lable at ScienceDirect

Journal of Environmental Radioactivity 138 (2014) 60e67

Contents lists avai

Journal of Environmental Radioactivity

journal homepage: www.elsevier .com/locate / jenvrad

The assumption of heterogeneous or homogeneous radioactivecontamination in soil/sediment: does it matter in terms of theexternal exposure of fauna?

K. Beaugelin-Seiller*

Institut de Radioprotection et de Sûret�e Nucl�eaire (IRSN), PRP-ENV, SERIS, LM2E, Cadarache, France

a r t i c l e i n f o

Article history:Received 31 January 2014Received in revised form24 July 2014Accepted 26 July 2014Available online

Keywords:EnvironmentDosimetryExternalContaminationHeterogeneity

* Centre of Cadarache bdg 159, BP3, 13115 SAINTTel.: þ33 442199416; fax: þ33 42199143.

E-mail address: karine.beaugelin@irsn.fr.

http://dx.doi.org/10.1016/j.jenvrad.2014.07.0270265-931X/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

The classical approach to environmental radioprotection is based on the assumption of homogeneouslycontaminated media. However, in soils and sediments there may be a significant variation of radioac-tivity with depth. The effect of this heterogeneity was investigated by examining the external exposure ofvarious sediment and soil organisms, and determining the resulting dose rates, assuming a realisticcombination of locations and radionuclides. The results were dependent on the exposure situation, i.e.,the organism, its location, and the quality and quantity of radionuclides. The dose rates ranged over threeorders of magnitude. The assumption of homogeneous contamination was not consistently conservative(if associated with a level of radioactivity averaged over the full thickness of soil or sediment that wassampled). Dose assessment for screening purposes requires consideration of the highest activity con-centration measured in a soil/sediment that is considered to be homogeneously contaminated. A morerefined assessment (e.g., higher tier of a graded approach) should take into consideration a more realisticcontamination profile, and apply different dosimetric approaches.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Over the last decade, environmental radioprotection has been atopic of increasing interest in the field of radioecology. This has ledto methodological developments that seek to determine the po-tential for negative effects in organisms exposed to radioactivity(Beresford et al., 2008, 2009, 2010a; Vives I Battle et al., 2007, 2011;Yankovich et al., 2010). Radiological risk assessment for fauna andflora is a process that should be as far as possible consistent withthe existing methods used both for chemical risk assessment andhuman radioprotection. Based generally on a more or less explicitlytiered approach, most of the models commonly used to assessradiological risk to wildlife aim to be conservative, at least at thescreening level (US DOE, 2002; Copplestone et al., 2002; Beresfordet al., 2007, 2010b).

Although radiological concepts already exist, resulting frommore than half a century of research and development regardingdoses to humans, it is necessary to adapt them to new fields of

PAUL LES DURANCE, France.

investigation. With regard to dosimetric calculations, this meansintegrating a large variety of types of organism, habitats, ways oflife, and a multitude of exposure scenarios. Because a realisticdescription of the true natural world is not possible, numeroussimplifications have been applied to perform these calculations(Vives I Battle et al., 2007). One of these simplifications is toconsider the exposure medium as homogeneous, both in itscomposition and its contamination. This assumption, although it iswidely adopted, does not necessarily reflect the actual situation,especially when considering soil or sediment contamination. Thesecompartments usually present contamination profiles that varywith depth, due to the combination of various processes (e.g., thedeposition of suspended matter, radionuclide migration). This hascaused assessors to consider the effect of the assumption of a ho-mogeneous medium on dose rates to exposed organisms. Thecommonly used tools assume that organisms are either on, or in,homogeneously contaminated soil or sediment (US DOE, 2002;Copplestone et al., 2002; Beresford et al., 2007). Investigations ofthis issue were initiated during the EMRAS II (EnvironmentalModelling for Radiation Safety) programme (IAEA, in press), usingthe EDEN (Elementary Dose evaluation for Naturel Environment)dosimetric tool (Beaugelin-Seiller et al., 2006), which allows for theconsideration of alternative approaches rather than relying on the

Table 2The mass and dimensions of selected organisms.

Case study Organism Mass (kg) X (cm) Y (cm) Z (cm)

Canadian lakescenario

Insect larva 1.8E-05 1.5Eþ00 1.5E-01 1.5E-01Benthic fish 1.5Eþ00 5.0Eþ01 8.0Eþ00 7.0Eþ00

Soil scenario Bee 5.9E-04 2.0Eþ00 7.5E-01 7.5E-01Rat 3.1E-01 2.0Eþ01 5.0Eþ00 6.0Eþ00Earthworm 2.6E-02 1.0Eþ01 1.0Eþ00 1.0Eþ00

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e67 61

“classic” approach to media contamination and biota occupancyscenarios (Vives I Battle et al., 2007, 2011).

2. Materials and methods

Most of the dosimetric approaches used in environmentalradioprotection (Copplestone et al., 2002; US-DOE, 2002; Beresfordet al., 2007; ICRP, 2008) apply a kind of dose coefficient (designedhereafter as the Dose Conversion Coefficient or DCC) to convert anactivity concentration into a dose rate (e.g., tomove from Bq kg�1 orBq L�1 to Gy/unit of time), by considering a homogeneouscontamination of the exposure source. These DCCs are generallytabulated as default values, which at best allows for some extrap-olation, for example based on size ratios (US-DOE, 2002; Beresfordet al., 2007). Soils or sediments are described as a one volumesource, characterised by a single contamination value, to which thecorresponding DCC is applied to determine the dose rate absorbedby the target organism. A few tools, such as EDEN, allow thecalculation of a specific DCC for each case study. Version 3 of EDENis now available (IAEA, in press), and can be used to calculate DCCs,or doses, to any organism, from any radionuclide and any exposurescenario, by runningMonte-Carlo simulations. All the required dataare user-defined with the exception of nuclear data, which aretaken from the JEFF (Joint Evaluated Fission and Fusion File) data-base (OCDE-NEA, 1997).

The effect of heterogeneous vs. homogeneous contamination ofsoil/sediment was investigated in this study by increasing thecomplexity of the description of these compartments, in two casestudies. The first was run in the framework of the IAEA EMRAS IIprogramme (IAEA, in press), according to the Canadian U mines &mills scenario. Measurements from the environmental monitoringprograms of operational mines and mills, as well as decom-missioned sites were gathered. Sediment profiles from BeaverlodgeLake, a remote lake in northern Saskatchewan, located east ofUranium City, displayed a large heterogeneity, i.e., from 0 to amaximum of 20 cm (at approximately 2 cm intervals) for 226Ra,210Pb, 210Po, and thorium and uranium.Where specific isotopic datawere missing, parts of the decay chains were considered to be atequilibrium (238U/234Th/234mPa; 226Ra/230Th; 210Po/210Pb) and/ortheir isotopic ratios with 238U were preserved (IAEA, in press).Profiles from two contrasting sites were used for this exercise(Table 1). Three exposure scenarios were adopted, considering twoaquatic organisms of various dimensions, body shapes and location(Table 2, Fig. 1).

It may be important to consider the radioactive decay productsin the radiological risk assessment. The Canadian lake scenario

Table 1Layered (from IAEA, in press) and aggregated radionuclide concentrations in sediments

Beaverlodge Ace Bay (BAB) Dub

238U 234U 226Ra 210Pb 238U

0e2 cm 6484 6452 11,500 15,700 4702e4 cm 17,414 17,329 30,886 42,166 634e6 cm 13,770 13,703 24,424 33,344 186e8 cm 12,103 12,044 21,467 29,3078e10 cm 2631 2618 4666 637010e20 cma 1662 1653 2947 402320e2 cm 692 688 1227 1675Simplified descriptionSurfaceb 6484 6452 11,500 15,700 470layer 1c 11,712 11,655 20,773 28,360 470layer 2d 1349 1342 2392 3266 34

Italics: use of the ratio of each radionuclide to 238U in the sediment layer for which dataa Missing values were estimated from the mean of adjacent layers.b 0e2 cm.c 2e8 cm (BAB) 0e2 cm (DLD).d 8e22 cm (BAB), 2e6 cm (DLD).

offered the opportunity to explore this issue for the isotope 234Th,which is in secular equilibrium with its daughter 234mPa, in com-bination with the effect of the heterogeneous contamination.

To complete the initial study, a second case study was laterundertaken using the soil contamination profiles published bySrnick et al. (2008), in relation to plutonium isotopes, 241Am, 137Csand 90Sr. Soil samples were collected on an alpine pasture of Austriain summer 1999. The global fallout was identified as the source oftransuranic contaminationwhen caesium and strontium stem fromthe Chernobyl accident. Our study did not aim to produce realisticdose assessments: data were used as presented by the authors,without correcting them for radioactive decay to the same date (seethe original publication for details). Characteristic isotopic ratios forthe years 1986 and 2006 were applied in the same dosimetricsimulation. Two contrasting profiles were selected (Table 3),considering two exposure scenarios (Fig. 2) for three terrestrialorganisms with different characteristics (Table 2).

Finally DCCs were calculated using EDEN, without applying anyradiation weighting factor, for 12 scenarios per site for sedimentand eight scenarios per soil location. A correction factor of 2.6 wasapplied to convert the dry weight of sediment into wet weight,considering a total bulk density of 1300 kg m�3 and a volumefraction of water of 0.8 (EC, 2003). These DCCs were applied tomeasurements at a given depth, or to an activity averaged overseveral layers, in accordance with the description ofcontamination.

3. Results and discussion

3.1. Dose rates to aquatic organisms

Total external doses rates were calculated for fish and insectlarva with and without consideration of the 234Th daughter prod-uct, 234mPa (Fig. 3). The corresponding values have no significancein themselves; only their relative comparison was considered inthis study. The pattern was similar between the four scenarios

(Bq kg�1 d.w.).

yna Lake Deep (DLD)

234U 230Th 226Ra 210Po 210Pb

,041 467,757 1740 5480 29,800 30,600,010 62,704 1090 1840 11,340 12,680,550 18,460 300 930 3600 5440

,041 467,757 1740 5480 29,800 30,600,041 467,757 1740 5480 29,800 30,600,188 34,022 572 1308 6389 8305

was available.

Fig. 1. Description of the sediment compartment (A: simplest approach; B: intermediate approach; C: realistic approach), and the location of organisms for the two Canadian sites.

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e6762

studied, regardless of which lake and progeny were considered.When an organism (fish or insect larva) was present on the sedi-ment, the maximum absorbed external dose rate (up to three or-ders of magnitude higher) was always obtained with anintermediate level of complexity (scenario B, Fig. 1). The homoge-neous (scenario A) and most realistic (scenario C) approachesgenerated similar results, with each producing the highest valuesdepending on the site. Once located in the subsurface layer (from 1to 4 cm under the surface), the insect larva received a similar doserate (within a factor of about two), regardless of the description ofthe sediment. Where organisms burrowed deeper into the sedi-ment (4e15 cm below the surface) similar results were obtained forthe two complex approaches (B and C), and the values wereapproximately one order of magnitude lower than those obtainedfor the homogeneous sediment (A). Considering these results andthe associated hypotheses, it was not possible to identify a sys-tematically conservative approach for the three ways we depictedthe sediment compartment.

The addition of 234mPa, the decay product of 234Th in secularequilibrium with its parent, increased logically the total externaldose rates, by a factor of up to about 100 (insect larva on thesediment at Dubyna Lake Deep). However, this depended on theexposure scenario, and tended to reduce the discrepancies betweenthe three approaches.

These results indicate that external dose rates for exposed faunaare affected by the way the sediment contamination is described,and indicate a dependence on the sediment contamination (natureand location of radionuclides) and the organism (location).

The effect of the heterogeneity of the sediment contaminationappeared to be significant, but tended to alternatively increase or

Table 3Layered (from Srnick et al., 2008) and aggregated radionuclide concentrations in soils (B

Trench T2

238Pu 239Pu 240Pu 241Am 137Cs 90Sr

0e1.1 cm 4320 1951.1e2.0 cm 0.53 14.7 0.84 4.34 4380 972.0e3.0 cm 0.79 30.4 1.32 8.88 1780 873.0e3.7 cm 0.25 11.9 0.71 3.53 743 583.7e4.5 cm 0.08 2.71 0.19 0.81 394 574.5e5.5 cm 0.05 1.07 0.05 0.31 230 535.5e6.6 cm 0.48 0.03 0.14 1426.6e8.1 cm 0.32 0.02 0.1 113Total 0.19 6.93 0.35 2.04 1465 65

decrease the external dose rates absorbed by the organisms for agiven site, depending on their location. The portion of the totalexternal dose rates due to each radionuclide was determined forthe three possible locations of the organism, by considering onlythe insect larva (Fig. 4). At Beaverlodge Ace Bay the contribution ofradionuclides to the total external dose rate absorbed by the or-ganism only differed when they were located on the sediment.When considering homogeneous contamination, exposure wasmainly due to seven radionuclides, whereas it was five in the het-erogeneous approach, with the main contributions being from226Ra (about 40%) or 210Po (ca 30%). For larva in the sediment, therewas no difference. Six radionuclides contributed significantly to thetotal external dose rate absorbed by the organism, with the maincontribution being from 210Po (about 40%). At Dubyna Lake Deep,five patterns were observed for the six situations of interest. Forheterogeneous contamination, each radionuclide had a differentcontribution for each location of organism. In contrast, when ho-mogeneity was assumed, two patterns of radionuclide contributionwere apparent, one for locations on the sediment surface, and theother for organisms buried at any depth in the sediment. Three tosix radionuclides contributed significantly to the dose rate absor-bed by the insect larva. Themain contributor was 234U in four of thefive cases (from about 50% to 60%). In the fifth case 234Th (ca 55%)was the predominant contributor, with very little contribution from210Po and 226Ra to the total external dose rate (3% and 1%, respec-tively). Conversely, the two isotopes of uranium (234 and 238)made a large contribution (25% and 16% respectively).

The contribution of 234mPa to the external exposure of organ-isms was explored for fish. Accounting for 234Th radioactive decayby considering its daughter did not change the general pattern of

q kg�1 d.w.).

Trench T8

238Pu 239Pu 240Pu 241Am 137Cs 90Sr

0e0.8 cm 0.20 0.77 0.11 0.25 6010 880.8e1.5 cm 0.23 1.01 0.06 0.30 7760 691.5e2.8 cm 0.53 8.15 0.35 2.38 5910 1082.8e3.9 cm 1.04 22.8 0.87 6.64 3630 1873.9e5.2 cm 0.84 21.1 0.98 6.19 2960 2085.2e6.4 cm 2330 1766.4e7.3 cm 0.17 3.62 0.18 1.06 1496 1637.3e8.2 cm 0.09 1.46 0.07 0.43 1131 80Total 0.42 8.42 0.37 2.46 3771 142

Fig. 2. Description of the soil compartment (heterogeneous vs. homogeneous), and thelocation of organisms for the two Austrian sites.

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e67 63

the radionuclides contribution in the heterogeneous scenario(Fig. 4). At both sites, 234mPa contributed to the dose rate absorbedby the fish (6%e11%), but to a much lesser extent than the maincontributors (230Th, 226Ra and 210Po at Beaverlodge Ace Bay; 234Uand 238U at Dubyna Lake Deep, as was also the case for insect larva).

Fig. 3. Total external dose rates (mGy h�1) absorbed by fish and insect larva from Beaverlogeneous (A; black bar), simplified heterogeneous (B; graded bar) or realistic heterogeneousthe right).

Conversely, introducing the progeny had a significant impact whenconsidering homogeneous contamination. At both sites, 234mPawasthe dominant radionuclide, with a contribution that ranged fromabout 50% to 75% of the total external dose rate. In secular equi-libriumwith 234Th, 234mPa should not be neglected when assuminghomogeneous contamination.

3.2. Dose rates to terrestrial organisms

To extend these results, total external dose rates were calculatedfor a bee in the air, a rat on the soil surface and an earthworm on orin the soil, by considering a homogeneous or heterogeneous dis-tribution of 238Pu, 239Pu, 240Pu, 241Am, 137Cs and 90Sr in soil layers atthe two selected sites (T2 and T8; Srnick et al., 2008). No attenua-tion of alpha particles by inert layers (e.g., the fur or skin) wasconsidered. The highest dose rates were obtained for earthwormsin soil, but the overall pattern was similar for all the organisms,their location, and the site (Fig. 5). The dose rates associated withalpha emitters are always orders of magnitude lower than thoseresulting from exposure to 137Cs and 90Sr. The further the organismis from the soil, the more pronounced is this effect, because of thereduced penetrative power of alpha radiation compared to beta andgamma radiation. Consequently, the total external dose rate waspredominantly due to the contribution from 137Cs (at least 93%). Amore detailed analysis revealed some differences between the twoapproaches which suggest that the assumption that a heteroge-neous description of soil contamination leads to the highest doserates it is not necessarily true. Two combinations (plutonium iso-topes for the earthworm on soil, at sites T2 and T8; and 137Cs, 90Srand total dose rates for the T8 site) generated higher dose rateswith a homogeneous description of soil contamination.

The ratio between the external dose rates obtained using the twoapproaches was calculated (Table 4). The highest ratios were ob-tained for alpha emitters, and reached in some cases a value of 50(earthworm exposure to 238Pu and 240Pu in T8 soil). They were

dge Ace Bay (upper graph) and Dubyna Lake Deep (lower graph) considering homo-(C; white bar) contamination of the sediment, without (on the left) or with 234mPa (on

Fig. 4. Analysis of the radionuclide contribution (%) to the total external dose rates at Beaverlodge Ace Bay (left) and Dubyna Lake Deep (right), considering heterogeneous orhomogeneous contamination of the sediment (A: depending on the location of the insect larva, without considering 234mPa, B: with and without consideration of 234mPa, for thefish).

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e6764

generally lower for the T2 station, with amaximal value of about 13.The ratio decreased for beta (90Sr) radiation,with amaximal value ofthree. Finally, for mainly gamma (137Cs) emitters, the results werewithin a factor of about two, regardless of the organism, its location,and the site. The assumption of homogeneous contaminationgenerally leads to an underestimation of the external dose ratesabsorbed by the organisms. The effect depends on the organism, itslocation, the radionuclide, and the site (a ratio of 14 for the beeexposed to 239Pu on T8 vs. a ratio of 2 on T2; a ratio of 1.5 for theearthworm on soil exposed to 137Cs on T2 vs. a ratio of 0.7 on T8).

The results obtained for terrestrial organisms confirm that thedepiction of the medium may impact on the level of external doserates absorbed by organisms exposed to a gradient of radioactivecontamination, depending on the exposure scenario (mainly theorganism location and source term).

3.3. Discussion

This study has considered if the assumption of homogeneouscontamination is an acceptable approach to assess the exposure oforganisms to radionuclides associated with soil/sediment. Withregard to the location of organisms on the surface of a sediment, thesimilarity between the dose rates obtained with a single homoge-neous layer of sediment (case A) and those resulting from a realisticdescription (case C) suggests that the intermediate approach (caseB) gives too much weight to the surface contamination. This couldbe explained by an overestimation of the corresponding DCCs, ormore likely of the surface contamination. This data was estimatedroughly from the volume activity in the first 2 cm of sediment(Bq kg�1 converted into Bq m�2 using a default density of1500 kg m�3 dry weight). For organisms living on the sediment

Fig. 5. External dose rates (Gy d�1) calculated for the two terrestrial sites (left: T2; right: T8) for different organisms and exposure scenarios, considering homogeneous (white bar)and heterogeneous (black bar) contamination of soil.

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e67 65

surface, the homogeneous approach provides the closest results tothose obtained using the realistic approach, which was consideredas the reference. However, for organisms buried in the sediment,dose rates more similar to the reference calculation (within a factorof two) were obtained using the intermediate approach.

Finally, when comparing the homogeneous approach to therealistic one, the total external dose rates absorbed by fish werewithin a factor of two. The effect of an assumption of heteroge-neous contamination increased with the depth that the insect larvais in the sediment. In our case study, the difference was at most oneorder of magnitude, and the assumption of homogeneouscontamination, applied considering the average contamination ofthe sediment layer, corresponds to a conservative approach.

The soil case study confirmed that the effect of a realisticdepiction of the contamination profile is closely linked to the na-ture and location of the radionuclides and the location of theorganism.

In this case, due to the huge predominance of 137Cs in the soilcontamination, the effect in terms of total dose rate was not sen-sitive. The more penetrating the radiation, the less the externaldose rate was influenced by the heterogeneity of the contamina-tion. This is explained by the shielding effect of the soil layers,which is more “efficient” for a single layer of given thickness thanfor a series of finer layers of a total same thickness, as long as theradioactivity is mainly located in the upper layers.

Finally, the ratios observed for alpha emitters (Table 4) have tobe put in perspective of the contribution of the external exposure tothe total dose rate absorbed by the organisms. For illustration, it hasbeen calculated for earthworms in the middle of the homogeneoussoil layer, for a unit concentration of each radionuclide of interest.Concentration ratios for annelids were taken from the WildlifeDatabase (Howard et al., 2013). External contribution to the totaldose rate absorbed by worms exposed to plutonium isotopes in soilwas about 7%, and it decreased to less than 2% for 241Am. The effect

Table 4The ratio between the external dose rates obtained for terrestrial organismsconsidering heterogeneous or homogeneous contamination of the soil for the twotrenches.

Trench Organism 238Pu 239Pu 240Pu 241Am 137Cs 90Sr Total

T2 Bee 2.4 2.2 1.9 1.8 2.3 1.2 2.3Rat 12.9 2.7 11.1 1.1 1.7 2.9 1.7Earthworm In soil 2.2 2.6 2.6 2.4 0.9 1.1 0.9

On soil 0.7 0.5 0.7 1.1 1.5 3.0 1.5T8 Bee 2.4 13.7 4.1 12.1 0.7 1.6 0.7

Rat 22.6 28.8 32.6 9.9 0.6 0.8 0.6Earthworm In soil 4.1 27.2 9.1 22.5 0.5 2.4 0.5

On soil 0.02 0.1 0.02 2.3 0.7 0.4 0.7

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e6766

of heterogeneity is at the end negligible regarding the total expo-sure of the animal.

4. Conclusions

The primary purpose of this study was to account for the actualdistribution of radionuclides in soil/sediment with regard to itseffect on fauna in terms of dosimetry, because most of the avail-able tools and approaches do not account for heterogeneity ofcontamination. Radiological risk assessment is usually conductedby assuming a single homogeneous volume source, but it isquestionable if this is appropriate. By taking advantage of thepotential of the EDEN tool, this was investigated by describing asoil/sediment compartment with an increasing complexity, fromthe usual single uniform compartment through to a multilayerrepresentation. For exposure to sediments, depending on thecompartment depiction, the total external dose rates varied by upto three orders of magnitude. This range was explained by thevariable contribution of each radionuclide that changes with theconfiguration of the exposure scene. The degree to which eachsediment depiction is satisfactory depended on the exposure sit-uation, i.e., the organism, its location as well as the quality andquantity of radionuclides. There were little divergences betweenthe results of the different case studies. A maximum of one orderof magnitude was observed between the total external dose ratesobtained using the homogeneous and realistic approaches. Thesoil case study, introducing an exposure from alpha emitters,revealed the extent of the shielding effect of the soil in combi-nation with the contamination profile, which was particularlysensitive with less penetrative alpha radiation. It confirmed theinfluence of the location of the organism, as well as the nature andlocation of the radionuclides.

The way heterogeneous contamination should be describedwith regard to the dose calculation for exposed organisms dependson the goal of the calculation. For a conservative assessment of totaldose rates, it is sufficient to maintain the usual homogeneous dis-tribution, in combinationwith the maximal activity reported in theprofile, rather than the average value. This assumption is used inmost approaches to environmental radioprotection. It is well suitedto the screening stage of the radiological risk assessment. Ifnecessary, the upper tiers of the assessment should be refined byconsidering the real contamination profile. These profiles shouldalso be considered when looking to understand and interpret doserates in terms of effect. These conclusions, similar to those fromother studies, indicate two ways for the future of environmentaldosimetry. Simple and robust dosimetric approaches are needed tomeet radiological risk assessment requirements. At the same time,more refined and accurate approaches should be made available, toprecisely determine the dose rates when assessing their effects onwildlife. Such approaches may also be useful for the upper tiers ofrisk assessment.

Acknowledgements

This work was partly undertaken within the framework of theIAEA EMRAS II programme. The author would like to thank theCNSC (Canadian Nuclear Safety Commission), and more especiallyits representatives in the EMRAS II BMG, R. Goulet and S. Mihok,who suggested the initial topic, as well as N. Beresford, the groupleader, who encouraged the study. Thanks also go to all members ofthe group, for the interesting and fruitful discussions the pre-sentations of this work generated, as well as to the reviewers, fortheir contribution to the improvement of the paper.

References

Beaugelin-Seiller, K., Jasserand, F., Garnier-Laplace, J., Gariel, J.C., 2006. Modellingthe radiological dose in non-human species: principles, computerization andapplication. Health Phys. 90, 485e493.

Beresford, N., Brown, J., Copplestone, D., Garnier-Laplace, J., Howard, B.J., Larsson, C.-M., Oughton, O., Pr€ohl, G., Zinger, I. (Eds.), 2007. D-ERICA: an IntegratedApproachto the Assessment and Management of Environmental Risks fromIonising Radiation. Description of Purpose, Methodology and Application.Contract Number: FI6R-CT-2003e508847, 82 pp.þ annexes.

Beresford, N.A., Balonov, M., Beaugelin-Seiller, K., Brown, J., Copplestone, D.,Hingston, J.L., Horyna, J., Hosseini, A., Howard, B.J., Kamboj, S., Nedveckaite, T.,Olyslaegers, G., Sazykina, T., Vives i Batlle, J., Yankovich, T.L., Yu, C., 2008. Aninternational comparison of models and approaches for the estimation of theradiological exposure of non-human biota. Appl. Radiat. Isot. 66, 1745e1749.

Beresford, N.A., Barnett, C.L., Beaugelin-Seiller, K., Brown, J.E., Cheng, J.-J.,Copplestone, D., Gaschak, S., Hingston, J.L., Horyna, J., Hosseini, A., Howard, B.J.,Kamboj, S., Kryshev, A., Nedveckaite, T., Olyslaegers, G., Sazykina, T., Smith, J.T.,Telleria, D., Vives i Batlle, J., Yankovich, T.L., Heling, R., Wood, M.D., Yu, C., 2009.Findings and recommendations from an international comparison of modelsand approaches for the estimation of radiological exposure to non-humanbiota. Radioprotection 44, 565e570.

Beresford, N.A., Barnett, C.L., Brown, J.E., Cheng, J.-J., Copplestone, D., Gaschak, S.,Hosseini, A., Howard, B.J., Kamboj, S., Nedveckaite, T., Olyslaegers, G., Smith, J.T.,Vives I Batlle, J., Vives-Lynch, S., Yu, C., 2010a. Predicting the radiation exposureof terrestrial wildlife in the Chernobyl exclusion zone: an international com-parison of approaches. J. Radiol. Prot. 30, 341e373.

Beresford, N.A., Hosseini, A., Brown, J.E., Cailes, C., Beaugelin-Seiller, K., Barnett, C.L.,Copplestone, D., 2010b. Assessment of risk to wildlife from ionising radiation:can initial screening tiers be used with a high level of confidence? J. Radiol. Prot.30, 265e281.

Copplestone, D., Bielby, S., Jones, S.R., Patton, D., Daniel, D., Gize, I., 2002. ImpactAssessment of Ionising Radiation on Wildlife, vol. 128. Environment Agency(Bristol, UK), R&D Publication, p. 222.

EC, 2003. Technical Guidance Document in Support of Commission Directive 93/67/EEC on Risk Assessment for New Notified Substances, Commission Regulation(EC) No 1488/94 on Risk Assessment for Existing Substances, Directive 98/8/ECof the European Parliament and of the Council Concerning the Placing ofBiocidal Products on the Market. Part II. European Commission, Joint ResearchCenter. Office for Official Publications of the European Communities,Luxembourg.

Howard, B.J., Beresford, N.A., Copplestone, D., Telleria, D., Proehl, G., Fesenko, S.,Jeffree, R., Yankovich, T., Brown, J., Higley, K., Johansen, M., Mulye, H.,Vandenhove, H., Gashchak, S., Wood, M.D., Takata, H., Andersson, P., Dale, P.,Ryan, J., Bollh€ofer, A., Doering, C., Barnett, C.L., Wells, C., 2013. The IAEAhandbook on radionuclide transfer to wildlife. J. Environ. Radioact. 121, 55e74.

IAEA, 2014. Modelling the Exposure of Wildlife to Radiation: Evaluation of CurrentApproaches and Identification of Future Requirements. Report of WorkingGroup 4. Biota Modelling Group of EMRAS II Topical Heading Reference Ap-proaches for Biota Dose Assessment Environmental Modelling for RadiationSafety (EMRAS II) Programme. Technical Document XXXX. IAEA, Vienna (inpress).

Organisation for Economic Cooperation and Development, 1997. JEF-PC version 2.0,OECD and Nuclear Energy Agency (NEA).

Srnick, M., Hrnecek, E., Steier, P., Wallner, A., Wallner, G., Bossew, P., 2008. Verticaldistribution of 238Pu, 239(240)Pu, 241Am, 90Sr and 137Cs in Austrian soilprofiles. Radiochim. Acta 96, 733e739.

US DOE, 2002. A Graded Approach for Evaluating Radiation Doses to Aquatic andTerrestrial Biota. DOE Standard DOE-STD-1153e2002.

Vives i Batlle, J., Balonov, M., Beaugelin-Seiller, K., Beresford, N.A., Brown, J.,Cheng, J.-J., Copplestone, D., Doi, M., Filistovic, V., Golikov, V., Horyna, J.,Hosseini, A., Howard, B.J., Jones, S.R., Kamboj, S., Kryshev, A., Nedveckaite, T.,Olyslaegers, G., Pr€ohl, G., Sazykina, T., Ulanovsky, A., Vives Lynch, S.,Yankovich, T., Yu, C., 2007. Inter-comparison of unweighted absorbed dose ratesfor non-human biota. Radiat. Environ. Biophys. 46, 349e373.

Vives i Batlle, J., Beaugelin-Seiller, K., Beresford, N.A., Copplestone, D., Horyna, J.,Hosseini, A., Johansen, M., Kamboj, S., Keum, D.-K., Kurosawa, N., Newsome, L.,Olyslaegers, G., Vandenhove, H., Ryufuku, S., Vives Lynch, S., Wood, M.D., Yu, C.,

K. Beaugelin-Seiller / Journal of Environmental Radioactivity 138 (2014) 60e67 67

2011. The estimation of absorbed dose rates for non-human biota: an extendedintercomparison. Radiat. Environ. Biophys. 50, 231e251.

Yankovich, T.L., Vives i Batlle, J., Vives-Lynch, S., Beresford, N.A., Barnett, C.L.,Beaugelin-Seiller, K., Brown, J.E., Cheng, J.-J., Copplestone, D., Heling, R.,

Hosseini, A., Howard, B.J., Kamboj, S., Kryshev, A.I., Nedveckaite, T.,Smith, J.T., Wood, M.D., 2010. An international model validation exercise onradionuclide transfer and doses to freshwater biota. J. Radiol. Prot. 30,299e340.