Journal of Environmental Radioactivity 175-176 (2017) 105e114

Contents lists available at ScienceDirect

Journal of Environmental Radioactivity

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

Review

Dose assessment in environmental radiological protection: State of

the art and perspectives
Karolina Stark a, *, Jose

M. Gome

z-Ros b, Jordi Vives i Batlle c, Elisabeth Lindbo Hansen d, e,
Karine Beaugelin-Seiller f, Lawrence A. Kapustka g, Michael D. Wood h, Clare Bradshaw a,
Almudena Real b, Corynne McGuire i, Thomas G. Hinton j
a
Department of Ecology, Environment, and Plant Sciences, Stockholm University, 10691 Stockholm, Sweden
b
Spanish Research Centre in Energy, Environment and Technology, CIEMAT, Avenida Complutense 40, 28040 Madrid, Spain
c
Biosphere Impact Studies Unit, Belgian Nuclear Research Centre SCK
CEN, Boeretang 200, 2400 Mol, Belgium
d
Norwegian Radiation Protection Authority, Department of Research, P.O. Box 55, NO-1332 Østerås, Norway
e
CERAD Centre of Excellence in Environmental Radioactivity, P.O. Box 5003, No-1432 Ås, Norway
f
Institut de Radioprotection et de Sûret

e Nucl
eaire, IRSN, PRP-ENV, SERIS, LRTE, Cadarache, 13115 Saint Paul Lez Durance Cedex, France
g
LK Consultancy, P.O. Box 373, Turner Valley, AB T0L 2A0, Canada
h
School of Environment and Life Sciences, University of Salford, Manchester M5 4WT, UK
i
Scottish Environment Protection Agency, Strathallan House, Castle Business Park, Stirling FK9 4TZ, UK
j
Institute of Environmental Radioactivity, Fukushima University, 1 Kanayagawa, Fukushima 960-1296, Japan

a r t i c l e i n f o a b s t r a c t

Article history: Exposure to radiation is a potential hazard to humans and the environment. The Fukushima accident
Received 12 April 2016 reminded the world of the importance of a reliable risk management system that incorporates the dose
Received in revised form received from radiation exposures. The dose to humans from exposure to radiation can be quantified
9 April 2017
using a well-defined system; its environmental equivalent, however, is still in a developmental state.
Accepted 1 May 2017
Additionally, the results of several papers published over the last decade have been criticized because of
poor dosimetry. Therefore, a workshop on environmental dosimetry was organized by the STAR (Strategy
for Allied Radioecology) Network of Excellence to review the state of the art in environmental dosimetry
Keywords:
Radiation dosimetry
and prioritize areas of methodological and guidance development. Herein, we report the key findings
Radioecology from that international workshop, summarise parameters that affect the dose animals and plants receive
Radiation exposure when exposed to radiation, and identify further research needs. Current dosimetry practices for deter-
Radiation protection mining environmental protection are based on simple screening dose assessments using knowledge of
Environmental dose assessment fundamental radiation physics, source-target geometry relationships, the influence of organism shape
and size, and knowledge of how radionuclide distributions in the body and in the soil profile alter dose.
In screening model calculations that estimate whole-body dose to biota the shapes of organisms are
simply represented as ellipsoids, while recently developed complex voxel phantom models allow organ-
specific dose estimates. We identified several research and guidance development priorities for
dosimetry. For external exposures, the uncertainty in dose estimates due to spatially heterogeneous
distributions of radionuclide contamination is currently being evaluated. Guidance is needed on the level
of dosimetry that is required when screening benchmarks are exceeded and how to report exposure in
dose-effect studies, including quantification of uncertainties. Further research is needed to establish
whether and how dosimetry should account for differences in tissue physiology, organism life stages,
seasonal variability (in ecology, physiology and radiation field), species life span, and the proportion of a
population that is actually exposed. We contend that, although major advances have recently been made
in environmental radiation protection, substantive improvements are required to reduce uncertainties
and increase the reliability of environmental dosimetry.
© 2017 Elsevier Ltd. All rights reserved.

* Corresponding author. Present address: Swedish Radiation Safety Authority, 171 16 Stockholm, Sweden.
E-mail address: [email protected] (K. Stark).

http://dx.doi.org/10.1016/j.jenvrad.2017.05.001
0265-931X/© 2017 Elsevier Ltd. All rights reserved.
106 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
2. Basic concepts and parameters in dosimetry for environmental protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
2.1. Parameters affecting radiation dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
2.2. Radiation dose in screening dose assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
3. Influence of parameters related to internal distribution of source on dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
3.1. Influence of radionuclide distribution within the body of an organism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
3.2. Relative biological effectiveness and “radiation weighting factors” in the context of protection of the environment . . . . . . . . . . . . . . . . . . . . . . 108
4. Influence of biota shape and size on dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5. Influence of parameters related to external distribution of source and location of the organism on dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.1. Influence of soil characteristics and the vertical distribution of radionuclides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
5.2. Spatial heterogeneity of the radionuclide contamination in the environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6. Influence of biota life stage and life span on dose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
7. Is the current environmental dosimetry methodology fit for purpose? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
8. Research needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

1. Introduction radiation dose received by the animals or plants (Pentreath et al.,

2015). stated that to apply the ICRP environmental protection
The potential risk posed by radionuclides in the environment approach in an actual exposure situation the dose assessment
due to uranium mining, nuclear waste repositories, routine and would ideally include: “Estimates of dose likely to be received by the
accidental discharges from nuclear plants, nuclear weapons and relevant biota under those environmental exposure situations, in
“dirty bombs” is a highly topical issue of great public interest. The terms of tissue, organs, and life stages most likely to be at risk with
2011 Fukushima-Daiichi nuclear accident in Japan reminded the regard to the relevant biological endpoints” and “the number of in-
world of how important it is to have a reliable emergency and long- dividuals, or fraction of the relevant population, that would be likely to
term risk management system for protecting humans and the receive such dose rates, and when”. The state of environmental
environment (animals and plants) from harmful exposure to ion- dosimetry has not yet reached the level of sophistication suggested
ising radiation. Such protection systems depend on reliable as- by (Pentreath et al., 2015), and recommendations to improve the
sessments of the radiation dose received. ICRP framework have been made (Brechignac et al., 2011);
The human radiation protection system is well developed and (Bradshaw et al., 2014).
internationally accepted to the point that the probability of a Dosimetry for animals and plants can be complex because of
deleterious event per unit of dose can be predicted (e.g. risk of variations in internal and external radiation exposures in both
cancer induction/milliSievert). Numerous revisions over the years space and time. Dosimetry using current screening dose assess-
have made human radiation dosimetry and the corresponding risk ment tasks that aim to demonstrate protection of the environment
estimates increasingly realistic and reliable (ICRP, 1979, 2001, 2002, from planned releases has proven to be challenging (Johansen et al.,
2004, 2007, 2012). In contrast, the environmental radiation pro- 2012; Stark et al., 2015). There are numerous environmental pa-
tection system is relatively new (ICRP, 2003, 2008, 2009; Pentreath, rameters and variabilities among species, their life cycles and at-
2012; ICRP, 2014) and radiation dosimetry for animals and plants is tributes of the environments in which they live that contribute to
not developed to the level of that used for humans (ICRP, 2008). For uncertainty in estimated doses. To determine the dose that free-
example, current knowledge is inadequate to develop probabilistic ranging animals receive as they move through environments that
risk factors that relate a dose to specific deleterious events in ani- are highly variable in contamination levels is a major challenge in
mal and plant populations. Nonetheless, international recommen- field research (Hinton et al., 2015). Unfortunately, robust dose as-
dations call for a more rigorous demonstration of radiation sessments are not always made by those reporting effects in
protection for the environment (ICRP, 2007). Thus, there is an radionuclide contaminated areas (e.g., (Hiyama et al., 2012; Moller
increased need to quantify radiation's risk to animals and plants. To et al., 2012; Akimoto, 2014; Boratynski et al., 2014; Murase et al.,
do so, reliable dosimetry is required in dose assessment tools, and 2015; Taira et al., 2015).
reliable estimates of dose are needed in studies considering effects This disparity between information needs and available dose
so that improved predictive dose-effect relationships can be data for animals and plants in the environment prompted us to
developed (Hinton et al., 2013). offer this review based on discussions from a dosimetry workshop
The International Commission on Radiological Protection's organized by the STAR Network of Excellence in radioecology
(ICRP) Committee 5 has recommended an environmental protec- (www.star-radioecology.org). The workshop examined the current
tion framework based on 12 reference animals and plants (referred state of dosimetry in environmental radiological protection and
to as RAPs) to which dose can be estimated using simplistic models included experts from 12 countries (Belgium, Canada, France,
(ICRP, 2008). The models rely heavily on empirical ratios that es- Germany, Japan, Norway, Portugal, Russia, Spain, Sweden, United
timate the radioactivity concentrations in animals and plants based Kingdom and USA). Here, we summarise the different parameters
on radioactivity concentrations in their environment (most often that affect radiation dose to animals and plants and review the state
based on radioactivity concentrations in soils). Tabulated Dose of the art in environmental dosimetry. We also respond to the
Concentration Coefficients (DCCs) are then used to convert the question: are the current screening dose assessment tools for biota
radioactivity concentration in biota or surrounding media to a fit for purpose? Finally, we formulate recommendations for dose
 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114 107

assessments, and highlight research needs. Organisms exhibit a great diversity in shapes and sizes with
diverse physiologies, all of which can vary within a species due to
2. Basic concepts and parameters in dosimetry for different life stages or seasons and can influence the received dose.
environmental protection An organism's life span will influence the total absorbed dose and
age can influence its radiosensitivity. Additionally, the location or
2.1. Parameters affecting radiation dose the geometric relationship between the radiation source and the
exposed organism is important because the intensity of the radia-
In radiological protection, absorbed dose is defined as the tion field around a source decreases with the distance.
transferred mean energy from ionising radiation to a matter
divided by the mass of that matter (ICRU, 2011). The SI unit for 2.2. Radiation dose in screening dose assessments
absorbed dose is the Joule kg1 otherwise known as Gray (Gy). The
dose rate is most often reported in units of mGy h1 or mGy d1 and To pragmatically simplify the complexity of determining dose to
can be compared to suggested environmental benchmarks [FASSET animals and plants, most screening dose assessment tools that aim
(Larsson, 2004); ERICA (Larsson, 2008); PROTECT (Howard et al., to demonstrate environmental protection from planned releases
2010)] of potential effects. Generally, the dose or dose rate esti- use dose conversion coefficients or dose conversion factors (DCC/
mates reported for animals and plants are at the level of whole DCF; mGy h1 per Bq kg1 fresh mass (fm)) to estimate internal and
organisms and rarely at the level of dose to specific organs. external dose. In the simplest case, an organism is assumed to be
Radionuclides in the environment and their progeny that emit infinitely large for internal dose estimates, infinitely small for
radiations of different type and energy lead to internal exposure of external dose estimates, and the radioactivity is distributed
animals and plants from contaminated food, water or air that they homogenously throughout its body and the medium (US DOE,
intake, and external exposure from their contaminated habitat. In 2002). Internal DCCint and external DCCext can be defined as
general, the key factors affecting radiation exposure and dose to absorbed dose rate (mGy h1) per unit activity concentration in
animals and plants can be divided into four groups: I) the internal organism (Bq kg1 fm) or medium (Bq per unit media fm). Methods
distribution of the source, II) the external distribution of the source, available to derive DCC values are described elsewhere (Taranenko
III) the characteristics of the organism, and IV) the location of the et al., 2004; Ulanovsky and Pro €hl, 2012) and tables of DCCs/DCFs for
organism with regards to exposure sources (Fig. 1). Internal radia- representative geometries of various reference animals and plants
tion exposure depends on the activity, distribution, and biological (RAPs) are available (Taranenko et al., 2004; ICRP, 2008) RAPs are
half-life of the radionuclide in organisms, and differences in phys- defined as ‘‘entities that provide a basis for the estimation of radiation
iology (tissue variations, metabolism). External radiation exposure dose rate to a range of organisms which are typical, or representative,
depends on distribution and activity concentrations in the envi- of a contaminated environment’’ (ICRP, 2008).
ronment, shielding properties of the medium, physical properties A key quantity for estimating internal absorbed dose is the
of the radionuclides, and species-specific habitat preferences. absorbed fraction (AF), defined as the fraction of energy emitted by

Fig. 1. Parameters affecting radiation dose to wildlife in a radionuclide contaminated site.

108 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114

a radiation source that is absorbed by an organism. In most ap- small organ.

proaches (such as the ERICA Tool and RESRAD-BIOTA; (US DOE, Anatomically detailed numerical phantoms have also been
2002; Brown et al., 2008, 2016), the AFs for photon and electron developed for some animals by several research groups (Kinase,
sources are calculated using Monte Carlo simulations. 2008; Ruedig et al., 2014). Most of these phantoms are voxel
In screening dose assessment models a tiered approach is used. models created from computerized tomography (CT) images, pro-
The Tier-1 assessment uses simple exposure scenario assumptions cessed and segmented to identify and localise individual organs. In
such as maximum values to purposely produce overly conservative modelling studies for clinical and radiopharmaceutical research,
dose estimates to organisms. This causes the dose rate predicted by stylized phantoms with analytical surfaces (Hindorf et al., 2004;
the model to be greater than the likely field exposure of a plant or Martinez et al., 2014) or non-uniform rational basis spline
animal. If the model-derived dose rate is below a benchmark cri- (NURBS) (Padilla et al., 2008) have been created for determination
terion of rejection, then the assessor has a high degree of confi- of absorbed dose in effect studies for mice (Hindorf et al., 2004;
dence that the real (but unknown) dose experienced by the Stabin et al., 2006; Bitar et al., 2007; Dogdas et al., 2007), rats
organism is also less and no higher tier assessment is required. (Stabin et al., 2006) and dogs (Padilla et al., 2008; Kramer et al.,
However, if the Tier-1 assessment produces dose estimates above 2012). More recently, voxel models specifically designed for RAPs
the screening benchmark then higher tiered assessments (Tier-2 in environmental protection were created and researchers have
and Tier-3) are required with increasing exposure scenario realism reported differences in the calculated AFs and DCCs when using
and site-specific measurements added to the model input values to voxels instead of ellipsoids for rat (Mohammadi et al., 2012), frog
improve the dose estimate (Brown et al., 2008). (Kinase, 2008; Mohammadi et al., 2011, 2012), crab (Caffrey and
The ERICA tool (Brown et al., 2008, 2016) does not have the Higley, 2013; Ruedig et al., 2015), rainbow trout (Martinez et al.,
capability to assess exposure to noble gases. Therefore, the England 2014; Ruedig et al., 2014, 2015), and flatfish (Ruedig et al., 2015).
and Wales Environment Agency R&D 128 method (Copplestone A comparison of whole-body doses estimated by ellipsoidal
et al., 2001); (Vives i Batlle and Jones, 2003) was recently adapted models in screening tools versus voxelised models’ considered a-
to calculate doses for all environmentally relevant Ar, Kr and Xe (210Po), b (14C, 36Cl, 90Sr/90Y) and mixed b/g- (60Co, 131I, 134Cs, 137Cs)
isotopes (Vives i Batlle et al., 2015). The assessment tool is freely emitters, heterogeneously distributed in the organs of a flatfish, in a
accessible (from https://wiki.ceh.ac.uk/x/4wWDDQ) and the trout, and in a crab (Ruedig et al., 2015). The results showed that
mathematical approach is fully described elsewhere (Copplestone differences between voxel-calculated and ellipsoid-calculated
et al., 2001; Vives i Batlle et al., 2012). To assess dose from radon whole-body dose rates agreed within a factor of two to three and
is different from the other gases, because it is not the radon (222Rn) that in most cases (exception 14C) the ellipsoidal models provide
that contributes significantly to dose, but the radon daughters, conservative values compared to the voxelised model.
whereupon dose from internal exposure becomes dominant. Radon Calculation of absorbed dose to specific organs from heteroge-
respiration needs to be explicitly modelled, as does the rapid neously distributed radionuclides is a more complex issue. For
radioactive decay producing the radon progeny, which delivers its small animals and high-energy photons it may result in dose var-
dose to the inner lining of the respiratory system. The method is iations between organs of up to one order of magnitude (Ruedig
fully described elsewhere (Vives i Batlle et al., 2008; Vives i Batlle et al., 2015). This is because the self-absorbed fraction for high
et al., 2012). energy electrons and photons can be lower than one for many or-
gans (Stabin et al., 2006), and is dependent on organ density. Stabin
3. Influence of parameters related to internal distribution of et al. (2006) calculated organ self- and cross- AFs for 10 segmented
source on dose organs in a mouse and a rat voxel model for photons and electrons
(with 0.01e4 MeV) and the variability in specific AFs were up to
3.1. Influence of radionuclide distribution within the body of an four orders of magnitude.
organism Using ellipsoids for organ dose estimates is only applicable
when calculating the absorbed dose in the source organ and cannot
Screening dose assessment models as described in Section 2 be used to determine the irradiation of adjacent organs. In some
predict dose to the whole body of ellipsoidal shaped organisms cases (high energy radiation, small animals), the contribution of
and not to internal organs, and assume a uniform distribution of cross irradiation among neighbouring organs to the total absorbed
radionuclides within the organism's body. Uncertainties in whole- dose rate experienced by an organ may not be negligible and its
body DCCs due to a possible non-homogeneous radionuclide dis- relevance should be explored.
tribution in the reference animals and plants would be less than
30% for photons and lower for electrons (Gomez-Ros et al., 2008). 3.2. Relative biological effectiveness and “radiation weighting
These uncertainties seem acceptable for conservative regulatory factors” in the context of protection of the environment
purposes of environmental protection.
Nevertheless, heterogeneous distributions in the body also The effectiveness of a given radiation to induce specific biolog-
change the organ doses when compared to an assumed homoge- ical effects in organisms depends on several factors, including the
neous distribution. Dose rates to an organ from radionuclides radiation quality that is commonly represented by the linear energy
concentrated in that organ can be significantly higher than a whole transfer (LET). LET indicates the average energy deposition along
body dose rate. An example is 131I. Unlike 137Cs, which is dispersed the radiation track at subcellular level. While g- and b-emitters are
over a large mass in the body (especially in muscles), 131I accu- classed as low LET radiations, a-emitters have high LET and are
mulates in the animal's thyroid, and thus can give a concentrated linked with a relatively higher adverse biological effect due to a
high dose to this organ. more concentrated deposition of energy in tissues. The Relative
In general, organ doses are not well characterised in environ- Biological Effectiveness (RBE) is defined as the ratio of the absorbed
mental protection. An estimation of organ dose rate can be ob- dose of a reference radiation (high-voltage X-rays or 60Co) to the
tained from a mass-ratio approximation, if the ratio of the AFs for absorbed dose of the compared radiation that is required to attain
the emitted energies is close to unity (Gomez-Ros et al., 2008; ICRP, the same biological effect. Thus, RBE is an experimentally obtained
2008; Ruedig et al., 2015). Although, this method may overestimate number to compare the damaging effect of radiation having
organ doses for high-energy emitters, especially if concentrated in a different LETs under specific conditions.
 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114 109

In human radiation protection, the radiation weighting factors energy photons (54Mn, 58Co, 60Co, 65Zn, 75Se, 95Nb, 110mAg), for
(wR) are 1 for photons (g) and electrons (b), and 20 for a-particles which DCCs can differ by up to a factor of 10 (Ulanovsky and Pro €hl,
(ICRP, 2007). Obviously, the RBE phenomenon exists in organisms 2008).
other than humans. In fact, most of the RBE information used in For terrestrial RAPs, the situation is more complex because they
human protection was derived from animal studies. However, no can be surrounded by materials with different composition and
consensus has been reached on the way to derive corresponding density. For a-and b-emitters, external exposure from soil may be
Radiation Weighting Factors (RWF) for biota in environmental neglected due to their short range in this media. For internal ex-
protection, nor the values to be applied. posures, variation of the DCCs with the RAPs size and shape de-
Several authors have reviewed the radiobiological literature pends on the energy spectrum. For instance, variation in DCCs
looking for RBE data for a emitters and tritium b-particles for between “bee” (0.589 g mass ellipsoid, 4-1.5-1.5 cm axis lengths)
deriving RWF for environmental risk assessment (SENES, 2005; and “adult deer” (245 kg mass ellipsoid, 260-120-120 cm axis
Chambers et al., 2006; ERICA, 2006; UNSCEAR, 2008; Higley lengths) is practically negligible for 3H whilst it is nearly 40% for 32P
et al., 2012). The RBE for comparable effects can vary significantly (Ulanovsky and Pro €hl, 2008).
between organism groups. For example, substantially higher RBEs For g-emitters, variability in DCCs is significantly higher. For
for fission neutrons were found for plant generic endpoints as example, internal exposure DCCs increase by about 30% between
compared to insects and mammals (Sinclair, 1985). “bee” and “adult deer” for 109Cd, a factor of 13 for 60Co, and a factor
It is clear from the literature on RBEs that only a small fraction of of 37 for 54Mn. Changes in the external exposure DCCs between
the sources is suitable for deriving RWFs for biota. Based on the “bee” and “adult deer” (both on soil surface, for a volume source of
limited valid information for biota on RBEs, several organizations soil) are not so drastic, decreasing by a factor 3 for 109Cd and a factor
recommended values of RWFs for a of 10e20 and low-energy b- 2 for 60Co and 54Mn.
emitters of 1e3 (Moiseenko et al.; FASSET, 2003; Vives i Batlle et al., The same influence of size on DCCs was found for some of the
2004; Chambers et al., 2006; UNSCEAR, 2008). The reason for ICRP RAPs exposed externally to radioisotopes of noble gases (41Ar,
applying a RWF of 3 for low-energy b-emitters is that the experi- 85
Kr, 88Kr, 131mXe and 133Xe; (Vives i Batlle et al., 2015). The largest
mental RBEs for tritium reflect elevated LET values for low energy difference, by a factor of 19, in the external exposure DCCs between
b-particles and electrons (Moiseenko et al., 2001; Vives i Batlle the smallest and largest RAPs was obtained for 85Kr, practically a
et al., 2004; Bellamy et al., 2015). pure b-emitter.

4. Influence of biota shape and size on dose 5. Influence of parameters related to external distribution of
source and location of the organism on dose
The shape and size of an organism can influence its received
dose and needs to be considered in environmental dosimetry. One 5.1. Influence of soil characteristics and the vertical distribution of
approach to estimate absorbed dose can be done as described in radionuclides
screening tools (Brown et al., 2008; Vives i Batlle et al., 2011) (in
section 2) by using a procedure based on DCCs calculated for The description of radionuclide contamination in soil/sediment
representative geometries of various animals and plants (RAPs) has been simplified to facilitate the calculation of external dose rate
(Taranenko et al., 2004; Ulanovsky and Pro €hl, 2008; Ulanovsky in screening dose assessment tools. The most important parame-
et al., 2008). ICRP's DCCs have been calculated by Monte Carlo ters influencing dose from soil is the variability of radioactivity with
simulation assuming uniform distribution of b-and g-emitters soil depth, the soil bulk density and water content. The vertical
either inside or outside the RAPs, modelled as ellipsoids (ICRP, distribution of radioactivity impacts external dose to surface or-
2008). For the ICRP RAPs, the mass of organisms varies over eight ganisms due to effects of self-attenuation. However, this aspect is
orders of magnitude (five orders of magnitude for aquatic organ- generally not considered in screening dose assessments. The most
isms and almost seven for terrestrial ones). Moreover, shapes of widespread approach in screening tools is to define soil/sediment
organisms vary from a sphere (e.g. trout egg/crab larvae) to elon- as a semi-infinite layer of given thickness for which composition
gated ellipsoids (e.g. earthworms) to flat ellipsoids (e.g. flatfish). and contamination are assumed to be homogeneous (Copplestone
Detailed tables with DCCs have been calculated for 36 RAPs et al., 2001; US DOE, 2002; Beresford et al., 2007).
(Ulanovsky and Pro €hl, 2008) considering 75 radionuclides (ICRP, The effect of a vertical heterogeneous distribution of radionu-
1983). clides in soil was investigated in the IAEA EMRAS II programme by
For aquatic organisms, calculated DCCs are related to the AFs considering a multilayer representation (Beaugelin-Seiller, 2014).
(Ulanovsky and Pro €hl, 2008; Ulanovsky et al., 2008). Calculated AFs The EDEN dosimetric tool (Beaugelin-Seiller et al., 2006) is capable
for electrons and photons in spherical organisms show a consistent of modelling dose from heterogeneous contamination in a 9-
and smooth behaviour as a function of energy and mass. For elec- layered soil profile. EDEN was applied to different combinations of
trons, the value of AF decreases with energy and increases with organism locations and radionuclides to determine external dose
mass, being practically 1 for low energies (below 100 keV) and rates to fauna in sediments and soils. A sediment scenario, inspired
organisms whose masses are above 1 kg. by a Canadian uranium mining industry scenario (IAEA, 2014), and
Variation of DCCs for aquatic RAPs depends mainly on the ra- a soil scenario, based on soil profiles with plutonium isotopes,
241
diation energy and the ratio between the mean free path or range of Am, 137Cs, and 90Sr (Srncik et al., 2008) were studied. Increasing
radiation and the size of the ellipsoid. For internal exposure, DCCs the realism of the sediment profile led to variations in external dose
are practically independent of organism shape and size for a- (e.g. rates up to several orders of magnitude for a given radionuclide
210
Po, 226Ra, Th, 231Pa, U, Np, Pu, 241Am, Cm, 252Cf) and low-energy (Beaugelin-Seiller, 2014).
(<10 keV) b-emitters (3H, 14C, 33P, 35S), where the range of the ra- The importance of calculating the external dose from sediment
diation is shorter than the dimension of the organism. As a result correctly was also demonstrated in a marine dose assessment at
the DCCs have the same values for three very different organisms: Fukushima. Here, a dynamic sediment transport model, including
“zooplankton”, “crab” and “flat fish” (ICRP, 2008). Similar behaviour vertical transport of radionuclides in the sediment, was used
is found for low energy g-emitters (e.g. 109Cd). However, variations instead of the traditional assumption of equilibrium. The dynamic
in size and shape can be important for internal exposure to high model reduced external dose rates to benthic organisms by a factor
110 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114

of ~20 (Vives i Batlle, 2016). patterns that exist over an area (Jenerette and Wu, 2010) and build
The homogeneous distribution assumption is generally appro- from classical ecology measures of communities, life-forms, and the
priate for screening assessments but, to ensure conservatism at Tier distribution and abundance of species. Such methods may help
1, if depth profile data are available then the maximal activity overcome some of the spatial and temporal variation problems
concentration in the depth profile should be used (Beaugelin- with dose assessments, and advance practices and policies of ra-
Seiller, 2014). The real contamination profile would have to be diation safety (Brechignac et al., 2011; Bradshaw et al., 2014);
considered in higher assessment tiers to make the modelled however, the landscape methods need to be tested for settings with
exposure scenario more realistic, or when calculating dose rates in radioactive contamination.
dose-effect studies. Another exposure situation where the contamination is highly
The shielding effect of water and influence on dose is well heterogeneous is the presence of discrete radioactive particles,
known. Soil moisture is an important soil property that can vary often referred to as ‘hot particles’. Current screening assessment
drastically from one day to the next, and is a user-defined param- tools are not appropriate for assessing this type of exposure sce-
eter in the ERICA Tool (Brown et al., 2008). Different assessors may nario. Deriving an activity concentration for soil when the radio-
assume a range of values (in %), which in turn directly affects the activity is in the form of discrete particles will not be representative
calculated external dose rates. This became apparent during a of a bulk sample. Also, the exposure of animals and plants is not
model comparison study using data from contaminated wetland certain to occur. In human dose assessments the probability of
ecosystems (Stark et al., 2015). The soil moisture's possible effect on exposure can be estimated based on the number of particles pre-
external DCCs was tested in the IAEA MODARIA programme sent and the habits of the individual (Dale et al., 2008). However, for
(Beaugelin-Seiller, 2016). From the 42 scenarios modelled with environmental protection where the aim is to protect populations,
varying water content, soil porosities, sizes of organism, and ra- the first question to be addressed should be if a specific method-
dionuclides, external DCCs varied from a factor of 1 (no effect) to ology is needed? For radioactive particles to have an impact on a
generally about 1.5, but reached up to a factor of 3 for pure b population they may need to be present in such high numbers that,
emitters. Thus, soil moisture effects on DCCs are generally not the contamination could be considered homogeneous and then a
significant in terms of total external dose rates to organisms, specific methodology is not required. Conversely, if the potentially
especially relative to the much larger uncertainties in transfer pa- affected population is small or endangered a specific assessment
rameters. Soil water content (in %) could be considered as a may be needed as doses to individuals become relevant. Either way,
refinement for higher assessment tiers, or in precise dose-effect at present regulatory agencies do not have the data, tools or guid-
studies, especially for burrowing organisms inhabiting the soil/ ance to make a judgement regarding whether or not a specific
sediment column. assessment of environmental exposure to particles is required.
The benefit of increasing the dosimetry complexity depends on
5.2. Spatial heterogeneity of the radionuclide contamination in the the research or assessment goals. For screening dose assessments
environment such additional complexity is not warranted; however, more ac-
curate dosimetry would be advantageous to reduce uncertainties in
The distribution of radionuclides in the environment is always studies to determine dose-effect relationships. Hinton et al. (2015)
heterogeneous, due to variations in deposition or distance from a have recently developed a combined GPS-dosimeter technique for
source, and to redistribution by physical, geochemical, and bio- free-ranging animals that enables the spatial-temporal exposure
logical processes. In general, this spatial heterogeneity is not aspects of habitat use and contaminant heterogeneity to be studied
accounted for in screening dose assessments for environmental simultaneously. In addition, two projects [the IAEA's MODARIA
protection. At best, current approaches consider spatial differences programme, http://www-ns.iaea.org/projects/modaria/default.asp,
in activity concentrations by calculating exposure for different using modelling scenarios; and the UK-funded research pro-
areas of contamination and then estimating the fraction of time gramme TREE (TRansfer - Exposure e Effects, http://tree.ceh.ac.uk/)
that an organism spends in each area based on habitat preferences using field measurements of exposure] are examining if the current
derived from the literature (Johansen et al., 2012). However, in simplistic assumptions in screening tools, which ignore how ani-
reality animals respond in a complex way in time and space to mals utilise their environment, are sufficiently protective.
differences in landscape features.
Approaches exist for making more realistic exposure estimates 6. Influence of biota life stage and life span on dose
to organisms in a heterogeneously contaminated environment. One
such approach, often called spatially explicit exposure modelling, In the current ICRP approach for radiological protection of ani-
combines spatial information about contamination and habitat mals and plants, the RAPs are chosen to cover a range of species,
types with information on animal home ranges and resources habitats, and shapes, also considering species of economic value
(habitat quality), often using geographic information systems (GIS)- and interest in research. However, the focus in environmental
based analysis. This method was used to assess doses to raccoons protection has so far been mainly on adult life stages. ICRP has
(Chow et al., 2005) and wild hogs (Gaines et al., 2005) at the included other life stages in their approach but also identified a lack
Savannah River Site, USA. Gaines et al. (2005) concluded that of data for these stages. Another exception is the specialized models
traditional methods that did not take spatial heterogeneity into for plants that have been developed, that incorporate dynamic
account overestimated radiation exposures in hogs by a factor of up activity distributions and geometries that are life-stage dependent
to 25. (Yoschenko et al., 2011; Biermans et al., 2014). The general lack of
Another group of modelling approaches, common in ecology data and models for different life-stages is a shortcoming in the
and animal behaviour studies, simulates the movement of one or current protection system because radiation exposure and in turn
more animals through the landscape. A few attempts to use such dose may change with the life stage of the organism as different life
methods have been made in ecotoxicology to model the movement stages may live in different habitats and have different diets (ICRP,
of animals through heterogeneously contaminated landscapes and 2008). For example, a fish may have pelagic juveniles but can be
estimate exposures (Loos et al., 2006; Van den Brink et al., 2007; benthic as an adult. Different life stages of a species may also have
Meli et al., 2013). different radiosensitivity with embryonic and juvenile stages often
Landscape ecology provides methods to characterize spatial being the most sensitive (Panter, 1986). As an example of a complex
 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114 111

life-cycle, consider the frog's life-cycle starting in a pond with a tool for studying limitations in the use of RAPs for regulatory pur-
shell-less egg that hatches into a swimming herbivorous larva with poses, as well as to provide additional details for site specific as-
gills. After metamorphosis the larvae becomes a lung-breathing sessments in highly contaminated areas and for dose-effect studies.
terrestrial carnivorous juvenile frog that when mature as adult For external exposures, the heterogeneous spatial distribution
will return to the pond to breed and lay new eggs. For such a life- of radionuclide contamination and how it affects dose is currently
cycle, each life stage would need its own dose assessment. being evaluated by several research teams. Also, for highly het-
Assessment tools such as the ERICA Tool allow an assessor to adjust erogeneous contamination such as radioactive particles, there is a
organism parameters, including geometry, occupancy, and radio- lack of guidance on when a specific dose assessment is required and
nuclide transfer to calculate doses to new organisms with different there is a lack of a suitable assessment methodology.
characteristics than found in the default library, so life-stage spe- Current screening dose assessments are performed on an indi-
cific dose assessments can be developed. However, this makes the vidual level for adults averaged out over a year (Brown et al., 2008,
knowledge and assumptions made by the assessor critical and re- 2016). Thus, further developments and data for other life stages in
lies on appropriate data being available (e.g. radionuclide transfer environmental dosimetry are needed to increase the abilities to
parameters for tadpoles). estimate over what time period in the life-cycle of the organism
The majority of current dose assessments calculate dose rate to that the dose will be received and also the fraction of the popula-
animals and plants and not (as for humans) the total dose received tion that will receive it. In general, dosimetry issues affected by
during a lifetime. Hence, the organism's life span is not considered. physiological differences among life stages and life span are not
A life span can vary from a few days in some plankton or insect included in current guidance, and need further attention in
species to several hundred years for some tree species. An organism dosimetry research.
with a short life span will not accumulate as high a total dose as an In environmental assessments for existing exposure situations,
organism with a long life span, and in turn short lived organisms for which activity concentrations in media and biota may be
may not accumulate as much damage as a long lived one. This can available, conservative screening dose assessments can be suffi-
be of particular importance in emergency assessments when doses cient if doses are below screening benchmarks. However, if
rates from different released radionuclides with a short half-life can benchmarks are exceeded more detailed dose assessments can be
be high. Organisms with a short life span may recover from this required including, for example, in situ measurements. For this
exposure when these radionuclides have declined during the post- there is little guidance at present. Detectors that have been
emergency phase while organisms with a long life span still live employed in the field for measuring dose to animals include opti-
with the effects from the emergency exposure. Temporal changes cally stimulated luminescence dosimeters (OSL (Reft, 2009);),
may also occur during an organism's lifespan, for example, the in- thermoluminescence dosimeters (TLDs; (Woodhead, 1973; Chesser
ternal and external dose change absolutely and relative to one et al., 2000; Beresford et al., 2008; Rodgers and Holmes, 2008; Stark
another as an organism grows, if the transfer remains constant. and Pettersson, 2008), radiophotoluminescence glass dosimeters
Thus, screening dose assessments may be misleading for the (RPLGDs; (Fuma et al., 2015; Kubota et al., 2015), and electron
assessor as they consider the (adult) organisms that get the highest paramagnetic resonance (EPR) measurements on materials such as
dose rate, but these are not necessarily the ones that receive the alanine and lithium formate (Waldeland and Malinen, 2011) or
highest total dose. Also, the reference organism that gets the hydroxyapatite (in tooth enamel and bone) (Fattibene and Callens,
highest dose rate is not necessarily the one with the most sensitive 2010; Ivanov et al., 2015). Such systems should have calibration
life stages, even though radiosensitivity is indirectly considered in certificates for the dose quantity that they measure that are
derived benchmarks. So far, there is no particular guidance on how traceable to a suitable primary standard and relevant for the
to deal with these issues in environmental dose assessments. exposure scenario (ICRU, 2006; Desrosiers et al., 2013). Assessors
would also benefit from guidance on how specific the dosimetry
7. Is the current environmental dosimetry methodology fit needs to be when screening benchmarks are exceeded.
for purpose? In effects oriented research dose assessments always need to be
detailed and reliable, simple assumptions should be avoided, and
Dosimetry for animals and plants is clearly necessary in envi- uncertainties should be accounted for. Internationally-agreed best
ronmental risk management and for regulatory purposes of plan- practice guidance and a standardisation of measuring and report-
ned exposures, existing and emergency situations, and in dose- ing dosimetry would benefit the field greatly. In laboratory exper-
effect studies including laboratory and field exposures. However, iments the radiation exposure should be verified with dosimeter
are the current dosimetry methods and tools fit for purpose? measurements or equivalent. In field effect studies it is essential to
The answer to this question depends on the goals of the dose verify the radiation dose (considering internal and external path-
assessment. In environmental risk management, where the aim is ways) and to quantify the uncertainties. To prove causality in field
to protect populations, a screening dose assessment should give a dose-effect investigations, a gradient of contaminated sites and
reliable conservative dose value for a worst case scenario (based on comparison with several control sites is preferable, provided that
simplified assumptions) that can be compared to screening associated doses are appropriately determined (i.e. considering all
benchmark values with confidence. For such screening dose as- essential pathways and the radionuclides that contribute the most
sessments our review of current dosimetry approaches in freely to received dose).
available tools (such as the ERICA Tool and RESRAD-BIOTA) shows Yet to be tested in a setting with radionuclide contamination is
that the radiation physics is well-defined and we have good to increase the ecological realism by using a landscape perspective
knowledge of how dose is affected by the shape and size of or- to consider the spatial and temporal variation in current dose as-
ganisms, heterogeneous distributions of radionuclides in the body sessments. In turn, evaluations of introduced uncertainty by
and in the soil profile, and effects of soil moisture. To improve the including more ecological parameters are then needed.
appropriate estimation of weighted absorbed doses, the RBE and In conclusion, current screening tools can calculate a conser-
appropriate RWFs for animals and plants needs to be determined. vative dose rate. However, the refinement of the assessment is
There is a need to further investigate non-uniform dose distribu- dependent on the site-specific knowledge of the assessor about
tions (organ doses) that may require modelling using advanced species present, their exposure pathways and life history traits.
methods such voxel models. Voxelised models can be useful as a Dose assessments can be complemented with dosimeter
112 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114

measurements. However, substantive improvements in environ- Norway, contract numbers: 209101 and 209102.
mental dosimetry are warranted to overcome the remaining un-
certainties connected to the ecology and physiology of animals and References
plants so that the reliability increases of the environmental pro-
tection system. The level of dosimetry refinement depends on the Akimoto, S., 2014. Morphological abnormalities in gall-forming aphids in a
radiation-contaminated area near Fukushima Daiichi: selective impact of
objective of the dose assessment. We recommend that when per- fallout? Ecol. Evol. 4, 355e369.
forming a dose assessment for animals and plants that goes beyond Beaugelin-Seiller, K., 2014. The assumption of heterogeneous or homogeneous
a Tier-1 screening dose assessment one should ideally: radioactive contamination in soil/sediment: does it matter in terms of the
external exposure of fauna? J. Environ. Radioact. 138, 60e67.
Beaugelin-Seiller, K., 2016. Effects of soil water content on the external exposure of
➢ have knowledge about the life-cycle and all essential exposure fauna to radioactive isotopes. J. Environ. Radioact. 151, 204e208.
pathways for the organism, Beaugelin-Seiller, K., Jasserand, F., Garnier-Laplace, J., Gariel, J.C., 2006. Modeling
radiological dose in non-human species: principles, computerization, and
➢ know and consider the seasonal habitat preferences and
application. Health Phys. 90, 485e493.
movement patterns of the study organisms, Bellamy, M., Puskin, J., Hertel, N., Eckerman, K., 2015. An empirical method for
➢ have knowledge of which are the most sensitive organisms at deriving RBE values associated with electrons, photons and radionuclides.
the site (e.g. protected species, most radiosensitive species) and Radiat. Prot. Dosim. 167, 664e670.
Beresford, N.A., Brown, J., Copplestone, D., Garnier-Laplace, J., Howard, B.,
if they have sensitive life stages, Larsson, C.-M., Oughton, D., Pro € hl, G., Zinger, I., 2007. A deliverable of the ERICA
➢ take into account the radiation field (i.e., the external and in- project FI6R-CT-2004e508847. D-ERICA: an Integrated Approach to the
ternal distribution of the radiation source, source strength, type Assessment and Management of Environmental Risks from Ionising Radiation,
vol. 88. Available from: https://wiki.ceh.ac.uk/download/attachments/
of radiation emitted, etc.) where the organism lives, 115017395/D-Erica.pdf?version¼1 [Accessed 6 March 2015].
➢ calculate and/or measure the radiation dose to the organisms in Beresford, N.A., Gaschak, S., Barnett, C.L., Howard, B.J., Chizhevsky, I., Stromman, G.,
their particular habitats, considering temporal and spatial Oughton, D.H., Wright, S.M., Maksimenko, A., Copplestone, D., 2008. Estimating
the exposure of small mammals at three sites within the Chernobyl exclusion
heterogeneity. zone - a test application of the ERICA Tool. J. Environ. Radioact. 99, 1496e1502.
Biermans, G., Horemans, N., Hens, N., Vives i Batlle, J., Vandenhove, H., Cuypers, A.,
2014. A dynamic dosimetry model for radioactive exposure scenarios in Ara-
bidopsis thaliana. J. Theor. Biol. 347, 54e62.
8. Research needs Bitar, A., Lisbona, A., Thedrez, P., Maurel, C.S., Le Forestier, D., Barbet, J., Bardies, M.,
2007. A voxel-based mouse for internal dose calculations using Monte Carlo
In this review we identified the following issues for which simulations (MCNP). Phys. Med. Biol. 52, 1013e1025.
Boratynski, Z., Lehmann, P., Mappes, T., Mousseau, T.A., Moller, A.P., 2014. Increased
further research is needed to reduce or quantify uncertainties in
radiation from Chernobyl decreases the expression of red colouration in natural
environmental dosimetry: populations of bank voles (Myodes glareolus). Sci. Rep. 4, 4.
Bradshaw, C., Kapustka, L., Barnthouse, L., Brown, J., Ciffroy, P., Forbes, V.,
➢ Improvement of methods for assessing when doses will be Geras'kin, S., Kautsky, U., Brechignac, F., 2014. Using an Ecosystem Approach to
complement protection schemes based on organism-level endpoints. J. Environ.
received and what fraction of the population will be exposed, for Radioact. 136, 98e104.
specific species considered in the assessment Brechignac, F., Bradshaw, C., Carroll, S., Jaworska, A., Kapustka, L., Monte, L.,
➢ Development of guidance on how to measure dose to animals Oughton, D., 2011. Recommendations from the International Union of Radio-
ecology to improve guidance on radiation protection. Integr. Environ. Assess.
and plants and development of a standardized system for how Manag. 7, 411e413.
to report dosimetry methods and dose results Brown, J.E., Alfonso, B., Avila, R., Beresford, N.A., Copplestone, D., Pro €hl, G.,
➢ Development of guidance on what level of additional dosimetry Ulanovsky, A., 2008. The ERICA tool. J. Environ. Radioact. 99, 1371e1383.
Brown, J.E., Alfonso, B., Avila, R., Beresford, N.A., Copplestone, D., Hosseini, A., 2016.
is required when screening benchmarks are exceeded A new version of the ERICA tool to facilitate impact assessments of radioactivity
➢ Further investigations of dosimetry issues affected by physio- on wild plants and animals. J. Environ. Radioact. 153, 141e148.
logical differences among tissues, life stages, seasonal changes, Caffrey, E.A., Higley, K.A., 2013. Creation of a voxel phantom of the ICRP reference
crab. J. Environ. Radioact. 120, 14e18.
and effects of a species' life span
Chambers, D.B., Osborne, R.V., Garva, A.L., 2006. Choosing an alpha radiation
➢ Further exploration of several biophysical factors for high LET weighting factor for doses to non-human biota. J. Environ. Radioact. 87, 1e14.
radiation, and especially at low levels of exposure, such as the Chesser, R.K., Sugg, D.W., Lomakin, M.D., Van den Bussche, R.A., DeWoody, J.A.,
Jagoe, C.H., Dallas, C.E., Whicker, F.W., Smith, M.H., Gaschak, S.P.,
heterogeneity of deposited radionuclides in a tissue or organ
Chizhevsky, I.V., Lyabik, V.V., Buntova, E.G., Holloman, K., Baker, R.J., 2000.
➢ Exploration of the contribution of cross irradiation from Concentrations and dose rate estimates of (134,137)cesium and (90)strontium
neighbouring organs to the total absorbed dose rate to an organ in small mammals at Chornobyl, Ukraine. Environ. Toxicol. Chem. 19, 305e312.
➢ Testing whether greater ecological realism using a landscape Chow, T.E., Gaines, K.F., Hodgson, M.E., Wilson, M.D., 2005. Habitat and exposure
modelling for ecological risk assessment: a case study for the raccoon on the
perspective to incorporate spatial and temporal variability, Savannah River Site. Ecol. Model 189, 151e167.
would improve current dose assessments Copplestone, D., Bielby, S., Jones, S.R., Patton, D., Daniel, C.P., Gize, I., 2001. Impact
➢ Development of guidance for highly heterogeneous contami- Assessment of Ionising Radiation on Wildlife, vol. 128. R&D Publication (Envi-
ronment Agency, UK and English Nature).
nation such as in the case of radioactive particles (e.g. when is an Dale, P., Robertson, I., Toner, M., 2008. Radioactive particles in dose assessments.
assessment required) and development of a suitable assessment J. Environ. Radioact. 99, 1589e1595.
methodology Desrosiers, M., DeWerd, L., Deye, J., Lindsay, P., Murphy, M.K., Mitch, M.,
Macchiarini, F., Stojadinovic, S., Stone, H., 2013. The importance of dosimetry
standardization in radiobiology. J. Res. Natl. Inst. Stand. Technol. 118, 16.
Acknowledgements Dogdas, B., Stout, D., Chatziioannou, A.F., Leahy, R.M., 2007. Digimouse: a 3D whole
body mouse atlas from CT and cryosection data. Phys. Med. Biol. 52, 577e587.
ERICA, 2006. Environmental risk for ionising contaminants: assessment and man-
We acknowledge the contributions to discussions from N. agement. In: Garnier-Laplace, J., Gilbin, R. (Eds.), Derivation of Predicted No-
Beresford, L. Braby, J. Brown, C. Burbidge, D. Copplestone, K. Higley, effect Dose Rates Values for Ecosystems and Their Sub-organisational Level
V. Ioshchenko, S. Kinase, A. Kryshev, H. Rabus, E. Shishkina, M. Exposed to Radioactive Substances. Deliverable 5. European Commission, 6th
Framework.
Steiner, R. Tanner, and A. Ulanovsky during the dosimetry work-
FASSET, 2003. Dosimetric Models and Data for Assessing Radiation Exposures to
shop. We are thankful to two anonymous reviewers whose sug- Biota. Deliverable 3, FASSET report.
gestions improved the manuscript. This work was performed Fattibene, P., Callens, F., 2010. EPR dosimetry with tooth enamel: a review. Appl.
within the STAR (Strategy for Allied Radioecology) Network of Radiat. Isot. 68, 2033e2116.
Fuma, S., Ihara, S., Kawaguchi, I., Ishikawa, T., Watanabe, Y., Kubota, Y., Sato, Y.,
Excellence supported by the European Commission, contract Takahashi, H., Aono, T., Ishii, N., Soeda, H., Matsui, K., Une, Y., Minamiya, Y.,
number: Fission-2010-3.5.1-269672, and the Research Council of Yoshida, S., 2015. Dose rate estimation of the Tohoku hynobiid salamander,
 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114 113

Hynobius lichenatus, in Fukushima. J. Environ. Radioact. 143, 123e134. Prot. 24, A1.
Gaines, K.F., Porter, D.E., Punshon, T., Brisbin, I.L., 2005. A spatially explicit model of Larsson, C.M., 2008. An overview of the ERICA Integrated Approach to the assess-
the wild hog for ecological risk assessment activities at the department of ment and management of environmental risks from ionising contaminants.
energy's Savannah river site. Hum. Ecol. Risk Assess. 11, 567e589. J. Environ. Radioact. 99, 1364e1370.
Gomez-Ros, J.M., Pro €hl, G., Ulanovsky, A., Lis, M., 2008. Uncertainties of internal Loos, M., Ragas, A.M.J., Schipper, A.M., Lopes, J.P.C., 2006. A Spatially Explicit
dose assessment for animals and plants due to non-homogeneously distributed Individual-based Controlled Random Walk Model to Determine Exposure
radionuclides. J. Environ. Radioact. 99, 1449e1455. Levels and Risks of Environmental Contaminants for Terrestrial Organisms in
Higley, K.A., Kocher, D.C., Real, A.G., Chambers, D.B., 2012. Relative biological River Floodplains. NoMiracle report Deliverable 4.2.1., the Netherlands, p. 82.
effectiveness and radiation weighting factors in the context of animals and Martinez, N.E., Johnson, T.E., Capello, K., Pinder III, J.E., 2014. Development and
plants. Ann. ICRP 41, 233e245. comparison of computational models for estimation of absorbed organ radia-
Hindorf, C., Ljungberg, M., Strand, S.E., 2004. Evaluation of parameters influencing S tion dose in rainbow trout (Oncorhynchus mykiss) from uptake of iodine-131.
values in mouse dosimetry. J. Nucl. Med. 45, 1960e1965. J. Environ. Radioact. 138, 50e59.
Hinton, T.G., Garnier-Laplace, J., Vandenhove, H., Dowdall, M., Adam-Guillermin, C., Meli, M., Auclerc, A., Palmqvist, A., Forbes, V.E., Grimm, V., 2013. Population-level
Alonzo, F., Barnett, C., Beaugelin-Seiller, K., Beresford, N.A., Bradshaw, C., consequences of spatially heterogeneous exposure to heavy metals in soil: an
Brown, J., Eyrolle, F., Fevrier, L., Gariel, J.C., Gilbin, R., Hertel-Aas, T., individual-based model of springtails. Ecol. Model 250, 338e351.
Horemans, N., Howard, B.J., Ikaheimonen, T., Mora, J.C., Oughton, D., Real, A., Mohammadi, A., Kinase, S., Saito, K., 2011. Comparison of photon and electron
Salbu, B., Simon-Cornu, M., Steiner, M., Sweeck, L., Batlle, J.V.I., 2013. An invi- absorbed fractions in voxel-based and simplified phantoms for small animals.
tation to contribute to a strategic research agenda in radioecology. J. Environ. Prog. Nucl. Sci. Technol. 2, 365e368.
Radioact. 115, 73e82. Mohammadi, A., Kinase, S., Saito, K., 2012. Evaluation of absorbed doses in voxel-
Hinton, T.G., Byrne, M.E., Webster, S., Beasley, J.C., 2015. Quantifying the spatial and based and simplified models for small animals. Radiat. Prot. Dosim. 150,
temporal variation in dose from external exposure to radiation: a new tool for 283e291.
use on free-ranging wildlife. J. Environ. Radioact. 145, 58e65. Moiseenko, V.V., Waker, J.A., Hamm, N.R., Prestwich, V.W., 2001. Calculation of
Hiyama, A., Nohara, C., Kinjo, S., Taira, W., Gima, S., Tanahara, A., Otaki, J.M., 2012. radiation-induced DNA damage from photons and tritium beta-particles.
The biological impacts of the Fukushima nuclear accident on the pale grass blue Radiat. Environ. Biophys. 40, 33e38.
butterfly. Sci. Rep. 2, 10. Moller, A.P., Hagiwara, A., Matsui, S., Kasahara, S., Kawatsu, K., Nishiumi, I.,
Howard, B.J., Beresford, N.A., Andersson, P., Brown, J.E., Copplestone, D., Beaugelin- Suzuki, H., Ueda, K., Mousseau, T.A., 2012. Abundance of birds in Fukushima as
Seiller, K., Garnier-Laplace, J., Howe, P.D., Oughton, D., Whitehouse, P., 2010. judged from chernobyl. Environ. Pollut. 164, 36e39.
Protection of the environment from ionising radiation in a regulatory context- Murase, K., Murase, J., Horie, R., Endo, K., 2015. Effects of the Fukushima Daiichi
an overview of the PROTECT coordinated action project. J. Radiol. Prot. 30, nuclear accident on goshawk reproduction. Sci. Rep. 5, 7.
195e214. Padilla, L., Lee, C., Milner, R., Shahlaee, A., Bolch, W.E., 2008. Canine anatomic
IAEA, 2014. Modelling the Exposure of Wildlife to Radiation: Evaluation of Current phantom for preclinical dosimetry in internal emitter therapy. J. Nucl. Med. 49
Approaches and Identification of Future Requirements. Report of Working (3), 446e452.
Group 4. Biota Modelling Group of EMRAS II Topical Heading Reference Ap- Panter, H.C., 1986. Variations in radiosensitivity during development of the frog
proaches for Biota Dose Assessment. Environmental Modelling for Radiation Limnodynastes tasmaniensis. J. Exp. Zool. 238, 193e199.
Safety (EMRAS II) Programme. Technical Document. IAEA, Vienna. Pentreath, R.J., 2012. Radiation and protection of the environment: the work of
ICRP, 1979. Radionuclide Release into the Environment - Assessment of Doses to Committee 5. Ann. ICRP 41, 45e56.
Man, vol. 29. ICRP publication. Annals of the ICRP 2. Pentreath, R.J., Larsson, C.-M., Copplestone, D., 2015. ICRP's approach to protection
ICRP, 1983. Radionuclide Transformations - Energy and Intensity of Transmissions, of the living environment under different exposure situations. Ann. ICRP 44.
vol. 38. ICRP Publication, Oxford (Annals of the ICRP 11), Pergamon Press. Reft, C.S., 2009. The energy dependence and dose response of a commercial opti-
ICRP, 2001. Annals of the ICRP 31. Doses to the Embryo and Fetus from Intakes of cally stimulated luminescent detector for kilovoltage photon, megavoltage
Radionuclides by the Mother, vol. 88. ICRP publication. photon, and electron, proton, and carbon beams. Med. Phys. 36, 1690e1699.
ICRP, 2002. Annals of the ICRP 32. Basic Anatomical and Physiological Data for Use Rodgers, B.E., Holmes, K.M., 2008. Radio-adaptive response to environmental ex-
in Radiological Protection Reference Values, vol. 89. ICRP publication. posures at Chernobyl. Dose-Response 6, 209e221.
ICRP, 2003. Annals of the ICRP 33. A Framework for Assessing the Impact of Ionising Ruedig, E., Caffrey, E., Hess, C., Higley, K., 2014. Monte Carlo derived absorbed
Radiation on Non-human Species, vol. 91. ICRP publication. fractions for a voxelized model of Oncorhynchus mykiss, a rainbow trout.
ICRP, 2004. Annals of the ICRP 34. Doses to Infants from Ingestion of Radionuclides Radiat. Environ. Biophys. 53, 581.
in Mothers' Milk, vol. 95. ICRP publication. Ruedig, E., Beresford, N.A., Gomez Fernandez, M.E., Higley, K., 2015. A comparison of
ICRP, 2007. Annals of the ICRP 37. The 2007 Recommendations of the International the ellipsoidal and voxelized dosimetric methodologies for internal, heteroge-
Commission on Radiological Protection, vol. 103. ICRP publication. neous radionuclide sources. J. Environ. Radioact. 140, 70e77.
ICRP, 2008. Annals of the ICRP 38. Environmental Protection - the Concept and Use SENES, 2005. Review and Evaluation of Published Literature on Alpha Radiation
of Reference Animals and Plants, vol. 108. ICRP publication. Weighting Factors for Non-human Biota. SENES Consultants Limited, Richmond
ICRP, 2009. Annals of the ICRP 39.. Environmental Protection: Transfer Parameters Hill, Ontario.
for Reference Animals and Plants, vol. 114. ICRP publication. Sinclair, W.K., 1985. Experimental RBE values of high let radiations at low doses and
ICRP, 2012. Annals of the ICRP 41. Compendium of Dose Coefficients Based on ICRP the implications for quality factor assignment. Radiat. Prot. Dosim. 13, 319e326.
Publication 60, vol. 119. ICRP publication, p. 129. Srncik, M., Hrnecek, E., Steier, P., Wallner, A., Wallner, G., Bossew, P., 2008. Vertical
ICRP, 2014. Annals of the ICRP 43. Protection of the Environment under Different distribution of 238Pu, 239(40)Pu, 241Am, 90Sr and 137Cs in Austrian soil
Exposure Situations, vol. 124. ICRP publication. profiles. Ract 96, 733.
ICRU, 2006. Measurement quality assurance for ionizing radiation dosimetry. ICRU Stabin, M.G., Peterson, T.E., Holburn, G.E., Emmons, M.A., 2006. Voxel-based mouse
Report 76 J. ICRU 6 (NP). and rat models for internal dose calculations. J. Nucl. Med. 47, 655e659.
ICRU, 2011. Fundamental quantities and units for ionizing radiation. ICRU Report 85 Stark, K., Pettersson, H.B.L., 2008. External radiation doses from Cs-137 to frog
J. ICRU 11. NP. phantoms in a wetland area: in situ measurements and dose model calcula-
Ivanov, D.V., Shishkina, E.A., Osipov, D.I., Razumeev, R.A., Pryakhin, E.A., 2015. In- tions. Radiat. Environ. Biophys. 47, 481e489.
ternal in vitro dosimetry for fish using hydroxyapatite-based EPR detectors. Stark, K., Andersson, P., Beresford, N.A., Yankovich, T.L., Wood, M.D., Johansen, M.P.,
Radiat. Environ. Biophys. 54, 257e263. Batlle, J.V.I., Twining, J., Keum, D.K., Bollhofer, A., Doering, C., Ryan, B.,
Jenerette, G.D., Wu, J., 2010. Quantitative measures and ecological hierarchy. In: Grzechnik, M., Vandenhove, H., 2015. Predicting exposure of wildlife in radio-
Kapustka, L., Landis, W.G. (Eds.), Environmental Risk Assessment and Man- nuclide contaminated wetland ecosystems. Environ. Pollut. 196, 201e213.
agement from a Landscape Perspective. John Wiley and Sons, Inc., New York, Taira, W., Hiyama, A., Nohara, C., Sakauchi, K., Otaki, J.M., 2015. Ingestional and
NY, USA, p. 396. transgenerational effects of the Fukushima nuclear accident on the pale grass
Johansen, M.P., Barnett, C.L., Beresford, N.A., Brown, J.E., Cerne, M., Howard, B.J., blue butterfly. J. Radiat. Res. 56, i2ei18.
Kamboj, S., Keum, D.K., Smodis, B., Twining, J.R., Vandenhove, H., Batlle, J.V.I., Taranenko, V., Prohl, G., Gomez-Ros, J.M., 2004. Absorbed dose rate conversion
Wood, M.D., Yu, C., 2012. Assessing doses to terrestrial wildlife at a radioactive coefficients for reference terrestrial biota for external photon and internal ex-
waste disposal site: inter-comparison of modelling approaches. Sci. Total En- posures. J. Radiol. Prot. 24, A35eA62.
viron. 427, 238e246. Ulanovsky, A., Pro €hl, G., 2008. Tables of dose conversion coefficients for estimating
Kinase, S., 2008. Voxel-based frog phantom for internal dose evaluation. J. Nucl. Sci. internal and external radiation exposures to terrestrial and aquatic biota.
Technol. 45, 1049e1052. Radiat. Environ. Biophys. 47, 195e203.
Kramer, G.H., Capello, K., Strocchi, S., Bearrs, B., Leung, K., Martinez, N., 2012. The Ulanovsky, A., Pro €hl, G., 2012. Dosimetry for reference animals and plants: current
HML's new voxel phantoms: two human males, one human female, and two state and prospects. Ann. ICRP 41, 218e232.
male canines. Health Phys. 103, 802e807. Ulanovsky, A., Pro €hl, G., Gomez-Ros, J.M., 2008. Methods for calculating dose con-
Kubota, Y., Takahashi, H., Watanabe, Y., Fuma, S., Kawaguchi, I., Aoki, M., Kubota, M., version coefficients for terrestrial and aquatic biota. J. Environ. Radioact. 99,
Furuhata, Y., Shigemura, Y., Yamada, F., Ishikawa, T., Obara, S., Yoshida, S., 2015. 1440e1448.
Estimation of absorbed radiation dose rates in wild rodents inhabiting a site UNSCEAR, 2008. Annex E: Effects of Non-ionizing Radiation on Non-human Biota.
severely contaminated by the Fukushima Dai-ichi nuclear power plant accident. United Nations sales publication.
J. Environ. Radioact. 142, 124e131. US DOE, 2002. A Graded Approach for Evaluating Radiation Doses to Aquatic and
Larsson, C.-M., 2004. The FASSET Framework for assessment of environmental Terrestrial Biota. DOE Standard. US DOE.
impact of ionising radiation in European ecosystemsdan overview. J. Radiol. Van den Brink, P.J., Baveco, J.M., Verboom, J., Heimbach, F., 2007. An individual-
114 K. Stark et al. / Journal of Environmental Radioactivity 175-176 (2017) 105e114

based approach to model spatial population dynamics of invertebrates in generic plant geometry. Radiat. Environ. Biophys. 50, 513e529.
aquatic ecosystems after pesticide contamination. Environ. Toxicol. Chem. 26, Vives i Batlle, J., Copplestone, D., Jones, S.R., Williams, C., 2012. Allometric meth-
2226e2236. odology for the assessment of radon exposures to wildlife. Sci. Total Environ.
Vives i Batlle, J., 2016. Dynamic modelling of radionuclide uptake by marine biota: 427e428, 50e59.
application to the Fukushima nuclear power plant accident. J. Environ. Radioact. Vives i Batlle, J., Jones, S.R., Copplestone, D., 2015. A methodology for 41Ar, 85,88Kr
151, 502e511. and 131m,133Xe wildlife dose assessment. J. Environ. Radioact. 144, 152e161.
Vives i Batlle, J., Jones, S.R., 2003. A methodology for the assessment of doses to Waldeland, E., Malinen, E., 2011. Review of the dose-to-water energy dependence of
terrestrial biota arising from external exposure to 41Ar and 85Kr. (Accompanying alanine and lithium formate EPR dosimeters and LiF TL-dosimeters e com-
CD). In: Copplestone, D., et al. (Eds.), Impact Assessment of Ionising Radiation parison with Monte Carlo simulations. Radiat. Meas. 46, 945e951.
on Wildlife, vol. 128. Environment Agency R&D Publication. March 2003 up- Woodhead, D.S., 1973. Radiation-dose received by plaice (Pleuronectes platessa)
date. Bristol, Environment Agency. from waste discharged into Northeast Irish sea from fuel reprocessing plant at

mez-Ros, J.M., 2004. A method for calculation of dose
Vives i Batlle, J., Jones, S.R., Go Windscale. Health Phys. 25, 115.
per unit concentration values for aquatic biota. J. Radiol. Prot. 24, A13. Yoschenko, V.I., Kashparov, V.A., Melnychuk, M.D., Levchuk, S.E., Bondar, Y.O.,
Vives i Batlle, J., Jones, S.R., Copplestone, D., 2008. Dosimetric Model for Biota Lazarev, M., Yoschenko, M.I., Farfa
n, E.B., Jannik, G.T., 2011. Chronic irradiation of
Exposure to Inhaled Radon Daughters. Environment Agency Science Report scots pine trees (Pinus Sylvestris) in the Chernobyl exclusion zone: dosimetry
SC060080, 34 pp. and radiobiological effects. Health Phys. 101, 393e408, 310.1097/
Vives i Batlle, J., Smith, A.D., Vives-Lynch, S., Copplestone, D., Prohl, G., Strand, T., HP.1090b1013e3182118094.
2011. Model-derived dose rates per unit concentration of radon in air in a