is not supported by the recent result of an inci­ dence study of leukaemias in the entire region for persons of all ages [3]. It was shown that a leukaemia increase had occurred since 1984 on both sides of the river, which was significant for the chronic myeloid type in the 5 km region for both sexes and all ages (relative risk 1.94) and also for all types up to the age of 65 (relative risk 1.78). POSTER PRESENTATIONS 3 5 9

TABLE I. BIOLOGICAL DOSIMETRY

Date of blood No. of analysed XT R No. No. ofс dicentrics j ' 316 of * sample metaphases dicentrics x 10

A. Inhabitants of Elbmarsch

1-19 1992-1993 18 371 34 1.85 ± 0.35a

B. Control group

1-15 1988-1993 10 351 3 0.3 ± 0.17a a Standard error of the mean.

TABLE П. TRITIUM IN TREES

Tritium Tree Location Annual ring zone concentration Laboratory (Bq/kg)

Chestnut Elbmarsch, riverside ca. 1984-1992 14.9 ± 0.3 Г opposite B W R K

Chestnut Elbmarsch, riverside ca. 1986-1992 33.0 ± 1.0 2b opposite B W R K

Apple Elbmarsch, Tespe ca. 1984-1992 33.9 ± 0.3 1

Spruce Grünhof, 1.6 km east of ca. 1984-1992 5.5 ± 0.5 2 BWRK

Chestnut Bremen ca. 1986-1992 < 4 .0 2 a Institute for Radiochemistry, Technical University of Munich. b Central Laboratory for Isotope Technology, Federal Research Centre for Nutrition, Karlsruhe.

2. RESULTS

The necessary mean bone marrow dose for induction of the effect in the children of Elbmarsch was estimated by us to be about 100 mSv. Our conclusion after 2 years of investigation was that several exposures short lived rare gas isotopes 3 6 0 POSTER PRESENTATIONS must have occurred which were above legal limits. This was derived on the basis of the following results: (1) The rate of dicentric chromosomes in peripheral lymphocytes of 19 adults in the affected region was studied in 1992 and 1993 and showed a sixfold eleva­ tion (Table I), corresponding to at least 40 mSv of mean whole body dose. (2) Elevated contamination of the soil in the vicinity of the BWRK by 137Cs, a progeny of I37Xe (3.9 min), without an increase of the isotope 134Cs, must be interpreted as rare gas emissions above legal limits. Caesium-137 is often accompanied by 90Sr activity in the same order of magnitude (progeny of 90Kr). Sporadically other progeny of short lived rare gases such as 89Sr, 85Zr and I41Ce were found. (3) Elevated concentrations of tritium were found in trees in the neighbourhood of the BWRK, which can only be explained by illegal gaseous emissions (Table II). (4) In our opinion, chronic leakages of radioactivity from the turbine building can be proved in the monitoring systems of the plant operator and the supervising ministry.

3. DISCUSSION

Our results have not been accepted up to now by the officials, and not only because of the well known influence of lobbyism. Another reason is the nature of risk perception by relevant sectors of the scientific community. The latter — influenced by more than 20 years of nuclear debate, by opinion leaders connected with the former nuclear research centres, and by the traditions of the ICRP — con­ tinue to believe in only a ‘hypothetical’ risk. Even after dose revision in Hiroshima and Nagasaki, the late effects of the Chernobyl event, and the findings in nuclear workers exposed within dose limits, they are not able to realize that low dose effects do occur and in their neighbourhood.

REFERENCES

[1] HOFFMANN, W ., KUNI, H., et al., “ Leukämiefälle in Birkenfeld und Umgebung” , Niedrigdosisstrahlung und Gesundheit (KÖHNLEIN, W., et al., Eds), Springer- Verlag, Berlin (1990) 175-181. [2] SCHMITZ-FEUERHAKE, I., VON BOETTICHER, H., et al., “ Strahlenbelastung durch Röntgendiagnostik bei Leukämiefallen in Sittensen im Landkreis Rotenburg/Wümme” , Neue Bewertung des Strahlenrisikos (LENGFELDER, E., WENDHAUSEN, H., Eds), MMV Medizin Verlag, München (1993) 93-101. [3] HOFFMANN, W ., GREISER, E., Report to the Committee, Bremer Institut für Präventionsforschung und Sozialmedizin, July 1994. POSTER PRESENTATIONS 3 6 1

IAEA-CN-54/83P

CLUSTER OF CHILDWOOD LEUKAEMIA IN ELBMARSCH, GERMANY

D. HARDER Institute of Medical Physics and Biophysics, University of Göttingen, Göttingen, Germany

1. INTRODUCTION

The cluster of leukaemia cases in the four villages making up the community of Elbmarsch occurred within less than 24 months (see Fig. 1). Five children and one young adult were taken ill with leukaemia and one child with aplastic anemia, which can develop into leukaemia. During the search for possible causes, all of the known exogenic risk factors, especially of a chemical and pharmacological nature,

Case Date Birth Sex Residence Diagnosis 1 t 12/89 1982 F Tespe Aplastic anaemia 2 02/90 1986 F Avendorf (C)-ALL 3 03/90 1981 M Tespe ALL 4 t 04/90 1981 M Rönne Monoblastocytic leukaemia 5 01/90 1989 F Tespe (C)-ALL 6 t 04/91 1970 M Marschacht AML 7 05/91 1988 F Avendorf (C)-ALL Nuclear power ^ station Krümmel

Nuclear research centre Geesthacht GKSS

7

Avendorf

FIG.l. Extension of the Elbmarsch community along the Elbe River and number of leukaemia cases. 3 6 2 POSTER PRESENTATIONS were checked, with negative results. All children lived in rural conditions in privately owned houses of recent construction, ate food grown on their families’ land, and had contact with animals. Of the exogenic factors, radiation was investigated in great detail since the 1316 MW Krümmel boiling water reactor, at a distance of between 200 m and 6 km from the cases, has been in operation since 1983, forming part of the electrical power generating system of Schleswig-Holstein, which is 83.5% nuclear (1991). In order to use even unconventional sources of information in addition to knowledge of the physical system of emission and effluent control, the possible depo­ sition of radioactivity in plants was checked, biological dosimetry comparisons were

TABLE I. BIOLOGICAL DOSIMETRY (1993)a

Yield/1000 cells Area Dicentrics/rings Cells and 9 5 % Cl

Children (four laboratories)

Elbmarsch 14 32 580 0.430 (0.235/0.721) ( n = 42)

Plön ( n = 30) 17 24 065 0.706 (0.412/1.487)

Elbmarsch 5 18 299 0.273 (0.089/0.638) ( n = 24), local food products

Elbmarsch 9 14 281 0.630 (0.288/1.196) ( n = 18), usual food products

Adults, female (three laboratories)

Elbmarsch 24 b 30 204 0.795 (0.509/1.182) ( n = 30)

Plön 20 28 556 0.700 (0.428/1.083) ( n = 30)

a Result: No significant difference with the control area Plön (see Fig. 2). D N A fingerprint identification in 23 samples. Present debate over scoring efficiency of one of the laboratories is unjustified, statistical fluctuation only. b One person with seven dicentrics/rings in 1008 cells. POSTER PRESENTATIONS 3 6 3

FIG. 2. Elbe River and the Elbmarsch community.

performed and the incidence of leukaemia in adults was investigated for cross-check (Table I, Figs 2 and 3). Concern of the population in Elbmarsch was deep, and pub­ lic debates were intense.

2. PHYSICAL INVESTIGATIONS

In a population of 1500 children, about one child is likely to fall ill with leukaemia within about 15 years. Giving appropriate consideration to a case of con­ genital heart disease with repeated fluoroscopy and to a case of previous embryonic carcinoma with operation and chemotherapy, the remaining excess of about five cases means that a possibly causative dose would have to be 5 times the doubling dose for leukaemia, which is close to 100 mSv for children; however, 500 mSv is a dose that cannot escape detection. Since the legal limit for planning nuclear power stations ensures that 0.3 mSv for air or water emissions is not exceeded, Krümmel emissions have been principally limited to 0.05 mSv/a, and the radioactivities emitted do not exceed small fractions 3 6 4 POSTER PRESENTATIONS

FIG. 3. Leukaemia, lymphoma and myeloma in adults. Incidence study (1994) by W. Hoffmann and E. Greiser (Bremen). Area: three districts (Kreise) around Elbmarsch (hatched areas on map). Diagnoses: (1) acute leukaemia, (2) chronic myeloid leukaemia, (3) chronic lymphocytic leukaemia, (4) non-Hodgkin ’s lymphoma, (5) Hodgkin ’s disease, (6) mul­ tiple myeloma, (7) myeloproliferative syndrome. Subgroups: Seven diagnoses, two sexes, 39 communities = 546 subgroups. Reference: Age standardized average incidence in the three districts. Results: Statistically significant (p < 0.05) increments or decrements in 25 communi­ ties (points marked on map). Expected number o f false positive results is approximately 27. Significant increase o f chronic myeloid leukaemia in Geesthacht (near Elbmarsch). No clear dependence on distance from Krümmel nuclear power station. No causal relationship to be deduced. POSTER PRESENTATIONS 3 6 5 of the permitted values. Dose rate, aerosol and noble gas monitoring inside the plant as well as external dose rate monitoring (e.g. a remote survey system, 160 passive environmental dosimeters, and soil and water probing) ensures gapless monitoring. In three expert evaluations, one of which was obtained from a known anti- nuclear organization, it was established that emissions into the air have been safely below one tenth of the permitted values, and that there is confidence in the emer­ gency emission monitoring system. It was concluded that no dose of a magnitude relevant to leukaemia causation can be deduced.

3. TR ITIUM IN PLANTS?

In 1992, leukaemia causation by enhanced tritium emission from Krümmel was a favoured hypothesis. Tritium measurement in the wood of trees has been a well tried method since the atmospheric nuclear weapons tests. The first Elbmarsch trees evaluated in July 1992 using a tritium liquid scintillation counter gave values of about 15 and 35 Bq/kg, but the wood samples had been taken as averages over the trunk cross-section. This circumstance caused two groups to try autoradiography to demonstrate radioactivity accumulation in discrete annual rings of trunk cross-sections. In November 1992 they showed blackening patterns resembling the annual ring pat­ terns, and in spite of many warnings to first exclude artefacts such as chemography they immediately published these findings widely in the press. Public reaction was very strong, but only 4 weeks later the first annual rings were separately measured in the scintillation counter and it was shown that the rings’ tritium content was by many orders of magnitude too small to cause film blackening. The threshold for 1 week film exposure time was 200 MBq/kg. While the artefact discussion is still not settled, stable scintillation counter values of about 15 Bq/kg for trunk cross-sections are now regularly produced, and a small excess over values for control trees seems to reflect permitted tritium releases, probably with participation of the water path. A tritium programme, now based on measurement uncertainties of about 2 Bq/kg, with coded wood samples and with analysis of single annual rings, is still under way.

4. CONCLUSIONS

Unconventional techniques of environmental monitoring have therefore not given any evidence contradicting the clear results of physical monitoring as described in Table I. The possibility that further research will detect past exposures of the population sufficient to cause the Elbmarsch leukaemia cluster appears to be very small. This does not, however, mean that current research programmes should be 3 6 6 POSTER PRESENTATIONS stopped, and one of them is performance of a close survey of the doses that might have been contributed by the second nuclear facility near Elbmarsch, the GKSS research centre (see Fig. 1). But all this will not solve the basic problem of finding the true reason for child­ hood leukaemia clustering, which is a worldwide phenomenon not at all limited to the vicinity of nuclear facilities. The German Radiation Protection Commission has just finished an investigation into these questions. The role played by incidence fluc­ tuations due to pure chance has been analysed, and a Poisson fluctuation with a small but finite probability for what appear as clusters has turned up. The alternative ‘infection hypothesis’, based on accumulating clinical and epidemiological evidence, but still not disentangled from the complexities of cancer initiation, promotion and progression, has also been discussed [1].

REFERENCE

[1] R A D IA T IO N P R O T E C T IO N C O M M IS S IO N , Ionisierende Strahlung und Leukämie­ erkrankungen bei Kindern und Jugendlichen, Bundesanzeiger, No. 155 (1994). Case Study 3 RADON IN HOMES

IAEA-CN-54/3P

THE OPTIMIZATION OF COMMUNICATION AND DECISION MAKING IN RADON POLICIES

G.X. EGGERMONT Radiation Protection Office (VUB), Brussels

A.J. POFFUN Laboratory of Nuclear Physics (RUG), University of Ghent Belgium

1. INTRODUCTION

Enhanced radon exposure at home and at work is an important collective health risk. Prevention and remediation potentials exist. At local and regional levels, high cancer risk, normally considered as unacceptable, can occur. Risk management of this complex problem could integrate a differentiated communication strategy in optimization scenarios, taking into account risk perception.

2. RADON EXPOSURE AT WORK

The collective dose gathering for the European Commission (CEC) ALARA conference in 1993 [1] illustrates that the collective radon exposure for workers in the non-nuclear Polish mining industry, 1337 man • Sv/a [2], is higher than the provi­ sional, slightly underestimated dose, 534 man-Sv/a, to all nuclear industrial and medical workers in the European Communities as a whole. Extrapolation of the data for the United Kingdom, with its most complete registry for collective doses in both non-nuclear and nuclear industry and services (respectively 164 man-Sv and 116 man-Sv in 1987), indicates that approximately 50% more dose occurs in non-nuclear activities than in nuclear activities. This is essentially due to radon. Analysis of the phosphate industries in Belgium and generally in Europe [3] has clearly pointed out that the extent of the problem of exposure at work is not limited to mines. In Belgium no systematic studies have been done on radon exposure at work (mines, schools, phosphate industry, gypsum industry, nuclear and non-nuclear

3 6 9 3 7 0 POSTER PRESENTATIONS waste treatment and transport), except for some specific interventions in former radium industries. The ICRP and national authorities are looking for coherence in the application of ionizing radiation safety standards in all activities involving exposure at work, but they are confronted with a paradox [4]. Coherent regulation of the radon exposure problem at work could increase considerably the cost or economic value of radiation protection in the future, resulting however in more effective detriment reduction. The cost increase due to radiation protection against radon exposure in numerous cleanup activities, particularly in eastern Europe, could be a constraint for this evolution. The draft CEC directive [5], applying ICRP-60, does not yet extend the regula­ tion to these so called non-nuclear sectors. However, Article 44 of this directive requires the national authorities to identify working conditions where significant exposure can occur, related to materials containing natural nuclides. National authorities will have the possibility of intervention to regulate and to set up radon registries. The CEC is not yet proposing ALARA for these practices. The opportunities for improving safety at work through optimization of radon exposure in non-nuclear activities are higher than in the nuclear industry. Consider­ able dose reduction capacity exists and risk levels at work higher than 10~3/a could be eliminated. Traditional actors in risk management at work, the trades unions, having influence on risk communication, have already studied the problem [6], but are not yet giving priority to this problem in safety at work action programmes.

3. RADON EXPOSURE AT HOME

Radon is a dominant pollution factor in the microenvironment of human sheltering, being responsible for almost 50% of the annual dose to the general popu­ lation [7]. Indoor air quality in general and radon exposure particularly could be an important factor for future health, needing adequate hygienic measures. Since healthy housing is a priority in human needs, present technological opportunities could help in reducing indoor air exposure to radon. The objective of reaching natural outdoor activity concentration standards, as proposed by United States Environmental Protection Agency, should be considered as a long term goal (annual risk level < 10“6) in the development of sustainable sheltering techniques. In Belgium, the average indoor concentration in homes is about 50 Bq/m3, with generally lower values in the northern part (average 39 Bq/m3) than in the southern part (average 77 Bq/m3) of the country. Furthermore, in some 1000 houses, radon activity concentrations of more than 4000 Bq/m3 are expected to occur [8]. Exposure levels of this magnitude represent an annual risk level of 10~2 for the inhabitants. The latter value is normally considered unacceptable for mem­ bers of the public. POSTER PRESENTATIONS 3 7 1

Detailed measuring campaigns conducted over the last years indicate quite clearly that in some locations (Table I) a significant fraction of the population is liv­ ing in houses with levels of more than 400 Bq/m3 (the CEC action level). This cor­ responds to a risk level of 10'3 per year. At work a risk level of this order is the limit for authorized and regulated activities. Therefore, as a first step to a coherent approach, optimization was also applied to radon at home. Four scenarios were developed: three about remediation of situa­ tions with concentrations respectively >4000, 400 and 150 Bq/m3 (the EPA action level), and one about the prevention of levels of more than 50 Bq/m3. The results as presented in Table II indicate that they all may be considered as highly justified cost effective actions.

TABLE I. CHARACTERISTICS OF SOME TOWNS EXPOSED TO HIGH RADON LEVELS IN BELGIUM3

Mean Enhanced _ Area Homes H0Usi"g Mean > 400 effective collective stock 3. M a X n ; 3 0Ca 1 ^ (km2) investigated (%) (Bq/m) (Bq/m3) dose dose (mSv/a) (man-Sv/a)

Visé 29 160 3 90 8000 2 3.6(1.5) 27(11) Neufchât 113 250 11 200 4200 10 7.9(3.4) 36(15) a In the calculations 50 fiSv/a per Bq/m3 is used as the value for the conversion factor [9] and 0.8 as the value for the occupancy factor. The values between brackets represent the results based upon the conver­ sion coefficients proposed in ICRP-65 [10]. The notion of enhanced collective dose represents the frac­ tion to be considered as more than ‘normal’. The latter corresponds to an exposure level of 50 Bq/m3. This notion has to be compared to the collective dose reduction obtained as a result of the intensive optimization effort in the nuclear energy sector, where by steam generator replacement some 10 man • Sv could be saved.

TABLE П. COST EFFECTIVENESS ACTION SCENARIOS a

Total cost Dose saved a Scenario (106 US $) (man • Sv/a) (US $/man • Sv)

Mitigation > 4000 Bq/m3 3 520 135

Mitigation > 4 00 Bq/m3 27 1005 1 285

Mitigation > 150 Bq/m 3 267 2625 2 500

Prevention > 5 0 Bq/m3 200 156 30 000 a In the estimations the cost for mitigation and the cost of the large scale measuring campaign, necessary to identify the problem dwellings, have been taken into account. 372 POSTER PRESENTATIONS

4. RADON RISK ASSESSMENT, COMMUNICATION AND MANAGEMENT

Risk assessment of a societal hazard for purposes of risk management starts with quantification and hazard estimation, but cannot be limited to its physical consequences. Radon exposure occurs together with other hazards. The dominant smoking hazard makes the epidemiological risk estimate of indoor radon difficult. Considering only the radiological hazard, risk is usually represented by the product of the severity of the effect g and the probability of occurrence p. For radon exposure the major effect is lung cancer with early death as the magnitude. The probability is mainly derived from workers’ exposure data [10]. Leukaem ia may also be related to radon exposure; however, little evidence exists as to the probability. The so called ‘objective’ estimation of the radon lung cancer risk, g •p , contains judgements, conditioning its risk assessment. The dose conversion fac­ tor, being a first estimate of the related risk, is currently estimated by experts with a spread of a factor of 3. This should normally have ample influence on decision making in risk management, considering the overwhelming radiological risk for radon in non-accidental conditions. The Governmental advisory councils for science and radiation protection in Belgium [11] have recommended authorities to study and measure the hazard of radon and to optimize protection. As a result, optimization scenarios were developed considering costs and benefits. A multidisciplinary analysis of radon risk done for the Belgian Health Council, however, has pointed out that the qualitative dimension of risk perception and the social and cultural interaction should be taken into account and integrated in optimi­ zation scenarios. For purposes of risk management and risk communication, the notion of risk can be represented by:

R = f ( g , p, a)

a being a set of mostly qualitative attributes, characterizing the significance or value given by the concerned people subjected to the hazard. Different approaches to risk perception with a theoretical and em pirical scien­ tific base are analysis and structuring of this perception of reality [12]. The Royal Society and Sandman [13] have grouped attributes related to percep­ tion in a binary set of components, such as natural/industrial, controllable/ uncontrollable, familiar/exotic, old/new, responsible/non-responsible, voluntary/ involuntary. In several radon perception studies made for the EPA , Sandman notes low ‘outrage’ (characterizing subjective response) and an underestimation of the radon POSTER PRESENTATIONS 373 hazard, clearly contrasting with reactions to several nuclear energy hazards. Our experience shows that the radon hazard is perceived as natural and fam iliar, as con­ trollable and not coerced. It has always existed. A lack of outrage am plifying actors and minor scientific disagreement are noticed. Limited equity problems arise between industrial interests and charges put on the collectivity.

5. RADON MANAGEMENT PROPOSAL

Extension of the mentioned optimization study from doses and costs to percep­ tion attributes can be done by using multi-attribute decision aiding tools, allowing qualitative factors to be weighted. Considering the importance of relevant actors in risk communication (RC) and management (RM ), and the empirical fact [12] that hazard experts are less open to qualitative factors, representative groups should be associated in evaluating scenarios, in structuring and weighting attributes and in sensitivity analysis. W hile working out this complete RM approach, an intermediate optimization scheme was developed for the Health Policy Council, integrating R C. Differences in perception, related to the source term, the hazard level and the individual and col­ lective nature of the risk, were taken into account. The responsibility of various com ­ petent authorities is integrated into the optimization scheme at different levels. Based upon previous considerations, the following strategy is put forward. Radon management starts with a general information campaign in combination with a massive measuring programme. The proposed action scenarios are accompanied by different selected information carriers to the concerned public. An important role is attributed to local authorities and consumer organizations:

— <150 Bq/m3: comprehensible factual information sheets, offering risk quantifiyng opportunities and practical references; — 150-400 Bq/m3: information motivating people to reconsider their risk and perception, offering practical and technical guidance to decrease voluntarily their exposure levels; — 400-4000 Bq/m3: incentive information, stimulating people to act, to take responsibility for their own fam ily’s health as well as for collective conse­ quences, with logistic support of regional authorities, offering technical and financial assistance for intervention(s) to eliminate high individual risks and to reduce collective doses through local authority A LA R A planning; — >4000 Bq/m3: information requiring action, focusing on magnitude and high probability of risk with financial assistance. For these exposures Governmental authorities should intervene, requiring certification and constraining house selling and renting. 374 POSTER PRESENTATIONS

Parallel to this, radon prevention actions should be started. Training courses for architects and contractors should be organized, promoting radon resistant build­ ing techniques in new constructions and renovations. Priority should be given to highly exposed regions.

6. CONCLUSION

Radon is the most important exposure factor of ionizing radiation at work and in the environment. Protection can be based on A LA R A . Optimization techniques are offering m ultidisciplinary tools to integrate technical and social assessment. The proposed R C and RM approach for radon can be more effective than recent results in nuclear energy protection. Clear, coherent risk communication objectives, beyond actions such as reassurance and warning, are necessary to eliminate the radon paradox. Influencing perception could include outrage mitigation as well as amplification.

REFERENCES

[1] CROFT, J., Role of optimization in the management of workers’ exposure (Proc. 4th European Conference on Radiation Protection Optimization), EUR-15234 (1994) 21-72. [2] DOMANSKI, T., Radiological recognition and classification of coal mines — Criteria and recommendations for surveillance (Proc. 4th European Conference on Radiation Protection Optimization), EUR-15234 (1994) 109-120. [3] BAETSLÉ, L., Study of the Radionuclides Contained in Waste Produced by the Phos­ phate Industry and Their Impact on the Environment, CEC Nuclear Science and Tech­ nology Rep., EUR-13262 (1991). [4] EGGERMONT, G., POFFIJN, A ., Radon exposure standards: A paradox in radiation protection (Proc. Radon in Our Euregio), Liège, 1993, Special Issue of Ann. Assoc. Belg. Radioprot. (1993) 409-425. [5] Amended proposal for a council directive for basic safety standards for protection against ionizing radiation, CECCOM 349 (30 July 1993). [6] EGGERMONT, G., et al., A union and consumer proposal for the regulation of enhanced natural radiation, Sei. Total Environ. 45 (1985) 631-638. [7] UNITED NATIONS, Sources, Effects and Risks of Ionizing Radiation (Report to the General Assembly), Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), UN, New York (1988). [8] POFFUN, A ., et al., Radon in Belgium: The current situation and plans for the future (Proc. 1991 Symp. on Radon and Radon Reduction Technology), Philadelphia (1991). [9] VANMARCKE, H., et al., Radon versus Rn-daughters, Health Phys. 56 (1989) 229-231. POSTER PRESENTATIONS 375

[10] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Protec­ tion against Radon-222 at Home and at Work, Rep. No. 65, Pergamon Press, Oxford and New York (1992). [11] EGGERMONT, G., The recommendations of the Belgian National Science Policy Council on radon research (Proc. Conf. Radon et gaz rares dans les sciences de la terre et de l’environnement, Mons, 1990) 17-20. [12] UNITED KINGDOM ROYAL SOCIETY, Risk Analysis, Perception and Manage­ ment, Report of a Royal Society Study Group, London (1992). [13] SANDMAN, P., Hazard versus outrage: Responding to public concern about radiologi­ cal risk (Advanced Workshop on Occupational and Environmental Radiation Protec­ tion, Boston, 1994). 376 POSTER PRESENTATIONS

IAEA-CN -54/20P

RADON PROGRAMM E IN THE CZECH REPUBLIC

E . T Y L O V Á National Radon Commission, Ministry for the Environment, Prague, Czech Republic

1. LEGISLATION

The large radon programme on decreasing the concentrations of 222Rn and short lived decay products of 222Rn (radon progeny) in homes was started in 1990 and is co-ordinated by the National Radon Commission. Recommendations to decrease radon and radon progeny concentrations in Czech homes came to force in 1990 and concentration limits in 1991. The requested level for new constructions is 100 Bq/m3 and for existing houses is 200 Bq/m3 (the action level). By approving this recommendation the Government accepted the solving of radon related problems and compensation for the cost of risk reduction measures in existing houses from the national budget.

2. DETAILED SURVEY OF SOURCES OF RADON

The elevated indoor radon levels in many houses result from uranium-rich ground, use of 226Ra-rich building materials and the 222Rn-rich tap water. Derived maps of radon risk have been elaborated, which divide the country into three categories according to the degree of radon risk from the soil. The maps serve as a basis for decision m aking concerning further surveys, which w ill measure radon and radon progeny concentrations in buildings, and for measuring radon activity in the soil before new building [1]. The further survey made by the Centre for Radiation Hygiene of the State Health Institute is focused on measurement of radon and radon progeny in selected buildings and water supplies of the population in areas with high radon risk, and comparisons to required lim its are made. Special attention is paid to areas with very high radon risk (such as Jáchymov and Petrovice u Sedlcan). A standardized measurement method is used in most surveys: track detectors, one in every living room, are located for a period of 1 year. If the estimated average value is higher than the action level, people living in the house are offered a further POSTER PRESENTATIONS 377 measurement to find the sources of radon. More than 48 000 measurements have been made since 1991. The result of the survey is that more than 3% of inhabitants have been found to live in homes with an equilibrium equivalent concentration of radon (EER ) of more than 200 Bq/m3 (the action level) [2]. One of the sources of radon in living quarters is building materials. In order to prevent the use of unsuitable materials in new buildings it was necessary to ensure testing of building materials. Data on mass activity of 226Ra in building materials and in some light ashes are collected in the database R A STA .

3. ESTABLISHM ENT OF INSTITUTIONAL SUPPORT

The M inistry for the Environment has appointed the National Radon Com m is­ sion. Its members are representatives of m inistries and agencies participating in solv­ ing the radon problem and selected experts. The radon programme is effected in districts and towns by district authorities and magistrates of large cities, which closely co-operate with the Hygiene Service. District authorities and city magistrates co-ordinate the search for objects with high radon risk in their area and the taking of remedial measures and their financial sup­ port, co-ordinating their actions with the State radon programme. They also allocate funds for remedial measures to municipalities and individuals who apply for State contributions for realization of anti-radon measures.

4. REALIZATION OF REM EDIAL MEASURES

A set of remedial measures has been prepared for living quarters with a volume activity of radon higher than the allowed 200 Bq/m3, and for public water mains sources with a volume activity higher than 50 kBq/m3. Each individual project of remedial measures is based on results and recommendations of preceding radon diagnosis of the object. M ore than 95 % of the first applied remedial measures were successful, meaning that the level of radon decreased below the lim it (200 Bq/m 3). The failed projects are also analysed in order to find alternative solutions and to col­ lect sufficient data for elaboration of standard remedial procedures. Rem edial actions have been realized in more than 1300 houses, 130 schools and kindergartens and 180 water sources since 1991 [3].

5. SOCIAL ASPECTS OF THE RADON PROGRAMME

An information and educational programme is centred on communication with public, providing factual information about the radon programme, and the appropri- 378 POSTER PRESENTATIONS ate perception of radon risk (this perception is sometimes distorted in comparison with perception of risks from other activities) is an important part of the radon p r o g ra m m e .

It became clear that an active participation of inhabitants is important for the effective functioning of the radon programme. The degree of participation depends on appropriate perception of radon risk. Underestimation of risks leads to passivity, overestimation to unnecessary stress of exposed inhabitants. This is why the factors influencing risk perception are investigated. The most important factors are:

(1) the involuntary nature of the risk (frequently connected with the unavailability of improvements); (2) the moral value, when the risk is caused by negligence (usage of building materials with high activity of 226Ra); (3) the effects on children, connected with anxieties about families; (4) the keeping secret of information about the real situation which leads to overes­ timation of the problem and inappropriate perception of the danger.

Analysis of these factors serves as a basis for the decision making process. From the psychological point of view it is important that everybody who fears height­ ened radon exposure is offered measurements to check whether remediation is indeed re q u ire d . Also, the public and media participation has a great impact on the radon programme and attention should be paid to the provision of objective information without sensationalism.

6. CONCLUSION

The experiences of the radon programme have shown that it is possible to decrease radon concentration in most existing houses to below the lim it. The whole programme is at present being evaluated, with the aim being to reach the optimum utilization of available funds in order to achieve maximum remedial effects.

REFERENCES

[1] CZECH GEOLOGICAL SURVEY, Derived Maps of Radon Risk in the Czech Repub­ lic, 1:200 000, Prague (1991). [2] BURIAN, J., Results of Investigation of Houses with Higher Radon Risk, Appendix 5, Year Book of the Radon Programme (1994) 1-5. [3] ZATOCIL, J., Realization of Remedial Actions, Appendix 7, Year Book of the Radon Programme (1994) 1. POSTER PRESENTATIONS 379

IAEA-CN-54/31P

RADON IN LATVIA’S DW ELLINGS

M . D A M B I S Radiation and Nuclear Safety Inspection of the Environmental State Inspection of Latvia, Riga, Latvia

The indoor radon situation, before this joint Latvian-Sw edish project started, had essentially not been investigated. There generally was a vague concern about whether indoor radon is a problem in Latvia, although it is well known that radon is the dominant source of ionizing radiation dose for the ‘average’ inhabitant in most countries [1, 2].

1. RADIOGEOLOGICAL SITUATION

The territory of Latvia is completely covered with loose sedimentary rock. The prim ary rock lies from 300 m to 1.8 km deep. The upper part consists of Quaternary sediments. From a radiogeological viewpoint [3] we can divide these into three g r o u p s :

(1) sand and peat (uranium content 0.1-1.5 ppm); (2) rocky moraine loam and sandy loam (0.5-4 ppm); (3) clay and clayey gravel (2.3-3.5 ppm).

It is evident that in the upper layers the uranium content is low. According to the classification given in Ref. [4], most of Latvia is a low risk area; only the places with uranium content above 2.8 ppm correspond to a normal risk area. The limno- glacial clay sediments of the Baltic suite occupy large areas in Latgale-Vidzem e, as well as in Kurzem e, but only in clays in Kurzem e is there a relatively enhanced con­ tent of uranium and thorium (in the Kurzeme heights). This corresponds very well with data given in Ref. [5]. In other parts of Latvia there are some local places with increased uranium content [4], especially near Cèsis in a building material pit (clay for bricks), where the effective activity is up to 3000 Bq/kg (it is permitted to use in civil construction only materials whose activity does not exceed 370 Bq/kg).

2. METHOD FOR RANDOM CHOICE OF HOUSES

The total number of detached houses in Latvia is 195 995, of which 32 235 are situated in the six biggest cities, the other 163 760 in the countryside and in small 380 POSTER PRESENTATIONS

TABLE I. AVERAGE RADON CONCENTRATION IN RANDOMLY SELECTED DETACHED HOUSES IN LATVIA

Radon Number of Number of small houses District concentration random (Bq/m3) points Total In country

Aizkraukle 55.9 6 5 114 3 356 Alúksne 41.8 7 4 471 3 654 Balvi 16.1 11 7 079 6 145 Bauska 98.25 8 4 832 4 327 Cèsis 82.8 9 6 217 4 943 Daugavpils 58.6 14 10 101 9 509 Dobele 82.4 6 3 519 2 308 Gulbene 52.4 7 3 875 3 056 Jelgava 38.9 7 4 102 3 643 Jèkabpils 91.9 10 7 626 4 915 Krâslava 86.0 12 9 385 7 822 Kuldïga 100.7 7 4 452 3 450 Liepâja 86.2 12 7 685 6 201 Limbazi 30.2 7 4 627 3 557 Ludza 81.2 15 10 287 8 412 Madona 60.2 10 6 197 4 731 Ogre 27.6 6 6 343 3 297 Preili 60.3 10 9 026 7 514 Rèzekne 95.0 13 10 437 10 014 Rïga 38.3 14 9 977 8 128 Saldus 90.1 7 3 750 2 588 Talsu 126.7 9 6 221 4 860 Tukuma 65.9 9 6 326 4 464 Valkas 35.7 6 3 963 2 828 Valmieras 76.5 8 5 852 3 405 Ventspils 87.5 6 2 296 2 193

Total 68.5 236 163 760 129 320 POSTER PRESENTATIONS 381 towns. For selection of random houses we had a problem, because we could not use the Swedish experience in this field — more than half of our houses do not have a telephone and there is no computerized house register. Therefore, the method we chose was the following: a computer was used to find on the map of Latvia random points at a density proportional to the number of sm all houses in each region (aproxi- mately one point for 780 houses was chosen). This caused later complications in the placement of detectors, because we had no opportunity to notify the people living in these houses. W e did not generally know if at the place where the computer put a random point a house is located. Therefore, an additional two random points in each district were chosen. W e tried to find a house as close as possible to a random point on the map of Latvia, but sometimes it was impossible, because the point was situated in the middle of a big lake or swamp.

3. INSTRUMENTS FOR RADON MEASUREMENT

The radon measurements were made with the E-PER M system [6, 7], consist­ ing of 60 standard 200 m l ionization chambers with an on/off mechanism (‘S ’ cham­ ber), 40 short term electrets of high sensitivity and 20 long term electrets, and two SPER-1 electret readers for measuring the surface potential of an electret, and two sets of reference electrets (each consisting of two superstable, long term electrets). These instruments belong to the Swedish Radiation Protection Institute (SSI) and were kindly lent to Latvia for this investigation. More than U S $200 of SSI money was used for the purchase of car fuel to place the electrets in selected houses.

4 . R E S U L T S

The results of the study are given in Table I. During this investigation about 22 000 km was travelled by car and the radon concentration in about 300 detached houses was measured. Table I shows only the radon concentration measurements in randomly selected detached houses in the countryside (the situation in the six biggest cities is not discussed in this paper). The average indoor radon concentration in detached houses is estimated as 68.5 Bq/m3, but this level differs rather strongly in different districts. The average level in Balvi District (in the eastern part of the country) is less than 20 Bq/m3, but it is more than 120 in Talsi District (in the western part of the country — Kurzeme). The North Kurzeme height is situated in Talsi District, and accordingly there is the greatest number of radioactive anomalies both in rock and in groundwater (the highest uranium content is given as 4 .2 X 10“4 g/L) [5]. The situation in Liepâja District (the biggest district in Latvia, in the southwestern part of the country) is very interesting. The estimated average level is 86.2 Bq/m3, but if we separately consider the West Kurzem e height and the 382 POSTER PRESENTATIONS coastal lowland (in each of these two areas were distributed 6 random points), then the radon levels are 29.7 Bq/m3 and 142.7 Bq/m3, respectively. So we can con­ clude that the heights in Kurzem e are regions in Latvia where the radon concentra­ tion can exceed 200 Bq/m3 comparatively often. To avoid additional health risk, supplementary investigations should be made there. On the other hand, this investigation showed that high radon concentrations are mostly found in houses having limited air exchange with outdoor air due to an addi­ tional brick layer around a wooden frame (such a house is ‘Skapari’ in Cèsu District, where the average radon concentration during 4 days exceeded 1500 Bq/m3, but this house was not randomly selected). Houses of such type we find everywhere in Latvia. Therefore, we conclude that the building regulations must be revised.

REFERENCES

[1] UNITED NATIONS ENVIRONMENT PROGRAMME, Radiation, Doses, Effects, Risks, UNEP (1985). [2] WORLD HEALTH ORGANIZATION, Air Quality Guidelines for Europe (WHO Regional Publications European Series No. 23), WHO, World Health Organization, Regional Office for Europe (1987). [3] Report about Results of Specific Geophysical Investigations in Territory of Latvia, Leningrad (1991) (in Russian). [4] ÀKERBLOM, G., “ Investigation and mapping of radon risk areas” (Proc. Int. Symp. on Geology for Environmental Planning, Trondheim, 1986). [5] STRAUME, J., BERNA, I., The Natural Radioactivity of Rocks and Groundwater in Latvia, Riga (1993) 38pp (in Latvian). [6] KOTRAPPA, P., et al., An electret passive environmental Rn-222 monitor based on ionization measurements, Health Phys. 5 4 1 (1988) 47. [7] KOTRAPPA, P., et al., A practical E-PERM (electret passive environmental radon monitor) system for indoor 222Rn measurement, Health Phys. 58 4 (1990) 461. POSTER PRESENTATIONS 383

IAEA-CN-54/36P

AN EXAM PLE OF ELEVATED INDOOR RADON CONCENTRATION IN A KINDERGARTEN IN SLOVENIA

J. VAUPOTIC J. Stefan Institute

I. KOBAL, P. JOVANOVIC Institute of Occupational Safety

Ljubljana, Slovenia

J. PLANINIC Faculty of Education, JJ . Strossmayer University, Osijek, Croatia

1. INTRODUCTION

Four years ago, Slovenia started a national radon programme. So far, radon concentrations in all the kindergartens, elementary schools, and high schools and in 1000 randomly selected flats have been measured. In the survey of indoor radon in kindergartens (730 buildings) [1, 2] and schools (888 buildings), instantaneous air concentrations were obtained under closed conditions in wintertime by use of modi­ fied Lucas cells [3]. In 69% of the buildings the values were below 100 Bq/m3 and in 33 buildings (2%) they exceeded 1000 Bq/m3. Geometrical means of 58 and 73 Bq/m3 were obtained in kindergartens and schools, respectively. The kinder­ gartens with more than 1000 Bq/m 3 were further investigated. In two of them, with winter average radon concentrations of about 1500 and 2000 Bq/m3, conditions were ameliorated and the radon level was substantially reduced. One example of an elevated concentration is a kindergarten situated in the Karst Region of Slovenia with an instantaneous radon concentration of about 2 400 Bq/m3 obtained early in the morning in wintertime in a room closed for 12 hours prior to sampling. The kindergarten is a single storey building, 10 years old, with four play rooms for about 50 children aged 1-2 years. Children usually arrive at the kindergarten between 7 and 8 a.m. and spend 6-8 hours there. The building was thoroughly examined by use of the following measuring techniques: etched track detectors, charcoal adsorption detectors, alpha spectroscopy (EDA/W LM -30 alpha spectrometer) and an ionization chamber counter (Genitron Instruments/Alpha GUARD). 384 POSTER PRESENTATIONS

FIG. 1. Daily variations of indoor radon concentrations in wintertime.

J u ly 1994 FIG. 2. Daily variations of indoor radon concentrations in summertime. POSTER PRESENTATIONS 385

2. RESULTS AND DISCUSSION

So far, track etch detectors [4, 5] have been exposed four times for 3 months (September 1993 to November 1993, December 1993 to February 1994, M arch 1994 to M ay 1994, June 1994 to August 1994) in three play rooms and twice for 6 months (M arch 1993 to August 1993, September 1993 to February 1994) in one play room. The 3 month winter average concentrations ranged from 490 to 660 Bq/m3 and were about 35 % higher than those obtained in autumn and about 25 % higher than the spring ones. Six month average indoor radon concentrations in the spring- summer period were 490 ± 22 Bq/m3 with an equilibrium factor of 0.2 [5] and 710 + 30 Bq/m3 in the autumn-winter period with the same equilibrium factor. The equilibrium factor is relatively low, and could be explained by the clim atic con­ ditions. Six month radon measurements were also made in the home of one of the children. A yearly indoor radon concentration of 130 ± 10 Bq/m3 with an equilibrium factor of 0.55 was obtained [5]. Even though a track etch detector is a good indicator of the average indoor radon concentration in a home, it could overestimate the exposure in the kinder­ garten, because the rooms are closed overnight and on weekends. In order to see daily variations and to determine the actual radon and radon progeny concentrations to which the children and the personnel are exposed in the kindergarten, it is neces­ sary to measure the concentrations of radon and radon progeny continuously during the working time. In our case, radon and radon progeny concentrations were fol­ lowed over a period of 1-2 weeks in each season of the year in the room with the highest indoor radon concentration. A s seen from Figs 1 and 2, the high early morn­ ing concentration dropped substantially during the working day and the air quality conditions are probably much better than expected on the basis of either instantane­ ous (alpha scintillation cell) or average (track etch) concentrations. For an accurate evaluation of the radon dose it is necessary to know the radon progeny concentrations and equilibrium factor values during a working day. These measurements have not yet been finished, and thus the radon dose received in the kindergarten was estimated on the basis of the 6 month average concentrations (given at the beginning of this section). It was assumed that the child is 1 year old, spends 5 hours per day in a play room during the spring-summer (500 hours per year) and 6 hours during the autumn-winter (600 hours per year). It was also assumed that the child spends half of its time in the kindergarten sleeping and the other half in light exercise, and thus a dose conversion factor of 9.06 m Gy per W LM (for rest) for radon progeny with A M TD = 0.15 /¿m was taken as an average [6]. The annual dose equivalent received by the basal cell nuclei in the bronchi of this 1 year old child was estimated to be 39 m Sv. An effective dose equivalent of 2.6 mSv/a was calcu­ lated assuming a dose ratio H p/Hb of 1/8 [7]. In addition, for the same 1 year old child the effective dose equivalent received in the home was calculated [7]. Using an equilibrium equivalent radon concentration 386 POSTER PRESENTATIONS of 71.5 Bq/m3 (annual average indoor radon concentration of 130 Bq/m3 and an equilibrium factor of 0.55) at home, an equilibrium equivalent radon concentration of 23 Bq/m 3 (annual average indoor radon concentration in Slovenian homes) [8, 9] indoors other than in the kindergarten or at home, 9 Bq/m3 [8] outdoors, and the dose in the kindergarten of 2.6 mSv/a, an effective dose equivalent of 9.5 m Sv/a was estimated. The result shows that the home average radon concentration is also rela­ tively high, and the child receives about 69% of the annual dose at home and 27% in the kindergarten. But following the same procedure and taking the annual average indoor radon concentration as that for Slovenian homes (23 Bq/m3) [8,9], this child would receive an effective dose equivalent of 5 mSv/a, with a 53% contribution from the kindergarten.

3. CONCLUSIONS

The results presented here show that exposure to radon progeny in the kinder­ garten investigated is relatively high. The calculated annual effective dose in the kin­ dergarten of 2.6 m Sv is higher than those reported in the literature [10] for areas of normal background. The results of continuous measurements of radon and radon progeny during working days will show whether the dose, calculated on the basis of the 6 month average radon concentrations and equilibrium factors, is overesti­ mated. Until then, the personnel of the kindergarten have been advised on appro­ priate ventilation (opening windows) of the play rooms, especially in the morning and in the wintertime.

REFERENCES

[1] VAUPOTIC, J., KRIZMAN, M., PLANINIC, J., PEZDIC, J., ADAMIC, K., STEGNAR, P., KOBAL, I., Systematic indoor radon and gamma measurements in kindergartens and play schools in Slovenia, Health Phys. 66 (1994) 550. [2] VAUPOTIC, J., KRIZMAN, M., PLANINIC, J., KOBAL, I., Radon level reduction in two kindergartens in Slovenia, Health Phys. 66 (1994) 568. [3] VAUPOTIC, J., ANICK, M., SKOFLJANEC, M., KOBAL, I., Alpha scintillation cell for direct measurement of indoor radon. J. Environ. Sei. Health. A27 (6) (1992) 1535. [4] URBAN, M., SCHMITZ, J., “ The KfK passive personal dosemeter for exposure to radon and external gamma-radiation” , (7th Int. Congr. Int. Radiation Protection Assoc., Sydney, April, 1988, IRPA 7. [5] PLANINIC, J., FAJ, Z., The equilibrium factor F between radon and its daughters, Nucl. Instr. Meth. A278 (1989) 550. [6] NATIONAL ACADEM Y OF SCIENCES, Comparative Dosimetry of Radon in Homes and Mines, National Academy Press, Washington, DC (1991). POSTER PRESENTATIONS 387

[7] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Lung Cancer Risk from Indoor Radon Exposures to Radon Daughters, ICRP Publication 50 (1987). [8] KRIZMAN, M ., MLJAC, L., Indoor Radon Concentrations in Slovenia: Methodology of Dose Assessment (US Rep. No. 7074), J. Stefan Institute, Ljubljana, Slovenia (1994). [9] HUMAR, M ., SKVARC, J., ILIC, R., Indoor Radon Concentrations in Slovenia (US Report No. 6977), J. Stefan Institute, Ljubljana, Slovenia (1994). [10] UNITED NATIONS, Sources and Effects of Ionizing Radiation (Report to the General Assembly with Scientific Annexes), United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR), UN, New York (1993). 388 POSTER PRESENTATIONS

IAEA-CN-S4/38P

UNCERTAINTIES IN EXPOSURES TO RADON PROGENY: THEIR IM PACT ON THE EVALUATION OF RISK ESTIM ATES

P. DUPORT, R. JANICA Atomic Energy Control Board, Ottawa, Ontario, Canada

Evaluation of the risk of lung cancer due to exposure to radon progeny, in underground miners and the general population, is based on risk assessments derived from epidemiological studies in mine populations [1]. In addition, some of the risk factors associated with other inhaled alpha emitters (e.g. transuranic), because of lack of information, are also based on the assumed effects of radon and its progeny [2]. The risk factor is derived from the amount of lung cancer in populations of underground miners who are known to have been exposed to radon progeny and from the exposure received by the individuals who make up these populations. In general, data on lung cancer incidence among miners are good, but many uncertain­ ties make exposure data less reliable. Therefore, the reliability of risk estimates is highly dependent on the reliability of exposure data. In this study, uncertainties in exposures received by individual workers were evaluated and compared, for two methods of measurement: grab sampling (GS) and integrating personal dosimetry (use of a personal alpha dosimeter, or PAD ). Uncertainties were evaluated for a hypothetical but realistic worker analytically and by means of Monte Carlo simulation.

1. ANALYTICAL DETERMINATION OF UNCERTAINTIES IN ANNUAL EXPOSURES OBTAINED FROM GRAB SAMPLING MEASUREM ENTS

When determined by means of GS measurements, annual exposures result from estimations of exposure times at work places and in travelway s, combined with short duration, generally infrequent working level (W L) measurements (Eq. (1)). (One W L is any combination of the short lived decay products of radon in 1 litre of air that w ill result in the ultimate emission by them of 1.3 X 10~5 M eV of alpha ray energy.) Uncertainties in exposures to radon progeny have been evaluated for a hypothetical but realistic worker who worked, on average, every month of the year equal amounts of time in four stopes, and travelled equal amounts of time to and from each stope. For the purpose of uncertainty evaluation, it has been assumed that W L values in the different stopes are such that the annual exposure would be about 1 W LM (an exposure of 1 W LM results from the product of W L POSTER PRESENTATIONS 389 and duration of exposure, normalized to a 1 month exposure basis), that the coeffi­ cient of variability C V = a(x)/x in W L values measured at the stopes is 0.30 (larger than С V s generally observed in real mine operations), and that the C V in travel and work time in stopes is also 0.30. The classical theory of propagation of errors was used to determine the uncertainty in the radon progeny exposure (Eq. (2)):

ERn>m = £ W LxTk + t WL-T (1) k=l

c \ 2 / с \ 2 0.5 WLk \ , / ST, ‘T 2kW L^ + 1 - ^ + W L | t 2 WL T k SEft ■k=l (2) -'R n.m £ W LkTk + tWLr

k = l where W Lk and T k are the monthly averages of the W L values and of the time spent in stope k, respectively, and W Lj and t are the monthly averages of the W L values and of the time spent in travel ways, respectively.

2. ANALYTICAL DETERMINATION OF UNCERTAINTIES IN EXPOSURES OBTAINED FROM PERSONAL DOSIM ETRY

The PA D on which this work is based has been described elsewhere [3, 4]. Using PAD measurements, exposure to radon progeny is obtained from Eq. (3). With PADs, there is no need to make assumptions regarding the distribution of instantaneous W L values at the work place and uncertainties in time spent at work places or in travelways. The classical theory of propagation of errors was used to determine the uncertainty in the annual exposure (Eq. (4)):

7.69 n2 + 6(nt - 0.5 n3) -'Rn.rn (3 ) ( 1 .3 x 1 0 5) 1 7 0 e f E Q

0.5 6 2n ? 7 .6 9 n 2 sn2 ^ERnm _ [6nx + 7.69n2- 3n3] [6П] + 7.69n2- 3n3] \n2 ERn.m 3 2n 2 s n3 [6щ + 7 .6 9 n 2 - 3 n 3] \ n 3 +<ïKtm+(t

(4) 390 POSTER PRESENTATIONS where n¡ are the number of alpha tracks in the different areas of the detector, e, E , f, and Q are the geometric efficiency of the detection assembly, the aerosol collection efficiency of the PA D , the ratio of the counted area to the total irradiated area of the detector, and the pump flow rate, respectively. The coefficients of variability in the parameters used in Eq. (3) were also taken from real life observations [5], i.e. CV(n,) = CV(n2) = CV(n3) = 0.10; CV(e) = 0.01; CV(E) = 0.03; CV (F) = 0.035; CV(Q ) = 0.05. Conservatively, the assumption has been made that the errors in the number of alpha particle tracks in the PA D detector result in a Gaussian distribution of track number, instead of the narrower Poisson distribution that would result from the random process of radioactive decay alone.

3. DETERM INATION OF UNCERTAINTIES IN EXPOSURES BY MEANS OF MONTE CARLO SIM ULATION (GS)

The probability distribution of exposures to radon progeny received by the hypothetical worker was determined by Monte Carlo simulations. In each run of the simulation, monthly exposures were calculated using Eq. (1). The assumed distribu­ tions of W L values at work places and in travel ways, as well as С Vs in these values, were the same as in the analytical determination of uncertainties, but the distribution of the time spent in travelways was assumed to be uniform, within the range 0.5-1.00 h. The time spent in stopes was 8 h minus time in travelways.

4. DETERM INATION OF UNCERTAINTIES IN EXPOSURES BY M EANS OF MONTE CARLO SIM ULATION (PAD)

The uncertainty in the exposure obtained from the use of personal monitors was also determined by Monte Carlo simulations (Eq. (3)). The assumed numbers of tracks in the PAD detector, as well as the other PAD parameters from which exposures were calculated, were taken from real life cases.

5. COMPARISON OF THEORETICAL UNCERTAINTIES IN ANNUAL EXPOSURES FROM GS AND PAD W ITH ACTUAL OBSERVATIONS

The annual exposure received by the hypothetical worker is the sum, over 12 months, of the monthly exposures, and the coefficient of variation in the annual exposure is 1/VT2 times that in monthly exposures. Theoretical uncertainties in annual exposures obtained analytically and by Monte Carlo simulations are given in Table I. In view of Table I, personal monitor­ ing provides a more precise estimation of the annual exposure received by a uranium POSTER PRESENTATIONS 391

TABLE I. THEORETICAL UNCERTAINTY IN THE ANNUAL EXPOSURE (GRAB SAM PLING AND PERSONAL DOSIM ETRY) RECEIVED BY A HYPOTHETICAL MINER, DETERM INED ANALYTICALLY AND BY MONTE CARLO SIM ULATION

Annual exposure Coefficient of Dosimetry method Uncertainty analysis (WLM) variability (%)

GS Analytical 1.17 6.1 Monte Carlo 1.19 6.9

PAD Analytical 1.17 3.3 Monte Carlo 1.20 3.0

TABLE II. SOURCES OF KNOWN AND UNKNOWN ERRORS IN GRAB SAMPLING AND PERSONAL DOSIM ETRY

Sources of error Personal dosimetry GS dosimetry

Instrumentation and procedures Known Known Spatial variability of radon Unknown No error progeny concentration

Temporal variability of radon Unknown No error progeny concentration

Time spent at work places Unknown No error

worker than grab sampling (CV of about 3% for PAD and 6% for GS) but both methods are, in theory, quite precise. Therefore, when they are used simultaneously methods are, in theory, quite precise. Therefore, when they are used simultaneously on the same workers, exposure values provided by either method should be in close agreement. In fact, in the few instances where both methods were used simultane­ ously, there was no agreement between G S and PAD individual exposures [6, 7]. The observed lack of agreement could be explained by comparing which sources of errors are known or unknown in each method (Table II). In both methods, instrumen­ tal errors can be, and have been, evaluated. However, in personal dosimetry, spatial and temporal variability of the concentration of radon progeny, as well as the estima­ tion of the time spent at various places by a worker, is not a cause of uncertainty because the continuous sampling instrument is constantly worn by the worker and 392 POSTER PRESENTATIONS

it integrates, in real time, all elemental exposures. Conversely, uncertainties due to these parameters are not known in GS dosimetry. Therefore, it is reasonable to postulate that G S dosimetry is, inherently, a less reliable method for determining exposures to radon progeny than personal dosimetry.

6. PRACTICAL IMPLICATIONS

A major implication of the above observations is that risk estimates derived from worker populations whose exposures were determined by G S methods may not be quite reliable and that, consequently, it would be prudent to keep in mind the likelihood of large and unknown uncertainties in exposures when interpreting risk estimates obtained from epidemiology. To illustrate this last point, the ratio between the lowest and highest excess relative risk per unit exposure reported in 11 major cohorts of miners exposed to radon progeny is 1:31 [8]. From a regulatory standpoint, uncertainties in estimated individual exposures to radon progeny are not critical so long as the median total effective dose of a group of workers does not exceed about one third of the dose lim it, because there is only a small probability that a member of such a group may reach the dose limit. However, it must be remembered that in uranium mines radon progeny exposure contributes only a fraction of the total effective radiation dose delivered to the lung by the combination of inhaled radon progeny and radioactive long lived dust, and by gamma radiation. A ll components of the total effective dose being affected by uncertainties, the uncertainty in radon progeny exposures is only one of several sources of uncertainty that affect the total effective dose received by a worker. Site specific routine monitoring procedures, based on the results of surveys from which the distribution of individual exposures in various job categories (and the uncertain­ ties attached to them) could be determined, would, in the future, allow the determina­ tion of individual exposures with reduced uncertainties from which more reliable risk factors could be derived.

REFERENCES

[1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Protec­ tion Against Radon-222 at Home and at Work, ICRP Publication 65 (1993). [2] COMMITTEE ON THE BIOLOGICAL EFFECTS OF IONIZING RADIATIONS, Health Risks of Radon and other Internally Deposited Alpha-Emitters — BEIR IV, National Research Council (1988). [3] DUPORT, P.J., MADELAINE, G., ZETTWOOG, P., PINEAU, J.F., “ Enregistre­ ment des rayonnements alpha dans le dosimètre individuel et le dosimètre de site du Commissariat à l ’Energie Atomique” (10th Int. Conf. on Solid State Nuclear Tracks Detectors, Lyon, France, 1979 (FRANÇOIS, H., et al., Eds). POSTER PRESENTATIONS 393

[4] BERNHARD, S., PINEAU, J.F., RANNOU, A., ZETTWOOG, P., 1983: “One year of individual dosimetry in French mines” (Proc. Int. Conf. on Occupational Radiation Safety in Mining) Toronto, Canada, 1984 (STOCKER, H., Ed.). [5] DUPORT, P.J., “ Errors and uncertainties in air monitoring measurements and dose determination in uranium mines and mills (Int. Conf. on Radiation Safety in Mining, Saskatoon, Sask., Canada, 1992). [6] PIECHOWSKI, W., LE GAC, J., BRENOT, J., NÉNOT, J.C., ZETTWOOG, P., “ Exposure to short lived radon daughters: Comparison of individual and ambient monitoring in a French uranium mine” (Proc. Int. Conf. on Radiation Hazards in Mining: Control, Measurement, and Medical Aspects, Golden, CO, 1981) (GOMEZ, M., Ed.). [7] DUPORT, P., STOCKER, H., DALKOWSKI, E. Implications of dose distribution on monitoring requirements in uranium mines, Health Phys. 55 2 (1988) 407. [8] LUBIN, J.H., BOICE, D.J., HORNUNG, R.W., EDLING, C., HOWE, G.R., KUNZ, E., KUSIAK, R.A., MORRISSON, H.I., RADFORD, E.P., SAMET, J.M., TIRMARCHE, M., WOODWARD, A., XIANG, Y.S., PIERCE, D.A., Radon and Lung Cancer Risk: A Joint Analysis of 11 Underground Miners Studies, National Cancer Institute (1994). 394 POSTER PRESENTATIONS

IAEA-CN-54/87P

M ESURE DU RADON DANS LES HABITATIONS EN FRANCE

M.H. EL JAM M AL, J.P. GAMBARD Institut de protection et de sûreté nucléaire (IPSN ), F-92665 Fontenay-aux-Roses Cedex

J . C A R M E S Direction générale de la santé, F-75350 Paris 07 SP

F r a n c e

L ’Institut de protection et de sûreté nucléaire (IPSN ) mène, en collaboration avec le Ministère de la santé, une campagne nationale de mesure de l’exposition domestique au radon sur l’ensemble du territoire français. Cette campagne a pour but de connaître la concentration de radon à l’intérieur des habitations ainsi que sa répartition par département et par type de logements. Les informations collectées vont permettre de mieux évaluer l’impact sanitaire de l’exposition au radon de la population française et de mesurer l’efficacité des actions que chacun peut prendre pour limiter la présence de ce gaz.

ORGANISATION DE LA CAMPAGNE

La sélection des points de mesure dans un département, réalisée par l’IPSN , s’appuie sur un ou plusieurs critères. Concernant la population, il y a environ un point de mesure pour 1 500 habitants; pratiquement, il existe un point de mesure dans chaque commune dépassant 500 habitants en zone rurale et dépassant 1 000 à 1 500 habitants en zone urbaine. Au plan géographique, la couverture du territoire ne doit pas présenter une trop grande hétérogénéité (c’est-à-dire trop de grandes zones sans mesure); un nombre moyen de 200 à 250 points de mesure est considéré comme suffisant pour couvrir un département. Concernant l’habitat, seul le logement privé est concerné par la campagne de mesure. Le dosimètre utilisé, KO D ALPH A, est un film capable d’enregistrer la concentration du gaz présent dans l’habitation. Il doit être exposé pendant environ deux mois. La fourniture et la lecture du dosimètre sont faites par l’IPSN. Un POSTER PRESENTATIONS 395

' ’-л ' ii i i Campagne à réaliser Ш Ш Résultats depuis 1982 Е Ш Résultats attendus pour 1994

FIG. 1. Cartographie radon: moyenne en Bq/m3 des mesures effectuées.

questionnaire décrivant les conditions de la pose et le type d’habitat accompagne le d o s im è tre . Les ingénieurs sanitaires du Ministère de la santé réalisent l’enquête dans les départements: sélection des habitations à mesurer, information des élus locaux et des particuliers, pose et recueil des dosimètres et questionnaires, et, enfin, communica­ tion des résultats aux particuliers.

RESULTATS

Les départements a priori les plus concernés, sur la base de critères géolo­ giques, ont été choisis en premier dès 1982. A ce jour, plus de 4 200 mesures ont 396 POSTER PRESENTATIONS

été effectuées dans 46 départements. Certaines habitations (en général, des rési­ dences secondaires peu aérées) peuvent avoir des concentrations en radon très supérieures. Dans cinq départements, la moyenne des valeurs relevées dépasse 150 becquerels par mètre cube (Bq/m3): il s’agit de la Corrèze, de la Creuse, du Finistère, de la Loire et de la Haute-Vienne. La valeur maximale (4 787 Bq/m3) a été enregistrée dans ce dernier département. L ’augmentation de la concentration du radon dans les maisons est généralement due à une teneur plus élevée d’uranium dans le sol (régions granitiques). Une analyse des données a montré l’influence de la nature du sol, du type de logement (ancien ou moderne) et de l’étage où la mesure a été effectuée. On peut calculer une «moyenne par habitant» en pondérant les moyennes des départements par leur nombre d’habitants. Elle est, à ce stade de l’enquête, de 61 Bq/m3 d’air. Sur la base des données admises actuellement au plan international, il en résulte une dose individuelle annuelle par inhalation de radon de l’ordre de 1,5 millisievert (mSv). L ’exposition moyenne a été estimée à 3,7 m Sv par an en Suède et à 1,3 en Grande-Bretagne. POSTER PRESENTATIONS 397

IAEA-CN-54/88P

RADON ET RISQUE DE CANCER: ETUDES EPIDEM IOLOGIQUES APRES EXPOSITION PROFESSIONNELLE OU DOM ESTIQUE

M. TIRM ARCHE Institut de protection et de sûreté nucléaire (IPSN ), Paris, France

L ’évaluation du risque de cancer après exposition au radon est basée essen­ tiellement sur les résultats des études épidémiologiques des mineurs, notamment des mineurs d’uranium. Une étude sur les mineurs d’uranium français, ayant commencé à travailler au fond avant 1972, a permis d’étudier le risque de cancer lié à des expositions cumulées faibles, comparativement à celles des études américaines et tchèques, et étalées sur une longe période d’exposition. Les résultats de cette étude démontrent une augmentation du risque de décès par cancer du poumon et par cancer du larynx comparativement à la population française; une relation linéaire entre l’ex­ position au radon et l’excès de risque relatif par unité d’exposition est décrite pour les décès par cancer du poumon (régression poissonnienne). Cette étude fait partie d’une analyse conjointe menée sur 11 cohortes de mineurs, afin d’évaluer de façon précise les différents facteurs pouvant influencer la relation dose-effet radon-décès par cancer du poumon; les facteurs étudiés sont: l’âge depuis la première exposition, l’âge atteint, le temps depuis l’exposition, la manière dont cette exposition est cumulée dans le temps, le tabagisme. L ’analyse est réalisée sous la conduite du National Cancer Institute (Etats-Unis d’Amérique); parallèlement, sur ces mêmes cohortes de mineurs, les causes de décès autres que le cancer du poumon sont étudiées; ceci afin de vérifier si certaines causes, en excès de façon non significative au niveau des études individuelles, pourraient être en rela­ tion avec le radon (Données analysées par l’université d’Oxford). L ’extrapolation du risque à partir d’études en milieu professionnel pour évaluer le risque au niveau de la population générale, exposée à l’inhalation du radon dans les habitations, demande quelques remarques: l’exposition radiologique des mineurs est en fait multiple: le radon et ses descendants, l’irradiation externe gamma et les poussières d’uranium à vie longue ainsi que d’autres produits, présents dans les mines, mais absents dans les habitations, peuvent agir sur le risque de cancer du poumon. Il est difficile de donner une estimation précise du risque de cancer lié au radon pour des non-fumeurs ainsi que pour la population féminine à partir des données des mineurs. 398 POSTER PRESENTATIONS

Une étude cas-témoin internationale, coordonnée par la Com m ission des Com ­ munautés européennes, est actuellement conduite en France, afin d’évaluer le risque de cancer du poumon en fonction de la concentration du radon dans les habitations. Ces études permettent de vérifier sur le terrain l’évaluation du risque estimé à partir de données en milieu professionnel, en tenant compte notamment de la consomma­ tion tabagique. En effet, l’extrapolation du risque au niveau d’une population exposée au niveau domestique est fortement tributaire de l’interaction radon-tabac. POSTER PRESENTATIONS 399

IAEA-CN-54/99P

COUNTERMEASURES AGAINST EXCESSIVELY ENHANCED LEVELS OF RADON IN HOM ES IN M INING AREAS

R. CZARW INSKI, W. KRAUS, R. LEHM ANN, W. RÖHNSCH Federal Office for Radiation Protection, Berlin, Germany

1. INTRODUCTION

The remediation of houses with high indoor radon concentrations often means very expensive individual measures dependent on entry pathways, the distribution of radon inside the building, the type of construction, the structural condition and the living habits of the occupants. Special consideration must be given to the influence of near surface drivings of abandoned or existing mines beneath such houses, because in this case several buildings can be influenced by the same radon source. It must be considered whether the totality of remedial measures in individual build­ ings or the m inim ization of the source influence, e.g. continuation of the ventilation of a decommissioned mine beneath residential areas and its optimization, would be more cost efficient with respect to the reduction of the radon concentration. In any case, decisions must be based not only on an optimization but also on recommended reference levels. The German Commission on Radiological Protection has recom­ mended radon levels with different priority classes for such decisions.

2. RECOMMENDATIONS OF THE GERMAN COMMISSION ON RADIOLOGICAL PROTECTION

Based on the new recommendations of the IC R P [1] and taking into account the range of radon concentrations in German houses, the German Commission on Radiological Protection recommends the following radon reference levels for dwellings (annual average of the measured concentration):

— The upper end of the ‘normal range’ is a radon concentration in occupied rooms of 250 Bq/m3. Below this radon concentration, remedial measures are not justified (‘no action level’). This level is taken as a goal for the construction of new houses as well as for remediation of existing houses. — The range between 250 and 1000 Bq/m3 is the ‘discretion range’ in which the occupant can act on his own discretion using simple remedial measures, such as change of room utilization, more frequent and intensified ventilation and 400 POSTER PRESENTATIONS

exploration Schlema creek street shaft / w w m m

I I Jj^borehole shaft “Gang Miet" I I I I I I Л ,Х.+ 3501___

“ t J sublevel j I I (- — - т Л - I------I ______VA-. --i b I i i i +33-1.-1 road "Sana Miet° n ..... r e r ■ Markus-Semmler bottom- 330.0 m sea level

FIG. 1. Radon concentration in a building constructed above shallow drivings depending on mine ventilation [2]. Hatched bars, mine ventilation switched off. POSTER PRESENTATIONS 401

sealing of entry pathways. Measures such as installation of radon impeding layers or change of pressure conditions in the building should be done only in co-operation with technical advisers. — The range above 1000 Bq/m3 must be considered as the ‘remediation range’. Depending on the radon concentrations, optimized remedial measures should be taken, up to expensive and technologically complicated ones. If the annual average of the radon concentration exceeds 15 000 Bq/m3, the exposure must be reduced immediately, i.e. within 1 year.

3. NEAR SURFACE MINING AND INDOOR RADON CONCENTRATIONS

Excessively enhanced radon concentrations may be caused by shallow drivings and mine openings underneath houses. Measurements in a building situated in the uranium m ining town of Schlema immediately above such shallow drivings show the influence of mine ventilation on the radon concentration. During weekend cessation of the ventilation, the radon concentration was enhanced by 2 orders of magnitude, up to more than 30 000 Bq/m3. Figure 1 shows these concentration variations. The mine openings are characterized by a large potential and high availability of radon. In the m ining regions of the western Erzgebirge, radon concentrations of several hundred kBq/m3 were observed in underground mines. Therefore, large quantities of radon can be released to the surface in a short time. Convective gas transport is responsible for the radon concentration in buildings above or near the drivings or openings. If the artificial mine ventilation is out of operation, a typical winter and summer air stream direction can be observed, especially in the case of different heights of the involved openings.

4. MITIGATION MEASURES IN MINING AREAS

The choice of remedial measures to reduce the radon concentration is dependent on the local and individual situation. In any case this problem is an optimization task. According to the measurements described above, it could be assumed that many buildings of Schlema are strongly influenced by the shallow drivings. Actually, the m ining has been discontinued there and decommissioning is going on with flooding of the cavities. Because of the rising water level, parts of the mine including the near surface drives beneath the houses may be cut off from ventilation. For the ongoing underground decommissioning work, it is necessary to install a new ventilation system to reduce the underground radiation exposure. A t the moment, the ventilated cavity is approximately 30 X 106 m 3. The necessary operational costs amount to 0.8-1.7 million DM per year. After decommissioning of the mine, the 402 POSTER PRESENTATIONS

TABLE I. VARIATION OF THE RADON CONCENTRATION (Bq/m3) IN SELECTED BUILDINGS (MEAN VALUE OF MEASUREM ENTS ON GROUND FLOOR) BEFORE, DURING AND AFTER SW ITCHING OFF OF THE MINE VENTILATION (MEASURED BY TRACK ETCH DETECTORS FOR 2 W EEKS)

Ventilation Ventilation House3 Active switched off reactivated

1 500 19 500 1150 2 300 500 530 3 260 150 640 4 480 5 470 420 5 270 500 430 6 260 490 350 7 410 420 490 8 1200 2 000 1530 9 100 230 540 10 250 300 1470 11 130 200 210 12 210 100 60 13 50 370 300 14 70 120 410

a Houses 1-10 with supposed connections to the underground mining area (tunnels, shafts and drivings). Houses 11-14 where probably no connections to the underground mining area exist.

residual mining cavity will be nearly 2 x 106 m 3. Corresponding to first estima­ tions, the costs for the operation of the reinstalled ventilation up to the year 2000 amount to 1.5 million DM [4]. In accordance with the experiences of the project ‘Remediation Models for Radon Burdened Dw ellings in Schneeberg’ [5], the remediation costs for one house in this region are about 50 000 DM . A rough com parison of both these values shows that the costs for mine ventilation are equiva­ lent to the remediation of about 30 individual buildings. Therefore, it must be checked whether after finishing of the underground remediation the ventilation system could be used to reduce the radiation exposure over the whole or at least a considerable part of the settlement area. For this reason, a measurement programme in selected homes of Schlema was conducted during switching off of the mine venti­ lation over a period of 2 weeks. The results confirmed the influence in only a few POSTER PRESENTATIONS 403 houses, as shown in Table I. However, up to now it has been impossible to simulate the final condition of the decommissioned mine. Presently there are many unsealed mine openings influencing the air stream. Therefore, the decision whether artificial or natural underground ventilation may be used instead of the remediation of a number of individual buildings can be taken only after a comprehensive measure­ ment programme providing statistically significant results. The idea to exhaust radon enriched mine air from shallow drivings and thus to decrease simultaneously the indoor radon concentration was applied in the old mining town of Schneeberg, where districts with high building density may be affected by mining. The cavity volume of the mine openings is more than 1 x 106 m 3. Because of missing mine maps, the shafts and underground roads are only partly accessible. Therefore, calculation of air quantity and air circulation as a part of ventilation such as in newly built mines is not possible. Nevertheless, an access to the opening was made and a large fan was installed, operating by alternate pressure and suction. A significant effect on indoor radon concentration was found in certain buildings up to a distance of about 1 km. Depending on the underground air stream, the radon concentration varied by up to 2 orders of magnitude. Additional investigations are planned to acquire knowledge for the development of cheaper natural mine air ventilation, because the effect may be increased by reopening of formerly installed shafts or horizontal passages (adits) and the direction of the air stre a m .

5. CONCLUSIONS

A s mentioned above, the target for new buildings is an annual radon concentra­ tion of 250 Bq/m3, the upper limit of the normal range. In the well known mining districts, radon protected construction of new houses should generally be done independent of future decisions about artificial or natural mine ventilation. There­ fore, for decisions on the installation and operation of underground ventilation sys­ tems, this problem is of secondary importance. W ith respect to the remediation of existing houses in such areas, the ‘pros’ and ‘cons’ of the use of this general method or of individual methods (remediation of single buildings) must be considered. The advantages and disadvantages of underground ventilation are as follows:

Pros: — possible involvement of many houses; — directed release of the underground radon; — avoidance of uncontrolled radon exhalation from mainly unknown cracks or g a p s; — no necessity for alternative areas for the construction of new houses;

Cons:— permanence of maintenance and operation costs, particularly with artificial ventilation; 4 04 POSTER PRESENTATIONS

— uncertainty about the total number of involved houses and the individual reduction of the radon concentration;

— possible increase of radon exhalation from rock; — release of large amounts of radon into the open air (however, monitoring is possible).

The results of measurements with switched off mine ventilation show that the initial expectation of a simple solution turned out to be a deception. Therefore, it is necessary to conduct detailed measurements in all possibly affected buildings of the community. This means that general conditions cannot be described and that the conditions in each community with its special peculiarities must be investigated separately.

REFERENCES

[1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Protec­ tion against radon-222 at home and at work, ICRP Publication 65, Pergamon Press, Oxford and New York (1993). [2] CZARWINSKI, R., LEHMANN, R., RÖHNSCH, W ., Measurements of radon con­ centrations in buildings — results of investigations in mining regions of Saxony and Thuringia (Proc. IRPA8 Congr. Montreal, Canada, 1992). [3] LEHMANN, R., CZARWINSKI, R., Influence of shallow drivings on the radon concentrations in houses — a problem of old mining regions (Proc. Workshop on Indoor Radon Remedial Actions, Rimini, Italy, 1993). [4] WISMUT GmbH, personal communication. [5] FEDERAL OFFICE FOR RADIATION PROTECTION, Estimation of Requirements and Costs of Radon Mitigation of German Residential Buildings, Int. Rep. of the Federal Office for Radiation Protection (1994). 405 POSTER PRESENTATIONS

IAEA-CN-54/126P

STUDY OF INDOOR RADON CONCENTRATION IN POLAND1

K. MAM ONT-CIESLA, J. JAGIELAK, S. ROSIÑSKI, A. SOSIÑSKA, M. BYSIEK, J. HENSCHKE Central Laboratory for Radiological Protection, Warsaw, Poland

1. INTRODUCTION

Investigations of indoor radon concentration and radioactivity of building materials were initiated at the Central Laboratory for Radiological Protection (CLRP) as early as 1968 [1-8]. In 1985 [9] we applied an integrating method using track detectors (Kodak LR -1 15 type II) in almost 500 dwellings for pilot studies in selected areas in Poland. Com parison of the average radon concentrations for four seasons of the year and in three groups of buildings (masonry, concrete and wood) revealed that the ground beneath the building structure is the dominant source of indoor radon. Since the National Atom ic Energy Agency in its regulations of 31 M arch 1988 set the perm issible lim it for the equilibrium equivalent concentration of radon in new buildings (100 Bq/m3), a nationwide survey project for radon in buildings has been undertaken. This study has three objectives: — to estimate the radiation exposure of the Polish population due to radon p r o g e n y ; — to identify radon prone areas in Poland; — to investigate the dependence of indoor radon concentration on such parameters as type of construction material, presence (or absence) of a cellar under the building, number of storeys.

2 . M E T H O D

An integrating method using a solid state track detector based on Pershore CR-39 foil placed at the bottom of a Karlsruhe type diffusion chamber was applied. The time of exposure was 12 months. After exposure, CR-39 detectors were chemi­ cally etched in 10 N NaOH solution at 70°C for 8 hours. The alpha tracks were

1 This work has been partially supported by the IAEA under research contract No. 6161/RI/RB. 406 POSTER PRESENTATIONS

counted by means of an automatic computerized system [10], in an area of 4.16 mm2 corresponding to 16 fields of view. The average background density were 1.62 ± 1.39 tracks per mm. Calibration was done using a 320 L calibration chamber and dry Pylon 226Ra source of activity 3.71 X 106 pCi, with an error of 4% at the 99% confidence level. On the basis of ten calibration exposures, the sensitivity

к = (29.6 ± 0.7) kBq-h-m “3/tr-mm“2

was found with the background track density equal to 1.6 tr/mm2. The low level detection lim it of this method (assumed to be 3 standard deviations of the background track density) was estimated to be 34 kBq-h-m '3, which corresponds, for the exposure time of 12 months, to a radon concentration of 4 Bq/m3. The method was verified in the First International IA EA -U S EPA Passive Radon Detector Intercomparison (October 1991) with an average performance ratio equal to 1.07.

3. RESULTS AND CONCLUSIONS

The radon diffusion chambers are distributed by the staff of the network of meteorological stations evenly covering an area of the country. The results presented refer to three macro regions: Biaiystok, W rociaw and Katowice (one third of the area of Poland); 1060 dwellings were included in the measurements.

Katowice Wroclaw Biaiystok 3 regions Region

FIG. 1. Radon concentration vs. type o f building (dark bar, with cellar; dashed bar, without cellar; hatched bar, all). POSTER PRESENTATIONS 407

60 г-

50

о* ш 40 с о СО Н 30 ф о с о о с 20 о ■Ол Œ 10

J__I__L J__ I__L 1 2 3 4 I 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 I Storey FIG. 2. Radon concentration vs. storey.

Radon concentration (Bq/m3) 500 400 300 200 100

W/R Wood R R/H H W/R/H Prefabr.

Arithmetic mean 68.4 64.6 58.1 50.7 48.4 40.5 37.4 Max. value 318 152 286 432 106 92 168 Type of material FIG. 3. Radon concentration vs. type of material (WIR, wood and red brick; Wood, wood only; R, red brick only; R/H, red brick and hollow brick; H, hollow brick only; W/R/H, wood, red brick and hollow brick; Prefabricated, prefabricated concrete). Solid bar, arithmetic mean; hatched bar, maximum value. 408 POSTER PRESENTATIONS

Conclusions are as follows:

— The log normal distribution is a good approximation to the distribution of radon concentration, with a geometric mean of about 35 Bq/m3 and an arith­ metic mean of 46 Bq/m3, corresponding to an effective dose of about 1.2 mSv/a [11]. The highest values, from 200 to 500 Bq/m 3, constitute about 2% of all results. Most of them appear in Kotlina Kiodzka in the southern part of the W roclaw region. — The mean values for buildings without a cellar are about 55 % higher than for buildings with a cellar (Fig. 1). (Cellars themselves were not tested because they are usually not liveable in Poland). — Radon concentration decreases when number of storeys increases (Fig.2). It is true for each of the five ‘type of material’ groups of buildings. — The mean value is the highest in the group of buildings constructed of wood and red brick and is the lowest in the group of concrete buildings (called ‘prefabricated’) (Fig. 3).

REFERENCES

[1] PENSKO, J., MAMONT, K., WARDASZKO, T., Measurements of ionization radia­ tion in some dwellings in Poland, Nukleonika 14 4/69 (1969) 415. [2] PENSKO, J., Somatic and genetic risks for population due to the natural radiation and type of the building development in Poland, Kerntechnik 17 (1975) 537. [3] MAMONT-CIESLA, K., GWIAZDOWSKI, B., BIERNACKA, M., DZIKIEWICZ- SAPIECHA, H., HENSCHKE, J., Proposal of the Recommendations Limiting the Permissible Concentrations of Natural Radionuclides in Building Materials (Internal CLOR Report 24/77/Z-II, 1977) 20. [4] GWIAZDOWSKI, B., MAMONT-CIESLA, K., BIERNACKA, M., SAPIECHA- DZIKIEWICZ, H., Investigation of Radioactivity of Building Materials, Gamma Radi­ ation Doses and Radon Concentrations in Dwelling-Houses (Report of Polish Academy of Sciences, Warsaw, 1978) 87. [5] GWIAZDOWSKI, B., MAMONT-CIESLA, K., BIERNACKA, M., “ Theassessment of indoor exposure from gamma emitters and radon-222 in Poland” (Proc. Specialists Meeting on the Assessment of Radon and Daughter Exposure and Related Biological Effects, Rome (CNEN), 1980). [6] MAMONT-CIESLA, K., GWIAZDOWSKI, B., BIERNACKA, M., A. ZAK, “ Radioactivity of building materials in Poland” (Second Special Symp. on Natural Radiation Environment, Bombay, 1981). [7] GWIAZDOWSKI, B., MAMONT-CIESLA, K., BIERNACKA, M., “ Assessment of indoor exposure from gamma emitters and radon-222 in Poland” (Second Special Symp. on Natural Radiation Environment, Bombay, 1981). [8] BIERNACKA, M., BRUNARSKI, L., GWIAZDOWSKI, B., KRAWCZYK, M., MAMONT-CIESLA, K ., ZAK, A ., Investigation of the natural radioactivity of build­ ing materials, Probl. Tech. Med. 13 3-4 (1982) 17. POSTER PRESENTATIONS 409

[9] BIERNACKA, M., HENSCHKE, J., JAGIELAK, J., KOCZYÑSKI, A., MAMONT- CIESLA, K., Preliminary measurements of the natural ionization radiation in three types of buildings in Poland, Post. Fiz. Med. 26 1-2 (1991) 55. [10] ROSIÑSKI, S.W ., A computerized track detector reader, Nukleonika 38 2 (1993) 57. [11] UNITED NATIONS, Report of the Scientific Committee on the Effects of Radiation (UNSCEAR) UN, New York (1993).

Case Study 4

RADIOACTIVE WASTE DISPOSAL AND THE ENVIRONMENT

IAEA-CN-54/5P

OVERCOM ING THE RISK PERCEPTION GAP: A SOCIALLY ACCEPTABLE APPROACH TO W ASTE MANAGEMENT

E.R. FRECH, M.A. GREBER Atomic Energy of Canada Limited (A ECL), Ottawa, Ontario, Canada

1. DIFFERENT W AYS OF ASSESSING RISK

Risk is usually defined by experts in the field as the product of the probability of occurrence of an event P times the severity of the potential harm or consequences C, or R = P x С [1]. By this definition, the risk associated with managing high level nuclear waste by deep geological disposal is negligible, even for thousands of years into the future. Most members of the public, however, view nuclear waste as uniquely hazardous and difficult to handle. M any see the risk as absolute rather than relative, and feel that unknown catastrophic events are somehow inevitable [2]. Researchers who conducted a word association test in the U SA in 1991 on the words ‘nuclear waste repository’ found that the most frequent single associations were dangerous, danger, death and pollution [3]. They concluded that the responses revealed perva­ s ive dread, revulsion and anger. A survey conducted for Health and Welfare Canada in 1993 [4] found that Canadians rank nuclear waste as the 11th highest risk to their health. This is almost equal to their perception o f risks from motor vehicle accidents, which kill about 4500 Canadians a year. This is despite the fact that in Canada nuclear fuel waste is managed very safely and there is no public suggestion that there have been any specific victim s of nuclear waste. The same study showed that Canadians had a spe­ cial aversion to carcinogenic agents in their drinking water. Seventy-four per cent agreed (46% strongly) with the statement, “ If even a tiny amount of a substance that can cause cancer were found in my tap water, I wouldn’t drink it” . When asked for their response to the statement, “ No matter how low the level of exposure to radia­ tion, it can still cause cancer” , 62% agreed, 21% strongly.

2. UNDERSTANDING THE GAP

This difference between the general public’s view and that of experts on risk has been of some interest to social scientists, and the overall conclusion of the litera­ ture on the subject is that risk is not an objective phenomenon perceived in the same

413 4 14 POSTER PRESENTATIONS way by all parties. Instead, it is a psychological and social construct, its roots deeply embedded in the workings of the human mind and in a specific social context. Each individual and group assigns a different meaning to risk information. There are mul­ tiple truths about risks, and multiple ways of seeing, perceiving and interpreting risk events. Each interested party — including those who generate the risk, those who attempt to manage it and those who experience it — sees it in different ways [5]. Am ong the factors that influence perception of risk, the following appear to be the most important in relation to radiological risks:

— voluntariness, — personal controllability, — catastrophic potential, — familiarity, — scientific uncertainty, — effects on future generations, — trust in governments and institutions, — equity, fairness and benefits, — reversibility, — personal stake, — o r ig in , — personal values, — accident history, — effects on children.

Considering the many factors that influence the public, researcher Paul Slovic concluded that there is wisdom as well as error in public attitudes and perceptions and offered some hope for resolution of the quandary:

“ Lay people sometimes lack certain information about hazards. However, their basic conceptualization of risk is much richer than that of the experts and reflects legitimate concerns that are typically omitted from expert risk assess­ ments. As a result, risk communication and risk management efforts are des­ tined to fail unless they are structured as a two-way process. Each side, expert and public, has something valid to contribute. Each side must respect the insights and intelligence of the other” [6].

3. BRIDGING THE GAP

How can society best be served in this situation? M any risk experts suggest that the public’s perceptions should be ‘corrected’ so that available funds can be applied to risk reduction in areas where the greatest good can be obtained [7]. Meanwhile, through the political process, the public often insists on expending vast sums on the POSTER PRESENTATIONS 415 further reduction of risks that are already relatively small, while virtually ignoring risks that kill thousands of people a year. One option is to look behind both the experts’ and the public’s assessment of the risks, and determine whether the characteristics of proposed projects can be changed so that they conform more closely to the values people use when making their assessments of risks. This would involve dealing with such factors as voluntari­ ness, controllability, reversibility, equity and fairness, benefits and trust in institutions. Increased voluntariness can be achieved by adopting approaches that allow considerable public involvement in decision making, such as voluntary approaches to site selection, and using agreement building techniques for decision making and problem solving. In Canada, A E C L is recommending a site selection process for a future nuclear fuel waste disposal vault in which the proponent would share decision making power with potential host communities. Such a process would essentially provide for a host community veto on siting decisions. Increased individual control can be accomplished by providing for community participation in monitoring, by establishing community advisory committees with authority to shut down non-conforming facilities [8], equipping the population with the means to detect radiation and to protect themselves from some of its effects, or other similar means. The public’s rationale is that when projects or proposals involve making irrevocable decisions with potentially irreversible impacts, it is better to delay the action that might cause such impacts, even in the face of strong arguments that the status quo is undesirable. The Canadian nuclear fuel waste disposal concept proposes a long range stepwise implementation plan with multiple decision points and monitoring phases. There is a recognition that we cannot in any case prevent future generations from exercising control over what they inherit, or control whether they modify or even reverse today’s decisions if that is what they deem the right thing to d o . Addressing equity requires processes to deal with impacts and their distribu­ tion. Impact management program s that involve the co-ordinated application of m iti­ gation, enhancement, compensation, monitoring and contingency measures can contribute to the siting of locally unwanted facilities and may enhance the acceptabil­ ity of the risks associated with nuclear waste management facilities. Guaranteeing property values is one such potential mitigation technique. Recent research shows that when most members of the public evaluate a risk, what they perceive may be a net value (the difference between risks and benefits) rather than a gross indicator of potential harm [9]. Governments and other propo­ nents can enhance benefits by recognizing that there is a value to society generally from having a place to put generating stations, waste disposal sites and other such facilities, and sharing some of this value with host communities. 416 POSTER PRESENTATIONS

One of the key factors in the perception of risk is the public’s trust in the insti­ tutions that w ill be managing the risk. In the Canadian nuclear fuel waste disposal concept, it has been proposed that a site selection process based on safety and environmental protection, fairness, openness, voluntarism and shared decision m ak­ ing should enhance the likelihood of achieving mutual respect and trust [10]. Such a process would seek a site where technical suitability and public acceptance coin­ cide, rather than applying a set of predetermined technical criteria to find the ‘best’ technical site and then im posing a disposal facility on a potentially unw illing popula­ tion in the name of the public interest.

4. CONCLUSION

How a technology is to be implemented is not strictly a technical determina­ tion, and decisions about what ought to be done involve much more than confidence in scientific and technological ability. Important factors are the need to respect differ­ ent viewpoints and perspectives, and to consult the affected public and involve it in the decision m aking process. The public acceptance of projects and proposals involv­ ing radioactive waste management is dominated by questions related to values, fair­ ness, rights and responsibilities. Using radioactive material will always involve some degree of uncertainty, which the public may view in a way that ascribes an inflated value to the risks involved. Experience seems to suggest that attempts to bring the public’s perception of the risks more in line with the scientifically calculated values simply by providing information (education) have little positive effect. It may be more productive to recognize that the public w ill invariably incorporate non-scientific values in its deci­ sion making processes. Under such circumstances, nuclear waste management projects and other proposals dealing with the use of radioactive material could be made more publicly acceptable by ensuring that their characteristics reflect the values used by most members of the public in making risk assessments.

REFERENCES

[1] CAMPBELL, C .T ., “ Chemical carcinogens and human risk assessment” , Special arti­ cles in Federation Proceedings (1980). [2] FOSTER, D., Nuclear Fuel Waste Management: A Preliminary Evaluation of the Social Issues, AECL Technical Record TR-332, AECL, Ottawa, Ontario (1985). [3] SLOVIC, P., LAYM AN, P., FLYNN, J., Perceived risk, trust and the politics of nuclear waste, Science 254 (1991) 1603-1607. [4] SLOVIC, P., FLYNN, J., MERTZ, C.K ., MULLICAN, L., Health Risk Perception in Canada: A Research Report to the Department of National Health and Welfare, Government of Canada 93-EHD-170 (1993). POSTER PRESENTATIONS 417

[5] COVELLO, V .T., The perception of technological risk: a literature review, Technol. Forecast. Soc. Change 23 2 (1983) 285-297. [6] SLOVIC, P., Perception of risk, Science 236 (1987) 280-285. [7] LIND, N .C., Health and Safety Policies: Guiding Principles for Risk Management, Rep. JCHS 93-1, Joint Committee on Health and Safety, Royal Society of Canada and the Canadian Academy of Engineering (1993). [8] SLOVIC, P., Perceived risk, trust and democracy, Risk Anal. 13 6 (1993) 675-682. [9] GREGORY, R., MENDELSOHN, R., Perceived risk, dread and benefits, Risk Anal. 13 3 (1993). [10] GREBER, M .A., FRECH, E.R., HILLIER, J.A., The Disposal of Canada’s Nuclear Fuel Waste: Public Involvement and Social Aspects, Rep. AECL-1072, COG-93-2 (1994). 418 POSTER PRESENTATIONS

IAEA-CN-S4/37P

RADIOACTIVE W ASTE DISPOSAL AND THE ENVIRONM ENT IN LATVIA

A . S A L M I N S M inistry of Environmental Protection and Regional Development of Latvia, Riga, Latvia

1. INTRODUCTION

Latvia has one central final repository of radioactive waste, the State Enter­ prise ‘Radon’, which is financed from the State budget and has been in operation since 1962. The total amount of radioactive waste is 678 m 3. The average annual amount of waste is 10 m 3 plus the same amount of concrete. The total activity is 5200 C i (1.9 x 105 GBq); the main isotopes are 137Cs, 3H , ^ C o and 90Sr (Fig. 1).

2 3 9 p u i 36C I ! 133B a i B i 210 i 26A I 1 ^ R n i a e R a i 210P b i 14C 1 9 0 y ■ д д w C o 90S r BM “ N i + 3H

137mB a ШШШШШШШШШШШШШШШЯЯШШШШЩШШШШШ 137C s 0 400 800 1200 1600 2000

Activity (Ci)

FIG. 1. Total activity of radioactive waste. POSTER PRESENTATIONS 419

Radioactive waste is disposed of in nearby underground surface monolithic concrete reservoirs:

— three for solid waste, each for 200 m 3; — two for contaminated biological objects, each for 40 m 3; — one (steel) for temporary storage of liquid waste, 200 m 3; — one for solid waste, 1250 m 3.

1.1. Conditioning of waste

Until 1994 sealed sources were disposed of with their containers. The new procedure states that such sources should be disposed of without containers into two underground 0.2 m 3 stainless steel reservoirs in monolithic concrete and further conditioned with melted lead. A ll solid waste is disposed of in reservoirs and made monolithic with concrete or cemented liquid waste. Biological waste is conditioned by chemical treatment for protection against biological processes, with further cementation. Such cemented blocks are disposed of in reservoirs and made monolithic with concrete. Until 1985 all liquid waste was stored, and then conditioned by using a water cleaning system for research reactors. Ion exchange resins and active carbon from these procedures were cemented and disposed of with solid waste. Under the new regulations, liquids must be cemented, but during the wintertime they are stored in stainless steel reservoirs.

1.2. State institutions

The Ministry of Environmental Protection and Regional Development is responsible for drafting laws, regulations and Governmental decisions. Radiation and Nuclear Safety Inspection’s main task is supervision and grant­ ing of licenses and permits for use of radioactive materials and upgrading of the national database on radioactive waste and radioactive materials in the country.

1.3. Legislation and regulations

The first law ‘Radiation safety and nuclear safety’ is to be enforced in 1994. It contains a chapter on radioactive waste, which assigns responsibility to the owners of radioactive materials. It is prohibited for Inspection to grant a license for using materials which cannot be disposed of, e.g. if the activity is higher than acceptable or there is lack of agreement with the waste repository. The next level draft documents are ‘Basic regulation for protection against ionizing radiation’ and ‘Regulation on radioactive waste management’. A s a basis for these regulations, the C E C , IA EA and ICR P recommendations are used, and also the Nordic countries’ experiences in this field. 42 0 POSTER PRESENTATIONS

1.4. Investigations and supervision of environmental impact

Supervision is done at two levels: at the State level by the Radiation and Nuclear Safety Inspection and at the local level by the Regional Environmental Pro­ tection Committee. The Radiation Safety Laboratory at the waste repository has its own investigation system. According to regulations, the Radiation Safety Unit is an independent part of the enterprise. During the State level inspections, representatives of municipal Government and of the public are invited to participate in order to increase the controllability of the facility and to dissipate negative opinions. For clarification of the future environmental impact of radioactive waste dis­ posal, modelling was done together with Swedish experts using BIOM OVES. Results show that also in the far future additional doses for inhabitants w ill be less than those acceptable according to the IC R P recommendations.

2. PUBLIC OPINION

After 1986, mostly during 1988-1991, the public had a strong negative opinion against radioactive waste disposal. This situation was due to at least four m ain issues:

(1) lack of information, related to the K G B of the former Soviet Union, who pro­ hibited public information about radioactive materials (only a small number of professionals could receive such information); (2) general problems with radiophobia after the Chernobyl disaster; (3) public opinion that waste disposal could be anywhere except near one’s home (NIM BY syndrome); (4) use of this situation by some politicians as a way to make their image, some­ times with incorrect information having no scientific basis.

Tw o Governmental commissions investigated the situation. The first commis­ sion clarified technological procedures and examined the eventual impact on the environment, the second developed examinations of design, technology and radio­ active contamination. Both commissions found that the radioactive waste repository is not a source of environmental contamination. Disposal technology requires some changes in procedures, and the State has developed a supervision system. POSTER PRESENTATIONS 421

CONCLUSIONS

The radiological investigation programmes have reached these conclusions:

W ide investigations in surrounding territories are necessary. Measurements of gamma background levels at sampling points must be done. Only simple measurements of total beta activity of samples w ill be done. Detailed investigations must be done if anomalous high levels or systematic increase of results from simplified measurements are found. 422 POSTER PRESENTATIONS

IAEA-CN-54/80P

TRANSM UTATION OF NUCLEAR W ASTE: SAFETY ASPECTS

T . T H E D É E N Centre for Safety Research, Royal Institute of Technology (KTH ), Stockholm, Sweden

Several ideas for accelerator driven undercritical reactors have been presented during recent years. The technical breakthrough in accelerators which can generate intensive and highly energetic proton flows has made the concept feasible. Protons hit a target — fluid lead, tungsten or molten salt has been proposed — and then gener­ ate fast neutrons by spallation. These neutrons or moderated thermic ones w ill trans­ mute, say, nuclear waste into short lived and stable isotopes. These isotopes must be separated from the remaining long lived isotopes using chemical and/or physical methods. The input material could be in an aqueous or molten salt form, each of these giving rise to technical and safety related problems. Together with the transmu­ tation the plant w ill also generate energy. It should be pointed out that no real analy­ sis of the economics of transmutation has been performed up to now. Accelerator driven transmutation can be used for three purposes:

— to transform the long lived isotopes in the nuclear waste into stable or short lived ones; — to burn weapons grade plutonium and get stable and short lived isotopes; — to produce energy.

There is work going on with several different concepts today. At C E R N , Nobel Prize winner C. de Rubbia has proposed a thorium based undercritical reactor driven by a series of cyclotrons. A group under the lead of C. Bowman at Los Alamos has worked for some years on a concept with a linear proton accelerator as spallation source. The safety aspects of the system compared with other ways to take care of nuclear waste are of course fundamental. The safety has been studied in a report by W . Gudowski et al. [1]. It is important to consider the whole system, including:

— transport from the nuclear plant or intermediate storage facility; — storage at the transmutation plant; — transformation of the material into an aqueous or molten salt form; — various aspects of the transmutation facility, including risks connected with repair and replacement of components; — storage of short lived material for a period of some hundred years; — transport of the remaining long lived material to suitable definite storage; POSTER PRESENTATIONS 423

— storage deep in rock of the remaining long lived material (this will be only some small percentage of the original waste, the quantity depending mostly on economic considerations); — after a lifetime of some decades, decommissioning of the plant.

Since no actual plant exists, the result of safety analysis is qualitative. The main requirements are:

(1) The system is subcritical and can switched off at any moment. (2) There is a small inventory of actinides and fission products. (3) The negative temperature coefficient is large. (4) The system is passively safe in the event of a coolant accident. (5) The fuel cycle is proliferation resistant. (6) The remaining isotopes are short lived or stable.

Some safety related problems are still unsolved. If the molten salt alternative is chosen, that might cause material problems and then also risks for control and repair personnel. The same goes for the separation of long lived materials from the short lived and stable ones. Possible risk takers are those working in the plant on control, repair and replacement. Some small health and environmental effects might be found in the close vicinity of the plant. Risks for future generations (with the exception of the next 200 years) can be neglected. If one compares the risk with that of storage of nuclear waste deep in rock, the transmutation alternative will imply an exchange of very small risks in the future for a higher risk to people now living and some few genera­ tions ahead. Quantitative safety analysis must be conducted together with the development and construction of transmutation plants.

REFERENCE

[1] GUDOWSKI, W., PETTERSSON, K., THEDÉEN, T., Accelerator Transmutation of Wastes (ATW ): Prospects and Safety.

Case Study 5

CHERNOBYL HEALTH EFFECTS

IAEA-CN-S4/22P

INTERNAL DOSIM ETRY, BIOLOGICAL DOSIM ETRY, AND BIOM EDICAL STUDIES IN CHILDREN FROM THE FORM ER USSR EXPOSED TO THE CONSEQUENCES OF THE CHERNOBYL ACCIDENT

F. MAURO, R. DE VITA, L. PADOVANI, M. SPANÖ, G. TARRONI, S. BAZZARRI, P. M ETALLI Dipartimento Ambiente, EN EA Casaccia

G. SERLUPI CRESCENZI Circolo San Pietro

Rome, Italy

Upon request of humanitarian organizations (initially the Sovereign M ilitary Order of Malta, and then the Circolo San Pietro in Rom e), a programme was started in 1991 with the aim of monitoring children living in (and in some cases evacuated from) areas contaminated by the fallout of the Chernobyl accident. The groups of children were selected and assembled by local organizations or authorities of Bela­ rus, the Russian Federation and Ukraine. In the first phase of the study (1991-1993), some 1000 children travelled to Italy from the three Republics and remained in Italy for about one month. A ll children underwent paediatric checkups and routine labora­ tory medical analyses in several collaborating Italian hospitals and institutions (Casa Sollievo della Sofferenza, Istituto Superiore di Sanità, Ospedale Bambin Gesù, etc.). Alm ost 500 of these children (from 16 different localities) were examined by whole body counter (W BC) and urine radiotoxicological analyses, carried out at EN EA (National Agency for New Technologies, Energy and Environment) laboratories. Cytogenetic analyses were also performed at EN EA (in collaboration with the 2nd University of Rome ‘Tor Vergata’) in a selected fraction (42 children from Navrovl’a, Stolin, and the Belarus section of the evacuated area around the Cher­ nobyl plant) of the latter group. All children exhibited ,37Cs contamination (minimum 100 Bq, maximum 33 000 Bq) with very large variations among and within areas (all areas are marked as contaminated in the IA E A [1] and locally issued maps). Related levels of 134Cs (and naturally occurring 40K ) were also observed. The variations could be attributed to different life settings and styles and, in particular, to food habits, im plying non- compliance with officially issued countermeasures. In fact, a consistent (one third to one half) decrease of internal contamination was observed by the end of the stay in Italy [2] because of the use of (non-contaminated) average Italian diet.

427 428 POSTER PRESENTATIONS

Because of the very low fraction of chromosomal aberrations observed and the time elapsed since the accident, dicentric based cytogenetic dosimetry was not possible; however, stable chromosomal rearrangements and breaks [3] that could be attributed [4] to the fallout were recorded. In particular, the cytogenetic data indicate that the three groups above mentioned — selected on the basis of the 137C s internal contamination measured by W BC and urine radiotoxicology — exhibit frequencies of acentric fragments statistically higher than those observed in the (Italian) control. Dicentrics and stable translocations have also been observed — in the instance of the dicentrics, their number compares to the zero in the control — but their number is too low to allow either a statistical analysis or a biological estimation of the dose. The cytogenetic data have been analysed against the individual W BC results, and in the instance of one group only (Navrovl’a, Belarus) a statistically significant correla­ tion (p < 0.05) was observed between the internal contamination and the overall fre­ quency of acentric fragments, dicentrics and translocations. The integrated use of these methods, in parallel with medical examinations, is relatively new as far as the application to an exposed population of children is concerned. Thus, the IC R P caesium metabolic model [5] for the two sexes of this age range has been confirmed with actual measurements. Furthermore, the use of W BC as a means to detect and quantify previous exposures to radioactive materials has recently been reconsidered [6]. In general, these children did not show evident signs of pathologies which could be attributed to ionizing radiation [7]. Pathologies related to the specific age range, to demographic and territorial features, and to the hygienic, social and administrative situations (including the essentially rural state of the contaminated or evacuated areas) were observed. However, some deviations in the values concerning CD 4 and CD 8 lymphocyte subpopulations, and also some functional and structural thyroid changes, were observed in a small fraction of the children. The reasons for these deviations were not evident and would need further studies. The latter observa­ tion could be in agreement with the reports on thyroid cancer induction after the Chernobyl accident [8, 9]. The first phase of the programme indicated that such an approach could not give information of statistical validity on the consequences of the Chernobyl accident mainly because of the lack of (an) adequate local control group(s), and the scarce information on the criteria for selection of the children. However, the approach was useful to select individuals for personalized follow-up and to verify local environ­ mental radioactivity data and compliance with countermeasures. A ll relevant infor­ mation was transmitted to the humanitarian organizations and to local authorities, physicians and families. Accordingly, a second phase of the programme has been planned and is cur­ rently (1994) under way. This phase is characterized by:

(1) The very careful selection of a relatively small (250 persons), homogeneous sample of exposed and control children, with plenty of information on their POSTER PRESENTATIONS 429

background (and, therefore, deeper interaction with local operators). This sample will be extracted from a few locations of the three Republics where indications are available on the continuing exposure to Chernobyl related con­ tamination. The members of the control group are extracted from certainly non-contaminated areas and have sim ilar social and ethnic backgrounds. (2) Medical (paediatrics, haematology, immunology, thyroid) and laboratory examinations are being performed. These examinations are being carried out at the Casa Sollievo della Sofferenza and Bambin Gesù hospitals, where particular attention is given to the use of standardized, intercomparable and effective methods of analysis. (3) Determination of internal contamination (W BC and urine radiotoxicology) is being carried out at EN EA laboratories. In the instance of the W BC, the apparatus of the Casa Sollievo della Sofferenza is also available (using E N EA as a reference laboratory). (4) Multiple biodosimetric end points are being measured at the EN EA laborato­ ries (in collaboration with Dr. R. Jensen, University of California at San Fran­ cisco). The following methods are being applied:

— traditional cytogenetics (for the scoring of chromosomal aberrations); — chromosome painting (this assay is based on the use of human chromosome libraries that selectively hybridize specific chromosomes and can be recog­ nized by fluorescent labelling; in this way, chromosomal translocations can be adequately detected) [10]; — induction of the glycophorin A (GPA) mutation (this mutagenesis assay detects variant erythrocytes that occur as a result of in vivo allele loss at the GPA locus on chromosome 4 in humans heterozygous for the M and N allelic blood types) [11].

Previous work, carried out on nuclear bomb survivors [12], on adult popula­ tions involved in the Chernobyl accidents and on occupational groups [11], has suggested that m ultiple end points correlate with one another, point to different persisting effects, and yield data which may differ from physical dose measure­ ments or reconstructions. (5) Statistical prescriptions and evaluations w ill be performed by the group of the Istituto Superiore di Sanità which already analysed the results of the first phase of the programme.

Until now, some 200 children have entered this second phase of the study. Analyses and evaluations are still being carried out. Briefly, these groups of children (from Gomel, Belarus; evacuated from Pripjat, Ukraine; from Novosybkov, Russian Federation, in collaboration with D r. L . Baleva; and the control group) so far appear more homogeneous in terms of contamination levels (except for particular individuals with specific life styles); these levels seem to be sim ilar to those observed in the first phase of the study. Some thyroid pathologies have again been observed. 430 POSTER PRESENTATIONS

Traditional cytogenetic results appear prelim inarily to be sim ilar to those of the first phase of the study. Finally, G PA tests have been carried out on 29 of 74 children who exhibited the M N blood type; the relative analysis is still under way, but it appears that this mutation may not be a relevant phenomenon, probably because of the relatively low doses of exposure. Local authorities and humanitarian organizations have asked us to extend this programme to groups from areas in Siberia and Kazakhstan contaminated by nuclear explosion tests and radiological accidents. It is unfortunate that, at least in our experience, international co-operation in this field is operationally very difficult, because of a variety of problems, including the geopolitical evolution of the former USSR.

REFERENCES

[1] INTERNATIONAL ATOMIC ENERGY AGENCY, Surface Contamination Maps (The International Chernobyl Project), IAEA, Vienna (1991). [2] TARRONI, G., CASTELLANI, C.M., MELANDRI, C., BATTISTI, P., RAMPA, E., TICCONI, R., FORMIGNANI, M., BAZZARRI, S., SANTORI, G., DE VITA, R., SERLUPI CRESCENZI, G., Evaluation of 137Cs internal contamination in children by means of whole body counter measurements, Radiat. Protect. Dosimetry 41 (1992) 223. [3] PADOVANI, L ., CAPOROSSI, D., TEDESCHI, B., VERNOLE, P., NICOLETTI, B., MAURO, F., Cytogenetic study in lymphocytes from children exposed to ionizing radiation after the Chernobyl accident, Mutation Res. 319 (1993) 55. [4] AW A, A .A ., “ Chromosome damage in atomic bomb survivors and their offspring — Hiroshima and Nagasaki” , Radiation-Induced Chromosome Damage in Man (ISHIHARA, T., SASAKI, M., Eds), Liss, New York (1983) 433-453. [5] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Age Dependent Doses to Members of the Public from Intake of Radionuclide, Part 1, Publi­ cation 56, Pergamon Press, Oxford and New York (1989). [6] TOOHEY, R., PALMER, E., ANDERSON, L., BERGER, C., COHEN, N., EISELE, G ., WACHHOLZ, В., BURR, W ., Jr., Current status of whole-body count­ ing as a means to detect and quantify previous exposures to radioactive materials, Health Phys. 60 Suppl. 1 (1991) 7. [7] CIRCOLO SAN PIETRO, ENEA, Convegno sullo Stato di Salute di Bambini Provenienti dalle Zone Contaminate di Chernobyl, Circolo San Pietro, Rome (1993). [8] KAZAKOV, V.S., DEMIDCHIK, E.P., ASTAKHOVA, L.N., Thyroid cancer after Chernobyl, Science 359 (1992) 21. [9] BAVERSTOCK, K., EGLOFF, B., PINCHERA, A., RUTCHI, C., WILLIAMS, D., Thyroid cancer after Chernobyl, Science 359 (1992) 21. [10] GRAY, J.W., KALLIONIEMI, O., KALLIONIEMI, A., KUO, W.L., STRAUME, T., TKACHUKD, D., TENJIN, T., WEIER, H.U., PINKEL, D., “ Applications of fluorescence in situ hybridization in biological dosimetry and detection of disease- specific chromosome aberrations” , New Horizons in Biological Dosimetry (GLED- HILL, B.L., MAURO, F., Eds), Wiley Liss, New York (1991) 399-413. POSTER PRESENTATIONS 431

JENSEN, R.H., GRANT, S.G., LANGLOIS, R.G., BIGBEE, W .L., “ Somatic cell genotoxicity at the glycophorin A locus in humans” , New Horizons in Biological Dosimetry (GLEDHILL, B .L., MAURO, F., Eds), Wiley Liss, New York (1991) 329-339. NAKAMURA, N., UMEKI, U., HORAI, Y., KYOIZUMI, S., KUSHIRO, J.I., KUSUNOKI, Y ., AKIYAM A, M., Evaluation of four somatic mutation assays for biological dosimetry of radiation-exposed people including atomic bomb survivors, New Horizons in Biological Dosimetry, (GLEDHILL, B.L., MAURO, F., Eds), Wiley Liss, New York (1991) 341-350. 432 POSTER PRESENTATIONS

IAEA-CN-S4/27P

CANCER AND NON-CANCER DISEASES OF THE THYROID GLAND AND THEIR DOSE DEPENDENCE IN CHILDREN AND ADOLESCENTS AFFECTED AS A RESULT OF THE CHERNOBYL ACCIDENT

A.F. TSYB, E.M. PARSHKOV, V.K. IVANOV M edical Radiological Research Centre of the Russian Academy of Medical Sciences, Obninsk, Russian Federation

Over 30 000 children and adolescents from four rayons of Bryansk oblast and from three rayons of Kaluga oblast have been examined. The cohort is being fol­ lowed up dynamically. In four rayons of Bryansk oblast, 24.5 % of all children are residing in the terri­ tory with contamination density over 555 GBq/km 2. At the same time, the exam­ ined children of Kaluga oblast reside in rayons with 137Cs contamination density less than 481 GBq/km2. Mean absorbed doses to the thyroid of children under 7 years old living in different settlements with 137Cs contamination over 555 GBq/km 2 range from 70 to 220 cGy [1]. The distribution of individual doses in Kaluga oblast is characterized by the relatively long ‘tail’ in the range of large doses. Few persons received doses exceed­ ing 1000 cGy. At the same time the thyroid of more than half of the children exam­ ined (54.6% ) was exposed to low doses of radiation [2]. An age dependence of the radiation doses to the thyroid was clearly detected. The largest doses were found to be in children under 3 years old. Three per cent of children received doses above 200 cGy. The results of studies of thyroid gland function were characterized by the levels of certain hormones in the blood of children and adolescents. Significant increase of T3 and T4 levels occurred only during the first 2 years after the accident. This means that the thyroid gland function was stimulated, most likely because of radiation. Such changes have not been observed in the following years. The ratio between TSH and free T4 levels in blood can indicate hyperfunction or hypofunction of the thyroid gland. The ratio shows that the fluctuation in function took place in a small percentage of children, as was usually observed in the control group also. The number of children of Bryansk oblast with the increased titre of antibodies to thyroglobulin varied between 1.5 and 5.5% irrespective of the density of radiation contamination. It did not depend on the level of radiation contamination of the areas. TABLE I. DISTRIBUTION OF THYROID CANCER IN CHILDREN OF BRYANSK OBLAST (15 APRIL 1994) *о ú¿ (D «5 Э и - ’ со ’S -8 S S i 'ст*g _ е E •о > Æ *о о л 3 b о 2 8 ¿* о С ¿Г* с О « SP 6 Н кп с Л О с л Л о Z СЛ л S E 1)0> - § g- n О <4— ú¿ О 03 >> с d

о I } ViГ} о л с й 1) О

о OTR RSNAIN 433 PRESENTATIONSPOSTER O o Tt o O O со cu Z ХЭ 00 O o > O N >> o > 8- n n n O ON O tj n - O'O'5'O'O'O'O'C'fl'O'O' O ' O ' l f ' C ' O ' O ' O ' O ' 5 ' O ' O O r- ID «—i ЛЛ Л Л Л ЛЛ *S, —— je •o СЧ X ( C/3 a Л c3 & o n

s- 's cu O се È* n CU ‘a- O & n u- Д 3 d j N ^ и О O о о U L 2 5L o > n i -S _ ' á CU ‘S- —ч н-н OOO Л Л л ^ Л CQ c¡t b

‘cL _ÇJ a. 5 > r-. en en r- + Л c3 8 I P 6 (A 4> b n *cL .52 IX O b «3 n O 00 VO ON ON О ' O Os (N ON ON СП ON ON n

In addition, there were numerous head X rays. 4 3 4 POSTER PRESENTATIONS

The number of children with increased titre of antibodies to thyroglobulin in Kaluga oblast was found to be even lower [3]. In the first 2 years after the accident, from 3.6 to 4.8% of children were found to have increased titre of antibodies to microsomal fraction of thyroid (MAB). The number of children with increased levels of MAB was significantly higher in the group with irradiated thyroid than in the control group. The diagnostic test is very sensitive to detection of autoimmune thyroiditis developing without clinical symptoms of disease [4]. Thyroid gland function was studied together with its structure by means of ultrasound screening. Among the examined children, abnormalities of the thyroid were found in 15-22%, mainly hyperplasia and euthyroid goitre. The incidences of autoimmune thyroiditis and nodule formation remained in the range from 0.3 to 0.5%. Thus, they are at the same level in various periods since the accident. A rela­ tionship between the incidence of thyroiditis and nodules and the radiation dose to thyroid was not found. No cancer of the thyroid gland has been detected in children of Kaluga oblast since the accident. At the same time, 17 cases of cancer were detected in children of Bryansk oblast from 1990 to 1994 (Table I). Significantly, this rise was registered in radiation contaminated rayons of Bryansk oblast in 1993 (eight cases). Four cases of thyroid cancer were found within 4 months in 1994. In the period before the Chernobyl accident, the spontaneous level of the incidence of thyroid cancer in children of this region did not exceed one case in 10 years. The papillary type of cancer was morphologically verified in 15 children and the follicular type in one child. Regional lymph node métastasés developed in five children and métastasés to the lung in three children. Among the all detected cases of thyroid cancer there were nine girls and eight boys. Their age distribution is as follows: year of birth 1975 — 1, 1976 — 2, 1977 - 3, 1979 - 1, 1980 - 1, 1982 - 1, 1983 - 1, 1984 - 3, 1985 - 2, 1986 — 1 (irradiated in utero). Dosimetric investigation was performed in nine out of 17 cases. The most probable values of the reconstructed doses of 131I radiation (reference values) to the thyroid ranged from 70 to 140 cGy in six cases. In three children they were less than 3 cGy. There is an impression that the development of thyroid cancer in children is aggressive. The period from the last examination of a child to the diagnosis of cancer is not more than 1 year. As was mentioned above, a quarter of the children already had métastasés. These results emphasize the importance of early diagnosis and of treatment in due time of abnormalities of the thyroid gland, in different dose groups. To determine the dose dependence of incidence rates of the thyroid abnormali­ ties in Kaluga oblast, a cohort of 5766 children and adolescents was set up. In May and June 1986 individual absorbed doses to the thyroid due to 131I were estimated in them on the basis of direct radiometric measurements. The collective dose to chil- POSTER PRESENTATIONS 4 3 5

ш с_

(D > ■Р JP GD Œ

D o s e ( c G y )

FIG. I. Estimation of relative risk of thyroid diseases (except cancer).

dren and adolescents included in the cohort is up to 300 000 man -cGy and the mean individual dose is 55 cGy. On the basis of the ICRP recommendations, the lifetime risk in this group was found to be not less than 15 additional cases of radiation induced thyroid cancer. Application of the special methods for calculation of radia­ tion risks [5] enabled identification of the dose dependence of the morbidity inci­ dence rate in this cohort for 8 years since the accident (Fig.l). Attributive risk, i.e. share of radiation induced thyroid abnormalities among all revealed cases of thyroid disease, is 12% (4.1-18.7% ).

REFERENCES

[1 ] TSYB, A .F ., STEPANENKO, V .F., OMELCHENKO, V .N ., Radiation and Risk, Vol. 4, Bull. State Medical and Epidemiological Registry, Moscow-Obninsk (1994) 87-93.

[2 ] TSYB, A.F., MATVEENKO, E.G., GOROBETS, V .F., et al., Radiation and Risk, Vol. 4, Bull. State Medical and Epidemiological Registry, Moscow-Obninsk (1994) 62-73. [3] TSYB, A .F ., MATVEENKO, E.G., GOROBETS, V.F., Med. Radiol. 7 (1991) 4. [4] SHINKARKINA, A.P., PODGORODNICHENKO, V.K., POVERENNY, A.M., Radiobiol. Radioecol. 1 (1994) 3. [5] PRESTON, D .L., LUBIN, J.I., PIERCE, D .A., Epicure Command Summary, Hirosoft, Seattle, W A (1990). 436 POSTER PRESENTATIONS

IAEA-CN-54/28P

DOSE-MORBIDITY DEPENDENCE OF

CHERNOBYL ACCIDENT EMERGENCY W ORKERS

V.K. IVANOV, E.M. RASTOPCHIN, A.F. TSYB Medical Radiological Research Centre of the Russian Academy of Medical Sciences, Obninsk, Russian Federation

During 1986-1989, thousands of people participated in the cleanup after the Chernobyl accident. In order to provide medical treatment and monitoring of the emergency workers, the National Medical Dosimetric Registry has been established in the Russian Federation [1]. At present, the Registry contains information on 235 888 residents of Russia exposed to radiation due to the Chernobyl accident, including 144 762 Chernobyl emergency workers. Of these 144 762 emergency workers, 99 475 have official documents verify­ ing the doses of external exposure received during the period of restoration activities within the 30 km zone around the Chernobyl nuclear power plant. For 131 persons among this number, no information about any medical checkup after the cleanup in the 30 km zone is available. Thus the size of the cohort under study is 99 344 per­ sons. Men constitute more than 99% of these emergency workers. The average age of the emergency workers is 33.7 years; at the time of the accident, 90% of the emer­ gency workers were in the age group of 18-40 years. Eighty-two per cent of the emergency workers from the 99 344 person cohort under consideration took part in the restoration activities within 2 years after the Chernobyl accident. Figures 1 and 2 show the age and external exposure dose distributions of the emergency workers. All the participants of the Chernobyl cleanup are subjected to annual medical checkups in the hospitals at their places of residence. Special Registry documents are completed in accordance with the results of medical checkups, the diseases revealed being encoded by ICD-9. Thus the National Registry contains morbidity rate dynamics of emergency workers for the period 1986-1993. Medical and dosimetric information on every emergency worker in the Regis­ try provides ample grounds for the determination of dose-morbidity dependence, i.e. for the determination of radiation risk coefficients [2]. The Poisson regression method and the EPICURE [3] code are used to deter­ mine the dose-morbidity dependence for general classes of diseases. The relative risk RR for the dose D is defined as follows:

RR (D) = 1 + bD where b is a constant. Parameter b is called the excess relative risk per dose unit and POSTER PRESENTATIONS 4 3 7

с

A g e

FIG. 1. Age distribution of emergency workers.

0-5 5-10 10-15 15-20 20-25 25,+

c G y

FIG. 2. External dose distribution of emergency workers.

is inverse in dimension to the exposure dose. Figure 3 shows estimates of excess rela­ tive risk per 1 cGy at the 95 % confidence interval. The figure shows increase of rela­ tive risk values depending on the dose for some general classes of diseases for emergency workers: malignant neoplasms, diseases of blood and blood forming organs, diseases of the endocrine system, mental disorders, diseases of the nervous system and sensory organs, and diseases of the circulatory system. 4 3 8 POSTER PRESENTATIONS

Skin and subcutaneous tissue - Genitourinary system - Digestive system - Respiratory system - Circulatory system- Nervous system and sensory organs Mental disorders Blood and blood-forming organs E ndocrine diseases Malignant neoplasms Neoplasms Infectious and parasitic diseases

-0.04 0.00 0.04 0.08

b ( o G y " ‘ )

FIG. 3. Estimation of excess relative risk per 1 cGy for various classes of diseases.

It should be noted that a number of unfavourable factors, emotional stress among them, affected the health of emergency workers along with the radiation exposure. Therefore, the results obtained emphasize the necessity of long term monitoring of the state of health of emergency workers, primarily of those exposed to higher doses.

REFERENCES

[1] TSYB, A.F., et al., Med. Radiol. N7 (1989) 3 ( in Russian). [2] TSYB, A.F., et al., Proceedings of the Fukui Workshop on Health Risk, Kutsuxama, Fukui, Japan (1992) 176-195 . [3] PRESTON, D .L., et al., EPICURE User’s Guide, Hirosoft, Seattle, WA (1992). POSTER PRESENTATIONS 439

IAEA-CN-54/29P

DELAYED RADIATION EFFECTS OF THE ACCIDENT AT THE

CHERNOBYL NUCLEAR POW ER PLANT ON THE

EXPOSED POPULATION IN THE RUSSIAN FEDERATION

A.I. GORSKY, V.K. IVANOV, A.F. TSYB Medical Radiological Research Centre of the Russian Academy of Medical Science, Obninsk, Russian Federation

1. INTRODUCTION

In order to predict long term effects of radiation exposure on human health, predictive models based on extrapolation of data obtained during limited periods in the lives of people were used. This modern methodology of prediction and the data necessary for it are described in Refs [1, 2]. In compliance with ICRP expert recommendations [1], the multiplicative model of excess cancer mortaility prediction is preferable for the majority of sites. The predictions contained in the present paper were made according to this model. For prediction of late effects, the calculated model employed approximate excess radiation risk (ERR) values for organ specific cancer presented in BEIR-V [2] and the total mortality rates and organ specific mortality rates for the unexposed cohort differentiated by age and sex given in Ref. [3]. The minimum latency period and time interval (‘plateau’) of expected long term exposure effects conform to values recommended in Ref. [1].

2. MATERIALS AND METHODS

2.1. Cohort

The investigation was based on the data from two cohorts.

Emergency workers

Emergency workers make up a major group within the Russian Medical and Dosimetric State Registry (RMDSR). The cohort is of special interest for study of the effect of radiation from the Chernobyl accident on the population because medi­ cal and dosimetric data are documented, consistent and reliable. Doses received by emergency workers during a relatively short time span (1-2 months) for the most 4 4 0 POSTER PRESENTATIONS

7 0 0

60 0

И 500

I 40 0

g 300 я U 200

1 0 0

0 0 10 20 30 40 50 60 70

Post-exposure period (years)

FIG. 1. Distribution of estimated total number of radiation induced cancer deaths among emergency workers (employed in 1986) per 100 000people (average dose 165 mSv) depending on post-exposure interval.

Post-exposure period (years)

FIG. 2. Distribution of estimated annual radiation induced cancer death probability rates among emergency workers (employed in 1986) per 100 000 people (average dose 165 mSv) depending on post-exposure interval. POSTER PRESENTATIONS 4 4 1

FIG. 3. Distribution o f expected number of radiation induced cancer deaths among residents of contaminated areas (555-740 GBq/km2, lifetime dose 105 mSv) per 100 000 people depending on the time since the Chernobyl accident.

и я Q i >> ce J3 •4-1 СО 'O U <1> CJ C ce

FIG. 4. Distribution of estimated excess annual cancer mortality rates among residents of contaminated areas (555-740 GBq/km2, lifetime dose 105 mSv) per 100 000people depend­ ing on the time since the Chernobyl accident. 4 4 2 POSTER PRESENTATIONS part exceed predicted lifetime doses for residents of contaminated areas; hence the dynamics of late stochastic effects of radiation exposure with time should be more intensive for the emergency worker cohort than for the residents of contaminated areas. According to the RMSDR data, emergency workers in the contaminated area in 1986 received the highest doses on average (about 165 mSv) and predictions were made for this group.

Male residents o f areas of the Russian Federation contaminated as a result of the Chernobyl accident

In our calculations we used the age distributions of the Russian population con­ tained in Ref. [4]. The time-dose distribution was calculated in compliance with the model described in Ref. [5], taking into account doses from external and internal sources in 1988 described in Ref. [6]. The dose was calculated for the residents of areas with surface contamination of 555-740 GBq/km2 (150 000 people are esti­ mated to reside in the area with these contamination levels). The estimated average lifetime dose for a member of the population is about 105 mSv.

3. RESULTS AND DISCUSSION

Predictions of excess cancer deaths among emergency workers are shown in Figs 1 and 2. Figure 1 shows the radiation induced organ specific cancer deaths distribution according to post-exposure interval. The estimated number of excess deaths among emergency workers will be approximately 650, which is in compliance with a lifetime mortality probability coefficient of about 4.0 x 10'2/Sv. Figure 2 shows the number of excess annual cancer deaths among emergency workers depending on the time since their employment in the exposure zone. It is clear from Fig. 2 that the maximum of annual excess deaths is expected 20-25 years after the exposure and will constitute 18 deaths a year. The estimated lifetime excess number of cancer deaths among the residents of contaminated areas per 100 000 people will constitute 270. This number corresponds to a lifetime mortality probability coefficient of about 2.6 X 10“2/Sv. The distribu­ tion of the expected number of radiation induced cancer deaths among residents depending on the time since the Chernobyl accident is shown in Fig. 3. The maxi­ mum of excess deaths among the residents is expected to be five persons 50 years after the exposure (Fig. 4). The dependence shown in Figs 2 and 4 is used by ICRP experts to produce restrictions on radiation risk. POSTER PRESENTATIONS 4 4 3

In compliance with ICRP recommendations, the level of risk unacceptable to the population is 3 x 1СГ5 per year; hence risk values will be unacceptable 12-15 years after cleanup operations for residents of areas with 555-740 GBq/km2 contamination (see Fig. 4) and 5-8 years (see Fig. 2) for non-professional emer­ gency workers (whose occupation is not radiation related). Also, the dose received by emergency workers from the cohort under study for 1-2 months exceeds the dose limit recommended by ICRP experts for people whose occupation is radiation related (the dose should not exceed 50 mSv/a).

4. CONCLUSION

Excess radiation induced cancer mortality among residents of the Russian Fed­ eration resulting from the consequences of the Chernobyl accident or from residence in areas contaminated after the accident was predicted in compliance with ICRP recommendations. For emergency workers the excess lifetime radiation induced cancer deaths are expected to amount to 650 per 100 000 people (or 3.3% of natural cancer mortality). For residents of radionuclide contaminated areas (with contamination of 555-740 GBq/km2) this number is predicted to be 270 deaths per 100 000 people (or 1.4% of natural cancer mortality). As for leukaemia mortality, the number is 86 per 100 000 people (or 21 % of the natural leukaemia mortality) for emergency workers and 42 per 100 000 people (or 11 % of the natural mortality) for residents.

REFERENCES

[1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, 1990 Recommendations of the ICRP, ICRP Publication 60, Pergamon Press, Oxford and London (1991). [2] NATIONAL ACADEM Y OF SCIENCES, Health Effects of Exposure to Low Levels of Ionizing Radiation, BEIR-V Rep. NAS, Washington, DC (1990). [3] AXEL, E.M ., DVOIRIN, V .V ., Cancer statistics (morbidity, mortality, trends, socio­ economic damage, life expectancy), M: VONC AMS USSR (1992) 9-56. [4] Population of USSR 1987, Statistical Records, Finansy i Statistika, Moscow (1988). [5] BARKHUDAROV, R.M ., GORDEEV, K.I.,.SAVKIN, M.N., Long Term Prediction of Population Exposure in the Areas Contaminated After the Chernobyl Accident, reprint from Enviromental Contamination Following a Major Nuclear Accident, Vol. 2, IAEA, Vienna (1990). [6] Problems of Ecological Monitoring (Proc. of the Radiological and Biological Con­ ference), Part I, Briansk (1991). 4 4 4 POSTER PRESENTATIONS

IAEA-CN-54/30P

RECONSTRUCTION OF THE ABSORBED EXTERNAL DOSES TO THE POPULATION LIVING IN AREAS OF THE RUSSIAN FEDERATION CONTAMINATED AS A RESULT OF THE CHERNOBYL ACCIDENT

V.A. PITKEVICH, V.K. IVANOV, A.F. TSYB Medical Radiological Research Centre of the Russian Academy of Medical Sciences

V.M. SHESHAKOV, A.V. GOLUBENKOV, R.V. BORODIN, V.S. KOSYKH Science and Production Association ‘Typhoon’

Obninsk, Russian Federation

1. INTRODUCTION

For epidemiological studies aimed at understanding how the radiation factor affected the people living in the areas contaminated following the Chernobyl acci­ dent, knowledge is required of radiation loads for human organs and tissues from external and internal exposure. In this respect, of great importance are data about absorbed doses received in the first year after the accident. Methods and assessments available in the literature (e.g. Refs [1, 2]) are based on a simple approximation of one time entrance of radionuclides into the environment. Space-time characteristics of depositions have been reconstructed with varying accuracy, primarily for 137Cs and 131I. Therefore, now 8 years after the accident, it seems important to recon­ struct the complete dynamic picture of the radioactive contamination of the Russian Federation with consideration of newly published data [3] about the source term.

2. MATERIALS AND METHODS

2.1. Reconstruction of external absorbed doses from radioactive clouds and fallout

For reconstruction of the time dependence of the exposure dose rate in the populated areas, considering that hourly measurements of the exposure dose rate were not made from the beginning of the radioactive contamination, it is appropriate to use results of modelling of atmospheric dispersion of radionuclides (MADR). POSTER PRESENTATIONS 4 4 5

To obtain space-time characteristics of the radioactive contamination, where q (t, x, y) is the density of surface contamination with a radionuclide and p (t, x, y) its volumetric concentration in the surface atmospheric layer at the point with geographical co-ordinates (x, y), we used a model of regional transport with allowance for three dimensional turbulent diffusion of aerosol and gaseous con­ taminant in the atmosphere. This model is described at length in Ref. [4]. The height of the atmospheric boundary layer was taken to be 2 km. The modelling covered the period from 26 April to 21 May 1986 and results were obtained for every 6 h. The results of MADR are strongly dependent on the source term Q(t, h, d, s, a), where h is release height and d is size of aerosol particles carrying the s-th radionu­ clide (for the aerosol part of the release). For selection of the Q (t, h, d, s, a) function for the key radionuclides 137Cs, 131I and 144Ce, data of Refs [1, 3] were used. We used the results of MADR only for space-time description of the contami­ nation. For taking into account local non-uniformities in depositions, a model of ‘local effective precipitation’ (LEP) was developed. It is used to allow corrections as a factor to model functions q(t, x, y, s, a) and p (t, x, y, s, a), matching modelling results with measured and reconstructed deposition densities. The reconstruction of deposition density of major dose forming nuclides in the populated areas relies on detailed statistical analysis of gamma spectrometric data of 2867 soil samples col­ lected in 1986-1988. The analytical results are described in Ref. [5]. A stepwise approximation of the deposition rate was used for this purpose. As a result, functions q(t, x, y, s, a) and p(t, x, y, s, a) for 137Cs, 134Cs, 136Cs, 131I, 132Te, 133I, 140Ba + 140La, 95Zr + 95Nb, 103Ru, 106Ru, 141Ce, 143Ce, 144Ce and 125Sb were reconstructed for the contaminated territories of the Russian Federation. Based on volumetric activity of radionuclides, the absorbed dose rate (ADR) for external exposure of the public from a radioactive cloud was assessed. To describe ADR from the radioactive fallout in the populated area, we used reconstructed functions qr(t, s). When the model time dependence of ADR was constructed, account was taken of the following processes: deposition of radio­ nuclides on the soil without structures and terrain features, qr(t, x, y, s, a); radio­ active decay of radionuclides, Xs; gamma radiation attenuation by snow cover, S(t); vertical migration of radionuclides in soil, M (t, x, y); and measurements of exposure dose rate (EDR) in the considered populated area in 1986.

3. RESULTS AND DISCUSSION

Figure 1 shows results of ADR reconstruction for Novozybkov, Bryansk Region (numerous early measurements of EDR) and Mileevo, Khvastovichsky District, Kaluga Region (no early measurements of EDR). As is seen from Fig. 1, the reconstruction model adequately describes the available measurement data and fully reconstructs the ADR time dependence from the accident time to date. 4 4 6 POSTER PRESENTATIONS

t (days)

t (days)

FIG. 1. Reconstructed dependence of ADR on the interval t between the current date and the date o f the Chernobyl accident. Solid line, model exposure dose rate (EDR) from soil surface depositions; shaded areas, ADR from radioactive clouds; marks, measured EDR with correc­ tion for the energy dependence of dosimeter sensitivity. POSTER PRESENTATIONS 447

TABLE I. ESTIMATED EXTERNAL RADIATION DOSES FOR THE POPULATIONS OF SOME AREAS OF THE RUSSIAN FEDERATION FROM THE TIME OF RADIOACTIVE CONTAMINATION TO LATE 1986a

Populated area ^mln О ^max ^exp Dcl

Zlynka 11.5-28.6-77.9 5.000 0.083 Vyshkov 9.6-29.1-47.5 6.466 0.124 Barki 10.7-29.1-59.4 9.361 0.052 Zarech’e 27.0-39.3-57.0 11.692 0.120

Klintsy 0.4-4.2-26.8 0.522 0.005 Novozybkov 1.8-16.6-38.8 1.053 0.033 Zhizdra 0.2-2.2-5.1 0.203 0.015

Ul’yanovo 2.8-4.1-5.4 0.530 0.002

Dudorovsky 1.3-7.8-21.8 1.013 0.004 Khvastovichi 0.2-1.8-5.6 0.701 0.003

Kolodyassy 2.7-7.9-16.2 1.411 0.026

a ff-ffmax is the minimum-average-maximum density of 137Cs deposition measured in the populated area, expressed as Ci/km2 (3.7 x 104 Bq/m2); Dexp is the exposure dose in air at 1 m height from radioactive depositions with allowance for dose from natural sources of gamma radiation, expressed as p(2.58 X 10"4 С /kg); Dc] is the whole body absorbed dose from passing radioactive clouds, expressed as cSv.

Table I, by way of example, includes preliminary estimates for exposure doses from deposition and absorbed doses from the time of the accident to late 1986 for several population groups in populated areas of the Bryansk and Kaluga regions.

4. CONCLUSION

The proposed methodology allows the main characteristics of the radiological situation to be reconstructed in the populated areas of the territory contaminated after the Chernobyl accident. 4 4 8 POSTER PRESENTATIONS

REFERENCES

[1] IZRAEL, YU .A. (Ed.), Chernobyl: Radioactive Contamination of the Environments, Hydrometeoizdat, Leningrad (1990) (in Russian). [2] GOLYKOV, V .YU ., BALONOV, M.I., PONOMAREV, A .V ., “ Estimation of exter­ nal gamma radiation doses to the population after the Chernobyl accident” , The Cher­ nobyl Papers, Vol. 1: Doses to the Soviet Population and Early Health Effects Studies (MERWIN, S.E., BALONOV, M., Eds), REPS, Washington, DC (1993) 247-288. [3] BUSULUKOV, YU .P., DOBRYNIN, YU .L., “ Release of radionuclides during the Chernobyl accident” , The Chernobyl Papers, Vol. 1: Doses to the Soviet Population and Early Health Effects Studies (MERWIN, S.E., BALONOV, M., Eds), REPS, Washington, DC (1993) 3-22. [4] VAKULOVSKY, S.M., SHERSHAKOV, V.M., GOLUBENKOV, A.V., et al., Com­ puter information support system in problems of analysis of the radiological situation in territories contaminated after the Chernobyl accident, Radiation and Risk 3 (1993) 36 (in Russian). [5] PITKEVICH, V .A ., SHERSHAKOV, V.M ., DUBA, V .V ., et al., Reconstruction of radionuclide composition of fallout on the territory of Russia as a result of the Chernobyl accident, Radiation and Risk 3 (1993) 62 (in Russian). POSTER PRESENTATIONS 4 4 9

IAEA-CN-54-33P

CONSEQUENCES OF THE CHERNOBYL ACCIDENT IN GEORGIA

D. MANDZHGALADZE Radiology Research Institute, Georgia

M. SHAVDIYA Principal Public Health Physician of Georgia

The accident at the Chernobyl NPP, which caused general contamination of nearby regions, also affected the Black Sea coast of Georgia. On 2 May 1986, a few days after the accident, Georgian meteorological stations recorded a considerable increase in the radiation background due to short lived isotopes almost everywhere in Georgia. The heavy rains which fell along the Black Sea coast of Georgia early in May 1986 contributed to the environmental radioactive contamination. In official documents, Georgia appears as one of the most contaminated regions after Belarus, Ukraine and the Russian Federation.

TABLE I. CONTAMINATION LEVELS IN GEORGIA3

Contamination level (Ci/km2) Number of Town/village control point l34Cs l37Cs 4°k ^Sr

1 Gulripshi 0.21 1.0 0.21 0.16 2 Leselidze 0.18 0.59 0.15 0.16

3 Zugdidi 0.17 0.83 0.19 0.07 4 Chakva 0.20 1.02 0.24 0.12

5 Poti 0.18 0.91 — 0.18 6 Ochamchire 0.24 0.25 0.68 0.05 7 Machara 0.31 1.5 0.21

8 Batumi 0.27 0.97 0.34 9 Tkibuli 0.25 1.12 1.02 0.28

10 Tsikhisdziri 0.41 1.82 1.10 —

a The maximum 239Pu concentration was 4.58 Bq/kg of soil, near the village of Grigoleti. 4 5 0 POSTER PRESENTATIONS

Systematic monitoring of the levels of radionuclides resulting from the Chernobyl accident by our Institute and the Georgian Ministry of Health showed that the contamination was ‘spotty’, reaching 5.8 Ci/km2 in some places (see Table I). From the data it is clear that the contamination level in western Georgia was higher than 1 Ci/km 2, reaching 3 Ci/km2 in parts o f Abkhazia and Adjaria. Radio­ nuclide concentrations in agricultural products were measured. Very high concentra­ tions were found in some products: milk, ‘funduk’ nuts, grass, fruit, leaves of citrus plants and — especially — leaves of tea plants, in some samples of which the radio­ nuclide concentration was higher than 56 000 Bq/kg. Unfortunately, owing to the inadequacy of the material resources available, it was not possible to carry out a detailed study of the nature of the contamination. However, work is continuing. The results obtained show that there is an objective need to carry out a careful, comprehensive study even if the radioecological situation following the Chernobyl disaster has become somewhat more stable on the Black Sea coast of Georgia. POSTER PRESENTATIONS 4 5 1

IAEA-CN-54/35P

CONCENTRATIONS OF THE MOST IMPORTANT RADIONUCLIDES IN MILK IN SLOVENIA AFTER THE CHERNOBYL NUCLEAR ACCIDENT

P. JOVANOVIC, M. KANDUC Institute of Occupational Safety of Slovenia, Bohoriceva, Ljubljana, Slovenia

1. INTRODUCTION

As part of the permanent radioactivity monitoring and surveillance of the human environment in Slovenia, activity concentration measurements of radionu­ clides in samples of air, atmospheric precipitation, soil, drinking water, food and fodder were carried out [1]. This paper presents the results of the activity concentra­ tion measurements of radionuclides in milk, which is on the critical pathway for doses received by man. Four locations for collecting milk samples were chosen: Kobarid (northwest region), Bohinjska Bistrica (northwest region), Murska Sobota (northwest region) and Ljubljana (central region), where, owing to the varied con­ figuration of the Earth’s surface, atmospheric conditions differ. The average monthly activity concentrations of 134Cs, 137Cs, 89Sr and 90Sr in the samples of fresh milk taken in these locations for the period from the nuclear accident at Cher­ nobyl in 1986 until the end of 1993 are presented. For Ljubljana, activity concentra­ tions of 131I in milk are also given.

2. RESULTS

The highest activity concentrations of 137Cs and 134Cs were measured in Kobarid: 200 Bq/L and 99 Bq/L, respectively; the activity concentrations in the other locations were approximately 4 times lower. During the first month after the nuclear accident, the ratio between 137Cs and 134Cs was 2:1. The ratio of average monthly activity concentrations between 89Sr and 90Sr was 6:1. The highest average monthly 89Sr activity was measured in May 1986 in Bohinjska Bistrica at 6.5 Bq/L, whereas the daily activity concentrations were several times higher. Thus the differences in the contents of isotopes 134Cs, l37Cs and 90Sr in the samples of milk from various locations result from different concentrations of radionuclides in various elements of the food chain, as well as from the differences in the manner of feeding and in the type of fodder. In the months following the accident, the activity concentrations of radionuclides in milk were lower, owing to the recommendation of Governmental 4 5 2 POSTER PRESENTATIONS institutions that cattle should be fed only dry fodder. During November 1986 cattle were fed contaminated grass because of the lack of dry fodder and other feed, which resulted in higher concentrations during the winter months. 89Sr and 13iI were present in the milk during the first months after the accident only, whereas the other radionuclides are, owing to their longer decay times, still present. It should be emphasized that the concentrations of radionuclides in milk originating from mountainous regions (Kobarid and Bohinjska Bistrica) are higher than those from Ljubljana and Murska Sobota. In 1993 the average 137Cs activity concentration in milk from Kobarid and Bohinjska Bistrica was 2.9 Bq/L and from Ljubljana and Murska Sobota only 0.47 Bq/L and 0.36 Bq/L, respectively. The average annual 90Sr activity concentration in 1993 was lowest in Ljubljana, 0.16 Bq/L, and highest in Kobarid, 0.26 Bq/L. The 137Cs concentrations in milk are still slightly higher than they were before the Chernobyl accident. Because of their consumption of milk, the committed effective doses for children up to the age of 1 year, calculated for the period 1986-1993, were 0.51 mSv in Kobarid, 0.15 mSv in Ljubljana, 0.27 mSv in Bohinjska Bistrica and 0.10 mSv in Murska Sobota ( 131I is not included). About 45% of the dose was reached in the first year after the Chernobyl accident.

3. CONCLUSION

The ingestion doses calculated for 1 year old children show that the Govern­ mental recommendations for the restricted use of contaminated milk and vegetables reduced effective doses by at least a factor of 5. Yearly effective doses after 1993 due to ingestion of food are in the range of 5 p Sv, which is very similar to what we have estimated for the years before the Chernobyl accident.

REFERENCES

[1] KANDUC, M ., JOVANOVIC, P., KUHAR, B., “ Estimates of effective equivalent dose commitments for Slovene population following the Chernobyl accident” (Proc. Int. Symp. on Post-Chernobyl Environmental Radioactivity Studies in East European Countries, Kazimierz, Poland, 1990), Zaklad Poligrafii, Lublin (1991) 86-95. POSTER PRESENTATIONS 453

IAEA-CN-54/43P

COMPUTER SOLUTION FOR A CANCER EPIDEMIOLOGY RESEARCH DEPARTMENT IN THE REPUBLIC OF BELARUS

M.P. GÜNTER, J.P. BLEUER, T. ABELIN Department of Social and Preventive Medicine, University of Berne, Berne, Switzerland

J. AVERKIN, I. PRUDYVUS, S. RUBTSOV State Research Institute of Oncology and Medical Radiology, Lesnoy/Minsk, Belarus

1. INTRODUCTION

In 1965, a cancer registry based on handwritten cards was established in the Republic of Belarus. As the registration system was already computerized in 1978, hardware and software now need to be upgraded to meet international standards. There are 12 oncological dispensaries in Belarus, specialized in diagnosing and treating cancer patients. Patients suspected to suffer from a malignant disease are transferred to these centres for further diagnostic assessment. If the cancer diagnosis is confirmed, the case is reported to the central registry by means of a special form and additionally on computer floppy disk. The data include family name, first name, father’s name, sex, age, date of birth, nationality, profession, address, registration and diagnosis date, ICD-9 code, method of verification of the diagnosis, and the registration number in the dispensary.

2. OBJECTIVES

The main objective is to establish a research centre meeting international stan­ dards for cancer registration. It should serve as a basic tool for epidemiological research in the field of Chernobyl related health effects. Additional surveys concern­ ing other environmental hazards are currently being discussed. As a base for further scientific work, adequate computer equipment is crucial. Estimates based on case reporting in the past led to the following technical requirements: The system must handle 30 000 new entries per year, and data of at least 10 years must be accessible on line. High flexibility for easy adaptation to new retrieval tasks is required; various keys will be provided for linking the system with other databases. ICD-9, ICD-10 4 5 4 POSTER PRESENTATIONS as well as ICD-O (2) classifications must be handled. Data security requires an auto­ matic backup system, elaborate access controls and field encryption for sensitive data. To guarantee data consistency, automatic integrity checks will be implemented as far as technically feasible.

3. EVALUATION

The requirements mentioned above can be met by systems ranging from main­ frames to UNIX based workstations and PC networks. An evaluation was therefore carried out, focusing on database management software and disregarding the hard­ ware platform. Only widely used relational database products, with certain guarantee for support and update possibilities, were taken into account. Solutions demanding lesser hardware capabilities were preferred. Attention was paid to the presence of powerful development tools (4th generation language or similar) and extendability: Further extensions such as implementation of a client/server architecture should be possible without loss of hardware and software, and with only a few changes in the programmed code. Each software product meeting the criteria mentioned above was grouped together with its best suitable hardware platform, operating system, network and network operating system. The systems were then evaluated, with focus on com­ parison of performance and cost of the various options.

4. RESULTS AND DISCUSSION

Comparison of cost and perfomance favoured a PC based system. As Macintosh is not well known in Belarus, Intel based systems were chosen. There is good general knowledge of PCs in Belarus; this therefore seemed an ideal solution also with regard to maintenance. Speed of access and processing time were less of a limitation than availability of repair services and replacement hardware. Further­ more, in case of hardware failure, the server can be easily replaced and the whole system will not be out of operation for a long period. Widespread programming skills in XBase made us favour software of this standard. Since WHO had defined it as one of the standards in the framework of IPHECA1, compatibility with dBase IV was yet another reason for joining the XBase family. Last but not least, the widespread knowledge of the chosen software ensures continuity of the work in case of changes in staff. The final system design includes a file server for central data storage. Meant solely for these limited tasks, the system will include data entry stations and workstations for analysis and maintenance. All computers will be inter-

1 WHO International Programme on the Health Effects of the Chernobyl Accident. POSTER PRESENTATIONS 4 5 5 connected with an Ethernet, thus providing access to the main database. Data backup must be appropriate to the high level of importance and great volume of data. Trans­ action logging during the day and incremental backup of the data every night will therefore be implemented. FoxPro provides powerful development tools, similar to a 4th generation language, which guarantee rapid progress in development of applications. The exten­ sion capacity is increased by the possiblity of linking FoxPro with one of the most powerful programming languages, C. The possibility to generate executable pro­ gram files and high performance in processing allow the integration of older machines into the system. The requested extendability towards a client/server architecture is provided in version 2.6 of FoxPro.

ACKNOWLEDGEMENTS

Thanks are due I.M. Drobyshewskaya, Minister of Health of Belarus, and K.V. Kasakov, former Minister of Health, for supporting international collaboration in Chernobyl related health research; J.A. Korotkevich, Research Institute for Oncology and Medical Radiology, for his collaboration; F. Gurtner, M. Egger, W. Schäppi, and J. Lambert, University of Berne, for their work in the Project Group; and the Swiss Federal Office of Public Health for financial support.

REFERENCES

[1] FERNANDEZ, E.B., et al., Database Security and Integrity, Addison-Wesley, Read­ ing, MA (1981). [2] INTERNATIONAL AGENCY FOR RESEARCH IN CANCER (WHO), Cancer Registration, Lyon (1991). [3] KORTH, H.F., et al., Data Base System Concepts, McGraw-Hill, New York (1988). [4] OPPLIGER, R., Introduction to Computer Security, Vieweg, Braunschweig (1992). [5] STATISTICAL ANALYSIS AND QUALITY CONTROL CENTER, Quality Control for Cancer Registries, Seattle, WA (1985). [6] VORKAUF, H., Anonymization of Data as a Measure to Fulfil the New Privacy Act, Federal Office of Public Health, Head of the Unit for Biostatistics, Berne (unpublished).

HOWTO ORD R IAEA PUBLICATIONS

No. 3, July 1996

☆ ☆ In the United States of America and Canada, the exclusive sales agent for IAEA publica­ tions, to whom all orders and inquiries should be addressed, is: UNIPUB, 4611-F Assembly Drive, Lanham, MD 20706-4391, USA ☆ ☆ In the following countries IAEA publications may be purchased from the sources listed below, or from major local booksellers. Payment may be made in local currency or with UNESCO coupons. AUSTRALIA Hunter Publications, 58A Gipps Street, Collingwood, Victoria 3066 BELGIUM Jean de Lannoy, 202 Avenue du Roi, B-1060 Brussels BRUNEI Parry’s Book Center Sdn. Bhd., P.O. Box 10960, 50730 Kuala Lumpur, Malaysia CHINA IAEA Publications in Chinese: China Nuclear Energy Industry Corporation, Translation Section, P.O. Box 2103, Beijing CZECH REPUBLIC Artia Pegas Press Ltd., Palác Metro, Narodni tr. 25, P.O. Box 825, CZ-111 21 Prague 1 DENMARK Munksgaard International Publishers Ltd., P.O. Box 2148, DK-1016 Copenhagen К EGYPT The Middle East Observer, 41 Sherif Street, Cairo FRANCE Office International de Documentation et Librairie, 48, rue Gay-Lussac, F-75240 Paris Cedex 05 GERMANY UNO-Verlag, Vertriebs- und Verlags GmbH, Dag Hammarskjöld-Haus, Poppelsdorfer Allee 55, D-53115 Bonn HUNGARY Librotrade Ltd., Book Import, P.O. Box 126, H-1656 Budapest INDIA Viva Books Private Limited, 4325/3, Ansari Road, Darya Ganj, New Delhi-110002 ISRAEL YOZMOT Literature Ltd., P.O. Box 56055, IL-61560 Tel Aviv ITALY Librería Scientifica Dott. Lucio di Biasio “AEIOU”, Via Coronelli 6,1-20146 Milan JAPAN Maruzen Company, Ltd., P.O. Box 5050, 100-31 Tokyo International MALAYSIA Parry’s Book Center Sdn. Bhd., P. O. Box 10960, 50730 Kuala Lumpur NETHERLANDS Martinus Nijhoff International, P.O. Box 269, NL-2501 AXThe Hague Swets and Zeitlinger b.v., P.O. Box 830, NL-2610 SZ Lisse POLAND Ars Polona, Foreign Trade Enterprise, Krakowskie Przedmieácie 7, PL-00-068 Warsaw SINGAPORE Parry’s Book Center Pte. Ltd., P.O. Box 1165, Singapore 913415 SLOVAKIA Alfa Press Publishers, Hurbanovo námestie 3, SQ-815 89 Bratislava SPAIN Díaz de Santos, Lagasca 95, E-28006 Madrid Díaz de Santos, Balmes 417, E-08022 Barcelona SWEDEN Fritzes Customer Service, S-106 47 Stockholm UNITED KINGDOM HMSO, Publications Centre, Agency Section, 51 Nine Elms Lane, London SW8 5DR

☆ ☆ Orders (except for customers in Canada and the USA) and requests for information may also be addressed directly to: Sales and Promotion Unit International Atomic Energy Agency Wagramerstrasse 5, P.O. Box 100, A-1400 Vienna, Austria Telephone: +43 1 2060 22529 (or 22530) Facsimile: +43 1 2060 29302 Electronic mail: [email protected] (О < t 00 <м 9 in О)

VIENNA' ( ш а 92=0=103096=7 ÍE E ffl 0 0 7 4 = 1 8 8 4