IAEA-TECDOC-460

CONCEPT AND APPROACHES USED IN ASSESSING INDIVIDUA COLLECTIVD LAN E DOSES FROM RELEASE RADIOACTIVF SO E EFFLUENTS

A TECHNICAL DOCUMENT ISSUEE TH Y DB INTERNATIONAL ATOMIC ENERGY AGENCY, VIENNA, 1988 CONCEPT AND APPROACHES USED IN ASSESSING INDIVIDUAL AND COLLECTIVE DOSES FROM RELEASES OF RADIOACTIVE EFFLUENTS IAEA, VIENNA, 1988 IAEA-TECDOC-460

Printed by the IAEA in Austria June 1988 PLEASE BE AWARE THAT ALL OF THE MISSING PAGES IN THIS DOCUMENT WERE ORIGINALLY BLANK IAEe Th A doe t normallsno y maintain stock f reportso thin si s series. However, microfiche copie f thesso e reportobtainee b n sca d from

INIS Clearinghouse International Atomic Energy Agency Wagramerstrasse5 P.O. Box 100 A-1400 Vienna, Austria

Orders shoul accompaniee db prepaymeny db f Austriao t n Schillings 100,- in the form of a cheque or in the form of IAEA microfiche service coupons which may be ordered separately from the INIS Clearinghouse. FOREWORD

n implementatioa s A e IAEth A f Basio n s Safety Standard r Radiatiofo s n Protectio Agence th ] y[1 ninitiate expanden a d d programme th 198n i en o 3 Radiation Protectio Generae th f o nl Public. This sub-programm Agence th f yo e has three components, namely (1) Limitation of the Releases of Radioactive Effluents into the Environment, (2) Monitoring for the Radiation Protection of the General Public, and (3) Application of the Radiation Protection Principles to the Sources of Potential Exposure.

Unde e firsth r t component Limitatioe th n ,o f Releaseo n Agence th s y issue 198n i d 6Safeta y e PrincipleGuidth n o e r Limitinfo s g Releasef so Radioactive Effluents into the Environment, Safety Series 77. This Safety Guide was the revised version of the earlier issued Safety Series 45. In addition e Agencth , y issue 198n i d 5Safeta y Guide, Safety n Serieo , 67 s Assigning a Value for Transboundary Radiation Exposure.

e applicationo guidT th n o ee principleth f o s r limitinfo s g radioactive releases contained in Safety Series 77, the Agency is in the process of preparing a number of safety guides. The first one is this present document which deals with the principles aspects of the methods for the assessment of e individuath d collectivan l e dose. Another document whic undes i h r preparation will be a safety guide on the detailed methdologies for the assessmen f individuao t d collectivan l e dose d thian ss documen beins i t g prepare y revisinb d e existinth g g Safet e Generiyth Serien o c7 5 sModel d an s Parameter Assessinr fo s Environmentae th g l Transfe f Radionuclideo r s from Routine Releases o safetTw . y guide e undear s r preparatio e applicatioth n o n n of the principles for limiting radioactive releases to specific cases. The first one is on the mining and milling of radioactive ores, which is now under publication nucleae th n o e other Th e .poweon r r plant reprocessind an s g plants, which is in an advance stage of completion. Another safety guide, which is under preparation is on the Establishment of Upper Bounds to Doses to Individuals from Global and Regional Sources. Several other documents are relate o thit d s present documen d thes an e tSafet ar e y Serie . 50-SG-SsMo n 3o the Atmospheric dispersio Nuclean i n r Power Plant Sitin Safetd an g y Serie. sNo 50-SG-S Hydrologican o 6 l Dispersio Radioactivf o n e Materia Relation i l o t n Nuclear Power Plant Siting. e presenTh t document aim t givina s a generag l guidanc o thost e e responsible for establishing programmes for the determination of individual dose s wela ss collectiva l e dose n connectioi s n with licensina a sitgr fo e nuclear installation. This document will be supplemented by a series of document n guidanco s d specifian e c procedure r thosfo s e responsiblr fo e implementin e programmth g f doso e e assessmen e environmentth n i t .

The document is concerned with the principles applied for calculating individual and collective doses from routine releases of radionuclides to the atmosphere and hydrosphere but not releases directly to the geosphere, as in waste management. These areas will be covered by other Agency publications.

The document has been prepared with the help of two Advisory Group meetings and Consultants meetings. The final compilation of the document was the responsibility of Mr. J.U. Ahmed of the Division of Nuclear Safety.

EDITORIAL NOTE

preparingIn this materialpress, the International the for staff of Atomic Energy Agency have mounted and paginated the original manuscripts and given some attention to presentation. The views expressed necessarilynot do reflect governmentsthosethe of Memberthe of States or organizations under whose auspices the manuscripts were produced. The use in this book of particular designations of countries or territories does not imply any judgement by the publisher, the IAEA, as to the legal status of such countries or territories, of their authorities institutions and delimitation the of or theirof boundaries. The mention specificof companies theirof or products brandor names does implynot any endorsement or recommendation on the part of the IAEA. CONTENTS

Chapte . 1 INTRODUCTIOr N ...... 7 .

1.1. General principle f sourco s individuad ean l related dose assessments ...... 7 . 1.2. Principle f assessino s g radiation exposure ...... 9 .

Chapter 2. ASSESSMENT OF DOSE TO THE CRITICAL GROUP ...... 19

2.1. Principle f calculatinso g exposure criticae th o t s l group ...... 9 1 . assessmene 2.2Th . f exposureo t s from radionuclides released to the atmosphere ...... 19 2.3. The assessment of exposures from radionuclides released to the aquatic environment ...... 26

Chapter3. ASSESSMEN COLLECTIVF TO E DOSE ...... 1 3 .

3.1. The concept of collective dose ...... 31 3.2. Methods for calculating collective doses ...... 33

Chapte . 4 rAPPROPRIATENES MODEF SO L SELECTIO UNCERTAINTIED NAN S IN DOSE ASSESSMENTS ...... 41

4.1. Introduction ...... 1 4 . 4.2. General problems encountere dosn di e assessments base modeln do s ...... 2 4 . 4.3. Factors affectin uncertainte th g f resultyo s ...... 4 4 . 4.4. Sensitivity analysis ...... 6 4 . 4.5. Uncertainty analysis ...... 7 4 .

REFERENCES ...... 49

LIS PARTICIPANTF TO S ...... 5 5 . Chapter 1

INTRODUCTION

l.1 General principles of source and individual related dose assessments

A practice involving exposure o radiatiot s n usually gives a ris o t e distribution of doses, and thus of detriment, and to a distribution of benefits fropracticee th m n generaI . l thes o distributiontw e e quitar s e different and, therefore e distributio,th f benefit o nusee b o justif t dn ca s y e distributioth f detrimento n e sris th o onleac t kf i yh individua s smalli l , not exceeding levels which could be regarded as acceptable. The dose limits recommende e ICRth P y b d[11 e intende]ar o ensurt d e this leve f protectioo l n even for the most highly exposed individual.

The dose limits, however, are not intended to be values used for design installations, or for planning operations, but are rather the lower boundary regioa f o unacceptablf o n e value norman i s l circumstances. Values above th e limits are specifically not permitted, but values below the limits are not automatically permitted. In this sense, the limits are only the constraints with the overriding consideration that the radiation doses should be kept as low as reasonably achievable, economic and social factors being taken into account ("Optimizatio protection)f o n . dose th e f o limit e constraints us a s e Th r optimizatiofo s n presents some conceptual difficulties, especiall e cas th f genera o en i y l exposuref o s membere publicth e dosf Th o s.e n individual-relatelimia s i t d requirement, while optimizatio source-relatea s i n d requirement. Becaus e dosth e e limit appliee combineth o t s d exposure from many practices cannot i , usee tb o t d limit a given single practice even when it is more restrictive than results of optimization. In fact, exposures at the limit from one single practice would leav margio en r othefo n r practices which expose r coul,o d expos e samth ee critical group.

e probleTh f overlamo f exposurepo s from different practicet no s i s restricte y givean o nt d instan f timeo t . Each yea f operatioo r a f o n continuing practice can cause exposures which would be delivered in the future and which would add to the contributions of other years of operation e practiceoth f e dosTh e. rate resulting froe combinemth d l effecal f o t such annual discharges will increase, eventually reachin a steadg y state. This maximu me annua valuth f lo e dose from continuing practices could occur far in the future, and could also be maintained over considerable periods of time.

e purposth r f constrainino eFo g optimizatio e protectioth f o n f o n specific practices t seemi , s reasonable, therefore, that national authorities select "dose upper bounds" whic e smalar h l fraction e dosth e f o s limit, allowing for the overlap in exposure from various continuing practice d alsan so reserving some margi r unforeseefo n n future requirements which otherwise woul precludede b d .

The limitation of radiation exposures from the operation of specific nuclear fuel cycle installations therefore involve combinatioa s f o n assessments of individual exposures from the installations and the contribution of individual doses from other sources, usually estimated by source-related assessments (i.e. calculation of collective exposures). Beninson [12 discusses ]ha e limitatioe applieb th w thi n r ho dca s fo d f o n future radiation exposure from the present operation of the nuclear fuel cycle. The Agency also discusses some of these problems in its publication principlee oth n r settinfo s g global dose upper bounds (8).

The limitation of radiation exposure is discussed in more detail in the Agency's report on principles for limiting releases of radioactive effluents into the environment [3]. This document is concerned with the general principles for assessing individual and collective doses, and the practical guidance wil e founb l othen i d r Agency documents.

As discussed in the IAEA Safety Series report No. 77, [3], optimization should also be used in establishing the annual limit of discharg f radioactivo e e materials inte environmenth o orden i t o keet e r pth resulting exposures as low as reasonably achievable. Most optimization techniques requir e estimatioth e e totath lf o nradiatio n detriment, whics i h generally assumed to be proportional to the collective effective dose equivalent commitment (defined as the time-integration of the collective effective dose equivalent rate). Some of the components of the collective effective dose equivalent commitment y als e optimizatioe needema sb o th n i d n procedures :

- when very long-lived radionuclides are involved, it is often assumed thae long-terth t m componen e collectivth f to e effective dose equivalent commitmen e sameth s ,i t irrespective of the form and location of release; it follows that, in that case, onle incompletyth e collective effective dose equivalent commitment needs to be calculated, up to a point in time where e site-specifith l al c estimate f collectivo s e effective dose equivalent rate have merged. variatioe th n- collective witth time f th ho e e effective dose equivalent rat y als ema needee b o r releasefo d f long-liveso d radionuclides, whic y exposhma e populations over long periods of time, if the question is raised of whether detriment expressed at future times should be given less, the same, or more weight than detriment expresse neae th r n i futured .

In addition, the distribution of individual doses may also be an input in the optimization procedures if it is deemed that the cost of radiation detrimen componena s ha t t functioa tha e individuas i tth f o n l dose.

1.2 Principles of assessing radiation exposure

Schematically, the source-dose relationship can be represented in the following way.

Table 1.1 Schematic representation of the source-dose relationship

Stage Processes Quantities

Source Transport Annual Dispersion Releases Deposition Sedimentation Bio-accumulation

Environmental Environment Food, air and water concentrations intakes occupancy and radiation levels

Exposure Dose/intake conversion

Dose Annual dose Collective to Critical dose Group commitment per year of release The transfer processes between the source and the environment are represente y predictivb d e models, each involvin a largg e numbef o r parameters.

e transferTh s betwee e alse environmenar th no n discussema d an t n i d the following section where they constitute a common factor between the two type f assessmento s . Ther e threar e e main mode f exposureo s : external irradiation, internal irradiation following inhalatio d ingestioan n f o n radionuclides.

e calculatiod pointh en f o te Th ne effectiv th wil e b l e dose equivalent, as this quantity is taken to be proportional to the radiation risk [5J.

1.2.1 Overview of Environmental Transfer Models

The estimation of doses for the purpose of comparison with upper bound generalls i s y base modeln o d d selectean s d parameter values expected o lea t o conservativt d e estimate f actuao s le othedosesth rn O .hand , optimization needs comparisons base n beso d t estimates, using modela d an s data base adapte o providt d e expecteth e d dose s realisticalla s s a y possible. Thes o purposetw e f doso s e assessment therefore nee differena d t selection of parameter values and in many cases also qualitatively different models methode Th . s recommende r individuafo 3 chapten d i dd an an - l 2 r collective dose assessments respectivel e intendear y o covet d r theso tw e types of applications.

Radioactive materials released to the environment give rise to radiation doses to man through a variety of pathways. Representations of some of these pathways are shown in figures 1.1 and 1.2, relating to atmospheri d aquatian c c discharges, respectively e typ Th modef .o e l that shoul usee b d d depende informatioth n o s n required e characteristicth n o , s of the radionuclide concerned and on the way in which it is released into e environmentmaie th th n f o consideration e On . whethes i s r time-dependent informatio requires i n r whetheo d r some for f steadmo y e b stat n ca e assumed e latteth n i r. case e calleModelus ar e r dfo s concentration factor or equilibrium models while time dependent models are referred to as dynamic or systems analysis type models. A publication by ICRP [13] discusses these o typetw modelf o s d theiran s limitation d strengthan s s when appliea o t d variet f differeno y t situations.

10 ORIGINAL PRIMARY EXPOSURE PATHWAYS DOSE SOURCF EO RECIPIENT EXPOSURE

EXT. IPR. FRO A DISTANCM E (PLUME) EXT. IRR. FROM SUBMERSION

SOIL EXT. IRR. FROE SURFACEMTH , -} F EARTO H TRANSFER

VEGETATION

BIOACCUMULATION ~> CONSUMPTION

INHALATION (BIOACCUMULATION)

Figure 1.1: Schematic representatio f exposuro n e pathways followine th g releas f radionuclideeo atmospheree th o t s .

The concentration factor (CF) metho moss i d t widely user fo d assessing the consequences of planned chronic releases of radioactive materials n thiI . s method, simple multiplicative coefficient e useo ar st d obtai e concentratioth n f radionuclideo n poine th f intako t a s many b e n I . the general case these coefficients called transfer coefficients, concentration factors or, in some cases, bioaccumulation factors, are defined as quotients between time integrals of quantities in different parts of the system.

In the case of continuous release at a constant rate, and provided that the environmental conditions governing the transfer processes remain relatively constant with timr tha(o ee adequatelt b then ca y y characterized by average parameters), it can be shown that the same transfer coefficients cover the steady state condition.

Thi usualls i s e situatioyth r continuoufo n s routine releases, where for relatively short-lived radionuclides constant relationships can be assumed between the rate of discharge and the concentration of the radionuclides in the environment, since the steady state may be reached rapidly.

11 PRIMARY EXPOSURE PATHWAYS DOSE RECIPIENT

0 f-

0 Q

Figure 1.2: Schematic representation of exposure pathways following the releas f radionuclideo e s int watea o r body.

e effectivTh e dose equivalent fro givea m n discharg a singl f eo e radionuclide will arise from many exposure pathway sn serie i whic d e an sar h in parallel r pathwayFo . n seriei s e totath s l transfer facto branca f ro h is the product of the transfer factors involved. The total transfer factor f severao e transfel th branche f o rm su parallen factore i s e th th f o s i l branches.

12 The assumption of a constant release rate together with the use of the concentration factor approach should be appropriate, in most cases, for predicting the average exposure per year. However, if extreme seasonal variation in the dispersion is combined with a variable release, averaging over both parameters might lea o significant d t over-or under-estimatiof o n the exposure. Care must therefore be taken in choosing representative value f dispersioo s n parameters when there significanknob e ar o ewt t fluctuations. Examples of this are seasonal variations in river flow rate or in prevailing wind directions.

e disadvantagTh methoF C e thas i th d doet t i trepresenf o eno s e tth time-dependent behaviou f nuclideo r environmentan i s l materials i.e. their rat f build-uo e r removapo l froenvironmene th m t followin e introductioth g n r stoppino practicea f o g . Thi particularls i s y importan r shortfo t , unplanned releases of radionuclides, where control decisions might be based on the time dependence of the activity of various nuclides in specified materialse applicablb F approact C no e Th y thin .i eh ma s situation. Also, when radionuclides from planned releases accumulate in the environment or persist for long times after their introduction, it may be necessary to introduce time dependence into the CF approach or to use more complex time-dependent models.

n summar I consideres i t i y d tha generan i methoF tC e s adequatli th d e r dosfo e assessments following planned release f radionuclideo s s inte th o environment particularl calculatinn i y g Derived release Limitsth f n I ca e. be assume essentialle b o t d y continuous F typC e ,th mode adequates i l d ,an relativs it o t ee simplicitdu y probably preferabl e estimatioth r fo e f o n transfer throug environmene th h mann i t y cases.

Some radionuclide uniformle b n s ca suc C tritius ya h d an m 14 distributed in the environment. In these cases specific activity models can als usede b o , recognizin limitatioe th g f suco n h models [14].

1.2.2. Dose Assessments

In the following sections general discussion on the problem of calculation dosef so s from externa d internalan l irradiation pathwaye ar s given. It is placed here as it contains features that are common to both individua d collectivan l e dose assessments.

13 1.2.2.1 External irradiation

Externa le presenc th irradiatio o t f bete o d gamm du an a s ai n emitter environmentan i s l materials (air, water, soil, etc.). Models taking into accoun e characteristicth t e emittinth f o s ge sourc th nuclide f o e d an s geometry yiel n estimata de absorbe th f e o point th d t f dosa o s r e ai rat n i e interest e conversioTh . n from absorbed effectivo dost er ai rat n ei dose equivalent can then be carried out.

Factors that are of importance in converting the absorbed dose in air to effective dose equivalent are:

a conversion factor effectivo frot mr ai dos n ei e dose equivalent;

shielding froexternae th m l radiatio buildingy nb other o s r barriers terrain (site specific);

the time spent indoor yeara n i s.

Detailed information and discussion on the different physical parameters that are of importance for the conversion factor can be found in [14-20] e shieldinTh . g factor buildingr fo s s will vary considerably from place to place depending upon structure of the building, the composition on the building material and how the activity is depositied inside the building and on roofs and walls. The time spent indoors per year will also depend on local factors such as climate as well as on occupation and particular habits. More informatio e founb referencen n i d ca n s [20].

1.2.2.2 Internal Irradiation

Internal irradiation occurs followin e intakth g f radionuclideeo s into the body, primarily through inhalation or ingestion. It should be noted that internal irradiation followin n intaka g f radioactivo e e materials is protracte n timi d e afte e intaketh r e effectiv.Th e dose equivalenr tpe unit intake by an individual will depend on the metabolism, age and life-expectanc e individuath f o yhalf-life th wels e la s th a l f o e radionuclide concerned. Assessment usuallt no e ysar r carriefo t ou d particular individual t rathebu s mea e th rn absorbed dos calculateds i e r Fo .

14 this calculation, representative varioue valueth r ssfo factors, sucs ha those relate metabolismo t d e take,ar n eithe completr rfo e populationr so for particular subgroup populatione th n si .

A number of factors have to be taken into account in determining the e effectivvaluth f eo e dose equivalen r unipe t t intak f giveo e n radionuclide: e retentioTh e radionuclidth f ) o ni e variou th n i es organs whica s i h functio f individuao n l metabolis wels a ms radioactiv a l e decay.

ii) An assessment of the irradiation of individual organs and tissues. Organs are irradiated both from the radiations resulting from transformations occurring in the organ itself and from those occurring in surrounding organs. Irradiated organs are referred to as target organs and those in which transformations occur are referred to as source or&ans. The absorbed dose in both target and source organs depende physicath n o s l propertie e radionuclidth f o s e as e siz d spacine variouwelth an eth s la f o gs organs.

iii) The absorbed doses in the various organs are converted to the effective dose equivalent. This procedure involves weightine th g organ dose d applicatioan s e appropriatth f o n e qualitye factoth r fo r radiation of interest.

This procedure has been adopted by ICRP [21] to calculate dose equivalents in organs; approximately 30 organs are considered as source and/or target organs e effectivTh . e dose equivalen alss i t o calculated according to the definition given in ICRP 26 [11]. ICRP publication 30 and supplements [21] give effective dose equivalents per unit intake by inhalation and ingestion for a large number of radionuclides and for an integration period of 50 years following the time of intake; these are called committed effective dose equivalent r unipe s t intake.

The committed effective dose equivalents per unit intake given in ICRP -30 are intended for use in estimating the exposure of adult radiation workers. In assessing the exposure of the general public account has to be taken of the age of the exposed population as well as the likely chemical form of radioactive material in the environment. Due to their smaller body size, effective doses per unit intake by children are generally greater than thos f adultso e . Thi oftes i s n counteracte y loweb d r inhalatiod an n

15 ingestion rates, the exposure of infants or children can lead to their receiving higher indiviudal doses than adults. In assessing the exposure of critical group e intaketh s d effectivan s e dose r unipe s t intake appropriate o different t ages should ideall e usedb y . Thi oftes i st practica no n a d an l suggested approach is to use a correction factor applied to published values of the Annual Limits of Intake ALIs, for occupational exposure. For example e IAE] hav,th [4 A e suggested usin e hundredtfactoa g on e f th o r f o h occupational ALI for critical groups involving infants and children. However, such a factor leads to predictions of exposure which are over estimates in many cases or which, alternatively in a few cases may be under - estimates. Such correction factors should therefore be used with caution and where possible actual effect-ive dose e groupequivalentag f e o sth r fo s interest should be calculated.

n assessinI g collective generalls dosei t i s y adequate us o t e effective dose equivalent r unipe s t intak r x adultfo ese d s an only e Ag . differences may affect the accuracy of the commonly used dose conversion factors provided in ICRP Publication No. 30 [21]. In addition, some evidenc bees ha en reported thae ICRth tP value y under-estimatsma e th e committed effective dose equivalen o referenct n resultinema g from intake of several radionuclides [22]. Nonetheless, it is recommended that the standard ICRP value f doso s e conversion factor usede b s , pending international acceptance of newer scientific data.

The most important intake routes are inhalation and ingestion. For both these exposure pathways formulas and methods exists for converting the ground leve r concentratioai l e intakth differena r vi eo n t foodstufn a o t f effective dose equivalent from one year of exposure.

Ingestion rate f variouo s s terrestria d aquatian l c foods will vary widely with location depending on the availability of the particular foodstuff as well as individual preference. It should be noted that variation within individual s greatt countriegreatera no e b f i , y s ,a sma those observed between geographical regiono economit e du sr sociao c l factors and residence location. In most countries, Government agencies are likely to have information on the average intake of the important components of the diet.

16 r estimatinFo g individual exposure, higher than average ingestion rates should preferably be obtained on local basis by means of dietary surveys. Such surveys o identifshoult m ai e dcritica th y l group, i.ee .th individuals in the population who as a result of their habits or location are likely to be representative of those people receiving the highest radiation expsoures from a given source. In the absence of relevant information, conservative estimates of ingestion rates should be used for the assessment of individual doses in the critical group whereas realistic estimates are required in order to optimize the level of protection.

17 Chapter 2

ASSESSMENT OF DOSE TO THE CRITICAL GROUP

1 2. Principle calculatinf so g exposure criticae th o b s l group

As outlined in Chapter 1 radionuclides released into the environment can potentially lead to the exposure of man to radiation via a number of pathways. Since the determination of release limits requires the prediction of future exposures of the public, mathematical models of these pathways are required to predict the radiation doses following releases of radionuclides intenvironmente th o . These environmental models, whic e simplifiear h d mathematical representation of the actual transport, transfer, and exposure processes describee ,ar generan i d l term n i this s Chapter A significan. t fraction of the relevant information has been taken from IAEA Safety Series Ho. 57 [5].

e actuaTh l doses receive membery publie b d th f co s will vary widely depending on such factors as age, metabolism, dietary and other habits, as variationn o wel s a l n theii s r environment normae th s li t practicI . n i e radiation protection to account for this variability by identifying an appropriate critical group e grouTh . p shoul e representativb d f thoso e e individuals in the population expected to receive the highest dose equivalents frosource th m f radiatioeo n under consideration must I e .b t small, enouge relativelb o t h y homogeneous with respec ageo t t , died an t those aspect behaviouf so r that affec dosee th t s received.

One important quantity to assess, when considering releases into the environment, is the dose committed in the critical group by one year of release maie Th .n purpos f theseo e assessment ensuro t s si e compliance with source upper bounds, as discussed in detail in the IAEA Safety Series 198, Mo77 . 6 [3].

Assessmene Th 2 2. Exposuref to s from Radionuclides Releasee th o t d Atmosphere

Radionuclide releasee b y sma d inte atmospherth o e routinely eithes a r the result of operations from various types of installations or through the emanatio f gaseo n s from radioactive material e grounth n i ds (e.g. radon from

19 mill tailings). As the cloud travels downwind the exposed population will initiall e irradiateb y o principatw y b d l routes: external irradiatioe du n o radioactivt e material e clou d th interna an n di s l irradiation following inhalatio f suco n h materials e transfeTh . materia f o r l frocloue th mo t d the groun y resule furthema dth n i t r exposur f peoplo e y threb e e other routes: external irradiation by the deposited radioactive materials, the inhalatio y sucan hf o material-subsequentln y resuspende e atmosphereth n i d , e transfeanth d materiaf o r l throug e environmenth h wateo t d foodstuff an r s e consumeb whic y hma many b d.

2.2.1 Atmospheric Transfer Processes

e airbornTh e radionuclides wil e transporteb l d downwin d dispersean d y b d e normath l atmospheric mixing processes. Radioactive material wile b l removed froe plummth e durin transis depositiod it g an y bott b t we h n processes, which include gravitational settling sorption onto ground and vegetation, and rainout or washout in precipitation. Figure 2.1 illustrates several of the relevant concepts.

r -•..-. ' '

RAINOUT

WASHOUT Fig. 2.1 Atmospheric dispersion and removal processes.

e distancTh e over which radionuclide e transporteb y e ma s th n i d atmosphere depend mann o s y factors, suc radioactivs a h e half-lifee th , physical and chemical states of the nuclides, meteorological conditions, topography d depositioan , n processes r manFo .y nuclides releasen i d airborne effluents as particulates the majority of the released material may

20 be removed froplume mth e withiw hundrefe a n d kilometers, eithey b r radioactive depositioy decab r o y n processes r sucFo .h nuclides only local and regional exposure sconsiderede b nee o t dlimitea r Fo .d numbef o r nuclides, because of their inert chemical behaviour, long radioactive half-life rapir ,o d exchange betwee e atmospherth n othed an e r sectore th f so environment, a significant fraction of the released material may be much more widely dispersed r such.nuclidesFo . r ,K especiall , C , H y , globaI andl 12populatio9 n exposur y alsema oe considered b nee o t d . Dispersion model r calculatiofo s f locao n regiona d an l l exposuref o s critical groups are considered in this chapter, while models for global dispersion and consequent exposures are discussed in chapter 3.

Material discharged into the atmosphere is transported longitudinally by wine dispersed th an d d both laterall verticalld an y turbuleny yb t diffusion. Vertical diffusio e limitee temperaturb th y y nma b d e gradients e atmosphereith n effluenn a f I . t plum insertes i e d inte atmospherth o e temperaturr ai wher e eth decreasins i e g rapidly with altitude, vertical diffusion upward is essentially unbounded. However, where the plume is inserted int n atmosphera o temperaturr e ai wher e th e decreasins i e g only slowl temperaturr ai wherr y o e th e increasins i e g with altitude, vertical diffusion wil confinee lb limitea o t d d laye f atmosphereo r . Other factors which effect both lateral and vertical diffusion rates include the temperatur velocitd effluene an e th f yo t stream presence ,th f atmospherio e c disturbances, and the effects of topography and man-made structures [23,24,4,25,26]. In any case it should be kept in mind that large uncertainties in calculated airborne concentrations will often exist for any particular locatio r poino n timen i tt distanc A . e beyon tena d f so kilometers o order, tw thi r f o o ss e uncertaint on mucs a s a he b y yma magnitude for the calculated short-term concentrations [21}. However, for calculated annual concentrations uncertainte ,th mann i ys i ycase s much smaller, sometimes only by a factor of 2.

There are various models in existence to determine the degree of atmospheric dispersion and the amount of deposition onto the ground following release f radionuclideo s e atmosphereth o t s . These modelf o e ar s varying type d degreean s f complexito s type th e t thabu ys foun ha t d widest application, particularly in relation to routine releases, is the gaussian plume model. Descriptions pf gaussian plume models and values for associated paramete e givenumbea ar rn i nf publication ro s [5,25,28,29].

21 The endpoints of atmospheric dispersion models required for critical group dose assessment are the average concentration of a radionuclide in air at a point and the total deposition rate of the radionuclide. These two quantities then form an input to- the calculation of critical group doses froe varioumth s exposure pathways mose Th .t importan f theso te ar e external irradiation from material in the air and on ground, internal irradiation from inhalation of material in the cloud and resuspended from the ground and internal irradiation from ingestion of contaminated foods.

2.2.2 External irradiation from materia e clouth n di l

As discussed in section 1.2.2.1 the effective dose equivalent from external irradiatio calculates i n d fro e absorbemth dr takin ai dos n i ge account of the shielding afforded by being indoors for part of the time.

Two approaches are available for the estimation of the absorbed dose in air from a dispersing cloud; those are the semi-infinite and finite cloud models. The former is only adequate when the radionuclide is uniformly distributed in the atmosphere or when the photon energy is sufficiently low that reasonabla thi s i s e approximation ovevolume plumea th r r f o eFo . j emittin e mosth f gto radionuclides this situation only occur severat a s l tens of kilometres from the release or at even greater distances for release particulan i s r conditions manr Fo y. situations therefore, particularly releas e closth o t ee point finita , e cloud model mus usede tb .

e finitTh e cloud model involves simulatin e clounumbea th gy b df o r small volume sources and integrating over these sources. The integration is carrie evaluates t ovei l spacou dal rd an e d numerically e evaluatioTh . s i n extremely complex but it is possible to simplify calculations of critical group dose thiy b s s metho makiny b d numbea g f approximationo r s [30].f I doses calculated using this procedure approach the dose limit or appropriate source upper bound then fuller calculations shoul undertakee b d usiny b n g r examplfo modee th e l give IAEn i n A Safet Refd an y . ] Serie[9][5 7 .5 s

The semi-infinite cloud mode e radiatio bases th i l n o d n froe clouth m d being in radiative equilibrium so that the energy abosrbed by a given volume elemencloue th df to equal s that emittee samth e y b elementd . Neae th r ground, however, the radiation source represents only one half of the space, so that energy absorbed in a given volume element is only one half of the energy emitted by the same element.

22 A modified version of this model can be used to evaluate the beta irradiation dose to the skin» as beta particles have a range of only a few e pointh t a beinmeterr ai g n i sconsidered . References shoul e consulteb d d for details [5,31,32] .

2.2.3 Internal irradiation following inhalation of material in the cloud

e calculatioTh e committeth f o n d effective dose e equivalenon o t e tdu year inhalation of material in the cloud only depends upon the time-integrated activity concentration of ground level air over one year, the breathing rate and the committed effective dose equivalent per unit activity inhaled.

2.2.4 Internal irradiation following inhalation of resuspended material

Once radioactive material is deposited on the ground, particles can become resuspended as a result of disturbances caused by wind or human activities, for example digging or ploughing.

The applicatio f resuspensioo n n models shows tha e contributiotth f o n resuspensio e totath lo t n inhalation exposur f individualo e practicalls i s y negligible wite possiblth h e exceptio f somo n e nuclides depositen i d semi-arid environments [5] or of unusual circumstances. Even then it is onl relevana y t pathwa materialr yfo st readil no whic e ar hy incorporated into biological systems, e.g. the transuranium elements. Nevertheless it is a pathway tha s includei t mann i d y analyses, primaril r completenessfo y . Appendix A contains a discussion of existing resuspension models and gives illustrative value r sucfo sh models.

2.2.5 External irradiation from deposited radionuclides

The estimation of the external dose from material deposited onto the groun relativela s i d y straightforward procedur numbea d modelf an eo r s exist for this purpose. The simplest way to calculate the dose rate in air abov contaminatea e d surfac o assumt s i ee tha tn infinit a thi s i s e plane source wite activitth h y uniformly distribute e surfaceth n o d . This method is only appropriat r depositfo e f radionuclideo s s wit shora h t radioactive half-life. For longer lived radionuclides it is necessary to model the migration down through the soil column. A number of models have been developed to predict this downward movement but the experimental data to

23 support the models are limited to a few nuclides and a few soil types. However, this proces f considerablo s i s e significanc determininn i e e th g external exposure from deposited radionuclides and needs to be taken into account.

e timTh e variatio e verticath f no l profil materiaf o eusee b o t dn ca l estimate the exposure above the soil surface. A technique that can be used o dividit s e soith e l column int numbea o calculato f layert ro d san e eth contribution to the exposure from each [31]. Standard methods are available to calculate the absorbed dose rate in air from such a distribution. The absorbed dose in air is then converted to an effective dose equivalent.

Account should also be taken of the shielding afforded by buildings, occupancy factors etc., when estimating individual doses from this exposure route [20].

2.2.6 Internal Irradiation Following the Ingestion of Terrestrial Food

e basiTh c principle calculatinr fo s committee th g d effective dose equivalent froe yearon m s intak f foos eo bee ha d n discusse n Sectioi d n 1.2.2.2. A model for the transfer of radionuclides through the terrestrial foodchain is required to calculate the radionuclide concentration in the food [33,34,35].

The transfer of radionuclides through the terrestrial environment into foodchains is complex. Many processes are involved and much depends on the characteristic particulae th nuclide f th o d f so rean environment e morTh .e important processes are illustrated schematically in Figure 2.2.

RADIONUCLIDE DEPOSITUM

majoe FigurTh r2 e2. processe transfee th r f radionuclidefo so r s through terrestrial foodchains.

24 Radionuclides depositing from the atmosphere may be intercepted by the foliage of vegetation. In general, they are removed from the surfaces of plan naturay sb l loss processes, suc weatherins a h g wit half-lifha e ranging severao t w frofe l ma ten daysf so .surface th Par f o te deposie b y tma abosrbed and transferred to other parts of the plant; this process is known as translocation and is far more significant for some nuclides, notably caesium, than for others, for example plutonium. Another important process by which plants may become contaminated is by absorption from deposits of radionuclide n soili s .

Plant y alssma o become contaminated with radionuclides depositee th n o d soil be resuspension processes or by splashing due to rainfall. In condition f continouo s s deposition these route e insignificanar s t compared witdirece th h t contamination processes from atmosphere. When deposition ceases thee onlar yy importan r thosfo t e nuclides whic e relativelar h y insoluble in soil and hence are not taken up to any extent by the roots, e.g. plutonium.

Radionuclides are lost from the system by migration down the soil column rooe th t f o zone n t somanI .ou d e cases long-lived radionuclide soin i s l modifiee b y ma d progressivel y biochemicab y l charge n soid thii sy an l sma change the extent to which they are absorbed by the plant's root system. e fixatioTh f caesiuo n y clab m y particle somn i s e soilwela s li s known example of such a process where root uptake is drastically reduced for some type f soilo s .

The transfe f radioactivito r o animalt y anothes i s r important route e exposurwhic th lean o ca h t d manf o emose .Th t well know d studiean n d pathway is the pasture-cow-milk pathway. This is important where cows graze a large surface ared hencan a e hav substantiaa e l intak f depositeeo d material. The transfer of radionuclides from pasture grass or other fodder crop o variout s s type f livestoco s d hencan ko huma t e n meat supplien ca s also be an important exposure pathway, particularly when the livestock graze pasture. Other pathways have been found to be important in some countries, r examplfo e lichen-reindeeth e r pathwa sub-artin i y c regions mose Th .t important route of intake of radionucides is by the consumption of contaminated gras r foddero s . Other possible route f intako s froe ar em contaminated water supplies y inhalatio,b f radionuclideo n e th n i s atmosphere and by the inadvertent consumption of contaminated soil. In

25 general, non f theso e e route f intakso e importanar e t compared wite th h direct ingestion of radionucildes in fodder particularly when continous depostion is taking place. Inhalation is only likely to be significant for those nuclides whose transfe n animaa f smalls o i lr e t acrosTh gu . e th s inadvertent consumptioy becomma i eso importanf o n t after depositios ha n cease t onl r thosbu dyfo e radionuclides whic e inefficientlar h y takep nu from the soil by the plants roots.

2.3. The assessment of exposures from radionuclides released to the aquatic environment

Radionuclide e releaseb y sma d inte aquatith o c environment either directly, by routine discharges to a water body, or indirectly due to the movemen f radionuclideo t s from othe re environment partth f o s . Once release e aquatith o t dc environment radioactive material wil transportee lb d and disperse y advectivb d d turbulenan e t processes occurin e receivinth n i g g water body. The magnitude of these processes is strongly dependent on the type of water body concerned and on the use man makes of the water in terms of drinking water, harves f aquatito c foods d recreationaan , l activities, the radiological impact of discharges of raionuclides to the aquatic environment will therefore vary markelde th made y f us wito etype d th han e receiving water body (e.g. lake, river, sea). Interactions with susupended matter and sidements are important physico-chemical processes which affect the transport of radionuclides in the aquatic environment and under some circumstances interaction with biota may also provide a transport mechanism. There are many models that have been developed for modelling aquatic transport. However most of these have been developed for some specific place such as a particular river system or estuary. It is not possible to have a general model to predict the transfer of radionuclides release e aquatith o t dc envrionmen possibls i s a tr release fo e o t s atmosphere.

As discussed chapter 1 radionuclides released into the aquatic environment may lead to the exposure of man by various routes. The three main onee ingestioar s f drinkino n g water, ingestio f contaminateo n d aquatic foodstuffs d externaan , l irradiation from radionuclides depositee th n o d shoreline. In some locations the use of water to irrigate agriculatural land may also be important. Other possible exposure routes include inhalatio f sea-sprao n resuspender yo d material d externa,an l irradiation while swimmin r handlino g g fishing gear.

26 2.3.1 Models for the dispersion of radionuclides in water bodies

modele Th s use o estimatt d e transporth e diffusiod an t f radionuclideo n s released to the aquatic environment are very dependent on the characteristic e receivinth f o s g water body. This dependence reflecte th s boundary conditions imposed on the dispersion process due to constraints introduced by the geometry of the water body. For example, in a river channel the cross current and vertical dispersion are limited while for a lak r oceao e n site crosth e s current dispersio largels i n y unlimitedA . number of hydrological dispersion models for various types of water body have been postulated by the US Nuclear Regulatory Commission (USNRC) [36], Other available models have been reviewed by Hoffman et al [37], IAEA Safety describe7 Serie5 . no s s relatively simpl ee dispersio modelth r fo s n of radionuclides into various water bodies [5]. Four type watef o s r body e consideredar : rivers, lakes, estuarie d coastaan s l sers. These categories tend to merge together and in particular, there is no clear interface between estuaries and the sea. The USNRC models and those given in Safety series 57 do not include losses of material due to sedimentation or through interaction with biological materials. Thes potentialle ar e y important processes, particularly losses due to sedimentation and a simple model for this is given in Safety Series 57. UNSCEAR [20] also discuss a simple model for dispersion in isolated water bodies which includes losses due to sedimentation processes. A detailed review of models for dispersion of radionuclide waten i s r bodie y als sma e foun b o n IAEi d A Safety Guid6 S e [6]. This report describes model varyinf o s g degree f complexito s r fo y dispersio e differenth n i n t type f wateo s r body.

2.3.2 Internal Irradiation due to Consumption of Drinking Water.

The committed effective dose equivalent from one year of intake by ingestion of dirinking water is calculated as the product of the radionuclide concentratio watere th watee n i n,th r intake eth ratd an e effective dose equivalen r unitpe t intak y ingestionb e e shoulOn . d keen i p mind that the intake rates in a critical group will be different for differen groupse ag t .

n principlI necessars i t i e y decontaminatio o alloan yt r wfo watey b n r treatment processes effece th ; f thio t s will vary witphysico-chemicae th h l processes employe treatinn i d nuclide th watee d th g an er concerned [38].

27 Also it may be necessary to allow for radioactive decay during the time that wate s storei r r transporteo d e poin th f consumption o to t d ; wate oftes i r n store reservoirn i d s lon a s severa a gr fo s l months before use n somI . e circumstances desalinated sea water may be used as drinking water and the effective decontaminatio desalinatioe th y b n n procesmuce b hy greatesma r than that produced by fresh water treatment. In practice decontamination losses are often ignored leading to a conservative estimate of the critical group dose from this pathway.

2.3.3 Internal irradiation due to consumption of aquatic foods

e effectivTh e dose equivalent froe consumptioth m f aquatio n c foods i s calculated using equation 1.3 with suitable values for the intake rate of the food of interest and the committed effective dose equivalent per unit intake by ingestion as mentioned in Chapter 1. The activity concentration in the aquatic food is obtained from the concentration in water.

Radionuclides may be incoprorated into the tissues of aquatic biota following ingestion, absorption or adsorption from the water. The transfer to aquatic biota is generallly estimated by use of concentration factors, also called bio-accumulation factors, between the water and the relevant organism. f bio-accumulatioo e us e Th n factor valis i s d only whe a steady-statn e distribution exists between all relevant parts of the aquatic environment (e.g. the water, sediments, the aquatic food web, and the biota). The bio-accumulation factor is quite variable with values ranging over several order magnitudf o s n somi e e givea case r nfo s radionuclid d organisan e m [39]. This variation is due to a number of factors including the composition of the water, sediment water interactions, the chemical state of the released radionuclides, and the characteristics of the aquatic organism. In addition, bio-accumulation factors are determined in a number of ways which may be one reason for the wide range of values reported in the literature.

Compilations of values for bio-accumulation factors for a wide range of nuclides in a variety of different species are available [40,41], IAEA Safety ] alsSerie[5 o 7 give5 s compilatioa s f bio-accumulatioo n n factors.

28 Caution should be used on whether the bio-accumulation factors apply to the whole plant or animal, or only to its edible part. When the bioaccumulation factor swhole applth eo t yplan r animaalss o ti ot i l necessary to consider the fraction of the organism which is edible. The edible fraction e differensth varr fo y t fresh wate marind an r e flord an a fauna.

2.3.4 Ingestio terrestriaf o n l foods contaminated with radionuclides from water

Radionuclide e transferre b waten i sy rma o terrestriat d l foodchains through animals drinking water and through the irrigation of pasture and food crops. The steps involved in the calculation of the dose to the critical group from these pathways are, first to calculate the concentation e relevanith n t food d the an o calculatst n e ingestioth e n dose using equatio discusses a 3 1. n fore chapten i dmodef Th o m . l1 r use o calculatt d e e concentratioth f radionuclideo n terrestrian i s l foods froe watemth r pathway is the same as for release to atmosphere.

An important route by which activity can reach man from irrigation is by spray irrigation, particularly of cultivated crops. The form of model used o assest se sam th s thathi a ey othes i tsan userr fo depositiod n process. However ,watee becausth e siz rf th o e droplet f e intensito e th d an sf o y application the fraction deposited on the external surface of vegetation may be substantially lower thadepositior fo n n from atmosphere [42,43].

2.3.5 Internal irradiation following inhalatio f radionuclideo n s with aquatic origins

Inhalatio f airborno n e activit y occuyma r following releasee th o t s aquatic environment. This could arise either from the resuspension of contaminated sedimen e transfeth t o particle t f radionuclideo re du r o s s from the sea surface by sae-spray [44,45]. The extent to which radionuclides will be transferred via sea-spray depends on the particular nuclide and the circumstances prevailin e sit f th interesto e t a g . This pathwa bees ha yn investigated along the North-West Coast of England [45]. In general, inhalation is not an important exposure pathway following releases of radionuclides to the aquactic environment.

29 2.3.6 External irradiation

External irradiation from contaminated sediments on tidal flats and along rive n importana re b bank n tca s exposure pathwa r somfo ye individuals [46,47,48]. Other possible external irradiation pathway occun r ca s fo r example while handling fishing gear, swimming or boating. In each case account has to be taken of the occupancy factor for the location and activity of concern. As an input to the assessment of exposure it is necessary to know the activity concentration in water or the activity concentration n sedimeni s s appropriatea t .

Occupancy times characteristi f criticao c l groupe sitth f o et a s interest should be used where possible but for a preliminary assessment or where data are not available, default values have been suggested in Safety Series 57 [5]. Chapte3 r

ASSESSMEN COLLECTIVF TO E DOSE

3 . 1 The concept of collective dose

e collectivTh produce numbee th th s f individualef i ro o t dos ) (S e s exposef theio d ran daverag e radiatio purposee nth dose r f Fo radiatios.o n protection, "radiation dose s usuall"i effective yth takee b o nt e dose equivalent (H ) as defined by the International Commission on Radiological c» Protection (ICRP)collective e th uni Th f .to e dos thes man.sievere i e th n t (man.Sv). Assessment f collectivo s n inpu a eusee b asseso s dos t a tdy ema s the justificatio practica f o n e considere n thi(i ds case normae th , l operation f nucleao r installations optimizo t r )o e leve th eprotectiof lo e applieb o t n d in order to keep the exposures as low as reasonably achievable.

Since the collective dose estimates may be used to justify the practice considered or to optimize the level of protection to be applied in order to kee s reasonablexposuree a th pw lo s sa y achievable resulte th , s a s e havb o t e accurat possibles ea consequencea s A . valuee th , svarioue choseth r sfo n parameters used in the models should be realistic. This is different from the practice recommende e individuath r fo d l dose assessments whicn s i i , t hi advise adopo t d t conservative value orden i s guaranteo t r certaiea n margif no safet proteco t y e individualth t .

Where there is a spectrum of doses over a population, from a given source, the collective effective dose equivalent is the weighted product of the effective dose equivalent from that source and the number of individuals in the exposed population. In this document, collective dose will be used to mean collective effective dose equivalent unless otherwise indicated. In integral form the collective dose is given by:

H d ) H N( H f (3.1 o S ) e numbeth s f i individualo r H wher)d H H( e s receivin effectivn ga e dose E K equivalent between H and H +dH from the given source. £ £ E ES

In practice ,simplea r summation formul oftes ai n used which represents the addition of individual doses in groups of the population. This is given by

31 S = l N. H. (3.2) i i e averagr caputh pe s wher i f to e populatio. a doseH n i e . N n d grouan , i p e numbeith s f people groupo r th e summatio n Th i e. oves l groupi n al r . i s Whichever formulation is used, the collective dose is an extensive quantity: in other words, various components of the collective dose can be added y togethethawa ta individuan i r l doses cannot.

In some cases, the exposure of the population is delivered at a varying rate ove perioda r n thesI . e conveniens casei t i s o defint collectiva e e dose rate (collective effective dose equivalent rate at time t), S (t), as K the weighted product of effective dose equivalent rate due to the source and numbe f individualo r e populationth n i s :

S = l HE N(Hg) d Hg (3.3)

The assessment of the collective dose rate is obtained by including in e populatioth n under consideratio l individualal n s receivin dosa g e froe th m given e sourcesummatioth s A .r integrao n l remains unchangee th f i d populatio mads i n e arbitrarily larger thae actuath n l exposed grou y addinb p g unexposed persons s convenieni e purpost i ,th r f assessino fo et e totath g l collective effective dose equivalent, or the total collective effective dose equivalent rate, fro a sourcem o specif,t populatioe th y worle th ds a n population. This specificatio t necessarno e exposes i nth f i yd grou s smali p l and well defined, and every exposed individual can be accounted for.

In order to have a measure of the total exposure of the population, caused by a given source, the collective dose commitment (collective effective ç dose equivalent commitment useds e )i collectiv Th . e dose commitmente du ,S to a given source is defined as the infinite time-integral of the collective effective dose equivalent rate S ,(t) , cause y thab d tmeasura sources i t eI . of the total detriment to health (as a first approximation) from the exposures that result from that source. It may be expressed as:

SC = J S (t) dt (3.4)

32 3.2 Methods for calculating collective doses

Collective doses can be calculated according to two types of methods:

- by summation, over space and time, of the individual dose rates obtained in the individual-related assessments. A schematic representation of this method for atmospheric and liquid discharges is shown in Fig. 3.1 and 3.1 (b).

model- f o e througus s e specificallth h y devise o estimatt d e collective doses A schemeati. c representatio f thio n s metho shwos i d Fign i n 2 .3. 2 (b)3. .d an ) (a

In addition, there are models of an intermediate type which incorporate variou methodso stw aspecte th .f o s

The procedure r implementinfo s e summatioth g n metho e straightforwarar d d but, in practice, they may be cumbersome for large population groups, if no simplifying assumptions are used. In general the method uses mathematical models to assess dispersion in the environment, along with transfer and bioaccumulation pathways o giv,t e concentration functioa s a s f distancno d an e time.

Models usin e summatioth g n metho e commonlar d y use o assest d s a s realistic as possible the local and regional contribution to the collective effective dose equivalent commitment from environmental releaes. Further information of this method can be found in [9,49].

The second method assumes complete deposition of discharged activity up distanca o t e which specifiede doeb t nee o no st d e utilizatioTh . e th f o n environment by an individual coupled with average population density would lead to collective dose assessment through inhalation and external pathways. n casI f ingestiono e , entire contaminated food material assumes i s o havt d e been consumed by the regional population. Although this method is elegant and simple for computation, it uses average as well as constant transfer parameters throughou depositionae th t l area d thuan , s woul t givy no dan e indication of the distribution of collective dose. Such models have been developed for both local, regional and global assessment. The local and regiona llarga o modelt e e extensar t site-independen averagf o makd e tan e us e global value mosr f sfo theio t r parameters .usee b s screenin Thea dy yma g

33 DISCHARGE

Dispersion model

AIR CONCENTRATION

Meteorological data

TIME INTEGRATED CONCENTRATIONS, DEPOSITION RATES AND EXTERNAL IRRADIATION, IN SECTORS

Environmental transfer d bioaccumulatioan n data

DOSIMETRY

INDIVIDUAL DOSES FROM RADIONUCLIDE N FOODSTUFFSI S , EXTERNAL IRRADIATION IN SECTORS D INHALATIONAN , IN SECTORS Agricultural production data

Population COLLECTIVE INTAKE data INGESTION

Dosimetry COLLECTIVE DOSES FROM EXTERNAL IRRADIATION COLLECTIVE DOSES AND INHALATION FROM INGESTION

TOTAL COLLECTIVE DOSE

Fig. 3.1(a) Schematic diagram for the summation method atmospheric discharge

34 LIQUID DISCHARGE DATA

Dispersion model

CONCENTRATIONS IN WATER AND SEDIMENT SECTORN SI S F RIVEO OCEAR O R N

Bioaccuraulation Transfer data factor

CONCENTRATION I N TIME INTEGRATED RADIONUCLIDE AQUATIC FOODSTUFFS CONCENTRATIONS WATEN I , R Irrigation IN IRRIGATED + EXTERNAL DOSE RATE and transfer FOODSTUFFS FROM SEDIMENTS data

Total Total Agricultural consumption consumption Dosimetry production rate of drinking data water

COLLECTIVE COLLECTIVE INDIVIDUAL DOSE COLLECTIVE INTAKF EO INTAKE FROM EXTERNAL DOSE INTAKE FOODSTUFF IRRADIATION INGESTION

Dosimetry Dosimetry Population Dosimetry data

_l ' COLLECTIVE COLLECTIVE COLLECTIVE DOSE COLLECTIVE DOSE DOSE (EXTERNAL) DOSE (INGESTION) (INGESTION)

TOTAL COLLECTIVE DOSE

Fig. 3.Kb) Schematic diagram for the summation method liquid discharge

35 ATMOSPHERIC DISCHARGE DATA

Simple model

TIME INTEGRATED AIR Transfer TOTAL ACTIVITY CONCENTRATION coefficients DEPOSITED yield

dosimetry dosimetry inhalation rates

PER CAPUT PER CAPUT Dosimetry INHALATION EXTERNAL DOSE DOSE

population population data data

COLLECTIVE COLLECTIVE COLLECTIVE DOSE DOSE DOSE

TOTAL COLLECTIVE DOSE

Fig. 3.2(a) Schematic specifie diagrath r fo mc method - atmospheris c discharRe

36 LIQUID DISCHARGE DATA

simple model

TIME INTEGRATED COMCENTRATIOMS Bioaccumulation IN WATE D SEDIMENTAN R S Irrigation and data transfer data \/ TIME INTEGRATED RADIONUCLIDES CONCENTRATION IN total consumption dosimetry IN IRRIGATED AQUATIC FOODSTUFFS datdrinkinr fo a g FOODSTUFFS water

Consumption datr fo a foodstuffs

V COLLECTIVE COLLECTIVE INDIVIDUAL DOSE COLLECTIVE INTAKE FROM INTAKE OF FROM EXTERNAL INTAKE FOODSTUFFS DRINKING WATER IRRADIATION (INGESTION)

Dosimetry Dosimetry Population Dosimetry data

COLLECTIVE COLLECTIVE COLLECTIVE COLLECTIVE DOSE DOSE DOSE DOSE (INGESTION) (INGESTION) (EXTERNAL) (EXTERNAL)

TOTAL COLLECTIVE DOSE

Fig. 3.2(b) Schematic e specifidiagrath r fo mc metho r liquifo d d discharges

37 models to determine the most important radionuclides as well as for a first estimate of the magnitude of the collective dose in order to see if further more refined calculations need to be carried out. They are not recommended a detaile r fo d assessmen e loca th d regiona an lf to l contributio e totath lo t n collective effective dose equivalent commitment. Further referencee b n sca foun n [50]i d .

In principle, collective doses should be calculated for the entire world's population mosr Fo .t radionuclides, however mobilite e th , th n i y environmen hamperes i t a shor y b dt physical half-lif y removab r eo l processes such as sedimentation or migration into the deep layers of the soil. The calculation is thus only carried out for those radionuclides from the local and regional zone, whic y extenhma d fropoine th mreleasf o t a distanc o t e e varying from about 100 km to several thousand kilometers, according to the model used. The radionuclides that are of importance for the global 5 8 4 1 3 contributio collective th o d t n an er K dos e, C commitmen , H e ar t 129 I. Krypton-85 n isotopa s ,a f noblo e e gas, remains confinee th n i d atmosphere afte airbornn a r e release becomet I . s disperse a relativel n i d y uniform way over the atmosphere of the entire globe in a matter of a few years and acts durin followine th g g yearlong-tera s sa m sourc f irradiatioeo f no the world's population. Tritium, carbon-14, and iodine-129 are other long-lived radionuclides which, upon release into the environment, become incorporate chemican i d l compoundsCH„Id an ,O C tha , ,t HO suc s a h presen higta h environmental mobility. These three radionuclides folloe th w water, carbon, and iodine cycles, respectively; their distribution in the 85 environment, however, is not as uniform as that of Kr because the concentrations of water vapour and of iodine in air present a substantial degre f variabilito e d becausan y oceae th e n n environmentasedimenta s a t sac l sink for carbon-14.

99 237 Other long-lived radionuclides y alsma o, becomNp er o , c likT e dispersed, once released, throughout the world and therefore be candidates for global assessments. They have, however, been little studied so far.

The IAEA has prepared in recent years two documents on radionuclides of regional and world-wide interest [51,52] in which detailed information can be found. l casesal e calculation i th , , e globaAs th f lo n component assumea s uniform distribution in at least one large sector of the environment, the global componen e collectivth f o t e effective dose equivalent commitmeno t s i t

38 be added to the local and regional component, which corresponds to the first-pass circulation. The global components of the collective effective 14 85 129 muce ar h I higheequivalen d r an thae , tth nKr commitment , C r sfo loca regionad an l l components releasesH r ;fo , however e globa,th l componen e samth e estimates f i to orde e magnitudb f o r o e t locad th d s an la e regional component.

Two types of model are used to carry out the global assessments:

compartment models, which yield the variation of the collective effective dose equivalent rate, and summatioy b , n over tim thif eo s quantity, the collective effective dose equivalent commitment,

3 14 - for radionuclides produced naturally ( H and C) specific models that make use of the knowledge of the production rate and doses related to the natural origin of those radionuclides.

Further information about those model r assessmenfo s e globath f lto 9 12 5 8 4 1 3 followine foune th b n n i d ca gI d an componen, Kr , C t fro, H m references [20,53,54,55,56,57,58].

39 Chapte4 r

APPROPRIATENESS OF MODEL SELECTION AND UNCERTAINTIE DOSN SI E ASSESSMENTS

1 Introductio4. n

Environmental concentration f radionuclideo s d radiatioan s n exposures resulting from routine operation f nucleao s r fuel cycle facilitie usualle ar s y d extremelan verw lo y y difficult t impossibleno f i , measureo t , . Consequently, individua d collectivan l e dose assessment usualle ar s y derived from models. Applying the models requires the knowledge of various parameters relating in particular to the transport of the radionuclides into the environment, to the human uses of the environment, or to biological processes in the human body.

Any mathematical model simulating actual processes in the real world can do so only imperfectly and the realization of a model in most cases make certain numerica r algorithmilo c approximations necessary. These approximation e unavoidablear s d thepresene an ,ar y o different t degreen i s modely an n spitI . f theso e e necessite problem th vien i f o wd an mako syt e predictive calculations, e.g r licensinfo . g releases, assumption r thesfo s e processes and variables have to be made. These assumptions may be based, on scientific knowledg f naturao e l lawd principlesan s n experimentao , l measurement n laboratoriesi s n fielo , d measurements under similar conditions, n inferenceo s from related event r paso s t experience lasth tn i resort r o e n ,o expert judgement.

Because model n onlca sy approximate real physica d biologicaan l l systems, their predictions are frequently uncertain. The extent of this uncertainty may vary from less than a factor of two to several orders of magnitude depending on the specific problem the model is to address and the quality of data availabl o quantift e e parameterth y modele th n i .s

Despite the potential for large uncertainties in the predictions of models o othe,n r mechanism exist r guidinfo s g decision d evaluatinan s g the possible significance of future releases of radioactivity. It is therefore necessary to formulate procedures to evaluate the validity of model predictions in order to increase confidence in the decision-making process [10] .

41 e assessmen th modeA r fo l f potentiao t r reao l l radiation exposures of population groups is usually expected to give one of two types of results:

n "over-estimateA ) (a " (conservative) value, i.e resula . t whics i h unlikely to underpredict actual exposures, possibly with a stated probabilit o underpredict y t actual exposure a state y b s d factor.

(b) A "best-estimate" value, i.e. a result which is intended to be as realistic as possible, possibly with stated error estimates.

The first type of results ("overestimates") are easier obtainable and are usually employed, for example, in the individual dose assessments. In this case, conservative values are adopted for the various parameters o thas , t overestimate f individuao s l dosee ar s obtained; this provide n additionaa s l margi f safety o ne calculate th s a , d dose e requirear s o remait d n belo e invidiuawth l dose limits imposey b d national regulations.

The second type of results ("best-estimates") would be needed, e.g. in assessments of collective doses. These have to be as accurate as possible usee b o justif t do t f thee practice i , th yar y e considereo t r o d optimize the level of protection to be applied in order to keep the exposures as low as reasonably achievable. In addition, information on e uncertaintieth e collectivth n i s e dose estimat helpfus i et point i s a ls e areath whicn i o st h improvemen t gives desirabli i t s sa somd an ee indications on the validity of the estimate.

4.2. General Problems encountered in Dose Assessments based on Models

It appears worthwhile to stress several facts of modelling and related general problems of importance with regard to validity and accurac f doso y e assessments:

modeA usualls ) i l(a y designe o simulatt d e thosd e an part , of s those processe e reath l, worldin s , which previously have been identified as the important and relevant ones, in sufficient a sufficien o detait d an l t degre f accuracyo e . Thuse ,th

42 "reliability" of a model or its results should only be assessed s intendeit r fo d are f applicatioa.o n space(i n r ,fo tim d an e the endpoints considered).

modeA conceptuall) s li (b y identica workina o lt g hypothesir fo s the explanatio f interactionso n , transport processes, etc.s A a hypothesis, a model should be able to:

) (i possible result quantitativa n i s e formd ,an (iivalidatee b ) d where possible.

(c) There exist only very few reliable and useful experimental data pointe largeth n i s, multi-dimensional observation space (dimensions of which are, e.g. time, location, pathway, radionuclide, organ, chemical form, atmospheric conditions, agricultural practices, etc.) against which model structure, and input parameter modea n actuallf o slca d objectivelan y e yb tested under realistic conditions. Host validation tests performed hitherto were either indirect (base chemican o d r lo physical similarities r subjectiv)o e inferences from evidence for similat identicano t bu r l situations (e.g. from fall-out measurements). mose th t f o seriou e On s ) problem (d e e lac th th f test ko s r i sfo s completeness of the exposure pathways considered in a model. There is no procedure available yet to ensure for a given simulation task tharelevanl al t t exposure pathways have actually been identified and properly taken into account in a model.

(e) Processes in environmental systems and their status are subject larga o t e natural variabilit timn d spacei y an e .e th Man f yo important parameters describing these processes and status are known from experimen r theorto y only with large marginf o s uncertainty.

(f) Some factors needed in such models (e.g. agricultural practices, consumption habits d locaan , l population densitn i y r thfuturefa e , hydrogeological propertie unexploref o s d underground strata) are even completely unknown.

43 An evaluation of model uncertainty is recommended. Estimation of the potential exten f mispredictioto recommendes i n d througe th h comparison of model predictions against independent sets of observations (model validation). If this is not possible, sensitivity and uncertainty analyses are recommended to identify model components that are the most important contributor e uncertaintth o t s predictionsn i y . These procedures are recognized as necessary to improve the confidence with which model predictions can be used as tools for decision-making.

n inadequatA r incompleto e e specificatio e probleth f o nm (scenario) describee b modee o t th ly b dmigh t produce wrong informatior nfo decision-maker se correspondin th eve f i n g model structur d parameteran e s should be completely adequate and accurate for the specified scenario.

n adequata r Fo e scenario n inadequata , e model structure sucs a h wrong definition of compartments, usage of inadequate approximations (e.g. Gaussian plume approximation whe t applicable)nno r neglectin,o g important exposure pathways can lead to misleading results even if most parameters are correct.

The most appropriate specificatio problee th moded f o nan m l structure wil t ensurlno e adequate accurac f resulto y s from model calculation e parameteth f i s r values chose e inappropriatar n e th r fo e problem under consideration (e.g. the interactions of compartments in the eco-system, deposition velocity, wind direction).

In the following paragraphs all three factors affecting the accuracy of results will be considered.

3 Factor4. s Affectin Uncertainte th g Resultf o y s

least a Ther e t ar ethre e majn features affectin qualite th g f o y results from environmental transfer models:

(a) The assessment problem modee Th l ) structur(b e chosen (c) The parameter values used in the calculation.

44 The factors affectin e reliabilitth g modef o y l predictions have been identified as belonging to those three distinct categories: (a) uncertainty due to improper definition and conceptualization of the assessment problem, (b) uncertainty due to improper formulation of the mathematical ) uncertaintmodel(c d e estimatio,an th n i y parametef o n r values.

4.3.1. Influenc Specificatioe th f eo Problee th f o nm

e firsTh t consideration shoul directee b d d toward e specificatioth s n of the problem which is then simulated by an approximative model. This will depen numbea n o df factor o r s including:

(i) The type of results required (overestimates for licensing decisions, best-estimates for optimizations or technology assessments);

(ii) The temporal resolution required (equilibrium or dynamic model, consideratio releasa yeae e f th tim o rth n i eef o nand/o r time after a release, predictions only for present times or for times far in the future, lifetime averages or annual exposures, etc.);

(iii)The spatial resolution required (local, regional, global assessment, generi r site-specifio c c models.)

The evaluation of reliability in model predictions is extremely difficult when the uncertainty in model predictions is due to errors in the formulation of the assessment problem, including incorrect specificatio e scenarioth f o n f releaso s d exposuran e d incorrecan e t specificatio e quantitieth f predictede o nb o t s n thiI . s case, reliability assessment will be entirely based on judgement.

There is no mathematical or other objective technique available (besides comparison with appropriate experimental data) with which to tes r analysto e quantitativel e degreth y f influenco e e finath ln o e result usage modea th f o ef lo s develope slightla r fo d y different scenario. Nevertheless thorouga , h consideratio e applicabilitth f o n f o y the scenario underlying a certain model for the situation where it will be used is strongly recommended.

45 4.3.2 Model Structure

The validity of model formulation will also be subject to judgemental assessment procedures unless date obtaineaar d from experiments that are specifically designed to test model predictions over the intended rang f modeeo l application.

Knowledge of this subject have bearings on the choice for different assessment purposes of appropriate model principles as well as adequate element modee th ln i s structure.

4.3.3 Determinatio f Parameteo n r Value Associated an s d Uncertainties

The most astute selection of environmental scenario and model structure wil t ensurlno e optimum accuracy from model calculatione th f i s parameter values chose e inappropriatar n problee th r mfo e under consideration. Previous sections of this report have pointed at the variety of environmental and human parameters that must be quantified for dose calculations to be made.

Further, the processes in and between compartments in environmental systems s wel ,a s humaa l n interactions with those systems e subjecar , o t a large natural variability in time and space. Many of the important parameters describing these processes are known or estimated only with large uncertainties. Some factors neede r dosfo de calculations (e.g. agricultural practices, consumption habits d locaan , l population density in the far future, or hydrogeological properties of unexplored underground strata y eve )ma completele b n y unknown n spitI .f thes o e e problems, assumptions and approximations for these processes and variable se availabl b hav o t e o providt e e quantitative dose estimates.

These problems have been treated in several publications [59,60,61,62].

4.4. Sensitivity Analysis

Method f sensivito s y analysi typicalle ar s y performe o identift d y the effec modef to l parameter moden o s l predictions e classica.Th l method of sensitivity analysis [63] is to take the partial derivatives of the

46 model response with respect to each model parameter. It is useful to normalize these values by numerically estimating the percent change in model response resulting fro percena m t perturbatio modee th ln i n parameter.

Those parameters which are well known (i.e. low uncertainties) and w differentiahavlo a e l sensitivit unlikele ar y mako yt significana e t contribution to overall uncertainties. Conversely, parameters which may wele b lt knowno d hav an nhiga e h differential sensitivit e likelar y o yt greatly affec e uncertaintth t modef o y l predictions.

An additional screening technique is the use of a sensitivity index formed froe rati th mmodef o l predictions using plausible maximud man minimum valueparametea r sfo r [64). High value f thiso s quotient indicat e importancth e parametea f eo r ovee fulth rl rang valuef eo s likely to be used for the assessment, and not just the importance of the parameter in the region of its nominal value.

These sensitivity methods are simple to use and require very little information to perform. However, it is important to remember: (1) all sensitivity indices are a function of the assessment scenario and model, the prediction of interest (and any time and spatial dynamics associated with the prediction) and the nominal values of all parameters. Any changes in these conditions requires that new sensitivity index values be calculated. (2)Actual model uncertainties due to parameter variation are a function of the simultaneous variability (and hence, interactions) of all model parameters. But despite this limitation, the sensitivity analysi mann i s y cases shoul e abl b do giv t e e information abou mose th t t relevant parameters. Now looking for the quantification of these parameter e sidd providinon an en o s mora g e sophisticated methof o d parameter analysis on the other, it is necessary to use the method of uncertainty analysis.

4.5 Uncertainty Analysis

In order to determine parameter values together with their uncertaintie quantifo t d an uncertaintse th y e ordeth f o r y o conservatism of the resulting dose, procedure referred to as a "parameter imprecision analysis usee b dn "ca [65] . This procedure involves

47 estimating the variability associated with each model parameter to ascertain the influence on the model output of the combined variability of all model parameters. Analytical error propagation formulae can be used to perform parameter imprecision analyses on relatively simple models [66,67,683- For more complex models, numerical techniques computea more f b o employin ey e convienienrma us e th g t than complex analytical solutions (69,70,71,72,73,74,75].

48 REFERENCES

] Internationa[I l Atomic Energy Agency, Basic Safety Standard Radiatior sfo n Protection, 1982 Edition, Safety IAEA, Serie9 . , Ho sVienn a (1982).

[2J International Atomic Energy Agency, Principle Establishinr fo s g Limits Release foth r f Radioactiveo e Materials intEnvironmente th o , Safety Series No. 45, IAEA Vienna (1978).

[3] International Atomic Energy Agency, Principles for Limiting Releases of Radioactive Effluents into the Environment, Safety Series Mo. 77, IAEA, Vienna (1986).

] Internationa[4 l Atomic Energy Agency, Atmospheric Dispersio Nuclean ni r Power Plant Siting SafetA : y guide Safety Serie . 50-SG-53sNo , IAEA, Vienna (1980).

[5] International Atomic Energy Agency, Generic Models and Parameters for Assessin Environmentae th g l Transfe Radionuclideo rf s from Routine Releases, Safety IAEA , Serie57 ,. ViennsNo a (1982).

[6] International Atomic Energy Agency, Hydrological Dispersion of Radioactive Materia Relation i l Nucleao t n r Power Plant Siting, Safety Serie 50-SG-S6. sNo , IAEA, Vienna (1985).

[7] International Atomic Energy Agency, Applciation of the Principles for Limiting Radioactive Release Case Nucleath f eo n si r Power Plantd san Reprocessing Plants (under preparation).

[83 International Atomic Energy Agency e Problem,Th Settinf so g Global Dose Upper Bounds (under finalizatio publication)r fo n .

[93 International Atomic Energy Agency ,PracticaA l Methodologe th r fo y Assessment of Individual and Collective Radiation Doses from Radioactivity in the Environment (under preparation).

[103 International Atomic Energy Agency, Procedure Assessinr fo s e th g Reliabilit f radionuclidyo e Environmental Transfer Model Predictions, IAEA, Vienna (Draft, 1987).

[II] International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection. ICRP Publication 26. Annals of the ICRP 1: No. 3 (1977).

[12J Beninson, D, Limitation of future radiation exposures from the present operation of nuclear fuel cycle installations. IAEA - SM- 258/52, Proceeding IAEe th A f Seminarso , Madrid, (1981).

[133 International Commissio n Radiologicano l Proteciton. Radionuclide Release intEnvironmente th o : Assessmen Dosef to Mano t s . ICRP Publication 29. Annals of the ICRP 2: No. 2, (1979).

[14] NRPB/CEA, Methodology for evaluating the radiological consequences of radioactive effluents released in normal operation. Joint report by the National Radiological Protection Board and the Commissariat à l'Energie Atomique. Luxembourg, CEC, Doc. No. V/3865/79 (1979).

49 [15] Kramer, R and Drexler, G, On the calculation of the effective dose equivalent. Rad. Prot. Dosim. 3, 13-24 (1982).

[16] Kramer Zanki, ,R William, ,M Drexlerd e calculatioan Th ,G . ,G f o n dose from external photon exposures using reference human phantomd san Monte Carlo methods male , Th Pare . (ADAM1 t d femal)an e (EVA) adult mathematical phantoms. GSF report S-885, (Munich, Gesellschaft für Strahle Umweld nan t forschuhg mbH,1982).

[17] Willaims Zanki, ,G Drexler, ,M d Eckerle calculatioan Th , G , ,H f no dose from external exposures using reference human phatoms and Monte Carlo methods Part 2. The dose to organs and tissues from occupational exposures F reporpublishede GS b .o t .

[18] International Commissio Radiologican o n l Protection. e Reporth f o t Task Grou Referencn o p e Man .A repor t prepare a tas y kb d grouf o p Committee 2. ICRP Publication 23. Pergamon Press, Oxford, 1975.

[19] Poston, J.W d Snyderan . , W.S mode.A exposurr fo l semi-inifinita o t e e cloud of photon emitter. Health Phys. 26: 287 (1974).

[20] United Mations Scientific Committee on the Effects of Atomic Radiation, Source d Effectsan f Ionizinso g Radiation: 1982 Generae reporth o t l Assembly, United Nations Yorw ,Me k (1982).

[21] ICRP, Limit r Intakfo s Radionuclidef eo Workersy sb , ICRP Publication 30, Par Ann. 1 t 4 (1979).3/ ICR , 2 P, Par Ann2 t .4 (1980) ICR3/ 4 P , Part 3 Ann. ICRP 6 2/3 (1981).

[22] NRPB, Metabolic and dosimetric models for application to members of the public. National Radiological Protection Board., Chilton, NRPB-G-53 (1984) (London, HMSO).

[23] Pasquill, F, Atmosphere Diffusion (2nd Edn.), Ellis Horwood Ltd., Chichester (1974).

[24] Gifford, F.A., in Meteorology and Atomic Energy 1968 (Slade, D.H. Ed.), USAEC, Rep. TID-2 0 (1968)419 .

[25] Till, J.E. and Meyer, H.R. (Editors), Radiological Assessment. A textbook on environmental dose analysis. United States Nuclear Regulatory Commission, NUREG/C 3332- R , ORNL-5968 (1983).

[26] Weber, E, Air Pollution: Assessment Methodology and Modelling (Plenum Press Yorkw ,Ne , London, 1982).

[27] Hoffman, F.O. and Baes, C.F. Ill (Eds.), A statistical analysis of selected parameters for predicting foodchain transport and internal dose of radionuclides. Oak Ridge National Laboratory report, (ORNL/NUREG/TM-282)(1979).

[28] Vogt, K.J. Atmosphärische Ausbretitunge Ablagerunged un n n radioaktiver Stoffe. IN Radioökologie, Berichtsband der Fachtagung Radioökologie des Deutschen Atomforums e.V., vom 2-3 Okt. 1979 in Wissenschaftszentrum Bonn. (Vulkan-verlay, Essen, 1979).

[29] Clark, R.H., A model for short and medium range dispersion of radionuclides release atmospheree th o t d e firsTh . t repora f o t working group on atmospheric dispersion. NRPB-R91 (1979).

50 [30] Jones, J.A., ESCLOUD, A computer program to calculate the air concentration, deposition rate and external dose rate from a continuous discharg f radioactiveo e materia atmospheree th o t l , Chilton, National Radiological Protection Board, NRPB-R101 (1981) (London, HMSO).

[31] NRPB/CEA, Methodolog evaluatinr fo y radiologicae th g l consequencef so radioactive effluents release norman i d l oepration. Joint repory tb the National radiological Protection Board and the Commissariat à l'Energie Atomique. Luxembourg, CEC, Doc. No. V/3865/79.

[32] Brenk Dosi, D 0 ,radionulkidM s 80 faktore. ca r Berechnunr nfü ezu r gde exterrnen strahlen exposition durch photonen - und elecktronen Strahlung. Bericht der Fa. Brenk Systemplanung in auftrag des bundesgesundheitsamtes 7904/1- U BS , ,. Der AacheNr .- n (1983).

[33] Simmonds, J.R., Linsley, G.S. and Jones, J.A., A general model for the transfer of radioactive materials in terrestrial food chains. Chilton, NRPB-R89 (1979) (London, HMSO).

[34] Booth, R.S., Kaye, S.V., Preliminary systems analysis modef lo radioactivity transfer to man from deposition in a terrestrial environment. ORNL-TM-3135 (1971).

[35] Matthies, M, Eisfeld, K, Paretzke, H.G., Prohl, and Wirth, E, Simulatio Radionulkidtransfers nde n landwirtschaftlichei s n nahruhgsketten GSR-S-882. (1982).

[36] USNRC* NUclear Regulatory Commission, Methods for estimating aquatic dispersion of liquid effluents from routine reactor releases for the prupos f implementineo g appendi regulator, I x y guide 1.113, Rev.l, (1977)

[37] Hoffman, P.O., Miller, C.W., Shaeffer, D.L., et al, Compilation of documentaed computer codes applicable to environmental assessment of radioactivity releases. ORNL/TM-5830 (1977).

[38] Denham, D.H d Soldat.an , J.K. ,A stud selectef o y d parameters affecting the radiation dose from radionuclides in drinking water downstream of the Hanford project, Health Physic s9 (1974) 13 Vol ; ..28

[39] Ophel, I.L., "Aquatic foodchain transpor radionuclides"f to , Proc. workshop on the evaluation of models used for the environmental assessmen radionuclidf to e release, Gatlonburg 1977, (Hoffman, P.O., Shaeffer, O.K., Miller, C.W., Garten, C.T., Jr., Eds.) Conf-770901 (1978).

[40] International Atomic Energy Agency, The radiological basis of the IAEA revised definition and recommendations concerning high-level radioactive waste unsuitabl dumpinr fo e seat a g , IAEA Tech c 211Do ., IAEA, Vienna (1978).

[41] Owen, P., Hoffman, P.O., Bases III, A statistical analysis of selected parameters for predicting food chain transport and internal dose of radionuclides. (ORNL/NUREG/TM-282) (October 1979).

[42] Delmas, J., Disdier, R., Grauby, A., et al, Radiocontamination expérimental quelquee ed s espèces cultivées soumise sl'irrigatioa r npa aspersion. Proc. Int. Symp Radioecologyn o . , CEM, Cadarache, France, 1969.

51 [43] Delmas Grauby, d Disdier,J. an . ,EtudeA , ,R. s expérimentalee l r su s transfert dancultures le s quelquee sd s radionucleides présents dans effluents le s centralesde s electronucleaires 321-33. ,p n i 2 Environmental behaviour of radionuclides released in the nuclear industry. IAEA, Vienna, 1973.

[44] Pattenden, N.J., Cambray, R.S., Playford, K., et al. Atmospheric measurements on radionuclides previously discharged to sea. International Symposium on the impacts of radionuclide releases into the marin environment, Vienna, 1980.

[45] Eakins, J.D., and Lally, A., E., The transfer to land of actinide-bearing sediments y sprayfrob e Iris th a m .hSe Science th f o e Total Environment 23-35 ,3 2 (1984).

[46] Jefferies, D.F., Exposure of radiation from gamma-emitting fission-product radionuclide estuarinn i s e sediments froNorthease th m t Irish Sea. Environment surveillance in the vicinity of Nuclear Facilities, (W.C/ Reining, ed.) E.G. Thomas, Springfield, Illinois (1970).

[47] Cowser, K.E., et al, Evaluation of radiation dose to man from radionuclides release Clince th o ht d River. Disposa f radioactivlo e wastes into seas. Oceans surfacd ,an e waters, IAEA, Vienna (1966).

[48] Foster, R.F d Soldat.an , J.K., Evaluatio exposuree th f o n s resulting frodisposae th m f radioactivlo e wastes into seas, oceans d surfac,an e waters, IAEA, Vienna (1966)

[49] Bouville: Personal Contribution

[50] UNSCEAR - Sources and Effects of Ionizing Radiation. United Nations Scientific Committe Effecte th n Atomif o eo s c Radiation 1977 reporo t the General Assembly, with Annexes Yorw ,Ne 1977, k .

[51] IAEA, Réport of the Adviosry Group meeting on "Models and Radiological Basi Recommendationr sfo Sadionuclidn so e Release Regionaf so d an l World-Wide Interest" held in Vienna, 1-4 June 1982.

[52] IAEA, Regional and Global Environmental Behaviour of radionuclides from the Nuclear Fuel Cycle. IAEA-TECDOC-279. Vienna, 1983.

[53] Kelly, G.N., Jones, J.A., Bryant, P.J., Morley (1975), ,F. e Th , predicted radiation exposur populatioe th e Europeaf eo th f o n n community resulting from discharges of Krypton-85, tritium carbon-14 and iodine-129 from the nuclear power industry to the year 2000, Commission of the European Communities report V/2676/75.

[54] Killough, G.G., (1980) ,dynamiA c mode estimatinr fo l g radiation dose to the world population from releases of ^4C to the atmosphere, Health Phys. 38, 269.

[55] Bergman McEwand an . , (1977),R ,C. , Dos d dosean e commitmeno t e du t Carbon-14 from the Nuclear Industry, aktiebolaget Atominergie Report S-548.

[56] Bonka, H., Grndler, D., Horn, H.G., (1979), Prolems occuring while carrying out cost-benefit analysis for ^H, l^C, 85jcr> iodine and aerosol nuclean i s r facilities n topicai , l seminaPracticae th n ro l

52 Impliactions of the ICRP Recommendations (1977) and the Revised IAEA Basic Standar Radiatior fo d n Protection, IAEA Publication IAEA-SR-36/14.

[57] Bergman, R., Bergstrom, U., Evans, S., Environmental Transport and Long Term Dose PreliminarA . y stud f Iodine-129o y , p.4 Proceedingn 3i f so SpecialisA anNB t Meetin 1-129n go , (1977).

[58] Kocher, D.C. ,A dynami cglobae modeth f lo l iodine cycl d estimatioan e n of dose to the world population from releases of ioden 129 to the environment, Environment International 5: 15-31 (1981).

[59] Huffman, P.O Baesd .an , C.F. (1979) .A statistica l Analysi Selectef so d Parameters for predicting Food Chain Transport and Internal Dose of Radionuclides, Report NUREG/CR-1004 (Oak Ridge National Laboratoryk ,Oa Ridge, Tennessee).

[60] Little, C.A Millerd .an , C.W. (1979) Uncertainte Th . y Associated with selected Environmental Transport Models, Report Mo. ORNL-5528 (Oak Ridge National Laboratory k Ridge,Oa , Tennessee).

[61] Onishi, Y. Yabusaki, S.B., Cole, C.R., Davis, W.E. and Whelan, G. (1982). "Multimedia Contaminant Environmental Exposure Assessment (MCEA) Methodology for Coal-Fired Power Plants, Report prepared for the Rand Corporation (Battelle, Pacific Northwest Laboratories, Richland, Washington).

[62] Onishi, Y. , R.I. Serne, E.M. Arnold et al. Critical Review: Radionuclide Transport, Sediment Transport and Water Quality Mathematical Modelin d Radionuclidan g e Adsorption, Desorption Mechanisims. Report Nos. NUREG/CR-1332, PN2-290K1981).

[63] TOMOVIC, R. (1963). Sensitivity Analysis of Dynamic Systems. McGraw Hill, N.Y.

[64] Hoffman, P.O. R. Gardner (1983) Evaluation of Uncertainties in Radiological Assessment. A text book on Environmental Dose Analysis. U.S. NRC, HUREG/CR-3332.

[65] Schwarz, G. (1980). "General aspects of accuracy in dose calculations." Proceedings of a Workshop on Accuracy in Dose Calculations for Radionuclides Release e Environmentth o t d , Lindokers, K.Hd .an Bonnenberg, H. J., Eds. (GUW Gesellschaft für Umweltuberwachung mbH, Aldenhoven, Federal Republic of Germany).

[66] Shaeffer, D.L. (1980). "A Model evaluation methodology applicable to environmental assessment models, "Ecological Modeling 8, 275.

[67] Collée, R., Abee H.H., Cohen, L.K., Elsenhower, E.H., Jarvis, A.N., Fisenne, I.M., Jackson Johnson.. ,M , R.H., Olsond Peelan . . ,J ,D (1980). "Reporting environmental radiation measurements data," page 6-1 in Upgrading Environmental Radiation Data, Health Physics Society Committee Report Nos. HPSR-1 520/11-80-01A ,EP 2 (U.S. Environmental Protection Agency, Washington, D.C.).

[68] Schubert, J., Brodsky, A. and Tyler, S. (1967)."The lognormal function as a stochastic model of the distributuon of Strontium-90 and other fission products in humans," Health Phys. 14, 1187.

53 [69] Rubenstein, R.Y. (1981). Simulation and thé Monte Carlo Method (John Wile Sonsd an y , Inc. Yorkw ,Ne w York) ,Ne .

[70] Iman, R.L., Conover, W.J Campbelld an . , J.E.. (1980). Risk Methodology for Geological Disposal of Radioactive Waste: Small Sample Sensitivity Analysis Technique r Computefo s r Models, wit Application a h Riso t n k Assessment, Report NUREG/CR-1397 (U.S. Nuclear Regulatory Commission, Washington, D.C.).

[71] Iman, R.L., Helton, J.C. and Campbell, J.E. (1981). "An approach to sensitivity analysis of computer models: Part I-Introduction, input variable selection and preliminary variable assessment," J. Quality Tech , 17413 . .

[72] Iman, R.L., Helton, J.C. and Campbell, J.E. (1981). " An approach to sensitivity analysi computef so r models" Part II-Rankin f inpugo t variables, response surface validation, distribution effect and technique synopsis, Qualit. "J y Tech , 23213 . .

[73] Scharz, G. and Hoffinan, F.O. (1981). "Imprecision of dose predictions for radionuclides release environmente th o t d n applicatio:A a f no Montecarlo simulation technique," Envir, Int 289, 4 . .

[74] O'Neill, R.V. (1979) A revie." f stochastio w c modellin ecology,n i g " ISEM Journa 104, 1 l .

[75] McKay, M.D., Conover, W.J d Beckman.an , R.J. (1979) A compariso." f no three methods for selecting values of input variables in the analysis of output fro computea m r code," Technometric 239, s21 .

54 LIS PARTICIPANTF TO S

First Consultant meeting to prepare the working document for the first Advisory Group meeting on Recommendations on Methodologies for the Source - and Individual - Related Assessments in Relation to the Limitation of Releases of Radioactive Effluents into the Environment. Vienna, 14-18 May 1984

FRANCE

Bouvill. A . Dr e CEA BP No. 6 92260 Fontenay-aur-Roses

UNITED KINGDOM

. SimmondJ . Ms s National Radiological Protection Board Harwell England

IAEA

. J.UMr . Ahmed (Scientific Secretary) Division of Nuclear Safety

First Advisory Group meeting, Vienna, 27-31 August 1984

ARGENTINA

Beninso. D . Dr n (Chairman) Comision Nacional de Energia Atomica Avenida Libertador 8250 Buenos Aires 1429

AUSTRIA

Dr. Manfred Tschurlovits Atominstitut der Oesterr. Hochschulen Schuettelstr. 115 1020 Vienna

FRANCE

Bouvill. A . Dr e CEA BP No. 6 92260 Fontenay-aux-Roses

55 GERMANY. FEDERAL REPUBLIC OF

Dr. A. Bayer Kernforschungzentrum (KfK) INR Weibergass5 e 7500 Karlsruhe-INR

HUNGARY

L.B. Dr . Sztanyik National Institut r Radiobiologefo y and Radiation Hygiene 1221 Budapest XXII Pentz Karol5 yU.

INDIA

Mr. T.M. Krishnamoorthy Health Physics Division Bhabha Atomic Research Centre Bombay

JAPAN

Makin. Mr o Gambette Ru 3 1 a 92260 Fontenay-aux-Roses France

SWEDEN

Dr. R. Bergman National Defence Research Institute FAO 4 S-90 UME2 18 A

Mr. T. Godas National Swedish Institutr fo e Radiation Protection Box 60204 StockholS-101 0 4 m

UNITED KINGDOM

Ms. J. Simmonds National Radiological Protection Board Harwell England

UNIO SOVIEF NO T SOCIALIST REPUBLICS

Mr. Kuzjmin Institute of Atomic Energy Kurchatov Moscow

56 UNITED STATES OF AMERICA Corle. J . yMr Pacific Northwest Laboratory Batteile Blvd. P.O. Box 999 Richland Washington 99352

IAEA

Mr. J.U. Ahmed (Scientific Secretary) Division of Nuclear Safety

Mr. A. Gonzalez Division of Nuclear Safety

Stroha. P . Mr l Divisio Nucleaf o n r Safety

Second Consultant Meetin Compilo t g drafe th e t document produced furine th g First Advisory Group Meeting: Chilton, England, 12-14 December 1984.

FRANCE Dr. A. Bouville CEA BP No. 6 92260 Fontenay-aux-Roses

UNITED KINGDOM

Simmond. J . Ms s National Radiological Protection Board Harwell England

IAEA

Mr. J.U. Ahmed (Scientific Secretary) Division of Nuclear Safety

57 Second Advisory Group meeting, Vienna, 17-21 June 1985

ARGENTINA

Dr. D. Beninson (Chairman) Comision Naciona Energie d l a Atomica Avenida Libertador 8250 Buenos Aires 1429

AUSTRIA

Dr. Manfred Tschurlovits Atominstitut der Oesterr. Hochschulen Schuettelstr. 115 1020 Vienna

FRANCE

Dr. A. Bouville CEA 6 . BPNo 92260 Fontenay-aux-Roses

GERMANY. FEDERAL REPUBLIP CO

Dr. Bachner Gesellschaft für Reaktrosicherheit Glockengasse l D-5000 Köln

HUNGARY

Dr. L.B. Sztanyik National Institut r Radiobiologfo e y and Radiation Hygiene 1221 Budapest XXII Pentz Karol5 yU.

INDIA

Mr. T.M. Krishnamoorthy Health Physics Division Bhabha Atomic Research Centre Bombay

SWEDEN

Bergma. R . Dr n National Defence Research Institute FAO 4 S-901 82 UMEA

Mr. G. Johansson National Swedish Institute for Radiation Protection Box 60204 S-10 Stockhol1 0 4 m

58 UNITED KINGDOM

Dr. M.J. Clark National Radiological Protection Board Chilton Didcot OXON OX11 ORQ England

UNITED STATE AMERICF SO A

Corle. J . yMr Pacific Northwest Laboratory Battelle Blvd, 9 99 x P.OBo . Richland Washington 99352

IAEA

Mr. J.U. Ahmed (Scientific Secretary) Divisio Nucleaf o n r Safety

Mr. F.N. Flakus Divisio Nucleaf o n r Safety

Mr. J. Inaba Divisio Nucleaf no r Safety

Third Consultant Meetin finalizo t g documene eth r Publicationtfo : Vienna, 9-13 March 1987.

SWEDEN

Mr. R. Böge National Institut Radiatiof eo n Protection Box 60204 S-104 01 Stockholm

IAEA

Mr. J.U. Ahmed (Scientific Secretary) Division of Nuclear Safety

59