Long-term environmental behaviour of radionucleides€¦ · LONG-TERM ENVIRONMENTAL BEHAVIOUR OF...

LONG-TERM ENVIRONMENTAL BEHAVIOUR OF RADIONUCLIDESFrançois Bréchignac, Leif Moberg, Matti Suomela

C E C - I P S N A s s o c i a t i o n F i n a l R e p o r t

RECENT ADVANCES IN EUROPE

I n December 1996, the Institut de Protection et de Sûreté Nucléaire(IPSN) signed an Association contract (F14P-CT96-0039) with theEuropean Commission (EC), Directorate General Research.

Within the framework of the Association Contract, research in radiationprotection was carried out in the fields covered by the Fourth FrameworkProgramme on Nuclear Fission Safety. The contract running over a period of36 months from January 1, 1997 to December 31, 1999, contained 3 pro-jects, which were executed by a total of 18 partners from all over Europeincluding Norway as a non EU member state funded by its own government.The contract value amounted to 4,2 MECU, with the contribution from thecommission of up to 2,3 MECU.

IPSN has born the overall responsibility for coordination of the workwithin three multinational projects. Each project was managed by a scientificproject leader coordinating the scientific work ; Dr François Bréchignac,IPSN, FR, for the PEACE project, Dr Leif Moberg, SSI, SE, for theLANDSCAPE project and Dr Matti Suomela, STUK, SF, for the EPORAproject. The implementation of the association contract was supervised by a steering committee consisting of EC and IPSN representatives. IPSN wasproviding the secretariat and administrative support to the committee.Chairmanship was assumed by Mr. Ezio Andreta, then Mr. Hans Forsströmfor EC and Pr. Henri Métivier for IPSN.

The association contract covered the following topic "Evaluation of radiation risks: quantification of parameters which determine the fluxes of radionuclides in ecosystems, and long-term consequences of accidentalcontamination in semi-natural environments".

The overall objectives of the PEACE project aimed at assessing the impactof the radioactive contamination on agricultural environments by accidentalreleases of caesium and strontium radioisotopes from a nuclear power plant.Research efforts focused on studies of the soil-to-plant interface processeswith special emphasis on the quantification of key parameters which influen-ce and govern the fate of Cs and Sr in soils and their transfer to plant crops.

The objectives of the LANDSCAPE project were to obtain a basis forreliable assessments of the radiation exposure to man from radionuclides inplant and animal systems of semi-natural (forest) ecosystems in Europe underdifferent time scales. The work included studies of the seasonal and annualvariations of the radionuclide content in the main above-ground biological

I P S N - C E C A S S O C I A T I O N F I N A L R E P O R T 1

I P S N - C E C A S S O C I A T I O N F I N A L R E P O R T2

components of different representative forest ecosystems, the estimation ofradioactive caesium flow between these biological components and the soil,and quantification of the impact of classical forest management on the radio-nuclide contents and fluxes within the vegetation.

The objective of the EPORA project was to evaluate the potential effectof industrial pollution on migration of radionuclides in soil, on radionuclidesassociation in different soil constituents and on transfer of radionuclides fromsoil to plants. The potential impact of industrial pollution on runoff of radio-nuclides as well as radiation doses has also been studied. Therefore, an assess-ment was made of the current pollution status in the Kola peninsula with particular emphasis on the Monchegorsk area in Northern Fennoscandia.

The integration of these three projects within this association created aunique opportunity to link research efforts together, by bringing about expe-rimental data under laboratory conditions and research concerned with theproduction of field data bases, therefore allowing for a confrontation betweenthe different approaches with their underlying concepts for modelling. Thisreport shows the permanent effort of integration of the three projects and pro-vides a good illustration of the european synergy. In this report an overviewof the output results from this synergy is given. For more information on thethree projects readers should refer to the final project reports listed at the endof this document.

The success of this integration is mainly due to the constant efforts of thethree projects leaders, François Bréchignac, Leif Moberg and Matti Suomela.Working with them was a constant pleasure and a permanent enrichment.Wewish to thank them for their European spirit along the three years and thanksto such scientists we can be confident in the future of the scientific Europe.

The association chairmen wish to gratefully acknowledge the three projectleaders but also, the members of the association, Jean-Claude Barescut,Martine Contini, Gilbert Desmet, Laurent Le Yanc, Jack Sinnaeve, MartineWauters, Ernst H. Schulte, Remi Seydoux, and more especially our secretaryStéphane Lorthioir and Martine Jacquard who prepared all the associationmeetings with her usual efficacy.

Pr. Henri MÉTIVIER Mr Hans FORSSTRÖMDirector of Research IPSN Head of Unit DG RTD-DII-3 EC

■ It is of paramount importance to know the soil solution composition in order to under-stand radionuclide transfer to plants.

■ On untilled mineral agricultural soils, the migration of 137Cs and of 90Sr remain simi-lar, and the depth profiles are established early on, with very slow further evolution.

■ On the forest podzolic soils studied, migration of 137Cs is pronounced, with increasedamounts in the mineral layer ten years after an aerial deposition.

■ The classically used KD, as determined in vitro, does not allow an acceptable predictionof migration. The initial soil structure and hydric conditions upon contamination areimportant.

■ During the past decade, there is a general decrease of 60-80 % in 137Cs concentrationsin ground vegetation, fungi, mosses and lichens at both temperate and boreal forestsites studied.

■ More than a decade after Chernobyl, the total inventory of 137Cs is still rising in pinetrees of boreal forests.

■ The forest models currently developed yield agreeing estimates of the 137Cs concen-trations in the different compartments of the ecosystem, taking natural variability intoaccount.

■ An increasing load of heavy-metal pollution in boreal forests modifies the radionuclidesstatus in podzolic soils and their subsequent transfer to plants.

■ Fertilisation can be a beneficial tool for restoring contaminated forests, as well as forthe remediation of the state of agricultural lands in severe fallout situations.

■ There is almost no 137Cs loss via runoff water from boreal forest ecosystems exceptfrom the wetter portions of bogs.

■ Spatial variations in radionuclide deposition are dominant and largely cancel out evo-lution over time and the influence of climatic intermittence.

■ Variability of 137Cs uptake by moose is caused more by spatial variation in depositionand the normal movements of the animal than by variations in diet selection or intake.

■ The 90Sr contamination of plant food products is important for internal dose assess-ment and may be affected by additional non-radioactive pollution.

What have we learnt?



PEACE

PEACE

PEACE

PEACE

EPORA

EPORA

Norway

Belgium

France

Germany

UnitedKingdom

LANDSCAPE

LANDSCAPE

Spain


PEACE

LANDSCAPE

EPORA

EPORA

EPORA

Finland

Russia

Italy

Sweden

Kola Peninsula

LANDSCAPE

LANDSCAPE

LANDSCAPE

PEACE


François Bréchignac: Institut de Protection et de Sûreté Nucléaire, IPSN, France.Leif Moberg: Swedish Radiation Protection Institute, SSI, Sweden.

Matti Suomela: Radiation and Nuclear Safety Authority, STUK, Finland.

Moose photo on cover page: Per Claesson, Bildhuset

Copyright 2000 - Mission communication IPSN

Conception: Penez Communication - Tel: (33) 01 48 78 07 70

S U M M A R Y


■ 8 | Introduction

■ 10 | Redistribution in the soil-plant system

■ 29 | Modelling

■ 38 | Countermeasures (fertilisation)

■ 44 | Runoff

■ 49 | Spatial variations

■ 56 | Dose assessment implications

■ 60 | «What is still needed ?»


I N T R O D U C T I O N

The widespread radioactive environmental pollution which resulted from atmos-pheric nuclear weapons testing during the mid-years of this century prompted

the first concern among the scientific community about its potential impact onhuman health. Although the development of the civilian nuclear industry has evol-ved with a parallel effort of mastering the associated risks, the Chernobyl accidentin 1986 revealed, on a global scale, the radiological impact of radioactive pollutionof the environment. Assessing the resulting radiation doses that humans might recei-ve, either by exposure from radionuclides on the ground or by internal ingestion (viafood), requires a good understanding of the long-term behaviour of radionuclides inthe environment. This knowledge constitutes a basis for the prevention goals ofradiation protection.

This document reports on a joint European effort to advance this understanding,placing particular emphasis on the soil-plant system of semi-natural ecosystems andagricultural land. The problem has been tackled through various converging andinnovative approaches to the long-term behaviour of the radionuclides 137Cs, 90Srand 239/240Pu. The approaches range from in situ experiments, designed to exploitpast contamination events, to laboratory simulations in controlled conditions ofaccidental contamination, in view of mastering the wide variability of the naturalenvironment. Particular attention has been devoted to the philosophy of experi-mental data acquisition in order to secure the various modelling developmentsundertaken with appropriate validations. A broad scope of different ecosystems,ranging from arctic and boreal regions down to Mediterranean ones, has been consi-dered, with emphasis on forests and agricultural land, where food production takesplace. Up to 22 different site locations scattered throughout this territory have beeninvestigated, reflecting a wide, and representative, variety of European soils and cli-matic conditions. A brief description of the various ecosystems and soils consideredis given in Table 1.

Long-term behaviour of radionuclides in these ecosystems are most susceptible totheir vertical distribution and migration in the soils, as well as to their availabilityand uptake by plants and animals, processes which have been examined particularlyclosely, within time-frames ranging between a few years to several decades. Finally,the occurrence of additional non-radioactive pollutants in many ecosystems, whichcould invalidate our current prediction pertinence, has been questioned for the firsttime with respect to their potential influence on radionuclide movements.

This report is focused on six different themes, each of which having been stu-died at least in two of the three projects: redistribution in the soil-plant system,modelling, countermeasures, runoff, spatial variations, and dose assessment.More extensive information on the project results can be found in the finalreports of the individual projects (see short reference list).

Table 1

I N T R O D U C T I O N


It should be borne in mind that forest podzols are fundamentally different fromagricultural soils. Forest podzols present a clear multilayered vertical structurecharacterised mainly by a clay-poor mineral layer (sand) which supports anorganic layer rich in organic matter (>50 %). In contrast, agricultural soils pre-sent reduced amounts of organic matter (a few %) associated with higher claycontents.

Geographical Ecosystem Soil type Soil Particle size pH CEC OrganicSite Type Soil texture layer % clay % silt % sand (cmol/kg) matter %

■◗ Table 1 : Main soil characteristics of the various ecosystem/site locations investigated.

1 Tilled layer a pH (KCl)2 Organic layer b pH (CaCl2)3 Mineral layer n.d. Not determined

Mol (Belgium) Agricultural Podzol P1 4.1 14.6 78.3 4.6a 11.7 3

Sand

Jülich (Germany) Agricultural Luvisol P1 11.2 78.8 8.4 7.1a 13.4 1.5

Silt loam

Barcelona (Spain) Agricultural Luvisol P1 13.9 28.8 47.7 7.3a 16.4 2.3

Loam calc.

Belleville (France) Agricultural Fluvisol P1 5.8 13.7 79.0 4.1a 5.8 1.5

Loamy sand

Wellesbourne (U.K.) Agricultural Fluvisol P1 9.5 19.7 66.9 6.3a 17.8 3.9

Sandy loam

Hille (Sweden) Forest Podzol O2 n.d. n.d. n.d. 3.5b n.d. 58.2

Ferric M3 1.9 8.9 89.2 4.2b 2.0

Långsjön (Sweden) Forest Podzol O2 n.d. n.d. n.d. 3.6b n.d. 43.8

M3 3.7 28.1 67.5 4.1b 5.0

Prylen (Sweden) Forest Podzol O2 n.d. n.d. n.d. 3.7b n.d. 53.6

M3 3.4 17.1 79.5 3.9b 5.1

Vindeln (Sweden) Forest Regosol O2 n.d. n.d. n.d. 3.4b n.d. 66.8

M3 <11 26.1 69.7 4.4b 5.2

Liesineva (Finland) Forest Peatland O2 n.d. n.d. n.d. 3.2b n.d. 95.2

M3

Siikakangas (Finland) Forest Podzol O2 n.d. n.d. n.d. 3.1b n.d. 59.1

Ferric M3 2.6 17.5 66.3 4.1b 6.8

Lapland (Finland) Forest Podzol O2 n.d. n.d. n.d. 3.0b 193 79

M3 2.4 7.5 90.1 3.7b 24 3.2

Monchegorsk (Russia) Forest Podzol O2 n.d. n.d. n.d. 3.0b 206 76

M3 1.3 8.2 90.5 3.8b 30 8.6


● A g r i c u l t u r a l e c o s y s t e m s

● F o r e s t e c o s y s t e m s ( p o d z o l s )

● F o r e s t e c o s y s t e m s s u b m i t t e d

t o add i t i ona l po l l u t i on by h eavy me ta l s

REDISTRIBUTION IN THE SOIL-PLANT SYSTEM1

On a contaminated soil, the long-termpollution of plant food products occursvia root uptake. Prediction modelshave first tackled this problem byapplying a “transfer factor” coefficientdefined as the ratio between the speci-fic activity in the plant and the specificactivity in the soil. This empiricalapproach, however, suffers from highand poorly documented variabilities,which can largely be accounted for bythe plant species considered, but alsoby the complexity and diversity of thesoil compartment. This is illustrated bythe 137Cs and 90Sr transfer factorswhich have been measured simulta-neously on five distinct soil types, withvariable respective contents of clay,loam and sand, for three different plantconsumables: barley grains, bean podsand lettuce.

These data have been obtained fromlong-term experiments carried out in adedicated facility designed for simula-ting, in fully controlled conditions, anaccidental deposition of radioactiveaerosol contaminants. The facilityhosts various soil monoliths sampledthroughout Europe, which are moni-tored in lysimeters with advancedcontrol over hydrological fluxes, andconducted under artificially recons-tructed and computer-controlled cli-matic conditions.

The classification of the observedtransfer factors does not always agreewith soil physico-chemical characteris-tics alone. For 137Cs in particular, thisis due to complex variations of the soilsolution composition, which is a func-tion, not only of soil type, but also ofa number of other parameters such assoil moisture, climate, fertilisation androot uptake activity. The “soil solu-tion” composition is of paramountimportance, since this is the very placewhere roots absorb radionuclidesalong with dissolved mineralnutrients. Predictions have been refi-ned by introducing the “KD” concept,defined as the radionuclide partitioncoefficient between the soil liquid andsolid phases, reflecting either theimmobilization potential of a givensoil for a radionuclide (retention onthe solid matrix), or its availability toplants (presence as solute in the liquid

A g r i c u l t u r a l e c o s y s t e m s

A G R I C U LT U R A L E C O S Y S T E M S | Redistribution in the soil-plant system


figure 1


Redistribution in the soil-plant system | A G R I C U LT U R A L E C O S Y S T E M S

■◗ figure 1137Cs and 90Sr root transfer

(as concentration ratios,

Bq.kg-1 in plants /Bq.kg-1 in

soil) obtained on five distinct

agricultural soils for lettuce,

bean pods and barley grains.

PEACE

■◗ figure 2

Influence of soil moisture

on KDCs.

PEACE

137Cs Root Transfer (Concentration Ratio)

CR

(B

q/kg

) / (

Bq/

kg)

CR

(B

q/kg

) / (

Bq/

kg)

KD

134 C

s (%

cha

nge

wit

h re

spec

t to

sat

urat

ion)

Soils types

Soils types

Soil water content (% with respect to saturation)

90Sr Root Transfer (Concentration Ratio)



phase). This approach has made itpossible to take into account the well-known competition occurring bet-ween Cs and K for fixation on clays,or between Sr and Ca for adsorptionon the argilo-humic complex, whichboth influence the soil solutioncontent. Limits are imposed on thisconcept by the equilibrium nature ofthe KD coefficient, which is furtheroften determined in water-saturatedconditions. Such conditions are rarelyactually encountered in nature, due forexample to the continuous pumpingactivity of the roots or climate-drivenvariations in soil moisture.Based on a series of in vitro experi-ments, soil moisture is indeeddemonstrated to have a markedinfluence on the radionuclides concen-tration in the soil solution. For

example, reducing soil moisture pro-motes an increase in Cs concentration,thereby reducing its KD, tending tofavour root uptake. In parallel, the K concentration is alsoincreased, to an extent proportionallyeven greater than for Cs (Cs/K lowerfor reduced soil moisture).Furthermore, considering the rootuptake, it has been confirmed thatincreasing K in the soil solutionreduces 137Cs uptake due to competi-tion for the transport systems. Theultimate 137Cs root uptake, and sub-sequent plant contamination rate,results from a combination of theseeffects. This demonstrates the centralrole played by the soil solution com-position as a key element contributingto the variability of transfer factors.This is further supported by modelling

figure 2


Redistribution in the soil-plant system | A G R I C U LT U R A L E C O S Y S T E M S

■◗ figure 3137Cs and 90Sr migration

profiles observed 4 years

after contamination on

various agricultural soils.

PEACE

■◗ figure 4137Cs and 90Sr migration evolution

during three successive years on an

untilled silt loam soil installed in an

advanced lysimeter facility with

hydrological and climatic control.

Experimental contamination by

aerosols was performed in 1994.

PEACE

■◗ figure 5

Cs and Sr profiles observed

after 1,3.5 and 8 months fol-

lowing contamination in an

undisturbed soil driven in

columns submitted to a high

hydrological flux.

PEACE

137Cs Vertical Migration

137Cs Migration Evolution

Silwood Caesum and Strontium Activity ProfilesActivity concentration (Bq/g)

90Sr Migration Evolution

90Sr Vertical Migration

Dep

th (

cm)

Dep

th (

cm)

«Silt

loam

» so

il de

pth

(cm

)

«Silt

loam

» so

il de

pth

(cm

)D

epth

(cm

)



developments of root uptake, whichreveal that soil type and climaticconditions act on the soil-to-planttransfer essentially via the chemicalcomposition of the soil solution andthe physiological features of theplant: K and plant growth for 137Cstransfer, Ca and transpiration for 90Srtransfer.

When attempting to understandmigration patterns, the first elementsare furnished by chemistry, whichindicates that the high potentialmobility of 137Cs, due to its signifi-cant solubility, is counteracted by itsvery specific and efficient capturewithin clay particles (illite, in particu-lar). Conversely, the lower potentialmobility of 90Sr, due to its lower solu-bility, is not counteracted to the sameextent by specific fixation sites on thesoil matrix. This results in clearly dis-tinct KD values for the two radionu-clides, as determined in vitro, fromwhich a higher migration rate isexpected for 90Sr than for 137Cs afterseveral years. However, experimentalobservations do not fit this theoreti-cal approach.

On the one hand, the two radionu-clides developed migration profiles,on a given agricultural soil (untilledplots), which still show similar shapesfour years after the contamination

deposition. Even the soil type, whichis traditionally characterized by aradionuclide-specific KD coefficient,does not significantly alter the obser-ved profiles.On the other hand, the low rates ofmigration observed in the long term(around 1 cm/year in the years follo-wing contamination) cannot explainthe shape of the profiles. In the shorter term, when experimen-tally forcing migration on soilcolumns with a high hydrologicalflux, the two radionuclides exhibitprofiles which are established withinone month, with very limited furtherevolution. Altogether, these observations stron-gly suggest that the migration profilesare established very early after conta-mination, under the influence of ini-tial processes that do not depend pri-marily on the soils’ KD coefficients, asdetermined in vitro. Their evolution issubsequently very slow. This indicatesthe importance of the early conditionsprevailing immediately after contami-nation, such as soil moisture and firstrain events, which may be paramountin determining the extent to whichradionuclides will penetrate in depth.This aspect particularly deserves fur-ther consideration.

figure 5

figure 3

figure 4


Redistribution in the soil-plant system | F O R E S T E C O S Y S T E M S ( P O D Z O L S )

■◗ figure 1

The vertical distribution

of 137Cs (%) in the various

horizons of forest soils in

Fennoscandia:

a) Hille, b) Långsjön,

c) Prylen, d) Siikakangas and

e) Vindeln.

LANDSCAPE

a) Hille Cs-137 (%)

c) Prylen Cs-137 (%)

e) Vindeln Cs-137 (%)

d) Siikakangas Cs-137 (%)

b) Langsjön Cs-137 (%)

F O R E S T E C O S Y S T E M S ( P O D Z O L S ) | Redistribution in the soil-plant system


In contrast to agricultural soils, theforest podzol soils are characterizedby a multi-layer structure, or hori-zons: organic horizons (Ol=non-frag-mented litter, Of=fragmented organicmatter, Oh=humified organic matterand the OAh and Ah horizons whichare organic-rich mineral horizons),and mineral horizons (A and B).From studies after the Chernobyl acci-dent, it is known that most of the cae-sium during the first years was locatedin the raw humus layers. The verticaldistributions of 137Cs in podzol soilsin the regions of this study show atrend towards redistribution from thevegetation and raw humus layersdown to the mineral layer. By 1995 to 1998 after deposition, between 20 and 40% of the originaldeposit had reached the mineral layer.Although there is a general trend todownward migration, the depth distri-bution varies between different loca-tions. The soil particle distributionaffects the hydraulic conductivity andcapillarity. Dense soil with low hydrau-lic conductivity and high capillaritywill probably be more effective in retai-ning caesium in shallow layers due toslow percolation, high ion-exchangecapacity and an upward water flowduring warm periods. Biotic factorsmay also lead to variations in soilmicroflora which may influence thecaesium distribution. The depth distri-bution is further affected by differences

in clay content in the mineral layerwhich, in areas with high contents, maytrap caesium in the mineral fraction.

The vertical redistribution of radiocae-sium in podzol soils is also reflected asa decreasing rate in photon fluence asmeasured by in situ gamma spectrome-try. Between 1986 and the late 1990´s,the photon fluency at 662 keV decrea-sed faster by a factor of at least twothan what would have been expectedfrom the physical half life of 137Cs.Except for the exponential decrease,temporal fluctuations up to a factor oftwo were observed. These are due tovariations in soil moisture and lead tolower external doses during winter,spring and autumn compared to thesummer.

F o r e s t e c o s y s t e m s ( p o d z o l s )

figure 1



■◗ figure 2

Temporal changes of 137Cs (Tag)

in green parts of heather and

current shoots of young Scots

pine for three regions:

Saint-Léger, Belgium (red);

Hille, Sweden (blue) and

Vindeln, Sweden (green).

LANDSCAPETag

(Bq/

kg)

/ (kB

q/m

2 )T

ag(B

q/kg

) / (

kBq/

m2 )

Time (years after the Chernobyl accident)

Time (years after the Chernobyl accident)



The changes in the vertical distributionof 137Cs in soil have implications forplant root uptake, which, depending onspecies, can cause either an increase ora decrease in the uptake of caesium. Atboth boreal (Hille) and temperate sites(Belgium and Luxembourg), plants onaverage lost 70% of their activity overthe past decade. In relative terms,however, major differences were obser-ved in connection with this loss, whichsurprisingly tallied with the taxonomicfamilies. For instance, most of theplants in the Asteraceae, Fabaceae andLiliaceae families were characterised bya loss exceeding 70%. In contrast, theloss in most of the grasses and Rubusspecies was considerably less than70%. The loss in Ericales and ferns wasin the middle range.

The relationship between rooting depthand radiocaesium migration may partly explain the differences in thecontamination trends of plants. This isundoubtedly the most likely explana-tion in the case of the shallow-rootingOxalis acetosella and in pine, whichhas a deep tap root. Furthermore, thecontribution of successive cohorts ofever deeper-rooting mycorrhizal fungiin this species cannot be excluded.Most of the radiocaesium stored in theLiliaceae runners has probably been lostto the soil surface horizon, in whichthere has been a general decrease of137Cs due to migration. On the other

hand, the fibrous rootsystem of grassesenables them to maintain firm anchora-ge in deeper humiferous horizons thatare still relatively heavily contaminated.Ericales and ferns often have runningroots. However, they also producenumerous, more or less fibrous, adven-titious roots that explore deep soillayers. One should also bear in mindthat these plants are mycotrophic, andtherefore the buffer role of mycorrhizalfungi in the decontamination of thesespecies cannot be precluded.Aggregated transfer factors, Tag(Bq.kg-1 per kBq.m-2) can be used tocompare data from different sites.Such a comparison shows that the Tagvalues in vegetation ranged from 0.2to 241 (10-3 m2.kg-1). The lowestvalues were found in young plants ofdeciduous trees and in herbs, while thehighest values were found in dwarfshrubs. On average, the general trendfor 137Cs in the vascular plants inves-tigated from seven studied sites was:Salix spp < Betula spp < Epilobiumangustifolium < Deschampsia flexuosa< Pinus sylvestris < Vaccinium spp <Calluna vulgaris, with the same varia-tions between the sites.

A comparison of the long-term annualchanges in aggregated transfer factorsshows that the activity in heather(Calluna vulgaris) changed in a verysimilar way over time for the sites studied. figure 2



For Scots pine (Pinus sylvestris), the Tagvalues have increased dramaticallysince a few years after the deposit.Although the Tag amplitudes are diffe-rent, the rate of increase is equal bet-ween the sites, indicating (as in the caseof Calluna vulgaris) that a general pro-cess is at work. This can be a protractedredistribution of Cs in the forest floorfrom the mosses and litter layers downto the humus and mineral soil, wherepine roots are found. The monotonous-ly increasing Tag values in young pinesneed further investigation, since thismay be of significance for externaldoses, internal doses via meat frommoose and gallinaceous birds, for fores-try in general (biofuels) and the steady-state levels for the different compoundsin the forest ecosystem.Mushrooms are the most heavily

contaminated organisms in forest eco-systems and many of them play a directand significant role in the contamina-tion of food chains by radioactivity.The contamination and decontamina-tion curves of mushrooms reflect theevolution of the activities in the varioussoil horizons. This characteristic has tobe considered in the light of the specificuse of soil horizons by the mycelium ofthese organisms. The data availableindicate an overall reduction of 137Csconcentrations in mushrooms between1991 and 1998. When comparing theaverage activity, during this period, ofobligate mycorrhizal species (usuallycharacterised by a deep mycelium)with that of facultative mycorrhizalspecies (which have a more superficialmycelium), a reduction of 60% and80% is observed, respectively.



A rough assessment of the proportionof caesium retained in fruitbodies intheir growing area shows that the ave-rage value in Cortinarius alboviolaceu-sis was about 5% of the deposition,whereas it varied from 1.3 to 1.7% inother mushrooms. In mushrooms thathave lost 70% of their contaminationover the past seven years, one can esti-mate that at least 5% of the totaldeposition had been retained at thetime of maximum fruitbody contami-nation. There is a need for continuingthe investigations on the role ofmycorrhizae in the active radiocae-sium transfer from the soil to hostplants.

Trees are often the most significantcomponent in a forest ecosystem froman economical point of view.Concentrations of radiocaesium in dif-ferent parts of trees vary, and are chan-ging with time. The use of trees inforest industry and the question ofaccess of people to heavily contamina-ted forests motivate analysing the dis-tribution of man-made radionuclidesin trees in the short and long term. Root uptake adds to the radionuclidecontent of the tree roughly in propor-tion to growth intensity and thusincreases the radiocaesium content ofthe tree which originates from metabo-lised, initially deposited, radiocaesium.

This caesium will be found in bothabove-ground parts and in roots, andvaries with the season of the year.

Growth dynamics cause fluctuation inradiocaesium contents in coniferneedles of different ages during thegrowth period. Different growthconditions add to the variation bet-ween sites.

There is an overall increase in 137Csconcentration in the stem wood of pinebetween 1991 and 1997. The concen-tration of 137Cs in the new growth oftwigs and needles of these treeshas also increased during the period bet-ween the two samplings, due to rootuptake. The caesium concentrations arehighest in the youngest twigs, which isconsistent with other measurements.The total activity, in percentage of theground deposition, stored in the above-ground parts of the trees of the olderstand has increased from 1.1% in 1991to 2.0% in 1997.

figure 2


Redistribution in the soil-plant system | FOREST ECOSYSTEMS SUBMITTED TO ADDITIONAL POLLUTION BY HEAVY METALS

■◗ figure 1

Characterisation of the

heavy metal pollution

gradient generated around

the smelter. Illustration of

the soil contents in Cu and

exchangeable K observed at

7 up to 152 km from the

smelter, and the correspon-

ding plant content in Cu and

K as measured within 4

understorey plant species.

EPORA

(mg.

kg-1

)(m

g.kg

-1)

(mg.

kg-1

)(m

g.kg

-1)

Distance from smelter (km)

Soils

Plants

The Severonikel smelter atMonchegorsk, Kola Peninsula, hasreleased huge amounts of SO2, Ni andCu into the atmosphere since itsconstruction in 1938. As an examplethe annual emissions released in 1984amounted to about 270,000 tons forSO2, 3,200 tons for Ni and 2,400 tonsfor Cu. Annual emissions had beenreduced by about one half in 1994 butalthough decreasing, considerable pol-lution of the surrounding environmenthas been going on. This pollution hascaused a dramatic alteration in thevegetation status as observed in thevicinity of the plant.

A strong pollution gradient has beengenerated around the smelter, as illus-trated in . The concentrations

of heavy metals in the upper soil horizons(litter and O horizons) show a verysharp increase as distance from thesmelter decreases (the reference site

was located 152 km from the smelter).The total concentrations of Ni and Cuin the organic layer increase regularly,from about 10 mg.kg-1 at the referencesite up to about 5,000 mg.kg-1 at themost polluted site. Near the smelter (7 km),concentrations in the underlying mine-ral horizons are also higher, indicatingsome vertical migration from the surfacelayers, but the trends observed indicatethat only a very small fraction of themetals has leached to levels lower thanthe maximum sampling depth. Thetotal amounts of nickel or copper at agiven site may therefore be used torepresent the integrated pollution effectat that site, which can be compared tothe other sites. Only minor changes areobserved in soil pH and cation-exchange capacity with distance from

the smelter, their respective vertical distribution being usual for suchpodzols. The heavy-metal accumulationin plants displays a similar trend, with

Fo r e s t e c o s y s t e m s s u b m i t t e d t o a d d i t i o n a l p o l l u t i o n b y h e av y m e t a l s

FOREST ECOSYSTEMS SUBMITTED TO ADDITIONAL POLLUTION BY HEAVY METALS | Redistribution in the soil-plant system


figure 1



■◗ figure 2

Influence of heavy metal loads

on the distribution of 137Cs, 90Sr

and 239+240Pu in the litter, the

root zone (O and E horizons)

and below the root zone,expressed

as percentages relative to the

total inventories in the overall

soil profile.

EPORA

■◗ figure 3

Influence of heavy metal loads

in the soil organic layer on

the 137Cs exchangeable and

persistently bound fractions as

determined through sequential

extractions.

EPORA

Litter

Root zone

Below root zone

Cs sequential extractions

% a

ctiv

ity

wit

hre

spec

t to

tot

al%

act

ivit

y w

ith

resp

ect

to t

otal

% a

ctiv

ity

wit

hre

spec

t to

tot

al%

of

tota

l Cs

inth

e or

gani

c la

yer



higher concentrations in the vicinity ofthe smelter.In view of this pollution gradient, it isof great interest to examine how thepresence of heavy metals is affectingthe soil nutrients status and their uptake by plants. Concentrations of exchangeable K, Mg, and Ca allexhibit a strong depletion in the O-layer(organic) with increasing pollution

levels, whereas there is an increase in thelower horizons (E and B).The corresponding contents of thesenutrients within plants follow a trendsimilar to that observed for the exchan-geable fraction in the O horizon, withhighly depleted concentrations near thesmelter.

The additional pollution by heavymetals (Cu and Ni) clearly affects thedistribution of radionuclides (137Cs,90Sr and 239+240Pu) in the various soilhorizons. In the litter layer, the radio-nuclide content (expressed as a per-

centage of the total inventory in theoverall soil profile) increases withdecreasing distance from the smelter(higher load of heavy-metal pollution). This increase is especially sharp for239+240Pu, which rises from 1% at thereference site up to 56% at the mostpolluted site. Conversely, in the root zone(O and E horizons), the radionuclidecontent decreases versus decreasing

distance from the smelter.These observations most likely resultfrom an increased thickness of the litterlayer at the most polluted sites (vicinityof the smelter), with a parallel thicknessreduction in the organic layers (rootzone), both being promoted by aninhibition (by heavy metals) of themicrobial activity which converts litterinto humic substances. Below the rootzone, 90Sr and 137Cs are also affected.This is most evident for 90Sr, thecontent of which increases in the B-layerfrom about 10 % at the reference site(152 km from the smelter) up to 50 %



figure 2

figure 1



■◗ figure 4

Concentrations of 137Cs (Bq.kg-1

dry weight) and aggregated trans-

fer factors (Tag,m2.kg-1) of 137Cs,

90Sr and 239+240Pu from the global

fall-out for four plant species

as affected by the heavy metal

pollution load at various distances

from the Cu-Ni smelter.

EPORA Distance from smelter (km)

(Bq.

kg-1

)T

ag(m

2 .kg

-1)

Tag

(m2 .

kg-1

)T

ag(m

2 .kg

-1)



at the most contaminated site. Noeffect, however, can be observed for239+240Pu. It is to be stressed thatthese results are also due to differencesin the solubility of the radionuclides inthe soils studied.A clear influence of this additional pol-lution by heavy metals has also beenobserved on the availability of radionu-clides in the soil’s organic layer (O hori-zons), as examined through sequentialextractions of radionuclides from thesoils. The effect is most marked for137Cs, the exchangeable fraction ofwhich is significantly reduced forincreased heavy-metal loads. Also, for 90Sr and 239+240Pu, theexchangeable fractions were lower atthe most polluted site (35 % and 1 %,respectively) as compared to thosefound at the other sites (46 % and 4-6 %, respectively). At each site, 90Srwas essentially distributed between theexchangeable, bound-to-oxides andbound-to-organic matter fractions,with only 1-2 % persistently bound,whilst 85-90 % of 239+240Pu wasbound to the organic fraction.

Radionuclide behaviour in plants alongthe heavy-metal pollution gradient wasobserved by measuring the 137Cs, 90Srand 239+240Pu content of four unders-torey species at the five sites, rangingfrom 7 up to 152 km from the smelter,each site consisting of five distinctplots. The plant activity concentrations

ranged from 10 to 100 Bq.kg-1

(dry weight) for 137Cs, from 1 to 17 Bq.kg-1 for 90Sr and from 0.001 to0.019 Bq.kg-1 for 239+240Pu, with considerable intra-site variability.Vaccinium myrtillus, Empetrumnigrum and Vaccinium vitis-idaea (to alesser extent) all displayed reduced137Cs contamination levels at higherpollution loads, whereas Deschampsiaflexuosa, the only grass-like species,exhibited higher contamination levels.

Except for Vaccinium vitis-idaea,which tended to remain unaffected, the90Sr activity concentration also decreasedin all plants at higher pollution loads incontrast to 239+240Pu, which increased.

When taking into account the totalradionuclide inventory in the soil profileconsidered, the aggregated transferfactors can be calculated for each plant(Tag, m2.kg-1) (Figure 4). Rangingfrom 0.01 to 0.11 m2.kg-1, the 137CsTag values agreed with the relativelylarge range reported in the literaturefor understorey forest vegetation.Except for Deschampsia flexuosaagain, the median Tag values for 137Cswere observed to decrease at higherpollution loads (reduced distancesfrom the smelter). When excluding thereference site and only considering thefour polluted sites, the effect was sta-tistically significant, except forVaccinium vitis-idaea. These results

figure 4

figure 3



agree with the data on the easilyexchangeable 137Cs fractions in theroot zone, previously described, whichalso decrease at higher pollution loads(Figure 3). It is interesting to point outthat this is quite similar to the Kbehaviour. In contrast, transfer factors ofDeschampsia flexuosa tended toincrease at higher pollution loads, butit is important to note that this plantexhibited considerably clearer physio-logical alterations than did the others,especially at the most polluted site. Theaggregated 90Sr transfer factors do notappear to be as sensitive to the pollu-tion load as 137Cs is. Only Vacciniummyrtillus and Empetrum nigrum fromthe two most polluted sites show atrend similar to that observed for137Cs. These results broadly agree withthe easily exchangeable 90Sr fractionsin the organic horizon, as reportedabove. In contrast to the general trendsobserved for 137Cs and 90Sr, the aggre-gated 239+240Pu transfer factors increa-sed significantly at higher pollutionloads. This seems particularly clear forVaccinium myrtillus, Empetrumnigrum, and Deschampsia flexuosa, noeffect being detectable for Vacciniumvitis-idaea. However, it should be stres-sed that this could be an artefact resul-ting from the higher amounts of

239+240Pu in the litter layer at the mostpolluted sites, which might have led tosuperficial contamination of the plants.(Due to its extremely small soil-to-plant transfer, the Tag values for Pu areespecially sensitive to superficial conta-mination of plants by soil).

It is of great interest to compare thetrends observed for radionuclidebehaviour along the pollution gradientwith those observed for the distributionof nutrients and heavy metals in thesoil-plant system. It appears thathigher inputs of additional pollutantspromote a reduction of 137Cs and90Sr transfer factors (at least forsome plants), which parallels aconcomitant decrease in nutrientcations, such as K, Ca, and Mg, inthe organic horizon. Conversely, themost polluted sites also exhibithigher contents in exchangeablemetallic cations (Al, Fe, Ni, Cu) inthe organic layer, where root colonisa-tion is most pronounced. However, itshould be stressed that the effectsobserved are not similar for all plantsand all radionuclides considered,demonstrating that some specificplant and radionuclide features areinvolved which are still unresolved.

● R o o t u p t a k e i n m i n e r a l

a g r i c u l t u r a l s o i l s

● M i g r a t i o n w i t h i n m i n e r a l

a g r i c u l t u r a l s o i l s

● R e d i s t r i b u t i o n o f r a d i o c a e s i u m

i n f o r e s t e c o s y s t e m s

MODELLING2


I P S N - C C E A S S O C I A T I O N F I N A L R E P O R T30

Modelling | R O O T U P T A K E I N M I N E R A L A G R I C U LT U R A L S O I L S

■◗ figure 1

Soil solution 137Cs and K

evolution in the root vicinity

during a barley crop, and

their observed/predicted

accumulation in plants.

PEACE


[Cs]

at

root

sur

face

(B

q.l-1

)A

ccum

ulat

ed C

s (k

Bq.

m-2

)

Time (days)

[K]

at r

oot

surf

ace

(mM

)A

ccum

ulat

ed K

(g.

m-2

)

R o o t u p t a k e i n m i n e r a l a g r i c u l t u r a l s o i l s


If the soil solution composition isaffected by soil parameters (such astexture, moisture content, counter-ions availability, etc…), it is alsoinfluenced by the root uptake activityitself, which is primarily designed tosupply such nutrients as K and Ca tothe plant in amounts suitable to sup-port appropriate growth. Based on thefinding, derived from recent advancesin plant biology, that 137Cs and 90Srare withdrawn respectively by the K- and Ca-transport mechanisms, theanalytical model developed (RUR:Root Uptake of Radionuclides) takesthe physiology of plant growth intoaccount in that it reproduces theplant’s ability to adjust the rate of sup-ply to its demand. By this means, thedominant process leading to 137Csaccumulation in plants can vary accor-ding to soil type, and even changeduring crop growth on certain soils.Large temporal variations of radionu-clide accumulation in plants have

indeed been observed during growth,and these are accurately predicted bythe model. But also, in the event of drastic Kdepletion that roots may promote intheir vicinity under certain soil condi-tions, plants can regulate the efficiencyof their transport systems in such away that soils with high initial 137Csand K concentrations in the solutionmay lead to 137Cs accumulations inplants that are similar to those foundin soils with lower 137Cs and K. Theempirical relationship, often reported,between the uptake of a radionuclideand its nutrient analogue concentra-tion is now reaching an appropriateresolution, which clarifies the influen-ce of soil type, agricultural practicesand time. The main conclusion towhich the RUR model leads is thatradionuclide bioavailability is not aconstant for a given soil, but rather isvariable depending on crop type andon time.

Assessment models currently used forprediction of radionuclide behaviourhave generally been derived empiricallyfrom given contamination situationsagainst which they have been calibra-ted. Largely ignoring the actual pro-cesses involved, they often lack a moregeneric nature, which would allowtheir broader and more secure applica-tion. This is where analytical model

development can provide substantialimprovements, since they provide abetter understanding of what is actual-ly occurring. This strategy is illustratedby model-based clarification of therelevant mechanisms involved in plantcontamination via root uptake andmigration through confrontation withexperimental observations.

figure 1

R O O T U P T A K E I N M I N E R A L A G R I C U LT U R A L S O I L S | Modelling


Modelling | M I G R A T I O N W I T H I N M I N E R A L A G R I C U LT U R A L S O I L S

■◗ figure 2

Model-based predictions of

137Cs and 90Sr migration against

experimental data obtained

in a «silt loam» agricultural soil

(three replicates).

PEACE

137 C

s ac

tivi

ty (

% o

f to

tal d

epos

itio

n)90

Sr

acti

vity

(%

of

tota

l dep

osit

ion)

Depth (cm)

Depth (cm)

M I G R A T I O N W I T H I N M I N E R A L A G R I C U LT U R A L S O I L S | Modelling


M i g r a t i o n w i t h i n m i n e r a l a g r i c u l t u r a l s o i l s

The simulation of the hourly-to-yearlyvertical migration of 137Cs and 90Sr ina saturated/unsaturated soil columnwith vegetation is undertaken based ona one-dimensional, dynamic physical-ly-based model formulation (TRANS-SOL model). The major processesinvolved in water flow and transfer ofradionuclides have been taken intoaccount. Mechanisms such as down-ward migration enhancement due tomacropores, water exchanges with sur-face atmospheric layer and time-depen-dent sorption of radionuclides onto thesoil matrix are accounted for, as well asthe slow initial dissolution of aerosolsand the influence of soil moisture onthe radionuclide concentrations in thesoil solution. Despite this accurate des-cription effort, introducing the radio-nuclides’ respective KD coefficients, asderived from in vitro sorption/desorp-tion experiments, shows migrationprofiles which are underestimated for137Cs, and overestimated for 90Sr.

However, the model also demonstratesthat prediction of the observed profilesis improved within the first 15 cm soillayer for both radionuclides when apartition coefficient of about 600 isintroduced. This suggests that migra-tion is governed not only by physico-chemical features accounted for by thein vitro radionuclide KD, but also byadditional soil properties which, whentaken into account, result in a more

appropriate overall “global retentioncoefficient”. In addition to soil type, asdescribed by the respective KD valuesof the radionuclides, migration alsoprobably depends on the soil structure(particle aggregation) and the associatedbiological activity of the soil (microbialand root activity), which can beaccounted for by a “global retentioncoefficient”.

figure 2


Modelling | R E D I S T R I B U T I O N O F R A D I O C A E S I U M I N F O R E S T E C O S Y S T E M S

■◗ Table 1

Reliability Indices (RI),

showing the level of agree-

ment between the model

predictions themselves

(Km-m) and between the

models predictions and the

experimental data (Km-d).

A RI of 2 corresponds to an

agreement within a factor

of 2.The smaller the RI,

the better the agreement.

LANDSCAPE

Variables Km-m Km-d

All variables - 2.1

Litter 1.9 -

Litter/Soil 1.9 1.15

Needles 2.2 1.8

Tree 1.9 3.0

Wood 2.9 3.2

Berries 1.6 1.5

Moose 1.4 -

1- Swedish Defence Research Establishment 2 - University of Trieste3 - Swedish Radiation Protection Institute. The development of FORESTLAND was performed in close co-operation with the Russian Institute of AgriculturalRadiology and Agroecology, Obninsk, Russia (SSI Project RYS 6.15).4 - It should be noted that part of the data used for parameterisation of FORESTLAND was collected in experimental plots located in the Bryansk area.

R E D I S T R I B U T I O N O F R A D I O C A E S I U M I N F O R E S T E C O S Y S T E M S | Modelling


R e d i s t r i b u t i o n o f r a d i o c a e s i u m i n f o r e s t e c o s y s t e m s

The modelling of radionuclide migra-tion and transfer in forest ecosystemshas been an area of scientific investiga-tion since the advent of radioecology inthe late 1950’s and early 1960’s. TheChernobyl accident led to radioactivecontamination of vast forested areas.The knowledge obtained from studiescarried out in these forests has beenused in the development of several newmodels.

Based on a general conceptual modeland the results of a review of existingmodels, three models were developed:the FOA model 1, the LOGNAT model 2(LGN) and FORESTLAND 3 (FRLD).These three models are based on thecompartment model principles andfirst-order kinetics for the turnover ofcaesium in the boreal forest. However,owing to the individual strategies cho-sen for describing and structuring theecosystem and the different emphasisput on certain transfer and turnoverprocesses, the resulting models exhibit,in essence, different characteristics.

The three models are also based onexperimental data obtained in differentgeographical areas. This has resulted indifferences in the model structures andin the values assigned to differentmodel parameters. To what extent dothese differences lead to disagreementin the predictions made with themodels? To what degree would model

predictions agree with experimentalobservations made independently inother areas, i.e. not used for calibratingthe models? To answer these questions,a two-step inter-comparison exercisewas carried out with the following spe-cific objectives:

1. To determine the level of agreementbetween the predictions of radiocaesiumlevels in different forest componentsafter an aerial deposition (model-modelcomparison).2. To assess the agreement of the modelpredictions with experimental dataobtained in field studies (model-datacomparison).

The experimental data for step 2 wereobtained from pine forests in theBryansk 4 region of Russia that werecontaminated by the Chernobyl acci-dent, and these data were collectedindependently from 1986 to 1994. Theexperimental values were not discloseduntil after the results of the calculationshad been submitted.

The model predictions were made on ayearly basis for the first 50 years follo-wing the contamination event. Theresults were presented for thevariables/endpoints: Litter, Litter/soil,Needles, Tree, Wood, Berries, Moose.Table 1 shows the level of agreementbetween the predictions for both steps 1and 2 of the inter-comparison. Table 1


Modelling | R E D I S T R I B U T I O N O F R A D I O C A E S I U M I N F O R E S T E C O S Y S T E M S

■◗ figure 1

Experimental values versus model

predictions for the variables

Needles,Wood,Tree and Berries

and for all three models (up).

Model predictions and observed

values of the radiocaesium

concentration in needles (down).

The dotted lines correspond

to the interval of variation

of the experimental values.

LANDSCAPE

Mod

el p

redi

ctio

ns

137 C

s co

ncen

trat

ion

(Bq.

kg-1

/ kB

q.m

-2)

Experimental data

Years

R E D I S T R I B U T I O N O F R A D I O C A E S I U M I N F O R E S T E C O S Y S T E M S | Modelling


Model predictions for all three modelsversus the experimental data forneedles, wood, tree and berries, areshown in Figure 1 (up). Figure 1 (down)also shows the results for needles inclu-ding the 50-year predictions, the experi-mental values and the natural variabili-ty in the experimental data.

The predictions made with the modelsfor the scenario studied were in relati-vely good agreement (within a factor of1.4 to 2.9) for all the variables conside-red. There was also good agreementbetween the model predictions and theexperimental data (within a factor of1.15 to 3.2). There were no significantdifferences between the models studiedas to how they reproduced the experi-mental data. This suggests that, at leastfor the scenario concerned and for thefirst ten years after the deposition, allthree models are equally serviceable ifthe final aim is to estimate absoluteconcentrations in different forest com-

ponents. The agreement between themodels, however, decreases with time,and there were differences in the formof the time dependencies predicted bythe models, especially for wood. Thismay lead to larger differences,although probably within a factor of10, between the model predictions andthe experimental data for time intervalsin excess of the period for which datawere available for comparison. Fromthe experimental data (due to the highvariability) it was not possible to dis-cern which model predicts best thekinetics.

The comparison with the experimentaldata confirms the fact that a majorquestion remaining to be solved is whe-ther the models correctly predict thekinetics of radiocaesium levels in diffe-rent forest components. The experi-mental data used in the comparison didnot allow this question to be answereddue to the high spatial variability ascompared to the observed time varia-tions. Any similar data set collected for vali-dation purposes will probably encoun-ter the same problem. To answer thisquestion, longer time series need to beused. Alternatively, field or laboratoryexperiments can be carried out to testthe hypothesis underlying the models,and from them conclusions drawnabout the accuracy of the time dyna-mics predicted.

figure 1


COUNTERMEASURES (FERTILISATION)3

● O n a g r i c u l t u r a l s o i l s

● O n f o r e s t s o i l s

O n a g r i c u l t u r a l s o i l s

O N A G R I C U LT U R A L S O I L S | Countermeasures (fertilisation)

Fertilisation of agricultural crops hasoften been reported as a potentialmeans for reducing the contaminationof plant food products. From initial,sometimes confusing, experimentalresults, a consensus has evolved thatthis can be expected to be most effecti-ve in soils with low retention capacity(low KD or CEC) and reduced fertility(especially, low K and Ca concentra-tions in the soil solution). Focusing onthe soil solution composition providesan efficient approach to understandingthis feature. The 137Cs and 90Srconcentration factors in lettuce andbeans (CF, expressed with respect to thesoil solution) decrease logarithmicallywith increasing concentration of theirrespective analogue ions, K and Ca+Mg.

The influence of analogue ions is there-fore stronger in the low range ofconcentrations-most likely in soils withreduced fertility. Measured on varioussoil types, the degree of correlation bet-ween CF and the concentration in ana-

logues remains high, suggesting thatplant contamination is primarily sus-ceptible to the status of analogues inthe soil solution, irrespective of the soiltype. This agrees with the principle ofradionuclide competition with its ana-logue ion occurring at the root level formembrane-based ionic transport sys-tems.

It is of interest to note that concurrentwith this competition effect at the rootlevel where plant radionuclide uptake(CF) is reduced by increasing concentra-tions of analogue ions, the radionuclideavailability at the soil matrix/liquidinterface (KD) is increased. The resultingplant contamination is therefore a com-bination of both effects, CF being moresensitive to analogues than KD, especiallyin the low concentration range. This iswhy the fertilisation effect on CR(expressed with respect to the bulk soil)is most often limited on traditionally fertilized agricultural soils.

figure 1


figure 2

figure 3


Countermeasures (fertilisation) | O N A G R I C U LT U R A L S O I L S

■◗ figure 1

Influence of the K

concentration on the 137Cs

plant contamination

(CF: concentration factor,

Bq.kg-1 in plants / Bq.l-1 in soil

solution) for lettuce and beans.

PEACE

■◗ figure 2

Influence of the Ca+Mg

concentration on the 90Sr

plant contamination

(CF: concentration factor, Bq.kg-1

in plants / Bq.l-1 in soil solution)

for lettuce and beans.

PEACE

■◗ figure 3

Fertilisation effect on the

137Cs Concentration Ratio (CR,

Bq.kg-1 in plants / Bq.kg -1 in soil)

in lettuce observed on various soils.

PEACE

137 C

s C

F (

l/kg)

90S

r C

F (

l/kg)

K (mM)

Ca + Mg (mM)

137 C

s C

R o

rK

(m

Mx1

0-1 )

Soil type

O N F O R E S T S O I L S | Countermeasures (fertilisation)


O n f o r e s t s o i l s

When assessing the influence of the dif-ferent forest ecosystems on the exposu-re of man to radiation, it is also neces-sary to understand how radionuclidescirculate in forest ecosystems that havebeen subjected to various managementmeasures. The most common manage-ment measures are fertilisation, limingand site preparation. Potassium is pro-bably the most interesting fertiliseraffecting the distribution of radiocae-sium in forest ecosystems. Generally,peat soils contain quite small amountsof potassium compared to the othermain nutrients and compared to theamounts bound in the tree stands.Mineral soils, on the other hand, nor-mally contain enough potassium forthe tree growth. Instead, nitrogen isquite often a growth-limiting factor onpodzol soils.

It is therefore of interest to investigatethe effects of potassium fertilisation onthe radiocaesium distribution in Scotspine stands on peat and mineral soilsand to assess whether fertilisation can bea possible countermeasure or restora-tion technique in a radioactive falloutsituation. Such fertilisation experimentswere located in western Finland, wherethe radioactive fallout resulting from theChernobyl accident was relatively high.

The 137Cs concentrations in bark,wood and needles of Scots pine, aswell as in many samples of branches of

different ages, were found to be lowerin the fertilised plots than in thecontrol plots, both on mineral soil andon peat-land. For dead branches anddead portions of living branches, somesamples of branches of different agesand of needles more than four yearsold, 137Cs concentrations did not differ between treatments.

On the mineral soil site, the 137Csconcentration was lower inDeschampsia flexuosa and Vacciniummyrtillus on the fertilised plots compa-red to the control plots. For grasses,herbs, lichens, mosses and other dwarfshrubs (Empetrum nigrum andCalluna vulgaris) no statistically signifi-cant difference in the 137Cs concentra-tion was observed between the treat-ments. However, at the fertilised plots,the concentrations were only 15-50 %of those at the control plots in thesespecies.

On the peat-land site, Vaccinium myr-tillus, tree seedlings, lichens, mossesand other dwarf shrubs (Empetrumnigrum and Calluna vulgaris) hadlower concentrations of 137Cs than onfertilised plots. In herbs, grasses, deadvegetation, Dryopteris carthusianaand Vaccinium vitis-idaea, theconcentrations at fertilised plots were38–68 % of those at control plots,although the differences were not sta-tistically significant.


Countermeasures (fertilisation) | O N F O R E S T S O I L S

■◗ figure 1

The relative distribution of

137Cs in pine-dominated

forests on mineral soil and

on peat-land (up), and the

distribution of 137Cs in

vegetation (down) as a

percent of the 137Cs

budget of pine-dominated

forests on mineral soil and

on peat-land.

LANDSCAPE

137 C

s di

stri

buti

on (

%)

137 C

s di

stri

buti

on (

%)

O N F O R E S T S O I L S | Countermeasures (fertilisation)


The 137Cs distributions on fertilisedand control plots on the mineral soilsite and on the peat-land site (Figure 1)show significant changes due to thetreatments. The activity fraction invegetation decreased and the activity inthe soil increased on fertilised plots. Onthe mineral soil, fertilised vegetationcontained less than 20 % of the 137Csactivity found in vegetation on controlplots; on peat-land, the correspondingfigure was 60 %. In above-groundparts of vegetation the reduction in137Cs activity in vegetation was slight-ly more than when roots and stumpswere included in vegetation.

In general the results show that radio-caesium uptake in vegetation can bereduced by site-specific nutrient fertili-sation on nutrient-poor sites. The moreefficient reduction of 137Cs content invegetation on fertilised mineral soilcompared to peat-land may reflect theexcess of nutrients. On peat-land, thedose of K-fertiliser was less abundantin relation to the prevailing nutrientstatus of the stand. Dry-mass accumu-lation can be increased by fertilisation,especially in stem-wood. The results onmineral soil site show clearly that thereduction in Cs uptake cannot beexplained by growth response alone,but that is also related to root uptake.

The growth response of trees to fertili-sation is normally observed in the

growing season immediately followingthat of the treatment. As to the timingof treatments in relation to Chernobylfallout, the observed reduction inuptake of 137Cs on the mineral soil sitemay represent nearly the maximumreduction that can be achieved withthe amounts of fertiliser used.However, the effect of fertilisation onthe dry-mass accumulation of vegeta-tion will last only a few years on sub-dry mineral soil sites, and 15-20 yearson peat-land sites poor in mineralnutrients.

In summary, the results indicate thebenefits of fertilisation for restoringcontaminated forests in a severe falloutsituation. The availability of timber tothe forest industry can be increasedconsiderably by long-term treatmentsof forests. Through multiple use offorests, wild berry and mushroomgatherers, as well as hunters, receiveless radiocaesium through foodstuffsfrom fertilised forests than from unfer-tilised ones.

The results of the study motivate fur-ther research on the applicability offorest management methods as coun-termeasures in a fallout situation.Such methods offer possibilities forrespecting the principles of sustai-nable forestry, as the treatments donot radically alter the ecosystem.

figure 1


● F r o m f o r e s t e c o s y s t e m s

● F r o m f o r e s t e c o s y s t e m s s u b m i t t e d

t o a d d i t i o n a l p o l l u t i o n b y h e a v y m e t a l s

RUNOFF4

F R O M F O R E S T E C O S Y S T E M S | Runoff


F r o m f o r e s t e c o s y s t e m s

In northern Sweden, snowmelt andrunoff had achieved their maximumannual intensity at the time of the maincaesium deposition of Chernobyl,which occurred on April 29, 1986. Theamount of 137Cs discharged duringthis period from a 0.5 km catchmentthat was studied was about 600 MBq,corresponding to 5 % of the totaldeposition in the area. In the followingyears only relatively low levels of radio-active caesium were detected in thestream water, and in total about 10 %has been lost over the period 1986-1994.

New and more detailed calculationsshow that the initial loss during snow-melt in 1986 essentially occurred fromthe peat areas and amounted to about

40 % of the total deposition in theseareas. No significant loss from the pineand spruce forests occurred during thesame period. From autumn 1986onwards, the annual loss amounted to30% from the “wetter” part of the peatbog-i.e., the part often contributing tothe surface runoff-while it is about 2%from the “drier” part of the peat area.The loss from areas of unsaturatedmineral soil type is less than 0.03%.The corresponding ecological half-life(Teco, not taking into account radioac-tive decay) for the boreal type ecosys-tem was estimated from measurementson groundwater and exceeded 4000years (Teff = 30 years, including decay),while the Teco for the drier fractions ofbogs was about 34 years (Teff = 16years). On the wet fractions of bogs,

figure 1


Runoff | F R O M F O R E S T E C O S Y S T E M S

■◗ figure 1

The cumulative amount (%)

of 137Cs in stream water

discharged from

a catchment.

LANDSCAPE

■◗ figure 2

The activity concentration (Bq/l)

of 137Cs in stream- and ground-

water (GW) in January 1999 at

the Vindeln site (the error bars

represent 1 σ).

LANDSCAPE

Cum

ulat

ive

Dis

char

ge (

%)

137 C

s (B

q/l)

Time after 29 april 1986 (years)

Aquifer

FROM FOREST ECOSYSTEMS SUBMITTED TO ADDITIONAL POLLUTION BY HEAVY METALS | Runoff


From forest ecosystems submit ted to addi t ional po l lut ion by heavy met als

In addition to soil type, vegetation,geography and climate, a chemicalindustrial pollution of the catchmentarea may also affect the runoff pro-cess. This question has been addres-sed through investigations on twocatchment areas located at 7 km(heavily pollution load) and 21 km(moderately pollution load) from theMonchegorsk Cu-Ni smelter. Thereference catchment area was selectedin northern Finland, about 137 kmfrom the smelter.

Of the total 90Sr deposition, onlyabout one half is still present in theorganic horizons and the upper 15 cm

which often contribute to saturated sur-face discharge, Teco, was about two tothree years.

The measurements on surface andgroundwater show that a small fractionof the 137Cs deposition has reached thegroundwater at a depth of about 1 m.Assuming that the total annual dischar-ge of 330 mm exceeds this depth, witha constant activity of 0.0018 Bq.l-1(Figure 2), the corresponding dischargevia groundwater would be 6 x 10-5%

per year, which can be neglected com-pared to the discharge from the bog,which on average has been 1-2% peryear over a ten-year period.

In conclusion, more than a decade afterthe Chernobyl deposition, the amountof 137Cs leaving the forest ecosystemvia runoff is negligible except from thewetter fraction of bogs. The caesiumhas now reached the groundwater at adepth of 1 m, but the discharge fromthe system remains negligible.

figure 2


Runoff | FROM FOREST ECOSYSTEMS SUBMITTED TO ADDITIONAL POLLUTION BY HEAVY METALS

of the underlying mineral soil, both atthe reference and moderately-pollu-ted catchments.Conversely, at the most pollutedcatchment, close to Monchegorsk,only 15 % of the estimated totaldeposition of 90Sr remains in soil atthese depths. Yet, the fraction in theorganic horizon of the total in thesampled soil is very similar at all threesites, indicating that loss from theorganic fraction probably governs theresidence time and distribution bet-ween organic and mineral fractions ofthe sampled soil.

The activity concentrations of 137Csin stream water correlate well with arelationship based on the fraction ofthe upstream catchment area contai-ning deep organic soil found to bevalid at several Scandinavian sitesfree of notable impact from regionalsources of chemical pollution. Thus,no significant influence of chemicalpollution on the loss of 137Cs byrunoff can be seen. In contrast, theactivity concentration of 90Sr in thestream water at the most pollutedcatchment is about twice as high asexpected from the 90Sr/137Cs activityratios at the two other catchmentareas (reference and moderately-pol-luted). This coincides, moreover, with

a loss of about 85 % of the initialdeposit from the sampled soilcolumns at the most polluted catch-ment (i.e., a much greater loss thanfrom the others).

Severe wind erosion of the organicsoil is obvious at the most pollutedcatchment, giving rise to an airbornetransfer to the stream water.However, the loss of 137Cs in thiscatchment does not agree, apart fromthat likely to be related to a verysmall peat area. This indicates thatthe contributions of 137Cs and 90Sr tothe transfer of organic matter fromthe atmosphere to the stream waterare not the major cause of the notablychanged 90Sr/137Cs ratio in water. Amore probable explanation for therelatively high loss of 90Sr from themost polluted catchment is thereforetransfer by shallow groundwater inthe steep slopes of the stream.

The increased rate of 90Sr transfer tostream water in the heavily pollutedcatchment emphasises the importanceof considering concomitant input todownstream aquatic recipients inlandscapes affected by chemical pol-lution, and the potentially increasedcontamination in freshwater foodchains leading to man.


● O n a g r i c u l t u r a l s o i l s

● I n f o r e s t e c o s y s t e m s - r a d i o c a e s i u m i n m o o s e

SPATIAL VARIATIONS5


■◗ figure 1

Gamma emission mapping

(103 counts/s) over a

lysimeter soil surface two

years after contamination.

PEACE

Spatial variations | O N A G R I C U LT U R A L S O I L S

Lysi

met

er w

est

side

(cm

)

Lysimeter south side (cm)


O n a g r i c u l t u r a l s o i l s

After the Chernobyl accidental releasein the atmosphere, large spatial varia-tions in radionuclide deposition wereobserved due to rain events over certainareas, that washed the radioactivecloud to a varying extent. On a smallerscale, large spatial variations resultedfrom the deposition of hot particles(essentially in the vicinity of the plant),and also, in the longer term, from hori-zontal redistribution by runoff. Itshould be noted that large spatial varia-tions are still present on a small scale.For example, records show the metricto decimetric surface deposition of cali-brated aerosols (3 µm on average)simulated under controlled conditions(no wind, no rain) to be heterogeneous.The subsequent influence of rain in

washing off the radionuclides intercep-ted by the canopies further contributesto this heterogeneity, as illustrated bygamma emission mapping of the lysi-meter soils’ surface.Large spatial variations in radionuclidemigration have also been observed inthe soil profiles, which may evenexceed that caused by soil type or cli-matic intermittence. The soil structure(particle aggregation resulting from thesoil’s biological activity), already citedto explain the patterns of migrationobserved, most likely contributes to thisvariability. Soil microfauna (earth-worms, in particular) and microflora(decomposition and turnover of orga-nic matter) are key elements governingthe soil structure.

figure 1

O N A G R I C U LT U R A L S O I L | Spatial variations


Spatial variations | I N F O R E S T E C O S Y S T E M S — R A D I O C A E S I U M I N M O O S E

I n f o r e s t e c o s y s t e m s - r a d i o c a e s i u m i n m o o s e

The semi-natural ecosystems of nor-thern Europe have abundant largewild herbivores that are a significantfood source for man. After theChernobyl accident, large variations inradiocaesium levels were found in thetissues of free-ranging herbivores cul-led from the affected areas. Althoughin herbivores there is relatively littlescope for variation in the degree ofabsorption and metabolism of pollu-tant radionuclides, the ability to pre-dict the intake of radionuclides by her-bivores living in heterogeneous envi-ronments has been poor. The radionu-clide intake by herbivores was expec-ted to depend critically on diet compo-sition, in terms of plant parts and plantspecies, and the overall levels of envi-ronmental contamination, as it variesover a range of spatial scales. Theneed to quantify diet composition isespecially acute where herbivoresingest fungal fruiting bodies, whichcan contain very high concentrationsof radiocaesium and other pollutants.These are most abundant and may beingested in late summer and autumn,prior to the culling seasons for largeruminants, whose radionuclide conta-mination levels may hence be stronglyaffected. In this project, methods weredeveloped and applied (i) to measurethe intake and diet composition ofmoose in the Boreal forest, and (ii) toassess the use of space by moose, andto consider these in relation to likely

patterns of deposition and distributionof radionuclides in vegetation. Thisfacilitates examination of the likelycontributions of these factors to varia-bility in radionuclide intake and conta-mination levels in moose, against thewider background of variability due todeposition pattern.

A method had been previously deve-loped to quantify diet compositionand intake by domestic ruminantsgrazing on temperate grassland. Thisnon-invasive technique relies on com-paring faecal concentrations of natu-ral plant hydrocarbons with the simi-lar compounds-usually long-chain n-alkanes which are orally dosed atknown rates. The method had notheretofore been applied to wild rumi-nants because of the requirement toadminister a daily oral dose, nor hadit been applied to ruminants foragingon diets of woody browse or mixturesof woody and non-woody plants,typical of wild boreal herbivores. Theavailability of intra-ruminal, control-led release devices (CRD’s), whichrelease the n-alkane marker over aperiod of several weeks, provided theopportunity to apply the method towild herbivores. Experiments wereconducted to test and validate themethod using CRD’s under conditionsrelevant to foraging by boreal forestruminants before it was applied towild moose.

I N F O R E S T E C O S Y S T E M S — R A D I O C A E S I U M I N M O O S E | Spatial variations


Previous results had sug-gested that the behaviourof the dosed hydrocarbonsin the digestive tract is simi-lar for all species of rumi-nant, and is similar to thatof natural plant waxconstituents. The expecta-tion that the method wouldapply equally well to bothmoose (Alces alces) anddomestic ruminants wasborne out in the experi-ment, but further work isneeded on the quantifica-tion of the exact faecalrecoveries of all the mar-kers used, in order to obtain accuratemeasurements of food intake.Validation experiments conductedwith red deer (Cervus elaphus) andcaptive moose suggested that themethodology is equally applicable toruminants ingesting diets of woodybrowse plants, in both summer andwinter seasons. There was good quan-titative agreement between the resultsof the hydrocarbon marker methodand measurements of diet composi-tion by weighing, in the case ofmoose. There was also good agree-ment between the hydrocarbon mar-ker method of intake estimation withdetailed calibration of mass intake viabrowsing damage, in the case of reddeer.

The experiments further showed thatn-alkanes are suitable dietary markersfor birch in leaf, and, although theconcentrations in twig material arelow, they are sufficient to be used asmarkers in winter, when leaves areabsent. Scots pine did not containsignificant amounts of n-alkanes.Hence the use of long-chain fatty alco-hols is suggested when this is a likelydietary constituent. In a separate expe-riment, using housed goats, the com-mon fungal sterol, ergosterol, was tes-ted as a potential faecal marker for die-tary fungi. Because of the negligiblerecovery of ergosterol itself in thefaeces of goats, it is unlikely that it canbe used as a suitable marker for dieta-ry fungi in ruminants.


Spatial variations | I N F O R E S T E C O S Y S T E M S — R A D I O C A E S I U M I N M O O S E

■◗ figure 1

Trajectories of two female

moose in mid-March 1997.

LANDSCAPE

■◗ figure 2

Home ranges for five moose

for the period late April

to late June 1998.

LANDSCAPE

I N F O R E S T E C O S Y S T E M S — R A D I O C A E S I U M I N M O O S E | Spatial variations


The plant hydrocarbon markermethod was applied to wild moose innorthern Sweden, with the objective ofestimating the intake and diet compo-sition of moose in two seasons, winterand summer. The moose were locatedfrom helicopter and shot with ananaesthetic dart. The immobilization,the CRD insertion and the attachmentof GPS (Geographical PositionSystem) collars were carried out in1997 and 1998. Inall, CRD’s wereinserted into eight animals and 20 ani-mals were equipped with GPS-collars.Location of the alkane-dosed moosewas carried out using the radio teleme-try facility of the GPS collars in thewinter of 1997, whereas in winter andsummer 1998, the GPS system wascalled on to locate the dosed moose forthe collection of faeces.

Samples of the vegetation available tothe moose were collected in all threeseasons in order to provide informa-tion on the range of hydrocarbons andlong-chain fatty alcohols likely to beingested by moose.The conclusion on the basis of theresults for the intensively studiedmoose in northern Sweden is that thewinter diet of both moose bulls and

cows was dominated by Scots pine. Incontrast the summer diet was domina-ted by broad-leaved tree species, pri-marily birch (Betula spp.). Somesamples showed evidence that heather(Calluna vulgaris) was present in thediet, but the levels were such that itwas unlikely to have significantradioecological consequences at thepopulation level.

Overall, however, it is unlikely thatvariation in diet composition of moosein late summer and autumn, prior toand during the hunting season, is amajor determinant of variation inradionuclide uptake. The exception tothis remains the ingestion of fungalfruiting bodies, which can lead to signi-ficant uptake of radionuclides, but theingestion of which is unpredictable dueto their ephemeral temporal and spatialdistribution.

It seems likely that the greatest variabi-lity in radionuclide uptake by moose isbrought about by spatial variation indeposition, in conjunction with normalmovements of the animals, rather thanbeing due to variations in diet selectionor intake per se. figure 2figure 1


● F o r e s t e c o s y s t e m s

● F o r e s t e c o s y s t e m s s u b m i t t e d

t o a d d i t i o n a l p o l l u t i o n b y h e a v y m e t a l s

● A g r i c u l t u r a l l a n d s

DOSE ASSESSMENT IMPLICATIONS6

F O R E S T E C O S Y S T E M S | Dose assessment implications


F o r e s t e c o s y s t e m s

Semi-natural ecosystems can makesubstantial contributions to the radia-tion exposure of man. This fact becameknown among radioecologists as aresult of research performed following

the atmospheric nuclear weapons testsin the 1950’s and 1960’s. This know-ledge was broadened after theChernobyl accident, and today it is alsowell known to the general public livingin contaminated regions. In particular,it has been shown that forests can deli-ver large radiation doses through theconsumption of berries, mushroomsand game, but also through the indus-trial use of forest products. TheLANDSCAPE project focussed oncreating a basis for reliable assessmentsof the radiation exposure of humans.Knowledge about the distribution andconcentrations of radionuclides(137Cs) in the forest ecosystem andchanges taking place over time is neces-sary for estimating the doses. However,in addition, assumptions also have to

be made concerning consumptionhabits and time spent in the contami-nated environment in order to calcula-te the internal and external doses. The results of the project give 137Csevolution over time in important com-partments of the forest, as well as thechange in the radiation field expressedas photon fluence, and necessarymigration models have been tested.

One particular exposure case wasinvestigated: the radiological conse-quences following energy productionusing radioactively contaminated bio-fuels from forests in Sweden. The doseestimates used the available dataconcerning the transfer of 137Cs fromsoil to forest flora and fauna. Severalexposure pathways were considered,including the enrichment of 137Cs inthe waste products and their recyclingback to the forest as fertilizer. With anactivity concentration of 5 kBq.kg-1137Cs in the ashes, which is the recom-mended limit for ash recycling inSweden, the highest doses among thepublic, about 0.02 mSv.y-1, are receivedby a critical group composed of huntersand gatherers in forests. The averageSwedish population in the same forestconditions receives a tenth of that dose.Considering the occupational doses, thehighest doses-on the order of 0.1 to 0.5mSv.y-1-are received by persons wor-king on an ash deposit.


Dose assessment implications | FOREST ECOSYSTEMS SUBMITTED TO ADDITIONAL POLLUTION BY HEAVY METALS

Forest ecosystems submit ted to addi t ional po l lut ion by heavy met als

The extensive data generated on 137Csredistribution in the podzolic soils ofboreal forest ecosystems have beenexploited to assess the potentialinfluence of heavy-metal pollution onexternal radiation doses. The externalgamma radiation due to 137Cs in thesoil depends on the vertical distributionof this radionuclide as well as on thewet-bulk density of the various soillayers. The gamma dose rates were cal-culated at all plots from each site alongthe pollution gradient.

The contributions of 137Cs fromnuclear bomb and Chernobyl falloutsto the external dose rate are different.When normalising the data on a depo-sition of 1 kBq.m-2, the 137Cs-genera-ted external dose rate from Chernobylis about 1.6 times higher than thatfrom the global fallout. In these podzo-lic soils, it is true, the Chernobyl-deri-ved 137Cs remains closer to the soilsurface than that from the global fal-lout. However, since total deposition ofChernobyl 137Cs is only about 10 % ofthe total inventory in the soil, the totalexternal dose rate from 137Cs in soils ismainly due to the global fallout.Although the pollution load was clear-ly affecting the vertical 137Cs distribu-tion in the soils, the total external doserate was not affected, with an estima-ted amount of about 1.5 nGy.h-1. Thismight be due to the very low wet bulk

densities of the litter and the organichorizons of these podzolic soils (0.1g.cm-3). As a consequence, the attenua-tion of the gamma rays is quite small,and different depth profiles of 137Cs inthe soil produce very small changes inthe dose rates at a height of 1 m abovethe ground.

It must be stressed, however, thatconsidering the influence of heavymetals on the radionuclide uptake offour understorey plant species, a pio-neer aspect examined in EPORA, aninfluence of the pollution load onexternal and internal dose rates fromthe vegetation is not at all precluded.This is a question which particularlydeserves further attention.

A G R I C U LT U R A L L A N D S | Dose assessment implications


A g r i c u l t u r a l l a n d s

Agricultural lands, which supporthuman food production, are centralfor assessing the potential radiologi-cal impact to humans of radioactivecontaminants. The improved unders-tanding of their behaviour in the soil-plant system, as derived from thePEACE project achievements, is suchas to contribute to refining radiationprotection issues, with a particularcontribution made to improving pre-diction performances. As an example,in-depth soil migration profiles ofboth 137Cs and 90Sr appear to bemost strongly influenced by the initialclimatic conditions prevailing uponcontamination-such as soil humidityand the first rain events. 90Sr does notpenetrate in depth as fast and deeply

as existing prediction models wouldindicate; it therefore remains in theroot zone of most crops, with poten-tial impact on their further contami-nation. Indeed, 90Sr is shown to betransferred to food crops in higheramounts than 137Cs, irrespective ofthe plants considered or of soil type.Such features illustrate how the betterunderstanding of radionuclide move-ments thus procured is conducive tothe improvement of assessmentmodels, whilst providing useful infor-mation for optimising the currentphilosophy on post-accidental inter-vention.

■ Reaching a better understanding of the concept of “bioavailability” and its applicabilityin different ecosystems.

■ Clarifying the importance of climatic conditions immediately after radioactive deposition.

■ Identifying what is needed for the assessment of radiation doses and where basicresearch can afford support.

■ Investigating the scientific pertinence of dose to ecosystems.

■ Distinguishing dose distribution over time from different ecosystems.

■ Unravelling the influence of chemical industrial pollutants on the behaviour of radionuclides in the environment.

■ Further model development and testing, including other lines of enquiry related towaste repositories and other radionuclides, for example.

■ Practical countermeasures in different time perspectives and different terrestrial ecosystems.

■ Clarifying current knowledge and identifying research needs through a systematicexpert evaluation of all experimental data.

■ Collection of strategic data series for model validation.

What is still needed?


S H O R T R E F E R E N C E L I S T


Bréchignac F. et al. 1999 PEACE Final Report. Programme for Evaluating the impact ofAccidents Contaminating the Environment – Soil-radionuclides processes of interaction andmodelling of their impact on contamination of plant food products. CEC-IPSN Associationcontract n° F14P-CT96-0039a within the 4th EURATOM Framework Programme on NuclearFission Safety. IPSN Report Series DPRE/SERLAB/99-017(P), Paris, France.

Moberg L. et al. 1999 LANDSCAPE Final Report. An integrated approach to radionuclideflow in semi-natural ecosystems underlying exposure pathways to man. CEC-IPSN Associationcontract n° F14P-CT96-0039b within the 4th EURATOM Framework Programme on NuclearFission Safety. SSI Report 99:19, Stockholm, Sweden. ISSN 0282-4434.

Suomela M. et al. 1999 EPORA Final Report. Effect of industrial pollution on the dynamics ofradionuclides in boreal understorey ecosystems. CEC-IPSN Association contract n° F14P-CT96-0039c within the 4th EURATOM Framework Programme on Nuclear Fission Safety.STUK Report Series A168, Helsinki, Finland. ISSN 0781-1705


A B S T R A C T

T his document provides an overview on the recent advances in Europe concerningthe long-term environmental behaviour of radionuclides. It describes and illus-

trates the most significant scientific achievements gathered through three Europeanprojects, PEACE, LANDSCAPE and EPORA, which have been co-funded by theEuropean Commission (DGXII) under the Nuclear Fission Safety Programme.

The improvement of radiation protection issues requires an accurate understanding ofthe radionuclide contaminants behaviour within ecosystems. The three projects havetherefore been focused on the soil-plant system of agricultural lands and semi-naturalforest ecosystems based on both, experimental approaches in controlled conditions andin the field. A particular emphasis is devoted to unravelling intra-compartments loadsand inter-compartments movements of 137Cs, 90Sr and 239+240Pu, addressing also forthe first time the potential influence of non-radioactive additional pollutants (heavymetals). The variety of conditions encountered in Europe is emcompassed through awide selection of experimental sites ranging from Mediterranean up to Boreal ecosys-tems. Experimental data acquisition and mechanistic modelling are synergisticallydeveloped in order to promote access to improved assessment predictions in an acci-dental situation.

The knowledge generated first includes the redistribution of radionuclides in the soil-plant system with emphasis on vertical migration in soils and root uptake by plants. Itprovides next recent advances into the mechanistic modelling of these features. Fromthe new data gathered through three years of investigations, essential conclusions oncontamination discharge via run-off water are drawn, and insights discussed with res-pect to counter-measure, and spatial variations. Preliminary conclusions are finallyderived in the perspective of dose assessment, and areas deserving further researchidentified.


RADIATION AND NUCLEAR TUA RAHOLA

SAFETY AUTHORITY, HELSINKI, FINLAND : KRISTINA RISSANEN

MARKETTA PUHAKAINEN

TARJA HEIKKINEN

COORDINATOR : MATTI SUOMELA

NATIONAL DEFENCE RESEARCH RONNY BERGMAN

ESTABLISHMENT, UMEÅ, SWEDEN : THORBJÖRN NYLÉN

NATIONAL RESEARCH CENTER KURT BUNZL

FOR ENVIRONMENTAL HEALTH, WOLFGANG SCHIMMACK

NEUHERBERG, GERMANY :

UNIVERSITY OF HELSINKI, TIMO JAAKKOLA

FINLAND : LISA RIEKKINEN

MICHAEL TILLANDER

NORWEGIAN UNIVERSITY OF SCIENCE EILIV STEINNES

AND TECHNOLOGY,TRONDHEIM, NORWAY : TORIL HAUGLAND

HÅVARD THØRRING

EPORA


SWEDISH RADIATION RODOLFO AVILA

PROTECTION INSTITUTE, LYNN HUBBARD

STOCKHOLM, SWEDEN : LENA WALLBERG

COORDINATOR : LEIF MOBERG

UNIVERSITY OF TRIESTE, ITALY : ENRICO FEOLI

MAURO SCIMONE

CHRISTINA MILESI

MACAULAY LAND USE RESEARCH BOB MAYES

INSTITUTE, ABERDEEN, U.K. : GLENN IASON

RADIATION AND NUCLEAR SAFETY AINO RANTAVAARA

AUTHORITY, HELSINKI, FINLAND : VIRVE VETIKKO

NATIONAL DEFENCE RESEARCH RONNY BERGMAN

ESTABLISHMENT, UMEÅ, SWEDEN : TORBJÖRN NYLÈN

SWEDISH UNIVERSITY OF AGRICULTURAL THOMAS PALO

SCIENCES, UMEÅ, SWEDEN : NEIL WHITE

THE FINNISH FOREST RESEARCH HANNU RAITIO

INSTITUTE, PARKANO, FINLAND : LASSE ARO

SEPPO KAUNISTO

FACULTÉ UNIVERSITAIRE DES SCIENCES OLIVIER GUILLITTE

AGRONOMIQUES, GEMBLOUX, BELGIUM :

LANDSCAPE


INSTITUTE FOR NUCLEAR PROTECTION CHANTAL MADOZ-ESCANDE

AND SAFETY, DEPARTEMENT FOR CLAUDE COLLE

ENVIRONMENTAL PROTECTION, MARC-ANDRÉ GONZE

SERLAB, CADARACHE, FRANCE : THOMAS PERRIER

CHRISTIAN TAMPONNET

VIRGINIE CAMILLERI

HÉLÈNE JUHAN

ERIC DUBOIS

COORDINATOR : FRANÇOIS BRECHIGNAC

SCK-CENTRE D’ÉTUDES NUCLÉAIRES NADIA WAEGENEERS

DE MOL, BELGIUM : YVES THIRY

CHRISTIAN VANDECASTEELE

UNIVERSITY OF BARCELONA, SPAIN : RAMON VALLEJO

JAUME CASADESUS

CINTA ROCA

TERESA SAURAS

SWEDISH UNIVERSITY OF AGRICULTURAL K.J. JOHANSON

SCIENCES, UPPSALA, SWEDEN : KLAS ROSEN

SVERKER FORSBERG

IMPERIAL COLLEGE OF SCIENCE GEORGE SHAW

TECHNOLOGY AND MEDICINE, JOANNA MARCHANT

LONDON, U.K.: ADRIAN BUTLER

JOHN TOMPKINS

CEC-DGXII, BRUSSELS, BELGUIM ERNST-HERMANN SCHULTE

(DETACHED TO CADARACHE) :

PEACE

The Institut de protection et de sûreté nucléaire (French institute forprotection and nuclear safety) is assigned the task of conducting researchand expert appraisals in all disciplines implicated in the control of nuclearrisks and of their impact on man and the environment. Its field of concernprimarily addresses the safety of installations, the study and prevention ofaccidents, the protection of human health, transport safety, the inspectionof nuclear materials, and organization and training in crisis management.

Under the joint authority of the Ministries of Industry and of theEnvironment, IPSN has, since 1990, been granted a specific status as wellas its own budget. IPSN’s staff numbers about 1300, two-thirds of whomare engineers and research personnel specialized in a variety of branches(physics, chemistry, biology, medicine, agronomy, etc.). Within the frame-work of an ongoing reflection concerning the French nuclear inspectionsystem, IPSN has the vocation to become a fully independent public esta-blishment in its own right.

Informat ion center on nuc lear safety

77-83, avenue du Général -de-Gaul le

92140 Clamar t , France

Mai l ing address :

BP 6 – 92265 Fontenay-aux-Roses Cedex

Tel . : (33) 01 46 54 80 07

Fax: (33) 01 46 54 79 49

© É

ditio

n 20

00 /

Miss

ion

com

mun

icat

ion

IPSN

- D

épôt

léga

l Avr

il 20

00

Date post:	15-Sep-2018
Category:	Documents
Upload:	dinhnga
View:	218 times
Download:	0 times

Long-term environmental behaviour of radionucleides€¦ · LONG-TERM ENVIRONMENTAL BEHAVIOUR OF...

Documents